2023-01-16 12:30:05,009 - mmrotate - INFO - Environment info: ------------------------------------------------------------ sys.platform: linux Python: 3.8.5 (default, Sep 4 2020, 07:30:14) [GCC 7.3.0] CUDA available: True GPU 0,1,2,3,4,5,6,7: GeForce RTX 3090 CUDA_HOME: /usr/local/cuda NVCC: Cuda compilation tools, release 11.0, V11.0.221 GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 PyTorch: 1.7.1+cu110 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 v1.6.0 (Git Hash 5ef631a030a6f73131c77892041042805a06064f) - OpenMP 201511 (a.k.a. OpenMP 4.5) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.0 - 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 - CuDNN 8.0.5 - Magma 2.5.2 - Build settings: BLAS=MKL, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_VULKAN_WRAPPER -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, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=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: 0.8.2+cu110 OpenCV: 4.6.0 MMCV: 1.7.0 MMCV Compiler: GCC 7.3 MMCV CUDA Compiler: 11.0 MMRotate: 0.3.3+efe4abb ------------------------------------------------------------ 2023-01-16 12:30:05,456 - mmrotate - INFO - Distributed training: True 2023-01-16 12:30:05,665 - mmrotate - INFO - Config: dataset_type = 'DOTADataset' data_root = '/opt/data/private/LYX/data/split_ms_dota/' img_norm_cfg = dict( mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', with_bbox=True), dict(type='RResize', img_scale=(1024, 1024)), dict( type='RRandomFlip', flip_ratio=[0.25, 0.25, 0.25], direction=['horizontal', 'vertical', 'diagonal'], version='le90'), dict( type='PolyRandomRotate', rotate_ratio=0.5, angles_range=180, auto_bound=False, rect_classes=[9, 11], version='le90'), dict( type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']) ] test_pipeline = [ dict(type='LoadImageFromFile'), dict( type='MultiScaleFlipAug', img_scale=(1024, 1024), flip=False, transforms=[ dict(type='RResize'), dict( type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img']) ]) ] data = dict( samples_per_gpu=1, workers_per_gpu=2, train=dict( type='DOTADataset', ann_file= '/opt/data/private/LYX/data/split_ms_dota/trainval_2/annfiles/', img_prefix= '/opt/data/private/LYX/data/split_ms_dota/trainval_2/images/', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', with_bbox=True), dict(type='RResize', img_scale=(1024, 1024)), dict( type='RRandomFlip', flip_ratio=[0.25, 0.25, 0.25], direction=['horizontal', 'vertical', 'diagonal'], version='le90'), dict( type='PolyRandomRotate', rotate_ratio=0.5, angles_range=180, auto_bound=False, rect_classes=[9, 11], version='le90'), dict( type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']) ], version='le90'), val=dict( type='DOTADataset', ann_file='/opt/data/private/LYX/data/split_ms_dota/val/annfiles/', img_prefix='/opt/data/private/LYX/data/split_ms_dota/val/images/', pipeline=[ dict(type='LoadImageFromFile'), dict( type='MultiScaleFlipAug', img_scale=(1024, 1024), flip=False, transforms=[ dict(type='RResize'), dict( type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img']) ]) ], version='le90'), test=dict( type='DOTADataset', ann_file='/opt/data/private/LYX/data/split_ms_dota/test/images/', img_prefix='/opt/data/private/LYX/data/split_ms_dota/test/images/', pipeline=[ dict(type='LoadImageFromFile'), dict( type='MultiScaleFlipAug', img_scale=(1024, 1024), flip=False, transforms=[ dict(type='RResize'), dict( type='Normalize', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), dict(type='Pad', size_divisor=32), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img']) ]) ], version='le90')) evaluation = dict(interval=3, metric='mAP') optimizer = dict( type='AdamW', lr=0.0002, betas=(0.9, 0.999), weight_decay=0.05) optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2)) lr_config = dict( policy='step', warmup='linear', warmup_iters=500, warmup_ratio=0.3333333333333333, step=[8, 11]) runner = dict(type='EpochBasedRunner', max_epochs=12) checkpoint_config = dict(interval=1) log_config = dict(interval=500, hooks=[dict(type='TextLoggerHook')]) dist_params = dict(backend='nccl') log_level = 'INFO' load_from = None resume_from = '/opt/data/private/LYX/selective-large-kernel/work_dirs/lsk9b1_dota/epoch_6.pth' workflow = [('train', 1)] opencv_num_threads = 0 mp_start_method = 'fork' angle_version = 'le90' gpu_number = 8 fp16 = dict(loss_scale='dynamic') model = dict( type='OrientedRCNN', backbone=dict( type='LSK9Net', embed_dims=[64, 128, 320, 512], drop_rate=0.1, drop_path_rate=0.1, depths=[2, 2, 4, 2], init_cfg=dict( type='Pretrained', checkpoint='/opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar' ), norm_cfg=dict(type='SyncBN', requires_grad=True)), neck=dict( type='FPN', in_channels=[64, 128, 320, 512], out_channels=256, num_outs=5), rpn_head=dict( type='OrientedRPNHead', in_channels=256, feat_channels=256, version='le90', anchor_generator=dict( type='AnchorGenerator', scales=[8], ratios=[0.5, 1.0, 2.0], strides=[4, 8, 16, 32, 64]), bbox_coder=dict( type='MidpointOffsetCoder', angle_range='le90', target_means=[0.0, 0.0, 0.0, 0.0, 0.0, 0.0], target_stds=[1.0, 1.0, 1.0, 1.0, 0.5, 0.5]), loss_cls=dict( type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0), loss_bbox=dict( type='SmoothL1Loss', beta=0.1111111111111111, loss_weight=1.0)), roi_head=dict( type='OrientedStandardRoIHead', bbox_roi_extractor=dict( type='RotatedSingleRoIExtractor', roi_layer=dict( type='RoIAlignRotated', out_size=7, sample_num=2, clockwise=True), out_channels=256, featmap_strides=[4, 8, 16, 32]), bbox_head=dict( type='RotatedShared2FCBBoxHead', in_channels=256, fc_out_channels=1024, roi_feat_size=7, num_classes=15, bbox_coder=dict( type='DeltaXYWHAOBBoxCoder', angle_range='le90', norm_factor=None, edge_swap=True, proj_xy=True, target_means=(0.0, 0.0, 0.0, 0.0, 0.0), target_stds=(0.1, 0.1, 0.2, 0.2, 0.1)), reg_class_agnostic=True, loss_cls=dict( type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0), loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0))), train_cfg=dict( rpn=dict( assigner=dict( type='MaxIoUAssigner', pos_iou_thr=0.7, neg_iou_thr=0.3, min_pos_iou=0.3, match_low_quality=True, gpu_assign_thr=800, ignore_iof_thr=-1), sampler=dict( type='RandomSampler', num=256, pos_fraction=0.5, neg_pos_ub=-1, add_gt_as_proposals=False), allowed_border=0, pos_weight=-1, debug=False), rpn_proposal=dict( nms_pre=2000, max_per_img=2000, nms=dict(type='nms', iou_threshold=0.8), min_bbox_size=0), rcnn=dict( assigner=dict( type='MaxIoUAssigner', pos_iou_thr=0.5, neg_iou_thr=0.5, min_pos_iou=0.5, match_low_quality=False, iou_calculator=dict(type='RBboxOverlaps2D'), gpu_assign_thr=800, ignore_iof_thr=-1), sampler=dict( type='RRandomSampler', num=512, pos_fraction=0.25, neg_pos_ub=-1, add_gt_as_proposals=True), pos_weight=-1, debug=False)), test_cfg=dict( rpn=dict( nms_pre=2000, max_per_img=2000, nms=dict(type='nms', iou_threshold=0.8), min_bbox_size=0), rcnn=dict( nms_pre=2000, min_bbox_size=0, score_thr=0.05, nms=dict(iou_thr=0.1), max_per_img=2000))) work_dir = './work_dirs/lsk9b1_dota' auto_resume = False gpu_ids = range(0, 8) 2023-01-16 12:30:05,665 - mmrotate - INFO - Set random seed to 0, deterministic: False 2023-01-16 12:30:05,929 - mmrotate - INFO - initialize LSK9Net with init_cfg {'type': 'Pretrained', 'checkpoint': '/opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar'} 2023-01-16 12:30:06,999 - mmrotate - INFO - initialize FPN with init_cfg {'type': 'Xavier', 'layer': 'Conv2d', 'distribution': 'uniform'} 2023-01-16 12:30:07,026 - mmrotate - INFO - initialize OrientedRPNHead with init_cfg {'type': 'Normal', 'layer': 'Conv2d', 'std': 0.01} 2023-01-16 12:30:07,036 - mmrotate - INFO - initialize RotatedShared2FCBBoxHead with init_cfg [{'type': 'Normal', 'std': 0.01, 'override': {'name': 'fc_cls'}}, {'type': 'Normal', 'std': 0.001, 'override': {'name': 'fc_reg'}}, {'type': 'Xavier', 'layer': 'Linear', 'override': [{'name': 'shared_fcs'}, {'name': 'cls_fcs'}, {'name': 'reg_fcs'}]}] Name of parameter - Initialization information backbone.patch_embed1.proj.weight - torch.Size([64, 3, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed1.proj.