2023/09/06 14:32:14 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.8.17 (default, Jul 5 2023, 21:04:15) [GCC 11.2.0] CUDA available: True numpy_random_seed: 1506048241 GPU 0,1,2,3,4,5,6,7: Tesla V100-SXM2-32GB CUDA_HOME: /usr/local/cuda NVCC: Cuda compilation tools, release 11.2, V11.2.152 GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) PyTorch: 1.9.0+cu111 PyTorch compiling details: PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.1.2 (Git Hash 98be7e8afa711dc9b66c8ff3504129cb82013cdb) - OpenMP 201511 (a.k.a. OpenMP 4.5) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.1 - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86 - CuDNN 8.0.5 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.9.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, TorchVision: 0.10.0+cu111 OpenCV: 4.8.0 MMEngine: 0.8.4 Runtime environment: cudnn_benchmark: True mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: 1506048241 Distributed launcher: pytorch Distributed training: True GPU number: 4 ------------------------------------------------------------ 2023/09/06 14:32:14 - mmengine - INFO - Config: crop_size = ( 640, 640, ) data_preprocessor = dict( bgr_to_rgb=True, mean=[ 122.7709, 116.746, 104.0937, ], pad_val=0, seg_pad_val=255, size_divisor=640, std=[ 68.5005, 66.6322, 70.3232, ], test_cfg=dict(size_divisor=32), type='SegDataPreProcessor') data_root = 'data/coco_stuff164k' dataset_type = 'COCOStuffDataset' default_hooks = dict( checkpoint=dict( by_epoch=False, interval=10000, save_best='mIoU', type='CheckpointHook'), logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), param_scheduler=dict(type='ParamSchedulerHook'), sampler_seed=dict(type='DistSamplerSeedHook'), timer=dict(type='IterTimerHook'), visualization=dict(type='SegVisualizationHook')) default_scope = 'mmseg' env_cfg = dict( cudnn_benchmark=True, dist_cfg=dict(backend='nccl'), mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) img_ratios = [ 0.5, 0.75, 1.0, 1.25, 1.5, 1.75, ] launcher = 'pytorch' load_from = None log_level = 'INFO' log_processor = dict(by_epoch=False) model = dict( asymetric_input=True, data_preprocessor=dict( bgr_to_rgb=True, mean=[ 122.7709, 116.746, 104.0937, ], pad_val=0, seg_pad_val=255, size_divisor=640, std=[ 68.5005, 66.6322, 70.3232, ], test_cfg=dict(size_divisor=32), type='SegDataPreProcessor'), decode_head=dict( align_corners=False, deep_supervision_idxs=[ 7, ], loss_decode=[ dict( class_weight=[ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.1, ], loss_name='loss_cls_ce', loss_weight=2.0, naive_reduction=True, type='CrossEntropyLoss'), dict( loss_name='loss_mask_ce', loss_weight=5.0, type='CrossEntropyLoss', use_sigmoid=True), dict( eps=1, ignore_index=None, loss_name='loss_mask_dice', loss_weight=5.0, naive_dice=True, type='DiceLoss'), ], maskgen_cfg=dict( act_cfg=dict(type='QuickGELU'), cross_attn=False, embed_dims=768, final_norm=True, frozen_exclude=[], mlp_ratio=4, norm_cfg=dict(eps=1e-05, type='LN'), num_heads=12, num_layers=3, out_dims=512, qkv_bias=True, sos_token_format='cls_token', sos_token_num=100), num_classes=171, san_cfg=dict( cfg_decoder=dict( embed_channels=256, mlp_channels=256, num_heads=12, num_layers=1, num_mlp=3, rescale=True), cfg_encoder=dict(mlp_ratio=4, num_encode_layer=8, num_heads=6), clip_channels=768, embed_dims=240, fusion_index=[ 0, 1, 2, 3, ], in_channels=3, norm_cfg=dict(eps=1e-06, type='LN'), num_queries=100, patch_bias=True, patch_size=16), train_cfg=dict( assigner=dict( match_costs=[ dict(type='ClassificationCost', weight=2.0), dict( type='CrossEntropyLossCost', use_sigmoid=True, weight=5.0), dict(eps=1.0, pred_act=True, type='DiceCost', weight=5.0), ], type='HungarianAssigner'), importance_sample_ratio=0.75, num_points=12544, oversample_ratio=3.0), type='SideAdapterCLIPHead'), encoder_resolution=0.5, image_encoder=dict( act_cfg=dict(type='QuickGELU'), attn_drop_rate=0.0, drop_path_rate=0.0, drop_rate=0.0, embed_dims=768, frozen_exclude=[ 'pos_embed', ], img_size=( 224, 224, ), in_channels=3, interpolate_mode='bicubic', mlp_ratio=4, norm_cfg=dict(eps=1e-05, type='LN'), norm_eval=False, num_heads=12, num_layers=9, out_indices=( 2, 5, 8, ), out_origin=True, output_cls_token=True, patch_bias=False, patch_pad=0, patch_size=16, pre_norm=True, qkv_bias=True, type='VisionTransformer', with_cls_token=True), pretrained='pretrain/clip_vit_base_patch16_224.pth', test_cfg=dict(mode='whole'), text_encoder=dict( cache_feature=True, cat_bg=True, dataset_name='coco-stuff164k', embed_dims=512, mlp_ratio=4, norm_cfg=dict(eps=1e-05, type='LN'), num_heads=8, num_layers=12, output_dims=512, templates='vild', type='CLIPTextEncoder'), train_cfg=dict(), type='MultimodalEncoderDecoder') norm_cfg = dict(requires_grad=True, type='SyncBN') num_classes = 171 optim_wrapper = dict( clip_grad=dict(max_norm=0.01, norm_type=2), loss_scale='dynamic', optimizer=dict( betas=( 0.9, 0.999, ), lr=0.0001, type='AdamW', weight_decay=0.0001), paramwise_cfg=dict( custom_keys=dict( cls_token=dict(decay_mult=0.0), img_encoder=dict(decay_mult=1.0, lr_mult=0.1), norm=dict(decay_mult=0.0), pos_embed=dict(decay_mult=0.0))), type='AmpOptimWrapper') optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) param_scheduler = [ dict( begin=0, by_epoch=False, end=60000, eta_min=0.0, power=1.0, type='PolyLR'), ] resume = False test_cfg = dict(type='TestLoop') test_dataloader = dict( batch_size=1, dataset=dict( data_prefix=dict( img_path='images/val2017', seg_map_path='annotations/val2017'), data_root='data/coco_stuff164k', pipeline=[ dict(type='LoadImageFromFile'), dict(max_size=2560, scale=( 640, 640, ), type='ResizeShortestEdge'), dict(type='LoadAnnotations'), dict(type='PackSegInputs'), ], type='COCOStuffDataset'), num_workers=4, persistent_workers=True, sampler=dict(shuffle=False, type='DefaultSampler')) test_evaluator = dict( iou_metrics=[ 'mIoU', ], type='IoUMetric') test_pipeline = [ dict(type='LoadImageFromFile'), dict(max_size=2560, scale=( 640, 640, ), type='ResizeShortestEdge'), dict(type='LoadAnnotations'), dict(type='PackSegInputs'), ] train_cfg = dict( max_iters=60000, type='IterBasedTrainLoop', val_begin=55000, val_interval=500) train_dataloader = dict( batch_size=8, dataset=dict( data_prefix=dict( img_path='images/train2017', seg_map_path='annotations/train2017'), data_root='data/coco_stuff164k', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), dict( max_size=2560, resize_type='ResizeShortestEdge', scales=[ 320, 384, 448, 512, 576, 640, 704, 768, 832, 896, 960, 1024, ], type='RandomChoiceResize'), dict(cat_max_ratio=1.0, crop_size=( 640, 640, ), type='RandomCrop'), dict(type='PhotoMetricDistortion'), dict(prob=0.5, type='RandomFlip'), dict(type='PackSegInputs'), ], type='COCOStuffDataset'), num_workers=4, persistent_workers=True, sampler=dict(shuffle=True, type='InfiniteSampler')) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations'), dict( max_size=2560, resize_type='ResizeShortestEdge', scales=[ 320, 384, 448, 512, 576, 640, 704, 768, 832, 896, 960, 1024, ], type='RandomChoiceResize'), dict(cat_max_ratio=1.0, crop_size=( 640, 640, ), type='RandomCrop'), dict(type='PhotoMetricDistortion'), dict(prob=0.5, type='RandomFlip'), dict(type='PackSegInputs'), ] tta_model = dict(type='SegTTAModel') tta_pipeline = [ dict(backend_args=None, type='LoadImageFromFile'), dict( transforms=[ [ dict(keep_ratio=True, scale_factor=0.5, type='Resize'), dict(keep_ratio=True, scale_factor=0.75, type='Resize'), dict(keep_ratio=True, scale_factor=1.0, type='Resize'), dict(keep_ratio=True, scale_factor=1.25, type='Resize'), dict(keep_ratio=True, scale_factor=1.5, type='Resize'), dict(keep_ratio=True, scale_factor=1.75, type='Resize'), ], [ dict(direction='horizontal', prob=0.0, type='RandomFlip'), dict(direction='horizontal', prob=1.0, type='RandomFlip'), ], [ dict(type='LoadAnnotations'), ], [ dict(type='PackSegInputs'), ], ], type='TestTimeAug'), ] val_cfg = dict(type='ValLoop') val_dataloader = dict( batch_size=1, dataset=dict( data_prefix=dict( img_path='images/val2017', seg_map_path='annotations/val2017'), data_root='data/coco_stuff164k', pipeline=[ dict(type='LoadImageFromFile'), dict(max_size=2560, scale=( 640, 640, ), type='ResizeShortestEdge'), dict(type='LoadAnnotations'), dict(type='PackSegInputs'), ], type='COCOStuffDataset'), num_workers=4, persistent_workers=True, sampler=dict(shuffle=False, type='DefaultSampler')) val_evaluator = dict( iou_metrics=[ 'mIoU', ], type='IoUMetric') vis_backends = [ dict(type='LocalVisBackend'), ] visualizer = dict( name='visualizer', type='SegLocalVisualizer', vis_backends=[ dict(type='LocalVisBackend'), ]) work_dir = './work_dirs/train_60k_amp_fixed' 2023/09/06 14:32:19 - mmengine - INFO - Hooks will be executed in the following order: before_run: (VERY_HIGH ) RuntimeInfoHook (BELOW_NORMAL) LoggerHook -------------------- before_train: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (VERY_LOW ) CheckpointHook -------------------- before_train_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (NORMAL ) DistSamplerSeedHook -------------------- before_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook -------------------- after_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (NORMAL ) SegVisualizationHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_train_epoch: (NORMAL ) IterTimerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_val: (VERY_HIGH ) RuntimeInfoHook -------------------- before_val_epoch: (NORMAL ) IterTimerHook -------------------- before_val_iter: (NORMAL ) IterTimerHook -------------------- after_val_iter: (NORMAL ) IterTimerHook (NORMAL ) SegVisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_val_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_val: (VERY_HIGH ) RuntimeInfoHook -------------------- after_train: (VERY_HIGH ) RuntimeInfoHook (VERY_LOW ) CheckpointHook -------------------- before_test: (VERY_HIGH ) RuntimeInfoHook -------------------- before_test_epoch: (NORMAL ) IterTimerHook -------------------- before_test_iter: (NORMAL ) IterTimerHook -------------------- after_test_iter: (NORMAL ) IterTimerHook (NORMAL ) SegVisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_test_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_test: (VERY_HIGH ) RuntimeInfoHook -------------------- after_run: (BELOW_NORMAL) LoggerHook -------------------- 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- image_encoder.pos_embed:lr=0.0001 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- image_encoder.pos_embed:weight_decay=0.0 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- image_encoder.pos_embed:decay_mult=0.0 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.pos_embed:lr=0.0001 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.pos_embed:weight_decay=0.0 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.pos_embed:decay_mult=0.0 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.query_pos_embed:lr=0.0001 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.query_pos_embed:weight_decay=0.0 2023/09/06 14:32:26 - mmengine - INFO - paramwise_options -- decode_head.side_adapter_network.query_pos_embed:decay_mult=0.0 2023/09/06 14:32:27 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. Name of parameter - Initialization information image_encoder.cls_token - torch.Size([1, 1, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.pos_embed - torch.Size([1, 197, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.patch_embed.projection.weight - torch.Size([768, 3, 16, 16]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.ln_pre.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.ln_pre.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.0.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.1.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.2.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.3.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.4.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.5.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.6.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.7.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ln1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ln1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.attn.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.attn.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.attn.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.attn.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ln2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ln2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ffn.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ffn.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ffn.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in VisionTransformer image_encoder.layers.8.ffn.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in VisionTransformer text_encoder.positional_embedding - torch.Size([77, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.text_projection - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.logit_scale - torch.Size([]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.bg_embed - torch.Size([1, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.token_embedding.weight - torch.Size([49408, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.0.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.1.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.2.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.3.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.4.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.5.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.6.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.7.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.8.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.9.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.10.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.attentions.0.attn.in_proj_weight - torch.Size([1536, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.attentions.0.attn.in_proj_bias - torch.Size([1536]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.attentions.0.attn.out_proj.weight - torch.Size([512, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.attentions.0.attn.out_proj.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.ffns.0.layers.0.0.weight - torch.Size([2048, 512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.ffns.0.layers.0.0.bias - torch.Size([2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.ffns.0.layers.1.weight - torch.Size([512, 2048]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.ffns.0.layers.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.norms.0.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.norms.0.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.norms.1.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.transformer.11.norms.1.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.ln_final.weight - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder text_encoder.ln_final.bias - torch.Size([512]): Initialized by user-defined `init_weights` in CLIPTextEncoder decode_head.side_adapter_network.pos_embed - torch.Size([1, 1600, 240]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.query_pos_embed - torch.Size([1, 100, 240]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.query_embed - torch.Size([1, 100, 240]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.patch_embed.projection.weight - torch.Size([240, 3, 16, 16]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.patch_embed.projection.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.0.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.1.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.2.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.3.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.4.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.5.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.6.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ln1.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ln1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.attn.attn.in_proj_weight - torch.Size([720, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.attn.attn.in_proj_bias - torch.Size([720]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.attn.attn.out_proj.weight - torch.Size([240, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.attn.attn.out_proj.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ln2.weight - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ln2.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ffn.layers.0.0.weight - torch.Size([960, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ffn.layers.0.0.bias - torch.Size([960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ffn.layers.1.weight - torch.Size([240, 960]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.encode_layers.7.ffn.layers.1.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.0.0.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.0.0.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.0.1.conv.weight - torch.Size([240, 768, 1, 1]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.conv_clips.0.1.conv.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.1.0.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.1.0.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.1.1.conv.weight - torch.Size([240, 768, 1, 1]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.conv_clips.1.1.conv.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.2.0.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.2.0.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.2.1.conv.weight - torch.Size([240, 768, 1, 1]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.conv_clips.2.1.conv.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.3.0.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.3.0.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.conv_clips.3.1.conv.weight - torch.Size([240, 768, 1, 1]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.side_adapter_network.conv_clips.3.1.conv.bias - torch.Size([240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.0.weight - torch.Size([256, 240]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.0.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.1.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.2.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.query_mlp.layers.2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.0.weight - torch.Size([256, 240, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.0.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.1.weight - torch.Size([256, 256, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.2.weight - torch.Size([256, 256, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.pix_mlp.layers.2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.0.weight - torch.Size([256, 240, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.0.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.1.weight - torch.Size([256, 256, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.2.weight - torch.Size([3072, 256, 1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.attn_mlp.layers.2.bias - torch.Size([3072]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.bias_scaling.weight - torch.Size([1, 1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.side_adapter_network.mask_decoder.bias_scaling.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of MultimodalEncoderDecoder decode_head.rec_with_attnbias.layers.0.attentions.0.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.attentions.0.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.attentions.0.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.attentions.0.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.ffns.0.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.ffns.0.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.ffns.0.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.ffns.0.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.norms.0.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.norms.0.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.norms.1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.0.norms.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.attentions.0.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.attentions.0.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.attentions.0.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.attentions.0.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.ffns.0.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.ffns.0.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.ffns.0.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.ffns.0.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.norms.0.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.norms.0.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.norms.1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.1.norms.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.attentions.0.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.attentions.0.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.attentions.0.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.attentions.0.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.ffns.0.layers.0.0.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.ffns.0.layers.0.0.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.ffns.0.layers.1.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.ffns.0.layers.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.norms.0.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.norms.0.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.norms.1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.layers.2.norms.1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.ln_post.weight - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.ln_post.bias - torch.Size([768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead decode_head.rec_with_attnbias.proj.weight - torch.Size([512, 768]): Initialized by user-defined `init_weights` in SideAdapterCLIPHead 2023/09/06 14:32:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io 2023/09/06 14:32:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. 2023/09/06 14:32:28 - mmengine - INFO - Checkpoints will be saved to /home/caoanqi/mmsegmentation/work_dirs/train_60k_amp_fixed. 2023/09/06 14:32:52 - mmengine - INFO - Iter(train) [ 50/60000] base_lr: 9.9918e-05 lr: 9.9918e-05 eta: 8:06:36 time: 0.4448 data_time: 0.0213 memory: 16050 grad_norm: nan loss: 49.9432 decode.loss_cls_ce: 19.1963 decode.loss_mask_ce: 1.5234 decode.loss_mask_dice: 4.2863 decode.d7.loss_cls_ce: 19.1429 decode.d7.loss_mask_ce: 1.5060 decode.d7.loss_mask_dice: 4.2885 2023/09/06 14:33:15 - mmengine - INFO - Iter(train) [ 100/60000] base_lr: 9.9835e-05 lr: 9.9835e-05 eta: 7:47:29 time: 0.4448 data_time: 0.0212 memory: 16012 grad_norm: 134.4031 loss: 29.4718 decode.loss_cls_ce: 9.5541 decode.loss_mask_ce: 1.5808 decode.loss_mask_dice: 3.5920 decode.d7.loss_cls_ce: 9.5475 decode.d7.loss_mask_ce: 1.5749 decode.d7.loss_mask_dice: 3.6225 2023/09/06 14:33:37 - mmengine - INFO - Iter(train) [ 150/60000] base_lr: 9.9752e-05 lr: 9.9752e-05 eta: 7:40:15 time: 0.4518 data_time: 0.0212 memory: 15781 grad_norm: 69.6754 loss: 21.9089 decode.loss_cls_ce: 5.8466 decode.loss_mask_ce: 1.5304 decode.loss_mask_dice: 3.5886 decode.d7.loss_cls_ce: 5.8358 decode.d7.loss_mask_ce: 1.5245 decode.d7.loss_mask_dice: 3.5830 2023/09/06 14:34:00 - mmengine - INFO - Iter(train) [ 200/60000] base_lr: 9.9668e-05 lr: 9.9668e-05 eta: 7:37:00 time: 0.4483 data_time: 0.0213 memory: 15872 grad_norm: 47.1404 loss: 21.3601 decode.loss_cls_ce: 5.6902 decode.loss_mask_ce: 1.5571 decode.loss_mask_dice: 3.4515 decode.d7.loss_cls_ce: 5.6411 decode.d7.loss_mask_ce: 1.5585 decode.d7.loss_mask_dice: 3.4616 2023/09/06 14:34:22 - mmengine - INFO - Iter(train) [ 250/60000] base_lr: 9.9585e-05 lr: 9.9585e-05 eta: 7:34:36 time: 0.4452 data_time: 0.0203 memory: 15733 grad_norm: 41.5377 loss: 19.6612 decode.loss_cls_ce: 4.9887 decode.loss_mask_ce: 1.5919 decode.loss_mask_dice: 3.2624 decode.d7.loss_cls_ce: 4.9545 decode.d7.loss_mask_ce: 1.5794 decode.d7.loss_mask_dice: 3.2843 2023/09/06 14:34:45 - mmengine - INFO - Iter(train) [ 300/60000] base_lr: 9.9502e-05 lr: 9.9502e-05 eta: 7:32:37 time: 0.4497 data_time: 0.0202 memory: 15758 grad_norm: 36.0429 loss: 17.8750 decode.loss_cls_ce: 4.7427 decode.loss_mask_ce: 1.4580 decode.loss_mask_dice: 2.7637 decode.d7.loss_cls_ce: 4.6998 decode.d7.loss_mask_ce: 1.4436 decode.d7.loss_mask_dice: 2.7672 2023/09/06 14:35:07 - mmengine - INFO - Iter(train) [ 350/60000] base_lr: 9.9418e-05 lr: 9.9418e-05 eta: 7:31:26 time: 0.4498 data_time: 0.0204 memory: 15771 grad_norm: 38.1561 loss: 16.7370 decode.loss_cls_ce: 4.4657 decode.loss_mask_ce: 1.3246 decode.loss_mask_dice: 2.5676 decode.d7.loss_cls_ce: 4.4650 decode.d7.loss_mask_ce: 1.3164 decode.d7.loss_mask_dice: 2.5977 2023/09/06 14:35:29 - mmengine - INFO - Iter(train) [ 400/60000] base_lr: 9.9335e-05 lr: 9.9335e-05 eta: 7:30:00 time: 0.4417 data_time: 0.0217 memory: 15913 grad_norm: 32.6137 loss: 15.8843 decode.loss_cls_ce: 4.2488 decode.loss_mask_ce: 1.2495 decode.loss_mask_dice: 2.4525 decode.d7.loss_cls_ce: 4.2109 decode.d7.loss_mask_ce: 1.2555 decode.d7.loss_mask_dice: 2.4670 2023/09/06 14:35:52 - mmengine - INFO - Iter(train) [ 450/60000] base_lr: 9.9252e-05 lr: 9.9252e-05 eta: 7:28:45 time: 0.4458 data_time: 0.0217 memory: 15774 grad_norm: 36.7149 loss: 16.4495 decode.loss_cls_ce: 4.2685 decode.loss_mask_ce: 1.3021 decode.loss_mask_dice: 2.6560 decode.d7.loss_cls_ce: 4.2425 decode.d7.loss_mask_ce: 1.3086 decode.d7.loss_mask_dice: 2.6718 2023/09/06 14:36:14 - mmengine - INFO - Iter(train) [ 500/60000] base_lr: 9.9168e-05 lr: 9.9168e-05 eta: 7:28:11 time: 0.4493 data_time: 0.0220 memory: 16003 grad_norm: 42.6107 loss: 16.8330 decode.loss_cls_ce: 4.1814 decode.loss_mask_ce: 1.4110 decode.loss_mask_dice: 2.8253 decode.d7.loss_cls_ce: 4.1735 decode.d7.loss_mask_ce: 1.4044 decode.d7.loss_mask_dice: 2.8373 2023/09/06 14:36:37 - mmengine - INFO - Iter(train) [ 550/60000] base_lr: 9.9085e-05 lr: 9.9085e-05 eta: 7:27:29 time: 0.4473 data_time: 0.0228 memory: 15832 grad_norm: 32.7700 loss: 16.2677 decode.loss_cls_ce: 4.1491 decode.loss_mask_ce: 1.3212 decode.loss_mask_dice: 2.6655 decode.d7.loss_cls_ce: 4.1515 decode.d7.loss_mask_ce: 1.3026 decode.d7.loss_mask_dice: 2.6777 2023/09/06 14:36:59 - mmengine - INFO - Iter(train) [ 600/60000] base_lr: 9.9002e-05 lr: 9.9002e-05 eta: 7:26:38 time: 0.4500 data_time: 0.0220 memory: 15786 grad_norm: 35.0112 loss: 16.2942 decode.loss_cls_ce: 4.0717 decode.loss_mask_ce: 1.3293 decode.loss_mask_dice: 2.7471 decode.d7.loss_cls_ce: 4.0758 decode.d7.loss_mask_ce: 1.3142 decode.d7.loss_mask_dice: 2.7561 2023/09/06 14:37:21 - mmengine - INFO - Iter(train) [ 650/60000] base_lr: 9.8918e-05 lr: 9.8918e-05 eta: 7:25:53 time: 0.4464 data_time: 0.0214 memory: 15761 grad_norm: 35.8911 loss: 14.0531 decode.loss_cls_ce: 3.5720 decode.loss_mask_ce: 1.2292 decode.loss_mask_dice: 2.2078 decode.d7.loss_cls_ce: 3.6041 decode.d7.loss_mask_ce: 1.2192 decode.d7.loss_mask_dice: 2.2208 2023/09/06 14:37:43 - mmengine - INFO - Iter(train) [ 700/60000] base_lr: 9.8835e-05 lr: 9.8835e-05 eta: 7:25:10 time: 0.4444 data_time: 0.0219 memory: 15770 grad_norm: 36.7134 loss: 13.0628 decode.loss_cls_ce: 3.3259 decode.loss_mask_ce: 1.1860 decode.loss_mask_dice: 2.0429 decode.d7.loss_cls_ce: 3.2978 decode.d7.loss_mask_ce: 1.1639 decode.d7.loss_mask_dice: 2.0463 2023/09/06 14:38:06 - mmengine - INFO - Iter(train) [ 750/60000] base_lr: 9.8752e-05 lr: 9.8752e-05 eta: 7:24:28 time: 0.4459 data_time: 0.0220 memory: 16264 grad_norm: 31.9370 loss: 15.1007 decode.loss_cls_ce: 3.7726 decode.loss_mask_ce: 1.3082 decode.loss_mask_dice: 2.4720 decode.d7.loss_cls_ce: 3.7450 decode.d7.loss_mask_ce: 1.3055 decode.d7.loss_mask_dice: 2.4973 2023/09/06 14:38:28 - mmengine - INFO - Iter(train) [ 800/60000] base_lr: 9.8668e-05 lr: 9.8668e-05 eta: 7:23:49 time: 0.4453 data_time: 0.0219 memory: 15874 grad_norm: 31.5125 loss: 15.8664 decode.loss_cls_ce: 3.7975 decode.loss_mask_ce: 1.4471 decode.loss_mask_dice: 2.7029 decode.d7.loss_cls_ce: 3.7831 decode.d7.loss_mask_ce: 1.4274 decode.d7.loss_mask_dice: 2.7085 2023/09/06 14:38:50 - mmengine - INFO - Iter(train) [ 850/60000] base_lr: 9.8585e-05 lr: 9.8585e-05 eta: 7:23:15 time: 0.4482 data_time: 0.0215 memory: 15823 grad_norm: 31.2147 loss: 13.7286 decode.loss_cls_ce: 3.2747 decode.loss_mask_ce: 1.2009 decode.loss_mask_dice: 2.4045 decode.d7.loss_cls_ce: 3.2317 decode.d7.loss_mask_ce: 1.2101 decode.d7.loss_mask_dice: 2.4068 2023/09/06 14:39:13 - mmengine - INFO - Iter(train) [ 900/60000] base_lr: 9.8502e-05 lr: 9.8502e-05 eta: 7:22:42 time: 0.4484 data_time: 0.0215 memory: 15859 grad_norm: 28.5899 loss: 13.5793 decode.loss_cls_ce: 3.3569 decode.loss_mask_ce: 1.1405 decode.loss_mask_dice: 2.2819 decode.d7.loss_cls_ce: 3.3636 decode.d7.loss_mask_ce: 1.1416 decode.d7.loss_mask_dice: 2.2947 2023/09/06 14:39:35 - mmengine - INFO - Iter(train) [ 950/60000] base_lr: 9.8418e-05 lr: 9.8418e-05 eta: 7:22:21 time: 0.4514 data_time: 0.0222 memory: 15884 grad_norm: 28.4470 loss: 14.4977 decode.loss_cls_ce: 3.4233 decode.loss_mask_ce: 1.2594 decode.loss_mask_dice: 2.5600 decode.d7.loss_cls_ce: 3.4204 decode.d7.loss_mask_ce: 1.2543 decode.d7.loss_mask_dice: 2.5803 2023/09/06 14:39:58 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 14:39:58 - mmengine - INFO - Iter(train) [ 1000/60000] base_lr: 9.8335e-05 lr: 9.8335e-05 eta: 7:21:55 time: 0.4474 data_time: 0.0216 memory: 15799 grad_norm: 36.4814 loss: 14.4144 decode.loss_cls_ce: 3.5990 decode.loss_mask_ce: 1.2023 decode.loss_mask_dice: 2.4002 decode.d7.loss_cls_ce: 3.6133 decode.d7.loss_mask_ce: 1.1886 decode.d7.loss_mask_dice: 2.4110 2023/09/06 14:40:20 - mmengine - INFO - Iter(train) [ 1050/60000] base_lr: 9.8252e-05 lr: 9.8252e-05 eta: 7:21:27 time: 0.4459 data_time: 0.0218 memory: 15939 grad_norm: 25.4708 loss: 13.3742 decode.loss_cls_ce: 3.3272 decode.loss_mask_ce: 1.1320 decode.loss_mask_dice: 2.2495 decode.d7.loss_cls_ce: 3.3032 decode.d7.loss_mask_ce: 1.1240 decode.d7.loss_mask_dice: 2.2383 2023/09/06 14:40:42 - mmengine - INFO - Iter(train) [ 1100/60000] base_lr: 9.8168e-05 lr: 9.8168e-05 eta: 7:21:00 time: 0.4504 data_time: 0.0215 memory: 15832 grad_norm: 29.5959 loss: 13.7746 decode.loss_cls_ce: 3.3848 decode.loss_mask_ce: 1.2092 decode.loss_mask_dice: 2.2968 decode.d7.loss_cls_ce: 3.3751 decode.d7.loss_mask_ce: 1.2068 decode.d7.loss_mask_dice: 2.3017 2023/09/06 14:41:05 - mmengine - INFO - Iter(train) [ 1150/60000] base_lr: 9.8085e-05 lr: 9.8085e-05 eta: 7:20:41 time: 0.4503 data_time: 0.0225 memory: 15848 grad_norm: 29.6184 loss: 14.8111 decode.loss_cls_ce: 3.4884 decode.loss_mask_ce: 1.3416 decode.loss_mask_dice: 2.5663 decode.d7.loss_cls_ce: 3.4991 decode.d7.loss_mask_ce: 1.3414 decode.d7.loss_mask_dice: 2.5743 2023/09/06 14:41:27 - mmengine - INFO - Iter(train) [ 1200/60000] base_lr: 9.8002e-05 lr: 9.8002e-05 eta: 7:20:17 time: 0.4501 data_time: 0.0216 memory: 15794 grad_norm: 30.1019 loss: 12.4614 decode.loss_cls_ce: 2.9301 decode.loss_mask_ce: 1.1567 decode.loss_mask_dice: 2.1313 decode.d7.loss_cls_ce: 2.9667 decode.d7.loss_mask_ce: 1.1493 decode.d7.loss_mask_dice: 2.1272 2023/09/06 14:41:50 - mmengine - INFO - Iter(train) [ 1250/60000] base_lr: 9.7918e-05 lr: 9.7918e-05 eta: 7:19:54 time: 0.4464 data_time: 0.0222 memory: 15967 grad_norm: 24.2694 loss: 12.9473 decode.loss_cls_ce: 3.1477 decode.loss_mask_ce: 1.0309 decode.loss_mask_dice: 2.2823 decode.d7.loss_cls_ce: 3.1651 decode.d7.loss_mask_ce: 1.0296 decode.d7.loss_mask_dice: 2.2916 2023/09/06 14:42:12 - mmengine - INFO - Iter(train) [ 1300/60000] base_lr: 9.7835e-05 lr: 9.7835e-05 eta: 7:19:25 time: 0.4478 data_time: 0.0224 memory: 15898 grad_norm: 27.1823 loss: 12.1809 decode.loss_cls_ce: 3.0925 decode.loss_mask_ce: 1.1459 decode.loss_mask_dice: 1.8604 decode.d7.loss_cls_ce: 3.0741 decode.d7.loss_mask_ce: 1.1386 decode.d7.loss_mask_dice: 1.8694 2023/09/06 14:42:34 - mmengine - INFO - Iter(train) [ 1350/60000] base_lr: 9.7752e-05 lr: 9.7752e-05 eta: 7:18:57 time: 0.4459 data_time: 0.0231 memory: 16003 grad_norm: 27.3527 loss: 12.4835 decode.loss_cls_ce: 2.9439 decode.loss_mask_ce: 1.1111 decode.loss_mask_dice: 2.1801 decode.d7.loss_cls_ce: 2.9332 decode.d7.loss_mask_ce: 1.1166 decode.d7.loss_mask_dice: 2.1987 2023/09/06 14:42:57 - mmengine - INFO - Iter(train) [ 1400/60000] base_lr: 9.7668e-05 lr: 9.7668e-05 eta: 7:18:32 time: 0.4479 data_time: 0.0229 memory: 15848 grad_norm: 23.5943 loss: 11.7337 decode.loss_cls_ce: 2.6858 decode.loss_mask_ce: 1.1576 decode.loss_mask_dice: 2.0126 decode.d7.loss_cls_ce: 2.7059 decode.d7.loss_mask_ce: 1.1466 decode.d7.loss_mask_dice: 2.0253 2023/09/06 14:43:19 - mmengine - INFO - Iter(train) [ 1450/60000] base_lr: 9.7585e-05 lr: 9.7585e-05 eta: 7:18:06 time: 0.4473 data_time: 0.0231 memory: 16081 grad_norm: 27.3379 loss: 14.1682 decode.loss_cls_ce: 3.2096 decode.loss_mask_ce: 1.2932 decode.loss_mask_dice: 2.5747 decode.d7.loss_cls_ce: 3.2153 decode.d7.loss_mask_ce: 1.2920 decode.d7.loss_mask_dice: 2.5834 2023/09/06 14:43:41 - mmengine - INFO - Iter(train) [ 1500/60000] base_lr: 9.7502e-05 lr: 9.7502e-05 eta: 7:17:39 time: 0.4491 data_time: 0.0223 memory: 15785 grad_norm: 26.2744 loss: 13.0754 decode.loss_cls_ce: 3.0937 decode.loss_mask_ce: 1.1692 decode.loss_mask_dice: 2.2690 decode.d7.loss_cls_ce: 3.1118 decode.d7.loss_mask_ce: 1.1696 decode.d7.loss_mask_dice: 2.2622 2023/09/06 14:44:04 - mmengine - INFO - Iter(train) [ 1550/60000] base_lr: 9.7418e-05 lr: 9.7418e-05 eta: 7:17:15 time: 0.4476 data_time: 0.0222 memory: 15861 grad_norm: 23.1042 loss: 12.0938 decode.loss_cls_ce: 2.9588 decode.loss_mask_ce: 1.0684 decode.loss_mask_dice: 2.0201 decode.d7.loss_cls_ce: 2.9447 decode.d7.loss_mask_ce: 1.0619 decode.d7.loss_mask_dice: 2.0399 2023/09/06 14:44:26 - mmengine - INFO - Iter(train) [ 1600/60000] base_lr: 9.7335e-05 lr: 9.7335e-05 eta: 7:16:53 time: 0.4517 data_time: 0.0222 memory: 15835 grad_norm: 29.4386 loss: 13.9515 decode.loss_cls_ce: 3.2342 decode.loss_mask_ce: 1.2014 decode.loss_mask_dice: 2.5302 decode.d7.loss_cls_ce: 3.2240 decode.d7.loss_mask_ce: 1.2158 decode.d7.loss_mask_dice: 2.5460 2023/09/06 14:44:49 - mmengine - INFO - Iter(train) [ 1650/60000] base_lr: 9.7252e-05 lr: 9.7252e-05 eta: 7:16:30 time: 0.4459 data_time: 0.0221 memory: 15884 grad_norm: 24.2650 loss: 12.7271 decode.loss_cls_ce: 2.9280 decode.loss_mask_ce: 1.1299 decode.loss_mask_dice: 2.3197 decode.d7.loss_cls_ce: 2.8942 decode.d7.loss_mask_ce: 1.1277 decode.d7.loss_mask_dice: 2.3276 2023/09/06 14:45:11 - mmengine - INFO - Iter(train) [ 1700/60000] base_lr: 9.7168e-05 lr: 9.7168e-05 eta: 7:16:04 time: 0.4463 data_time: 0.0227 memory: 15950 grad_norm: 23.9483 loss: 13.6465 decode.loss_cls_ce: 3.0633 decode.loss_mask_ce: 1.2465 decode.loss_mask_dice: 2.5105 decode.d7.loss_cls_ce: 3.0384 decode.d7.loss_mask_ce: 1.2578 decode.d7.loss_mask_dice: 2.5300 2023/09/06 14:45:33 - mmengine - INFO - Iter(train) [ 1750/60000] base_lr: 9.7085e-05 lr: 9.7085e-05 eta: 7:15:38 time: 0.4483 data_time: 0.0220 memory: 15823 grad_norm: 21.8616 loss: 12.3280 decode.loss_cls_ce: 2.9842 decode.loss_mask_ce: 1.0702 decode.loss_mask_dice: 2.1124 decode.d7.loss_cls_ce: 2.9931 decode.d7.loss_mask_ce: 1.0605 decode.d7.loss_mask_dice: 2.1077 2023/09/06 14:45:56 - mmengine - INFO - Iter(train) [ 1800/60000] base_lr: 9.7002e-05 lr: 9.7002e-05 eta: 7:15:18 time: 0.4465 data_time: 0.0227 memory: 15882 grad_norm: 23.6923 loss: 12.4180 decode.loss_cls_ce: 2.8415 decode.loss_mask_ce: 1.1470 decode.loss_mask_dice: 2.2116 decode.d7.loss_cls_ce: 2.8468 decode.d7.loss_mask_ce: 1.1585 decode.d7.loss_mask_dice: 2.2126 2023/09/06 14:46:18 - mmengine - INFO - Iter(train) [ 1850/60000] base_lr: 9.6918e-05 lr: 9.6918e-05 eta: 7:14:53 time: 0.4475 data_time: 0.0223 memory: 15937 grad_norm: 23.6331 loss: 10.9587 decode.loss_cls_ce: 2.5551 decode.loss_mask_ce: 1.0756 decode.loss_mask_dice: 1.8601 decode.d7.loss_cls_ce: 2.5342 decode.d7.loss_mask_ce: 1.0722 decode.d7.loss_mask_dice: 1.8614 2023/09/06 14:46:41 - mmengine - INFO - Iter(train) [ 1900/60000] base_lr: 9.6835e-05 lr: 9.6835e-05 eta: 7:14:28 time: 0.4493 data_time: 0.0233 memory: 16106 grad_norm: 26.8451 loss: 11.4594 decode.loss_cls_ce: 2.7550 decode.loss_mask_ce: 1.0318 decode.loss_mask_dice: 1.9379 decode.d7.loss_cls_ce: 2.7723 decode.d7.loss_mask_ce: 1.0225 decode.d7.loss_mask_dice: 1.9399 2023/09/06 14:47:03 - mmengine - INFO - Iter(train) [ 1950/60000] base_lr: 9.6752e-05 lr: 9.6752e-05 eta: 7:14:08 time: 0.4485 data_time: 0.0231 memory: 15731 grad_norm: 23.1569 loss: 12.5544 decode.loss_cls_ce: 2.8388 decode.loss_mask_ce: 1.2123 decode.loss_mask_dice: 2.2342 decode.d7.loss_cls_ce: 2.7976 decode.d7.loss_mask_ce: 1.2246 decode.d7.loss_mask_dice: 2.2469 2023/09/06 14:47:26 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 14:47:26 - mmengine - INFO - Iter(train) [ 2000/60000] base_lr: 9.6668e-05 lr: 9.6668e-05 eta: 7:13:42 time: 0.4470 data_time: 0.0228 memory: 15974 grad_norm: 24.5868 loss: 12.4040 decode.loss_cls_ce: 2.9114 decode.loss_mask_ce: 1.1303 decode.loss_mask_dice: 2.1470 decode.d7.loss_cls_ce: 2.9113 decode.d7.loss_mask_ce: 1.1289 decode.d7.loss_mask_dice: 2.1751 2023/09/06 14:47:48 - mmengine - INFO - Iter(train) [ 2050/60000] base_lr: 9.6585e-05 lr: 9.6585e-05 eta: 7:13:19 time: 0.4502 data_time: 0.0224 memory: 15833 grad_norm: 25.8221 loss: 13.0426 decode.loss_cls_ce: 3.0120 decode.loss_mask_ce: 1.1731 decode.loss_mask_dice: 2.3376 decode.d7.loss_cls_ce: 3.0090 decode.d7.loss_mask_ce: 1.1716 decode.d7.loss_mask_dice: 2.3395 2023/09/06 14:48:10 - mmengine - INFO - Iter(train) [ 2100/60000] base_lr: 9.6502e-05 lr: 9.6502e-05 eta: 7:12:54 time: 0.4478 data_time: 0.0224 memory: 15846 grad_norm: 21.2454 loss: 10.9602 decode.loss_cls_ce: 2.5450 decode.loss_mask_ce: 1.0369 decode.loss_mask_dice: 1.9102 decode.d7.loss_cls_ce: 2.5127 decode.d7.loss_mask_ce: 1.0358 decode.d7.loss_mask_dice: 1.9196 2023/09/06 14:48:33 - mmengine - INFO - Iter(train) [ 2150/60000] base_lr: 9.6418e-05 lr: 9.6418e-05 eta: 7:12:33 time: 0.4467 data_time: 0.0228 memory: 15846 grad_norm: 21.5769 loss: 11.7186 decode.loss_cls_ce: 2.6885 decode.loss_mask_ce: 1.1073 decode.loss_mask_dice: 2.0477 decode.d7.loss_cls_ce: 2.7138 decode.d7.loss_mask_ce: 1.1072 decode.d7.loss_mask_dice: 2.0542 2023/09/06 14:48:55 - mmengine - INFO - Iter(train) [ 2200/60000] base_lr: 9.6335e-05 lr: 9.6335e-05 eta: 7:12:09 time: 0.4461 data_time: 0.0228 memory: 15819 grad_norm: 24.8681 loss: 11.3910 decode.loss_cls_ce: 2.6481 decode.loss_mask_ce: 1.1111 decode.loss_mask_dice: 1.9573 decode.d7.loss_cls_ce: 2.6104 decode.d7.loss_mask_ce: 1.1129 decode.d7.loss_mask_dice: 1.9513 2023/09/06 14:49:18 - mmengine - INFO - Iter(train) [ 2250/60000] base_lr: 9.6252e-05 lr: 9.6252e-05 eta: 7:11:46 time: 0.4485 data_time: 0.0221 memory: 15769 grad_norm: 18.5987 loss: 12.0587 decode.loss_cls_ce: 2.8422 decode.loss_mask_ce: 1.0656 decode.loss_mask_dice: 2.1113 decode.d7.loss_cls_ce: 2.8474 decode.d7.loss_mask_ce: 1.0717 decode.d7.loss_mask_dice: 2.1206 2023/09/06 14:49:40 - mmengine - INFO - Iter(train) [ 2300/60000] base_lr: 9.6168e-05 lr: 9.6168e-05 eta: 7:11:25 time: 0.4491 data_time: 0.0224 memory: 15872 grad_norm: 22.5986 loss: 11.2587 decode.loss_cls_ce: 2.5079 decode.loss_mask_ce: 1.0521 decode.loss_mask_dice: 2.0627 decode.d7.loss_cls_ce: 2.5191 decode.d7.loss_mask_ce: 1.0504 decode.d7.loss_mask_dice: 2.0665 2023/09/06 14:50:03 - mmengine - INFO - Iter(train) [ 2350/60000] base_lr: 9.6085e-05 lr: 9.6085e-05 eta: 7:11:05 time: 0.4543 data_time: 0.0224 memory: 15795 grad_norm: 20.4761 loss: 12.2446 decode.loss_cls_ce: 2.7829 decode.loss_mask_ce: 1.1126 decode.loss_mask_dice: 2.2348 decode.d7.loss_cls_ce: 2.7726 decode.d7.loss_mask_ce: 1.1136 decode.d7.loss_mask_dice: 2.2281 2023/09/06 14:50:25 - mmengine - INFO - Iter(train) [ 2400/60000] base_lr: 9.6002e-05 lr: 9.6002e-05 eta: 7:10:45 time: 0.4546 data_time: 0.0215 memory: 16013 grad_norm: 25.2345 loss: 12.8678 decode.loss_cls_ce: 2.9232 decode.loss_mask_ce: 1.2208 decode.loss_mask_dice: 2.2776 decode.d7.loss_cls_ce: 2.9519 decode.d7.loss_mask_ce: 1.2279 decode.d7.loss_mask_dice: 2.2663 2023/09/06 14:50:48 - mmengine - INFO - Iter(train) [ 2450/60000] base_lr: 9.5918e-05 lr: 9.5918e-05 eta: 7:10:23 time: 0.4479 data_time: 0.0228 memory: 15782 grad_norm: 20.9262 loss: 12.0582 decode.loss_cls_ce: 2.7765 decode.loss_mask_ce: 1.0901 decode.loss_mask_dice: 2.1598 decode.d7.loss_cls_ce: 2.7808 decode.d7.loss_mask_ce: 1.0830 decode.d7.loss_mask_dice: 2.1680 2023/09/06 14:51:10 - mmengine - INFO - Iter(train) [ 2500/60000] base_lr: 9.5835e-05 lr: 9.5835e-05 eta: 7:10:01 time: 0.4499 data_time: 0.0233 memory: 15885 grad_norm: 19.7614 loss: 12.8512 decode.loss_cls_ce: 3.0194 decode.loss_mask_ce: 1.0746 decode.loss_mask_dice: 2.3390 decode.d7.loss_cls_ce: 3.0016 decode.d7.loss_mask_ce: 1.0600 decode.d7.loss_mask_dice: 2.3565 2023/09/06 14:51:32 - mmengine - INFO - Iter(train) [ 2550/60000] base_lr: 9.5752e-05 lr: 9.5752e-05 eta: 7:09:36 time: 0.4467 data_time: 0.0226 memory: 15872 grad_norm: 20.6343 loss: 11.9862 decode.loss_cls_ce: 2.8029 decode.loss_mask_ce: 1.1127 decode.loss_mask_dice: 2.0851 decode.d7.loss_cls_ce: 2.7782 decode.d7.loss_mask_ce: 1.1188 decode.d7.loss_mask_dice: 2.0885 2023/09/06 14:51:55 - mmengine - INFO - Iter(train) [ 2600/60000] base_lr: 9.5668e-05 lr: 9.5668e-05 eta: 7:09:14 time: 0.4528 data_time: 0.0225 memory: 15807 grad_norm: 21.5329 loss: 12.4098 decode.loss_cls_ce: 2.9030 decode.loss_mask_ce: 1.0620 decode.loss_mask_dice: 2.2322 decode.d7.loss_cls_ce: 2.9052 decode.d7.loss_mask_ce: 1.0667 decode.d7.loss_mask_dice: 2.2406 2023/09/06 14:52:17 - mmengine - INFO - Iter(train) [ 2650/60000] base_lr: 9.5585e-05 lr: 9.5585e-05 eta: 7:08:51 time: 0.4477 data_time: 0.0226 memory: 15736 grad_norm: 20.7479 loss: 11.4289 decode.loss_cls_ce: 2.6253 decode.loss_mask_ce: 1.0356 decode.loss_mask_dice: 2.0536 decode.d7.loss_cls_ce: 2.6246 decode.d7.loss_mask_ce: 1.0334 decode.d7.loss_mask_dice: 2.0563 2023/09/06 14:52:40 - mmengine - INFO - Iter(train) [ 2700/60000] base_lr: 9.5502e-05 lr: 9.5502e-05 eta: 7:08:29 time: 0.4469 data_time: 0.0228 memory: 15785 grad_norm: 22.2292 loss: 13.4084 decode.loss_cls_ce: 3.0531 decode.loss_mask_ce: 1.1371 decode.loss_mask_dice: 2.5265 decode.d7.loss_cls_ce: 3.0360 decode.d7.loss_mask_ce: 1.1343 decode.d7.loss_mask_dice: 2.5214 2023/09/06 14:53:02 - mmengine - INFO - Iter(train) [ 2750/60000] base_lr: 9.5418e-05 lr: 9.5418e-05 eta: 7:08:04 time: 0.4454 data_time: 0.0224 memory: 15810 grad_norm: 22.5522 loss: 11.8605 decode.loss_cls_ce: 2.8812 decode.loss_mask_ce: 0.9500 decode.loss_mask_dice: 2.0763 decode.d7.loss_cls_ce: 2.9133 decode.d7.loss_mask_ce: 0.9573 decode.d7.loss_mask_dice: 2.0825 2023/09/06 14:53:24 - mmengine - INFO - Iter(train) [ 2800/60000] base_lr: 9.5335e-05 lr: 9.5335e-05 eta: 7:07:41 time: 0.4490 data_time: 0.0222 memory: 15809 grad_norm: 19.9896 loss: 11.1236 decode.loss_cls_ce: 2.5910 decode.loss_mask_ce: 1.0163 decode.loss_mask_dice: 1.9475 decode.d7.loss_cls_ce: 2.6018 decode.d7.loss_mask_ce: 1.0120 decode.d7.loss_mask_dice: 1.9548 2023/09/06 14:53:47 - mmengine - INFO - Iter(train) [ 2850/60000] base_lr: 9.5252e-05 lr: 9.5252e-05 eta: 7:07:21 time: 0.4502 data_time: 0.0223 memory: 15974 grad_norm: 20.9265 loss: 12.3126 decode.loss_cls_ce: 2.8431 decode.loss_mask_ce: 1.0572 decode.loss_mask_dice: 2.2567 decode.d7.loss_cls_ce: 2.8300 decode.d7.loss_mask_ce: 1.0692 decode.d7.loss_mask_dice: 2.2565 2023/09/06 14:54:09 - mmengine - INFO - Iter(train) [ 2900/60000] base_lr: 9.5168e-05 lr: 9.5168e-05 eta: 7:06:59 time: 0.4478 data_time: 0.0225 memory: 15897 grad_norm: 24.4725 loss: 13.6789 decode.loss_cls_ce: 2.9885 decode.loss_mask_ce: 1.2166 decode.loss_mask_dice: 2.6417 decode.d7.loss_cls_ce: 2.9714 decode.d7.loss_mask_ce: 1.2147 decode.d7.loss_mask_dice: 2.6460 2023/09/06 14:54:32 - mmengine - INFO - Iter(train) [ 2950/60000] base_lr: 9.5085e-05 lr: 9.5085e-05 eta: 7:06:36 time: 0.4472 data_time: 0.0227 memory: 15849 grad_norm: 20.8547 loss: 11.2520 decode.loss_cls_ce: 2.6547 decode.loss_mask_ce: 1.0553 decode.loss_mask_dice: 1.9084 decode.d7.loss_cls_ce: 2.6467 decode.d7.loss_mask_ce: 1.0605 decode.d7.loss_mask_dice: 1.9264 2023/09/06 14:54:54 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 14:54:54 - mmengine - INFO - Iter(train) [ 3000/60000] base_lr: 9.5002e-05 lr: 9.5002e-05 eta: 7:06:13 time: 0.4481 data_time: 0.0231 memory: 15857 grad_norm: 19.5626 loss: 12.0419 decode.loss_cls_ce: 2.6613 decode.loss_mask_ce: 1.1521 decode.loss_mask_dice: 2.2082 decode.d7.loss_cls_ce: 2.6591 decode.d7.loss_mask_ce: 1.1543 decode.d7.loss_mask_dice: 2.2069 2023/09/06 14:55:17 - mmengine - INFO - Iter(train) [ 3050/60000] base_lr: 9.4918e-05 lr: 9.4918e-05 eta: 7:05:51 time: 0.4507 data_time: 0.0228 memory: 15719 grad_norm: 21.7010 loss: 11.6431 decode.loss_cls_ce: 2.7600 decode.loss_mask_ce: 1.0599 decode.loss_mask_dice: 2.0187 decode.d7.loss_cls_ce: 2.7365 decode.d7.loss_mask_ce: 1.0581 decode.d7.loss_mask_dice: 2.0100 2023/09/06 14:55:39 - mmengine - INFO - Iter(train) [ 3100/60000] base_lr: 9.4835e-05 lr: 9.4835e-05 eta: 7:05:31 time: 0.4483 data_time: 0.0227 memory: 15925 grad_norm: 23.6851 loss: 12.3931 decode.loss_cls_ce: 2.7563 decode.loss_mask_ce: 1.1526 decode.loss_mask_dice: 2.2961 decode.d7.loss_cls_ce: 2.7550 decode.d7.loss_mask_ce: 1.1431 decode.d7.loss_mask_dice: 2.2902 2023/09/06 14:56:02 - mmengine - INFO - Iter(train) [ 3150/60000] base_lr: 9.4752e-05 lr: 9.4752e-05 eta: 7:05:11 time: 0.4472 data_time: 0.0226 memory: 15844 grad_norm: 18.5297 loss: 11.1861 decode.loss_cls_ce: 2.5578 decode.loss_mask_ce: 1.1053 decode.loss_mask_dice: 1.9382 decode.d7.loss_cls_ce: 2.5580 decode.d7.loss_mask_ce: 1.1003 decode.d7.loss_mask_dice: 1.9265 2023/09/06 14:56:24 - mmengine - INFO - Iter(train) [ 3200/60000] base_lr: 9.4668e-05 lr: 9.4668e-05 eta: 7:04:46 time: 0.4449 data_time: 0.0220 memory: 15787 grad_norm: 22.9864 loss: 13.0905 decode.loss_cls_ce: 2.8866 decode.loss_mask_ce: 1.2477 decode.loss_mask_dice: 2.3936 decode.d7.loss_cls_ce: 2.9172 decode.d7.loss_mask_ce: 1.2501 decode.d7.loss_mask_dice: 2.3953 2023/09/06 14:56:46 - mmengine - INFO - Iter(train) [ 3250/60000] base_lr: 9.4585e-05 lr: 9.4585e-05 eta: 7:04:22 time: 0.4470 data_time: 0.0237 memory: 15899 grad_norm: 21.0835 loss: 11.2904 decode.loss_cls_ce: 2.5251 decode.loss_mask_ce: 1.0463 decode.loss_mask_dice: 2.0853 decode.d7.loss_cls_ce: 2.4899 decode.d7.loss_mask_ce: 1.0511 decode.d7.loss_mask_dice: 2.0927 2023/09/06 14:57:09 - mmengine - INFO - Iter(train) [ 3300/60000] base_lr: 9.4502e-05 lr: 9.4502e-05 eta: 7:03:59 time: 0.4490 data_time: 0.0234 memory: 15846 grad_norm: 20.1351 loss: 11.7605 decode.loss_cls_ce: 2.7281 decode.loss_mask_ce: 1.0688 decode.loss_mask_dice: 2.0681 decode.d7.loss_cls_ce: 2.7319 decode.d7.loss_mask_ce: 1.0782 decode.d7.loss_mask_dice: 2.0854 2023/09/06 14:57:32 - mmengine - INFO - Iter(train) [ 3350/60000] base_lr: 9.4418e-05 lr: 9.4418e-05 eta: 7:03:40 time: 0.4518 data_time: 0.0227 memory: 15799 grad_norm: 23.2904 loss: 12.2437 decode.loss_cls_ce: 2.8372 decode.loss_mask_ce: 1.0702 decode.loss_mask_dice: 2.2124 decode.d7.loss_cls_ce: 2.8416 decode.d7.loss_mask_ce: 1.0729 decode.d7.loss_mask_dice: 2.2094 2023/09/06 14:57:54 - mmengine - INFO - Iter(train) [ 3400/60000] base_lr: 9.4335e-05 lr: 9.4335e-05 eta: 7:03:18 time: 0.4476 data_time: 0.0232 memory: 15823 grad_norm: 19.3978 loss: 13.0482 decode.loss_cls_ce: 2.9661 decode.loss_mask_ce: 1.1631 decode.loss_mask_dice: 2.3854 decode.d7.loss_cls_ce: 2.9993 decode.d7.loss_mask_ce: 1.1583 decode.d7.loss_mask_dice: 2.3759 2023/09/06 14:58:16 - mmengine - INFO - Iter(train) [ 3450/60000] base_lr: 9.4252e-05 lr: 9.4252e-05 eta: 7:02:55 time: 0.4480 data_time: 0.0230 memory: 15835 grad_norm: 20.8763 loss: 11.4398 decode.loss_cls_ce: 2.6165 decode.loss_mask_ce: 1.0244 decode.loss_mask_dice: 2.0820 decode.d7.loss_cls_ce: 2.6121 decode.d7.loss_mask_ce: 1.0297 decode.d7.loss_mask_dice: 2.0752 2023/09/06 14:58:39 - mmengine - INFO - Iter(train) [ 3500/60000] base_lr: 9.4168e-05 lr: 9.4168e-05 eta: 7:02:34 time: 0.4502 data_time: 0.0223 memory: 15833 grad_norm: 19.1083 loss: 11.4851 decode.loss_cls_ce: 2.5341 decode.loss_mask_ce: 1.1205 decode.loss_mask_dice: 2.0919 decode.d7.loss_cls_ce: 2.5189 decode.d7.loss_mask_ce: 1.1171 decode.d7.loss_mask_dice: 2.1026 2023/09/06 14:59:01 - mmengine - INFO - Iter(train) [ 3550/60000] base_lr: 9.4085e-05 lr: 9.4085e-05 eta: 7:02:13 time: 0.4487 data_time: 0.0227 memory: 15870 grad_norm: 19.2340 loss: 11.5839 decode.loss_cls_ce: 2.6490 decode.loss_mask_ce: 1.1043 decode.loss_mask_dice: 2.0490 decode.d7.loss_cls_ce: 2.6243 decode.d7.loss_mask_ce: 1.0925 decode.d7.loss_mask_dice: 2.0648 2023/09/06 14:59:24 - mmengine - INFO - Iter(train) [ 3600/60000] base_lr: 9.4002e-05 lr: 9.4002e-05 eta: 7:01:51 time: 0.4474 data_time: 0.0225 memory: 15785 grad_norm: 20.5671 loss: 11.8989 decode.loss_cls_ce: 2.5653 decode.loss_mask_ce: 1.2226 decode.loss_mask_dice: 2.1741 decode.d7.loss_cls_ce: 2.5702 decode.d7.loss_mask_ce: 1.2020 decode.d7.loss_mask_dice: 2.1648 2023/09/06 14:59:46 - mmengine - INFO - Iter(train) [ 3650/60000] base_lr: 9.3918e-05 lr: 9.3918e-05 eta: 7:01:29 time: 0.4481 data_time: 0.0225 memory: 16171 grad_norm: 19.4057 loss: 11.5672 decode.loss_cls_ce: 2.4400 decode.loss_mask_ce: 1.1318 decode.loss_mask_dice: 2.2051 decode.d7.loss_cls_ce: 2.4599 decode.d7.loss_mask_ce: 1.1296 decode.d7.loss_mask_dice: 2.2008 2023/09/06 15:00:09 - mmengine - INFO - Iter(train) [ 3700/60000] base_lr: 9.3835e-05 lr: 9.3835e-05 eta: 7:01:07 time: 0.4528 data_time: 0.0237 memory: 15992 grad_norm: 23.2637 loss: 10.8173 decode.loss_cls_ce: 2.5398 decode.loss_mask_ce: 1.0259 decode.loss_mask_dice: 1.8470 decode.d7.loss_cls_ce: 2.5402 decode.d7.loss_mask_ce: 1.0190 decode.d7.loss_mask_dice: 1.8454 2023/09/06 15:00:31 - mmengine - INFO - Iter(train) [ 3750/60000] base_lr: 9.3752e-05 lr: 9.3752e-05 eta: 7:00:44 time: 0.4457 data_time: 0.0230 memory: 15822 grad_norm: 19.3870 loss: 11.8259 decode.loss_cls_ce: 2.5362 decode.loss_mask_ce: 1.1280 decode.loss_mask_dice: 2.2521 decode.d7.loss_cls_ce: 2.5194 decode.d7.loss_mask_ce: 1.1328 decode.d7.loss_mask_dice: 2.2573 2023/09/06 15:00:54 - mmengine - INFO - Iter(train) [ 3800/60000] base_lr: 9.3668e-05 lr: 9.3668e-05 eta: 7:00:22 time: 0.4479 data_time: 0.0236 memory: 15810 grad_norm: 18.2685 loss: 11.9517 decode.loss_cls_ce: 2.7117 decode.loss_mask_ce: 1.0063 decode.loss_mask_dice: 2.2634 decode.d7.loss_cls_ce: 2.7132 decode.d7.loss_mask_ce: 1.0064 decode.d7.loss_mask_dice: 2.2507 2023/09/06 15:01:16 - mmengine - INFO - Iter(train) [ 3850/60000] base_lr: 9.3585e-05 lr: 9.3585e-05 eta: 7:00:04 time: 0.4534 data_time: 0.0217 memory: 15818 grad_norm: 23.9957 loss: 11.1480 decode.loss_cls_ce: 2.6155 decode.loss_mask_ce: 0.9325 decode.loss_mask_dice: 2.0484 decode.d7.loss_cls_ce: 2.5793 decode.d7.loss_mask_ce: 0.9321 decode.d7.loss_mask_dice: 2.0403 2023/09/06 15:01:39 - mmengine - INFO - Iter(train) [ 3900/60000] base_lr: 9.3502e-05 lr: 9.3502e-05 eta: 6:59:42 time: 0.4462 data_time: 0.0231 memory: 15772 grad_norm: 20.4908 loss: 10.1894 decode.loss_cls_ce: 2.2984 decode.loss_mask_ce: 1.0083 decode.loss_mask_dice: 1.7825 decode.d7.loss_cls_ce: 2.2978 decode.d7.loss_mask_ce: 1.0070 decode.d7.loss_mask_dice: 1.7954 2023/09/06 15:02:01 - mmengine - INFO - Iter(train) [ 3950/60000] base_lr: 9.3418e-05 lr: 9.3418e-05 eta: 6:59:19 time: 0.4464 data_time: 0.0232 memory: 15861 grad_norm: 18.8205 loss: 10.7811 decode.loss_cls_ce: 2.4986 decode.loss_mask_ce: 1.0132 decode.loss_mask_dice: 1.8698 decode.d7.loss_cls_ce: 2.5025 decode.d7.loss_mask_ce: 1.0161 decode.d7.loss_mask_dice: 1.8810 2023/09/06 15:02:24 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:02:24 - mmengine - INFO - Iter(train) [ 4000/60000] base_lr: 9.3335e-05 lr: 9.3335e-05 eta: 6:58:56 time: 0.4506 data_time: 0.0223 memory: 15871 grad_norm: 20.2953 loss: 12.1550 decode.loss_cls_ce: 2.7079 decode.loss_mask_ce: 1.1521 decode.loss_mask_dice: 2.2176 decode.d7.loss_cls_ce: 2.6998 decode.d7.loss_mask_ce: 1.1568 decode.d7.loss_mask_dice: 2.2209 2023/09/06 15:02:46 - mmengine - INFO - Iter(train) [ 4050/60000] base_lr: 9.3252e-05 lr: 9.3252e-05 eta: 6:58:36 time: 0.4519 data_time: 0.0234 memory: 15861 grad_norm: 19.5760 loss: 11.4777 decode.loss_cls_ce: 2.5351 decode.loss_mask_ce: 1.1360 decode.loss_mask_dice: 2.0603 decode.d7.loss_cls_ce: 2.5427 decode.d7.loss_mask_ce: 1.1302 decode.d7.loss_mask_dice: 2.0734 2023/09/06 15:03:09 - mmengine - INFO - Iter(train) [ 4100/60000] base_lr: 9.3168e-05 lr: 9.3168e-05 eta: 6:58:14 time: 0.4504 data_time: 0.0220 memory: 15785 grad_norm: 21.3189 loss: 12.2620 decode.loss_cls_ce: 2.8074 decode.loss_mask_ce: 1.1255 decode.loss_mask_dice: 2.2209 decode.d7.loss_cls_ce: 2.7692 decode.d7.loss_mask_ce: 1.1187 decode.d7.loss_mask_dice: 2.2203 2023/09/06 15:03:31 - mmengine - INFO - Iter(train) [ 4150/60000] base_lr: 9.3085e-05 lr: 9.3085e-05 eta: 6:57:54 time: 0.4506 data_time: 0.0218 memory: 15948 grad_norm: 20.7168 loss: 12.3416 decode.loss_cls_ce: 2.7949 decode.loss_mask_ce: 1.0147 decode.loss_mask_dice: 2.3514 decode.d7.loss_cls_ce: 2.7958 decode.d7.loss_mask_ce: 1.0220 decode.d7.loss_mask_dice: 2.3628 2023/09/06 15:03:54 - mmengine - INFO - Iter(train) [ 4200/60000] base_lr: 9.3002e-05 lr: 9.3002e-05 eta: 6:57:33 time: 0.4507 data_time: 0.0222 memory: 15864 grad_norm: 19.7178 loss: 10.8216 decode.loss_cls_ce: 2.4388 decode.loss_mask_ce: 1.0372 decode.loss_mask_dice: 1.9297 decode.d7.loss_cls_ce: 2.4628 decode.d7.loss_mask_ce: 1.0372 decode.d7.loss_mask_dice: 1.9158 2023/09/06 15:04:17 - mmengine - INFO - Iter(train) [ 4250/60000] base_lr: 9.2918e-05 lr: 9.2918e-05 eta: 6:57:12 time: 0.4519 data_time: 0.0227 memory: 15849 grad_norm: 19.0038 loss: 12.5515 decode.loss_cls_ce: 2.7556 decode.loss_mask_ce: 1.1571 decode.loss_mask_dice: 2.3591 decode.d7.loss_cls_ce: 2.7655 decode.d7.loss_mask_ce: 1.1622 decode.d7.loss_mask_dice: 2.3520 2023/09/06 15:04:39 - mmengine - INFO - Iter(train) [ 4300/60000] base_lr: 9.2835e-05 lr: 9.2835e-05 eta: 6:56:50 time: 0.4496 data_time: 0.0228 memory: 15835 grad_norm: 18.4728 loss: 12.2636 decode.loss_cls_ce: 2.6981 decode.loss_mask_ce: 1.2223 decode.loss_mask_dice: 2.2241 decode.d7.loss_cls_ce: 2.6807 decode.d7.loss_mask_ce: 1.2178 decode.d7.loss_mask_dice: 2.2206 2023/09/06 15:05:02 - mmengine - INFO - Iter(train) [ 4350/60000] base_lr: 9.2752e-05 lr: 9.2752e-05 eta: 6:56:27 time: 0.4509 data_time: 0.0230 memory: 15938 grad_norm: 22.0336 loss: 11.4200 decode.loss_cls_ce: 2.6928 decode.loss_mask_ce: 1.0162 decode.loss_mask_dice: 2.0003 decode.d7.loss_cls_ce: 2.7008 decode.d7.loss_mask_ce: 1.0110 decode.d7.loss_mask_dice: 1.9988 2023/09/06 15:05:24 - mmengine - INFO - Iter(train) [ 4400/60000] base_lr: 9.2668e-05 lr: 9.2668e-05 eta: 6:56:06 time: 0.4560 data_time: 0.0226 memory: 15796 grad_norm: 22.3886 loss: 11.7397 decode.loss_cls_ce: 2.5659 decode.loss_mask_ce: 1.0684 decode.loss_mask_dice: 2.2275 decode.d7.loss_cls_ce: 2.5853 decode.d7.loss_mask_ce: 1.0627 decode.d7.loss_mask_dice: 2.2299 2023/09/06 15:05:47 - mmengine - INFO - Iter(train) [ 4450/60000] base_lr: 9.2585e-05 lr: 9.2585e-05 eta: 6:55:43 time: 0.4473 data_time: 0.0227 memory: 15859 grad_norm: 23.6507 loss: 12.0803 decode.loss_cls_ce: 2.6684 decode.loss_mask_ce: 1.1103 decode.loss_mask_dice: 2.2464 decode.d7.loss_cls_ce: 2.6929 decode.d7.loss_mask_ce: 1.1152 decode.d7.loss_mask_dice: 2.2471 2023/09/06 15:06:09 - mmengine - INFO - Iter(train) [ 4500/60000] base_lr: 9.2502e-05 lr: 9.2502e-05 eta: 6:55:22 time: 0.4504 data_time: 0.0217 memory: 15856 grad_norm: 19.0309 loss: 12.0529 decode.loss_cls_ce: 2.7447 decode.loss_mask_ce: 1.0734 decode.loss_mask_dice: 2.2054 decode.d7.loss_cls_ce: 2.7533 decode.d7.loss_mask_ce: 1.0815 decode.d7.loss_mask_dice: 2.1946 2023/09/06 15:06:31 - mmengine - INFO - Iter(train) [ 4550/60000] base_lr: 9.2418e-05 lr: 9.2418e-05 eta: 6:54:59 time: 0.4488 data_time: 0.0238 memory: 15718 grad_norm: 20.0172 loss: 11.5091 decode.loss_cls_ce: 2.6450 decode.loss_mask_ce: 1.1161 decode.loss_mask_dice: 2.0118 decode.d7.loss_cls_ce: 2.6097 decode.d7.loss_mask_ce: 1.1022 decode.d7.loss_mask_dice: 2.0244 2023/09/06 15:06:54 - mmengine - INFO - Iter(train) [ 4600/60000] base_lr: 9.2335e-05 lr: 9.2335e-05 eta: 6:54:38 time: 0.4542 data_time: 0.0222 memory: 15859 grad_norm: 19.9266 loss: 12.8479 decode.loss_cls_ce: 3.0498 decode.loss_mask_ce: 1.1133 decode.loss_mask_dice: 2.2828 decode.d7.loss_cls_ce: 3.0142 decode.d7.loss_mask_ce: 1.1138 decode.d7.loss_mask_dice: 2.2740 2023/09/06 15:07:17 - mmengine - INFO - Iter(train) [ 4650/60000] base_lr: 9.2252e-05 lr: 9.2252e-05 eta: 6:54:17 time: 0.4486 data_time: 0.0228 memory: 15793 grad_norm: 20.8302 loss: 10.5815 decode.loss_cls_ce: 2.3473 decode.loss_mask_ce: 1.0967 decode.loss_mask_dice: 1.8410 decode.d7.loss_cls_ce: 2.3423 decode.d7.loss_mask_ce: 1.1068 decode.d7.loss_mask_dice: 1.8474 2023/09/06 15:07:39 - mmengine - INFO - Iter(train) [ 4700/60000] base_lr: 9.2168e-05 lr: 9.2168e-05 eta: 6:53:55 time: 0.4472 data_time: 0.0228 memory: 15783 grad_norm: 18.3629 loss: 11.3521 decode.loss_cls_ce: 2.5270 decode.loss_mask_ce: 1.0420 decode.loss_mask_dice: 2.1157 decode.d7.loss_cls_ce: 2.5358 decode.d7.loss_mask_ce: 1.0373 decode.d7.loss_mask_dice: 2.0943 2023/09/06 15:08:02 - mmengine - INFO - Iter(train) [ 4750/60000] base_lr: 9.2085e-05 lr: 9.2085e-05 eta: 6:53:35 time: 0.4525 data_time: 0.0231 memory: 15784 grad_norm: 18.0797 loss: 11.5520 decode.loss_cls_ce: 2.6095 decode.loss_mask_ce: 1.1246 decode.loss_mask_dice: 2.0326 decode.d7.loss_cls_ce: 2.6378 decode.d7.loss_mask_ce: 1.1250 decode.d7.loss_mask_dice: 2.0226 2023/09/06 15:08:24 - mmengine - INFO - Iter(train) [ 4800/60000] base_lr: 9.2002e-05 lr: 9.2002e-05 eta: 6:53:13 time: 0.4553 data_time: 0.0223 memory: 15949 grad_norm: 18.9162 loss: 11.1250 decode.loss_cls_ce: 2.5041 decode.loss_mask_ce: 1.1092 decode.loss_mask_dice: 1.9493 decode.d7.loss_cls_ce: 2.5079 decode.d7.loss_mask_ce: 1.1083 decode.d7.loss_mask_dice: 1.9462 2023/09/06 15:08:47 - mmengine - INFO - Iter(train) [ 4850/60000] base_lr: 9.1918e-05 lr: 9.1918e-05 eta: 6:52:51 time: 0.4485 data_time: 0.0236 memory: 15794 grad_norm: 20.8161 loss: 11.4244 decode.loss_cls_ce: 2.4840 decode.loss_mask_ce: 1.0714 decode.loss_mask_dice: 2.1487 decode.d7.loss_cls_ce: 2.4987 decode.d7.loss_mask_ce: 1.0732 decode.d7.loss_mask_dice: 2.1486 2023/09/06 15:09:09 - mmengine - INFO - Iter(train) [ 4900/60000] base_lr: 9.1835e-05 lr: 9.1835e-05 eta: 6:52:29 time: 0.4488 data_time: 0.0226 memory: 15773 grad_norm: 18.7028 loss: 10.6690 decode.loss_cls_ce: 2.4233 decode.loss_mask_ce: 0.9880 decode.loss_mask_dice: 1.9403 decode.d7.loss_cls_ce: 2.4114 decode.d7.loss_mask_ce: 0.9828 decode.d7.loss_mask_dice: 1.9233 2023/09/06 15:09:32 - mmengine - INFO - Iter(train) [ 4950/60000] base_lr: 9.1752e-05 lr: 9.1752e-05 eta: 6:52:07 time: 0.4490 data_time: 0.0230 memory: 15861 grad_norm: 17.5467 loss: 11.7201 decode.loss_cls_ce: 2.6784 decode.loss_mask_ce: 1.0333 decode.loss_mask_dice: 2.1340 decode.d7.loss_cls_ce: 2.6979 decode.d7.loss_mask_ce: 1.0421 decode.d7.loss_mask_dice: 2.1342 2023/09/06 15:09:54 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:09:54 - mmengine - INFO - Iter(train) [ 5000/60000] base_lr: 9.1668e-05 lr: 9.1668e-05 eta: 6:51:47 time: 0.4544 data_time: 0.0223 memory: 15933 grad_norm: 18.4047 loss: 11.2704 decode.loss_cls_ce: 2.5052 decode.loss_mask_ce: 0.9865 decode.loss_mask_dice: 2.1394 decode.d7.loss_cls_ce: 2.5034 decode.d7.loss_mask_ce: 0.9869 decode.d7.loss_mask_dice: 2.1489 2023/09/06 15:10:17 - mmengine - INFO - Iter(train) [ 5050/60000] base_lr: 9.1585e-05 lr: 9.1585e-05 eta: 6:51:25 time: 0.4489 data_time: 0.0229 memory: 15799 grad_norm: 19.8723 loss: 10.2502 decode.loss_cls_ce: 2.2533 decode.loss_mask_ce: 1.0008 decode.loss_mask_dice: 1.8751 decode.d7.loss_cls_ce: 2.2498 decode.d7.loss_mask_ce: 0.9953 decode.d7.loss_mask_dice: 1.8758 2023/09/06 15:10:39 - mmengine - INFO - Iter(train) [ 5100/60000] base_lr: 9.1502e-05 lr: 9.1502e-05 eta: 6:51:03 time: 0.4514 data_time: 0.0226 memory: 15798 grad_norm: 17.5174 loss: 11.2706 decode.loss_cls_ce: 2.6298 decode.loss_mask_ce: 1.0250 decode.loss_mask_dice: 1.9616 decode.d7.loss_cls_ce: 2.6415 decode.d7.loss_mask_ce: 1.0392 decode.d7.loss_mask_dice: 1.9735 2023/09/06 15:11:02 - mmengine - INFO - Iter(train) [ 5150/60000] base_lr: 9.1418e-05 lr: 9.1418e-05 eta: 6:50:42 time: 0.4506 data_time: 0.0228 memory: 15847 grad_norm: 19.1303 loss: 10.9012 decode.loss_cls_ce: 2.4469 decode.loss_mask_ce: 1.0131 decode.loss_mask_dice: 1.9922 decode.d7.loss_cls_ce: 2.4470 decode.d7.loss_mask_ce: 1.0178 decode.d7.loss_mask_dice: 1.9843 2023/09/06 15:11:25 - mmengine - INFO - Iter(train) [ 5200/60000] base_lr: 9.1335e-05 lr: 9.1335e-05 eta: 6:50:21 time: 0.4484 data_time: 0.0222 memory: 16039 grad_norm: 18.0673 loss: 12.6393 decode.loss_cls_ce: 2.8072 decode.loss_mask_ce: 1.0936 decode.loss_mask_dice: 2.4323 decode.d7.loss_cls_ce: 2.7888 decode.d7.loss_mask_ce: 1.0984 decode.d7.loss_mask_dice: 2.4191 2023/09/06 15:11:47 - mmengine - INFO - Iter(train) [ 5250/60000] base_lr: 9.1252e-05 lr: 9.1252e-05 eta: 6:49:58 time: 0.4502 data_time: 0.0227 memory: 15787 grad_norm: 17.2132 loss: 11.2454 decode.loss_cls_ce: 2.6497 decode.loss_mask_ce: 1.0649 decode.loss_mask_dice: 1.9011 decode.d7.loss_cls_ce: 2.6470 decode.d7.loss_mask_ce: 1.0656 decode.d7.loss_mask_dice: 1.9171 2023/09/06 15:12:10 - mmengine - INFO - Iter(train) [ 5300/60000] base_lr: 9.1168e-05 lr: 9.1168e-05 eta: 6:49:36 time: 0.4502 data_time: 0.0231 memory: 15836 grad_norm: 19.2574 loss: 9.9352 decode.loss_cls_ce: 2.2454 decode.loss_mask_ce: 0.9126 decode.loss_mask_dice: 1.8005 decode.d7.loss_cls_ce: 2.2706 decode.d7.loss_mask_ce: 0.9049 decode.d7.loss_mask_dice: 1.8012 2023/09/06 15:12:32 - mmengine - INFO - Iter(train) [ 5350/60000] base_lr: 9.1085e-05 lr: 9.1085e-05 eta: 6:49:13 time: 0.4504 data_time: 0.0225 memory: 15795 grad_norm: 18.1446 loss: 11.3878 decode.loss_cls_ce: 2.6773 decode.loss_mask_ce: 0.9979 decode.loss_mask_dice: 2.0094 decode.d7.loss_cls_ce: 2.6908 decode.d7.loss_mask_ce: 1.0129 decode.d7.loss_mask_dice: 1.9996 2023/09/06 15:12:55 - mmengine - INFO - Iter(train) [ 5400/60000] base_lr: 9.1002e-05 lr: 9.1002e-05 eta: 6:48:53 time: 0.4541 data_time: 0.0241 memory: 15858 grad_norm: 17.6372 loss: 11.3382 decode.loss_cls_ce: 2.6291 decode.loss_mask_ce: 1.0568 decode.loss_mask_dice: 1.9829 decode.d7.loss_cls_ce: 2.6218 decode.d7.loss_mask_ce: 1.0628 decode.d7.loss_mask_dice: 1.9848 2023/09/06 15:13:17 - mmengine - INFO - Iter(train) [ 5450/60000] base_lr: 9.0918e-05 lr: 9.0918e-05 eta: 6:48:31 time: 0.4517 data_time: 0.0237 memory: 16004 grad_norm: 17.2812 loss: 10.7681 decode.loss_cls_ce: 2.4574 decode.loss_mask_ce: 0.9984 decode.loss_mask_dice: 1.9313 decode.d7.loss_cls_ce: 2.4438 decode.d7.loss_mask_ce: 1.0032 decode.d7.loss_mask_dice: 1.9340 2023/09/06 15:13:40 - mmengine - INFO - Iter(train) [ 5500/60000] base_lr: 9.0835e-05 lr: 9.0835e-05 eta: 6:48:09 time: 0.4489 data_time: 0.0225 memory: 15913 grad_norm: 18.2878 loss: 10.7661 decode.loss_cls_ce: 2.4361 decode.loss_mask_ce: 0.9431 decode.loss_mask_dice: 2.0028 decode.d7.loss_cls_ce: 2.4393 decode.d7.loss_mask_ce: 0.9428 decode.d7.loss_mask_dice: 2.0019 2023/09/06 15:14:02 - mmengine - INFO - Iter(train) [ 5550/60000] base_lr: 9.0752e-05 lr: 9.0752e-05 eta: 6:47:47 time: 0.4529 data_time: 0.0232 memory: 15872 grad_norm: 21.6066 loss: 11.0546 decode.loss_cls_ce: 2.5149 decode.loss_mask_ce: 1.0397 decode.loss_mask_dice: 1.9680 decode.d7.loss_cls_ce: 2.5325 decode.d7.loss_mask_ce: 1.0349 decode.d7.loss_mask_dice: 1.9647 2023/09/06 15:14:25 - mmengine - INFO - Iter(train) [ 5600/60000] base_lr: 9.0668e-05 lr: 9.0668e-05 eta: 6:47:25 time: 0.4479 data_time: 0.0232 memory: 15898 grad_norm: 17.2598 loss: 11.2556 decode.loss_cls_ce: 2.5198 decode.loss_mask_ce: 1.0493 decode.loss_mask_dice: 2.0550 decode.d7.loss_cls_ce: 2.5040 decode.d7.loss_mask_ce: 1.0575 decode.d7.loss_mask_dice: 2.0701 2023/09/06 15:14:47 - mmengine - INFO - Iter(train) [ 5650/60000] base_lr: 9.0585e-05 lr: 9.0585e-05 eta: 6:47:03 time: 0.4532 data_time: 0.0218 memory: 15884 grad_norm: 19.6109 loss: 11.0811 decode.loss_cls_ce: 2.5732 decode.loss_mask_ce: 1.0410 decode.loss_mask_dice: 1.9164 decode.d7.loss_cls_ce: 2.5912 decode.d7.loss_mask_ce: 1.0343 decode.d7.loss_mask_dice: 1.9250 2023/09/06 15:15:10 - mmengine - INFO - Iter(train) [ 5700/60000] base_lr: 9.0502e-05 lr: 9.0502e-05 eta: 6:46:42 time: 0.4494 data_time: 0.0229 memory: 15842 grad_norm: 17.7261 loss: 12.3743 decode.loss_cls_ce: 2.7859 decode.loss_mask_ce: 1.0490 decode.loss_mask_dice: 2.3406 decode.d7.loss_cls_ce: 2.8006 decode.d7.loss_mask_ce: 1.0424 decode.d7.loss_mask_dice: 2.3559 2023/09/06 15:15:32 - mmengine - INFO - Iter(train) [ 5750/60000] base_lr: 9.0418e-05 lr: 9.0418e-05 eta: 6:46:20 time: 0.4483 data_time: 0.0225 memory: 15978 grad_norm: 17.7938 loss: 11.5705 decode.loss_cls_ce: 2.6672 decode.loss_mask_ce: 0.9998 decode.loss_mask_dice: 2.1249 decode.d7.loss_cls_ce: 2.6503 decode.d7.loss_mask_ce: 1.0058 decode.d7.loss_mask_dice: 2.1225 2023/09/06 15:15:55 - mmengine - INFO - Iter(train) [ 5800/60000] base_lr: 9.0335e-05 lr: 9.0335e-05 eta: 6:45:57 time: 0.4493 data_time: 0.0235 memory: 15884 grad_norm: 17.3705 loss: 10.0982 decode.loss_cls_ce: 2.3039 decode.loss_mask_ce: 0.9487 decode.loss_mask_dice: 1.8169 decode.d7.loss_cls_ce: 2.2795 decode.d7.loss_mask_ce: 0.9418 decode.d7.loss_mask_dice: 1.8076 2023/09/06 15:16:17 - mmengine - INFO - Iter(train) [ 5850/60000] base_lr: 9.0252e-05 lr: 9.0252e-05 eta: 6:45:35 time: 0.4477 data_time: 0.0233 memory: 15872 grad_norm: 18.1273 loss: 11.1875 decode.loss_cls_ce: 2.4757 decode.loss_mask_ce: 0.9402 decode.loss_mask_dice: 2.1586 decode.d7.loss_cls_ce: 2.5287 decode.d7.loss_mask_ce: 0.9310 decode.d7.loss_mask_dice: 2.1533 2023/09/06 15:16:40 - mmengine - INFO - Iter(train) [ 5900/60000] base_lr: 9.0168e-05 lr: 9.0168e-05 eta: 6:45:12 time: 0.4477 data_time: 0.0232 memory: 15885 grad_norm: 16.7631 loss: 11.0368 decode.loss_cls_ce: 2.6089 decode.loss_mask_ce: 0.9709 decode.loss_mask_dice: 1.9479 decode.d7.loss_cls_ce: 2.5998 decode.d7.loss_mask_ce: 0.9602 decode.d7.loss_mask_dice: 1.9492 2023/09/06 15:17:02 - mmengine - INFO - Iter(train) [ 5950/60000] base_lr: 9.0085e-05 lr: 9.0085e-05 eta: 6:44:49 time: 0.4486 data_time: 0.0240 memory: 15872 grad_norm: 17.4013 loss: 11.5507 decode.loss_cls_ce: 2.4982 decode.loss_mask_ce: 1.1397 decode.loss_mask_dice: 2.1564 decode.d7.loss_cls_ce: 2.5037 decode.d7.loss_mask_ce: 1.1202 decode.d7.loss_mask_dice: 2.1325 2023/09/06 15:17:25 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:17:25 - mmengine - INFO - Iter(train) [ 6000/60000] base_lr: 9.0002e-05 lr: 9.0002e-05 eta: 6:44:27 time: 0.4507 data_time: 0.0228 memory: 15872 grad_norm: 16.1077 loss: 11.3457 decode.loss_cls_ce: 2.4476 decode.loss_mask_ce: 1.0804 decode.loss_mask_dice: 2.1412 decode.d7.loss_cls_ce: 2.4400 decode.d7.loss_mask_ce: 1.0901 decode.d7.loss_mask_dice: 2.1464 2023/09/06 15:17:47 - mmengine - INFO - Iter(train) [ 6050/60000] base_lr: 8.9918e-05 lr: 8.9918e-05 eta: 6:44:04 time: 0.4491 data_time: 0.0226 memory: 15872 grad_norm: 17.0190 loss: 10.5091 decode.loss_cls_ce: 2.5164 decode.loss_mask_ce: 0.8927 decode.loss_mask_dice: 1.8752 decode.d7.loss_cls_ce: 2.4595 decode.d7.loss_mask_ce: 0.9001 decode.d7.loss_mask_dice: 1.8650 2023/09/06 15:18:10 - mmengine - INFO - Iter(train) [ 6100/60000] base_lr: 8.9835e-05 lr: 8.9835e-05 eta: 6:43:42 time: 0.4497 data_time: 0.0231 memory: 15911 grad_norm: 17.3525 loss: 10.2974 decode.loss_cls_ce: 2.3242 decode.loss_mask_ce: 0.9493 decode.loss_mask_dice: 1.8786 decode.d7.loss_cls_ce: 2.3297 decode.d7.loss_mask_ce: 0.9401 decode.d7.loss_mask_dice: 1.8755 2023/09/06 15:18:32 - mmengine - INFO - Iter(train) [ 6150/60000] base_lr: 8.9751e-05 lr: 8.9751e-05 eta: 6:43:20 time: 0.4559 data_time: 0.0234 memory: 15961 grad_norm: 18.0463 loss: 11.2073 decode.loss_cls_ce: 2.4212 decode.loss_mask_ce: 1.1457 decode.loss_mask_dice: 2.0524 decode.d7.loss_cls_ce: 2.3946 decode.d7.loss_mask_ce: 1.1474 decode.d7.loss_mask_dice: 2.0461 2023/09/06 15:18:55 - mmengine - INFO - Iter(train) [ 6200/60000] base_lr: 8.9668e-05 lr: 8.9668e-05 eta: 6:42:59 time: 0.4493 data_time: 0.0229 memory: 15781 grad_norm: 16.8560 loss: 10.7646 decode.loss_cls_ce: 2.4331 decode.loss_mask_ce: 0.9670 decode.loss_mask_dice: 1.9835 decode.d7.loss_cls_ce: 2.4292 decode.d7.loss_mask_ce: 0.9679 decode.d7.loss_mask_dice: 1.9839 2023/09/06 15:19:17 - mmengine - INFO - Iter(train) [ 6250/60000] base_lr: 8.9585e-05 lr: 8.9585e-05 eta: 6:42:37 time: 0.4479 data_time: 0.0223 memory: 15799 grad_norm: 18.6798 loss: 10.8075 decode.loss_cls_ce: 2.3960 decode.loss_mask_ce: 1.0215 decode.loss_mask_dice: 1.9603 decode.d7.loss_cls_ce: 2.4284 decode.d7.loss_mask_ce: 1.0210 decode.d7.loss_mask_dice: 1.9802 2023/09/06 15:19:40 - mmengine - INFO - Iter(train) [ 6300/60000] base_lr: 8.9501e-05 lr: 8.9501e-05 eta: 6:42:15 time: 0.4495 data_time: 0.0240 memory: 15807 grad_norm: 18.0776 loss: 10.9180 decode.loss_cls_ce: 2.5266 decode.loss_mask_ce: 1.0187 decode.loss_mask_dice: 1.9384 decode.d7.loss_cls_ce: 2.4759 decode.d7.loss_mask_ce: 1.0177 decode.d7.loss_mask_dice: 1.9408 2023/09/06 15:20:02 - mmengine - INFO - Iter(train) [ 6350/60000] base_lr: 8.9418e-05 lr: 8.9418e-05 eta: 6:41:53 time: 0.4506 data_time: 0.0235 memory: 15899 grad_norm: 19.9182 loss: 12.0852 