HTTP/2 302 date: Sat, 03 Jan 2026 13:45:55 GMT content-type: text/html; charset=utf-8 content-length: 0 vary: X-PJAX, X-PJAX-Container, Turbo-Visit, Turbo-Frame, X-Requested-With,Accept-Encoding, Accept, X-Requested-With location: https://release-assets.githubusercontent.com/github-production-release-asset/546174873/ca0d5428-fa42-465b-82c9-15c4bbb148e4?sp=r&sv=2018-11-09&sr=b&spr=https&se=2026-01-03T14%3A38%3A47Z&rscd=attachment%3B+filename%3Dr50-fp16_phase2.log&rsct=application%2Foctet-stream&skoid=96c2d410-5711-43a1-aedd-ab1947aa7ab0&sktid=398a6654-997b-47e9-b12b-9515b896b4de&skt=2026-01-03T13%3A38%3A15Z&ske=2026-01-03T14%3A38%3A47Z&sks=b&skv=2018-11-09&sig=XswilnpxXypa%2F45aYGTWEj4NQR6KV1JVODjORviXCgw%3D&jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmVsZWFzZS1hc3NldHMuZ2l0aHVidXNlcmNvbnRlbnQuY29tIiwia2V5Ijoia2V5MSIsImV4cCI6MTc2NzQ0ODI1NSwibmJmIjoxNzY3NDQ3OTU1LCJwYXRoIjoicmVsZWFzZWFzc2V0cHJvZHVjdGlvbi5ibG9iLmNvcmUud2luZG93cy5uZXQifQ.FxrOhzjZEx47KA7y_813JSjiRqiMOhL7j6yy8oyQWd4&response-content-disposition=attachment%3B%20filename%3Dr50-fp16_phase2.log&response-content-type=application%2Foctet-stream cache-control: no-cache strict-transport-security: max-age=31536000; includeSubdomains; preload x-frame-options: deny x-content-type-options: nosniff x-xss-protection: 0 referrer-policy: no-referrer-when-downgrade content-security-policy: default-src 'none'; base-uri 'self'; child-src github.githubassets.com github.com/assets-cdn/worker/ github.com/assets/ gist.github.com/assets-cdn/worker/; connect-src 'self' uploads.github.com www.githubstatus.com collector.github.com raw.githubusercontent.com api.github.com github-cloud.s3.amazonaws.com github-production-repository-file-5c1aeb.s3.amazonaws.com github-production-upload-manifest-file-7fdce7.s3.amazonaws.com github-production-user-asset-6210df.s3.amazonaws.com *.rel.tunnels.api.visualstudio.com wss://*.rel.tunnels.api.visualstudio.com github.githubassets.com objects-origin.githubusercontent.com copilot-proxy.githubusercontent.com proxy.individual.githubcopilot.com proxy.business.githubcopilot.com proxy.enterprise.githubcopilot.com *.actions.githubusercontent.com wss://*.actions.githubusercontent.com productionresultssa0.blob.core.windows.net/ productionresultssa1.blob.core.windows.net/ productionresultssa2.blob.core.windows.net/ productionresultssa3.blob.core.windows.net/ productionresultssa4.blob.core.windows.net/ productionresultssa5.blob.core.windows.net/ productionresultssa6.blob.core.windows.net/ productionresultssa7.blob.core.windows.net/ productionresultssa8.blob.core.windows.net/ productionresultssa9.blob.core.windows.net/ productionresultssa10.blob.core.windows.net/ productionresultssa11.blob.core.windows.net/ productionresultssa12.blob.core.windows.net/ productionresultssa13.blob.core.windows.net/ productionresultssa14.blob.core.windows.net/ productionresultssa15.blob.core.windows.net/ productionresultssa16.blob.core.windows.net/ productionresultssa17.blob.core.windows.net/ productionresultssa18.blob.core.windows.net/ productionresultssa19.blob.core.windows.net/ github-production-repository-image-32fea6.s3.amazonaws.com github-production-release-asset-2e65be.s3.amazonaws.com insights.github.com wss://alive.github.com wss://alive-staging.github.com api.githubcopilot.com api.individual.githubcopilot.com api.business.githubcopilot.com api.enterprise.githubcopilot.com; font-src github.githubassets.com; form-action 'self' github.com gist.github.com copilot-workspace.githubnext.com objects-origin.githubusercontent.com; frame-ancestors 'none'; frame-src viewscreen.githubusercontent.com notebooks.githubusercontent.com; img-src 'self' data: blob: github.githubassets.com media.githubusercontent.com camo.githubusercontent.com identicons.github.com avatars.githubusercontent.com private-avatars.githubusercontent.com github-cloud.s3.amazonaws.com objects.githubusercontent.com release-assets.githubusercontent.com secured-user-images.githubusercontent.com/ user-images.githubusercontent.com/ private-user-images.githubusercontent.com opengraph.githubassets.com marketplace-screenshots.githubusercontent.com/ copilotprodattachments.blob.core.windows.net/github-production-copilot-attachments/ github-production-user-asset-6210df.s3.amazonaws.com customer-stories-feed.github.com spotlights-feed.github.com objects-origin.githubusercontent.com *.githubusercontent.com; manifest-src 'self'; media-src github.com user-images.githubusercontent.com/ secured-user-images.githubusercontent.com/ private-user-images.githubusercontent.com github-production-user-asset-6210df.s3.amazonaws.com gist.github.com github.githubassets.com; script-src github.githubassets.com; style-src 'unsafe-inline' github.githubassets.com; upgrade-insecure-requests; worker-src github.githubassets.com github.com/assets-cdn/worker/ github.com/assets/ gist.github.com/assets-cdn/worker/ server: github.com x-github-request-id: D69E:296706:A735C8:C01A59:69591D93 HTTP/2 200 last-modified: Tue, 14 Mar 2023 15:42:36 GMT etag: "0x8DB24A2BC089C25" server: Windows-Azure-Blob/1.0 Microsoft-HTTPAPI/2.0 x-ms-request-id: 98731145-a01e-0051-50b7-7c7cd9000000 x-ms-version: 2018-11-09 x-ms-creation-time: Tue, 14 Mar 2023 15:42:36 GMT x-ms-blob-content-md5: aj/vELpzh0jVUnaDyETqBQ== x-ms-lease-status: unlocked x-ms-lease-state: available x-ms-blob-type: BlockBlob x-ms-server-encrypted: true via: 1.1 varnish, 1.1 varnish accept-ranges: bytes age: 0 date: Sat, 03 Jan 2026 13:45:55 GMT x-served-by: cache-iad-kcgs7200139-IAD, cache-bom-vanm7210032-BOM x-cache: MISS, MISS x-cache-hits: 0, 0 x-timer: S1767447955.427023,VS0,VE290 content-disposition: attachment; filename=r50-fp16_phase2.log content-type: application/octet-stream content-length: 345829 2023-03-13 02:02:12,393 - mmdet - INFO - Environment info: ------------------------------------------------------------ sys.platform: linux Python: 3.8.16 (default, Mar 2 2023, 03:21:46) [GCC 11.2.0] CUDA available: True GPU 0,1,2,3: NVIDIA RTX A6000 CUDA_HOME: /usr/local/cuda NVCC: Build cuda_11.8.r11.8/compiler.31833905_0 GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0 PyTorch: 1.10.1 PyTorch compiling details: PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.2.3 (Git Hash 7336ca9f055cf1bfa13efb658fe15dc9b41f0740) - OpenMP 201511 (a.k.a. OpenMP 4.5) - LAPACK is enabled (usually provided by MKL) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.3 - 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_61,code=sm_61;-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;-gencode;arch=compute_37,code=compute_37 - CuDNN 8.2 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, 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 -DEDGE_PROFILER_USE_KINETO -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=1.10.1, 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: 0.11.2 OpenCV: 4.7.0 MMCV: 1.3.16 MMCV Compiler: GCC 7.3 MMCV CUDA Compiler: 11.3 MMDetection: 2.14.0 MMSegmentation: 0.14.1 MMDetection3D: 0.17.2+2919b99 ------------------------------------------------------------ 2023-03-13 02:02:13,281 - mmdet - INFO - Distributed training: True 2023-03-13 02:02:14,205 - mmdet - INFO - Config: point_cloud_range = [-51.2, -51.2, -5.0, 51.2, 51.2, 3.0] class_names = [ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ] dataset_type = 'NuScenesDataset' data_root = 'data/nuscenes/' input_modality = dict( use_lidar=False, use_camera=True, use_radar=False, use_map=False, use_external=False) file_client_args = dict(backend='disk') train_pipeline = [ dict( type='LoadMultiViewImageFromFiles_BEVDet', is_train=True, data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='LoadPointsFromFile', coord_type='LIDAR', load_dim=5, use_dim=5, file_client_args=dict(backend='disk')), dict(type='LoadAnnotations3D', with_bbox_3d=True, with_label_3d=True), dict( type='GlobalRotScaleTrans', rot_range=[-0.3925, 0.3925], scale_ratio_range=[0.95, 1.05], translation_std=[0, 0, 0], update_img2lidar=True), dict( type='RandomFlip3D', sync_2d=False, flip_ratio_bev_horizontal=0.5, flip_ratio_bev_vertical=0.5, update_img2lidar=True), dict( type='PointToMultiViewDepth', grid_config=dict( xbound=[-51.2, 51.2, 0.8], ybound=[-51.2, 51.2, 0.8], zbound=[-10.0, 10.0, 20.0], dbound=[2.0, 58.0, 0.5])), dict( type='ObjectRangeFilter', point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]), dict( type='ObjectNameFilter', classes=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ]), dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ]), dict( type='Collect3D', keys=['img_inputs', 'gt_bboxes_3d', 'gt_labels_3d']) ] test_pipeline = [ dict( type='LoadMultiViewImageFromFiles_BEVDet', data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='MultiScaleFlipAug3D', img_scale=(1333, 800), pts_scale_ratio=1, flip=False, transforms=[ dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], with_label=False), dict(type='Collect3D', keys=['img_inputs']) ]) ] eval_pipeline = [ dict( type='LoadMultiViewImageFromFiles_BEVDet', data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], with_label=False), dict(type='Collect3D', keys=['img_inputs']) ] data = dict( samples_per_gpu=16, workers_per_gpu=16, train=dict( type='NuScenesDataset', data_root='data/nuscenes/', ann_file='data/nuscenes/nuscenes_infos_train.pkl', pipeline=[ dict( type='LoadMultiViewImageFromFiles_BEVDet', is_train=True, data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='LoadPointsFromFile', coord_type='LIDAR', load_dim=5, use_dim=5, file_client_args=dict(backend='disk')), dict( type='LoadAnnotations3D', with_bbox_3d=True, with_label_3d=True), dict( type='GlobalRotScaleTrans', rot_range=[-0.3925, 0.3925], scale_ratio_range=[0.95, 1.05], translation_std=[0, 0, 0], update_img2lidar=True), dict( type='RandomFlip3D', sync_2d=False, flip_ratio_bev_horizontal=0.5, flip_ratio_bev_vertical=0.5, update_img2lidar=True), dict( type='PointToMultiViewDepth', grid_config=dict( xbound=[-51.2, 51.2, 0.8], ybound=[-51.2, 51.2, 0.8], zbound=[-10.0, 10.0, 20.0], dbound=[2.0, 58.0, 0.5])), dict( type='ObjectRangeFilter', point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]), dict( type='ObjectNameFilter', classes=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ]), dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ]), dict( type='Collect3D', keys=['img_inputs', 'gt_bboxes_3d', 'gt_labels_3d']) ], classes=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], modality=dict( use_lidar=False, use_camera=True, use_radar=False, use_map=False, use_external=False), test_mode=False, box_type_3d='LiDAR', use_valid_flag=True, speed_mode=None, max_interval=None, min_interval=None, prev_only=None, fix_direction=None, img_info_prototype='bevdet', use_sequence_group_flag=True, sequences_split_num=2, filter_empty_gt=False), val=dict( type='NuScenesDataset', data_root='data/nuscenes/', ann_file='data/nuscenes/nuscenes_infos_val.pkl', pipeline=[ dict( type='LoadMultiViewImageFromFiles_BEVDet', data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='MultiScaleFlipAug3D', img_scale=(1333, 800), pts_scale_ratio=1, flip=False, transforms=[ dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], with_label=False), dict(type='Collect3D', keys=['img_inputs']) ]) ], classes=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], modality=dict( use_lidar=False, use_camera=True, use_radar=False, use_map=False, use_external=False), test_mode=True, box_type_3d='LiDAR', img_info_prototype='bevdet', use_sequence_group_flag=True, sequences_split_num=1), test=dict( type='NuScenesDataset', data_root='data/nuscenes/', ann_file='data/nuscenes/nuscenes_infos_val.pkl', pipeline=[ dict( type='LoadMultiViewImageFromFiles_BEVDet', data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='MultiScaleFlipAug3D', img_scale=(1333, 800), pts_scale_ratio=1, flip=False, transforms=[ dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], with_label=False), dict(type='Collect3D', keys=['img_inputs']) ]) ], classes=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], modality=dict( use_lidar=False, use_camera=True, use_radar=False, use_map=False, use_external=False), test_mode=True, box_type_3d='LiDAR', img_info_prototype='bevdet', use_sequence_group_flag=True, sequences_split_num=1)) evaluation = dict( interval=10536, pipeline=[ dict( type='LoadMultiViewImageFromFiles_BEVDet', data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04)), dict( type='DefaultFormatBundle3D', class_names=[ 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone' ], with_label=False), dict(type='Collect3D', keys=['img_inputs']) ]) checkpoint_config = dict(interval=1756) log_config = dict( interval=50, hooks=[dict(type='TextLoggerHook'), dict(type='TensorboardLoggerHook')]) dist_params = dict(backend='nccl') log_level = 'INFO' work_dir = './work_dirs/r50-fp16_phase2' load_from = None resume_from = 'work_dirs/r50-fp16_phase1/iter_2634.pth' workflow = [('train', 1)] resume_optimizer = False find_unused_parameters = False num_gpus = 4 batch_size = 16 num_iters_per_epoch = 439 num_epochs = 24 checkpoint_epoch_interval = 4 train_sequences_split_num = 2 test_sequences_split_num = 1 filter_empty_gt = False with_cp = False base_bev_channels = 80 do_history = True history_cat_num = 16 history_cat_conv_out_channels = 160 do_history_stereo_fusion = True stereo_out_feats = 64 history_stereo_prev_step = 1 stereo_sampling_num = 7 bev_encoder_in_channels = 160 depth_loss_weight = 3.0 velocity_code_weight = 1.0 data_config = dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04) grid_config = dict( xbound=[-51.2, 51.2, 0.8], ybound=[-51.2, 51.2, 0.8], zbound=[-10.0, 10.0, 20.0], dbound=[2.0, 58.0, 0.5]) voxel_size = [0.1, 0.1, 0.2] model = dict( type='SOLOFusion', do_history=True, history_cat_num=16, history_cat_conv_out_channels=160, do_history_stereo_fusion=True, history_stereo_prev_step=1, img_backbone=dict( pretrained='torchvision://resnet50', type='ResNet', depth=50, num_stages=4, out_indices=(0, 1, 2, 3), frozen_stages=0, norm_cfg=dict(type='BN', requires_grad=True), norm_eval=False, with_cp=False, style='pytorch'), img_neck=dict( type='SECONDFPN', in_channels=[256, 512, 1024, 2048], upsample_strides=[0.25, 0.5, 1, 2], out_channels=[128, 128, 128, 128]), stereo_neck=dict( type='SECONDFPN', in_channels=[256, 512, 1024, 2048], upsample_strides=[1, 2, 4, 8], out_channels=[64, 64, 64, 64], final_conv_feature_dim=64), img_view_transformer=dict( type='ViewTransformerSOLOFusion', do_history_stereo_fusion=True, stereo_sampling_num=7, loss_depth_weight=3.0, grid_config=dict( xbound=[-51.2, 51.2, 0.8], ybound=[-51.2, 51.2, 0.8], zbound=[-10.0, 10.0, 20.0], dbound=[2.0, 58.0, 0.5]), data_config=dict( cams=[ 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT' ], Ncams=6, input_size=(256, 704), src_size=(900, 1600), resize=(-0.06, 0.11), rot=(-5.4, 5.4), flip=True, crop_h=(0.0, 0.0), resize_test=0.04), numC_Trans=80, extra_depth_net=dict( type='ResNetForBEVDet', numC_input=256, num_layer=[3], num_channels=[256], stride=[1])), pre_process=dict( type='ResNetForBEVDet', numC_input=80, num_layer=[2], num_channels=[80], stride=[1], backbone_output_ids=[0]), img_bev_encoder_backbone=dict( type='ResNetForBEVDet', numC_input=160, num_channels=[160, 320, 640], backbone_output_ids=[-1, 0, 1, 2]), img_bev_encoder_neck=dict( type='SECONDFPN', in_channels=[160, 160, 320, 640], upsample_strides=[1, 2, 4, 8], out_channels=[64, 64, 64, 64]), pts_bbox_head=dict( type='CenterHead', in_channels=256, tasks=[ dict(num_class=1, class_names=['car']), dict(num_class=2, class_names=['truck', 'construction_vehicle']), dict(num_class=2, class_names=['bus', 'trailer']), dict(num_class=1, class_names=['barrier']), dict(num_class=2, class_names=['motorcycle', 'bicycle']), dict(num_class=2, class_names=['pedestrian', 'traffic_cone']) ], common_heads=dict( reg=(2, 2), height=(1, 2), dim=(3, 2), rot=(2, 2), vel=(2, 2)), share_conv_channel=64, bbox_coder=dict( type='CenterPointBBoxCoder', pc_range=[-51.2, -51.2], post_center_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0], max_num=500, score_threshold=0.1, out_size_factor=8, voxel_size=[0.1, 0.1], code_size=9), separate_head=dict( type='SeparateHead', init_bias=-2.19, final_kernel=3), loss_cls=dict(type='GaussianFocalLoss', reduction='mean'), loss_bbox=dict(type='L1Loss', reduction='mean', loss_weight=0.25), norm_bbox=True), train_cfg=dict( pts=dict( point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0], grid_size=[1024, 1024, 40], voxel_size=[0.1, 0.1, 0.2], out_size_factor=8, dense_reg=1, gaussian_overlap=0.1, max_objs=500, min_radius=2, code_weights=[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0])), test_cfg=dict( pts=dict( pc_range=[-51.2, -51.2], post_center_limit_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0], max_per_img=500, max_pool_nms=False, min_radius=[4, 12, 10, 1, 0.85, 0.175], score_threshold=0.1, out_size_factor=8, voxel_size=[0.1, 0.1], pre_max_size=1000, post_max_size=83, nms_type=[ 'rotate', 'rotate', 'rotate', 'circle', 'rotate', 'rotate' ], nms_thr=[0.2, 0.2, 0.2, 0.2, 0.2, 0.5], nms_rescale_factor=[ 1.0, [0.7, 0.7], [0.4, 0.55], 1.1, [1.0, 1.0], [4.5, 9.0] ]))) lr = 0.0004 optimizer = dict(type='AdamW', lr=0.0002, weight_decay=1e-07) optimizer_config = dict( type='WarmupFp16OptimizerHook', grad_clip=dict(max_norm=5, norm_type=2), warmup_loss_scale_value=1.0, warmup_loss_scale_iters=109, loss_scale=512.0) lr_config = None runner = dict(type='IterBasedRunner', max_iters=10536) custom_hooks = [ dict( type='ExpMomentumEMAHook', resume_from='work_dirs/r50-fp16_phase1/iter_2634.pth', resume_optimizer=False, momentum=0.001, priority=49) ] gpu_ids = range(0, 4) 2023-03-13 02:02:14,206 - mmdet - INFO - Set random seed to 0, deterministic: False 2023-03-13 02:02:14,573 - mmdet - INFO - initialize ResNet with init_cfg {'type': 'Pretrained', 'checkpoint': 'torchvision://resnet50'} 2023-03-13 02:02:17,388 - mmdet - INFO - initialize SECONDFPN with init_cfg [{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] 2023-03-13 02:02:17,399 - mmdet - INFO - initialize SECONDFPN with init_cfg [{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] 2023-03-13 02:02:17,413 - mmdet - INFO - initialize SECONDFPN with init_cfg [{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] Name of parameter - Initialization information pts_bbox_head.shared_conv.conv.weight - torch.Size([64, 256, 3, 3]): Initialized by user-defined `init_weights` in ConvModule pts_bbox_head.shared_conv.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.shared_conv.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.heatmap.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.0.heatmap.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.heatmap.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.1.heatmap.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.heatmap.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.2.heatmap.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.heatmap.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.3.heatmap.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.heatmap.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.4.heatmap.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.reg.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.reg.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.reg.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.reg.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.reg.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.height.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.height.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.height.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.height.1.weight - torch.Size([1, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.height.1.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.dim.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.dim.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.dim.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.dim.1.weight - torch.Size([3, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.dim.1.bias - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.rot.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.rot.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.rot.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.rot.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.rot.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.vel.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.vel.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.vel.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.vel.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.vel.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.heatmap.0.conv.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.heatmap.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.heatmap.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.heatmap.1.weight - torch.Size([2, 64, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pts_bbox_head.task_heads.5.heatmap.1.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SOLOFusion img_backbone.conv1.weight - torch.Size([64, 3, 7, 7]): PretrainedInit: load from torchvision://resnet50 img_backbone.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.downsample.1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.0.downsample.1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.1.