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed1.norm.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed1.norm.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.layer_scale_1 - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.layer_scale_2 - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.norm1.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.norm1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.norm2.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.norm2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.proj_1.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.proj_1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv0.weight - torch.Size([64, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv0.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([64, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv1.weight - torch.Size([32, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv1.bias - torch.Size([32]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv2.weight - torch.Size([32, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv2.bias - torch.Size([32]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv.weight - torch.Size([64, 32, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.spatial_gating_unit.conv.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.proj_2.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.attn.proj_2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.fc1.weight - torch.Size([512, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.fc1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.dwconv.dwconv.weight - torch.Size([512, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.dwconv.dwconv.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.fc2.weight - torch.Size([64, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.0.mlp.fc2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.layer_scale_1 - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.layer_scale_2 - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.norm1.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.norm1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.norm2.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.norm2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.proj_1.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.proj_1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv0.weight - torch.Size([64, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv0.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([64, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv1.weight - torch.Size([32, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv1.bias - torch.Size([32]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv2.weight - torch.Size([32, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv2.bias - torch.Size([32]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv.weight - torch.Size([64, 32, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.spatial_gating_unit.conv.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.proj_2.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.attn.proj_2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.fc1.weight - torch.Size([512, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.fc1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.dwconv.dwconv.weight - torch.Size([512, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.dwconv.dwconv.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.fc2.weight - torch.Size([64, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block1.1.mlp.fc2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm1.weight - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed2.proj.weight - torch.Size([128, 64, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed2.proj.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed2.norm.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed2.norm.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.layer_scale_1 - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.layer_scale_2 - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.norm1.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.norm1.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.norm2.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.norm2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.proj_1.weight - torch.Size([128, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.proj_1.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv0.weight - torch.Size([128, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv0.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([128, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv1.weight - torch.Size([64, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv2.weight - torch.Size([64, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv.weight - torch.Size([128, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.spatial_gating_unit.conv.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.proj_2.weight - torch.Size([128, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.attn.proj_2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.fc1.weight - torch.Size([1024, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.fc1.bias - torch.Size([1024]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.dwconv.dwconv.weight - torch.Size([1024, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.dwconv.dwconv.bias - torch.Size([1024]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.fc2.weight - torch.Size([128, 1024, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.0.mlp.fc2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.layer_scale_1 - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.layer_scale_2 - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.norm1.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.norm1.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.norm2.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.norm2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.proj_1.weight - torch.Size([128, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.proj_1.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv0.weight - torch.Size([128, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv0.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([128, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv1.weight - torch.Size([64, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv1.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv2.weight - torch.Size([64, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv2.bias - torch.Size([64]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv.weight - torch.Size([128, 64, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.spatial_gating_unit.conv.