decode.loss_cls_ce: 2.6464 decode.loss_mask_ce: 1.1191 decode.loss_mask_dice: 2.2870 decode.d7.loss_cls_ce: 2.6294 decode.d7.loss_mask_ce: 1.1264 decode.d7.loss_mask_dice: 2.2769 2023/09/06 15:20:25 - mmengine - INFO - Iter(train) [ 6400/60000] base_lr: 8.9335e-05 lr: 8.9335e-05 eta: 6:41:31 time: 0.4505 data_time: 0.0232 memory: 15991 grad_norm: 18.0879 loss: 12.6499 decode.loss_cls_ce: 2.8325 decode.loss_mask_ce: 1.0962 decode.loss_mask_dice: 2.3807 decode.d7.loss_cls_ce: 2.8341 decode.d7.loss_mask_ce: 1.1028 decode.d7.loss_mask_dice: 2.4035 2023/09/06 15:20:47 - mmengine - INFO - Iter(train) [ 6450/60000] base_lr: 8.9251e-05 lr: 8.9251e-05 eta: 6:41:08 time: 0.4532 data_time: 0.0229 memory: 16004 grad_norm: 20.1821 loss: 9.7508 decode.loss_cls_ce: 2.2606 decode.loss_mask_ce: 0.9368 decode.loss_mask_dice: 1.6762 decode.d7.loss_cls_ce: 2.2722 decode.d7.loss_mask_ce: 0.9421 decode.d7.loss_mask_dice: 1.6629 2023/09/06 15:21:10 - mmengine - INFO - Iter(train) [ 6500/60000] base_lr: 8.9168e-05 lr: 8.9168e-05 eta: 6:40:46 time: 0.4488 data_time: 0.0231 memory: 15845 grad_norm: 18.6714 loss: 10.7603 decode.loss_cls_ce: 2.2136 decode.loss_mask_ce: 1.1168 decode.loss_mask_dice: 2.0492 decode.d7.loss_cls_ce: 2.2149 decode.d7.loss_mask_ce: 1.1144 decode.d7.loss_mask_dice: 2.0515 2023/09/06 15:21:32 - mmengine - INFO - Iter(train) [ 6550/60000] base_lr: 8.9085e-05 lr: 8.9085e-05 eta: 6:40:24 time: 0.4541 data_time: 0.0232 memory: 15800 grad_norm: 20.5086 loss: 12.0446 decode.loss_cls_ce: 2.7387 decode.loss_mask_ce: 1.1156 decode.loss_mask_dice: 2.1643 decode.d7.loss_cls_ce: 2.7143 decode.d7.loss_mask_ce: 1.1282 decode.d7.loss_mask_dice: 2.1834 2023/09/06 15:21:55 - mmengine - INFO - Iter(train) [ 6600/60000] base_lr: 8.9001e-05 lr: 8.9001e-05 eta: 6:40:02 time: 0.4530 data_time: 0.0234 memory: 15769 grad_norm: 18.0969 loss: 10.8676 decode.loss_cls_ce: 2.5999 decode.loss_mask_ce: 1.0110 decode.loss_mask_dice: 1.8225 decode.d7.loss_cls_ce: 2.5927 decode.d7.loss_mask_ce: 1.0184 decode.d7.loss_mask_dice: 1.8231 2023/09/06 15:22:18 - mmengine - INFO - Iter(train) [ 6650/60000] base_lr: 8.8918e-05 lr: 8.8918e-05 eta: 6:39:40 time: 0.4484 data_time: 0.0227 memory: 15946 grad_norm: 20.1088 loss: 10.5061 decode.loss_cls_ce: 2.4101 decode.loss_mask_ce: 1.0130 decode.loss_mask_dice: 1.8431 decode.d7.loss_cls_ce: 2.3891 decode.d7.loss_mask_ce: 1.0132 decode.d7.loss_mask_dice: 1.8375 2023/09/06 15:22:40 - mmengine - INFO - Iter(train) [ 6700/60000] base_lr: 8.8835e-05 lr: 8.8835e-05 eta: 6:39:18 time: 0.4522 data_time: 0.0231 memory: 15848 grad_norm: 19.1538 loss: 10.3599 decode.loss_cls_ce: 2.3412 decode.loss_mask_ce: 0.9477 decode.loss_mask_dice: 1.8732 decode.d7.loss_cls_ce: 2.3576 decode.d7.loss_mask_ce: 0.9613 decode.d7.loss_mask_dice: 1.8788 2023/09/06 15:23:03 - mmengine - INFO - Iter(train) [ 6750/60000] base_lr: 8.8751e-05 lr: 8.8751e-05 eta: 6:38:55 time: 0.4497 data_time: 0.0231 memory: 15896 grad_norm: 19.6436 loss: 10.4729 decode.loss_cls_ce: 2.3310 decode.loss_mask_ce: 0.9899 decode.loss_mask_dice: 1.9250 decode.d7.loss_cls_ce: 2.3092 decode.d7.loss_mask_ce: 0.9943 decode.d7.loss_mask_dice: 1.9235 2023/09/06 15:23:25 - mmengine - INFO - Iter(train) [ 6800/60000] base_lr: 8.8668e-05 lr: 8.8668e-05 eta: 6:38:34 time: 0.4537 data_time: 0.0222 memory: 15783 grad_norm: 16.8926 loss: 10.9249 decode.loss_cls_ce: 2.4583 decode.loss_mask_ce: 1.0101 decode.loss_mask_dice: 1.9999 decode.d7.loss_cls_ce: 2.4525 decode.d7.loss_mask_ce: 1.0107 decode.d7.loss_mask_dice: 1.9935 2023/09/06 15:23:48 - mmengine - INFO - Iter(train) [ 6850/60000] base_lr: 8.8585e-05 lr: 8.8585e-05 eta: 6:38:11 time: 0.4482 data_time: 0.0233 memory: 15874 grad_norm: 16.5945 loss: 10.2152 decode.loss_cls_ce: 2.2554 decode.loss_mask_ce: 0.9643 decode.loss_mask_dice: 1.8929 decode.d7.loss_cls_ce: 2.2394 decode.d7.loss_mask_ce: 0.9707 decode.d7.loss_mask_dice: 1.8926 2023/09/06 15:24:10 - mmengine - INFO - Iter(train) [ 6900/60000] base_lr: 8.8501e-05 lr: 8.8501e-05 eta: 6:37:48 time: 0.4493 data_time: 0.0225 memory: 15794 grad_norm: 19.2695 loss: 11.1022 decode.loss_cls_ce: 2.4538 decode.loss_mask_ce: 1.0327 decode.loss_mask_dice: 2.0657 decode.d7.loss_cls_ce: 2.4666 decode.d7.loss_mask_ce: 1.0220 decode.d7.loss_mask_dice: 2.0614 2023/09/06 15:24:33 - mmengine - INFO - Iter(train) [ 6950/60000] base_lr: 8.8418e-05 lr: 8.8418e-05 eta: 6:37:26 time: 0.4490 data_time: 0.0232 memory: 15798 grad_norm: 16.8134 loss: 11.3285 decode.loss_cls_ce: 2.4370 decode.loss_mask_ce: 1.1327 decode.loss_mask_dice: 2.0918 decode.d7.loss_cls_ce: 2.4441 decode.d7.loss_mask_ce: 1.1314 decode.d7.loss_mask_dice: 2.0915 2023/09/06 15:24:55 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:24:55 - mmengine - INFO - Iter(train) [ 7000/60000] base_lr: 8.8335e-05 lr: 8.8335e-05 eta: 6:37:04 time: 0.4500 data_time: 0.0229 memory: 15914 grad_norm: 18.2179 loss: 11.1665 decode.loss_cls_ce: 2.4488 decode.loss_mask_ce: 1.0181 decode.loss_mask_dice: 2.1169 decode.d7.loss_cls_ce: 2.4499 decode.d7.loss_mask_ce: 1.0182 decode.d7.loss_mask_dice: 2.1146 2023/09/06 15:25:18 - mmengine - INFO - Iter(train) [ 7050/60000] base_lr: 8.8251e-05 lr: 8.8251e-05 eta: 6:36:41 time: 0.4479 data_time: 0.0235 memory: 15812 grad_norm: 18.9454 loss: 10.7869 decode.loss_cls_ce: 2.3324 decode.loss_mask_ce: 1.0259 decode.loss_mask_dice: 2.0387 decode.d7.loss_cls_ce: 2.3241 decode.d7.loss_mask_ce: 1.0313 decode.d7.loss_mask_dice: 2.0346 2023/09/06 15:25:40 - mmengine - INFO - Iter(train) [ 7100/60000] base_lr: 8.8168e-05 lr: 8.8168e-05 eta: 6:36:18 time: 0.4491 data_time: 0.0228 memory: 16002 grad_norm: 17.9039 loss: 11.5399 decode.loss_cls_ce: 2.5278 decode.loss_mask_ce: 1.0461 decode.loss_mask_dice: 2.1757 decode.d7.loss_cls_ce: 2.5526 decode.d7.loss_mask_ce: 1.0431 decode.d7.loss_mask_dice: 2.1947 2023/09/06 15:26:03 - mmengine - INFO - Iter(train) [ 7150/60000] base_lr: 8.8085e-05 lr: 8.8085e-05 eta: 6:35:56 time: 0.4552 data_time: 0.0219 memory: 15821 grad_norm: 16.6558 loss: 10.3142 decode.loss_cls_ce: 2.3368 decode.loss_mask_ce: 0.9097 decode.loss_mask_dice: 1.9150 decode.d7.loss_cls_ce: 2.3408 decode.d7.loss_mask_ce: 0.9025 decode.d7.loss_mask_dice: 1.9094 2023/09/06 15:26:25 - mmengine - INFO - Iter(train) [ 7200/60000] base_lr: 8.8001e-05 lr: 8.8001e-05 eta: 6:35:34 time: 0.4492 data_time: 0.0220 memory: 15746 grad_norm: 17.0900 loss: 10.9623 decode.loss_cls_ce: 2.4916 decode.loss_mask_ce: 1.0531 decode.loss_mask_dice: 1.9322 decode.d7.loss_cls_ce: 2.4938 decode.d7.loss_mask_ce: 1.0512 decode.d7.loss_mask_dice: 1.9404 2023/09/06 15:26:48 - mmengine - INFO - Iter(train) [ 7250/60000] base_lr: 8.7918e-05 lr: 8.7918e-05 eta: 6:35:13 time: 0.4542 data_time: 0.0226 memory: 15992 grad_norm: 17.7207 loss: 9.5868 decode.loss_cls_ce: 2.1583 decode.loss_mask_ce: 0.9231 decode.loss_mask_dice: 1.6798 decode.d7.loss_cls_ce: 2.1806 decode.d7.loss_mask_ce: 0.9357 decode.d7.loss_mask_dice: 1.7094 2023/09/06 15:27:10 - mmengine - INFO - Iter(train) [ 7300/60000] base_lr: 8.7835e-05 lr: 8.7835e-05 eta: 6:34:52 time: 0.4502 data_time: 0.0231 memory: 15909 grad_norm: 16.3490 loss: 11.7150 decode.loss_cls_ce: 2.6916 decode.loss_mask_ce: 1.0262 decode.loss_mask_dice: 2.1495 decode.d7.loss_cls_ce: 2.6494 decode.d7.loss_mask_ce: 1.0334 decode.d7.loss_mask_dice: 2.1650 2023/09/06 15:27:33 - mmengine - INFO - Iter(train) [ 7350/60000] base_lr: 8.7751e-05 lr: 8.7751e-05 eta: 6:34:29 time: 0.4485 data_time: 0.0231 memory: 15797 grad_norm: 17.2989 loss: 10.3982 decode.loss_cls_ce: 2.2972 decode.loss_mask_ce: 0.9674 decode.loss_mask_dice: 1.9337 decode.d7.loss_cls_ce: 2.3109 decode.d7.loss_mask_ce: 0.9663 decode.d7.loss_mask_dice: 1.9228 2023/09/06 15:27:55 - mmengine - INFO - Iter(train) [ 7400/60000] base_lr: 8.7668e-05 lr: 8.7668e-05 eta: 6:34:07 time: 0.4500 data_time: 0.0243 memory: 16053 grad_norm: 17.5474 loss: 11.1366 decode.loss_cls_ce: 2.5306 decode.loss_mask_ce: 1.0372 decode.loss_mask_dice: 1.9855 decode.d7.loss_cls_ce: 2.5547 decode.d7.loss_mask_ce: 1.0375 decode.d7.loss_mask_dice: 1.9912 2023/09/06 15:28:18 - mmengine - INFO - Iter(train) [ 7450/60000] base_lr: 8.7585e-05 lr: 8.7585e-05 eta: 6:33:45 time: 0.4519 data_time: 0.0233 memory: 15845 grad_norm: 17.6760 loss: 10.5120 decode.loss_cls_ce: 2.3265 decode.loss_mask_ce: 0.9861 decode.loss_mask_dice: 1.9392 decode.d7.loss_cls_ce: 2.3386 decode.d7.loss_mask_ce: 0.9861 decode.d7.loss_mask_dice: 1.9354 2023/09/06 15:28:40 - mmengine - INFO - Iter(train) [ 7500/60000] base_lr: 8.7501e-05 lr: 8.7501e-05 eta: 6:33:23 time: 0.4502 data_time: 0.0229 memory: 15761 grad_norm: 17.3675 loss: 10.5379 decode.loss_cls_ce: 2.3848 decode.loss_mask_ce: 0.9268 decode.loss_mask_dice: 1.9392 decode.d7.loss_cls_ce: 2.3961 decode.d7.loss_mask_ce: 0.9311 decode.d7.loss_mask_dice: 1.9601 2023/09/06 15:29:03 - mmengine - INFO - Iter(train) [ 7550/60000] base_lr: 8.7418e-05 lr: 8.7418e-05 eta: 6:33:02 time: 0.4483 data_time: 0.0231 memory: 15924 grad_norm: 19.4061 loss: 10.9009 decode.loss_cls_ce: 2.4141 decode.loss_mask_ce: 1.0206 decode.loss_mask_dice: 1.9865 decode.d7.loss_cls_ce: 2.4366 decode.d7.loss_mask_ce: 1.0457 decode.d7.loss_mask_dice: 1.9974 2023/09/06 15:29:26 - mmengine - INFO - Iter(train) [ 7600/60000] base_lr: 8.7335e-05 lr: 8.7335e-05 eta: 6:32:40 time: 0.4537 data_time: 0.0226 memory: 15885 grad_norm: 16.9512 loss: 10.0907 decode.loss_cls_ce: 2.2834 decode.loss_mask_ce: 0.9557 decode.loss_mask_dice: 1.8113 decode.d7.loss_cls_ce: 2.2779 decode.d7.loss_mask_ce: 0.9560 decode.d7.loss_mask_dice: 1.8064 2023/09/06 15:29:48 - mmengine - INFO - Iter(train) [ 7650/60000] base_lr: 8.7251e-05 lr: 8.7251e-05 eta: 6:32:18 time: 0.4501 data_time: 0.0235 memory: 15889 grad_norm: 17.4616 loss: 10.9091 decode.loss_cls_ce: 2.4271 decode.loss_mask_ce: 1.0118 decode.loss_mask_dice: 2.0031 decode.d7.loss_cls_ce: 2.4372 decode.d7.loss_mask_ce: 1.0086 decode.d7.loss_mask_dice: 2.0213 2023/09/06 15:30:11 - mmengine - INFO - Iter(train) [ 7700/60000] base_lr: 8.7168e-05 lr: 8.7168e-05 eta: 6:31:57 time: 0.4516 data_time: 0.0237 memory: 15806 grad_norm: 19.8572 loss: 9.8383 decode.loss_cls_ce: 2.2268 decode.loss_mask_ce: 0.9540 decode.loss_mask_dice: 1.7371 decode.d7.loss_cls_ce: 2.2522 decode.d7.loss_mask_ce: 0.9445 decode.d7.loss_mask_dice: 1.7237 2023/09/06 15:30:33 - mmengine - INFO - Iter(train) [ 7750/60000] base_lr: 8.7085e-05 lr: 8.7085e-05 eta: 6:31:34 time: 0.4481 data_time: 0.0226 memory: 15907 grad_norm: 17.1993 loss: 9.3892 decode.loss_cls_ce: 2.0853 decode.loss_mask_ce: 0.8836 decode.loss_mask_dice: 1.7121 decode.d7.loss_cls_ce: 2.0753 decode.d7.loss_mask_ce: 0.9027 decode.d7.loss_mask_dice: 1.7302 2023/09/06 15:30:56 - mmengine - INFO - Iter(train) [ 7800/60000] base_lr: 8.7001e-05 lr: 8.7001e-05 eta: 6:31:12 time: 0.4483 data_time: 0.0224 memory: 15812 grad_norm: 18.3152 loss: 9.3761 decode.loss_cls_ce: 2.0925 decode.loss_mask_ce: 0.8816 decode.loss_mask_dice: 1.7060 decode.d7.loss_cls_ce: 2.0924 decode.d7.loss_mask_ce: 0.8855 decode.d7.loss_mask_dice: 1.7180 2023/09/06 15:31:18 - mmengine - INFO - Iter(train) [ 7850/60000] base_lr: 8.6918e-05 lr: 8.6918e-05 eta: 6:30:50 time: 0.4525 data_time: 0.0231 memory: 15822 grad_norm: 21.2029 loss: 10.9937 decode.loss_cls_ce: 2.5159 decode.loss_mask_ce: 0.9901 decode.loss_mask_dice: 1.9697 decode.d7.loss_cls_ce: 2.5468 decode.d7.loss_mask_ce: 0.9986 decode.d7.loss_mask_dice: 1.9726 2023/09/06 15:31:41 - mmengine - INFO - Iter(train) [ 7900/60000] base_lr: 8.6835e-05 lr: 8.6835e-05 eta: 6:30:27 time: 0.4502 data_time: 0.0227 memory: 15772 grad_norm: 18.3470 loss: 10.0973 decode.loss_cls_ce: 2.2780 decode.loss_mask_ce: 0.8995 decode.loss_mask_dice: 1.8706 decode.d7.loss_cls_ce: 2.3021 decode.d7.loss_mask_ce: 0.8897 decode.d7.loss_mask_dice: 1.8574 2023/09/06 15:32:03 - mmengine - INFO - Iter(train) [ 7950/60000] base_lr: 8.6751e-05 lr: 8.6751e-05 eta: 6:30:05 time: 0.4500 data_time: 0.0226 memory: 15784 grad_norm: 17.7976 loss: 11.4766 decode.loss_cls_ce: 2.4239 decode.loss_mask_ce: 1.0868 decode.loss_mask_dice: 2.2375 decode.d7.loss_cls_ce: 2.4287 decode.d7.loss_mask_ce: 1.0816 decode.d7.loss_mask_dice: 2.2181 2023/09/06 15:32:26 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:32:26 - mmengine - INFO - Iter(train) [ 8000/60000] base_lr: 8.6668e-05 lr: 8.6668e-05 eta: 6:29:42 time: 0.4497 data_time: 0.0228 memory: 15861 grad_norm: 16.9897 loss: 10.2225 decode.loss_cls_ce: 2.2473 decode.loss_mask_ce: 0.9718 decode.loss_mask_dice: 1.8728 decode.d7.loss_cls_ce: 2.2549 decode.d7.loss_mask_ce: 0.9856 decode.d7.loss_mask_dice: 1.8902 2023/09/06 15:32:48 - mmengine - INFO - Iter(train) [ 8050/60000] base_lr: 8.6585e-05 lr: 8.6585e-05 eta: 6:29:20 time: 0.4536 data_time: 0.0236 memory: 15821 grad_norm: 20.3044 loss: 10.9453 decode.loss_cls_ce: 2.4296 decode.loss_mask_ce: 1.0683 decode.loss_mask_dice: 1.9636 decode.d7.loss_cls_ce: 2.4615 decode.d7.loss_mask_ce: 1.0627 decode.d7.loss_mask_dice: 1.9596 2023/09/06 15:33:11 - mmengine - INFO - Iter(train) [ 8100/60000] base_lr: 8.6501e-05 lr: 8.6501e-05 eta: 6:28:58 time: 0.4506 data_time: 0.0234 memory: 15745 grad_norm: 17.3407 loss: 11.0128 decode.loss_cls_ce: 2.4187 decode.loss_mask_ce: 1.0678 decode.loss_mask_dice: 2.0160 decode.d7.loss_cls_ce: 2.4174 decode.d7.loss_mask_ce: 1.0789 decode.d7.loss_mask_dice: 2.0139 2023/09/06 15:33:34 - mmengine - INFO - Iter(train) [ 8150/60000] base_lr: 8.6418e-05 lr: 8.6418e-05 eta: 6:28:36 time: 0.4546 data_time: 0.0221 memory: 15871 grad_norm: 16.8417 loss: 9.9086 decode.loss_cls_ce: 2.2228 decode.loss_mask_ce: 0.9945 decode.loss_mask_dice: 1.7321 decode.d7.loss_cls_ce: 2.2287 decode.d7.loss_mask_ce: 0.9954 decode.d7.loss_mask_dice: 1.7351 2023/09/06 15:33:56 - mmengine - INFO - Iter(train) [ 8200/60000] base_lr: 8.6335e-05 lr: 8.6335e-05 eta: 6:28:14 time: 0.4516 data_time: 0.0231 memory: 15924 grad_norm: 16.9323 loss: 10.7127 decode.loss_cls_ce: 2.3309 decode.loss_mask_ce: 0.9984 decode.loss_mask_dice: 2.0023 decode.d7.loss_cls_ce: 2.3746 decode.d7.loss_mask_ce: 0.9991 decode.d7.loss_mask_dice: 2.0074 2023/09/06 15:34:19 - mmengine - INFO - Iter(train) [ 8250/60000] base_lr: 8.6251e-05 lr: 8.6251e-05 eta: 6:27:53 time: 0.4553 data_time: 0.0225 memory: 15822 grad_norm: 17.8490 loss: 10.0405 decode.loss_cls_ce: 2.0184 decode.loss_mask_ce: 1.0293 decode.loss_mask_dice: 1.9741 decode.d7.loss_cls_ce: 2.0076 decode.d7.loss_mask_ce: 1.0371 decode.d7.loss_mask_dice: 1.9741 2023/09/06 15:34:41 - mmengine - INFO - Iter(train) [ 8300/60000] base_lr: 8.6168e-05 lr: 8.6168e-05 eta: 6:27:31 time: 0.4503 data_time: 0.0234 memory: 15849 grad_norm: 16.5251 loss: 11.7938 decode.loss_cls_ce: 2.6250 decode.loss_mask_ce: 1.0418 decode.loss_mask_dice: 2.2149 decode.d7.loss_cls_ce: 2.6636 decode.d7.loss_mask_ce: 1.0279 decode.d7.loss_mask_dice: 2.2207 2023/09/06 15:35:04 - mmengine - INFO - Iter(train) [ 8350/60000] base_lr: 8.6085e-05 lr: 8.6085e-05 eta: 6:27:09 time: 0.4494 data_time: 0.0233 memory: 15872 grad_norm: 17.4606 loss: 10.4576 decode.loss_cls_ce: 2.3910 decode.loss_mask_ce: 0.8990 decode.loss_mask_dice: 1.9364 decode.d7.loss_cls_ce: 2.3954 decode.d7.loss_mask_ce: 0.8939 decode.d7.loss_mask_dice: 1.9420 2023/09/06 15:35:27 - mmengine - INFO - Iter(train) [ 8400/60000] base_lr: 8.6001e-05 lr: 8.6001e-05 eta: 6:26:47 time: 0.4561 data_time: 0.0226 memory: 15831 grad_norm: 16.9387 loss: 11.1871 decode.loss_cls_ce: 2.3985 decode.loss_mask_ce: 1.0705 decode.loss_mask_dice: 2.1137 decode.d7.loss_cls_ce: 2.4079 decode.d7.loss_mask_ce: 1.0787 decode.d7.loss_mask_dice: 2.1179 2023/09/06 15:35:49 - mmengine - INFO - Iter(train) [ 8450/60000] base_lr: 8.5918e-05 lr: 8.5918e-05 eta: 6:26:26 time: 0.4533 data_time: 0.0229 memory: 15883 grad_norm: 17.8626 loss: 10.9259 decode.loss_cls_ce: 2.3636 decode.loss_mask_ce: 1.0064 decode.loss_mask_dice: 2.0674 decode.d7.loss_cls_ce: 2.4134 decode.d7.loss_mask_ce: 1.0052 decode.d7.loss_mask_dice: 2.0700 2023/09/06 15:36:12 - mmengine - INFO - Iter(train) [ 8500/60000] base_lr: 8.5835e-05 lr: 8.5835e-05 eta: 6:26:04 time: 0.4519 data_time: 0.0234 memory: 15859 grad_norm: 20.7808 loss: 11.1073 decode.loss_cls_ce: 2.4766 decode.loss_mask_ce: 0.9982 decode.loss_mask_dice: 2.0805 decode.d7.loss_cls_ce: 2.5077 decode.d7.loss_mask_ce: 0.9921 decode.d7.loss_mask_dice: 2.0521 2023/09/06 15:36:34 - mmengine - INFO - Iter(train) [ 8550/60000] base_lr: 8.5751e-05 lr: 8.5751e-05 eta: 6:25:42 time: 0.4509 data_time: 0.0233 memory: 15732 grad_norm: 16.0487 loss: 10.9601 decode.loss_cls_ce: 2.4122 decode.loss_mask_ce: 1.0295 decode.loss_mask_dice: 2.0358 decode.d7.loss_cls_ce: 2.4360 decode.d7.loss_mask_ce: 1.0242 decode.d7.loss_mask_dice: 2.0223 2023/09/06 15:36:57 - mmengine - INFO - Iter(train) [ 8600/60000] base_lr: 8.5668e-05 lr: 8.5668e-05 eta: 6:25:20 time: 0.4512 data_time: 0.0229 memory: 15860 grad_norm: 18.5839 loss: 8.8553 decode.loss_cls_ce: 2.0803 decode.loss_mask_ce: 0.8435 decode.loss_mask_dice: 1.5033 decode.d7.loss_cls_ce: 2.0781 decode.d7.loss_mask_ce: 0.8444 decode.d7.loss_mask_dice: 1.5057 2023/09/06 15:37:19 - mmengine - INFO - Iter(train) [ 8650/60000] base_lr: 8.5585e-05 lr: 8.5585e-05 eta: 6:24:57 time: 0.4516 data_time: 0.0223 memory: 15831 grad_norm: 15.9713 loss: 11.0095 decode.loss_cls_ce: 2.5048 decode.loss_mask_ce: 1.0307 decode.loss_mask_dice: 1.9727 decode.d7.loss_cls_ce: 2.5075 decode.d7.loss_mask_ce: 1.0278 decode.d7.loss_mask_dice: 1.9660 2023/09/06 15:37:42 - mmengine - INFO - Iter(train) [ 8700/60000] base_lr: 8.5501e-05 lr: 8.5501e-05 eta: 6:24:35 time: 0.4494 data_time: 0.0232 memory: 15870 grad_norm: 19.8889 loss: 11.2443 decode.loss_cls_ce: 2.4972 decode.loss_mask_ce: 1.0132 decode.loss_mask_dice: 2.1275 decode.d7.loss_cls_ce: 2.4705 decode.d7.loss_mask_ce: 1.0191 decode.d7.loss_mask_dice: 2.1169 2023/09/06 15:38:05 - mmengine - INFO - Iter(train) [ 8750/60000] base_lr: 8.5418e-05 lr: 8.5418e-05 eta: 6:24:13 time: 0.4488 data_time: 0.0231 memory: 15761 grad_norm: 16.7302 loss: 10.9935 decode.loss_cls_ce: 2.4049 decode.loss_mask_ce: 1.0903 decode.loss_mask_dice: 1.9955 decode.d7.loss_cls_ce: 2.4028 decode.d7.loss_mask_ce: 1.0908 decode.d7.loss_mask_dice: 2.0092 2023/09/06 15:38:27 - mmengine - INFO - Iter(train) [ 8800/60000] base_lr: 8.5335e-05 lr: 8.5335e-05 eta: 6:23:50 time: 0.4498 data_time: 0.0238 memory: 15745 grad_norm: 18.3129 loss: 10.9417 decode.loss_cls_ce: 2.5596 decode.loss_mask_ce: 0.9778 decode.loss_mask_dice: 1.9280 decode.d7.loss_cls_ce: 2.5776 decode.d7.loss_mask_ce: 0.9759 decode.d7.loss_mask_dice: 1.9228 2023/09/06 15:38:50 - mmengine - INFO - Iter(train) [ 8850/60000] base_lr: 8.5251e-05 lr: 8.5251e-05 eta: 6:23:28 time: 0.4485 data_time: 0.0230 memory: 15810 grad_norm: 20.5574 loss: 11.2988 decode.loss_cls_ce: 2.5667 decode.loss_mask_ce: 0.9790 decode.loss_mask_dice: 2.0900 decode.d7.loss_cls_ce: 2.6002 decode.d7.loss_mask_ce: 0.9756 decode.d7.loss_mask_dice: 2.0874 2023/09/06 15:39:12 - mmengine - INFO - Iter(train) [ 8900/60000] base_lr: 8.5168e-05 lr: 8.5168e-05 eta: 6:23:06 time: 0.4543 data_time: 0.0223 memory: 16051 grad_norm: 16.2314 loss: 11.9221 decode.loss_cls_ce: 2.5841 decode.loss_mask_ce: 1.0076 decode.loss_mask_dice: 2.3783 decode.d7.loss_cls_ce: 2.5734 decode.d7.loss_mask_ce: 1.0131 decode.d7.loss_mask_dice: 2.3657 2023/09/06 15:39:35 - mmengine - INFO - Iter(train) [ 8950/60000] base_lr: 8.5085e-05 lr: 8.5085e-05 eta: 6:22:44 time: 0.4506 data_time: 0.0232 memory: 15987 grad_norm: 17.2003 loss: 9.6249 decode.loss_cls_ce: 2.0801 decode.loss_mask_ce: 0.9142 decode.loss_mask_dice: 1.8214 decode.d7.loss_cls_ce: 2.0685 decode.d7.loss_mask_ce: 0.9052 decode.d7.loss_mask_dice: 1.8355 2023/09/06 15:39:57 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:39:57 - mmengine - INFO - Iter(train) [ 9000/60000] base_lr: 8.5001e-05 lr: 8.5001e-05 eta: 6:22:21 time: 0.4501 data_time: 0.0239 memory: 15780 grad_norm: 18.3240 loss: 9.6895 decode.loss_cls_ce: 2.2244 decode.loss_mask_ce: 0.8846 decode.loss_mask_dice: 1.7312 decode.d7.loss_cls_ce: 2.2237 decode.d7.loss_mask_ce: 0.8860 decode.d7.loss_mask_dice: 1.7396 2023/09/06 15:40:20 - mmengine - INFO - Iter(train) [ 9050/60000] base_lr: 8.4918e-05 lr: 8.4918e-05 eta: 6:21:59 time: 0.4508 data_time: 0.0239 memory: 15885 grad_norm: 17.4756 loss: 11.2388 decode.loss_cls_ce: 2.5183 decode.loss_mask_ce: 1.0234 decode.loss_mask_dice: 2.0784 decode.d7.loss_cls_ce: 2.5065 decode.d7.loss_mask_ce: 1.0325 decode.d7.loss_mask_dice: 2.0797 2023/09/06 15:40:42 - mmengine - INFO - Iter(train) [ 9100/60000] base_lr: 8.4835e-05 lr: 8.4835e-05 eta: 6:21:38 time: 0.4488 data_time: 0.0224 memory: 15862 grad_norm: 17.4938 loss: 10.8219 decode.loss_cls_ce: 2.3661 decode.loss_mask_ce: 0.9944 decode.loss_mask_dice: 2.0691 decode.d7.loss_cls_ce: 2.3309 decode.d7.loss_mask_ce: 0.9872 decode.d7.loss_mask_dice: 2.0742 2023/09/06 15:41:05 - mmengine - INFO - Iter(train) [ 9150/60000] base_lr: 8.4751e-05 lr: 8.4751e-05 eta: 6:21:15 time: 0.4506 data_time: 0.0229 memory: 15961 grad_norm: 15.8692 loss: 9.9520 decode.loss_cls_ce: 2.3625 decode.loss_mask_ce: 0.8796 decode.loss_mask_dice: 1.7577 decode.d7.loss_cls_ce: 2.3159 decode.d7.loss_mask_ce: 0.8733 decode.d7.loss_mask_dice: 1.7631 2023/09/06 15:41:28 - mmengine - INFO - Iter(train) [ 9200/60000] base_lr: 8.4668e-05 lr: 8.4668e-05 eta: 6:20:54 time: 0.4593 data_time: 0.0236 memory: 15833 grad_norm: 18.2125 loss: 11.2439 decode.loss_cls_ce: 2.4401 decode.loss_mask_ce: 0.9902 decode.loss_mask_dice: 2.1806 decode.d7.loss_cls_ce: 2.4647 decode.d7.loss_mask_ce: 0.9891 decode.d7.loss_mask_dice: 2.1793 2023/09/06 15:41:50 - mmengine - INFO - Iter(train) [ 9250/60000] base_lr: 8.4585e-05 lr: 8.4585e-05 eta: 6:20:32 time: 0.4514 data_time: 0.0229 memory: 15936 grad_norm: 19.2106 loss: 9.9103 decode.loss_cls_ce: 2.1486 decode.loss_mask_ce: 0.9401 decode.loss_mask_dice: 1.8569 decode.d7.loss_cls_ce: 2.1537 decode.d7.loss_mask_ce: 0.9488 decode.d7.loss_mask_dice: 1.8622 2023/09/06 15:42:13 - mmengine - INFO - Iter(train) [ 9300/60000] base_lr: 8.4501e-05 lr: 8.4501e-05 eta: 6:20:11 time: 0.4538 data_time: 0.0227 memory: 15858 grad_norm: 18.3978 loss: 11.0885 decode.loss_cls_ce: 2.5592 decode.loss_mask_ce: 1.0022 decode.loss_mask_dice: 1.9730 decode.d7.loss_cls_ce: 2.5750 decode.d7.loss_mask_ce: 0.9982 decode.d7.loss_mask_dice: 1.9809 2023/09/06 15:42:35 - mmengine - INFO - Iter(train) [ 9350/60000] base_lr: 8.4418e-05 lr: 8.4418e-05 eta: 6:19:48 time: 0.4498 data_time: 0.0235 memory: 15861 grad_norm: 16.5345 loss: 9.7767 decode.loss_cls_ce: 2.2432 decode.loss_mask_ce: 0.9316 decode.loss_mask_dice: 1.7049 decode.d7.loss_cls_ce: 2.2616 decode.d7.loss_mask_ce: 0.9288 decode.d7.loss_mask_dice: 1.7065 2023/09/06 15:42:58 - mmengine - INFO - Iter(train) [ 9400/60000] base_lr: 8.4335e-05 lr: 8.4335e-05 eta: 6:19:26 time: 0.4511 data_time: 0.0234 memory: 15833 grad_norm: 19.8068 loss: 9.6046 decode.loss_cls_ce: 2.1706 decode.loss_mask_ce: 0.9205 decode.loss_mask_dice: 1.7054 decode.d7.loss_cls_ce: 2.1782 decode.d7.loss_mask_ce: 0.9213 decode.d7.loss_mask_dice: 1.7085 2023/09/06 15:43:20 - mmengine - INFO - Iter(train) [ 9450/60000] base_lr: 8.4251e-05 lr: 8.4251e-05 eta: 6:19:03 time: 0.4501 data_time: 0.0239 memory: 15910 grad_norm: 16.7589 loss: 9.5303 decode.loss_cls_ce: 2.1573 decode.loss_mask_ce: 0.8310 decode.loss_mask_dice: 1.7716 decode.d7.loss_cls_ce: 2.1542 decode.d7.loss_mask_ce: 0.8336 decode.d7.loss_mask_dice: 1.7827 2023/09/06 15:43:43 - mmengine - INFO - Iter(train) [ 9500/60000] base_lr: 8.4168e-05 lr: 8.4168e-05 eta: 6:18:41 time: 0.4494 data_time: 0.0231 memory: 15696 grad_norm: 17.8644 loss: 9.4733 decode.loss_cls_ce: 2.2001 decode.loss_mask_ce: 0.9058 decode.loss_mask_dice: 1.6212 decode.d7.loss_cls_ce: 2.1870 decode.d7.loss_mask_ce: 0.9061 decode.d7.loss_mask_dice: 1.6530 2023/09/06 15:44:06 - mmengine - INFO - Iter(train) [ 9550/60000] base_lr: 8.4085e-05 lr: 8.4085e-05 eta: 6:18:19 time: 0.4499 data_time: 0.0231 memory: 15952 grad_norm: 16.5849 loss: 10.0465 decode.loss_cls_ce: 2.3767 decode.loss_mask_ce: 0.9318 decode.loss_mask_dice: 1.7082 decode.d7.loss_cls_ce: 2.3857 decode.d7.loss_mask_ce: 0.9335 decode.d7.loss_mask_dice: 1.7106 2023/09/06 15:44:28 - mmengine - INFO - Iter(train) [ 9600/60000] base_lr: 8.4001e-05 lr: 8.4001e-05 eta: 6:17:56 time: 0.4501 data_time: 0.0235 memory: 15874 grad_norm: 15.6860 loss: 9.7324 decode.loss_cls_ce: 2.2273 decode.loss_mask_ce: 0.8584 decode.loss_mask_dice: 1.7779 decode.d7.loss_cls_ce: 2.2310 decode.d7.loss_mask_ce: 0.8713 decode.d7.loss_mask_dice: 1.7665 2023/09/06 15:44:51 - mmengine - INFO - Iter(train) [ 9650/60000] base_lr: 8.3918e-05 lr: 8.3918e-05 eta: 6:17:34 time: 0.4489 data_time: 0.0231 memory: 15926 grad_norm: 17.4653 loss: 9.7639 decode.loss_cls_ce: 2.1729 decode.loss_mask_ce: 0.8609 decode.loss_mask_dice: 1.8653 decode.d7.loss_cls_ce: 2.1548 decode.d7.loss_mask_ce: 0.8569 decode.d7.loss_mask_dice: 1.8531 2023/09/06 15:45:13 - mmengine - INFO - Iter(train) [ 9700/60000] base_lr: 8.3835e-05 lr: 8.3835e-05 eta: 6:17:12 time: 0.4497 data_time: 0.0237 memory: 15872 grad_norm: 17.9974 loss: 11.2715 decode.loss_cls_ce: 2.3840 decode.loss_mask_ce: 1.0672 decode.loss_mask_dice: 2.1712 decode.d7.loss_cls_ce: 2.3931 decode.d7.loss_mask_ce: 1.0820 decode.d7.loss_mask_dice: 2.1742 2023/09/06 15:45:36 - mmengine - INFO - Iter(train) [ 9750/60000] base_lr: 8.3751e-05 lr: 8.3751e-05 eta: 6:16:49 time: 0.4500 data_time: 0.0231 memory: 15731 grad_norm: 17.2205 loss: 10.5322 decode.loss_cls_ce: 2.4068 decode.loss_mask_ce: 0.9692 decode.loss_mask_dice: 1.8923 decode.d7.loss_cls_ce: 2.3780 decode.d7.loss_mask_ce: 0.9845 decode.d7.loss_mask_dice: 1.9014 2023/09/06 15:45:58 - mmengine - INFO - Iter(train) [ 9800/60000] base_lr: 8.3668e-05 lr: 8.3668e-05 eta: 6:16:27 time: 0.4495 data_time: 0.0234 memory: 16065 grad_norm: 17.7410 loss: 11.3063 decode.loss_cls_ce: 2.5728 decode.loss_mask_ce: 0.9685 decode.loss_mask_dice: 2.1376 decode.d7.loss_cls_ce: 2.5179 decode.d7.loss_mask_ce: 0.9694 decode.d7.loss_mask_dice: 2.1401 2023/09/06 15:46:21 - mmengine - INFO - Iter(train) [ 9850/60000] base_lr: 8.3585e-05 lr: 8.3585e-05 eta: 6:16:05 time: 0.4506 data_time: 0.0235 memory: 15735 grad_norm: 17.4582 loss: 10.0421 decode.loss_cls_ce: 2.3836 decode.loss_mask_ce: 0.8788 decode.loss_mask_dice: 1.7630 decode.d7.loss_cls_ce: 2.3772 decode.d7.loss_mask_ce: 0.8809 decode.d7.loss_mask_dice: 1.7585 2023/09/06 15:46:43 - mmengine - INFO - Iter(train) [ 9900/60000] base_lr: 8.3501e-05 lr: 8.3501e-05 eta: 6:15:42 time: 0.4492 data_time: 0.0235 memory: 15910 grad_norm: 15.6258 loss: 9.9153 decode.loss_cls_ce: 2.0995 decode.loss_mask_ce: 0.9596 decode.loss_mask_dice: 1.8942 decode.d7.loss_cls_ce: 2.1207 decode.d7.loss_mask_ce: 0.9581 decode.d7.loss_mask_dice: 1.8834 2023/09/06 15:47:06 - mmengine - INFO - Iter(train) [ 9950/60000] base_lr: 8.3418e-05 lr: 8.3418e-05 eta: 6:15:20 time: 0.4531 data_time: 0.0235 memory: 15934 grad_norm: 17.4380 loss: 11.0310 decode.loss_cls_ce: 2.4287 decode.loss_mask_ce: 1.0359 decode.loss_mask_dice: 2.0542 decode.d7.loss_cls_ce: 2.4088 decode.d7.loss_mask_ce: 1.0381 decode.d7.loss_mask_dice: 2.0652 2023/09/06 15:47:28 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:47:28 - mmengine - INFO - Iter(train) [10000/60000] base_lr: 8.3335e-05 lr: 8.3335e-05 eta: 6:14:58 time: 0.4529 data_time: 0.0229 memory: 15846 grad_norm: 16.9902 loss: 11.0647 decode.loss_cls_ce: 2.3863 decode.loss_mask_ce: 1.0457 decode.loss_mask_dice: 2.0915 decode.d7.loss_cls_ce: 2.3966 decode.d7.loss_mask_ce: 1.0306 decode.d7.loss_mask_dice: 2.1141 2023/09/06 15:47:28 - mmengine - INFO - Saving checkpoint at 10000 iterations 2023/09/06 15:47:54 - mmengine - INFO - Iter(train) [10050/60000] base_lr: 8.3251e-05 lr: 8.3251e-05 eta: 6:14:51 time: 0.4500 data_time: 0.0232 memory: 15935 grad_norm: 20.3147 loss: 9.6385 decode.loss_cls_ce: 2.2157 decode.loss_mask_ce: 0.8698 decode.loss_mask_dice: 1.7077 decode.d7.loss_cls_ce: 2.2537 decode.d7.loss_mask_ce: 0.8730 decode.d7.loss_mask_dice: 1.7186 2023/09/06 15:48:17 - mmengine - INFO - Iter(train) [10100/60000] base_lr: 8.3168e-05 lr: 8.3168e-05 eta: 6:14:29 time: 0.4580 data_time: 0.0225 memory: 15761 grad_norm: 16.0314 loss: 10.6184 decode.loss_cls_ce: 2.3650 decode.loss_mask_ce: 0.9793 decode.loss_mask_dice: 1.9558 decode.d7.loss_cls_ce: 2.3737 decode.d7.loss_mask_ce: 0.9791 decode.d7.loss_mask_dice: 1.9654 2023/09/06 15:48:39 - mmengine - INFO - Iter(train) [10150/60000] base_lr: 8.3085e-05 lr: 8.3085e-05 eta: 6:14:07 time: 0.4490 data_time: 0.0240 memory: 16015 grad_norm: 20.5088 loss: 8.9724 decode.loss_cls_ce: 2.1083 decode.loss_mask_ce: 0.8172 decode.loss_mask_dice: 1.5579 decode.d7.loss_cls_ce: 2.1098 decode.d7.loss_mask_ce: 0.8054 decode.d7.loss_mask_dice: 1.5737 2023/09/06 15:49:02 - mmengine - INFO - Iter(train) [10200/60000] base_lr: 8.3001e-05 lr: 8.3001e-05 eta: 6:13:45 time: 0.4524 data_time: 0.0231 memory: 15987 grad_norm: 17.6418 loss: 11.5160 decode.loss_cls_ce: 2.5691 decode.loss_mask_ce: 1.0289 decode.loss_mask_dice: 2.1615 decode.d7.loss_cls_ce: 2.5538 decode.d7.loss_mask_ce: 1.0427 decode.d7.loss_mask_dice: 2.1600 2023/09/06 15:49:24 - mmengine - INFO - Iter(train) [10250/60000] base_lr: 8.2918e-05 lr: 8.2918e-05 eta: 6:13:23 time: 0.4503 data_time: 0.0232 memory: 15874 grad_norm: 15.7753 loss: 10.5888 decode.loss_cls_ce: 2.3617 decode.loss_mask_ce: 0.9559 decode.loss_mask_dice: 1.9762 decode.d7.loss_cls_ce: 2.3624 decode.d7.loss_mask_ce: 0.9460 decode.d7.loss_mask_dice: 1.9868 2023/09/06 15:49:47 - mmengine - INFO - Iter(train) [10300/60000] base_lr: 8.2835e-05 lr: 8.2835e-05 eta: 6:13:00 time: 0.4488 data_time: 0.0234 memory: 15923 grad_norm: 16.2527 loss: 10.7840 decode.loss_cls_ce: 2.5094 decode.loss_mask_ce: 0.9712 decode.loss_mask_dice: 1.9107 decode.d7.loss_cls_ce: 2.5189 decode.d7.loss_mask_ce: 0.9704 decode.d7.loss_mask_dice: 1.9034 2023/09/06 15:50:10 - mmengine - INFO - Iter(train) [10350/60000] base_lr: 8.2751e-05 lr: 8.2751e-05 eta: 6:12:38 time: 0.4506 data_time: 0.0238 memory: 15870 grad_norm: 17.3049 loss: 11.0339 decode.loss_cls_ce: 2.3844 decode.loss_mask_ce: 0.9773 decode.loss_mask_dice: 2.1313 decode.d7.loss_cls_ce: 2.4221 decode.d7.loss_mask_ce: 0.9738 decode.d7.loss_mask_dice: 2.1451 2023/09/06 15:50:32 - mmengine - INFO - Iter(train) [10400/60000] base_lr: 8.2668e-05 lr: 8.2668e-05 eta: 6:12:15 time: 0.4490 data_time: 0.0233 memory: 15820 grad_norm: 16.7230 loss: 9.6549 decode.loss_cls_ce: 2.0860 decode.loss_mask_ce: 0.9982 decode.loss_mask_dice: 1.7432 decode.d7.loss_cls_ce: 2.0720 decode.d7.loss_mask_ce: 0.9985 decode.d7.loss_mask_dice: 1.7568 2023/09/06 15:50:55 - mmengine - INFO - Iter(train) [10450/60000] base_lr: 8.2585e-05 lr: 8.2585e-05 eta: 6:11:53 time: 0.4529 data_time: 0.0231 memory: 15844 grad_norm: 16.8054 loss: 9.2842 decode.loss_cls_ce: 2.0946 decode.loss_mask_ce: 0.8699 decode.loss_mask_dice: 1.6852 decode.d7.loss_cls_ce: 2.0925 decode.d7.loss_mask_ce: 0.8670 decode.d7.loss_mask_dice: 1.6751 2023/09/06 15:51:17 - mmengine - INFO - Iter(train) [10500/60000] base_lr: 8.2501e-05 lr: 8.2501e-05 eta: 6:11:30 time: 0.4490 data_time: 0.0232 memory: 15831 grad_norm: 16.8498 loss: 10.7653 decode.loss_cls_ce: 2.3858 decode.loss_mask_ce: 1.0377 decode.loss_mask_dice: 1.9499 decode.d7.loss_cls_ce: 2.4074 decode.d7.loss_mask_ce: 1.0343 decode.d7.loss_mask_dice: 1.9503 2023/09/06 15:51:40 - mmengine - INFO - Iter(train) [10550/60000] base_lr: 8.2418e-05 lr: 8.2418e-05 eta: 6:11:07 time: 0.4492 data_time: 0.0233 memory: 16167 grad_norm: 17.4124 loss: 10.9956 decode.loss_cls_ce: 2.5214 decode.loss_mask_ce: 0.9259 decode.loss_mask_dice: 2.0602 decode.d7.loss_cls_ce: 2.5229 decode.d7.loss_mask_ce: 0.9224 decode.d7.loss_mask_dice: 2.0427 2023/09/06 15:52:02 - mmengine - INFO - Iter(train) [10600/60000] base_lr: 8.2335e-05 lr: 8.2335e-05 eta: 6:10:45 time: 0.4502 data_time: 0.0230 memory: 15833 grad_norm: 19.8119 loss: 11.7115 decode.loss_cls_ce: 2.5437 decode.loss_mask_ce: 1.0130 decode.loss_mask_dice: 2.3017 decode.d7.loss_cls_ce: 2.5254 decode.d7.loss_mask_ce: 1.0127 decode.d7.loss_mask_dice: 2.3149 2023/09/06 15:52:25 - mmengine - INFO - Iter(train) [10650/60000] base_lr: 8.2251e-05 lr: 8.2251e-05 eta: 6:10:22 time: 0.4519 data_time: 0.0230 memory: 15784 grad_norm: 17.1864 loss: 10.0561 decode.loss_cls_ce: 2.2674 decode.loss_mask_ce: 0.9866 decode.loss_mask_dice: 1.7593 decode.d7.loss_cls_ce: 2.2969 decode.d7.loss_mask_ce: 0.9885 decode.d7.loss_mask_dice: 1.7573 2023/09/06 15:52:47 - mmengine - INFO - Iter(train) [10700/60000] base_lr: 8.2168e-05 lr: 8.2168e-05 eta: 6:10:00 time: 0.4512 data_time: 0.0232 memory: 15872 grad_norm: 17.0909 loss: 10.2686 decode.loss_cls_ce: 2.2727 decode.loss_mask_ce: 0.9532 decode.loss_mask_dice: 1.9011 decode.d7.loss_cls_ce: 2.2767 decode.d7.loss_mask_ce: 0.9632 decode.d7.loss_mask_dice: 1.9016 2023/09/06 15:53:10 - mmengine - INFO - Iter(train) [10750/60000] base_lr: 8.2085e-05 lr: 8.2085e-05 eta: 6:09:38 time: 0.4562 data_time: 0.0236 memory: 15861 grad_norm: 17.4407 loss: 10.6024 decode.loss_cls_ce: 2.3054 decode.loss_mask_ce: 1.0185 decode.loss_mask_dice: 1.9634 decode.d7.loss_cls_ce: 2.3217 decode.d7.loss_mask_ce: 1.0204 decode.d7.loss_mask_dice: 1.9731 2023/09/06 15:53:32 - mmengine - INFO - Iter(train) [10800/60000] base_lr: 8.2001e-05 lr: 8.2001e-05 eta: 6:09:16 time: 0.4527 data_time: 0.0234 memory: 15820 grad_norm: 17.1471 loss: 9.8510 decode.loss_cls_ce: 2.1227 decode.loss_mask_ce: 0.9321 decode.loss_mask_dice: 1.8470 decode.d7.loss_cls_ce: 2.1500 decode.d7.loss_mask_ce: 0.9447 decode.d7.loss_mask_dice: 1.8545 2023/09/06 15:53:55 - mmengine - INFO - Iter(train) [10850/60000] base_lr: 8.1918e-05 lr: 8.1918e-05 eta: 6:08:54 time: 0.4471 data_time: 0.0234 memory: 15844 grad_norm: 16.5501 loss: 9.4697 decode.loss_cls_ce: 2.2161 decode.loss_mask_ce: 0.8915 decode.loss_mask_dice: 1.6182 decode.d7.loss_cls_ce: 2.2176 decode.d7.loss_mask_ce: 0.8904 decode.d7.loss_mask_dice: 1.6358 2023/09/06 15:54:18 - mmengine - INFO - Iter(train) [10900/60000] base_lr: 8.1835e-05 lr: 8.1835e-05 eta: 6:08:32 time: 0.4489 data_time: 0.0228 memory: 15774 grad_norm: 16.4447 loss: 10.1800 decode.loss_cls_ce: 2.3223 decode.loss_mask_ce: 0.9341 decode.loss_mask_dice: 1.8425 decode.d7.loss_cls_ce: 2.3083 decode.d7.loss_mask_ce: 0.9330 decode.d7.loss_mask_dice: 1.8398 2023/09/06 15:54:40 - mmengine - INFO - Iter(train) [10950/60000] base_lr: 8.1751e-05 lr: 8.1751e-05 eta: 6:08:09 time: 0.4489 data_time: 0.0238 memory: 15785 grad_norm: 17.0268 loss: 9.9092 decode.loss_cls_ce: 2.3238 decode.loss_mask_ce: 0.9262 decode.loss_mask_dice: 1.7061 decode.d7.loss_cls_ce: 2.3227 decode.d7.loss_mask_ce: 0.9283 decode.d7.loss_mask_dice: 1.7020 2023/09/06 15:55:03 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 15:55:03 - mmengine - INFO - Iter(train) [11000/60000] base_lr: 8.1668e-05 lr: 8.1668e-05 eta: 6:07:47 time: 0.4505 data_time: 0.0237 memory: 15857 grad_norm: 17.5236 loss: 9.6791 decode.loss_cls_ce: 2.1143 decode.loss_mask_ce: 0.9095 decode.loss_mask_dice: 1.8218 decode.d7.loss_cls_ce: 2.1025 decode.d7.loss_mask_ce: 0.9085 decode.d7.loss_mask_dice: 1.8226 2023/09/06 15:55:25 - mmengine - INFO - Iter(train) [11050/60000] base_lr: 8.1585e-05 lr: 8.1585e-05 eta: 6:07:24 time: 0.4491 data_time: 0.0234 memory: 15868 grad_norm: 17.1404 loss: 10.2243 decode.loss_cls_ce: 2.2862 decode.loss_mask_ce: 0.9832 decode.loss_mask_dice: 1.8543 decode.d7.loss_cls_ce: 2.2568 decode.d7.loss_mask_ce: 0.9839 decode.d7.loss_mask_dice: 1.8598 2023/09/06 15:55:48 - mmengine - INFO - Iter(train) [11100/60000] base_lr: 8.1501e-05 lr: 8.1501e-05 eta: 6:07:01 time: 0.4499 data_time: 0.0238 memory: 15834 grad_norm: 15.4026 loss: 9.7133 decode.loss_cls_ce: 2.2096 decode.loss_mask_ce: 0.8635 decode.loss_mask_dice: 1.7801 decode.d7.loss_cls_ce: 2.2103 decode.d7.loss_mask_ce: 0.8645 decode.d7.loss_mask_dice: 1.7852 2023/09/06 15:56:10 - mmengine - INFO - Iter(train) [11150/60000] base_lr: 8.1418e-05 lr: 8.1418e-05 eta: 6:06:39 time: 0.4478 data_time: 0.0228 memory: 15772 grad_norm: 17.4586 loss: 10.2127 decode.loss_cls_ce: 2.1856 decode.loss_mask_ce: 1.0365 decode.loss_mask_dice: 1.8722 decode.d7.loss_cls_ce: 2.2150 decode.d7.loss_mask_ce: 1.0361 decode.d7.loss_mask_dice: 1.8673 2023/09/06 15:56:33 - mmengine - INFO - Iter(train) [11200/60000] base_lr: 8.1335e-05 lr: 8.1335e-05 eta: 6:06:16 time: 0.4518 data_time: 0.0230 memory: 15871 grad_norm: 17.0823 loss: 9.9863 decode.loss_cls_ce: 2.3587 decode.loss_mask_ce: 0.8552 decode.loss_mask_dice: 1.7663 decode.d7.loss_cls_ce: 2.3614 decode.d7.loss_mask_ce: 0.8522 decode.d7.loss_mask_dice: 1.7924 2023/09/06 15:56:55 - mmengine - INFO - Iter(train) [11250/60000] base_lr: 8.1251e-05 lr: 8.1251e-05 eta: 6:05:54 time: 0.4539 data_time: 0.0233 memory: 16003 grad_norm: 17.6986 loss: 11.0490 decode.loss_cls_ce: 2.3545 decode.loss_mask_ce: 1.0983 decode.loss_mask_dice: 2.0706 decode.d7.loss_cls_ce: 2.3483 decode.d7.loss_mask_ce: 1.0948 decode.d7.loss_mask_dice: 2.0824 2023/09/06 15:57:18 - mmengine - INFO - Iter(train) [11300/60000] base_lr: 8.1168e-05 lr: 8.1168e-05 eta: 6:05:32 time: 0.4525 data_time: 0.0236 memory: 15888 grad_norm: 16.8859 loss: 9.8327 decode.loss_cls_ce: 2.0335 decode.loss_mask_ce: 1.0213 decode.loss_mask_dice: 1.8540 decode.d7.loss_cls_ce: 2.0301 decode.d7.loss_mask_ce: 1.0266 decode.d7.loss_mask_dice: 1.8672 2023/09/06 15:57:40 - mmengine - INFO - Iter(train) [11350/60000] base_lr: 8.1085e-05 lr: 8.1085e-05 eta: 6:05:09 time: 0.4507 data_time: 0.0232 memory: 15924 grad_norm: 16.5521 loss: 10.7878 decode.loss_cls_ce: 2.3571 decode.loss_mask_ce: 1.0322 decode.loss_mask_dice: 2.0006 decode.d7.loss_cls_ce: 2.3692 decode.d7.loss_mask_ce: 1.0265 decode.d7.loss_mask_dice: 2.0021 2023/09/06 15:58:03 - mmengine - INFO - Iter(train) [11400/60000] base_lr: 8.1001e-05 lr: 8.1001e-05 eta: 6:04:47 time: 0.4498 data_time: 0.0229 memory: 15988 grad_norm: 17.6116 loss: 10.2402 decode.loss_cls_ce: 2.3287 decode.loss_mask_ce: 0.9679 decode.loss_mask_dice: 1.8290 decode.d7.loss_cls_ce: 2.3191 decode.d7.loss_mask_ce: 0.9610 decode.d7.loss_mask_dice: 1.8346 2023/09/06 15:58:26 - mmengine - INFO - Iter(train) [11450/60000] base_lr: 8.0918e-05 lr: 8.0918e-05 eta: 6:04:25 time: 0.4482 data_time: 0.0233 memory: 15846 grad_norm: 15.7617 loss: 10.7073 decode.loss_cls_ce: 2.3222 decode.loss_mask_ce: 0.9909 decode.loss_mask_dice: 2.0250 decode.d7.loss_cls_ce: 2.3229 decode.d7.loss_mask_ce: 1.0047 decode.d7.loss_mask_dice: 2.0417 2023/09/06 15:58:48 - mmengine - INFO - Iter(train) [11500/60000] base_lr: 8.0835e-05 lr: 8.0835e-05 eta: 6:04:02 time: 0.4506 data_time: 0.0234 memory: 15809 grad_norm: 15.3036 loss: 8.7905 decode.loss_cls_ce: 2.0707 decode.loss_mask_ce: 0.8189 decode.loss_mask_dice: 1.5031 decode.d7.loss_cls_ce: 2.0630 decode.d7.loss_mask_ce: 0.8182 decode.d7.loss_mask_dice: 1.5166 2023/09/06 15:59:11 - mmengine - INFO - Iter(train) [11550/60000] base_lr: 8.0751e-05 lr: 8.0751e-05 eta: 6:03:40 time: 0.4487 data_time: 0.0233 memory: 15886 grad_norm: 17.7535 loss: 10.6063 decode.loss_cls_ce: 2.2935 decode.loss_mask_ce: 1.0182 decode.loss_mask_dice: 1.9861 decode.d7.loss_cls_ce: 2.2888 decode.d7.loss_mask_ce: 1.0161 decode.d7.loss_mask_dice: 2.0037 2023/09/06 15:59:33 - mmengine - INFO - Iter(train) [11600/60000] base_lr: 8.0668e-05 lr: 8.0668e-05 eta: 6:03:17 time: 0.4490 data_time: 0.0232 memory: 15914 grad_norm: 16.5834 loss: 9.6976 decode.loss_cls_ce: 2.1770 decode.loss_mask_ce: 0.9528 decode.loss_mask_dice: 1.7334 decode.d7.loss_cls_ce: 2.1534 decode.d7.loss_mask_ce: 0.9474 decode.d7.loss_mask_dice: 1.7336 2023/09/06 15:59:56 - mmengine - INFO - Iter(train) [11650/60000] base_lr: 8.0585e-05 lr: 8.0585e-05 eta: 6:02:55 time: 0.4550 data_time: 0.0236 memory: 15951 grad_norm: 17.1554 loss: 10.4411 decode.loss_cls_ce: 2.3516 decode.loss_mask_ce: 0.9669 decode.loss_mask_dice: 1.9060 decode.d7.loss_cls_ce: 2.3362 decode.d7.loss_mask_ce: 0.9661 decode.d7.loss_mask_dice: 1.9142 2023/09/06 16:00:19 - mmengine - INFO - Iter(train) [11700/60000] base_lr: 8.0501e-05 lr: 8.0501e-05 eta: 6:02:34 time: 0.4531 data_time: 0.0245 memory: 15784 grad_norm: 17.1437 loss: 9.7122 decode.loss_cls_ce: 2.2112 decode.loss_mask_ce: 0.9399 decode.loss_mask_dice: 1.7001 decode.d7.loss_cls_ce: 2.2013 decode.d7.loss_mask_ce: 0.9463 decode.d7.loss_mask_dice: 1.7135 2023/09/06 16:00:41 - mmengine - INFO - Iter(train) [11750/60000] base_lr: 8.0418e-05 lr: 8.0418e-05 eta: 6:02:11 time: 0.4536 data_time: 0.0231 memory: 15756 grad_norm: 18.8233 loss: 11.3000 decode.loss_cls_ce: 2.4927 decode.loss_mask_ce: 1.0656 decode.loss_mask_dice: 2.1054 decode.d7.loss_cls_ce: 2.4804 decode.d7.loss_mask_ce: 1.0700 decode.d7.loss_mask_dice: 2.0858 2023/09/06 16:01:04 - mmengine - INFO - Iter(train) [11800/60000] base_lr: 8.0335e-05 lr: 8.0335e-05 eta: 6:01:49 time: 0.4531 data_time: 0.0233 memory: 15719 grad_norm: 16.3838 loss: 10.1232 decode.loss_cls_ce: 2.1933 decode.loss_mask_ce: 0.9479 decode.loss_mask_dice: 1.9197 decode.d7.loss_cls_ce: 2.2139 decode.d7.loss_mask_ce: 0.9401 decode.d7.loss_mask_dice: 1.9084 2023/09/06 16:01:26 - mmengine - INFO - Iter(train) [11850/60000] base_lr: 8.0251e-05 lr: 8.0251e-05 eta: 6:01:26 time: 0.4520 data_time: 0.0230 memory: 15822 grad_norm: 16.9473 loss: 10.8604 decode.loss_cls_ce: 2.3502 decode.loss_mask_ce: 1.0343 decode.loss_mask_dice: 2.0474 decode.d7.loss_cls_ce: 2.3375 decode.d7.loss_mask_ce: 1.0440 decode.d7.loss_mask_dice: 2.0471 2023/09/06 16:01:49 - mmengine - INFO - Iter(train) [11900/60000] base_lr: 8.0168e-05 lr: 8.0168e-05 eta: 6:01:04 time: 0.4494 data_time: 0.0233 memory: 15847 grad_norm: 16.8052 loss: 10.9270 decode.loss_cls_ce: 2.3885 decode.loss_mask_ce: 0.9717 decode.loss_mask_dice: 2.0974 decode.d7.loss_cls_ce: 2.3870 decode.d7.loss_mask_ce: 0.9778 decode.d7.loss_mask_dice: 2.1046 2023/09/06 16:02:11 - mmengine - INFO - Iter(train) [11950/60000] base_lr: 8.0085e-05 lr: 8.0085e-05 eta: 6:00:42 time: 0.4473 data_time: 0.0227 memory: 15808 grad_norm: 18.0798 loss: 10.7272 decode.loss_cls_ce: 2.3893 decode.loss_mask_ce: 1.0060 decode.loss_mask_dice: 1.9546 decode.d7.loss_cls_ce: 2.4110 decode.d7.loss_mask_ce: 1.0086 decode.d7.loss_mask_dice: 1.9578 2023/09/06 16:02:34 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:02:34 - mmengine - INFO - Iter(train) [12000/60000] base_lr: 8.0001e-05 lr: 8.0001e-05 eta: 6:00:19 time: 0.4495 data_time: 0.0244 memory: 15821 grad_norm: 15.6656 loss: 10.6109 decode.loss_cls_ce: 2.1863 decode.loss_mask_ce: 1.0178 decode.loss_mask_dice: 2.0844 decode.d7.loss_cls_ce: 2.2358 decode.d7.loss_mask_ce: 1.0124 decode.d7.loss_mask_dice: 2.0740 2023/09/06 16:02:56 - mmengine - INFO - Iter(train) [12050/60000] base_lr: 7.9918e-05 lr: 7.9918e-05 eta: 5:59:56 time: 0.4496 data_time: 0.0230 memory: 16027 grad_norm: 17.8697 loss: 10.2693 decode.loss_cls_ce: 2.2598 decode.loss_mask_ce: 1.0284 decode.loss_mask_dice: 1.8642 decode.d7.loss_cls_ce: 2.2407 decode.d7.loss_mask_ce: 1.0327 decode.d7.loss_mask_dice: 1.8434 2023/09/06 16:03:19 - mmengine - INFO - Iter(train) [12100/60000] base_lr: 7.9835e-05 lr: 7.9835e-05 eta: 5:59:34 time: 0.4504 data_time: 0.0232 memory: 15691 grad_norm: 16.9104 loss: 9.1990 decode.loss_cls_ce: 2.0390 decode.loss_mask_ce: 0.8815 decode.loss_mask_dice: 1.6910 decode.d7.loss_cls_ce: 2.0147 decode.d7.loss_mask_ce: 0.8821 decode.d7.loss_mask_dice: 1.6908 2023/09/06 16:03:41 - mmengine - INFO - Iter(train) [12150/60000] base_lr: 7.9751e-05 lr: 7.9751e-05 eta: 5:59:11 time: 0.4499 data_time: 0.0230 memory: 15783 grad_norm: 15.9369 loss: 10.7467 decode.loss_cls_ce: 2.4434 decode.loss_mask_ce: 0.9196 decode.loss_mask_dice: 2.0181 decode.d7.loss_cls_ce: 2.4242 decode.d7.loss_mask_ce: 0.9253 decode.d7.loss_mask_dice: 2.0161 2023/09/06 16:04:04 - mmengine - INFO - Iter(train) [12200/60000] base_lr: 7.9668e-05 lr: 7.9668e-05 eta: 5:58:49 time: 0.4505 data_time: 0.0243 memory: 15968 grad_norm: 17.4585 loss: 10.0605 decode.loss_cls_ce: 2.2400 decode.loss_mask_ce: 1.0064 decode.loss_mask_dice: 1.7896 decode.d7.loss_cls_ce: 2.2461 decode.d7.loss_mask_ce: 0.9978 decode.d7.loss_mask_dice: 1.7806 2023/09/06 16:04:26 - mmengine - INFO - Iter(train) [12250/60000] base_lr: 7.9585e-05 lr: 7.9585e-05 eta: 5:58:26 time: 0.4496 data_time: 0.0235 memory: 15884 grad_norm: 18.0736 loss: 10.5257 decode.loss_cls_ce: 2.2207 decode.loss_mask_ce: 0.9303 decode.loss_mask_dice: 2.0870 decode.d7.loss_cls_ce: 2.2462 decode.d7.loss_mask_ce: 0.9416 decode.d7.loss_mask_dice: 2.1001 2023/09/06 16:04:49 - mmengine - INFO - Iter(train) [12300/60000] base_lr: 7.9501e-05 lr: 7.9501e-05 eta: 5:58:04 time: 0.4532 data_time: 0.0233 memory: 15926 grad_norm: 19.2007 loss: 9.4506 decode.loss_cls_ce: 1.9285 decode.loss_mask_ce: 0.9428 decode.loss_mask_dice: 1.8518 decode.d7.loss_cls_ce: 1.9432 decode.d7.loss_mask_ce: 0.9353 decode.d7.loss_mask_dice: 1.8491 2023/09/06 16:05:12 - mmengine - INFO - Iter(train) [12350/60000] base_lr: 7.9418e-05 lr: 7.9418e-05 eta: 5:57:42 time: 0.4514 data_time: 0.0233 memory: 15858 grad_norm: 16.2749 loss: 11.1828 decode.loss_cls_ce: 2.5120 decode.loss_mask_ce: 0.9705 decode.loss_mask_dice: 2.1082 decode.d7.loss_cls_ce: 2.5141 decode.d7.loss_mask_ce: 0.9733 decode.d7.loss_mask_dice: 2.1048 2023/09/06 16:05:34 - mmengine - INFO - Iter(train) [12400/60000] base_lr: 7.9335e-05 lr: 7.9335e-05 eta: 5:57:20 time: 0.4497 data_time: 0.0229 memory: 15883 grad_norm: 16.2487 loss: 10.1624 decode.loss_cls_ce: 2.1930 decode.loss_mask_ce: 0.9768 decode.loss_mask_dice: 1.9162 decode.d7.loss_cls_ce: 2.1693 decode.d7.loss_mask_ce: 0.9819 decode.d7.loss_mask_dice: 1.9252 2023/09/06 16:05:57 - mmengine - INFO - Iter(train) [12450/60000] base_lr: 7.9251e-05 lr: 7.9251e-05 eta: 5:56:57 time: 0.4490 data_time: 0.0237 memory: 15781 grad_norm: 17.3132 loss: 9.5742 decode.loss_cls_ce: 2.1907 decode.loss_mask_ce: 0.9596 decode.loss_mask_dice: 1.6594 decode.d7.loss_cls_ce: 2.1588 decode.d7.loss_mask_ce: 0.9589 decode.d7.loss_mask_dice: 1.6468 2023/09/06 16:06:19 - mmengine - INFO - Iter(train) [12500/60000] base_lr: 7.9168e-05 lr: 7.9168e-05 eta: 5:56:35 time: 0.4549 data_time: 0.0228 memory: 15821 grad_norm: 18.3039 loss: 9.6812 decode.loss_cls_ce: 2.1999 decode.loss_mask_ce: 0.9226 decode.loss_mask_dice: 1.7023 decode.d7.loss_cls_ce: 2.2246 decode.d7.loss_mask_ce: 0.9159 decode.d7.loss_mask_dice: 1.7159 2023/09/06 16:06:42 - mmengine - INFO - Iter(train) [12550/60000] base_lr: 7.9085e-05 lr: 7.9085e-05 eta: 5:56:13 time: 0.4492 data_time: 0.0236 memory: 15784 grad_norm: 15.6386 loss: 10.5861 decode.loss_cls_ce: 2.3602 decode.loss_mask_ce: 0.9762 decode.loss_mask_dice: 1.9506 decode.d7.loss_cls_ce: 2.3626 decode.d7.loss_mask_ce: 0.9719 decode.d7.loss_mask_dice: 1.9646 2023/09/06 16:07:05 - mmengine - INFO - Iter(train) [12600/60000] base_lr: 7.9001e-05 lr: 7.9001e-05 eta: 5:55:51 time: 0.4527 data_time: 0.0233 memory: 15770 grad_norm: 16.0580 loss: 9.8182 decode.loss_cls_ce: 2.1704 decode.loss_mask_ce: 0.9698 decode.loss_mask_dice: 1.7740 decode.d7.loss_cls_ce: 2.1450 decode.d7.loss_mask_ce: 0.9821 decode.d7.loss_mask_dice: 1.7768 2023/09/06 16:07:27 - mmengine - INFO - Iter(train) [12650/60000] base_lr: 7.8918e-05 lr: 7.8918e-05 eta: 5:55:28 time: 0.4505 data_time: 0.0233 memory: 16094 grad_norm: 18.6182 loss: 10.8482 decode.loss_cls_ce: 2.3782 decode.loss_mask_ce: 1.0407 decode.loss_mask_dice: 1.9963 decode.d7.loss_cls_ce: 2.3945 decode.d7.loss_mask_ce: 1.0440 decode.d7.loss_mask_dice: 1.9946 2023/09/06 16:07:50 - mmengine - INFO - Iter(train) [12700/60000] base_lr: 7.8835e-05 lr: 7.8835e-05 eta: 5:55:06 time: 0.4525 data_time: 0.0232 memory: 16014 grad_norm: 17.5329 loss: 10.1215 decode.loss_cls_ce: 2.1967 decode.loss_mask_ce: 0.9989 decode.loss_mask_dice: 1.8853 decode.d7.loss_cls_ce: 2.1597 decode.d7.loss_mask_ce: 1.0022 decode.d7.loss_mask_dice: 1.8787 2023/09/06 16:08:12 - mmengine - INFO - Iter(train) [12750/60000] base_lr: 7.8751e-05 lr: 7.8751e-05 eta: 5:54:44 time: 0.4521 data_time: 0.0226 memory: 15964 grad_norm: 19.1453 loss: 9.8370 decode.loss_cls_ce: 2.1859 decode.loss_mask_ce: 0.9555 decode.loss_mask_dice: 1.7760 decode.d7.loss_cls_ce: 2.1741 decode.d7.loss_mask_ce: 0.9609 decode.d7.loss_mask_dice: 1.7847 2023/09/06 16:08:35 - mmengine - INFO - Iter(train) [12800/60000] base_lr: 7.8668e-05 lr: 7.8668e-05 eta: 5:54:22 time: 0.4499 data_time: 0.0231 memory: 16065 grad_norm: 17.3055 loss: 10.3914 decode.loss_cls_ce: 2.1940 decode.loss_mask_ce: 0.9706 decode.loss_mask_dice: 2.0324 decode.d7.loss_cls_ce: 2.1689 decode.d7.loss_mask_ce: 0.9855 decode.d7.loss_mask_dice: 2.0401 2023/09/06 16:08:58 - mmengine - INFO - Iter(train) [12850/60000] base_lr: 7.8585e-05 lr: 7.8585e-05 eta: 5:53:59 time: 0.4489 data_time: 0.0240 memory: 15733 grad_norm: 17.2468 loss: 9.9678 decode.loss_cls_ce: 2.2563 decode.loss_mask_ce: 0.9708 decode.loss_mask_dice: 1.7514 decode.d7.loss_cls_ce: 2.2614 decode.d7.loss_mask_ce: 0.9779 decode.d7.loss_mask_dice: 1.7500 2023/09/06 16:09:20 - mmengine - INFO - Iter(train) [12900/60000] base_lr: 7.8501e-05 lr: 7.8501e-05 eta: 5:53:37 time: 0.4503 data_time: 0.0234 memory: 15937 grad_norm: 16.8371 loss: 9.5628 decode.loss_cls_ce: 2.1186 decode.loss_mask_ce: 0.9306 decode.loss_mask_dice: 1.7355 decode.d7.loss_cls_ce: 2.1108 decode.d7.loss_mask_ce: 0.9262 decode.d7.loss_mask_dice: 1.7411 2023/09/06 16:09:43 - mmengine - INFO - Iter(train) [12950/60000] base_lr: 7.8418e-05 lr: 7.8418e-05 eta: 5:53:14 time: 0.4527 data_time: 0.0234 memory: 15869 grad_norm: 17.2507 loss: 10.3030 decode.loss_cls_ce: 2.2724 decode.loss_mask_ce: 0.9305 decode.loss_mask_dice: 1.9458 decode.d7.loss_cls_ce: 2.2863 decode.d7.loss_mask_ce: 0.9138 decode.d7.loss_mask_dice: 1.9542 2023/09/06 16:10:05 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:10:05 - mmengine - INFO - Iter(train) [13000/60000] base_lr: 7.8335e-05 lr: 7.8335e-05 eta: 5:52:52 time: 0.4540 data_time: 0.0230 memory: 15771 grad_norm: 17.2654 loss: 10.6194 decode.loss_cls_ce: 2.3685 decode.loss_mask_ce: 0.9648 decode.loss_mask_dice: 1.9596 decode.d7.loss_cls_ce: 2.4083 decode.d7.loss_mask_ce: 0.9611 decode.d7.loss_mask_dice: 1.9571 2023/09/06 16:10:28 - mmengine - INFO - Iter(train) [13050/60000] base_lr: 7.8251e-05 lr: 7.8251e-05 eta: 5:52:30 time: 0.4502 data_time: 0.0227 memory: 15967 grad_norm: 17.2577 loss: 9.4542 decode.loss_cls_ce: 2.0844 decode.loss_mask_ce: 0.8872 decode.loss_mask_dice: 1.7623 decode.d7.loss_cls_ce: 2.0776 decode.d7.loss_mask_ce: 0.8812 decode.d7.loss_mask_dice: 1.7614 2023/09/06 16:10:50 - mmengine - INFO - Iter(train) [13100/60000] base_lr: 7.8168e-05 lr: 7.8168e-05 eta: 5:52:07 time: 0.4514 data_time: 0.0237 memory: 15759 grad_norm: 18.3523 loss: 9.8208 decode.loss_cls_ce: 2.2036 decode.loss_mask_ce: 0.9294 decode.loss_mask_dice: 1.7682 decode.d7.loss_cls_ce: 2.2114 decode.d7.loss_mask_ce: 0.9389 decode.d7.loss_mask_dice: 1.7691 2023/09/06 16:11:13 - mmengine - INFO - Iter(train) [13150/60000] base_lr: 7.8085e-05 lr: 7.8085e-05 eta: 5:51:45 time: 0.4515 data_time: 0.0237 memory: 15884 grad_norm: 16.9889 loss: 10.1794 decode.loss_cls_ce: 2.2912 decode.loss_mask_ce: 0.9678 decode.loss_mask_dice: 1.8154 decode.d7.loss_cls_ce: 2.3103 decode.d7.loss_mask_ce: 0.9695 decode.d7.loss_mask_dice: 1.8252 2023/09/06 16:11:36 - mmengine - INFO - Iter(train) [13200/60000] base_lr: 7.8001e-05 lr: 7.8001e-05 eta: 5:51:23 time: 0.4548 data_time: 0.0242 memory: 15926 grad_norm: 15.9027 loss: 9.3682 decode.loss_cls_ce: 2.1343 decode.loss_mask_ce: 0.8640 decode.loss_mask_dice: 1.6817 decode.d7.loss_cls_ce: 2.1418 decode.d7.loss_mask_ce: 0.8627 decode.d7.loss_mask_dice: 1.6837 2023/09/06 16:11:58 - mmengine - INFO - Iter(train) [13250/60000] base_lr: 7.7918e-05 lr: 7.7918e-05 eta: 5:51:01 time: 0.4518 data_time: 0.0233 memory: 15859 grad_norm: 15.6218 loss: 10.0953 decode.loss_cls_ce: 2.2720 decode.loss_mask_ce: 0.9137 decode.loss_mask_dice: 1.8580 decode.d7.loss_cls_ce: 2.2751 decode.d7.loss_mask_ce: 0.9135 decode.d7.loss_mask_dice: 1.8630 2023/09/06 16:12:21 - mmengine - INFO - Iter(train) [13300/60000] base_lr: 7.7835e-05 lr: 7.7835e-05 eta: 5:50:39 time: 0.4504 data_time: 0.0235 memory: 15822 grad_norm: 17.1530 loss: 10.4176 decode.loss_cls_ce: 2.2817 decode.loss_mask_ce: 0.9960 decode.loss_mask_dice: 1.9402 decode.d7.loss_cls_ce: 2.2638 decode.d7.loss_mask_ce: 0.9995 decode.d7.loss_mask_dice: 1.9364 2023/09/06 16:12:44 - mmengine - INFO - Iter(train) [13350/60000] base_lr: 7.7751e-05 lr: 7.7751e-05 eta: 5:50:16 time: 0.4503 data_time: 0.0233 memory: 15990 grad_norm: 17.9369 loss: 10.9841 decode.loss_cls_ce: 2.4567 decode.loss_mask_ce: 1.0079 decode.loss_mask_dice: 2.0265 decode.d7.loss_cls_ce: 2.4442 decode.d7.loss_mask_ce: 1.0113 decode.d7.loss_mask_dice: 2.0376 2023/09/06 16:13:06 - mmengine - INFO - Iter(train) [13400/60000] base_lr: 7.7668e-05 lr: 7.7668e-05 eta: 5:49:54 time: 0.4512 data_time: 0.0234 memory: 15937 grad_norm: 15.7818 loss: 10.0381 decode.loss_cls_ce: 2.2684 decode.loss_mask_ce: 0.9106 decode.loss_mask_dice: 1.8297 decode.d7.loss_cls_ce: 2.3020 decode.d7.loss_mask_ce: 0.9041 decode.d7.loss_mask_dice: 1.8234 2023/09/06 16:13:29 - mmengine - INFO - Iter(train) [13450/60000] base_lr: 7.7585e-05 lr: 7.7585e-05 eta: 5:49:31 time: 0.4506 data_time: 0.0235 memory: 15902 grad_norm: 15.8248 loss: 10.6397 decode.loss_cls_ce: 2.3147 decode.loss_mask_ce: 1.0282 decode.loss_mask_dice: 1.9837 decode.d7.loss_cls_ce: 2.3253 decode.d7.loss_mask_ce: 1.0157 decode.d7.loss_mask_dice: 1.9722 2023/09/06 16:13:51 - mmengine - INFO - Iter(train) [13500/60000] base_lr: 7.7501e-05 lr: 7.7501e-05 eta: 5:49:09 time: 0.4511 data_time: 0.0238 memory: 15821 grad_norm: 18.2353 loss: 9.9892 decode.loss_cls_ce: 2.2634 decode.loss_mask_ce: 0.9464 decode.loss_mask_dice: 1.7849 decode.d7.loss_cls_ce: 2.2727 decode.d7.loss_mask_ce: 0.9515 decode.d7.loss_mask_dice: 1.7703 2023/09/06 16:14:14 - mmengine - INFO - Iter(train) [13550/60000] base_lr: 7.7418e-05 lr: 7.7418e-05 eta: 5:48:47 time: 0.4525 data_time: 0.0232 memory: 16002 grad_norm: 16.3001 loss: 10.4497 decode.loss_cls_ce: 2.2943 decode.loss_mask_ce: 0.9870 decode.loss_mask_dice: 1.9240 decode.d7.loss_cls_ce: 2.3275 decode.d7.loss_mask_ce: 0.9937 decode.d7.loss_mask_dice: 1.9232 2023/09/06 16:14:36 - mmengine - INFO - Iter(train) [13600/60000] base_lr: 7.7335e-05 lr: 7.7335e-05 eta: 5:48:25 time: 0.4493 data_time: 0.0234 memory: 15899 grad_norm: 17.8390 loss: 9.9693 decode.loss_cls_ce: 2.2724 decode.loss_mask_ce: 0.8727 decode.loss_mask_dice: 1.8666 decode.d7.loss_cls_ce: 2.2107 decode.d7.loss_mask_ce: 0.8775 decode.d7.loss_mask_dice: 1.8694 2023/09/06 16:14:59 - mmengine - INFO - Iter(train) [13650/60000] base_lr: 7.7251e-05 lr: 7.7251e-05 eta: 5:48:02 time: 0.4545 data_time: 0.0226 memory: 15822 grad_norm: 17.0478 loss: 11.0155 decode.loss_cls_ce: 2.4914 decode.loss_mask_ce: 0.9192 decode.loss_mask_dice: 2.0969 decode.d7.loss_cls_ce: 2.4925 decode.d7.loss_mask_ce: 0.9288 decode.d7.loss_mask_dice: 2.0867 2023/09/06 16:15:22 - mmengine - INFO - Iter(train) [13700/60000] base_lr: 7.7168e-05 lr: 7.7168e-05 eta: 5:47:40 time: 0.4569 data_time: 0.0229 memory: 15835 grad_norm: 17.9707 loss: 10.3856 decode.loss_cls_ce: 2.4212 decode.loss_mask_ce: 0.9455 decode.loss_mask_dice: 1.8330 decode.d7.loss_cls_ce: 2.3980 decode.d7.loss_mask_ce: 0.9524 decode.d7.loss_mask_dice: 1.8356 2023/09/06 16:15:44 - mmengine - INFO - Iter(train) [13750/60000] base_lr: 7.7085e-05 lr: 7.7085e-05 eta: 5:47:18 time: 0.4503 data_time: 0.0239 memory: 15757 grad_norm: 15.7097 loss: 9.7565 decode.loss_cls_ce: 2.1226 decode.loss_mask_ce: 0.9947 decode.loss_mask_dice: 1.7575 decode.d7.loss_cls_ce: 2.1553 decode.d7.loss_mask_ce: 0.9786 decode.d7.loss_mask_dice: 1.7479 2023/09/06 16:16:07 - mmengine - INFO - Iter(train) [13800/60000] base_lr: 7.7001e-05 lr: 7.7001e-05 eta: 5:46:55 time: 0.4482 data_time: 0.0232 memory: 15794 grad_norm: 17.7540 loss: 10.1570 decode.loss_cls_ce: 2.1314 decode.loss_mask_ce: 1.0334 decode.loss_mask_dice: 1.9233 decode.d7.loss_cls_ce: 2.1239 decode.d7.loss_mask_ce: 1.0306 decode.d7.loss_mask_dice: 1.9143 2023/09/06 16:16:29 - mmengine - INFO - Iter(train) [13850/60000] base_lr: 7.6918e-05 lr: 7.6918e-05 eta: 5:46:32 time: 0.4498 data_time: 0.0233 memory: 15845 grad_norm: 17.0699 loss: 10.0541 decode.loss_cls_ce: 2.2858 decode.loss_mask_ce: 0.9457 decode.loss_mask_dice: 1.7919 decode.d7.loss_cls_ce: 2.2926 decode.d7.loss_mask_ce: 0.9417 decode.d7.loss_mask_dice: 1.7964 2023/09/06 16:16:52 - mmengine - INFO - Iter(train) [13900/60000] base_lr: 7.6835e-05 lr: 7.6835e-05 eta: 5:46:10 time: 0.4527 data_time: 0.0230 memory: 15951 grad_norm: 16.2045 loss: 10.7251 decode.loss_cls_ce: 2.3883 decode.loss_mask_ce: 0.8985 decode.loss_mask_dice: 2.0675 decode.d7.loss_cls_ce: 2.3876 decode.d7.loss_mask_ce: 0.8972 decode.d7.loss_mask_dice: 2.0861 2023/09/06 16:17:14 - mmengine - INFO - Iter(train) [13950/60000] base_lr: 7.6751e-05 lr: 7.6751e-05 eta: 5:45:47 time: 0.4499 data_time: 0.0229 memory: 15771 grad_norm: 16.4991 loss: 10.5419 decode.loss_cls_ce: 2.3917 decode.loss_mask_ce: 0.9310 decode.loss_mask_dice: 1.9642 decode.d7.loss_cls_ce: 2.3795 decode.d7.loss_mask_ce: 0.9312 decode.d7.loss_mask_dice: 1.9444 2023/09/06 16:17:37 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:17:37 - mmengine - INFO - Iter(train) [14000/60000] base_lr: 7.6668e-05 lr: 7.6668e-05 eta: 5:45:25 time: 0.4497 data_time: 0.0235 memory: 15870 grad_norm: 16.2477 loss: 10.4240 decode.loss_cls_ce: 2.1821 decode.loss_mask_ce: 1.0173 decode.loss_mask_dice: 1.9844 decode.d7.loss_cls_ce: 2.2130 decode.d7.loss_mask_ce: 1.0237 decode.d7.loss_mask_dice: 2.0035 2023/09/06 16:17:59 - mmengine - INFO - Iter(train) [14050/60000] base_lr: 7.6585e-05 lr: 7.6585e-05 eta: 5:45:02 time: 0.4515 data_time: 0.0229 memory: 15818 grad_norm: 17.5225 loss: 10.5000 decode.loss_cls_ce: 2.3384 decode.loss_mask_ce: 0.9828 decode.loss_mask_dice: 1.9167 decode.d7.loss_cls_ce: 2.3354 decode.d7.loss_mask_ce: 0.9781 decode.d7.loss_mask_dice: 1.9487 2023/09/06 16:18:22 - mmengine - INFO - Iter(train) [14100/60000] base_lr: 7.6501e-05 lr: 7.6501e-05 eta: 5:44:40 time: 0.4501 data_time: 0.0229 memory: 15758 grad_norm: 16.6730 loss: 10.8459 decode.loss_cls_ce: 2.3669 decode.loss_mask_ce: 0.9984 decode.loss_mask_dice: 2.0719 decode.d7.loss_cls_ce: 2.3364 decode.d7.loss_mask_ce: 0.9940 decode.d7.loss_mask_dice: 2.0783 2023/09/06 16:18:45 - mmengine - INFO - Iter(train) [14150/60000] base_lr: 7.6418e-05 lr: 7.6418e-05 eta: 5:44:17 time: 0.4527 data_time: 0.0235 memory: 15807 grad_norm: 16.8897 loss: 10.7966 decode.loss_cls_ce: 2.3184 decode.loss_mask_ce: 1.0001 decode.loss_mask_dice: 2.0838 decode.d7.loss_cls_ce: 2.3023 decode.d7.loss_mask_ce: 1.0033 decode.d7.loss_mask_dice: 2.0888 2023/09/06 16:19:07 - mmengine - INFO - Iter(train) [14200/60000] base_lr: 7.6335e-05 lr: 7.6335e-05 eta: 5:43:55 time: 0.4491 data_time: 0.0234 memory: 15822 grad_norm: 16.2741 loss: 10.3187 decode.loss_cls_ce: 2.2213 decode.loss_mask_ce: 0.9534 decode.loss_mask_dice: 1.9969 decode.d7.loss_cls_ce: 2.2053 decode.d7.loss_mask_ce: 0.9481 decode.d7.loss_mask_dice: 1.9937 2023/09/06 16:19:30 - mmengine - INFO - Iter(train) [14250/60000] base_lr: 7.6251e-05 lr: 7.6251e-05 eta: 5:43:32 time: 0.4505 data_time: 0.0228 memory: 15873 grad_norm: 16.2463 loss: 9.4880 decode.loss_cls_ce: 2.0232 decode.loss_mask_ce: 0.9163 decode.loss_mask_dice: 1.7935 decode.d7.loss_cls_ce: 2.0332 decode.d7.loss_mask_ce: 0.9232 decode.d7.loss_mask_dice: 1.7986 2023/09/06 16:19:52 - mmengine - INFO - Iter(train) [14300/60000] base_lr: 7.6168e-05 lr: 7.6168e-05 eta: 5:43:10 time: 0.4487 data_time: 0.0222 memory: 15846 grad_norm: 16.7588 loss: 8.6315 decode.loss_cls_ce: 1.9666 decode.loss_mask_ce: 0.8129 decode.loss_mask_dice: 1.5512 decode.d7.loss_cls_ce: 1.9520 decode.d7.loss_mask_ce: 0.8057 decode.d7.loss_mask_dice: 1.5432 2023/09/06 16:20:15 - mmengine - INFO - Iter(train) [14350/60000] base_lr: 7.6085e-05 lr: 7.6085e-05 eta: 5:42:48 time: 0.4487 data_time: 0.0237 memory: 16064 grad_norm: 16.8331 loss: 10.7042 decode.loss_cls_ce: 2.2522 decode.loss_mask_ce: 0.9910 decode.loss_mask_dice: 2.1110 decode.d7.loss_cls_ce: 2.2676 decode.d7.loss_mask_ce: 0.9890 decode.d7.loss_mask_dice: 2.0934 2023/09/06 16:20:37 - mmengine - INFO - Iter(train) [14400/60000] base_lr: 7.6001e-05 lr: 7.6001e-05 eta: 5:42:25 time: 0.4524 data_time: 0.0224 memory: 15897 grad_norm: 17.9764 loss: 10.7731 decode.loss_cls_ce: 2.2892 decode.loss_mask_ce: 1.0181 decode.loss_mask_dice: 2.0679 decode.d7.loss_cls_ce: 2.3160 decode.d7.loss_mask_ce: 1.0187 decode.d7.loss_mask_dice: 2.0631 2023/09/06 16:21:00 - mmengine - INFO - Iter(train) [14450/60000] base_lr: 7.5918e-05 lr: 7.5918e-05 eta: 5:42:03 time: 0.4508 data_time: 0.0233 memory: 15798 grad_norm: 18.5450 loss: 9.8238 decode.loss_cls_ce: 2.1754 decode.loss_mask_ce: 0.9136 decode.loss_mask_dice: 1.8237 decode.d7.loss_cls_ce: 2.1847 decode.d7.loss_mask_ce: 0.9177 decode.d7.loss_mask_dice: 1.8087 2023/09/06 16:21:22 - mmengine - INFO - Iter(train) [14500/60000] base_lr: 7.5835e-05 lr: 7.5835e-05 eta: 5:41:40 time: 0.4499 data_time: 0.0234 memory: 15939 grad_norm: 18.1782 loss: 9.6692 decode.loss_cls_ce: 2.2409 decode.loss_mask_ce: 0.8632 decode.loss_mask_dice: 1.7416 decode.d7.loss_cls_ce: 2.2236 decode.d7.loss_mask_ce: 0.8678 decode.d7.loss_mask_dice: 1.7321 2023/09/06 16:21:45 - mmengine - INFO - Iter(train) [14550/60000] base_lr: 7.5751e-05 lr: 7.5751e-05 eta: 5:41:18 time: 0.4560 data_time: 0.0253 memory: 15911 grad_norm: 16.6384 loss: 9.1466 decode.loss_cls_ce: 2.0927 decode.loss_mask_ce: 0.8293 decode.loss_mask_dice: 1.6551 decode.d7.loss_cls_ce: 2.1154 decode.d7.loss_mask_ce: 0.8276 decode.d7.loss_mask_dice: 1.6266 2023/09/06 16:22:08 - mmengine - INFO - Iter(train) [14600/60000] base_lr: 7.5668e-05 lr: 7.5668e-05 eta: 5:40:55 time: 0.4491 data_time: 0.0227 memory: 15869 grad_norm: 18.7088 loss: 10.7330 decode.loss_cls_ce: 2.3028 decode.loss_mask_ce: 1.0135 decode.loss_mask_dice: 2.0571 decode.d7.loss_cls_ce: 2.2879 decode.d7.loss_mask_ce: 1.0247 decode.d7.loss_mask_dice: 2.0472 2023/09/06 16:22:30 - mmengine - INFO - Iter(train) [14650/60000] base_lr: 7.5585e-05 lr: 7.5585e-05 eta: 5:40:33 time: 0.4491 data_time: 0.0232 memory: 15819 grad_norm: 17.5402 loss: 10.5839 decode.loss_cls_ce: 2.2609 decode.loss_mask_ce: 1.0029 decode.loss_mask_dice: 2.0234 decode.d7.loss_cls_ce: 2.2643 decode.d7.loss_mask_ce: 1.0009 decode.d7.loss_mask_dice: 2.0314 2023/09/06 16:22:53 - mmengine - INFO - Iter(train) [14700/60000] base_lr: 7.5501e-05 lr: 7.5501e-05 eta: 5:40:10 time: 0.4528 data_time: 0.0233 memory: 15903 grad_norm: 16.9276 loss: 9.1153 decode.loss_cls_ce: 2.0445 decode.loss_mask_ce: 0.8365 decode.loss_mask_dice: 1.6794 decode.d7.loss_cls_ce: 2.0302 decode.d7.loss_mask_ce: 0.8425 decode.d7.loss_mask_dice: 1.6823 2023/09/06 16:23:15 - mmengine - INFO - Iter(train) [14750/60000] base_lr: 7.5418e-05 lr: 7.5418e-05 eta: 5:39:48 time: 0.4508 data_time: 0.0234 memory: 15935 grad_norm: 15.4278 loss: 10.1484 decode.loss_cls_ce: 2.1692 decode.loss_mask_ce: 0.9566 decode.loss_mask_dice: 1.9565 decode.d7.loss_cls_ce: 2.1482 decode.d7.loss_mask_ce: 0.9550 decode.d7.loss_mask_dice: 1.9629 2023/09/06 16:23:31 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:23:38 - mmengine - INFO - Iter(train) [14800/60000] base_lr: 7.5335e-05 lr: 7.5335e-05 eta: 5:39:25 time: 0.4521 data_time: 0.0222 memory: 16026 grad_norm: 16.4917 loss: 10.5619 decode.loss_cls_ce: 2.2784 decode.loss_mask_ce: 0.9765 decode.loss_mask_dice: 2.0319 decode.d7.loss_cls_ce: 2.2413 decode.d7.loss_mask_ce: 0.9845 decode.d7.loss_mask_dice: 2.0494 2023/09/06 16:24:00 - mmengine - INFO - Iter(train) [14850/60000] base_lr: 7.5251e-05 lr: 7.5251e-05 eta: 5:39:03 time: 0.4514 data_time: 0.0238 memory: 15786 grad_norm: 17.1009 loss: 10.4137 decode.loss_cls_ce: 2.3867 decode.loss_mask_ce: 0.8947 decode.loss_mask_dice: 1.9362 decode.d7.loss_cls_ce: 2.3501 decode.d7.loss_mask_ce: 0.9015 decode.d7.loss_mask_dice: 1.9445 2023/09/06 16:24:23 - mmengine - INFO - Iter(train) [14900/60000] base_lr: 7.5168e-05 lr: 7.5168e-05 eta: 5:38:40 time: 0.4508 data_time: 0.0233 memory: 15845 grad_norm: 15.9533 loss: 10.7521 decode.loss_cls_ce: 2.3302 decode.loss_mask_ce: 0.9968 decode.loss_mask_dice: 2.0382 decode.d7.loss_cls_ce: 2.3602 decode.d7.loss_mask_ce: 0.9870 decode.d7.loss_mask_dice: 2.0398 2023/09/06 16:24:45 - mmengine - INFO - Iter(train) [14950/60000] base_lr: 7.5085e-05 lr: 7.5085e-05 eta: 5:38:18 time: 0.4530 data_time: 0.0231 memory: 15782 grad_norm: 18.4140 loss: 9.3537 decode.loss_cls_ce: 2.0624 decode.loss_mask_ce: 0.9142 decode.loss_mask_dice: 1.6888 decode.d7.loss_cls_ce: 2.0717 decode.d7.loss_mask_ce: 0.9166 decode.d7.loss_mask_dice: 1.7000 2023/09/06 16:25:08 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:25:08 - mmengine - INFO - Iter(train) [15000/60000] base_lr: 7.5001e-05 lr: 7.5001e-05 eta: 5:37:55 time: 0.4532 data_time: 0.0230 memory: 15885 grad_norm: 18.9974 loss: 9.6591 decode.loss_cls_ce: 2.1311 decode.loss_mask_ce: 0.8517 decode.loss_mask_dice: 1.8268 decode.d7.loss_cls_ce: 2.1648 decode.d7.loss_mask_ce: 0.8504 decode.d7.loss_mask_dice: 1.8342 2023/09/06 16:25:31 - mmengine - INFO - Iter(train) [15050/60000] base_lr: 7.4918e-05 lr: 7.4918e-05 eta: 5:37:33 time: 0.4512 data_time: 0.0240 memory: 15896 grad_norm: 17.7015 loss: 9.2487 decode.loss_cls_ce: 2.0032 decode.loss_mask_ce: 0.9458 decode.loss_mask_dice: 1.6730 decode.d7.loss_cls_ce: 2.0081 decode.d7.loss_mask_ce: 0.9478 decode.d7.loss_mask_dice: 1.6708 2023/09/06 16:25:53 - mmengine - INFO - Iter(train) [15100/60000] base_lr: 7.4835e-05 lr: 7.4835e-05 eta: 5:37:11 time: 0.4529 data_time: 0.0232 memory: 15785 grad_norm: 17.1871 loss: 10.0806 decode.loss_cls_ce: 2.2054 decode.loss_mask_ce: 0.9707 decode.loss_mask_dice: 1.8782 decode.d7.loss_cls_ce: 2.1790 decode.d7.loss_mask_ce: 0.9660 decode.d7.loss_mask_dice: 1.8813 2023/09/06 16:26:16 - mmengine - INFO - Iter(train) [15150/60000] base_lr: 7.4751e-05 lr: 7.4751e-05 eta: 5:36:49 time: 0.4485 data_time: 0.0234 memory: 15980 grad_norm: 16.1058 loss: 8.4964 decode.loss_cls_ce: 1.9279 decode.loss_mask_ce: 0.8346 decode.loss_mask_dice: 1.4841 decode.d7.loss_cls_ce: 1.9120 decode.d7.loss_mask_ce: 0.8405 decode.d7.loss_mask_dice: 1.4974 2023/09/06 16:26:38 - mmengine - INFO - Iter(train) [15200/60000] base_lr: 7.4668e-05 lr: 7.4668e-05 eta: 5:36:26 time: 0.4509 data_time: 0.0234 memory: 15849 grad_norm: 16.6337 loss: 10.2619 decode.loss_cls_ce: 2.2165 decode.loss_mask_ce: 0.8807 decode.loss_mask_dice: 2.0252 decode.d7.loss_cls_ce: 2.2188 decode.d7.loss_mask_ce: 0.8960 decode.d7.loss_mask_dice: 2.0248 2023/09/06 16:27:01 - mmengine - INFO - Iter(train) [15250/60000] base_lr: 7.4585e-05 lr: 7.4585e-05 eta: 5:36:04 time: 0.4505 data_time: 0.0232 memory: 15884 grad_norm: 16.3839 loss: 10.1903 decode.loss_cls_ce: 2.2845 decode.loss_mask_ce: 0.9338 decode.loss_mask_dice: 1.8749 decode.d7.loss_cls_ce: 2.3009 decode.d7.loss_mask_ce: 0.9247 decode.d7.loss_mask_dice: 1.8714 2023/09/06 16:27:24 - mmengine - INFO - Iter(train) [15300/60000] base_lr: 7.4501e-05 lr: 7.4501e-05 eta: 5:35:41 time: 0.4524 data_time: 0.0230 memory: 15780 grad_norm: 16.5081 loss: 10.7474 decode.loss_cls_ce: 2.4350 decode.loss_mask_ce: 0.9476 decode.loss_mask_dice: 1.9954 decode.d7.loss_cls_ce: 2.4176 decode.d7.loss_mask_ce: 0.9612 decode.d7.loss_mask_dice: 1.9906 2023/09/06 16:27:46 - mmengine - INFO - Iter(train) [15350/60000] base_lr: 7.4418e-05 lr: 7.4418e-05 eta: 5:35:19 time: 0.4510 data_time: 0.0235 memory: 15811 grad_norm: 15.6745 loss: 9.8270 decode.loss_cls_ce: 2.1957 decode.loss_mask_ce: 0.8946 decode.loss_mask_dice: 1.8163 decode.d7.loss_cls_ce: 2.2375 decode.d7.loss_mask_ce: 0.8787 decode.d7.loss_mask_dice: 1.8041 2023/09/06 16:28:09 - mmengine - INFO - Iter(train) [15400/60000] base_lr: 7.4335e-05 lr: 7.4335e-05 eta: 5:34:57 time: 0.4513 data_time: 0.0230 memory: 15756 grad_norm: 18.2139 loss: 9.5956 decode.loss_cls_ce: 2.2158 decode.loss_mask_ce: 0.8559 decode.loss_mask_dice: 1.7192 decode.d7.loss_cls_ce: 2.2243 decode.d7.loss_mask_ce: 0.8618 decode.d7.loss_mask_dice: 1.7187 2023/09/06 16:28:31 - mmengine - INFO - Iter(train) [15450/60000] base_lr: 7.4251e-05 lr: 7.4251e-05 eta: 5:34:34 time: 0.4511 data_time: 0.0229 memory: 16121 grad_norm: 14.6232 loss: 10.0576 decode.loss_cls_ce: 2.2432 decode.loss_mask_ce: 0.9534 decode.loss_mask_dice: 1.8081 decode.d7.loss_cls_ce: 2.2859 decode.d7.loss_mask_ce: 0.9557 decode.d7.loss_mask_dice: 1.8113 2023/09/06 16:28:54 - mmengine - INFO - Iter(train) [15500/60000] base_lr: 7.4168e-05 lr: 7.4168e-05 eta: 5:34:11 time: 0.4500 data_time: 0.0240 memory: 15792 grad_norm: 17.8300 loss: 10.5063 decode.loss_cls_ce: 2.2572 decode.loss_mask_ce: 1.0491 decode.loss_mask_dice: 1.9446 decode.d7.loss_cls_ce: 2.2658 decode.d7.loss_mask_ce: 1.0409 decode.d7.loss_mask_dice: 1.9489 2023/09/06 16:29:16 - mmengine - INFO - Iter(train) [15550/60000] base_lr: 7.4085e-05 lr: 7.4085e-05 eta: 5:33:49 time: 0.4511 data_time: 0.0235 memory: 15781 grad_norm: 16.9186 loss: 9.5920 decode.loss_cls_ce: 2.0363 decode.loss_mask_ce: 0.9324 decode.loss_mask_dice: 1.7976 decode.d7.loss_cls_ce: 2.0750 decode.d7.loss_mask_ce: 0.9324 decode.d7.loss_mask_dice: 1.8182 2023/09/06 16:29:39 - mmengine - INFO - Iter(train) [15600/60000] base_lr: 7.4001e-05 lr: 7.4001e-05 eta: 5:33:26 time: 0.4520 data_time: 0.0237 memory: 15846 grad_norm: 13.9074 loss: 10.7142 decode.loss_cls_ce: 2.4362 decode.loss_mask_ce: 0.9371 decode.loss_mask_dice: 1.9695 decode.d7.loss_cls_ce: 2.4804 decode.d7.loss_mask_ce: 0.9356 decode.d7.loss_mask_dice: 1.9554 2023/09/06 16:30:01 - mmengine - INFO - Iter(train) [15650/60000] base_lr: 7.3918e-05 lr: 7.3918e-05 eta: 5:33:04 time: 0.4524 data_time: 0.0235 memory: 16004 grad_norm: 17.2660 loss: 10.9306 decode.loss_cls_ce: 2.4510 decode.loss_mask_ce: 0.9635 decode.loss_mask_dice: 2.0508 decode.d7.loss_cls_ce: 2.4602 decode.d7.loss_mask_ce: 