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer1.2.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.downsample.1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.0.downsample.1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.1.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.2.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer2.3.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.downsample.1.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.0.downsample.1.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.1.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.2.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.3.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.4.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer3.5.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.downsample.1.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.0.downsample.1.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.1.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_backbone.layer4.2.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 img_neck.deblocks.0.0.weight - torch.Size([128, 256, 4, 4]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.0.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.0.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.1.0.weight - torch.Size([128, 512, 2, 2]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.1.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.1.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.2.0.weight - torch.Size([1024, 128, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_neck.deblocks.2.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.2.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.3.0.weight - torch.Size([2048, 128, 2, 2]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_neck.deblocks.3.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_neck.deblocks.3.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dx - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.bx - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.nx - torch.Size([3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.frustum - torch.Size([112, 16, 44, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.log_inv_gaussians - torch.Size([112, 112]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.bin_centers - torch.Size([112]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.depthnet.weight - torch.Size([112, 256, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.depthnet.bias - torch.Size([112]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.conv1.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.bn1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.bn1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.conv2.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.bn2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.bn2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.downsample.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.0.downsample.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.conv1.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.bn1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.bn1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.conv2.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.bn2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.1.bn2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.conv1.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.bn1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.bn1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.conv2.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.bn2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.extra_depthnet.layers.0.2.bn2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.featnet.weight - torch.Size([80, 512, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.featnet.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dcn.0.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dcn.0.conv_offset.weight - torch.Size([27, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dcn.0.conv_offset.bias - torch.Size([27]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dcn.1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.dcn.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.input_conv.weight - torch.Size([256, 512, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.input_conv.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.fc.0.weight - torch.Size([33]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.fc.0.bias - torch.Size([33]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.fc.1.weight - torch.Size([256, 33]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.se.fc.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.0.conv.weight - torch.Size([16, 8, 1, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.0.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.0.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.1.conv.weight - torch.Size([8, 16, 1, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.1.bn.weight - torch.Size([8]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.1.bn.bias - torch.Size([8]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.2.weight - torch.Size([1, 8, 1, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion img_view_transformer.similarity_net.2.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.conv1.weight - torch.Size([160, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.bn1.weight - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.bn1.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.conv2.weight - torch.Size([160, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.bn2.weight - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.bn2.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.downsample.weight - torch.Size([160, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.0.downsample.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.conv1.weight - torch.Size([160, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.bn1.weight - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.bn1.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.conv2.weight - torch.Size([160, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.bn2.weight - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.0.1.bn2.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.conv1.weight - torch.Size([320, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.bn1.weight - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.bn1.bias - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.conv2.weight - torch.Size([320, 320, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.bn2.weight - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.bn2.bias - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.downsample.weight - torch.Size([320, 160, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.0.downsample.bias - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.conv1.weight - torch.Size([320, 320, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.bn1.weight - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.bn1.bias - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.conv2.weight - torch.Size([320, 320, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.bn2.weight - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.1.1.bn2.bias - torch.Size([320]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.conv1.weight - torch.Size([640, 320, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.bn1.weight - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.bn1.bias - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.conv2.weight - torch.Size([640, 640, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.bn2.weight - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.bn2.bias - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.downsample.weight - torch.Size([640, 320, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.0.downsample.bias - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.conv1.weight - torch.Size([640, 640, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.bn1.weight - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.bn1.bias - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.conv2.weight - torch.Size([640, 640, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.bn2.weight - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_backbone.layers.2.1.bn2.bias - torch.Size([640]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.0.0.weight - torch.Size([160, 64, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_bev_encoder_neck.deblocks.0.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.0.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.1.0.weight - torch.Size([160, 64, 2, 2]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_bev_encoder_neck.deblocks.1.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.1.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.2.0.weight - torch.Size([320, 64, 4, 4]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_bev_encoder_neck.deblocks.2.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.2.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.3.0.weight - torch.Size([640, 64, 8, 8]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 img_bev_encoder_neck.deblocks.3.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion img_bev_encoder_neck.deblocks.3.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion embed.0.weight - torch.Size([80, 80, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion embed.0.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion embed.1.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion embed.1.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion embed.3.weight - torch.Size([80, 80, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion embed.3.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion embed.4.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion embed.4.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.conv1.weight - torch.Size([80, 80, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.bn1.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.bn1.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.conv2.weight - torch.Size([80, 80, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.bn2.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.bn2.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.downsample.weight - torch.Size([80, 80, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.0.downsample.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.conv1.weight - torch.Size([80, 80, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.bn1.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.bn1.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.conv2.weight - torch.Size([80, 80, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.bn2.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion pre_process_net.layers.0.1.bn2.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_time_conv.0.weight - torch.Size([80, 81, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_time_conv.0.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_time_conv.1.weight - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_time_conv.1.bias - torch.Size([80]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_cat_conv.0.weight - torch.Size([160, 1360, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_cat_conv.0.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_cat_conv.1.weight - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion history_keyframe_cat_conv.1.bias - torch.Size([160]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.0.0.weight - torch.Size([256, 64, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 stereo_neck.deblocks.0.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.0.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.1.0.weight - torch.Size([512, 64, 2, 2]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 stereo_neck.deblocks.1.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.1.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.2.0.weight - torch.Size([1024, 64, 4, 4]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 stereo_neck.deblocks.2.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.2.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.3.0.weight - torch.Size([2048, 64, 8, 8]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 stereo_neck.deblocks.3.1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.deblocks.3.1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.final_conv.0.weight - torch.Size([128, 256, 3, 3]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.final_conv.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.final_conv.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SOLOFusion stereo_neck.final_conv.3.weight - torch.Size([64, 128, 1, 1]): The value is the same before and after calling `init_weights` of SOLOFusion 2023-03-13 02:02:17,461 - mmdet - INFO - Model: SOLOFusion( (pts_bbox_head): CenterHead( (loss_cls): GaussianFocalLoss() (loss_bbox): L1Loss() (shared_conv): ConvModule( (conv): Conv2d(256, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (task_heads): ModuleList( (0): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} (1): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} (2): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} (3): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} (4): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} (5): SeparateHead( (reg): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (height): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (dim): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (rot): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (vel): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (heatmap): Sequential( (0): ConvModule( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): Conv2d(64, 2, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) ) init_cfg={'type': 'Kaiming', 'layer': 'Conv2d'} ) ) (img_backbone): ResNet( (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False) (layer1): ResLayer( (0): Bottleneck( (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Sequential( (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (2): Bottleneck( (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) (layer2): ResLayer( (0): Bottleneck( (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Sequential( (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (2): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (3): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) (layer3): ResLayer( (0): Bottleneck( (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Sequential( (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (2): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (3): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (4): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (5): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) (layer4): ResLayer( (0): Bottleneck( (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Sequential( (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (2): Bottleneck( (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) ) init_cfg={'type': 'Pretrained', 'checkpoint': 'torchvision://resnet50'} (img_neck): SECONDFPN( (deblocks): ModuleList( (0): Sequential( (0): Conv2d(256, 128, kernel_size=(4, 4), stride=(4, 4), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (1): Sequential( (0): Conv2d(512, 128, kernel_size=(2, 2), stride=(2, 2), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (2): Sequential( (0): ConvTranspose2d(1024, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (3): Sequential( (0): ConvTranspose2d(2048, 128, kernel_size=(2, 2), stride=(2, 2), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) ) ) init_cfg=[{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] (img_view_transformer): ViewTransformerSOLOFusion( (depthnet): Conv2d(256, 112, kernel_size=(1, 1), stride=(1, 1)) (extra_depthnet): ResNetForBEVDet( (layers): Sequential( (0): Sequential( (0): BasicBlock( (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (1): BasicBlock( (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) (2): BasicBlock( (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) ) ) (featnet): Conv2d(512, 80, kernel_size=(1, 1), stride=(1, 1)) (dcn): Sequential( (0): ModulatedDeformConv2dPack( (conv_offset): Conv2d(256, 27, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (se): SELikeModule( (input_conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1)) (fc): Sequential( (0): BatchNorm1d(33, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (1): Linear(in_features=33, out_features=256, bias=True) (2): Sigmoid() ) ) (similarity_net): Sequential( (0): ConvModule( (conv): Conv3d(8, 16, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False) (bn): BatchNorm3d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (1): ConvModule( (conv): Conv3d(16, 8, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False) (bn): BatchNorm3d(8, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (activate): ReLU(inplace=True) ) (2): Conv3d(8, 1, kernel_size=(1, 1, 1), stride=(1, 1, 1)) ) ) (img_bev_encoder_backbone): ResNetForBEVDet( (layers): Sequential( (0): Sequential( (0): BasicBlock( (conv1): Conv2d(160, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Conv2d(160, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) ) (1): BasicBlock( (conv1): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) (1): Sequential( (0): BasicBlock( (conv1): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(320, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) ) (1): BasicBlock( (conv1): Conv2d(320, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(320, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) (2): Sequential( (0): BasicBlock( (conv1): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn1): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(640, 640, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) ) (1): BasicBlock( (conv1): Conv2d(640, 640, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(640, 640, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) ) ) (img_bev_encoder_neck): SECONDFPN( (deblocks): ModuleList( (0): Sequential( (0): ConvTranspose2d(160, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (1): Sequential( (0): ConvTranspose2d(160, 64, kernel_size=(2, 2), stride=(2, 2), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (2): Sequential( (0): ConvTranspose2d(320, 64, kernel_size=(4, 4), stride=(4, 4), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (3): Sequential( (0): ConvTranspose2d(640, 64, kernel_size=(8, 8), stride=(8, 8), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) ) ) init_cfg=[{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] (embed): Sequential( (0): Conv2d(80, 80, kernel_size=(1, 1), stride=(1, 1)) (1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): ReLU(inplace=True) (3): Conv2d(80, 80, kernel_size=(1, 1), stride=(1, 1)) (4): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (5): ReLU(inplace=True) ) (pre_process_net): ResNetForBEVDet( (layers): Sequential( (0): Sequential( (0): BasicBlock( (conv1): Conv2d(80, 80, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(80, 80, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) (downsample): Conv2d(80, 80, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) ) (1): BasicBlock( (conv1): Conv2d(80, 80, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(80, 80, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace=True) ) ) ) ) (history_keyframe_time_conv): Sequential( (0): Conv2d(81, 80, kernel_size=(1, 1), stride=(1, 1)) (1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (history_keyframe_cat_conv): Sequential( (0): Conv2d(1360, 160, kernel_size=(1, 1), stride=(1, 1)) (1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (stereo_neck): SECONDFPN( (deblocks): ModuleList( (0): Sequential( (0): ConvTranspose2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (1): Sequential( (0): ConvTranspose2d(512, 64, kernel_size=(2, 2), stride=(2, 2), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (2): Sequential( (0): ConvTranspose2d(1024, 64, kernel_size=(4, 4), stride=(4, 4), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) (3): Sequential( (0): ConvTranspose2d(2048, 64, kernel_size=(8, 8), stride=(8, 8), bias=False) (1): BatchNorm2d(64, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) ) ) (final_conv): Sequential( (0): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) (3): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) ) ) init_cfg=[{'type': 'Kaiming', 'layer': 'ConvTranspose2d'}, {'type': 'Constant', 'layer': 'NaiveSyncBatchNorm2d', 'val': 1.0}] ) 2023-03-13 02:02:27,060 - mmdet - INFO - load checkpoint from work_dirs/r50-fp16_phase1/iter_2634.pth 2023-03-13 02:02:27,060 - mmdet - INFO - Use load_from_local loader 2023-03-13 02:02:27,812 - mmdet - WARNING - The model and loaded state dict do not match exactly size mismatch for img_bev_encoder_backbone.layers.0.0.conv1.weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for