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.proj_2.weight - torch.Size([128, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.attn.proj_2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.fc1.weight - torch.Size([1024, 128, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.fc1.bias - torch.Size([1024]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.dwconv.dwconv.weight - torch.Size([1024, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.dwconv.dwconv.bias - torch.Size([1024]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.fc2.weight - torch.Size([128, 1024, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block2.1.mlp.fc2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm2.weight - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm2.bias - torch.Size([128]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed3.proj.weight - torch.Size([320, 128, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed3.proj.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed3.norm.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed3.norm.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.layer_scale_1 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.layer_scale_2 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.norm1.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.norm1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.norm2.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.norm2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.proj_1.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.proj_1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv0.weight - torch.Size([320, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv0.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([320, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv1.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv1.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv2.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv2.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv.weight - torch.Size([320, 160, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.spatial_gating_unit.conv.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.proj_2.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.attn.proj_2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.fc1.weight - torch.Size([1280, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.fc1.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.dwconv.dwconv.weight - torch.Size([1280, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.dwconv.dwconv.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.fc2.weight - torch.Size([320, 1280, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.0.mlp.fc2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.layer_scale_1 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.layer_scale_2 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.norm1.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.norm1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.norm2.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.norm2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.proj_1.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.proj_1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv0.weight - torch.Size([320, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv0.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([320, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv1.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv1.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv2.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv2.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv.weight - torch.Size([320, 160, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.spatial_gating_unit.conv.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.proj_2.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.attn.proj_2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.fc1.weight - torch.Size([1280, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.fc1.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.dwconv.dwconv.weight - torch.Size([1280, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.dwconv.dwconv.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.fc2.weight - torch.Size([320, 1280, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.1.mlp.fc2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.layer_scale_1 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.layer_scale_2 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.norm1.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.norm1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.norm2.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.norm2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.proj_1.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.proj_1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv0.weight - torch.Size([320, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv0.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([320, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv1.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv1.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv2.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv2.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv.weight - torch.Size([320, 160, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.spatial_gating_unit.conv.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.proj_2.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.attn.proj_2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.fc1.weight - torch.Size([1280, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.fc1.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.dwconv.dwconv.weight - torch.Size([1280, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.dwconv.dwconv.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.fc2.weight - torch.Size([320, 1280, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.2.mlp.fc2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.layer_scale_1 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.layer_scale_2 - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.norm1.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.norm1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.norm2.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.norm2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.proj_1.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.proj_1.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv0.weight - torch.Size([320, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv0.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([320, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv1.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv1.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv2.weight - torch.Size([160, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv2.bias - torch.Size([160]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv.weight - torch.Size([320, 160, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.spatial_gating_unit.conv.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.proj_2.weight - torch.Size([320, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.attn.proj_2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.fc1.weight - torch.Size([1280, 320, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.fc1.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.dwconv.dwconv.weight - torch.Size([1280, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.dwconv.dwconv.bias - torch.Size([1280]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.fc2.weight - torch.Size([320, 1280, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block3.3.mlp.fc2.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm3.weight - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm3.bias - torch.Size([320]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed4.proj.weight - torch.Size([512, 320, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed4.proj.