0.9643 decode.d7.loss_mask_dice: 2.0407 2023/09/06 16:30:24 - mmengine - INFO - Iter(train) [15700/60000] base_lr: 7.3835e-05 lr: 7.3835e-05 eta: 5:32:42 time: 0.4501 data_time: 0.0227 memory: 15794 grad_norm: 15.5816 loss: 9.5005 decode.loss_cls_ce: 1.9966 decode.loss_mask_ce: 0.9439 decode.loss_mask_dice: 1.8157 decode.d7.loss_cls_ce: 1.9791 decode.d7.loss_mask_ce: 0.9427 decode.d7.loss_mask_dice: 1.8225 2023/09/06 16:30:47 - mmengine - INFO - Iter(train) [15750/60000] base_lr: 7.3751e-05 lr: 7.3751e-05 eta: 5:32:19 time: 0.4505 data_time: 0.0228 memory: 15824 grad_norm: 17.7606 loss: 10.1242 decode.loss_cls_ce: 2.2635 decode.loss_mask_ce: 0.9513 decode.loss_mask_dice: 1.8369 decode.d7.loss_cls_ce: 2.2571 decode.d7.loss_mask_ce: 0.9539 decode.d7.loss_mask_dice: 1.8615 2023/09/06 16:31:09 - mmengine - INFO - Iter(train) [15800/60000] base_lr: 7.3668e-05 lr: 7.3668e-05 eta: 5:31:57 time: 0.4503 data_time: 0.0240 memory: 15808 grad_norm: 17.0603 loss: 9.2901 decode.loss_cls_ce: 2.0061 decode.loss_mask_ce: 0.9425 decode.loss_mask_dice: 1.6840 decode.d7.loss_cls_ce: 2.0180 decode.d7.loss_mask_ce: 0.9425 decode.d7.loss_mask_dice: 1.6971 2023/09/06 16:31:32 - mmengine - INFO - Iter(train) [15850/60000] base_lr: 7.3585e-05 lr: 7.3585e-05 eta: 5:31:34 time: 0.4496 data_time: 0.0240 memory: 15837 grad_norm: 16.3030 loss: 9.6911 decode.loss_cls_ce: 2.1357 decode.loss_mask_ce: 0.9575 decode.loss_mask_dice: 1.7570 decode.d7.loss_cls_ce: 2.1250 decode.d7.loss_mask_ce: 0.9597 decode.d7.loss_mask_dice: 1.7564 2023/09/06 16:31:54 - mmengine - INFO - Iter(train) [15900/60000] base_lr: 7.3501e-05 lr: 7.3501e-05 eta: 5:31:12 time: 0.4524 data_time: 0.0236 memory: 15773 grad_norm: 17.9596 loss: 10.4150 decode.loss_cls_ce: 2.3598 decode.loss_mask_ce: 0.9359 decode.loss_mask_dice: 1.9108 decode.d7.loss_cls_ce: 2.3441 decode.d7.loss_mask_ce: 0.9390 decode.d7.loss_mask_dice: 1.9254 2023/09/06 16:32:17 - mmengine - INFO - Iter(train) [15950/60000] base_lr: 7.3418e-05 lr: 7.3418e-05 eta: 5:30:49 time: 0.4511 data_time: 0.0239 memory: 15823 grad_norm: 15.3774 loss: 9.6202 decode.loss_cls_ce: 2.0272 decode.loss_mask_ce: 0.9670 decode.loss_mask_dice: 1.8039 decode.d7.loss_cls_ce: 2.0683 decode.d7.loss_mask_ce: 0.9649 decode.d7.loss_mask_dice: 1.7888 2023/09/06 16:32:40 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:32:40 - mmengine - INFO - Iter(train) [16000/60000] base_lr: 7.3335e-05 lr: 7.3335e-05 eta: 5:30:27 time: 0.4511 data_time: 0.0230 memory: 15965 grad_norm: 16.7750 loss: 10.5580 decode.loss_cls_ce: 2.3291 decode.loss_mask_ce: 1.0251 decode.loss_mask_dice: 1.9127 decode.d7.loss_cls_ce: 2.3400 decode.d7.loss_mask_ce: 1.0181 decode.d7.loss_mask_dice: 1.9330 2023/09/06 16:33:02 - mmengine - INFO - Iter(train) [16050/60000] base_lr: 7.3251e-05 lr: 7.3251e-05 eta: 5:30:05 time: 0.4523 data_time: 0.0228 memory: 15921 grad_norm: 16.2246 loss: 9.1278 decode.loss_cls_ce: 2.0271 decode.loss_mask_ce: 0.8641 decode.loss_mask_dice: 1.6635 decode.d7.loss_cls_ce: 2.0258 decode.d7.loss_mask_ce: 0.8709 decode.d7.loss_mask_dice: 1.6764 2023/09/06 16:33:25 - mmengine - INFO - Iter(train) [16100/60000] base_lr: 7.3168e-05 lr: 7.3168e-05 eta: 5:29:43 time: 0.4511 data_time: 0.0233 memory: 15978 grad_norm: 16.4183 loss: 9.7398 decode.loss_cls_ce: 2.1479 decode.loss_mask_ce: 0.9243 decode.loss_mask_dice: 1.7868 decode.d7.loss_cls_ce: 2.1685 decode.d7.loss_mask_ce: 0.9196 decode.d7.loss_mask_dice: 1.7926 2023/09/06 16:33:47 - mmengine - INFO - Iter(train) [16150/60000] base_lr: 7.3085e-05 lr: 7.3085e-05 eta: 5:29:20 time: 0.4514 data_time: 0.0228 memory: 15809 grad_norm: 15.5875 loss: 9.8121 decode.loss_cls_ce: 2.1832 decode.loss_mask_ce: 0.8834 decode.loss_mask_dice: 1.8216 decode.d7.loss_cls_ce: 2.2196 decode.d7.loss_mask_ce: 0.8829 decode.d7.loss_mask_dice: 1.8214 2023/09/06 16:34:10 - mmengine - INFO - Iter(train) [16200/60000] base_lr: 7.3001e-05 lr: 7.3001e-05 eta: 5:28:58 time: 0.4509 data_time: 0.0241 memory: 15909 grad_norm: 16.3953 loss: 10.0968 decode.loss_cls_ce: 2.3075 decode.loss_mask_ce: 0.9451 decode.loss_mask_dice: 1.8225 decode.d7.loss_cls_ce: 2.2846 decode.d7.loss_mask_ce: 0.9314 decode.d7.loss_mask_dice: 1.8057 2023/09/06 16:34:33 - mmengine - INFO - Iter(train) [16250/60000] base_lr: 7.2918e-05 lr: 7.2918e-05 eta: 5:28:36 time: 0.4525 data_time: 0.0236 memory: 16025 grad_norm: 16.3114 loss: 8.7084 decode.loss_cls_ce: 1.9658 decode.loss_mask_ce: 0.8622 decode.loss_mask_dice: 1.5204 decode.d7.loss_cls_ce: 1.9873 decode.d7.loss_mask_ce: 0.8581 decode.d7.loss_mask_dice: 1.5146 2023/09/06 16:34:55 - mmengine - INFO - Iter(train) [16300/60000] base_lr: 7.2835e-05 lr: 7.2835e-05 eta: 5:28:13 time: 0.4525 data_time: 0.0234 memory: 15804 grad_norm: 15.6097 loss: 9.7053 decode.loss_cls_ce: 2.2384 decode.loss_mask_ce: 0.8703 decode.loss_mask_dice: 1.7421 decode.d7.loss_cls_ce: 2.2441 decode.d7.loss_mask_ce: 0.8721 decode.d7.loss_mask_dice: 1.7383 2023/09/06 16:35:18 - mmengine - INFO - Iter(train) [16350/60000] base_lr: 7.2751e-05 lr: 7.2751e-05 eta: 5:27:51 time: 0.4540 data_time: 0.0231 memory: 15820 grad_norm: 15.9073 loss: 9.6767 decode.loss_cls_ce: 2.0951 decode.loss_mask_ce: 0.9960 decode.loss_mask_dice: 1.7570 decode.d7.loss_cls_ce: 2.0811 decode.d7.loss_mask_ce: 0.9900 decode.d7.loss_mask_dice: 1.7575 2023/09/06 16:35:40 - mmengine - INFO - Iter(train) [16400/60000] base_lr: 7.2668e-05 lr: 7.2668e-05 eta: 5:27:28 time: 0.4519 data_time: 0.0231 memory: 15758 grad_norm: 17.3067 loss: 9.9540 decode.loss_cls_ce: 2.1599 decode.loss_mask_ce: 0.8878 decode.loss_mask_dice: 1.9326 decode.d7.loss_cls_ce: 2.1720 decode.d7.loss_mask_ce: 0.8796 decode.d7.loss_mask_dice: 1.9221 2023/09/06 16:36:03 - mmengine - INFO - Iter(train) [16450/60000] base_lr: 7.2585e-05 lr: 7.2585e-05 eta: 5:27:06 time: 0.4532 data_time: 0.0233 memory: 15755 grad_norm: 17.0347 loss: 8.9922 decode.loss_cls_ce: 2.0410 decode.loss_mask_ce: 0.8468 decode.loss_mask_dice: 1.6119 decode.d7.loss_cls_ce: 2.0196 decode.d7.loss_mask_ce: 0.8575 decode.d7.loss_mask_dice: 1.6153 2023/09/06 16:36:25 - mmengine - INFO - Iter(train) [16500/60000] base_lr: 7.2501e-05 lr: 7.2501e-05 eta: 5:26:43 time: 0.4498 data_time: 0.0229 memory: 15859 grad_norm: 16.9062 loss: 10.9978 decode.loss_cls_ce: 2.4487 decode.loss_mask_ce: 1.0490 decode.loss_mask_dice: 2.0136 decode.d7.loss_cls_ce: 2.4356 decode.d7.loss_mask_ce: 1.0440 decode.d7.loss_mask_dice: 2.0069 2023/09/06 16:36:48 - mmengine - INFO - Iter(train) [16550/60000] base_lr: 7.2418e-05 lr: 7.2418e-05 eta: 5:26:21 time: 0.4516 data_time: 0.0239 memory: 15913 grad_norm: 15.7707 loss: 10.8013 decode.loss_cls_ce: 2.5205 decode.loss_mask_ce: 0.9119 decode.loss_mask_dice: 1.9859 decode.d7.loss_cls_ce: 2.4731 decode.d7.loss_mask_ce: 0.9221 decode.d7.loss_mask_dice: 1.9878 2023/09/06 16:37:11 - mmengine - INFO - Iter(train) [16600/60000] base_lr: 7.2335e-05 lr: 7.2335e-05 eta: 5:25:58 time: 0.4499 data_time: 0.0245 memory: 15810 grad_norm: 16.5131 loss: 9.9500 decode.loss_cls_ce: 2.2101 decode.loss_mask_ce: 0.9881 decode.loss_mask_dice: 1.7706 decode.d7.loss_cls_ce: 2.2175 decode.d7.loss_mask_ce: 0.9940 decode.d7.loss_mask_dice: 1.7697 2023/09/06 16:37:33 - mmengine - INFO - Iter(train) [16650/60000] base_lr: 7.2251e-05 lr: 7.2251e-05 eta: 5:25:36 time: 0.4510 data_time: 0.0240 memory: 15910 grad_norm: 16.7253 loss: 9.8220 decode.loss_cls_ce: 2.2080 decode.loss_mask_ce: 0.9220 decode.loss_mask_dice: 1.7772 decode.d7.loss_cls_ce: 2.2223 decode.d7.loss_mask_ce: 0.9251 decode.d7.loss_mask_dice: 1.7675 2023/09/06 16:37:56 - mmengine - INFO - Iter(train) [16700/60000] base_lr: 7.2168e-05 lr: 7.2168e-05 eta: 5:25:13 time: 0.4572 data_time: 0.0224 memory: 16064 grad_norm: 16.7669 loss: 9.4315 decode.loss_cls_ce: 2.0275 decode.loss_mask_ce: 0.9144 decode.loss_mask_dice: 1.7631 decode.d7.loss_cls_ce: 2.0486 decode.d7.loss_mask_ce: 0.9143 decode.d7.loss_mask_dice: 1.7636 2023/09/06 16:38:18 - mmengine - INFO - Iter(train) [16750/60000] base_lr: 7.2085e-05 lr: 7.2085e-05 eta: 5:24:51 time: 0.4495 data_time: 0.0236 memory: 15821 grad_norm: 16.8054 loss: 9.8131 decode.loss_cls_ce: 2.2783 decode.loss_mask_ce: 0.9160 decode.loss_mask_dice: 1.7437 decode.d7.loss_cls_ce: 2.2381 decode.d7.loss_mask_ce: 0.9114 decode.d7.loss_mask_dice: 1.7256 2023/09/06 16:38:41 - mmengine - INFO - Iter(train) [16800/60000] base_lr: 7.2001e-05 lr: 7.2001e-05 eta: 5:24:28 time: 0.4500 data_time: 0.0241 memory: 16000 grad_norm: 15.7413 loss: 10.2549 decode.loss_cls_ce: 2.3047 decode.loss_mask_ce: 0.8667 decode.loss_mask_dice: 1.9599 decode.d7.loss_cls_ce: 2.3016 decode.d7.loss_mask_ce: 0.8668 decode.d7.loss_mask_dice: 1.9551 2023/09/06 16:39:03 - mmengine - INFO - Iter(train) [16850/60000] base_lr: 7.1918e-05 lr: 7.1918e-05 eta: 5:24:06 time: 0.4606 data_time: 0.0230 memory: 15716 grad_norm: 16.1619 loss: 9.8707 decode.loss_cls_ce: 2.2353 decode.loss_mask_ce: 0.9034 decode.loss_mask_dice: 1.7995 decode.d7.loss_cls_ce: 2.1861 decode.d7.loss_mask_ce: 0.9185 decode.d7.loss_mask_dice: 1.8281 2023/09/06 16:39:26 - mmengine - INFO - Iter(train) [16900/60000] base_lr: 7.1835e-05 lr: 7.1835e-05 eta: 5:23:44 time: 0.4509 data_time: 0.0238 memory: 15873 grad_norm: 16.5440 loss: 10.3900 decode.loss_cls_ce: 2.2621 decode.loss_mask_ce: 0.9534 decode.loss_mask_dice: 1.9695 decode.d7.loss_cls_ce: 2.2738 decode.d7.loss_mask_ce: 0.9596 decode.d7.loss_mask_dice: 1.9715 2023/09/06 16:39:49 - mmengine - INFO - Iter(train) [16950/60000] base_lr: 7.1751e-05 lr: 7.1751e-05 eta: 5:23:21 time: 0.4491 data_time: 0.0231 memory: 15784 grad_norm: 16.6152 loss: 9.9372 decode.loss_cls_ce: 2.1739 decode.loss_mask_ce: 0.9924 decode.loss_mask_dice: 1.7896 decode.d7.loss_cls_ce: 2.2025 decode.d7.loss_mask_ce: 0.9847 decode.d7.loss_mask_dice: 1.7941 2023/09/06 16:40:11 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:40:11 - mmengine - INFO - Iter(train) [17000/60000] base_lr: 7.1668e-05 lr: 7.1668e-05 eta: 5:22:59 time: 0.4505 data_time: 0.0230 memory: 15759 grad_norm: 16.0357 loss: 9.3939 decode.loss_cls_ce: 2.0891 decode.loss_mask_ce: 0.9233 decode.loss_mask_dice: 1.6672 decode.d7.loss_cls_ce: 2.1163 decode.d7.loss_mask_ce: 0.9161 decode.d7.loss_mask_dice: 1.6820 2023/09/06 16:40:34 - mmengine - INFO - Iter(train) [17050/60000] base_lr: 7.1585e-05 lr: 7.1585e-05 eta: 5:22:37 time: 0.4510 data_time: 0.0239 memory: 15822 grad_norm: 17.5059 loss: 8.7432 decode.loss_cls_ce: 2.0479 decode.loss_mask_ce: 0.8694 decode.loss_mask_dice: 1.4531 decode.d7.loss_cls_ce: 2.0397 decode.d7.loss_mask_ce: 0.8705 decode.d7.loss_mask_dice: 1.4627 2023/09/06 16:40:56 - mmengine - INFO - Iter(train) [17100/60000] base_lr: 7.1501e-05 lr: 7.1501e-05 eta: 5:22:14 time: 0.4508 data_time: 0.0240 memory: 15835 grad_norm: 15.4312 loss: 9.2749 decode.loss_cls_ce: 2.0173 decode.loss_mask_ce: 0.8942 decode.loss_mask_dice: 1.7232 decode.d7.loss_cls_ce: 2.0097 decode.d7.loss_mask_ce: 0.8947 decode.d7.loss_mask_dice: 1.7359 2023/09/06 16:41:19 - mmengine - INFO - Iter(train) [17150/60000] base_lr: 7.1418e-05 lr: 7.1418e-05 eta: 5:21:51 time: 0.4511 data_time: 0.0232 memory: 15861 grad_norm: 17.7382 loss: 9.7101 decode.loss_cls_ce: 2.1807 decode.loss_mask_ce: 0.8979 decode.loss_mask_dice: 1.7701 decode.d7.loss_cls_ce: 2.1776 decode.d7.loss_mask_ce: 0.9011 decode.d7.loss_mask_dice: 1.7827 2023/09/06 16:41:42 - mmengine - INFO - Iter(train) [17200/60000] base_lr: 7.1335e-05 lr: 7.1335e-05 eta: 5:21:29 time: 0.4511 data_time: 0.0227 memory: 15849 grad_norm: 15.3894 loss: 9.3728 decode.loss_cls_ce: 2.0729 decode.loss_mask_ce: 0.8753 decode.loss_mask_dice: 1.7303 decode.d7.loss_cls_ce: 2.0932 decode.d7.loss_mask_ce: 0.8682 decode.d7.loss_mask_dice: 1.7329 2023/09/06 16:42:04 - mmengine - INFO - Iter(train) [17250/60000] base_lr: 7.1251e-05 lr: 7.1251e-05 eta: 5:21:07 time: 0.4506 data_time: 0.0223 memory: 15988 grad_norm: 16.3454 loss: 11.6509 decode.loss_cls_ce: 2.5021 decode.loss_mask_ce: 1.0454 decode.loss_mask_dice: 2.2754 decode.d7.loss_cls_ce: 2.5036 decode.d7.loss_mask_ce: 1.0417 decode.d7.loss_mask_dice: 2.2827 2023/09/06 16:42:27 - mmengine - INFO - Iter(train) [17300/60000] base_lr: 7.1168e-05 lr: 7.1168e-05 eta: 5:20:44 time: 0.4497 data_time: 0.0227 memory: 15885 grad_norm: 17.2049 loss: 10.3053 decode.loss_cls_ce: 2.3301 decode.loss_mask_ce: 0.9521 decode.loss_mask_dice: 1.8781 decode.d7.loss_cls_ce: 2.3239 decode.d7.loss_mask_ce: 0.9564 decode.d7.loss_mask_dice: 1.8648 2023/09/06 16:42:49 - mmengine - INFO - Iter(train) [17350/60000] base_lr: 7.1085e-05 lr: 7.1085e-05 eta: 5:20:22 time: 0.4501 data_time: 0.0222 memory: 15924 grad_norm: 18.4025 loss: 9.6299 decode.loss_cls_ce: 2.1696 decode.loss_mask_ce: 0.9158 decode.loss_mask_dice: 1.7295 decode.d7.loss_cls_ce: 2.1818 decode.d7.loss_mask_ce: 0.9132 decode.d7.loss_mask_dice: 1.7200 2023/09/06 16:43:12 - mmengine - INFO - Iter(train) [17400/60000] base_lr: 7.1001e-05 lr: 7.1001e-05 eta: 5:19:59 time: 0.4528 data_time: 0.0220 memory: 15886 grad_norm: 16.7550 loss: 10.8779 decode.loss_cls_ce: 2.4453 decode.loss_mask_ce: 1.0066 decode.loss_mask_dice: 1.9615 decode.d7.loss_cls_ce: 2.4783 decode.d7.loss_mask_ce: 1.0122 decode.d7.loss_mask_dice: 1.9740 2023/09/06 16:43:34 - mmengine - INFO - Iter(train) [17450/60000] base_lr: 7.0918e-05 lr: 7.0918e-05 eta: 5:19:36 time: 0.4506 data_time: 0.0221 memory: 15784 grad_norm: 17.1943 loss: 10.2069 decode.loss_cls_ce: 2.3255 decode.loss_mask_ce: 0.8937 decode.loss_mask_dice: 1.8997 decode.d7.loss_cls_ce: 2.3066 decode.d7.loss_mask_ce: 0.8964 decode.d7.loss_mask_dice: 1.8849 2023/09/06 16:43:57 - mmengine - INFO - Iter(train) [17500/60000] base_lr: 7.0835e-05 lr: 7.0835e-05 eta: 5:19:14 time: 0.4542 data_time: 0.0215 memory: 15846 grad_norm: 15.4682 loss: 9.6663 decode.loss_cls_ce: 2.1150 decode.loss_mask_ce: 0.9542 decode.loss_mask_dice: 1.7574 decode.d7.loss_cls_ce: 2.1004 decode.d7.loss_mask_ce: 0.9661 decode.d7.loss_mask_dice: 1.7731 2023/09/06 16:44:20 - mmengine - INFO - Iter(train) [17550/60000] base_lr: 7.0751e-05 lr: 7.0751e-05 eta: 5:18:52 time: 0.4495 data_time: 0.0220 memory: 15743 grad_norm: 16.3798 loss: 10.7161 decode.loss_cls_ce: 2.3420 decode.loss_mask_ce: 0.9719 decode.loss_mask_dice: 2.0377 decode.d7.loss_cls_ce: 2.3556 decode.d7.loss_mask_ce: 0.9694 decode.d7.loss_mask_dice: 2.0395 2023/09/06 16:44:42 - mmengine - INFO - Iter(train) [17600/60000] base_lr: 7.0668e-05 lr: 7.0668e-05 eta: 5:18:29 time: 0.4489 data_time: 0.0224 memory: 15874 grad_norm: 16.8000 loss: 9.6857 decode.loss_cls_ce: 2.1636 decode.loss_mask_ce: 0.9262 decode.loss_mask_dice: 1.7473 decode.d7.loss_cls_ce: 2.1761 decode.d7.loss_mask_ce: 0.9286 decode.d7.loss_mask_dice: 1.7438 2023/09/06 16:45:05 - mmengine - INFO - Iter(train) [17650/60000] base_lr: 7.0585e-05 lr: 7.0585e-05 eta: 5:18:07 time: 0.4526 data_time: 0.0215 memory: 15861 grad_norm: 16.4040 loss: 9.6191 decode.loss_cls_ce: 2.1317 decode.loss_mask_ce: 0.8982 decode.loss_mask_dice: 1.7759 decode.d7.loss_cls_ce: 2.1288 decode.d7.loss_mask_ce: 0.9037 decode.d7.loss_mask_dice: 1.7808 2023/09/06 16:45:27 - mmengine - INFO - Iter(train) [17700/60000] base_lr: 7.0501e-05 lr: 7.0501e-05 eta: 5:17:45 time: 0.4503 data_time: 0.0220 memory: 15759 grad_norm: 18.2525 loss: 9.3661 decode.loss_cls_ce: 2.0430 decode.loss_mask_ce: 0.9195 decode.loss_mask_dice: 1.7195 decode.d7.loss_cls_ce: 2.0505 decode.d7.loss_mask_ce: 0.9184 decode.d7.loss_mask_dice: 1.7151 2023/09/06 16:45:50 - mmengine - INFO - Iter(train) [17750/60000] base_lr: 7.0418e-05 lr: 7.0418e-05 eta: 5:17:23 time: 0.4518 data_time: 0.0226 memory: 15759 grad_norm: 17.2183 loss: 9.8645 decode.loss_cls_ce: 2.2161 decode.loss_mask_ce: 0.9667 decode.loss_mask_dice: 1.7448 decode.d7.loss_cls_ce: 2.2218 decode.d7.loss_mask_ce: 0.9641 decode.d7.loss_mask_dice: 1.7510 2023/09/06 16:46:13 - mmengine - INFO - Iter(train) [17800/60000] base_lr: 7.0335e-05 lr: 7.0335e-05 eta: 5:17:00 time: 0.4526 data_time: 0.0224 memory: 15847 grad_norm: 16.5322 loss: 9.3534 decode.loss_cls_ce: 2.0125 decode.loss_mask_ce: 0.8902 decode.loss_mask_dice: 1.7743 decode.d7.loss_cls_ce: 2.0266 decode.d7.loss_mask_ce: 0.8768 decode.d7.loss_mask_dice: 1.7729 2023/09/06 16:46:35 - mmengine - INFO - Iter(train) [17850/60000] base_lr: 7.0251e-05 lr: 7.0251e-05 eta: 5:16:38 time: 0.4508 data_time: 0.0222 memory: 15781 grad_norm: 16.7643 loss: 9.5763 decode.loss_cls_ce: 2.1639 decode.loss_mask_ce: 0.8881 decode.loss_mask_dice: 1.7190 decode.d7.loss_cls_ce: 2.2231 decode.d7.loss_mask_ce: 0.8689 decode.d7.loss_mask_dice: 1.7133 2023/09/06 16:46:58 - mmengine - INFO - Iter(train) [17900/60000] base_lr: 7.0168e-05 lr: 7.0168e-05 eta: 5:16:15 time: 0.4499 data_time: 0.0218 memory: 15787 grad_norm: 16.8581 loss: 9.7662 decode.loss_cls_ce: 2.1464 decode.loss_mask_ce: 0.9390 decode.loss_mask_dice: 1.7909 decode.d7.loss_cls_ce: 2.1682 decode.d7.loss_mask_ce: 0.9349 decode.d7.loss_mask_dice: 1.7868 2023/09/06 16:47:20 - mmengine - INFO - Iter(train) [17950/60000] base_lr: 7.0085e-05 lr: 7.0085e-05 eta: 5:15:53 time: 0.4515 data_time: 0.0226 memory: 16122 grad_norm: 15.7257 loss: 10.2735 decode.loss_cls_ce: 2.1793 decode.loss_mask_ce: 0.9409 decode.loss_mask_dice: 2.0085 decode.d7.loss_cls_ce: 2.1974 decode.d7.loss_mask_ce: 0.9270 decode.d7.loss_mask_dice: 2.0203 2023/09/06 16:47:43 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:47:43 - mmengine - INFO - Iter(train) [18000/60000] base_lr: 7.0001e-05 lr: 7.0001e-05 eta: 5:15:30 time: 0.4495 data_time: 0.0225 memory: 15797 grad_norm: 17.1526 loss: 9.8001 decode.loss_cls_ce: 2.2668 decode.loss_mask_ce: 0.8303 decode.loss_mask_dice: 1.8192 decode.d7.loss_cls_ce: 2.2283 decode.d7.loss_mask_ce: 0.8281 decode.d7.loss_mask_dice: 1.8273 2023/09/06 16:48:05 - mmengine - INFO - Iter(train) [18050/60000] base_lr: 6.9918e-05 lr: 6.9918e-05 eta: 5:15:08 time: 0.4509 data_time: 0.0229 memory: 15875 grad_norm: 20.0923 loss: 9.9680 decode.loss_cls_ce: 2.2400 decode.loss_mask_ce: 0.8963 decode.loss_mask_dice: 1.8378 decode.d7.loss_cls_ce: 2.2450 decode.d7.loss_mask_ce: 0.9036 decode.d7.loss_mask_dice: 1.8452 2023/09/06 16:48:28 - mmengine - INFO - Iter(train) [18100/60000] base_lr: 6.9834e-05 lr: 6.9834e-05 eta: 5:14:45 time: 0.4522 data_time: 0.0236 memory: 15846 grad_norm: 17.1696 loss: 8.8805 decode.loss_cls_ce: 2.0250 decode.loss_mask_ce: 0.8468 decode.loss_mask_dice: 1.5783 decode.d7.loss_cls_ce: 2.0141 decode.d7.loss_mask_ce: 0.8468 decode.d7.loss_mask_dice: 1.5695 2023/09/06 16:48:51 - mmengine - INFO - Iter(train) [18150/60000] base_lr: 6.9751e-05 lr: 6.9751e-05 eta: 5:14:23 time: 0.4504 data_time: 0.0230 memory: 15860 grad_norm: 17.9293 loss: 10.0398 decode.loss_cls_ce: 2.2532 decode.loss_mask_ce: 0.8750 decode.loss_mask_dice: 1.8961 decode.d7.loss_cls_ce: 2.2388 decode.d7.loss_mask_ce: 0.8739 decode.d7.loss_mask_dice: 1.9028 2023/09/06 16:49:13 - mmengine - INFO - Iter(train) [18200/60000] base_lr: 6.9668e-05 lr: 6.9668e-05 eta: 5:14:00 time: 0.4517 data_time: 0.0229 memory: 15937 grad_norm: 17.4445 loss: 8.9399 decode.loss_cls_ce: 1.9856 decode.loss_mask_ce: 0.8765 decode.loss_mask_dice: 1.5981 decode.d7.loss_cls_ce: 1.9873 decode.d7.loss_mask_ce: 0.8797 decode.d7.loss_mask_dice: 1.6126 2023/09/06 16:49:36 - mmengine - INFO - Iter(train) [18250/60000] base_lr: 6.9584e-05 lr: 6.9584e-05 eta: 5:13:38 time: 0.4560 data_time: 0.0242 memory: 15808 grad_norm: 17.7394 loss: 8.6933 decode.loss_cls_ce: 1.9409 decode.loss_mask_ce: 0.8749 decode.loss_mask_dice: 1.5345 decode.d7.loss_cls_ce: 1.9535 decode.d7.loss_mask_ce: 0.8617 decode.d7.loss_mask_dice: 1.5278 2023/09/06 16:49:58 - mmengine - INFO - Iter(train) [18300/60000] base_lr: 6.9501e-05 lr: 6.9501e-05 eta: 5:13:16 time: 0.4491 data_time: 0.0228 memory: 15781 grad_norm: 18.1063 loss: 10.4040 decode.loss_cls_ce: 2.3817 decode.loss_mask_ce: 0.8991 decode.loss_mask_dice: 1.9481 decode.d7.loss_cls_ce: 2.3277 decode.d7.loss_mask_ce: 0.9009 decode.d7.loss_mask_dice: 1.9464 2023/09/06 16:50:21 - mmengine - INFO - Iter(train) [18350/60000] base_lr: 6.9418e-05 lr: 6.9418e-05 eta: 5:12:53 time: 0.4517 data_time: 0.0228 memory: 15831 grad_norm: 14.8972 loss: 10.5549 decode.loss_cls_ce: 2.2508 decode.loss_mask_ce: 1.0024 decode.loss_mask_dice: 2.0240 decode.d7.loss_cls_ce: 2.2654 decode.d7.loss_mask_ce: 0.9976 decode.d7.loss_mask_dice: 2.0148 2023/09/06 16:50:44 - mmengine - INFO - Iter(train) [18400/60000] base_lr: 6.9334e-05 lr: 6.9334e-05 eta: 5:12:31 time: 0.4547 data_time: 0.0227 memory: 15792 grad_norm: 20.6350 loss: 9.7783 decode.loss_cls_ce: 2.1738 decode.loss_mask_ce: 0.9221 decode.loss_mask_dice: 1.7871 decode.d7.loss_cls_ce: 2.1988 decode.d7.loss_mask_ce: 0.9127 decode.d7.loss_mask_dice: 1.7836 2023/09/06 16:51:06 - mmengine - INFO - Iter(train) [18450/60000] base_lr: 6.9251e-05 lr: 6.9251e-05 eta: 5:12:09 time: 0.4491 data_time: 0.0230 memory: 15858 grad_norm: 17.5125 loss: 9.8772 decode.loss_cls_ce: 2.1501 decode.loss_mask_ce: 0.9053 decode.loss_mask_dice: 1.8767 decode.d7.loss_cls_ce: 2.1761 decode.d7.loss_mask_ce: 0.8944 decode.d7.loss_mask_dice: 1.8746 2023/09/06 16:51:29 - mmengine - INFO - Iter(train) [18500/60000] base_lr: 6.9168e-05 lr: 6.9168e-05 eta: 5:11:47 time: 0.4538 data_time: 0.0232 memory: 15860 grad_norm: 16.0390 loss: 8.5876 decode.loss_cls_ce: 2.0390 decode.loss_mask_ce: 0.7860 decode.loss_mask_dice: 1.4722 decode.d7.loss_cls_ce: 2.0216 decode.d7.loss_mask_ce: 0.7864 decode.d7.loss_mask_dice: 1.4824 2023/09/06 16:51:52 - mmengine - INFO - Iter(train) [18550/60000] base_lr: 6.9084e-05 lr: 6.9084e-05 eta: 5:11:24 time: 0.4521 data_time: 0.0238 memory: 15908 grad_norm: 16.5664 loss: 10.4955 decode.loss_cls_ce: 2.3908 decode.loss_mask_ce: 0.9522 decode.loss_mask_dice: 1.9115 decode.d7.loss_cls_ce: 2.3896 decode.d7.loss_mask_ce: 0.9525 decode.d7.loss_mask_dice: 1.8989 2023/09/06 16:52:14 - mmengine - INFO - Iter(train) [18600/60000] base_lr: 6.9001e-05 lr: 6.9001e-05 eta: 5:11:02 time: 0.4532 data_time: 0.0233 memory: 15900 grad_norm: 14.0230 loss: 9.6734 decode.loss_cls_ce: 2.1634 decode.loss_mask_ce: 0.8552 decode.loss_mask_dice: 1.8155 decode.d7.loss_cls_ce: 2.1552 decode.d7.loss_mask_ce: 0.8552 decode.d7.loss_mask_dice: 1.8287 2023/09/06 16:52:37 - mmengine - INFO - Iter(train) [18650/60000] base_lr: 6.8918e-05 lr: 6.8918e-05 eta: 5:10:39 time: 0.4547 data_time: 0.0227 memory: 15870 grad_norm: 19.4073 loss: 10.3676 decode.loss_cls_ce: 2.2861 decode.loss_mask_ce: 1.0039 decode.loss_mask_dice: 1.9095 decode.d7.loss_cls_ce: 2.2730 decode.d7.loss_mask_ce: 0.9961 decode.d7.loss_mask_dice: 1.8991 2023/09/06 16:53:00 - mmengine - INFO - Iter(train) [18700/60000] base_lr: 6.8834e-05 lr: 6.8834e-05 eta: 5:10:17 time: 0.4539 data_time: 0.0224 memory: 15859 grad_norm: 16.1074 loss: 10.0304 decode.loss_cls_ce: 2.3285 decode.loss_mask_ce: 0.8739 decode.loss_mask_dice: 1.8076 decode.d7.loss_cls_ce: 2.3408 decode.d7.loss_mask_ce: 0.8754 decode.d7.loss_mask_dice: 1.8042 2023/09/06 16:53:22 - mmengine - INFO - Iter(train) [18750/60000] base_lr: 6.8751e-05 lr: 6.8751e-05 eta: 5:09:55 time: 0.4517 data_time: 0.0227 memory: 15832 grad_norm: 16.4554 loss: 9.8496 decode.loss_cls_ce: 2.2646 decode.loss_mask_ce: 0.8730 decode.loss_mask_dice: 1.7764 decode.d7.loss_cls_ce: 2.3014 decode.d7.loss_mask_ce: 0.8690 decode.d7.loss_mask_dice: 1.7653 2023/09/06 16:53:45 - mmengine - INFO - Iter(train) [18800/60000] base_lr: 6.8668e-05 lr: 6.8668e-05 eta: 5:09:32 time: 0.4523 data_time: 0.0233 memory: 15773 grad_norm: 15.5004 loss: 9.8342 decode.loss_cls_ce: 2.1852 decode.loss_mask_ce: 0.9232 decode.loss_mask_dice: 1.8171 decode.d7.loss_cls_ce: 2.1618 decode.d7.loss_mask_ce: 0.9274 decode.d7.loss_mask_dice: 1.8194 2023/09/06 16:54:07 - mmengine - INFO - Iter(train) [18850/60000] base_lr: 6.8584e-05 lr: 6.8584e-05 eta: 5:09:10 time: 0.4537 data_time: 0.0233 memory: 15938 grad_norm: 18.2684 loss: 9.4838 decode.loss_cls_ce: 2.1217 decode.loss_mask_ce: 0.8125 decode.loss_mask_dice: 1.8085 decode.d7.loss_cls_ce: 2.1040 decode.d7.loss_mask_ce: 0.8074 decode.d7.loss_mask_dice: 1.8296 2023/09/06 16:54:30 - mmengine - INFO - Iter(train) [18900/60000] base_lr: 6.8501e-05 lr: 6.8501e-05 eta: 5:08:48 time: 0.4554 data_time: 0.0232 memory: 15795 grad_norm: 15.6990 loss: 8.6430 decode.loss_cls_ce: 1.8983 decode.loss_mask_ce: 0.8184 decode.loss_mask_dice: 1.6154 decode.d7.loss_cls_ce: 1.8832 decode.d7.loss_mask_ce: 0.8208 decode.d7.loss_mask_dice: 1.6069 2023/09/06 16:54:53 - mmengine - INFO - Iter(train) [18950/60000] base_lr: 6.8418e-05 lr: 6.8418e-05 eta: 5:08:26 time: 0.4558 data_time: 0.0236 memory: 15887 grad_norm: 17.3889 loss: 9.6922 decode.loss_cls_ce: 2.1257 decode.loss_mask_ce: 0.9416 decode.loss_mask_dice: 1.7831 decode.d7.loss_cls_ce: 2.1116 decode.d7.loss_mask_ce: 0.9444 decode.d7.loss_mask_dice: 1.7858 2023/09/06 16:55:16 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 16:55:16 - mmengine - INFO - Iter(train) [19000/60000] base_lr: 6.8334e-05 lr: 6.8334e-05 eta: 5:08:03 time: 0.4551 data_time: 0.0235 memory: 15757 grad_norm: 17.5551 loss: 9.6434 decode.loss_cls_ce: 2.0329 decode.loss_mask_ce: 0.9692 decode.loss_mask_dice: 1.8081 decode.d7.loss_cls_ce: 2.0540 decode.d7.loss_mask_ce: 0.9728 decode.d7.loss_mask_dice: 1.8065 2023/09/06 16:55:38 - mmengine - INFO - Iter(train) [19050/60000] base_lr: 6.8251e-05 lr: 6.8251e-05 eta: 5:07:41 time: 0.4513 data_time: 0.0241 memory: 15887 grad_norm: 16.8466 loss: 10.2256 decode.loss_cls_ce: 2.3773 decode.loss_mask_ce: 0.8724 decode.loss_mask_dice: 1.8668 decode.d7.loss_cls_ce: 2.3690 decode.d7.loss_mask_ce: 0.8766 decode.d7.loss_mask_dice: 1.8634 2023/09/06 16:56:01 - mmengine - INFO - Iter(train) [19100/60000] base_lr: 6.8168e-05 lr: 6.8168e-05 eta: 5:07:18 time: 0.4517 data_time: 0.0235 memory: 15809 grad_norm: 16.7196 loss: 9.8489 decode.loss_cls_ce: 2.2675 decode.loss_mask_ce: 0.9466 decode.loss_mask_dice: 1.7088 decode.d7.loss_cls_ce: 2.2717 decode.d7.loss_mask_ce: 0.9504 decode.d7.loss_mask_dice: 1.7039 2023/09/06 16:56:23 - mmengine - INFO - Iter(train) [19150/60000] base_lr: 6.8084e-05 lr: 6.8084e-05 eta: 5:06:56 time: 0.4490 data_time: 0.0227 memory: 15769 grad_norm: 17.7476 loss: 9.4938 decode.loss_cls_ce: 2.0822 decode.loss_mask_ce: 0.8649 decode.loss_mask_dice: 1.7906 decode.d7.loss_cls_ce: 2.1057 decode.d7.loss_mask_ce: 0.8658 decode.d7.loss_mask_dice: 1.7845 2023/09/06 16:56:46 - mmengine - INFO - Iter(train) [19200/60000] base_lr: 6.8001e-05 lr: 6.8001e-05 eta: 5:06:33 time: 0.4489 data_time: 0.0227 memory: 15948 grad_norm: 16.6649 loss: 9.9625 decode.loss_cls_ce: 2.2686 decode.loss_mask_ce: 0.8639 decode.loss_mask_dice: 1.8656 decode.d7.loss_cls_ce: 2.2318 decode.d7.loss_mask_ce: 0.8633 decode.d7.loss_mask_dice: 1.8693 2023/09/06 16:57:08 - mmengine - INFO - Iter(train) [19250/60000] base_lr: 6.7918e-05 lr: 6.7918e-05 eta: 5:06:11 time: 0.4556 data_time: 0.0233 memory: 15858 grad_norm: 18.0742 loss: 9.8537 decode.loss_cls_ce: 2.2688 decode.loss_mask_ce: 0.9031 decode.loss_mask_dice: 1.7609 decode.d7.loss_cls_ce: 2.2465 decode.d7.loss_mask_ce: 0.9140 decode.d7.loss_mask_dice: 1.7606 2023/09/06 16:57:31 - mmengine - INFO - Iter(train) [19300/60000] base_lr: 6.7834e-05 lr: 6.7834e-05 eta: 5:05:48 time: 0.4524 data_time: 0.0231 memory: 15960 grad_norm: 16.1688 loss: 10.0334 decode.loss_cls_ce: 2.2894 decode.loss_mask_ce: 0.9134 decode.loss_mask_dice: 1.8153 decode.d7.loss_cls_ce: 2.2852 decode.d7.loss_mask_ce: 0.9186 decode.d7.loss_mask_dice: 1.8115 2023/09/06 16:57:53 - mmengine - INFO - Iter(train) [19350/60000] base_lr: 6.7751e-05 lr: 6.7751e-05 eta: 5:05:25 time: 0.4490 data_time: 0.0229 memory: 15833 grad_norm: 17.0850 loss: 9.1240 decode.loss_cls_ce: 2.0191 decode.loss_mask_ce: 0.8302 decode.loss_mask_dice: 1.7085 decode.d7.loss_cls_ce: 2.0543 decode.d7.loss_mask_ce: 0.8209 decode.d7.loss_mask_dice: 1.6910 2023/09/06 16:58:16 - mmengine - INFO - Iter(train) [19400/60000] base_lr: 6.7668e-05 lr: 6.7668e-05 eta: 5:05:03 time: 0.4495 data_time: 0.0234 memory: 15847 grad_norm: 16.7720 loss: 10.1039 decode.loss_cls_ce: 2.3142 decode.loss_mask_ce: 0.9219 decode.loss_mask_dice: 1.8204 decode.d7.loss_cls_ce: 2.3131 decode.d7.loss_mask_ce: 0.9222 decode.d7.loss_mask_dice: 1.8121 2023/09/06 16:58:38 - mmengine - INFO - Iter(train) [19450/60000] base_lr: 6.7584e-05 lr: 6.7584e-05 eta: 5:04:40 time: 0.4511 data_time: 0.0240 memory: 15846 grad_norm: 18.0889 loss: 10.3773 decode.loss_cls_ce: 2.3605 decode.loss_mask_ce: 0.9347 decode.loss_mask_dice: 1.8836 decode.d7.loss_cls_ce: 2.3901 decode.d7.loss_mask_ce: 0.9369 decode.d7.loss_mask_dice: 1.8715 2023/09/06 16:59:01 - mmengine - INFO - Iter(train) [19500/60000] base_lr: 6.7501e-05 lr: 6.7501e-05 eta: 5:04:18 time: 0.4477 data_time: 0.0234 memory: 15820 grad_norm: 15.5456 loss: 9.6468 decode.loss_cls_ce: 2.1291 decode.loss_mask_ce: 0.9277 decode.loss_mask_dice: 1.7414 decode.d7.loss_cls_ce: 2.1755 decode.d7.loss_mask_ce: 0.9282 decode.d7.loss_mask_dice: 1.7449 2023/09/06 16:59:24 - mmengine - INFO - Iter(train) [19550/60000] base_lr: 6.7418e-05 lr: 6.7418e-05 eta: 5:03:55 time: 0.4487 data_time: 0.0231 memory: 16017 grad_norm: 15.6216 loss: 9.8875 decode.loss_cls_ce: 2.2320 decode.loss_mask_ce: 0.8794 decode.loss_mask_dice: 1.8142 decode.d7.loss_cls_ce: 2.2384 decode.d7.loss_mask_ce: 0.8895 decode.d7.loss_mask_dice: 1.8340 2023/09/06 16:59:46 - mmengine - INFO - Iter(train) [19600/60000] base_lr: 6.7334e-05 lr: 6.7334e-05 eta: 5:03:33 time: 0.4507 data_time: 0.0235 memory: 15773 grad_norm: 16.6786 loss: 9.6287 decode.loss_cls_ce: 2.2149 decode.loss_mask_ce: 0.8634 decode.loss_mask_dice: 1.7346 decode.d7.loss_cls_ce: 2.2085 decode.d7.loss_mask_ce: 0.8685 decode.d7.loss_mask_dice: 1.7388 2023/09/06 17:00:09 - mmengine - INFO - Iter(train) [19650/60000] base_lr: 6.7251e-05 lr: 6.7251e-05 eta: 5:03:11 time: 0.4552 data_time: 0.0226 memory: 15909 grad_norm: 15.6782 loss: 10.4574 decode.loss_cls_ce: 2.2828 decode.loss_mask_ce: 0.9973 decode.loss_mask_dice: 1.9563 decode.d7.loss_cls_ce: 2.2854 decode.d7.loss_mask_ce: 0.9940 decode.d7.loss_mask_dice: 1.9416 2023/09/06 17:00:32 - mmengine - INFO - Iter(train) [19700/60000] base_lr: 6.7168e-05 lr: 6.7168e-05 eta: 5:02:48 time: 0.4541 data_time: 0.0225 memory: 15872 grad_norm: 17.0107 loss: 9.0755 decode.loss_cls_ce: 1.9381 decode.loss_mask_ce: 0.8693 decode.loss_mask_dice: 1.7381 decode.d7.loss_cls_ce: 1.9048 decode.d7.loss_mask_ce: 0.8820 decode.d7.loss_mask_dice: 1.7431 2023/09/06 17:00:54 - mmengine - INFO - Iter(train) [19750/60000] base_lr: 6.7084e-05 lr: 6.7084e-05 eta: 5:02:26 time: 0.4551 data_time: 0.0229 memory: 15811 grad_norm: 15.1165 loss: 10.1764 decode.loss_cls_ce: 2.2042 decode.loss_mask_ce: 0.9626 decode.loss_mask_dice: 1.9049 decode.d7.loss_cls_ce: 2.2298 decode.d7.loss_mask_ce: 0.9626 decode.d7.loss_mask_dice: 1.9124 2023/09/06 17:01:17 - mmengine - INFO - Iter(train) [19800/60000] base_lr: 6.7001e-05 lr: 6.7001e-05 eta: 5:02:04 time: 0.4507 data_time: 0.0233 memory: 15899 grad_norm: 16.5586 loss: 10.0049 decode.loss_cls_ce: 2.1709 decode.loss_mask_ce: 0.9436 decode.loss_mask_dice: 1.8761 decode.d7.loss_cls_ce: 2.2022 decode.d7.loss_mask_ce: 0.9354 decode.d7.loss_mask_dice: 1.8768 2023/09/06 17:01:39 - mmengine - INFO - Iter(train) [19850/60000] base_lr: 6.6918e-05 lr: 6.6918e-05 eta: 5:01:41 time: 0.4521 data_time: 0.0237 memory: 15762 grad_norm: 17.9236 loss: 9.4186 decode.loss_cls_ce: 2.0786 decode.loss_mask_ce: 0.8957 decode.loss_mask_dice: 1.7241 decode.d7.loss_cls_ce: 2.0881 decode.d7.loss_mask_ce: 0.8957 decode.d7.loss_mask_dice: 1.7364 2023/09/06 17:02:02 - mmengine - INFO - Iter(train) [19900/60000] base_lr: 6.6834e-05 lr: 6.6834e-05 eta: 5:01:19 time: 0.4549 data_time: 0.0229 memory: 16041 grad_norm: 16.4409 loss: 9.5481 decode.loss_cls_ce: 2.1025 decode.loss_mask_ce: 0.9039 decode.loss_mask_dice: 1.7614 decode.d7.loss_cls_ce: 2.1115 decode.d7.loss_mask_ce: 0.9009 decode.d7.loss_mask_dice: 1.7679 2023/09/06 17:02:25 - mmengine - INFO - Iter(train) [19950/60000] base_lr: 6.6751e-05 lr: 6.6751e-05 eta: 5:00:57 time: 0.4577 data_time: 0.0234 memory: 15845 grad_norm: 17.8130 loss: 9.1222 decode.loss_cls_ce: 2.1545 decode.loss_mask_ce: 0.8311 decode.loss_mask_dice: 1.5864 decode.d7.loss_cls_ce: 2.1301 decode.d7.loss_mask_ce: 0.8326 decode.d7.loss_mask_dice: 1.5876 2023/09/06 17:02:47 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:02:47 - mmengine - INFO - Iter(train) [20000/60000] base_lr: 6.6668e-05 lr: 6.6668e-05 eta: 5:00:34 time: 0.4573 data_time: 0.0230 memory: 15819 grad_norm: 16.2255 loss: 9.5594 decode.loss_cls_ce: 2.2938 decode.loss_mask_ce: 0.8429 decode.loss_mask_dice: 1.6427 decode.d7.loss_cls_ce: 2.2892 decode.d7.loss_mask_ce: 0.8457 decode.d7.loss_mask_dice: 1.6452 2023/09/06 17:02:47 - mmengine - INFO - Saving checkpoint at 20000 iterations 2023/09/06 17:03:14 - mmengine - INFO - Iter(train) [20050/60000] base_lr: 6.6584e-05 lr: 6.6584e-05 eta: 5:00:19 time: 0.4538 data_time: 0.0241 memory: 15845 grad_norm: 17.0509 loss: 10.2359 decode.loss_cls_ce: 2.2504 decode.loss_mask_ce: 0.9686 decode.loss_mask_dice: 1.9069 decode.d7.loss_cls_ce: 2.2311 decode.d7.loss_mask_ce: 0.9652 decode.d7.loss_mask_dice: 1.9136 2023/09/06 17:03:36 - mmengine - INFO - Iter(train) [20100/60000] base_lr: 6.6501e-05 lr: 6.6501e-05 eta: 4:59:56 time: 0.4512 data_time: 0.0232 memory: 15948 grad_norm: 17.6597 loss: 10.1477 decode.loss_cls_ce: 2.2716 decode.loss_mask_ce: 0.9601 decode.loss_mask_dice: 1.8431 decode.d7.loss_cls_ce: 2.2716 decode.d7.loss_mask_ce: 0.9524 decode.d7.loss_mask_dice: 1.8489 2023/09/06 17:03:59 - mmengine - INFO - Iter(train) [20150/60000] base_lr: 6.6418e-05 lr: 6.6418e-05 eta: 4:59:34 time: 0.4490 data_time: 0.0235 memory: 15786 grad_norm: 18.2629 loss: 9.0376 decode.loss_cls_ce: 1.9276 decode.loss_mask_ce: 0.8855 decode.loss_mask_dice: 1.7019 decode.d7.loss_cls_ce: 1.9323 decode.d7.loss_mask_ce: 0.8913 decode.d7.loss_mask_dice: 1.6989 2023/09/06 17:04:21 - mmengine - INFO - Iter(train) [20200/60000] base_lr: 6.6334e-05 lr: 6.6334e-05 eta: 4:59:11 time: 0.4540 data_time: 0.0222 memory: 15883 grad_norm: 16.3721 loss: 10.2911 decode.loss_cls_ce: 2.2292 decode.loss_mask_ce: 0.8904 decode.loss_mask_dice: 2.0149 decode.d7.loss_cls_ce: 2.2691 decode.d7.loss_mask_ce: 0.8786 decode.d7.loss_mask_dice: 2.0089 2023/09/06 17:04:44 - mmengine - INFO - Iter(train) [20250/60000] base_lr: 6.6251e-05 lr: 6.6251e-05 eta: 4:58:49 time: 0.4513 data_time: 0.0236 memory: 15964 grad_norm: 17.2932 loss: 10.4348 decode.loss_cls_ce: 2.3225 decode.loss_mask_ce: 0.9713 decode.loss_mask_dice: 1.9140 decode.d7.loss_cls_ce: 2.3331 decode.d7.loss_mask_ce: 0.9688 decode.d7.loss_mask_dice: 1.9251 2023/09/06 17:05:06 - mmengine - INFO - Iter(train) [20300/60000] base_lr: 6.6168e-05 lr: 6.6168e-05 eta: 4:58:26 time: 0.4474 data_time: 0.0237 memory: 15834 grad_norm: 17.5264 loss: 9.2985 decode.loss_cls_ce: 2.1483 decode.loss_mask_ce: 0.8839 decode.loss_mask_dice: 1.6487 decode.d7.loss_cls_ce: 2.0948 decode.d7.loss_mask_ce: 0.8835 decode.d7.loss_mask_dice: 1.6393 2023/09/06 17:05:29 - mmengine - INFO - Iter(train) [20350/60000] base_lr: 6.6084e-05 lr: 6.6084e-05 eta: 4:58:03 time: 0.4482 data_time: 0.0236 memory: 15758 grad_norm: 18.1706 loss: 9.3991 decode.loss_cls_ce: 2.0987 decode.loss_mask_ce: 0.9327 decode.loss_mask_dice: 1.6796 decode.d7.loss_cls_ce: 2.0936 decode.d7.loss_mask_ce: 0.9296 decode.d7.loss_mask_dice: 1.6650 2023/09/06 17:05:51 - mmengine - INFO - Iter(train) [20400/60000] base_lr: 6.6001e-05 lr: 6.6001e-05 eta: 4:57:40 time: 0.4526 data_time: 0.0234 memory: 15873 grad_norm: 16.7629 loss: 8.6192 decode.loss_cls_ce: 1.8283 decode.loss_mask_ce: 0.8486 decode.loss_mask_dice: 1.6324 decode.d7.loss_cls_ce: 1.8357 decode.d7.loss_mask_ce: 0.8429 decode.d7.loss_mask_dice: 1.6313 2023/09/06 17:06:14 - mmengine - INFO - Iter(train) [20450/60000] base_lr: 6.5918e-05 lr: 6.5918e-05 eta: 4:57:18 time: 0.4482 data_time: 0.0234 memory: 15784 grad_norm: 17.1949 loss: 9.3174 decode.loss_cls_ce: 1.9772 decode.loss_mask_ce: 0.9052 decode.loss_mask_dice: 1.7610 decode.d7.loss_cls_ce: 2.0031 decode.d7.loss_mask_ce: 0.9037 decode.d7.loss_mask_dice: 1.7671 2023/09/06 17:06:36 - mmengine - INFO - Iter(train) [20500/60000] base_lr: 6.5834e-05 lr: 6.5834e-05 eta: 4:56:55 time: 0.4493 data_time: 0.0235 memory: 15859 grad_norm: 17.3620 loss: 9.1914 decode.loss_cls_ce: 1.9931 decode.loss_mask_ce: 0.8869 decode.loss_mask_dice: 1.6991 decode.d7.loss_cls_ce: 2.0222 decode.d7.loss_mask_ce: 0.8858 decode.d7.loss_mask_dice: 1.7044 2023/09/06 17:06:59 - mmengine - INFO - Iter(train) [20550/60000] base_lr: 6.5751e-05 lr: 6.5751e-05 eta: 4:56:33 time: 0.4537 data_time: 0.0232 memory: 15879 grad_norm: 16.4050 loss: 9.4780 decode.loss_cls_ce: 2.2032 decode.loss_mask_ce: 0.8951 decode.loss_mask_dice: 1.6502 decode.d7.loss_cls_ce: 2.1607 decode.d7.loss_mask_ce: 0.9017 decode.d7.loss_mask_dice: 1.6669 2023/09/06 17:07:22 - mmengine - INFO - Iter(train) [20600/60000] base_lr: 6.5668e-05 lr: 6.5668e-05 eta: 4:56:11 time: 0.4528 data_time: 0.0236 memory: 15935 grad_norm: 16.8784 loss: 8.8120 decode.loss_cls_ce: 2.0478 decode.loss_mask_ce: 0.8680 decode.loss_mask_dice: 1.4835 decode.d7.loss_cls_ce: 2.0458 decode.d7.loss_mask_ce: 0.8702 decode.d7.loss_mask_dice: 1.4968 2023/09/06 17:07:44 - mmengine - INFO - Iter(train) [20650/60000] base_lr: 6.5584e-05 lr: 6.5584e-05 eta: 4:55:48 time: 0.4488 data_time: 0.0241 memory: 15847 grad_norm: 17.0880 loss: 10.2807 decode.loss_cls_ce: 2.1394 decode.loss_mask_ce: 1.0380 decode.loss_mask_dice: 1.9557 decode.d7.loss_cls_ce: 2.1521 decode.d7.loss_mask_ce: 1.0337 decode.d7.loss_mask_dice: 1.9619 2023/09/06 17:08:07 - mmengine - INFO - Iter(train) [20700/60000] base_lr: 6.5501e-05 lr: 6.5501e-05 eta: 4:55:25 time: 0.4569 data_time: 0.0220 memory: 15858 grad_norm: 16.7916 loss: 9.6657 decode.loss_cls_ce: 2.1312 decode.loss_mask_ce: 0.8679 decode.loss_mask_dice: 1.8204 decode.d7.loss_cls_ce: 2.1376 decode.d7.loss_mask_ce: 0.8793 decode.d7.loss_mask_dice: 1.8293 2023/09/06 17:08:29 - mmengine - INFO - Iter(train) [20750/60000] base_lr: 6.5418e-05 lr: 6.5418e-05 eta: 4:55:03 time: 0.4513 data_time: 0.0241 memory: 15896 grad_norm: 16.8108 loss: 10.4616 decode.loss_cls_ce: 2.2643 decode.loss_mask_ce: 1.0322 decode.loss_mask_dice: 1.9397 decode.d7.loss_cls_ce: 2.2443 decode.d7.loss_mask_ce: 1.0423 decode.d7.loss_mask_dice: 1.9388 2023/09/06 17:08:52 - mmengine - INFO - Iter(train) [20800/60000] base_lr: 6.5334e-05 lr: 6.5334e-05 eta: 4:54:40 time: 0.4508 data_time: 0.0235 memory: 16024 grad_norm: 19.1134 loss: 9.9356 decode.loss_cls_ce: 2.2691 decode.loss_mask_ce: 0.8606 decode.loss_mask_dice: 1.8429 decode.d7.loss_cls_ce: 2.2607 decode.d7.loss_mask_ce: 0.8627 decode.d7.loss_mask_dice: 1.8396 2023/09/06 17:09:14 - mmengine - INFO - Iter(train) [20850/60000] base_lr: 6.5251e-05 lr: 6.5251e-05 eta: 4:54:18 time: 0.4501 data_time: 0.0237 memory: 15835 grad_norm: 15.9171 loss: 9.1387 decode.loss_cls_ce: 1.9993 decode.loss_mask_ce: 0.8804 decode.loss_mask_dice: 1.6841 decode.d7.loss_cls_ce: 2.0023 decode.d7.loss_mask_ce: 0.8780 decode.d7.loss_mask_dice: 1.6946 2023/09/06 17:09:37 - mmengine - INFO - Iter(train) [20900/60000] base_lr: 6.5168e-05 lr: 6.5168e-05 eta: 4:53:55 time: 0.4510 data_time: 0.0234 memory: 15848 grad_norm: 16.9515 loss: 10.6563 decode.loss_cls_ce: 2.3608 decode.loss_mask_ce: 0.9578 decode.loss_mask_dice: 1.9986 decode.d7.loss_cls_ce: 2.3642 decode.d7.loss_mask_ce: 0.9630 decode.d7.loss_mask_dice: 2.0119 2023/09/06 17:10:00 - mmengine - INFO - Iter(train) [20950/60000] base_lr: 6.5084e-05 lr: 6.5084e-05 eta: 4:53:33 time: 0.4564 data_time: 0.0228 memory: 15941 grad_norm: 16.9218 loss: 9.9108 decode.loss_cls_ce: 2.2044 decode.loss_mask_ce: 0.9810 decode.loss_mask_dice: 1.7588 decode.d7.loss_cls_ce: 2.2184 decode.d7.loss_mask_ce: 0.9722 decode.d7.loss_mask_dice: 1.7761 2023/09/06 17:10:22 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:10:22 - mmengine - INFO - Iter(train) [21000/60000] base_lr: 6.5001e-05 lr: 6.5001e-05 eta: 4:53:10 time: 0.4482 data_time: 0.0234 memory: 15821 grad_norm: 17.4309 loss: 11.0319 decode.loss_cls_ce: 2.5124 decode.loss_mask_ce: 0.9403 decode.loss_mask_dice: 2.0754 decode.d7.loss_cls_ce: 2.4670 decode.d7.loss_mask_ce: 0.9560 decode.d7.loss_mask_dice: 2.0808 2023/09/06 17:10:45 - mmengine - INFO - Iter(train) [21050/60000] base_lr: 6.4918e-05 lr: 6.4918e-05 eta: 4:52:48 time: 0.4539 data_time: 0.0228 memory: 15769 grad_norm: 18.0546 loss: 10.6334 decode.loss_cls_ce: 2.4920 decode.loss_mask_ce: 0.8943 decode.loss_mask_dice: 1.9332 decode.d7.loss_cls_ce: 2.4857 decode.d7.loss_mask_ce: 0.8994 decode.d7.loss_mask_dice: 1.9288 2023/09/06 17:11:07 - mmengine - INFO - Iter(train) [21100/60000] base_lr: 6.4834e-05 lr: 6.4834e-05 eta: 4:52:26 time: 0.4497 data_time: 0.0238 memory: 15799 grad_norm: 16.2286 loss: 10.3600 decode.loss_cls_ce: 2.2799 decode.loss_mask_ce: 0.9304 decode.loss_mask_dice: 1.9628 decode.d7.loss_cls_ce: 2.3174 decode.d7.loss_mask_ce: 0.9209 decode.d7.loss_mask_dice: 1.9485 2023/09/06 17:11:30 - mmengine - INFO - Iter(train) [21150/60000] base_lr: 6.4751e-05 lr: 6.4751e-05 eta: 4:52:03 time: 0.4527 data_time: 0.0227 memory: 15873 grad_norm: 16.1752 loss: 9.2131 decode.loss_cls_ce: 2.0243 decode.loss_mask_ce: 0.8903 decode.loss_mask_dice: 1.7001 decode.d7.loss_cls_ce: 2.0220 decode.d7.loss_mask_ce: 0.8843 decode.d7.loss_mask_dice: 1.6920 2023/09/06 17:11:53 - mmengine - INFO - Iter(train) [21200/60000] base_lr: 6.4668e-05 lr: 6.4668e-05 eta: 4:51:41 time: 0.4533 data_time: 0.0234 memory: 15870 grad_norm: 15.6891 loss: 9.6581 decode.loss_cls_ce: 2.1615 decode.loss_mask_ce: 0.9234 decode.loss_mask_dice: 1.7288 decode.d7.loss_cls_ce: 2.1586 decode.d7.loss_mask_ce: 0.9390 decode.d7.loss_mask_dice: 1.7468 2023/09/06 17:12:15 - mmengine - INFO - Iter(train) [21250/60000] base_lr: 6.4584e-05 lr: 6.4584e-05 eta: 4:51:18 time: 0.4486 data_time: 0.0238 memory: 15963 grad_norm: 17.5453 loss: 9.5387 decode.loss_cls_ce: 2.1480 decode.loss_mask_ce: 0.8672 decode.loss_mask_dice: 1.7382 decode.d7.loss_cls_ce: 2.1653 decode.d7.loss_mask_ce: 0.8734 decode.d7.loss_mask_dice: 1.7467 2023/09/06 17:12:38 - mmengine - INFO - Iter(train) [21300/60000] base_lr: 6.4501e-05 lr: 6.4501e-05 eta: 4:50:56 time: 0.4507 data_time: 0.0232 memory: 15955 grad_norm: 16.1474 loss: 9.6933 decode.loss_cls_ce: 2.2287 decode.loss_mask_ce: 0.8369 decode.loss_mask_dice: 1.7893 decode.d7.loss_cls_ce: 2.2125 decode.d7.loss_mask_ce: 0.8405 decode.d7.loss_mask_dice: 1.7854 2023/09/06 17:13:01 - mmengine - INFO - Iter(train) [21350/60000] base_lr: 6.4418e-05 lr: 6.4418e-05 eta: 4:50:33 time: 0.4510 data_time: 0.0245 memory: 15786 grad_norm: 15.7225 loss: 10.0307 decode.loss_cls_ce: 2.2569 decode.loss_mask_ce: 0.8899 decode.loss_mask_dice: 1.8674 decode.d7.loss_cls_ce: 2.2452 decode.d7.loss_mask_ce: 0.8871 decode.d7.loss_mask_dice: 1.8842 2023/09/06 17:13:23 - mmengine - INFO - Iter(train) [21400/60000] base_lr: 6.4334e-05 lr: 6.4334e-05 eta: 4:50:11 time: 0.4540 data_time: 0.0232 memory: 15910 grad_norm: 17.5064 loss: 9.9031 decode.loss_cls_ce: 2.1689 decode.loss_mask_ce: 0.9227 decode.loss_mask_dice: 1.8607 decode.d7.loss_cls_ce: 2.1480 decode.d7.loss_mask_ce: 0.9344 decode.d7.loss_mask_dice: 1.8684 2023/09/06 17:13:46 - mmengine - INFO - Iter(train) [21450/60000] base_lr: 6.4251e-05 lr: 6.4251e-05 eta: 4:49:48 time: 0.4475 data_time: 0.0237 memory: 15862 grad_norm: 18.9149 loss: 10.4875 decode.loss_cls_ce: 2.2723 decode.loss_mask_ce: 0.9052 decode.loss_mask_dice: 2.0707 decode.d7.loss_cls_ce: 2.2718 decode.d7.loss_mask_ce: 0.9181 decode.d7.loss_mask_dice: 2.0495 2023/09/06 17:14:08 - mmengine - INFO - Iter(train) [21500/60000] base_lr: 6.4168e-05 lr: 6.4168e-05 eta: 4:49:26 time: 0.4514 data_time: 0.0235 memory: 15849 grad_norm: 16.7655 loss: 9.4083 decode.loss_cls_ce: 2.1242 decode.loss_mask_ce: 0.9061 decode.loss_mask_dice: 1.6829 decode.d7.loss_cls_ce: 2.0958 decode.d7.loss_mask_ce: 0.9033 decode.d7.loss_mask_dice: 1.6961 2023/09/06 17:14:31 - mmengine - INFO - Iter(train) [21550/60000] base_lr: 6.4084e-05 lr: 6.4084e-05 eta: 4:49:04 time: 0.4494 data_time: 0.0241 memory: 15806 grad_norm: 17.1035 loss: 10.5011 decode.loss_cls_ce: 2.2622 decode.loss_mask_ce: 0.9999 decode.loss_mask_dice: 1.9761 decode.d7.loss_cls_ce: 2.2893 decode.d7.loss_mask_ce: 0.9972 decode.d7.loss_mask_dice: 1.9764 2023/09/06 17:14:54 - mmengine - INFO - Iter(train) [21600/60000] base_lr: 6.4001e-05 lr: 6.4001e-05 eta: 4:48:41 time: 0.4522 data_time: 0.0242 memory: 15833 grad_norm: 16.8646 loss: 8.6877 decode.loss_cls_ce: 1.9043 decode.loss_mask_ce: 0.8902 decode.loss_mask_dice: 1.5471 decode.d7.loss_cls_ce: 1.9097 decode.d7.loss_mask_ce: 0.8863 decode.d7.loss_mask_dice: 1.5502 2023/09/06 17:15:16 - mmengine - INFO - Iter(train) [21650/60000] base_lr: 6.3918e-05 lr: 6.3918e-05 eta: 4:48:18 time: 0.4484 data_time: 0.0237 memory: 15833 grad_norm: 17.3014 loss: 9.6521 decode.loss_cls_ce: 2.1680 decode.loss_mask_ce: 0.9382 decode.loss_mask_dice: 1.7027 decode.d7.loss_cls_ce: 2.1952 decode.d7.loss_mask_ce: 0.9441 decode.d7.loss_mask_dice: 1.7040 2023/09/06 17:15:38 - mmengine - INFO - Iter(train) [21700/60000] base_lr: 6.3834e-05 lr: 6.3834e-05 eta: 4:47:55 time: 0.4500 data_time: 0.0238 memory: 16015 grad_norm: 18.2346 loss: 11.5657 decode.loss_cls_ce: 2.5350 decode.loss_mask_ce: 0.9941 decode.loss_mask_dice: 2.2430 decode.d7.loss_cls_ce: 2.5638 decode.d7.loss_mask_ce: 0.9895 decode.d7.loss_mask_dice: 2.2403 2023/09/06 17:16:01 - mmengine - INFO - Iter(train) [21750/60000] base_lr: 6.3751e-05 lr: 6.3751e-05 eta: 4:47:33 time: 0.4470 data_time: 0.0235 memory: 15858 grad_norm: 16.7840 loss: 10.1498 decode.loss_cls_ce: 2.3408 decode.loss_mask_ce: 0.9367 decode.loss_mask_dice: 1.7995 decode.d7.loss_cls_ce: 2.3223 decode.d7.loss_mask_ce: 0.9419 decode.d7.loss_mask_dice: 1.8086 2023/09/06 17:16:24 - mmengine - INFO - Iter(train) [21800/60000] base_lr: 6.3668e-05 lr: 6.3668e-05 eta: 4:47:11 time: 0.4597 data_time: 0.0231 memory: 15872 grad_norm: 16.3066 loss: 8.5318 decode.loss_cls_ce: 1.8932 decode.loss_mask_ce: 0.8259 decode.loss_mask_dice: 1.5484 decode.d7.loss_cls_ce: 1.8793 decode.d7.loss_mask_ce: 0.8247 decode.d7.loss_mask_dice: 1.5602 2023/09/06 17:16:46 - mmengine - INFO - Iter(train) [21850/60000] base_lr: 6.3584e-05 lr: 6.3584e-05 eta: 4:46:48 time: 0.4496 data_time: 0.0250 memory: 15820 grad_norm: 16.2650 loss: 9.3243 decode.loss_cls_ce: 2.0127 decode.loss_mask_ce: 0.8730 decode.loss_mask_dice: 1.7711 decode.d7.loss_cls_ce: 1.9912 decode.d7.loss_mask_ce: 0.8913 decode.d7.loss_mask_dice: 1.7850 2023/09/06 17:17:09 - mmengine - INFO - Iter(train) [21900/60000] base_lr: 6.3501e-05 lr: 6.3501e-05 eta: 4:46:26 time: 0.4578 data_time: 0.0237 memory: 15833 grad_norm: 15.3762 loss: 9.3451 decode.loss_cls_ce: 2.0860 decode.loss_mask_ce: 0.8661 decode.loss_mask_dice: 1.7091 decode.d7.loss_cls_ce: 2.1080 decode.d7.loss_mask_ce: 0.8634 decode.d7.loss_mask_dice: 1.7126 2023/09/06 17:17:32 - mmengine - INFO - Iter(train) [21950/60000] base_lr: 6.3418e-05 lr: 6.3418e-05 eta: 4:46:03 time: 0.4522 data_time: 0.0243 memory: 15933 grad_norm: 17.4907 loss: 9.5058 decode.loss_cls_ce: 2.1564 decode.loss_mask_ce: 0.8866 decode.loss_mask_dice: 1.7135 decode.d7.loss_cls_ce: 2.1341 decode.d7.loss_mask_ce: 0.8887 decode.d7.loss_mask_dice: 1.7265 2023/09/06 17:17:54 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:17:54 - mmengine - INFO - Iter(train) [22000/60000] base_lr: 6.3334e-05 lr: 6.3334e-05 eta: 4:45:40 time: 0.4479 data_time: 0.0234 memory: 15883 grad_norm: 16.8011 loss: 8.9852 decode.loss_cls_ce: 1.9732 decode.loss_mask_ce: 0.8796 decode.loss_mask_dice: 1.6439 decode.d7.loss_cls_ce: 1.9661 decode.d7.loss_mask_ce: 0.8833 decode.d7.loss_mask_dice: 1.6390 2023/09/06 17:18:17 - mmengine - INFO - Iter(train) [22050/60000] base_lr: 6.3251e-05 lr: 6.3251e-05 eta: 4:45:18 time: 0.4512 data_time: 0.0239 memory: 15898 grad_norm: 17.0822 loss: 9.9573 decode.loss_cls_ce: 2.2798 decode.loss_mask_ce: 0.9472 decode.loss_mask_dice: 1.7362 decode.d7.loss_cls_ce: 2.3025 decode.d7.loss_mask_ce: 0.9469 decode.d7.loss_mask_dice: 1.7447 2023/09/06 17:18:39 - mmengine - INFO - Iter(train) [22100/60000] base_lr: 6.3168e-05 lr: 6.3168e-05 eta: 4:44:55 time: 0.4474 data_time: 0.0236 memory: 15900 grad_norm: 15.5420 loss: 9.4679 decode.loss_cls_ce: 2.0590 decode.loss_mask_ce: 0.8836 decode.loss_mask_dice: 1.7988 decode.d7.loss_cls_ce: 2.0527 decode.d7.loss_mask_ce: 0.8848 decode.d7.loss_mask_dice: 1.7890 2023/09/06 17:19:02 - mmengine - INFO - Iter(train) [22150/60000] base_lr: 6.3084e-05 lr: 6.3084e-05 eta: 4:44:33 time: 0.4525 data_time: 0.0232 memory: 15809 grad_norm: 18.3381 loss: 10.5321 decode.loss_cls_ce: 2.3579 decode.loss_mask_ce: 0.9593 decode.loss_mask_dice: 1.9606 decode.d7.loss_cls_ce: 2.3209 decode.d7.loss_mask_ce: 0.9630 decode.d7.loss_mask_dice: 1.9704 2023/09/06 17:19:24 - mmengine - INFO - Iter(train) [22200/60000] base_lr: 6.3001e-05 lr: 6.3001e-05 eta: 4:44:10 time: 0.4577 data_time: 0.0230 memory: 15912 grad_norm: 16.1843 loss: 10.0042 decode.loss_cls_ce: 2.2278 decode.loss_mask_ce: 0.9561 decode.loss_mask_dice: 1.8094 decode.d7.loss_cls_ce: 2.2573 decode.d7.loss_mask_ce: 0.9453 decode.d7.loss_mask_dice: 1.8083 2023/09/06 17:19:47 - mmengine - INFO - Iter(train) [22250/60000] base_lr: 6.2918e-05 lr: 6.2918e-05 eta: 4:43:48 time: 0.4489 data_time: 0.0232 memory: 15770 grad_norm: 19.6893 loss: 9.8050 decode.loss_cls_ce: 2.0929 decode.loss_mask_ce: 0.9367 decode.loss_mask_dice: 1.8672 decode.d7.loss_cls_ce: 2.1083 decode.d7.loss_mask_ce: 0.9339 decode.d7.loss_mask_dice: 1.8660 2023/09/06 17:20:10 - mmengine - INFO - Iter(train) [22300/60000] base_lr: 6.2834e-05 lr: 6.2834e-05 eta: 4:43:26 time: 0.4493 data_time: 0.0234 memory: 15935 grad_norm: 19.0808 loss: 10.1949 decode.loss_cls_ce: 2.3029 decode.loss_mask_ce: 0.8980 decode.loss_mask_dice: 1.8870 decode.d7.loss_cls_ce: 2.3117 decode.d7.loss_mask_ce: 0.8936 decode.d7.loss_mask_dice: 1.9017 2023/09/06 17:20:32 - mmengine - INFO - Iter(train) [22350/60000] base_lr: 6.2751e-05 lr: 6.2751e-05 eta: 4:43:03 time: 0.4503 data_time: 0.0239 memory: 15845 grad_norm: 17.7178 loss: 9.5953 decode.loss_cls_ce: 2.1198 decode.loss_mask_ce: 0.9061 decode.loss_mask_dice: 1.7647 decode.d7.loss_cls_ce: 2.1295 decode.d7.loss_mask_ce: 0.9027 decode.d7.loss_mask_dice: 1.7726 2023/09/06 17:20:55 - mmengine - INFO - Iter(train) [22400/60000] base_lr: 6.2668e-05 lr: 6.2668e-05 eta: 4:42:40 time: 0.4494 data_time: 0.0232 memory: 15795 grad_norm: 14.9404 loss: 9.0495 decode.loss_cls_ce: 2.0179 decode.loss_mask_ce: 0.8563 decode.loss_mask_dice: 1.6554 decode.d7.loss_cls_ce: 1.9947 decode.d7.loss_mask_ce: 0.8613 decode.d7.loss_mask_dice: 1.6638 2023/09/06 17:21:17 - mmengine - INFO - Iter(train) [22450/60000] base_lr: 6.2584e-05 lr: 6.2584e-05 eta: 4:42:18 time: 0.4527 data_time: 0.0232 memory: 15975 grad_norm: 16.5184 loss: 10.2293 decode.loss_cls_ce: 2.3467 decode.loss_mask_ce: 0.9403 decode.loss_mask_dice: 1.8253 decode.d7.loss_cls_ce: 2.3319 decode.d7.loss_mask_ce: 0.9443 decode.d7.loss_mask_dice: 1.8409 2023/09/06 17:21:40 - mmengine - INFO - Iter(train) [22500/60000] base_lr: 6.2501e-05 lr: 6.2501e-05 eta: 4:41:56 time: 0.4482 data_time: 0.0238 memory: 15951 grad_norm: 16.9444 loss: 8.7798 decode.loss_cls_ce: 1.9842 decode.loss_mask_ce: 0.8239 decode.loss_mask_dice: 1.5821 decode.d7.loss_cls_ce: 1.9751 decode.d7.loss_mask_ce: 0.8215 decode.d7.loss_mask_dice: 1.5931 2023/09/06 17:22:02 - mmengine - INFO - Iter(train) [22550/60000] base_lr: 6.2418e-05 lr: 6.2418e-05 eta: 4:41:33 time: 0.4489 data_time: 0.0234 memory: 15942 grad_norm: 16.9949 loss: 9.2603 decode.loss_cls_ce: 2.0205 decode.loss_mask_ce: 0.8276 decode.loss_mask_dice: 1.7716 decode.d7.loss_cls_ce: 2.0244 decode.d7.loss_mask_ce: 0.8336 decode.d7.loss_mask_dice: 1.7826 2023/09/06 17:22:25 - mmengine - INFO - Iter(train) [22600/60000] base_lr: 6.2334e-05 lr: 6.2334e-05 eta: 4:41:10 time: 0.4479 data_time: 0.0233 memory: 15822 grad_norm: 17.7739 loss: 10.4202 decode.loss_cls_ce: 2.2589 decode.loss_mask_ce: 0.9575 decode.loss_mask_dice: 1.9828 decode.d7.loss_cls_ce: 2.2985 decode.d7.loss_mask_ce: 0.9535 decode.d7.loss_mask_dice: 1.9691 2023/09/06 17:22:48 - mmengine - INFO - Iter(train) [22650/60000] base_lr: 6.2251e-05 lr: 6.2251e-05 eta: 4:40:48 time: 0.4499 data_time: 0.0235 memory: 15732 grad_norm: 16.0402 loss: 9.0910 decode.loss_cls_ce: 1.9913 decode.loss_mask_ce: 0.9223 decode.loss_mask_dice: 1.6294 decode.d7.loss_cls_ce: 1.9927 decode.d7.loss_mask_ce: 0.9212 decode.d7.loss_mask_dice: 1.6341 2023/09/06 17:23:10 - mmengine - INFO - Iter(train) [22700/60000] base_lr: 6.2168e-05 lr: 6.2168e-05 eta: 4:40:25 time: 0.4481 data_time: 0.0233 memory: 15899 grad_norm: 17.7843 loss: 10.4894 decode.loss_cls_ce: 2.2258 decode.loss_mask_ce: 1.0515 decode.loss_mask_dice: 1.9574 decode.d7.loss_cls_ce: 2.2487 decode.d7.loss_mask_ce: 1.0511 decode.d7.loss_mask_dice: 1.9549 2023/09/06 17:23:33 - mmengine - INFO - Iter(train) [22750/60000] base_lr: 6.2084e-05 lr: 6.2084e-05 eta: 4:40:02 time: 0.4472 data_time: 0.0232 memory: 15845 grad_norm: 17.3858 loss: 9.8373 decode.loss_cls_ce: 2.1505 decode.loss_mask_ce: 0.9280 decode.loss_mask_dice: 1.8371 decode.d7.loss_cls_ce: 2.1340 decode.d7.loss_mask_ce: 0.9326 decode.d7.loss_mask_dice: 1.8550 2023/09/06 17:23:55 - mmengine - INFO - Iter(train) [22800/60000] base_lr: 6.2001e-05 lr: 6.2001e-05 eta: 4:39:40 time: 0.4494 data_time: 0.0236 memory: 16183 grad_norm: 16.4196 loss: 10.0720 decode.loss_cls_ce: 2.2016 decode.loss_mask_ce: 0.8922 decode.loss_mask_dice: 1.9236 decode.d7.loss_cls_ce: 2.2262 decode.d7.loss_mask_ce: 0.9016 decode.d7.loss_mask_dice: 1.9267 2023/09/06 17:24:18 - mmengine - INFO - Iter(train) [22850/60000] base_lr: 6.1918e-05 lr: 6.1918e-05 eta: 4:39:17 time: 0.4499 data_time: 0.0234 memory: 16065 grad_norm: 18.2932 loss: 9.3930 decode.loss_cls_ce: 2.1374 decode.loss_mask_ce: 0.8829 decode.loss_mask_dice: 1.6693 decode.d7.loss_cls_ce: 2.1572 decode.d7.loss_mask_ce: 0.8828 decode.d7.loss_mask_dice: 1.6634 2023/09/06 17:24:40 - mmengine - INFO - Iter(train) [22900/60000] base_lr: 6.1834e-05 lr: 6.1834e-05 eta: 4:38:54 time: 0.4481 data_time: 0.0241 memory: 15809 grad_norm: 16.2152 loss: 8.9376 decode.loss_cls_ce: 1.9716 decode.loss_mask_ce: 0.8519 decode.loss_mask_dice: 1.6447 decode.d7.loss_cls_ce: 1.9770 decode.d7.loss_mask_ce: 0.8526 decode.d7.loss_mask_dice: 1.6397 2023/09/06 17:25:03 - mmengine - INFO - Iter(train) [22950/60000] base_lr: 6.1751e-05 lr: 6.1751e-05 eta: 4:38:32 time: 0.4555 data_time: 0.0229 memory: 15949 grad_norm: 16.4061 loss: 10.2424 decode.loss_cls_ce: 2.3100 decode.loss_mask_ce: 0.9425 decode.loss_mask_dice: 1.8582 decode.d7.loss_cls_ce: 2.3092 decode.d7.loss_mask_ce: 0.9493 decode.d7.loss_mask_dice: 1.8733 2023/09/06 17:25:25 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:25:25 - mmengine - INFO - Iter(train) [23000/60000] base_lr: 6.1668e-05 lr: 6.1668e-05 eta: 4:38:09 time: 0.4479 data_time: 0.0234 memory: 15846 grad_norm: 17.5284 loss: 8.5392 decode.loss_cls_ce: 1.9211 decode.loss_mask_ce: 0.8209 decode.loss_mask_dice: 1.5146 decode.d7.loss_cls_ce: 1.9468 decode.d7.loss_mask_ce: 0.8255 decode.d7.loss_mask_dice: 1.5103 2023/09/06 17:25:48 - mmengine - INFO - Iter(train) [23050/60000] base_lr: 6.1584e-05 lr: 6.1584e-05 eta: 4:37:47 time: 0.4558 data_time: 0.0224 memory: 15909 grad_norm: 17.9554 loss: 10.2602 decode.loss_cls_ce: 2.1412 decode.loss_mask_ce: 1.0279 decode.loss_mask_dice: 1.9473 decode.d7.loss_cls_ce: 2.1690 decode.d7.loss_mask_ce: 1.0237 decode.d7.loss_mask_dice: 1.9512 2023/09/06 17:26:11 - mmengine - INFO - Iter(train) [23100/60000] base_lr: 6.1501e-05 lr: 6.1501e-05 eta: 4:37:25 time: 0.4550 data_time: 0.0235 memory: 15923 grad_norm: 15.1162 loss: 10.0088 decode.loss_cls_ce: 2.1268 decode.loss_mask_ce: 0.9672 decode.loss_mask_dice: 1.9146 decode.d7.loss_cls_ce: 2.1177 decode.d7.loss_mask_ce: 0.9627 decode.d7.loss_mask_dice: 1.9198 2023/09/06 17:26:33 - mmengine - INFO - Iter(train) [23150/60000] base_lr: 6.1418e-05 lr: 6.1418e-05 eta: 4:37:02 time: 0.4485 data_time: 0.0232 memory: 15834 grad_norm: 17.8791 loss: 9.4797 decode.loss_cls_ce: 2.1658 decode.loss_mask_ce: 0.8739 decode.loss_mask_dice: 1.7096 decode.d7.loss_cls_ce: 2.1447 decode.d7.loss_mask_ce: 0.8659 decode.d7.loss_mask_dice: 1.7197 2023/09/06 17:26:56 - mmengine - INFO - Iter(train) [23200/60000] base_lr: 6.1334e-05 lr: 6.1334e-05 eta: 4:36:39 time: 0.4512 data_time: 0.0238 memory: 15989 grad_norm: 16.0628 loss: 9.6141 decode.loss_cls_ce: 2.1588 decode.loss_mask_ce: 0.9316 decode.loss_mask_dice: 1.7353 decode.d7.loss_cls_ce: 2.1254 decode.d7.loss_mask_ce: 0.9363 decode.d7.loss_mask_dice: 1.7267 2023/09/06 17:27:18 - mmengine - INFO - Iter(train) [23250/60000] base_lr: 6.1251e-05 lr: 6.1251e-05 eta: 4:36:17 time: 0.4497 data_time: 0.0238 memory: 15872 grad_norm: 19.7140 loss: 9.4366 decode.loss_cls_ce: 1.9963 decode.loss_mask_ce: 0.9627 decode.loss_mask_dice: 1.7639 decode.d7.loss_cls_ce: 1.9795 decode.d7.loss_mask_ce: 0.9568 decode.d7.loss_mask_dice: 1.7772 2023/09/06 17:27:41 - mmengine - INFO - Iter(train) [23300/60000] base_lr: 6.1168e-05 lr: 6.1168e-05 eta: 4:35:54 time: 0.4562 data_time: 0.0233 memory: 15796 grad_norm: 15.5242 loss: 9.9951 decode.loss_cls_ce: 2.0641 decode.loss_mask_ce: 0.9471 decode.loss_mask_dice: 1.9918 decode.d7.loss_cls_ce: 2.0516 decode.d7.loss_mask_ce: 0.9459 decode.d7.loss_mask_dice: 1.9946 2023/09/06 17:28:03 - mmengine - INFO - Iter(train) [23350/60000] base_lr: 6.1084e-05 lr: 6.1084e-05 eta: 4:35:32 time: 0.4523 data_time: 0.0235 memory: 15859 grad_norm: 18.0793 loss: 9.1549 decode.loss_cls_ce: 2.0428 decode.loss_mask_ce: 0.8381 decode.loss_mask_dice: 1.6724 decode.d7.loss_cls_ce: 2.0748 decode.d7.loss_mask_ce: 0.8480 decode.d7.loss_mask_dice: 1.6789 2023/09/06 17:28:26 - mmengine - INFO - Iter(train) [23400/60000] base_lr: 6.1001e-05 lr: 6.1001e-05 eta: 4:35:09 time: 0.4561 data_time: 0.0228 memory: 15730 grad_norm: 16.9455 loss: 9.2726 decode.loss_cls_ce: 2.1102 decode.loss_mask_ce: 0.8334 decode.loss_mask_dice: 1.6893 decode.d7.loss_cls_ce: 2.1018 decode.d7.loss_mask_ce: 0.8437 decode.d7.loss_mask_dice: 1.6942 2023/09/06 17:28:48 - mmengine - INFO - Iter(train) [23450/60000] base_lr: 6.0918e-05 lr: 6.0918e-05 eta: 4:34:46 time: 0.4500 data_time: 0.0243 memory: 15795 grad_norm: 16.7241 loss: 8.5323 decode.loss_cls_ce: 1.9669 decode.loss_mask_ce: 0.7702 decode.loss_mask_dice: 1.5243 decode.d7.loss_cls_ce: 1.9533 decode.d7.loss_mask_ce: 0.7857 decode.d7.loss_mask_dice: 1.5320 2023/09/06 17:29:11 - mmengine - INFO - Iter(train) [23500/60000] base_lr: 6.0834e-05 lr: 6.0834e-05 eta: 4:34:24 time: 0.4496 data_time: 0.0239 memory: 15759 grad_norm: 14.7284 loss: 9.5848 decode.loss_cls_ce: 2.1721 decode.loss_mask_ce: 0.9178 decode.loss_mask_dice: 1.7096 decode.d7.loss_cls_ce: 2.1590 decode.d7.loss_mask_ce: 0.9190 decode.d7.loss_mask_dice: 1.7073 2023/09/06 17:29:33 - mmengine - INFO - Iter(train) [23550/60000] base_lr: 6.0751e-05 lr: 6.0751e-05 eta: 4:34:01 time: 0.4492 data_time: 0.0236 memory: 15965 grad_norm: 15.1340 loss: 9.4502 decode.loss_cls_ce: 2.0688 decode.loss_mask_ce: 0.8986 decode.loss_mask_dice: 1.7646 decode.d7.loss_cls_ce: 2.0606 decode.d7.loss_mask_ce: 0.8940 decode.d7.loss_mask_dice: 1.7636 2023/09/06 17:29:56 - mmengine - INFO - Iter(train) [23600/60000] base_lr: 6.0668e-05 lr: 6.0668e-05 eta: 4:33:38 time: 0.4539 data_time: 0.0234 memory: 15926 grad_norm: 16.6503 loss: 10.1661 decode.loss_cls_ce: 2.1909 decode.loss_mask_ce: 1.0103 decode.loss_mask_dice: 1.8785 decode.d7.loss_cls_ce: 2.2002 decode.d7.loss_mask_ce: 1.0123 decode.d7.loss_mask_dice: 1.8738 2023/09/06 17:30:18 - mmengine - INFO - Iter(train) [23650/60000] base_lr: 6.0584e-05 lr: 6.0584e-05 eta: 4:33:16 time: 0.4481 data_time: 0.0241 memory: 15923 grad_norm: 16.8082 loss: 9.4666 decode.loss_cls_ce: 2.0096 decode.loss_mask_ce: 0.9143 decode.loss_mask_dice: 1.7894 decode.d7.loss_cls_ce: 2.0358 decode.d7.loss_mask_ce: 0.9180 decode.d7.loss_mask_dice: 1.7995 2023/09/06 17:30:41 - mmengine - INFO - Iter(train) [23700/60000] base_lr: 6.0501e-05 lr: 6.0501e-05 eta: 4:32:54 time: 0.4575 data_time: 0.0231 memory: 15860 grad_norm: 17.6502 loss: 9.7733 decode.loss_cls_ce: 2.1497 decode.loss_mask_ce: 0.9835 decode.loss_mask_dice: 1.7606 decode.d7.loss_cls_ce: 2.1445 decode.d7.loss_mask_ce: 0.9816 decode.d7.loss_mask_dice: 1.7534 2023/09/06 17:31:04 - mmengine - INFO - Iter(train) [23750/60000] base_lr: 6.0418e-05 lr: 6.0418e-05 eta: 4:32:31 time: 0.4474 data_time: 0.0232 memory: 15836 grad_norm: 16.3025 loss: 9.6990 decode.loss_cls_ce: 2.1440 decode.loss_mask_ce: 0.9083 decode.loss_mask_dice: 1.7894 decode.d7.loss_cls_ce: 2.1713 decode.d7.loss_mask_ce: 0.8990 decode.d7.loss_mask_dice: 1.7869 2023/09/06 17:31:26 - mmengine - INFO - Iter(train) [23800/60000] base_lr: 6.0334e-05 lr: 6.0334e-05 eta: 4:32:09 time: 0.4557 data_time: 0.0221 memory: 15772 grad_norm: 20.1858 loss: 10.0985 decode.loss_cls_ce: 2.1650 decode.loss_mask_ce: 0.9845 decode.loss_mask_dice: 1.9127 decode.d7.loss_cls_ce: 2.1291 decode.d7.loss_mask_ce: 0.9818 decode.d7.loss_mask_dice: 1.9254 2023/09/06 17:31:49 - mmengine - INFO - Iter(train) [23850/60000] base_lr: 6.0251e-05 lr: 6.0251e-05 eta: 4:31:46 time: 0.4571 data_time: 0.0220 memory: 16131 grad_norm: 18.4139 loss: 11.0700 decode.loss_cls_ce: 2.3225 decode.loss_mask_ce: 1.0650 decode.loss_mask_dice: 2.1381 decode.d7.loss_cls_ce: 2.3453 decode.d7.loss_mask_ce: 1.0641 decode.d7.loss_mask_dice: 2.1351 2023/09/06 17:32:12 - mmengine - INFO - Iter(train) [23900/60000] base_lr: 6.0168e-05 lr: 6.0168e-05 eta: 4:31:24 time: 0.4570 data_time: 0.0229 memory: 16029 grad_norm: 16.0881 loss: 9.8081 decode.loss_cls_ce: 2.1811 decode.loss_mask_ce: 0.9399 decode.loss_mask_dice: 1.7700 decode.d7.loss_cls_ce: 2.1861 decode.d7.loss_mask_ce: 0.9483 decode.d7.loss_mask_dice: 1.7827 2023/09/06 17:32:34 - mmengine - INFO - Iter(train) [23950/60000] base_lr: 6.0084e-05 lr: 6.0084e-05 eta: 4:31:01 time: 0.4498 data_time: 0.0235 memory: 15794 grad_norm: 16.1854 loss: 8.8742 decode.loss_cls_ce: 1.9765 decode.loss_mask_ce: 0.8003 decode.loss_mask_dice: 1.6742 decode.d7.loss_cls_ce: 1.9697 decode.d7.loss_mask_ce: 0.8006 decode.d7.loss_mask_dice: 1.6528 2023/09/06 17:32:57 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:32:57 - mmengine - INFO - Iter(train) [24000/60000] base_lr: 6.0001e-05 lr: 6.0001e-05 eta: 4:30:39 time: 0.4497 data_time: 0.0235 memory: 15833 grad_norm: 20.7573 loss: 8.7612 decode.loss_cls_ce: 2.0466 decode.loss_mask_ce: 0.7710 decode.loss_mask_dice: 1.5744 decode.d7.loss_cls_ce: 2.0456 decode.d7.loss_mask_ce: 0.7705 decode.d7.loss_mask_dice: 1.5531 2023/09/06 17:33:19 - mmengine - INFO - Iter(train) [24050/60000] base_lr: 5.9918e-05 lr: 5.9918e-05 eta: 4:30:16 time: 0.4507 data_time: 0.0239 memory: 15756 grad_norm: 18.2176 loss: 9.9496 decode.loss_cls_ce: 2.0939 decode.loss_mask_ce: 1.0245 decode.loss_mask_dice: 1.8539 decode.d7.loss_cls_ce: 2.0989 decode.d7.loss_mask_ce: 1.0182 decode.d7.loss_mask_dice: 1.8601 2023/09/06 17:33:42 - mmengine - INFO - Iter(train) [24100/60000] base_lr: 5.9834e-05 lr: 5.9834e-05 eta: 4:29:53 time: 0.4500 data_time: 0.0236 memory: 16157 grad_norm: 19.6953 loss: 9.2907 decode.loss_cls_ce: 2.0669 decode.loss_mask_ce: 0.8291 decode.loss_mask_dice: 1.7337 decode.d7.loss_cls_ce: 2.0895 decode.d7.loss_mask_ce: 0.8334 decode.d7.loss_mask_dice: 1.7381 2023/09/06 17:34:04 - mmengine - INFO - Iter(train) [24150/60000] base_lr: 5.9751e-05 lr: 5.9751e-05 eta: 4:29:31 time: 0.4510 data_time: 0.0241 memory: 15801 grad_norm: 19.5124 loss: 9.0526 decode.loss_cls_ce: 2.0400 decode.loss_mask_ce: 0.8979 decode.loss_mask_dice: 1.5970 decode.d7.loss_cls_ce: 2.0098 decode.d7.loss_mask_ce: 0.8979 decode.d7.loss_mask_dice: 1.6100 2023/09/06 17:34:27 - mmengine - INFO - Iter(train) [24200/60000] base_lr: 5.9668e-05 lr: 5.9668e-05 eta: 4:29:08 time: 0.4487 data_time: 0.0234 memory: 15850 grad_norm: 16.3658 loss: 9.0918 decode.loss_cls_ce: 2.0852 decode.loss_mask_ce: 0.8241 decode.loss_mask_dice: 1.6490 decode.d7.loss_cls_ce: 2.0705 decode.d7.loss_mask_ce: 0.8259 decode.d7.loss_mask_dice: 1.6370 2023/09/06 17:34:49 - mmengine - INFO - Iter(train) [24250/60000] base_lr: 5.9584e-05 lr: 5.9584e-05 eta: 4:28:45 time: 0.4501 data_time: 0.0241 memory: 15807 grad_norm: 16.3682 loss: 9.7703 decode.loss_cls_ce: 2.1813 decode.loss_mask_ce: 0.9109 decode.loss_mask_dice: 1.8087 decode.d7.loss_cls_ce: 2.1559 decode.d7.loss_mask_ce: 0.9144 decode.d7.loss_mask_dice: 1.7990 2023/09/06 17:35:12 - mmengine - INFO - Iter(train) [24300/60000] base_lr: 5.9501e-05 lr: 5.9501e-05 eta: 4:28:23 time: 0.4588 data_time: 0.0232 memory: 16069 grad_norm: 16.1429 loss: 9.3538 decode.loss_cls_ce: 2.1342 decode.loss_mask_ce: 0.8509 decode.loss_mask_dice: 1.6877 decode.d7.loss_cls_ce: 2.1226 decode.d7.loss_mask_ce: 0.8518 decode.d7.loss_mask_dice: 1.7064 2023/09/06 17:35:34 - mmengine - INFO - Iter(train) [24350/60000] base_lr: 5.9418e-05 lr: 5.9418e-05 eta: 4:28:00 time: 0.4478 data_time: 0.0235 memory: 15950 grad_norm: 16.2294 loss: 9.2903 decode.loss_cls_ce: 2.0397 decode.loss_mask_ce: 0.8459 decode.loss_mask_dice: 1.7546 decode.d7.loss_cls_ce: 2.0437 decode.d7.loss_mask_ce: 0.8415 decode.d7.loss_mask_dice: 1.7649 2023/09/06 17:35:57 - mmengine - INFO - Iter(train) [24400/60000] base_lr: 5.9334e-05 lr: 5.9334e-05 eta: 4:27:38 time: 0.4500 data_time: 0.0245 memory: 15913 grad_norm: 16.2123 loss: 8.7368 decode.loss_cls_ce: 1.8967 decode.loss_mask_ce: 0.8814 decode.loss_mask_dice: 1.5923 decode.d7.loss_cls_ce: 1.8812 decode.d7.loss_mask_ce: 0.8858 decode.d7.loss_mask_dice: 1.5993 2023/09/06 17:36:19 - mmengine - INFO - Iter(train) [24450/60000] base_lr: 5.9251e-05 lr: 5.9251e-05 eta: 4:27:15 time: 0.4489 data_time: 0.0236 memory: 15784 grad_norm: 16.0868 loss: 8.5996 decode.loss_cls_ce: 1.7232 decode.loss_mask_ce: 0.9431 decode.loss_mask_dice: 1.6273 decode.d7.loss_cls_ce: 1.7583 decode.d7.loss_mask_ce: 0.9331 decode.d7.loss_mask_dice: 1.6146 2023/09/06 17:36:42 - mmengine - INFO - Iter(train) [24500/60000] base_lr: 5.9168e-05 lr: 5.9168e-05 eta: 4:26:52 time: 0.4531 data_time: 0.0237 memory: 15769 grad_norm: 18.0519 loss: 10.4794 decode.loss_cls_ce: 2.1633 decode.loss_mask_ce: 1.0370 decode.loss_mask_dice: 2.0484 decode.d7.loss_cls_ce: 2.1540 decode.d7.loss_mask_ce: 1.0391 decode.d7.loss_mask_dice: 2.0377 2023/09/06 17:37:04 - mmengine - INFO - Iter(train) [24550/60000] base_lr: 5.9084e-05 lr: 5.9084e-05 eta: 4:26:30 time: 0.4489 data_time: 0.0240 memory: 15823 grad_norm: 17.3955 loss: 8.9311 decode.loss_cls_ce: 1.9147 decode.loss_mask_ce: 0.9131 decode.loss_mask_dice: 1.6401 decode.d7.loss_cls_ce: 1.9161 decode.d7.loss_mask_ce: 0.9100 decode.d7.loss_mask_dice: 1.6370 2023/09/06 17:37:27 - mmengine - INFO - Iter(train) [24600/60000] base_lr: 5.9001e-05 lr: 5.9001e-05 eta: 4:26:08 time: 0.4581 data_time: 0.0226 memory: 15783 grad_norm: 16.6176 loss: 10.4486 decode.loss_cls_ce: 2.3348 decode.loss_mask_ce: 0.9397 decode.loss_mask_dice: 1.9461 decode.d7.loss_cls_ce: 2.3572 decode.d7.loss_mask_ce: 0.9400 decode.d7.loss_mask_dice: 1.9309 2023/09/06 17:37:50 - mmengine - INFO - Iter(train) [24650/60000] base_lr: 5.8918e-05 lr: 5.8918e-05 eta: 4:25:45 time: 0.4530 data_time: 0.0233 memory: 15850 grad_norm: 17.3718 loss: 9.9071 decode.loss_cls_ce: 2.1798 decode.loss_mask_ce: 0.9081 decode.loss_mask_dice: 1.8836 decode.d7.loss_cls_ce: 2.1578 decode.d7.loss_mask_ce: 0.9031 decode.d7.loss_mask_dice: 1.8747 2023/09/06 17:38:13 - mmengine - INFO - Iter(train) [24700/60000] base_lr: 5.8834e-05 lr: 5.8834e-05 eta: 4:25:23 time: 0.4505 data_time: 0.0242 memory: 15884 grad_norm: 19.0561 loss: 9.2237 decode.loss_cls_ce: 2.0009 decode.loss_mask_ce: 0.9100 decode.loss_mask_dice: 1.6957 decode.d7.loss_cls_ce: 2.0249 decode.d7.loss_mask_ce: 0.9009 decode.d7.loss_mask_dice: 1.6914 2023/09/06 17:38:35 - mmengine - INFO - Iter(train) [24750/60000] base_lr: 5.8751e-05 lr: 5.8751e-05 eta: 4:25:00 time: 0.4518 data_time: 0.0236 memory: 15733 grad_norm: 18.6380 loss: 8.2381 decode.loss_cls_ce: 1.7489 decode.loss_mask_ce: 0.8703 decode.loss_mask_dice: 1.5091 decode.d7.loss_cls_ce: 1.7424 decode.d7.loss_mask_ce: 0.8614 decode.d7.loss_mask_dice: 1.5060 2023/09/06 17:38:58 - mmengine - INFO - Iter(train) [24800/60000] base_lr: 5.8668e-05 lr: 5.8668e-05 eta: 4:24:38 time: 0.4552 data_time: 0.0229 memory: 15821 grad_norm: 15.3178 loss: 9.3762 decode.loss_cls_ce: 2.1092 decode.loss_mask_ce: 0.8622 decode.loss_mask_dice: 1.7139 decode.d7.loss_cls_ce: 2.0867 decode.d7.loss_mask_ce: 0.8681 decode.d7.loss_mask_dice: 1.7361 2023/09/06 17:39:21 - mmengine - INFO - Iter(train) [24850/60000] base_lr: 5.8584e-05 lr: 5.8584e-05 eta: 4:24:16 time: 0.4560 data_time: 0.0235 memory: 16031 grad_norm: 16.5239 loss: 8.6829 decode.loss_cls_ce: 1.8706 decode.loss_mask_ce: 0.7946 decode.loss_mask_dice: 1.6863 decode.d7.loss_cls_ce: 1.8526 decode.d7.loss_mask_ce: 0.7962 decode.d7.loss_mask_dice: 1.6826 2023/09/06 17:39:43 - mmengine - INFO - Iter(train) [24900/60000] base_lr: 5.8501e-05 lr: 5.8501e-05 eta: 4:23:53 time: 0.4515 data_time: 0.0235 memory: 15859 grad_norm: 16.1842 loss: 9.1984 decode.loss_cls_ce: 1.9818 decode.loss_mask_ce: 0.8810 decode.loss_mask_dice: 1.7194 decode.d7.loss_cls_ce: 2.0119 decode.d7.loss_mask_ce: 0.8830 decode.d7.loss_mask_dice: 1.7212 2023/09/06 17:40:06 - mmengine - INFO - Iter(train) [24950/60000] base_lr: 5.8418e-05 lr: 5.8418e-05 eta: 4:23:31 time: 0.4488 data_time: 0.0236 memory: 15960 grad_norm: 18.5711 loss: 9.9125 decode.loss_cls_ce: 2.2520 decode.loss_mask_ce: 0.8704 decode.loss_mask_dice: 1.8349 decode.d7.loss_cls_ce: 2.2425 decode.d7.loss_mask_ce: 0.8682 decode.d7.loss_mask_dice: 1.8445 2023/09/06 17:40:28 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:40:28 - mmengine - INFO - Iter(train) [25000/60000] base_lr: 5.8334e-05 lr: 5.8334e-05 eta: 4:23:08 time: 0.4489 data_time: 0.0237 memory: 15680 grad_norm: 