img_bev_encoder_backbone.layers.0.0.downsample.weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for img_bev_encoder_neck.deblocks.0.0.weight: copying a param with shape torch.Size([80, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([160, 64, 1, 1]). unexpected key in source state_dict: ema_pts_bbox_head_shared_conv_conv_weight, ema_pts_bbox_head_shared_conv_bn_weight, ema_pts_bbox_head_shared_conv_bn_bias, ema_pts_bbox_head_shared_conv_bn_running_mean, ema_pts_bbox_head_shared_conv_bn_running_var, ema_pts_bbox_head_shared_conv_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_reg_0_conv_weight, ema_pts_bbox_head_task_heads_0_reg_0_bn_weight, ema_pts_bbox_head_task_heads_0_reg_0_bn_bias, ema_pts_bbox_head_task_heads_0_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_0_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_reg_1_weight, ema_pts_bbox_head_task_heads_0_reg_1_bias, ema_pts_bbox_head_task_heads_0_height_0_conv_weight, ema_pts_bbox_head_task_heads_0_height_0_bn_weight, ema_pts_bbox_head_task_heads_0_height_0_bn_bias, ema_pts_bbox_head_task_heads_0_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_height_0_bn_running_var, ema_pts_bbox_head_task_heads_0_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_height_1_weight, ema_pts_bbox_head_task_heads_0_height_1_bias, ema_pts_bbox_head_task_heads_0_dim_0_conv_weight, ema_pts_bbox_head_task_heads_0_dim_0_bn_weight, ema_pts_bbox_head_task_heads_0_dim_0_bn_bias, ema_pts_bbox_head_task_heads_0_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_0_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_dim_1_weight, ema_pts_bbox_head_task_heads_0_dim_1_bias, ema_pts_bbox_head_task_heads_0_rot_0_conv_weight, ema_pts_bbox_head_task_heads_0_rot_0_bn_weight, ema_pts_bbox_head_task_heads_0_rot_0_bn_bias, ema_pts_bbox_head_task_heads_0_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_0_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_rot_1_weight, ema_pts_bbox_head_task_heads_0_rot_1_bias, ema_pts_bbox_head_task_heads_0_vel_0_conv_weight, ema_pts_bbox_head_task_heads_0_vel_0_bn_weight, ema_pts_bbox_head_task_heads_0_vel_0_bn_bias, ema_pts_bbox_head_task_heads_0_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_0_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_vel_1_weight, ema_pts_bbox_head_task_heads_0_vel_1_bias, ema_pts_bbox_head_task_heads_0_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_0_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_0_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_0_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_0_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_0_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_0_heatmap_1_weight, ema_pts_bbox_head_task_heads_0_heatmap_1_bias, ema_pts_bbox_head_task_heads_1_reg_0_conv_weight, ema_pts_bbox_head_task_heads_1_reg_0_bn_weight, ema_pts_bbox_head_task_heads_1_reg_0_bn_bias, ema_pts_bbox_head_task_heads_1_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_1_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_reg_1_weight, ema_pts_bbox_head_task_heads_1_reg_1_bias, ema_pts_bbox_head_task_heads_1_height_0_conv_weight, ema_pts_bbox_head_task_heads_1_height_0_bn_weight, ema_pts_bbox_head_task_heads_1_height_0_bn_bias, ema_pts_bbox_head_task_heads_1_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_height_0_bn_running_var, ema_pts_bbox_head_task_heads_1_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_height_1_weight, ema_pts_bbox_head_task_heads_1_height_1_bias, ema_pts_bbox_head_task_heads_1_dim_0_conv_weight, ema_pts_bbox_head_task_heads_1_dim_0_bn_weight, ema_pts_bbox_head_task_heads_1_dim_0_bn_bias, ema_pts_bbox_head_task_heads_1_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_1_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_dim_1_weight, ema_pts_bbox_head_task_heads_1_dim_1_bias, ema_pts_bbox_head_task_heads_1_rot_0_conv_weight, ema_pts_bbox_head_task_heads_1_rot_0_bn_weight, ema_pts_bbox_head_task_heads_1_rot_0_bn_bias, ema_pts_bbox_head_task_heads_1_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_1_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_rot_1_weight, ema_pts_bbox_head_task_heads_1_rot_1_bias, ema_pts_bbox_head_task_heads_1_vel_0_conv_weight, ema_pts_bbox_head_task_heads_1_vel_0_bn_weight, ema_pts_bbox_head_task_heads_1_vel_0_bn_bias, ema_pts_bbox_head_task_heads_1_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_1_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_vel_1_weight, ema_pts_bbox_head_task_heads_1_vel_1_bias, ema_pts_bbox_head_task_heads_1_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_1_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_1_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_1_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_1_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_1_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_1_heatmap_1_weight, ema_pts_bbox_head_task_heads_1_heatmap_1_bias, ema_pts_bbox_head_task_heads_2_reg_0_conv_weight, ema_pts_bbox_head_task_heads_2_reg_0_bn_weight, ema_pts_bbox_head_task_heads_2_reg_0_bn_bias, ema_pts_bbox_head_task_heads_2_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_2_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_reg_1_weight, ema_pts_bbox_head_task_heads_2_reg_1_bias, ema_pts_bbox_head_task_heads_2_height_0_conv_weight, ema_pts_bbox_head_task_heads_2_height_0_bn_weight, ema_pts_bbox_head_task_heads_2_height_0_bn_bias, ema_pts_bbox_head_task_heads_2_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_height_0_bn_running_var, ema_pts_bbox_head_task_heads_2_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_height_1_weight, ema_pts_bbox_head_task_heads_2_height_1_bias, ema_pts_bbox_head_task_heads_2_dim_0_conv_weight, ema_pts_bbox_head_task_heads_2_dim_0_bn_weight, ema_pts_bbox_head_task_heads_2_dim_0_bn_bias, ema_pts_bbox_head_task_heads_2_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_2_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_dim_1_weight, ema_pts_bbox_head_task_heads_2_dim_1_bias, ema_pts_bbox_head_task_heads_2_rot_0_conv_weight, ema_pts_bbox_head_task_heads_2_rot_0_bn_weight, ema_pts_bbox_head_task_heads_2_rot_0_bn_bias, ema_pts_bbox_head_task_heads_2_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_2_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_rot_1_weight, ema_pts_bbox_head_task_heads_2_rot_1_bias, ema_pts_bbox_head_task_heads_2_vel_0_conv_weight, ema_pts_bbox_head_task_heads_2_vel_0_bn_weight, ema_pts_bbox_head_task_heads_2_vel_0_bn_bias, ema_pts_bbox_head_task_heads_2_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_2_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_vel_1_weight, ema_pts_bbox_head_task_heads_2_vel_1_bias, ema_pts_bbox_head_task_heads_2_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_2_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_2_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_2_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_2_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_2_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_2_heatmap_1_weight, ema_pts_bbox_head_task_heads_2_heatmap_1_bias, ema_pts_bbox_head_task_heads_3_reg_0_conv_weight, ema_pts_bbox_head_task_heads_3_reg_0_bn_weight, ema_pts_bbox_head_task_heads_3_reg_0_bn_bias, ema_pts_bbox_head_task_heads_3_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_3_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_reg_1_weight, ema_pts_bbox_head_task_heads_3_reg_1_bias, ema_pts_bbox_head_task_heads_3_height_0_conv_weight, ema_pts_bbox_head_task_heads_3_height_0_bn_weight, ema_pts_bbox_head_task_heads_3_height_0_bn_bias, ema_pts_bbox_head_task_heads_3_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_height_0_bn_running_var, ema_pts_bbox_head_task_heads_3_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_height_1_weight, ema_pts_bbox_head_task_heads_3_height_1_bias, ema_pts_bbox_head_task_heads_3_dim_0_conv_weight, ema_pts_bbox_head_task_heads_3_dim_0_bn_weight, ema_pts_bbox_head_task_heads_3_dim_0_bn_bias, ema_pts_bbox_head_task_heads_3_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_3_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_dim_1_weight, ema_pts_bbox_head_task_heads_3_dim_1_bias, ema_pts_bbox_head_task_heads_3_rot_0_conv_weight, ema_pts_bbox_head_task_heads_3_rot_0_bn_weight, ema_pts_bbox_head_task_heads_3_rot_0_bn_bias, ema_pts_bbox_head_task_heads_3_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_3_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_rot_1_weight, ema_pts_bbox_head_task_heads_3_rot_1_bias, ema_pts_bbox_head_task_heads_3_vel_0_conv_weight, ema_pts_bbox_head_task_heads_3_vel_0_bn_weight, ema_pts_bbox_head_task_heads_3_vel_0_bn_bias, ema_pts_bbox_head_task_heads_3_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_3_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_vel_1_weight, ema_pts_bbox_head_task_heads_3_vel_1_bias, ema_pts_bbox_head_task_heads_3_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_3_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_3_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_3_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_3_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_3_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_3_heatmap_1_weight, ema_pts_bbox_head_task_heads_3_heatmap_1_bias, ema_pts_bbox_head_task_heads_4_reg_0_conv_weight, ema_pts_bbox_head_task_heads_4_reg_0_bn_weight, ema_pts_bbox_head_task_heads_4_reg_0_bn_bias, ema_pts_bbox_head_task_heads_4_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_4_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_reg_1_weight, ema_pts_bbox_head_task_heads_4_reg_1_bias, ema_pts_bbox_head_task_heads_4_height_0_conv_weight, ema_pts_bbox_head_task_heads_4_height_0_bn_weight, ema_pts_bbox_head_task_heads_4_height_0_bn_bias, ema_pts_bbox_head_task_heads_4_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_height_0_bn_running_var, ema_pts_bbox_head_task_heads_4_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_height_1_weight, ema_pts_bbox_head_task_heads_4_height_1_bias, ema_pts_bbox_head_task_heads_4_dim_0_conv_weight, ema_pts_bbox_head_task_heads_4_dim_0_bn_weight, ema_pts_bbox_head_task_heads_4_dim_0_bn_bias, ema_pts_bbox_head_task_heads_4_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_4_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_dim_1_weight, ema_pts_bbox_head_task_heads_4_dim_1_bias, ema_pts_bbox_head_task_heads_4_rot_0_conv_weight, ema_pts_bbox_head_task_heads_4_rot_0_bn_weight, ema_pts_bbox_head_task_heads_4_rot_0_bn_bias, ema_pts_bbox_head_task_heads_4_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_4_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_rot_1_weight, ema_pts_bbox_head_task_heads_4_rot_1_bias, ema_pts_bbox_head_task_heads_4_vel_0_conv_weight, ema_pts_bbox_head_task_heads_4_vel_0_bn_weight, ema_pts_bbox_head_task_heads_4_vel_0_bn_bias, ema_pts_bbox_head_task_heads_4_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_4_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_vel_1_weight, ema_pts_bbox_head_task_heads_4_vel_1_bias, ema_pts_bbox_head_task_heads_4_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_4_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_4_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_4_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_4_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_4_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_4_heatmap_1_weight, ema_pts_bbox_head_task_heads_4_heatmap_1_bias, ema_pts_bbox_head_task_heads_5_reg_0_conv_weight, ema_pts_bbox_head_task_heads_5_reg_0_bn_weight, ema_pts_bbox_head_task_heads_5_reg_0_bn_bias, ema_pts_bbox_head_task_heads_5_reg_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_reg_0_bn_running_var, ema_pts_bbox_head_task_heads_5_reg_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_reg_1_weight, ema_pts_bbox_head_task_heads_5_reg_1_bias, ema_pts_bbox_head_task_heads_5_height_0_conv_weight, ema_pts_bbox_head_task_heads_5_height_0_bn_weight, ema_pts_bbox_head_task_heads_5_height_0_bn_bias, ema_pts_bbox_head_task_heads_5_height_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_height_0_bn_running_var, ema_pts_bbox_head_task_heads_5_height_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_height_1_weight, ema_pts_bbox_head_task_heads_5_height_1_bias, ema_pts_bbox_head_task_heads_5_dim_0_conv_weight, ema_pts_bbox_head_task_heads_5_dim_0_bn_weight, ema_pts_bbox_head_task_heads_5_dim_0_bn_bias, ema_pts_bbox_head_task_heads_5_dim_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_dim_0_bn_running_var, ema_pts_bbox_head_task_heads_5_dim_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_dim_1_weight, ema_pts_bbox_head_task_heads_5_dim_1_bias, ema_pts_bbox_head_task_heads_5_rot_0_conv_weight, ema_pts_bbox_head_task_heads_5_rot_0_bn_weight, ema_pts_bbox_head_task_heads_5_rot_0_bn_bias, ema_pts_bbox_head_task_heads_5_rot_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_rot_0_bn_running_var, ema_pts_bbox_head_task_heads_5_rot_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_rot_1_weight, ema_pts_bbox_head_task_heads_5_rot_1_bias, ema_pts_bbox_head_task_heads_5_vel_0_conv_weight, ema_pts_bbox_head_task_heads_5_vel_0_bn_weight, ema_pts_bbox_head_task_heads_5_vel_0_bn_bias, ema_pts_bbox_head_task_heads_5_vel_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_vel_0_bn_running_var, ema_pts_bbox_head_task_heads_5_vel_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_vel_1_weight, ema_pts_bbox_head_task_heads_5_vel_1_bias, ema_pts_bbox_head_task_heads_5_heatmap_0_conv_weight, ema_pts_bbox_head_task_heads_5_heatmap_0_bn_weight, ema_pts_bbox_head_task_heads_5_heatmap_0_bn_bias, ema_pts_bbox_head_task_heads_5_heatmap_0_bn_running_mean, ema_pts_bbox_head_task_heads_5_heatmap_0_bn_running_var, ema_pts_bbox_head_task_heads_5_heatmap_0_bn_num_batches_tracked, ema_pts_bbox_head_task_heads_5_heatmap_1_weight, ema_pts_bbox_head_task_heads_5_heatmap_1_bias, ema_img_backbone_conv1_weight, ema_img_backbone_bn1_weight, ema_img_backbone_bn1_bias, ema_img_backbone_bn1_running_mean, ema_img_backbone_bn1_running_var, ema_img_backbone_bn1_num_batches_tracked, ema_img_backbone_layer1_0_conv1_weight, ema_img_backbone_layer1_0_bn1_weight, ema_img_backbone_layer1_0_bn1_bias, ema_img_backbone_layer1_0_bn1_running_mean, ema_img_backbone_layer1_0_bn1_running_var, ema_img_backbone_layer1_0_bn1_num_batches_tracked, ema_img_backbone_layer1_0_conv2_weight, ema_img_backbone_layer1_0_bn2_weight, ema_img_backbone_layer1_0_bn2_bias, ema_img_backbone_layer1_0_bn2_running_mean, ema_img_backbone_layer1_0_bn2_running_var, ema_img_backbone_layer1_0_bn2_num_batches_tracked, ema_img_backbone_layer1_0_conv3_weight, ema_img_backbone_layer1_0_bn3_weight, ema_img_backbone_layer1_0_bn3_bias, ema_img_backbone_layer1_0_bn3_running_mean, ema_img_backbone_layer1_0_bn3_running_var, ema_img_backbone_layer1_0_bn3_num_batches_tracked, ema_img_backbone_layer1_0_downsample_0_weight, ema_img_backbone_layer1_0_downsample_1_weight, ema_img_backbone_layer1_0_downsample_1_bias, ema_img_backbone_layer1_0_downsample_1_running_mean, ema_img_backbone_layer1_0_downsample_1_running_var, ema_img_backbone_layer1_0_downsample_1_num_batches_tracked, ema_img_backbone_layer1_1_conv1_weight, ema_img_backbone_layer1_1_bn1_weight, ema_img_backbone_layer1_1_bn1_bias, ema_img_backbone_layer1_1_bn1_running_mean, ema_img_backbone_layer1_1_bn1_running_var, ema_img_backbone_layer1_1_bn1_num_batches_tracked, ema_img_backbone_layer1_1_conv2_weight, ema_img_backbone_layer1_1_bn2_weight, ema_img_backbone_layer1_1_bn2_bias, ema_img_backbone_layer1_1_bn2_running_mean, ema_img_backbone_layer1_1_bn2_running_var, ema_img_backbone_layer1_1_bn2_num_batches_tracked, ema_img_backbone_layer1_1_conv3_weight, ema_img_backbone_layer1_1_bn3_weight, ema_img_backbone_layer1_1_bn3_bias, ema_img_backbone_layer1_1_bn3_running_mean, ema_img_backbone_layer1_1_bn3_running_var, ema_img_backbone_layer1_1_bn3_num_batches_tracked, ema_img_backbone_layer1_2_conv1_weight, ema_img_backbone_layer1_2_bn1_weight, ema_img_backbone_layer1_2_bn1_bias, ema_img_backbone_layer1_2_bn1_running_mean, ema_img_backbone_layer1_2_bn1_running_var, ema_img_backbone_layer1_2_bn1_num_batches_tracked, ema_img_backbone_layer1_2_conv2_weight, ema_img_backbone_layer1_2_bn2_weight, ema_img_backbone_layer1_2_bn2_bias, ema_img_backbone_layer1_2_bn2_running_mean, ema_img_backbone_layer1_2_bn2_running_var, ema_img_backbone_layer1_2_bn2_num_batches_tracked, ema_img_backbone_layer1_2_conv3_weight, ema_img_backbone_layer1_2_bn3_weight, ema_img_backbone_layer1_2_bn3_bias, ema_img_backbone_layer1_2_bn3_running_mean, ema_img_backbone_layer1_2_bn3_running_var, ema_img_backbone_layer1_2_bn3_num_batches_tracked, ema_img_backbone_layer2_0_conv1_weight, ema_img_backbone_layer2_0_bn1_weight, ema_img_backbone_layer2_0_bn1_bias, ema_img_backbone_layer2_0_bn1_running_mean, ema_img_backbone_layer2_0_bn1_running_var, ema_img_backbone_layer2_0_bn1_num_batches_tracked, ema_img_backbone_layer2_0_conv2_weight, ema_img_backbone_layer2_0_bn2_weight, ema_img_backbone_layer2_0_bn2_bias, ema_img_backbone_layer2_0_bn2_running_mean, ema_img_backbone_layer2_0_bn2_running_var, ema_img_backbone_layer2_0_bn2_num_batches_tracked, ema_img_backbone_layer2_0_conv3_weight, ema_img_backbone_layer2_0_bn3_weight, ema_img_backbone_layer2_0_bn3_bias, ema_img_backbone_layer2_0_bn3_running_mean, ema_img_backbone_layer2_0_bn3_running_var, ema_img_backbone_layer2_0_bn3_num_batches_tracked, ema_img_backbone_layer2_0_downsample_0_weight, ema_img_backbone_layer2_0_downsample_1_weight, ema_img_backbone_layer2_0_downsample_1_bias, ema_img_backbone_layer2_0_downsample_1_running_mean, ema_img_backbone_layer2_0_downsample_1_running_var, ema_img_backbone_layer2_0_downsample_1_num_batches_tracked, ema_img_backbone_layer2_1_conv1_weight, ema_img_backbone_layer2_1_bn1_weight, ema_img_backbone_layer2_1_bn1_bias, ema_img_backbone_layer2_1_bn1_running_mean, ema_img_backbone_layer2_1_bn1_running_var, ema_img_backbone_layer2_1_bn1_num_batches_tracked, ema_img_backbone_layer2_1_conv2_weight, ema_img_backbone_layer2_1_bn2_weight, ema_img_backbone_layer2_1_bn2_bias, ema_img_backbone_layer2_1_bn2_running_mean, ema_img_backbone_layer2_1_bn2_running_var, ema_img_backbone_layer2_1_bn2_num_batches_tracked, ema_img_backbone_layer2_1_conv3_weight, ema_img_backbone_layer2_1_bn3_weight, ema_img_backbone_layer2_1_bn3_bias, ema_img_backbone_layer2_1_bn3_running_mean, ema_img_backbone_layer2_1_bn3_running_var, ema_img_backbone_layer2_1_bn3_num_batches_tracked, ema_img_backbone_layer2_2_conv1_weight, ema_img_backbone_layer2_2_bn1_weight, ema_img_backbone_layer2_2_bn1_bias, ema_img_backbone_layer2_2_bn1_running_mean, ema_img_backbone_layer2_2_bn1_running_var, ema_img_backbone_layer2_2_bn1_num_batches_tracked, ema_img_backbone_layer2_2_conv2_weight, ema_img_backbone_layer2_2_bn2_weight, ema_img_backbone_layer2_2_bn2_bias, ema_img_backbone_layer2_2_bn2_running_mean, ema_img_backbone_layer2_2_bn2_running_var, ema_img_backbone_layer2_2_bn2_num_batches_tracked, ema_img_backbone_layer2_2_conv3_weight, ema_img_backbone_layer2_2_bn3_weight, ema_img_backbone_layer2_2_bn3_bias, ema_img_backbone_layer2_2_bn3_running_mean, ema_img_backbone_layer2_2_bn3_running_var, ema_img_backbone_layer2_2_bn3_num_batches_tracked, ema_img_backbone_layer2_3_conv1_weight, ema_img_backbone_layer2_3_bn1_weight, ema_img_backbone_layer2_3_bn1_bias, ema_img_backbone_layer2_3_bn1_running_mean, ema_img_backbone_layer2_3_bn1_running_var, ema_img_backbone_layer2_3_bn1_num_batches_tracked, ema_img_backbone_layer2_3_conv2_weight, ema_img_backbone_layer2_3_bn2_weight, ema_img_backbone_layer2_3_bn2_bias, ema_img_backbone_layer2_3_bn2_running_mean, ema_img_backbone_layer2_3_bn2_running_var, ema_img_backbone_layer2_3_bn2_num_batches_tracked, ema_img_backbone_layer2_3_conv3_weight, ema_img_backbone_layer2_3_bn3_weight, ema_img_backbone_layer2_3_bn3_bias, ema_img_backbone_layer2_3_bn3_running_mean, ema_img_backbone_layer2_3_bn3_running_var, ema_img_backbone_layer2_3_bn3_num_batches_tracked, ema_img_backbone_layer3_0_conv1_weight, ema_img_backbone_layer3_0_bn1_weight, ema_img_backbone_layer3_0_bn1_bias, ema_img_backbone_layer3_0_bn1_running_mean, ema_img_backbone_layer3_0_bn1_running_var, ema_img_backbone_layer3_0_bn1_num_batches_tracked, ema_img_backbone_layer3_0_conv2_weight, ema_img_backbone_layer3_0_bn2_weight, ema_img_backbone_layer3_0_bn2_bias, ema_img_backbone_layer3_0_bn2_running_mean, ema_img_backbone_layer3_0_bn2_running_var, ema_img_backbone_layer3_0_bn2_num_batches_tracked, ema_img_backbone_layer3_0_conv3_weight, ema_img_backbone_layer3_0_bn3_weight, ema_img_backbone_layer3_0_bn3_bias, ema_img_backbone_layer3_0_bn3_running_mean, ema_img_backbone_layer3_0_bn3_running_var, ema_img_backbone_layer3_0_bn3_num_batches_tracked, ema_img_backbone_layer3_0_downsample_0_weight, ema_img_backbone_layer3_0_downsample_1_weight, ema_img_backbone_layer3_0_downsample_1_bias, ema_img_backbone_layer3_0_downsample_1_running_mean, ema_img_backbone_layer3_0_downsample_1_running_var, ema_img_backbone_layer3_0_downsample_1_num_batches_tracked, ema_img_backbone_layer3_1_conv1_weight, ema_img_backbone_layer3_1_bn1_weight, ema_img_backbone_layer3_1_bn1_bias, ema_img_backbone_layer3_1_bn1_running_mean, ema_img_backbone_layer3_1_bn1_running