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed4.norm.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.patch_embed4.norm.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.layer_scale_1 - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.layer_scale_2 - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.norm1.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.norm1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.norm2.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.norm2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.proj_1.weight - torch.Size([512, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.proj_1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv0.weight - torch.Size([512, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv0.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([512, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv1.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv1.bias - torch.Size([256]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv2.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv2.bias - torch.Size([256]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv.weight - torch.Size([512, 256, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.spatial_gating_unit.conv.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.proj_2.weight - torch.Size([512, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.attn.proj_2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.fc1.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.fc1.bias - torch.Size([2048]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.dwconv.dwconv.weight - torch.Size([2048, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.dwconv.dwconv.bias - torch.Size([2048]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.fc2.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.0.mlp.fc2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.layer_scale_1 - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.layer_scale_2 - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.norm1.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.norm1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.norm2.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.norm2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.proj_1.weight - torch.Size([512, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.proj_1.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv0.weight - torch.Size([512, 1, 5, 5]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv0.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv_spatial.weight - torch.Size([512, 1, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv_spatial.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv1.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv1.bias - torch.Size([256]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv2.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv2.bias - torch.Size([256]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv_squeeze.weight - torch.Size([2, 2, 7, 7]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv_squeeze.bias - torch.Size([2]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv.weight - torch.Size([512, 256, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.spatial_gating_unit.conv.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.proj_2.weight - torch.Size([512, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.attn.proj_2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.fc1.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.fc1.bias - torch.Size([2048]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.dwconv.dwconv.weight - torch.Size([2048, 1, 3, 3]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.dwconv.dwconv.bias - torch.Size([2048]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.fc2.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.block4.1.mlp.fc2.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm4.weight - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar backbone.norm4.bias - torch.Size([512]): PretrainedInit: load from /opt/data/private/LYX/data/pretrained/lsk9_b1.pth.tar neck.lateral_convs.0.conv.weight - torch.Size([256, 64, 1, 1]): XavierInit: gain=1, distribution=uniform, bias=0 neck.lateral_convs.0.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.lateral_convs.1.conv.weight - torch.Size([256, 128, 1, 1]): XavierInit: gain=1, distribution=uniform, bias=0 neck.lateral_convs.1.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.lateral_convs.2.conv.weight - torch.Size([256, 320, 1, 1]): XavierInit: gain=1, distribution=uniform, bias=0 neck.lateral_convs.2.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.lateral_convs.3.conv.weight - torch.Size([256, 512, 1, 1]): XavierInit: gain=1, distribution=uniform, bias=0 neck.lateral_convs.3.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.fpn_convs.0.conv.weight - torch.Size([256, 256, 3, 3]): XavierInit: gain=1, distribution=uniform, bias=0 neck.fpn_convs.0.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.fpn_convs.1.conv.weight - torch.Size([256, 256, 3, 3]): XavierInit: gain=1, distribution=uniform, bias=0 neck.fpn_convs.1.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.fpn_convs.2.conv.weight - torch.Size([256, 256, 3, 3]): XavierInit: gain=1, distribution=uniform, bias=0 neck.fpn_convs.2.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN neck.fpn_convs.3.conv.weight - torch.Size([256, 256, 3, 3]): XavierInit: gain=1, distribution=uniform, bias=0 neck.fpn_convs.3.conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of OrientedRCNN rpn_head.rpn_conv.weight - torch.Size([256, 256, 3, 3]): NormalInit: mean=0, std=0.01, bias=0 rpn_head.rpn_conv.bias - torch.Size([256]): NormalInit: mean=0, std=0.01, bias=0 rpn_head.rpn_cls.weight - torch.Size([3, 256, 1, 1]): NormalInit: mean=0, std=0.01, bias=0 rpn_head.rpn_cls.bias - torch.Size([3]): NormalInit: mean=0, std=0.01, bias=0 rpn_head.rpn_reg.weight - torch.Size([18, 256, 1, 1]): NormalInit: mean=0, std=0.01, bias=0 rpn_head.rpn_reg.bias - torch.Size([18]): NormalInit: mean=0, std=0.01, bias=0 roi_head.bbox_head.fc_cls.weight - torch.Size([16, 1024]): XavierInit: gain=1, distribution=normal, bias=0 roi_head.bbox_head.fc_cls.bias - torch.Size([16]): NormalInit: mean=0, std=0.01, bias=0 roi_head.bbox_head.fc_reg.weight - torch.Size([5, 1024]): XavierInit: gain=1, distribution=normal, bias=0 roi_head.bbox_head.fc_reg.bias - torch.Size([5]): NormalInit: mean=0, std=0.001, bias=0 roi_head.bbox_head.shared_fcs.0.weight - torch.Size([1024, 12544]): XavierInit: gain=1, distribution=normal, bias=0 roi_head.bbox_head.shared_fcs.0.bias - torch.Size([1024]): XavierInit: gain=1, distribution=normal, bias=0 roi_head.bbox_head.shared_fcs.1.weight - torch.Size([1024, 1024]): XavierInit: gain=1, distribution=normal, bias=0 roi_head.bbox_head.shared_fcs.1.bias - torch.Size([1024]): XavierInit: gain=1, distribution=normal, bias=0 2023-01-16 12:33:37,784 - mmrotate - INFO - load checkpoint from local path: /opt/data/private/LYX/selective-large-kernel/work_dirs/lsk9b1_dota/epoch_6.pth 2023-01-16 12:33:38,425 - mmrotate - INFO - resumed epoch 6, iter 51246 2023-01-16 12:33:38,427 - mmrotate - INFO - Start running, host: root@interactive43916, work_dir: /opt/data/private/LYX/selective-large-kernel/work_dirs/lsk9b1_dota 2023-01-16 12:33:38,428 - mmrotate - INFO - Hooks will be executed in the following order: before_run: (VERY_HIGH ) StepLrUpdaterHook (ABOVE_NORMAL) Fp16OptimizerHook (NORMAL ) CheckpointHook (LOW ) DistEvalHook (VERY_LOW ) TextLoggerHook -------------------- before_train_epoch: (VERY_HIGH ) StepLrUpdaterHook (NORMAL ) DistSamplerSeedHook (LOW ) IterTimerHook (LOW ) DistEvalHook (VERY_LOW ) TextLoggerHook -------------------- before_train_iter: (VERY_HIGH ) StepLrUpdaterHook (LOW ) IterTimerHook (LOW ) DistEvalHook -------------------- after_train_iter: (ABOVE_NORMAL) Fp16OptimizerHook (NORMAL ) CheckpointHook (LOW ) IterTimerHook (LOW ) DistEvalHook (VERY_LOW ) TextLoggerHook -------------------- after_train_epoch: (NORMAL ) CheckpointHook (LOW ) DistEvalHook (VERY_LOW ) TextLoggerHook -------------------- before_val_epoch: (NORMAL ) DistSamplerSeedHook (LOW ) IterTimerHook (VERY_LOW ) TextLoggerHook -------------------- before_val_iter: (LOW ) IterTimerHook -------------------- after_val_iter: (LOW ) IterTimerHook -------------------- after_val_epoch: (VERY_LOW ) TextLoggerHook -------------------- after_run: (VERY_LOW ) TextLoggerHook -------------------- 2023-01-16 12:33:38,429 - mmrotate - INFO - workflow: [('train', 1)], max: 12 epochs 2023-01-16 12:33:38,431 - mmrotate - INFO - Checkpoints will be saved to /opt/data/private/LYX/selective-large-kernel/work_dirs/lsk9b1_dota by HardDiskBackend. 