17.3921 loss: 8.8291 decode.loss_cls_ce: 2.0121 decode.loss_mask_ce: 0.8624 decode.loss_mask_dice: 1.5365 decode.d7.loss_cls_ce: 2.0271 decode.d7.loss_mask_ce: 0.8541 decode.d7.loss_mask_dice: 1.5370 2023/09/06 17:40:51 - mmengine - INFO - Iter(train) [25050/60000] base_lr: 5.8251e-05 lr: 5.8251e-05 eta: 4:22:46 time: 0.4577 data_time: 0.0234 memory: 15833 grad_norm: 17.6725 loss: 9.8792 decode.loss_cls_ce: 2.1720 decode.loss_mask_ce: 0.8925 decode.loss_mask_dice: 1.8906 decode.d7.loss_cls_ce: 2.1441 decode.d7.loss_mask_ce: 0.8997 decode.d7.loss_mask_dice: 1.8803 2023/09/06 17:41:14 - mmengine - INFO - Iter(train) [25100/60000] base_lr: 5.8168e-05 lr: 5.8168e-05 eta: 4:22:24 time: 0.4610 data_time: 0.0237 memory: 15770 grad_norm: 15.6326 loss: 10.3974 decode.loss_cls_ce: 2.2453 decode.loss_mask_ce: 0.9980 decode.loss_mask_dice: 1.9462 decode.d7.loss_cls_ce: 2.2497 decode.d7.loss_mask_ce: 1.0023 decode.d7.loss_mask_dice: 1.9559 2023/09/06 17:41:37 - mmengine - INFO - Iter(train) [25150/60000] base_lr: 5.8084e-05 lr: 5.8084e-05 eta: 4:22:01 time: 0.4516 data_time: 0.0241 memory: 15949 grad_norm: 17.1077 loss: 10.4395 decode.loss_cls_ce: 2.3260 decode.loss_mask_ce: 0.9522 decode.loss_mask_dice: 1.9326 decode.d7.loss_cls_ce: 2.3438 decode.d7.loss_mask_ce: 0.9511 decode.d7.loss_mask_dice: 1.9339 2023/09/06 17:42:00 - mmengine - INFO - Iter(train) [25200/60000] base_lr: 5.8001e-05 lr: 5.8001e-05 eta: 4:21:39 time: 0.4524 data_time: 0.0229 memory: 15964 grad_norm: 15.5967 loss: 10.1021 decode.loss_cls_ce: 2.2523 decode.loss_mask_ce: 0.9186 decode.loss_mask_dice: 1.8772 decode.d7.loss_cls_ce: 2.2609 decode.d7.loss_mask_ce: 0.9156 decode.d7.loss_mask_dice: 1.8775 2023/09/06 17:42:22 - mmengine - INFO - Iter(train) [25250/60000] base_lr: 5.7918e-05 lr: 5.7918e-05 eta: 4:21:17 time: 0.4492 data_time: 0.0248 memory: 15730 grad_norm: 21.0165 loss: 9.4590 decode.loss_cls_ce: 2.0520 decode.loss_mask_ce: 0.9462 decode.loss_mask_dice: 1.7442 decode.d7.loss_cls_ce: 2.0257 decode.d7.loss_mask_ce: 0.9478 decode.d7.loss_mask_dice: 1.7432 2023/09/06 17:42:45 - mmengine - INFO - Iter(train) [25300/60000] base_lr: 5.7834e-05 lr: 5.7834e-05 eta: 4:20:54 time: 0.4480 data_time: 0.0237 memory: 15794 grad_norm: 17.5311 loss: 10.6994 decode.loss_cls_ce: 2.3357 decode.loss_mask_ce: 1.0472 decode.loss_mask_dice: 1.9475 decode.d7.loss_cls_ce: 2.3710 decode.d7.loss_mask_ce: 1.0409 decode.d7.loss_mask_dice: 1.9572 2023/09/06 17:43:08 - mmengine - INFO - Iter(train) [25350/60000] base_lr: 5.7751e-05 lr: 5.7751e-05 eta: 4:20:32 time: 0.4545 data_time: 0.0239 memory: 15938 grad_norm: 18.3987 loss: 8.8810 decode.loss_cls_ce: 2.0030 decode.loss_mask_ce: 0.8966 decode.loss_mask_dice: 1.5495 decode.d7.loss_cls_ce: 2.0020 decode.d7.loss_mask_ce: 0.8941 decode.d7.loss_mask_dice: 1.5358 2023/09/06 17:43:30 - mmengine - INFO - Iter(train) [25400/60000] base_lr: 5.7668e-05 lr: 5.7668e-05 eta: 4:20:09 time: 0.4529 data_time: 0.0242 memory: 15845 grad_norm: 17.3057 loss: 9.0183 decode.loss_cls_ce: 1.9789 decode.loss_mask_ce: 0.8547 decode.loss_mask_dice: 1.6644 decode.d7.loss_cls_ce: 1.9976 decode.d7.loss_mask_ce: 0.8537 decode.d7.loss_mask_dice: 1.6691 2023/09/06 17:43:53 - mmengine - INFO - Iter(train) [25450/60000] base_lr: 5.7584e-05 lr: 5.7584e-05 eta: 4:19:47 time: 0.4491 data_time: 0.0234 memory: 15861 grad_norm: 18.2913 loss: 9.6459 decode.loss_cls_ce: 2.0923 decode.loss_mask_ce: 0.9928 decode.loss_mask_dice: 1.7310 decode.d7.loss_cls_ce: 2.1114 decode.d7.loss_mask_ce: 0.9872 decode.d7.loss_mask_dice: 1.7312 2023/09/06 17:44:15 - mmengine - INFO - Iter(train) [25500/60000] base_lr: 5.7501e-05 lr: 5.7501e-05 eta: 4:19:24 time: 0.4538 data_time: 0.0239 memory: 16106 grad_norm: 16.9376 loss: 9.9503 decode.loss_cls_ce: 2.3252 decode.loss_mask_ce: 0.8633 decode.loss_mask_dice: 1.8030 decode.d7.loss_cls_ce: 2.2778 decode.d7.loss_mask_ce: 0.8682 decode.d7.loss_mask_dice: 1.8128 2023/09/06 17:44:38 - mmengine - INFO - Iter(train) [25550/60000] base_lr: 5.7418e-05 lr: 5.7418e-05 eta: 4:19:02 time: 0.4541 data_time: 0.0239 memory: 15784 grad_norm: 17.8777 loss: 8.7073 decode.loss_cls_ce: 1.9435 decode.loss_mask_ce: 0.8583 decode.loss_mask_dice: 1.5529 decode.d7.loss_cls_ce: 1.9483 decode.d7.loss_mask_ce: 0.8588 decode.d7.loss_mask_dice: 1.5455 2023/09/06 17:45:01 - mmengine - INFO - Iter(train) [25600/60000] base_lr: 5.7334e-05 lr: 5.7334e-05 eta: 4:18:39 time: 0.4546 data_time: 0.0249 memory: 15894 grad_norm: 16.2685 loss: 9.3498 decode.loss_cls_ce: 2.0302 decode.loss_mask_ce: 0.9420 decode.loss_mask_dice: 1.7084 decode.d7.loss_cls_ce: 2.0138 decode.d7.loss_mask_ce: 0.9476 decode.d7.loss_mask_dice: 1.7078 2023/09/06 17:45:23 - mmengine - INFO - Iter(train) [25650/60000] base_lr: 5.7251e-05 lr: 5.7251e-05 eta: 4:18:17 time: 0.4541 data_time: 0.0235 memory: 15857 grad_norm: 16.5088 loss: 9.6038 decode.loss_cls_ce: 2.0762 decode.loss_mask_ce: 0.8776 decode.loss_mask_dice: 1.8320 decode.d7.loss_cls_ce: 2.1034 decode.d7.loss_mask_ce: 0.8800 decode.d7.loss_mask_dice: 1.8347 2023/09/06 17:45:46 - mmengine - INFO - Iter(train) [25700/60000] base_lr: 5.7168e-05 lr: 5.7168e-05 eta: 4:17:55 time: 0.4550 data_time: 0.0235 memory: 15882 grad_norm: 17.1219 loss: 9.4437 decode.loss_cls_ce: 2.1392 decode.loss_mask_ce: 0.8631 decode.loss_mask_dice: 1.7170 decode.d7.loss_cls_ce: 2.1503 decode.d7.loss_mask_ce: 0.8651 decode.d7.loss_mask_dice: 1.7091 2023/09/06 17:46:09 - mmengine - INFO - Iter(train) [25750/60000] base_lr: 5.7084e-05 lr: 5.7084e-05 eta: 4:17:32 time: 0.4593 data_time: 0.0225 memory: 15770 grad_norm: 15.3283 loss: 9.0931 decode.loss_cls_ce: 1.9721 decode.loss_mask_ce: 0.9013 decode.loss_mask_dice: 1.6686 decode.d7.loss_cls_ce: 1.9750 decode.d7.loss_mask_ce: 0.9049 decode.d7.loss_mask_dice: 1.6712 2023/09/06 17:46:32 - mmengine - INFO - Iter(train) [25800/60000] base_lr: 5.7001e-05 lr: 5.7001e-05 eta: 4:17:10 time: 0.4499 data_time: 0.0244 memory: 15911 grad_norm: 16.3550 loss: 9.1926 decode.loss_cls_ce: 2.0062 decode.loss_mask_ce: 0.8415 decode.loss_mask_dice: 1.7328 decode.d7.loss_cls_ce: 2.0360 decode.d7.loss_mask_ce: 0.8391 decode.d7.loss_mask_dice: 1.7370 2023/09/06 17:46:54 - mmengine - INFO - Iter(train) [25850/60000] base_lr: 5.6918e-05 lr: 5.6918e-05 eta: 4:16:47 time: 0.4501 data_time: 0.0241 memory: 15871 grad_norm: 15.7942 loss: 10.1528 decode.loss_cls_ce: 2.1548 decode.loss_mask_ce: 0.9664 decode.loss_mask_dice: 1.9569 decode.d7.loss_cls_ce: 2.1709 decode.d7.loss_mask_ce: 0.9520 decode.d7.loss_mask_dice: 1.9518 2023/09/06 17:47:17 - mmengine - INFO - Iter(train) [25900/60000] base_lr: 5.6834e-05 lr: 5.6834e-05 eta: 4:16:25 time: 0.4531 data_time: 0.0230 memory: 15772 grad_norm: 15.9525 loss: 8.8961 decode.loss_cls_ce: 1.9234 decode.loss_mask_ce: 0.8952 decode.loss_mask_dice: 1.6242 decode.d7.loss_cls_ce: 1.9242 decode.d7.loss_mask_ce: 0.8959 decode.d7.loss_mask_dice: 1.6331 2023/09/06 17:47:39 - mmengine - INFO - Iter(train) [25950/60000] base_lr: 5.6751e-05 lr: 5.6751e-05 eta: 4:16:02 time: 0.4489 data_time: 0.0239 memory: 16050 grad_norm: 17.1963 loss: 9.8955 decode.loss_cls_ce: 2.1259 decode.loss_mask_ce: 0.9455 decode.loss_mask_dice: 1.8839 decode.d7.loss_cls_ce: 2.1249 decode.d7.loss_mask_ce: 0.9399 decode.d7.loss_mask_dice: 1.8754 2023/09/06 17:48:02 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:48:02 - mmengine - INFO - Iter(train) [26000/60000] base_lr: 5.6668e-05 lr: 5.6668e-05 eta: 4:15:40 time: 0.4537 data_time: 0.0239 memory: 15769 grad_norm: 17.2900 loss: 10.2124 decode.loss_cls_ce: 2.2521 decode.loss_mask_ce: 0.9387 decode.loss_mask_dice: 1.9372 decode.d7.loss_cls_ce: 2.2096 decode.d7.loss_mask_ce: 0.9450 decode.d7.loss_mask_dice: 1.9298 2023/09/06 17:48:25 - mmengine - INFO - Iter(train) [26050/60000] base_lr: 5.6584e-05 lr: 5.6584e-05 eta: 4:15:17 time: 0.4590 data_time: 0.0230 memory: 15923 grad_norm: 17.8967 loss: 10.7320 decode.loss_cls_ce: 2.4076 decode.loss_mask_ce: 0.9464 decode.loss_mask_dice: 2.0070 decode.d7.loss_cls_ce: 2.4411 decode.d7.loss_mask_ce: 0.9309 decode.d7.loss_mask_dice: 1.9990 2023/09/06 17:48:47 - mmengine - INFO - Iter(train) [26100/60000] base_lr: 5.6501e-05 lr: 5.6501e-05 eta: 4:14:55 time: 0.4552 data_time: 0.0232 memory: 15731 grad_norm: 20.1794 loss: 9.3380 decode.loss_cls_ce: 2.0403 decode.loss_mask_ce: 0.9372 decode.loss_mask_dice: 1.6980 decode.d7.loss_cls_ce: 2.0213 decode.d7.loss_mask_ce: 0.9390 decode.d7.loss_mask_dice: 1.7021 2023/09/06 17:49:10 - mmengine - INFO - Iter(train) [26150/60000] base_lr: 5.6418e-05 lr: 5.6418e-05 eta: 4:14:33 time: 0.4499 data_time: 0.0243 memory: 15937 grad_norm: 16.1375 loss: 9.6409 decode.loss_cls_ce: 2.0064 decode.loss_mask_ce: 0.9091 decode.loss_mask_dice: 1.9004 decode.d7.loss_cls_ce: 2.0032 decode.d7.loss_mask_ce: 0.9072 decode.d7.loss_mask_dice: 1.9145 2023/09/06 17:49:33 - mmengine - INFO - Iter(train) [26200/60000] base_lr: 5.6334e-05 lr: 5.6334e-05 eta: 4:14:11 time: 0.4605 data_time: 0.0233 memory: 16040 grad_norm: 19.0563 loss: 9.3466 decode.loss_cls_ce: 2.0510 decode.loss_mask_ce: 0.8528 decode.loss_mask_dice: 1.7509 decode.d7.loss_cls_ce: 2.0881 decode.d7.loss_mask_ce: 0.8530 decode.d7.loss_mask_dice: 1.7508 2023/09/06 17:49:56 - mmengine - INFO - Iter(train) [26250/60000] base_lr: 5.6251e-05 lr: 5.6251e-05 eta: 4:13:48 time: 0.4594 data_time: 0.0235 memory: 15846 grad_norm: 17.6853 loss: 8.3580 decode.loss_cls_ce: 1.8438 decode.loss_mask_ce: 0.8031 decode.loss_mask_dice: 1.5147 decode.d7.loss_cls_ce: 1.8869 decode.d7.loss_mask_ce: 0.7977 decode.d7.loss_mask_dice: 1.5119 2023/09/06 17:50:18 - mmengine - INFO - Iter(train) [26300/60000] base_lr: 5.6168e-05 lr: 5.6168e-05 eta: 4:13:26 time: 0.4483 data_time: 0.0232 memory: 15911 grad_norm: 17.7510 loss: 8.5415 decode.loss_cls_ce: 1.7902 decode.loss_mask_ce: 0.8975 decode.loss_mask_dice: 1.5762 decode.d7.loss_cls_ce: 1.7892 decode.d7.loss_mask_ce: 0.8967 decode.d7.loss_mask_dice: 1.5916 2023/09/06 17:50:41 - mmengine - INFO - Iter(train) [26350/60000] base_lr: 5.6084e-05 lr: 5.6084e-05 eta: 4:13:03 time: 0.4479 data_time: 0.0237 memory: 15809 grad_norm: 15.7961 loss: 9.4671 decode.loss_cls_ce: 1.9854 decode.loss_mask_ce: 0.9283 decode.loss_mask_dice: 1.8186 decode.d7.loss_cls_ce: 1.9762 decode.d7.loss_mask_ce: 0.9282 decode.d7.loss_mask_dice: 1.8305 2023/09/06 17:51:04 - mmengine - INFO - Iter(train) [26400/60000] base_lr: 5.6001e-05 lr: 5.6001e-05 eta: 4:12:41 time: 0.4582 data_time: 0.0230 memory: 15896 grad_norm: 16.9947 loss: 9.2943 decode.loss_cls_ce: 2.0786 decode.loss_mask_ce: 0.8637 decode.loss_mask_dice: 1.7070 decode.d7.loss_cls_ce: 2.0629 decode.d7.loss_mask_ce: 0.8672 decode.d7.loss_mask_dice: 1.7149 2023/09/06 17:51:27 - mmengine - INFO - Iter(train) [26450/60000] base_lr: 5.5918e-05 lr: 5.5918e-05 eta: 4:12:19 time: 0.4568 data_time: 0.0233 memory: 15885 grad_norm: 16.5360 loss: 9.8140 decode.loss_cls_ce: 2.1257 decode.loss_mask_ce: 0.9365 decode.loss_mask_dice: 1.8408 decode.d7.loss_cls_ce: 2.1277 decode.d7.loss_mask_ce: 0.9401 decode.d7.loss_mask_dice: 1.8431 2023/09/06 17:51:49 - mmengine - INFO - Iter(train) [26500/60000] base_lr: 5.5834e-05 lr: 5.5834e-05 eta: 4:11:57 time: 0.4592 data_time: 0.0233 memory: 15791 grad_norm: 17.1304 loss: 10.9244 decode.loss_cls_ce: 2.4131 decode.loss_mask_ce: 0.9577 decode.loss_mask_dice: 2.0867 decode.d7.loss_cls_ce: 2.4115 decode.d7.loss_mask_ce: 0.9648 decode.d7.loss_mask_dice: 2.0906 2023/09/06 17:52:12 - mmengine - INFO - Iter(train) [26550/60000] base_lr: 5.5751e-05 lr: 5.5751e-05 eta: 4:11:34 time: 0.4513 data_time: 0.0245 memory: 15860 grad_norm: 16.5284 loss: 10.4220 decode.loss_cls_ce: 2.1868 decode.loss_mask_ce: 1.0047 decode.loss_mask_dice: 2.0239 decode.d7.loss_cls_ce: 2.1672 decode.d7.loss_mask_ce: 1.0091 decode.d7.loss_mask_dice: 2.0302 2023/09/06 17:52:35 - mmengine - INFO - Iter(train) [26600/60000] base_lr: 5.5668e-05 lr: 5.5668e-05 eta: 4:11:11 time: 0.4541 data_time: 0.0243 memory: 15900 grad_norm: 16.7781 loss: 9.3289 decode.loss_cls_ce: 2.0832 decode.loss_mask_ce: 0.9070 decode.loss_mask_dice: 1.6593 decode.d7.loss_cls_ce: 2.0948 decode.d7.loss_mask_ce: 0.9137 decode.d7.loss_mask_dice: 1.6710 2023/09/06 17:52:57 - mmengine - INFO - Iter(train) [26650/60000] base_lr: 5.5584e-05 lr: 5.5584e-05 eta: 4:10:49 time: 0.4596 data_time: 0.0231 memory: 15951 grad_norm: 17.1865 loss: 9.0746 decode.loss_cls_ce: 2.0205 decode.loss_mask_ce: 0.8955 decode.loss_mask_dice: 1.6153 decode.d7.loss_cls_ce: 2.0426 decode.d7.loss_mask_ce: 0.8859 decode.d7.loss_mask_dice: 1.6147 2023/09/06 17:53:20 - mmengine - INFO - Iter(train) [26700/60000] base_lr: 5.5501e-05 lr: 5.5501e-05 eta: 4:10:27 time: 0.4548 data_time: 0.0238 memory: 15820 grad_norm: 17.5308 loss: 10.1787 decode.loss_cls_ce: 2.2204 decode.loss_mask_ce: 0.9661 decode.loss_mask_dice: 1.8955 decode.d7.loss_cls_ce: 2.2343 decode.d7.loss_mask_ce: 0.9671 decode.d7.loss_mask_dice: 1.8954 2023/09/06 17:53:43 - mmengine - INFO - Iter(train) [26750/60000] base_lr: 5.5418e-05 lr: 5.5418e-05 eta: 4:10:04 time: 0.4543 data_time: 0.0237 memory: 15824 grad_norm: 17.1443 loss: 10.0275 decode.loss_cls_ce: 2.2036 decode.loss_mask_ce: 0.8942 decode.loss_mask_dice: 1.9212 decode.d7.loss_cls_ce: 2.2064 decode.d7.loss_mask_ce: 0.9006 decode.d7.loss_mask_dice: 1.9016 2023/09/06 17:54:05 - mmengine - INFO - Iter(train) [26800/60000] base_lr: 5.5334e-05 lr: 5.5334e-05 eta: 4:09:42 time: 0.4508 data_time: 0.0236 memory: 15845 grad_norm: 16.7956 loss: 9.5114 decode.loss_cls_ce: 2.1002 decode.loss_mask_ce: 0.8610 decode.loss_mask_dice: 1.7935 decode.d7.loss_cls_ce: 2.1137 decode.d7.loss_mask_ce: 0.8567 decode.d7.loss_mask_dice: 1.7863 2023/09/06 17:54:28 - mmengine - INFO - Iter(train) [26850/60000] base_lr: 5.5251e-05 lr: 5.5251e-05 eta: 4:09:19 time: 0.4498 data_time: 0.0236 memory: 15884 grad_norm: 15.3581 loss: 8.6017 decode.loss_cls_ce: 1.8536 decode.loss_mask_ce: 0.9038 decode.loss_mask_dice: 1.5477 decode.d7.loss_cls_ce: 1.8482 decode.d7.loss_mask_ce: 0.8985 decode.d7.loss_mask_dice: 1.5498 2023/09/06 17:54:50 - mmengine - INFO - Iter(train) [26900/60000] base_lr: 5.5168e-05 lr: 5.5168e-05 eta: 4:08:57 time: 0.4581 data_time: 0.0239 memory: 15847 grad_norm: 14.5624 loss: 9.3325 decode.loss_cls_ce: 1.9730 decode.loss_mask_ce: 0.9821 decode.loss_mask_dice: 1.6980 decode.d7.loss_cls_ce: 2.0084 decode.d7.loss_mask_ce: 0.9766 decode.d7.loss_mask_dice: 1.6943 2023/09/06 17:55:13 - mmengine - INFO - Iter(train) [26950/60000] base_lr: 5.5084e-05 lr: 5.5084e-05 eta: 4:08:34 time: 0.4539 data_time: 0.0236 memory: 15844 grad_norm: 15.3715 loss: 9.8429 decode.loss_cls_ce: 2.0271 decode.loss_mask_ce: 0.9597 decode.loss_mask_dice: 1.9524 decode.d7.loss_cls_ce: 2.0172 decode.d7.loss_mask_ce: 0.9510 decode.d7.loss_mask_dice: 1.9356 2023/09/06 17:55:36 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 17:55:36 - mmengine - INFO - Iter(train) [27000/60000] base_lr: 5.5001e-05 lr: 5.5001e-05 eta: 4:08:11 time: 0.4491 data_time: 0.0240 memory: 15835 grad_norm: 18.7973 loss: 9.9639 decode.loss_cls_ce: 2.1834 decode.loss_mask_ce: 0.9225 decode.loss_mask_dice: 1.8560 decode.d7.loss_cls_ce: 2.2025 decode.d7.loss_mask_ce: 0.9296 decode.d7.loss_mask_dice: 1.8699 2023/09/06 17:55:58 - mmengine - INFO - Iter(train) [27050/60000] base_lr: 5.4918e-05 lr: 5.4918e-05 eta: 4:07:49 time: 0.4470 data_time: 0.0235 memory: 15925 grad_norm: 15.4636 loss: 10.1612 decode.loss_cls_ce: 2.1707 decode.loss_mask_ce: 0.9657 decode.loss_mask_dice: 1.9321 decode.d7.loss_cls_ce: 2.1895 decode.d7.loss_mask_ce: 0.9787 decode.d7.loss_mask_dice: 1.9245 2023/09/06 17:56:21 - mmengine - INFO - Iter(train) [27100/60000] base_lr: 5.4834e-05 lr: 5.4834e-05 eta: 4:07:26 time: 0.4500 data_time: 0.0236 memory: 15849 grad_norm: 17.5478 loss: 10.1918 decode.loss_cls_ce: 2.2826 decode.loss_mask_ce: 0.9525 decode.loss_mask_dice: 1.8552 decode.d7.loss_cls_ce: 2.2909 decode.d7.loss_mask_ce: 0.9549 decode.d7.loss_mask_dice: 1.8557 2023/09/06 17:56:43 - mmengine - INFO - Iter(train) [27150/60000] base_lr: 5.4751e-05 lr: 5.4751e-05 eta: 4:07:04 time: 0.4477 data_time: 0.0238 memory: 16017 grad_norm: 17.0676 loss: 9.3404 decode.loss_cls_ce: 2.1015 decode.loss_mask_ce: 0.8776 decode.loss_mask_dice: 1.6938 decode.d7.loss_cls_ce: 2.1088 decode.d7.loss_mask_ce: 0.8676 decode.d7.loss_mask_dice: 1.6911 2023/09/06 17:57:06 - mmengine - INFO - Iter(train) [27200/60000] base_lr: 5.4668e-05 lr: 5.4668e-05 eta: 4:06:41 time: 0.4482 data_time: 0.0241 memory: 15860 grad_norm: 17.0066 loss: 9.4414 decode.loss_cls_ce: 2.0938 decode.loss_mask_ce: 0.8777 decode.loss_mask_dice: 1.7532 decode.d7.loss_cls_ce: 2.0937 decode.d7.loss_mask_ce: 0.8716 decode.d7.loss_mask_dice: 1.7515 2023/09/06 17:57:28 - mmengine - INFO - Iter(train) [27250/60000] base_lr: 5.4584e-05 lr: 5.4584e-05 eta: 4:06:19 time: 0.4530 data_time: 0.0252 memory: 15773 grad_norm: 17.3364 loss: 9.9880 decode.loss_cls_ce: 2.1114 decode.loss_mask_ce: 0.9824 decode.loss_mask_dice: 1.8853 decode.d7.loss_cls_ce: 2.1450 decode.d7.loss_mask_ce: 0.9748 decode.d7.loss_mask_dice: 1.8891 2023/09/06 17:57:51 - mmengine - INFO - Iter(train) [27300/60000] base_lr: 5.4501e-05 lr: 5.4501e-05 eta: 4:05:56 time: 0.4534 data_time: 0.0234 memory: 15797 grad_norm: 16.9305 loss: 9.2868 decode.loss_cls_ce: 1.9527 decode.loss_mask_ce: 0.9226 decode.loss_mask_dice: 1.7721 decode.d7.loss_cls_ce: 1.9488 decode.d7.loss_mask_ce: 0.9083 decode.d7.loss_mask_dice: 1.7824 2023/09/06 17:58:14 - mmengine - INFO - Iter(train) [27350/60000] base_lr: 5.4418e-05 lr: 5.4418e-05 eta: 4:05:33 time: 0.4505 data_time: 0.0235 memory: 15810 grad_norm: 16.4363 loss: 9.1609 decode.loss_cls_ce: 2.0062 decode.loss_mask_ce: 0.8795 decode.loss_mask_dice: 1.6832 decode.d7.loss_cls_ce: 2.0314 decode.d7.loss_mask_ce: 0.8767 decode.d7.loss_mask_dice: 1.6841 2023/09/06 17:58:36 - mmengine - INFO - Iter(train) [27400/60000] base_lr: 5.4334e-05 lr: 5.4334e-05 eta: 4:05:11 time: 0.4497 data_time: 0.0240 memory: 15731 grad_norm: 17.7834 loss: 9.2253 decode.loss_cls_ce: 2.1204 decode.loss_mask_ce: 0.8805 decode.loss_mask_dice: 1.6187 decode.d7.loss_cls_ce: 2.1112 decode.d7.loss_mask_ce: 0.8801 decode.d7.loss_mask_dice: 1.6144 2023/09/06 17:58:59 - mmengine - INFO - Iter(train) [27450/60000] base_lr: 5.4251e-05 lr: 5.4251e-05 eta: 4:04:48 time: 0.4603 data_time: 0.0232 memory: 16055 grad_norm: 16.4723 loss: 10.2531 decode.loss_cls_ce: 2.2489 decode.loss_mask_ce: 0.9977 decode.loss_mask_dice: 1.8875 decode.d7.loss_cls_ce: 2.2694 decode.d7.loss_mask_ce: 0.9828 decode.d7.loss_mask_dice: 1.8668 2023/09/06 17:59:22 - mmengine - INFO - Iter(train) [27500/60000] base_lr: 5.4168e-05 lr: 5.4168e-05 eta: 4:04:26 time: 0.4520 data_time: 0.0231 memory: 15755 grad_norm: 18.9015 loss: 9.5308 decode.loss_cls_ce: 2.1887 decode.loss_mask_ce: 0.8396 decode.loss_mask_dice: 1.7376 decode.d7.loss_cls_ce: 2.1685 decode.d7.loss_mask_ce: 0.8476 decode.d7.loss_mask_dice: 1.7489 2023/09/06 17:59:45 - mmengine - INFO - Iter(train) [27550/60000] base_lr: 5.4084e-05 lr: 5.4084e-05 eta: 4:04:04 time: 0.4569 data_time: 0.0235 memory: 15806 grad_norm: 16.7287 loss: 9.6839 decode.loss_cls_ce: 2.0608 decode.loss_mask_ce: 0.9575 decode.loss_mask_dice: 1.8099 decode.d7.loss_cls_ce: 2.0665 decode.d7.loss_mask_ce: 0.9646 decode.d7.loss_mask_dice: 1.8247 2023/09/06 18:00:07 - mmengine - INFO - Iter(train) [27600/60000] base_lr: 5.4001e-05 lr: 5.4001e-05 eta: 4:03:42 time: 0.4527 data_time: 0.0230 memory: 15911 grad_norm: 17.6122 loss: 9.1249 decode.loss_cls_ce: 2.0955 decode.loss_mask_ce: 0.8560 decode.loss_mask_dice: 1.6075 decode.d7.loss_cls_ce: 2.0994 decode.d7.loss_mask_ce: 0.8573 decode.d7.loss_mask_dice: 1.6092 2023/09/06 18:00:30 - mmengine - INFO - Iter(train) [27650/60000] base_lr: 5.3918e-05 lr: 5.3918e-05 eta: 4:03:19 time: 0.4511 data_time: 0.0242 memory: 15923 grad_norm: 16.7303 loss: 10.6197 decode.loss_cls_ce: 2.2150 decode.loss_mask_ce: 1.0249 decode.loss_mask_dice: 2.0717 decode.d7.loss_cls_ce: 2.2153 decode.d7.loss_mask_ce: 1.0215 decode.d7.loss_mask_dice: 2.0712 2023/09/06 18:00:53 - mmengine - INFO - Iter(train) [27700/60000] base_lr: 5.3834e-05 lr: 5.3834e-05 eta: 4:02:57 time: 0.4571 data_time: 0.0239 memory: 15810 grad_norm: 18.2429 loss: 8.8698 decode.loss_cls_ce: 1.8394 decode.loss_mask_ce: 0.9386 decode.loss_mask_dice: 1.6458 decode.d7.loss_cls_ce: 1.8600 decode.d7.loss_mask_ce: 0.9371 decode.d7.loss_mask_dice: 1.6489 2023/09/06 18:01:15 - mmengine - INFO - Iter(train) [27750/60000] base_lr: 5.3751e-05 lr: 5.3751e-05 eta: 4:02:34 time: 0.4522 data_time: 0.0243 memory: 15784 grad_norm: 17.1829 loss: 8.4367 decode.loss_cls_ce: 1.9507 decode.loss_mask_ce: 0.7957 decode.loss_mask_dice: 1.4755 decode.d7.loss_cls_ce: 1.9492 decode.d7.loss_mask_ce: 0.7943 decode.d7.loss_mask_dice: 1.4713 2023/09/06 18:01:38 - mmengine - INFO - Iter(train) [27800/60000] base_lr: 5.3668e-05 lr: 5.3668e-05 eta: 4:02:12 time: 0.4545 data_time: 0.0224 memory: 16051 grad_norm: 15.4802 loss: 9.6845 decode.loss_cls_ce: 2.1427 decode.loss_mask_ce: 0.8798 decode.loss_mask_dice: 1.8228 decode.d7.loss_cls_ce: 2.1507 decode.d7.loss_mask_ce: 0.8801 decode.d7.loss_mask_dice: 1.8084 2023/09/06 18:02:01 - mmengine - INFO - Iter(train) [27850/60000] base_lr: 5.3584e-05 lr: 5.3584e-05 eta: 4:01:50 time: 0.4548 data_time: 0.0241 memory: 15694 grad_norm: 18.6817 loss: 9.1036 decode.loss_cls_ce: 2.1338 decode.loss_mask_ce: 0.8261 decode.loss_mask_dice: 1.5914 decode.d7.loss_cls_ce: 2.1337 decode.d7.loss_mask_ce: 0.8220 decode.d7.loss_mask_dice: 1.5966 2023/09/06 18:02:24 - mmengine - INFO - Iter(train) [27900/60000] base_lr: 5.3501e-05 lr: 5.3501e-05 eta: 4:01:27 time: 0.4558 data_time: 0.0237 memory: 15924 grad_norm: 17.7306 loss: 9.8119 decode.loss_cls_ce: 2.2396 decode.loss_mask_ce: 0.9446 decode.loss_mask_dice: 1.7308 decode.d7.loss_cls_ce: 2.2372 decode.d7.loss_mask_ce: 0.9407 decode.d7.loss_mask_dice: 1.7190 2023/09/06 18:02:46 - mmengine - INFO - Iter(train) [27950/60000] base_lr: 5.3418e-05 lr: 5.3418e-05 eta: 4:01:05 time: 0.4511 data_time: 0.0237 memory: 15787 grad_norm: 18.2396 loss: 9.8812 decode.loss_cls_ce: 2.0730 decode.loss_mask_ce: 0.9407 decode.loss_mask_dice: 1.9294 decode.d7.loss_cls_ce: 2.0824 decode.d7.loss_mask_ce: 0.9267 decode.d7.loss_mask_dice: 1.9290 2023/09/06 18:03:09 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:03:09 - mmengine - INFO - Iter(train) [28000/60000] base_lr: 5.3334e-05 lr: 5.3334e-05 eta: 4:00:42 time: 0.4546 data_time: 0.0237 memory: 16130 grad_norm: 17.1403 loss: 8.5793 decode.loss_cls_ce: 1.9741 decode.loss_mask_ce: 0.7934 decode.loss_mask_dice: 1.5135 decode.d7.loss_cls_ce: 1.9932 decode.d7.loss_mask_ce: 0.7863 decode.d7.loss_mask_dice: 1.5188 2023/09/06 18:03:32 - mmengine - INFO - Iter(train) [28050/60000] base_lr: 5.3251e-05 lr: 5.3251e-05 eta: 4:00:20 time: 0.4604 data_time: 0.0233 memory: 15962 grad_norm: 19.3031 loss: 9.3391 decode.loss_cls_ce: 2.0225 decode.loss_mask_ce: 0.9267 decode.loss_mask_dice: 1.7162 decode.d7.loss_cls_ce: 2.0385 decode.d7.loss_mask_ce: 0.9185 decode.d7.loss_mask_dice: 1.7166 2023/09/06 18:03:55 - mmengine - INFO - Iter(train) [28100/60000] base_lr: 5.3168e-05 lr: 5.3168e-05 eta: 3:59:58 time: 0.4544 data_time: 0.0237 memory: 15881 grad_norm: 20.0937 loss: 9.5510 decode.loss_cls_ce: 2.1119 decode.loss_mask_ce: 0.8485 decode.loss_mask_dice: 1.8187 decode.d7.loss_cls_ce: 2.0935 decode.d7.loss_mask_ce: 0.8505 decode.d7.loss_mask_dice: 1.8278 2023/09/06 18:04:17 - mmengine - INFO - Iter(train) [28150/60000] base_lr: 5.3084e-05 lr: 5.3084e-05 eta: 3:59:35 time: 0.4504 data_time: 0.0247 memory: 15784 grad_norm: 18.7946 loss: 8.7743 decode.loss_cls_ce: 1.7964 decode.loss_mask_ce: 0.9219 decode.loss_mask_dice: 1.6560 decode.d7.loss_cls_ce: 1.8069 decode.d7.loss_mask_ce: 0.9269 decode.d7.loss_mask_dice: 1.6661 2023/09/06 18:04:40 - mmengine - INFO - Iter(train) [28200/60000] base_lr: 5.3001e-05 lr: 5.3001e-05 eta: 3:59:13 time: 0.4607 data_time: 0.0231 memory: 15809 grad_norm: 16.8185 loss: 8.6609 decode.loss_cls_ce: 1.9933 decode.loss_mask_ce: 0.8040 decode.loss_mask_dice: 1.5435 decode.d7.loss_cls_ce: 1.9609 decode.d7.loss_mask_ce: 0.8054 decode.d7.loss_mask_dice: 1.5538 2023/09/06 18:05:03 - mmengine - INFO - Iter(train) [28250/60000] base_lr: 5.2918e-05 lr: 5.2918e-05 eta: 3:58:51 time: 0.4523 data_time: 0.0239 memory: 15877 grad_norm: 17.1128 loss: 8.7444 decode.loss_cls_ce: 1.9130 decode.loss_mask_ce: 0.8601 decode.loss_mask_dice: 1.6014 decode.d7.loss_cls_ce: 1.9092 decode.d7.loss_mask_ce: 0.8631 decode.d7.loss_mask_dice: 1.5976 2023/09/06 18:05:26 - mmengine - INFO - Iter(train) [28300/60000] base_lr: 5.2834e-05 lr: 5.2834e-05 eta: 3:58:28 time: 0.4605 data_time: 0.0228 memory: 15861 grad_norm: 18.2779 loss: 9.9547 decode.loss_cls_ce: 2.1916 decode.loss_mask_ce: 0.9390 decode.loss_mask_dice: 1.8481 decode.d7.loss_cls_ce: 2.1842 decode.d7.loss_mask_ce: 0.9383 decode.d7.loss_mask_dice: 1.8535 2023/09/06 18:05:49 - mmengine - INFO - Iter(train) [28350/60000] base_lr: 5.2751e-05 lr: 5.2751e-05 eta: 3:58:06 time: 0.4588 data_time: 0.0234 memory: 15795 grad_norm: 19.2691 loss: 9.7780 decode.loss_cls_ce: 2.2758 decode.loss_mask_ce: 0.8351 decode.loss_mask_dice: 1.7761 decode.d7.loss_cls_ce: 2.2683 decode.d7.loss_mask_ce: 0.8433 decode.d7.loss_mask_dice: 1.7794 2023/09/06 18:06:12 - mmengine - INFO - Iter(train) [28400/60000] base_lr: 5.2668e-05 lr: 5.2668e-05 eta: 3:57:44 time: 0.4502 data_time: 0.0237 memory: 15730 grad_norm: 20.2877 loss: 9.7382 decode.loss_cls_ce: 2.1962 decode.loss_mask_ce: 0.8533 decode.loss_mask_dice: 1.8359 decode.d7.loss_cls_ce: 2.1765 decode.d7.loss_mask_ce: 0.8475 decode.d7.loss_mask_dice: 1.8289 2023/09/06 18:06:34 - mmengine - INFO - Iter(train) [28450/60000] base_lr: 5.2584e-05 lr: 5.2584e-05 eta: 3:57:22 time: 0.4597 data_time: 0.0233 memory: 15821 grad_norm: 18.0633 loss: 9.3098 decode.loss_cls_ce: 1.9910 decode.loss_mask_ce: 0.8399 decode.loss_mask_dice: 1.8202 decode.d7.loss_cls_ce: 2.0139 decode.d7.loss_mask_ce: 0.8387 decode.d7.loss_mask_dice: 1.8061 2023/09/06 18:06:57 - mmengine - INFO - Iter(train) [28500/60000] base_lr: 5.2501e-05 lr: 5.2501e-05 eta: 3:56:59 time: 0.4502 data_time: 0.0242 memory: 15813 grad_norm: 17.1623 loss: 9.0199 decode.loss_cls_ce: 1.9534 decode.loss_mask_ce: 0.8785 decode.loss_mask_dice: 1.6671 decode.d7.loss_cls_ce: 1.9582 decode.d7.loss_mask_ce: 0.8770 decode.d7.loss_mask_dice: 1.6856 2023/09/06 18:07:20 - mmengine - INFO - Iter(train) [28550/60000] base_lr: 5.2418e-05 lr: 5.2418e-05 eta: 3:56:37 time: 0.4599 data_time: 0.0229 memory: 15834 grad_norm: 16.8512 loss: 8.6431 decode.loss_cls_ce: 1.9949 decode.loss_mask_ce: 0.7750 decode.loss_mask_dice: 1.5597 decode.d7.loss_cls_ce: 1.9828 decode.d7.loss_mask_ce: 0.7755 decode.d7.loss_mask_dice: 1.5553 2023/09/06 18:07:43 - mmengine - INFO - Iter(train) [28600/60000] base_lr: 5.2334e-05 lr: 5.2334e-05 eta: 3:56:14 time: 0.4545 data_time: 0.0235 memory: 15845 grad_norm: 17.0347 loss: 8.9780 decode.loss_cls_ce: 1.8930 decode.loss_mask_ce: 0.8412 decode.loss_mask_dice: 1.7650 decode.d7.loss_cls_ce: 1.8964 decode.d7.loss_mask_ce: 0.8452 decode.d7.loss_mask_dice: 1.7372 2023/09/06 18:08:05 - mmengine - INFO - Iter(train) [28650/60000] base_lr: 5.2251e-05 lr: 5.2251e-05 eta: 3:55:52 time: 0.4477 data_time: 0.0232 memory: 15870 grad_norm: 17.3860 loss: 9.5317 decode.loss_cls_ce: 2.0415 decode.loss_mask_ce: 0.9125 decode.loss_mask_dice: 1.8151 decode.d7.loss_cls_ce: 2.0286 decode.d7.loss_mask_ce: 0.9205 decode.d7.loss_mask_dice: 1.8134 2023/09/06 18:08:28 - mmengine - INFO - Iter(train) [28700/60000] base_lr: 5.2168e-05 lr: 5.2168e-05 eta: 3:55:29 time: 0.4572 data_time: 0.0275 memory: 15925 grad_norm: 16.8816 loss: 10.0251 decode.loss_cls_ce: 2.1695 decode.loss_mask_ce: 0.9250 decode.loss_mask_dice: 1.9270 decode.d7.loss_cls_ce: 2.1477 decode.d7.loss_mask_ce: 0.9333 decode.d7.loss_mask_dice: 1.9227 2023/09/06 18:08:50 - mmengine - INFO - Iter(train) [28750/60000] base_lr: 5.2084e-05 lr: 5.2084e-05 eta: 3:55:07 time: 0.4559 data_time: 0.0230 memory: 15821 grad_norm: 18.3903 loss: 9.0150 decode.loss_cls_ce: 2.0401 decode.loss_mask_ce: 0.8712 decode.loss_mask_dice: 1.5840 decode.d7.loss_cls_ce: 2.0446 decode.d7.loss_mask_ce: 0.8830 decode.d7.loss_mask_dice: 1.5922 2023/09/06 18:09:13 - mmengine - INFO - Iter(train) [28800/60000] base_lr: 5.2001e-05 lr: 5.2001e-05 eta: 3:54:44 time: 0.4540 data_time: 0.0238 memory: 15807 grad_norm: 17.4767 loss: 10.0740 decode.loss_cls_ce: 2.2180 decode.loss_mask_ce: 0.9342 decode.loss_mask_dice: 1.8891 decode.d7.loss_cls_ce: 2.2321 decode.d7.loss_mask_ce: 0.9372 decode.d7.loss_mask_dice: 1.8633 2023/09/06 18:09:36 - mmengine - INFO - Iter(train) [28850/60000] base_lr: 5.1918e-05 lr: 5.1918e-05 eta: 3:54:22 time: 0.4540 data_time: 0.0241 memory: 15808 grad_norm: 16.0879 loss: 9.3361 decode.loss_cls_ce: 2.0638 decode.loss_mask_ce: 0.9081 decode.loss_mask_dice: 1.6952 decode.d7.loss_cls_ce: 2.0726 decode.d7.loss_mask_ce: 0.9066 decode.d7.loss_mask_dice: 1.6897 2023/09/06 18:09:58 - mmengine - INFO - Iter(train) [28900/60000] base_lr: 5.1834e-05 lr: 5.1834e-05 eta: 3:53:59 time: 0.4517 data_time: 0.0242 memory: 15808 grad_norm: 15.4593 loss: 8.8733 decode.loss_cls_ce: 1.9525 decode.loss_mask_ce: 0.8216 decode.loss_mask_dice: 1.6564 decode.d7.loss_cls_ce: 1.9628 decode.d7.loss_mask_ce: 0.8138 decode.d7.loss_mask_dice: 1.6661 2023/09/06 18:10:21 - mmengine - INFO - Iter(train) [28950/60000] base_lr: 5.1751e-05 lr: 5.1751e-05 eta: 3:53:37 time: 0.4602 data_time: 0.0235 memory: 15783 grad_norm: 16.6674 loss: 9.9718 decode.loss_cls_ce: 2.2092 decode.loss_mask_ce: 0.9848 decode.loss_mask_dice: 1.7856 decode.d7.loss_cls_ce: 2.2148 decode.d7.loss_mask_ce: 0.9822 decode.d7.loss_mask_dice: 1.7953 2023/09/06 18:10:44 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:10:44 - mmengine - INFO - Iter(train) [29000/60000] base_lr: 5.1668e-05 lr: 5.1668e-05 eta: 3:53:15 time: 0.4490 data_time: 0.0244 memory: 15869 grad_norm: 17.4036 loss: 9.4091 decode.loss_cls_ce: 2.1510 decode.loss_mask_ce: 0.9126 decode.loss_mask_dice: 1.6344 decode.d7.loss_cls_ce: 2.1645 decode.d7.loss_mask_ce: 0.9106 decode.d7.loss_mask_dice: 1.6361 2023/09/06 18:11:06 - mmengine - INFO - Iter(train) [29050/60000] base_lr: 5.1584e-05 lr: 5.1584e-05 eta: 3:52:52 time: 0.4493 data_time: 0.0240 memory: 15809 grad_norm: 18.6063 loss: 9.9810 decode.loss_cls_ce: 2.1095 decode.loss_mask_ce: 0.9690 decode.loss_mask_dice: 1.9148 decode.d7.loss_cls_ce: 2.1383 decode.d7.loss_mask_ce: 0.9568 decode.d7.loss_mask_dice: 1.8926 2023/09/06 18:11:29 - mmengine - INFO - Iter(train) [29100/60000] base_lr: 5.1501e-05 lr: 5.1501e-05 eta: 3:52:29 time: 0.4524 data_time: 0.0235 memory: 15900 grad_norm: 17.1049 loss: 9.4596 decode.loss_cls_ce: 2.1278 decode.loss_mask_ce: 0.8810 decode.loss_mask_dice: 1.7164 decode.d7.loss_cls_ce: 2.1184 decode.d7.loss_mask_ce: 0.8901 decode.d7.loss_mask_dice: 1.7259 2023/09/06 18:11:52 - mmengine - INFO - Iter(train) [29150/60000] base_lr: 5.1418e-05 lr: 5.1418e-05 eta: 3:52:07 time: 0.4591 data_time: 0.0239 memory: 15783 grad_norm: 16.8402 loss: 8.3634 decode.loss_cls_ce: 1.8368 decode.loss_mask_ce: 0.8075 decode.loss_mask_dice: 1.5318 decode.d7.loss_cls_ce: 1.8490 decode.d7.loss_mask_ce: 0.8076 decode.d7.loss_mask_dice: 1.5307 2023/09/06 18:12:15 - mmengine - INFO - Iter(train) [29200/60000] base_lr: 5.1334e-05 lr: 5.1334e-05 eta: 3:51:45 time: 0.4562 data_time: 0.0227 memory: 15694 grad_norm: 16.1880 loss: 8.6756 decode.loss_cls_ce: 2.0206 decode.loss_mask_ce: 0.8036 decode.loss_mask_dice: 1.5063 decode.d7.loss_cls_ce: 2.0160 decode.d7.loss_mask_ce: 0.8039 decode.d7.loss_mask_dice: 1.5251 2023/09/06 18:12:38 - mmengine - INFO - Iter(train) [29250/60000] base_lr: 5.1251e-05 lr: 5.1251e-05 eta: 3:51:22 time: 0.4606 data_time: 0.0237 memory: 15872 grad_norm: 17.8845 loss: 10.0791 decode.loss_cls_ce: 2.2023 decode.loss_mask_ce: 0.9631 decode.loss_mask_dice: 1.8905 decode.d7.loss_cls_ce: 2.1759 decode.d7.loss_mask_ce: 0.9557 decode.d7.loss_mask_dice: 1.8916 2023/09/06 18:13:00 - mmengine - INFO - Iter(train) [29300/60000] base_lr: 5.1168e-05 lr: 5.1168e-05 eta: 3:51:00 time: 0.4496 data_time: 0.0237 memory: 15872 grad_norm: 16.8623 loss: 9.7391 decode.loss_cls_ce: 2.0838 decode.loss_mask_ce: 0.9602 decode.loss_mask_dice: 1.8244 decode.d7.loss_cls_ce: 2.0713 decode.d7.loss_mask_ce: 0.9596 decode.d7.loss_mask_dice: 1.8398 2023/09/06 18:13:23 - mmengine - INFO - Iter(train) [29350/60000] base_lr: 5.1084e-05 lr: 5.1084e-05 eta: 3:50:38 time: 0.4593 data_time: 0.0232 memory: 15952 grad_norm: 16.8895 loss: 9.1929 decode.loss_cls_ce: 1.9363 decode.loss_mask_ce: 0.8999 decode.loss_mask_dice: 1.7403 decode.d7.loss_cls_ce: 1.9648 decode.d7.loss_mask_ce: 0.8960 decode.d7.loss_mask_dice: 1.7554 2023/09/06 18:13:46 - mmengine - INFO - Iter(train) [29400/60000] base_lr: 5.1001e-05 lr: 5.1001e-05 eta: 3:50:15 time: 0.4610 data_time: 0.0231 memory: 16003 grad_norm: 17.3830 loss: 9.4928 decode.loss_cls_ce: 2.1077 decode.loss_mask_ce: 0.9470 decode.loss_mask_dice: 1.6886 decode.d7.loss_cls_ce: 2.1205 decode.d7.loss_mask_ce: 0.9508 decode.d7.loss_mask_dice: 1.6781 2023/09/06 18:14:09 - mmengine - INFO - Iter(train) [29450/60000] base_lr: 5.0918e-05 lr: 5.0918e-05 eta: 3:49:53 time: 0.4518 data_time: 0.0235 memory: 15882 grad_norm: 18.0514 loss: 9.4443 decode.loss_cls_ce: 2.0064 decode.loss_mask_ce: 0.9153 decode.loss_mask_dice: 1.7936 decode.d7.loss_cls_ce: 2.0276 decode.d7.loss_mask_ce: 0.9096 decode.d7.loss_mask_dice: 1.7919 2023/09/06 18:14:31 - mmengine - INFO - Iter(train) [29500/60000] base_lr: 5.0834e-05 lr: 5.0834e-05 eta: 3:49:30 time: 0.4512 data_time: 0.0236 memory: 16053 grad_norm: 16.7358 loss: 9.4338 decode.loss_cls_ce: 2.0721 decode.loss_mask_ce: 0.8862 decode.loss_mask_dice: 1.7377 decode.d7.loss_cls_ce: 2.0928 decode.d7.loss_mask_ce: 0.8892 decode.d7.loss_mask_dice: 1.7558 2023/09/06 18:14:54 - mmengine - INFO - Iter(train) [29550/60000] base_lr: 5.0751e-05 lr: 5.0751e-05 eta: 3:49:08 time: 0.4532 data_time: 0.0239 memory: 15885 grad_norm: 16.9315 loss: 10.7381 decode.loss_cls_ce: 2.4834 decode.loss_mask_ce: 0.9344 decode.loss_mask_dice: 1.9471 decode.d7.loss_cls_ce: 2.4829 decode.d7.loss_mask_ce: 0.9361 decode.d7.loss_mask_dice: 1.9543 2023/09/06 18:15:17 - mmengine - INFO - Iter(train) [29600/60000] base_lr: 5.0668e-05 lr: 5.0668e-05 eta: 3:48:45 time: 0.4526 data_time: 0.0249 memory: 15845 grad_norm: 17.7338 loss: 8.9027 decode.loss_cls_ce: 1.9599 decode.loss_mask_ce: 0.8228 decode.loss_mask_dice: 1.6752 decode.d7.loss_cls_ce: 1.9387 decode.d7.loss_mask_ce: 0.8288 decode.d7.loss_mask_dice: 1.6773 2023/09/06 18:15:39 - mmengine - INFO - Iter(train) [29650/60000] base_lr: 5.0584e-05 lr: 5.0584e-05 eta: 3:48:23 time: 0.4504 data_time: 0.0238 memory: 15757 grad_norm: 18.4332 loss: 9.9393 decode.loss_cls_ce: 2.2451 decode.loss_mask_ce: 0.8959 decode.loss_mask_dice: 1.8395 decode.d7.loss_cls_ce: 2.2137 decode.d7.loss_mask_ce: 0.9006 decode.d7.loss_mask_dice: 1.8445 2023/09/06 18:16:02 - mmengine - INFO - Iter(train) [29700/60000] base_lr: 5.0501e-05 lr: 5.0501e-05 eta: 3:48:01 time: 0.4623 data_time: 0.0236 memory: 15923 grad_norm: 16.3540 loss: 10.5939 decode.loss_cls_ce: 2.3377 decode.loss_mask_ce: 0.9391 decode.loss_mask_dice: 2.0274 decode.d7.loss_cls_ce: 2.3286 decode.d7.loss_mask_ce: 0.9482 decode.d7.loss_mask_dice: 2.0129 2023/09/06 18:16:25 - mmengine - INFO - Iter(train) [29750/60000] base_lr: 5.0418e-05 lr: 5.0418e-05 eta: 3:47:38 time: 0.4513 data_time: 0.0226 memory: 15836 grad_norm: 17.5705 loss: 8.2912 decode.loss_cls_ce: 1.8885 decode.loss_mask_ce: 0.7731 decode.loss_mask_dice: 1.4926 decode.d7.loss_cls_ce: 1.8767 decode.d7.loss_mask_ce: 0.7672 decode.d7.loss_mask_dice: 1.4931 2023/09/06 18:16:47 - mmengine - INFO - Iter(train) [29800/60000] base_lr: 5.0334e-05 lr: 5.0334e-05 eta: 3:47:16 time: 0.4493 data_time: 0.0237 memory: 15849 grad_norm: 20.3523 loss: 10.0266 decode.loss_cls_ce: 2.2347 decode.loss_mask_ce: 0.9666 decode.loss_mask_dice: 1.8066 decode.d7.loss_cls_ce: 2.2319 decode.d7.loss_mask_ce: 0.9663 decode.d7.loss_mask_dice: 1.8206 2023/09/06 18:17:10 - mmengine - INFO - Iter(train) [29850/60000] base_lr: 5.0251e-05 lr: 5.0251e-05 eta: 3:46:53 time: 0.4526 data_time: 0.0232 memory: 15807 grad_norm: 17.1255 loss: 10.3044 decode.loss_cls_ce: 2.1518 decode.loss_mask_ce: 0.9961 decode.loss_mask_dice: 2.0027 decode.d7.loss_cls_ce: 2.1511 decode.d7.loss_mask_ce: 0.9965 decode.d7.loss_mask_dice: 2.0062 2023/09/06 18:17:33 - mmengine - INFO - Iter(train) [29900/60000] base_lr: 5.0168e-05 lr: 5.0168e-05 eta: 3:46:31 time: 0.4618 data_time: 0.0235 memory: 15783 grad_norm: 16.9391 loss: 9.1287 decode.loss_cls_ce: 1.9211 decode.loss_mask_ce: 0.9068 decode.loss_mask_dice: 1.7214 decode.d7.loss_cls_ce: 1.9438 decode.d7.loss_mask_ce: 0.9067 decode.d7.loss_mask_dice: 1.7289 2023/09/06 18:17:56 - mmengine - INFO - Iter(train) [29950/60000] base_lr: 5.0084e-05 lr: 5.0084e-05 eta: 3:46:08 time: 0.4518 data_time: 0.0234 memory: 15885 grad_norm: 16.8834 loss: 8.6748 decode.loss_cls_ce: 1.9186 decode.loss_mask_ce: 0.8037 decode.loss_mask_dice: 1.6093 decode.d7.loss_cls_ce: 1.9121 decode.d7.loss_mask_ce: 0.8099 decode.d7.loss_mask_dice: 1.6211 2023/09/06 18:18:18 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:18:18 - mmengine - INFO - Iter(train) [30000/60000] base_lr: 5.0001e-05 lr: 5.0001e-05 eta: 3:45:46 time: 0.4533 data_time: 0.0242 memory: 15754 grad_norm: 17.5341 loss: 9.2248 decode.loss_cls_ce: 2.0537 decode.loss_mask_ce: 0.8661 decode.loss_mask_dice: 1.6935 decode.d7.loss_cls_ce: 2.0611 decode.d7.loss_mask_ce: 0.8693 decode.d7.loss_mask_dice: 1.6811 2023/09/06 18:18:18 - mmengine - INFO - Saving checkpoint at 30000 iterations 2023/09/06 18:18:44 - mmengine - INFO - Iter(train) [30050/60000] base_lr: 4.9917e-05 lr: 4.9917e-05 eta: 3:45:26 time: 0.4539 data_time: 0.0237 memory: 15787 grad_norm: 16.7223 loss: 8.9366 decode.loss_cls_ce: 1.8680 decode.loss_mask_ce: 0.9400 decode.loss_mask_dice: 1.6644 decode.d7.loss_cls_ce: 1.8703 decode.d7.loss_mask_ce: 0.9412 decode.d7.loss_mask_dice: 1.6527 2023/09/06 18:19:07 - mmengine - INFO - Iter(train) [30100/60000] base_lr: 4.9834e-05 lr: 4.9834e-05 eta: 3:45:04 time: 0.4586 data_time: 0.0235 memory: 16025 grad_norm: 15.9309 loss: 9.0135 decode.loss_cls_ce: 1.9685 decode.loss_mask_ce: 0.8649 decode.loss_mask_dice: 1.6508 decode.d7.loss_cls_ce: 2.0161 decode.d7.loss_mask_ce: 0.8711 decode.d7.loss_mask_dice: 1.6421 2023/09/06 18:19:30 - mmengine - INFO - Iter(train) [30150/60000] base_lr: 4.9751e-05 lr: 4.9751e-05 eta: 3:44:42 time: 0.4580 data_time: 0.0232 memory: 15935 grad_norm: 19.5912 loss: 9.8132 decode.loss_cls_ce: 2.1005 decode.loss_mask_ce: 0.9291 decode.loss_mask_dice: 1.8560 decode.d7.loss_cls_ce: 2.1333 decode.d7.loss_mask_ce: 0.9270 decode.d7.loss_mask_dice: 1.8674 2023/09/06 18:19:53 - mmengine - INFO - Iter(train) [30200/60000] base_lr: 4.9667e-05 lr: 4.9667e-05 eta: 3:44:19 time: 0.4565 data_time: 0.0226 memory: 15717 grad_norm: 17.4372 loss: 8.9291 decode.loss_cls_ce: 1.9166 decode.loss_mask_ce: 0.8992 decode.loss_mask_dice: 1.6406 decode.d7.loss_cls_ce: 1.9207 decode.d7.loss_mask_ce: 0.9037 decode.d7.loss_mask_dice: 1.6483 2023/09/06 18:20:15 - mmengine - INFO - Iter(train) [30250/60000] base_lr: 4.9584e-05 lr: 4.9584e-05 eta: 3:43:57 time: 0.4549 data_time: 0.0231 memory: 15858 grad_norm: 16.6604 loss: 9.3164 decode.loss_cls_ce: 2.0136 decode.loss_mask_ce: 0.8837 decode.loss_mask_dice: 1.7598 decode.d7.loss_cls_ce: 2.0148 decode.d7.loss_mask_ce: 0.8887 decode.d7.loss_mask_dice: 1.7558 2023/09/06 18:20:38 - mmengine - INFO - Iter(train) [30300/60000] base_lr: 4.9501e-05 lr: 4.9501e-05 eta: 3:43:34 time: 0.4523 data_time: 0.0252 memory: 15744 grad_norm: 17.6614 loss: 9.6315 decode.loss_cls_ce: 2.0857 decode.loss_mask_ce: 0.8770 decode.loss_mask_dice: 1.8488 decode.d7.loss_cls_ce: 2.0635 decode.d7.loss_mask_ce: 0.8831 decode.d7.loss_mask_dice: 1.8735 2023/09/06 18:21:01 - mmengine - INFO - Iter(train) [30350/60000] base_lr: 4.9417e-05 lr: 4.9417e-05 eta: 3:43:12 time: 0.4499 data_time: 0.0238 memory: 15694 grad_norm: 17.1201 loss: 9.4952 decode.loss_cls_ce: 2.0501 decode.loss_mask_ce: 0.9090 decode.loss_mask_dice: 1.7927 decode.d7.loss_cls_ce: 2.0561 decode.d7.loss_mask_ce: 0.9012 decode.d7.loss_mask_dice: 1.7860 2023/09/06 18:21:24 - mmengine - INFO - Iter(train) [30400/60000] base_lr: 4.9334e-05 lr: 4.9334e-05 eta: 3:42:50 time: 0.4598 data_time: 0.0229 memory: 16017 grad_norm: 17.2418 loss: 9.0247 decode.loss_cls_ce: 2.0202 decode.loss_mask_ce: 0.8145 decode.loss_mask_dice: 1.6670 decode.d7.loss_cls_ce: 2.0092 decode.d7.loss_mask_ce: 0.8229 decode.d7.loss_mask_dice: 1.6909 2023/09/06 18:21:47 - mmengine - INFO - Iter(train) [30450/60000] base_lr: 4.9251e-05 lr: 4.9251e-05 eta: 3:42:27 time: 0.4587 data_time: 0.0233 memory: 15848 grad_norm: 18.2186 loss: 9.3957 decode.loss_cls_ce: 2.0363 decode.loss_mask_ce: 0.8988 decode.loss_mask_dice: 1.7625 decode.d7.loss_cls_ce: 2.0086 decode.d7.loss_mask_ce: 0.9096 decode.d7.loss_mask_dice: 1.7798 2023/09/06 18:22:09 - mmengine - INFO - Iter(train) [30500/60000] base_lr: 4.9167e-05 lr: 4.9167e-05 eta: 3:42:05 time: 0.4502 data_time: 0.0242 memory: 15771 grad_norm: 16.1643 loss: 9.5487 decode.loss_cls_ce: 2.0573 decode.loss_mask_ce: 0.9269 decode.loss_mask_dice: 1.7844 decode.d7.loss_cls_ce: 2.0483 decode.d7.loss_mask_ce: 0.9304 decode.d7.loss_mask_dice: 1.8013 2023/09/06 18:22:32 - mmengine - INFO - Iter(train) [30550/60000] base_lr: 4.9084e-05 lr: 4.9084e-05 eta: 3:41:42 time: 0.4589 data_time: 0.0239 memory: 15949 grad_norm: 16.1359 loss: 8.4586 decode.loss_cls_ce: 1.8558 decode.loss_mask_ce: 0.8328 decode.loss_mask_dice: 1.5380 decode.d7.loss_cls_ce: 1.8578 decode.d7.loss_mask_ce: 0.8337 decode.d7.loss_mask_dice: 1.5405 2023/09/06 18:22:55 - mmengine - INFO - Iter(train) [30600/60000] base_lr: 4.9001e-05 lr: 4.9001e-05 eta: 3:41:20 time: 0.4506 data_time: 0.0239 memory: 15898 grad_norm: 15.5436 loss: 9.7818 decode.loss_cls_ce: 2.0175 decode.loss_mask_ce: 0.9235 decode.loss_mask_dice: 1.9438 decode.d7.loss_cls_ce: 2.0134 decode.d7.loss_mask_ce: 0.9282 decode.d7.loss_mask_dice: 1.9555 2023/09/06 18:23:17 - mmengine - INFO - Iter(train) [30650/60000] base_lr: 4.8917e-05 lr: 4.8917e-05 eta: 3:40:57 time: 0.4579 data_time: 0.0230 memory: 15797 grad_norm: 15.9312 loss: 8.5381 decode.loss_cls_ce: 1.9355 decode.loss_mask_ce: 0.8558 decode.loss_mask_dice: 1.4668 decode.d7.loss_cls_ce: 1.9552 decode.d7.loss_mask_ce: 0.8607 decode.d7.loss_mask_dice: 1.4641 2023/09/06 18:23:40 - mmengine - INFO - Iter(train) [30700/60000] base_lr: 4.8834e-05 lr: 4.8834e-05 eta: 3:40:35 time: 0.4587 data_time: 0.0231 memory: 15805 grad_norm: 16.7953 loss: 8.4750 decode.loss_cls_ce: 1.8690 decode.loss_mask_ce: 0.8381 decode.loss_mask_dice: 1.5220 decode.d7.loss_cls_ce: 1.8790 decode.d7.loss_mask_ce: 0.8462 decode.d7.loss_mask_dice: 1.5207 2023/09/06 18:24:03 - mmengine - INFO - Iter(train) [30750/60000] base_lr: 4.8751e-05 lr: 4.8751e-05 eta: 3:40:13 time: 0.4500 data_time: 0.0241 memory: 15873 grad_norm: 17.9708 loss: 10.0206 decode.loss_cls_ce: 2.2321 decode.loss_mask_ce: 0.9484 decode.loss_mask_dice: 1.8285 decode.d7.loss_cls_ce: 2.2150 decode.d7.loss_mask_ce: 0.9561 decode.d7.loss_mask_dice: 1.8405 2023/09/06 18:24:26 - mmengine - INFO - Iter(train) [30800/60000] base_lr: 4.8667e-05 lr: 4.8667e-05 eta: 3:39:50 time: 0.4515 data_time: 0.0248 memory: 15741 grad_norm: 16.7785 loss: 9.4486 decode.loss_cls_ce: 2.0829 decode.loss_mask_ce: 0.8806 decode.loss_mask_dice: 1.7446 decode.d7.loss_cls_ce: 2.1020 decode.d7.loss_mask_ce: 0.8847 decode.d7.loss_mask_dice: 1.7537 2023/09/06 18:24:48 - mmengine - INFO - Iter(train) [30850/60000] base_lr: 4.8584e-05 lr: 4.8584e-05 eta: 3:39:28 time: 0.4597 data_time: 0.0228 memory: 15963 grad_norm: 16.9367 loss: 8.4758 decode.loss_cls_ce: 1.8869 decode.loss_mask_ce: 0.8001 decode.loss_mask_dice: 1.5499 decode.d7.loss_cls_ce: 1.8929 decode.d7.loss_mask_ce: 0.8076 decode.d7.loss_mask_dice: 1.5383 2023/09/06 18:25:11 - mmengine - INFO - Iter(train) [30900/60000] base_lr: 4.8501e-05 lr: 4.8501e-05 eta: 3:39:05 time: 0.4594 data_time: 0.0233 memory: 15823 grad_norm: 18.3123 loss: 9.4169 decode.loss_cls_ce: 2.1279 decode.loss_mask_ce: 0.8711 decode.loss_mask_dice: 1.7115 decode.d7.loss_cls_ce: 2.1582 decode.d7.loss_mask_ce: 0.8639 decode.d7.loss_mask_dice: 1.6844 2023/09/06 18:25:34 - mmengine - INFO - Iter(train) [30950/60000] base_lr: 4.8417e-05 lr: 4.8417e-05 eta: 3:38:43 time: 0.4496 data_time: 0.0235 memory: 16092 grad_norm: 17.7511 loss: 10.0436 decode.loss_cls_ce: 2.1468 decode.loss_mask_ce: 0.9794 decode.loss_mask_dice: 1.8907 decode.d7.loss_cls_ce: 2.1685 decode.d7.loss_mask_ce: 0.9765 decode.d7.loss_mask_dice: 1.8817 2023/09/06 18:25:57 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:25:57 - mmengine - INFO - Iter(train) [31000/60000] base_lr: 4.8334e-05 lr: 4.8334e-05 eta: 3:38:20 time: 0.4513 data_time: 0.0246 memory: 15859 grad_norm: 17.4391 loss: 8.9189 decode.loss_cls_ce: 1.9724 decode.loss_mask_ce: 0.8078 decode.loss_mask_dice: 1.6780 decode.d7.loss_cls_ce: 1.9731 decode.d7.loss_mask_ce: 0.8054 decode.d7.loss_mask_dice: 1.6822 2023/09/06 18:26:19 - mmengine - INFO - Iter(train) [31050/60000] base_lr: 4.8251e-05 lr: 4.8251e-05 eta: 3:37:58 time: 0.4503 data_time: 0.0243 memory: 15862 grad_norm: 17.0191 loss: 9.3611 decode.loss_cls_ce: 2.0828 decode.loss_mask_ce: 0.8279 decode.loss_mask_dice: 1.7544 decode.d7.loss_cls_ce: 2.1011 decode.d7.loss_mask_ce: 0.8329 decode.d7.loss_mask_dice: 1.7620 2023/09/06 18:26:42 - mmengine - INFO - Iter(train) [31100/60000] base_lr: 4.8167e-05 lr: 4.8167e-05 eta: 3:37:35 time: 0.4612 data_time: 0.0240 memory: 15835 grad_norm: 16.4910 loss: 9.6312 decode.loss_cls_ce: 1.9997 decode.loss_mask_ce: 0.9245 decode.loss_mask_dice: 1.8893 decode.d7.loss_cls_ce: 2.0238 decode.d7.loss_mask_ce: 0.9168 decode.d7.loss_mask_dice: 1.8771 2023/09/06 18:27:05 - mmengine - INFO - Iter(train) [31150/60000] base_lr: 4.8084e-05 lr: 4.8084e-05 eta: 3:37:13 time: 0.4606 data_time: 0.0232 memory: 15757 grad_norm: 17.2510 loss: 8.7135 decode.loss_cls_ce: 1.8911 decode.loss_mask_ce: 0.8529 decode.loss_mask_dice: 1.5983 decode.d7.loss_cls_ce: 1.9109 decode.d7.loss_mask_ce: 0.8559 decode.d7.loss_mask_dice: 1.6044 2023/09/06 18:27:28 - mmengine - INFO - Iter(train) [31200/60000] base_lr: 4.8001e-05 lr: 4.8001e-05 eta: 3:36:51 time: 0.4592 data_time: 0.0229 memory: 15809 grad_norm: 18.2580 loss: 9.8256 decode.loss_cls_ce: 2.2445 decode.loss_mask_ce: 0.8607 decode.loss_mask_dice: 1.8051 decode.d7.loss_cls_ce: 2.2426 decode.d7.loss_mask_ce: 0.8520 decode.d7.loss_mask_dice: 1.8206 2023/09/06 18:27:51 - mmengine - INFO - Iter(train) [31250/60000] base_lr: 4.7917e-05 lr: 4.7917e-05 eta: 3:36:29 time: 0.4582 data_time: 0.0234 memory: 15819 grad_norm: 15.1010 loss: 9.4553 decode.loss_cls_ce: 2.0462 decode.loss_mask_ce: 0.8973 decode.loss_mask_dice: 1.7879 decode.d7.loss_cls_ce: 2.0524 decode.d7.loss_mask_ce: 0.8988 decode.d7.loss_mask_dice: 1.7726 2023/09/06 18:28:14 - mmengine - INFO - Iter(train) [31300/60000] base_lr: 4.7834e-05 lr: 4.7834e-05 eta: 3:36:06 time: 0.4510 data_time: 0.0241 memory: 15808 grad_norm: 17.9844 loss: 9.2289 decode.loss_cls_ce: 1.9308 decode.loss_mask_ce: 0.8904 decode.loss_mask_dice: 1.7890 decode.d7.loss_cls_ce: 1.9478 decode.d7.loss_mask_ce: 0.8878 decode.d7.loss_mask_dice: 1.7832 2023/09/06 18:28:36 - mmengine - INFO - Iter(train) [31350/60000] base_lr: 4.7751e-05 lr: 4.7751e-05 eta: 3:35:43 time: 0.4512 data_time: 0.0244 memory: 15818 grad_norm: 17.7908 loss: 9.1993 decode.loss_cls_ce: 2.0083 decode.loss_mask_ce: 0.8629 decode.loss_mask_dice: 1.7245 decode.d7.loss_cls_ce: 2.0195 decode.d7.loss_mask_ce: 0.8638 decode.d7.loss_mask_dice: 1.7202 2023/09/06 18:28:59 - mmengine - INFO - Iter(train) [31400/60000] base_lr: 4.7667e-05 lr: 4.7667e-05 eta: 3:35:21 time: 0.4617 data_time: 0.0233 memory: 15821 grad_norm: 18.5069 loss: 9.1943 decode.loss_cls_ce: 1.9603 decode.loss_mask_ce: 0.9017 decode.loss_mask_dice: 1.7269 decode.d7.loss_cls_ce: 1.9771 decode.d7.loss_mask_ce: 0.9018 decode.d7.loss_mask_dice: 1.7265 2023/09/06 18:29:22 - mmengine - INFO - Iter(train) [31450/60000] base_lr: 4.7584e-05 lr: 4.7584e-05 eta: 3:34:58 time: 0.4514 data_time: 0.0244 memory: 15781 grad_norm: 16.5941 loss: 9.1137 decode.loss_cls_ce: 2.0072 decode.loss_mask_ce: 0.8709 decode.loss_mask_dice: 1.6848 decode.d7.loss_cls_ce: 1.9883 decode.d7.loss_mask_ce: 0.8687 decode.d7.loss_mask_dice: 1.6937 2023/09/06 18:29:44 - mmengine - INFO - Iter(train) [31500/60000] base_lr: 4.7501e-05 lr: 4.7501e-05 eta: 3:34:36 time: 0.4597 data_time: 0.0230 memory: 15986 grad_norm: 16.8258 loss: 8.7122 decode.loss_cls_ce: 1.8806 decode.loss_mask_ce: 0.8558 decode.loss_mask_dice: 1.6023 decode.d7.loss_cls_ce: 1.8924 decode.d7.loss_mask_ce: 0.8586 decode.d7.loss_mask_dice: 1.6224 2023/09/06 18:30:07 - mmengine - INFO - Iter(train) [31550/60000] base_lr: 4.7417e-05 lr: 4.7417e-05 eta: 3:34:14 time: 0.4525 data_time: 0.0246 memory: 15807 grad_norm: 16.7136 loss: 8.3431 decode.loss_cls_ce: 1.9736 decode.loss_mask_ce: 0.7861 decode.loss_mask_dice: 1.4026 decode.d7.loss_cls_ce: 1.9594 decode.d7.loss_mask_ce: 0.7859 decode.d7.loss_mask_dice: 1.4354 2023/09/06 18:30:30 - mmengine - INFO - Iter(train) [31600/60000] base_lr: 4.7334e-05 lr: 4.7334e-05 eta: 3:33:51 time: 0.4519 data_time: 0.0233 memory: 15989 grad_norm: 15.9375 loss: 10.0808 decode.loss_cls_ce: 2.0930 decode.loss_mask_ce: 0.9208 decode.loss_mask_dice: 1.9955 decode.d7.loss_cls_ce: 2.1501 decode.d7.loss_mask_ce: 0.9205 decode.d7.loss_mask_dice: 2.0009 2023/09/06 18:30:53 - mmengine - INFO - Iter(train) [31650/60000] base_lr: 4.7251e-05 lr: 4.7251e-05 eta: 3:33:29 time: 0.4529 data_time: 0.0252 memory: 15705 grad_norm: 17.3117 loss: 8.6391 decode.loss_cls_ce: 1.9117 decode.loss_mask_ce: 0.9006 decode.loss_mask_dice: 1.5041 decode.d7.loss_cls_ce: 1.9285 decode.d7.loss_mask_ce: 0.8907 decode.d7.loss_mask_dice: 1.5035 2023/09/06 18:31:15 - mmengine - INFO - Iter(train) [31700/60000] base_lr: 4.7167e-05 lr: 4.7167e-05 eta: 3:33:06 time: 0.4527 data_time: 0.0247 memory: 15861 grad_norm: 16.7369 loss: 9.7801 decode.loss_cls_ce: 2.1718 decode.loss_mask_ce: 0.9242 decode.loss_mask_dice: 1.7850 decode.d7.loss_cls_ce: 2.1666 decode.d7.loss_mask_ce: 0.9246 decode.d7.loss_mask_dice: 1.8079 2023/09/06 18:31:38 - mmengine - INFO - Iter(train) [31750/60000] base_lr: 4.7084e-05 lr: 4.7084e-05 eta: 3:32:43 time: 0.4541 data_time: 0.0234 memory: 15911 grad_norm: 17.4909 loss: 8.2200 decode.loss_cls_ce: 1.7361 decode.loss_mask_ce: 0.8501 decode.loss_mask_dice: 1.5274 decode.d7.loss_cls_ce: 1.7107 decode.d7.loss_mask_ce: 0.8713 decode.d7.loss_mask_dice: 1.5245 2023/09/06 18:32:00 - mmengine - INFO - Iter(train) [31800/60000] base_lr: 4.7001e-05 lr: 4.7001e-05 eta: 3:32:21 time: 0.4483 data_time: 0.0238 memory: 16145 grad_norm: 17.0650 loss: 9.8966 decode.loss_cls_ce: 2.1942 decode.loss_mask_ce: 0.9327 decode.loss_mask_dice: 1.8061 decode.d7.loss_cls_ce: 2.2173 decode.d7.loss_mask_ce: 0.9420 decode.d7.loss_mask_dice: 1.8044 2023/09/06 18:32:23 - mmengine - INFO - Iter(train) [31850/60000] base_lr: 4.6917e-05 lr: 4.6917e-05 eta: 3:31:59 time: 0.4543 data_time: 0.0250 memory: 15758 grad_norm: 17.2279 loss: 9.3293 decode.loss_cls_ce: 2.0970 decode.loss_mask_ce: 0.9069 decode.loss_mask_dice: 1.6564 decode.d7.loss_cls_ce: 2.1074 decode.d7.loss_mask_ce: 0.9042 decode.d7.loss_mask_dice: 1.6575 2023/09/06 18:32:46 - mmengine - INFO - Iter(train) [31900/60000] base_lr: 4.6834e-05 lr: 4.6834e-05 eta: 3:31:36 time: 0.4613 data_time: 0.0236 memory: 15895 grad_norm: 17.7825 loss: 8.7760 decode.loss_cls_ce: 1.9401 decode.loss_mask_ce: 0.8497 decode.loss_mask_dice: 1.5980 decode.d7.loss_cls_ce: 1.9454 decode.d7.loss_mask_ce: 0.8487 decode.d7.loss_mask_dice: 1.5941 2023/09/06 18:33:09 - mmengine - INFO - Iter(train) [31950/60000] base_lr: 4.6751e-05 lr: 4.6751e-05 eta: 3:31:14 time: 0.4552 data_time: 0.0234 memory: 15871 grad_norm: 16.6430 loss: 9.8310 decode.loss_cls_ce: 2.2410 decode.loss_mask_ce: 0.8700 decode.loss_mask_dice: 1.7973 decode.d7.loss_cls_ce: 2.2487 decode.d7.loss_mask_ce: 0.8601 decode.d7.loss_mask_dice: 1.8138 2023/09/06 18:33:32 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:33:32 - mmengine - INFO - Iter(train) [32000/60000] base_lr: 4.6667e-05 lr: 4.6667e-05 eta: 3:30:51 time: 0.4621 data_time: 0.0238 memory: 15743 grad_norm: 16.9797 loss: 9.0509 decode.loss_cls_ce: 1.9728 decode.loss_mask_ce: 0.8895 decode.loss_mask_dice: 1.6741 decode.d7.loss_cls_ce: 1.9399 decode.d7.loss_mask_ce: 0.9057 decode.d7.loss_mask_dice: 1.6689 2023/09/06 18:33:55 - mmengine - INFO - Iter(train) [32050/60000] base_lr: 4.6584e-05 lr: 4.6584e-05 eta: 3:30:29 time: 0.4589 data_time: 0.0236 memory: 15819 grad_norm: 16.7789 loss: 8.6485 decode.loss_cls_ce: 1.9934 decode.loss_mask_ce: 0.8651 decode.loss_mask_dice: 1.4652 decode.d7.loss_cls_ce: 1.9971 decode.d7.loss_mask_ce: 0.8641 decode.d7.loss_mask_dice: 1.4636 2023/09/06 18:34:18 - mmengine - INFO - Iter(train) [32100/60000] base_lr: 4.6501e-05 lr: 4.6501e-05 eta: 3:30:07 time: 0.4589 data_time: 0.0223 memory: 15808 grad_norm: 16.1540 loss: 9.4447 decode.loss_cls_ce: 2.1367 decode.loss_mask_ce: 0.8989 decode.loss_mask_dice: 1.6995 decode.d7.loss_cls_ce: 2.1284 decode.d7.loss_mask_ce: 0.8897 decode.d7.loss_mask_dice: 1.6914 2023/09/06 18:34:40 - mmengine - INFO - Iter(train) [32150/60000] base_lr: 4.6417e-05 lr: 4.6417e-05 eta: 3:29:44 time: 0.4499 data_time: 0.0242 memory: 15898 grad_norm: 17.9979 loss: 9.5169 decode.loss_cls_ce: 2.0361 decode.loss_mask_ce: 0.9369 decode.loss_mask_dice: 1.7957 decode.d7.loss_cls_ce: 2.0212 decode.d7.loss_mask_ce: 0.9333 decode.d7.loss_mask_dice: 1.7937 2023/09/06 18:35:03 - mmengine - INFO - Iter(train) [32200/60000] base_lr: 4.6334e-05 lr: 4.6334e-05 eta: 3:29:22 time: 0.4495 data_time: 0.0234 memory: 15940 grad_norm: 18.4739 loss: 9.5462 decode.loss_cls_ce: 2.1381 decode.loss_mask_ce: 0.9061 decode.loss_mask_dice: 1.7193 decode.d7.loss_cls_ce: 2.1423 decode.d7.loss_mask_ce: 0.9038 decode.d7.loss_mask_dice: 1.7366 2023/09/06 18:35:26 - mmengine - INFO - Iter(train) [32250/60000] base_lr: 4.6251e-05 lr: 4.6251e-05 eta: 3:28:59 time: 0.4595 data_time: 0.0235 memory: 15797 grad_norm: 16.2905 loss: 9.3170 decode.loss_cls_ce: 2.0322 decode.loss_mask_ce: 0.8737 decode.loss_mask_dice: 1.7614 decode.d7.loss_cls_ce: 2.0089 decode.d7.loss_mask_ce: 0.8805 decode.d7.loss_mask_dice: 1.7602 2023/09/06 18:35:48 - mmengine - INFO - Iter(train) [32300/60000] base_lr: 4.6167e-05 lr: 4.6167e-05 eta: 3:28:37 time: 0.4517 data_time: 0.0232 memory: 15978 grad_norm: 16.7877 loss: 9.8736 decode.loss_cls_ce: 2.1398 decode.loss_mask_ce: 0.8975 decode.loss_mask_dice: 1.9002 decode.d7.loss_cls_ce: 2.1270 decode.d7.loss_mask_ce: 0.9024 decode.d7.loss_mask_dice: 1.9068 2023/09/06 18:36:11 - mmengine - INFO - Iter(train) [32350/60000] base_lr: 4.6084e-05 lr: 4.6084e-05 eta: 3:28:14 time: 0.4531 data_time: 0.0234 memory: 16029 grad_norm: 16.1647 loss: 9.0964 decode.loss_cls_ce: 2.0116 decode.loss_mask_ce: 0.8771 decode.loss_mask_dice: 1.6652 decode.d7.loss_cls_ce: 2.0055 decode.d7.loss_mask_ce: 0.8752 decode.d7.loss_mask_dice: 1.6619 2023/09/06 18:36:34 - mmengine - INFO - Iter(train) [32400/60000] base_lr: 4.6001e-05 lr: 4.6001e-05 eta: 3:27:52 time: 0.4508 data_time: 0.0236 memory: 15872 grad_norm: 16.4057 loss: 8.6177 decode.loss_cls_ce: 1.8763 decode.loss_mask_ce: 0.8505 decode.loss_mask_dice: 1.5672 decode.d7.loss_cls_ce: 1.8742 decode.d7.loss_mask_ce: 0.8614 decode.d7.loss_mask_dice: 1.5882 2023/09/06 18:36:56 - mmengine - INFO - Iter(train) [32450/60000] base_lr: 4.5917e-05 lr: 4.5917e-05 eta: 3:27:29 time: 0.4493 data_time: 0.0237 memory: 15949 grad_norm: 17.4718 loss: 10.1777 decode.loss_cls_ce: 2.1872 decode.loss_mask_ce: 0.9336 decode.loss_mask_dice: 1.9636 decode.d7.loss_cls_ce: 2.1850 decode.d7.loss_mask_ce: 0.9302 decode.d7.loss_mask_dice: 1.9781 2023/09/06 18:37:19 - mmengine - INFO - Iter(train) [32500/60000] base_lr: 4.5834e-05 lr: 4.5834e-05 eta: 3:27:07 time: 0.4608 data_time: 0.0230 memory: 15908 grad_norm: 16.3243 loss: 9.6107 decode.loss_cls_ce: 2.0254 decode.loss_mask_ce: 0.9230 decode.loss_mask_dice: 1.8563 decode.d7.loss_cls_ce: 2.0300 decode.d7.loss_mask_ce: 0.9280 decode.d7.loss_mask_dice: 1.8480 2023/09/06 18:37:42 - mmengine - INFO - Iter(train) [32550/60000] base_lr: 4.5751e-05 lr: 4.5751e-05 eta: 3:26:44 time: 0.4608 data_time: 0.0234 memory: 16117 grad_norm: 17.1038 loss: 9.2023 decode.loss_cls_ce: 1.9904 decode.loss_mask_ce: 0.9045 decode.loss_mask_dice: 1.7075 decode.d7.loss_cls_ce: 1.9948 decode.d7.loss_mask_ce: 0.8988 decode.d7.loss_mask_dice: 1.7064 2023/09/06 18:38:05 - mmengine - INFO - Iter(train) [32600/60000] base_lr: 4.5667e-05 lr: 4.5667e-05 eta: 3:26:22 time: 0.4509 data_time: 0.0241 memory: 15845 grad_norm: 16.3953 loss: 8.8786 decode.loss_cls_ce: 2.0145 decode.loss_mask_ce: 0.8341 decode.loss_mask_dice: 1.5792 decode.d7.loss_cls_ce: 2.0413 decode.d7.loss_mask_ce: 0.8330 decode.d7.loss_mask_dice: 1.5764 2023/09/06 18:38:28 - mmengine - INFO - Iter(train) [32650/60000] base_lr: 4.5584e-05 lr: 4.5584e-05 eta: 3:26:00 time: 0.4606 data_time: 0.0229 memory: 15797 grad_norm: 17.4127 loss: 9.4503 decode.loss_cls_ce: 2.0059 decode.loss_mask_ce: 0.9357 decode.loss_mask_dice: 1.7795 decode.d7.loss_cls_ce: 1.9978 decode.d7.loss_mask_ce: 0.9375 decode.d7.loss_mask_dice: 1.7939 2023/09/06 18:38:51 - mmengine - INFO - Iter(train) [32700/60000] base_lr: 4.5501e-05 lr: 4.5501e-05 eta: 3:25:37 time: 0.4499 data_time: 0.0241 memory: 15988 grad_norm: 18.3310 loss: 8.7395 decode.loss_cls_ce: 1.8288 decode.loss_mask_ce: 0.8989 decode.loss_mask_dice: 1.6326 decode.d7.loss_cls_ce: 1.8287 decode.d7.loss_mask_ce: 0.9075 decode.d7.loss_mask_dice: 1.6429 2023/09/06 18:39:14 - mmengine - INFO - Iter(train) [32750/60000] base_lr: 4.5417e-05 lr: 4.5417e-05 eta: 3:25:15 time: 0.4496 data_time: 0.0246 memory: 15964 grad_norm: 17.8355 loss: 9.0918 decode.loss_cls_ce: 2.0402 decode.loss_mask_ce: 0.8819 decode.loss_mask_dice: 1.6123 decode.d7.loss_cls_ce: 2.0435 decode.d7.loss_mask_ce: 0.8879 decode.d7.loss_mask_dice: 1.6260 2023/09/06 18:39:36 - mmengine - INFO - Iter(train) [32800/60000] base_lr: 4.5334e-05 lr: 4.5334e-05 eta: 3:24:52 time: 0.4516 data_time: 0.0244 memory: 15823 grad_norm: 16.6022 loss: 9.2547 decode.loss_cls_ce: 2.0486 decode.loss_mask_ce: 0.9088 decode.loss_mask_dice: 1.6775 decode.d7.loss_cls_ce: 2.0218 decode.d7.loss_mask_ce: 0.9127 decode.d7.loss_mask_dice: 1.6853 2023/09/06 18:39:59 - mmengine - INFO - Iter(train) [32850/60000] base_lr: 4.5251e-05 lr: 4.5251e-05 eta: 3:24:30 time: 0.4623 data_time: 0.0229 memory: 15921 grad_norm: 17.4074 loss: 8.8388 decode.loss_cls_ce: 1.8913 decode.loss_mask_ce: 0.8765 decode.loss_mask_dice: 1.6426 decode.d7.loss_cls_ce: 1.9064 decode.d7.loss_mask_ce: 0.8739 decode.d7.loss_mask_dice: 1.6482 2023/09/06 18:40:22 - mmengine - INFO - Iter(train) [32900/60000] base_lr: 4.5167e-05 lr: 4.5167e-05 eta: 3:24:07 time: 0.4509 data_time: 0.0238 memory: 15847 grad_norm: 17.2613 loss: 8.9249 decode.loss_cls_ce: 2.1182 decode.loss_mask_ce: 0.8073 decode.loss_mask_dice: 1.5116 decode.d7.loss_cls_ce: 2.1750 decode.d7.loss_mask_ce: 0.8033 decode.d7.loss_mask_dice: 1.5096 2023/09/06 18:40:45 - mmengine - INFO - Iter(train) [32950/60000] base_lr: 4.5084e-05 lr: 4.5084e-05 eta: 3:23:45 time: 0.4590 data_time: 0.0231 memory: 15954 grad_norm: 18.4269 loss: 10.2478 decode.loss_cls_ce: 2.3065 decode.loss_mask_ce: 0.8891 decode.loss_mask_dice: 1.9371 decode.d7.loss_cls_ce: 2.2781 decode.d7.loss_mask_ce: 0.8910 decode.d7.loss_mask_dice: 1.9460 2023/09/06 18:41:07 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:41:07 - mmengine - INFO - Iter(train) [33000/60000] base_lr: 4.5001e-05 lr: 4.5001e-05 eta: 3:23:22 time: 0.4496 data_time: 0.0239 memory: 15859 grad_norm: 16.9767 loss: 8.3409 decode.loss_cls_ce: 1.8513 decode.loss_mask_ce: 0.8192 decode.loss_mask_dice: 1.5010 decode.d7.loss_cls_ce: 1.8616 decode.d7.loss_mask_ce: 0.8138 decode.d7.loss_mask_dice: 1.4939 2023/09/06 18:41:30 - mmengine - INFO - Iter(train) [33050/60000] base_lr: 4.4917e-05 lr: 4.4917e-05 eta: 3:23:00 time: 0.4509 data_time: 0.0242 memory: 15859 grad_norm: 17.9826 loss: 9.7217 decode.loss_cls_ce: 2.1396 decode.loss_mask_ce: 0.8830 decode.loss_mask_dice: 1.8429 decode.d7.loss_cls_ce: 2.1448 decode.d7.loss_mask_ce: 0.8732 decode.d7.loss_mask_dice: 1.8381 2023/09/06 18:41:53 - mmengine - INFO - Iter(train) [33100/60000] base_lr: 4.4834e-05 lr: 4.4834e-05 eta: 3:22:37 time: 0.4600 data_time: 0.0229 memory: 15924 grad_norm: 17.9175 loss: 8.6300 decode.loss_cls_ce: 1.8084 decode.loss_mask_ce: 0.8408 decode.loss_mask_dice: 1.6603 decode.d7.loss_cls_ce: 1.8203 decode.d7.loss_mask_ce: 0.8360 decode.d7.loss_mask_dice: 1.6642 2023/09/06 18:42:16 - mmengine - INFO - Iter(train) [33150/60000] base_lr: 4.4751e-05 lr: 4.4751e-05 eta: 3:22:15 time: 0.4512 data_time: 0.0241 memory: 15769 grad_norm: 17.6220 loss: 9.5691 decode.loss_cls_ce: 2.1012 decode.loss_mask_ce: 0.9335 decode.loss_mask_dice: 1.7410 decode.d7.loss_cls_ce: 2.1139 decode.d7.loss_mask_ce: 0.9317 decode.d7.loss_mask_dice: 1.7479 2023/09/06 18:42:38 - mmengine - INFO - Iter(train) [33200/60000] base_lr: 4.4667e-05 lr: 4.4667e-05 eta: 3:21:52 time: 0.4517 data_time: 0.0238 memory: 15836 grad_norm: 16.5909 loss: 9.8234 decode.loss_cls_ce: 2.1354 decode.loss_mask_ce: 0.8865 decode.loss_mask_dice: 1.8791 decode.d7.loss_cls_ce: 2.1553 decode.d7.loss_mask_ce: 0.8849 decode.d7.loss_mask_dice: 1.8823 2023/09/06 18:43:01 - mmengine - INFO - Iter(train) [33250/60000] base_lr: 4.4584e-05 lr: 4.4584e-05 eta: 3:21:30 time: 0.4608 data_time: 0.0232 memory: 15887 grad_norm: 18.2826 loss: 10.1220 decode.loss_cls_ce: 2.1655 decode.loss_mask_ce: 0.9288 decode.loss_mask_dice: 1.9533 decode.d7.loss_cls_ce: 2.1768 decode.d7.loss_mask_ce: 0.9381 decode.d7.loss_mask_dice: 1.9596 2023/09/06 18:43:24 - mmengine - INFO - Iter(train) [33300/60000] base_lr: 4.4501e-05 lr: 4.4501e-05 eta: 3:21:08 time: 0.4530 data_time: 0.0234 memory: 15888 grad_norm: 17.5050 loss: 10.6315 decode.loss_cls_ce: 2.3484 decode.loss_mask_ce: 0.9332 decode.loss_mask_dice: 2.0188 decode.d7.loss_cls_ce: 2.3758 decode.d7.loss_mask_ce: 0.9287 decode.d7.loss_mask_dice: 2.0268 2023/09/06 18:43:47 - mmengine - INFO - Iter(train) [33350/60000] base_lr: 4.4417e-05 lr: 4.4417e-05 eta: 3:20:45 time: 0.4610 data_time: 0.0238 memory: 15888 grad_norm: 18.6046 loss: 9.0623 decode.loss_cls_ce: 1.8913 decode.loss_mask_ce: 0.9223 decode.loss_mask_dice: 1.7079 decode.d7.loss_cls_ce: 1.9006 decode.d7.loss_mask_ce: 0.9203 decode.d7.loss_mask_dice: 1.7199 2023/09/06 18:44:10 - mmengine - INFO - Iter(train) [33400/60000] base_lr: 4.4334e-05 lr: 4.4334e-05 eta: 3:20:23 time: 0.4504 data_time: 0.0239 memory: 15834 grad_norm: 15.4299 loss: 8.1689 decode.loss_cls_ce: 1.8827 decode.loss_mask_ce: 0.7977 decode.loss_mask_dice: 1.4057 decode.d7.loss_cls_ce: 1.8771 decode.d7.loss_mask_ce: 0.8026 decode.d7.loss_mask_dice: 1.4030 2023/09/06 18:44:33 - mmengine - INFO - Iter(train) [33450/60000] base_lr: 4.4251e-05 lr: 4.4251e-05 eta: 3:20:01 time: 0.4620 data_time: 0.0231 memory: 15784 grad_norm: 18.1666 loss: 9.1343 decode.loss_cls_ce: 2.0327 decode.loss_mask_ce: 0.8750 decode.loss_mask_dice: 1.6624 decode.d7.loss_cls_ce: 2.0148 decode.d7.loss_mask_ce: 0.8746 decode.d7.loss_mask_dice: 1.6748 2023/09/06 18:44:56 - mmengine - INFO - Iter(train) [33500/60000] base_lr: 4.4167e-05 lr: 4.4167e-05 eta: 3:19:38 time: 0.4593 data_time: 0.0229 memory: 15775 grad_norm: 19.3608 loss: 9.0711 decode.loss_cls_ce: 1.8562 decode.loss_mask_ce: 0.8724 decode.loss_mask_dice: 1.8107 decode.d7.loss_cls_ce: 1.8375 decode.d7.loss_mask_ce: 0.8786 decode.d7.loss_mask_dice: 1.8157 2023/09/06 18:45:19 - mmengine - INFO - Iter(train) [33550/60000] base_lr: 4.4084e-05 lr: 4.4084e-05 eta: 3:19:16 time: 0.4622 data_time: 0.0235 memory: 15847 grad_norm: 17.6059 loss: 10.7373 decode.loss_cls_ce: 2.2345 decode.loss_mask_ce: 1.0574 decode.loss_mask_dice: 2.0625 decode.d7.loss_cls_ce: 2.2415 decode.d7.loss_mask_ce: 1.0640 decode.d7.loss_mask_dice: 2.0773 2023/09/06 18:45:42 - mmengine - INFO - Iter(train) [33600/60000] base_lr: 4.4001e-05 lr: 4.4001e-05 eta: 3:18:54 time: 0.4619 data_time: 0.0233 memory: 15947 grad_norm: 15.8873 loss: 9.3821 decode.loss_cls_ce: 1.9405 decode.loss_mask_ce: 0.9042 decode.loss_mask_dice: 1.8484 decode.d7.loss_cls_ce: 1.9098 decode.d7.loss_mask_ce: 0.9071 decode.d7.loss_mask_dice: 1.8721 2023/09/06 18:46:05 - mmengine - INFO - Iter(train) [33650/60000] base_lr: 4.3917e-05 lr: 4.3917e-05 eta: 3:18:31 time: 0.4561 data_time: 0.0236 memory: 15953 grad_norm: 16.3004 loss: 9.1298 decode.loss_cls_ce: 2.0121 decode.loss_mask_ce: 0.8387 decode.loss_mask_dice: 1.7106 decode.d7.loss_cls_ce: 2.0029 decode.d7.loss_mask_ce: 0.8465 decode.d7.loss_mask_dice: 1.7190 2023/09/06 18:46:27 - mmengine - INFO - Iter(train) [33700/60000] base_lr: 4.3834e-05 lr: 4.3834e-05 eta: 3:18:09 time: 0.4496 data_time: 0.0235 memory: 15859 grad_norm: 18.5066 loss: 8.5774 decode.loss_cls_ce: 1.8034 decode.loss_mask_ce: 0.8993 decode.loss_mask_dice: 1.5946 decode.d7.loss_cls_ce: 1.7852 decode.d7.loss_mask_ce: 0.8958 decode.d7.loss_mask_dice: 1.5990 2023/09/06 18:46:50 - mmengine - INFO - Iter(train) [33750/60000] base_lr: 4.3751e-05 lr: 4.3751e-05 eta: 3:17:46 time: 0.4585 data_time: 0.0232 memory: 15849 grad_norm: 19.8098 loss: 10.2925 decode.loss_cls_ce: 2.2635 decode.loss_mask_ce: 0.9605 decode.loss_mask_dice: 1.9221 decode.d7.loss_cls_ce: 2.2722 decode.d7.loss_mask_ce: 0.9629 decode.d7.loss_mask_dice: 1.9113 2023/09/06 18:47:13 - mmengine - INFO - Iter(train) [33800/60000] base_lr: 4.3667e-05 lr: 4.3667e-05 eta: 3:17:24 time: 0.4568 data_time: 0.0232 memory: 15747 grad_norm: 18.6651 loss: 9.0303 decode.loss_cls_ce: 2.0876 decode.loss_mask_ce: 0.7866 decode.loss_mask_dice: 1.6213 decode.d7.loss_cls_ce: 2.1164 decode.d7.loss_mask_ce: 0.7913 decode.d7.loss_mask_dice: 1.6271 2023/09/06 18:47:36 - mmengine - INFO - Iter(train) [33850/60000] base_lr: 4.3584e-05 lr: 4.3584e-05 eta: 3:17:01 time: 0.4514 data_time: 0.0237 memory: 15821 grad_norm: 17.0210 loss: 9.3199 decode.loss_cls_ce: 2.0354 decode.loss_mask_ce: 0.8115 decode.loss_mask_dice: 1.8025 decode.d7.loss_cls_ce: 2.0338 decode.d7.loss_mask_ce: 0.8198 decode.d7.loss_mask_dice: 1.8169 2023/09/06 18:47:59 - mmengine - INFO - Iter(train) [33900/60000] base_lr: 4.3501e-05 lr: 4.3501e-05 eta: 3:16:39 time: 0.4527 data_time: 0.0236 memory: 15900 grad_norm: 16.2375 loss: 9.9594 decode.loss_cls_ce: 2.0794 decode.loss_mask_ce: 0.9248 decode.loss_mask_dice: 1.9865 decode.d7.loss_cls_ce: 2.0352 decode.d7.loss_mask_ce: 0.9368 decode.d7.loss_mask_dice: 1.9967 2023/09/06 18:48:22 - mmengine - INFO - Iter(train) [33950/60000] base_lr: 4.3417e-05 lr: 4.3417e-05 eta: 3:16:17 time: 0.4591 data_time: 0.0235 memory: 15848 grad_norm: 16.8890 loss: 9.2435 decode.loss_cls_ce: 1.9628 decode.loss_mask_ce: 0.8960 decode.loss_mask_dice: 1.7640 decode.d7.loss_cls_ce: 1.9589 decode.d7.loss_mask_ce: 0.9004 decode.d7.loss_mask_dice: 1.7613 2023/09/06 18:48:45 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:48:45 - mmengine - INFO - Iter(train) [34000/60000] base_lr: 4.3334e-05 lr: 4.3334e-05 eta: 3:15:54 time: 0.4546 data_time: 0.0242 memory: 15823 grad_norm: 16.1502 loss: 9.1134 decode.loss_cls_ce: 2.0251 decode.loss_mask_ce: 0.9025 decode.loss_mask_dice: 1.6281 decode.d7.loss_cls_ce: 2.0321 decode.d7.loss_mask_ce: 0.8932 decode.d7.loss_mask_dice: 1.6325 2023/09/06 18:49:08 - mmengine - INFO - Iter(train) [34050/60000] base_lr: 4.3251e-05 lr: 4.3251e-05 eta: 3:15:32 time: 0.4599 data_time: 0.0233 memory: 15861 grad_norm: 16.5213 loss: 9.7126 decode.loss_cls_ce: 2.2591 decode.loss_mask_ce: 0.8724 decode.loss_mask_dice: 1.7144 decode.d7.loss_cls_ce: 2.2692 decode.d7.loss_mask_ce: 0.8778 decode.d7.loss_mask_dice: 1.7197 2023/09/06 18:49:31 - mmengine - INFO - Iter(train) [34100/60000] base_lr: 4.3167e-05 lr: 4.3167e-05 eta: 3:15:09 time: 0.4481 data_time: 0.0236 memory: 16066 grad_norm: 17.2580 loss: 9.7664 decode.loss_cls_ce: 2.1515 decode.loss_mask_ce: 0.9293 decode.loss_mask_dice: 1.7950 decode.d7.loss_cls_ce: 2.1460 decode.d7.loss_mask_ce: 0.9367 decode.d7.loss_mask_dice: 1.8079 2023/09/06 18:49:53 - mmengine - INFO - Iter(train) [34150/60000] base_lr: 4.3084e-05 lr: 4.3084e-05 eta: 3:14:47 time: 0.4598 data_time: 0.0234 memory: 15848 grad_norm: 17.1755 loss: 10.4686 decode.loss_cls_ce: 2.3805 decode.loss_mask_ce: 0.9651 decode.loss_mask_dice: 1.8899 decode.d7.loss_cls_ce: 2.3936 decode.d7.loss_mask_ce: 0.9649 decode.d7.loss_mask_dice: 1.8746 2023/09/06 18:50:16 - mmengine - INFO - Iter(train) [34200/60000] base_lr: 4.3001e-05 lr: 4.3001e-05 eta: 3:14:25 time: 0.4601 data_time: 0.0233 memory: 15781 grad_norm: 16.6123 loss: 8.8225 decode.loss_cls_ce: 1.9221 decode.loss_mask_ce: 0.8235 decode.loss_mask_dice: 1.6564 decode.d7.loss_cls_ce: 1.9374 decode.d7.loss_mask_ce: 0.8295 decode.d7.loss_mask_dice: 1.6535 2023/09/06 18:50:39 - mmengine - INFO - Iter(train) [34250/60000] base_lr: 4.2917e-05 lr: 4.2917e-05 eta: 3:14:02 time: 0.4591 data_time: 0.0237 memory: 15794 grad_norm: 16.9853 loss: 8.3771 decode.loss_cls_ce: 1.8258 decode.loss_mask_ce: 0.7996 decode.loss_mask_dice: 1.5562 decode.d7.loss_cls_ce: 1.8294 decode.d7.loss_mask_ce: 0.8029 decode.d7.loss_mask_dice: 1.5632 2023/09/06 18:51:02 - mmengine - INFO - Iter(train) [34300/60000] base_lr: 4.2834e-05 lr: 4.2834e-05 eta: 3:13:40 time: 0.4574 data_time: 0.0230 memory: 15812 grad_norm: 18.6471 loss: 9.7354 decode.loss_cls_ce: 2.2176 decode.loss_mask_ce: 0.8626 decode.loss_mask_dice: 1.7685 decode.d7.loss_cls_ce: 2.2310 decode.d7.loss_mask_ce: 0.8594 decode.d7.loss_mask_dice: 1.7963 2023/09/06 18:51:25 - mmengine - INFO - Iter(train) [34350/60000] base_lr: 4.2751e-05 lr: 4.2751e-05 