_var, ema_img_backbone_layer3_1_bn1_num_batches_tracked, ema_img_backbone_layer3_1_conv2_weight, ema_img_backbone_layer3_1_bn2_weight, ema_img_backbone_layer3_1_bn2_bias, ema_img_backbone_layer3_1_bn2_running_mean, ema_img_backbone_layer3_1_bn2_running_var, ema_img_backbone_layer3_1_bn2_num_batches_tracked, ema_img_backbone_layer3_1_conv3_weight, ema_img_backbone_layer3_1_bn3_weight, ema_img_backbone_layer3_1_bn3_bias, ema_img_backbone_layer3_1_bn3_running_mean, ema_img_backbone_layer3_1_bn3_running_var, ema_img_backbone_layer3_1_bn3_num_batches_tracked, ema_img_backbone_layer3_2_conv1_weight, ema_img_backbone_layer3_2_bn1_weight, ema_img_backbone_layer3_2_bn1_bias, ema_img_backbone_layer3_2_bn1_running_mean, ema_img_backbone_layer3_2_bn1_running_var, ema_img_backbone_layer3_2_bn1_num_batches_tracked, ema_img_backbone_layer3_2_conv2_weight, ema_img_backbone_layer3_2_bn2_weight, ema_img_backbone_layer3_2_bn2_bias, ema_img_backbone_layer3_2_bn2_running_mean, ema_img_backbone_layer3_2_bn2_running_var, ema_img_backbone_layer3_2_bn2_num_batches_tracked, ema_img_backbone_layer3_2_conv3_weight, ema_img_backbone_layer3_2_bn3_weight, ema_img_backbone_layer3_2_bn3_bias, ema_img_backbone_layer3_2_bn3_running_mean, ema_img_backbone_layer3_2_bn3_running_var, ema_img_backbone_layer3_2_bn3_num_batches_tracked, ema_img_backbone_layer3_3_conv1_weight, ema_img_backbone_layer3_3_bn1_weight, ema_img_backbone_layer3_3_bn1_bias, ema_img_backbone_layer3_3_bn1_running_mean, ema_img_backbone_layer3_3_bn1_running_var, ema_img_backbone_layer3_3_bn1_num_batches_tracked, ema_img_backbone_layer3_3_conv2_weight, ema_img_backbone_layer3_3_bn2_weight, ema_img_backbone_layer3_3_bn2_bias, ema_img_backbone_layer3_3_bn2_running_mean, ema_img_backbone_layer3_3_bn2_running_var, ema_img_backbone_layer3_3_bn2_num_batches_tracked, ema_img_backbone_layer3_3_conv3_weight, ema_img_backbone_layer3_3_bn3_weight, ema_img_backbone_layer3_3_bn3_bias, ema_img_backbone_layer3_3_bn3_running_mean, ema_img_backbone_layer3_3_bn3_running_var, ema_img_backbone_layer3_3_bn3_num_batches_tracked, ema_img_backbone_layer3_4_conv1_weight, ema_img_backbone_layer3_4_bn1_weight, ema_img_backbone_layer3_4_bn1_bias, ema_img_backbone_layer3_4_bn1_running_mean, ema_img_backbone_layer3_4_bn1_running_var, ema_img_backbone_layer3_4_bn1_num_batches_tracked, ema_img_backbone_layer3_4_conv2_weight, ema_img_backbone_layer3_4_bn2_weight, ema_img_backbone_layer3_4_bn2_bias, ema_img_backbone_layer3_4_bn2_running_mean, ema_img_backbone_layer3_4_bn2_running_var, ema_img_backbone_layer3_4_bn2_num_batches_tracked, ema_img_backbone_layer3_4_conv3_weight, ema_img_backbone_layer3_4_bn3_weight, ema_img_backbone_layer3_4_bn3_bias, ema_img_backbone_layer3_4_bn3_running_mean, ema_img_backbone_layer3_4_bn3_running_var, ema_img_backbone_layer3_4_bn3_num_batches_tracked, ema_img_backbone_layer3_5_conv1_weight, ema_img_backbone_layer3_5_bn1_weight, ema_img_backbone_layer3_5_bn1_bias, ema_img_backbone_layer3_5_bn1_running_mean, ema_img_backbone_layer3_5_bn1_running_var, ema_img_backbone_layer3_5_bn1_num_batches_tracked, ema_img_backbone_layer3_5_conv2_weight, ema_img_backbone_layer3_5_bn2_weight, ema_img_backbone_layer3_5_bn2_bias, ema_img_backbone_layer3_5_bn2_running_mean, ema_img_backbone_layer3_5_bn2_running_var, ema_img_backbone_layer3_5_bn2_num_batches_tracked, ema_img_backbone_layer3_5_conv3_weight, ema_img_backbone_layer3_5_bn3_weight, ema_img_backbone_layer3_5_bn3_bias, ema_img_backbone_layer3_5_bn3_running_mean, ema_img_backbone_layer3_5_bn3_running_var, ema_img_backbone_layer3_5_bn3_num_batches_tracked, ema_img_backbone_layer4_0_conv1_weight, ema_img_backbone_layer4_0_bn1_weight, ema_img_backbone_layer4_0_bn1_bias, ema_img_backbone_layer4_0_bn1_running_mean, ema_img_backbone_layer4_0_bn1_running_var, ema_img_backbone_layer4_0_bn1_num_batches_tracked, ema_img_backbone_layer4_0_conv2_weight, ema_img_backbone_layer4_0_bn2_weight, ema_img_backbone_layer4_0_bn2_bias, ema_img_backbone_layer4_0_bn2_running_mean, ema_img_backbone_layer4_0_bn2_running_var, ema_img_backbone_layer4_0_bn2_num_batches_tracked, ema_img_backbone_layer4_0_conv3_weight, ema_img_backbone_layer4_0_bn3_weight, ema_img_backbone_layer4_0_bn3_bias, ema_img_backbone_layer4_0_bn3_running_mean, ema_img_backbone_layer4_0_bn3_running_var, ema_img_backbone_layer4_0_bn3_num_batches_tracked, ema_img_backbone_layer4_0_downsample_0_weight, ema_img_backbone_layer4_0_downsample_1_weight, ema_img_backbone_layer4_0_downsample_1_bias, ema_img_backbone_layer4_0_downsample_1_running_mean, ema_img_backbone_layer4_0_downsample_1_running_var, ema_img_backbone_layer4_0_downsample_1_num_batches_tracked, ema_img_backbone_layer4_1_conv1_weight, ema_img_backbone_layer4_1_bn1_weight, ema_img_backbone_layer4_1_bn1_bias, ema_img_backbone_layer4_1_bn1_running_mean, ema_img_backbone_layer4_1_bn1_running_var, ema_img_backbone_layer4_1_bn1_num_batches_tracked, ema_img_backbone_layer4_1_conv2_weight, ema_img_backbone_layer4_1_bn2_weight, ema_img_backbone_layer4_1_bn2_bias, ema_img_backbone_layer4_1_bn2_running_mean, ema_img_backbone_layer4_1_bn2_running_var, ema_img_backbone_layer4_1_bn2_num_batches_tracked, ema_img_backbone_layer4_1_conv3_weight, ema_img_backbone_layer4_1_bn3_weight, ema_img_backbone_layer4_1_bn3_bias, ema_img_backbone_layer4_1_bn3_running_mean, ema_img_backbone_layer4_1_bn3_running_var, ema_img_backbone_layer4_1_bn3_num_batches_tracked, ema_img_backbone_layer4_2_conv1_weight, ema_img_backbone_layer4_2_bn1_weight, ema_img_backbone_layer4_2_bn1_bias, ema_img_backbone_layer4_2_bn1_running_mean, ema_img_backbone_layer4_2_bn1_running_var, ema_img_backbone_layer4_2_bn1_num_batches_tracked, ema_img_backbone_layer4_2_conv2_weight, ema_img_backbone_layer4_2_bn2_weight, ema_img_backbone_layer4_2_bn2_bias, ema_img_backbone_layer4_2_bn2_running_mean, ema_img_backbone_layer4_2_bn2_running_var, ema_img_backbone_layer4_2_bn2_num_batches_tracked, ema_img_backbone_layer4_2_conv3_weight, ema_img_backbone_layer4_2_bn3_weight, ema_img_backbone_layer4_2_bn3_bias, ema_img_backbone_layer4_2_bn3_running_mean, ema_img_backbone_layer4_2_bn3_running_var, ema_img_backbone_layer4_2_bn3_num_batches_tracked, ema_img_neck_deblocks_0_0_weight, ema_img_neck_deblocks_0_1_weight, ema_img_neck_deblocks_0_1_bias, ema_img_neck_deblocks_0_1_running_mean, ema_img_neck_deblocks_0_1_running_var, ema_img_neck_deblocks_0_1_num_batches_tracked, ema_img_neck_deblocks_1_0_weight, ema_img_neck_deblocks_1_1_weight, ema_img_neck_deblocks_1_1_bias, ema_img_neck_deblocks_1_1_running_mean, ema_img_neck_deblocks_1_1_running_var, ema_img_neck_deblocks_1_1_num_batches_tracked, ema_img_neck_deblocks_2_0_weight, ema_img_neck_deblocks_2_1_weight, ema_img_neck_deblocks_2_1_bias, ema_img_neck_deblocks_2_1_running_mean, ema_img_neck_deblocks_2_1_running_var, ema_img_neck_deblocks_2_1_num_batches_tracked, ema_img_neck_deblocks_3_0_weight, ema_img_neck_deblocks_3_1_weight, ema_img_neck_deblocks_3_1_bias, ema_img_neck_deblocks_3_1_running_mean, ema_img_neck_deblocks_3_1_running_var, ema_img_neck_deblocks_3_1_num_batches_tracked, ema_img_view_transformer_dx, ema_img_view_transformer_bx, ema_img_view_transformer_nx, ema_img_view_transformer_frustum, ema_img_view_transformer_log_inv_gaussians, ema_img_view_transformer_bin_centers, ema_img_view_transformer_depthnet_weight, ema_img_view_transformer_depthnet_bias, ema_img_view_transformer_extra_depthnet_layers_0_0_conv1_weight, ema_img_view_transformer_extra_depthnet_layers_0_0_bn1_weight, ema_img_view_transformer_extra_depthnet_layers_0_0_bn1_bias, ema_img_view_transformer_extra_depthnet_layers_0_0_bn1_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_0_bn1_running_var, ema_img_view_transformer_extra_depthnet_layers_0_0_bn1_num_batches_tracked, ema_img_view_transformer_extra_depthnet_layers_0_0_conv2_weight, ema_img_view_transformer_extra_depthnet_layers_0_0_bn2_weight, ema_img_view_transformer_extra_depthnet_layers_0_0_bn2_bias, ema_img_view_transformer_extra_depthnet_layers_0_0_bn2_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_0_bn2_running_var, ema_img_view_transformer_extra_depthnet_layers_0_0_bn2_num_batches_tracked, ema_img_view_transformer_extra_depthnet_layers_0_0_downsample_weight, ema_img_view_transformer_extra_depthnet_layers_0_0_downsample_bias, ema_img_view_transformer_extra_depthnet_layers_0_1_conv1_weight, ema_img_view_transformer_extra_depthnet_layers_0_1_bn1_weight, ema_img_view_transformer_extra_depthnet_layers_0_1_bn1_bias, ema_img_view_transformer_extra_depthnet_layers_0_1_bn1_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_1_bn1_running_var, ema_img_view_transformer_extra_depthnet_layers_0_1_bn1_num_batches_tracked, ema_img_view_transformer_extra_depthnet_layers_0_1_conv2_weight, ema_img_view_transformer_extra_depthnet_layers_0_1_bn2_weight, ema_img_view_transformer_extra_depthnet_layers_0_1_bn2_bias, ema_img_view_transformer_extra_depthnet_layers_0_1_bn2_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_1_bn2_running_var, ema_img_view_transformer_extra_depthnet_layers_0_1_bn2_num_batches_tracked, ema_img_view_transformer_extra_depthnet_layers_0_2_conv1_weight, ema_img_view_transformer_extra_depthnet_layers_0_2_bn1_weight, ema_img_view_transformer_extra_depthnet_layers_0_2_bn1_bias, ema_img_view_transformer_extra_depthnet_layers_0_2_bn1_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_2_bn1_running_var, ema_img_view_transformer_extra_depthnet_layers_0_2_bn1_num_batches_tracked, ema_img_view_transformer_extra_depthnet_layers_0_2_conv2_weight, ema_img_view_transformer_extra_depthnet_layers_0_2_bn2_weight, ema_img_view_transformer_extra_depthnet_layers_0_2_bn2_bias, ema_img_view_transformer_extra_depthnet_layers_0_2_bn2_running_mean, ema_img_view_transformer_extra_depthnet_layers_0_2_bn2_running_var, ema_img_view_transformer_extra_depthnet_layers_0_2_bn2_num_batches_tracked, ema_img_view_transformer_featnet_weight, ema_img_view_transformer_featnet_bias, ema_img_view_transformer_dcn_0_weight, ema_img_view_transformer_dcn_0_conv_offset_weight, ema_img_view_transformer_dcn_0_conv_offset_bias, ema_img_view_transformer_dcn_1_weight, ema_img_view_transformer_dcn_1_bias, ema_img_view_transformer_dcn_1_running_mean, ema_img_view_transformer_dcn_1_running_var, ema_img_view_transformer_dcn_1_num_batches_tracked, ema_img_view_transformer_se_input_conv_weight, ema_img_view_transformer_se_input_conv_bias, ema_img_view_transformer_se_fc_0_weight, ema_img_view_transformer_se_fc_0_bias, ema_img_view_transformer_se_fc_0_running_mean, ema_img_view_transformer_se_fc_0_running_var, ema_img_view_transformer_se_fc_0_num_batches_tracked, ema_img_view_transformer_se_fc_1_weight, ema_img_view_transformer_se_fc_1_bias, ema_img_view_transformer_similarity_net_0_conv_weight, ema_img_view_transformer_similarity_net_0_bn_weight, ema_img_view_transformer_similarity_net_0_bn_bias, ema_img_view_transformer_similarity_net_0_bn_running_mean, ema_img_view_transformer_similarity_net_0_bn_running_var, ema_img_view_transformer_similarity_net_0_bn_num_batches_tracked, ema_img_view_transformer_similarity_net_1_conv_weight, ema_img_view_transformer_similarity_net_1_bn_weight, ema_img_view_transformer_similarity_net_1_bn_bias, ema_img_view_transformer_similarity_net_1_bn_running_mean, ema_img_view_transformer_similarity_net_1_bn_running_var, ema_img_view_transformer_similarity_net_1_bn_num_batches_tracked, ema_img_view_transformer_similarity_net_2_weight, ema_img_view_transformer_similarity_net_2_bias, ema_img_bev_encoder_backbone_layers_0_0_conv1_weight, ema_img_bev_encoder_backbone_layers_0_0_bn1_weight, ema_img_bev_encoder_backbone_layers_0_0_bn1_bias, ema_img_bev_encoder_backbone_layers_0_0_bn1_running_mean, ema_img_bev_encoder_backbone_layers_0_0_bn1_running_var, ema_img_bev_encoder_backbone_layers_0_0_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_0_0_conv2_weight, ema_img_bev_encoder_backbone_layers_0_0_bn2_weight, ema_img_bev_encoder_backbone_layers_0_0_bn2_bias, ema_img_bev_encoder_backbone_layers_0_0_bn2_running_mean, ema_img_bev_encoder_backbone_layers_0_0_bn2_running_var, ema_img_bev_encoder_backbone_layers_0_0_bn2_num_batches_tracked, ema_img_bev_encoder_backbone_layers_0_0_downsample_weight, ema_img_bev_encoder_backbone_layers_0_0_downsample_bias, ema_img_bev_encoder_backbone_layers_0_1_conv1_weight, ema_img_bev_encoder_backbone_layers_0_1_bn1_weight, ema_img_bev_encoder_backbone_layers_0_1_bn1_bias, ema_img_bev_encoder_backbone_layers_0_1_bn1_running_mean, ema_img_bev_encoder_backbone_layers_0_1_bn1_running_var, ema_img_bev_encoder_backbone_layers_0_1_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_0_1_conv2_weight, ema_img_bev_encoder_backbone_layers_0_1_bn2_weight, ema_img_bev_encoder_backbone_layers_0_1_bn2_bias, ema_img_bev_encoder_backbone_layers_0_1_bn2_running_mean, ema_img_bev_encoder_backbone_layers_0_1_bn2_running_var, ema_img_bev_encoder_backbone_layers_0_1_bn2_num_batches_tracked, ema_img_bev_encoder_backbone_layers_1_0_conv1_weight, ema_img_bev_encoder_backbone_layers_1_0_bn1_weight, ema_img_bev_encoder_backbone_layers_1_0_bn1_bias, ema_img_bev_encoder_backbone_layers_1_0_bn1_running_mean, ema_img_bev_encoder_backbone_layers_1_0_bn1_running_var, ema_img_bev_encoder_backbone_layers_1_0_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_1_0_conv2_weight, ema_img_bev_encoder_backbone_layers_1_0_bn2_weight, ema_img_bev_encoder_backbone_layers_1_0_bn2_bias, ema_img_bev_encoder_backbone_layers_1_0_bn2_running_mean, ema_img_bev_encoder_backbone_layers_1_0_bn2_running_var, ema_img_bev_encoder_backbone_layers_1_0_bn2_num_batches_tracked, ema_img_bev_encoder_backbone_layers_1_0_downsample_weight, ema_img_bev_encoder_backbone_layers_1_0_downsample_bias, ema_img_bev_encoder_backbone_layers_1_1_conv1_weight, ema_img_bev_encoder_backbone_layers_1_1_bn1_weight, ema_img_bev_encoder_backbone_layers_1_1_bn1_bias, ema_img_bev_encoder_backbone_layers_1_1_bn1_running_mean, ema_img_bev_encoder_backbone_layers_1_1_bn1_running_var, ema_img_bev_encoder_backbone_layers_1_1_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_1_1_conv2_weight, ema_img_bev_encoder_backbone_layers_1_1_bn2_weight, ema_img_bev_encoder_backbone_layers_1_1_bn2_bias, ema_img_bev_encoder_backbone_layers_1_1_bn2_running_mean, ema_img_bev_encoder_backbone_layers_1_1_bn2_running_var, ema_img_bev_encoder_backbone_layers_1_1_bn2_num_batches_tracked, ema_img_bev_encoder_backbone_layers_2_0_conv1_weight, ema_img_bev_encoder_backbone_layers_2_0_bn1_weight, ema_img_bev_encoder_backbone_layers_2_0_bn1_bias, ema_img_bev_encoder_backbone_layers_2_0_bn1_running_mean, ema_img_bev_encoder_backbone_layers_2_0_bn1_running_var, ema_img_bev_encoder_backbone_layers_2_0_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_2_0_conv2_weight, ema_img_bev_encoder_backbone_layers_2_0_bn2_weight, ema_img_bev_encoder_backbone_layers_2_0_bn2_bias, ema_img_bev_encoder_backbone_layers_2_0_bn2_running_mean, ema_img_bev_encoder_backbone_layers_2_0_bn2_running_var, ema_img_bev_encoder_backbone_layers_2_0_bn2_num_batches_tracked, ema_img_bev_encoder_backbone_layers_2_0_downsample_weight, ema_img_bev_encoder_backbone_layers_2_0_downsample_bias, ema_img_bev_encoder_backbone_layers_2_1_conv1_weight, ema_img_bev_encoder_backbone_layers_2_1_bn1_weight, ema_img_bev_encoder_backbone_layers_2_1_bn1_bias, ema_img_bev_encoder_backbone_layers_2_1_bn1_running_mean, ema_img_bev_encoder_backbone_layers_2_1_bn1_running_var, ema_img_bev_encoder_backbone_layers_2_1_bn1_num_batches_tracked, ema_img_bev_encoder_backbone_layers_2_1_conv2_weight, ema_img_bev_encoder_backbone_layers_2_1_bn2_weight, ema_img_bev_encoder_backbone_layers_2_1_bn2_bias, ema_img_bev_encoder_backbone_layers_2_1_bn2_running_mean, ema_img_bev_encoder_backbone_layers_2_1_bn2_running_var, ema_img_bev_encoder_backbone_layers_2_1_bn2_num_batches_tracked, ema_img_bev_encoder_neck_deblocks_0_0_weight, ema_img_bev_encoder_neck_deblocks_0_1_weight, ema_img_bev_encoder_neck_deblocks_0_1_bias, ema_img_bev_encoder_neck_deblocks_0_1_running_mean, ema_img_bev_encoder_neck_deblocks_0_1_running_var, ema_img_bev_encoder_neck_deblocks_0_1_num_batches_tracked, ema_img_bev_encoder_neck_deblocks_1_0_weight, ema_img_bev_encoder_neck_deblocks_1_1_weight, ema_img_bev_encoder_neck_deblocks_1_1_bias, ema_img_bev_encoder_neck_deblocks_1_1_running_mean, ema_img_bev_encoder_neck_deblocks_1_1_running_var, ema_img_bev_encoder_neck_deblocks_1_1_num_batches_tracked, ema_img_bev_encoder_neck_deblocks_2_0_weight, ema_img_bev_encoder_neck_deblocks_2_1_weight, ema_img_bev_encoder_neck_deblocks_2_1_bias, ema_img_bev_encoder_neck_deblocks_2_1_running_mean, ema_img_bev_encoder_neck_deblocks_2_1_running_var, ema_img_bev_encoder_neck_deblocks_2_1_num_batches_tracked, ema_img_bev_encoder_neck_deblocks_3_0_weight, ema_img_bev_encoder_neck_deblocks_3_1_weight, ema_img_bev_encoder_neck_deblocks_3_1_bias, ema_img_bev_encoder_neck_deblocks_3_1_running_mean, ema_img_bev_encoder_neck_deblocks_3_1_running_var, ema_img_bev_encoder_neck_deblocks_3_1_num_batches_tracked, ema_embed_0_weight, ema_embed_0_bias, ema_embed_1_weight, ema_embed_1_bias, ema_embed_1_running_mean, ema_embed_1_running_var, ema_embed_1_num_batches_tracked, ema_embed_3_weight, ema_embed_3_bias, ema_embed_4_weight, ema_embed_4_bias, ema_embed_4_running_mean, ema_embed_4_running_var, ema_embed_4_num_batches_tracked, ema_pre_process_net_layers_0_0_conv1_weight, ema_pre_process_net_layers_0_0_bn1_weight, ema_pre_process_net_layers_0_0_bn1_bias, ema_pre_process_net_layers_0_0_bn1_running_mean, ema_pre_process_net_layers_0_0_bn1_running_var, ema_pre_process_net_layers_0_0_bn1_num_batches_tracked, ema_pre_process_net_layers_0_0_conv2_weight, ema_pre_process_net_layers_0_0_bn2_weight, ema_pre_process_net_layers_0_0_bn2_bias, ema_pre_process_net_layers_0_0_bn2_running_mean, ema_pre_process_net_layers_0_0_bn2_running_var, ema_pre_process_net_layers_0_0_bn2_num_batches_tracked, ema_pre_process_net_layers_0_0_downsample_weight, ema_pre_process_net_layers_0_0_downsample_bias, ema_pre_process_net_layers_0_1_conv1_weight, ema_pre_process_net_layers_0_1_bn1_weight, ema_pre_process_net_layers_0_1_bn1_bias, ema_pre_process_net_layers_0_1_bn1_running_mean, ema_pre_process_net_layers_0_1_bn1_running_var, ema_pre_process_net_layers_0_1_bn1_num_batches_tracked, ema_pre_process_net_layers_0_1_conv2_weight, ema_pre_process_net_layers_0_1_bn2_weight, ema_pre_process_net_layers_0_1_bn2_bias, ema_pre_process_net_layers_0_1_bn2_running_mean, ema_pre_process_net_layers_0_1_bn2_running_var, ema_pre_process_net_layers_0_1_bn2_num_batches_tracked, ema_stereo_neck_deblocks_0_0_weight, ema_stereo_neck_deblocks_0_1_weight, ema_stereo_neck_deblocks_0_1_bias, ema_stereo_neck_deblocks_0_1_running_mean, ema_stereo_neck_deblocks_0_1_running_var, ema_stereo_neck_deblocks_0_1_num_batches_tracked, ema_stereo_neck_deblocks_1_0_weight, ema_stereo_neck_deblocks_1_1_weight, ema_stereo_neck_deblocks_1_1_bias, ema_stereo_neck_deblocks_1_1_running_mean, ema_stereo_neck_deblocks_1_1_running_var, ema_stereo_neck_deblocks_1_1_num_batches_tracked, ema_stereo_neck_deblocks_2_0_weight, ema_stereo_neck_deblocks_2_1_weight, ema_stereo_neck_deblocks_2_1_bias, ema_stereo_neck_deblocks_2_1_running_mean, ema_stereo_neck_deblocks_2_1_running_var, ema_stereo_neck_deblocks_2_1_num_batches_tracked, ema_stereo_neck_deblocks_3_0_weight, ema_stereo_neck_deblocks_3_1_weight, ema_stereo_neck_deblocks_3_1_bias, ema_stereo_neck_deblocks_3_1_running_mean, ema_stereo_neck_deblocks_3_1_running_var, ema_stereo_neck_deblocks_3_1_num_batches_tracked, ema_stereo_neck_final_conv_0_weight, ema_stereo_neck_final_conv_1_weight, ema_stereo_neck_final_conv_1_bias, ema_stereo_neck_final_conv_1_running_mean, ema_stereo_neck_final_conv_1_running_var, ema_stereo_neck_final_conv_1_num_batches_tracked, ema_stereo_neck_final_conv_3_weight missing keys in source state_dict: history_keyframe_time_conv.0.weight, history_keyframe_time_conv.0.bias, history_keyframe_time_conv.1.weight, history_keyframe_time_conv.1.bias, history_keyframe_time_conv.1.running_mean, history_keyframe_time_conv.1.running_var, history_keyframe_cat_conv.0.weight, history_keyframe_cat_conv.0.bias, history_keyframe_cat_conv.1.weight, history_keyframe_cat_conv.1.bias, history_keyframe_cat_conv.1.running_mean, history_keyframe_cat_conv.1.running_var 2023-03-13 02:02:27,814 - mmdet - INFO - resumed from epoch: 1, iter 2633 2023-03-13 02:02:27,817 - mmdet - INFO - Start running, host: jinhyun1@klab-neuron, work_dir: /home/jinhyun1/src/S22/SOLOFusion/work_dirs/r50-fp16_phase2 2023-03-13 02:02:27,817 - mmdet - INFO - Hooks will be executed in the following order: before_run: (ABOVE_NORMAL) WarmupFp16OptimizerHook (49 ) ExpMomentumEMAHook (NORMAL ) CheckpointHook (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- before_train_epoch: (49 ) ExpMomentumEMAHook (LOW ) IterTimerHook (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- before_train_iter: (LOW ) IterTimerHook -------------------- after_train_iter: (ABOVE_NORMAL) WarmupFp16OptimizerHook (49 ) ExpMomentumEMAHook (NORMAL ) CheckpointHook (LOW ) IterTimerHook (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- after_train_epoch: (49 ) ExpMomentumEMAHook (NORMAL ) CheckpointHook (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- before_val_epoch: (LOW ) IterTimerHook (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- before_val_iter: (LOW ) IterTimerHook -------------------- after_val_iter: (LOW ) IterTimerHook -------------------- after_val_epoch: (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- after_run: (VERY_LOW ) TextLoggerHook (VERY_LOW ) TensorboardLoggerHook -------------------- 2023-03-13 02:02:27,817 - mmdet - INFO - workflow: [('train', 1)], max: 10536 iters 2023-03-13 02:02:27,830 - mmdet - INFO - load checkpoint from work_dirs/r50-fp16_phase1/iter_2634.pth 2023-03-13 02:02:27,830 - mmdet - INFO - Use load_from_local loader 2023-03-13 02:02:28,482 - mmdet - WARNING - The model and loaded state dict do not match exactly size mismatch for ema_img_bev_encoder_backbone_layers_0_0_conv1_weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for ema_img_bev_encoder_backbone_layers_0_0_downsample_weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for ema_img_bev_encoder_neck_deblocks_0_0_weight: copying a param with shape torch.Size([80, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([160, 64, 1, 1]). size mismatch for img_bev_encoder_backbone.layers.0.0.conv1.weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for img_bev_encoder_backbone.layers.0.0.downsample.weight: copying a param with shape torch.Size([160, 80, 3, 3]) from checkpoint, the shape in current model is torch.Size([160, 160, 3, 3]). size mismatch for img_bev_encoder_neck.deblocks.0.0.weight: copying a param with shape torch.Size([80, 64, 1, 1]) from checkpoint, the shape in current model is torch.Size([160, 64, 1, 1]). missing keys in source state_dict: ema_history_keyframe_time_conv_0_weight, ema_history_keyframe_time_conv_0_bias, ema_history_keyframe_time_conv_1_weight, ema_history_keyframe_time_conv_1_bias, ema_history_keyframe_time_conv_1_running_mean, ema_history_keyframe_time_conv_1_running_var, ema_history_keyframe_cat_conv_0_weight, ema_history_keyframe_cat_conv_0_bias, ema_history_keyframe_cat_conv_1_weight, ema_history_keyframe_cat_conv_1_bias, ema_history_keyframe_cat_conv_1_running_mean, ema_history_keyframe_cat_conv_1_running_var, history_keyframe_time_conv.0.weight, history_keyframe_time_conv.0.bias, history_keyframe_time_conv.1.weight, history_keyframe_time_conv.1.bias, history_keyframe_time_conv.1.running_mean, history_keyframe_time_conv.1.running_var, history_keyframe_cat_conv.0.weight, history_keyframe_cat_conv.0.bias, history_keyframe_cat_conv.1.weight, history_keyframe_cat_conv.1.bias, history_keyframe_cat_conv.1.running_mean, history_keyframe_cat_conv.1.running_var 2023-03-13 02:02:28,489 - mmdet - INFO - resumed from epoch: 1, iter 2633 2023-03-13 02:02:28,492 - mmdet - INFO - Checkpoints will be saved to /home/jinhyun1/src/S22/SOLOFusion/work_dirs/r50-fp16_phase2 by HardDiskBackend. 