2023-01-16 12:39:05,808 - mmrotate - INFO - Epoch [7][500/8541] lr: 2.000e-04, eta: 9:13:43, time: 0.655, data_time: 0.011, memory: 7885, loss_rpn_cls: 0.0105, loss_rpn_bbox: 0.0442, loss_cls: 0.1072, acc: 95.6162, loss_bbox: 0.1051, loss: 0.2669, grad_norm: nan 2023-01-16 12:46:04,233 - mmrotate - INFO - Epoch [7][1000/8541] lr: 2.000e-04, eta: 10:24:31, time: 0.837, data_time: 0.004, memory: 7885, loss_rpn_cls: 0.0111, loss_rpn_bbox: 0.0467, loss_cls: 0.1067, acc: 95.6218, loss_bbox: 0.1053, loss: 0.2697, grad_norm: 1.4324 2023-01-16 12:51:26,788 - mmrotate - INFO - Epoch [7][1500/8541] lr: 2.000e-04, eta: 9:50:29, time: 0.645, data_time: 0.004, memory: 8506, loss_rpn_cls: 0.0110, loss_rpn_bbox: 0.0423, loss_cls: 0.1115, acc: 95.4699, loss_bbox: 0.1090, loss: 0.2738, grad_norm: inf 2023-01-16 12:56:51,889 - mmrotate - INFO - Epoch [7][2000/8541] lr: 2.000e-04, eta: 9:31:49, time: 0.650, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0138, loss_rpn_bbox: 0.0437, loss_cls: 0.1106, acc: 95.5330, loss_bbox: 0.1045, loss: 0.2726, grad_norm: 1.4713 2023-01-16 13:02:15,730 - mmrotate - INFO - Epoch [7][2500/8541] lr: 2.000e-04, eta: 9:18:03, time: 0.648, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0423, loss_cls: 0.1052, acc: 95.6837, loss_bbox: 0.1042, loss: 0.2619, grad_norm: 1.3958 2023-01-16 13:07:36,503 - mmrotate - INFO - Epoch [7][3000/8541] lr: 2.000e-04, eta: 9:06:15, time: 0.642, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0106, loss_rpn_bbox: 0.0445, loss_cls: 0.1064, acc: 95.6915, loss_bbox: 0.1049, loss: 0.2664, grad_norm: 1.3969 2023-01-16 13:13:24,577 - mmrotate - INFO - Epoch [7][3500/8541] lr: 2.000e-04, eta: 9:02:30, time: 0.696, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0109, loss_rpn_bbox: 0.0437, loss_cls: 0.1053, acc: 95.7007, loss_bbox: 0.1044, loss: 0.2643, grad_norm: 1.4337 2023-01-16 13:18:47,984 - mmrotate - INFO - Epoch [7][4000/8541] lr: 2.000e-04, eta: 8:53:23, time: 0.647, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0450, loss_cls: 0.1063, acc: 95.6348, loss_bbox: 0.1043, loss: 0.2660, grad_norm: 1.4196 2023-01-16 13:24:12,483 - mmrotate - INFO - Epoch [7][4500/8541] lr: 2.000e-04, eta: 8:45:17, time: 0.649, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0110, loss_rpn_bbox: 0.0429, loss_cls: 0.1097, acc: 95.4657, loss_bbox: 0.1067, loss: 0.2703, grad_norm: 1.4124 2023-01-16 13:29:36,003 - mmrotate - INFO - Epoch [7][5000/8541] lr: 2.000e-04, eta: 8:37:34, time: 0.647, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0468, loss_cls: 0.1080, acc: 95.5782, loss_bbox: 0.1065, loss: 0.2716, grad_norm: nan 2023-01-16 13:35:00,717 - mmrotate - INFO - Epoch [7][5500/8541] lr: 2.000e-04, eta: 8:30:26, time: 0.649, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0109, loss_rpn_bbox: 0.0423, loss_cls: 0.1071, acc: 95.6368, loss_bbox: 0.1042, loss: 0.2644, grad_norm: 1.3724 2023-01-16 13:40:24,364 - mmrotate - INFO - Epoch [7][6000/8541] lr: 2.000e-04, eta: 8:23:28, time: 0.647, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0101, loss_rpn_bbox: 0.0409, loss_cls: 0.1049, acc: 95.6752, loss_bbox: 0.1024, loss: 0.2584, grad_norm: 1.3724 2023-01-16 13:47:34,291 - mmrotate - INFO - Epoch [7][6500/8541] lr: 2.000e-04, eta: 8:28:56, time: 0.860, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0105, loss_rpn_bbox: 0.0427, loss_cls: 0.1050, acc: 95.6889, loss_bbox: inf, loss: inf, grad_norm: nan 2023-01-16 13:52:55,817 - mmrotate - INFO - Epoch [7][7000/8541] lr: 2.000e-04, eta: 8:21:10, time: 0.643, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0109, loss_rpn_bbox: 0.0447, loss_cls: 0.1084, acc: 95.5708, loss_bbox: 0.1056, loss: 0.2696, grad_norm: 1.3966 2023-01-16 13:58:16,875 - mmrotate - INFO - Epoch [7][7500/8541] lr: 2.000e-04, eta: 8:13:41, time: 0.642, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0098, loss_rpn_bbox: 0.0440, loss_cls: 0.1051, acc: 95.6967, loss_bbox: 0.1038, loss: 0.2626, grad_norm: 1.3773 2023-01-16 14:05:06,217 - mmrotate - INFO - Epoch [7][8000/8541] lr: 2.000e-04, eta: 8:14:25, time: 0.819, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0097, loss_rpn_bbox: 0.0404, loss_cls: 0.1043, acc: 95.7313, loss_bbox: 0.1019, loss: 0.2562, grad_norm: 1.3063 2023-01-16 14:10:29,634 - mmrotate - INFO - Epoch [7][8500/8541] lr: 2.000e-04, eta: 8:07:03, time: 0.647, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0102, loss_rpn_bbox: 0.0434, loss_cls: 0.1050, acc: 95.6685, loss_bbox: 0.1027, loss: 0.2613, grad_norm: 1.3818 2023-01-16 14:10:56,014 - mmrotate - INFO - Saving checkpoint at 7 epochs 2023-01-16 14:16:24,003 - mmrotate - INFO - Epoch [8][500/8541] lr: 2.000e-04, eta: 7:57:28, time: 0.652, data_time: 0.011, memory: 8731, loss_rpn_cls: 0.0094, loss_rpn_bbox: 0.0392, loss_cls: 0.1024, acc: 95.8203, loss_bbox: 0.1009, loss: 0.2519, grad_norm: 1.3635 2023-01-16 14:23:19,601 - mmrotate - INFO - Epoch [8][1000/8541] lr: 2.000e-04, eta: 7:57:21, time: 0.831, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0106, loss_rpn_bbox: 0.0432, loss_cls: 0.1011, acc: 95.8651, loss_bbox: 0.1000, loss: 0.2550, grad_norm: 1.3537 2023-01-16 14:28:42,933 - mmrotate - INFO - Epoch [8][1500/8541] lr: 2.000e-04, eta: 7:50:16, time: 0.647, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0103, loss_rpn_bbox: 0.0442, loss_cls: 0.1049, acc: 95.6736, loss_bbox: 0.1033, loss: 0.2627, grad_norm: 1.3740 2023-01-16 14:34:07,620 - mmrotate - INFO - Epoch [8][2000/8541] lr: 2.000e-04, eta: 7:43:25, time: 0.649, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0099, loss_rpn_bbox: 0.0421, loss_cls: 0.1040, acc: 95.7295, loss_bbox: 0.1033, loss: 0.2592, grad_norm: nan 2023-01-16 14:41:58,429 - mmrotate - INFO - Epoch [8][2500/8541] lr: 2.000e-04, eta: 7:45:34, time: 0.942, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0112, loss_rpn_bbox: 0.0420, loss_cls: 0.1028, acc: 95.7882, loss_bbox: 0.1025, loss: 0.2585, grad_norm: 1.3895 2023-01-16 14:47:21,354 - mmrotate - INFO - Epoch [8][3000/8541] lr: 2.000e-04, eta: 7:38:22, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0110, loss_rpn_bbox: 0.0433, loss_cls: 0.1062, acc: 95.6355, loss_bbox: 0.1027, loss: 0.2633, grad_norm: 1.4000 2023-01-16 14:52:44,120 - mmrotate - INFO - Epoch [8][3500/8541] lr: 2.000e-04, eta: 7:31:19, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0098, loss_rpn_bbox: 0.0414, loss_cls: 0.1041, acc: 95.7422, loss_bbox: 0.1027, loss: 0.2579, grad_norm: 1.3671 2023-01-16 14:58:06,984 - mmrotate - INFO - Epoch [8][4000/8541] lr: 2.000e-04, eta: 7:24:25, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0109, loss_rpn_bbox: 0.0432, loss_cls: 0.1076, acc: 