eta: 3:13:18 time: 0.4584 data_time: 0.0234 memory: 15871 grad_norm: 16.5727 loss: 9.1607 decode.loss_cls_ce: 2.0143 decode.loss_mask_ce: 0.8372 decode.loss_mask_dice: 1.7048 decode.d7.loss_cls_ce: 2.0455 decode.d7.loss_mask_ce: 0.8377 decode.d7.loss_mask_dice: 1.7211 2023/09/06 18:51:48 - mmengine - INFO - Iter(train) [34400/60000] base_lr: 4.2667e-05 lr: 4.2667e-05 eta: 3:12:55 time: 0.4570 data_time: 0.0228 memory: 15834 grad_norm: nan loss: 9.7854 decode.loss_cls_ce: 2.0644 decode.loss_mask_ce: 0.8817 decode.loss_mask_dice: 1.9469 decode.d7.loss_cls_ce: 2.0673 decode.d7.loss_mask_ce: 0.8763 decode.d7.loss_mask_dice: 1.9487 2023/09/06 18:52:11 - mmengine - INFO - Iter(train) [34450/60000] base_lr: 4.2584e-05 lr: 4.2584e-05 eta: 3:12:33 time: 0.4581 data_time: 0.0227 memory: 15809 grad_norm: 15.8152 loss: 9.2548 decode.loss_cls_ce: 2.0526 decode.loss_mask_ce: 0.8503 decode.loss_mask_dice: 1.7129 decode.d7.loss_cls_ce: 2.0756 decode.d7.loss_mask_ce: 0.8487 decode.d7.loss_mask_dice: 1.7148 2023/09/06 18:52:34 - mmengine - INFO - Iter(train) [34500/60000] base_lr: 4.2501e-05 lr: 4.2501e-05 eta: 3:12:10 time: 0.4593 data_time: 0.0232 memory: 15732 grad_norm: 17.5463 loss: 8.6911 decode.loss_cls_ce: 1.8779 decode.loss_mask_ce: 0.8526 decode.loss_mask_dice: 1.6074 decode.d7.loss_cls_ce: 1.8938 decode.d7.loss_mask_ce: 0.8510 decode.d7.loss_mask_dice: 1.6084 2023/09/06 18:52:57 - mmengine - INFO - Iter(train) [34550/60000] base_lr: 4.2417e-05 lr: 4.2417e-05 eta: 3:11:48 time: 0.4617 data_time: 0.0236 memory: 15746 grad_norm: 18.3184 loss: 9.4515 decode.loss_cls_ce: 1.9827 decode.loss_mask_ce: 0.9410 decode.loss_mask_dice: 1.8068 decode.d7.loss_cls_ce: 1.9783 decode.d7.loss_mask_ce: 0.9365 decode.d7.loss_mask_dice: 1.8061 2023/09/06 18:53:20 - mmengine - INFO - Iter(train) [34600/60000] base_lr: 4.2334e-05 lr: 4.2334e-05 eta: 3:11:26 time: 0.4584 data_time: 0.0231 memory: 15991 grad_norm: 16.5720 loss: 8.6151 decode.loss_cls_ce: 1.8008 decode.loss_mask_ce: 0.9193 decode.loss_mask_dice: 1.6014 decode.d7.loss_cls_ce: 1.7742 decode.d7.loss_mask_ce: 0.9206 decode.d7.loss_mask_dice: 1.5987 2023/09/06 18:53:43 - mmengine - INFO - Iter(train) [34650/60000] base_lr: 4.2251e-05 lr: 4.2251e-05 eta: 3:11:03 time: 0.4589 data_time: 0.0229 memory: 15823 grad_norm: 17.7010 loss: 9.4533 decode.loss_cls_ce: 2.0725 decode.loss_mask_ce: 0.9434 decode.loss_mask_dice: 1.7140 decode.d7.loss_cls_ce: 2.0891 decode.d7.loss_mask_ce: 0.9351 decode.d7.loss_mask_dice: 1.6991 2023/09/06 18:54:06 - mmengine - INFO - Iter(train) [34700/60000] base_lr: 4.2167e-05 lr: 4.2167e-05 eta: 3:10:41 time: 0.4611 data_time: 0.0232 memory: 16015 grad_norm: 16.9761 loss: 9.1010 decode.loss_cls_ce: 2.0209 decode.loss_mask_ce: 0.8793 decode.loss_mask_dice: 1.6425 decode.d7.loss_cls_ce: 2.0451 decode.d7.loss_mask_ce: 0.8783 decode.d7.loss_mask_dice: 1.6350 2023/09/06 18:54:29 - mmengine - INFO - Iter(train) [34750/60000] base_lr: 4.2084e-05 lr: 4.2084e-05 eta: 3:10:18 time: 0.4589 data_time: 0.0233 memory: 15935 grad_norm: 17.7540 loss: 8.6323 decode.loss_cls_ce: 1.8285 decode.loss_mask_ce: 0.8613 decode.loss_mask_dice: 1.6095 decode.d7.loss_cls_ce: 1.8494 decode.d7.loss_mask_ce: 0.8611 decode.d7.loss_mask_dice: 1.6226 2023/09/06 18:54:51 - mmengine - INFO - Iter(train) [34800/60000] base_lr: 4.2001e-05 lr: 4.2001e-05 eta: 3:09:56 time: 0.4576 data_time: 0.0227 memory: 15774 grad_norm: 20.9571 loss: 9.8027 decode.loss_cls_ce: 2.1423 decode.loss_mask_ce: 0.9118 decode.loss_mask_dice: 1.8203 decode.d7.loss_cls_ce: 2.1880 decode.d7.loss_mask_ce: 0.9131 decode.d7.loss_mask_dice: 1.8272 2023/09/06 18:55:14 - mmengine - INFO - Iter(train) [34850/60000] base_lr: 4.1917e-05 lr: 4.1917e-05 eta: 3:09:34 time: 0.4511 data_time: 0.0246 memory: 15848 grad_norm: 16.2254 loss: 9.0712 decode.loss_cls_ce: 1.9617 decode.loss_mask_ce: 0.9497 decode.loss_mask_dice: 1.6254 decode.d7.loss_cls_ce: 1.9797 decode.d7.loss_mask_ce: 0.9403 decode.d7.loss_mask_dice: 1.6143 2023/09/06 18:55:37 - mmengine - INFO - Iter(train) [34900/60000] base_lr: 4.1834e-05 lr: 4.1834e-05 eta: 3:09:11 time: 0.4605 data_time: 0.0231 memory: 15811 grad_norm: 16.3020 loss: 9.5630 decode.loss_cls_ce: 2.0300 decode.loss_mask_ce: 0.9545 decode.loss_mask_dice: 1.8009 decode.d7.loss_cls_ce: 2.0495 decode.d7.loss_mask_ce: 0.9505 decode.d7.loss_mask_dice: 1.7777 2023/09/06 18:56:00 - mmengine - INFO - Iter(train) [34950/60000] base_lr: 4.1751e-05 lr: 4.1751e-05 eta: 3:08:49 time: 0.4574 data_time: 0.0228 memory: 15734 grad_norm: 18.3904 loss: 8.9213 decode.loss_cls_ce: 2.0159 decode.loss_mask_ce: 0.9066 decode.loss_mask_dice: 1.5641 decode.d7.loss_cls_ce: 1.9739 decode.d7.loss_mask_ce: 0.9056 decode.d7.loss_mask_dice: 1.5552 2023/09/06 18:56:23 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 18:56:23 - mmengine - INFO - Iter(train) [35000/60000] base_lr: 4.1667e-05 lr: 4.1667e-05 eta: 3:08:26 time: 0.4588 data_time: 0.0229 memory: 15923 grad_norm: 15.9058 loss: 8.5907 decode.loss_cls_ce: 1.7733 decode.loss_mask_ce: 0.8562 decode.loss_mask_dice: 1.6633 decode.d7.loss_cls_ce: 1.7717 decode.d7.loss_mask_ce: 0.8591 decode.d7.loss_mask_dice: 1.6671 2023/09/06 18:56:46 - mmengine - INFO - Iter(train) [35050/60000] base_lr: 4.1584e-05 lr: 4.1584e-05 eta: 3:08:04 time: 0.4524 data_time: 0.0242 memory: 15847 grad_norm: 15.9346 loss: 9.8499 decode.loss_cls_ce: 2.1491 decode.loss_mask_ce: 0.8671 decode.loss_mask_dice: 1.9164 decode.d7.loss_cls_ce: 2.1568 decode.d7.loss_mask_ce: 0.8625 decode.d7.loss_mask_dice: 1.8980 2023/09/06 18:57:08 - mmengine - INFO - Iter(train) [35100/60000] base_lr: 4.1501e-05 lr: 4.1501e-05 eta: 3:07:41 time: 0.4497 data_time: 0.0241 memory: 15848 grad_norm: 16.1776 loss: 9.1967 decode.loss_cls_ce: 2.0372 decode.loss_mask_ce: 0.8838 decode.loss_mask_dice: 1.6653 decode.d7.loss_cls_ce: 2.0736 decode.d7.loss_mask_ce: 0.8811 decode.d7.loss_mask_dice: 1.6556 2023/09/06 18:57:31 - mmengine - INFO - Iter(train) [35150/60000] base_lr: 4.1417e-05 lr: 4.1417e-05 eta: 3:07:19 time: 0.4567 data_time: 0.0226 memory: 15835 grad_norm: 16.5179 loss: 9.0157 decode.loss_cls_ce: 2.0544 decode.loss_mask_ce: 0.7739 decode.loss_mask_dice: 1.6613 decode.d7.loss_cls_ce: 2.0798 decode.d7.loss_mask_ce: 0.7728 decode.d7.loss_mask_dice: 1.6736 2023/09/06 18:57:54 - mmengine - INFO - Iter(train) [35200/60000] base_lr: 4.1334e-05 lr: 4.1334e-05 eta: 3:06:56 time: 0.4594 data_time: 0.0231 memory: 15783 grad_norm: 18.6999 loss: 9.3806 decode.loss_cls_ce: 2.0191 decode.loss_mask_ce: 0.8759 decode.loss_mask_dice: 1.7852 decode.d7.loss_cls_ce: 2.0203 decode.d7.loss_mask_ce: 0.8823 decode.d7.loss_mask_dice: 1.7978 2023/09/06 18:58:17 - mmengine - INFO - Iter(train) [35250/60000] base_lr: 4.1251e-05 lr: 4.1251e-05 eta: 3:06:34 time: 0.4549 data_time: 0.0242 memory: 15873 grad_norm: 18.6570 loss: 9.0822 decode.loss_cls_ce: 2.0106 decode.loss_mask_ce: 0.8580 decode.loss_mask_dice: 1.6652 decode.d7.loss_cls_ce: 2.0163 decode.d7.loss_mask_ce: 0.8623 decode.d7.loss_mask_dice: 1.6698 2023/09/06 18:58:40 - mmengine - INFO - Iter(train) [35300/60000] base_lr: 4.1167e-05 lr: 4.1167e-05 eta: 3:06:11 time: 0.4517 data_time: 0.0236 memory: 15786 grad_norm: 16.7910 loss: 8.8109 decode.loss_cls_ce: 1.9916 decode.loss_mask_ce: 0.7818 decode.loss_mask_dice: 1.6184 decode.d7.loss_cls_ce: 2.0094 decode.d7.loss_mask_ce: 0.7881 decode.d7.loss_mask_dice: 1.6217 2023/09/06 18:59:03 - mmengine - INFO - Iter(train) [35350/60000] base_lr: 4.1084e-05 lr: 4.1084e-05 eta: 3:05:49 time: 0.4590 data_time: 0.0229 memory: 15832 grad_norm: 17.4553 loss: 7.6174 decode.loss_cls_ce: 1.7865 decode.loss_mask_ce: 0.7141 decode.loss_mask_dice: 1.3209 decode.d7.loss_cls_ce: 1.7812 decode.d7.loss_mask_ce: 0.7004 decode.d7.loss_mask_dice: 1.3143 2023/09/06 18:59:26 - mmengine - INFO - Iter(train) [35400/60000] base_lr: 4.1001e-05 lr: 4.1001e-05 eta: 3:05:27 time: 0.4611 data_time: 0.0237 memory: 16077 grad_norm: 17.3604 loss: 8.1969 decode.loss_cls_ce: 1.7584 decode.loss_mask_ce: 0.9028 decode.loss_mask_dice: 1.4298 decode.d7.loss_cls_ce: 1.7679 decode.d7.loss_mask_ce: 0.9117 decode.d7.loss_mask_dice: 1.4263 2023/09/06 18:59:49 - mmengine - INFO - Iter(train) [35450/60000] base_lr: 4.0917e-05 lr: 4.0917e-05 eta: 3:05:04 time: 0.4587 data_time: 0.0233 memory: 15898 grad_norm: 18.3786 loss: 9.4064 decode.loss_cls_ce: 2.0822 decode.loss_mask_ce: 0.8846 decode.loss_mask_dice: 1.7476 decode.d7.loss_cls_ce: 2.0615 decode.d7.loss_mask_ce: 0.8863 decode.d7.loss_mask_dice: 1.7443 2023/09/06 19:00:12 - mmengine - INFO - Iter(train) [35500/60000] base_lr: 4.0834e-05 lr: 4.0834e-05 eta: 3:04:42 time: 0.4489 data_time: 0.0236 memory: 15805 grad_norm: 17.2886 loss: 9.7659 decode.loss_cls_ce: 2.2153 decode.loss_mask_ce: 0.9038 decode.loss_mask_dice: 1.7658 decode.d7.loss_cls_ce: 2.2085 decode.d7.loss_mask_ce: 0.9087 decode.d7.loss_mask_dice: 1.7638 2023/09/06 19:00:34 - mmengine - INFO - Iter(train) [35550/60000] base_lr: 4.0751e-05 lr: 4.0751e-05 eta: 3:04:19 time: 0.4591 data_time: 0.0228 memory: 15845 grad_norm: 17.5493 loss: 10.0603 decode.loss_cls_ce: 2.1448 decode.loss_mask_ce: 0.9074 decode.loss_mask_dice: 1.9802 decode.d7.loss_cls_ce: 2.1328 decode.d7.loss_mask_ce: 0.9099 decode.d7.loss_mask_dice: 1.9852 2023/09/06 19:00:57 - mmengine - INFO - Iter(train) [35600/60000] base_lr: 4.0667e-05 lr: 4.0667e-05 eta: 3:03:57 time: 0.4518 data_time: 0.0249 memory: 15786 grad_norm: nan loss: 8.9199 decode.loss_cls_ce: 2.0107 decode.loss_mask_ce: 0.8902 decode.loss_mask_dice: 1.5727 decode.d7.loss_cls_ce: 1.9844 decode.d7.loss_mask_ce: 0.8993 decode.d7.loss_mask_dice: 1.5626 2023/09/06 19:01:20 - mmengine - INFO - Iter(train) [35650/60000] base_lr: 4.0584e-05 lr: 4.0584e-05 eta: 3:03:34 time: 0.4494 data_time: 0.0244 memory: 15810 grad_norm: 18.8367 loss: 9.5209 decode.loss_cls_ce: 2.0283 decode.loss_mask_ce: 0.9135 decode.loss_mask_dice: 1.7920 decode.d7.loss_cls_ce: 2.0671 decode.d7.loss_mask_ce: 0.9135 decode.d7.loss_mask_dice: 1.8065 2023/09/06 19:01:42 - mmengine - INFO - Iter(train) [35700/60000] base_lr: 4.0501e-05 lr: 4.0501e-05 eta: 3:03:11 time: 0.4498 data_time: 0.0243 memory: 15683 grad_norm: 17.4548 loss: 9.1482 decode.loss_cls_ce: 2.0996 decode.loss_mask_ce: 0.8193 decode.loss_mask_dice: 1.6407 decode.d7.loss_cls_ce: 2.1244 decode.d7.loss_mask_ce: 0.8230 decode.d7.loss_mask_dice: 1.6412 2023/09/06 19:02:05 - mmengine - INFO - Iter(train) [35750/60000] base_lr: 4.0417e-05 lr: 4.0417e-05 eta: 3:02:49 time: 0.4511 data_time: 0.0244 memory: 15743 grad_norm: 19.4641 loss: 8.6582 decode.loss_cls_ce: 1.9087 decode.loss_mask_ce: 0.8131 decode.loss_mask_dice: 1.6140 decode.d7.loss_cls_ce: 1.9060 decode.d7.loss_mask_ce: 0.8032 decode.d7.loss_mask_dice: 1.6132 2023/09/06 19:02:28 - mmengine - INFO - Iter(train) [35800/60000] base_lr: 4.0334e-05 lr: 4.0334e-05 eta: 3:02:26 time: 0.4558 data_time: 0.0235 memory: 15900 grad_norm: 16.3445 loss: 9.4923 decode.loss_cls_ce: 2.1553 decode.loss_mask_ce: 0.8409 decode.loss_mask_dice: 1.7621 decode.d7.loss_cls_ce: 2.1349 decode.d7.loss_mask_ce: 0.8375 decode.d7.loss_mask_dice: 1.7616 2023/09/06 19:02:51 - mmengine - INFO - Iter(train) [35850/60000] base_lr: 4.0251e-05 lr: 4.0251e-05 eta: 3:02:04 time: 0.4598 data_time: 0.0234 memory: 15898 grad_norm: 16.8739 loss: 9.5400 decode.loss_cls_ce: 2.0669 decode.loss_mask_ce: 0.9535 decode.loss_mask_dice: 1.7632 decode.d7.loss_cls_ce: 2.0629 decode.d7.loss_mask_ce: 0.9537 decode.d7.loss_mask_dice: 1.7399 2023/09/06 19:03:14 - mmengine - INFO - Iter(train) [35900/60000] base_lr: 4.0167e-05 lr: 4.0167e-05 eta: 3:01:41 time: 0.4581 data_time: 0.0228 memory: 15845 grad_norm: 17.5959 loss: 9.6181 decode.loss_cls_ce: 2.0654 decode.loss_mask_ce: 0.9025 decode.loss_mask_dice: 1.8454 decode.d7.loss_cls_ce: 2.0601 decode.d7.loss_mask_ce: 0.8950 decode.d7.loss_mask_dice: 1.8498 2023/09/06 19:03:36 - mmengine - INFO - Iter(train) [35950/60000] base_lr: 4.0084e-05 lr: 4.0084e-05 eta: 3:01:19 time: 0.4576 data_time: 0.0230 memory: 15784 grad_norm: 16.4668 loss: 8.7781 decode.loss_cls_ce: 1.8459 decode.loss_mask_ce: 0.8245 decode.loss_mask_dice: 1.7100 decode.d7.loss_cls_ce: 1.8457 decode.d7.loss_mask_ce: 0.8309 decode.d7.loss_mask_dice: 1.7212 2023/09/06 19:04:00 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:04:00 - mmengine - INFO - Iter(train) [36000/60000] base_lr: 4.0001e-05 lr: 4.0001e-05 eta: 3:00:57 time: 0.4593 data_time: 0.0236 memory: 15794 grad_norm: 18.9063 loss: 9.1685 decode.loss_cls_ce: 2.0922 decode.loss_mask_ce: 0.8457 decode.loss_mask_dice: 1.6541 decode.d7.loss_cls_ce: 2.0770 decode.d7.loss_mask_ce: 0.8366 decode.d7.loss_mask_dice: 1.6628 2023/09/06 19:04:23 - mmengine - INFO - Iter(train) [36050/60000] base_lr: 3.9917e-05 lr: 3.9917e-05 eta: 3:00:34 time: 0.4587 data_time: 0.0231 memory: 15872 grad_norm: 19.6576 loss: 9.2042 decode.loss_cls_ce: 2.0489 decode.loss_mask_ce: 0.8998 decode.loss_mask_dice: 1.6385 decode.d7.loss_cls_ce: 2.0720 decode.d7.loss_mask_ce: 0.9083 decode.d7.loss_mask_dice: 1.6368 2023/09/06 19:04:46 - mmengine - INFO - Iter(train) [36100/60000] base_lr: 3.9834e-05 lr: 3.9834e-05 eta: 3:00:12 time: 0.4608 data_time: 0.0236 memory: 15860 grad_norm: 20.2960 loss: 10.1909 decode.loss_cls_ce: 2.2528 decode.loss_mask_ce: 0.9824 decode.loss_mask_dice: 1.8545 decode.d7.loss_cls_ce: 2.2557 decode.d7.loss_mask_ce: 0.9845 decode.d7.loss_mask_dice: 1.8609 2023/09/06 19:05:08 - mmengine - INFO - Iter(train) [36150/60000] base_lr: 3.9751e-05 lr: 3.9751e-05 eta: 2:59:50 time: 0.4615 data_time: 0.0234 memory: 15935 grad_norm: 17.3176 loss: 9.1689 decode.loss_cls_ce: 1.9826 decode.loss_mask_ce: 0.8639 decode.loss_mask_dice: 1.7431 decode.d7.loss_cls_ce: 1.9626 decode.d7.loss_mask_ce: 0.8632 decode.d7.loss_mask_dice: 1.7536 2023/09/06 19:05:31 - mmengine - INFO - Iter(train) [36200/60000] base_lr: 3.9667e-05 lr: 3.9667e-05 eta: 2:59:27 time: 0.4514 data_time: 0.0244 memory: 15797 grad_norm: 17.8467 loss: 8.8352 decode.loss_cls_ce: 1.9426 decode.loss_mask_ce: 0.8532 decode.loss_mask_dice: 1.6161 decode.d7.loss_cls_ce: 1.9474 decode.d7.loss_mask_ce: 0.8534 decode.d7.loss_mask_dice: 1.6224 2023/09/06 19:05:54 - mmengine - INFO - Iter(train) [36250/60000] base_lr: 3.9584e-05 lr: 3.9584e-05 eta: 2:59:05 time: 0.4624 data_time: 0.0238 memory: 15974 grad_norm: 16.6697 loss: 8.5058 decode.loss_cls_ce: 1.8155 decode.loss_mask_ce: 0.8553 decode.loss_mask_dice: 1.5859 decode.d7.loss_cls_ce: 1.7993 decode.d7.loss_mask_ce: 0.8500 decode.d7.loss_mask_dice: 1.5998 2023/09/06 19:06:17 - mmengine - INFO - Iter(train) [36300/60000] base_lr: 3.9501e-05 lr: 3.9501e-05 eta: 2:58:42 time: 0.4572 data_time: 0.0233 memory: 15861 grad_norm: 17.4681 loss: 8.8432 decode.loss_cls_ce: 1.9118 decode.loss_mask_ce: 0.8494 decode.loss_mask_dice: 1.6426 decode.d7.loss_cls_ce: 1.9410 decode.d7.loss_mask_ce: 0.8485 decode.d7.loss_mask_dice: 1.6499 2023/09/06 19:06:40 - mmengine - INFO - Iter(train) [36350/60000] base_lr: 3.9417e-05 lr: 3.9417e-05 eta: 2:58:20 time: 0.4559 data_time: 0.0232 memory: 15940 grad_norm: 19.6618 loss: 9.0971 decode.loss_cls_ce: 1.9418 decode.loss_mask_ce: 0.9225 decode.loss_mask_dice: 1.6606 decode.d7.loss_cls_ce: 1.9891 decode.d7.loss_mask_ce: 0.9162 decode.d7.loss_mask_dice: 1.6668 2023/09/06 19:07:03 - mmengine - INFO - Iter(train) [36400/60000] base_lr: 3.9334e-05 lr: 3.9334e-05 eta: 2:57:57 time: 0.4624 data_time: 0.0238 memory: 15859 grad_norm: 17.6656 loss: 8.7488 decode.loss_cls_ce: 1.8706 decode.loss_mask_ce: 0.8664 decode.loss_mask_dice: 1.6425 decode.d7.loss_cls_ce: 1.8693 decode.d7.loss_mask_ce: 0.8630 decode.d7.loss_mask_dice: 1.6370 2023/09/06 19:07:26 - mmengine - INFO - Iter(train) [36450/60000] base_lr: 3.9251e-05 lr: 3.9251e-05 eta: 2:57:35 time: 0.4511 data_time: 0.0242 memory: 15849 grad_norm: 18.8576 loss: 9.1550 decode.loss_cls_ce: 2.0517 decode.loss_mask_ce: 0.8711 decode.loss_mask_dice: 1.6443 decode.d7.loss_cls_ce: 2.0566 decode.d7.loss_mask_ce: 0.8686 decode.d7.loss_mask_dice: 1.6628 2023/09/06 19:07:49 - mmengine - INFO - Iter(train) [36500/60000] base_lr: 3.9167e-05 lr: 3.9167e-05 eta: 2:57:12 time: 0.4613 data_time: 0.0239 memory: 15870 grad_norm: 18.6234 loss: 8.6409 decode.loss_cls_ce: 1.8786 decode.loss_mask_ce: 0.8700 decode.loss_mask_dice: 1.5737 decode.d7.loss_cls_ce: 1.8590 decode.d7.loss_mask_ce: 0.8751 decode.d7.loss_mask_dice: 1.5844 2023/09/06 19:08:11 - mmengine - INFO - Iter(train) [36550/60000] base_lr: 3.9084e-05 lr: 3.9084e-05 eta: 2:56:50 time: 0.4509 data_time: 0.0243 memory: 16013 grad_norm: 17.7490 loss: 8.5735 decode.loss_cls_ce: 1.9166 decode.loss_mask_ce: 0.7826 decode.loss_mask_dice: 1.5921 decode.d7.loss_cls_ce: 1.9238 decode.d7.loss_mask_ce: 0.7727 decode.d7.loss_mask_dice: 1.5856 2023/09/06 19:08:34 - mmengine - INFO - Iter(train) [36600/60000] base_lr: 3.9001e-05 lr: 3.9001e-05 eta: 2:56:27 time: 0.4597 data_time: 0.0231 memory: 15909 grad_norm: 17.9423 loss: 8.9640 decode.loss_cls_ce: 1.9310 decode.loss_mask_ce: 0.8605 decode.loss_mask_dice: 1.6969 decode.d7.loss_cls_ce: 1.9180 decode.d7.loss_mask_ce: 0.8589 decode.d7.loss_mask_dice: 1.6988 2023/09/06 19:08:57 - mmengine - INFO - Iter(train) [36650/60000] base_lr: 3.8917e-05 lr: 3.8917e-05 eta: 2:56:05 time: 0.4637 data_time: 0.0247 memory: 15950 grad_norm: 19.4603 loss: 9.7322 decode.loss_cls_ce: 2.1246 decode.loss_mask_ce: 0.9306 decode.loss_mask_dice: 1.8116 decode.d7.loss_cls_ce: 2.1283 decode.d7.loss_mask_ce: 0.9348 decode.d7.loss_mask_dice: 1.8023 2023/09/06 19:09:20 - mmengine - INFO - Iter(train) [36700/60000] base_lr: 3.8834e-05 lr: 3.8834e-05 eta: 2:55:42 time: 0.4570 data_time: 0.0243 memory: 15758 grad_norm: 16.9045 loss: 8.8791 decode.loss_cls_ce: 1.9103 decode.loss_mask_ce: 0.8915 decode.loss_mask_dice: 1.6410 decode.d7.loss_cls_ce: 1.9018 decode.d7.loss_mask_ce: 0.8985 decode.d7.loss_mask_dice: 1.6360 2023/09/06 19:09:43 - mmengine - INFO - Iter(train) [36750/60000] base_lr: 3.8751e-05 lr: 3.8751e-05 eta: 2:55:20 time: 0.4499 data_time: 0.0240 memory: 15899 grad_norm: 19.8495 loss: 8.8424 decode.loss_cls_ce: 1.9347 decode.loss_mask_ce: 0.8460 decode.loss_mask_dice: 1.6319 decode.d7.loss_cls_ce: 1.9391 decode.d7.loss_mask_ce: 0.8519 decode.d7.loss_mask_dice: 1.6388 2023/09/06 19:10:06 - mmengine - INFO - Iter(train) [36800/60000] base_lr: 3.8667e-05 lr: 3.8667e-05 eta: 2:54:57 time: 0.4612 data_time: 0.0237 memory: 15794 grad_norm: 18.7955 loss: 9.1088 decode.loss_cls_ce: 2.0833 decode.loss_mask_ce: 0.8403 decode.loss_mask_dice: 1.6463 decode.d7.loss_cls_ce: 2.0637 decode.d7.loss_mask_ce: 0.8352 decode.d7.loss_mask_dice: 1.6400 2023/09/06 19:10:28 - mmengine - INFO - Iter(train) [36850/60000] base_lr: 3.8584e-05 lr: 3.8584e-05 eta: 2:54:35 time: 0.4505 data_time: 0.0248 memory: 15746 grad_norm: 18.2832 loss: 9.0094 decode.loss_cls_ce: 1.9334 decode.loss_mask_ce: 0.9303 decode.loss_mask_dice: 1.6276 decode.d7.loss_cls_ce: 1.9573 decode.d7.loss_mask_ce: 0.9267 decode.d7.loss_mask_dice: 1.6342 2023/09/06 19:10:51 - mmengine - INFO - Iter(train) [36900/60000] base_lr: 3.8501e-05 lr: 3.8501e-05 eta: 2:54:12 time: 0.4585 data_time: 0.0238 memory: 15872 grad_norm: 18.1367 loss: 8.6565 decode.loss_cls_ce: 1.9008 decode.loss_mask_ce: 0.8621 decode.loss_mask_dice: 1.5747 decode.d7.loss_cls_ce: 1.8982 decode.d7.loss_mask_ce: 0.8541 decode.d7.loss_mask_dice: 1.5666 2023/09/06 19:11:14 - mmengine - INFO - Iter(train) [36950/60000] base_lr: 3.8417e-05 lr: 3.8417e-05 eta: 2:53:50 time: 0.4549 data_time: 0.0243 memory: 16094 grad_norm: 18.0483 loss: 8.7948 decode.loss_cls_ce: 1.9085 decode.loss_mask_ce: 0.8118 decode.loss_mask_dice: 1.6789 decode.d7.loss_cls_ce: 1.9041 decode.d7.loss_mask_ce: 0.8129 decode.d7.loss_mask_dice: 1.6786 2023/09/06 19:11:37 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:11:37 - mmengine - INFO - Iter(train) [37000/60000] base_lr: 3.8334e-05 lr: 3.8334e-05 eta: 2:53:27 time: 0.4573 data_time: 0.0246 memory: 15785 grad_norm: 17.4930 loss: 10.1535 decode.loss_cls_ce: 2.1251 decode.loss_mask_ce: 0.9787 decode.loss_mask_dice: 1.9487 decode.d7.loss_cls_ce: 2.1395 decode.d7.loss_mask_ce: 0.9811 decode.d7.loss_mask_dice: 1.9803 2023/09/06 19:12:00 - mmengine - INFO - Iter(train) [37050/60000] base_lr: 3.8251e-05 lr: 3.8251e-05 eta: 2:53:05 time: 0.4590 data_time: 0.0242 memory: 15807 grad_norm: 16.7114 loss: 8.2993 decode.loss_cls_ce: 1.8370 decode.loss_mask_ce: 0.8517 decode.loss_mask_dice: 1.4692 decode.d7.loss_cls_ce: 1.8113 decode.d7.loss_mask_ce: 0.8511 decode.d7.loss_mask_dice: 1.4791 2023/09/06 19:12:23 - mmengine - INFO - Iter(train) [37100/60000] base_lr: 3.8167e-05 lr: 3.8167e-05 eta: 2:52:42 time: 0.4603 data_time: 0.0233 memory: 15897 grad_norm: 18.3303 loss: 9.5258 decode.loss_cls_ce: 2.0045 decode.loss_mask_ce: 0.9154 decode.loss_mask_dice: 1.8222 decode.d7.loss_cls_ce: 2.0515 decode.d7.loss_mask_ce: 0.9165 decode.d7.loss_mask_dice: 1.8157 2023/09/06 19:12:46 - mmengine - INFO - Iter(train) [37150/60000] base_lr: 3.8084e-05 lr: 3.8084e-05 eta: 2:52:20 time: 0.4531 data_time: 0.0233 memory: 15883 grad_norm: 17.9633 loss: 9.2522 decode.loss_cls_ce: 1.9170 decode.loss_mask_ce: 0.9107 decode.loss_mask_dice: 1.7872 decode.d7.loss_cls_ce: 1.9287 decode.d7.loss_mask_ce: 0.9041 decode.d7.loss_mask_dice: 1.8045 2023/09/06 19:13:09 - mmengine - INFO - Iter(train) [37200/60000] base_lr: 3.8001e-05 lr: 3.8001e-05 eta: 2:51:57 time: 0.4517 data_time: 0.0245 memory: 15861 grad_norm: 16.6607 loss: 9.9117 decode.loss_cls_ce: 2.0698 decode.loss_mask_ce: 0.8801 decode.loss_mask_dice: 2.0136 decode.d7.loss_cls_ce: 2.0746 decode.d7.loss_mask_ce: 0.8719 decode.d7.loss_mask_dice: 2.0017 2023/09/06 19:13:31 - mmengine - INFO - Iter(train) [37250/60000] base_lr: 3.7917e-05 lr: 3.7917e-05 eta: 2:51:35 time: 0.4496 data_time: 0.0238 memory: 15909 grad_norm: 17.5172 loss: 8.4225 decode.loss_cls_ce: 1.8191 decode.loss_mask_ce: 0.8538 decode.loss_mask_dice: 1.5349 decode.d7.loss_cls_ce: 1.8169 decode.d7.loss_mask_ce: 0.8624 decode.d7.loss_mask_dice: 1.5353 2023/09/06 19:13:54 - mmengine - INFO - Iter(train) [37300/60000] base_lr: 3.7834e-05 lr: 3.7834e-05 eta: 2:51:12 time: 0.4518 data_time: 0.0244 memory: 15847 grad_norm: 17.1824 loss: 8.5253 decode.loss_cls_ce: 1.9936 decode.loss_mask_ce: 0.7563 decode.loss_mask_dice: 1.5361 decode.d7.loss_cls_ce: 1.9714 decode.d7.loss_mask_ce: 0.7574 decode.d7.loss_mask_dice: 1.5105 2023/09/06 19:14:16 - mmengine - INFO - Iter(train) [37350/60000] base_lr: 3.7751e-05 lr: 3.7751e-05 eta: 2:50:49 time: 0.4516 data_time: 0.0243 memory: 15857 grad_norm: 16.2453 loss: 9.9365 decode.loss_cls_ce: 2.1033 decode.loss_mask_ce: 0.9217 decode.loss_mask_dice: 1.9295 decode.d7.loss_cls_ce: 2.1290 decode.d7.loss_mask_ce: 0.9183 decode.d7.loss_mask_dice: 1.9347 2023/09/06 19:14:39 - mmengine - INFO - Iter(train) [37400/60000] base_lr: 3.7667e-05 lr: 3.7667e-05 eta: 2:50:27 time: 0.4597 data_time: 0.0233 memory: 15886 grad_norm: 18.2872 loss: 8.9194 decode.loss_cls_ce: 2.0211 decode.loss_mask_ce: 0.7895 decode.loss_mask_dice: 1.6417 decode.d7.loss_cls_ce: 2.0359 decode.d7.loss_mask_ce: 0.7888 decode.d7.loss_mask_dice: 1.6424 2023/09/06 19:15:02 - mmengine - INFO - Iter(train) [37450/60000] base_lr: 3.7584e-05 lr: 3.7584e-05 eta: 2:50:04 time: 0.4575 data_time: 0.0236 memory: 15846 grad_norm: 15.8921 loss: 8.7490 decode.loss_cls_ce: 1.8727 decode.loss_mask_ce: 0.8477 decode.loss_mask_dice: 1.6545 decode.d7.loss_cls_ce: 1.8725 decode.d7.loss_mask_ce: 0.8496 decode.d7.loss_mask_dice: 1.6520 2023/09/06 19:15:25 - mmengine - INFO - Iter(train) [37500/60000] base_lr: 3.7501e-05 lr: 3.7501e-05 eta: 2:49:42 time: 0.4552 data_time: 0.0233 memory: 15858 grad_norm: 16.2268 loss: 9.5915 decode.loss_cls_ce: 2.0975 decode.loss_mask_ce: 0.8397 decode.loss_mask_dice: 1.8484 decode.d7.loss_cls_ce: 2.1314 decode.d7.loss_mask_ce: 0.8328 decode.d7.loss_mask_dice: 1.8417 2023/09/06 19:15:48 - mmengine - INFO - Iter(train) [37550/60000] base_lr: 3.7417e-05 lr: 3.7417e-05 eta: 2:49:20 time: 0.4534 data_time: 0.0240 memory: 16027 grad_norm: 16.3284 loss: 9.6229 decode.loss_cls_ce: 2.0635 decode.loss_mask_ce: 0.9508 decode.loss_mask_dice: 1.7977 decode.d7.loss_cls_ce: 2.0775 decode.d7.loss_mask_ce: 0.9543 decode.d7.loss_mask_dice: 1.7792 2023/09/06 19:16:11 - mmengine - INFO - Iter(train) [37600/60000] base_lr: 3.7334e-05 lr: 3.7334e-05 eta: 2:48:57 time: 0.4518 data_time: 0.0244 memory: 15791 grad_norm: 18.2552 loss: 8.1998 decode.loss_cls_ce: 1.8031 decode.loss_mask_ce: 0.7534 decode.loss_mask_dice: 1.5245 decode.d7.loss_cls_ce: 1.8212 decode.d7.loss_mask_ce: 0.7624 decode.d7.loss_mask_dice: 1.5352 2023/09/06 19:16:33 - mmengine - INFO - Iter(train) [37650/60000] base_lr: 3.7251e-05 lr: 3.7251e-05 eta: 2:48:34 time: 0.4593 data_time: 0.0242 memory: 15975 grad_norm: 18.5258 loss: 9.1872 decode.loss_cls_ce: 1.9753 decode.loss_mask_ce: 0.8874 decode.loss_mask_dice: 1.7228 decode.d7.loss_cls_ce: 2.0053 decode.d7.loss_mask_ce: 0.8824 decode.d7.loss_mask_dice: 1.7140 2023/09/06 19:16:56 - mmengine - INFO - Iter(train) [37700/60000] base_lr: 3.7167e-05 lr: 3.7167e-05 eta: 2:48:12 time: 0.4519 data_time: 0.0247 memory: 15656 grad_norm: 19.1084 loss: 7.9007 decode.loss_cls_ce: 1.7236 decode.loss_mask_ce: 0.8041 decode.loss_mask_dice: 1.4153 decode.d7.loss_cls_ce: 1.7228 decode.d7.loss_mask_ce: 0.8033 decode.d7.loss_mask_dice: 1.4316 2023/09/06 19:17:18 - mmengine - INFO - Iter(train) [37750/60000] base_lr: 3.7084e-05 lr: 3.7084e-05 eta: 2:47:49 time: 0.4516 data_time: 0.0243 memory: 15773 grad_norm: 18.4617 loss: 8.9048 decode.loss_cls_ce: 1.9537 decode.loss_mask_ce: 0.8431 decode.loss_mask_dice: 1.6518 decode.d7.loss_cls_ce: 1.9592 decode.d7.loss_mask_ce: 0.8421 decode.d7.loss_mask_dice: 1.6549 2023/09/06 19:17:41 - mmengine - INFO - Iter(train) [37800/60000] base_lr: 3.7001e-05 lr: 3.7001e-05 eta: 2:47:26 time: 0.4567 data_time: 0.0235 memory: 15963 grad_norm: 17.7333 loss: 9.2067 decode.loss_cls_ce: 2.0063 decode.loss_mask_ce: 0.8946 decode.loss_mask_dice: 1.7023 decode.d7.loss_cls_ce: 2.0214 decode.d7.loss_mask_ce: 0.8946 decode.d7.loss_mask_dice: 1.6875 2023/09/06 19:18:04 - mmengine - INFO - Iter(train) [37850/60000] base_lr: 3.6917e-05 lr: 3.6917e-05 eta: 2:47:04 time: 0.4588 data_time: 0.0226 memory: 15835 grad_norm: 17.9216 loss: 7.7442 decode.loss_cls_ce: 1.7714 decode.loss_mask_ce: 0.7563 decode.loss_mask_dice: 1.3467 decode.d7.loss_cls_ce: 1.7988 decode.d7.loss_mask_ce: 0.7520 decode.d7.loss_mask_dice: 1.3191 2023/09/06 19:18:27 - mmengine - INFO - Iter(train) [37900/60000] base_lr: 3.6834e-05 lr: 3.6834e-05 eta: 2:46:41 time: 0.4596 data_time: 0.0229 memory: 15886 grad_norm: 17.6308 loss: 8.6874 decode.loss_cls_ce: 1.8798 decode.loss_mask_ce: 0.8508 decode.loss_mask_dice: 1.6036 decode.d7.loss_cls_ce: 1.9096 decode.d7.loss_mask_ce: 0.8483 decode.d7.loss_mask_dice: 1.5954 2023/09/06 19:18:50 - mmengine - INFO - Iter(train) [37950/60000] base_lr: 3.6751e-05 lr: 3.6751e-05 eta: 2:46:19 time: 0.4627 data_time: 0.0233 memory: 15897 grad_norm: 17.3632 loss: 9.5755 decode.loss_cls_ce: 2.0371 decode.loss_mask_ce: 0.9471 decode.loss_mask_dice: 1.7900 decode.d7.loss_cls_ce: 2.0713 decode.d7.loss_mask_ce: 0.9472 decode.d7.loss_mask_dice: 1.7827 2023/09/06 19:19:13 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:19:13 - mmengine - INFO - Iter(train) [38000/60000] base_lr: 3.6667e-05 lr: 3.6667e-05 eta: 2:45:57 time: 0.4632 data_time: 0.0241 memory: 15923 grad_norm: 16.8511 loss: 9.0449 decode.loss_cls_ce: 2.0746 decode.loss_mask_ce: 0.8123 decode.loss_mask_dice: 1.6329 decode.d7.loss_cls_ce: 2.0832 decode.d7.loss_mask_ce: 0.8070 decode.d7.loss_mask_dice: 1.6349 2023/09/06 19:19:36 - mmengine - INFO - Iter(train) [38050/60000] base_lr: 3.6584e-05 lr: 3.6584e-05 eta: 2:45:34 time: 0.4615 data_time: 0.0233 memory: 15911 grad_norm: 18.6341 loss: 8.9888 decode.loss_cls_ce: 2.0042 decode.loss_mask_ce: 0.8705 decode.loss_mask_dice: 1.6190 decode.d7.loss_cls_ce: 2.0115 decode.d7.loss_mask_ce: 0.8661 decode.d7.loss_mask_dice: 1.6173 2023/09/06 19:19:59 - mmengine - INFO - Iter(train) [38100/60000] base_lr: 3.6501e-05 lr: 3.6501e-05 eta: 2:45:12 time: 0.4530 data_time: 0.0243 memory: 15883 grad_norm: 18.6951 loss: 8.1569 decode.loss_cls_ce: 1.7160 decode.loss_mask_ce: 0.8713 decode.loss_mask_dice: 1.4914 decode.d7.loss_cls_ce: 1.6914 decode.d7.loss_mask_ce: 0.8807 decode.d7.loss_mask_dice: 1.5062 2023/09/06 19:20:22 - mmengine - INFO - Iter(train) [38150/60000] base_lr: 3.6417e-05 lr: 3.6417e-05 eta: 2:44:49 time: 0.4512 data_time: 0.0238 memory: 15923 grad_norm: 18.5376 loss: 9.7346 decode.loss_cls_ce: 2.1138 decode.loss_mask_ce: 0.8688 decode.loss_mask_dice: 1.8702 decode.d7.loss_cls_ce: 2.1603 decode.d7.loss_mask_ce: 0.8687 decode.d7.loss_mask_dice: 1.8528 2023/09/06 19:20:44 - mmengine - INFO - Iter(train) [38200/60000] base_lr: 3.6334e-05 lr: 3.6334e-05 eta: 2:44:27 time: 0.4521 data_time: 0.0242 memory: 15766 grad_norm: 19.8310 loss: 9.3666 decode.loss_cls_ce: 1.9747 decode.loss_mask_ce: 0.9357 decode.loss_mask_dice: 1.7629 decode.d7.loss_cls_ce: 1.9787 decode.d7.loss_mask_ce: 0.9445 decode.d7.loss_mask_dice: 1.7702 2023/09/06 19:21:07 - mmengine - INFO - Iter(train) [38250/60000] base_lr: 3.6251e-05 lr: 3.6251e-05 eta: 2:44:04 time: 0.4628 data_time: 0.0234 memory: 15861 grad_norm: 20.6123 loss: 10.2338 decode.loss_cls_ce: 2.1164 decode.loss_mask_ce: 0.9699 decode.loss_mask_dice: 2.0299 decode.d7.loss_cls_ce: 2.1444 decode.d7.loss_mask_ce: 0.9688 decode.d7.loss_mask_dice: 2.0044 2023/09/06 19:21:30 - mmengine - INFO - Iter(train) [38300/60000] base_lr: 3.6167e-05 lr: 3.6167e-05 eta: 2:43:41 time: 0.4527 data_time: 0.0252 memory: 15860 grad_norm: 18.5453 loss: 9.6200 decode.loss_cls_ce: 2.0662 decode.loss_mask_ce: 0.8787 decode.loss_mask_dice: 1.8517 decode.d7.loss_cls_ce: 2.0720 decode.d7.loss_mask_ce: 0.8932 decode.d7.loss_mask_dice: 1.8583 2023/09/06 19:21:52 - mmengine - INFO - Iter(train) [38350/60000] base_lr: 3.6084e-05 lr: 3.6084e-05 eta: 2:43:19 time: 0.4490 data_time: 0.0237 memory: 15922 grad_norm: 17.6490 loss: 7.9843 decode.loss_cls_ce: 1.8835 decode.loss_mask_ce: 0.7477 decode.loss_mask_dice: 1.3518 decode.d7.loss_cls_ce: 1.8749 decode.d7.loss_mask_ce: 0.7591 decode.d7.loss_mask_dice: 1.3672 2023/09/06 19:22:15 - mmengine - INFO - Iter(train) [38400/60000] base_lr: 3.6001e-05 lr: 3.6001e-05 eta: 2:42:56 time: 0.4572 data_time: 0.0230 memory: 15822 grad_norm: 18.1602 loss: 9.6611 decode.loss_cls_ce: 2.1020 decode.loss_mask_ce: 0.9019 decode.loss_mask_dice: 1.8255 decode.d7.loss_cls_ce: 2.0916 decode.d7.loss_mask_ce: 0.9041 decode.d7.loss_mask_dice: 1.8360 2023/09/06 19:22:38 - mmengine - INFO - Iter(train) [38450/60000] base_lr: 3.5917e-05 lr: 3.5917e-05 eta: 2:42:34 time: 0.4543 data_time: 0.0243 memory: 15998 grad_norm: 16.9421 loss: 9.7725 decode.loss_cls_ce: 2.1486 decode.loss_mask_ce: 0.9086 decode.loss_mask_dice: 1.8380 decode.d7.loss_cls_ce: 2.1285 decode.d7.loss_mask_ce: 0.9067 decode.d7.loss_mask_dice: 1.8421 2023/09/06 19:23:01 - mmengine - INFO - Iter(train) [38500/60000] base_lr: 3.5834e-05 lr: 3.5834e-05 eta: 2:42:11 time: 0.4556 data_time: 0.0233 memory: 15768 grad_norm: 17.1257 loss: 9.0806 decode.loss_cls_ce: 1.9495 decode.loss_mask_ce: 0.8609 decode.loss_mask_dice: 1.7092 decode.d7.loss_cls_ce: 1.9912 decode.d7.loss_mask_ce: 0.8524 decode.d7.loss_mask_dice: 1.7175 2023/09/06 19:23:24 - mmengine - INFO - Iter(train) [38550/60000] base_lr: 3.5751e-05 lr: 3.5751e-05 eta: 2:41:49 time: 0.4584 data_time: 0.0235 memory: 16005 grad_norm: 16.7119 loss: 9.0628 decode.loss_cls_ce: 2.0194 decode.loss_mask_ce: 0.8666 decode.loss_mask_dice: 1.6466 decode.d7.loss_cls_ce: 2.0178 decode.d7.loss_mask_ce: 0.8719 decode.d7.loss_mask_dice: 1.6405 2023/09/06 19:23:47 - mmengine - INFO - Iter(train) [38600/60000] base_lr: 3.5667e-05 lr: 3.5667e-05 eta: 2:41:26 time: 0.4496 data_time: 0.0235 memory: 15895 grad_norm: 16.9291 loss: 9.6602 decode.loss_cls_ce: 2.1287 decode.loss_mask_ce: 0.8438 decode.loss_mask_dice: 1.8571 decode.d7.loss_cls_ce: 2.1273 decode.d7.loss_mask_ce: 0.8467 decode.d7.loss_mask_dice: 1.8566 2023/09/06 19:24:09 - mmengine - INFO - Iter(train) [38650/60000] base_lr: 3.5584e-05 lr: 3.5584e-05 eta: 2:41:03 time: 0.4512 data_time: 0.0243 memory: 15744 grad_norm: 17.0856 loss: 9.4081 decode.loss_cls_ce: 1.9665 decode.loss_mask_ce: 0.9532 decode.loss_mask_dice: 1.7821 decode.d7.loss_cls_ce: 1.9603 decode.d7.loss_mask_ce: 0.9524 decode.d7.loss_mask_dice: 1.7936 2023/09/06 19:24:32 - mmengine - INFO - Iter(train) [38700/60000] base_lr: 3.5501e-05 lr: 3.5501e-05 eta: 2:40:41 time: 0.4614 data_time: 0.0232 memory: 15898 grad_norm: 20.0510 loss: 9.7320 decode.loss_cls_ce: 2.1022 decode.loss_mask_ce: 0.9692 decode.loss_mask_dice: 1.7927 decode.d7.loss_cls_ce: 2.0814 decode.d7.loss_mask_ce: 0.9757 decode.d7.loss_mask_dice: 1.8109 2023/09/06 19:24:55 - mmengine - INFO - Iter(train) [38750/60000] base_lr: 3.5417e-05 lr: 3.5417e-05 eta: 2:40:19 time: 0.4605 data_time: 0.0237 memory: 16106 grad_norm: 16.7784 loss: 9.2979 decode.loss_cls_ce: 2.0309 decode.loss_mask_ce: 0.9139 decode.loss_mask_dice: 1.7038 decode.d7.loss_cls_ce: 2.0567 decode.d7.loss_mask_ce: 0.9097 decode.d7.loss_mask_dice: 1.6830 2023/09/06 19:25:18 - mmengine - INFO - Iter(train) [38800/60000] base_lr: 3.5334e-05 lr: 3.5334e-05 eta: 2:39:56 time: 0.4627 data_time: 0.0246 memory: 15886 grad_norm: 17.1766 loss: 10.0229 decode.loss_cls_ce: 2.1058 decode.loss_mask_ce: 0.9394 decode.loss_mask_dice: 1.9620 decode.d7.loss_cls_ce: 2.0966 decode.d7.loss_mask_ce: 0.9412 decode.d7.loss_mask_dice: 1.9779 2023/09/06 19:25:41 - mmengine - INFO - Iter(train) [38850/60000] base_lr: 3.5251e-05 lr: 3.5251e-05 eta: 2:39:34 time: 0.4572 data_time: 0.0232 memory: 15935 grad_norm: 16.9171 loss: 9.1993 decode.loss_cls_ce: 2.0279 decode.loss_mask_ce: 0.8823 decode.loss_mask_dice: 1.6785 decode.d7.loss_cls_ce: 2.0569 decode.d7.loss_mask_ce: 0.8755 decode.d7.loss_mask_dice: 1.6782 2023/09/06 19:26:04 - mmengine - INFO - Iter(train) [38900/60000] base_lr: 3.5167e-05 lr: 3.5167e-05 eta: 2:39:11 time: 0.4523 data_time: 0.0239 memory: 15872 grad_norm: 18.2794 loss: 9.0143 decode.loss_cls_ce: 1.9944 decode.loss_mask_ce: 0.8088 decode.loss_mask_dice: 1.6995 decode.d7.loss_cls_ce: 2.0098 decode.d7.loss_mask_ce: 0.8018 decode.d7.loss_mask_dice: 1.7000 2023/09/06 19:26:27 - mmengine - INFO - Iter(train) [38950/60000] base_lr: 3.5084e-05 lr: 3.5084e-05 eta: 2:38:49 time: 0.4605 data_time: 0.0238 memory: 15792 grad_norm: 17.8272 loss: 8.8504 decode.loss_cls_ce: 1.8007 decode.loss_mask_ce: 0.8755 decode.loss_mask_dice: 1.7414 decode.d7.loss_cls_ce: 1.8439 decode.d7.loss_mask_ce: 0.8773 decode.d7.loss_mask_dice: 1.7116 2023/09/06 19:26:50 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:26:50 - mmengine - INFO - Iter(train) [39000/60000] base_lr: 3.5001e-05 lr: 3.5001e-05 eta: 2:38:26 time: 0.4582 data_time: 0.0230 memory: 15775 grad_norm: 17.7529 loss: 9.2156 decode.loss_cls_ce: 2.0658 decode.loss_mask_ce: 0.8642 decode.loss_mask_dice: 1.6870 decode.d7.loss_cls_ce: 2.0391 decode.d7.loss_mask_ce: 0.8634 decode.d7.loss_mask_dice: 1.6962 2023/09/06 19:27:13 - mmengine - INFO - Iter(train) [39050/60000] base_lr: 3.4917e-05 lr: 3.4917e-05 eta: 2:38:04 time: 0.4633 data_time: 0.0239 memory: 15772 grad_norm: 16.6550 loss: 9.1892 decode.loss_cls_ce: 2.0768 decode.loss_mask_ce: 0.8698 decode.loss_mask_dice: 1.6430 decode.d7.loss_cls_ce: 2.0813 decode.d7.loss_mask_ce: 0.8732 decode.d7.loss_mask_dice: 1.6451 2023/09/06 19:27:36 - mmengine - INFO - Iter(train) [39100/60000] base_lr: 3.4834e-05 lr: 3.4834e-05 eta: 2:37:41 time: 0.4585 data_time: 0.0235 memory: 15961 grad_norm: 16.6306 loss: 9.0187 decode.loss_cls_ce: 1.9823 decode.loss_mask_ce: 0.8814 decode.loss_mask_dice: 1.6531 decode.d7.loss_cls_ce: 1.9472 decode.d7.loss_mask_ce: 0.8936 decode.d7.loss_mask_dice: 1.6611 2023/09/06 19:27:58 - mmengine - INFO - Iter(train) [39150/60000] base_lr: 3.4751e-05 lr: 3.4751e-05 eta: 2:37:18 time: 0.4500 data_time: 0.0239 memory: 15680 grad_norm: 17.2856 loss: 8.6999 decode.loss_cls_ce: 1.9364 decode.loss_mask_ce: 0.8339 decode.loss_mask_dice: 1.5895 decode.d7.loss_cls_ce: 1.9185 decode.d7.loss_mask_ce: 0.8383 decode.d7.loss_mask_dice: 1.5834 2023/09/06 19:28:21 - mmengine - INFO - Iter(train) [39200/60000] base_lr: 3.4667e-05 lr: 3.4667e-05 eta: 2:36:56 time: 0.4510 data_time: 0.0243 memory: 15770 grad_norm: 19.2258 loss: 9.0598 decode.loss_cls_ce: 1.9369 decode.loss_mask_ce: 0.8578 decode.loss_mask_dice: 1.7387 decode.d7.loss_cls_ce: 1.9305 decode.d7.loss_mask_ce: 0.8588 decode.d7.loss_mask_dice: 1.7371 2023/09/06 19:28:43 - mmengine - INFO - Iter(train) [39250/60000] base_lr: 3.4584e-05 lr: 3.4584e-05 eta: 2:36:33 time: 0.4563 data_time: 0.0236 memory: 15837 grad_norm: 17.2225 loss: 8.5893 decode.loss_cls_ce: 1.8612 decode.loss_mask_ce: 0.8268 decode.loss_mask_dice: 1.5973 decode.d7.loss_cls_ce: 1.9007 decode.d7.loss_mask_ce: 0.8237 decode.d7.loss_mask_dice: 1.5796 2023/09/06 19:29:06 - mmengine - INFO - Iter(train) [39300/60000] base_lr: 3.4501e-05 lr: 3.4501e-05 eta: 2:36:11 time: 0.4608 data_time: 0.0242 memory: 15785 grad_norm: 17.8892 loss: 8.8100 decode.loss_cls_ce: 1.8829 decode.loss_mask_ce: 0.8316 decode.loss_mask_dice: 1.6823 decode.d7.loss_cls_ce: 1.8958 decode.d7.loss_mask_ce: 0.8311 decode.d7.loss_mask_dice: 1.6862 2023/09/06 19:29:29 - mmengine - INFO - Iter(train) [39350/60000] base_lr: 3.4417e-05 lr: 3.4417e-05 eta: 2:35:48 time: 0.4524 data_time: 0.0243 memory: 15884 grad_norm: 16.9269 loss: 8.4972 decode.loss_cls_ce: 1.8645 decode.loss_mask_ce: 0.7828 decode.loss_mask_dice: 1.5958 decode.d7.loss_cls_ce: 1.8714 decode.d7.loss_mask_ce: 0.7807 decode.d7.loss_mask_dice: 1.6020 2023/09/06 19:29:52 - mmengine - INFO - Iter(train) [39400/60000] base_lr: 3.4334e-05 lr: 3.4334e-05 eta: 2:35:25 time: 0.4515 data_time: 0.0237 memory: 15811 grad_norm: 15.6056 loss: 8.5572 decode.loss_cls_ce: 1.8854 decode.loss_mask_ce: 0.8350 decode.loss_mask_dice: 1.5519 decode.d7.loss_cls_ce: 1.8821 decode.d7.loss_mask_ce: 0.8400 decode.d7.loss_mask_dice: 1.5627 2023/09/06 19:30:15 - mmengine - INFO - Iter(train) [39450/60000] base_lr: 3.4251e-05 lr: 3.4251e-05 eta: 2:35:03 time: 0.4616 data_time: 0.0239 memory: 15824 grad_norm: 16.9412 loss: 9.1337 decode.loss_cls_ce: 2.0444 decode.loss_mask_ce: 0.8179 decode.loss_mask_dice: 1.7137 decode.d7.loss_cls_ce: 2.0333 decode.d7.loss_mask_ce: 0.8174 decode.d7.loss_mask_dice: 1.7069 2023/09/06 19:30:38 - mmengine - INFO - Iter(train) [39500/60000] base_lr: 3.4167e-05 lr: 3.4167e-05 eta: 2:34:41 time: 0.4620 data_time: 0.0241 memory: 16027 grad_norm: 18.5000 loss: 8.8628 decode.loss_cls_ce: 1.9624 decode.loss_mask_ce: 0.8490 decode.loss_mask_dice: 1.6162 decode.d7.loss_cls_ce: 1.9557 decode.d7.loss_mask_ce: 0.8500 decode.d7.loss_mask_dice: 1.6295 2023/09/06 19:31:01 - mmengine - INFO - Iter(train) [39550/60000] base_lr: 3.4084e-05 lr: 3.4084e-05 eta: 2:34:18 time: 0.4517 data_time: 0.0244 memory: 15732 grad_norm: 17.9579 loss: 8.6870 decode.loss_cls_ce: 1.9445 decode.loss_mask_ce: 0.8256 decode.loss_mask_dice: 1.5604 decode.d7.loss_cls_ce: 1.9564 decode.d7.loss_mask_ce: 0.8297 decode.d7.loss_mask_dice: 1.5704 2023/09/06 19:31:23 - mmengine - INFO - Iter(train) [39600/60000] base_lr: 3.4001e-05 lr: 3.4001e-05 eta: 2:33:55 time: 0.4520 data_time: 0.0244 memory: 15846 grad_norm: 17.0686 loss: 9.3874 decode.loss_cls_ce: 2.1287 decode.loss_mask_ce: 0.8897 decode.loss_mask_dice: 1.6771 decode.d7.loss_cls_ce: 2.1090 decode.d7.loss_mask_ce: 0.8931 decode.d7.loss_mask_dice: 1.6897 2023/09/06 19:31:46 - mmengine - INFO - Iter(train) [39650/60000] base_lr: 3.3917e-05 lr: 3.3917e-05 eta: 2:33:33 time: 0.4569 data_time: 0.0235 memory: 15911 grad_norm: 17.2708 loss: 8.0945 decode.loss_cls_ce: 1.7598 decode.loss_mask_ce: 0.8211 decode.loss_mask_dice: 1.4614 decode.d7.loss_cls_ce: 1.7636 decode.d7.loss_mask_ce: 0.8248 decode.d7.loss_mask_dice: 1.4637 2023/09/06 19:32:09 - mmengine - INFO - Iter(train) [39700/60000] base_lr: 3.3834e-05 lr: 3.3834e-05 eta: 2:33:10 time: 0.4597 data_time: 0.0253 memory: 16027 grad_norm: 19.9367 loss: 10.0520 decode.loss_cls_ce: 2.1370 decode.loss_mask_ce: 0.9526 decode.loss_mask_dice: 1.9138 decode.d7.loss_cls_ce: 2.1969 decode.d7.loss_mask_ce: 0.9483 decode.d7.loss_mask_dice: 1.9035 2023/09/06 19:32:32 - mmengine - INFO - Iter(train) [39750/60000] base_lr: 3.3751e-05 lr: 3.3751e-05 eta: 2:32:48 time: 0.4642 data_time: 0.0247 memory: 16081 grad_norm: 17.3417 loss: 9.0192 decode.loss_cls_ce: 1.9312 decode.loss_mask_ce: 0.8897 decode.loss_mask_dice: 1.6851 decode.d7.loss_cls_ce: 1.9353 decode.d7.loss_mask_ce: 0.8876 decode.d7.loss_mask_dice: 1.6904 2023/09/06 19:32:55 - mmengine - INFO - Iter(train) [39800/60000] base_lr: 3.3667e-05 lr: 3.3667e-05 eta: 2:32:25 time: 0.4608 data_time: 0.0235 memory: 15773 grad_norm: 16.3057 loss: 8.8715 decode.loss_cls_ce: 1.8778 decode.loss_mask_ce: 0.8250 decode.loss_mask_dice: 1.7358 decode.d7.loss_cls_ce: 1.8837 decode.d7.loss_mask_ce: 0.8221 decode.d7.loss_mask_dice: 1.7272 2023/09/06 19:33:18 - mmengine - INFO - Iter(train) [39850/60000] base_lr: 3.3584e-05 lr: 3.3584e-05 eta: 2:32:03 time: 0.4605 data_time: 0.0243 memory: 15895 grad_norm: 17.1539 loss: 8.8348 decode.loss_cls_ce: 2.0133 decode.loss_mask_ce: 0.8691 decode.loss_mask_dice: 1.5445 decode.d7.loss_cls_ce: 1.9967 decode.d7.loss_mask_ce: 0.8727 decode.d7.loss_mask_dice: 1.5385 2023/09/06 19:33:41 - mmengine - INFO - Iter(train) [39900/60000] base_lr: 3.3501e-05 lr: 3.3501e-05 eta: 2:31:40 time: 0.4508 data_time: 0.0246 memory: 15880 grad_norm: 17.4890 loss: 8.3152 decode.loss_cls_ce: 1.7929 decode.loss_mask_ce: 0.7642 decode.loss_mask_dice: 1.5888 decode.d7.loss_cls_ce: 1.7981 decode.d7.loss_mask_ce: 0.7726 decode.d7.loss_mask_dice: 1.5986 2023/09/06 19:34:03 - mmengine - INFO - Iter(train) [39950/60000] base_lr: 3.3417e-05 lr: 3.3417e-05 eta: 2:31:18 time: 0.4497 data_time: 0.0235 memory: 16089 grad_norm: 16.0538 loss: 8.5506 decode.loss_cls_ce: 1.8509 decode.loss_mask_ce: 0.7987 decode.loss_mask_dice: 1.6257 decode.d7.loss_cls_ce: 1.8646 decode.d7.loss_mask_ce: 0.8025 decode.d7.loss_mask_dice: 1.6082 2023/09/06 19:34:26 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:34:26 - mmengine - INFO - Iter(train) [40000/60000] base_lr: 3.3334e-05 lr: 3.3334e-05 eta: 2:30:55 time: 0.4513 data_time: 0.0245 memory: 15821 grad_norm: 20.0617 loss: 8.7907 decode.loss_cls_ce: 1.8484 decode.loss_mask_ce: 0.8879 decode.loss_mask_dice: 1.6486 decode.d7.loss_cls_ce: 1.8645 decode.d7.loss_mask_ce: 0.8873 decode.d7.loss_mask_dice: 1.6539 2023/09/06 19:34:26 - mmengine - INFO - Saving checkpoint at 40000 iterations 2023/09/06 19:34:52 - mmengine - INFO - Iter(train) [40050/60000] base_lr: 3.3251e-05 lr: 3.3251e-05 eta: 2:30:34 time: 0.4627 data_time: 0.0237 memory: 15846 grad_norm: 16.7293 loss: 9.2395 decode.loss_cls_ce: 2.0186 decode.loss_mask_ce: 0.8540 decode.loss_mask_dice: 1.7490 decode.d7.loss_cls_ce: 2.0183 decode.d7.loss_mask_ce: 0.8563 decode.d7.loss_mask_dice: 1.7433 2023/09/06 19:35:15 - mmengine - INFO - Iter(train) [40100/60000] base_lr: 3.3167e-05 lr: 3.3167e-05 eta: 2:30:12 time: 0.4613 data_time: 0.0238 memory: 16110 grad_norm: 17.0804 loss: 8.4438 decode.loss_cls_ce: 1.8422 decode.loss_mask_ce: 0.8240 decode.loss_mask_dice: 1.5601 decode.d7.loss_cls_ce: 1.8570 decode.d7.loss_mask_ce: 0.8130 decode.d7.loss_mask_dice: 1.5475 2023/09/06 19:35:38 - mmengine - INFO - Iter(train) [40150/60000] base_lr: 3.3084e-05 lr: 3.3084e-05 eta: 2:29:49 time: 0.4544 data_time: 0.0249 memory: 15899 grad_norm: 16.0821 loss: 10.1383 decode.loss_cls_ce: 2.1980 decode.loss_mask_ce: 0.9360 decode.loss_mask_dice: 1.9486 decode.d7.loss_cls_ce: 2.1787 decode.d7.loss_mask_ce: 0.9368 decode.d7.loss_mask_dice: 1.9402 2023/09/06 19:36:01 - mmengine - INFO - Iter(train) [40200/60000] base_lr: 3.3001e-05 lr: 3.3001e-05 eta: 2:29:27 time: 0.4573 data_time: 0.0233 memory: 15868 grad_norm: 18.1862 loss: 9.9423 decode.loss_cls_ce: 2.1413 decode.loss_mask_ce: 0.9153 decode.loss_mask_dice: 1.8986 decode.d7.loss_cls_ce: 2.1766 decode.d7.loss_mask_ce: 0.9103 decode.d7.loss_mask_dice: 1.9002 2023/09/06 19:36:24 - mmengine - INFO - Iter(train) [40250/60000] base_lr: 3.2917e-05 lr: 3.2917e-05 eta: 2:29:04 time: 0.4607 data_time: 0.0234 memory: 15821 grad_norm: 18.3488 loss: 8.7449 decode.loss_cls_ce: 1.8784 decode.loss_mask_ce: 0.8713 decode.loss_mask_dice: 1.6066 decode.d7.loss_cls_ce: 1.9161 decode.d7.loss_mask_ce: 0.8697 decode.d7.loss_mask_dice: 1.6028 2023/09/06 19:36:47 - mmengine - INFO - Iter(train) [40300/60000] base_lr: 3.2834e-05 lr: 3.2834e-05 eta: 2:28:42 time: 0.4550 data_time: 0.0226 memory: 15861 grad_norm: 18.6975 loss: 8.8496 decode.loss_cls_ce: 1.9867 decode.loss_mask_ce: 0.8336 decode.loss_mask_dice: 1.6187 decode.d7.loss_cls_ce: 1.9802 decode.d7.loss_mask_ce: 0.8253 decode.d7.loss_mask_dice: 1.6052 2023/09/06 19:37:10 - mmengine - INFO - Iter(train) [40350/60000] base_lr: 3.2751e-05 lr: 3.2751e-05 eta: 2:28:19 time: 0.4554 data_time: 0.0242 memory: 15798 grad_norm: 17.6315 loss: 7.9994 decode.loss_cls_ce: 1.8083 decode.loss_mask_ce: 0.8068 decode.loss_mask_dice: 1.3756 decode.d7.loss_cls_ce: 1.8199 decode.d7.loss_mask_ce: 0.8054 decode.d7.loss_mask_dice: 1.3835 2023/09/06 19:37:33 - mmengine - INFO - Iter(train) [40400/60000] base_lr: 3.2667e-05 lr: 3.2667e-05 eta: 2:27:57 time: 0.4595 data_time: 0.0244 memory: 15791 grad_norm: 18.0684 loss: 8.8379 decode.loss_cls_ce: 1.9764 decode.loss_mask_ce: 0.8218 decode.loss_mask_dice: 1.6209 decode.d7.loss_cls_ce: 1.9892 decode.d7.loss_mask_ce: 0.8093 decode.d7.loss_mask_dice: 1.6203 2023/09/06 19:37:55 - mmengine - INFO - Iter(train) [40450/60000] base_lr: 3.2584e-05 lr: 3.2584e-05 eta: 2:27:34 time: 0.4539 data_time: 0.0232 memory: 15783 grad_norm: 17.0079 loss: 9.1090 decode.loss_cls_ce: 2.0324 decode.loss_mask_ce: 0.9078 decode.loss_mask_dice: 1.6080 decode.d7.loss_cls_ce: 2.0455 decode.d7.loss_mask_ce: 0.9033 decode.d7.loss_mask_dice: 1.6118 2023/09/06 19:38:18 - mmengine - INFO - Iter(train) [40500/60000] base_lr: 3.2501e-05 lr: 3.2501e-05 eta: 2:27:11 time: 0.4519 data_time: 0.0244 memory: 16105 grad_norm: 17.8438 loss: 8.6070 decode.loss_cls_ce: 1.9660 decode.loss_mask_ce: 0.7728 decode.loss_mask_dice: 1.5559 decode.d7.loss_cls_ce: 1.9917 decode.d7.loss_mask_ce: 0.7681 decode.d7.loss_mask_dice: 1.5527 2023/09/06 19:38:41 - mmengine - INFO - Iter(train) [40550/60000] base_lr: 3.2417e-05 lr: 3.2417e-05 eta: 2:26:49 time: 0.4499 data_time: 0.0236 memory: 15845 grad_norm: 18.5450 loss: 8.8549 decode.loss_cls_ce: 1.9770 decode.loss_mask_ce: 0.8506 decode.loss_mask_dice: 1.5858 decode.d7.loss_cls_ce: 2.0012 decode.d7.loss_mask_ce: 0.8591 decode.d7.loss_mask_dice: 1.5812 2023/09/06 19:39:04 - mmengine - INFO - Iter(train) [40600/60000] base_lr: 3.2334e-05 lr: 3.2334e-05 eta: 2:26:26 time: 0.4618 data_time: 0.0236 memory: 15836 grad_norm: 16.6302 loss: 8.7882 decode.loss_cls_ce: 1.9345 decode.loss_mask_ce: 0.8547 decode.loss_mask_dice: 1.6055 decode.d7.loss_cls_ce: 1.9251 decode.d7.loss_mask_ce: 0.8550 decode.d7.loss_mask_dice: 1.6134 2023/09/06 19:39:27 - mmengine - INFO - Iter(train) [40650/60000] base_lr: 3.2251e-05 lr: 3.2251e-05 eta: 2:26:04 time: 0.4541 data_time: 0.0234 memory: 15940 grad_norm: 22.9103 loss: 9.1382 decode.loss_cls_ce: 2.1887 decode.loss_mask_ce: 0.7855 decode.loss_mask_dice: 1.5968 decode.d7.loss_cls_ce: 2.1866 decode.d7.loss_mask_ce: 0.7796 decode.d7.loss_mask_dice: 1.6011 2023/09/06 19:39:50 - mmengine - INFO - Iter(train) [40700/60000] base_lr: 3.2167e-05 lr: 3.2167e-05 eta: 2:25:41 time: 0.4534 data_time: 0.0260 memory: 15989 grad_norm: 16.8368 loss: 9.5118 decode.loss_cls_ce: 2.1756 decode.loss_mask_ce: 0.8493 decode.loss_mask_dice: 1.7233 decode.d7.loss_cls_ce: 2.1687 decode.d7.loss_mask_ce: 0.8612 decode.d7.loss_mask_dice: 1.7338 2023/09/06 19:40:12 - mmengine - INFO - Iter(train) [40750/60000] base_lr: 3.2084e-05 lr: 3.2084e-05 eta: 2:25:18 time: 0.4508 data_time: 0.0242 memory: 15884 grad_norm: 18.1091 loss: 8.4288 decode.loss_cls_ce: 1.8304 decode.loss_mask_ce: 0.8217 decode.loss_mask_dice: 1.5478 decode.d7.loss_cls_ce: 1.8504 decode.d7.loss_mask_ce: 0.8208 decode.d7.loss_mask_dice: 1.5577 2023/09/06 19:40:35 - mmengine - INFO - Iter(train) [40800/60000] base_lr: 3.2001e-05 lr: 3.2001e-05 eta: 2:24:56 time: 0.4515 data_time: 0.0246 memory: 15771 grad_norm: 16.7481 loss: 9.1316 decode.loss_cls_ce: 1.9866 decode.loss_mask_ce: 0.8986 decode.loss_mask_dice: 1.6821 decode.d7.loss_cls_ce: 1.9784 decode.d7.loss_mask_ce: 0.8968 decode.d7.loss_mask_dice: 1.6891 2023/09/06 19:40:57 - mmengine - INFO - Iter(train) [40850/60000] base_lr: 3.1917e-05 lr: 3.1917e-05 eta: 2:24:33 time: 0.4543 data_time: 0.0240 memory: 15895 grad_norm: 18.0599 loss: 10.0546 decode.loss_cls_ce: 2.2118 decode.loss_mask_ce: 0.9317 decode.loss_mask_dice: 1.8839 decode.d7.loss_cls_ce: 2.1833 decode.d7.loss_mask_ce: 0.9414 decode.d7.loss_mask_dice: 1.9025 2023/09/06 19:41:20 - mmengine - INFO - Iter(train) [40900/60000] base_lr: 3.1834e-05 lr: 3.1834e-05 eta: 2:24:10 time: 0.4531 data_time: 0.0240 memory: 15858 grad_norm: 17.4400 loss: 8.8115 decode.loss_cls_ce: 1.9097 decode.loss_mask_ce: 0.8703 decode.loss_mask_dice: 1.6172 decode.d7.loss_cls_ce: 1.9169 decode.d7.loss_mask_ce: 0.8771 decode.d7.loss_mask_dice: 1.6203 2023/09/06 19:41:43 - mmengine - INFO - Iter(train) [40950/60000] base_lr: 3.1751e-05 lr: 3.1751e-05 eta: 2:23:48 time: 0.4505 data_time: 0.0244 memory: 15847 grad_norm: 15.5733 loss: 8.3107 decode.loss_cls_ce: 1.7677 decode.loss_mask_ce: 0.7972 decode.loss_mask_dice: 1.6068 decode.d7.loss_cls_ce: 1.7491 decode.d7.loss_mask_ce: 0.7975 decode.d7.loss_mask_dice: 1.5925 2023/09/06 19:42:05 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:42:05 - mmengine - INFO - Iter(train) [41000/60000] base_lr: 3.1667e-05 lr: 3.1667e-05 eta: 2:23:25 time: 0.4603 data_time: 0.0232 memory: 15870 grad_norm: 17.2595 loss: 8.9847 decode.loss_cls_ce: 1.9617 decode.loss_mask_ce: 0.8934 decode.loss_mask_dice: 1.6488 decode.d7.loss_cls_ce: 1.9657 decode.d7.loss_mask_ce: 0.8853 decode.d7.loss_mask_dice: 1.6298 2023/09/06 19:42:28 - mmengine - INFO - Iter(train) [41050/60000] base_lr: 3.1584e-05 lr: 3.1584e-05 eta: 2:23:03 time: 0.4605 data_time: 0.0231 memory: 15885 grad_norm: 17.5722 loss: 9.9447 decode.loss_cls_ce: 2.0766 decode.loss_mask_ce: 0.9506 decode.loss_mask_dice: 1.9421 decode.d7.loss_cls_ce: 2.0704 decode.d7.loss_mask_ce: 0.9531 decode.d7.loss_mask_dice: 1.9520 2023/09/06 19:42:51 - mmengine - INFO - Iter(train) [41100/60000] base_lr: 3.1501e-05 lr: 3.1501e-05 eta: 2:22:40 time: 0.4616 data_time: 0.0245 memory: 15845 grad_norm: 17.5427 loss: 9.1916 decode.loss_cls_ce: 1.9910 decode.loss_mask_ce: 0.9223 decode.loss_mask_dice: 1.6768 decode.d7.loss_cls_ce: 1.9846 decode.d7.loss_mask_ce: 0.9344 decode.d7.loss_mask_dice: 1.6824 2023/09/06 19:43:14 - mmengine - INFO - Iter(train) [41150/60000] base_lr: 3.1417e-05 lr: 3.1417e-05 eta: 2:22:18 time: 0.4615 data_time: 0.0240 memory: 15769 grad_norm: 19.7781 loss: 8.3361 decode.loss_cls_ce: 1.8294 decode.loss_mask_ce: 0.7814 decode.loss_mask_dice: 1.5406 decode.d7.loss_cls_ce: 1.8586 decode.d7.loss_mask_ce: 0.7843 decode.d7.loss_mask_dice: 1.5418 2023/09/06 19:43:37 - mmengine - INFO - Iter(train) [41200/60000] base_lr: 3.1334e-05 lr: 3.1334e-05 eta: 2:21:55 time: 0.4634 data_time: 0.0244 memory: 15858 grad_norm: 17.6111 loss: 8.9409 decode.loss_cls_ce: 2.0043 decode.loss_mask_ce: 0.8605 decode.loss_mask_dice: 1.6018 decode.d7.loss_cls_ce: 2.0117 decode.d7.loss_mask_ce: 0.8612 decode.d7.loss_mask_dice: 1.6014 2023/09/06 19:44:00 - mmengine - INFO - Iter(train) [41250/60000] base_lr: 3.1251e-05 lr: 3.1251e-05 eta: 2:21:33 time: 0.4617 data_time: 0.0232 memory: 15864 grad_norm: 16.4641 loss: 8.7475 decode.loss_cls_ce: 1.8230 decode.loss_mask_ce: 0.8709 decode.loss_mask_dice: 1.6738 decode.d7.loss_cls_ce: 1.8515 decode.d7.loss_mask_ce: 0.8679 decode.d7.loss_mask_dice: 1.6604 2023/09/06 19:44:23 - mmengine - INFO - Iter(train) [41300/60000] base_lr: 3.1167e-05 lr: 3.1167e-05 eta: 2:21:10 time: 0.4496 data_time: 0.0239 memory: 15935 grad_norm: 16.4861 loss: 8.3262 decode.loss_cls_ce: 1.8096 decode.loss_mask_ce: 0.8149 decode.loss_mask_dice: 1.5406 decode.d7.loss_cls_ce: 1.8025 decode.d7.loss_mask_ce: 0.8206 decode.d7.loss_mask_dice: 1.5380 2023/09/06 19:44:46 - mmengine - INFO - Iter(train) [41350/60000] base_lr: 3.1084e-05 lr: 3.1084e-05 eta: 2:20:47 time: 0.4596 data_time: 0.0231 memory: 15951 grad_norm: 17.5987 loss: 10.3608 decode.loss_cls_ce: 2.1597 decode.loss_mask_ce: 0.9744 decode.loss_mask_dice: 2.0328 decode.d7.loss_cls_ce: 2.1724 decode.d7.loss_mask_ce: 0.9842 decode.d7.loss_mask_dice: 2.0373 2023/09/06 19:45:09 - mmengine - INFO - Iter(train) [41400/60000] base_lr: 3.1001e-05 lr: 3.1001e-05 eta: 2:20:25 time: 0.4566 data_time: 0.0234 memory: 15846 grad_norm: 17.4442 loss: 9.7072 decode.loss_cls_ce: 2.1375 decode.loss_mask_ce: 0.9256 decode.loss_mask_dice: 1.7877 decode.d7.loss_cls_ce: 2.1452 decode.d7.loss_mask_ce: 0.9301 decode.d7.loss_mask_dice: 1.7812 2023/09/06 19:45:32 - mmengine - INFO - Iter(train) [41450/60000] base_lr: 3.0917e-05 lr: 3.0917e-05 eta: 2:20:02 time: 0.4614 data_time: 0.0233 memory: 15938 grad_norm: 18.2134 loss: 9.5543 decode.loss_cls_ce: 2.0668 decode.loss_mask_ce: 0.9466 decode.loss_mask_dice: 1.7609 decode.d7.loss_cls_ce: 2.0466 decode.d7.loss_mask_ce: 0.9594 decode.d7.loss_mask_dice: 1.7740 2023/09/06 19:45:55 - mmengine - INFO - Iter(train) [41500/60000] base_lr: 3.0834e-05 lr: 3.0834e-05 eta: 2:19:40 time: 0.4536 data_time: 0.0239 memory: 15950 grad_norm: 15.6167 loss: 8.5607 decode.loss_cls_ce: 1.8869 decode.loss_mask_ce: 0.9039 decode.loss_mask_dice: 1.4956 decode.d7.loss_cls_ce: 1.8786 decode.d7.loss_mask_ce: 0.9036 decode.d7.loss_mask_dice: 1.4922 2023/09/06 19:46:18 - mmengine - INFO - Iter(train) [41550/60000] base_lr: 3.0751e-05 lr: 3.0751e-05 eta: 2:19:17 time: 0.4536 data_time: 0.0248 memory: 15910 grad_norm: 16.7409 loss: 9.3797 decode.loss_cls_ce: 2.0237 decode.loss_mask_ce: 0.9317 decode.loss_mask_dice: 1.7392 decode.d7.loss_cls_ce: 2.0090 decode.d7.loss_mask_ce: 0.9305 decode.d7.loss_mask_dice: 1.7455 2023/09/06 19:46:41 - mmengine - INFO - Iter(train) [41600/60000] base_lr: 3.0667e-05 lr: 3.0667e-05 eta: 2:18:55 time: 0.4615 data_time: 0.0243 memory: 15848 grad_norm: 16.3629 loss: 9.5617 decode.loss_cls_ce: 2.1556 decode.loss_mask_ce: 0.8604 decode.loss_mask_dice: 1.7460 decode.d7.loss_cls_ce: 2.1747 decode.d7.loss_mask_ce: 0.8756 decode.d7.loss_mask_dice: 1.7493 2023/09/06 19:47:04 - mmengine - INFO - Iter(train) [41650/60000] base_lr: 3.0584e-05 lr: 3.0584e-05 eta: 2:18:32 time: 0.4619 data_time: 0.0232 memory: 15834 grad_norm: 16.9576 loss: 8.4920 decode.loss_cls_ce: 1.8387 decode.loss_mask_ce: 0.8313 decode.loss_mask_dice: 1.5813 decode.d7.loss_cls_ce: 1.8221 decode.d7.loss_mask_ce: 0.8260 decode.d7.loss_mask_dice: 1.5925 2023/09/06 19:47:27 - mmengine - INFO - Iter(train) [41700/60000] base_lr: 3.0501e-05 lr: 3.0501e-05 eta: 2:18:10 time: 0.4605 data_time: 0.0231 memory: 15871 grad_norm: 17.0712 loss: 8.8209 decode.loss_cls_ce: 2.0014 decode.loss_mask_ce: 0.8086 decode.loss_mask_dice: 1.5922 decode.d7.loss_cls_ce: 2.0158 decode.d7.loss_mask_ce: 0.8126 decode.d7.loss_mask_dice: 1.5903 2023/09/06 19:47:50 - mmengine - INFO - Iter(train) [41750/60000] base_lr: 3.0417e-05 lr: 3.0417e-05 eta: 2:17:47 time: 0.4606 data_time: 0.0231 memory: 15786 grad_norm: 17.2426 loss: 9.5481 decode.loss_cls_ce: 2.0551 decode.loss_mask_ce: 0.9071 decode.loss_mask_dice: 1.7946 decode.d7.loss_cls_ce: 2.0784 decode.d7.loss_mask_ce: 0.9166 decode.d7.loss_mask_dice: 1.7963 2023/09/06 19:48:13 - mmengine - INFO - Iter(train) [41800/60000] base_lr: 3.0334e-05 lr: 3.0334e-05 eta: 2:17:25 time: 0.4610 data_time: 0.0232 memory: 15922 grad_norm: 18.5386 loss: 8.7138 decode.loss_cls_ce: 1.9073 decode.loss_mask_ce: 0.8202 decode.loss_mask_dice: 1.6384 decode.d7.loss_cls_ce: 1.9192 decode.d7.loss_mask_ce: 0.8092 decode.d7.loss_mask_dice: 1.6196 2023/09/06 19:48:36 - mmengine - INFO - Iter(train) [41850/60000] base_lr: 3.0251e-05 lr: 3.0251e-05 eta: 2:17:02 time: 0.4624 data_time: 0.0234 memory: 15856 grad_norm: 17.4248 loss: 9.2479 decode.loss_cls_ce: 2.0144 decode.loss_mask_ce: 0.8598 decode.loss_mask_dice: 1.7434 decode.d7.loss_cls_ce: 2.0062 decode.d7.loss_mask_ce: 0.8760 decode.d7.loss_mask_dice: 1.7481 2023/09/06 19:48:59 - mmengine - INFO - Iter(train) [41900/60000] base_lr: 3.0167e-05 lr: 3.0167e-05 eta: 2:16:40 time: 0.4588 data_time: 0.0228 memory: 15769 grad_norm: 16.4687 loss: 9.2424 decode.loss_cls_ce: 2.0946 decode.loss_mask_ce: 0.8161 decode.loss_mask_dice: 1.7108 decode.d7.loss_cls_ce: 2.0947 decode.d7.loss_mask_ce: 0.8099 decode.d7.loss_mask_dice: 1.7163 2023/09/06 19:49:22 - mmengine - INFO - Iter(train) [41950/60000] base_lr: 3.0084e-05 lr: 3.0084e-05 eta: 2:16:17 time: 0.4600 data_time: 0.0232 memory: 15807 grad_norm: 19.2898 loss: 8.9389 decode.loss_cls_ce: 1.9884 decode.loss_mask_ce: 0.8725 decode.loss_mask_dice: 1.6034 decode.d7.loss_cls_ce: 1.9878 decode.d7.loss_mask_ce: 0.8628 decode.d7.loss_mask_dice: 1.6239 2023/09/06 19:49:45 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:49:45 - mmengine - INFO - Iter(train) [42000/60000] base_lr: 3.0001e-05 lr: 3.0001e-05 eta: 2:15:55 time: 0.4526 data_time: 0.0242 memory: 15872 grad_norm: 18.8463 loss: 8.5351 decode.loss_cls_ce: 1.8180 decode.loss_mask_ce: 0.8567 decode.loss_mask_dice: 1.5872 decode.d7.loss_cls_ce: 1.8557 decode.d7.loss_mask_ce: 0.8497 decode.d7.loss_mask_dice: 1.5679 2023/09/06 19:50:08 - mmengine - INFO - Iter(train) [42050/60000] base_lr: 2.9917e-05 lr: 2.9917e-05 eta: 2:15:32 time: 0.4620 data_time: 0.0233 memory: 15756 grad_norm: 18.3601 loss: 9.4657 decode.loss_cls_ce: 2.0726 decode.loss_mask_ce: 0.9599 decode.loss_mask_dice: 1.6914 decode.d7.loss_cls_ce: 2.0996 decode.d7.loss_mask_ce: 0.9486 decode.d7.loss_mask_dice: 1.6937 2023/09/06 19:50:31 - mmengine - INFO - Iter(train) [42100/60000] base_lr: 2.9834e-05 lr: 2.9834e-05 eta: 2:15:10 time: 0.4596 data_time: 0.0228 memory: 15911 grad_norm: 17.5436 loss: 9.1918 decode.loss_cls_ce: 2.0263 decode.loss_mask_ce: 0.8437 decode.loss_mask_dice: 1.7094 decode.d7.loss_cls_ce: 2.0540 decode.d7.loss_mask_ce: 0.8432 decode.d7.loss_mask_dice: 1.7152 2023/09/06 19:50:54 - mmengine - INFO - Iter(train) [42150/60000] base_lr: 2.9750e-05 lr: 2.9750e-05 eta: 2:14:47 time: 0.4630 data_time: 0.0239 memory: 15925 grad_norm: 22.2807 loss: 8.5389 decode.loss_cls_ce: 1.9862 decode.loss_mask_ce: 0.7988 decode.loss_mask_dice: 1.4779 decode.d7.loss_cls_ce: 1.9962 decode.d7.loss_mask_ce: 0.7987 decode.d7.loss_mask_dice: 1.4811 2023/09/06 19:51:17 - mmengine - INFO - Iter(train) [42200/60000] base_lr: 2.9667e-05 lr: 2.9667e-05 eta: 2:14:25 time: 0.4632 data_time: 0.0238 memory: 15762 grad_norm: 16.9812 loss: 8.4765 decode.loss_cls_ce: 1.9110 decode.loss_mask_ce: 0.7935 decode.loss_mask_dice: 1.5181 decode.d7.loss_cls_ce: 1.9302 decode.d7.loss_mask_ce: 0.7960 decode.d7.loss_mask_dice: 1.5277 2023/09/06 19:51:40 - mmengine - INFO - Iter(train) [42250/60000] base_lr: 2.9584e-05 lr: 2.9584e-05 eta: 2:14:02 time: 0.4553 data_time: 0.0243 memory: 15822 grad_norm: 17.2809 loss: 8.2633 decode.loss_cls_ce: 1.8903 decode.loss_mask_ce: 0.7507 decode.loss_mask_dice: 1.4935 decode.d7.loss_cls_ce: 1.8724 decode.d7.loss_mask_ce: 0.7660 decode.d7.loss_mask_dice: 1.4905 2023/09/06 19:52:02 - mmengine - INFO - Iter(train) [42300/60000] base_lr: 2.9500e-05 lr: 2.9500e-05 eta: 2:13:39 time: 0.4526 data_time: 0.0242 memory: 15834 grad_norm: 17.2348 loss: 9.9274 decode.loss_cls_ce: 2.0826 decode.loss_mask_ce: 0.9314 decode.loss_mask_dice: 1.9596 decode.d7.loss_cls_ce: 2.0863 decode.d7.loss_mask_ce: 0.9265 decode.d7.loss_mask_dice: 1.9410 2023/09/06 19:52:25 - mmengine - INFO - Iter(train) [42350/60000] base_lr: 2.9417e-05 lr: 2.9417e-05 eta: 2:13:17 time: 0.4588 data_time: 0.0230 memory: 15990 grad_norm: 16.1197 loss: 9.1000 decode.loss_cls_ce: 1.9741 decode.loss_mask_ce: 0.8883 decode.loss_mask_dice: 1.6939 decode.d7.loss_cls_ce: 1.9646 decode.d7.loss_mask_ce: 0.8817 decode.d7.loss_mask_dice: 1.6974 2023/09/06 19:52:48 - mmengine - INFO - Iter(train) [42400/60000] base_lr: 2.9334e-05 lr: 2.9334e-05 eta: 2:12:54 time: 0.4613 data_time: 0.0236 memory: 15796 grad_norm: 18.0339 loss: 9.4304 decode.loss_cls_ce: 1.9820 decode.loss_mask_ce: 0.9506 decode.loss_mask_dice: 1.7963 decode.d7.loss_cls_ce: 1.9676 decode.d7.loss_mask_ce: 0.9444 decode.d7.loss_mask_dice: 1.7895 2023/09/06 19:53:11 - mmengine - INFO - Iter(train) [42450/60000] base_lr: 2.9250e-05 lr: 2.9250e-05 eta: 2:12:32 time: 0.4617 data_time: 0.0241 memory: 15835 grad_norm: 17.1778 loss: 8.6317 decode.loss_cls_ce: 1.8241 decode.loss_mask_ce: 0.8471 decode.loss_mask_dice: 1.6376 decode.d7.loss_cls_ce: 1.8333 decode.d7.loss_mask_ce: 0.8509 decode.d7.loss_mask_dice: 1.6386 2023/09/06 19:53:34 - mmengine - INFO - Iter(train) [42500/60000] base_lr: 2.9167e-05 lr: 2.9167e-05 eta: 2:12:09 time: 0.4578 data_time: 0.0236 memory: 16027 grad_norm: 17.8658 loss: 9.4336 decode.loss_cls_ce: 2.1014 decode.loss_mask_ce: 0.8740 decode.loss_mask_dice: 1.7626 decode.d7.loss_cls_ce: 2.0757 decode.d7.loss_mask_ce: 0.8696 decode.d7.loss_mask_dice: 1.7503 2023/09/06 19:53:57 - mmengine - INFO - Iter(train) [42550/60000] base_lr: 2.9084e-05 lr: 2.9084e-05 eta: 2:11:47 time: 0.4633 data_time: 0.0236 memory: 15884 grad_norm: 15.5561 loss: 9.8907 decode.loss_cls_ce: 2.1131 decode.loss_mask_ce: 0.9375 decode.loss_mask_dice: 1.8894 decode.d7.loss_cls_ce: 2.1054 decode.d7.loss_mask_ce: 0.9324 decode.d7.loss_mask_dice: 1.9129 2023/09/06 19:54:20 - mmengine - INFO - Iter(train) [42600/60000] base_lr: 2.9000e-05 lr: 2.9000e-05 eta: 2:11:24 time: 0.4555 data_time: 0.0250 memory: 15810 grad_norm: 17.3567 loss: 7.9503 decode.loss_cls_ce: 1.7169 decode.loss_mask_ce: 0.7728 decode.loss_mask_dice: 1.4714 decode.d7.loss_cls_ce: 1.7359 decode.d7.loss_mask_ce: 0.7747 decode.d7.loss_mask_dice: 1.4786 2023/09/06 19:54:43 - mmengine - INFO - Iter(train) [42650/60000] base_lr: 2.8917e-05 lr: 2.8917e-05 eta: 2:11:02 time: 0.4628 data_time: 0.0239 memory: 15783 grad_norm: 22.4626 loss: 9.0946 decode.loss_cls_ce: 1.9913 decode.loss_mask_ce: 0.9030 decode.loss_mask_dice: 1.6573 decode.d7.loss_cls_ce: 1.9819 decode.d7.loss_mask_ce: 0.8988 decode.d7.loss_mask_dice: 1.6623 2023/09/06 19:55:06 - mmengine - INFO - Iter(train) [42700/60000] base_lr: 2.8834e-05 lr: 2.8834e-05 eta: 2:10:39 time: 0.4617 data_time: 0.0233 memory: 15798 grad_norm: 19.7102 loss: 9.2119 decode.loss_cls_ce: 2.1423 decode.loss_mask_ce: 0.8753 decode.loss_mask_dice: 1.5976 decode.d7.loss_cls_ce: 2.1374 decode.d7.loss_mask_ce: 0.8704 decode.d7.loss_mask_dice: 1.5889 2023/09/06 19:55:30 - mmengine - INFO - Iter(train) [42750/60000] base_lr: 2.8750e-05 lr: 2.8750e-05 eta: 2:10:17 time: 0.4567 data_time: 0.0228 memory: 15846 grad_norm: 18.1106 loss: 7.6972 decode.loss_cls_ce: 1.8294 decode.loss_mask_ce: 0.7327 decode.loss_mask_dice: 1.3004 decode.d7.loss_cls_ce: 1.8062 decode.d7.loss_mask_ce: 0.7332 decode.d7.loss_mask_dice: 1.2953 2023/09/06 19:55:53 - mmengine - INFO - Iter(train) [42800/60000] base_lr: 2.8667e-05 lr: 2.8667e-05 eta: 2:09:54 time: 0.4625 data_time: 0.0238 memory: 15825 grad_norm: 19.5181 loss: 10.3350 decode.loss_cls_ce: 2.2757 decode.loss_mask_ce: 0.9367 decode.loss_mask_dice: 1.9489 decode.d7.loss_cls_ce: 2.2850 decode.d7.loss_mask_ce: 0.9371 decode.d7.loss_mask_dice: 1.9516 2023/09/06 19:56:16 - mmengine - INFO - Iter(train) [42850/60000] base_lr: 2.8584e-05 lr: 2.8584e-05 eta: 2:09:32 time: 0.4632 data_time: 0.0236 memory: 15775 grad_norm: 17.8550 loss: 9.4655 decode.loss_cls_ce: 2.0387 decode.loss_mask_ce: 0.9327 decode.loss_mask_dice: 1.7586 decode.d7.loss_cls_ce: 2.0421 decode.d7.loss_mask_ce: 0.9271 decode.d7.loss_mask_dice: 1.7662 2023/09/06 19:56:39 - mmengine - INFO - Iter(train) [42900/60000] base_lr: 2.8500e-05 lr: 2.8500e-05 eta: 2:09:09 time: 0.4640 data_time: 0.0240 memory: 15832 grad_norm: 19.0509 loss: 10.1475 decode.loss_cls_ce: 2.2101 decode.loss_mask_ce: 0.9449 decode.loss_mask_dice: 1.9049 decode.d7.loss_cls_ce: 2.2229 decode.d7.loss_mask_ce: 0.9505 decode.d7.loss_mask_dice: 1.9143 2023/09/06 19:57:02 - mmengine - INFO - Iter(train) [42950/60000] base_lr: 2.8417e-05 lr: 2.8417e-05 eta: 2:08:47 time: 0.4591 data_time: 0.0235 memory: 15794 grad_norm: 16.6960 loss: 7.5054 decode.loss_cls_ce: 1.6924 decode.loss_mask_ce: 0.7281 decode.loss_mask_dice: 1.3398 decode.d7.loss_cls_ce: 1.6867 decode.d7.loss_mask_ce: 0.7274 decode.d7.loss_mask_dice: 1.3310 2023/09/06 19:57:25 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 19:57:25 - mmengine - INFO - Iter(train) [43000/60000] base_lr: 2.8334e-05 lr: 2.8334e-05 eta: 2:08:24 time: 0.4612 data_time: 0.0233 memory: 15744 grad_norm: 23.6846 loss: 8.9162 decode.loss_cls_ce: 1.9672 decode.loss_mask_ce: 0.8464 decode.loss_mask_dice: 1.6434 decode.d7.loss_cls_ce: 1.9751 decode.d7.loss_mask_ce: 0.8409 decode.d7.loss_mask_dice: 1.6433 2023/09/06 19:57:48 - mmengine - INFO - Iter(train) [43050/60000] base_lr: 2.8250e-05 lr: 2.8250e-05 eta: 2:08:02 time: 0.4638 data_time: 0.0239 memory: 15772 grad_norm: 18.4715 loss: 10.2636 decode.loss_cls_ce: 2.3383 decode.loss_mask_ce: 0.9275 decode.loss_mask_dice: 1.8629 decode.d7.loss_cls_ce: 2.3509 decode.d7.loss_mask_ce: 0.9238 decode.d7.loss_mask_dice: 1.8602 2023/09/06 19:58:11 - mmengine - INFO - Iter(train) [43100/60000] base_lr: 2.8167e-05 lr: 2.8167e-05 eta: 2:07:39 time: 0.4616 data_time: 0.0238 memory: 15697 grad_norm: 17.5332 loss: 8.3659 decode.loss_cls_ce: 1.5996 decode.loss_mask_ce: 0.9205 decode.loss_mask_dice: 1.6494 decode.d7.loss_cls_ce: 1.6227 decode.d7.loss_mask_ce: 0.9257 decode.d7.loss_mask_dice: 1.6480 2023/09/06 19:58:34 - mmengine - INFO - Iter(train) [43150/60000] base_lr: 2.8084e-05 lr: 2.8084e-05 eta: 2:07:17 time: 0.4630 data_time: 0.0237 memory: 15682 grad_norm: 18.0515 loss: 9.2764 decode.loss_cls_ce: 2.0646 decode.loss_mask_ce: 0.9032 decode.loss_mask_dice: 1.6575 decode.d7.loss_cls_ce: 2.0726 decode.d7.loss_mask_ce: 0.9039 decode.d7.loss_mask_dice: 1.6747 2023/09/06 19:58:57 - mmengine - INFO - Iter(train) [43200/60000] base_lr: 2.8000e-05 lr: 2.8000e-05 eta: 2:06:54 time: 0.4517 data_time: 0.0245 memory: 15771 grad_norm: 18.5061 loss: 9.4179 decode.loss_cls_ce: 2.0876 decode.loss_mask_ce: 0.8822 decode.loss_mask_dice: 1.7154 decode.d7.loss_cls_ce: 2.1147 decode.d7.loss_mask_ce: 0.8911 decode.d7.loss_mask_dice: 1.7268 2023/09/06 19:59:20 - mmengine - INFO - Iter(train) [43250/60000] base_lr: 2.7917e-05 lr: 2.7917e-05 eta: 2:06:32 time: 0.4533 data_time: 0.0242 memory: 15858 grad_norm: 17.6723 loss: 8.5349 decode.loss_cls_ce: 1.7889 decode.loss_mask_ce: 0.8547 decode.loss_mask_dice: 1.6206 decode.d7.loss_cls_ce: 1.7863 decode.d7.loss_mask_ce: 0.8647 decode.d7.loss_mask_dice: 1.6197 2023/09/06 19:59:42 - mmengine - INFO - Iter(train) [43300/60000] base_lr: 2.7834e-05 lr: 2.7834e-05 eta: 2:06:09 time: 0.4557 data_time: 0.0242 memory: 15911 grad_norm: 16.7704 loss: 9.8224 decode.loss_cls_ce: 2.1314 decode.loss_mask_ce: 0.8699 decode.loss_mask_dice: 1.9106 decode.d7.loss_cls_ce: 2.1300 decode.d7.loss_mask_ce: 0.8620 decode.d7.loss_mask_dice: 1.9184 2023/09/06 20:00:05 - mmengine - INFO - Iter(train) [43350/60000] base_lr: 2.7750e-05 lr: 2.7750e-05 eta: 2:05:46 time: 0.4509 data_time: 0.0245 memory: 16050 grad_norm: 19.4569 loss: 8.7848 decode.loss_cls_ce: 1.9414 decode.loss_mask_ce: 0.7674 decode.loss_mask_dice: 1.6805 decode.d7.loss_cls_ce: 1.9553 decode.d7.loss_mask_ce: 0.7586 decode.d7.loss_mask_dice: 1.6816 2023/09/06 20:00:28 - mmengine - INFO - Iter(train) [43400/60000] base_lr: 2.7667e-05 lr: 2.7667e-05 eta: 2:05:24 time: 0.4538 data_time: 0.0236 memory: 15771 grad_norm: 18.0256 loss: 8.9157 decode.loss_cls_ce: 1.9650 decode.loss_mask_ce: 0.8164 decode.loss_mask_dice: 1.6690 decode.d7.loss_cls_ce: 1.9584 decode.d7.loss_mask_ce: 0.8171 decode.d7.loss_mask_dice: 1.6897 2023/09/06 20:00:51 - mmengine - INFO - Iter(train) [43450/60000] base_lr: 2.7584e-05 lr: 2.7584e-05 eta: 2:05:01 time: 0.4607 data_time: 0.0237 memory: 15733 grad_norm: 17.7656 loss: 8.8226 decode.loss_cls_ce: 1.9850 decode.loss_mask_ce: 0.8648 decode.loss_mask_dice: 1.5396 decode.d7.loss_cls_ce: 2.0277 decode.d7.loss_mask_ce: 0.8672 decode.d7.loss_mask_dice: 1.5382 2023/09/06 20:01:14 - mmengine - INFO - Iter(train) [43500/60000] base_lr: 2.7500e-05 lr: 2.7500e-05 eta: 2:04:39 time: 0.4611 data_time: 0.0234 memory: 15845 grad_norm: 17.6686 loss: 9.7924 decode.loss_cls_ce: 2.0705 decode.loss_mask_ce: 0.9278 decode.loss_mask_dice: 1.8782 decode.d7.loss_cls_ce: 2.1125 decode.d7.loss_mask_ce: 0.9306 decode.d7.loss_mask_dice: 1.8729 2023/09/06 20:01:37 - mmengine - INFO - Iter(train) [43550/60000] base_lr: 2.7417e-05 lr: 2.7417e-05 eta: 2:04:16 time: 0.4568 data_time: 0.0246 memory: 15989 grad_norm: 17.1764 loss: 9.0283 decode.loss_cls_ce: 1.9592 decode.loss_mask_ce: 0.8568 decode.loss_mask_dice: 1.6859 decode.d7.loss_cls_ce: 1.9747 decode.d7.loss_mask_ce: 0.8549 decode.d7.loss_mask_dice: 1.6968 2023/09/06 20:01:59 - mmengine - INFO - Iter(train) [43600/60000] base_lr: 2.7334e-05 lr: 2.7334e-05 eta: 2:03:53 time: 0.4535 data_time: 0.0254 memory: 15796 grad_norm: 18.0006 loss: 10.0636 decode.loss_cls_ce: 2.2473 decode.loss_mask_ce: 0.9298 decode.loss_mask_dice: 1.8487 decode.d7.loss_cls_ce: 2.2557 decode.d7.loss_mask_ce: 0.9324 decode.d7.loss_mask_dice: 1.8497 2023/09/06 20:02:22 - mmengine - INFO - Iter(train) [43650/60000] base_lr: 2.7250e-05 lr: 2.7250e-05 eta: 2:03:31 time: 0.4610 data_time: 0.0240 memory: 15836 grad_norm: 18.0438 loss: 9.4791 decode.loss_cls_ce: 2.1435 decode.loss_mask_ce: 0.8893 decode.loss_mask_dice: 1.7072 decode.d7.loss_cls_ce: 2.1409 decode.d7.loss_mask_ce: 0.8892 decode.d7.loss_mask_dice: 1.7090 2023/09/06 20:02:45 - mmengine - INFO - Iter(train) [43700/60000] base_lr: 2.7167e-05 lr: 2.7167e-05 eta: 2:03:08 time: 0.4624 data_time: 0.0237 memory: 15823 grad_norm: 16.9180 loss: 7.8678 decode.loss_cls_ce: 1.7171 decode.loss_mask_ce: 0.7716 decode.loss_mask_dice: 1.4506 decode.d7.loss_cls_ce: 1.7101 decode.d7.loss_mask_ce: 0.7763 decode.d7.loss_mask_dice: 1.4421 2023/09/06 20:03:08 - mmengine - INFO - Iter(train) [43750/60000] base_lr: 2.7084e-05 lr: 2.7084e-05 eta: 2:02:46 time: 0.4611 data_time: 0.0241 memory: 15883 grad_norm: 16.9551 loss: 8.8551 decode.loss_cls_ce: 1.9701 decode.loss_mask_ce: 0.8335 decode.loss_mask_dice: 1.6126 decode.d7.loss_cls_ce: 1.9838 decode.d7.loss_mask_ce: 0.8331 decode.d7.loss_mask_dice: 1.6220 2023/09/06 20:03:31 - mmengine - INFO - Iter(train) [43800/60000] base_lr: 2.7000e-05 lr: 2.7000e-05 eta: 2:02:23 time: 0.4596 data_time: 0.0244 memory: 15822 grad_norm: 17.9561 loss: 8.0572 decode.loss_cls_ce: 1.7725 decode.loss_mask_ce: 0.7555 decode.loss_mask_dice: 1.4841 decode.d7.loss_cls_ce: 1.7924 decode.d7.loss_mask_ce: 0.7612 decode.d7.loss_mask_dice: 1.4914 2023/09/06 20:03:54 - mmengine - INFO - Iter(train) [43850/60000] base_lr: 2.6917e-05 lr: 2.6917e-05 eta: 2:02:00 time: 0.4601 data_time: 0.0239 memory: 15821 grad_norm: 17.7377 loss: 8.1865 decode.loss_cls_ce: 1.8713 decode.loss_mask_ce: 0.7984 decode.loss_mask_dice: 1.4184 decode.d7.loss_cls_ce: 1.8719 decode.d7.loss_mask_ce: 0.8034 decode.d7.loss_mask_dice: 1.4231 2023/09/06 20:04:17 - mmengine - INFO - Iter(train) [43900/60000] base_lr: 2.6834e-05 lr: 2.6834e-05 eta: 2:01:38 time: 0.4626 data_time: 0.0233 memory: 15910 grad_norm: 18.0627 loss: 9.6309 decode.loss_cls_ce: 2.0137 decode.loss_mask_ce: 0.9656 decode.loss_mask_dice: 1.8280 decode.d7.loss_cls_ce: 2.0135 decode.d7.loss_mask_ce: 0.9740 decode.d7.loss_mask_dice: 1.8359 2023/09/06 20:04:40 - mmengine - INFO - Iter(train) [43950/60000] base_lr: 2.6750e-05 lr: 2.6750e-05 eta: 2:01:15 time: 0.4647 data_time: 0.0238 memory: 15818 grad_norm: 20.2326 loss: 9.4676 decode.loss_cls_ce: 2.0983 decode.loss_mask_ce: 0.8770 decode.loss_mask_dice: 1.7491 decode.d7.loss_cls_ce: 2.1175 decode.d7.loss_mask_ce: 0.8775 decode.d7.loss_mask_dice: 1.7481 2023/09/06 20:05:03 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:05:03 - mmengine - INFO - Iter(train) [44000/60000] base_lr: 2.6667e-05 lr: 2.6667e-05 eta: 2:00:53 time: 0.4498 data_time: 0.0239 memory: 15849 grad_norm: 17.2224 loss: 9.9645 decode.loss_cls_ce: 2.2487 decode.loss_mask_ce: 0.8816 decode.loss_mask_dice: 1.8623 decode.d7.loss_cls_ce: 2.2352 decode.d7.loss_mask_ce: 0.8741 decode.d7.loss_mask_dice: 1.8627 2023/09/06 20:05:26 - mmengine - INFO - Iter(train) [44050/60000] base_lr: 2.6584e-05 lr: 2.6584e-05 eta: 2:00:30 time: 0.4599 data_time: 0.0245 memory: 15847 grad_norm: 18.2355 loss: 9.4291 decode.loss_cls_ce: 2.1665 decode.loss_mask_ce: 0.8671 decode.loss_mask_dice: 1.6715 decode.d7.loss_cls_ce: 2.1917 decode.d7.loss_mask_ce: 0.8674 decode.d7.loss_mask_dice: 1.6649 2023/09/06 20:05:49 - mmengine - INFO - Iter(train) [44100/60000] base_lr: 2.6500e-05 lr: 2.6500e-05 eta: 2:00:08 time: 0.4537 data_time: 0.0248 memory: 15832 grad_norm: 16.8557 loss: 9.5383 decode.loss_cls_ce: 2.1205 decode.loss_mask_ce: 0.8350 decode.loss_mask_dice: 1.7905 decode.d7.loss_cls_ce: 2.1467 decode.d7.loss_mask_ce: 0.8432 decode.d7.loss_mask_dice: 1.8024 2023/09/06 20:06:11 - mmengine - INFO - Iter(train) [44150/60000] base_lr: 2.6417e-05 lr: 2.6417e-05 eta: 1:59:45 time: 0.4527 data_time: 0.0248 memory: 15773 grad_norm: 16.2980 loss: 8.6942 decode.loss_cls_ce: 1.8728 decode.loss_mask_ce: 0.8522 decode.loss_mask_dice: 1.6192 decode.d7.loss_cls_ce: 1.8914 decode.d7.loss_mask_ce: 0.8503 decode.d7.loss_mask_dice: 1.6082 2023/09/06 20:06:34 - mmengine - INFO - Iter(train) [44200/60000] base_lr: 2.6334e-05 lr: 2.6334e-05 eta: 1:59:22 time: 0.4504 data_time: 0.0240 memory: 15899 grad_norm: 16.3991 loss: 9.7249 decode.loss_cls_ce: 2.0941 decode.loss_mask_ce: 0.9281 decode.loss_mask_dice: 1.8374 decode.d7.loss_cls_ce: 2.0867 decode.d7.loss_mask_ce: 0.9283 decode.d7.loss_mask_dice: 1.8504 2023/09/06 20:06:57 - mmengine - INFO - Iter(train) [44250/60000] base_lr: 2.6250e-05 lr: 2.6250e-05 eta: 1:58:59 time: 0.4553 data_time: 0.0235 memory: 15784 grad_norm: 18.1117 loss: 9.3050 decode.loss_cls_ce: 1.9805 decode.loss_mask_ce: 0.8603 decode.loss_mask_dice: 1.8157 decode.d7.loss_cls_ce: 1.9631 decode.d7.loss_mask_ce: 0.8687 decode.d7.loss_mask_dice: 1.8167 2023/09/06 20:07:20 - mmengine - INFO - Iter(train) [44300/60000] base_lr: 2.6167e-05 lr: 2.6167e-05 eta: 1:58:37 time: 0.4579 data_time: 0.0237 memory: 15731 grad_norm: 16.3103 loss: 8.9490 decode.loss_cls_ce: 1.9781 decode.loss_mask_ce: 0.8149 decode.loss_mask_dice: 1.6845 decode.d7.loss_cls_ce: 1.9636 decode.d7.loss_mask_ce: 0.8095 decode.d7.loss_mask_dice: 1.6985 2023/09/06 20:07:42 - mmengine - INFO - Iter(train) [44350/60000] base_lr: 2.6084e-05 lr: 2.6084e-05 eta: 1:58:14 time: 0.4546 data_time: 0.0244 memory: 15920 grad_norm: 19.4001 loss: 9.8336 decode.loss_cls_ce: 2.2806 decode.loss_mask_ce: 0.8564 decode.loss_mask_dice: 1.7827 decode.d7.loss_cls_ce: 2.2612 decode.d7.loss_mask_ce: 0.8641 decode.d7.loss_mask_dice: 1.7886 2023/09/06 20:08:05 - mmengine - INFO - Iter(train) [44400/60000] base_lr: 2.6000e-05 lr: 2.6000e-05 eta: 1:57:52 time: 0.4591 data_time: 0.0234 memory: 15834 grad_norm: 23.0319 loss: 8.7494 decode.loss_cls_ce: 1.8856 decode.loss_mask_ce: 0.8560 decode.loss_mask_dice: 1.6474 decode.d7.loss_cls_ce: 1.8760 decode.d7.loss_mask_ce: 0.8509 decode.d7.loss_mask_dice: 1.6335 2023/09/06 20:08:28 - mmengine - INFO - Iter(train) [44450/60000] base_lr: 2.5917e-05 lr: 2.5917e-05 eta: 1:57:29 time: 0.4622 data_time: 0.0236 memory: 15884 grad_norm: 18.9313 loss: 8.1056 decode.loss_cls_ce: 1.7657 decode.loss_mask_ce: 0.8250 decode.loss_mask_dice: 1.4648 decode.d7.loss_cls_ce: 1.7687 decode.d7.loss_mask_ce: 0.8178 decode.d7.loss_mask_dice: 1.4636 2023/09/06 20:08:51 - mmengine - INFO - Iter(train) [44500/60000] base_lr: 2.5834e-05 lr: 2.5834e-05 eta: 1:57:07 time: 0.4565 data_time: 0.0237 memory: 15796 grad_norm: 17.8500 loss: 9.8376 decode.loss_cls_ce: 2.0033 decode.loss_mask_ce: 0.9673 decode.loss_mask_dice: 1.9519 decode.d7.loss_cls_ce: 1.9978 decode.d7.loss_mask_ce: 0.9692 decode.d7.loss_mask_dice: 1.9481 2023/09/06 20:09:14 - mmengine - INFO - Iter(train) [44550/60000] base_lr: 2.5750e-05 lr: 2.5750e-05 eta: 1:56:44 time: 0.4526 data_time: 0.0242 memory: 15886 grad_norm: 16.5655 loss: 9.3533 decode.loss_cls_ce: 2.0143 decode.loss_mask_ce: 0.9077 decode.loss_mask_dice: 1.7556 decode.d7.loss_cls_ce: 2.0268 decode.d7.loss_mask_ce: 0.9105 decode.d7.loss_mask_dice: 1.7384 2023/09/06 20:09:37 - mmengine - INFO - Iter(train) [44600/60000] base_lr: 2.5667e-05 lr: 2.5667e-05 eta: 1:56:21 time: 0.4525 data_time: 0.0240 memory: 15693 grad_norm: 17.4893 loss: 9.2057 decode.loss_cls_ce: 2.0272 decode.loss_mask_ce: 0.8878 decode.loss_mask_dice: 1.6896 decode.d7.loss_cls_ce: 2.0684 decode.d7.loss_mask_ce: 0.8731 decode.d7.loss_mask_dice: 1.6597 2023/09/06 20:09:59 - mmengine - INFO - Iter(train) [44650/60000] base_lr: 2.5584e-05 lr: 2.5584e-05 eta: 1:55:59 time: 0.4581 data_time: 0.0239 memory: 15758 grad_norm: 20.6573 loss: 8.8238 decode.loss_cls_ce: 1.9405 decode.loss_mask_ce: 0.8376 decode.loss_mask_dice: 1.6297 decode.d7.loss_cls_ce: 1.9518 decode.d7.loss_mask_ce: 0.8380 decode.d7.loss_mask_dice: 1.6262 2023/09/06 20:10:22 - mmengine - INFO - Iter(train) [44700/60000] base_lr: 2.5500e-05 lr: 2.5500e-05 eta: 1:55:36 time: 0.4576 data_time: 0.0242 memory: 15885 grad_norm: 14.8128 loss: 8.6547 decode.loss_cls_ce: 1.9510 decode.loss_mask_ce: 0.8448 decode.loss_mask_dice: 1.5466 decode.d7.loss_cls_ce: 1.9281 decode.d7.loss_mask_ce: 0.8491 decode.d7.loss_mask_dice: 1.5351 2023/09/06 20:10:45 - mmengine - INFO - Iter(train) [44750/60000] base_lr: 2.5417e-05 lr: 2.5417e-05 eta: 1:55:13 time: 0.4516 data_time: 0.0239 memory: 15642 grad_norm: 21.6993 loss: 7.7502 decode.loss_cls_ce: 1.7748 decode.loss_mask_ce: 0.6755 decode.loss_mask_dice: 1.4306 decode.d7.loss_cls_ce: 1.7479 decode.d7.loss_mask_ce: 0.6902 decode.d7.loss_mask_dice: 1.4312 2023/09/06 20:11:08 - mmengine - INFO - Iter(train) [44800/60000] base_lr: 2.5334e-05 lr: 2.5334e-05 eta: 1:54:51 time: 0.4599 data_time: 0.0228 memory: 15764 grad_norm: 17.3340 loss: 8.1564 decode.loss_cls_ce: 1.7521 decode.loss_mask_ce: 0.7817 decode.loss_mask_dice: 1.5450 decode.d7.loss_cls_ce: 1.7416 decode.d7.loss_mask_ce: 0.7893 decode.d7.loss_mask_dice: 1.5467 2023/09/06 20:11:31 - mmengine - INFO - Iter(train) [44850/60000] base_lr: 2.5250e-05 lr: 2.5250e-05 eta: 1:54:28 time: 0.4536 data_time: 0.0240 memory: 15925 grad_norm: 17.9855 loss: 9.5381 decode.loss_cls_ce: 2.0801 decode.loss_mask_ce: 0.8643 decode.loss_mask_dice: 1.8254 decode.d7.loss_cls_ce: 2.0604 decode.d7.loss_mask_ce: 0.8768 decode.d7.loss_mask_dice: 1.8310 2023/09/06 20:11:54 - mmengine - INFO - Iter(train) [44900/60000] base_lr: 2.5167e-05 lr: 2.5167e-05 eta: 1:54:06 time: 0.4630 data_time: 0.0243 memory: 15819 grad_norm: 16.9358 loss: 8.8701 decode.loss_cls_ce: 2.0074 decode.loss_mask_ce: 0.8786 decode.loss_mask_dice: 1.5693 decode.d7.loss_cls_ce: 1.9750 decode.d7.loss_mask_ce: 0.8767 decode.d7.loss_mask_dice: 1.5631 2023/09/06 20:12:16 - mmengine - INFO - Iter(train) [44950/60000] base_lr: 2.5084e-05 lr: 2.5084e-05 eta: 1:53:43 time: 0.4531 data_time: 0.0243 memory: 16006 grad_norm: 17.2600 loss: 9.3964 decode.loss_cls_ce: 2.0534 decode.loss_mask_ce: 0.9003 decode.loss_mask_dice: 1.7487 decode.d7.loss_cls_ce: 2.0333 decode.d7.loss_mask_ce: 0.8990 decode.d7.loss_mask_dice: 1.7617 2023/09/06 20:12:39 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:12:39 - mmengine - INFO - Iter(train) [45000/60000] base_lr: 2.5000e-05 lr: 2.5000e-05 eta: 1:53:20 time: 0.4601 data_time: 0.0229 memory: 16106 grad_norm: 19.1394 loss: 9.3772 decode.loss_cls_ce: 2.0597 decode.loss_mask_ce: 0.9127 decode.loss_mask_dice: 1.7189 decode.d7.loss_cls_ce: 2.0577 decode.d7.loss_mask_ce: 0.9106 decode.d7.loss_mask_dice: 1.7175 2023/09/06 20:13:02 - mmengine - INFO - Iter(train) [45050/60000] base_lr: 2.4917e-05 lr: 2.4917e-05 eta: 1:52:58 time: 0.4498 data_time: 0.0225 memory: 15861 grad_norm: 17.6070 loss: 9.1171 decode.loss_cls_ce: 1.9242 decode.loss_mask_ce: 0.8999 decode.loss_mask_dice: 1.7331 decode.d7.loss_cls_ce: 1.8983 decode.d7.loss_mask_ce: 0.9106 decode.d7.loss_mask_dice: 1.7511 2023/09/06 20:13:25 - mmengine - INFO - Iter(train) [45100/60000] base_lr: 2.4834e-05 lr: 2.4834e-05 eta: 1:52:35 time: 0.4569 data_time: 0.0233 memory: 15950 grad_norm: 18.9003 loss: 9.9041 decode.loss_cls_ce: 2.2031 decode.loss_mask_ce: 0.8407 decode.loss_mask_dice: 1.9135 decode.d7.loss_cls_ce: 2.1839 decode.d7.loss_mask_ce: 0.8451 decode.d7.loss_mask_dice: 1.9178 2023/09/06 20:13:48 - mmengine - INFO - Iter(train) [45150/60000] base_lr: 2.4750e-05 lr: 2.4750e-05 eta: 1:52:12 time: 0.4530 data_time: 0.0249 memory: 15844 grad_norm: 20.0488 loss: 9.2410 decode.loss_cls_ce: 2.0490 decode.loss_mask_ce: 0.8687 decode.loss_mask_dice: 1.7060 decode.d7.loss_cls_ce: 2.0642 decode.d7.loss_mask_ce: 0.8674 decode.d7.loss_mask_dice: 1.6857 2023/09/06 20:14:11 - mmengine - INFO - Iter(train) [45200/60000] base_lr: 2.4667e-05 lr: 2.4667e-05 eta: 1:51:50 time: 0.4529 data_time: 0.0246 memory: 15923 grad_norm: 17.2542 loss: 8.8545 decode.loss_cls_ce: 2.0217 decode.loss_mask_ce: 0.8632 decode.loss_mask_dice: 1.5276 decode.d7.loss_cls_ce: 2.0309 decode.d7.loss_mask_ce: 0.8692 decode.d7.loss_mask_dice: 1.5419 2023/09/06 20:14:33 - mmengine - INFO - Iter(train) [45250/60000] base_lr: 2.4584e-05 lr: 2.4584e-05 eta: 1:51:27 time: 0.4610 data_time: 0.0231 memory: 15823 grad_norm: 20.9126 loss: 9.6032 decode.loss_cls_ce: 2.1243 decode.loss_mask_ce: 0.9168 decode.loss_mask_dice: 1.7657 decode.d7.loss_cls_ce: 2.1212 decode.d7.loss_mask_ce: 0.9004 decode.d7.loss_mask_dice: 1.7749 2023/09/06 20:14:57 - mmengine - INFO - Iter(train) [45300/60000] base_lr: 2.4500e-05 lr: 2.4500e-05 eta: 1:51:05 time: 0.4591 data_time: 0.0229 memory: 15769 grad_norm: 18.9224 loss: 8.9669 decode.loss_cls_ce: 1.9691 decode.loss_mask_ce: 0.8679 decode.loss_mask_dice: 1.6426 decode.d7.loss_cls_ce: 2.0083 decode.d7.loss_mask_ce: 0.8565 decode.d7.loss_mask_dice: 1.6224 2023/09/06 20:15:19 - mmengine - INFO - Iter(train) [45350/60000] base_lr: 2.4417e-05 lr: 2.4417e-05 eta: 1:50:42 time: 0.4514 data_time: 0.0248 memory: 15898 grad_norm: 18.3357 loss: 7.9621 decode.loss_cls_ce: 1.8042 decode.loss_mask_ce: 0.7948 decode.loss_mask_dice: 1.3782 decode.d7.loss_cls_ce: 1.8222 decode.d7.loss_mask_ce: 0.7827 decode.d7.loss_mask_dice: 1.3800 2023/09/06 20:15:42 - mmengine - INFO - Iter(train) [45400/60000] base_lr: 2.4334e-05 lr: 2.4334e-05 eta: 1:50:19 time: 0.4516 data_time: 0.0247 memory: 15948 grad_norm: 19.3282 loss: 7.8220 decode.loss_cls_ce: 1.7153 decode.loss_mask_ce: 0.7893 decode.loss_mask_dice: 1.3940 decode.d7.loss_cls_ce: 1.7332 decode.d7.loss_mask_ce: 0.7888 decode.d7.loss_mask_dice: 1.4014 2023/09/06 20:16:05 - mmengine - INFO - Iter(train) [45450/60000] base_lr: 2.4250e-05 lr: 2.4250e-05 eta: 1:49:57 time: 0.4620 data_time: 0.0242 memory: 15924 grad_norm: 18.2957 loss: 8.7704 decode.loss_cls_ce: 1.8759 decode.loss_mask_ce: 0.8618 decode.loss_mask_dice: 1.6491 decode.d7.loss_cls_ce: 1.8808 decode.d7.loss_mask_ce: 0.8607 decode.d7.loss_mask_dice: 1.6421 2023/09/06 20:16:28 - mmengine - INFO - Iter(train) [45500/60000] base_lr: 2.4167e-05 lr: 2.4167e-05 eta: 1:49:34 time: 0.4594 data_time: 0.0238 memory: 15950 grad_norm: 16.9203 loss: 9.4063 decode.loss_cls_ce: 2.0945 decode.loss_mask_ce: 0.8156 decode.loss_mask_dice: 1.7880 decode.d7.loss_cls_ce: 2.0935 decode.d7.loss_mask_ce: 0.8217 decode.d7.loss_mask_dice: 1.7930 2023/09/06 20:16:51 - mmengine - INFO - Iter(train) [45550/60000] base_lr: 2.4084e-05 lr: 2.4084e-05 eta: 1:49:12 time: 0.4637 data_time: 0.0238 memory: 15783 grad_norm: 18.2359 loss: 9.4325 decode.loss_cls_ce: 2.0690 decode.loss_mask_ce: 0.8751 decode.loss_mask_dice: 1.7768 decode.d7.loss_cls_ce: 2.0440 decode.d7.loss_mask_ce: 0.8789 decode.d7.loss_mask_dice: 1.7886 2023/09/06 20:17:14 - mmengine - INFO - Iter(train) [45600/60000] base_lr: 2.4000e-05 lr: 2.4000e-05 eta: 1:48:49 time: 0.4573 data_time: 0.0237 memory: 15833 grad_norm: 20.5075 loss: 8.9812 decode.loss_cls_ce: 1.9332 decode.loss_mask_ce: 0.9227 decode.loss_mask_dice: 1.6217 decode.d7.loss_cls_ce: 1.9502 decode.d7.loss_mask_ce: 0.9243 decode.d7.loss_mask_dice: 1.6290 2023/09/06 20:17:37 - mmengine - INFO - Iter(train) [45650/60000] base_lr: 2.3917e-05 lr: 2.3917e-05 eta: 1:48:26 time: 0.4522 data_time: 0.0242 memory: 16014 grad_norm: 15.8993 loss: 8.4968 decode.loss_cls_ce: 1.8056 decode.loss_mask_ce: 0.8463 decode.loss_mask_dice: 1.5978 decode.d7.loss_cls_ce: 1.8055 decode.d7.loss_mask_ce: 0.8391 decode.d7.loss_mask_dice: 1.6025 2023/09/06 20:18:00 - mmengine - INFO - Iter(train) [45700/60000] base_lr: 2.3834e-05 lr: 2.3834e-05 eta: 1:48:04 time: 0.4624 data_time: 0.0236 memory: 15860 grad_norm: 19.4083 loss: 9.8741 decode.loss_cls_ce: 1.9939 decode.loss_mask_ce: 0.9502 decode.loss_mask_dice: 1.9730 decode.d7.loss_cls_ce: 1.9998 decode.d7.loss_mask_ce: 0.9654 decode.d7.loss_mask_dice: 1.9918 2023/09/06 20:18:23 - mmengine - INFO - Iter(train) [45750/60000] base_lr: 2.3750e-05 lr: 2.3750e-05 eta: 1:47:41 time: 0.4541 data_time: 0.0240 memory: 15912 grad_norm: 17.1936 loss: 8.8028 decode.loss_cls_ce: 1.9526 decode.loss_mask_ce: 0.8072 decode.loss_mask_dice: 1.6475 decode.d7.loss_cls_ce: 1.9506 decode.d7.loss_mask_ce: 0.8041 decode.d7.loss_mask_dice: 1.6408 2023/09/06 20:18:45 - mmengine - INFO - Iter(train) [45800/60000] base_lr: 2.3667e-05 lr: 2.3667e-05 eta: 1:47:19 time: 0.4532 data_time: 0.0246 memory: 15768 grad_norm: 17.5911 loss: 9.2025 decode.loss_cls_ce: 1.9505 decode.loss_mask_ce: 0.9094 decode.loss_mask_dice: 1.7288 decode.d7.loss_cls_ce: 1.9722 decode.d7.loss_mask_ce: 0.9053 decode.d7.loss_mask_dice: 1.7363 2023/09/06 20:19:08 - mmengine - INFO - Iter(train) [45850/60000] base_lr: 2.3584e-05 lr: 2.3584e-05 eta: 1:46:56 time: 0.4625 data_time: 0.0237 memory: 15782 grad_norm: 18.1712 loss: 10.5243 decode.loss_cls_ce: 2.4451 decode.loss_mask_ce: 0.8913 decode.loss_mask_dice: 1.9299 decode.d7.loss_cls_ce: 2.4457 decode.d7.loss_mask_ce: 0.8841 decode.d7.loss_mask_dice: 1.9282 2023/09/06 20:19:31 - mmengine - INFO - Iter(train) [45900/60000] base_lr: 2.3500e-05 lr: 2.3500e-05 eta: 1:46:33 time: 0.4508 data_time: 0.0240 memory: 15798 grad_norm: 25.4380 loss: 9.6432 decode.loss_cls_ce: 2.0776 decode.loss_mask_ce: 0.8913 decode.loss_mask_dice: 1.8522 decode.d7.loss_cls_ce: 2.0856 decode.d7.loss_mask_ce: 0.8860 decode.d7.loss_mask_dice: 1.8504 2023/09/06 20:19:54 - mmengine - INFO - Iter(train) [45950/60000] base_lr: 2.3417e-05 lr: 2.3417e-05 eta: 1:46:11 time: 0.4592 data_time: 0.0232 memory: 15932 grad_norm: 19.7624 loss: 8.5698 decode.loss_cls_ce: 1.8848 decode.loss_mask_ce: 0.8483 decode.loss_mask_dice: 1.5584 decode.d7.loss_cls_ce: 1.8909 decode.d7.loss_mask_ce: 0.8354 decode.d7.loss_mask_dice: 1.5519 2023/09/06 20:20:17 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:20:17 - mmengine - INFO - Iter(train) [46000/60000] base_lr: 2.3334e-05 lr: 2.3334e-05 eta: 1:45:48 time: 0.4540 data_time: 0.0240 memory: 15990 grad_norm: 19.4754 loss: 9.2905 decode.loss_cls_ce: 1.9802 decode.loss_mask_ce: 0.9378 decode.loss_mask_dice: 1.7131 decode.d7.loss_cls_ce: 1.9869 decode.d7.loss_mask_ce: 0.9470 decode.d7.loss_mask_dice: 1.7256 2023/09/06 20:20:40 - mmengine - INFO - Iter(train) [46050/60000] base_lr: 2.3250e-05 lr: 2.3250e-05 eta: 1:45:25 time: 0.4533 data_time: 0.0242 memory: 15819 grad_norm: 19.0444 loss: 8.6664 decode.loss_cls_ce: 1.8701 decode.loss_mask_ce: 0.8900 decode.loss_mask_dice: 1.5803 decode.d7.loss_cls_ce: 1.8428 decode.d7.loss_mask_ce: 0.8905 decode.d7.loss_mask_dice: 1.5927 2023/09/06 20:21:02 - mmengine - INFO - Iter(train) [46100/60000] base_lr: 2.3167e-05 lr: 2.3167e-05 eta: 1:45:03 time: 0.4510 data_time: 0.0246 memory: 15860 grad_norm: 17.4855 loss: 9.7198 decode.loss_cls_ce: 2.0189 decode.loss_mask_ce: 0.9874 decode.loss_mask_dice: 1.8434 decode.d7.loss_cls_ce: 2.0196 decode.d7.loss_mask_ce: 0.9872 decode.d7.loss_mask_dice: 1.8633 2023/09/06 20:21:25 - mmengine - INFO - Iter(train) [46150/60000] base_lr: 2.3084e-05 lr: 2.3084e-05 eta: 1:44:40 time: 0.4535 data_time: 0.0247 memory: 15783 grad_norm: 17.1705 loss: 9.1006 decode.loss_cls_ce: 2.0494 decode.loss_mask_ce: 0.8973 decode.loss_mask_dice: 1.6019 decode.d7.loss_cls_ce: 2.0400 decode.d7.loss_mask_ce: 0.9015 decode.d7.loss_mask_dice: 1.6104 2023/09/06 20:21:48 - mmengine - INFO - Iter(train) [46200/60000] base_lr: 2.3000e-05 lr: 2.3000e-05 eta: 1:44:17 time: 0.4569 data_time: 0.0234 memory: 15804 grad_norm: 18.0647 loss: 9.1485 decode.loss_cls_ce: 1.9187 decode.loss_mask_ce: 0.9297 decode.loss_mask_dice: 1.7143 decode.d7.loss_cls_ce: 1.9270 decode.d7.loss_mask_ce: 0.9292 decode.d7.loss_mask_dice: 1.7295 2023/09/06 20:22:11 - mmengine - INFO - Iter(train) [46250/60000] base_lr: 2.2917e-05 lr: 2.2917e-05 eta: 1:43:55 time: 0.4623 data_time: 0.0234 memory: 15780 grad_norm: 17.2874 loss: 8.9558 decode.loss_cls_ce: 2.0072 decode.loss_mask_ce: 0.7971 decode.loss_mask_dice: 1.6751 decode.d7.loss_cls_ce: 2.0086 decode.d7.loss_mask_ce: 0.7926 decode.d7.loss_mask_dice: 1.6751 2023/09/06 20:22:34 - mmengine - INFO - Iter(train) [46300/60000] base_lr: 2.2834e-05 lr: 2.2834e-05 eta: 1:43:32 time: 0.4540 data_time: 0.0251 memory: 15721 grad_norm: 17.9414 loss: 9.8344 decode.loss_cls_ce: 2.0450 decode.loss_mask_ce: 0.9499 decode.loss_mask_dice: 1.9153 decode.d7.loss_cls_ce: 2.0450 decode.d7.loss_mask_ce: 0.9549 decode.d7.loss_mask_dice: 1.9244 2023/09/06 20:22:57 - mmengine - INFO - Iter(train) [46350/60000] base_lr: 2.2750e-05 lr: 2.2750e-05 eta: 1:43:10 time: 0.4518 data_time: 0.0247 memory: 15759 grad_norm: 19.1014 loss: 8.4012 decode.loss_cls_ce: 1.7513 decode.loss_mask_ce: 0.9419 decode.loss_mask_dice: 1.4909 decode.d7.loss_cls_ce: 1.7693 decode.d7.loss_mask_ce: 0.9449 decode.d7.loss_mask_dice: 1.5029 2023/09/06 20:23:19 - mmengine - INFO - Iter(train) [46400/60000] base_lr: 2.2667e-05 lr: 2.2667e-05 eta: 1:42:47 time: 0.4618 data_time: 0.0240 memory: 15938 grad_norm: 18.4941 loss: 8.9871 decode.loss_cls_ce: 2.0555 decode.loss_mask_ce: 0.8468 decode.loss_mask_dice: 1.5951 decode.d7.loss_cls_ce: 2.0606 decode.d7.loss_mask_ce: 0.8460 decode.d7.loss_mask_dice: 1.5831 2023/09/06 20:23:42 - mmengine - INFO - Iter(train) [46450/60000] base_lr: 2.2584e-05 lr: 2.2584e-05 eta: 1:42:24 time: 0.4620 data_time: 0.0238 memory: 15797 grad_norm: 16.9141 loss: 9.7255 decode.loss_cls_ce: 2.1586 decode.loss_mask_ce: 0.8978 decode.loss_mask_dice: 1.8070 decode.d7.loss_cls_ce: 2.1469 decode.d7.loss_mask_ce: 0.8996 decode.d7.loss_mask_dice: 1.8155 2023/09/06 20:24:06 - mmengine - INFO - Iter(train) [46500/60000] base_lr: 2.2500e-05 lr: 2.2500e-05 eta: 1:42:02 time: 0.4629 data_time: 0.0235 memory: 15822 grad_norm: 16.8298 loss: 8.7475 decode.loss_cls_ce: 1.7558 decode.loss_mask_ce: 0.8627 decode.loss_mask_dice: 1.7564 decode.d7.loss_cls_ce: 1.7510 decode.d7.loss_mask_ce: 0.8662 decode.d7.loss_mask_dice: 1.7553 2023/09/06 20:24:28 - mmengine - INFO - Iter(train) [46550/60000] base_lr: 2.2417e-05 lr: 2.2417e-05 eta: 1:41:39 time: 0.4562 data_time: 0.0230 memory: 15978 grad_norm: 18.3787 loss: 10.6398 decode.loss_cls_ce: 2.1733 decode.loss_mask_ce: 1.1157 decode.loss_mask_dice: 2.0135 decode.d7.loss_cls_ce: 2.1836 decode.d7.loss_mask_ce: 1.1198 decode.d7.loss_mask_dice: 2.0339 2023/09/06 20:24:51 - mmengine - INFO - Iter(train) [46600/60000] base_lr: 2.2334e-05 lr: 2.2334e-05 eta: 1:41:17 time: 0.4545 data_time: 0.0250 memory: 15912 grad_norm: 18.2451 loss: 9.1058 decode.loss_cls_ce: 2.0010 decode.loss_mask_ce: 0.9218 decode.loss_mask_dice: 1.6304 decode.d7.loss_cls_ce: 1.9901 decode.d7.loss_mask_ce: 0.9138 decode.d7.loss_mask_dice: 1.6488 2023/09/06 20:25:14 - mmengine - INFO - Iter(train) [46650/60000] base_lr: 2.2250e-05 lr: 2.2250e-05 eta: 1:40:54 time: 0.4622 data_time: 0.0241 memory: 15774 grad_norm: 18.3268 loss: 9.2115 decode.loss_cls_ce: 2.0623 decode.loss_mask_ce: 0.8759 decode.loss_mask_dice: 1.6639 decode.d7.loss_cls_ce: 2.0529 decode.d7.loss_mask_ce: 0.8849 decode.d7.loss_mask_dice: 1.6717 2023/09/06 20:25:37 - mmengine - INFO - Iter(train) [46700/60000] base_lr: 2.2167e-05 lr: 2.2167e-05 eta: 1:40:31 time: 0.4518 data_time: 0.0237 memory: 15797 grad_norm: 17.6221 loss: 9.5039 decode.loss_cls_ce: 2.0517 decode.loss_mask_ce: 0.8930 decode.loss_mask_dice: 1.7984 decode.d7.loss_cls_ce: 2.0790 decode.d7.loss_mask_ce: 0.8946 decode.d7.loss_mask_dice: 1.7872 2023/09/06 20:26:00 - mmengine - INFO - Iter(train) [46750/60000] base_lr: 2.2084e-05 lr: 2.2084e-05 eta: 1:40:09 time: 0.4630 data_time: 0.0246 memory: 15848 grad_norm: 18.7692 loss: 10.0518 decode.loss_cls_ce: 2.2074 decode.loss_mask_ce: 0.9637 decode.loss_mask_dice: 1.8481 decode.d7.loss_cls_ce: 2.2171 decode.d7.loss_mask_ce: 0.9635 decode.d7.loss_mask_dice: 1.8521 2023/09/06 20:26:23 - mmengine - INFO - Iter(train) [46800/60000] base_lr: 2.2000e-05 lr: 2.2000e-05 eta: 1:39:46 time: 0.4500 data_time: 0.0240 memory: 15898 grad_norm: 17.6891 loss: 8.6107 decode.loss_cls_ce: 1.9070 decode.loss_mask_ce: 0.8361 decode.loss_mask_dice: 1.5636 decode.d7.loss_cls_ce: 1.8980 decode.d7.loss_mask_ce: 0.8345 decode.d7.loss_mask_dice: 1.5715 2023/09/06 20:26:46 - mmengine - INFO - Iter(train) [46850/60000] base_lr: 2.1917e-05 lr: 2.1917e-05 eta: 1:39:23 time: 0.4641 data_time: 0.0239 memory: 16027 grad_norm: 16.4182 loss: 8.8428 decode.loss_cls_ce: 1.9180 decode.loss_mask_ce: 0.8246 decode.loss_mask_dice: 1.6790 decode.d7.loss_cls_ce: 1.9278 decode.d7.loss_mask_ce: 0.8186 decode.d7.loss_mask_dice: 1.6749 2023/09/06 20:27:09 - mmengine - INFO - Iter(train) [46900/60000] base_lr: 2.1834e-05 lr: 2.1834e-05 eta: 1:39:01 time: 0.4502 data_time: 0.0244 memory: 15730 grad_norm: 18.3351 loss: 7.4436 decode.loss_cls_ce: 1.5427 decode.loss_mask_ce: 0.8291 decode.loss_mask_dice: 1.3370 decode.d7.loss_cls_ce: 1.5467 decode.d7.loss_mask_ce: 0.8361 decode.d7.loss_mask_dice: 1.3520 2023/09/06 20:27:31 - mmengine - INFO - Iter(train) [46950/60000] base_lr: 2.1750e-05 lr: 2.1750e-05 eta: 1:38:38 time: 0.4617 data_time: 0.0235 memory: 15870 grad_norm: 17.4999 loss: 8.4535 decode.loss_cls_ce: 1.8309 decode.loss_mask_ce: 0.8455 decode.loss_mask_dice: 1.5571 decode.d7.loss_cls_ce: 1.8220 decode.d7.loss_mask_ce: 0.8408 decode.d7.loss_mask_dice: 1.5573 2023/09/06 20:27:54 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:27:54 - mmengine - INFO - Iter(train) [47000/60000] base_lr: 2.1667e-05 lr: 2.1667e-05 eta: 1:38:16 time: 0.4533 data_time: 0.0253 memory: 15857 grad_norm: 19.0930 loss: 10.2777 decode.loss_cls_ce: 2.1397 decode.loss_mask_ce: 0.9809 decode.loss_mask_dice: 2.0207 decode.d7.loss_cls_ce: 2.1293 decode.d7.loss_mask_ce: 0.9911 decode.d7.loss_mask_dice: 2.0161 2023/09/06 20:28:17 - mmengine - INFO - Iter(train) [47050/60000] base_lr: 2.1584e-05 lr: 2.1584e-05 eta: 1:37:53 time: 0.4557 data_time: 0.0245 memory: 15949 grad_norm: 19.0442 loss: 9.8566 decode.loss_cls_ce: 2.1428 decode.loss_mask_ce: 1.0017 decode.loss_mask_dice: 1.8011 decode.d7.loss_cls_ce: 2.1010 decode.d7.loss_mask_ce: 1.0084 decode.d7.loss_mask_dice: 1.8016 2023/09/06 20:28:40 - mmengine - INFO - Iter(train) [47100/60000] base_lr: 2.1500e-05 lr: 2.1500e-05 eta: 1:37:30 time: 0.4561 data_time: 0.0225 memory: 15809 grad_norm: 21.4049 loss: 9.7167 decode.loss_cls_ce: 2.0724 decode.loss_mask_ce: 0.9266 decode.loss_mask_dice: 1.8634 decode.d7.loss_cls_ce: 2.0745 decode.d7.loss_mask_ce: 0.9281 decode.d7.loss_mask_dice: 1.8516 2023/09/06 20:29:03 - mmengine - INFO - Iter(train) [47150/60000] base_lr: 2.1417e-05 lr: 2.1417e-05 eta: 1:37:08 time: 0.4600 data_time: 0.0232 memory: 15938 grad_norm: 19.5276 loss: 8.7919 decode.loss_cls_ce: 2.0397 decode.loss_mask_ce: 0.7414 decode.loss_mask_dice: 1.6082 decode.d7.loss_cls_ce: 2.0587 decode.d7.loss_mask_ce: 0.7454 decode.d7.loss_mask_dice: 1.5986 2023/09/06 20:29:26 - mmengine - INFO - Iter(train) [47200/60000] base_lr: 2.1334e-05 lr: 2.1334e-05 eta: 1:36:45 time: 0.4545 data_time: 0.0244 memory: 15913 grad_norm: 17.5640 loss: 9.6973 decode.loss_cls_ce: 2.0351 decode.loss_mask_ce: 0.8967 decode.loss_mask_dice: 1.8972 decode.d7.loss_cls_ce: 2.0705 decode.d7.loss_mask_ce: 0.9009 decode.d7.loss_mask_dice: 1.8969 2023/09/06 20:29:48 - mmengine - INFO - Iter(train) [47250/60000] base_lr: 2.1250e-05 lr: 2.1250e-05 eta: 1:36:22 time: 0.4538 data_time: 0.0242 memory: 15796 grad_norm: 17.8572 loss: 8.7017 decode.loss_cls_ce: 1.8556 decode.loss_mask_ce: 0.8492 decode.loss_mask_dice: 1.6352 decode.d7.loss_cls_ce: 1.8921 decode.d7.loss_mask_ce: 0.8491 decode.d7.loss_mask_dice: 1.6204 2023/09/06 20:30:11 - mmengine - INFO - Iter(train) [47300/60000] base_lr: 2.1167e-05 lr: 2.1167e-05 eta: 1:36:00 time: 0.4529 data_time: 0.0240 memory: 15924 grad_norm: 15.9758 loss: 9.2486 decode.loss_cls_ce: 2.0559 decode.loss_mask_ce: 0.8414 decode.loss_mask_dice: 1.7191 decode.d7.loss_cls_ce: 2.0611 decode.d7.loss_mask_ce: 0.8368 decode.d7.loss_mask_dice: 1.7343 2023/09/06 20:30:34 - mmengine - INFO - Iter(train) [47350/60000] base_lr: 2.1084e-05 lr: 2.1084e-05 eta: 1:35:37 time: 0.4563 data_time: 0.0242 memory: 15742 grad_norm: 16.1025 loss: 9.7233 decode.loss_cls_ce: 2.1804 decode.loss_mask_ce: 0.8701 decode.loss_mask_dice: 1.8065 decode.d7.loss_cls_ce: 2.2055 decode.d7.loss_mask_ce: 0.8715 decode.d7.loss_mask_dice: 1.7894 2023/09/06 20:30:57 - mmengine - INFO - Iter(train) [47400/60000] base_lr: 2.1000e-05 lr: 2.1000e-05 eta: 1:35:14 time: 0.4620 data_time: 0.0252 memory: 15795 grad_norm: 19.3950 loss: 9.1423 decode.loss_cls_ce: 2.0130 decode.loss_mask_ce: 0.8613 decode.loss_mask_dice: 1.6897 decode.d7.loss_cls_ce: 2.0237 decode.d7.loss_mask_ce: 0.8674 decode.d7.loss_mask_dice: 1.6873 2023/09/06 20:31:20 - mmengine - INFO - Iter(train) [47450/60000] base_lr: 2.0917e-05 lr: 2.0917e-05 eta: 1:34:52 time: 0.4577 data_time: 0.0238 memory: 15988 grad_norm: 21.4339 loss: 9.7951 decode.loss_cls_ce: 2.0095 decode.loss_mask_ce: 0.9694 decode.loss_mask_dice: 1.9316 decode.d7.loss_cls_ce: 1.9945 decode.d7.loss_mask_ce: 0.9663 decode.d7.loss_mask_dice: 1.9237 2023/09/06 20:31:43 - mmengine - INFO - Iter(train) [47500/60000] base_lr: 2.0834e-05 lr: 2.0834e-05 eta: 1:34:29 time: 0.4621 data_time: 0.0235 memory: 15811 grad_norm: 18.6376 loss: 9.4012 decode.loss_cls_ce: 2.0696 decode.loss_mask_ce: 0.9524 decode.loss_mask_dice: 1.6584 decode.d7.loss_cls_ce: 2.0966 decode.d7.loss_mask_ce: 0.9584 decode.d7.loss_mask_dice: 1.6658 2023/09/06 20:32:06 - mmengine - INFO - Iter(train) [47550/60000] base_lr: 2.0750e-05 lr: 2.0750e-05 eta: 1:34:07 time: 0.4545 data_time: 0.0253 memory: 15924 grad_norm: 18.9758 loss: 8.9254 decode.loss_cls_ce: 1.9915 decode.loss_mask_ce: 0.8496 decode.loss_mask_dice: 1.6168 decode.d7.loss_cls_ce: 2.0116 decode.d7.loss_mask_ce: 0.8495 decode.d7.loss_mask_dice: 1.6064 2023/09/06 20:32:28 - mmengine - INFO - Iter(train) [47600/60000] base_lr: 2.0667e-05 lr: 2.0667e-05 eta: 1:33:44 time: 0.4519 data_time: 0.0242 memory: 15847 grad_norm: 18.5746 loss: 8.7608 decode.loss_cls_ce: 1.8739 decode.loss_mask_ce: 0.8402 decode.loss_mask_dice: 1.6532 decode.d7.loss_cls_ce: 1.8843 decode.d7.loss_mask_ce: 0.8491 decode.d7.loss_mask_dice: 1.6601 2023/09/06 20:32:51 - mmengine - INFO - Iter(train) [47650/60000] base_lr: 2.0584e-05 lr: 2.0584e-05 eta: 1:33:21 time: 0.4598 data_time: 0.0233 memory: 15949 grad_norm: 17.6044 loss: 9.2928 decode.loss_cls_ce: 1.9212 decode.loss_mask_ce: 0.9088 decode.loss_mask_dice: 1.8292 decode.d7.loss_cls_ce: 1.9070 decode.d7.loss_mask_ce: 0.9104 decode.d7.loss_mask_dice: 1.8162 2023/09/06 20:33:14 - mmengine - INFO - Iter(train) [47700/60000] base_lr: 2.0500e-05 lr: 2.0500e-05 eta: 1:32:59 time: 0.4524 data_time: 0.0243 memory: 15936 grad_norm: 17.2851 loss: 9.1120 decode.loss_cls_ce: 1.9519 decode.loss_mask_ce: 0.8446 decode.loss_mask_dice: 1.7608 decode.d7.loss_cls_ce: 1.9502 decode.d7.loss_mask_ce: 0.8362 decode.d7.loss_mask_dice: 1.7683 2023/09/06 20:33:36 - mmengine - INFO - Iter(train) [47750/60000] base_lr: 2.0417e-05 lr: 2.0417e-05 eta: 1:32:36 time: 0.4527 data_time: 0.0243 memory: 15963 grad_norm: 17.9888 loss: 10.0275 decode.loss_cls_ce: 2.1896 decode.loss_mask_ce: 0.9532 decode.loss_mask_dice: 1.8837 decode.d7.loss_cls_ce: 2.1552 decode.d7.loss_mask_ce: 0.9623 decode.d7.loss_mask_dice: 1.8836 2023/09/06 20:33:59 - mmengine - INFO - Iter(train) [47800/60000] base_lr: 2.0334e-05 lr: 2.0334e-05 eta: 1:32:13 time: 0.4534 data_time: 0.0252 memory: 15760 grad_norm: 19.8439 loss: 8.8994 decode.loss_cls_ce: 1.8801 decode.loss_mask_ce: 0.9307 decode.loss_mask_dice: 1.6334 decode.d7.loss_cls_ce: 1.8869 decode.d7.loss_mask_ce: 0.9353 decode.d7.loss_mask_dice: 1.6331 2023/09/06 20:34:22 - mmengine - INFO - Iter(train) [47850/60000] base_lr: 2.0250e-05 lr: 2.0250e-05 eta: 1:31:50 time: 0.4513 data_time: 0.0242 memory: 15886 grad_norm: 16.4209 loss: 8.5903 decode.loss_cls_ce: 1.9332 decode.loss_mask_ce: 0.7559 decode.loss_mask_dice: 1.6136 decode.d7.loss_cls_ce: 1.9209 decode.d7.loss_mask_ce: 0.7565 decode.d7.loss_mask_dice: 1.6103 2023/09/06 20:34:45 - mmengine - INFO - Iter(train) [47900/60000] base_lr: 2.0167e-05 lr: 2.0167e-05 eta: 1:31:28 time: 0.4577 data_time: 0.0238 memory: 15717 grad_norm: 16.7115 loss: 8.7144 decode.loss_cls_ce: 1.8309 decode.loss_mask_ce: 0.8733 decode.loss_mask_dice: 1.6313 decode.d7.loss_cls_ce: 1.8638 decode.d7.loss_mask_ce: 0.8754 decode.d7.loss_mask_dice: 1.6398 2023/09/06 20:35:08 - mmengine - INFO - Iter(train) [47950/60000] base_lr: 2.0084e-05 lr: 2.0084e-05 eta: 1:31:05 time: 0.4612 data_time: 0.0233 memory: 15896 grad_norm: 16.9610 loss: 7.9340 decode.loss_cls_ce: 1.7310 decode.loss_mask_ce: 0.7849 decode.loss_mask_dice: 1.4377 decode.d7.loss_cls_ce: 1.7454 decode.d7.loss_mask_ce: 0.7902 decode.d7.loss_mask_dice: 1.4448 2023/09/06 20:35:31 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:35:31 - mmengine - INFO - Iter(train) [48000/60000] base_lr: 2.0000e-05 lr: 2.0000e-05 eta: 1:30:43 time: 0.4598 data_time: 0.0231 memory: 15974 grad_norm: 19.3223 loss: 9.6390 decode.loss_cls_ce: 2.0692 decode.loss_mask_ce: 0.9041 decode.loss_mask_dice: 1.8298 decode.d7.loss_cls_ce: 2.0629 decode.d7.loss_mask_ce: 0.9106 decode.d7.loss_mask_dice: 1.8623 2023/09/06 20:35:53 - mmengine - INFO - Iter(train) [48050/60000] base_lr: 1.9917e-05 lr: 1.9917e-05 eta: 1:30:20 time: 0.4529 data_time: 0.0255 memory: 15871 grad_norm: 17.5176 loss: 8.8764 decode.loss_cls_ce: 1.8936 decode.loss_mask_ce: 0.8764 decode.loss_mask_dice: 1.6619 decode.d7.loss_cls_ce: 1.8943 decode.d7.loss_mask_ce: 0.8793 decode.d7.loss_mask_dice: 1.6709 2023/09/06 20:36:16 - mmengine - INFO - Iter(train) [48100/60000] base_lr: 1.9834e-05 lr: 1.9834e-05 eta: 1:29:57 time: 0.4544 data_time: 0.0252 memory: 15933 grad_norm: 18.4429 loss: 9.1498 decode.loss_cls_ce: 2.0188 decode.loss_mask_ce: 0.8733 decode.loss_mask_dice: 1.6839 decode.d7.loss_cls_ce: 2.0138 decode.d7.loss_mask_ce: 0.8755 decode.d7.loss_mask_dice: 1.6846 2023/09/06 20:36:39 - mmengine - INFO - Iter(train) [48150/60000] base_lr: 1.9750e-05 lr: 1.9750e-05 eta: 1:29:35 time: 0.4600 data_time: 0.0242 memory: 15910 grad_norm: 20.2490 loss: 8.7984 decode.loss_cls_ce: 1.9754 decode.loss_mask_ce: 0.8100 decode.loss_mask_dice: 1.6096 decode.d7.loss_cls_ce: 2.0012 decode.d7.loss_mask_ce: 0.8124 decode.d7.loss_mask_dice: 1.5898 2023/09/06 20:37:02 - mmengine - INFO - Iter(train) [48200/60000] base_lr: 1.9667e-05 lr: 1.9667e-05 eta: 1:29:12 time: 0.4533 data_time: 0.0239 memory: 15908 grad_norm: 17.1546 loss: 9.3294 decode.loss_cls_ce: 1.8540 decode.loss_mask_ce: 0.9857 decode.loss_mask_dice: 1.8218 decode.d7.loss_cls_ce: 1.8483 decode.d7.loss_mask_ce: 0.9898 decode.d7.loss_mask_dice: 1.8297 2023/09/06 20:37:24 - mmengine - INFO - Iter(train) [48250/60000] base_lr: 1.9584e-05 lr: 1.9584e-05 eta: 1:28:49 time: 0.4534 data_time: 0.0242 memory: 16118 grad_norm: 17.9602 loss: 8.6272 decode.loss_cls_ce: 1.9491 decode.loss_mask_ce: 0.7863 decode.loss_mask_dice: 1.5599 decode.d7.loss_cls_ce: 1.9641 decode.d7.loss_mask_ce: 0.7869 decode.d7.loss_mask_dice: 1.5809 2023/09/06 20:37:47 - mmengine - INFO - Iter(train) [48300/60000] base_lr: 1.9500e-05 lr: 1.9500e-05 eta: 1:28:27 time: 0.4563 data_time: 0.0242 memory: 15819 grad_norm: 17.3991 loss: 9.1131 decode.loss_cls_ce: 2.0343 decode.loss_mask_ce: 0.8431 decode.loss_mask_dice: 1.6700 decode.d7.loss_cls_ce: 2.0381 decode.d7.loss_mask_ce: 0.8497 decode.d7.loss_mask_dice: 1.6779 2023/09/06 20:38:10 - mmengine - INFO - Iter(train) [48350/60000] base_lr: 1.9417e-05 lr: 1.9417e-05 eta: 1:28:04 time: 0.4634 data_time: 0.0244 memory: 16026 grad_norm: 16.6323 loss: 8.8428 decode.loss_cls_ce: 1.9930 decode.loss_mask_ce: 0.8305 decode.loss_mask_dice: 1.5812 decode.d7.loss_cls_ce: 1.9976 decode.d7.loss_mask_ce: 0.8514 decode.d7.loss_mask_dice: 1.5891 2023/09/06 20:38:33 - mmengine - INFO - Iter(train) [48400/60000] base_lr: 1.9334e-05 lr: 1.9334e-05 eta: 1:27:41 time: 0.4589 data_time: 0.0240 memory: 15843 grad_norm: nan loss: 8.4879 decode.loss_cls_ce: 1.8339 decode.loss_mask_ce: 0.8101 decode.loss_mask_dice: 1.5728 decode.d7.loss_cls_ce: 1.8558 decode.d7.loss_mask_ce: 0.8199 decode.d7.loss_mask_dice: 1.5954 2023/09/06 20:38:56 - mmengine - INFO - Iter(train) [48450/60000] base_lr: 1.9250e-05 lr: 1.9250e-05 eta: 1:27:19 time: 0.4641 data_time: 0.0241 memory: 15872 grad_norm: 18.2276 loss: 9.1886 decode.loss_cls_ce: 1.9970 decode.loss_mask_ce: 0.8780 decode.loss_mask_dice: 1.7121 decode.d7.loss_cls_ce: 2.0334 decode.d7.loss_mask_ce: 0.8717 decode.d7.loss_mask_dice: 1.6963 2023/09/06 20:39:19 - mmengine - INFO - Iter(train) [48500/60000] base_lr: 1.9167e-05 lr: 1.9167e-05 eta: 1:26:56 time: 0.4551 data_time: 0.0237 memory: 15934 grad_norm: 17.3945 loss: 9.3135 decode.loss_cls_ce: 2.0757 decode.loss_mask_ce: 0.8667 decode.loss_mask_dice: 1.7165 decode.d7.loss_cls_ce: 2.0788 decode.d7.loss_mask_ce: 0.8660 decode.d7.loss_mask_dice: 1.7099 2023/09/06 20:39:42 - mmengine - INFO - Iter(train) [48550/60000] base_lr: 1.9084e-05 lr: 1.9084e-05 eta: 1:26:34 time: 0.4609 data_time: 0.0245 memory: 15894 grad_norm: 18.4914 loss: 8.5860 decode.loss_cls_ce: 1.9704 decode.loss_mask_ce: 0.8074 decode.loss_mask_dice: 1.5146 decode.d7.loss_cls_ce: 1.9739 decode.d7.loss_mask_ce: 0.8078 decode.d7.loss_mask_dice: 1.5119 2023/09/06 20:40:05 - mmengine - INFO - Iter(train) [48600/60000] base_lr: 1.9000e-05 lr: 1.9000e-05 eta: 1:26:11 time: 0.4558 data_time: 0.0270 memory: 15844 grad_norm: 17.6449 loss: 8.6523 decode.loss_cls_ce: 1.7651 decode.loss_mask_ce: 0.8635 decode.loss_mask_dice: 1.6826 decode.d7.loss_cls_ce: 1.7972 decode.d7.loss_mask_ce: 0.8627 decode.d7.loss_mask_dice: 1.6812 2023/09/06 20:40:28 - mmengine - INFO - Iter(train) [48650/60000] base_lr: 1.8917e-05 lr: 1.8917e-05 eta: 1:25:48 time: 0.4565 data_time: 0.0236 memory: 15809 grad_norm: 17.4007 loss: 8.7740 decode.loss_cls_ce: 1.8149 decode.loss_mask_ce: 0.9349 decode.loss_mask_dice: 1.6261 decode.d7.loss_cls_ce: 1.8383 decode.d7.loss_mask_ce: 0.9349 decode.d7.loss_mask_dice: 1.6249 2023/09/06 20:40:50 - mmengine - INFO - Iter(train) [48700/60000] base_lr: 1.8834e-05 lr: 1.8834e-05 eta: 1:25:26 time: 0.4545 data_time: 0.0250 memory: 15783 grad_norm: 18.1639 loss: 9.5437 decode.loss_cls_ce: 2.0868 decode.loss_mask_ce: 0.9213 decode.loss_mask_dice: 1.7557 decode.d7.loss_cls_ce: 2.0787 decode.d7.loss_mask_ce: 0.9332 decode.d7.loss_mask_dice: 1.7680 2023/09/06 20:41:14 - mmengine - INFO - Iter(train) [48750/60000] base_lr: 1.8750e-05 lr: 1.8750e-05 eta: 1:25:03 time: 0.4629 data_time: 0.0243 memory: 15720 grad_norm: 19.2118 loss: 9.9162 decode.loss_cls_ce: 2.1112 decode.loss_mask_ce: 0.9810 decode.loss_mask_dice: 1.8457 decode.d7.loss_cls_ce: 2.1515 decode.d7.loss_mask_ce: 0.9787 decode.d7.loss_mask_dice: 1.8480 2023/09/06 20:41:36 - mmengine - INFO - Iter(train) [48800/60000] base_lr: 1.8667e-05 lr: 1.8667e-05 eta: 1:24:40 time: 0.4543 data_time: 0.0243 memory: 15747 grad_norm: 16.1047 loss: 9.0413 decode.loss_cls_ce: 2.1015 decode.loss_mask_ce: 0.8532 decode.loss_mask_dice: 1.5668 decode.d7.loss_cls_ce: 2.1043 decode.d7.loss_mask_ce: 0.8499 decode.d7.loss_mask_dice: 1.5657 2023/09/06 20:41:59 - mmengine - INFO - Iter(train) [48850/60000] base_lr: 1.8584e-05 lr: 1.8584e-05 eta: 1:24:18 time: 0.4627 data_time: 0.0242 memory: 15871 grad_norm: 18.2914 loss: 8.8492 decode.loss_cls_ce: 1.9332 decode.loss_mask_ce: 0.8856 decode.loss_mask_dice: 1.6085 decode.d7.loss_cls_ce: 1.9192 decode.d7.loss_mask_ce: 0.8866 decode.d7.loss_mask_dice: 1.6161 2023/09/06 20:42:22 - mmengine - INFO - Iter(train) [48900/60000] base_lr: 1.8500e-05 lr: 1.8500e-05 eta: 1:23:55 time: 0.4626 data_time: 0.0256 memory: 16015 grad_norm: 16.6585 loss: 10.3070 decode.loss_cls_ce: 2.2501 decode.loss_mask_ce: 0.9570 decode.loss_mask_dice: 1.9401 decode.d7.loss_cls_ce: 2.2608 decode.d7.loss_mask_ce: 0.9604 decode.d7.loss_mask_dice: 1.9387 2023/09/06 20:42:45 - mmengine - INFO - Iter(train) [48950/60000] base_lr: 1.8417e-05 lr: 1.8417e-05 eta: 1:23:32 time: 0.4622 data_time: 0.0233 memory: 15948 grad_norm: 24.3941 loss: 9.2398 decode.loss_cls_ce: 1.9594 decode.loss_mask_ce: 0.8729 decode.loss_mask_dice: 1.7704 decode.d7.loss_cls_ce: 1.9875 decode.d7.loss_mask_ce: 0.8733 decode.d7.loss_mask_dice: 1.7764 2023/09/06 20:43:08 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:43:08 - mmengine - INFO - Iter(train) [49000/60000] base_lr: 1.8334e-05 lr: 1.8334e-05 eta: 1:23:10 time: 0.4590 data_time: 0.0236 memory: 15808 grad_norm: 20.5727 loss: 8.2021 decode.loss_cls_ce: 1.8443 decode.loss_mask_ce: 0.8082 decode.loss_mask_dice: 1.4483 decode.d7.loss_cls_ce: 1.8401 decode.d7.loss_mask_ce: 0.8113 decode.d7.loss_mask_dice: 1.4498 2023/09/06 20:43:31 - mmengine - INFO - Iter(train) [49050/60000] base_lr: 1.8250e-05 lr: 1.8250e-05 eta: 1:22:47 time: 0.4633 data_time: 0.0236 memory: 15861 grad_norm: 18.8966 loss: 9.6268 decode.loss_cls_ce: 2.1337 decode.loss_mask_ce: 0.9262 decode.loss_mask_dice: 1.7561 decode.d7.loss_cls_ce: 2.1003 decode.d7.loss_mask_ce: 0.9365 decode.d7.loss_mask_dice: 1.7739 2023/09/06 20:43:54 - mmengine - INFO - Iter(train) [49100/60000] base_lr: 1.8167e-05 lr: 1.8167e-05 eta: 1:22:24 time: 0.4499 data_time: 0.0243 memory: 15910 grad_norm: nan loss: 8.9654 decode.loss_cls_ce: 1.9344 decode.loss_mask_ce: 0.8688 decode.loss_mask_dice: 1.6765 decode.d7.loss_cls_ce: 1.9425 decode.d7.loss_mask_ce: 0.8666 decode.d7.loss_mask_dice: 1.6765 2023/09/06 20:44:17 - mmengine - INFO - Iter(train) [49150/60000] base_lr: 1.8084e-05 lr: 1.8084e-05 eta: 1:22:02 time: 0.4616 data_time: 0.0234 memory: 15734 grad_norm: 17.7427 loss: 8.8798 decode.loss_cls_ce: 1.9586 decode.loss_mask_ce: 0.8240 decode.loss_mask_dice: 1.6462 decode.d7.loss_cls_ce: 1.9730 decode.d7.loss_mask_ce: 0.8241 decode.d7.loss_mask_dice: 1.6539 2023/09/06 20:44:40 - mmengine - INFO - Iter(train) [49200/60000] base_lr: 1.8000e-05 lr: 1.8000e-05 eta: 1:21:39 time: 0.4629 data_time: 0.0240 memory: 15846 grad_norm: 18.8947 loss: 9.3067 decode.loss_cls_ce: 1.9296 decode.loss_mask_ce: 0.9324 decode.loss_mask_dice: 1.7752 decode.d7.loss_cls_ce: 1.9596 decode.d7.loss_mask_ce: 0.9294 decode.d7.loss_mask_dice: 1.7805 2023/09/06 20:45:03 - mmengine - INFO - Iter(train) [49250/60000] base_lr: 1.7917e-05 lr: 1.7917e-05 eta: 1:21:17 time: 0.4572 data_time: 0.0238 memory: 16093 grad_norm: 16.8077 loss: 8.4368 decode.loss_cls_ce: 1.9192 decode.loss_mask_ce: 0.7995 decode.loss_mask_dice: 1.5099 decode.d7.loss_cls_ce: 1.8804 decode.d7.loss_mask_ce: 0.8086 decode.d7.loss_mask_dice: 1.5192 2023/09/06 20:45:26 - mmengine - INFO - Iter(train) [49300/60000] base_lr: 1.7834e-05 lr: 1.7834e-05 eta: 1:20:54 time: 0.4582 data_time: 0.0248 memory: 16195 grad_norm: 21.3168 loss: 8.5068 decode.loss_cls_ce: 1.9180 decode.loss_mask_ce: 0.7718 decode.loss_mask_dice: 1.5680 decode.d7.loss_cls_ce: 1.9167 decode.d7.loss_mask_ce: 0.7679 decode.d7.loss_mask_dice: 1.5644 2023/09/06 20:45:49 - mmengine - INFO - Iter(train) [49350/60000] base_lr: 1.7750e-05 lr: 1.7750e-05 eta: 1:20:31 time: 0.4515 data_time: 0.0239 memory: 16000 grad_norm: 20.4985 loss: 9.4944 decode.loss_cls_ce: 1.9917 decode.loss_mask_ce: 0.8438 decode.loss_mask_dice: 1.9029 decode.d7.loss_cls_ce: 1.9879 decode.d7.loss_mask_ce: 0.8500 decode.d7.loss_mask_dice: 1.9180 2023/09/06 20:46:11 - mmengine - INFO - Iter(train) [49400/60000] base_lr: 1.7667e-05 lr: 1.7667e-05 eta: 1:20:09 time: 0.4518 data_time: 0.0237 memory: 15759 grad_norm: 17.9400 loss: 9.7532 decode.loss_cls_ce: 2.1202 decode.loss_mask_ce: 0.9263 decode.loss_mask_dice: 1.8298 decode.d7.loss_cls_ce: 2.1033 decode.d7.loss_mask_ce: 0.9306 decode.d7.loss_mask_dice: 1.8431 2023/09/06 20:46:34 - mmengine - INFO - Iter(train) [49450/60000] base_lr: 1.7584e-05 lr: 1.7584e-05 eta: 1:19:46 time: 0.4541 data_time: 0.0252 memory: 15912 grad_norm: 20.6480 loss: 7.8485 decode.loss_cls_ce: 1.7787 decode.loss_mask_ce: 0.7550 decode.loss_mask_dice: 1.3718 decode.d7.loss_cls_ce: 1.7956 decode.d7.loss_mask_ce: 0.7654 decode.d7.loss_mask_dice: 1.3820 2023/09/06 20:46:56 - mmengine - INFO - Iter(train) [49500/60000] base_lr: 1.7500e-05 lr: 1.7500e-05 eta: 1:19:23 time: 0.4510 data_time: 0.0241 memory: 15818 grad_norm: 17.4381 loss: 9.4394 decode.loss_cls_ce: 1.9990 decode.loss_mask_ce: 0.9314 decode.loss_mask_dice: 1.7863 decode.d7.loss_cls_ce: 1.9939 decode.d7.loss_mask_ce: 0.9373 decode.d7.loss_mask_dice: 1.7915 2023/09/06 20:47:19 - mmengine - INFO - Iter(train) [49550/60000] base_lr: 1.7417e-05 lr: 1.7417e-05 eta: 1:19:01 time: 0.4515 data_time: 0.0245 memory: 15963 grad_norm: 18.0118 loss: 8.9164 decode.loss_cls_ce: 1.9425 decode.loss_mask_ce: 0.9105 decode.loss_mask_dice: 1.6056 decode.d7.loss_cls_ce: 1.9389 decode.d7.loss_mask_ce: 0.9039 decode.d7.loss_mask_dice: 1.6151 2023/09/06 20:47:42 - mmengine - INFO - Iter(train) [49600/60000] base_lr: 1.7334e-05 lr: 1.7334e-05 eta: 1:18:38 time: 0.4626 data_time: 0.0240 memory: 15824 grad_norm: 17.2671 loss: 8.9309 decode.loss_cls_ce: 1.9543 decode.loss_mask_ce: 0.8629 decode.loss_mask_dice: 1.6527 decode.d7.loss_cls_ce: 1.9423 decode.d7.loss_mask_ce: 0.8650 decode.d7.loss_mask_dice: 1.6536 2023/09/06 20:48:05 - mmengine - INFO - Iter(train) [49650/60000] base_lr: 1.7250e-05 lr: 1.7250e-05 eta: 1:18:15 time: 0.4507 data_time: 0.0239 memory: 15975 grad_norm: 17.9690 loss: 8.8340 decode.loss_cls_ce: 2.0696 decode.loss_mask_ce: 0.8133 decode.loss_mask_dice: 1.5358 decode.d7.loss_cls_ce: 2.0745 decode.d7.loss_mask_ce: 0.8131 decode.d7.loss_mask_dice: 1.5277 2023/09/06 20:48:28 - mmengine - INFO - Iter(train) [49700/60000] base_lr: 1.7167e-05 lr: 1.7167e-05 eta: 1:17:53 time: 0.4609 data_time: 0.0233 memory: 15771 grad_norm: 17.4851 loss: 9.2281 decode.loss_cls_ce: 1.9779 decode.loss_mask_ce: 0.8905 decode.loss_mask_dice: 1.7355 decode.d7.loss_cls_ce: 1.9988 decode.d7.loss_mask_ce: 0.8946 decode.d7.loss_mask_dice: 1.7309 2023/09/06 20:48:51 - mmengine - INFO - Iter(train) [49750/60000] base_lr: 1.7084e-05 lr: 1.7084e-05 eta: 1:17:30 time: 0.4618 data_time: 0.0238 memory: 15993 grad_norm: 19.7240 loss: 8.6615 decode.loss_cls_ce: 1.8275 decode.loss_mask_ce: 0.8704 decode.loss_mask_dice: 1.6226 decode.d7.loss_cls_ce: 1.8385 decode.d7.loss_mask_ce: 0.8799 decode.d7.loss_mask_dice: 1.6227 2023/09/06 20:49:14 - mmengine - INFO - Iter(train) [49800/60000] base_lr: 1.7000e-05 lr: 1.7000e-05 eta: 1:17:07 time: 0.4544 data_time: 0.0250 memory: 15921 grad_norm: 16.3137 loss: 8.4527 decode.loss_cls_ce: 1.8684 decode.loss_mask_ce: 0.7997 decode.loss_mask_dice: 1.5537 decode.d7.loss_cls_ce: 1.8383 decode.d7.loss_mask_ce: 0.8132 decode.d7.loss_mask_dice: 1.5794 2023/09/06 20:49:37 - mmengine - INFO - Iter(train) [49850/60000] base_lr: 1.6917e-05 lr: 1.6917e-05 eta: 1:16:45 time: 0.4530 data_time: 0.0243 memory: 15834 grad_norm: 17.7160 loss: 8.4813 decode.loss_cls_ce: 1.7981 decode.loss_mask_ce: 0.8520 decode.loss_mask_dice: 1.5983 decode.d7.loss_cls_ce: 1.7943 decode.d7.loss_mask_ce: 0.8464 decode.d7.loss_mask_dice: 1.5922 2023/09/06 20:49:59 - mmengine - INFO - Iter(train) [49900/60000] base_lr: 1.6834e-05 lr: 1.6834e-05 eta: 1:16:22 time: 0.4525 data_time: 0.0237 memory: 15693 grad_norm: 16.3923 loss: 8.9682 decode.loss_cls_ce: 1.9735 decode.loss_mask_ce: 0.8192 decode.loss_mask_dice: 1.6805 decode.d7.loss_cls_ce: 1.9958 decode.d7.loss_mask_ce: 0.8146 decode.d7.loss_mask_dice: 1.6846 2023/09/06 20:50:22 - mmengine - INFO - Iter(train) [49950/60000] base_lr: 1.6750e-05 lr: 1.6750e-05 eta: 1:15:59 time: 0.4597 data_time: 0.0245 memory: 15774 grad_norm: 19.1668 loss: 8.0618 decode.loss_cls_ce: 1.8425 decode.loss_mask_ce: 0.7976 decode.loss_mask_dice: 1.3823 decode.d7.loss_cls_ce: 1.8666 decode.d7.loss_mask_ce: 0.7991 decode.d7.loss_mask_dice: 1.3737 2023/09/06 20:50:45 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:50:45 - mmengine - INFO - Iter(train) [50000/60000] base_lr: 1.6667e-05 lr: 1.6667e-05 eta: 1:15:37 time: 0.4551 data_time: 0.0242 memory: 15820 grad_norm: 17.5373 loss: 9.1897 decode.loss_cls_ce: 1.9261 decode.loss_mask_ce: 0.8843 decode.loss_mask_dice: 1.7958 decode.d7.loss_cls_ce: 1.9188 decode.d7.loss_mask_ce: 0.8765 decode.d7.loss_mask_dice: 1.7881 2023/09/06 20:50:45 - mmengine - INFO - Saving checkpoint at 50000 iterations 2023/09/06 20:51:11 - mmengine - INFO - Iter(train) [50050/60000] base_lr: 1.6584e-05 lr: 1.6584e-05 eta: 1:15:15 time: 0.4519 data_time: 0.0241 memory: 15977 grad_norm: 19.6421 loss: 8.8436 decode.loss_cls_ce: 1.8953 decode.loss_mask_ce: 0.8329 decode.loss_mask_dice: 1.6827 decode.d7.loss_cls_ce: 1.9040 decode.d7.loss_mask_ce: 0.8335 decode.d7.loss_mask_dice: 1.6952 2023/09/06 20:51:34 - mmengine - INFO - Iter(train) [50100/60000] base_lr: 1.6500e-05 lr: 1.6500e-05 eta: 1:14:52 time: 0.4502 data_time: 0.0242 memory: 15858 grad_norm: 16.5630 loss: 7.7918 decode.loss_cls_ce: 1.7346 decode.loss_mask_ce: 0.8114 decode.loss_mask_dice: 1.3707 decode.d7.loss_cls_ce: 1.6987 decode.d7.loss_mask_ce: 0.8123 decode.d7.loss_mask_dice: 1.3641 2023/09/06 20:51:56 - mmengine - INFO - Iter(train) [50150/60000] base_lr: 1.6417e-05 lr: 1.6417e-05 eta: 1:14:29 time: 0.4527 data_time: 0.0250 memory: 15936 grad_norm: 16.5261 loss: 8.8831 decode.loss_cls_ce: 1.8652 decode.loss_mask_ce: 0.8732 decode.loss_mask_dice: 1.6983 decode.d7.loss_cls_ce: 1.8803 decode.d7.loss_mask_ce: 0.8684 decode.d7.loss_mask_dice: 1.6977 2023/09/06 20:52:19 - mmengine - INFO - Iter(train) [50200/60000] base_lr: 1.6334e-05 lr: 1.6334e-05 eta: 1:14:07 time: 0.4621 data_time: 0.0240 memory: 15911 grad_norm: 18.2792 loss: 10.1952 decode.loss_cls_ce: 2.2122 decode.loss_mask_ce: 0.9851 decode.loss_mask_dice: 1.8836 decode.d7.loss_cls_ce: 2.2398 decode.d7.loss_mask_ce: 0.9832 decode.d7.loss_mask_dice: 1.8914 2023/09/06 20:52:42 - mmengine - INFO - Iter(train) [50250/60000] base_lr: 1.6250e-05 lr: 1.6250e-05 eta: 1:13:44 time: 0.4529 data_time: 0.0248 memory: 15833 grad_norm: 20.4173 loss: 8.5683 decode.loss_cls_ce: 1.9183 decode.loss_mask_ce: 0.8265 decode.loss_mask_dice: 1.5364 decode.d7.loss_cls_ce: 1.9071 decode.d7.loss_mask_ce: 0.8210 decode.d7.loss_mask_dice: 1.5589 2023/09/06 20:53:05 - mmengine - INFO - Iter(train) [50300/60000] base_lr: 1.6167e-05 lr: 1.6167e-05 eta: 1:13:21 time: 0.4619 data_time: 0.0237 memory: 15905 grad_norm: 17.7813 loss: 9.0700 decode.loss_cls_ce: 1.9967 decode.loss_mask_ce: 0.7999 decode.loss_mask_dice: 1.7376 decode.d7.loss_cls_ce: 1.9937 decode.d7.loss_mask_ce: 0.8048 decode.d7.loss_mask_dice: 1.7372 2023/09/06 20:53:28 - mmengine - INFO - Iter(train) [50350/60000] base_lr: 1.6084e-05 lr: 1.6084e-05 eta: 1:12:59 time: 0.4625 data_time: 0.0250 memory: 15812 grad_norm: 17.6272 loss: 8.5926 decode.loss_cls_ce: 1.9254 decode.loss_mask_ce: 0.8382 decode.loss_mask_dice: 1.5352 decode.d7.loss_cls_ce: 1.9307 decode.d7.loss_mask_ce: 0.8356 decode.d7.loss_mask_dice: 1.5276 2023/09/06 20:53:51 - mmengine - INFO - Iter(train) [50400/60000] base_lr: 1.6000e-05 lr: 1.6000e-05 eta: 1:12:36 time: 0.4642 data_time: 0.0243 memory: 16004 grad_norm: 19.2024 loss: 8.9049 decode.loss_cls_ce: 1.7979 decode.loss_mask_ce: 0.8665 decode.loss_mask_dice: 1.7693 decode.d7.loss_cls_ce: 1.8447 decode.d7.loss_mask_ce: 0.8601 decode.d7.loss_mask_dice: 1.7664 2023/09/06 20:54:14 - mmengine - INFO - Iter(train) [50450/60000] base_lr: 1.5917e-05 lr: 1.5917e-05 eta: 1:12:13 time: 0.4575 data_time: 0.0236 memory: 15870 grad_norm: 16.3427 loss: 8.4199 decode.loss_cls_ce: 1.8279 decode.loss_mask_ce: 0.8325 decode.loss_mask_dice: 1.5607 decode.d7.loss_cls_ce: 1.7987 decode.d7.loss_mask_ce: 0.8335 decode.d7.loss_mask_dice: 1.5666 2023/09/06 20:54:37 - mmengine - INFO - Iter(train) [50500/60000] base_lr: 1.5834e-05 lr: 1.5834e-05 eta: 1:11:51 time: 0.4588 data_time: 0.0243 memory: 15806 grad_norm: 16.2888 loss: 8.8950 decode.loss_cls_ce: 1.9331 decode.loss_mask_ce: 0.8804 decode.loss_mask_dice: 1.6295 decode.d7.loss_cls_ce: 1.9355 decode.d7.loss_mask_ce: 0.8859 decode.d7.loss_mask_dice: 1.6306 2023/09/06 20:55:00 - mmengine - INFO - Iter(train) [50550/60000] base_lr: 1.5750e-05 lr: 1.5750e-05 eta: 1:11:28 time: 0.4576 data_time: 0.0239 memory: 16001 grad_norm: 17.3399 loss: 9.1725 decode.loss_cls_ce: 2.0328 decode.loss_mask_ce: 0.8471 decode.loss_mask_dice: 1.7022 decode.d7.loss_cls_ce: 2.0284 decode.d7.loss_mask_ce: 0.8547 decode.d7.loss_mask_dice: 1.7073 2023/09/06 20:55:22 - mmengine - INFO - Iter(train) [50600/60000] base_lr: 1.5667e-05 lr: 1.5667e-05 eta: 1:11:05 time: 0.4535 data_time: 0.0247 memory: 15793 grad_norm: 17.9500 loss: 8.7558 decode.loss_cls_ce: 1.9619 decode.loss_mask_ce: 0.8185 decode.loss_mask_dice: 1.6013 decode.d7.loss_cls_ce: 1.9636 decode.d7.loss_mask_ce: 0.8189 decode.d7.loss_mask_dice: 1.5915 2023/09/06 20:55:45 - mmengine - INFO - Iter(train) [50650/60000] base_lr: 1.5584e-05 lr: 1.5584e-05 eta: 1:10:43 time: 0.4550 data_time: 0.0253 memory: 15782 grad_norm: 18.4876 loss: 9.2536 decode.loss_cls_ce: 2.0824 decode.loss_mask_ce: 0.8464 decode.loss_mask_dice: 1.6892 decode.d7.loss_cls_ce: 2.0938 decode.d7.loss_mask_ce: 0.8515 decode.d7.loss_mask_dice: 1.6904 2023/09/06 20:56:08 - mmengine - INFO - Iter(train) [50700/60000] base_lr: 1.5500e-05 lr: 1.5500e-05 eta: 1:10:20 time: 0.4519 data_time: 0.0241 memory: 15756 grad_norm: 18.5747 loss: 9.9386 decode.loss_cls_ce: 2.1139 decode.loss_mask_ce: 1.0034 decode.loss_mask_dice: 1.8428 decode.d7.loss_cls_ce: 2.1298 decode.d7.loss_mask_ce: 1.0078 decode.d7.loss_mask_dice: 1.8409 2023/09/06 20:56:31 - mmengine - INFO - Iter(train) [50750/60000] base_lr: 1.5417e-05 lr: 1.5417e-05 eta: 1:09:57 time: 0.4615 data_time: 0.0240 memory: 15848 grad_norm: 22.1352 loss: 7.9983 decode.loss_cls_ce: 1.7693 decode.loss_mask_ce: 0.7916 decode.loss_mask_dice: 1.4152 decode.d7.loss_cls_ce: 1.8018 decode.d7.loss_mask_ce: 0.7945 decode.d7.loss_mask_dice: 1.4259 2023/09/06 20:56:54 - mmengine - INFO - Iter(train) [50800/60000] base_lr: 1.5334e-05 lr: 1.5334e-05 eta: 1:09:35 time: 0.4560 data_time: 0.0232 memory: 15822 grad_norm: 18.5486 loss: 8.3081 decode.loss_cls_ce: 1.8062 decode.loss_mask_ce: 0.8355 decode.loss_mask_dice: 1.4981 decode.d7.loss_cls_ce: 1.8211 decode.d7.loss_mask_ce: 0.8403 decode.d7.loss_mask_dice: 1.5069 2023/09/06 20:57:17 - mmengine - INFO - Iter(train) [50850/60000] base_lr: 1.5250e-05 lr: 1.5250e-05 eta: 1:09:12 time: 0.4511 data_time: 0.0242 memory: 15717 grad_norm: 18.1760 loss: 9.6234 decode.loss_cls_ce: 2.0506 decode.loss_mask_ce: 0.9658 decode.loss_mask_dice: 1.7796 decode.d7.loss_cls_ce: 2.0721 decode.d7.loss_mask_ce: 0.9599 decode.d7.loss_mask_dice: 1.7955 2023/09/06 20:57:40 - mmengine - INFO - Iter(train) [50900/60000] base_lr: 1.5167e-05 lr: 1.5167e-05 eta: 1:08:49 time: 0.4526 data_time: 0.0246 memory: 15947 grad_norm: 17.7415 loss: 8.7790 decode.loss_cls_ce: 1.9494 decode.loss_mask_ce: 0.8133 decode.loss_mask_dice: 1.6092 decode.d7.loss_cls_ce: 1.9860 decode.d7.loss_mask_ce: 0.8127 decode.d7.loss_mask_dice: 1.6083 2023/09/06 20:58:02 - mmengine - INFO - Iter(train) [50950/60000] base_lr: 1.5084e-05 lr: 1.5084e-05 eta: 1:08:27 time: 0.4529 data_time: 0.0238 memory: 15806 grad_norm: 17.8265 loss: 9.3578 decode.loss_cls_ce: 2.0079 decode.loss_mask_ce: 0.8946 decode.loss_mask_dice: 1.7549 decode.d7.loss_cls_ce: 2.0302 decode.d7.loss_mask_ce: 0.9007 decode.d7.loss_mask_dice: 1.7695 2023/09/06 20:58:25 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 20:58:25 - mmengine - INFO - Iter(train) [51000/60000] base_lr: 1.5000e-05 lr: 1.5000e-05 eta: 1:08:04 time: 0.4527 data_time: 0.0246 memory: 15858 grad_norm: 16.2623 loss: 10.0604 decode.loss_cls_ce: 2.2083 decode.loss_mask_ce: 0.9492 decode.loss_mask_dice: 1.8907 decode.d7.loss_cls_ce: 2.1609 decode.d7.loss_mask_ce: 0.9502 decode.d7.loss_mask_dice: 1.9012 2023/09/06 20:58:48 - mmengine - INFO - Iter(train) [51050/60000] base_lr: 1.4917e-05 lr: 1.4917e-05 eta: 1:07:41 time: 0.4525 data_time: 0.0246 memory: 15897 grad_norm: 17.8549 loss: 8.4638 decode.loss_cls_ce: 1.8107 decode.loss_mask_ce: 0.8439 decode.loss_mask_dice: 1.5835 decode.d7.loss_cls_ce: 1.8081 decode.d7.loss_mask_ce: 0.8362 decode.d7.loss_mask_dice: 1.5813 2023/09/06 20:59:10 - mmengine - INFO - Iter(train) [51100/60000] base_lr: 1.4834e-05 lr: 1.4834e-05 eta: 1:07:19 time: 0.4526 data_time: 0.0241 memory: 15844 grad_norm: 18.4319 loss: 9.0628 decode.loss_cls_ce: 1.8829 decode.loss_mask_ce: 0.8971 decode.loss_mask_dice: 1.7543 decode.d7.loss_cls_ce: 1.8889 decode.d7.loss_mask_ce: 0.8852 decode.d7.loss_mask_dice: 1.7545 2023/09/06 20:59:33 - mmengine - INFO - Iter(train) [51150/60000] base_lr: 1.4750e-05 lr: 1.4750e-05 eta: 1:06:56 time: 0.4608 data_time: 0.0237 memory: 15784 grad_norm: 20.4235 loss: 8.2544 decode.loss_cls_ce: 1.8940 decode.loss_mask_ce: 0.7784 decode.loss_mask_dice: 1.4359 decode.d7.loss_cls_ce: 1.9263 decode.d7.loss_mask_ce: 0.7763 decode.d7.loss_mask_dice: 1.4437 2023/09/06 20:59:56 - mmengine - INFO - Iter(train) [51200/60000] base_lr: 1.4667e-05 lr: 1.4667e-05 eta: 1:06:33 time: 0.4515 data_time: 0.0247 memory: 15809 grad_norm: 19.0588 loss: 9.0142 decode.loss_cls_ce: 1.8978 decode.loss_mask_ce: 0.8441 decode.loss_mask_dice: 1.7447 decode.d7.loss_cls_ce: 1.9447 decode.d7.loss_mask_ce: 0.8388 decode.d7.loss_mask_dice: 1.7441 2023/09/06 21:00:19 - mmengine - INFO - Iter(train) [51250/60000] base_lr: 1.4584e-05 lr: 1.4584e-05 eta: 1:06:11 time: 0.4615 data_time: 0.0236 memory: 15847 grad_norm: 17.8632 loss: 8.7164 decode.loss_cls_ce: 1.9563 decode.loss_mask_ce: 0.7930 decode.loss_mask_dice: 1.6132 decode.d7.loss_cls_ce: 1.9505 decode.d7.loss_mask_ce: 0.7919 decode.d7.loss_mask_dice: 1.6115 2023/09/06 21:00:42 - mmengine - INFO - Iter(train) [51300/60000] base_lr: 1.4500e-05 lr: 1.4500e-05 eta: 1:05:48 time: 0.4530 data_time: 0.0244 memory: 15806 grad_norm: 17.0906 loss: 7.3912 decode.loss_cls_ce: 1.6675 decode.loss_mask_ce: 0.7450 decode.loss_mask_dice: 1.2818 decode.d7.loss_cls_ce: 1.6787 decode.d7.loss_mask_ce: 0.7422 decode.d7.loss_mask_dice: 1.2760 2023/09/06 21:01:05 - mmengine - INFO - Iter(train) [51350/60000] base_lr: 1.4417e-05 lr: 1.4417e-05 eta: 1:05:25 time: 0.4599 data_time: 0.0246 memory: 15961 grad_norm: 16.9782 loss: 9.2492 decode.loss_cls_ce: 2.0198 decode.loss_mask_ce: 0.8805 decode.loss_mask_dice: 1.7147 decode.d7.loss_cls_ce: 2.0311 decode.d7.loss_mask_ce: 0.8793 decode.d7.loss_mask_dice: 1.7237 2023/09/06 21:01:28 - mmengine - INFO - Iter(train) [51400/60000] base_lr: 1.4334e-05 lr: 1.4334e-05 eta: 1:05:03 time: 0.4625 data_time: 0.0233 memory: 15921 grad_norm: 18.7734 loss: 8.4569 decode.loss_cls_ce: 1.7072 decode.loss_mask_ce: 0.8569 decode.loss_mask_dice: 1.6572 decode.d7.loss_cls_ce: 1.7150 decode.d7.loss_mask_ce: 0.8672 decode.d7.loss_mask_dice: 1.6534 2023/09/06 21:01:50 - mmengine - INFO - Iter(train) [51450/60000] base_lr: 1.4250e-05 lr: 1.4250e-05 eta: 1:04:40 time: 0.4514 data_time: 0.0243 memory: 15769 grad_norm: 21.2256 loss: 8.8412 decode.loss_cls_ce: 1.9090 decode.loss_mask_ce: 0.8606 decode.loss_mask_dice: 1.6282 decode.d7.loss_cls_ce: 1.9094 decode.d7.loss_mask_ce: 0.8702 decode.d7.loss_mask_dice: 1.6636 2023/09/06 21:02:13 - mmengine - INFO - Iter(train) [51500/60000] base_lr: 1.4167e-05 lr: 1.4167e-05 eta: 1:04:17 time: 0.4587 data_time: 0.0254 memory: 15884 grad_norm: 20.2227 loss: 8.9748 decode.loss_cls_ce: 1.9751 decode.loss_mask_ce: 0.8558 decode.loss_mask_dice: 1.6532 decode.d7.loss_cls_ce: 1.9810 decode.d7.loss_mask_ce: 0.8507 decode.d7.loss_mask_dice: 1.6590 2023/09/06 21:02:36 - mmengine - INFO - Iter(train) [51550/60000] base_lr: 1.4084e-05 lr: 1.4084e-05 eta: 1:03:55 time: 0.4617 data_time: 0.0242 memory: 15895 grad_norm: 18.8757 loss: 9.3107 decode.loss_cls_ce: 2.0868 decode.loss_mask_ce: 0.8773 decode.loss_mask_dice: 1.6970 decode.d7.loss_cls_ce: 2.0893 decode.d7.loss_mask_ce: 0.8684 decode.d7.loss_mask_dice: 1.6919 2023/09/06 21:02:59 - mmengine - INFO - Iter(train) [51600/60000] base_lr: 1.4000e-05 lr: 1.4000e-05 eta: 1:03:32 time: 0.4519 data_time: 0.0244 memory: 15872 grad_norm: 17.0353 loss: 9.3575 decode.loss_cls_ce: 2.0174 decode.loss_mask_ce: 0.8935 decode.loss_mask_dice: 1.7444 decode.d7.loss_cls_ce: 2.0590 decode.d7.loss_mask_ce: 0.8962 decode.d7.loss_mask_dice: 1.7470 2023/09/06 21:03:22 - mmengine - INFO - Iter(train) [51650/60000] base_lr: 1.3917e-05 lr: 1.3917e-05 eta: 1:03:09 time: 0.4591 data_time: 0.0244 memory: 15882 grad_norm: 18.4861 loss: 9.9190 decode.loss_cls_ce: 2.1842 decode.loss_mask_ce: 0.9615 decode.loss_mask_dice: 1.8082 decode.d7.loss_cls_ce: 2.1816 decode.d7.loss_mask_ce: 0.9629 decode.d7.loss_mask_dice: 1.8205 2023/09/06 21:03:44 - mmengine - INFO - Iter(train) [51700/60000] base_lr: 1.3834e-05 lr: 1.3834e-05 eta: 1:02:47 time: 0.4566 data_time: 0.0238 memory: 15796 grad_norm: 16.6493 loss: 8.8225 decode.loss_cls_ce: 1.9480 decode.loss_mask_ce: 0.8162 decode.loss_mask_dice: 1.6283 decode.d7.loss_cls_ce: 1.9604 decode.d7.loss_mask_ce: 0.8198 decode.d7.loss_mask_dice: 1.6498 2023/09/06 21:04:07 - mmengine - INFO - Iter(train) [51750/60000] base_lr: 1.3750e-05 lr: 1.3750e-05 eta: 1:02:24 time: 0.4551 data_time: 0.0258 memory: 15979 grad_norm: 17.5675 loss: 9.7381 decode.loss_cls_ce: 2.0306 decode.loss_mask_ce: 0.9741 decode.loss_mask_dice: 1.8688 decode.d7.loss_cls_ce: 2.0284 decode.d7.loss_mask_ce: 0.9799 decode.d7.loss_mask_dice: 1.8561 2023/09/06 21:04:30 - mmengine - INFO - Iter(train) [51800/60000] base_lr: 1.3667e-05 lr: 1.3667e-05 eta: 1:02:01 time: 0.4529 data_time: 0.0244 memory: 15922 grad_norm: 18.6769 loss: 9.4111 decode.loss_cls_ce: 2.0454 decode.loss_mask_ce: 0.9315 decode.loss_mask_dice: 1.7168 decode.d7.loss_cls_ce: 2.0462 decode.d7.loss_mask_ce: 0.9381 decode.d7.loss_mask_dice: 1.7330 2023/09/06 21:04:53 - mmengine - INFO - Iter(train) [51850/60000] base_lr: 1.3584e-05 lr: 1.3584e-05 eta: 1:01:39 time: 0.4634 data_time: 0.0241 memory: 15822 grad_norm: 17.3136 loss: 9.7146 decode.loss_cls_ce: 2.0755 decode.loss_mask_ce: 0.9101 decode.loss_mask_dice: 1.8587 decode.d7.loss_cls_ce: 2.1012 decode.d7.loss_mask_ce: 0.9185 decode.d7.loss_mask_dice: 1.8507 2023/09/06 21:05:16 - mmengine - INFO - Iter(train) [51900/60000] base_lr: 1.3500e-05 lr: 1.3500e-05 eta: 1:01:16 time: 0.4627 data_time: 0.0239 memory: 15884 grad_norm: 21.2353 loss: 7.6905 decode.loss_cls_ce: 1.7743 decode.loss_mask_ce: 0.7670 decode.loss_mask_dice: 1.2950 decode.d7.loss_cls_ce: 1.7970 decode.d7.loss_mask_ce: 0.7640 decode.d7.loss_mask_dice: 1.2933 2023/09/06 21:05:39 - mmengine - INFO - Iter(train) [51950/60000] base_lr: 1.3417e-05 lr: 1.3417e-05 eta: 1:00:53 time: 0.4614 data_time: 0.0238 memory: 15784 grad_norm: 18.5961 loss: 8.3895 decode.loss_cls_ce: 1.7798 decode.loss_mask_ce: 0.8492 decode.loss_mask_dice: 1.5623 decode.d7.loss_cls_ce: 1.7926 decode.d7.loss_mask_ce: 0.8432 decode.d7.loss_mask_dice: 1.5625 2023/09/06 21:06:02 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:06:02 - mmengine - INFO - Iter(train) [52000/60000] base_lr: 1.3334e-05 lr: 1.3334e-05 eta: 1:00:31 time: 0.4643 data_time: 0.0235 memory: 15873 grad_norm: 19.6453 loss: 9.2559 decode.loss_cls_ce: 2.1666 decode.loss_mask_ce: 0.8649 decode.loss_mask_dice: 1.5988 decode.d7.loss_cls_ce: 2.1700 decode.d7.loss_mask_ce: 0.8550 decode.d7.loss_mask_dice: 1.6005 2023/09/06 21:06:25 - mmengine - INFO - Iter(train) [52050/60000] base_lr: 1.3250e-05 lr: 1.3250e-05 eta: 1:00:08 time: 0.4590 data_time: 0.0241 memory: 15884 grad_norm: 17.1572 loss: 8.6398 decode.loss_cls_ce: 1.8340 decode.loss_mask_ce: 0.8529 decode.loss_mask_dice: 1.6291 decode.d7.loss_cls_ce: 1.8403 decode.d7.loss_mask_ce: 0.8605 decode.d7.loss_mask_dice: 1.6230 2023/09/06 21:06:48 - mmengine - INFO - Iter(train) [52100/60000] base_lr: 1.3167e-05 lr: 1.3167e-05 eta: 0:59:45 time: 0.4610 data_time: 0.0233 memory: 15832 grad_norm: 17.1498 loss: 8.8736 decode.loss_cls_ce: 1.9571 decode.loss_mask_ce: 0.8485 decode.loss_mask_dice: 1.6367 decode.d7.loss_cls_ce: 1.9535 decode.d7.loss_mask_ce: 0.8350 decode.d7.loss_mask_dice: 1.6428 2023/09/06 21:07:11 - mmengine - INFO - Iter(train) [52150/60000] base_lr: 1.3084e-05 lr: 1.3084e-05 eta: 0:59:23 time: 0.4600 data_time: 0.0233 memory: 15834 grad_norm: 17.5960 loss: 8.3373 decode.loss_cls_ce: 1.8403 decode.loss_mask_ce: 0.8421 decode.loss_mask_dice: 1.4763 decode.d7.loss_cls_ce: 1.8435 decode.d7.loss_mask_ce: 0.8429 decode.d7.loss_mask_dice: 1.4921 2023/09/06 21:07:35 - mmengine - INFO - Iter(train) [52200/60000] base_lr: 1.3000e-05 lr: 1.3000e-05 eta: 0:59:00 time: 0.4607 data_time: 0.0231 memory: 15845 grad_norm: 18.4552 loss: 9.0725 decode.loss_cls_ce: 1.8052 decode.loss_mask_ce: 0.9652 decode.loss_mask_dice: 1.7719 decode.d7.loss_cls_ce: 1.8034 decode.d7.loss_mask_ce: 0.9579 decode.d7.loss_mask_dice: 1.7690 2023/09/06 21:07:58 - mmengine - INFO - Iter(train) [52250/60000] base_lr: 1.2917e-05 lr: 1.2917e-05 eta: 0:58:37 time: 0.4568 data_time: 0.0234 memory: 15800 grad_norm: 17.0143 loss: 9.3887 decode.loss_cls_ce: 2.0640 decode.loss_mask_ce: 0.8702 decode.loss_mask_dice: 1.7421 decode.d7.loss_cls_ce: 2.0936 decode.d7.loss_mask_ce: 0.8707 decode.d7.loss_mask_dice: 1.7481 2023/09/06 21:08:20 - mmengine - INFO - Iter(train) [52300/60000] base_lr: 1.2834e-05 lr: 1.2834e-05 eta: 0:58:15 time: 0.4530 data_time: 0.0251 memory: 16002 grad_norm: 16.4652 loss: 8.7973 decode.loss_cls_ce: 1.9851 decode.loss_mask_ce: 0.7908 decode.loss_mask_dice: 1.6203 decode.d7.loss_cls_ce: 1.9744 decode.d7.loss_mask_ce: 0.7956 decode.d7.loss_mask_dice: 1.6310 2023/09/06 21:08:43 - mmengine - INFO - Iter(train) [52350/60000] base_lr: 1.2750e-05 lr: 1.2750e-05 eta: 0:57:52 time: 0.4564 data_time: 0.0245 memory: 15925 grad_norm: 17.3649 loss: 9.2956 decode.loss_cls_ce: 2.0131 decode.loss_mask_ce: 0.8171 decode.loss_mask_dice: 1.7993 decode.d7.loss_cls_ce: 2.0605 decode.d7.loss_mask_ce: 0.8084 decode.d7.loss_mask_dice: 1.7972 2023/09/06 21:09:06 - mmengine - INFO - Iter(train) [52400/60000] base_lr: 1.2667e-05 lr: 1.2667e-05 eta: 0:57:29 time: 0.4649 data_time: 0.0240 memory: 15823 grad_norm: 19.3356 loss: 9.2119 decode.loss_cls_ce: 1.9837 decode.loss_mask_ce: 0.9307 decode.loss_mask_dice: 1.6743 decode.d7.loss_cls_ce: 2.0159 decode.d7.loss_mask_ce: 0.9273 decode.d7.loss_mask_dice: 1.6800 2023/09/06 21:09:29 - mmengine - INFO - Iter(train) [52450/60000] base_lr: 1.2584e-05 lr: 1.2584e-05 eta: 0:57:07 time: 0.4590 data_time: 0.0241 memory: 15807 grad_norm: 18.5462 loss: 8.9687 decode.loss_cls_ce: 1.9564 decode.loss_mask_ce: 0.8586 decode.loss_mask_dice: 1.6713 decode.d7.loss_cls_ce: 1.9348 decode.d7.loss_mask_ce: 0.8625 decode.d7.loss_mask_dice: 1.6851 2023/09/06 21:09:52 - mmengine - INFO - Iter(train) [52500/60000] base_lr: 1.2500e-05 lr: 1.2500e-05 eta: 0:56:44 time: 0.4544 data_time: 0.0243 memory: 15705 grad_norm: 19.4731 loss: 8.4468 decode.loss_cls_ce: 1.8321 decode.loss_mask_ce: 0.8178 decode.loss_mask_dice: 1.5717 decode.d7.loss_cls_ce: 1.8668 decode.d7.loss_mask_ce: 0.8081 decode.d7.loss_mask_dice: 1.5504 2023/09/06 21:10:15 - mmengine - INFO - Iter(train) [52550/60000] base_lr: 1.2417e-05 lr: 1.2417e-05 eta: 0:56:21 time: 0.4542 data_time: 0.0249 memory: 15773 grad_norm: 20.0671 loss: 9.1335 decode.loss_cls_ce: 1.9124 decode.loss_mask_ce: 0.9014 decode.loss_mask_dice: 1.7497 decode.d7.loss_cls_ce: 1.9333 decode.d7.loss_mask_ce: 0.8971 decode.d7.loss_mask_dice: 1.7397 2023/09/06 21:10:37 - mmengine - INFO - Iter(train) [52600/60000] base_lr: 1.2334e-05 lr: 1.2334e-05 eta: 0:55:59 time: 0.4539 data_time: 0.0231 memory: 15885 grad_norm: 20.1887 loss: 8.3438 decode.loss_cls_ce: 1.8613 decode.loss_mask_ce: 0.7635 decode.loss_mask_dice: 1.5344 decode.d7.loss_cls_ce: 1.8678 decode.d7.loss_mask_ce: 0.7615 decode.d7.loss_mask_dice: 1.5554 2023/09/06 21:11:00 - mmengine - INFO - Iter(train) [52650/60000] base_lr: 1.2250e-05 lr: 1.2250e-05 eta: 0:55:36 time: 0.4542 data_time: 0.0256 memory: 15786 grad_norm: 18.0290 loss: 7.9614 decode.loss_cls_ce: 1.6639 decode.loss_mask_ce: 0.8091 decode.loss_mask_dice: 1.5020 decode.d7.loss_cls_ce: 1.6744 decode.d7.loss_mask_ce: 0.8018 decode.d7.loss_mask_dice: 1.5101 2023/09/06 21:11:23 - mmengine - INFO - Iter(train) [52700/60000] base_lr: 1.2167e-05 lr: 1.2167e-05 eta: 0:55:13 time: 0.4576 data_time: 0.0243 memory: 15965 grad_norm: 17.3162 loss: 9.2886 decode.loss_cls_ce: 2.0322 decode.loss_mask_ce: 0.8329 decode.loss_mask_dice: 1.7597 decode.d7.loss_cls_ce: 2.0713 decode.d7.loss_mask_ce: 0.8282 decode.d7.loss_mask_dice: 1.7643 2023/09/06 21:11:46 - mmengine - INFO - Iter(train) [52750/60000] base_lr: 1.2084e-05 lr: 1.2084e-05 eta: 0:54:51 time: 0.4570 data_time: 0.0246 memory: 15935 grad_norm: 19.8252 loss: 8.7099 decode.loss_cls_ce: 1.8083 decode.loss_mask_ce: 0.9383 decode.loss_mask_dice: 1.5987 decode.d7.loss_cls_ce: 1.8387 decode.d7.loss_mask_ce: 0.9298 decode.d7.loss_mask_dice: 1.5961 2023/09/06 21:12:09 - mmengine - INFO - Iter(train) [52800/60000] base_lr: 1.2000e-05 lr: 1.2000e-05 eta: 0:54:28 time: 0.4525 data_time: 0.0240 memory: 15756 grad_norm: 16.4677 loss: 8.6476 decode.loss_cls_ce: 1.9161 decode.loss_mask_ce: 0.8525 decode.loss_mask_dice: 1.5546 decode.d7.loss_cls_ce: 1.9132 decode.d7.loss_mask_ce: 0.8555 decode.d7.loss_mask_dice: 1.5557 2023/09/06 21:12:31 - mmengine - INFO - Iter(train) [52850/60000] base_lr: 1.1917e-05 lr: 1.1917e-05 eta: 0:54:05 time: 0.4559 data_time: 0.0238 memory: 15706 grad_norm: 16.7379 loss: 8.8737 decode.loss_cls_ce: 1.8542 decode.loss_mask_ce: 0.8506 decode.loss_mask_dice: 1.7103 decode.d7.loss_cls_ce: 1.8762 decode.d7.loss_mask_ce: 0.8625 decode.d7.loss_mask_dice: 1.7200 2023/09/06 21:12:54 - mmengine - INFO - Iter(train) [52900/60000] base_lr: 1.1834e-05 lr: 1.1834e-05 eta: 0:53:43 time: 0.4540 data_time: 0.0252 memory: 15926 grad_norm: 19.3227 loss: 9.3988 decode.loss_cls_ce: 2.0073 decode.loss_mask_ce: 0.8957 decode.loss_mask_dice: 1.7833 decode.d7.loss_cls_ce: 2.0438 decode.d7.loss_mask_ce: 0.8909 decode.d7.loss_mask_dice: 1.7777 2023/09/06 21:13:17 - mmengine - INFO - Iter(train) [52950/60000] base_lr: 1.1750e-05 lr: 1.1750e-05 eta: 0:53:20 time: 0.4622 data_time: 0.0237 memory: 15821 grad_norm: 19.3585 loss: 8.7389 decode.loss_cls_ce: 1.8790 decode.loss_mask_ce: 0.7643 decode.loss_mask_dice: 1.7145 decode.d7.loss_cls_ce: 1.8933 decode.d7.loss_mask_ce: 0.7668 decode.d7.loss_mask_dice: 1.7210 2023/09/06 21:13:40 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:13:40 - mmengine - INFO - Iter(train) [53000/60000] base_lr: 1.1667e-05 lr: 1.1667e-05 eta: 0:52:57 time: 0.4516 data_time: 0.0242 memory: 15859 grad_norm: nan loss: 8.3788 decode.loss_cls_ce: 1.7350 decode.loss_mask_ce: 0.8355 decode.loss_mask_dice: 1.6203 decode.d7.loss_cls_ce: 1.7410 decode.d7.loss_mask_ce: 