2023-03-13 02:03:17,767 - mmdet - INFO - Iter [2650/10536] lr: 2.000e-04, eta: 17:56:53, time: 2.786, data_time: 0.353, memory: 32126, loss_depth: 9.2246, task0.loss_xy: 0.1178, task0.loss_z: 0.1301, task0.loss_whl: 0.0735, task0.loss_yaw: 0.3069, task0.loss_vel: 0.5081, task0.loss_heatmap: 1.7998, task1.loss_xy: 0.1217, task1.loss_z: 0.1583, task1.loss_whl: 0.1563, task1.loss_yaw: 0.2928, task1.loss_vel: 0.4557, task1.loss_heatmap: 2.4299, task2.loss_xy: 0.1264, task2.loss_z: 0.1533, task2.loss_whl: 0.1576, task2.loss_yaw: 0.2963, task2.loss_vel: 0.6007, task2.loss_heatmap: 2.5641, task3.loss_xy: 0.1262, task3.loss_z: 0.0996, task3.loss_whl: 0.1286, task3.loss_yaw: 0.2750, task3.loss_vel: 0.0212, task3.loss_heatmap: 2.0111, task4.loss_xy: 0.1275, task4.loss_z: 0.1244, task4.loss_whl: 0.1350, task4.loss_yaw: 0.3000, task4.loss_vel: 0.1897, task4.loss_heatmap: 2.3076, task5.loss_xy: 0.1243, task5.loss_z: 0.1198, task5.loss_whl: 0.1647, task5.loss_yaw: 0.3049, task5.loss_vel: 0.2037, task5.loss_heatmap: 2.0437, loss: 28.4807, grad_norm: 38.4696 2023-03-13 02:05:18,844 - mmdet - INFO - Iter [2700/10536] lr: 2.000e-04, eta: 8:27:33, time: 2.422, data_time: 0.066, memory: 32126, loss_depth: 9.2988, task0.loss_xy: 0.1122, task0.loss_z: 0.1070, task0.loss_whl: 0.0638, task0.loss_yaw: 0.2890, task0.loss_vel: 0.4745, task0.loss_heatmap: 1.3909, task1.loss_xy: 0.1174, task1.loss_z: 0.1423, task1.loss_whl: 0.1380, task1.loss_yaw: 0.2751, task1.loss_vel: 0.3175, task1.loss_heatmap: 1.9061, task2.loss_xy: 0.1195, task2.loss_z: 0.1412, task2.loss_whl: 0.1234, task2.loss_yaw: 0.2916, task2.loss_vel: 0.5052, task2.loss_heatmap: 1.7853, task3.loss_xy: 0.1188, task3.loss_z: 0.0758, task3.loss_whl: 0.1418, task3.loss_yaw: 0.2741, task3.loss_vel: 0.0186, task3.loss_heatmap: 1.1263, task4.loss_xy: 0.1131, task4.loss_z: 0.0873, task4.loss_whl: 0.1232, task4.loss_yaw: 0.3048, task4.loss_vel: 0.3789, task4.loss_heatmap: 1.4315, task5.loss_xy: 0.1178, task5.loss_z: 0.0993, task5.loss_whl: 0.1536, task5.loss_yaw: 0.3057, task5.loss_vel: 0.1848, task5.loss_heatmap: 1.5420, loss: 24.1962, grad_norm: 20.3803 2023-03-13 02:07:17,573 - mmdet - INFO - Iter [2750/10536] lr: 2.000e-04, eta: 7:00:28, time: 2.374, data_time: 0.062, memory: 32211, loss_depth: 9.2780, task0.loss_xy: 0.1091, task0.loss_z: 0.0959, task0.loss_whl: 0.0607, task0.loss_yaw: 0.2667, task0.loss_vel: 0.4770, task0.loss_heatmap: 1.2343, task1.loss_xy: 0.1139, task1.loss_z: 0.1158, task1.loss_whl: 0.1255, task1.loss_yaw: 0.2615, task1.loss_vel: 0.2974, task1.loss_heatmap: 1.7076, task2.loss_xy: 0.1180, task2.loss_z: 0.1180, task2.loss_whl: 0.1354, task2.loss_yaw: 0.2944, task2.loss_vel: 0.3491, task2.loss_heatmap: 1.8731, task3.loss_xy: 0.1143, task3.loss_z: 0.0726, task3.loss_whl: 0.1363, task3.loss_yaw: 0.2830, task3.loss_vel: 0.0219, task3.loss_heatmap: 1.0940, task4.loss_xy: 0.1107, task4.loss_z: 0.0902, task4.loss_whl: 0.1188, task4.loss_yaw: 0.3014, task4.loss_vel: 0.4670, task4.loss_heatmap: 1.4472, task5.loss_xy: 0.1158, task5.loss_z: 0.0894, task5.loss_whl: 0.1436, task5.loss_yaw: 0.3010, task5.loss_vel: 0.1973, task5.loss_heatmap: 1.4880, loss: 23.6238, grad_norm: 15.9416 2023-03-13 02:09:24,853 - mmdet - INFO - Iter [2800/10536] lr: 2.000e-04, eta: 6:30:57, time: 2.546, data_time: 0.065, memory: 32211, loss_depth: 9.2309, task0.loss_xy: 0.1091, task0.loss_z: 0.0948, task0.loss_whl: 0.0606, task0.loss_yaw: 0.2580, task0.loss_vel: 0.4436, task0.loss_heatmap: 1.2705, task1.loss_xy: 0.1139, task1.loss_z: 0.1193, task1.loss_whl: 0.1217, task1.loss_yaw: 0.2547, task1.loss_vel: 0.3659, task1.loss_heatmap: 1.7313, task2.loss_xy: 0.1197, task2.loss_z: 0.1161, task2.loss_whl: 0.1191, task2.loss_yaw: 0.2914, task2.loss_vel: 0.5350, task2.loss_heatmap: 1.7390, task3.loss_xy: 0.1150, task3.loss_z: 0.0798, task3.loss_whl: 0.1562, task3.loss_yaw: 0.2618, task3.loss_vel: 0.0145, task3.loss_heatmap: 1.1594, task4.loss_xy: 0.1075, task4.loss_z: 0.0840, task4.loss_whl: 0.1124, task4.loss_yaw: 0.3014, task4.loss_vel: 0.4287, task4.loss_heatmap: 1.6182, task5.loss_xy: 0.1153, task5.loss_z: 0.0889, task5.loss_whl: 0.1421, task5.loss_yaw: 0.2995, task5.loss_vel: 0.1802, task5.loss_heatmap: 1.3769, loss: 23.7363, grad_norm: 18.0103 2023-03-13 02:11:28,586 - mmdet - INFO - Iter [2850/10536] lr: 2.000e-04, eta: 6:11:58, time: 2.475, data_time: 0.066, memory: 32211, loss_depth: 9.2828, task0.loss_xy: 0.1087, task0.loss_z: 0.0924, task0.loss_whl: 0.0595, task0.loss_yaw: 0.2541, task0.loss_vel: 0.5225, task0.loss_heatmap: 1.2811, task1.loss_xy: 0.1149, task1.loss_z: 0.1136, task1.loss_whl: 0.1186, task1.loss_yaw: 0.2648, task1.loss_vel: 0.3189, task1.loss_heatmap: 1.8096, task2.loss_xy: 0.1180, task2.loss_z: 0.1142, task2.loss_whl: 0.1126, task2.loss_yaw: 0.2893, task2.loss_vel: 0.4943, task2.loss_heatmap: 1.7041, task3.loss_xy: 0.1159, task3.loss_z: 0.0657, task3.loss_whl: 0.1169, task3.loss_yaw: 0.2459, task3.loss_vel: 0.0144, task3.loss_heatmap: 1.1190, task4.loss_xy: 0.1108, task4.loss_z: 0.0863, task4.loss_whl: 0.1231, task4.loss_yaw: 0.3076, task4.loss_vel: 0.5077, task4.loss_heatmap: 1.6051, task5.loss_xy: 0.1150, task5.loss_z: 0.0826, task5.loss_whl: 0.1317, task5.loss_yaw: 0.2968, task5.loss_vel: 0.2155, task5.loss_heatmap: 1.4236, loss: 23.8578, grad_norm: 15.9451 2023-03-13 02:13:29,242 - mmdet - INFO - Iter [2900/10536] lr: 2.000e-04, eta: 5:57:51, time: 2.413, data_time: 0.064, memory: 32211, loss_depth: 9.2966, task0.loss_xy: 0.1081, task0.loss_z: 0.0921, task0.loss_whl: 0.0617, task0.loss_yaw: 0.2503, task0.loss_vel: 0.4513, task0.loss_heatmap: 1.2561, task1.loss_xy: 0.1116, task1.loss_z: 0.1150, task1.loss_whl: 0.1054, task1.loss_yaw: 0.2492, task1.loss_vel: 0.3515, task1.loss_heatmap: 1.7175, task2.loss_xy: 0.1142, task2.loss_z: 0.1144, task2.loss_whl: 0.1111, task2.loss_yaw: 0.2737, task2.loss_vel: 0.5886, task2.loss_heatmap: 1.7077, task3.loss_xy: 0.1134, task3.loss_z: 0.0708, task3.loss_whl: 0.1285, task3.loss_yaw: 0.2783, task3.loss_vel: 0.0176, task3.loss_heatmap: 1.1395, task4.loss_xy: 0.1073, task4.loss_z: 0.0732, task4.loss_whl: 0.1047, task4.loss_yaw: 0.3045, task4.loss_vel: 0.3137, task4.loss_heatmap: 1.3975, task5.loss_xy: 0.1149, task5.loss_z: 0.0933, task5.loss_whl: 0.1318, task5.loss_yaw: 0.2934, task5.loss_vel: 0.1898, task5.loss_heatmap: 1.4459, loss: 23.3943, grad_norm: 15.7535 2023-03-13 02:15:30,185 - mmdet - INFO - Iter [2950/10536] lr: 2.000e-04, eta: 5:47:41, time: 2.419, data_time: 0.060, memory: 32211, loss_depth: 9.2198, task0.loss_xy: 0.1073, task0.loss_z: 0.0917, task0.loss_whl: 0.0609, task0.loss_yaw: 0.2450, task0.loss_vel: 0.4981, task0.loss_heatmap: 1.2219, task1.loss_xy: 0.1130, task1.loss_z: 0.1159, task1.loss_whl: 0.1180, task1.loss_yaw: 0.2586, task1.loss_vel: 0.3249, task1.loss_heatmap: 1.7408, task2.loss_xy: 0.1189, task2.loss_z: 0.1266, task2.loss_whl: 0.1167, task2.loss_yaw: 0.2914, task2.loss_vel: 0.3785, task2.loss_heatmap: 1.8222, task3.loss_xy: 0.1136, task3.loss_z: 0.0697, task3.loss_whl: 0.1321, task3.loss_yaw: 0.2692, task3.loss_vel: 0.0205, task3.loss_heatmap: 1.0499, task4.loss_xy: 0.1033, task4.loss_z: 0.0748, task4.loss_whl: 0.1068, task4.loss_yaw: 0.3018, task4.loss_vel: 0.5099, task4.loss_heatmap: 1.2896, task5.loss_xy: 0.1142, task5.loss_z: 0.0811, task5.loss_whl: 0.1373, task5.loss_yaw: 0.2925, task5.loss_vel: 0.1939, task5.loss_heatmap: 1.3656, loss: 23.1960, grad_norm: 13.5957 2023-03-13 02:17:29,429 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 02:17:29,429 - mmdet - INFO - Iter [3000/10536] lr: 2.000e-04, eta: 5:39:08, time: 2.385, data_time: 0.062, memory: 32211, loss_depth: 9.2632, task0.loss_xy: 0.1066, task0.loss_z: 0.0883, task0.loss_whl: 0.0587, task0.loss_yaw: 0.2384, task0.loss_vel: 0.4472, task0.loss_heatmap: 1.2128, task1.loss_xy: 0.1107, task1.loss_z: 0.1109, task1.loss_whl: 0.1157, task1.loss_yaw: 0.2425, task1.loss_vel: 0.3612, task1.loss_heatmap: 1.6795, task2.loss_xy: 0.1184, task2.loss_z: 0.1067, task2.loss_whl: 0.1070, task2.loss_yaw: 0.2887, task2.loss_vel: 0.5123, task2.loss_heatmap: 1.6200, task3.loss_xy: 0.1117, task3.loss_z: 0.0711, task3.loss_whl: 0.1320, task3.loss_yaw: 0.2746, task3.loss_vel: 0.0153, task3.loss_heatmap: 1.0454, task4.loss_xy: 0.1028, task4.loss_z: 0.0816, task4.loss_whl: 0.1167, task4.loss_yaw: 0.2989, task4.loss_vel: 0.3566, task4.loss_heatmap: 1.3978, task5.loss_xy: 0.1145, task5.loss_z: 0.0865, task5.loss_whl: 0.1346, task5.loss_yaw: 0.2896, task5.loss_vel: 0.1906, task5.loss_heatmap: 1.3508, loss: 22.9599, grad_norm: 15.4288 2023-03-13 02:19:30,832 - mmdet - INFO - Iter [3050/10536] lr: 2.000e-04, eta: 5:32:49, time: 2.428, data_time: 0.069, memory: 32211, loss_depth: 9.2135, task0.loss_xy: 0.1056, task0.loss_z: 0.0860, task0.loss_whl: 0.0590, task0.loss_yaw: 0.2308, task0.loss_vel: 0.4138, task0.loss_heatmap: 1.1923, task1.loss_xy: 0.1126, task1.loss_z: 0.1072, task1.loss_whl: 0.1083, task1.loss_yaw: 0.2474, task1.loss_vel: 0.3914, task1.loss_heatmap: 1.6793, task2.loss_xy: 0.1175, task2.loss_z: 0.1038, task2.loss_whl: 0.1061, task2.loss_yaw: 0.2845, task2.loss_vel: 0.3818, task2.loss_heatmap: 1.5524, task3.loss_xy: 0.1127, task3.loss_z: 0.0716, task3.loss_whl: 0.1166, task3.loss_yaw: 0.2736, task3.loss_vel: 0.0139, task3.loss_heatmap: 1.0263, task4.loss_xy: 0.0995, task4.loss_z: 0.0718, task4.loss_whl: 0.1068, task4.loss_yaw: 0.2848, task4.loss_vel: 0.4143, task4.loss_heatmap: 1.3526, task5.loss_xy: 0.1136, task5.loss_z: 0.0868, task5.loss_whl: 0.1420, task5.loss_yaw: 0.2878, task5.loss_vel: 0.1842, task5.loss_heatmap: 1.4558, loss: 22.7080, grad_norm: 13.5090 2023-03-13 02:21:31,270 - mmdet - INFO - Iter [3100/10536] lr: 2.000e-04, eta: 5:27:09, time: 2.409, data_time: 0.064, memory: 32211, loss_depth: 9.1498, task0.loss_xy: 0.1054, task0.loss_z: 0.0904, task0.loss_whl: 0.0596, task0.loss_yaw: 0.2341, task0.loss_vel: 0.4232, task0.loss_heatmap: 1.1987, task1.loss_xy: 0.1111, task1.loss_z: 0.1010, task1.loss_whl: 0.1134, task1.loss_yaw: 0.2479, task1.loss_vel: 0.3953, task1.loss_heatmap: 1.7034, task2.loss_xy: 0.1179, task2.loss_z: 0.1023, task2.loss_whl: 0.1166, task2.loss_yaw: 0.2822, task2.loss_vel: 0.6019, task2.loss_heatmap: 1.5131, task3.loss_xy: 0.1138, task3.loss_z: 0.0716, task3.loss_whl: 0.1462, task3.loss_yaw: 0.2823, task3.loss_vel: 0.0219, task3.loss_heatmap: 1.0757, task4.loss_xy: 0.1009, task4.loss_z: 0.0753, task4.loss_whl: 0.1111, task4.loss_yaw: 0.2999, task4.loss_vel: 0.3637, task4.loss_heatmap: 1.2006, task5.loss_xy: 0.1131, task5.loss_z: 0.0801, task5.loss_whl: 0.1394, task5.loss_yaw: 0.2889, task5.loss_vel: 0.1861, task5.loss_heatmap: 1.3779, loss: 22.7157, grad_norm: 13.2846 2023-03-13 02:23:32,401 - mmdet - INFO - Iter [3150/10536] lr: 2.000e-04, eta: 5:22:22, time: 2.423, data_time: 0.061, memory: 32211, loss_depth: 9.1498, task0.loss_xy: 0.1061, task0.loss_z: 0.0854, task0.loss_whl: 0.0587, task0.loss_yaw: 0.2248, task0.loss_vel: 0.4389, task0.loss_heatmap: 1.1890, task1.loss_xy: 0.1116, task1.loss_z: 0.1048, task1.loss_whl: 0.1156, task1.loss_yaw: 0.2483, task1.loss_vel: 0.3555, task1.loss_heatmap: 1.6749, task2.loss_xy: 0.1169, task2.loss_z: 0.1054, task2.loss_whl: 0.1195, task2.loss_yaw: 0.2861, task2.loss_vel: 0.4547, task2.loss_heatmap: 1.5029, task3.loss_xy: 0.1107, task3.loss_z: 0.0615, task3.loss_whl: 0.1342, task3.loss_yaw: 0.2678, task3.loss_vel: 0.0155, task3.loss_heatmap: 0.9773, task4.loss_xy: 0.1042, task4.loss_z: 0.0763, task4.loss_whl: 0.1150, task4.loss_yaw: 0.2963, task4.loss_vel: 0.3649, task4.loss_heatmap: 1.2175, task5.loss_xy: 0.1113, task5.loss_z: 0.0756, task5.loss_whl: 0.1430, task5.loss_yaw: 0.2874, task5.loss_vel: 0.1713, task5.loss_heatmap: 1.3387, loss: 22.3177, grad_norm: 13.7523 2023-03-13 02:25:34,182 - mmdet - INFO - Iter [3200/10536] lr: 2.000e-04, eta: 5:18:13, time: 2.436, data_time: 0.059, memory: 32211, loss_depth: 8.9840, task0.loss_xy: 0.1072, task0.loss_z: 0.0877, task0.loss_whl: 0.0603, task0.loss_yaw: 0.2329, task0.loss_vel: 0.4372, task0.loss_heatmap: 1.2541, task1.loss_xy: 0.1133, task1.loss_z: 0.1069, task1.loss_whl: 0.1167, task1.loss_yaw: 0.2444, task1.loss_vel: 0.2992, task1.loss_heatmap: 1.7090, task2.loss_xy: 0.1164, task2.loss_z: 0.1061, task2.loss_whl: 0.1131, task2.loss_yaw: 0.2853, task2.loss_vel: 0.4185, task2.loss_heatmap: 1.5836, task3.loss_xy: 0.1106, task3.loss_z: 0.0632, task3.loss_whl: 0.1324, task3.loss_yaw: 0.2736, task3.loss_vel: 0.0157, task3.loss_heatmap: 0.9642, task4.loss_xy: 0.1047, task4.loss_z: 0.0803, task4.loss_whl: 0.1096, task4.loss_yaw: 0.3041, task4.loss_vel: 0.3460, task4.loss_heatmap: 1.2632, task5.loss_xy: 0.1125, task5.loss_z: 0.0820, task5.loss_whl: 0.1383, task5.loss_yaw: 0.2884, task5.loss_vel: 0.1676, task5.loss_heatmap: 1.3287, loss: 22.2607, grad_norm: 12.0939 2023-03-13 02:27:36,964 - mmdet - INFO - Iter [3250/10536] lr: 2.000e-04, eta: 5:14:36, time: 2.456, data_time: 0.060, memory: 32211, loss_depth: 8.9589, task0.loss_xy: 0.1065, task0.loss_z: 0.0862, task0.loss_whl: 0.0598, task0.loss_yaw: 0.2233, task0.loss_vel: 0.3447, task0.loss_heatmap: 1.1727, task1.loss_xy: 0.1114, task1.loss_z: 0.1123, task1.loss_whl: 0.1125, task1.loss_yaw: 0.2436, task1.loss_vel: 0.3456, task1.loss_heatmap: 1.6630, task2.loss_xy: 0.1164, task2.loss_z: 0.0953, task2.loss_whl: 0.1206, task2.loss_yaw: 0.2842, task2.loss_vel: 0.4879, task2.loss_heatmap: 1.6107, task3.loss_xy: 0.1120, task3.loss_z: 0.0664, task3.loss_whl: 0.1202, task3.loss_yaw: 0.2484, task3.loss_vel: 0.0154, task3.loss_heatmap: 1.0392, task4.loss_xy: 0.1040, task4.loss_z: 0.0756, task4.loss_whl: 0.1037, task4.loss_yaw: 0.2940, task4.loss_vel: 0.4187, task4.loss_heatmap: 1.2946, task5.loss_xy: 0.1138, task5.loss_z: 0.0850, task5.loss_whl: 0.1356, task5.loss_yaw: 0.2850, task5.loss_vel: 0.1828, task5.loss_heatmap: 1.3383, loss: 22.2883, grad_norm: 14.6395 2023-03-13 02:29:36,269 - mmdet - INFO - Iter [3300/10536] lr: 2.000e-04, eta: 5:10:35, time: 2.386, data_time: 0.059, memory: 32211, loss_depth: 9.2602, task0.loss_xy: 0.1052, task0.loss_z: 0.0815, task0.loss_whl: 0.0593, task0.loss_yaw: 0.2253, task0.loss_vel: 0.3526, task0.loss_heatmap: 1.1919, task1.loss_xy: 0.1111, task1.loss_z: 0.1047, task1.loss_whl: 0.1096, task1.loss_yaw: 0.2375, task1.loss_vel: 0.2933, task1.loss_heatmap: 1.5931, task2.loss_xy: 0.1173, task2.loss_z: 0.1102, task2.loss_whl: 0.1133, task2.loss_yaw: 0.2743, task2.loss_vel: 0.5100, task2.loss_heatmap: 1.6449, task3.loss_xy: 0.1132, task3.loss_z: 0.0642, task3.loss_whl: 0.1078, task3.loss_yaw: 0.2670, task3.loss_vel: 0.0189, task3.loss_heatmap: 1.0825, task4.loss_xy: 0.0996, task4.loss_z: 0.0666, task4.loss_whl: 0.1093, task4.loss_yaw: 0.2874, task4.loss_vel: 0.3379, task4.loss_heatmap: 1.1393, task5.loss_xy: 0.1143, task5.loss_z: 0.0831, task5.loss_whl: 0.1391, task5.loss_yaw: 0.2840, task5.loss_vel: 0.1804, task5.loss_heatmap: 1.4521, loss: 22.4418, grad_norm: 13.8242 2023-03-13 02:31:37,010 - mmdet - INFO - Iter [3350/10536] lr: 2.000e-04, eta: 5:07:06, time: 2.415, data_time: 0.060, memory: 32211, loss_depth: 9.1437, task0.loss_xy: 0.1051, task0.loss_z: 0.0809, task0.loss_whl: 0.0585, task0.loss_yaw: 0.2222, task0.loss_vel: 0.3699, task0.loss_heatmap: 1.1588, task1.loss_xy: 0.1080, task1.loss_z: 0.0973, task1.loss_whl: 0.1118, task1.loss_yaw: 0.2369, task1.loss_vel: 0.2648, task1.loss_heatmap: 1.5764, task2.loss_xy: 0.1143, task2.loss_z: 0.1123, task2.loss_whl: 0.1046, task2.loss_yaw: 0.2848, task2.loss_vel: 0.4251, task2.loss_heatmap: 1.5902, task3.loss_xy: 0.1135, task3.loss_z: 0.0594, task3.loss_whl: 0.1114, task3.loss_yaw: 0.2385, task3.loss_vel: 0.0163, task3.loss_heatmap: 1.0157, task4.loss_xy: 0.1004, task4.loss_z: 0.0663, task4.loss_whl: 0.1082, task4.loss_yaw: 0.2969, task4.loss_vel: 0.3713, task4.loss_heatmap: 1.2532, task5.loss_xy: 0.1123, task5.loss_z: 0.0764, task5.loss_whl: 0.1355, task5.loss_yaw: 0.2851, task5.loss_vel: 0.1797, task5.loss_heatmap: 1.3346, loss: 22.0405, grad_norm: 11.8122 2023-03-13 02:33:36,692 - mmdet - INFO - Iter [3400/10536] lr: 2.000e-04, eta: 5:03:38, time: 2.394, data_time: 0.060, memory: 32270, loss_depth: 9.1649, task0.loss_xy: 0.1058, task0.loss_z: 0.0837, task0.loss_whl: 0.0576, task0.loss_yaw: 0.2186, task0.loss_vel: 0.3670, task0.loss_heatmap: 1.1679, task1.loss_xy: 0.1112, task1.loss_z: 0.1091, task1.loss_whl: 0.1103, task1.loss_yaw: 0.2389, task1.loss_vel: 0.3175, task1.loss_heatmap: 1.7071, task2.loss_xy: 0.1148, task2.loss_z: 0.1118, task2.loss_whl: 0.1057, task2.loss_yaw: 0.2791, task2.loss_vel: 0.5051, task2.loss_heatmap: 1.5998, task3.loss_xy: 0.1107, task3.loss_z: 0.0654, task3.loss_whl: 0.1163, task3.loss_yaw: 0.2375, task3.loss_vel: 0.0226, task3.loss_heatmap: 1.0664, task4.loss_xy: 0.1044, task4.loss_z: 0.0790, task4.loss_whl: 0.1069, task4.loss_yaw: 0.2943, task4.loss_vel: 0.4994, task4.loss_heatmap: 1.3973, task5.loss_xy: 0.1138, task5.loss_z: 0.0833, task5.loss_whl: 0.1356, task5.loss_yaw: 0.2866, task5.loss_vel: 0.1836, task5.loss_heatmap: 1.3851, loss: 22.7643, grad_norm: 12.7516 2023-03-13 02:35:36,899 - mmdet - INFO - Iter [3450/10536] lr: 2.000e-04, eta: 5:00:26, time: 2.404, data_time: 0.067, memory: 32270, loss_depth: 9.1690, task0.loss_xy: 0.1050, task0.loss_z: 0.0821, task0.loss_whl: 0.0574, task0.loss_yaw: 0.2154, task0.loss_vel: 0.3776, task0.loss_heatmap: 1.1749, task1.loss_xy: 0.1110, task1.loss_z: 0.0965, task1.loss_whl: 0.1133, task1.loss_yaw: 0.2446, task1.loss_vel: 0.2843, task1.loss_heatmap: 1.6364, task2.loss_xy: 0.1139, task2.loss_z: 0.1027, task2.loss_whl: 0.1117, task2.loss_yaw: 0.2729, task2.loss_vel: 0.4826, task2.loss_heatmap: 1.4883, task3.loss_xy: 0.1096, task3.loss_z: 0.0587, task3.loss_whl: 0.1192, task3.loss_yaw: 0.2281, task3.loss_vel: 0.0175, task3.loss_heatmap: 0.9826, task4.loss_xy: 0.1045, task4.loss_z: 0.0701, task4.loss_whl: 0.1086, task4.loss_yaw: 0.2948, task4.loss_vel: 0.3466, task4.loss_heatmap: 1.2703, task5.loss_xy: 0.1121, task5.loss_z: 0.0718, task5.loss_whl: 0.1318, task5.loss_yaw: 0.2853, task5.loss_vel: 0.1740, task5.loss_heatmap: 1.3139, loss: 22.0392, grad_norm: 12.3869 2023-03-13 02:37:36,716 - mmdet - INFO - Iter [3500/10536] lr: 2.000e-04, eta: 4:57:19, time: 2.396, data_time: 0.062, memory: 32270, loss_depth: 9.1137, task0.loss_xy: 0.1067, task0.loss_z: 0.0851, task0.loss_whl: 0.0575, task0.loss_yaw: 0.2182, task0.loss_vel: 0.3154, task0.loss_heatmap: 1.2110, task1.loss_xy: 0.1119, task1.loss_z: 0.1008, task1.loss_whl: 0.1165, task1.loss_yaw: 0.2403, task1.loss_vel: 0.3166, task1.loss_heatmap: 1.7138, task2.loss_xy: 0.1170, task2.loss_z: 0.1095, task2.loss_whl: 0.1243, task2.loss_yaw: 0.2798, task2.loss_vel: 0.4808, task2.loss_heatmap: 1.6495, task3.loss_xy: 0.1110, task3.loss_z: 0.0697, task3.loss_whl: 0.1352, task3.loss_yaw: 