95.6352, loss_bbox: 0.1036, loss: 0.2652, grad_norm: nan 2023-01-16 15:03:29,883 - mmrotate - INFO - Epoch [8][4500/8541] lr: 2.000e-04, eta: 7:17:37, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0107, loss_rpn_bbox: 0.0427, loss_cls: 0.1063, acc: 95.6534, loss_bbox: 0.1035, loss: 0.2631, grad_norm: 1.3996 2023-01-16 15:08:52,864 - mmrotate - INFO - Epoch [8][5000/8541] lr: 2.000e-04, eta: 7:10:56, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0103, loss_rpn_bbox: 0.0404, loss_cls: 0.1023, acc: 95.8203, loss_bbox: 0.1014, loss: 0.2544, grad_norm: 1.3778 2023-01-16 15:14:15,089 - mmrotate - INFO - Epoch [8][5500/8541] lr: 2.000e-04, eta: 7:04:18, time: 0.644, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0097, loss_rpn_bbox: 0.0399, loss_cls: 0.1007, acc: 95.8857, loss_bbox: 0.1004, loss: 0.2507, grad_norm: 1.3196 2023-01-16 15:20:46,172 - mmrotate - INFO - Epoch [8][6000/8541] lr: 2.000e-04, eta: 7:00:40, time: 0.782, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0097, loss_rpn_bbox: 0.0434, loss_cls: 0.1052, acc: 95.7198, loss_bbox: inf, loss: inf, grad_norm: nan 2023-01-16 15:26:09,186 - mmrotate - INFO - Epoch [8][6500/8541] lr: 2.000e-04, eta: 6:54:06, time: 0.646, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0100, loss_rpn_bbox: 0.0444, loss_cls: 0.1052, acc: 95.6830, loss_bbox: 0.1032, loss: 0.2628, grad_norm: 1.3739 2023-01-16 15:31:33,359 - mmrotate - INFO - Epoch [8][7000/8541] lr: 2.000e-04, eta: 6:47:39, time: 0.648, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0096, loss_rpn_bbox: 0.0410, loss_cls: 0.1029, acc: 95.7716, loss_bbox: 0.1022, loss: 0.2557, grad_norm: 1.3359 2023-01-16 15:36:57,989 - mmrotate - INFO - Epoch [8][7500/8541] lr: 2.000e-04, eta: 6:41:17, time: 0.649, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0099, loss_rpn_bbox: 0.0425, loss_cls: 0.1052, acc: 95.6907, loss_bbox: 0.1017, loss: 0.2593, grad_norm: 1.3618 2023-01-16 15:42:22,039 - mmrotate - INFO - Epoch [8][8000/8541] lr: 2.000e-04, eta: 6:34:58, time: 0.648, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0104, loss_rpn_bbox: 0.0409, loss_cls: 0.1025, acc: 95.8146, loss_bbox: 0.1009, loss: 0.2547, grad_norm: 1.3687 2023-01-16 15:48:43,687 - mmrotate - INFO - Epoch [8][8500/8541] lr: 2.000e-04, eta: 6:30:37, time: 0.763, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0097, loss_rpn_bbox: 0.0436, loss_cls: 0.1059, acc: 95.6753, loss_bbox: 0.1047, loss: 0.2639, grad_norm: 1.3812 2023-01-16 15:49:10,349 - mmrotate - INFO - Saving checkpoint at 8 epochs 2023-01-16 15:54:40,782 - mmrotate - INFO - Epoch [9][500/8541] lr: 2.000e-05, eta: 6:23:05, time: 0.656, data_time: 0.011, memory: 8731, loss_rpn_cls: 0.0074, loss_rpn_bbox: 0.0385, loss_cls: 0.0928, acc: 96.1410, loss_bbox: 0.0945, loss: 0.2332, grad_norm: 1.1430 2023-01-16 16:01:53,062 - mmrotate - INFO - Epoch [9][1000/8541] lr: 2.000e-05, eta: 6:20:10, time: 0.865, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0074, loss_rpn_bbox: 0.0372, loss_cls: 0.0905, acc: 96.2472, loss_bbox: 0.0949, loss: 0.2301, grad_norm: 1.1428 2023-01-16 16:07:15,189 - mmrotate - INFO - Epoch [9][1500/8541] lr: 2.000e-05, eta: 6:13:48, time: 0.644, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0072, loss_rpn_bbox: 0.0371, loss_cls: 0.0888, acc: 96.3172, loss_bbox: 0.0923, loss: 0.2253, grad_norm: 1.1133 2023-01-16 16:14:54,396 - mmrotate - INFO - Epoch [9][2000/8541] lr: 2.000e-05, eta: 6:11:20, time: 0.918, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0069, loss_rpn_bbox: 0.0369, loss_cls: 0.0879, acc: 96.3652, loss_bbox: 0.0909, loss: 0.2226, grad_norm: 1.1200 2023-01-16 16:20:16,393 - mmrotate - INFO - Epoch [9][2500/8541] lr: 2.000e-05, eta: 6:04:54, time: 0.644, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0067, loss_rpn_bbox: 0.0361, loss_cls: 0.0862, acc: 96.4301, loss_bbox: 0.0911, loss: 0.2201, grad_norm: 1.1152 2023-01-16 16:25:38,761 - mmrotate - INFO - Epoch [9][3000/8541] lr: 2.000e-05, eta: 5:58:32, time: 0.645, data_time: 0.004, memory: 8731, loss_rpn_cls: 0.0060, loss_rpn_bbox: 0.0315, loss_cls: 0.0842, acc: 96.5156, loss_bbox: 0.0878, loss: 0.2095, grad_norm: 1.0915 2023-01-16 16:30:59,471 - mmrotate - INFO - Epoch [9][3500/8541] lr: 2.000e-05, eta: 5:52:10, time: 0.641, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0061, loss_rpn_bbox: 0.0327, loss_cls: 0.0834, acc: 96.5097, loss_bbox: 0.0883, loss: 0.2105, grad_norm: inf 2023-01-16 16:36:23,617 - mmrotate - INFO - Epoch [9][4000/8541] lr: 2.000e-05, eta: 5:45:57, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0061, loss_rpn_bbox: 0.0336, loss_cls: 0.0856, acc: 96.4224, loss_bbox: 0.0897, loss: 0.2150, grad_norm: nan 2023-01-16 16:41:45,455 - mmrotate - INFO - Epoch [9][4500/8541] lr: 2.000e-05, eta: 5:39:42, time: 0.644, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0060, loss_rpn_bbox: 0.0331, loss_cls: 0.0845, acc: 96.4795, loss_bbox: 0.0877, loss: 0.2113, grad_norm: 1.0898 2023-01-16 16:48:16,829 - mmrotate - INFO - Epoch [9][5000/8541] lr: 2.000e-05, eta: 5:35:02, time: 0.783, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0059, loss_rpn_bbox: 0.0319, loss_cls: 0.0836, acc: 96.5063, loss_bbox: 0.0880, loss: 0.2094, grad_norm: 1.0947 2023-01-16 16:54:52,732 - mmrotate - INFO - Epoch [9][5500/8541] lr: 2.000e-05, eta: 5:30:22, time: 0.792, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0061, loss_rpn_bbox: 0.0336, loss_cls: 0.0819, acc: 96.6006, loss_bbox: 0.0881, loss: 0.2098, grad_norm: 1.1083 2023-01-16 17:00:14,844 - mmrotate - INFO - Epoch [9][6000/8541] lr: 2.000e-05, eta: 5:24:07, time: 0.644, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0061, loss_rpn_bbox: 0.0315, loss_cls: 0.0838, acc: 96.5078, loss_bbox: 0.0875, loss: 0.2090, grad_norm: 1.1165 2023-01-16 17:06:32,113 - mmrotate - INFO - Epoch [9][6500/8541] lr: 2.000e-05, eta: 5:19:00, time: 0.755, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0062, loss_rpn_bbox: 0.0346, loss_cls: 0.0828, acc: 96.5755, loss_bbox: 0.0880, loss: 0.2116, grad_norm: 1.1218 2023-01-16 17:11:54,842 - mmrotate - INFO - Epoch [9][7000/8541] lr: 2.000e-05, eta: 5:12:48, time: 0.645, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0054, loss_rpn_bbox: 0.0328, loss_cls: 0.0807, acc: 96.6354, loss_bbox: 0.0867, loss: 0.2056, grad_norm: nan 2023-01-16 17:17:17,714 - mmrotate - INFO - Epoch [9][7500/8541] lr: 2.000e-05, eta: 5:06:38, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0059, loss_rpn_bbox: 0.0340, loss_cls: 0.0803, acc: 96.6694, loss_bbox: 0.0847, loss: 0.2049, grad_norm: 1.1012 2023-01-16 17:22:40,906 - mmrotate - INFO - Epoch [9][8000/8541] lr: 2.000e-05, eta: 5:00:30, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0061, loss_rpn_bbox: 0.0325, loss_cls: 0.0810, acc: 96.6431, loss_bbox: 0.0857, loss: 0.2054, grad_norm: 1.1013 2023-01-16 17:28:03,997 - mmrotate - INFO - Epoch [9][8500/8541] lr: 2.000e-05, eta: 4:54:24, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0058, loss_rpn_bbox: 0.0318, loss_cls: 0.0791, acc: 96.7154, loss_bbox: 0.0848, loss: 0.2015, grad_norm: 1.1058 2023-01-16 17:28:30,860 - mmrotate - INFO - Saving checkpoint at 9 epochs 2023-01-16 17:30:59,400 - mmrotate - INFO - +--------------------+-------+-------+--------+-------+ | class | gts | dets | recall | ap | +--------------------+-------+-------+--------+-------+ | plane | 23860 | 27080 | 0.959 | 0.907 | | baseball-diamond | 1727 | 2833 | 0.968 | 0.901 | | bridge | 4142 | 8929 | 0.852 | 0.758 | | ground-track-field | 1134 | 2865 | 0.991 | 0.893 | | small-vehicle | 33183 | 56725 | 0.898 | 0.772 | | large-vehicle | 29737 | 48279 | 0.968 | 0.892 | | ship | 80574 | 91330 | 0.929 | 0.900 | | tennis-court | 4389 | 5404 | 0.987 | 0.909 | | basketball-court | 1097 | 2165 | 0.994 | 0.894 | | storage-tank | 28751 | 26589 | 0.710 | 0.712 | | soccer-ball-field | 1242 | 2855 | 0.928 | 0.871 | | roundabout | 1536 | 3429 | 0.926 | 0.892 | | harbor | 15489 | 22012 | 0.922 | 0.876 | | swimming-pool | 3456 | 6009 | 0.884 | 0.752 | | helicopter | 765 | 1386 | 0.945 | 0.897 | +--------------------+-------+-------+--------+-------+ | mAP | | | | 0.855 | +--------------------+-------+-------+--------+-------+ 2023-01-16 17:30:59,532 - mmrotate - INFO - Exp name: lsk9b1_dota.py 2023-01-16 17:30:59,533 - mmrotate - INFO - Epoch(val) [9][2034] mAP: 0.8551 2023-01-16 17:38:43,599 - mmrotate - INFO - Epoch [10][500/8541] lr: 2.000e-05, eta: 4:49:40, time: 0.928, data_time: 0.011, memory: 9073, loss_rpn_cls: 0.0058, loss_rpn_bbox: 0.0326, loss_cls: 0.0806, acc: 96.6435, loss_bbox: 0.0849, loss: 0.2039, grad_norm: 1.1052 2023-01-16 17:44:24,905 - mmrotate - INFO - Epoch [10][1000/8541] lr: 2.000e-05, eta: 4:43:49, time: 0.683, data_time: 0.005, memory: 9073, loss_rpn_cls: 0.0055, loss_rpn_bbox: 0.0311, loss_cls: 0.0792, acc: 96.7005, loss_bbox: 0.0853, loss: 0.2011, grad_norm: 1.0827 2023-01-16 17:49:49,417 - mmrotate - INFO - Epoch [10][1500/8541] lr: 2.000e-05, eta: 4:37:45, time: 0.649, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0058, loss_rpn_bbox: 0.0316, loss_cls: 0.0800, acc: 96.6868, loss_bbox: 0.0855, loss: 0.2028, grad_norm: 1.1367 2023-01-16 17:55:13,052 - mmrotate - INFO - Epoch [10][2000/8541] lr: 2.000e-05, eta: 4:31:41, time: 0.647, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0059, loss_rpn_bbox: 0.0313, loss_cls: 0.0796, acc: 96.6879, loss_bbox: 0.0852, loss: 0.2020, grad_norm: 1.1216 2023-01-16 18:00:35,483 - mmrotate - INFO - Epoch [10][2500/8541] lr: 2.000e-05, eta: 4:25:37, time: 0.645, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0056, loss_rpn_bbox: 0.0323, loss_cls: 0.0792, acc: 96.7198, loss_bbox: 0.0853, loss: 0.2025, grad_norm: 1.1194 2023-01-16 18:06:01,246 - mmrotate - INFO - Epoch [10][3000/8541] lr: 2.000e-05, eta: 4:19:37, time: 0.651, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0054, loss_rpn_bbox: 0.0330, loss_cls: 0.0800, acc: 96.6473, loss_bbox: 0.0846, loss: 0.2030, grad_norm: nan 2023-01-16 18:11:26,322 - mmrotate - INFO - Epoch [10][3500/8541] lr: 2.000e-05, eta: 4:13:39, time: 0.650, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0057, loss_rpn_bbox: 0.0327, loss_cls: 0.0785, acc: 96.7331, loss_bbox: 0.0852, loss: 0.2021, grad_norm: 1.1059 2023-01-16 18:16:48,953 - mmrotate - INFO - Epoch [10][4000/8541] lr: 2.000e-05, eta: 4:07:39, time: 0.645, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0326, loss_cls: 0.0780, acc: 96.7281, loss_bbox: 0.0858, loss: 0.2017, grad_norm: 1.0937 2023-01-16 18:24:38,570 - mmrotate - INFO - Epoch [10][4500/8541] lr: 2.000e-05, eta: 4:03:24, time: 0.939, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0058, loss_rpn_bbox: 0.0321, loss_cls: 0.0762, acc: 96.8343, loss_bbox: 0.0842, loss: 0.1982, grad_norm: nan 2023-01-16 18:30:03,805 - mmrotate - INFO - Epoch [10][5000/8541] lr: 2.000e-05, eta: 3:57:24, time: 0.650, data_time: 0.005, memory: 9073, loss_rpn_cls: 0.0054, loss_rpn_bbox: 0.0324, loss_cls: 0.0790, acc: 96.6994, loss_bbox: 0.0846, loss: 0.2014, grad_norm: 1.1352 2023-01-16 18:35:29,870 - mmrotate - INFO - Epoch [10][5500/8541] lr: 2.000e-05, eta: 3:51:26, time: 0.652, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0055, loss_rpn_bbox: 0.0341, loss_cls: 0.0778, acc: 96.7485, loss_bbox: 0.0853, loss: 0.2027, grad_norm: 1.1262 2023-01-16 18:41:44,403 - mmrotate - INFO - Epoch [10][6000/8541] lr: 2.000e-05, eta: 3:46:00, time: 0.749, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0052, loss_rpn_bbox: 0.0327, loss_cls: 0.0761, acc: 96.8224, loss_bbox: 0.0837, loss: 0.1976, grad_norm: 1.1175 2023-01-16 18:47:08,592 - mmrotate - INFO - Epoch [10][6500/8541] lr: 2.000e-05, eta: 3:40:01, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0327, loss_cls: 0.0762, acc: 96.8398, loss_bbox: 0.0840, loss: 0.1979, grad_norm: 1.0829 2023-01-16 18:52:31,522 - mmrotate - INFO - Epoch [10][7000/8541] lr: 2.000e-05, eta: 3:34:04, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0055, loss_rpn_bbox: 0.0323, loss_cls: 0.0760, acc: 96.8362, loss_bbox: 0.0839, loss: 0.1977, grad_norm: 1.1215 2023-01-16 18:57:58,134 - mmrotate - INFO - Epoch [10][7500/8541] lr: 2.000e-05, eta: 3:28:09, time: 0.653, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0057, loss_rpn_bbox: 0.0327, loss_cls: 0.0770, acc: 96.8139, loss_bbox: 0.0831, loss: 0.1985, grad_norm: 1.1293 2023-01-16 19:03:24,335 - mmrotate - INFO - Epoch [10][8000/8541] lr: 2.000e-05, eta: 3:22:15, time: 0.652, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0055, loss_rpn_bbox: 0.0317, loss_cls: 0.0762, acc: 96.8146, loss_bbox: 0.0830, loss: 0.1963, grad_norm: 1.1022 2023-01-16 19:08:48,077 - mmrotate - INFO - Epoch [10][8500/8541] lr: 2.000e-05, eta: 3:16:20, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0048, loss_rpn_bbox: 0.0318, loss_cls: 0.0751, acc: 96.8808, loss_bbox: 0.0825, loss: 0.1943, grad_norm: 1.1023 2023-01-16 19:09:14,457 - mmrotate - INFO - Saving checkpoint at 10 epochs 2023-01-16 19:15:49,153 - mmrotate - INFO - Epoch [11][500/8541] lr: 2.000e-05, eta: 3:10:17, time: 0.785, data_time: 0.011, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0324, loss_cls: 0.0758, acc: 96.8466, loss_bbox: 0.0823, loss: 0.1957, grad_norm: inf 2023-01-16 19:21:13,370 - mmrotate - INFO - Epoch [11][1000/8541] lr: 2.000e-05, eta: 3:04:24, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0313, loss_cls: 0.0765, acc: 96.8116, loss_bbox: 0.0829, loss: 0.1959, grad_norm: 1.0924 2023-01-16 19:26:36,392 - mmrotate - INFO - Epoch [11][1500/8541] lr: 2.000e-05, eta: 2:58:31, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0315, loss_cls: 0.0740, acc: 96.9110, loss_bbox: 0.0810, loss: 0.1915, grad_norm: 1.1196 2023-01-16 19:31:59,562 - mmrotate - INFO - Epoch [11][2000/8541] lr: 2.000e-05, eta: 2:52:39, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0314, loss_cls: 0.0760, acc: 96.8350, loss_bbox: 0.0838, loss: 0.1965, grad_norm: 1.1304 2023-01-16 19:38:35,140 - mmrotate - INFO - Epoch [11][2500/8541] lr: 2.000e-05, eta: 2:47:16, time: 0.791, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0048, loss_rpn_bbox: 0.0319, loss_cls: 0.0763, acc: 96.8197, loss_bbox: 0.0844, loss: 0.1973, grad_norm: nan 2023-01-16 19:43:58,171 - mmrotate - INFO - Epoch [11][3000/8541] lr: 2.000e-05, eta: 2:41:24, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0054, loss_rpn_bbox: 0.0321, loss_cls: 0.0769, acc: 96.7816, loss_bbox: 0.0843, loss: 0.1987, grad_norm: 1.1281 2023-01-16 19:49:21,000 - mmrotate - INFO - Epoch [11][3500/8541] lr: 2.000e-05, eta: 2:35:32, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0323, loss_cls: 0.0746, acc: 96.9006, loss_bbox: 0.0836, loss: 0.1958, grad_norm: 1.1490 2023-01-16 19:54:43,274 - mmrotate - INFO - Epoch [11][4000/8541] lr: 2.000e-05, eta: 2:29:41, time: 0.645, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0303, loss_cls: 0.0752, acc: 96.8709, loss_bbox: 0.0818, loss: 0.1926, grad_norm: 1.1065 2023-01-16 20:00:05,586 - mmrotate - INFO - Epoch [11][4500/8541] lr: 2.000e-05, eta: 2:23:51, time: 0.645, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0319, loss_cls: 0.0745, acc: 96.9051, loss_bbox: 0.0810, loss: 0.1925, grad_norm: 1.1274 2023-01-16 20:05:27,834 - mmrotate - INFO - Epoch [11][5000/8541] lr: 2.000e-05, eta: 2:18:02, time: 0.644, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0057, loss_rpn_bbox: 0.0311, loss_cls: 0.0757, acc: 96.8461, loss_bbox: 0.0827, loss: 0.1951, grad_norm: 1.1321 2023-01-16 