0.8420 decode.d7.loss_mask_dice: 1.6051 2023/09/06 21:14:03 - mmengine - INFO - Iter(train) [53050/60000] base_lr: 1.1584e-05 lr: 1.1584e-05 eta: 0:52:34 time: 0.4520 data_time: 0.0246 memory: 15860 grad_norm: 19.3297 loss: 9.4852 decode.loss_cls_ce: 2.0550 decode.loss_mask_ce: 0.8665 decode.loss_mask_dice: 1.8214 decode.d7.loss_cls_ce: 2.0575 decode.d7.loss_mask_ce: 0.8620 decode.d7.loss_mask_dice: 1.8229 2023/09/06 21:14:25 - mmengine - INFO - Iter(train) [53100/60000] base_lr: 1.1500e-05 lr: 1.1500e-05 eta: 0:52:12 time: 0.4506 data_time: 0.0240 memory: 15908 grad_norm: 17.1920 loss: 8.8888 decode.loss_cls_ce: 1.9372 decode.loss_mask_ce: 0.8254 decode.loss_mask_dice: 1.6750 decode.d7.loss_cls_ce: 1.9549 decode.d7.loss_mask_ce: 0.8256 decode.d7.loss_mask_dice: 1.6707 2023/09/06 21:14:48 - mmengine - INFO - Iter(train) [53150/60000] base_lr: 1.1417e-05 lr: 1.1417e-05 eta: 0:51:49 time: 0.4598 data_time: 0.0243 memory: 15898 grad_norm: 18.5057 loss: 8.1503 decode.loss_cls_ce: 1.7522 decode.loss_mask_ce: 0.8647 decode.loss_mask_dice: 1.4565 decode.d7.loss_cls_ce: 1.7699 decode.d7.loss_mask_ce: 0.8567 decode.d7.loss_mask_dice: 1.4503 2023/09/06 21:15:11 - mmengine - INFO - Iter(train) [53200/60000] base_lr: 1.1334e-05 lr: 1.1334e-05 eta: 0:51:26 time: 0.4570 data_time: 0.0247 memory: 15924 grad_norm: 21.4836 loss: 8.2426 decode.loss_cls_ce: 1.7208 decode.loss_mask_ce: 0.8267 decode.loss_mask_dice: 1.5623 decode.d7.loss_cls_ce: 1.7178 decode.d7.loss_mask_ce: 0.8376 decode.d7.loss_mask_dice: 1.5773 2023/09/06 21:15:34 - mmengine - INFO - Iter(train) [53250/60000] base_lr: 1.1250e-05 lr: 1.1250e-05 eta: 0:51:04 time: 0.4574 data_time: 0.0250 memory: 16159 grad_norm: 16.6346 loss: 8.2544 decode.loss_cls_ce: 1.8216 decode.loss_mask_ce: 0.8294 decode.loss_mask_dice: 1.4735 decode.d7.loss_cls_ce: 1.8206 decode.d7.loss_mask_ce: 0.8265 decode.d7.loss_mask_dice: 1.4828 2023/09/06 21:15:57 - mmengine - INFO - Iter(train) [53300/60000] base_lr: 1.1167e-05 lr: 1.1167e-05 eta: 0:50:41 time: 0.4565 data_time: 0.0247 memory: 16040 grad_norm: 16.8839 loss: 9.5094 decode.loss_cls_ce: 2.0762 decode.loss_mask_ce: 0.9222 decode.loss_mask_dice: 1.7573 decode.d7.loss_cls_ce: 2.1051 decode.d7.loss_mask_ce: 0.9147 decode.d7.loss_mask_dice: 1.7339 2023/09/06 21:16:20 - mmengine - INFO - Iter(train) [53350/60000] base_lr: 1.1084e-05 lr: 1.1084e-05 eta: 0:50:18 time: 0.4596 data_time: 0.0231 memory: 15743 grad_norm: 16.2433 loss: 7.9352 decode.loss_cls_ce: 1.7948 decode.loss_mask_ce: 0.7834 decode.loss_mask_dice: 1.3762 decode.d7.loss_cls_ce: 1.8112 decode.d7.loss_mask_ce: 0.7840 decode.d7.loss_mask_dice: 1.3855 2023/09/06 21:16:42 - mmengine - INFO - Iter(train) [53400/60000] base_lr: 1.1000e-05 lr: 1.1000e-05 eta: 0:49:56 time: 0.4544 data_time: 0.0251 memory: 15897 grad_norm: 16.7928 loss: 9.0654 decode.loss_cls_ce: 1.9219 decode.loss_mask_ce: 0.8997 decode.loss_mask_dice: 1.7000 decode.d7.loss_cls_ce: 1.9321 decode.d7.loss_mask_ce: 0.9015 decode.d7.loss_mask_dice: 1.7102 2023/09/06 21:17:05 - mmengine - INFO - Iter(train) [53450/60000] base_lr: 1.0917e-05 lr: 1.0917e-05 eta: 0:49:33 time: 0.4522 data_time: 0.0243 memory: 15860 grad_norm: 19.1257 loss: 9.6841 decode.loss_cls_ce: 2.2219 decode.loss_mask_ce: 0.8442 decode.loss_mask_dice: 1.7804 decode.d7.loss_cls_ce: 2.2099 decode.d7.loss_mask_ce: 0.8502 decode.d7.loss_mask_dice: 1.7774 2023/09/06 21:17:28 - mmengine - INFO - Iter(train) [53500/60000] base_lr: 1.0834e-05 lr: 1.0834e-05 eta: 0:49:10 time: 0.4642 data_time: 0.0238 memory: 15833 grad_norm: 17.4880 loss: 8.9829 decode.loss_cls_ce: 1.9125 decode.loss_mask_ce: 0.8698 decode.loss_mask_dice: 1.6969 decode.d7.loss_cls_ce: 1.9271 decode.d7.loss_mask_ce: 0.8831 decode.d7.loss_mask_dice: 1.6934 2023/09/06 21:17:51 - mmengine - INFO - Iter(train) [53550/60000] base_lr: 1.0750e-05 lr: 1.0750e-05 eta: 0:48:48 time: 0.4543 data_time: 0.0238 memory: 15794 grad_norm: 17.8890 loss: 8.9794 decode.loss_cls_ce: 1.9218 decode.loss_mask_ce: 0.8781 decode.loss_mask_dice: 1.6726 decode.d7.loss_cls_ce: 1.9344 decode.d7.loss_mask_ce: 0.8861 decode.d7.loss_mask_dice: 1.6864 2023/09/06 21:18:14 - mmengine - INFO - Iter(train) [53600/60000] base_lr: 1.0667e-05 lr: 1.0667e-05 eta: 0:48:25 time: 0.4543 data_time: 0.0250 memory: 15835 grad_norm: 17.8452 loss: 9.1314 decode.loss_cls_ce: 1.9946 decode.loss_mask_ce: 0.8472 decode.loss_mask_dice: 1.7226 decode.d7.loss_cls_ce: 1.9864 decode.d7.loss_mask_ce: 0.8495 decode.d7.loss_mask_dice: 1.7312 2023/09/06 21:18:37 - mmengine - INFO - Iter(train) [53650/60000] base_lr: 1.0584e-05 lr: 1.0584e-05 eta: 0:48:02 time: 0.4532 data_time: 0.0243 memory: 15872 grad_norm: 16.7242 loss: 9.1138 decode.loss_cls_ce: 1.8850 decode.loss_mask_ce: 0.9025 decode.loss_mask_dice: 1.7524 decode.d7.loss_cls_ce: 1.9100 decode.d7.loss_mask_ce: 0.9041 decode.d7.loss_mask_dice: 1.7598 2023/09/06 21:19:00 - mmengine - INFO - Iter(train) [53700/60000] base_lr: 1.0500e-05 lr: 1.0500e-05 eta: 0:47:40 time: 0.4616 data_time: 0.0235 memory: 15821 grad_norm: 17.5647 loss: 9.1162 decode.loss_cls_ce: 2.0009 decode.loss_mask_ce: 0.8794 decode.loss_mask_dice: 1.6655 decode.d7.loss_cls_ce: 2.0134 decode.d7.loss_mask_ce: 0.8739 decode.d7.loss_mask_dice: 1.6831 2023/09/06 21:19:23 - mmengine - INFO - Iter(train) [53750/60000] base_lr: 1.0417e-05 lr: 1.0417e-05 eta: 0:47:17 time: 0.4507 data_time: 0.0237 memory: 15733 grad_norm: 18.5195 loss: 8.9428 decode.loss_cls_ce: 2.0022 decode.loss_mask_ce: 0.8037 decode.loss_mask_dice: 1.6594 decode.d7.loss_cls_ce: 2.0152 decode.d7.loss_mask_ce: 0.8070 decode.d7.loss_mask_dice: 1.6555 2023/09/06 21:19:45 - mmengine - INFO - Iter(train) [53800/60000] base_lr: 1.0334e-05 lr: 1.0334e-05 eta: 0:46:54 time: 0.4639 data_time: 0.0236 memory: 15845 grad_norm: 18.0801 loss: 9.2738 decode.loss_cls_ce: 1.9732 decode.loss_mask_ce: 0.8618 decode.loss_mask_dice: 1.8089 decode.d7.loss_cls_ce: 1.9735 decode.d7.loss_mask_ce: 0.8548 decode.d7.loss_mask_dice: 1.8016 2023/09/06 21:20:08 - mmengine - INFO - Iter(train) [53850/60000] base_lr: 1.0250e-05 lr: 1.0250e-05 eta: 0:46:32 time: 0.4587 data_time: 0.0243 memory: 15896 grad_norm: 17.5943 loss: 8.7841 decode.loss_cls_ce: 1.9031 decode.loss_mask_ce: 0.8279 decode.loss_mask_dice: 1.6614 decode.d7.loss_cls_ce: 1.8940 decode.d7.loss_mask_ce: 0.8327 decode.d7.loss_mask_dice: 1.6652 2023/09/06 21:20:31 - mmengine - INFO - Iter(train) [53900/60000] base_lr: 1.0167e-05 lr: 1.0167e-05 eta: 0:46:09 time: 0.4622 data_time: 0.0235 memory: 15962 grad_norm: 17.8137 loss: 8.4321 decode.loss_cls_ce: 1.7859 decode.loss_mask_ce: 0.8622 decode.loss_mask_dice: 1.5504 decode.d7.loss_cls_ce: 1.8234 decode.d7.loss_mask_ce: 0.8634 decode.d7.loss_mask_dice: 1.5469 2023/09/06 21:20:54 - mmengine - INFO - Iter(train) [53950/60000] base_lr: 1.0084e-05 lr: 1.0084e-05 eta: 0:45:46 time: 0.4530 data_time: 0.0243 memory: 15963 grad_norm: 16.5187 loss: 9.4661 decode.loss_cls_ce: 2.0430 decode.loss_mask_ce: 0.8477 decode.loss_mask_dice: 1.8216 decode.d7.loss_cls_ce: 2.0850 decode.d7.loss_mask_ce: 0.8403 decode.d7.loss_mask_dice: 1.8284 2023/09/06 21:21:17 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:21:17 - mmengine - INFO - Iter(train) [54000/60000] base_lr: 1.0000e-05 lr: 1.0000e-05 eta: 0:45:23 time: 0.4493 data_time: 0.0238 memory: 15898 grad_norm: 18.5208 loss: 8.7306 decode.loss_cls_ce: 1.7821 decode.loss_mask_ce: 0.9167 decode.loss_mask_dice: 1.6707 decode.d7.loss_cls_ce: 1.7791 decode.d7.loss_mask_ce: 0.9163 decode.d7.loss_mask_dice: 1.6656 2023/09/06 21:21:39 - mmengine - INFO - Iter(train) [54050/60000] base_lr: 9.9168e-06 lr: 9.9168e-06 eta: 0:45:01 time: 0.4548 data_time: 0.0241 memory: 15798 grad_norm: 18.7872 loss: 8.5512 decode.loss_cls_ce: 1.9040 decode.loss_mask_ce: 0.8897 decode.loss_mask_dice: 1.4744 decode.d7.loss_cls_ce: 1.9285 decode.d7.loss_mask_ce: 0.8835 decode.d7.loss_mask_dice: 1.4711 2023/09/06 21:22:02 - mmengine - INFO - Iter(train) [54100/60000] base_lr: 9.8335e-06 lr: 9.8335e-06 eta: 0:44:38 time: 0.4630 data_time: 0.0238 memory: 15786 grad_norm: 17.5966 loss: 8.4275 decode.loss_cls_ce: 1.8411 decode.loss_mask_ce: 0.8073 decode.loss_mask_dice: 1.5438 decode.d7.loss_cls_ce: 1.8806 decode.d7.loss_mask_ce: 0.8097 decode.d7.loss_mask_dice: 1.5449 2023/09/06 21:22:25 - mmengine - INFO - Iter(train) [54150/60000] base_lr: 9.7502e-06 lr: 9.7502e-06 eta: 0:44:15 time: 0.4620 data_time: 0.0236 memory: 15857 grad_norm: 19.7054 loss: 8.7100 decode.loss_cls_ce: 1.8589 decode.loss_mask_ce: 0.8679 decode.loss_mask_dice: 1.6293 decode.d7.loss_cls_ce: 1.8651 decode.d7.loss_mask_ce: 0.8615 decode.d7.loss_mask_dice: 1.6274 2023/09/06 21:22:48 - mmengine - INFO - Iter(train) [54200/60000] base_lr: 9.6668e-06 lr: 9.6668e-06 eta: 0:43:53 time: 0.4512 data_time: 0.0240 memory: 15732 grad_norm: 18.0144 loss: 8.5763 decode.loss_cls_ce: 1.7089 decode.loss_mask_ce: 0.8567 decode.loss_mask_dice: 1.7154 decode.d7.loss_cls_ce: 1.7251 decode.d7.loss_mask_ce: 0.8481 decode.d7.loss_mask_dice: 1.7221 2023/09/06 21:23:11 - mmengine - INFO - Iter(train) [54250/60000] base_lr: 9.5835e-06 lr: 9.5835e-06 eta: 0:43:30 time: 0.4526 data_time: 0.0243 memory: 15807 grad_norm: 17.1591 loss: 8.5403 decode.loss_cls_ce: 1.8723 decode.loss_mask_ce: 0.7751 decode.loss_mask_dice: 1.6145 decode.d7.loss_cls_ce: 1.8640 decode.d7.loss_mask_ce: 0.7848 decode.d7.loss_mask_dice: 1.6298 2023/09/06 21:23:34 - mmengine - INFO - Iter(train) [54300/60000] base_lr: 9.5002e-06 lr: 9.5002e-06 eta: 0:43:07 time: 0.4623 data_time: 0.0238 memory: 15887 grad_norm: 17.4482 loss: 8.5090 decode.loss_cls_ce: 1.8945 decode.loss_mask_ce: 0.7728 decode.loss_mask_dice: 1.5825 decode.d7.loss_cls_ce: 1.8986 decode.d7.loss_mask_ce: 0.7794 decode.d7.loss_mask_dice: 1.5812 2023/09/06 21:23:57 - mmengine - INFO - Iter(train) [54350/60000] base_lr: 9.4168e-06 lr: 9.4168e-06 eta: 0:42:45 time: 0.4568 data_time: 0.0240 memory: 15990 grad_norm: 18.4930 loss: 9.3977 decode.loss_cls_ce: 2.1662 decode.loss_mask_ce: 0.8291 decode.loss_mask_dice: 1.7102 decode.d7.loss_cls_ce: 2.1413 decode.d7.loss_mask_ce: 0.8352 decode.d7.loss_mask_dice: 1.7158 2023/09/06 21:24:20 - mmengine - INFO - Iter(train) [54400/60000] base_lr: 9.3335e-06 lr: 9.3335e-06 eta: 0:42:22 time: 0.4636 data_time: 0.0243 memory: 16038 grad_norm: 18.4172 loss: 9.7062 decode.loss_cls_ce: 2.0088 decode.loss_mask_ce: 0.9140 decode.loss_mask_dice: 1.9245 decode.d7.loss_cls_ce: 2.0228 decode.d7.loss_mask_ce: 0.9110 decode.d7.loss_mask_dice: 1.9250 2023/09/06 21:24:43 - mmengine - INFO - Iter(train) [54450/60000] base_lr: 9.2502e-06 lr: 9.2502e-06 eta: 0:41:59 time: 0.4627 data_time: 0.0236 memory: 15820 grad_norm: 20.3145 loss: 9.0436 decode.loss_cls_ce: 1.9328 decode.loss_mask_ce: 0.8953 decode.loss_mask_dice: 1.6911 decode.d7.loss_cls_ce: 1.9393 decode.d7.loss_mask_ce: 0.8976 decode.d7.loss_mask_dice: 1.6874 2023/09/06 21:25:06 - mmengine - INFO - Iter(train) [54500/60000] base_lr: 9.1668e-06 lr: 9.1668e-06 eta: 0:41:37 time: 0.4620 data_time: 0.0235 memory: 15808 grad_norm: 18.0753 loss: 9.0411 decode.loss_cls_ce: 2.0309 decode.loss_mask_ce: 0.8317 decode.loss_mask_dice: 1.6458 decode.d7.loss_cls_ce: 2.0392 decode.d7.loss_mask_ce: 0.8331 decode.d7.loss_mask_dice: 1.6605 2023/09/06 21:25:29 - mmengine - INFO - Iter(train) [54550/60000] base_lr: 9.0835e-06 lr: 9.0835e-06 eta: 0:41:14 time: 0.4560 data_time: 0.0240 memory: 15831 grad_norm: 20.6244 loss: 8.7193 decode.loss_cls_ce: 1.9395 decode.loss_mask_ce: 0.8379 decode.loss_mask_dice: 1.5819 decode.d7.loss_cls_ce: 1.9344 decode.d7.loss_mask_ce: 0.8444 decode.d7.loss_mask_dice: 1.5813 2023/09/06 21:25:51 - mmengine - INFO - Iter(train) [54600/60000] base_lr: 9.0002e-06 lr: 9.0002e-06 eta: 0:40:51 time: 0.4563 data_time: 0.0242 memory: 15748 grad_norm: 24.9042 loss: 9.0822 decode.loss_cls_ce: 1.9633 decode.loss_mask_ce: 0.8765 decode.loss_mask_dice: 1.6959 decode.d7.loss_cls_ce: 1.9665 decode.d7.loss_mask_ce: 0.8818 decode.d7.loss_mask_dice: 1.6981 2023/09/06 21:26:14 - mmengine - INFO - Iter(train) [54650/60000] base_lr: 8.9168e-06 lr: 8.9168e-06 eta: 0:40:29 time: 0.4589 data_time: 0.0243 memory: 15898 grad_norm: 16.5995 loss: 9.1979 decode.loss_cls_ce: 2.0214 decode.loss_mask_ce: 0.9228 decode.loss_mask_dice: 1.6487 decode.d7.loss_cls_ce: 2.0264 decode.d7.loss_mask_ce: 0.9217 decode.d7.loss_mask_dice: 1.6569 2023/09/06 21:26:37 - mmengine - INFO - Iter(train) [54700/60000] base_lr: 8.8335e-06 lr: 8.8335e-06 eta: 0:40:06 time: 0.4557 data_time: 0.0242 memory: 15974 grad_norm: 18.0234 loss: 8.9285 decode.loss_cls_ce: 1.9505 decode.loss_mask_ce: 0.8381 decode.loss_mask_dice: 1.6693 decode.d7.loss_cls_ce: 1.9551 decode.d7.loss_mask_ce: 0.8430 decode.d7.loss_mask_dice: 1.6726 2023/09/06 21:27:00 - mmengine - INFO - Iter(train) [54750/60000] base_lr: 8.7501e-06 lr: 8.7501e-06 eta: 0:39:43 time: 0.4594 data_time: 0.0231 memory: 15882 grad_norm: 17.0582 loss: 8.4602 decode.loss_cls_ce: 1.7905 decode.loss_mask_ce: 0.8130 decode.loss_mask_dice: 1.6121 decode.d7.loss_cls_ce: 1.8058 decode.d7.loss_mask_ce: 0.8158 decode.d7.loss_mask_dice: 1.6231 2023/09/06 21:27:23 - mmengine - INFO - Iter(train) [54800/60000] base_lr: 8.6668e-06 lr: 8.6668e-06 eta: 0:39:21 time: 0.4641 data_time: 0.0241 memory: 15800 grad_norm: 19.5494 loss: 9.8186 decode.loss_cls_ce: 2.1447 decode.loss_mask_ce: 0.9461 decode.loss_mask_dice: 1.8259 decode.d7.loss_cls_ce: 2.1395 decode.d7.loss_mask_ce: 0.9502 decode.d7.loss_mask_dice: 1.8120 2023/09/06 21:27:46 - mmengine - INFO - Iter(train) [54850/60000] base_lr: 8.5835e-06 lr: 8.5835e-06 eta: 0:38:58 time: 0.4613 data_time: 0.0244 memory: 15845 grad_norm: 20.7972 loss: 8.7897 decode.loss_cls_ce: 1.7903 decode.loss_mask_ce: 0.9131 decode.loss_mask_dice: 1.6887 decode.d7.loss_cls_ce: 1.8023 decode.d7.loss_mask_ce: 0.9127 decode.d7.loss_mask_dice: 1.6826 2023/09/06 21:28:09 - mmengine - INFO - Iter(train) [54900/60000] base_lr: 8.5001e-06 lr: 8.5001e-06 eta: 0:38:35 time: 0.4641 data_time: 0.0242 memory: 15989 grad_norm: 19.5677 loss: 9.8153 decode.loss_cls_ce: 2.1674 decode.loss_mask_ce: 0.8512 decode.loss_mask_dice: 1.8962 decode.d7.loss_cls_ce: 2.1373 decode.d7.loss_mask_ce: 0.8592 decode.d7.loss_mask_dice: 1.9041 2023/09/06 21:28:32 - mmengine - INFO - Iter(train) [54950/60000] base_lr: 8.4168e-06 lr: 8.4168e-06 eta: 0:38:13 time: 0.4575 data_time: 0.0244 memory: 15951 grad_norm: 21.3969 loss: 8.9910 decode.loss_cls_ce: 2.0888 decode.loss_mask_ce: 0.7667 decode.loss_mask_dice: 1.6293 decode.d7.loss_cls_ce: 2.0781 decode.d7.loss_mask_ce: 0.7693 decode.d7.loss_mask_dice: 1.6588 2023/09/06 21:28:55 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:28:55 - mmengine - INFO - Iter(train) [55000/60000] base_lr: 8.3335e-06 lr: 8.3335e-06 eta: 0:37:50 time: 0.4655 data_time: 0.0254 memory: 15964 grad_norm: 17.7708 loss: 8.5666 decode.loss_cls_ce: 1.8568 decode.loss_mask_ce: 0.8035 decode.loss_mask_dice: 1.6289 decode.d7.loss_cls_ce: 1.8575 decode.d7.loss_mask_ce: 0.7959 decode.d7.loss_mask_dice: 1.6240 2023/09/06 21:29:34 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:15:16 time: 0.5329 data_time: 0.0018 memory: 26158 2023/09/06 21:29:46 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:09:37 time: 0.3394 data_time: 0.0018 memory: 26153 2023/09/06 21:29:57 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:07:34 time: 0.1560 data_time: 0.0021 memory: 26158 2023/09/06 21:30:06 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:06:12 time: 0.1649 data_time: 0.0018 memory: 26147 2023/09/06 21:30:12 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:05:06 time: 0.0497 data_time: 0.0018 memory: 9129 2023/09/06 21:30:18 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:04:22 time: 0.1916 data_time: 0.0018 memory: 9123 2023/09/06 21:30:24 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:03:47 time: 0.1095 data_time: 0.0019 memory: 26146 2023/09/06 21:30:32 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:03:25 time: 0.3374 data_time: 0.0018 memory: 26149 2023/09/06 21:30:38 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:03:03 time: 0.1269 data_time: 0.0022 memory: 26146 2023/09/06 21:30:44 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:02:42 time: 0.0923 data_time: 0.0019 memory: 1528 2023/09/06 21:30:50 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:02:25 time: 0.1010 data_time: 0.0018 memory: 26149 2023/09/06 21:30:55 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:02:09 time: 0.0863 data_time: 0.0018 memory: 26141 2023/09/06 21:30:58 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:01:53 time: 0.1490 data_time: 0.0019 memory: 1528 2023/09/06 21:31:03 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:01:40 time: 0.0841 data_time: 0.0019 memory: 9136 2023/09/06 21:31:07 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:01:27 time: 0.0608 data_time: 0.0019 memory: 9139 2023/09/06 21:31:11 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:01:16 time: 0.0497 data_time: 0.0019 memory: 1484 2023/09/06 21:31:14 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:01:05 time: 0.0476 data_time: 0.0019 memory: 24305 2023/09/06 21:31:17 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:55 time: 0.0499 data_time: 0.0019 memory: 1420 2023/09/06 21:31:20 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:45 time: 0.0930 data_time: 0.0018 memory: 1705 2023/09/06 21:31:23 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:36 time: 0.0959 data_time: 0.0022 memory: 1574 2023/09/06 21:31:27 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:28 time: 0.1416 data_time: 0.0020 memory: 1484 2023/09/06 21:31:32 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:21 time: 0.0645 data_time: 0.0019 memory: 9129 2023/09/06 21:31:36 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:13 time: 0.0854 data_time: 0.0019 memory: 9133 2023/09/06 21:31:39 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:06 time: 0.0811 data_time: 0.0020 memory: 1528 2023/09/06 21:31:41 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0471 data_time: 0.0018 memory: 1528 2023/09/06 21:31:46 - mmengine - INFO - per class results: 2023/09/06 21:31:46 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.28 | 91.73 | | bicycle | 58.55 | 78.69 | | car | 61.4 | 80.4 | | motorcycle | 79.38 | 89.8 | | airplane | 72.19 | 89.16 | | bus | 77.18 | 88.5 | | train | 77.38 | 88.19 | | truck | 60.34 | 78.21 | | boat | 56.67 | 80.34 | | traffic light | 56.19 | 82.3 | | fire hydrant | 70.61 | 92.69 | | stop sign | 78.27 | 95.69 | | parking meter | 50.9 | 84.05 | | bench | 48.41 | 69.66 | | bird | 71.11 | 86.83 | | cat | 80.57 | 89.72 | | dog | 76.75 | 84.38 | | horse | 77.39 | 91.29 | | sheep | 86.12 | 94.11 | | cow | 83.04 | 90.23 | | elephant | 90.58 | 95.37 | | bear | 88.29 | 91.2 | | zebra | 90.22 | 94.31 | | giraffe | 82.69 | 90.66 | | backpack | 24.13 | 64.8 | | umbrella | 72.47 | 77.54 | | handbag | 22.72 | 36.27 | | tie | 10.19 | 32.86 | | suitcase | 71.41 | 81.92 | | frisbee | 60.42 | 87.73 | | skis | 30.57 | 63.83 | | snowboard | 52.69 | 73.35 | | sports ball | 52.81 | 71.93 | | kite | 49.84 | 72.6 | | baseball bat | 32.5 | 67.8 | | baseball glove | 48.34 | 88.75 | | skateboard | 45.61 | 83.63 | | surfboard | 74.43 | 89.49 | | tennis racket | 70.0 | 88.53 | | bottle | 44.0 | 69.82 | | wine glass | 40.19 | 68.0 | | cup | 39.9 | 58.3 | | fork | 35.23 | 56.3 | | knife | 21.64 | 26.95 | | spoon | 16.04 | 34.37 | | bowl | 31.18 | 41.58 | | banana | 61.16 | 84.53 | | apple | 41.05 | 52.33 | | sandwich | 46.14 | 63.5 | | orange | 65.3 | 76.2 | | broccoli | 52.73 | 67.88 | | carrot | 46.09 | 58.95 | | hot dog | 50.21 | 60.92 | | pizza | 69.07 | 80.86 | | donut | 64.3 | 83.55 | | cake | 70.74 | 83.1 | | chair | 42.55 | 66.85 | | couch | 53.99 | 73.06 | | potted plant | 23.49 | 32.87 | | bed | 61.57 | 80.43 | | dining table | 41.83 | 69.12 | | toilet | 68.65 | 91.62 | | tv | 68.69 | 80.24 | | laptop | 66.43 | 84.88 | | mouse | 60.2 | 69.32 | | remote | 39.48 | 72.32 | | keyboard | 56.29 | 70.21 | | cell phone | 67.7 | 82.65 | | microwave | 45.14 | 61.0 | | oven | 49.33 | 69.83 | | toaster | 69.43 | 81.06 | | sink | 39.7 | 81.9 | | refrigerator | 65.67 | 82.63 | | book | 42.22 | 61.43 | | clock | 71.23 | 87.05 | | vase | 42.97 | 80.25 | | scissors | 55.73 | 72.13 | | teddy bear | 77.45 | 85.47 | | hair drier | 15.59 | 54.62 | | toothbrush | 41.46 | 78.0 | | banner | 27.6 | 57.18 | | blanket | 10.94 | 15.82 | | branch | 11.81 | 27.32 | | bridge | 23.47 | 36.95 | | building-other | 51.87 | 71.38 | | bush | 32.49 | 51.0 | | cabinet | 43.33 | 64.5 | | cage | 13.92 | 19.98 | | cardboard | 34.61 | 50.02 | | carpet | 49.01 | 75.26 | | ceiling-other | 59.62 | 76.36 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.39 | 0.87 | | clothes | 13.99 | 19.08 | | clouds | 45.62 | 62.99 | | counter | 21.32 | 51.32 | | cupboard | 3.01 | 14.62 | | curtain | 56.42 | 75.53 | | desk-stuff | 34.24 | 52.42 | | dirt | 36.55 | 56.4 | | door-stuff | 30.61 | 51.79 | | fence | 33.2 | 65.61 | | floor-marble | 1.15 | 1.21 | | floor-other | 19.39 | 29.18 | | floor-stone | 2.94 | 3.91 | | floor-tile | 55.03 | 65.06 | | floor-wood | 53.47 | 78.61 | | flower | 43.01 | 69.66 | | fog | 12.16 | 13.48 | | food-other | 29.23 | 48.85 | | fruit | 27.31 | 54.64 | | furniture-other | 10.57 | 15.14 | | grass | 65.9 | 82.79 | | gravel | 22.6 | 38.01 | | ground-other | 6.23 | 8.47 | | hill | 14.0 | 24.34 | | house | 24.93 | 30.88 | | leaves | 17.02 | 22.47 | | light | 30.98 | 50.91 | | mat | 2.77 | 5.69 | | metal | 13.48 | 14.75 | | mirror-stuff | 37.78 | 63.91 | | moss | 0.0 | 0.0 | | mountain | 54.76 | 70.7 | | mud | 3.82 | 8.34 | | napkin | 7.32 | 20.9 | | net | 31.58 | 58.94 | | paper | 24.67 | 34.87 | | pavement | 49.51 | 69.93 | | pillow | 10.1 | 15.96 | | plant-other | 20.11 | 29.63 | | plastic | 8.11 | 9.07 | | platform | 18.44 | 26.04 | | playingfield | 64.0 | 81.11 | | railing | 4.57 | 15.68 | | railroad | 49.42 | 78.35 | | river | 50.62 | 77.81 | | road | 62.92 | 78.5 | | rock | 44.2 | 69.33 | | roof | 34.15 | 48.39 | | rug | 33.01 | 44.6 | | salad | 11.17 | 15.56 | | sand | 57.21 | 64.8 | | sea | 84.37 | 92.03 | | shelf | 21.02 | 32.2 | | sky-other | 68.42 | 83.34 | | skyscraper | 21.13 | 26.52 | | snow | 87.62 | 94.74 | | solid-other | 0.0 | 0.0 | | stairs | 18.52 | 32.5 | | stone | 0.38 | 0.39 | | straw | 11.86 | 13.01 | | structural-other | 1.25 | 2.57 | | table | 16.05 | 21.42 | | tent | 8.56 | 13.95 | | textile-other | 9.44 | 10.44 | | towel | 22.44 | 35.0 | | tree | 71.53 | 83.52 | | vegetable | 39.87 | 53.83 | | wall-brick | 40.6 | 50.94 | | wall-concrete | 51.3 | 69.49 | | wall-other | 17.24 | 29.25 | | wall-panel | 0.7 | 0.85 | | wall-stone | 25.1 | 36.14 | | wall-tile | 62.16 | 78.22 | | wall-wood | 33.14 | 46.79 | | water-other | 22.02 | 27.25 | | waterdrops | 0.04 | 0.08 | | window-blind | 43.39 | 54.68 | | window-other | 42.0 | 73.5 | | wood | 20.13 | 25.29 | +------------------+-------+-------+ 2023/09/06 21:31:46 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7700 mIoU: 41.7700 mAcc: 56.9600 data_time: 0.0023 time: 0.1327 2023/09/06 21:31:50 - mmengine - INFO - The best checkpoint with 41.7700 mIoU at 55000 iter is saved to best_mIoU_iter_55000.pth. 2023/09/06 21:32:16 - mmengine - INFO - Iter(train) [55050/60000] base_lr: 8.2501e-06 lr: 8.2501e-06 eta: 0:37:28 time: 0.4559 data_time: 0.0239 memory: 15771 grad_norm: 17.9792 loss: 9.7939 decode.loss_cls_ce: 2.0976 decode.loss_mask_ce: 0.9219 decode.loss_mask_dice: 1.8712 decode.d7.loss_cls_ce: 2.1045 decode.d7.loss_mask_ce: 0.9209 decode.d7.loss_mask_dice: 1.8777 2023/09/06 21:32:39 - mmengine - INFO - Iter(train) [55100/60000] base_lr: 8.1668e-06 lr: 8.1668e-06 eta: 0:37:06 time: 0.4576 data_time: 0.0239 memory: 15831 grad_norm: 16.5241 loss: 9.7579 decode.loss_cls_ce: 2.0411 decode.loss_mask_ce: 1.0139 decode.loss_mask_dice: 1.8188 decode.d7.loss_cls_ce: 2.0411 decode.d7.loss_mask_ce: 1.0109 decode.d7.loss_mask_dice: 1.8321 2023/09/06 21:33:02 - mmengine - INFO - Iter(train) [55150/60000] base_lr: 8.0835e-06 lr: 8.0835e-06 eta: 0:36:43 time: 0.4587 data_time: 0.0248 memory: 16028 grad_norm: 18.4950 loss: 10.5020 decode.loss_cls_ce: 2.3583 decode.loss_mask_ce: 0.9141 decode.loss_mask_dice: 1.9611 decode.d7.loss_cls_ce: 2.3892 decode.d7.loss_mask_ce: 0.9144 decode.d7.loss_mask_dice: 1.9649 2023/09/06 21:33:24 - mmengine - INFO - Iter(train) [55200/60000] base_lr: 8.0001e-06 lr: 8.0001e-06 eta: 0:36:20 time: 0.4554 data_time: 0.0241 memory: 15774 grad_norm: nan loss: 8.5045 decode.loss_cls_ce: 1.8991 decode.loss_mask_ce: 0.8750 decode.loss_mask_dice: 1.4867 decode.d7.loss_cls_ce: 1.8721 decode.d7.loss_mask_ce: 0.8766 decode.d7.loss_mask_dice: 1.4950 2023/09/06 21:33:47 - mmengine - INFO - Iter(train) [55250/60000] base_lr: 7.9168e-06 lr: 7.9168e-06 eta: 0:35:57 time: 0.4562 data_time: 0.0242 memory: 15949 grad_norm: 16.9699 loss: 8.3985 decode.loss_cls_ce: 1.7645 decode.loss_mask_ce: 0.8119 decode.loss_mask_dice: 1.6084 decode.d7.loss_cls_ce: 1.7983 decode.d7.loss_mask_ce: 0.8140 decode.d7.loss_mask_dice: 1.6015 2023/09/06 21:34:10 - mmengine - INFO - Iter(train) [55300/60000] base_lr: 7.8335e-06 lr: 7.8335e-06 eta: 0:35:35 time: 0.4554 data_time: 0.0244 memory: 15835 grad_norm: 17.8025 loss: 9.8376 decode.loss_cls_ce: 2.0893 decode.loss_mask_ce: 0.9208 decode.loss_mask_dice: 1.9162 decode.d7.loss_cls_ce: 2.1055 decode.d7.loss_mask_ce: 0.9097 decode.d7.loss_mask_dice: 1.8960 2023/09/06 21:34:33 - mmengine - INFO - Iter(train) [55350/60000] base_lr: 7.7501e-06 lr: 7.7501e-06 eta: 0:35:12 time: 0.4563 data_time: 0.0220 memory: 15872 grad_norm: 17.7952 loss: 9.2068 decode.loss_cls_ce: 1.9955 decode.loss_mask_ce: 0.8647 decode.loss_mask_dice: 1.7452 decode.d7.loss_cls_ce: 1.9854 decode.d7.loss_mask_ce: 0.8753 decode.d7.loss_mask_dice: 1.7407 2023/09/06 21:34:55 - mmengine - INFO - Iter(train) [55400/60000] base_lr: 7.6668e-06 lr: 7.6668e-06 eta: 0:34:49 time: 0.4474 data_time: 0.0231 memory: 15897 grad_norm: 18.5115 loss: 8.1721 decode.loss_cls_ce: 1.8666 decode.loss_mask_ce: 0.7849 decode.loss_mask_dice: 1.4454 decode.d7.loss_cls_ce: 1.8655 decode.d7.loss_mask_ce: 0.7782 decode.d7.loss_mask_dice: 1.4314 2023/09/06 21:35:18 - mmengine - INFO - Iter(train) [55450/60000] base_lr: 7.5835e-06 lr: 7.5835e-06 eta: 0:34:27 time: 0.4548 data_time: 0.0220 memory: 15926 grad_norm: 18.6150 loss: 8.8661 decode.loss_cls_ce: 1.9458 decode.loss_mask_ce: 0.8455 decode.loss_mask_dice: 1.6457 decode.d7.loss_cls_ce: 1.9398 decode.d7.loss_mask_ce: 0.8453 decode.d7.loss_mask_dice: 1.6440 2023/09/06 21:35:41 - mmengine - INFO - Iter(train) [55500/60000] base_lr: 7.5001e-06 lr: 7.5001e-06 eta: 0:34:04 time: 0.4523 data_time: 0.0215 memory: 15818 grad_norm: 18.0633 loss: 8.9258 decode.loss_cls_ce: 1.9422 decode.loss_mask_ce: 0.8283 decode.loss_mask_dice: 1.6858 decode.d7.loss_cls_ce: 1.9548 decode.d7.loss_mask_ce: 0.8278 decode.d7.loss_mask_dice: 1.6868 2023/09/06 21:35:43 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:57 time: 0.0456 data_time: 0.0016 memory: 1528 2023/09/06 21:35:45 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0483 data_time: 0.0017 memory: 1441 2023/09/06 21:35:48 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0479 data_time: 0.0019 memory: 1595 2023/09/06 21:35:50 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0484 data_time: 0.0016 memory: 1550 2023/09/06 21:35:53 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0501 data_time: 0.0017 memory: 1574 2023/09/06 21:35:55 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0501 data_time: 0.0017 memory: 1462 2023/09/06 21:35:58 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:44 time: 0.0502 data_time: 0.0017 memory: 1528 2023/09/06 21:36:00 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0505 data_time: 0.0017 memory: 1528 2023/09/06 21:36:03 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0485 data_time: 0.0017 memory: 2187 2023/09/06 21:36:05 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0474 data_time: 0.0017 memory: 1528 2023/09/06 21:36:08 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0480 data_time: 0.0016 memory: 1550 2023/09/06 21:36:10 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0497 data_time: 0.0017 memory: 1528 2023/09/06 21:36:12 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0479 data_time: 0.0017 memory: 1528 2023/09/06 21:36:15 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:27 time: 0.0499 data_time: 0.0018 memory: 1727 2023/09/06 21:36:18 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0489 data_time: 0.0017 memory: 1815 2023/09/06 21:36:20 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0491 data_time: 0.0017 memory: 1484 2023/09/06 21:36:22 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0477 data_time: 0.0017 memory: 2361 2023/09/06 21:36:25 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0488 data_time: 0.0017 memory: 1420 2023/09/06 21:36:27 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0538 data_time: 0.0017 memory: 1705 2023/09/06 21:36:30 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0461 data_time: 0.0016 memory: 1574 2023/09/06 21:36:32 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0459 data_time: 0.0018 memory: 1484 2023/09/06 21:36:35 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0673 data_time: 0.0019 memory: 1574 2023/09/06 21:36:37 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0493 data_time: 0.0017 memory: 1683 2023/09/06 21:36:40 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0464 data_time: 0.0020 memory: 1528 2023/09/06 21:36:42 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0462 data_time: 0.0016 memory: 1528 2023/09/06 21:36:45 - mmengine - INFO - per class results: 2023/09/06 21:36:45 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.28 | 91.81 | | bicycle | 58.58 | 78.34 | | car | 60.54 | 79.6 | | motorcycle | 79.61 | 89.72 | | airplane | 72.36 | 89.02 | | bus | 77.73 | 89.24 | | train | 77.83 | 88.22 | | truck | 59.37 | 76.99 | | boat | 56.75 | 79.8 | | traffic light | 55.21 | 81.71 | | fire hydrant | 70.36 | 92.38 | | stop sign | 74.69 | 95.18 | | parking meter | 51.27 | 85.11 | | bench | 48.22 | 69.12 | | bird | 70.47 | 85.07 | | cat | 80.98 | 89.5 | | dog | 77.49 | 84.4 | | horse | 78.52 | 91.03 | | sheep | 86.05 | 93.73 | | cow | 80.9 | 87.78 | | elephant | 89.41 | 95.14 | | bear | 89.29 | 92.23 | | zebra | 90.1 | 94.06 | | giraffe | 82.55 | 90.49 | | backpack | 24.25 | 65.03 | | umbrella | 72.72 | 77.56 | | handbag | 22.72 | 35.22 | | tie | 9.24 | 28.82 | | suitcase | 71.95 | 81.67 | | frisbee | 59.73 | 87.26 | | skis | 30.46 | 61.09 | | snowboard | 55.89 | 71.84 | | sports ball | 53.81 | 71.82 | | kite | 50.41 | 72.02 | | baseball bat | 33.97 | 65.99 | | baseball glove | 51.79 | 88.14 | | skateboard | 48.11 | 82.97 | | surfboard | 74.95 | 88.46 | | tennis racket | 70.95 | 88.02 | | bottle | 44.92 | 69.67 | | wine glass | 40.44 | 65.32 | | cup | 40.59 | 58.33 | | fork | 35.44 | 54.21 | | knife | 21.59 | 26.06 | | spoon | 20.31 | 34.26 | | bowl | 31.55 | 41.71 | | banana | 61.4 | 82.97 | | apple | 40.82 | 51.88 | | sandwich | 44.33 | 59.67 | | orange | 66.78 | 76.06 | | broccoli | 52.4 | 67.68 | | carrot | 45.7 | 58.28 | | hot dog | 49.96 | 60.33 | | pizza | 65.98 | 76.55 | | donut | 64.77 | 82.32 | | cake | 71.41 | 82.98 | | chair | 42.84 | 66.15 | | couch | 53.79 | 72.86 | | potted plant | 23.29 | 32.16 | | bed | 61.43 | 80.91 | | dining table | 40.95 | 72.28 | | toilet | 69.79 | 91.19 | | tv | 68.67 | 80.86 | | laptop | 66.74 | 84.41 | | mouse | 60.5 | 69.31 | | remote | 39.62 | 72.16 | | keyboard | 57.29 | 70.89 | | cell phone | 67.49 | 82.37 | | microwave | 48.52 | 61.54 | | oven | 46.21 | 68.94 | | toaster | 41.11 | 80.13 | | sink | 40.07 | 80.9 | | refrigerator | 64.37 | 83.73 | | book | 42.32 | 60.48 | | clock | 71.75 | 86.74 | | vase | 42.0 | 81.72 | | scissors | 51.92 | 65.41 | | teddy bear | 77.54 | 84.72 | | hair drier | 15.02 | 52.98 | | toothbrush | 40.9 | 73.35 | | banner | 27.74 | 56.33 | | blanket | 10.54 | 15.16 | | branch | 12.71 | 27.02 | | bridge | 23.09 | 37.13 | | building-other | 51.85 | 71.99 | | bush | 32.04 | 50.08 | | cabinet | 43.67 | 63.21 | | cage | 13.75 | 19.63 | | cardboard | 34.08 | 48.87 | | carpet | 49.85 | 75.49 | | ceiling-other | 59.49 | 76.25 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.21 | 0.44 | | clothes | 14.45 | 19.92 | | clouds | 45.11 | 60.98 | | counter | 19.75 | 48.01 | | cupboard | 1.75 | 8.78 | | curtain | 56.93 | 74.56 | | desk-stuff | 32.43 | 49.27 | | dirt | 36.52 | 55.74 | | door-stuff | 30.94 | 51.97 | | fence | 32.5 | 65.84 | | floor-marble | 0.99 | 1.03 | | floor-other | 19.38 | 29.24 | | floor-stone | 2.86 | 3.84 | | floor-tile | 54.83 | 64.76 | | floor-wood | 52.91 | 78.8 | | flower | 42.24 | 66.9 | | fog | 13.48 | 15.35 | | food-other | 29.24 | 45.71 | | fruit | 27.25 | 53.18 | | furniture-other | 10.66 | 15.61 | | grass | 66.0 | 83.25 | | gravel | 24.09 | 41.15 | | ground-other | 6.25 | 8.61 | | hill | 13.81 | 23.85 | | house | 24.39 | 29.81 | | leaves | 17.74 | 23.82 | | light | 30.39 | 48.51 | | mat | 2.8 | 5.76 | | metal | 13.36 | 14.49 | | mirror-stuff | 37.82 | 61.43 | | moss | 0.0 | 0.0 | | mountain | 54.89 | 70.16 | | mud | 3.75 | 8.52 | | napkin | 6.2 | 16.1 | | net | 29.7 | 56.65 | | paper | 22.37 | 30.68 | | pavement | 49.53 | 70.78 | | pillow | 11.69 | 16.97 | | plant-other | 20.43 | 29.86 | | plastic | 7.63 | 8.5 | | platform | 19.43 | 27.55 | | playingfield | 65.08 | 83.28 | | railing | 4.7 | 16.46 | | railroad | 49.46 | 79.61 | | river | 49.44 | 75.25 | | road | 62.77 | 77.49 | | rock | 44.34 | 68.94 | | roof | 33.8 | 48.56 | | rug | 34.29 | 45.89 | | salad | 13.28 | 14.53 | | sand | 57.63 | 65.18 | | sea | 84.49 | 92.36 | | shelf | 21.62 | 32.92 | | sky-other | 68.67 | 84.28 | | skyscraper | 21.1 | 26.62 | | snow | 87.67 | 95.2 | | solid-other | 0.0 | 0.0 | | stairs | 18.7 | 32.67 | | stone | 0.59 | 0.6 | | straw | 11.71 | 12.86 | | structural-other | 1.14 | 2.25 | | table | 13.91 | 18.1 | | tent | 8.13 | 13.48 | | textile-other | 10.61 | 11.78 | | towel | 23.63 | 35.46 | | tree | 71.57 | 83.05 | | vegetable | 39.66 | 52.13 | | wall-brick | 41.77 | 51.52 | | wall-concrete | 51.42 | 68.68 | | wall-other | 16.76 | 29.18 | | wall-panel | 0.41 | 0.5 | | wall-stone | 26.18 | 37.2 | | wall-tile | 62.23 | 78.84 | | wall-wood | 32.92 | 46.73 | | water-other | 23.24 | 29.4 | | waterdrops | 0.01 | 0.02 | | window-blind | 42.92 | 54.89 | | window-other | 41.95 | 74.55 | | wood | 19.42 | 24.4 | +------------------+-------+-------+ 2023/09/06 21:36:45 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7000 mIoU: 41.6300 mAcc: 56.4500 data_time: 0.0018 time: 0.0491 2023/09/06 21:37:08 - mmengine - INFO - Iter(train) [55550/60000] base_lr: 7.4168e-06 lr: 7.4168e-06 eta: 0:33:41 time: 0.4555 data_time: 0.0218 memory: 15819 grad_norm: 20.2557 loss: 9.6495 decode.loss_cls_ce: 2.1044 decode.loss_mask_ce: 0.9102 decode.loss_mask_dice: 1.8137 decode.d7.loss_cls_ce: 2.1202 decode.d7.loss_mask_ce: 0.9080 decode.d7.loss_mask_dice: 1.7930 2023/09/06 21:37:30 - mmengine - INFO - Iter(train) [55600/60000] base_lr: 7.3335e-06 lr: 7.3335e-06 eta: 0:33:19 time: 0.4549 data_time: 0.0219 memory: 15899 grad_norm: 19.4500 loss: 9.6134 decode.loss_cls_ce: 2.0706 decode.loss_mask_ce: 0.9350 decode.loss_mask_dice: 1.7951 decode.d7.loss_cls_ce: 2.0749 decode.d7.loss_mask_ce: 0.9390 decode.d7.loss_mask_dice: 1.7987 2023/09/06 21:37:53 - mmengine - INFO - Iter(train) [55650/60000] base_lr: 7.2501e-06 lr: 7.2501e-06 eta: 0:32:56 time: 0.4538 data_time: 0.0228 memory: 15921 grad_norm: 18.1416 loss: 8.8738 decode.loss_cls_ce: 1.9281 decode.loss_mask_ce: 0.7992 decode.loss_mask_dice: 1.7030 decode.d7.loss_cls_ce: 1.9466 decode.d7.loss_mask_ce: 0.7888 decode.d7.loss_mask_dice: 1.7082 2023/09/06 21:38:15 - mmengine - INFO - Iter(train) [55700/60000] base_lr: 7.1668e-06 lr: 7.1668e-06 eta: 0:32:33 time: 0.4483 data_time: 0.0231 memory: 15924 grad_norm: 19.0396 loss: 8.5999 decode.loss_cls_ce: 1.8310 decode.loss_mask_ce: 0.8532 decode.loss_mask_dice: 1.6191 decode.d7.loss_cls_ce: 1.8134 decode.d7.loss_mask_ce: 0.8522 decode.d7.loss_mask_dice: 1.6310 2023/09/06 21:38:38 - mmengine - INFO - Iter(train) [55750/60000] base_lr: 7.0835e-06 lr: 7.0835e-06 eta: 0:32:10 time: 0.4521 data_time: 0.0233 memory: 15794 grad_norm: 19.8931 loss: 8.0280 decode.loss_cls_ce: 1.8248 decode.loss_mask_ce: 0.8075 decode.loss_mask_dice: 1.3998 decode.d7.loss_cls_ce: 1.7951 decode.d7.loss_mask_ce: 0.8018 decode.d7.loss_mask_dice: 1.3991 2023/09/06 21:39:01 - mmengine - INFO - Iter(train) [55800/60000] base_lr: 7.0001e-06 lr: 7.0001e-06 eta: 0:31:48 time: 0.4550 data_time: 0.0222 memory: 15974 grad_norm: 16.4420 loss: 9.7190 decode.loss_cls_ce: 2.1498 decode.loss_mask_ce: 0.9174 decode.loss_mask_dice: 1.7753 decode.d7.loss_cls_ce: 2.1665 decode.d7.loss_mask_ce: 0.9229 decode.d7.loss_mask_dice: 1.7872 2023/09/06 21:39:23 - mmengine - INFO - Iter(train) [55850/60000] base_lr: 6.9168e-06 lr: 6.9168e-06 eta: 0:31:25 time: 0.4542 data_time: 0.0231 memory: 15820 grad_norm: 19.7629 loss: 9.7094 decode.loss_cls_ce: 2.2295 decode.loss_mask_ce: 0.8257 decode.loss_mask_dice: 1.8056 decode.d7.loss_cls_ce: 2.2270 decode.d7.loss_mask_ce: 0.8128 decode.d7.loss_mask_dice: 1.8089 2023/09/06 21:39:46 - mmengine - INFO - Iter(train) [55900/60000] base_lr: 6.8334e-06 lr: 6.8334e-06 eta: 0:31:02 time: 0.4538 data_time: 0.0224 memory: 15783 grad_norm: 18.3884 loss: 9.0867 decode.loss_cls_ce: 2.0482 decode.loss_mask_ce: 0.8392 decode.loss_mask_dice: 1.6704 decode.d7.loss_cls_ce: 2.0264 decode.d7.loss_mask_ce: 0.8437 decode.d7.loss_mask_dice: 1.6586 2023/09/06 21:40:09 - mmengine - INFO - Iter(train) [55950/60000] base_lr: 6.7501e-06 lr: 6.7501e-06 eta: 0:30:40 time: 0.4532 data_time: 0.0222 memory: 15882 grad_norm: 17.3541 loss: 9.8190 decode.loss_cls_ce: 2.1183 decode.loss_mask_ce: 0.8876 decode.loss_mask_dice: 1.8946 decode.d7.loss_cls_ce: 2.1283 decode.d7.loss_mask_ce: 0.8931 decode.d7.loss_mask_dice: 1.8970 2023/09/06 21:40:31 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:40:31 - mmengine - INFO - Iter(train) [56000/60000] base_lr: 6.6668e-06 lr: 6.6668e-06 eta: 0:30:17 time: 0.4516 data_time: 0.0224 memory: 15806 grad_norm: 23.5986 loss: 8.8934 decode.loss_cls_ce: 1.8119 decode.loss_mask_ce: 0.9147 decode.loss_mask_dice: 1.7208 decode.d7.loss_cls_ce: 1.7961 decode.d7.loss_mask_ce: 0.9169 decode.d7.loss_mask_dice: 1.7330 2023/09/06 21:40:34 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0457 data_time: 0.0019 memory: 1528 2023/09/06 21:40:36 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0470 data_time: 0.0017 memory: 1441 2023/09/06 21:40:39 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0462 data_time: 0.0016 memory: 1595 2023/09/06 21:40:41 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:50 time: 0.0469 data_time: 0.0016 memory: 1550 2023/09/06 21:40:43 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0492 data_time: 0.0016 memory: 1574 2023/09/06 21:40:46 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:45 time: 0.0491 data_time: 0.0016 memory: 1462 2023/09/06 21:40:48 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0496 data_time: 0.0017 memory: 1528 2023/09/06 21:40:50 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:40 time: 0.0490 data_time: 0.0016 memory: 1528 2023/09/06 21:40:53 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:38 time: 0.0471 data_time: 0.0017 memory: 2187 2023/09/06 21:40:55 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0462 data_time: 0.0017 memory: 1528 2023/09/06 21:40:58 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:33 time: 0.0466 data_time: 0.0016 memory: 1550 2023/09/06 21:41:00 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0492 data_time: 0.0017 memory: 1528 2023/09/06 21:41:03 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:28 time: 0.0466 data_time: 0.0017 memory: 1528 2023/09/06 21:41:05 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0483 data_time: 0.0017 memory: 1727 2023/09/06 21:41:07 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0484 data_time: 0.0016 memory: 1815 2023/09/06 21:41:10 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:21 time: 0.0494 data_time: 0.0018 memory: 1484 2023/09/06 21:41:12 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0471 data_time: 0.0017 memory: 2361 2023/09/06 21:41:15 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:16 time: 0.0484 data_time: 0.0016 memory: 1420 2023/09/06 21:41:17 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0535 data_time: 0.0017 memory: 1705 2023/09/06 21:41:20 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0461 data_time: 0.0016 memory: 1574 2023/09/06 21:41:22 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0455 data_time: 0.0016 memory: 1484 2023/09/06 21:41:24 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0468 data_time: 0.0017 memory: 1574 2023/09/06 21:41:27 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0493 data_time: 0.0017 memory: 1683 2023/09/06 21:41:29 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0478 data_time: 0.0016 memory: 1528 2023/09/06 21:41:32 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0478 data_time: 0.0016 memory: 1528 2023/09/06 21:41:35 - mmengine - INFO - per class results: 2023/09/06 21:41:35 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.24 | 91.64 | | bicycle | 59.0 | 78.89 | | car | 60.7 | 79.82 | | motorcycle | 79.17 | 88.95 | | airplane | 71.68 | 88.65 | | bus | 77.73 | 88.82 | | train | 77.3 | 87.89 | | truck | 59.65 | 77.37 | | boat | 56.8 | 79.62 | | traffic light | 54.61 | 81.24 | | fire hydrant | 72.27 | 92.66 | | stop sign | 77.36 | 94.73 | | parking meter | 51.82 | 83.63 | | bench | 48.52 | 68.87 | | bird | 70.47 | 86.52 | | cat | 80.68 | 89.71 | | dog | 77.94 | 84.37 | | horse | 79.43 | 91.0 | | sheep | 85.95 | 93.8 | | cow | 83.79 | 91.17 | | elephant | 90.55 | 95.2 | | bear | 88.6 | 91.61 | | zebra | 90.11 | 94.06 | | giraffe | 82.59 | 90.45 | | backpack | 23.81 | 65.32 | | umbrella | 72.5 | 77.31 | | handbag | 23.61 | 36.8 | | tie | 10.64 | 31.39 | | suitcase | 71.15 | 81.32 | | frisbee | 61.15 | 87.71 | | skis | 30.66 | 62.33 | | snowboard | 52.65 | 69.48 | | sports ball | 53.32 | 71.33 | | kite | 50.42 | 72.08 | | baseball bat | 33.15 | 66.93 | | baseball glove | 50.07 | 88.96 | | skateboard | 46.79 | 82.66 | | surfboard | 75.13 | 88.59 | | tennis racket | 70.42 | 88.03 | | bottle | 45.61 | 71.75 | | wine glass | 40.88 | 66.75 | | cup | 41.22 | 59.91 | | fork | 36.16 | 56.57 | | knife | 21.89 | 27.13 | | spoon | 16.51 | 34.32 | | bowl | 30.95 | 40.57 | | banana | 61.42 | 82.71 | | apple | 41.22 | 52.06 | | sandwich | 44.78 | 60.33 | | orange | 65.99 | 75.14 | | broccoli | 52.57 | 66.86 | | carrot | 45.89 | 57.14 | | hot dog | 50.09 | 60.38 | | pizza | 69.1 | 80.5 | | donut | 64.14 | 81.59 | | cake | 71.37 | 83.12 | | chair | 42.61 | 65.58 | | couch | 54.2 | 72.81 | | potted plant | 22.73 | 31.85 | | bed | 61.3 | 81.25 | | dining table | 41.71 | 70.83 | | toilet | 69.29 | 91.44 | | tv | 67.71 | 80.53 | | laptop | 65.55 | 82.91 | | mouse | 58.05 | 67.85 | | remote | 38.04 | 69.28 | | keyboard | 58.05 | 71.18 | | cell phone | 66.84 | 82.01 | | microwave | 47.68 | 61.46 | | oven | 49.15 | 71.56 | | toaster | 64.06 | 79.39 | | sink | 40.15 | 81.42 | | refrigerator | 64.89 | 83.72 | | book | 42.59 | 60.99 | | clock | 72.75 | 86.16 | | vase | 42.75 | 80.88 | | scissors | 57.83 | 73.42 | | teddy bear | 76.71 | 84.38 | | hair drier | 15.31 | 54.28 | | toothbrush | 42.62 | 75.67 | | banner | 29.35 | 59.57 | | blanket | 11.64 | 16.69 | | branch | 12.17 | 26.83 | | bridge | 23.65 | 37.09 | | building-other | 51.76 | 71.43 | | bush | 32.56 | 50.23 | | cabinet | 43.26 | 63.73 | | cage | 15.81 | 23.31 | | cardboard | 34.0 | 48.65 | | carpet | 50.22 | 76.28 | | ceiling-other | 58.91 | 76.28 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.09 | 0.2 | | clothes | 14.52 | 20.09 | | clouds | 44.36 | 59.82 | | counter | 20.49 | 50.41 | | cupboard | 1.87 | 9.16 | | curtain | 56.33 | 73.83 | | desk-stuff | 33.88 | 51.98 | | dirt | 36.29 | 56.25 | | door-stuff | 30.53 | 50.56 | | fence | 32.82 | 64.84 | | floor-marble | 5.45 | 5.83 | | floor-other | 19.28 | 28.32 | | floor-stone | 2.91 | 3.8 | | floor-tile | 55.32 | 65.34 | | floor-wood | 53.88 | 78.45 | | flower | 42.83 | 69.04 | | fog | 12.61 | 14.0 | | food-other | 29.39 | 49.68 | | fruit | 27.56 | 56.22 | | furniture-other | 9.95 | 14.17 | | grass | 65.95 | 82.91 | | gravel | 23.05 | 38.9 | | ground-other | 6.27 | 8.5 | | hill | 13.81 | 24.66 | | house | 24.15 | 28.84 | | leaves | 17.64 | 23.74 | | light | 30.47 | 48.34 | | mat | 2.37 | 4.88 | | metal | 12.5 | 13.54 | | mirror-stuff | 36.76 | 64.89 | | moss | 0.0 | 0.0 | | mountain | 54.3 | 71.36 | | mud | 3.91 | 8.5 | | napkin | 7.49 | 18.94 | | net | 30.22 | 56.6 | | paper | 23.49 | 32.68 | | pavement | 49.6 | 71.25 | | pillow | 13.73 | 20.76 | | plant-other | 20.26 | 30.01 | | plastic | 7.66 | 8.6 | | platform | 18.39 | 25.68 | | playingfield | 63.58 | 80.72 | | railing | 4.91 | 16.52 | | railroad | 49.39 | 78.78 | | river | 49.75 | 75.61 | | road | 62.45 | 77.24 | | rock | 43.94 | 68.83 | | roof | 33.39 | 50.87 | | rug | 33.18 | 43.65 | | salad | 11.51 | 16.4 | | sand | 57.29 | 64.6 | | sea | 84.93 | 92.84 | | shelf | 21.67 | 33.78 | | sky-other | 68.61 | 84.72 | | skyscraper | 22.5 | 29.53 | | snow | 87.71 | 94.84 | | solid-other | 0.0 | 0.0 | | stairs | 19.31 | 34.2 | | stone | 1.05 | 1.08 | | straw | 11.78 | 12.92 | | structural-other | 1.1 | 2.24 | | table | 14.96 | 19.62 | | tent | 8.13 | 13.63 | | textile-other | 10.54 | 11.56 | | towel | 23.53 | 35.57 | | tree | 71.47 | 83.45 | | vegetable | 39.29 | 54.35 | | wall-brick | 41.11 | 50.42 | | wall-concrete | 51.1 | 69.08 | | wall-other | 16.81 | 29.69 | | wall-panel | 0.23 | 0.27 | | wall-stone | 24.14 | 34.56 | | wall-tile | 61.8 | 77.82 | | wall-wood | 33.97 | 47.69 | | water-other | 22.25 | 28.11 | | waterdrops | 0.0 | 0.0 | | window-blind | 42.96 | 56.19 | | window-other | 42.16 | 73.86 | | wood | 19.21 | 24.15 | +------------------+-------+-------+ 2023/09/06 21:41:35 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7000 mIoU: 41.8500 mAcc: 56.7500 data_time: 0.0017 time: 0.0482 2023/09/06 21:41:35 - mmengine - INFO - The previous best checkpoint /home/caoanqi/mmsegmentation/work_dirs/train_60k_amp_fixed/best_mIoU_iter_55000.pth is removed 2023/09/06 21:41:38 - mmengine - INFO - The best checkpoint with 41.8500 mIoU at 56000 iter is saved to best_mIoU_iter_56000.pth. 2023/09/06 21:42:04 - mmengine - INFO - Iter(train) [56050/60000] base_lr: 6.5834e-06 lr: 6.5834e-06 eta: 0:29:55 time: 0.4542 data_time: 0.0227 memory: 15767 grad_norm: 17.0493 loss: 9.3232 decode.loss_cls_ce: 1.9212 decode.loss_mask_ce: 0.9889 decode.loss_mask_dice: 1.7389 decode.d7.loss_cls_ce: 1.9373 decode.d7.loss_mask_ce: 0.9918 decode.d7.loss_mask_dice: 1.7451 2023/09/06 21:42:26 - mmengine - INFO - Iter(train) [56100/60000] base_lr: 6.5001e-06 lr: 6.5001e-06 eta: 0:29:32 time: 0.4575 data_time: 0.0226 memory: 15746 grad_norm: 20.0242 loss: 9.6874 decode.loss_cls_ce: 2.1288 decode.loss_mask_ce: 0.8666 decode.loss_mask_dice: 1.8296 decode.d7.loss_cls_ce: 2.1653 decode.d7.loss_mask_ce: 0.8670 decode.d7.loss_mask_dice: 1.8302 2023/09/06 21:42:49 - mmengine - INFO - Iter(train) [56150/60000] base_lr: 6.4168e-06 lr: 6.4168e-06 eta: 0:29:09 time: 0.4535 data_time: 0.0220 memory: 15846 grad_norm: 18.5571 loss: 9.4954 decode.loss_cls_ce: 1.9679 decode.loss_mask_ce: 0.9126 decode.loss_mask_dice: 1.8543 decode.d7.loss_cls_ce: 1.9605 decode.d7.loss_mask_ce: 0.9290 decode.d7.loss_mask_dice: 1.8710 2023/09/06 21:43:12 - mmengine - INFO - Iter(train) [56200/60000] base_lr: 6.3334e-06 lr: 6.3334e-06 eta: 0:28:47 time: 0.4519 data_time: 0.0224 memory: 15975 grad_norm: 18.6561 loss: 9.3097 decode.loss_cls_ce: 2.0781 decode.loss_mask_ce: 0.9036 decode.loss_mask_dice: 1.6655 decode.d7.loss_cls_ce: 2.0647 decode.d7.loss_mask_ce: 0.9173 decode.d7.loss_mask_dice: 1.6805 2023/09/06 21:43:34 - mmengine - INFO - Iter(train) [56250/60000] base_lr: 6.2501e-06 lr: 6.2501e-06 eta: 0:28:24 time: 0.4544 data_time: 0.0225 memory: 15834 grad_norm: 18.9894 loss: 9.5356 decode.loss_cls_ce: 2.1169 decode.loss_mask_ce: 0.8429 decode.loss_mask_dice: 1.7901 decode.d7.loss_cls_ce: 2.1518 decode.d7.loss_mask_ce: 0.8448 decode.d7.loss_mask_dice: 1.7890 2023/09/06 21:43:57 - mmengine - INFO - Iter(train) [56300/60000] base_lr: 6.1668e-06 lr: 6.1668e-06 eta: 0:28:01 time: 0.4516 data_time: 0.0218 memory: 15949 grad_norm: 21.1207 loss: 8.6962 decode.loss_cls_ce: 1.8508 decode.loss_mask_ce: 0.8428 decode.loss_mask_dice: 1.6681 decode.d7.loss_cls_ce: 1.8176 decode.d7.loss_mask_ce: 0.8408 decode.d7.loss_mask_dice: 1.6760 2023/09/06 21:44:19 - mmengine - INFO - Iter(train) [56350/60000] base_lr: 6.0834e-06 lr: 6.0834e-06 eta: 0:27:38 time: 0.4545 data_time: 0.0225 memory: 15820 grad_norm: 19.0487 loss: 7.6781 decode.loss_cls_ce: 1.6456 decode.loss_mask_ce: 0.7846 decode.loss_mask_dice: 1.3957 decode.d7.loss_cls_ce: 1.6659 decode.d7.loss_mask_ce: 0.7882 decode.d7.loss_mask_dice: 1.3981 2023/09/06 21:44:42 - mmengine - INFO - Iter(train) [56400/60000] base_lr: 6.0001e-06 lr: 6.0001e-06 eta: 0:27:16 time: 0.4530 data_time: 0.0227 memory: 15898 grad_norm: 20.5115 loss: 9.4126 decode.loss_cls_ce: 2.1551 decode.loss_mask_ce: 0.8405 decode.loss_mask_dice: 1.7049 decode.d7.loss_cls_ce: 2.1649 decode.d7.loss_mask_ce: 0.8426 decode.d7.loss_mask_dice: 1.7046 2023/09/06 21:45:05 - mmengine - INFO - Iter(train) [56450/60000] base_lr: 5.9168e-06 lr: 5.9168e-06 eta: 0:26:53 time: 0.4521 data_time: 0.0227 memory: 15961 grad_norm: 21.9784 loss: 9.3357 decode.loss_cls_ce: 2.0135 decode.loss_mask_ce: 0.8404 decode.loss_mask_dice: 1.8035 decode.d7.loss_cls_ce: 2.0317 decode.d7.loss_mask_ce: 0.8435 decode.d7.loss_mask_dice: 1.8030 2023/09/06 21:45:27 - mmengine - INFO - Iter(train) [56500/60000] base_lr: 5.8334e-06 lr: 5.8334e-06 eta: 0:26:30 time: 0.4514 data_time: 0.0229 memory: 16145 grad_norm: 18.9109 loss: 8.9531 decode.loss_cls_ce: 1.9462 decode.loss_mask_ce: 0.8594 decode.loss_mask_dice: 1.6684 decode.d7.loss_cls_ce: 1.9614 decode.d7.loss_mask_ce: 0.8585 decode.d7.loss_mask_dice: 1.6591 2023/09/06 21:45:30 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0461 data_time: 0.0018 memory: 1528 2023/09/06 21:45:32 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0470 data_time: 0.0017 memory: 1441 2023/09/06 21:45:35 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0468 data_time: 0.0017 memory: 1595 2023/09/06 21:45:37 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:50 time: 0.0472 data_time: 0.0017 memory: 1550 2023/09/06 21:45:39 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0501 data_time: 0.0019 memory: 1574 2023/09/06 21:45:42 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:45 time: 0.0499 data_time: 0.0021 memory: 1462 2023/09/06 21:45:44 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0497 data_time: 0.0019 memory: 1528 2023/09/06 21:45:47 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0496 data_time: 0.0018 memory: 1528 2023/09/06 21:45:49 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:38 time: 0.0475 data_time: 0.0020 memory: 2187 2023/09/06 21:45:52 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0466 data_time: 0.0017 memory: 1528 2023/09/06 21:45:54 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:33 time: 0.0468 data_time: 0.0018 memory: 1550 2023/09/06 21:45:56 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0491 data_time: 0.0018 memory: 1528 2023/09/06 21:45:59 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0467 data_time: 0.0017 memory: 1528 2023/09/06 21:46:01 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0491 data_time: 0.0019 memory: 1727 2023/09/06 21:46:04 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0484 data_time: 0.0018 memory: 1815 2023/09/06 21:46:06 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:21 time: 0.0495 data_time: 0.0021 memory: 1484 2023/09/06 21:46:09 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0473 data_time: 0.0018 memory: 2361 2023/09/06 21:46:11 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0491 data_time: 0.0018 memory: 1420 2023/09/06 21:46:14 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0544 data_time: 0.0019 memory: 1705 2023/09/06 21:46:16 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0467 data_time: 0.0018 memory: 1574 2023/09/06 21:46:18 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0466 data_time: 0.0019 memory: 1484 2023/09/06 21:46:21 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0480 data_time: 0.0019 memory: 1574 2023/09/06 21:46:23 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0500 data_time: 0.0018 memory: 1683 2023/09/06 21:46:26 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0467 data_time: 0.0018 memory: 1528 2023/09/06 21:46:28 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0467 data_time: 0.0018 memory: 1528 2023/09/06 21:46:31 - mmengine - INFO - per class results: 2023/09/06 21:46:31 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.26 | 91.72 | | bicycle | 59.88 | 80.89 | | car | 60.34 | 79.72 | | motorcycle | 79.22 | 89.4 | | airplane | 71.89 | 88.64 | | bus | 77.58 | 88.45 | | train | 77.27 | 87.93 | | truck | 58.75 | 77.31 | | boat | 58.12 | 82.75 | | traffic light | 56.08 | 80.87 | | fire hydrant | 70.33 | 92.92 | | stop sign | 78.53 | 95.53 | | parking meter | 50.6 | 83.57 | | bench | 48.64 | 69.81 | | bird | 72.85 | 85.52 | | cat | 80.79 | 89.07 | | dog | 76.95 | 84.02 | | horse | 79.34 | 90.74 | | sheep | 85.93 | 93.7 | | cow | 83.84 | 91.18 | | elephant | 90.61 | 95.24 | | bear | 89.4 | 92.28 | | zebra | 90.11 | 93.99 | | giraffe | 82.6 | 90.42 | | backpack | 24.47 | 63.94 | | umbrella | 72.81 | 77.66 | | handbag | 23.04 | 36.44 | | tie | 12.1 | 29.41 | | suitcase | 71.22 | 81.98 | | frisbee | 60.82 | 87.83 | | skis | 31.14 | 61.25 | | snowboard | 55.9 | 72.21 | | sports ball | 53.27 | 72.41 | | kite | 50.18 | 71.83 | | baseball bat | 33.32 | 65.74 | | baseball glove | 51.63 | 88.46 | | skateboard | 47.2 | 82.03 | | surfboard | 75.52 | 88.61 | | tennis racket | 70.59 | 87.85 | | bottle | 45.58 | 71.78 | | wine glass | 39.73 | 66.03 | | cup | 40.99 | 59.81 | | fork | 35.78 | 55.77 | | knife | 22.47 | 27.18 | | spoon | 16.41 | 34.27 | | bowl | 31.15 | 41.99 | | banana | 61.98 | 82.71 | | apple | 42.21 | 53.77 | | sandwich | 44.51 | 60.21 | | orange | 65.66 | 74.21 | | broccoli | 52.78 | 67.26 | | carrot | 45.59 | 57.93 | | hot dog | 49.19 | 59.36 | | pizza | 67.27 | 78.54 | | donut | 65.04 | 83.09 | | cake | 70.8 | 83.25 | | chair | 42.75 | 65.4 | | couch | 54.12 | 73.41 | | potted plant | 23.3 | 32.68 | | bed | 61.06 | 81.32 | | dining table | 41.95 | 70.69 | | toilet | 69.13 | 91.62 | | tv | 68.09 | 79.93 | | laptop | 65.89 | 84.03 | | mouse | 58.63 | 67.78 | | remote | 38.32 | 69.99 | | keyboard | 56.51 | 71.48 | | cell phone | 66.99 | 82.24 | | microwave | 45.25 | 61.9 | | oven | 48.46 | 71.45 | | toaster | 68.14 | 79.68 | | sink | 39.92 | 81.9 | | refrigerator | 63.84 | 84.79 | | book | 43.25 | 63.58 | | clock | 72.33 | 86.81 | | vase | 41.17 | 81.32 | | scissors | 63.11 | 79.63 | | teddy bear | 76.86 | 83.81 | | hair drier | 14.93 | 52.72 | | toothbrush | 40.34 | 75.19 | | banner | 27.95 | 57.11 | | blanket | 11.46 | 16.67 | | branch | 12.35 | 28.48 | | bridge | 23.13 | 36.32 | | building-other | 51.64 | 72.17 | | bush | 32.67 | 49.27 | | cabinet | 44.13 | 64.64 | | cage | 14.0 | 19.96 | | cardboard | 34.48 | 50.06 | | carpet | 49.68 | 74.94 | | ceiling-other | 59.51 | 76.96 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.16 | 0.35 | | clothes | 14.38 | 20.16 | | clouds | 44.39 | 59.11 | | counter | 21.3 | 51.09 | | cupboard | 2.61 | 13.1 | | curtain | 56.4 | 74.63 | | desk-stuff | 33.92 | 51.65 | | dirt | 36.71 | 56.98 | | door-stuff | 30.66 | 51.02 | | fence | 33.12 | 65.37 | | floor-marble | 0.35 | 0.36 | | floor-other | 19.26 | 28.73 | | floor-stone | 2.84 | 3.77 | | floor-tile | 55.69 | 65.82 | | floor-wood | 53.77 | 78.78 | | flower | 43.55 | 68.36 | | fog | 12.93 | 14.2 | | food-other | 29.25 | 48.59 | | fruit | 28.0 | 56.29 | | furniture-other | 10.32 | 14.61 | | grass | 66.19 | 83.26 | | gravel | 23.38 | 37.93 | | ground-other | 6.2 | 8.25 | | hill | 14.35 | 25.9 | | house | 24.65 | 30.34 | | leaves | 17.71 | 25.11 | | light | 30.67 | 48.75 | | mat | 2.53 | 4.98 | | metal | 12.96 | 14.08 | | mirror-stuff | 37.39 | 62.88 | | moss | 0.0 | 0.0 | | mountain | 54.39 | 70.73 | | mud | 3.72 | 8.48 | | napkin | 8.0 | 21.42 | | net | 31.08 | 59.24 | | paper | 26.8 | 37.36 | | pavement | 49.78 | 72.24 | | pillow | 10.97 | 16.14 | | plant-other | 20.28 | 30.27 | | plastic | 7.8 | 8.62 | | platform | 18.34 | 25.72 | | playingfield | 64.86 | 81.93 | | railing | 4.78 | 15.62 | | railroad | 49.47 | 79.36 | | river | 48.77 | 75.41 | | road | 61.8 | 75.68 | | rock | 43.9 | 68.77 | | roof | 33.47 | 48.98 | | rug | 33.55 | 44.93 | | salad | 11.81 | 15.52 | | sand | 57.65 | 65.47 | | sea | 84.3 | 92.01 | | shelf | 21.56 | 33.44 | | sky-other | 68.97 | 85.23 | | skyscraper | 21.29 | 27.37 | | snow | 87.7 | 95.07 | | solid-other | 0.0 | 0.0 | | stairs | 19.93 | 35.34 | | stone | 0.84 | 0.86 | | straw | 11.52 | 12.62 | | structural-other | 1.13 | 2.27 | | table | 16.09 | 21.07 | | tent | 8.26 | 13.6 | | textile-other | 10.52 | 11.66 | | towel | 23.74 | 35.35 | | tree | 71.82 | 83.2 | | vegetable | 39.15 | 52.92 | | wall-brick | 41.03 | 50.55 | | wall-concrete | 51.59 | 69.63 | | wall-other | 16.82 | 28.58 | | wall-panel | 0.23 | 0.29 | | wall-stone | 25.47 | 35.91 | | wall-tile | 62.12 | 77.94 | | wall-wood | 33.12 | 46.58 | | water-other | 20.84 | 26.65 | | waterdrops | 0.03 | 0.07 | | window-blind | 43.54 | 56.65 | | window-other | 42.31 | 73.33 | | wood | 19.8 | 24.93 | +------------------+-------+-------+ 2023/09/06 21:46:31 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.8400 mIoU: 41.9300 mAcc: 56.8400 data_time: 0.0018 time: 0.0485 2023/09/06 21:46:31 - mmengine - INFO - The previous best checkpoint /home/caoanqi/mmsegmentation/work_dirs/train_60k_amp_fixed/best_mIoU_iter_56000.pth is removed 2023/09/06 21:46:34 - mmengine - INFO - The best checkpoint with 41.9300 mIoU at 56500 iter is saved to best_mIoU_iter_56500.pth. 2023/09/06 21:47:00 - mmengine - INFO - Iter(train) [56550/60000] base_lr: 5.7501e-06 lr: 5.7501e-06 eta: 0:26:08 time: 0.4523 data_time: 0.0230 memory: 15951 grad_norm: 18.9146 loss: 9.6504 decode.loss_cls_ce: 2.0395 decode.loss_mask_ce: 0.8588 decode.loss_mask_dice: 1.9241 decode.d7.loss_cls_ce: 2.0353 decode.d7.loss_mask_ce: 0.8624 decode.d7.loss_mask_dice: 1.9303 2023/09/06 21:47:23 - mmengine - INFO - Iter(train) [56600/60000] base_lr: 5.6668e-06 lr: 5.6668e-06 eta: 0:25:45 time: 0.4537 data_time: 0.0227 memory: 15796 grad_norm: 16.4600 loss: 8.9875 decode.loss_cls_ce: 1.9022 decode.loss_mask_ce: 0.8840 decode.loss_mask_dice: 1.6975 decode.d7.loss_cls_ce: 1.9400 decode.d7.loss_mask_ce: 0.8702 decode.d7.loss_mask_dice: 1.6936 2023/09/06 21:47:46 - mmengine - INFO - Iter(train) [56650/60000] base_lr: 5.5834e-06 lr: 5.5834e-06 eta: 0:25:23 time: 0.4575 data_time: 0.0234 memory: 15834 grad_norm: 20.4904 loss: 8.2773 decode.loss_cls_ce: 1.7846 decode.loss_mask_ce: 0.7577 decode.loss_mask_dice: 1.5855 decode.d7.loss_cls_ce: 1.7889 decode.d7.loss_mask_ce: 0.7638 decode.d7.loss_mask_dice: 1.5969 2023/09/06 21:48:09 - mmengine - INFO - Iter(train) [56700/60000] base_lr: 5.5001e-06 lr: 5.5001e-06 eta: 0:25:00 time: 0.4525 data_time: 0.0243 memory: 15975 grad_norm: 18.2823 loss: 7.7351 decode.loss_cls_ce: 1.7200 decode.loss_mask_ce: 0.7511 decode.loss_mask_dice: 1.4044 decode.d7.loss_cls_ce: 1.7099 decode.d7.loss_mask_ce: 0.7466 decode.d7.loss_mask_dice: 1.4029 2023/09/06 21:48:31 - mmengine - INFO - Iter(train) [56750/60000] base_lr: 5.4168e-06 lr: 5.4168e-06 eta: 0:24:37 time: 0.4536 data_time: 0.0230 memory: 15927 grad_norm: 18.5916 loss: 10.0569 decode.loss_cls_ce: 2.1847 decode.loss_mask_ce: 0.9068 decode.loss_mask_dice: 1.9387 decode.d7.loss_cls_ce: 2.1765 decode.d7.loss_mask_ce: 0.9068 decode.d7.loss_mask_dice: 1.9435 2023/09/06 21:48:54 - mmengine - INFO - Iter(train) [56800/60000] base_lr: 5.3334e-06 lr: 5.3334e-06 eta: 0:24:14 time: 0.4498 data_time: 0.0232 memory: 15806 grad_norm: 16.7837 loss: 9.2176 decode.loss_cls_ce: 2.1054 decode.loss_mask_ce: 0.8323 decode.loss_mask_dice: 1.6819 decode.d7.loss_cls_ce: 2.0821 decode.d7.loss_mask_ce: 0.8356 decode.d7.loss_mask_dice: 1.6804 2023/09/06 21:49:16 - mmengine - INFO - Iter(train) [56850/60000] base_lr: 5.2501e-06 lr: 5.2501e-06 eta: 0:23:52 time: 0.4515 data_time: 0.0231 memory: 15820 grad_norm: 17.6645 loss: 9.4230 decode.loss_cls_ce: 2.0918 decode.loss_mask_ce: 0.8424 decode.loss_mask_dice: 1.7727 decode.d7.loss_cls_ce: 2.1233 decode.d7.loss_mask_ce: 0.8416 decode.d7.loss_mask_dice: 1.7512 2023/09/06 21:49:39 - mmengine - INFO - Iter(train) [56900/60000] base_lr: 5.1668e-06 lr: 5.1668e-06 eta: 0:23:29 time: 0.4553 data_time: 0.0235 memory: 15759 grad_norm: 16.8232 loss: 8.7481 decode.loss_cls_ce: 1.9169 decode.loss_mask_ce: 0.8725 decode.loss_mask_dice: 1.6032 decode.d7.loss_cls_ce: 1.8902 decode.d7.loss_mask_ce: 0.8753 decode.d7.loss_mask_dice: 1.5900 2023/09/06 21:50:02 - mmengine - INFO - Iter(train) [56950/60000] base_lr: 5.0834e-06 lr: 5.0834e-06 eta: 0:23:06 time: 0.4554 data_time: 0.0238 memory: 15820 grad_norm: 18.7604 loss: 9.5213 decode.loss_cls_ce: 2.1958 decode.loss_mask_ce: 0.7703 decode.loss_mask_dice: 1.7813 decode.d7.loss_cls_ce: 2.2289 decode.d7.loss_mask_ce: 0.7699 decode.d7.loss_mask_dice: 1.7751 2023/09/06 21:50:25 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 21:50:25 - mmengine - INFO - Iter(train) [57000/60000] base_lr: 5.0001e-06 lr: 5.0001e-06 eta: 0:22:43 time: 0.4575 data_time: 0.0227 memory: 15900 grad_norm: 18.9834 loss: 8.7693 decode.loss_cls_ce: 1.8906 decode.loss_mask_ce: 0.8459 decode.loss_mask_dice: 1.6455 decode.d7.loss_cls_ce: 1.8980 decode.d7.loss_mask_ce: 0.8440 decode.d7.loss_mask_dice: 1.6453 2023/09/06 21:50:27 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0457 data_time: 0.0017 memory: 1528 2023/09/06 21:50:30 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0473 data_time: 0.0018 memory: 1441 2023/09/06 21:50:32 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0472 data_time: 0.0019 memory: 1595 2023/09/06 21:50:34 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:50 time: 0.0474 data_time: 0.0018 memory: 1550 2023/09/06 21:50:37 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0498 data_time: 0.0019 memory: 1574 2023/09/06 21:50:39 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0498 data_time: 0.0018 memory: 1462 2023/09/06 21:50:42 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0502 data_time: 0.0019 memory: 1528 2023/09/06 21:50:44 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0501 data_time: 0.0019 memory: 1528 2023/09/06 21:50:47 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0481 data_time: 0.0019 memory: 2187 2023/09/06 21:50:49 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0466 data_time: 0.0019 memory: 1528 2023/09/06 21:50:52 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0472 data_time: 0.0017 memory: 1550 2023/09/06 21:50:54 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0499 data_time: 0.0018 memory: 1528 2023/09/06 21:50:56 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0471 data_time: 0.0018 memory: 1528 2023/09/06 21:50:59 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0486 data_time: 0.0017 memory: 1727 2023/09/06 21:51:01 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0487 data_time: 0.0020 memory: 1815 2023/09/06 21:51:04 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:21 time: 0.0497 data_time: 0.0019 memory: 1484 2023/09/06 21:51:06 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0485 data_time: 0.0020 memory: 2361 2023/09/06 21:51:09 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0495 data_time: 0.0019 memory: 1420 2023/09/06 21:51:11 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0542 data_time: 0.0019 memory: 1705 2023/09/06 21:51:14 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0474 data_time: 0.0018 memory: 1574 2023/09/06 21:51:16 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0459 data_time: 0.0017 memory: 1484 2023/09/06 21:51:18 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0480 data_time: 0.0018 memory: 1574 2023/09/06 21:51:21 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0496 data_time: 0.0017 memory: 1683 2023/09/06 21:51:23 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0473 data_time: 0.0019 memory: 1528 2023/09/06 21:51:26 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0469 data_time: 0.0017 memory: 1528 2023/09/06 21:51:29 - mmengine - INFO - per class results: 2023/09/06 21:51:29 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.29 | 91.71 | | bicycle | 59.01 | 79.52 | | car | 60.63 | 79.61 | | motorcycle | 79.33 | 89.19 | | airplane | 71.7 | 88.73 | | bus | 77.57 | 89.0 | | train | 75.87 | 88.07 | | truck | 59.76 | 78.4 | | boat | 56.79 | 80.43 | | traffic light | 55.64 | 80.99 | | fire hydrant | 72.02 | 92.88 | | stop sign | 79.19 | 94.74 | | parking meter | 51.34 | 83.72 | | bench | 47.98 | 69.59 | | bird | 68.7 | 85.59 | | cat | 80.65 | 89.09 | | dog | 77.28 | 84.63 | | horse | 78.67 | 91.0 | | sheep | 85.96 | 93.92 | | cow | 82.75 | 90.24 | | elephant | 90.58 | 95.28 | | bear | 89.73 | 92.66 | | zebra | 90.13 | 94.16 | | giraffe | 82.57 | 90.64 | | backpack | 24.41 | 64.88 | | umbrella | 72.76 | 77.62 | | handbag | 22.84 | 36.24 | | tie | 11.63 | 33.23 | | suitcase | 71.34 | 81.49 | | frisbee | 61.13 | 87.89 | | skis | 30.34 | 63.76 | | snowboard | 52.6 | 68.87 | | sports ball | 53.5 | 72.89 | | kite | 49.2 | 71.99 | | baseball bat | 32.83 | 67.14 | | baseball glove | 45.38 | 88.81 | | skateboard | 47.06 | 82.82 | | surfboard | 75.34 | 88.83 | | tennis racket | 70.45 | 88.3 | | bottle | 44.69 | 71.19 | | wine glass | 41.31 | 67.85 | | cup | 41.4 | 59.86 | | fork | 35.79 | 55.85 | | knife | 22.09 | 27.15 | | spoon | 15.73 | 33.83 | | bowl | 31.38 | 41.81 | | banana | 61.15 | 82.64 | | apple | 40.7 | 51.46 | | sandwich | 45.78 | 61.86 | | orange | 66.53 | 75.79 | | broccoli | 52.34 | 66.32 | | carrot | 46.97 | 59.85 | | hot dog | 50.12 | 60.61 | | pizza | 68.94 | 80.71 | | donut | 65.13 | 83.19 | | cake | 71.34 | 83.61 | | chair | 42.56 | 65.11 | | couch | 54.0 | 73.41 | | potted plant | 22.92 | 31.81 | | bed | 61.1 | 81.14 | | dining table | 41.45 | 70.59 | | toilet | 68.87 | 90.98 | | tv | 66.98 | 79.95 | | laptop | 64.68 | 81.66 | | mouse | 59.58 | 70.4 | | remote | 38.29 | 70.95 | | keyboard | 57.72 | 71.64 | | cell phone | 67.5 | 82.44 | | microwave | 45.3 | 60.82 | | oven | 46.84 | 69.2 | | toaster | 35.82 | 79.4 | | sink | 40.15 | 81.58 | | refrigerator | 64.43 | 84.67 | | book | 43.16 | 62.28 | | clock | 72.28 | 86.3 | | vase | 41.11 | 80.67 | | scissors | 64.47 | 81.54 | | teddy bear | 77.01 | 84.74 | | hair drier | 16.03 | 51.51 | | toothbrush | 40.35 | 75.34 | | banner | 27.91 | 56.92 | | blanket | 11.44 | 16.71 | | branch | 12.1 | 29.33 | | bridge | 23.47 | 37.06 | | building-other | 51.41 | 71.68 | | bush | 32.55 | 50.15 | | cabinet | 44.59 | 64.37 | | cage | 15.44 | 22.35 | | cardboard | 34.23 | 49.35 | | carpet | 49.88 | 75.56 | | ceiling-other | 59.06 | 75.37 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.2 | 0.44 | | clothes | 14.27 | 19.68 | | clouds | 44.65 | 60.78 | | counter | 20.55 | 50.76 | | cupboard | 1.88 | 8.86 | | curtain | 56.21 | 74.13 | | desk-stuff | 33.38 | 50.91 | | dirt | 35.72 | 53.94 | | door-stuff | 31.4 | 50.53 | | fence | 33.02 | 64.77 | | floor-marble | 5.43 | 5.67 | | floor-other | 18.51 | 27.2 | | floor-stone | 2.89 | 3.82 | | floor-tile | 55.43 | 65.3 | | floor-wood | 54.22 | 78.19 | | flower | 43.24 | 68.05 | | fog | 14.93 | 16.52 | | food-other | 29.09 | 48.19 | | fruit | 28.41 | 58.32 | | furniture-other | 10.58 | 15.11 | | grass | 66.37 | 83.27 | | gravel | 24.03 | 39.94 | | ground-other | 6.07 | 8.39 | | hill | 14.18 | 25.24 | | house | 24.61 | 30.45 | | leaves | 18.23 | 25.6 | | light | 30.4 | 49.64 | | mat | 2.29 | 4.56 | | metal | 13.57 | 14.78 | | mirror-stuff | 36.27 | 65.91 | | moss | 0.0 | 0.0 | | mountain | 54.61 | 70.95 | | mud | 3.86 | 8.34 | | napkin | 7.54 | 18.03 | | net | 30.67 | 59.09 | | paper | 23.84 | 33.25 | | pavement | 49.11 | 70.4 | | pillow | 11.54 | 17.68 | | plant-other | 20.66 | 30.78 | | plastic | 7.85 | 8.84 | | platform | 19.5 | 27.24 | | playingfield | 64.8 | 82.89 | | railing | 4.58 | 15.12 | | railroad | 49.15 | 79.33 | | river | 48.72 | 75.85 | | road | 62.36 | 77.73 | | rock | 44.79 | 69.4 | | roof | 33.56 | 48.91 | | rug | 33.93 | 43.97 | | salad | 11.35 | 15.88 | | sand | 56.96 | 64.68 | | sea | 84.46 | 92.28 | | shelf | 22.17 | 34.04 | | sky-other | 68.64 | 84.18 | | skyscraper | 21.35 | 27.54 | | snow | 87.6 | 94.78 | | solid-other | 0.0 | 0.0 | | stairs | 19.12 | 33.69 | | stone | 0.84 | 0.86 | | straw | 11.56 | 12.69 | | structural-other | 1.16 | 2.27 | | table | 14.75 | 19.31 | | tent | 8.3 | 13.71 | | textile-other | 10.67 | 11.94 | | towel | 22.87 | 34.21 | | tree | 71.77 | 82.69 | | vegetable | 40.35 | 54.56 | | wall-brick | 41.06 | 50.42 | | wall-concrete | 51.73 | 69.96 | | wall-other | 17.22 | 29.48 | | wall-panel | 0.18 | 0.23 | | wall-stone | 26.19 | 37.18 | | wall-tile | 62.22 | 77.53 | | wall-wood | 32.82 | 46.14 | | water-other | 22.21 | 27.92 | | waterdrops | 0.0 | 0.0 | | window-blind | 40.12 | 51.11 | | window-other | 41.74 | 73.35 | | wood | 19.77 | 24.95 | +------------------+-------+-------+ 2023/09/06 21:51:29 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7500 mIoU: 41.6700 mAcc: 56.8600 data_time: 0.0019 time: 0.0488 2023/09/06 21:51:52 - mmengine - INFO - Iter(train) [57050/60000] base_lr: 4.9167e-06 lr: 4.9167e-06 eta: 0:22:21 time: 0.4569 data_time: 0.0220 memory: 15901 grad_norm: 18.9045 loss: 8.8487 decode.loss_cls_ce: 1.9996 decode.loss_mask_ce: 0.7919 decode.loss_mask_dice: 1.6179 decode.d7.loss_cls_ce: 2.0304 decode.d7.loss_mask_ce: 0.7962 decode.d7.loss_mask_dice: 1.6127 2023/09/06 21:52:14 - mmengine - INFO - Iter(train) [57100/60000] base_lr: 4.8334e-06 lr: 4.8334e-06 eta: 0:21:58 time: 0.4532 data_time: 0.0235 memory: 15892 grad_norm: 17.3663 loss: 8.9877 decode.loss_cls_ce: 1.9680 decode.loss_mask_ce: 0.8587 decode.loss_mask_dice: 1.6589 decode.d7.loss_cls_ce: 1.9812 decode.d7.loss_mask_ce: 0.8659 decode.d7.loss_mask_dice: 1.6550 2023/09/06 21:52:37 - mmengine - INFO - Iter(train) [57150/60000] base_lr: 4.7501e-06 lr: 4.7501e-06 eta: 0:21:35 time: 0.4529 data_time: 0.0240 memory: 15791 grad_norm: 17.9285 loss: 8.8511 decode.loss_cls_ce: 1.9546 decode.loss_mask_ce: 0.8427 decode.loss_mask_dice: 1.6227 decode.d7.loss_cls_ce: 1.9667 decode.d7.loss_mask_ce: 0.8408 decode.d7.loss_mask_dice: 1.6235 2023/09/06 21:53:00 - mmengine - INFO - Iter(train) [57200/60000] base_lr: 4.6667e-06 lr: 4.6667e-06 eta: 0:21:13 time: 0.4587 data_time: 0.0226 memory: 15794 grad_norm: 19.6129 loss: 9.9968 decode.loss_cls_ce: 2.1321 decode.loss_mask_ce: 0.9899 decode.loss_mask_dice: 1.8760 decode.d7.loss_cls_ce: 2.1462 decode.d7.loss_mask_ce: 0.9890 decode.d7.loss_mask_dice: 1.8637 2023/09/06 21:53:23 - mmengine - INFO - Iter(train) [57250/60000] base_lr: 4.5834e-06 lr: 4.5834e-06 eta: 0:20:50 time: 0.4575 data_time: 0.0237 memory: 15872 grad_norm: 21.2315 loss: 8.2555 decode.loss_cls_ce: 1.7955 decode.loss_mask_ce: 0.8217 decode.loss_mask_dice: 1.5005 decode.d7.loss_cls_ce: 1.8244 decode.d7.loss_mask_ce: 0.8220 decode.d7.loss_mask_dice: 1.4914 2023/09/06 21:53:45 - mmengine - INFO - Iter(train) [57300/60000] base_lr: 4.5001e-06 lr: 4.5001e-06 eta: 0:20:27 time: 0.4546 data_time: 0.0229 memory: 16063 grad_norm: 15.6348 loss: 9.3496 decode.loss_cls_ce: 2.0162 decode.loss_mask_ce: 0.9021 decode.loss_mask_dice: 1.7616 decode.d7.loss_cls_ce: 2.0086 decode.d7.loss_mask_ce: 0.8951 decode.d7.loss_mask_dice: 1.7660 2023/09/06 21:54:08 - mmengine - INFO - Iter(train) [57350/60000] base_lr: 4.4167e-06 lr: 4.4167e-06 eta: 0:20:04 time: 0.4528 data_time: 0.0232 memory: 15781 grad_norm: 16.8191 loss: 8.6224 decode.loss_cls_ce: 1.8051 decode.loss_mask_ce: 0.8755 decode.loss_mask_dice: 1.6153 decode.d7.loss_cls_ce: 1.8280 decode.d7.loss_mask_ce: 0.8801 decode.d7.loss_mask_dice: 1.6185 2023/09/06 21:54:31 - mmengine - INFO - Iter(train) [57400/60000] base_lr: 4.3334e-06 lr: 4.3334e-06 eta: 0:19:42 time: 0.4508 data_time: 0.0228 memory: 15909 grad_norm: 17.8727 loss: 8.5763 decode.loss_cls_ce: 1.8138 decode.loss_mask_ce: 0.7990 decode.loss_mask_dice: 1.6607 decode.d7.loss_cls_ce: 1.8371 decode.d7.loss_mask_ce: 0.8008 decode.d7.loss_mask_dice: 1.6650 2023/09/06 21:54:54 - mmengine - INFO - Iter(train) [57450/60000] base_lr: 4.2501e-06 lr: 4.2501e-06 eta: 0:19:19 time: 0.4569 data_time: 0.0233 memory: 15870 grad_norm: 17.4095 loss: 8.6384 decode.loss_cls_ce: 1.8451 decode.loss_mask_ce: 0.8475 decode.loss_mask_dice: 1.6260 decode.d7.loss_cls_ce: 1.8502 decode.d7.loss_mask_ce: 0.8527 decode.d7.loss_mask_dice: 1.6168 2023/09/06 21:55:16 - mmengine - INFO - Iter(train) [57500/60000] base_lr: 4.1667e-06 lr: 4.1667e-06 eta: 0:18:56 time: 0.4580 data_time: 0.0235 memory: 15706 grad_norm: 20.5311 loss: 8.5601 decode.loss_cls_ce: 1.8799 decode.loss_mask_ce: 0.8284 decode.loss_mask_dice: 1.5723 decode.d7.loss_cls_ce: 1.8829 decode.d7.loss_mask_ce: 0.8336 decode.d7.loss_mask_dice: 1.5631 2023/09/06 21:55:19 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0460 data_time: 0.0017 memory: 1528 2023/09/06 21:55:21 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0472 data_time: 0.0018 memory: 1441 2023/09/06 21:55:24 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0475 data_time: 0.0018 memory: 1595 2023/09/06 21:55:26 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0473 data_time: 0.0017 memory: 1550 2023/09/06 21:55:29 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0497 data_time: 0.0018 memory: 1574 2023/09/06 21:55:31 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0500 data_time: 0.0018 memory: 1462 2023/09/06 21:55:33 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0509 data_time: 0.0018 memory: 1528 2023/09/06 21:55:36 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0498 data_time: 0.0021 memory: 1528 2023/09/06 21:55:38 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0492 data_time: 0.0020 memory: 2187 2023/09/06 21:55:41 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0470 data_time: 0.0022 memory: 1528 2023/09/06 21:55:43 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0476 data_time: 0.0019 memory: 1550 2023/09/06 21:55:46 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0496 data_time: 0.0019 memory: 1528 2023/09/06 21:55:48 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0468 data_time: 0.0019 memory: 1528 2023/09/06 21:55:51 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0492 data_time: 0.0019 memory: 1727 2023/09/06 21:55:53 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0490 data_time: 0.0018 memory: 1815 2023/09/06 21:55:56 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0493 data_time: 0.0018 memory: 1484 2023/09/06 21:55:58 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0478 data_time: 0.0020 memory: 2361 2023/09/06 21:56:01 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0489 data_time: 0.0018 memory: 1420 2023/09/06 21:56:03 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0551 data_time: 0.0020 memory: 1705 2023/09/06 21:56:06 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0470 data_time: 0.0019 memory: 1574 2023/09/06 21:56:08 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0475 data_time: 0.0018 memory: 1484 2023/09/06 21:56:10 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0486 data_time: 0.0020 memory: 1574 2023/09/06 21:56:13 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0508 data_time: 0.0019 memory: 1683 2023/09/06 21:56:15 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0465 data_time: 0.0019 memory: 1528 2023/09/06 21:56:18 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0472 data_time: 0.0017 memory: 1528 2023/09/06 21:56:21 - mmengine - INFO - per class results: 2023/09/06 21:56:21 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.32 | 91.77 | | bicycle | 59.68 | 80.72 | | car | 61.2 | 79.92 | | motorcycle | 78.99 | 89.02 | | airplane | 71.38 | 88.73 | | bus | 77.21 | 88.75 | | train | 76.48 | 88.02 | | truck | 59.94 | 78.32 | | boat | 56.74 | 80.01 | | traffic light | 56.28 | 80.49 | | fire hydrant | 71.07 | 93.01 | | stop sign | 78.83 | 95.33 | | parking meter | 51.43 | 83.59 | | bench | 48.1 | 69.31 | | bird | 70.92 | 85.14 | | cat | 79.71 | 88.85 | | dog | 77.22 | 84.44 | | horse | 78.61 | 91.0 | | sheep | 85.97 | 93.98 | | cow | 82.79 | 90.07 | | elephant | 90.52 | 95.26 | | bear | 89.32 | 92.24 | | zebra | 90.12 | 94.17 | | giraffe | 82.6 | 90.46 | | backpack | 24.19 | 63.85 | | umbrella | 72.81 | 77.68 | | handbag | 22.84 | 36.78 | | tie | 10.13 | 29.9 | | suitcase | 71.78 | 82.13 | | frisbee | 60.9 | 87.4 | | skis | 30.04 | 61.0 | | snowboard | 53.18 | 68.44 | | sports ball | 53.32 | 72.26 | | kite | 49.52 | 71.73 | | baseball bat | 32.82 | 65.63 | | baseball glove | 50.63 | 87.82 | | skateboard | 45.37 | 81.25 | | surfboard | 74.88 | 88.5 | | tennis racket | 70.78 | 87.74 | | bottle | 45.14 | 71.0 | | wine glass | 40.99 | 66.72 | | cup | 41.39 | 59.62 | | fork | 35.31 | 55.91 | | knife | 21.93 | 27.29 | | spoon | 16.33 | 34.13 | | bowl | 31.53 | 42.33 | | banana | 61.72 | 82.64 | | apple | 40.53 | 51.33 | | sandwich | 45.22 | 60.85 | | orange | 63.66 | 72.52 | | broccoli | 52.81 | 67.81 | | carrot | 45.97 | 58.07 | | hot dog | 50.06 | 60.55 | | pizza | 67.63 | 79.13 | | donut | 64.96 | 83.3 | | cake | 70.79 | 83.27 | | chair | 42.42 | 65.65 | | couch | 54.05 | 72.97 | | potted plant | 22.67 | 31.65 | | bed | 61.04 | 81.05 | | dining table | 41.8 | 70.8 | | toilet | 69.57 | 91.36 | | tv | 66.69 | 79.99 | | laptop | 64.75 | 81.54 | | mouse | 60.19 | 70.21 | | remote | 38.12 | 70.87 | | keyboard | 58.04 | 72.36 | | cell phone | 67.26 | 81.99 | | microwave | 45.05 | 61.14 | | oven | 47.63 | 68.1 | | toaster | 36.77 | 79.92 | | sink | 40.64 | 81.54 | | refrigerator | 64.29 | 84.74 | | book | 43.37 | 62.94 | | clock | 72.15 | 86.22 | | vase | 41.4 | 81.28 | | scissors | 55.51 | 70.18 | | teddy bear | 76.82 | 84.14 | | hair drier | 15.52 | 52.88 | | toothbrush | 40.34 | 75.03 | | banner | 28.98 | 59.16 | | blanket | 11.57 | 16.74 | | branch | 12.21 | 29.39 | | bridge | 23.01 | 36.35 | | building-other | 51.82 | 72.43 | | bush | 32.9 | 50.13 | | cabinet | 44.21 | 64.2 | | cage | 11.26 | 16.22 | | cardboard | 34.06 | 49.52 | | carpet | 50.06 | 75.85 | | ceiling-other | 58.96 | 76.14 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.14 | 0.3 | | clothes | 14.3 | 19.67 | | clouds | 44.77 | 60.6 | | counter | 20.88 | 50.9 | | cupboard | 2.01 | 10.1 | | curtain | 56.14 | 74.71 | | desk-stuff | 33.72 | 50.7 | | dirt | 36.25 | 56.0 | | door-stuff | 31.06 | 51.18 | | fence | 32.82 | 65.33 | | floor-marble | 3.14 | 3.27 | | floor-other | 18.74 | 27.37 | | floor-stone | 2.84 | 3.8 | | floor-tile | 54.84 | 65.52 | | floor-wood | 53.48 | 77.95 | | flower | 43.38 | 68.91 | | fog | 14.2 | 15.92 | | food-other | 29.11 | 48.22 | | fruit | 27.84 | 57.6 | | furniture-other | 10.78 | 15.36 | | grass | 66.26 | 82.99 | | gravel | 23.03 | 38.48 | | ground-other | 6.2 | 8.31 | | hill | 14.21 | 24.8 | | house | 24.66 | 30.29 | | leaves | 16.94 | 23.42 | | light | 30.17 | 48.63 | | mat | 2.94 | 5.51 | | metal | 13.33 | 14.47 | | mirror-stuff | 37.05 | 63.8 | | moss | 0.0 | 0.0 | | mountain | 54.46 | 70.4 | | mud | 3.72 | 8.34 | | napkin | 8.0 | 21.4 | | net | 30.91 | 59.15 | | paper | 25.23 | 35.12 | | pavement | 49.59 | 70.79 | | pillow | 14.38 | 21.74 | | plant-other | 20.63 | 30.95 | | plastic | 8.0 | 8.92 | | platform | 19.49 | 27.45 | | playingfield | 64.61 | 81.9 | | railing | 4.8 | 16.1 | | railroad | 49.33 | 79.32 | | river | 47.83 | 74.01 | | road | 62.77 | 77.88 | | rock | 44.6 | 68.99 | | roof | 33.16 | 47.98 | | rug | 34.41 | 45.59 | | salad | 11.34 | 15.82 | | sand | 56.7 | 64.97 | | sea | 84.43 | 92.27 | | shelf | 22.17 | 34.19 | | sky-other | 68.63 | 84.22 | | skyscraper | 21.52 | 27.6 | | snow | 87.67 | 95.0 | | solid-other | 0.0 | 0.0 | | stairs | 19.23 | 33.94 | | stone | 0.78 | 0.8 | | straw | 10.69 | 11.72 | | structural-other | 1.1 | 2.16 | | table | 15.13 | 19.7 | | tent | 8.37 | 13.8 | | textile-other | 10.72 | 11.86 | | towel | 23.77 | 35.71 | | tree | 71.82 | 83.4 | | vegetable | 40.17 | 53.93 | | wall-brick | 41.01 | 50.37 | | wall-concrete | 50.32 | 67.27 | | wall-other | 16.63 | 29.68 | | wall-panel | 0.37 | 0.46 | | wall-stone | 26.29 | 37.23 | | wall-tile | 62.19 | 78.23 | | wall-wood | 32.86 | 46.27 | | water-other | 21.74 | 28.21 | | waterdrops | 0.0 | 0.01 | | window-blind | 40.52 | 52.32 | | window-other | 42.0 | 73.65 | | wood | 19.6 | 24.64 | +------------------+-------+-------+ 2023/09/06 21:56:21 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.6700 mIoU: 41.5900 mAcc: 56.7000 data_time: 0.0019 time: 0.0491 2023/09/06 21:56:43 - mmengine - INFO - Iter(train) [57550/60000] base_lr: 4.0834e-06 lr: 4.0834e-06 eta: 0:18:34 time: 0.4573 data_time: 0.0236 memory: 15759 grad_norm: 18.5691 loss: 7.7775 decode.loss_cls_ce: 1.7453 decode.loss_mask_ce: 0.7768 decode.loss_mask_dice: 1.3451 decode.d7.loss_cls_ce: 1.7622 decode.d7.loss_mask_ce: 0.7834 decode.d7.loss_mask_dice: 1.3646 2023/09/06 21:57:06 - mmengine - INFO - Iter(train) [57600/60000] base_lr: 4.0001e-06 lr: 4.0001e-06 eta: 0:18:11 time: 0.4529 data_time: 0.0232 memory: 15860 grad_norm: 19.7059 loss: 9.4314 decode.loss_cls_ce: 2.1627 decode.loss_mask_ce: 0.8610 decode.loss_mask_dice: 1.6874 decode.d7.loss_cls_ce: 2.1499 decode.d7.loss_mask_ce: 0.8667 decode.d7.loss_mask_dice: 1.7037 2023/09/06 21:57:29 - mmengine - INFO - Iter(train) [57650/60000] base_lr: 3.9167e-06 lr: 3.9167e-06 eta: 0:17:48 time: 0.4523 data_time: 0.0233 memory: 15870 grad_norm: 17.2225 loss: 7.8368 decode.loss_cls_ce: 1.7368 decode.loss_mask_ce: 0.7288 decode.loss_mask_dice: 1.4470 decode.d7.loss_cls_ce: 1.7309 decode.d7.loss_mask_ce: 0.7328 decode.d7.loss_mask_dice: 1.4605 2023/09/06 21:57:52 - mmengine - INFO - Iter(train) [57700/60000] base_lr: 3.8334e-06 lr: 3.8334e-06 eta: 0:17:25 time: 0.4542 data_time: 0.0242 memory: 15780 grad_norm: 17.9362 loss: 9.0623 decode.loss_cls_ce: 1.9310 decode.loss_mask_ce: 0.8402 decode.loss_mask_dice: 1.7486 decode.d7.loss_cls_ce: 1.9499 decode.d7.loss_mask_ce: 0.8409 decode.d7.loss_mask_dice: 1.7518 2023/09/06 21:58:14 - mmengine - INFO - Iter(train) [57750/60000] base_lr: 3.7501e-06 lr: 3.7501e-06 eta: 0:17:03 time: 0.4530 data_time: 0.0231 memory: 15883 grad_norm: 18.2175 loss: 8.6152 decode.loss_cls_ce: 1.9033 decode.loss_mask_ce: 0.8082 decode.loss_mask_dice: 1.5868 decode.d7.loss_cls_ce: 1.8940 decode.d7.loss_mask_ce: 0.8209 decode.d7.loss_mask_dice: 1.6019 2023/09/06 21:58:37 - mmengine - INFO - Iter(train) [57800/60000] base_lr: 3.6667e-06 lr: 3.6667e-06 eta: 0:16:40 time: 0.4577 data_time: 0.0241 memory: 15847 grad_norm: 18.4211 loss: 8.5787 decode.loss_cls_ce: 1.8757 decode.loss_mask_ce: 0.8289 decode.loss_mask_dice: 1.5809 decode.d7.loss_cls_ce: 1.8882 decode.d7.loss_mask_ce: 0.8198 decode.d7.loss_mask_dice: 1.5852 2023/09/06 21:59:00 - mmengine - INFO - Iter(train) [57850/60000] base_lr: 3.5834e-06 lr: 3.5834e-06 eta: 0:16:17 time: 0.4526 data_time: 0.0238 memory: 16051 grad_norm: 18.7683 loss: 9.1236 decode.loss_cls_ce: 1.9100 decode.loss_mask_ce: 0.8600 decode.loss_mask_dice: 1.7909 decode.d7.loss_cls_ce: 1.9032 decode.d7.loss_mask_ce: 0.8654 decode.d7.loss_mask_dice: 1.7941 2023/09/06 21:59:22 - mmengine - INFO - Iter(train) [57900/60000] base_lr: 3.5001e-06 lr: 3.5001e-06 eta: 0:15:54 time: 0.4509 data_time: 0.0236 memory: 16103 grad_norm: 17.5644 loss: 10.0305 decode.loss_cls_ce: 2.1888 decode.loss_mask_ce: 0.8985 decode.loss_mask_dice: 1.9336 decode.d7.loss_cls_ce: 2.1877 decode.d7.loss_mask_ce: 0.8931 decode.d7.loss_mask_dice: 1.9288 2023/09/06 21:59:45 - mmengine - INFO - Iter(train) [57950/60000] base_lr: 3.4167e-06 lr: 3.4167e-06 eta: 0:15:32 time: 0.4593 data_time: 0.0224 memory: 15978 grad_norm: 16.0239 loss: 8.5213 decode.loss_cls_ce: 1.8828 decode.loss_mask_ce: 0.8067 decode.loss_mask_dice: 1.5593 decode.d7.loss_cls_ce: 1.8906 decode.d7.loss_mask_ce: 0.8147 decode.d7.loss_mask_dice: 1.5672 2023/09/06 22:00:08 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 22:00:08 - mmengine - INFO - Iter(train) [58000/60000] base_lr: 3.3334e-06 lr: 3.3334e-06 eta: 0:15:09 time: 0.4573 data_time: 0.0223 memory: 15833 grad_norm: 17.5566 loss: 8.6134 decode.loss_cls_ce: 1.8327 decode.loss_mask_ce: 0.8633 decode.loss_mask_dice: 1.6283 decode.d7.loss_cls_ce: 1.8019 decode.d7.loss_mask_ce: 0.8650 decode.d7.loss_mask_dice: 1.6223 2023/09/06 22:00:11 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0462 data_time: 0.0020 memory: 1528 2023/09/06 22:00:13 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0477 data_time: 0.0017 memory: 1441 2023/09/06 22:00:15 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0471 data_time: 0.0018 memory: 1595 2023/09/06 22:00:18 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0478 data_time: 0.0019 memory: 1550 2023/09/06 22:00:20 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0499 data_time: 0.0022 memory: 1574 2023/09/06 22:00:23 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0502 data_time: 0.0019 memory: 1462 2023/09/06 22:00:25 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0501 data_time: 0.0018 memory: 1528 2023/09/06 22:00:28 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0499 data_time: 0.0018 memory: 1528 2023/09/06 22:00:30 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0477 data_time: 0.0017 memory: 2187 2023/09/06 22:00:33 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0472 data_time: 0.0021 memory: 1528 2023/09/06 22:00:35 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0475 data_time: 0.0019 memory: 1550 2023/09/06 22:00:37 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0498 data_time: 0.0019 memory: 1528 2023/09/06 22:00:40 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0473 data_time: 0.0018 memory: 1528 2023/09/06 22:00:42 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0488 data_time: 0.0019 memory: 1727 2023/09/06 22:00:45 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0491 data_time: 0.0019 memory: 1815 2023/09/06 22:00:47 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0496 data_time: 0.0018 memory: 1484 2023/09/06 22:00:50 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0489 data_time: 0.0018 memory: 2361 2023/09/06 22:00:52 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0498 data_time: 0.0019 memory: 1420 2023/09/06 22:00:55 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0539 data_time: 0.0018 memory: 1705 2023/09/06 22:00:57 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0467 data_time: 0.0018 memory: 1574 2023/09/06 22:01:00 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0461 data_time: 0.0018 memory: 1484 2023/09/06 22:01:02 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0473 data_time: 0.0019 memory: 1574 2023/09/06 22:01:05 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0504 data_time: 0.0020 memory: 1683 2023/09/06 22:01:07 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0467 data_time: 0.0017 memory: 1528 2023/09/06 22:01:09 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0468 data_time: 0.0017 memory: 1528 2023/09/06 22:01:12 - mmengine - INFO - per class results: 2023/09/06 22:01:12 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.24 | 91.75 | | bicycle | 60.09 | 80.63 | | car | 60.96 | 79.53 | | motorcycle | 79.71 | 89.76 | | airplane | 71.73 | 89.15 | | bus | 77.64 | 89.25 | | train | 75.85 | 88.28 | | truck | 60.22 | 78.65 | | boat | 56.6 | 80.62 | | traffic light | 56.41 | 80.76 | | fire hydrant | 71.8 | 92.85 | | stop sign | 79.19 | 95.42 | | parking meter | 51.72 | 83.32 | | bench | 47.87 | 69.3 | | bird | 69.25 | 85.48 | | cat | 78.07 | 87.58 | | dog | 77.03 | 84.29 | | horse | 78.98 | 90.88 | | sheep | 85.85 | 94.12 | | cow | 83.3 | 90.76 | | elephant | 90.58 | 95.28 | | bear | 89.34 | 92.59 | | zebra | 90.12 | 94.11 | | giraffe | 82.55 | 90.43 | | backpack | 24.69 | 64.03 | | umbrella | 72.76 | 77.68 | | handbag | 23.17 | 36.99 | | tie | 9.97 | 30.12 | | suitcase | 72.05 | 82.56 | | frisbee | 61.28 | 87.71 | | skis | 30.4 | 61.61 | | snowboard | 53.71 | 68.3 | | sports ball | 53.82 | 72.98 | | kite | 48.76 | 72.08 | | baseball bat | 33.22 | 66.21 | | baseball glove | 50.9 | 88.36 | | skateboard | 46.5 | 82.3 | | surfboard | 74.69 | 88.63 | | tennis racket | 70.87 | 87.83 | | bottle | 44.8 | 70.3 | | wine glass | 41.24 | 67.95 | | cup | 41.58 | 60.77 | | fork | 35.37 | 56.59 | | knife | 22.13 | 27.1 | | spoon | 16.04 | 33.95 | | bowl | 31.65 | 42.73 | | banana | 61.88 | 82.78 | | apple | 40.43 | 51.15 | | sandwich | 46.67 | 62.89 | | orange | 62.97 | 71.74 | | broccoli | 52.41 | 66.54 | | carrot | 46.68 | 59.26 | | hot dog | 50.1 | 60.76 | | pizza | 68.41 | 80.42 | | donut | 64.39 | 83.35 | | cake | 71.13 | 83.75 | | chair | 42.45 | 66.03 | | couch | 54.06 | 73.6 | | potted plant | 22.71 | 31.92 | | bed | 60.95 | 81.16 | | dining table | 42.01 | 70.48 | | toilet | 69.35 | 91.53 | | tv | 67.67 | 79.93 | | laptop | 66.11 | 83.39 | | mouse | 59.36 | 70.74 | | remote | 38.65 | 71.59 | | keyboard | 57.03 | 71.1 | | cell phone | 67.1 | 81.95 | | microwave | 47.88 | 60.71 | | oven | 49.34 | 70.62 | | toaster | 56.64 | 79.96 | | sink | 40.23 | 81.35 | | refrigerator | 64.03 | 84.16 | | book | 43.58 | 62.92 | | clock | 72.12 | 86.37 | | vase | 41.42 | 82.09 | | scissors | 56.52 | 71.5 | | teddy bear | 77.1 | 85.07 | | hair drier | 15.73 | 52.34 | | toothbrush | 40.68 | 75.9 | | banner | 27.85 | 58.04 | | blanket | 11.65 | 16.73 | | branch | 12.55 | 28.36 | | bridge | 23.25 | 36.77 | | building-other | 51.45 | 71.46 | | bush | 32.59 | 50.27 | | cabinet | 43.96 | 64.6 | | cage | 11.1 | 15.81 | | cardboard | 34.62 | 50.15 | | carpet | 50.26 | 76.21 | | ceiling-other | 59.23 | 76.2 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.18 | 0.4 | | clothes | 14.14 | 19.41 | | clouds | 44.96 | 60.17 | | counter | 20.78 | 51.44 | | cupboard | 1.67 | 8.21 | | curtain | 56.62 | 74.61 | | desk-stuff | 33.69 | 50.35 | | dirt | 36.08 | 55.86 | | door-stuff | 31.27 | 51.53 | | fence | 33.01 | 66.31 | | floor-marble | 0.97 | 1.01 | | floor-other | 19.05 | 27.84 | | floor-stone | 2.82 | 3.81 | | floor-tile | 55.21 | 66.03 | | floor-wood | 53.05 | 77.95 | | flower | 42.63 | 67.23 | | fog | 14.52 | 16.04 | | food-other | 28.57 | 47.79 | | fruit | 27.7 | 58.19 | | furniture-other | 10.66 | 15.33 | | grass | 66.16 | 83.05 | | gravel | 21.96 | 36.47 | | ground-other | 6.25 | 8.39 | | hill | 14.23 | 24.56 | | house | 24.69 | 30.78 | | leaves | 18.94 | 26.6 | | light | 30.43 | 49.09 | | mat | 2.8 | 5.39 | | metal | 13.91 | 15.18 | | mirror-stuff | 36.95 | 64.89 | | moss | 0.0 | 0.0 | | mountain | 54.57 | 70.24 | | mud | 3.81 | 8.33 | | napkin | 7.53 | 21.01 | | net | 30.49 | 59.12 | | paper | 23.91 | 33.39 | | pavement | 49.65 | 71.08 | | pillow | 14.08 | 20.93 | | plant-other | 20.54 | 31.03 | | plastic | 8.08 | 9.04 | | platform | 19.29 | 26.89 | | playingfield | 64.44 | 81.57 | | railing | 5.01 | 17.19 | | railroad | 49.11 | 78.87 | | river | 48.26 | 74.91 | | road | 62.58 | 77.06 | | rock | 44.54 | 69.13 | | roof | 33.2 | 47.62 | | rug | 33.58 | 44.05 | | salad | 11.04 | 15.58 | | sand | 56.83 | 64.8 | | sea | 84.42 | 92.17 | | shelf | 22.1 | 33.51 | | sky-other | 69.1 | 84.8 | | skyscraper | 21.54 | 27.8 | | snow | 87.68 | 95.06 | | solid-other | 0.0 | 0.0 | | stairs | 19.19 | 34.08 | | stone | 0.85 | 0.88 | | straw | 11.69 | 12.82 | | structural-other | 0.83 | 1.62 | | table | 14.61 | 19.07 | | tent | 8.42 | 13.89 | | textile-other | 10.69 | 11.85 | | towel | 23.07 | 34.54 | | tree | 71.76 | 83.18 | | vegetable | 39.72 | 54.44 | | wall-brick | 40.58 | 50.34 | | wall-concrete | 51.2 | 68.64 | | wall-other | 16.8 | 28.92 | | wall-panel | 0.47 | 0.58 | | wall-stone | 27.64 | 39.22 | | wall-tile | 62.01 | 78.04 | | wall-wood | 32.88 | 46.6 | | water-other | 21.97 | 27.98 | | waterdrops | 0.01 | 0.02 | | window-blind | 40.34 | 51.43 | | window-other | 42.06 | 73.61 | | wood | 19.14 | 24.06 | +------------------+-------+-------+ 2023/09/06 22:01:12 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7500 mIoU: 41.7600 mAcc: 56.7800 data_time: 0.0019 time: 0.0490 2023/09/06 22:01:35 - mmengine - INFO - Iter(train) [58050/60000] base_lr: 3.2501e-06 lr: 3.2501e-06 eta: 0:14:46 time: 0.4524 data_time: 0.0245 memory: 15821 grad_norm: 17.2750 loss: 8.1598 decode.loss_cls_ce: 1.8261 decode.loss_mask_ce: 0.7999 decode.loss_mask_dice: 1.4492 decode.d7.loss_cls_ce: 1.8138 decode.d7.loss_mask_ce: 0.8022 decode.d7.loss_mask_dice: 1.4685 2023/09/06 22:01:57 - mmengine - INFO - Iter(train) [58100/60000] base_lr: 3.1667e-06 lr: 3.1667e-06 eta: 0:14:24 time: 0.4562 data_time: 0.0230 memory: 15897 grad_norm: 18.4553 loss: 8.5427 decode.loss_cls_ce: 1.7694 decode.loss_mask_ce: 0.8927 decode.loss_mask_dice: 1.6045 decode.d7.loss_cls_ce: 1.7773 decode.d7.loss_mask_ce: 0.9019 decode.d7.loss_mask_dice: 1.5968 2023/09/06 22:02:20 - mmengine - INFO - Iter(train) [58150/60000] base_lr: 3.0834e-06 lr: 3.0834e-06 eta: 0:14:01 time: 0.4564 data_time: 0.0240 memory: 15807 grad_norm: 17.4227 loss: 8.0806 decode.loss_cls_ce: 1.7810 decode.loss_mask_ce: 0.7946 decode.loss_mask_dice: 1.4530 decode.d7.loss_cls_ce: 1.7943 decode.d7.loss_mask_ce: 0.7983 decode.d7.loss_mask_dice: 1.4593 2023/09/06 22:02:43 - mmengine - INFO - Iter(train) [58200/60000] base_lr: 3.0001e-06 lr: 3.0001e-06 eta: 0:13:38 time: 0.4518 data_time: 0.0236 memory: 15832 grad_norm: 17.6021 loss: 9.3519 decode.loss_cls_ce: 2.0581 decode.loss_mask_ce: 0.8839 decode.loss_mask_dice: 1.7295 decode.d7.loss_cls_ce: 2.0798 decode.d7.loss_mask_ce: 0.8809 decode.d7.loss_mask_dice: 1.7197 2023/09/06 22:03:06 - mmengine - INFO - Iter(train) [58250/60000] base_lr: 2.9167e-06 lr: 2.9167e-06 eta: 0:13:15 time: 0.4555 data_time: 0.0236 memory: 15844 grad_norm: 17.8355 loss: 9.0748 decode.loss_cls_ce: 1.9064 decode.loss_mask_ce: 0.8777 decode.loss_mask_dice: 1.7457 decode.d7.loss_cls_ce: 1.9114 decode.d7.loss_mask_ce: 0.8800 decode.d7.loss_mask_dice: 1.7536 2023/09/06 22:03:29 - mmengine - INFO - Iter(train) [58300/60000] base_lr: 2.8334e-06 lr: 2.8334e-06 eta: 0:12:53 time: 0.4647 data_time: 0.0235 memory: 15990 grad_norm: 18.6237 loss: 9.6468 decode.loss_cls_ce: 2.0600 decode.loss_mask_ce: 0.9375 decode.loss_mask_dice: 1.8167 decode.d7.loss_cls_ce: 2.0850 decode.d7.loss_mask_ce: 0.9380 decode.d7.loss_mask_dice: 1.8096 2023/09/06 22:03:51 - mmengine - INFO - Iter(train) [58350/60000] base_lr: 2.7500e-06 lr: 2.7500e-06 eta: 0:12:30 time: 0.4526 data_time: 0.0238 memory: 15926 grad_norm: 18.5152 loss: 8.7142 decode.loss_cls_ce: 1.8835 decode.loss_mask_ce: 0.8757 decode.loss_mask_dice: 1.5995 decode.d7.loss_cls_ce: 1.8705 decode.d7.loss_mask_ce: 0.8764 decode.d7.loss_mask_dice: 1.6086 2023/09/06 22:04:14 - mmengine - INFO - Iter(train) [58400/60000] base_lr: 2.6667e-06 lr: 2.6667e-06 eta: 0:12:07 time: 0.4532 data_time: 0.0238 memory: 16092 grad_norm: 20.3610 loss: 8.8388 decode.loss_cls_ce: 1.8936 decode.loss_mask_ce: 0.8651 decode.loss_mask_dice: 1.6664 decode.d7.loss_cls_ce: 1.8733 decode.d7.loss_mask_ce: 0.8762 decode.d7.loss_mask_dice: 1.6643 2023/09/06 22:04:37 - mmengine - INFO - Iter(train) [58450/60000] base_lr: 2.5834e-06 lr: 2.5834e-06 eta: 0:11:44 time: 0.4546 data_time: 0.0241 memory: 15846 grad_norm: 16.7509 loss: 9.1724 decode.loss_cls_ce: 1.9987 decode.loss_mask_ce: 0.9102 decode.loss_mask_dice: 1.6913 decode.d7.loss_cls_ce: 1.9932 decode.d7.loss_mask_ce: 0.8883 decode.d7.loss_mask_dice: 1.6907 2023/09/06 22:05:00 - mmengine - INFO - Iter(train) [58500/60000] base_lr: 2.5000e-06 lr: 2.5000e-06 eta: 0:11:22 time: 0.4534 data_time: 0.0242 memory: 16068 grad_norm: 20.0955 loss: 9.9062 decode.loss_cls_ce: 2.0711 decode.loss_mask_ce: 0.9631 decode.loss_mask_dice: 1.9034 decode.d7.loss_cls_ce: 2.0966 decode.d7.loss_mask_ce: 0.9593 decode.d7.loss_mask_dice: 1.9128 2023/09/06 22:05:02 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0472 data_time: 0.0018 memory: 1528 2023/09/06 22:05:04 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:56 time: 0.0478 data_time: 0.0019 memory: 1441 2023/09/06 22:05:07 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0468 data_time: 0.0018 memory: 1595 2023/09/06 22:05:09 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0478 data_time: 0.0018 memory: 1550 2023/09/06 22:05:12 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0511 data_time: 0.0019 memory: 1574 2023/09/06 22:05:14 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0497 data_time: 0.0018 memory: 1462 2023/09/06 22:05:17 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:44 time: 0.0503 data_time: 0.0019 memory: 1528 2023/09/06 22:05:19 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0502 data_time: 0.0019 memory: 1528 2023/09/06 22:05:22 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0476 data_time: 0.0019 memory: 2187 2023/09/06 22:05:24 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0472 data_time: 0.0019 memory: 1528 2023/09/06 22:05:27 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0477 data_time: 0.0022 memory: 1550 2023/09/06 22:05:29 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0498 data_time: 0.0019 memory: 1528 2023/09/06 22:05:31 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0479 data_time: 0.0021 memory: 1528 2023/09/06 22:05:34 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0490 data_time: 0.0019 memory: 1727 2023/09/06 22:05:36 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0491 data_time: 0.0019 memory: 1815 2023/09/06 22:05:39 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0500 data_time: 0.0019 memory: 1484 2023/09/06 22:05:41 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0481 data_time: 0.0021 memory: 2361 2023/09/06 22:05:44 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0496 data_time: 0.0019 memory: 1420 2023/09/06 22:05:46 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0552 data_time: 0.0019 memory: 1705 2023/09/06 22:05:49 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0488 data_time: 0.0019 memory: 1574 2023/09/06 22:05:51 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0467 data_time: 0.0019 memory: 1484 2023/09/06 22:05:54 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0480 data_time: 0.0019 memory: 1574 2023/09/06 22:05:56 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0505 data_time: 0.0021 memory: 1683 2023/09/06 22:05:59 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0468 data_time: 0.0019 memory: 1528 2023/09/06 22:06:01 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0471 data_time: 0.0017 memory: 1528 2023/09/06 22:06:04 - mmengine - INFO - per class results: 2023/09/06 22:06:04 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.31 | 91.72 | | bicycle | 60.11 | 80.93 | | car | 60.7 | 79.57 | | motorcycle | 79.55 | 89.68 | | airplane | 71.98 | 88.99 | | bus | 77.67 | 88.95 | | train | 76.05 | 88.17 | | truck | 60.04 | 78.43 | | boat | 56.6 | 80.62 | | traffic light | 55.74 | 81.01 | | fire hydrant | 71.62 | 92.95 | | stop sign | 79.5 | 95.47 | | parking meter | 51.22 | 83.66 | | bench | 47.95 | 69.52 | | bird | 69.31 | 85.51 | | cat | 80.26 | 89.48 | | dog | 77.18 | 84.46 | | horse | 79.2 | 90.97 | | sheep | 85.95 | 94.04 | | cow | 83.74 | 91.33 | | elephant | 90.63 | 95.34 | | bear | 89.8 | 92.86 | | zebra | 90.15 | 94.21 | | giraffe | 82.6 | 90.55 | | backpack | 24.49 | 64.34 | | umbrella | 72.91 | 77.74 | | handbag | 22.61 | 36.69 | | tie | 12.1 | 30.87 | | suitcase | 71.81 | 82.21 | | frisbee | 60.35 | 87.02 | | skis | 30.5 | 62.32 | | snowboard | 52.84 | 68.61 | | sports ball | 53.14 | 72.85 | | kite | 48.1 | 72.24 | | baseball bat | 33.32 | 66.76 | | baseball glove | 50.11 | 88.38 | | skateboard | 45.28 | 82.17 | | surfboard | 74.62 | 88.79 | | tennis racket | 70.58 | 88.02 | | bottle | 45.35 | 71.51 | | wine glass | 40.85 | 67.2 | | cup | 41.41 | 60.09 | | fork | 34.93 | 55.89 | | knife | 21.99 | 26.54 | | spoon | 16.02 | 34.36 | | bowl | 31.34 | 41.82 | | banana | 61.56 | 83.21 | | apple | 40.42 | 51.26 | | sandwich | 46.98 | 63.47 | | orange | 64.81 | 75.53 | | broccoli | 53.24 | 68.1 | | carrot | 47.39 | 60.36 | | hot dog | 50.09 | 60.86 | | pizza | 68.59 | 80.49 | | donut | 63.6 | 83.32 | | cake | 71.28 | 83.62 | | chair | 42.7 | 66.22 | | couch | 54.02 | 72.98 | | potted plant | 23.04 | 32.29 | | bed | 61.14 | 80.95 | | dining table | 41.84 | 70.51 | | toilet | 69.46 | 91.54 | | tv | 67.88 | 79.89 | | laptop | 66.67 | 84.03 | | mouse | 57.65 | 67.92 | | remote | 38.81 | 72.07 | | keyboard | 57.98 | 72.31 | | cell phone | 67.14 | 81.97 | | microwave | 45.73 | 61.11 | | oven | 49.22 | 70.49 | | toaster | 61.58 | 80.78 | | sink | 40.1 | 81.87 | | refrigerator | 64.93 | 85.17 | | book | 42.84 | 62.23 | | clock | 71.77 | 86.53 | | vase | 42.02 | 81.18 | | scissors | 66.01 | 86.14 | | teddy bear | 77.59 | 85.3 | | hair drier | 15.73 | 53.42 | | toothbrush | 40.0 | 76.7 | | banner | 27.86 | 57.33 | | blanket | 11.43 | 16.56 | | branch | 12.62 | 29.28 | | bridge | 22.98 | 36.89 | | building-other | 51.45 | 71.45 | | bush | 32.87 | 50.18 | | cabinet | 43.75 | 64.3 | | cage | 14.62 | 21.56 | | cardboard | 34.29 | 49.54 | | carpet | 49.82 | 75.68 | | ceiling-other | 59.27 | 76.39 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.29 | 0.62 | | clothes | 14.18 | 19.48 | | clouds | 44.67 | 60.35 | | counter | 20.98 | 51.83 | | cupboard | 1.63 | 8.1 | | curtain | 56.86 | 74.96 | | desk-stuff | 34.0 | 50.7 | | dirt | 36.25 | 55.72 | | door-stuff | 31.03 | 51.39 | | fence | 33.21 | 65.25 | | floor-marble | 1.92 | 2.01 | | floor-other | 19.23 | 27.99 | | floor-stone | 2.77 | 3.81 | | floor-tile | 54.91 | 66.02 | | floor-wood | 53.43 | 78.13 | | flower | 43.13 | 68.37 | | fog | 14.64 | 16.2 | | food-other | 28.74 | 47.64 | | fruit | 28.57 | 58.6 | | furniture-other | 10.41 | 14.81 | | grass | 66.27 | 83.45 | | gravel | 21.17 | 34.94 | | ground-other | 6.41 | 8.7 | | hill | 13.75 | 23.35 | | house | 24.66 | 30.64 | | leaves | 18.33 | 25.52 | | light | 30.48 | 48.73 | | mat | 3.27 | 5.97 | | metal | 13.95 | 15.22 | | mirror-stuff | 37.21 | 64.08 | | moss | 0.0 | 0.0 | | mountain | 54.41 | 70.74 | | mud | 3.7 | 8.22 | | napkin | 7.74 | 21.46 | | net | 29.88 | 59.07 | | paper | 23.12 | 32.32 | | pavement | 49.67 | 71.41 | | pillow | 13.2 | 20.55 | | plant-other | 20.62 | 31.07 | | plastic | 7.89 | 8.87 | | platform | 18.63 | 25.87 | | playingfield | 64.12 | 81.43 | | railing | 5.52 | 18.78 | | railroad | 48.68 | 79.03 | | river | 48.65 | 74.83 | | road | 62.28 | 76.55 | | rock | 44.63 | 69.06 | | roof | 33.3 | 47.74 | | rug | 33.96 | 45.78 | | salad | 10.62 | 15.34 | | sand | 57.18 | 64.79 | | sea | 84.42 | 92.12 | | shelf | 22.08 | 33.53 | | sky-other | 68.87 | 84.57 | | skyscraper | 21.99 | 28.47 | | snow | 87.7 | 94.93 | | solid-other | 0.0 | 0.0 | | stairs | 19.31 | 34.35 | | stone | 0.66 | 0.68 | | straw | 11.29 | 12.4 | | structural-other | 0.87 | 1.73 | | table | 15.17 | 19.94 | | tent | 8.42 | 13.9 | | textile-other | 10.62 | 11.84 | | towel | 23.24 | 34.84 | | tree | 71.86 | 83.23 | | vegetable | 39.9 | 54.87 | | wall-brick | 40.6 | 50.07 | | wall-concrete | 51.59 | 69.25 | | wall-other | 16.44 | 28.01 | | wall-panel | 0.57 | 0.7 | | wall-stone | 28.0 | 39.52 | | wall-tile | 62.36 | 77.96 | | wall-wood | 32.95 | 46.15 | | water-other | 21.93 | 28.28 | | waterdrops | 0.01 | 0.01 | | window-blind | 40.04 | 51.54 | | window-other | 42.15 | 73.51 | | wood | 20.1 | 25.35 | +------------------+-------+-------+ 2023/09/06 22:06:04 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.8000 mIoU: 41.8600 mAcc: 56.9900 data_time: 0.0019 time: 0.0491 2023/09/06 22:06:26 - mmengine - INFO - Iter(train) [58550/60000] base_lr: 2.4167e-06 lr: 2.4167e-06 eta: 0:10:59 time: 0.4542 data_time: 0.0243 memory: 15818 grad_norm: 18.3032 loss: 9.4116 decode.loss_cls_ce: 2.0710 decode.loss_mask_ce: 0.8759 decode.loss_mask_dice: 1.7461 decode.d7.loss_cls_ce: 2.0641 decode.d7.loss_mask_ce: 0.8784 decode.d7.loss_mask_dice: 1.7761 2023/09/06 22:06:49 - mmengine - INFO - Iter(train) [58600/60000] base_lr: 2.3334e-06 lr: 2.3334e-06 eta: 0:10:36 time: 0.4535 data_time: 0.0233 memory: 15779 grad_norm: 18.9955 loss: 8.2375 decode.loss_cls_ce: 1.7714 decode.loss_mask_ce: 0.8323 decode.loss_mask_dice: 1.5212 decode.d7.loss_cls_ce: 1.7582 decode.d7.loss_mask_ce: 0.8416 decode.d7.loss_mask_dice: 1.5127 2023/09/06 22:07:12 - mmengine - INFO - Iter(train) [58650/60000] base_lr: 2.2500e-06 lr: 2.2500e-06 eta: 0:10:14 time: 0.4546 data_time: 0.0238 memory: 15961 grad_norm: 18.7687 loss: 9.9808 decode.loss_cls_ce: 2.1104 decode.loss_mask_ce: 0.9000 decode.loss_mask_dice: 1.9799 decode.d7.loss_cls_ce: 2.1154 decode.d7.loss_mask_ce: 0.9012 decode.d7.loss_mask_dice: 1.9739 2023/09/06 22:07:35 - mmengine - INFO - Iter(train) [58700/60000] base_lr: 2.1667e-06 lr: 2.1667e-06 eta: 0:09:51 time: 0.4542 data_time: 0.0244 memory: 15897 grad_norm: 19.1366 loss: 8.5532 decode.loss_cls_ce: 1.8553 decode.loss_mask_ce: 0.8685 decode.loss_mask_dice: 1.5414 decode.d7.loss_cls_ce: 1.8643 decode.d7.loss_mask_ce: 0.8717 decode.d7.loss_mask_dice: 1.5520 2023/09/06 22:07:58 - mmengine - INFO - Iter(train) [58750/60000] base_lr: 2.0834e-06 lr: 2.0834e-06 eta: 0:09:28 time: 0.4532 data_time: 0.0240 memory: 15805 grad_norm: 17.8208 loss: 9.2081 decode.loss_cls_ce: 1.9339 decode.loss_mask_ce: 0.8454 decode.loss_mask_dice: 1.8185 decode.d7.loss_cls_ce: 1.9315 decode.d7.loss_mask_ce: 0.8442 decode.d7.loss_mask_dice: 1.8346 2023/09/06 22:08:20 - mmengine - INFO - Iter(train) [58800/60000] base_lr: 2.0000e-06 lr: 2.0000e-06 eta: 0:09:05 time: 0.4535 data_time: 0.0244 memory: 15921 grad_norm: 17.5270 loss: 7.7120 decode.loss_cls_ce: 1.8171 decode.loss_mask_ce: 0.7208 decode.loss_mask_dice: 1.3159 decode.d7.loss_cls_ce: 1.8412 decode.d7.loss_mask_ce: 0.7152 decode.d7.loss_mask_dice: 1.3017 2023/09/06 22:08:43 - mmengine - INFO - Iter(train) [58850/60000] base_lr: 1.9167e-06 lr: 1.9167e-06 eta: 0:08:43 time: 0.4591 data_time: 0.0241 memory: 15735 grad_norm: 17.9193 loss: 8.5304 decode.loss_cls_ce: 1.8066 decode.loss_mask_ce: 0.8180 decode.loss_mask_dice: 1.6525 decode.d7.loss_cls_ce: 1.7737 decode.d7.loss_mask_ce: 0.8232 decode.d7.loss_mask_dice: 1.6564 2023/09/06 22:09:06 - mmengine - INFO - Iter(train) [58900/60000] base_lr: 1.8334e-06 lr: 1.8334e-06 eta: 0:08:20 time: 0.4541 data_time: 0.0241 memory: 15908 grad_norm: 19.4910 loss: 8.6855 decode.loss_cls_ce: 1.8137 decode.loss_mask_ce: 0.9061 decode.loss_mask_dice: 1.6231 decode.d7.loss_cls_ce: 1.8332 decode.d7.loss_mask_ce: 0.9045 decode.d7.loss_mask_dice: 1.6050 2023/09/06 22:09:29 - mmengine - INFO - Iter(train) [58950/60000] base_lr: 1.7500e-06 lr: 1.7500e-06 eta: 0:07:57 time: 0.4529 data_time: 0.0238 memory: 15820 grad_norm: 17.3678 loss: 8.6087 decode.loss_cls_ce: 1.7849 decode.loss_mask_ce: 0.7784 decode.loss_mask_dice: 1.7432 decode.d7.loss_cls_ce: 1.7933 decode.d7.loss_mask_ce: 0.7868 decode.d7.loss_mask_dice: 1.7221 2023/09/06 22:09:52 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 22:09:52 - mmengine - INFO - Iter(train) [59000/60000] base_lr: 1.6667e-06 lr: 1.6667e-06 eta: 0:07:34 time: 0.4557 data_time: 0.0239 memory: 15848 grad_norm: 17.2661 loss: 10.3901 decode.loss_cls_ce: 2.2269 decode.loss_mask_ce: 0.9707 decode.loss_mask_dice: 1.9880 decode.d7.loss_cls_ce: 2.2342 decode.d7.loss_mask_ce: 0.9731 decode.d7.loss_mask_dice: 1.9973 2023/09/06 22:09:54 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:59 time: 0.0478 data_time: 0.0019 memory: 1528 2023/09/06 22:09:56 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:56 time: 0.0475 data_time: 0.0018 memory: 1441 2023/09/06 22:09:59 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:54 time: 0.0480 data_time: 0.0019 memory: 1595 2023/09/06 22:10:01 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0481 data_time: 0.0019 memory: 1550 2023/09/06 22:10:04 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:49 time: 0.0503 data_time: 0.0019 memory: 1574 2023/09/06 22:10:06 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0501 data_time: 0.0018 memory: 1462 2023/09/06 22:10:09 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:44 time: 0.0506 data_time: 0.0020 memory: 1528 2023/09/06 22:10:11 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0501 data_time: 0.0019 memory: 1528 2023/09/06 22:10:14 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0485 data_time: 0.0020 memory: 2187 2023/09/06 22:10:16 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0479 data_time: 0.0022 memory: 1528 2023/09/06 22:10:19 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0473 data_time: 0.0019 memory: 1550 2023/09/06 22:10:21 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0503 data_time: 0.0017 memory: 1528 2023/09/06 22:10:23 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0476 data_time: 0.0019 memory: 1528 2023/09/06 22:10:26 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0490 data_time: 0.0019 memory: 1727 2023/09/06 22:10:28 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0493 data_time: 0.0019 memory: 1815 2023/09/06 22:10:31 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0500 data_time: 0.0019 memory: 1484 2023/09/06 22:10:33 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0484 data_time: 0.0019 memory: 2361 2023/09/06 22:10:36 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0490 data_time: 0.0018 memory: 1420 2023/09/06 22:10:38 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0557 data_time: 0.0020 memory: 1705 2023/09/06 22:10:41 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0468 data_time: 0.0019 memory: 1574 2023/09/06 22:10:43 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0464 data_time: 0.0019 memory: 1484 2023/09/06 22:10:46 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0480 data_time: 0.0022 memory: 1574 2023/09/06 22:10:48 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0503 data_time: 0.0019 memory: 1683 2023/09/06 22:10:50 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0472 data_time: 0.0019 memory: 1528 2023/09/06 22:10:53 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0469 data_time: 0.0017 memory: 1528 2023/09/06 22:10:55 - mmengine - INFO - per class results: 2023/09/06 22:10:55 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.25 | 91.7 | | bicycle | 60.16 | 80.96 | | car | 60.91 | 79.83 | | motorcycle | 79.52 | 89.48 | | airplane | 72.0 | 88.75 | | bus | 77.38 | 88.73 | | train | 76.78 | 88.12 | | truck | 60.0 | 78.27 | | boat | 56.87 | 80.43 | | traffic light | 55.89 | 80.72 | | fire hydrant | 71.26 | 92.62 | | stop sign | 79.46 | 95.43 | | parking meter | 52.49 | 83.84 | | bench | 47.96 | 69.47 | | bird | 69.34 | 85.4 | | cat | 79.24 | 88.36 | | dog | 77.07 | 84.34 | | horse | 79.2 | 90.84 | | sheep | 85.95 | 94.03 | | cow | 83.76 | 91.28 | | elephant | 90.6 | 95.31 | | bear | 89.07 | 92.05 | | zebra | 90.14 | 94.16 | | giraffe | 82.59 | 90.45 | | backpack | 24.65 | 64.05 | | umbrella | 72.87 | 77.69 | | handbag | 22.38 | 36.3 | | tie | 9.95 | 28.21 | | suitcase | 71.78 | 82.47 | | frisbee | 60.83 | 87.52 | | skis | 30.21 | 60.76 | | snowboard | 53.39 | 68.39 | | sports ball | 53.22 | 72.86 | | kite | 48.58 | 71.89 | | baseball bat | 33.52 | 65.42 | | baseball glove | 50.97 | 88.32 | | skateboard | 45.71 | 81.76 | | surfboard | 74.54 | 88.52 | | tennis racket | 71.02 | 87.83 | | bottle | 45.33 | 70.72 | | wine glass | 40.52 | 66.63 | | cup | 40.71 | 58.79 | | fork | 34.87 | 55.36 | | knife | 22.1 | 26.61 | | spoon | 16.08 | 33.92 | | bowl | 31.05 | 41.73 | | banana | 61.61 | 82.92 | | apple | 40.5 | 51.34 | | sandwich | 46.73 | 63.26 | | orange | 64.97 | 75.52 | | broccoli | 52.8 | 67.32 | | carrot | 45.9 | 58.55 | | hot dog | 50.0 | 60.58 | | pizza | 68.12 | 79.66 | | donut | 64.55 | 83.31 | | cake | 71.03 | 83.64 | | chair | 42.67 | 66.06 | | couch | 53.88 | 73.18 | | potted plant | 22.84 | 32.09 | | bed | 60.81 | 81.29 | | dining table | 41.71 | 70.99 | | toilet | 69.36 | 91.52 | | tv | 67.89 | 80.0 | | laptop | 66.69 | 84.02 | | mouse | 57.65 | 67.72 | | remote | 38.61 | 71.7 | | keyboard | 57.75 | 71.77 | | cell phone | 67.07 | 81.78 | | microwave | 48.15 | 60.87 | | oven | 49.12 | 69.99 | | toaster | 60.15 | 80.46 | | sink | 40.22 | 81.58 | | refrigerator | 64.15 | 84.54 | | book | 42.59 | 61.78 | | clock | 71.78 | 86.34 | | vase | 41.99 | 81.21 | | scissors | 61.59 | 78.23 | | teddy bear | 77.6 | 85.21 | | hair drier | 16.01 | 53.19 | | toothbrush | 40.33 | 75.91 | | banner | 28.34 | 58.01 | | blanket | 11.39 | 16.39 | | branch | 12.14 | 28.07 | | bridge | 22.92 | 36.29 | | building-other | 51.8 | 72.54 | | bush | 32.84 | 50.83 | | cabinet | 43.91 | 64.38 | | cage | 13.65 | 19.8 | | cardboard | 34.16 | 49.83 | | carpet | 49.94 | 75.46 | | ceiling-other | 59.61 | 76.83 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.25 | 0.55 | | clothes | 14.01 | 19.19 | | clouds | 44.52 | 59.7 | | counter | 20.77 | 51.15 | | cupboard | 1.67 | 8.11 | | curtain | 56.68 | 74.78 | | desk-stuff | 34.17 | 50.52 | | dirt | 36.27 | 55.78 | | door-stuff | 31.05 | 51.71 | | fence | 33.51 | 66.59 | | floor-marble | 1.79 | 1.86 | | floor-other | 19.05 | 27.75 | | floor-stone | 2.76 | 3.8 | | floor-tile | 54.79 | 65.41 | | floor-wood | 53.73 | 77.63 | | flower | 42.57 | 66.71 | | fog | 14.41 | 16.06 | | food-other | 28.79 | 47.5 | | fruit | 28.37 | 58.1 | | furniture-other | 10.45 | 14.98 | | grass | 66.34 | 82.97 | | gravel | 20.6 | 33.75 | | ground-other | 6.41 | 8.69 | | hill | 13.96 | 24.37 | | house | 24.67 | 30.18 | | leaves | 17.71 | 24.58 | | light | 30.49 | 48.77 | | mat | 3.24 | 6.03 | | metal | 13.71 | 14.92 | | mirror-stuff | 37.4 | 63.49 | | moss | 0.0 | 0.0 | | mountain | 54.59 | 70.55 | | mud | 3.78 | 8.34 | | napkin | 8.02 | 21.57 | | net | 30.54 | 59.02 | | paper | 23.71 | 32.92 | | pavement | 49.62 | 71.01 | | pillow | 10.51 | 15.58 | | plant-other | 20.68 | 30.89 | | plastic | 7.9 | 8.84 | | platform | 20.37 | 28.51 | | playingfield | 64.72 | 82.13 | | railing | 5.26 | 17.88 | | railroad | 48.79 | 78.92 | | river | 48.23 | 73.63 | | road | 62.39 | 77.18 | | rock | 44.37 | 68.69 | | roof | 34.22 | 48.16 | | rug | 33.91 | 45.94 | | salad | 11.07 | 15.51 | | sand | 57.39 | 64.64 | | sea | 84.51 | 92.31 | | shelf | 22.51 | 34.49 | | sky-other | 68.83 | 84.74 | | skyscraper | 21.47 | 27.55 | | snow | 87.56 | 94.74 | | solid-other | 0.0 | 0.0 | | stairs | 19.42 | 34.59 | | stone | 0.85 | 0.88 | | straw | 11.05 | 12.13 | | structural-other | 0.74 | 1.45 | | table | 15.09 | 19.82 | | tent | 8.36 | 13.75 | | textile-other | 10.62 | 11.78 | | towel | 22.85 | 34.23 | | tree | 71.76 | 83.28 | | vegetable | 39.63 | 54.4 | | wall-brick | 40.54 | 49.74 | | wall-concrete | 50.96 | 68.46 | | wall-other | 16.66 | 29.08 | | wall-panel | 0.39 | 0.48 | | wall-stone | 27.63 | 39.22 | | wall-tile | 62.1 | 78.04 | | wall-wood | 32.5 | 44.86 | | water-other | 21.87 | 28.82 | | waterdrops | 0.0 | 0.0 | | window-blind | 40.49 | 51.1 | | window-other | 42.0 | 73.68 | | wood | 19.38 | 24.47 | +------------------+-------+-------+ 2023/09/06 22:10:56 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7500 mIoU: 41.8000 mAcc: 56.7200 data_time: 0.0019 time: 0.0490 2023/09/06 22:11:18 - mmengine - INFO - Iter(train) [59050/60000] base_lr: 1.5834e-06 lr: 1.5834e-06 eta: 0:07:12 time: 0.4575 data_time: 0.0244 memory: 15751 grad_norm: 20.0833 loss: 8.7016 decode.loss_cls_ce: 1.8953 decode.loss_mask_ce: 0.8963 decode.loss_mask_dice: 1.5548 decode.d7.loss_cls_ce: 1.9038 decode.d7.loss_mask_ce: 0.8964 decode.d7.loss_mask_dice: 1.5550 2023/09/06 22:11:41 - mmengine - INFO - Iter(train) [59100/60000] base_lr: 1.5000e-06 lr: 1.5000e-06 eta: 0:06:49 time: 0.4543 data_time: 0.0252 memory: 15780 grad_norm: 21.5456 loss: 9.0014 decode.loss_cls_ce: 1.8881 decode.loss_mask_ce: 0.9005 decode.loss_mask_dice: 1.7076 decode.d7.loss_cls_ce: 1.8886 decode.d7.loss_mask_ce: 0.9035 decode.d7.loss_mask_dice: 1.7131 2023/09/06 22:12:04 - mmengine - INFO - Iter(train) [59150/60000] base_lr: 1.4167e-06 lr: 1.4167e-06 eta: 0:06:26 time: 0.4553 data_time: 0.0241 memory: 15748 grad_norm: 16.6773 loss: 7.8193 decode.loss_cls_ce: 1.6749 decode.loss_mask_ce: 0.7579 decode.loss_mask_dice: 1.4716 decode.d7.loss_cls_ce: 1.6837 decode.d7.loss_mask_ce: 0.7604 decode.d7.loss_mask_dice: 1.4708 2023/09/06 22:12:27 - mmengine - INFO - Iter(train) [59200/60000] base_lr: 1.3334e-06 lr: 1.3334e-06 eta: 0:06:03 time: 0.4625 data_time: 0.0240 memory: 15761 grad_norm: 16.8359 loss: 8.7773 decode.loss_cls_ce: 1.8350 decode.loss_mask_ce: 0.8510 decode.loss_mask_dice: 1.6976 decode.d7.loss_cls_ce: 1.8571 decode.d7.loss_mask_ce: 0.8564 decode.d7.loss_mask_dice: 1.6802 2023/09/06 22:12:49 - mmengine - INFO - Iter(train) [59250/60000] base_lr: 1.2500e-06 lr: 1.2500e-06 eta: 0:05:41 time: 0.4518 data_time: 0.0235 memory: 15858 grad_norm: 17.7140 loss: 9.5937 decode.loss_cls_ce: 2.0905 decode.loss_mask_ce: 0.9067 decode.loss_mask_dice: 1.7965 decode.d7.loss_cls_ce: 2.0997 decode.d7.loss_mask_ce: 0.9075 decode.d7.loss_mask_dice: 1.7928 2023/09/06 22:13:12 - mmengine - INFO - Iter(train) [59300/60000] base_lr: 1.1667e-06 lr: 1.1667e-06 eta: 0:05:18 time: 0.4540 data_time: 0.0240 memory: 15806 grad_norm: 19.2327 loss: 8.8647 decode.loss_cls_ce: 1.9750 decode.loss_mask_ce: 0.8761 decode.loss_mask_dice: 1.5789 decode.d7.loss_cls_ce: 1.9584 decode.d7.loss_mask_ce: 0.8869 decode.d7.loss_mask_dice: 1.5894 2023/09/06 22:13:35 - mmengine - INFO - Iter(train) [59350/60000] base_lr: 1.0834e-06 lr: 1.0834e-06 eta: 0:04:55 time: 0.4550 data_time: 0.0242 memory: 15808 grad_norm: 18.4075 loss: 9.4813 decode.loss_cls_ce: 2.0339 decode.loss_mask_ce: 0.9133 decode.loss_mask_dice: 1.8014 decode.d7.loss_cls_ce: 2.0106 decode.d7.loss_mask_ce: 0.9216 decode.d7.loss_mask_dice: 1.8003 2023/09/06 22:13:57 - mmengine - INFO - Iter(train) [59400/60000] base_lr: 1.0000e-06 lr: 1.0000e-06 eta: 0:04:32 time: 0.4513 data_time: 0.0240 memory: 15760 grad_norm: 23.8395 loss: 8.6702 decode.loss_cls_ce: 1.8667 decode.loss_mask_ce: 0.8867 decode.loss_mask_dice: 1.5594 decode.d7.loss_cls_ce: 1.9049 decode.d7.loss_mask_ce: 0.8839 decode.d7.loss_mask_dice: 1.5686 2023/09/06 22:14:20 - mmengine - INFO - Iter(train) [59450/60000] base_lr: 9.1668e-07 lr: 9.1668e-07 eta: 0:04:10 time: 0.4505 data_time: 0.0235 memory: 15808 grad_norm: 18.3195 loss: 8.6200 decode.loss_cls_ce: 1.8822 decode.loss_mask_ce: 0.8487 decode.loss_mask_dice: 1.5831 decode.d7.loss_cls_ce: 1.8619 decode.d7.loss_mask_ce: 0.8529 decode.d7.loss_mask_dice: 1.5912 2023/09/06 22:14:43 - mmengine - INFO - Iter(train) [59500/60000] base_lr: 8.3335e-07 lr: 8.3335e-07 eta: 0:03:47 time: 0.4541 data_time: 0.0239 memory: 15707 grad_norm: 17.3507 loss: 8.1535 decode.loss_cls_ce: 1.8340 decode.loss_mask_ce: 0.7676 decode.loss_mask_dice: 1.4816 decode.d7.loss_cls_ce: 1.8238 decode.d7.loss_mask_ce: 0.7782 decode.d7.loss_mask_dice: 1.4683 2023/09/06 22:14:45 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:58 time: 0.0460 data_time: 0.0018 memory: 1528 2023/09/06 22:14:48 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:55 time: 0.0473 data_time: 0.0018 memory: 1441 2023/09/06 22:14:50 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:53 time: 0.0471 data_time: 0.0019 memory: 1595 2023/09/06 22:14:53 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0481 data_time: 0.0019 memory: 1550 2023/09/06 22:14:55 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:48 time: 0.0505 data_time: 0.0018 memory: 1574 2023/09/06 22:14:58 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0501 data_time: 0.0018 memory: 1462 2023/09/06 22:15:00 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:43 time: 0.0504 data_time: 0.0020 memory: 1528 2023/09/06 22:15:02 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0499 data_time: 0.0019 memory: 1528 2023/09/06 22:15:05 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0475 data_time: 0.0019 memory: 2187 2023/09/06 22:15:07 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0477 data_time: 0.0023 memory: 1528 2023/09/06 22:15:10 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0478 data_time: 0.0018 memory: 1550 2023/09/06 22:15:12 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0494 data_time: 0.0018 memory: 1528 2023/09/06 22:15:15 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0475 data_time: 0.0020 memory: 1528 2023/09/06 22:15:17 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:26 time: 0.0491 data_time: 0.0019 memory: 1727 2023/09/06 22:15:20 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0491 data_time: 0.0019 memory: 1815 2023/09/06 22:15:22 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0498 data_time: 0.0019 memory: 1484 2023/09/06 22:15:25 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0490 data_time: 0.0019 memory: 2361 2023/09/06 22:15:27 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0494 data_time: 0.0020 memory: 1420 2023/09/06 22:15:30 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0541 data_time: 0.0019 memory: 1705 2023/09/06 22:15:32 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0469 data_time: 0.0018 memory: 1574 2023/09/06 22:15:35 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0467 data_time: 0.0019 memory: 1484 2023/09/06 22:15:37 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0480 data_time: 0.0023 memory: 1574 2023/09/06 22:15:39 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0500 data_time: 0.0018 memory: 1683 2023/09/06 22:15:42 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0470 data_time: 0.0019 memory: 1528 2023/09/06 22:15:44 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0475 data_time: 0.0018 memory: 1528 2023/09/06 22:15:47 - mmengine - INFO - per class results: 2023/09/06 22:15:47 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.25 | 91.59 | | bicycle | 59.85 | 80.95 | | car | 61.07 | 79.76 | | motorcycle | 79.47 | 89.44 | | airplane | 72.01 | 88.74 | | bus | 77.62 | 89.01 | | train | 76.09 | 88.14 | | truck | 60.27 | 78.31 | | boat | 56.77 | 80.38 | | traffic light | 55.74 | 80.81 | | fire hydrant | 71.13 | 92.63 | | stop sign | 78.78 | 95.44 | | parking meter | 52.12 | 83.96 | | bench | 47.93 | 69.35 | | bird | 69.45 | 85.34 | | cat | 79.03 | 88.1 | | dog | 77.13 | 84.42 | | horse | 79.15 | 90.94 | | sheep | 85.93 | 94.02 | | cow | 83.74 | 91.26 | | elephant | 90.61 | 95.35 | | bear | 89.17 | 92.08 | | zebra | 90.13 | 94.16 | | giraffe | 82.59 | 90.48 | | backpack | 24.45 | 64.41 | | umbrella | 72.84 | 77.67 | | handbag | 22.6 | 36.69 | | tie | 10.55 | 29.76 | | suitcase | 71.73 | 82.25 | | frisbee | 60.92 | 87.68 | | skis | 30.32 | 61.52 | | snowboard | 52.92 | 68.69 | | sports ball | 53.09 | 72.84 | | kite | 48.49 | 71.86 | | baseball bat | 33.6 | 66.05 | | baseball glove | 50.32 | 88.66 | | skateboard | 47.6 | 82.12 | | surfboard | 74.41 | 88.66 | | tennis racket | 70.83 | 87.98 | | bottle | 45.27 | 71.13 | | wine glass | 40.87 | 67.41 | | cup | 41.23 | 59.81 | | fork | 34.82 | 55.49 | | knife | 22.01 | 26.56 | | spoon | 16.1 | 34.08 | | bowl | 31.04 | 41.85 | | banana | 61.82 | 82.89 | | apple | 40.56 | 51.43 | | sandwich | 46.75 | 63.22 | | orange | 64.99 | 75.54 | | broccoli | 52.94 | 67.83 | | carrot | 46.22 | 58.99 | | hot dog | 50.01 | 60.59 | | pizza | 68.02 | 79.65 | | donut | 64.63 | 83.35 | | cake | 70.91 | 83.62 | | chair | 42.65 | 66.15 | | couch | 53.9 | 73.04 | | potted plant | 22.74 | 32.1 | | bed | 61.18 | 81.05 | | dining table | 41.88 | 70.75 | | toilet | 69.3 | 91.49 | | tv | 68.22 | 80.27 | | laptop | 66.78 | 84.35 | | mouse | 59.81 | 70.32 | | remote | 38.99 | 72.43 | | keyboard | 57.67 | 71.49 | | cell phone | 67.0 | 81.81 | | microwave | 48.52 | 61.17 | | oven | 49.0 | 69.56 | | toaster | 51.83 | 80.61 | | sink | 40.27 | 81.54 | | refrigerator | 63.47 | 83.82 | | book | 42.66 | 61.71 | | clock | 71.85 | 86.32 | | vase | 41.98 | 80.92 | | scissors | 62.58 | 79.74 | | teddy bear | 77.67 | 85.39 | | hair drier | 15.56 | 53.32 | | toothbrush | 40.47 | 76.06 | | banner | 28.41 | 58.46 | | blanket | 11.54 | 16.64 | | branch | 12.09 | 28.37 | | bridge | 23.09 | 36.42 | | building-other | 51.55 | 72.16 | | bush | 32.53 | 51.05 | | cabinet | 43.69 | 64.13 | | cage | 10.98 | 15.72 | | cardboard | 34.29 | 49.93 | | carpet | 49.69 | 75.22 | | ceiling-other | 59.76 | 76.9 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.21 | 0.46 | | clothes | 14.14 | 19.47 | | clouds | 44.39 | 59.59 | | counter | 20.88 | 51.02 | | cupboard | 1.68 | 8.24 | | curtain | 56.62 | 74.7 | | desk-stuff | 34.19 | 50.25 | | dirt | 36.24 | 55.84 | | door-stuff | 31.02 | 52.1 | | fence | 33.41 | 66.85 | | floor-marble | 1.85 | 1.92 | | floor-other | 19.25 | 27.94 | | floor-stone | 2.76 | 3.8 | | floor-tile | 54.67 | 65.78 | | floor-wood | 53.88 | 77.99 | | flower | 42.52 | 66.7 | | fog | 14.5 | 16.01 | | food-other | 29.03 | 48.1 | | fruit | 28.45 | 58.19 | | furniture-other | 10.46 | 14.91 | | grass | 66.35 | 83.03 | | gravel | 20.88 | 34.34 | | ground-other | 6.37 | 8.7 | | hill | 14.06 | 24.46 | | house | 24.66 | 30.33 | | leaves | 17.62 | 24.31 | | light | 30.45 | 48.31 | | mat | 3.06 | 5.78 | | metal | 13.54 | 14.71 | | mirror-stuff | 37.75 | 62.75 | | moss | 0.0 | 0.0 | | mountain | 54.44 | 70.42 | | mud | 3.86 | 8.42 | | napkin | 7.88 | 21.51 | | net | 30.47 | 59.13 | | paper | 24.43 | 34.25 | | pavement | 49.47 | 70.99 | | pillow | 13.42 | 20.4 | | plant-other | 20.63 | 30.74 | | plastic | 7.89 | 8.77 | | platform | 20.55 | 28.76 | | playingfield | 64.5 | 81.76 | | railing | 5.12 | 17.14 | | railroad | 48.96 | 78.75 | | river | 49.14 | 74.5 | | road | 62.44 | 77.08 | | rock | 44.84 | 69.44 | | roof | 33.68 | 47.74 | | rug | 33.96 | 46.0 | | salad | 11.04 | 15.92 | | sand | 57.06 | 64.75 | | sea | 84.5 | 92.28 | | shelf | 22.24 | 33.86 | | sky-other | 68.88 | 84.83 | | skyscraper | 21.52 | 27.64 | | snow | 87.69 | 94.92 | | solid-other | 0.0 | 0.0 | | stairs | 19.49 | 34.53 | | stone | 0.84 | 0.86 | | straw | 11.69 | 12.84 | | structural-other | 0.82 | 1.59 | | table | 15.07 | 19.74 | | tent | 8.33 | 13.71 | | textile-other | 10.6 | 11.75 | | towel | 22.74 | 34.44 | | tree | 71.68 | 82.99 | | vegetable | 39.58 | 54.63 | | wall-brick | 40.75 | 50.29 | | wall-concrete | 50.61 | 68.24 | | wall-other | 16.53 | 29.05 | | wall-panel | 0.63 | 0.78 | | wall-stone | 27.72 | 39.2 | | wall-tile | 62.14 | 78.2 | | wall-wood | 33.09 | 46.11 | | water-other | 22.56 | 29.49 | | waterdrops | 0.0 | 0.0 | | window-blind | 40.05 | 51.39 | | window-other | 41.92 | 73.58 | | wood | 19.63 | 24.8 | +------------------+-------+-------+ 2023/09/06 22:15:47 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7300 mIoU: 41.7900 mAcc: 56.8400 data_time: 0.0020 time: 0.0490 2023/09/06 22:16:10 - mmengine - INFO - Iter(train) [59550/60000] base_lr: 7.5001e-07 lr: 7.5001e-07 eta: 0:03:24 time: 0.4541 data_time: 0.0235 memory: 15946 grad_norm: 18.3928 loss: 9.3748 decode.loss_cls_ce: 1.9687 decode.loss_mask_ce: 0.8500 decode.loss_mask_dice: 1.8686 decode.d7.loss_cls_ce: 1.9698 decode.d7.loss_mask_ce: 0.8527 decode.d7.loss_mask_dice: 1.8651 2023/09/06 22:16:33 - mmengine - INFO - Iter(train) [59600/60000] base_lr: 6.6668e-07 lr: 6.6668e-07 eta: 0:03:01 time: 0.4559 data_time: 0.0243 memory: 15911 grad_norm: 17.9104 loss: 8.7212 decode.loss_cls_ce: 1.8456 decode.loss_mask_ce: 0.8823 decode.loss_mask_dice: 1.6207 decode.d7.loss_cls_ce: 1.8518 decode.d7.loss_mask_ce: 0.8889 decode.d7.loss_mask_dice: 1.6320 2023/09/06 22:16:55 - mmengine - INFO - Iter(train) [59650/60000] base_lr: 5.8334e-07 lr: 5.8334e-07 eta: 0:02:39 time: 0.4548 data_time: 0.0247 memory: 15831 grad_norm: 19.2475 loss: 9.5101 decode.loss_cls_ce: 2.0842 decode.loss_mask_ce: 0.8960 decode.loss_mask_dice: 1.7658 decode.d7.loss_cls_ce: 2.1027 decode.d7.loss_mask_ce: 0.8984 decode.d7.loss_mask_dice: 1.7630 2023/09/06 22:17:18 - mmengine - INFO - Iter(train) [59700/60000] base_lr: 5.0001e-07 lr: 5.0001e-07 eta: 0:02:16 time: 0.4613 data_time: 0.0237 memory: 15895 grad_norm: 16.1448 loss: 8.8936 decode.loss_cls_ce: 1.9217 decode.loss_mask_ce: 0.8616 decode.loss_mask_dice: 1.6589 decode.d7.loss_cls_ce: 1.9117 decode.d7.loss_mask_ce: 0.8649 decode.d7.loss_mask_dice: 1.6748 2023/09/06 22:17:41 - mmengine - INFO - Iter(train) [59750/60000] base_lr: 4.1667e-07 lr: 4.1667e-07 eta: 0:01:53 time: 0.4527 data_time: 0.0235 memory: 15924 grad_norm: 18.6324 loss: 9.4692 decode.loss_cls_ce: 2.0364 decode.loss_mask_ce: 0.8606 decode.loss_mask_dice: 1.8262 decode.d7.loss_cls_ce: 2.0510 decode.d7.loss_mask_ce: 0.8589 decode.d7.loss_mask_dice: 1.8361 2023/09/06 22:18:04 - mmengine - INFO - Iter(train) [59800/60000] base_lr: 3.3334e-07 lr: 3.3334e-07 eta: 0:01:30 time: 0.4518 data_time: 0.0233 memory: 15845 grad_norm: 17.8013 loss: 9.7264 decode.loss_cls_ce: 2.0670 decode.loss_mask_ce: 0.9544 decode.loss_mask_dice: 1.8467 decode.d7.loss_cls_ce: 2.0451 decode.d7.loss_mask_ce: 0.9601 decode.d7.loss_mask_dice: 1.8531 2023/09/06 22:18:26 - mmengine - INFO - Iter(train) [59850/60000] base_lr: 2.5000e-07 lr: 2.5000e-07 eta: 0:01:08 time: 0.4557 data_time: 0.0241 memory: 15860 grad_norm: 21.2753 loss: 9.2833 decode.loss_cls_ce: 2.0681 decode.loss_mask_ce: 0.8462 decode.loss_mask_dice: 1.7164 decode.d7.loss_cls_ce: 2.1043 decode.d7.loss_mask_ce: 0.8437 decode.d7.loss_mask_dice: 1.7046 2023/09/06 22:18:49 - mmengine - INFO - Iter(train) [59900/60000] base_lr: 1.6667e-07 lr: 1.6667e-07 eta: 0:00:45 time: 0.4545 data_time: 0.0240 memory: 16001 grad_norm: 18.8886 loss: 9.3230 decode.loss_cls_ce: 2.0706 decode.loss_mask_ce: 0.8849 decode.loss_mask_dice: 1.7175 decode.d7.loss_cls_ce: 2.0278 decode.d7.loss_mask_ce: 0.8913 decode.d7.loss_mask_dice: 1.7309 2023/09/06 22:19:12 - mmengine - INFO - Iter(train) [59950/60000] base_lr: 8.3335e-08 lr: 8.3335e-08 eta: 0:00:22 time: 0.4530 data_time: 0.0239 memory: 15946 grad_norm: 15.5432 loss: 9.2116 decode.loss_cls_ce: 2.0086 decode.loss_mask_ce: 0.8844 decode.loss_mask_dice: 1.7051 decode.d7.loss_cls_ce: 2.0262 decode.d7.loss_mask_ce: 0.8855 decode.d7.loss_mask_dice: 1.7017 2023/09/06 22:19:35 - mmengine - INFO - Exp name: san-vit-b16_coco-stuff164k-640x640_20230906_143207 2023/09/06 22:19:35 - mmengine - INFO - Iter(train) [60000/60000] base_lr: 0.0000e+00 lr: 0.0000e+00 eta: 0:00:00 time: 0.4544 data_time: 0.0244 memory: 15756 grad_norm: 18.4039 loss: 8.1281 decode.loss_cls_ce: 1.8781 decode.loss_mask_ce: 0.7753 decode.loss_mask_dice: 1.4111 decode.d7.loss_cls_ce: 1.8661 decode.d7.loss_mask_ce: 0.7813 decode.d7.loss_mask_dice: 1.4162 2023/09/06 22:19:35 - mmengine - INFO - Saving checkpoint at 60000 iterations 2023/09/06 22:19:40 - mmengine - INFO - Iter(val) [ 50/1250] eta: 0:00:59 time: 0.0465 data_time: 0.0018 memory: 1528 2023/09/06 22:19:43 - mmengine - INFO - Iter(val) [ 100/1250] eta: 0:00:56 time: 0.0475 data_time: 0.0018 memory: 1441 2023/09/06 22:19:45 - mmengine - INFO - Iter(val) [ 150/1250] eta: 0:00:54 time: 0.0474 data_time: 0.0019 memory: 1595 2023/09/06 22:19:48 - mmengine - INFO - Iter(val) [ 200/1250] eta: 0:00:51 time: 0.0486 data_time: 0.0022 memory: 1550 2023/09/06 22:19:50 - mmengine - INFO - Iter(val) [ 250/1250] eta: 0:00:49 time: 0.0499 data_time: 0.0019 memory: 1574 2023/09/06 22:19:53 - mmengine - INFO - Iter(val) [ 300/1250] eta: 0:00:46 time: 0.0506 data_time: 0.0018 memory: 1462 2023/09/06 22:19:55 - mmengine - INFO - Iter(val) [ 350/1250] eta: 0:00:44 time: 0.0501 data_time: 0.0020 memory: 1528 2023/09/06 22:19:58 - mmengine - INFO - Iter(val) [ 400/1250] eta: 0:00:41 time: 0.0500 data_time: 0.0021 memory: 1528 2023/09/06 22:20:00 - mmengine - INFO - Iter(val) [ 450/1250] eta: 0:00:39 time: 0.0483 data_time: 0.0020 memory: 2187 2023/09/06 22:20:03 - mmengine - INFO - Iter(val) [ 500/1250] eta: 0:00:36 time: 0.0471 data_time: 0.0019 memory: 1528 2023/09/06 22:20:05 - mmengine - INFO - Iter(val) [ 550/1250] eta: 0:00:34 time: 0.0473 data_time: 0.0017 memory: 1550 2023/09/06 22:20:07 - mmengine - INFO - Iter(val) [ 600/1250] eta: 0:00:31 time: 0.0502 data_time: 0.0019 memory: 1528 2023/09/06 22:20:10 - mmengine - INFO - Iter(val) [ 650/1250] eta: 0:00:29 time: 0.0476 data_time: 0.0018 memory: 1528 2023/09/06 22:20:12 - mmengine - INFO - Iter(val) [ 700/1250] eta: 0:00:27 time: 0.0501 data_time: 0.0019 memory: 1727 2023/09/06 22:20:15 - mmengine - INFO - Iter(val) [ 750/1250] eta: 0:00:24 time: 0.0487 data_time: 0.0018 memory: 1815 2023/09/06 22:20:17 - mmengine - INFO - Iter(val) [ 800/1250] eta: 0:00:22 time: 0.0497 data_time: 0.0019 memory: 1484 2023/09/06 22:20:20 - mmengine - INFO - Iter(val) [ 850/1250] eta: 0:00:19 time: 0.0479 data_time: 0.0018 memory: 2361 2023/09/06 22:20:22 - mmengine - INFO - Iter(val) [ 900/1250] eta: 0:00:17 time: 0.0493 data_time: 0.0019 memory: 1420 2023/09/06 22:20:25 - mmengine - INFO - Iter(val) [ 950/1250] eta: 0:00:14 time: 0.0544 data_time: 0.0019 memory: 1705 2023/09/06 22:20:27 - mmengine - INFO - Iter(val) [1000/1250] eta: 0:00:12 time: 0.0469 data_time: 0.0018 memory: 1574 2023/09/06 22:20:30 - mmengine - INFO - Iter(val) [1050/1250] eta: 0:00:09 time: 0.0463 data_time: 0.0018 memory: 1484 2023/09/06 22:20:32 - mmengine - INFO - Iter(val) [1100/1250] eta: 0:00:07 time: 0.0475 data_time: 0.0018 memory: 1574 2023/09/06 22:20:35 - mmengine - INFO - Iter(val) [1150/1250] eta: 0:00:04 time: 0.0501 data_time: 0.0019 memory: 1683 2023/09/06 22:20:37 - mmengine - INFO - Iter(val) [1200/1250] eta: 0:00:02 time: 0.0470 data_time: 0.0017 memory: 1528 2023/09/06 22:20:39 - mmengine - INFO - Iter(val) [1250/1250] eta: 0:00:00 time: 0.0464 data_time: 0.0016 memory: 1528 2023/09/06 22:20:42 - mmengine - INFO - per class results: 2023/09/06 22:20:42 - mmengine - INFO - +------------------+-------+-------+ | Class | IoU | Acc | +------------------+-------+-------+ | person | 84.28 | 91.76 | | bicycle | 60.12 | 80.91 | | car | 61.02 | 79.71 | | motorcycle | 79.65 | 89.59 | | airplane | 72.03 | 88.73 | | bus | 77.68 | 88.98 | | train | 76.13 | 88.14 | | truck | 60.3 | 78.32 | | boat | 56.79 | 80.3 | | traffic light | 55.73 | 80.85 | | fire hydrant | 71.17 | 92.68 | | stop sign | 78.97 | 95.45 | | parking meter | 52.09 | 83.91 | | bench | 47.92 | 69.37 | | bird | 69.55 | 85.34 | | cat | 79.08 | 88.12 | | dog | 77.13 | 84.4 | | horse | 79.16 | 90.93 | | sheep | 85.96 | 93.97 | | cow | 83.77 | 91.27 | | elephant | 90.62 | 95.36 | | bear | 89.18 | 92.11 | | zebra | 90.14 | 94.17 | | giraffe | 82.6 | 90.5 | | backpack | 24.5 | 63.96 | | umbrella | 72.86 | 77.66 | | handbag | 22.62 | 36.52 | | tie | 9.9 | 27.52 | | suitcase | 71.75 | 82.44 | | frisbee | 60.79 | 87.51 | | skis | 30.45 | 61.67 | | snowboard | 53.25 | 68.62 | | sports ball | 53.15 | 72.73 | | kite | 48.51 | 71.73 | | baseball bat | 33.34 | 65.89 | | baseball glove | 50.78 | 88.43 | | skateboard | 47.72 | 82.04 | | surfboard | 74.48 | 88.59 | | tennis racket | 70.74 | 87.89 | | bottle | 45.26 | 71.24 | | wine glass | 40.68 | 67.11 | | cup | 41.34 | 59.92 | | fork | 34.86 | 55.59 | | knife | 21.93 | 26.62 | | spoon | 16.15 | 34.1 | | bowl | 31.0 | 41.67 | | banana | 61.54 | 82.73 | | apple | 40.56 | 51.45 | | sandwich | 46.38 | 62.61 | | orange | 65.15 | 75.71 | | broccoli | 52.79 | 67.7 | | carrot | 46.01 | 58.7 | | hot dog | 50.07 | 60.67 | | pizza | 68.3 | 80.01 | | donut | 64.62 | 83.29 | | cake | 70.91 | 83.64 | | chair | 42.67 | 66.12 | | couch | 54.01 | 73.17 | | potted plant | 22.7 | 32.0 | | bed | 61.23 | 81.09 | | dining table | 41.9 | 70.74 | | toilet | 69.35 | 91.48 | | tv | 68.18 | 80.27 | | laptop | 66.78 | 84.28 | | mouse | 59.36 | 69.76 | | remote | 38.85 | 72.0 | | keyboard | 57.7 | 71.48 | | cell phone | 67.13 | 81.92 | | microwave | 47.94 | 61.08 | | oven | 48.85 | 69.83 | | toaster | 54.75 | 80.54 | | sink | 40.29 | 81.55 | | refrigerator | 63.67 | 83.9 | | book | 42.73 | 61.79 | | clock | 71.91 | 86.36 | | vase | 42.07 | 80.94 | | scissors | 61.08 | 77.79 | | teddy bear | 77.46 | 85.16 | | hair drier | 15.78 | 53.1 | | toothbrush | 40.58 | 76.0 | | banner | 28.58 | 58.53 | | blanket | 11.52 | 16.59 | | branch | 12.14 | 28.28 | | bridge | 22.92 | 36.07 | | building-other | 51.51 | 72.11 | | bush | 32.6 | 51.08 | | cabinet | 43.8 | 64.28 | | cage | 11.22 | 16.07 | | cardboard | 34.34 | 49.99 | | carpet | 49.64 | 75.19 | | ceiling-other | 59.74 | 77.01 | | ceiling-tile | 0.0 | 0.0 | | cloth | 0.22 | 0.47 | | clothes | 14.11 | 19.33 | | clouds | 44.77 | 60.48 | | counter | 20.89 | 50.84 | | cupboard | 1.67 | 8.2 | | curtain | 56.62 | 74.81 | | desk-stuff | 34.1 | 50.26 | | dirt | 36.23 | 55.8 | | door-stuff | 31.07 | 52.12 | | fence | 33.33 | 66.72 | | floor-marble | 1.9 | 1.97 | | floor-other | 19.16 | 28.02 | | floor-stone | 2.79 | 3.8 | | floor-tile | 55.12 | 65.79 | | floor-wood | 53.82 | 77.94 | | flower | 42.54 | 66.68 | | fog | 14.61 | 16.14 | | food-other | 28.99 | 47.96 | | fruit | 28.45 | 57.94 | | furniture-other | 10.55 | 15.06 | | grass | 66.33 | 83.06 | | gravel | 20.94 | 34.42 | | ground-other | 6.38 | 8.72 | | hill | 14.17 | 24.61 | | house | 24.54 | 30.28 | | leaves | 17.51 | 24.15 | | light | 30.45 | 48.45 | | mat | 3.16 | 5.84 | | metal | 13.61 | 14.79 | | mirror-stuff | 37.77 | 62.98 | | moss | 0.0 | 0.0 | | mountain | 54.46 | 70.55 | | mud | 3.81 | 8.39 | | napkin | 8.0 | 21.47 | | net | 30.56 | 59.08 | | paper | 23.84 | 33.39 | | pavement | 49.56 | 70.95 | | pillow | 13.61 | 20.31 | | plant-other | 20.62 | 30.7 | | plastic | 7.92 | 8.87 | | platform | 19.83 | 27.63 | | playingfield | 64.68 | 82.0 | | railing | 5.03 | 16.93 | | railroad | 48.97 | 78.74 | | river | 49.57 | 75.52 | | road | 62.69 | 77.42 | | rock | 44.71 | 69.21 | | roof | 33.91 | 48.27 | | rug | 33.78 | 45.66 | | salad | 11.04 | 15.79 | | sand | 57.27 | 64.73 | | sea | 84.49 | 92.26 | | shelf | 22.19 | 33.97 | | sky-other | 68.86 | 84.59 | | skyscraper | 21.56 | 27.73 | | snow | 87.69 | 94.88 | | solid-other | 0.0 | 0.0 | | stairs | 19.49 | 34.51 | | stone | 0.79 | 0.81 | | straw | 11.65 | 12.79 | | structural-other | 0.82 | 1.6 | | table | 15.18 | 19.85 | | tent | 8.35 | 13.72 | | textile-other | 10.61 | 11.75 | | towel | 22.75 | 34.44 | | tree | 71.71 | 83.1 | | vegetable | 39.43 | 54.25 | | wall-brick | 40.76 | 50.27 | | wall-concrete | 50.88 | 68.46 | | wall-other | 16.62 | 29.06 | | wall-panel | 0.58 | 0.72 | | wall-stone | 27.68 | 39.16 | | wall-tile | 62.14 | 78.18 | | wall-wood | 33.04 | 46.04 | | water-other | 22.03 | 28.49 | | waterdrops | 0.0 | 0.0 | | window-blind | 40.0 | 51.41 | | window-other | 42.0 | 73.61 | | wood | 19.62 | 24.78 | +------------------+-------+-------+ 2023/09/06 22:20:42 - mmengine - INFO - Iter(val) [1250/1250] aAcc: 67.7700 mIoU: 41.8100 mAcc: 56.8000 data_time: 0.0019 time: 0.0491