0.2623, task3.loss_vel: 0.0188, task3.loss_heatmap: 1.0995, task4.loss_xy: 0.1028, task4.loss_z: 0.0697, task4.loss_whl: 0.1152, task4.loss_yaw: 0.2968, task4.loss_vel: 0.2954, task4.loss_heatmap: 1.1592, task5.loss_xy: 0.1135, task5.loss_z: 0.0841, task5.loss_whl: 0.1334, task5.loss_yaw: 0.2843, task5.loss_vel: 0.1815, task5.loss_heatmap: 1.3496, loss: 22.3506, grad_norm: 12.9666 2023-03-13 02:38:04,809 - mmdet - INFO - Saving checkpoint at 3512 iterations 2023-03-13 02:39:37,528 - mmdet - INFO - Iter [3550/10536] lr: 2.000e-04, eta: 4:54:27, time: 2.416, data_time: 0.060, memory: 32270, loss_depth: 8.9810, task0.loss_xy: 0.1034, task0.loss_z: 0.0817, task0.loss_whl: 0.0575, task0.loss_yaw: 0.2107, task0.loss_vel: 0.3646, task0.loss_heatmap: 1.1481, task1.loss_xy: 0.1111, task1.loss_z: 0.0977, task1.loss_whl: 0.0997, task1.loss_yaw: 0.2320, task1.loss_vel: 0.2897, task1.loss_heatmap: 1.5211, task2.loss_xy: 0.1149, task2.loss_z: 0.0960, task2.loss_whl: 0.0986, task2.loss_yaw: 0.2679, task2.loss_vel: 0.5162, task2.loss_heatmap: 1.4429, task3.loss_xy: 0.1102, task3.loss_z: 0.0649, task3.loss_whl: 0.1392, task3.loss_yaw: 0.2816, task3.loss_vel: 0.0138, task3.loss_heatmap: 0.9685, task4.loss_xy: 0.1034, task4.loss_z: 0.0664, task4.loss_whl: 0.1029, task4.loss_yaw: 0.2980, task4.loss_vel: 0.3953, task4.loss_heatmap: 1.1417, task5.loss_xy: 0.1125, task5.loss_z: 0.0722, task5.loss_whl: 0.1292, task5.loss_yaw: 0.2814, task5.loss_vel: 0.1911, task5.loss_heatmap: 1.3152, loss: 21.6221, grad_norm: 14.2269 2023-03-13 02:41:37,598 - mmdet - INFO - Iter [3600/10536] lr: 2.000e-04, eta: 4:51:35, time: 2.401, data_time: 0.059, memory: 32270, loss_depth: 9.1211, task0.loss_xy: 0.1043, task0.loss_z: 0.0786, task0.loss_whl: 0.0583, task0.loss_yaw: 0.2095, task0.loss_vel: 0.3901, task0.loss_heatmap: 1.1398, task1.loss_xy: 0.1094, task1.loss_z: 0.1060, task1.loss_whl: 0.1126, task1.loss_yaw: 0.2338, task1.loss_vel: 0.2505, task1.loss_heatmap: 1.6235, task2.loss_xy: 0.1143, task2.loss_z: 0.0932, task2.loss_whl: 0.1069, task2.loss_yaw: 0.2799, task2.loss_vel: 0.3472, task2.loss_heatmap: 1.4951, task3.loss_xy: 0.1081, task3.loss_z: 0.0597, task3.loss_whl: 0.1218, task3.loss_yaw: 0.2292, task3.loss_vel: 0.0172, task3.loss_heatmap: 0.9046, task4.loss_xy: 0.1023, task4.loss_z: 0.0672, task4.loss_whl: 0.1166, task4.loss_yaw: 0.2931, task4.loss_vel: 0.3668, task4.loss_heatmap: 1.1403, task5.loss_xy: 0.1137, task5.loss_z: 0.0765, task5.loss_whl: 0.1340, task5.loss_yaw: 0.2849, task5.loss_vel: 0.1822, task5.loss_heatmap: 1.3175, loss: 21.6100, grad_norm: 14.0039 2023-03-13 02:43:40,297 - mmdet - INFO - Iter [3650/10536] lr: 2.000e-04, eta: 4:49:05, time: 2.454, data_time: 0.056, memory: 32270, loss_depth: 9.0028, task0.loss_xy: 0.1048, task0.loss_z: 0.0764, task0.loss_whl: 0.0599, task0.loss_yaw: 0.2127, task0.loss_vel: 0.3687, task0.loss_heatmap: 1.1772, task1.loss_xy: 0.1099, task1.loss_z: 0.0947, task1.loss_whl: 0.1121, task1.loss_yaw: 0.2260, task1.loss_vel: 0.2521, task1.loss_heatmap: 1.5167, task2.loss_xy: 0.1149, task2.loss_z: 0.0917, task2.loss_whl: 0.1137, task2.loss_yaw: 0.2765, task2.loss_vel: 0.3922, task2.loss_heatmap: 1.4379, task3.loss_xy: 0.1111, task3.loss_z: 0.0609, task3.loss_whl: 0.1165, task3.loss_yaw: 0.2829, task3.loss_vel: 0.0189, task3.loss_heatmap: 1.0268, task4.loss_xy: 0.1007, task4.loss_z: 0.0686, task4.loss_whl: 0.1092, task4.loss_yaw: 0.2917, task4.loss_vel: 0.4582, task4.loss_heatmap: 1.1761, task5.loss_xy: 0.1122, task5.loss_z: 0.0733, task5.loss_whl: 0.1364, task5.loss_yaw: 0.2819, task5.loss_vel: 0.1672, task5.loss_heatmap: 1.3032, loss: 21.6366, grad_norm: 12.8562 2023-03-13 02:45:42,048 - mmdet - INFO - Iter [3700/10536] lr: 2.000e-04, eta: 4:46:33, time: 2.435, data_time: 0.058, memory: 32270, loss_depth: 9.0722, task0.loss_xy: 0.1042, task0.loss_z: 0.0864, task0.loss_whl: 0.0575, task0.loss_yaw: 0.2113, task0.loss_vel: 0.3245, task0.loss_heatmap: 1.1598, task1.loss_xy: 0.1100, task1.loss_z: 0.1024, task1.loss_whl: 0.1056, task1.loss_yaw: 0.2364, task1.loss_vel: 0.3079, task1.loss_heatmap: 1.5548, task2.loss_xy: 0.1176, task2.loss_z: 0.1049, task2.loss_whl: 0.1041, task2.loss_yaw: 0.2738, task2.loss_vel: 0.3707, task2.loss_heatmap: 1.4599, task3.loss_xy: 0.1083, task3.loss_z: 0.0556, task3.loss_whl: 0.1253, task3.loss_yaw: 0.2449, task3.loss_vel: 0.0146, task3.loss_heatmap: 1.0019, task4.loss_xy: 0.0995, task4.loss_z: 0.0728, task4.loss_whl: 0.1101, task4.loss_yaw: 0.2843, task4.loss_vel: 0.4577, task4.loss_heatmap: 1.2487, task5.loss_xy: 0.1128, task5.loss_z: 0.0753, task5.loss_whl: 0.1371, task5.loss_yaw: 0.2809, task5.loss_vel: 0.1649, task5.loss_heatmap: 1.2856, loss: 21.7442, grad_norm: 15.5407 2023-03-13 02:47:40,676 - mmdet - INFO - Iter [3750/10536] lr: 2.000e-04, eta: 4:43:44, time: 2.373, data_time: 0.058, memory: 32270, loss_depth: 9.0981, task0.loss_xy: 0.1046, task0.loss_z: 0.0832, task0.loss_whl: 0.0581, task0.loss_yaw: 0.2069, task0.loss_vel: 0.3229, task0.loss_heatmap: 1.1637, task1.loss_xy: 0.1117, task1.loss_z: 0.0941, task1.loss_whl: 0.1033, task1.loss_yaw: 0.2321, task1.loss_vel: 0.2815, task1.loss_heatmap: 1.6617, task2.loss_xy: 0.1152, task2.loss_z: 0.0985, task2.loss_whl: 0.0987, task2.loss_yaw: 0.2720, task2.loss_vel: 0.4516, task2.loss_heatmap: 1.5692, task3.loss_xy: 0.1125, task3.loss_z: 0.0642, task3.loss_whl: 0.1143, task3.loss_yaw: 0.2469, task3.loss_vel: 0.0183, task3.loss_heatmap: 1.1310, task4.loss_xy: 0.1012, task4.loss_z: 0.0747, task4.loss_whl: 0.1090, task4.loss_yaw: 0.2964, task4.loss_vel: 0.4811, task4.loss_heatmap: 1.2670, task5.loss_xy: 0.1127, task5.loss_z: 0.0823, task5.loss_whl: 0.1334, task5.loss_yaw: 0.2826, task5.loss_vel: 0.1767, task5.loss_heatmap: 1.3717, loss: 22.3030, grad_norm: 14.2727 2023-03-13 02:49:42,782 - mmdet - INFO - Iter [3800/10536] lr: 2.000e-04, eta: 4:41:19, time: 2.442, data_time: 0.058, memory: 32270, loss_depth: 8.9030, task0.loss_xy: 0.1049, task0.loss_z: 0.0791, task0.loss_whl: 0.0577, task0.loss_yaw: 0.2051, task0.loss_vel: 0.2688, task0.loss_heatmap: 1.1165, task1.loss_xy: 0.1099, task1.loss_z: 0.1049, task1.loss_whl: 0.1047, task1.loss_yaw: 0.2320, task1.loss_vel: 0.2600, task1.loss_heatmap: 1.6824, task2.loss_xy: 0.1148, task2.loss_z: 0.1055, task2.loss_whl: 0.1164, task2.loss_yaw: 0.2686, task2.loss_vel: 0.4157, task2.loss_heatmap: 1.4922, task3.loss_xy: 0.1097, task3.loss_z: 0.0591, task3.loss_whl: 0.1338, task3.loss_yaw: 0.2546, task3.loss_vel: 0.0191, task3.loss_heatmap: 0.9396, task4.loss_xy: 0.1022, task4.loss_z: 0.0696, task4.loss_whl: 0.0966, task4.loss_yaw: 0.2844, task4.loss_vel: 0.3651, task4.loss_heatmap: 1.1292, task5.loss_xy: 0.1119, task5.loss_z: 0.0770, task5.loss_whl: 0.1370, task5.loss_yaw: 0.2837, task5.loss_vel: 0.1708, task5.loss_heatmap: 1.3180, loss: 21.4033, grad_norm: 11.8109 2023-03-13 02:51:43,689 - mmdet - INFO - Iter [3850/10536] lr: 2.000e-04, eta: 4:38:49, time: 2.418, data_time: 0.064, memory: 32270, loss_depth: 9.0402, task0.loss_xy: 0.1050, task0.loss_z: 0.0856, task0.loss_whl: 0.0590, task0.loss_yaw: 0.2093, task0.loss_vel: 0.3764, task0.loss_heatmap: 1.1711, task1.loss_xy: 0.1101, task1.loss_z: 0.0983, task1.loss_whl: 0.1152, task1.loss_yaw: 0.2305, task1.loss_vel: 0.3432, task1.loss_heatmap: 1.5933, task2.loss_xy: 0.1122, task2.loss_z: 0.0993, task2.loss_whl: 0.1063, task2.loss_yaw: 0.2689, task2.loss_vel: 0.4561, task2.loss_heatmap: 1.4309, task3.loss_xy: 0.1097, task3.loss_z: 0.0540, task3.loss_whl: 0.1398, task3.loss_yaw: 0.2511, task3.loss_vel: 0.0174, task3.loss_heatmap: 0.9165, task4.loss_xy: 0.1018, task4.loss_z: 0.0678, task4.loss_whl: 0.1107, task4.loss_yaw: 0.2783, task4.loss_vel: 0.4437, task4.loss_heatmap: 1.1686, task5.loss_xy: 0.1122, task5.loss_z: 0.0726, task5.loss_whl: 0.1340, task5.loss_yaw: 0.2781, task5.loss_vel: 0.1726, task5.loss_heatmap: 1.3165, loss: 21.7563, grad_norm: 13.0648 2023-03-13 02:53:46,631 - mmdet - INFO - Iter [3900/10536] lr: 2.000e-04, eta: 4:36:33, time: 2.459, data_time: 0.063, memory: 32270, loss_depth: 8.9770, task0.loss_xy: 0.1044, task0.loss_z: 0.0826, task0.loss_whl: 0.0583, task0.loss_yaw: 0.2076, task0.loss_vel: 0.3233, task0.loss_heatmap: 1.1733, task1.loss_xy: 0.1095, task1.loss_z: 0.0989, task1.loss_whl: 0.1125, task1.loss_yaw: 0.2325, task1.loss_vel: 0.2999, task1.loss_heatmap: 1.6064, task2.loss_xy: 0.1130, task2.loss_z: 0.1044, task2.loss_whl: 0.1145, task2.loss_yaw: 0.2653, task2.loss_vel: 0.4832, task2.loss_heatmap: 1.5672, task3.loss_xy: 0.1086, task3.loss_z: 0.0673, task3.loss_whl: 0.1175, task3.loss_yaw: 0.2780, task3.loss_vel: 0.0181, task3.loss_heatmap: 0.9925, task4.loss_xy: 0.1027, task4.loss_z: 0.0759, task4.loss_whl: 0.1158, task4.loss_yaw: 0.2925, task4.loss_vel: 0.3282, task4.loss_heatmap: 1.2525, task5.loss_xy: 0.1117, task5.loss_z: 0.0806, task5.loss_whl: 0.1373, task5.loss_yaw: 0.2821, task5.loss_vel: 0.1660, task5.loss_heatmap: 1.2947, loss: 21.8559, grad_norm: 13.2363 2023-03-13 02:55:47,088 - mmdet - INFO - Iter [3950/10536] lr: 2.000e-04, eta: 4:34:05, time: 2.409, data_time: 0.061, memory: 32270, loss_depth: 8.9289, task0.loss_xy: 0.1044, task0.loss_z: 0.0808, task0.loss_whl: 0.0600, task0.loss_yaw: 0.2031, task0.loss_vel: 0.2960, task0.loss_heatmap: 1.1346, task1.loss_xy: 0.1092, task1.loss_z: 0.1026, task1.loss_whl: 0.1078, task1.loss_yaw: 0.2262, task1.loss_vel: 0.2677, task1.loss_heatmap: 1.5895, task2.loss_xy: 0.1159, task2.loss_z: 0.1062, task2.loss_whl: 0.1064, task2.loss_yaw: 0.2720, task2.loss_vel: 0.3660, task2.loss_heatmap: 1.5870, task3.loss_xy: 0.1105, task3.loss_z: 0.0620, task3.loss_whl: 0.1136, task3.loss_yaw: 0.2809, task3.loss_vel: 0.0183, task3.loss_heatmap: 0.9975, task4.loss_xy: 0.0967, task4.loss_z: 0.0606, task4.loss_whl: 0.0957, task4.loss_yaw: 0.2864, task4.loss_vel: 0.2560, task4.loss_heatmap: 0.9850, task5.loss_xy: 0.1121, task5.loss_z: 0.0754, task5.loss_whl: 0.1320, task5.loss_yaw: 0.2805, task5.loss_vel: 0.1787, task5.loss_heatmap: 1.3350, loss: 21.2411, grad_norm: 12.4626 2023-03-13 02:57:47,894 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 02:57:47,894 - mmdet - INFO - Iter [4000/10536] lr: 2.000e-04, eta: 4:31:41, time: 2.416, data_time: 0.059, memory: 32270, loss_depth: 8.9248, task0.loss_xy: 0.1036, task0.loss_z: 0.0737, task0.loss_whl: 0.0560, task0.loss_yaw: 0.1979, task0.loss_vel: 0.3042, task0.loss_heatmap: 1.0882, task1.loss_xy: 0.1082, task1.loss_z: 0.0887, task1.loss_whl: 0.1056, task1.loss_yaw: 0.2227, task1.loss_vel: 0.2753, task1.loss_heatmap: 1.5177, task2.loss_xy: 0.1102, task2.loss_z: 0.0931, task2.loss_whl: 0.1105, task2.loss_yaw: 0.2714, task2.loss_vel: 0.4502, task2.loss_heatmap: 1.3446, task3.loss_xy: 0.1089, task3.loss_z: 0.0556, task3.loss_whl: 0.1222, task3.loss_yaw: 0.2576, task3.loss_vel: 0.0169, task3.loss_heatmap: 0.9188, task4.loss_xy: 0.0966, task4.loss_z: 0.0622, task4.loss_whl: 0.1018, task4.loss_yaw: 0.2837, task4.loss_vel: 0.2881, task4.loss_heatmap: 0.8972, task5.loss_xy: 0.1104, task5.loss_z: 0.0708, task5.loss_whl: 0.1329, task5.loss_yaw: 0.2827, task5.loss_vel: 0.1574, task5.loss_heatmap: 1.2481, loss: 20.6585, grad_norm: 12.9023 2023-03-13 02:59:49,331 - mmdet - INFO - Iter [4050/10536] lr: 2.000e-04, eta: 4:29:21, time: 2.429, data_time: 0.060, memory: 32270, loss_depth: 8.9773, task0.loss_xy: 0.1040, task0.loss_z: 0.0795, task0.loss_whl: 0.0575, task0.loss_yaw: 0.2020, task0.loss_vel: 0.3184, task0.loss_heatmap: 1.1300, task1.loss_xy: 0.1088, task1.loss_z: 0.0920, task1.loss_whl: 0.1049, task1.loss_yaw: 0.2262, task1.loss_vel: 0.3336, task1.loss_heatmap: 1.5848, task2.loss_xy: 0.1153, task2.loss_z: 0.0915, task2.loss_whl: 0.1225, task2.loss_yaw: 0.2636, task2.loss_vel: 0.5126, task2.loss_heatmap: 1.5123, task3.loss_xy: 0.1092, task3.loss_z: 0.0575, task3.loss_whl: 0.1124, task3.loss_yaw: 0.2504, task3.loss_vel: 0.0181, task3.loss_heatmap: 0.9457, task4.loss_xy: 0.1041, task4.loss_z: 0.0747, task4.loss_whl: 0.0980, task4.loss_yaw: 0.2900, task4.loss_vel: 0.4307, task4.loss_heatmap: 1.1076, task5.loss_xy: 0.1113, task5.loss_z: 0.0734, task5.loss_whl: 0.1320, task5.loss_yaw: 0.2767, task5.loss_vel: 0.1907, task5.loss_heatmap: 1.3079, loss: 21.6271, grad_norm: 15.0364 2023-03-13 03:01:52,215 - mmdet - INFO - Iter [4100/10536] lr: 2.000e-04, eta: 4:27:09, time: 2.458, data_time: 0.060, memory: 32270, loss_depth: 8.9534, task0.loss_xy: 0.1041, task0.loss_z: 0.0750, task0.loss_whl: 0.0595, task0.loss_yaw: 0.1969, task0.loss_vel: 0.2635, task0.loss_heatmap: 1.1013, task1.loss_xy: 0.1077, task1.loss_z: 0.0917, task1.loss_whl: 0.1077, task1.loss_yaw: 0.2256, task1.loss_vel: 0.2199, task1.loss_heatmap: 1.5192, task2.loss_xy: 0.1153, task2.loss_z: 0.0977, task2.loss_whl: 0.1001, task2.loss_yaw: 0.2487, task2.loss_vel: 0.4330, task2.loss_heatmap: 1.4571, task3.loss_xy: 0.1092, task3.loss_z: 0.0543, task3.loss_whl: 0.1135, task3.loss_yaw: 0.2377, task3.loss_vel: 0.0159, task3.loss_heatmap: 0.9133, task4.loss_xy: 0.1003, task4.loss_z: 0.0675, task4.loss_whl: 0.1132, task4.loss_yaw: 0.2949, task4.loss_vel: 0.2002, task4.loss_heatmap: 1.1703, task5.loss_xy: 0.1116, task5.loss_z: 0.0745, task5.loss_whl: 0.1327, task5.loss_yaw: 0.2768, task5.loss_vel: 0.1837, task5.loss_heatmap: 1.3483, loss: 20.9955, grad_norm: 12.8513 2023-03-13 03:03:51,801 - mmdet - INFO - Iter [4150/10536] lr: 2.000e-04, eta: 4:24:44, time: 2.392, data_time: 0.062, memory: 32270, loss_depth: 8.9320, task0.loss_xy: 0.1032, task0.loss_z: 0.0807, task0.loss_whl: 0.0575, task0.loss_yaw: 0.1993, task0.loss_vel: 0.3254, task0.loss_heatmap: 1.1339, task1.loss_xy: 0.1109, task1.loss_z: 0.0980, task1.loss_whl: 0.1013, task1.loss_yaw: 0.2310, task1.loss_vel: 0.2542, task1.loss_heatmap: 1.5831, task2.loss_xy: 0.1126, task2.loss_z: 0.0904, task2.loss_whl: 0.1107, task2.loss_yaw: 0.2565, task2.loss_vel: 0.3379, task2.loss_heatmap: 1.3940, task3.loss_xy: 0.1071, task3.loss_z: 0.0560, task3.loss_whl: 0.1249, task3.loss_yaw: 0.2306, task3.loss_vel: 0.0184, task3.loss_heatmap: 0.8362, task4.loss_xy: 0.0981, task4.loss_z: 0.0625, task4.loss_whl: 0.1074, task4.loss_yaw: 0.2738, task4.loss_vel: 0.4240, task4.loss_heatmap: 1.1361, task5.loss_xy: 0.1129, task5.loss_z: 0.0760, task5.loss_whl: 0.1355, task5.loss_yaw: 0.2766, task5.loss_vel: 0.1678, task5.loss_heatmap: 1.3480, loss: 21.1047, grad_norm: 14.5913 2023-03-13 03:05:52,758 - mmdet - INFO - Iter [4200/10536] lr: 2.000e-04, eta: 4:22:26, time: 2.419, data_time: 0.063, memory: 32270, loss_depth: 8.8231, task0.loss_xy: 0.1040, task0.loss_z: 0.0731, task0.loss_whl: 0.0567, task0.loss_yaw: 0.1933, task0.loss_vel: 0.3467, task0.loss_heatmap: 1.1213, task1.loss_xy: 0.1082, task1.loss_z: 0.0926, task1.loss_whl: 0.1000, task1.loss_yaw: 0.2180, task1.loss_vel: 0.2706, task1.loss_heatmap: 1.4713, task2.loss_xy: 0.1114, task2.loss_z: 0.1058, task2.loss_whl: 0.0977, task2.loss_yaw: 0.2579, task2.loss_vel: 0.4007, task2.loss_heatmap: 1.3875, task3.loss_xy: 0.1097, task3.loss_z: 0.0546, task3.loss_whl: 0.1235, task3.loss_yaw: 0.2507, task3.loss_vel: 0.0192, task3.loss_heatmap: 0.8683, task4.loss_xy: 0.1007, task4.loss_z: 0.0658, task4.loss_whl: 0.1086, task4.loss_yaw: 0.2792, task4.loss_vel: 0.3392, task4.loss_heatmap: 1.0260, task5.loss_xy: 0.1127, task5.loss_z: 0.0741, task5.loss_whl: 0.1306, task5.loss_yaw: 0.2760, task5.loss_vel: 0.1768, task5.loss_heatmap: 1.3046, loss: 20.7601, grad_norm: 11.8200 2023-03-13 03:07:52,928 - mmdet - INFO - Iter [4250/10536] lr: 2.000e-04, eta: 4:20:06, time: 2.403, data_time: 0.068, memory: 32270, loss_depth: 8.9879, task0.loss_xy: 0.1044, task0.loss_z: 0.0778, task0.loss_whl: 0.0582, task0.loss_yaw: 0.1986, task0.loss_vel: 0.3512, task0.loss_heatmap: 1.1557, task1.loss_xy: 0.1109, task1.loss_z: 0.0974, task1.loss_whl: 0.1013, task1.loss_yaw: 0.2260, task1.loss_vel: 0.3559, task1.loss_heatmap: 1.5772, task2.loss_xy: 0.1114, task2.loss_z: 0.1018, task2.loss_whl: 0.1064, task2.loss_yaw: 0.2702, task2.loss_vel: 0.3460, task2.loss_heatmap: 1.3564, task3.loss_xy: 0.1086, task3.loss_z: 0.0541, task3.loss_whl: 0.1362, task3.loss_yaw: 0.2564, task3.loss_vel: 0.0148, task3.loss_heatmap: 0.9251, task4.loss_xy: 0.0981, task4.loss_z: 0.0644, task4.loss_whl: 0.1031, task4.loss_yaw: 0.2913, task4.loss_vel: 0.3310, task4.loss_heatmap: 1.1002, task5.loss_xy: 0.1117, task5.loss_z: 0.0720, task5.loss_whl: 0.1351, task5.loss_yaw: 0.2814, task5.loss_vel: 0.1620, task5.loss_heatmap: 1.2875, loss: 21.2280, grad_norm: 14.2873 2023-03-13 03:09:53,190 - mmdet - INFO - Iter [4300/10536] lr: 2.000e-04, eta: 4:17:47, time: 2.405, data_time: 0.067, memory: 32270, loss_depth: 9.0157, task0.loss_xy: 0.1046, task0.loss_z: 0.0851, task0.loss_whl: 0.0563, task0.loss_yaw: 0.1966, task0.loss_vel: 0.3249, task0.loss_heatmap: 1.1499, task1.loss_xy: 0.1085, task1.loss_z: 0.1065, task1.loss_whl: 0.1140, task1.loss_yaw: 0.2325, task1.loss_vel: 0.2801, task1.loss_heatmap: 1.5773, task2.loss_xy: 0.1145, task2.loss_z: 0.0949, task2.loss_whl: 0.0992, task2.loss_yaw: 0.2728, task2.loss_vel: 0.3231, task2.loss_heatmap: 1.4282, task3.loss_xy: 0.1085, task3.loss_z: 0.0521, task3.loss_whl: 0.1270, task3.loss_yaw: 0.2560, task3.loss_vel: 0.0163, task3.loss_heatmap: 0.9121, task4.loss_xy: 0.1038, task4.loss_z: 0.0645, task4.loss_whl: 0.1081, task4.loss_yaw: 0.2811, task4.loss_vel: 0.3423, task4.loss_heatmap: 1.0927, task5.loss_xy: 0.1109, task5.loss_z: 0.0719, task5.loss_whl: 0.1321, task5.loss_yaw: 0.2802, task5.loss_vel: 0.1645, task5.loss_heatmap: 1.2283, loss: 21.1371, grad_norm: 12.1659 2023-03-13 03:11:53,482 - mmdet - INFO - Iter [4350/10536] lr: 2.000e-04, eta: 4:15:29, time: 2.406, data_time: 0.067, memory: 32270, loss_depth: 8.9010, task0.loss_xy: 0.1049, task0.loss_z: 0.0827, task0.loss_whl: 0.0590, task0.loss_yaw: 0.2003, task0.loss_vel: 0.3174, task0.loss_heatmap: 1.1741, task1.loss_xy: 0.1106, task1.loss_z: 0.0976, task1.loss_whl: 0.1055, task1.loss_yaw: 0.2256, task1.loss_vel: 0.2387, task1.loss_heatmap: 1.5499, task2.loss_xy: 0.1140, task2.loss_z: 0.0975, task2.loss_whl: 0.1022, task2.loss_yaw: 0.2696, task2.loss_vel: 0.3564, task2.loss_heatmap: 1.4370, task3.loss_xy: 0.1108, task3.loss_z: 0.0612, task3.loss_whl: 0.1278, task3.loss_yaw: 0.2869, task3.loss_vel: 0.0220, task3.loss_heatmap: 1.1208, task4.loss_xy: 0.1043, task4.loss_z: 0.0738, task4.loss_whl: 0.1076, task4.loss_yaw: 0.2861, task4.loss_vel: 0.2966, task4.loss_heatmap: 1.1694, task5.loss_xy: 0.1120, task5.loss_z: 0.0736, task5.loss_whl: 0.1314, task5.loss_yaw: 0.2815, task5.loss_vel: 0.1717, task5.loss_heatmap: 1.3028, loss: 21.3845, grad_norm: 11.6332 2023-03-13 03:13:57,592 - mmdet - INFO - Iter [4400/10536] lr: 2.000e-04, eta: 4:13:26, time: 2.482, data_time: 0.062, memory: 32270, loss_depth: 8.8460, task0.loss_xy: 0.1024, task0.loss_z: 0.0787, task0.loss_whl: 0.0589, task0.loss_yaw: 0.1948, task0.loss_vel: 0.2826, task0.loss_heatmap: 1.1086, task1.loss_xy: 0.1079, task1.loss_z: 0.0883, task1.loss_whl: 0.1022, task1.loss_yaw: 0.2156, task1.loss_vel: 0.2082, task1.loss_heatmap: 1.4594, task2.loss_xy: 0.1136, task2.loss_z: 0.0933, task2.loss_whl: 0.1127, task2.loss_yaw: 0.2726, task2.loss_vel: 0.2383, task2.loss_heatmap: 1.4509, task3.loss_xy: 0.1082, task3.loss_z: 0.0606, task3.loss_whl: 0.1090, task3.loss_yaw: 0.2533, task3.loss_vel: 0.0169, task3.loss_heatmap: 0.9325, task4.loss_xy: 0.0998, task4.loss_z: 0.0667, task4.loss_whl: 0.1126, task4.loss_yaw: 0.2859, task4.loss_vel: 0.2393, task4.loss_heatmap: 1.1336, task5.loss_xy: 0.1101, task5.loss_z: 0.0726, task5.loss_whl: 0.1329, task5.loss_yaw: 0.2829, task5.loss_vel: 0.1566, task5.loss_heatmap: 1.2808, loss: 20.5892, grad_norm: 13.1189 2023-03-13 03:16:00,960 - mmdet - INFO - Iter [4450/10536] lr: 2.000e-04, eta: 4:11:20, time: 2.467, data_time: 0.064, memory: 32270, loss_depth: 8.8869, task0.loss_xy: 0.1036, task0.loss_z: 0.0726, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1940, task0.loss_vel: 0.2594, task0.loss_heatmap: 1.1139, task1.loss_xy: 0.1095, task1.loss_z: 0.0941, task1.loss_whl: 0.1015, task1.loss_yaw: 0.2218, task1.loss_vel: 0.2077, task1.loss_heatmap: 1.4897, task2.loss_xy: 0.1128, task2.loss_z: 0.0971, task2.loss_whl: 0.0995, task2.loss_yaw: 0.2610, task2.loss_vel: 0.3601, task2.loss_heatmap: 1.3539, task3.loss_xy: 0.1092, task3.loss_z: 0.0555, task3.loss_whl: 0.1167, task3.loss_yaw: 0.2666, task3.loss_vel: 0.0193, task3.loss_heatmap: 0.8909, task4.loss_xy: 0.0953, task4.loss_z: 0.0566, task4.loss_whl: 0.1077, task4.loss_yaw: 0.2826, task4.loss_vel: 0.2280, task4.loss_heatmap: 0.9437, task5.loss_xy: 0.1127, task5.loss_z: 0.0742, task5.loss_whl: 0.1229, task5.loss_yaw: 0.2752, task5.loss_vel: 0.1874, task5.loss_heatmap: 1.3029, loss: 20.4433, grad_norm: 11.7034 2023-03-13 03:17:59,925 - mmdet - INFO - Iter [4500/10536] lr: 2.000e-04, eta: 4:09:01, time: 2.379, data_time: 0.062, memory: 32270, loss_depth: 8.9394, task0.loss_xy: 0.1043, task0.loss_z: 0.0778, task0.loss_whl: 0.0585, task0.loss_yaw: 0.1970, task0.loss_vel: 0.2893, task0.loss_heatmap: 1.1653, task1.loss_xy: 0.1087, task1.loss_z: 0.0940, task1.loss_whl: 0.1085, task1.loss_yaw: 0.2123, task1.loss_vel: 0.2599, task1.loss_heatmap: 1.4999, task2.loss_xy: 0.1116, task2.loss_z: 0.0884, task2.loss_whl: 0.0956, task2.loss_yaw: 0.2471, task2.loss_vel: 0.3875, task2.loss_heatmap: 1.3434, task3.loss_xy: 0.1088, task3.loss_z: 0.0563, task3.loss_whl: 0.1213, task3.loss_yaw: 0.2342, task3.loss_vel: 0.0214, task3.loss_heatmap: 0.9920, task4.loss_xy: 0.0989, task4.loss_z: 