20:10:49,129 - mmrotate - INFO - Epoch [11][5500/8541] lr: 2.000e-05, eta: 2:12:13, time: 0.643, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0309, loss_cls: 0.0738, acc: 96.9301, loss_bbox: 0.0822, loss: 0.1920, grad_norm: 1.1026 2023-01-16 20:16:12,032 - mmrotate - INFO - Epoch [11][6000/8541] lr: 2.000e-05, eta: 2:06:25, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0051, loss_rpn_bbox: 0.0318, loss_cls: 0.0757, acc: 96.8518, loss_bbox: 0.0836, loss: 0.1962, grad_norm: inf 2023-01-16 20:21:36,197 - mmrotate - INFO - Epoch [11][6500/8541] lr: 2.000e-05, eta: 2:00:38, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0050, loss_rpn_bbox: 0.0303, loss_cls: 0.0749, acc: 96.9027, loss_bbox: 0.0817, loss: 0.1919, grad_norm: 1.1132 2023-01-16 20:26:59,265 - mmrotate - INFO - Epoch [11][7000/8541] lr: 2.000e-05, eta: 1:54:51, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0052, loss_rpn_bbox: 0.0315, loss_cls: 0.0753, acc: 96.8602, loss_bbox: 0.0823, loss: 0.1943, grad_norm: 1.1265 2023-01-16 20:32:20,730 - mmrotate - INFO - Epoch [11][7500/8541] lr: 2.000e-05, eta: 1:49:05, time: 0.643, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0048, loss_rpn_bbox: 0.0306, loss_cls: 0.0743, acc: 96.9041, loss_bbox: 0.0821, loss: 0.1917, grad_norm: 1.1034 2023-01-16 20:40:07,789 - mmrotate - INFO - Epoch [11][8000/8541] lr: 2.000e-05, eta: 1:43:50, time: 0.934, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0052, loss_rpn_bbox: 0.0321, loss_cls: 0.0742, acc: 96.8940, loss_bbox: 0.0814, loss: 0.1928, grad_norm: nan 2023-01-16 20:46:25,786 - mmrotate - INFO - Epoch [11][8500/8541] lr: 2.000e-05, eta: 1:38:14, time: 0.756, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0052, loss_rpn_bbox: 0.0308, loss_cls: 0.0746, acc: 96.8976, loss_bbox: 0.0812, loss: 0.1918, grad_norm: 1.1191 2023-01-16 20:46:52,348 - mmrotate - INFO - Saving checkpoint at 11 epochs 2023-01-16 20:52:20,015 - mmrotate - INFO - Epoch [12][500/8541] lr: 2.000e-06, eta: 1:31:54, time: 0.652, data_time: 0.011, memory: 9073, loss_rpn_cls: 0.0050, loss_rpn_bbox: 0.0305, loss_cls: 0.0727, acc: 96.9553, loss_bbox: 0.0816, loss: 0.1899, grad_norm: 1.1240 2023-01-16 20:57:42,852 - mmrotate - INFO - Epoch [12][1000/8541] lr: 2.000e-06, eta: 1:26:08, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0049, loss_rpn_bbox: 0.0302, loss_cls: 0.0750, acc: 96.8888, loss_bbox: 0.0825, loss: 0.1927, grad_norm: 1.1105 2023-01-16 21:03:07,000 - mmrotate - INFO - Epoch [12][1500/8541] lr: 2.000e-06, eta: 1:20:22, time: 0.648, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0046, loss_rpn_bbox: 0.0310, loss_cls: 0.0742, acc: 96.8938, loss_bbox: 0.0802, loss: 0.1900, grad_norm: nan 2023-01-16 21:08:30,546 - mmrotate - INFO - Epoch [12][2000/8541] lr: 2.000e-06, eta: 1:14:37, time: 0.647, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0049, loss_rpn_bbox: 0.0293, loss_cls: 0.0730, acc: 96.9900, loss_bbox: 0.0806, loss: 0.1879, grad_norm: 1.0920 2023-01-16 21:13:53,616 - mmrotate - INFO - Epoch [12][2500/8541] lr: 2.000e-06, eta: 1:08:52, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0050, loss_rpn_bbox: 0.0302, loss_cls: 0.0735, acc: 96.9462, loss_bbox: 0.0810, loss: 0.1897, grad_norm: 1.0981 2023-01-16 21:19:18,057 - mmrotate - INFO - Epoch [12][3000/8541] lr: 2.000e-06, eta: 1:03:08, time: 0.649, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0050, loss_rpn_bbox: 0.0296, loss_cls: 0.0734, acc: 96.9385, loss_bbox: 0.0800, loss: 0.1880, grad_norm: nan 2023-01-16 21:25:09,660 - mmrotate - INFO - Epoch [12][3500/8541] lr: 2.000e-06, eta: 0:57:27, time: 0.703, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0049, loss_rpn_bbox: 0.0318, loss_cls: 0.0740, acc: 96.9138, loss_bbox: 0.0828, loss: 0.1934, grad_norm: 1.0912 2023-01-16 21:30:32,871 - mmrotate - INFO - Epoch [12][4000/8541] lr: 2.000e-06, eta: 0:51:43, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0049, loss_rpn_bbox: 0.0310, loss_cls: 0.0720, acc: 97.0017, loss_bbox: 0.0800, loss: 0.1880, grad_norm: 1.0930 2023-01-16 21:35:57,899 - mmrotate - INFO - Epoch [12][4500/8541] lr: 2.000e-06, eta: 0:46:00, time: 0.650, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0043, loss_rpn_bbox: 0.0296, loss_cls: 0.0708, acc: 97.0672, loss_bbox: 0.0782, loss: 0.1829, grad_norm: 1.1069 2023-01-16 21:41:19,440 - mmrotate - INFO - Epoch [12][5000/8541] lr: 2.000e-06, eta: 0:40:17, time: 0.643, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0047, loss_rpn_bbox: 0.0314, loss_cls: 0.0718, acc: 97.0031, loss_bbox: 0.0799, loss: 0.1877, grad_norm: 1.1091 2023-01-16 21:46:40,924 - mmrotate - INFO - Epoch [12][5500/8541] lr: 2.000e-06, eta: 0:34:34, time: 0.643, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0046, loss_rpn_bbox: 0.0300, loss_cls: 0.0713, acc: 97.0208, loss_bbox: 0.0791, loss: 0.1851, grad_norm: 1.0825 2023-01-16 21:53:27,884 - mmrotate - INFO - Epoch [12][6000/8541] lr: 2.000e-06, eta: 0:28:57, time: 0.814, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0047, loss_rpn_bbox: 0.0305, loss_cls: 0.0728, acc: 96.9530, loss_bbox: 0.0806, loss: 0.1885, grad_norm: 1.0923 2023-01-16 21:58:51,175 - mmrotate - INFO - Epoch [12][6500/8541] lr: 2.000e-06, eta: 0:23:14, time: 0.647, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0047, loss_rpn_bbox: 0.0305, loss_cls: 0.0715, acc: 97.0273, loss_bbox: 0.0797, loss: 0.1864, grad_norm: 1.0857 2023-01-16 22:06:36,909 - mmrotate - INFO - Epoch [12][7000/8541] lr: 2.000e-06, eta: 0:17:36, time: 0.931, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0048, loss_rpn_bbox: 0.0333, loss_cls: 0.0734, acc: 96.9495, loss_bbox: 0.0816, loss: 0.1931, grad_norm: 1.1157 2023-01-16 22:11:59,823 - mmrotate - INFO - Epoch [12][7500/8541] lr: 2.000e-06, eta: 0:11:53, time: 0.646, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0049, loss_rpn_bbox: 0.0286, loss_cls: 0.0718, acc: 97.0180, loss_bbox: 0.0807, loss: 0.1860, grad_norm: nan 2023-01-16 22:17:25,330 - mmrotate - INFO - Epoch [12][8000/8541] lr: 2.000e-06, eta: 0:06:10, time: 0.651, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0048, loss_rpn_bbox: 0.0309, loss_cls: 0.0731, acc: 96.9715, loss_bbox: 0.0821, loss: 0.1909, grad_norm: 1.1237 2023-01-16 22:24:12,119 - mmrotate - INFO - Epoch [12][8500/8541] lr: 2.000e-06, eta: 0:00:28, time: 0.814, data_time: 0.004, memory: 9073, loss_rpn_cls: 0.0053, loss_rpn_bbox: 0.0322, loss_cls: 0.0756, acc: 96.8657, loss_bbox: 0.0829, loss: 0.1960, grad_norm: 1.1431 2023-01-16 22:24:38,964 - mmrotate - INFO - Saving checkpoint at 12 epochs 2023-01-16 22:27:11,918 - mmrotate - INFO - +--------------------+-------+-------+--------+-------+ | class | gts | dets | recall | ap | +--------------------+-------+-------+--------+-------+ | plane | 23860 | 26028 | 0.960 | 0.908 | | baseball-diamond | 1727 | 2300 | 0.976 | 0.904 | | bridge | 4142 | 7692 | 0.862 | 0.781 | | ground-track-field | 1134 | 2453 | 0.994 | 0.898 | | small-vehicle | 33183 | 48586 | 0.895 | 0.783 | | large-vehicle | 29737 | 46370 | 0.972 | 0.896 | | ship | 80574 | 87598 | 0.922 | 0.901 | | tennis-court | 4389 | 5216 | 0.991 | 0.908 | | basketball-court | 1097 | 1775 | 0.996 | 0.902 | | storage-tank | 28751 | 24724 | 0.696 | 0.635 | | soccer-ball-field | 1242 | 2400 | 0.935 | 0.884 | | roundabout | 1536 | 2408 | 0.932 | 0.902 | | harbor | 15489 | 20989 | 0.929 | 0.883 | | swimming-pool | 3456 | 5275 | 0.883 | 0.766 | | helicopter | 765 | 1123 | 0.937 | 0.903 | +--------------------+-------+-------+--------+-------+ | mAP | | | | 0.857 | +--------------------+-------+-------+--------+-------+ 2023-01-16 22:27:12,026 - mmrotate - INFO - Exp name: lsk9b1_dota.py 2023-01-16 22:27:12,027 - mmrotate - INFO - Epoch(val) [12][2034] mAP: 0.8569