0.0659, task4.loss_whl: 0.1174, task4.loss_yaw: 0.2721, task4.loss_vel: 0.4054, task4.loss_heatmap: 1.0076, task5.loss_xy: 0.1125, task5.loss_z: 0.0771, task5.loss_whl: 0.1203, task5.loss_yaw: 0.2780, task5.loss_vel: 0.1883, task5.loss_heatmap: 1.3157, loss: 20.9816, grad_norm: 14.8058 2023-03-13 03:20:03,095 - mmdet - INFO - Iter [4550/10536] lr: 2.000e-04, eta: 4:06:55, time: 2.464, data_time: 0.058, memory: 32270, loss_depth: 8.7683, task0.loss_xy: 0.1040, task0.loss_z: 0.0745, task0.loss_whl: 0.0584, task0.loss_yaw: 0.1907, task0.loss_vel: 0.2696, task0.loss_heatmap: 1.1078, task1.loss_xy: 0.1078, task1.loss_z: 0.0994, task1.loss_whl: 0.1038, task1.loss_yaw: 0.2108, task1.loss_vel: 0.2352, task1.loss_heatmap: 1.4703, task2.loss_xy: 0.1135, task2.loss_z: 0.0966, task2.loss_whl: 0.1105, task2.loss_yaw: 0.2666, task2.loss_vel: 0.2900, task2.loss_heatmap: 1.3827, task3.loss_xy: 0.1045, task3.loss_z: 0.0548, task3.loss_whl: 0.1397, task3.loss_yaw: 0.2604, task3.loss_vel: 0.0179, task3.loss_heatmap: 1.0077, task4.loss_xy: 0.0987, task4.loss_z: 0.0547, task4.loss_whl: 0.0938, task4.loss_yaw: 0.2819, task4.loss_vel: 0.2179, task4.loss_heatmap: 0.9015, task5.loss_xy: 0.1113, task5.loss_z: 0.0713, task5.loss_whl: 0.1340, task5.loss_yaw: 0.2761, task5.loss_vel: 0.1638, task5.loss_heatmap: 1.2723, loss: 20.3227, grad_norm: 12.5322 2023-03-13 03:22:05,137 - mmdet - INFO - Iter [4600/10536] lr: 2.000e-04, eta: 4:04:46, time: 2.441, data_time: 0.058, memory: 32270, loss_depth: 8.7256, task0.loss_xy: 0.1003, task0.loss_z: 0.0737, task0.loss_whl: 0.0575, task0.loss_yaw: 0.1826, task0.loss_vel: 0.3545, task0.loss_heatmap: 1.0804, task1.loss_xy: 0.1050, task1.loss_z: 0.0869, task1.loss_whl: 0.1012, task1.loss_yaw: 0.2039, task1.loss_vel: 0.2342, task1.loss_heatmap: 1.3909, task2.loss_xy: 0.1100, task2.loss_z: 0.0998, task2.loss_whl: 0.1100, task2.loss_yaw: 0.2580, task2.loss_vel: 0.3919, task2.loss_heatmap: 1.3112, task3.loss_xy: 0.1081, task3.loss_z: 0.0553, task3.loss_whl: 0.1200, task3.loss_yaw: 0.2122, task3.loss_vel: 0.0136, task3.loss_heatmap: 0.8706, task4.loss_xy: 0.0970, task4.loss_z: 0.0653, task4.loss_whl: 0.1026, task4.loss_yaw: 0.2705, task4.loss_vel: 0.3616, task4.loss_heatmap: 1.0020, task5.loss_xy: 0.1110, task5.loss_z: 0.0681, task5.loss_whl: 0.1402, task5.loss_yaw: 0.2757, task5.loss_vel: 0.1602, task5.loss_heatmap: 1.2412, loss: 20.2529, grad_norm: 12.4800 2023-03-13 03:24:06,502 - mmdet - INFO - Iter [4650/10536] lr: 2.000e-04, eta: 4:02:36, time: 2.427, data_time: 0.067, memory: 32270, loss_depth: 8.9246, task0.loss_xy: 0.1027, task0.loss_z: 0.0755, task0.loss_whl: 0.0572, task0.loss_yaw: 0.1867, task0.loss_vel: 0.3345, task0.loss_heatmap: 1.1320, task1.loss_xy: 0.1089, task1.loss_z: 0.0998, task1.loss_whl: 0.1065, task1.loss_yaw: 0.2223, task1.loss_vel: 0.2226, task1.loss_heatmap: 1.4906, task2.loss_xy: 0.1133, task2.loss_z: 0.0950, task2.loss_whl: 0.1044, task2.loss_yaw: 0.2558, task2.loss_vel: 0.3274, task2.loss_heatmap: 1.3928, task3.loss_xy: 0.1071, task3.loss_z: 0.0546, task3.loss_whl: 0.1212, task3.loss_yaw: 0.2484, task3.loss_vel: 0.0180, task3.loss_heatmap: 0.8643, task4.loss_xy: 0.1015, task4.loss_z: 0.0669, task4.loss_whl: 0.1035, task4.loss_yaw: 0.2862, task4.loss_vel: 0.2759, task4.loss_heatmap: 1.1487, task5.loss_xy: 0.1125, task5.loss_z: 0.0740, task5.loss_whl: 0.1300, task5.loss_yaw: 0.2762, task5.loss_vel: 0.1706, task5.loss_heatmap: 1.2827, loss: 20.7949, grad_norm: 13.5013 2023-03-13 03:26:05,053 - mmdet - INFO - Iter [4700/10536] lr: 2.000e-04, eta: 4:00:18, time: 2.371, data_time: 0.062, memory: 32270, loss_depth: 8.9087, task0.loss_xy: 0.1025, task0.loss_z: 0.0773, task0.loss_whl: 0.0588, task0.loss_yaw: 0.1835, task0.loss_vel: 0.3043, task0.loss_heatmap: 1.0867, task1.loss_xy: 0.1096, task1.loss_z: 0.0944, task1.loss_whl: 0.1085, task1.loss_yaw: 0.2194, task1.loss_vel: 0.2491, task1.loss_heatmap: 1.4860, task2.loss_xy: 0.1115, task2.loss_z: 0.0911, task2.loss_whl: 0.1001, task2.loss_yaw: 0.2624, task2.loss_vel: 0.3425, task2.loss_heatmap: 1.2830, task3.loss_xy: 0.1105, task3.loss_z: 0.0536, task3.loss_whl: 0.1263, task3.loss_yaw: 0.2835, task3.loss_vel: 0.0201, task3.loss_heatmap: 1.0493, task4.loss_xy: 0.0991, task4.loss_z: 0.0647, task4.loss_whl: 0.1077, task4.loss_yaw: 0.2846, task4.loss_vel: 0.3077, task4.loss_heatmap: 1.0738, task5.loss_xy: 0.1101, task5.loss_z: 0.0694, task5.loss_whl: 0.1307, task5.loss_yaw: 0.2768, task5.loss_vel: 0.1681, task5.loss_heatmap: 1.2696, loss: 20.7848, grad_norm: 13.7850 2023-03-13 03:28:08,339 - mmdet - INFO - Iter [4750/10536] lr: 2.000e-04, eta: 3:58:13, time: 2.466, data_time: 0.057, memory: 32270, loss_depth: 8.8969, task0.loss_xy: 0.1047, task0.loss_z: 0.0779, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1879, task0.loss_vel: 0.2698, task0.loss_heatmap: 1.1245, task1.loss_xy: 0.1101, task1.loss_z: 0.0907, task1.loss_whl: 0.1051, task1.loss_yaw: 0.2204, task1.loss_vel: 0.2596, task1.loss_heatmap: 1.5817, task2.loss_xy: 0.1146, task2.loss_z: 0.0895, task2.loss_whl: 0.1122, task2.loss_yaw: 0.2594, task2.loss_vel: 0.3549, task2.loss_heatmap: 1.5163, task3.loss_xy: 0.1076, task3.loss_z: 0.0535, task3.loss_whl: 0.1169, task3.loss_yaw: 0.2465, task3.loss_vel: 0.0180, task3.loss_heatmap: 0.9702, task4.loss_xy: 0.0956, task4.loss_z: 0.0575, task4.loss_whl: 0.1090, task4.loss_yaw: 0.2868, task4.loss_vel: 0.2895, task4.loss_heatmap: 0.9935, task5.loss_xy: 0.1104, task5.loss_z: 0.0675, task5.loss_whl: 0.1346, task5.loss_yaw: 0.2756, task5.loss_vel: 0.1557, task5.loss_heatmap: 1.2387, loss: 20.8601, grad_norm: 14.7060 2023-03-13 03:30:13,252 - mmdet - INFO - Iter [4800/10536] lr: 2.000e-04, eta: 3:56:13, time: 2.498, data_time: 0.059, memory: 32270, loss_depth: 8.8139, task0.loss_xy: 0.1025, task0.loss_z: 0.0776, task0.loss_whl: 0.0584, task0.loss_yaw: 0.1828, task0.loss_vel: 0.3023, task0.loss_heatmap: 1.0816, task1.loss_xy: 0.1073, task1.loss_z: 0.0907, task1.loss_whl: 0.1037, task1.loss_yaw: 0.2154, task1.loss_vel: 0.2919, task1.loss_heatmap: 1.4693, task2.loss_xy: 0.1131, task2.loss_z: 0.0880, task2.loss_whl: 0.0934, task2.loss_yaw: 0.2553, task2.loss_vel: 0.4227, task2.loss_heatmap: 1.3641, task3.loss_xy: 0.1085, task3.loss_z: 0.0531, task3.loss_whl: 0.1102, task3.loss_yaw: 0.2551, task3.loss_vel: 0.0158, task3.loss_heatmap: 0.8326, task4.loss_xy: 0.0999, task4.loss_z: 0.0699, task4.loss_whl: 0.0975, task4.loss_yaw: 0.2738, task4.loss_vel: 0.3159, task4.loss_heatmap: 1.1167, task5.loss_xy: 0.1104, task5.loss_z: 0.0699, task5.loss_whl: 0.1369, task5.loss_yaw: 0.2764, task5.loss_vel: 0.1521, task5.loss_heatmap: 1.2198, loss: 20.5485, grad_norm: 12.7333 2023-03-13 03:32:13,312 - mmdet - INFO - Iter [4850/10536] lr: 2.000e-04, eta: 3:54:01, time: 2.401, data_time: 0.060, memory: 32270, loss_depth: 8.8464, task0.loss_xy: 0.1030, task0.loss_z: 0.0742, task0.loss_whl: 0.0578, task0.loss_yaw: 0.1899, task0.loss_vel: 0.3368, task0.loss_heatmap: 1.1566, task1.loss_xy: 0.1089, task1.loss_z: 0.0897, task1.loss_whl: 0.1011, task1.loss_yaw: 0.2201, task1.loss_vel: 0.2540, task1.loss_heatmap: 1.5800, task2.loss_xy: 0.1101, task2.loss_z: 0.0902, task2.loss_whl: 0.1036, task2.loss_yaw: 0.2575, task2.loss_vel: 0.3122, task2.loss_heatmap: 1.3595, task3.loss_xy: 0.1067, task3.loss_z: 0.0563, task3.loss_whl: 0.1103, task3.loss_yaw: 0.2604, task3.loss_vel: 0.0212, task3.loss_heatmap: 0.9882, task4.loss_xy: 0.0992, task4.loss_z: 0.0612, task4.loss_whl: 0.0957, task4.loss_yaw: 0.2780, task4.loss_vel: 0.3114, task4.loss_heatmap: 1.0622, task5.loss_xy: 0.1123, task5.loss_z: 0.0697, task5.loss_whl: 0.1279, task5.loss_yaw: 0.2819, task5.loss_vel: 0.1770, task5.loss_heatmap: 1.3026, loss: 20.8736, grad_norm: 14.8002 2023-03-13 03:34:13,537 - mmdet - INFO - Iter [4900/10536] lr: 2.000e-04, eta: 3:51:49, time: 2.404, data_time: 0.060, memory: 32270, loss_depth: 8.7587, task0.loss_xy: 0.1034, task0.loss_z: 0.0787, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1878, task0.loss_vel: 0.3258, task0.loss_heatmap: 1.1451, task1.loss_xy: 0.1080, task1.loss_z: 0.0959, task1.loss_whl: 0.1080, task1.loss_yaw: 0.2229, task1.loss_vel: 0.2518, task1.loss_heatmap: 1.5431, task2.loss_xy: 0.1127, task2.loss_z: 0.0974, task2.loss_whl: 0.1012, task2.loss_yaw: 0.2417, task2.loss_vel: 0.3885, task2.loss_heatmap: 1.2465, task3.loss_xy: 0.1069, task3.loss_z: 0.0545, task3.loss_whl: 0.1214, task3.loss_yaw: 0.2591, task3.loss_vel: 0.0138, task3.loss_heatmap: 0.8594, task4.loss_xy: 0.1029, task4.loss_z: 0.0694, task4.loss_whl: 0.1010, task4.loss_yaw: 0.2814, task4.loss_vel: 0.2602, task4.loss_heatmap: 0.9983, task5.loss_xy: 0.1108, task5.loss_z: 0.0693, task5.loss_whl: 0.1323, task5.loss_yaw: 0.2780, task5.loss_vel: 0.1638, task5.loss_heatmap: 1.2326, loss: 20.3882, grad_norm: 12.3287 2023-03-13 03:36:16,689 - mmdet - INFO - Iter [4950/10536] lr: 2.000e-04, eta: 3:49:45, time: 2.463, data_time: 0.059, memory: 32270, loss_depth: 8.6191, task0.loss_xy: 0.1024, task0.loss_z: 0.0724, task0.loss_whl: 0.0577, task0.loss_yaw: 0.1812, task0.loss_vel: 0.2599, task0.loss_heatmap: 1.0749, task1.loss_xy: 0.1078, task1.loss_z: 0.0862, task1.loss_whl: 0.1077, task1.loss_yaw: 0.2099, task1.loss_vel: 0.2752, task1.loss_heatmap: 1.4805, task2.loss_xy: 0.1092, task2.loss_z: 0.0894, task2.loss_whl: 0.0976, task2.loss_yaw: 0.2448, task2.loss_vel: 0.3488, task2.loss_heatmap: 1.2531, task3.loss_xy: 0.1077, task3.loss_z: 0.0543, task3.loss_whl: 0.1177, task3.loss_yaw: 0.2956, task3.loss_vel: 0.0202, task3.loss_heatmap: 0.8816, task4.loss_xy: 0.0931, task4.loss_z: 0.0620, task4.loss_whl: 0.0979, task4.loss_yaw: 0.2804, task4.loss_vel: 0.2303, task4.loss_heatmap: 0.9333, task5.loss_xy: 0.1110, task5.loss_z: 0.0685, task5.loss_whl: 0.1266, task5.loss_yaw: 0.2734, task5.loss_vel: 0.1694, task5.loss_heatmap: 1.2515, loss: 19.9523, grad_norm: 12.4106 2023-03-13 03:38:17,158 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 03:38:17,158 - mmdet - INFO - Iter [5000/10536] lr: 2.000e-04, eta: 3:47:35, time: 2.410, data_time: 0.060, memory: 32270, loss_depth: 8.7906, task0.loss_xy: 0.1015, task0.loss_z: 0.0741, task0.loss_whl: 0.0571, task0.loss_yaw: 0.1809, task0.loss_vel: 0.2902, task0.loss_heatmap: 1.0959, task1.loss_xy: 0.1063, task1.loss_z: 0.0908, task1.loss_whl: 0.1037, task1.loss_yaw: 0.2105, task1.loss_vel: 0.2329, task1.loss_heatmap: 1.3846, task2.loss_xy: 0.1115, task2.loss_z: 0.0985, task2.loss_whl: 0.1166, task2.loss_yaw: 0.2557, task2.loss_vel: 0.3137, task2.loss_heatmap: 1.3389, task3.loss_xy: 0.1063, task3.loss_z: 0.0517, task3.loss_whl: 0.1106, task3.loss_yaw: 0.2660, task3.loss_vel: 0.0179, task3.loss_heatmap: 1.0178, task4.loss_xy: 0.0973, task4.loss_z: 0.0615, task4.loss_whl: 0.1118, task4.loss_yaw: 0.2722, task4.loss_vel: 0.2820, task4.loss_heatmap: 0.9946, task5.loss_xy: 0.1104, task5.loss_z: 0.0686, task5.loss_whl: 0.1320, task5.loss_yaw: 0.2759, task5.loss_vel: 0.1667, task5.loss_heatmap: 1.2257, loss: 20.3231, grad_norm: 13.0126 2023-03-13 03:40:18,270 - mmdet - INFO - Iter [5050/10536] lr: 2.000e-04, eta: 3:45:27, time: 2.422, data_time: 0.066, memory: 32270, loss_depth: 8.8657, task0.loss_xy: 0.1033, task0.loss_z: 0.0751, task0.loss_whl: 0.0570, task0.loss_yaw: 0.1830, task0.loss_vel: 0.2922, task0.loss_heatmap: 1.1002, task1.loss_xy: 0.1065, task1.loss_z: 0.0896, task1.loss_whl: 0.0976, task1.loss_yaw: 0.2032, task1.loss_vel: 0.2165, task1.loss_heatmap: 1.3707, task2.loss_xy: 0.1107, task2.loss_z: 0.0843, task2.loss_whl: 0.0864, task2.loss_yaw: 0.2490, task2.loss_vel: 0.2922, task2.loss_heatmap: 1.2871, task3.loss_xy: 0.1081, task3.loss_z: 0.0544, task3.loss_whl: 0.1089, task3.loss_yaw: 0.2407, task3.loss_vel: 0.0179, task3.loss_heatmap: 0.9070, task4.loss_xy: 0.0975, task4.loss_z: 0.0622, task4.loss_whl: 0.0998, task4.loss_yaw: 0.2723, task4.loss_vel: 0.3072, task4.loss_heatmap: 1.0218, task5.loss_xy: 0.1108, task5.loss_z: 0.0750, task5.loss_whl: 0.1333, task5.loss_yaw: 0.2752, task5.loss_vel: 0.1593, task5.loss_heatmap: 1.3229, loss: 20.2443, grad_norm: 11.9775 2023-03-13 03:42:19,395 - mmdet - INFO - Iter [5100/10536] lr: 2.000e-04, eta: 3:43:19, time: 2.422, data_time: 0.059, memory: 32270, loss_depth: 8.8564, task0.loss_xy: 0.1015, task0.loss_z: 0.0738, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1800, task0.loss_vel: 0.2610, task0.loss_heatmap: 1.0619, task1.loss_xy: 0.1076, task1.loss_z: 0.0855, task1.loss_whl: 0.1027, task1.loss_yaw: 0.2151, task1.loss_vel: 0.2508, task1.loss_heatmap: 1.4140, task2.loss_xy: 0.1122, task2.loss_z: 0.0897, task2.loss_whl: 0.1074, task2.loss_yaw: 0.2553, task2.loss_vel: 0.3607, task2.loss_heatmap: 1.3601, task3.loss_xy: 0.1034, task3.loss_z: 0.0515, task3.loss_whl: 0.1054, task3.loss_yaw: 0.2450, task3.loss_vel: 0.0199, task3.loss_heatmap: 0.7965, task4.loss_xy: 0.1002, task4.loss_z: 0.0624, task4.loss_whl: 0.0999, task4.loss_yaw: 0.2848, task4.loss_vel: 0.2957, task4.loss_heatmap: 0.9509, task5.loss_xy: 0.1097, task5.loss_z: 0.0702, task5.loss_whl: 0.1324, task5.loss_yaw: 0.2725, task5.loss_vel: 0.1566, task5.loss_heatmap: 1.2716, loss: 20.1807, grad_norm: 11.4891 2023-03-13 03:44:19,937 - mmdet - INFO - Iter [5150/10536] lr: 2.000e-04, eta: 3:41:10, time: 2.411, data_time: 0.058, memory: 32270, loss_depth: 8.8927, task0.loss_xy: 0.1023, task0.loss_z: 0.0742, task0.loss_whl: 0.0568, task0.loss_yaw: 0.1830, task0.loss_vel: 0.3194, task0.loss_heatmap: 1.0902, task1.loss_xy: 0.1070, task1.loss_z: 0.0980, task1.loss_whl: 0.1020, task1.loss_yaw: 0.2086, task1.loss_vel: 0.2664, task1.loss_heatmap: 1.5165, task2.loss_xy: 0.1067, task2.loss_z: 0.0827, task2.loss_whl: 0.1002, task2.loss_yaw: 0.2350, task2.loss_vel: 0.2722, task2.loss_heatmap: 1.1336, task3.loss_xy: 0.1053, task3.loss_z: 0.0467, task3.loss_whl: 0.1037, task3.loss_yaw: 0.2079, task3.loss_vel: 0.0167, task3.loss_heatmap: 0.8646, task4.loss_xy: 0.0951, task4.loss_z: 0.0602, task4.loss_whl: 0.1059, task4.loss_yaw: 0.2605, task4.loss_vel: 0.2647, task4.loss_heatmap: 0.9799, task5.loss_xy: 0.1113, task5.loss_z: 0.0704, task5.loss_whl: 0.1318, task5.loss_yaw: 0.2764, task5.loss_vel: 0.1626, task5.loss_heatmap: 1.2712, loss: 20.0826, grad_norm: 14.4067 2023-03-13 03:46:20,697 - mmdet - INFO - Iter [5200/10536] lr: 2.000e-04, eta: 3:39:01, time: 2.415, data_time: 0.062, memory: 32270, loss_depth: 8.7495, task0.loss_xy: 0.1021, task0.loss_z: 0.0757, task0.loss_whl: 0.0587, task0.loss_yaw: 0.1794, task0.loss_vel: 0.3156, task0.loss_heatmap: 1.0940, task1.loss_xy: 0.1085, task1.loss_z: 0.0966, task1.loss_whl: 0.1069, task1.loss_yaw: 0.2118, task1.loss_vel: 0.2262, task1.loss_heatmap: 1.5206, task2.loss_xy: 0.1142, task2.loss_z: 0.0880, task2.loss_whl: 0.1011, task2.loss_yaw: 0.2714, task2.loss_vel: 0.3384, task2.loss_heatmap: 1.4136, task3.loss_xy: 0.1085, task3.loss_z: 0.0586, task3.loss_whl: 0.1104, task3.loss_yaw: 0.2606, task3.loss_vel: 0.0251, task3.loss_heatmap: 0.8963, task4.loss_xy: 0.0986, task4.loss_z: 0.0655, task4.loss_whl: 0.1011, task4.loss_yaw: 0.2843, task4.loss_vel: 0.2235, task4.loss_heatmap: 0.9521, task5.loss_xy: 0.1114, task5.loss_z: 0.0711, task5.loss_whl: 0.1307, task5.loss_yaw: 0.2764, task5.loss_vel: 0.1713, task5.loss_heatmap: 1.2569, loss: 20.3748, grad_norm: 12.8955 2023-03-13 03:48:25,337 - mmdet - INFO - Iter [5250/10536] lr: 2.000e-04, eta: 3:37:01, time: 2.493, data_time: 0.059, memory: 32270, loss_depth: 8.8038, task0.loss_xy: 0.1034, task0.loss_z: 0.0743, task0.loss_whl: 0.0583, task0.loss_yaw: 0.1805, task0.loss_vel: 0.2555, task0.loss_heatmap: 1.0913, task1.loss_xy: 0.1062, task1.loss_z: 0.0915, task1.loss_whl: 0.0913, task1.loss_yaw: 0.1981, task1.loss_vel: 0.2279, task1.loss_heatmap: 1.4033, task2.loss_xy: 0.1117, task2.loss_z: 0.0902, task2.loss_whl: 0.1045, task2.loss_yaw: 0.2519, task2.loss_vel: 0.3708, task2.loss_heatmap: 1.2804, task3.loss_xy: 0.1014, task3.loss_z: 0.0521, task3.loss_whl: 0.1257, task3.loss_yaw: 0.2618, task3.loss_vel: 0.0131, task3.loss_heatmap: 1.0174, task4.loss_xy: 0.1006, task4.loss_z: 0.0546, task4.loss_whl: 0.1069, task4.loss_yaw: 0.2698, task4.loss_vel: 0.2835, task4.loss_heatmap: 1.0671, task5.loss_xy: 0.1109, task5.loss_z: 0.0702, task5.loss_whl: 0.1339, task5.loss_yaw: 0.2707, task5.loss_vel: 0.1697, task5.loss_heatmap: 1.2869, loss: 20.3912, grad_norm: 12.6131 2023-03-13 03:49:09,617 - mmdet - INFO - Saving checkpoint at 5268 iterations 2023-03-13 03:50:28,641 - mmdet - INFO - Iter [5300/10536] lr: 2.000e-04, eta: 3:34:58, time: 2.466, data_time: 0.057, memory: 32270, loss_depth: 8.7584, task0.loss_xy: 0.1029, task0.loss_z: 0.0754, task0.loss_whl: 0.0583, task0.loss_yaw: 0.1829, task0.loss_vel: 0.2324, task0.loss_heatmap: 1.1184, task1.loss_xy: 0.1049, task1.loss_z: 0.0847, task1.loss_whl: 0.0954, task1.loss_yaw: 0.1986, task1.loss_vel: 0.2181, task1.loss_heatmap: 1.4434, task2.loss_xy: 0.1144, task2.loss_z: 0.0961, task2.loss_whl: 0.1109, task2.loss_yaw: 0.2544, task2.loss_vel: 0.2957, task2.loss_heatmap: 1.3903, task3.loss_xy: 0.1066, task3.loss_z: 0.0516, task3.loss_whl: 0.1164, task3.loss_yaw: 0.2345, task3.loss_vel: 0.0185, task3.loss_heatmap: 0.9618, task4.loss_xy: 0.0953, task4.loss_z: 0.0579, task4.loss_whl: 0.1060, task4.loss_yaw: 0.2724, task4.loss_vel: 0.2501, task4.loss_heatmap: 0.9789, task5.loss_xy: 0.1120, task5.loss_z: 0.0700, task5.loss_whl: 0.1255, task5.loss_yaw: 0.2741, task5.loss_vel: 0.1721, task5.loss_heatmap: 1.2571, loss: 20.1965, grad_norm: 12.5100 2023-03-13 03:52:29,380 - mmdet - INFO - Iter [5350/10536] lr: 2.000e-04, eta: 3:32:51, time: 2.415, data_time: 0.056, memory: 32270, loss_depth: 8.7744, task0.loss_xy: 0.1027, task0.loss_z: 0.0754, task0.loss_whl: 0.0581, task0.loss_yaw: 0.1793, task0.loss_vel: 0.2848, task0.loss_heatmap: 1.1124, task1.loss_xy: 0.1071, task1.loss_z: 0.0868, task1.loss_whl: 0.1000, task1.loss_yaw: 0.2016, task1.loss_vel: 0.2196, task1.loss_heatmap: 1.4114, task2.loss_xy: 0.1121, task2.loss_z: 0.0875, task2.loss_whl: 0.1032, task2.loss_yaw: 0.2488, task2.loss_vel: 0.3816, task2.loss_heatmap: 1.2796, task3.loss_xy: 0.1085, task3.loss_z: 0.0499, task3.loss_whl: 0.1176, task3.loss_yaw: 0.2655, task3.loss_vel: 0.0173, task3.loss_heatmap: 1.0806, task4.loss_xy: 0.0949, task4.loss_z: 0.0577, task4.loss_whl: 0.1086, task4.loss_yaw: 0.2731, task4.loss_vel: 0.3401, task4.loss_heatmap: 1.0668, task5.loss_xy: 0.1095, task5.loss_z: 0.0656, task5.loss_whl: 0.1324, task5.loss_yaw: 0.2738, task5.loss_vel: 0.1642, task5.loss_heatmap: 1.2319, loss: 20.4844, grad_norm: 13.6018 2023-03-13 03:54:32,981 - mmdet - INFO - Iter [5400/10536] lr: 2.000e-04, eta: 3:30:48, time: 2.472, data_time: 0.091, memory: 32270, loss_depth: 8.7928, task0.loss_xy: 0.1023, task0.loss_z: 0.0737, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1778, task0.loss_vel: 0.3220, task0.loss_heatmap: 1.0829, task1.loss_xy: 0.1074, task1.loss_z: 0.0930, task1.loss_whl: 0.0975, task1.loss_yaw: 0.2077, task1.loss_vel: 0.2920, task1.loss_heatmap: 1.4635, task2.loss_xy: 0.1117, task2.loss_z: 0.1000, task2.loss_whl: 0.1035, task2.loss_yaw: 0.2402, task2.loss_vel: 0.3668, task2.loss_heatmap: 1.4030, task3.loss_xy: 0.1040, task3.loss_z: 0.0550, task3.loss_whl: 0.1138, task3.loss_yaw: 0.2640, task3.loss_vel: 0.0187, task3.loss_heatmap: 0.8584, task4.loss_xy: 0.0990, task4.loss_z: 0.0712, task4.loss_whl: 0.0948, task4.loss_yaw: 0.2700, task4.loss_vel: 0.3234, task4.loss_heatmap: 1.0210, task5.loss_xy: 0.1088, task5.loss_z: 0.0733, task5.loss_whl: 0.1356, task5.loss_yaw: 0.2762, task5.loss_vel: 0.1608, task5.loss_heatmap: 1.2487, loss: 20.4909, grad_norm: 11.8819 2023-03-13 03:56:35,282 - mmdet - INFO - Iter [5450/10536] lr: 2.000e-04, eta: 3:28:44, time: 2.446, data_time: 0.064, memory: 32270, loss_depth: 8.7383, task0.loss_xy: 0.1030, task0.loss_z: 0.0776, task0.loss_whl: 0.0571, task0.loss_yaw: 0.1823, task0.loss_vel: 0.2645, task0.loss_heatmap: 1.1064, task1.loss_xy: 0.1088, task1.loss_z: 0.0918, task1.loss_whl: 0.1000, task1.loss_yaw: 0.2052, task1.loss_vel: 0.2142, task1.loss_heatmap: 1.5073, task2.loss_xy: 0.1107, task2.loss_z: 0.0761, task2.loss_whl: 0.0915, task2.loss_yaw: 0.2420, task2.loss_vel: 0.3815, task2.loss_heatmap: 1.2919, task3.loss_xy: 0.1053, task3.loss_z: 0.0527, task3.loss_whl: 0.1091, task3.loss_yaw: 0.2547, task3.loss_vel: 0.0162, task3.loss_heatmap: 0.9836, task4.loss_xy: 0.1004, task4.loss_z: 0.0640, task4.loss_whl: 0.0968, task4.loss_yaw: 0.2721, task4.loss_vel: 0.2856, task4.loss_heatmap: 1.0873, task5.loss_xy: 0.1111, task5.loss_z: 0.0676, task5.loss_whl: 0.1321, task5.loss_yaw: 0.2674, task5.loss_vel: 0.1730, task5.loss_heatmap: 1.2504, loss: 20.3797, grad_norm: 12.4970 2023-03-13 03:58:37,242 - mmdet - INFO - Iter [5500/10536] lr: 2.000e-04, eta: 3:26:39, time: 2.439, data_time: 0.057, memory: 32270, loss_depth: 8.6735, task0.loss_xy: 0.1021, task0.loss_z: 0.0698, task0.loss_whl: 0.0571, task0.loss_yaw: 0.1732, task0.loss_vel: 0.2600, task0.loss_heatmap: 1.0600, task1.loss_xy: 0.1097, task1.loss_z: 0.0877, task1.loss_whl: 0.1057, task1.loss_yaw: 0.2146, task1.loss_vel: 0.2057, task1.loss_heatmap: 1.5430, task2.loss_xy: 0.1084, task2.loss_z: 0.0799, task2.loss_whl: 0.0990, task2.loss_yaw: 0.2486, task2.loss_vel: 0.3203, task2.loss_heatmap: 1.2715, task3.loss_xy: 0.1059, task3.loss_z: 0.0512, task3.loss_whl: 0.1124, task3.loss_yaw: 0.2421, task3.loss_vel: 0.0168, task3.loss_heatmap: 0.8951, task4.loss_xy: 0.0987, task4.loss_z: 0.0588, task4.loss_whl: 0.1032, task4.loss_yaw: 0.2696, task4.loss_vel: 0.2216, task4.loss_heatmap: 0.9661, task5.loss_xy: 0.1118, task5.loss_z: 0.0698, task5.loss_whl: 0.1307, task5.loss_yaw: 0.2759, task5.loss_vel: 0.1501, task5.loss_heatmap: 1.2521, loss: 19.9218, grad_norm: 11.0078 2023-03-13 04:00:40,838 - mmdet - INFO - Iter [5550/10536] lr: 2.000e-04, eta: 3:24:36, time: 2.472, data_time: 0.057, memory: 32270, loss_depth: 8.6790, task0.loss_xy: 0.1015, task0.loss_z: 0.0718, task0.loss_whl: 0.0568, task0.loss_yaw: 0.1751, task0.loss_vel: 0.3091, task0.loss_heatmap: 1.1042, task1.loss_xy: 0.1072, task1.loss_z: 0.0946, task1.loss_whl: 0.1106, task1.loss_yaw: 0.2054, task1.loss_vel: 0.2070, task1.loss_heatmap: 1.4433, task2.loss_xy: 0.1118, task2.loss_z: 0.0944, task2.loss_whl: 0.0968, task2.loss_yaw: 0.2501, task2.loss_vel: 0.3302, task2.loss_heatmap: 1.3233, task3.loss_xy: 0.1076, task3.loss_z: 0.0545, task3.loss_whl: 0.1136, task3.loss_yaw: 0.2449, task3.loss_vel: 0.0190, task3.loss_heatmap: 0.8654, task4.loss_xy: 0.0951, task4.loss_z: 0.0588, task4.loss_whl: 0.0931, task4.loss_yaw: 0.2856, task4.loss_vel: 0.2064, task4.loss_heatmap: 0.9578, task5.loss_xy: 0.1111, task5.loss_z: 0.0707, task5.loss_whl: 0.1265, task5.loss_yaw: 0.2705, task5.loss_vel: 0.1728, task5.loss_heatmap: 1.2237, loss: 19.9493, grad_norm: 11.3414 2023-03-13 04:02:42,376 - mmdet - INFO - Iter [5600/10536] lr: 2.000e-04, eta: 3:22:31, time: 2.431, data_time: 0.057, memory: 32270, loss_depth: 8.6109, task0.loss_xy: 0.1018, task0.loss_z: 0.0714, task0.loss_whl: 0.0556, task0.loss_yaw: 0.1763, task0.loss_vel: 0.2950, task0.loss_heatmap: 1.0952, task1.loss_xy: 0.1088, task1.loss_z: 0.0988, task1.loss_whl: 0.1088, task1.loss_yaw: 0.2148, task1.loss_vel: 0.2782, task1.loss_heatmap: 1.5776, task2.loss_xy: 0.1103, task2.loss_z: 0.0876, task2.loss_whl: 0.1132, task2.loss_yaw: 0.2537, task2.loss_vel: 0.2592, task2.loss_heatmap: 1.3962, task3.loss_xy: 0.1059, task3.loss_z: 0.0588, task3.loss_whl: 0.1104, task3.loss_yaw: 0.2590, task3.loss_vel: 0.0202, task3.loss_heatmap: 0.8809, task4.loss_xy: 0.1034, task4.loss_z: 0.0651, task4.loss_whl: 0.1044, task4.loss_yaw: 0.2828, task4.loss_vel: 0.3578, task4.loss_heatmap: 1.2261, task5.loss_xy: 0.1091, task5.loss_z: 0.0679, task5.loss_whl: 0.1281, task5.loss_yaw: 0.2705, task5.loss_vel: 0.1604, task5.loss_heatmap: 1.2545, loss: 20.5790, grad_norm: 13.1558 2023-03-13 04:04:41,962 - mmdet - INFO - Iter [5650/10536] lr: 2.000e-04, eta: 3:20:22, time: 2.392, data_time: 0.055, memory: 32270, loss_depth: 8.8565, task0.loss_xy: 0.1014, task0.loss_z: 0.0723, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1772, task0.loss_vel: 0.2648, task0.loss_heatmap: 1.0932, task1.loss_xy: 0.1075, task1.loss_z: 0.0821, task1.loss_whl: 0.0941, task1.loss_yaw: 0.1979, task1.loss_vel: 0.2469, task1.loss_heatmap: 1.3942, task2.loss_xy: 0.1123, task2.loss_z: 0.0862, task2.loss_whl: 0.0983, task2.loss_yaw: 0.2344, task2.loss_vel: 0.3113, task2.loss_heatmap: 1.3287, task3.loss_xy: 0.1102, task3.loss_z: 0.0550, task3.loss_whl: 0.1167, task3.loss_yaw: 0.2707, task3.loss_vel: 0.0178, task3.loss_heatmap: 0.9102, task4.loss_xy: 0.0965, task4.loss_z: 0.0647, task4.loss_whl: 0.1002, task4.loss_yaw: 0.2805, task4.loss_vel: 0.2012, task4.loss_heatmap: 1.0964, task5.loss_xy: 0.1098, task5.loss_z: 0.0718, task5.loss_whl: 0.1330, task5.loss_yaw: 0.2751, task5.loss_vel: 0.1439, task5.loss_heatmap: 1.2488, loss: 20.2185, grad_norm: 12.1551 2023-03-13 04:06:41,754 - mmdet - INFO - Iter [5700/10536] lr: 2.000e-04, eta: 3:18:14, time: 2.396, data_time: 0.057, memory: 32270, loss_depth: 8.8050, task0.loss_xy: 0.1031, task0.loss_z: 0.0719, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1761, task0.loss_vel: 0.2703, task0.loss_heatmap: 1.0983, task1.loss_xy: 0.1071, task1.loss_z: 0.0872, task1.loss_whl: 0.0989, task1.loss_yaw: 0.1943, task1.loss_vel: 0.2457, task1.loss_heatmap: 1.4355, task2.loss_xy: 0.1118, task2.loss_z: 0.0860, task2.loss_whl: 0.1062, task2.loss_yaw: 0.2480, task2.loss_vel: 0.2925, task2.loss_heatmap: 1.4211, task3.loss_xy: 0.1043, task3.loss_z: 0.0491, task3.loss_whl: 0.1238, task3.loss_yaw: 0.2529, task3.loss_vel: 0.0138, task3.loss_heatmap: 0.8619, task4.loss_xy: 0.0988, task4.loss_z: 0.0535, task4.loss_whl: 0.1049, task4.loss_yaw: 0.2720, task4.loss_vel: 0.2579, task4.loss_heatmap: 0.9699, task5.loss_xy: 0.1112, task5.loss_z: 0.0712, task5.loss_whl: 0.1274, task5.loss_yaw: 0.2737, task5.loss_vel: 0.1549, task5.loss_heatmap: 1.2967, loss: 20.2138, grad_norm: 11.8752 2023-03-13 04:08:42,886 - mmdet - INFO - Iter [5750/10536] lr: 2.000e-04, eta: 3:16:08, time: 2.423, data_time: 0.056, memory: 32270, loss_depth: 8.7750, task0.loss_xy: 0.1023, task0.loss_z: 0.0716, task0.loss_whl: 0.0573, task0.loss_yaw: 0.1750, task0.loss_vel: 0.2700, task0.loss_heatmap: 1.0955, task1.loss_xy: 0.1060, task1.loss_z: 0.0859, task1.loss_whl: 0.1030, task1.loss_yaw: 0.1959, task1.loss_vel: 0.2671, task1.loss_heatmap: 1.4286, task2.loss_xy: 0.1089, task2.loss_z: 0.0939, task2.loss_whl: 0.0985, task2.loss_yaw: 0.2309, task2.loss_vel: 0.3255, task2.loss_heatmap: 1.2275, task3.loss_xy: 0.1070, task3.loss_z: 0.0541, task3.loss_whl: 0.1133, task3.loss_yaw: 0.2485, task3.loss_vel: 0.0248, task3.loss_heatmap: 0.9454, task4.loss_xy: 0.0991, task4.loss_z: 0.0602, task4.loss_whl: 0.1027, task4.loss_yaw: 0.2667, task4.loss_vel: 0.2729, task4.loss_heatmap: 1.1189, task5.loss_xy: 0.1114, task5.loss_z: 0.0698, task5.loss_whl: 0.1373, task5.loss_yaw: 0.2713, task5.loss_vel: 0.1735, task5.loss_heatmap: 1.3278, loss: 20.3226, grad_norm: 12.8278 2023-03-13 04:10:43,507 - mmdet - INFO - Iter [5800/10536] lr: 2.000e-04, eta: 3:14:01, time: 2.412, data_time: 0.054, memory: 32270, loss_depth: 8.6194, task0.loss_xy: 0.1012, task0.loss_z: 0.0689, task0.loss_whl: 0.0571, task0.loss_yaw: 0.1756, task0.loss_vel: 0.2731, task0.loss_heatmap: 1.0663, task1.loss_xy: 0.1080, task1.loss_z: 0.0842, task1.loss_whl: 0.1018, task1.loss_yaw: 0.2129, task1.loss_vel: 0.1865, task1.loss_heatmap: 1.4723, task2.loss_xy: 0.1091, task2.loss_z: 0.0813, task2.loss_whl: 0.1006, task2.loss_yaw: 0.2428, task2.loss_vel: 0.2415, task2.loss_heatmap: 1.1452, task3.loss_xy: 0.1056, task3.loss_z: 0.0478, task3.loss_whl: 0.1134, task3.loss_yaw: 0.2365, task3.loss_vel: 0.0109, task3.loss_heatmap: 0.8967, task4.loss_xy: 0.0990, task4.loss_z: 0.0611, task4.loss_whl: 0.1027, task4.loss_yaw: 0.2593, task4.loss_vel: 0.3915, task4.loss_heatmap: 1.0491, task5.loss_xy: 0.1102, task5.loss_z: 0.0702, task5.loss_whl: 0.1315, task5.loss_yaw: 0.2752, task5.loss_vel: 0.1613, task5.loss_heatmap: 1.2411, loss: 19.8110, grad_norm: 11.3689 2023-03-13 04:12:45,851 - mmdet - INFO - Iter [5850/10536] lr: 2.000e-04, eta: 3:11:58, time: 2.447, data_time: 0.062, memory: 32270, loss_depth: 8.5994, task0.loss_xy: 0.1025, task0.loss_z: 0.0757, task0.loss_whl: 0.0581, task0.loss_yaw: 0.1745, task0.loss_vel: 0.2528, task0.loss_heatmap: 1.0881, task1.loss_xy: 0.1064, task1.loss_z: 0.0864, task1.loss_whl: 0.0983, task1.loss_yaw: 0.1971, task1.loss_vel: 0.2698, task1.loss_heatmap: 1.4394, task2.loss_xy: 0.1140, task2.loss_z: 0.0914, task2.loss_whl: 0.0985, task2.loss_yaw: 0.2420, task2.loss_vel: 0.3006, task2.loss_heatmap: 1.3933, task3.loss_xy: 0.1040, task3.loss_z: 0.0479, task3.loss_whl: 0.1068, task3.loss_yaw: 0.2228, task3.loss_vel: 0.0159, task3.loss_heatmap: 0.8237, task4.loss_xy: 0.0955, task4.loss_z: 0.0586, task4.loss_whl: 0.0915, task4.loss_yaw: 0.2666, task4.loss_vel: 0.2641, task4.loss_heatmap: 0.9620, task5.loss_xy: 0.1102, task5.loss_z: 0.0687, task5.loss_whl: 0.1237, task5.loss_yaw: 0.2705, task5.loss_vel: 0.1529, task5.loss_heatmap: 1.1988, loss: 19.7724, grad_norm: 11.5355 2023-03-13 04:14:45,466 - mmdet - INFO - Iter [5900/10536] lr: 2.000e-04, eta: 3:09:50, time: 2.392, data_time: 0.056, memory: 32270, loss_depth: 8.7040, task0.loss_xy: 0.1000, task0.loss_z: 0.0709, task0.loss_whl: 0.0561, task0.loss_yaw: 0.1660, task0.loss_vel: 0.2769, task0.loss_heatmap: 1.0594, task1.loss_xy: 0.1074, task1.loss_z: 0.0910, task1.loss_whl: 0.1023, task1.loss_yaw: 0.2024, task1.loss_vel: 0.2140, task1.loss_heatmap: 1.4643, task2.loss_xy: 0.1104, task2.loss_z: 0.0879, task2.loss_whl: 0.0967, task2.loss_yaw: 0.2392, task2.loss_vel: 0.3722, task2.loss_heatmap: 1.3702, task3.loss_xy: 0.1068, task3.loss_z: 0.0568, task3.loss_whl: 0.1223, task3.loss_yaw: 0.2674, task3.loss_vel: 0.0141, task3.loss_heatmap: 1.0542, task4.loss_xy: 0.0942, task4.loss_z: 0.0608, task4.loss_whl: 0.1079, task4.loss_yaw: 0.2801, task4.loss_vel: 0.1636, task4.loss_heatmap: 1.0071, task5.loss_xy: 0.1105, task5.loss_z: 0.0691, task5.loss_whl: 0.1339, task5.loss_yaw: 0.2714, task5.loss_vel: 0.1571, task5.loss_heatmap: 1.2451, loss: 20.2137, grad_norm: 11.9317 2023-03-13 04:16:50,130 - mmdet - INFO - Iter [5950/10536] lr: 2.000e-04, eta: 3:07:50, time: 2.493, data_time: 0.055, memory: 32270, loss_depth: 8.7060, task0.loss_xy: 0.1018, task0.loss_z: 0.0729, task0.loss_whl: 0.0560, task0.loss_yaw: 0.1676, task0.loss_vel: 0.3302, task0.loss_heatmap: 1.0634, task1.loss_xy: 0.1062, task1.loss_z: 0.0887, task1.loss_whl: 0.1041, task1.loss_yaw: 0.1984, task1.loss_vel: 0.2353, task1.loss_heatmap: 1.4268, task2.loss_xy: 0.1115, task2.loss_z: 0.0819, task2.loss_whl: 0.1019, task2.loss_yaw: 0.2458, task2.loss_vel: 0.3126, task2.loss_heatmap: 1.3738, task3.loss_xy: 0.1047, task3.loss_z: 0.0504, task3.loss_whl: 0.1070, task3.loss_yaw: 0.2498, task3.loss_vel: 0.0182, task3.loss_heatmap: 0.8150, task4.loss_xy: 0.0939, task4.loss_z: 0.0640, task4.loss_whl: 0.1028, task4.loss_yaw: 0.2807, task4.loss_vel: 0.1978, task4.loss_heatmap: 1.0064, task5.loss_xy: 0.1093, task5.loss_z: 0.0645, task5.loss_whl: 0.1259, task5.loss_yaw: 0.2704, task5.loss_vel: 0.1636, task5.loss_heatmap: 1.2213, loss: 19.9307, grad_norm: 12.4190 2023-03-13 04:18:51,449 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 04:18:51,450 - mmdet - INFO - Iter [6000/10536] lr: 2.000e-04, eta: 3:05:45, time: 2.426, data_time: 0.056, memory: 32270, loss_depth: 8.6374, task0.loss_xy: 0.1012, task0.loss_z: 0.0675, task0.loss_whl: 0.0556, task0.loss_yaw: 0.1696, task0.loss_vel: 0.2096, task0.loss_heatmap: 1.0518, task1.loss_xy: 0.1082, task1.loss_z: 0.0900, task1.loss_whl: 0.0999, task1.loss_yaw: 0.1990, task1.loss_vel: 0.2050, task1.loss_heatmap: 1.4688, task2.loss_xy: 0.1073, task2.loss_z: 0.0868, task2.loss_whl: 0.0978, task2.loss_yaw: 0.2391, task2.loss_vel: 0.2671, task2.loss_heatmap: 1.2525, task3.loss_xy: 0.1075, task3.loss_z: 0.0506, task3.loss_whl: 0.1268, task3.loss_yaw: 0.2563, task3.loss_vel: 0.0155, task3.loss_heatmap: 0.9008, task4.loss_xy: 0.0988, task4.loss_z: 0.0582, task4.loss_whl: 0.0961, task4.loss_yaw: 0.2690, task4.loss_vel: 0.1888, task4.loss_heatmap: 0.9231, task5.loss_xy: 0.1096, task5.loss_z: 0.0696, task5.loss_whl: 0.1285, task5.loss_yaw: 0.2680, task5.loss_vel: 0.1667, task5.loss_heatmap: 1.2669, loss: 19.6149, grad_norm: 11.3049 2023-03-13 04:20:54,512 - mmdet - INFO - Iter [6050/10536] lr: 2.000e-04, eta: 3:03:42, time: 2.461, data_time: 0.056, memory: 32270, loss_depth: 8.6490, task0.loss_xy: 0.1000, task0.loss_z: 0.0683, task0.loss_whl: 0.0580, task0.loss_yaw: 0.1666, task0.loss_vel: 0.2841, task0.loss_heatmap: 1.0433, task1.loss_xy: 0.1054, task1.loss_z: 0.0841, task1.loss_whl: 0.0997, task1.loss_yaw: 0.1946, task1.loss_vel: 0.2077, task1.loss_heatmap: 1.3737, task2.loss_xy: 0.1104, task2.loss_z: 0.0826, task2.loss_whl: 0.1056, task2.loss_yaw: 0.2360, task2.loss_vel: 0.3262, task2.loss_heatmap: 1.2483, task3.loss_xy: 0.1062, task3.loss_z: 0.0549, task3.loss_whl: 0.1062, task3.loss_yaw: 0.2600, task3.loss_vel: 0.0164, task3.loss_heatmap: 0.9139, task4.loss_xy: 0.0971, task4.loss_z: 0.0528, task4.loss_whl: 0.0954, task4.loss_yaw: 0.2621, task4.loss_vel: 0.2871, task4.loss_heatmap: 0.9996, task5.loss_xy: 0.1106, task5.loss_z: 0.0669, task5.loss_whl: 0.1350, task5.loss_yaw: 0.2660, task5.loss_vel: 0.1575, task5.loss_heatmap: 1.1928, loss: 19.7241, grad_norm: 12.1526 2023-03-13 04:22:56,775 - mmdet - INFO - Iter [6100/10536] lr: 2.000e-04, eta: 3:01:39, time: 2.445, data_time: 0.056, memory: 32270, loss_depth: 8.6847, task0.loss_xy: 0.1029, task0.loss_z: 0.0771, task0.loss_whl: 0.0567, task0.loss_yaw: 0.1772, task0.loss_vel: 0.2193, task0.loss_heatmap: 1.1162, task1.loss_xy: 0.1081, task1.loss_z: 0.0908, task1.loss_whl: 0.1004, task1.loss_yaw: 0.2032, task1.loss_vel: 0.2036, task1.loss_heatmap: 1.4695, task2.loss_xy: 0.1130, task2.loss_z: 0.0974, task2.loss_whl: 0.1150, task2.loss_yaw: 0.2519, task2.loss_vel: 0.2240, task2.loss_heatmap: 1.4225, task3.loss_xy: 0.1072, task3.loss_z: 0.0540, task3.loss_whl: 0.1302, task3.loss_yaw: 0.2518, task3.loss_vel: 0.0171, task3.loss_heatmap: 0.9394, task4.loss_xy: 0.0993, task4.loss_z: 0.0641, task4.loss_whl: 0.1042, task4.loss_yaw: 0.2751, task4.loss_vel: 0.2617, task4.loss_heatmap: 1.0517, task5.loss_xy: 0.1107, task5.loss_z: 0.0659, task5.loss_whl: 0.1300, task5.loss_yaw: 0.2707, task5.loss_vel: 0.1623, task5.loss_heatmap: 1.2020, loss: 20.1310, grad_norm: 11.0769 2023-03-13 04:24:56,130 - mmdet - INFO - Iter [6150/10536] lr: 2.000e-04, eta: 2:59:31, time: 2.387, data_time: 0.057, memory: 32270, loss_depth: 8.8259, task0.loss_xy: 0.1024, task0.loss_z: 0.0762, task0.loss_whl: 0.0581, task0.loss_yaw: 0.1683, task0.loss_vel: 0.2800, task0.loss_heatmap: 1.1284, task1.loss_xy: 0.1070, task1.loss_z: 0.0893, task1.loss_whl: 0.1053, task1.loss_yaw: 0.1910, task1.loss_vel: 0.2342, task1.loss_heatmap: 1.4437, task2.loss_xy: 0.1097, task2.loss_z: 0.0898, task2.loss_whl: 0.0964, task2.loss_yaw: 0.2188, task2.loss_vel: 0.3643, task2.loss_heatmap: 1.2600, task3.loss_xy: 0.1041, task3.loss_z: 0.0505, task3.loss_whl: 0.1141, task3.loss_yaw: 0.2144, task3.loss_vel: 0.0172, task3.loss_heatmap: 0.8616, task4.loss_xy: 0.0965, task4.loss_z: 0.0685, task4.loss_whl: 0.1044, task4.loss_yaw: 0.2689, task4.loss_vel: 0.2616, task4.loss_heatmap: 1.0111, task5.loss_xy: 0.1111, task5.loss_z: 0.0692, task5.loss_whl: 0.1282, task5.loss_yaw: 0.2714, task5.loss_vel: 0.1617, task5.loss_heatmap: 1.2129, loss: 20.0760, grad_norm: 11.9395 2023-03-13 04:26:56,614 - mmdet - INFO - Iter [6200/10536] lr: 2.000e-04, eta: 2:57:26, time: 2.410, data_time: 0.054, memory: 32270, loss_depth: 8.7457, task0.loss_xy: 0.1016, task0.loss_z: 0.0710, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1669, task0.loss_vel: 0.2713, task0.loss_heatmap: 1.0552, task1.loss_xy: 0.1066, task1.loss_z: 0.0865, task1.loss_whl: 0.1029, task1.loss_yaw: 0.2039, task1.loss_vel: 0.2371, task1.loss_heatmap: 1.4563, task2.loss_xy: 0.1090, task2.loss_z: 0.0834, task2.loss_whl: 0.1087, task2.loss_yaw: 0.2345, task2.loss_vel: 0.4013, task2.loss_heatmap: 1.2951, task3.loss_xy: 0.1036, task3.loss_z: 0.0482, task3.loss_whl: 0.1083, task3.loss_yaw: 0.2390, task3.loss_vel: 0.0192, task3.loss_heatmap: 0.7937, task4.loss_xy: 0.0942, task4.loss_z: 0.0527, task4.loss_whl: 0.1044, task4.loss_yaw: 0.2772, task4.loss_vel: 0.2030, task4.loss_heatmap: 0.9330, task5.loss_xy: 0.1113, task5.loss_z: 0.0665, task5.loss_whl: 0.1285, task5.loss_yaw: 0.2717, task5.loss_vel: 0.1647, task5.loss_heatmap: 1.1688, loss: 19.7805, grad_norm: 12.0882 2023-03-13 04:29:00,580 - mmdet - INFO - Iter [6250/10536] lr: 2.000e-04, eta: 2:55:24, time: 2.479, data_time: 0.085, memory: 32270, loss_depth: 8.6204, task0.loss_xy: 0.1010, task0.loss_z: 0.0676, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1659, task0.loss_vel: 0.2575, task0.loss_heatmap: 1.0616, task1.loss_xy: 0.1069, task1.loss_z: 0.0888, task1.loss_whl: 0.1029, task1.loss_yaw: 0.1916, task1.loss_vel: 0.2342, task1.loss_heatmap: 1.4356, task2.loss_xy: 0.1123, task2.loss_z: 0.0883, task2.loss_whl: 0.1099, task2.loss_yaw: 0.2498, task2.loss_vel: 0.2833, task2.loss_heatmap: 1.3201, task3.loss_xy: 0.1030, task3.loss_z: 0.0502, task3.loss_whl: 0.1357, task3.loss_yaw: 0.2700, task3.loss_vel: 0.0168, task3.loss_heatmap: 0.7783, task4.loss_xy: 0.0970, task4.loss_z: 0.0558, task4.loss_whl: 0.0971, task4.loss_yaw: 0.2663, task4.loss_vel: 0.1844, task4.loss_heatmap: 0.8581, task5.loss_xy: 0.1107, task5.loss_z: 0.0655, task5.loss_whl: 0.1261, task5.loss_yaw: 0.2712, task5.loss_vel: 0.1619, task5.loss_heatmap: 1.2389, loss: 19.5405, grad_norm: 11.9559 2023-03-13 04:31:02,906 - mmdet - INFO - Iter [6300/10536] lr: 2.000e-04, eta: 2:53:21, time: 2.446, data_time: 0.052, memory: 32270, loss_depth: 8.5504, task0.loss_xy: 0.0998, task0.loss_z: 0.0700, task0.loss_whl: 0.0574, task0.loss_yaw: 0.1634, task0.loss_vel: 0.2485, task0.loss_heatmap: 1.0372, task1.loss_xy: 0.1054, task1.loss_z: 0.0825, task1.loss_whl: 0.0935, task1.loss_yaw: 0.1933, task1.loss_vel: 0.1794, task1.loss_heatmap: 1.3538, task2.loss_xy: 0.1085, task2.loss_z: 0.0769, task2.loss_whl: 0.0915, task2.loss_yaw: 0.2295, task2.loss_vel: 0.2612, task2.loss_heatmap: 1.1869, task3.loss_xy: 0.1058, task3.loss_z: 0.0500, task3.loss_whl: 0.1136, task3.loss_yaw: 0.2656, task3.loss_vel: 0.0159, task3.loss_heatmap: 0.8824, task4.loss_xy: 0.0946, task4.loss_z: 0.0566, task4.loss_whl: 0.1066, task4.loss_yaw: 0.2743, task4.loss_vel: 0.3269, task4.loss_heatmap: 1.0992, task5.loss_xy: 0.1110, task5.loss_z: 0.0632, task5.loss_whl: 0.1254, task5.loss_yaw: 0.2667, task5.loss_vel: 0.1675, task5.loss_heatmap: 1.1867, loss: 19.5012, grad_norm: 12.4397 2023-03-13 04:33:08,669 - mmdet - INFO - Iter [6350/10536] lr: 2.000e-04, eta: 2:51:22, time: 2.515, data_time: 0.055, memory: 32270, loss_depth: 8.6898, task0.loss_xy: 0.1010, task0.loss_z: 0.0686, task0.loss_whl: 0.0574, task0.loss_yaw: 0.1673, task0.loss_vel: 0.2790, task0.loss_heatmap: 1.0652, task1.loss_xy: 0.1042, task1.loss_z: 0.0846, task1.loss_whl: 0.0996, task1.loss_yaw: 0.1997, task1.loss_vel: 0.1824, task1.loss_heatmap: 1.4348, task2.loss_xy: 0.1058, task2.loss_z: 0.0792, task2.loss_whl: 0.0922, task2.loss_yaw: 0.2114, task2.loss_vel: 0.3472, task2.loss_heatmap: 1.1772, task3.loss_xy: 0.1072, task3.loss_z: 0.0504, task3.loss_whl: 0.1075, task3.loss_yaw: 0.2668, task3.loss_vel: 0.0194, task3.loss_heatmap: 0.9380, task4.loss_xy: 0.0984, task4.loss_z: 0.0560, task4.loss_whl: 0.0991, task4.loss_yaw: 0.2619, task4.loss_vel: 0.2919, task4.loss_heatmap: 1.0050, task5.loss_xy: 0.1093, task5.loss_z: 0.0621, task5.loss_whl: 0.1256, task5.loss_yaw: 0.2664, task5.loss_vel: 0.1601, task5.loss_heatmap: 1.2000, loss: 19.7716, grad_norm: 14.0613 2023-03-13 04:35:12,100 - mmdet - INFO - Iter [6400/10536] lr: 2.000e-04, eta: 2:49:20, time: 2.469, data_time: 0.056, memory: 32270, loss_depth: 8.5708, task0.loss_xy: 0.1012, task0.loss_z: 0.0652, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1716, task0.loss_vel: 0.2550, task0.loss_heatmap: 1.0862, task1.loss_xy: 0.1058, task1.loss_z: 0.0759, task1.loss_whl: 0.0921, task1.loss_yaw: 0.1820, task1.loss_vel: 0.2386, task1.loss_heatmap: 1.4033, task2.loss_xy: 0.1119, task2.loss_z: 0.0766, task2.loss_whl: 0.0925, task2.loss_yaw: 0.2261, task2.loss_vel: 0.3316, task2.loss_heatmap: 1.2996, task3.loss_xy: 0.1038, task3.loss_z: 0.0452, task3.loss_whl: 0.1197, task3.loss_yaw: 0.2390, task3.loss_vel: 0.0162, task3.loss_heatmap: 0.8716, task4.loss_xy: 0.1024, task4.loss_z: 0.0561, task4.loss_whl: 0.1066, task4.loss_yaw: 0.2557, task4.loss_vel: 0.3304, task4.loss_heatmap: 1.0739, task5.loss_xy: 0.1094, task5.loss_z: 0.0637, task5.loss_whl: 0.1275, task5.loss_yaw: 0.2648, task5.loss_vel: 0.1673, task5.loss_heatmap: 1.2770, loss: 19.8727, grad_norm: 12.8005 2023-03-13 04:37:11,820 - mmdet - INFO - Iter [6450/10536] lr: 2.000e-04, eta: 2:47:13, time: 2.394, data_time: 0.055, memory: 32270, loss_depth: 8.8133, task0.loss_xy: 0.1027, task0.loss_z: 0.0752, task0.loss_whl: 0.0553, task0.loss_yaw: 0.1702, task0.loss_vel: 0.2561, task0.loss_heatmap: 1.0825, task1.loss_xy: 0.1073, task1.loss_z: 0.0848, task1.loss_whl: 0.0951, task1.loss_yaw: 0.1910, task1.loss_vel: 0.1926, task1.loss_heatmap: 1.3837, task2.loss_xy: 0.1110, task2.loss_z: 0.0835, task2.loss_whl: 0.1024, task2.loss_yaw: 0.2514, task2.loss_vel: 0.3182, task2.loss_heatmap: 1.2939, task3.loss_xy: 0.1085, task3.loss_z: 0.0516, task3.loss_whl: 0.1103, task3.loss_yaw: 0.2349, task3.loss_vel: 0.0219, task3.loss_heatmap: 0.9499, task4.loss_xy: 0.0955, task4.loss_z: 0.0637, task4.loss_whl: 0.1067, task4.loss_yaw: 0.2712, task4.loss_vel: 0.3197, task4.loss_heatmap: 1.0880, task5.loss_xy: 0.1100, task5.loss_z: 0.0682, task5.loss_whl: 0.1369, task5.loss_yaw: 0.2683, task5.loss_vel: 0.1551, task5.loss_heatmap: 1.2647, loss: 20.1952, grad_norm: 11.3408 2023-03-13 04:39:13,276 - mmdet - INFO - Iter [6500/10536] lr: 2.000e-04, eta: 2:45:09, time: 2.429, data_time: 0.056, memory: 32270, loss_depth: 8.6216, task0.loss_xy: 0.1015, task0.loss_z: 0.0767, task0.loss_whl: 0.0584, task0.loss_yaw: 0.1692, task0.loss_vel: 0.2625, task0.loss_heatmap: 1.0844, task1.loss_xy: 0.1062, task1.loss_z: 0.0893, task1.loss_whl: 0.0984, task1.loss_yaw: 0.1858, task1.loss_vel: 0.2413, task1.loss_heatmap: 1.3919, task2.loss_xy: 0.1062, task2.loss_z: 0.0755, task2.loss_whl: 0.0917, task2.loss_yaw: 0.2055, task2.loss_vel: 0.2900, task2.loss_heatmap: 1.1496, task3.loss_xy: 0.1061, task3.loss_z: 0.0521, task3.loss_whl: 0.1194, task3.loss_yaw: 0.2595, task3.loss_vel: 0.0163, task3.loss_heatmap: 0.8750, task4.loss_xy: 0.0982, task4.loss_z: 0.0685, task4.loss_whl: 0.0992, task4.loss_yaw: 0.2617, task4.loss_vel: 0.2564, task4.loss_heatmap: 1.0302, task5.loss_xy: 0.1102, task5.loss_z: 0.0706, task5.loss_whl: 0.1291, task5.loss_yaw: 0.2733, task5.loss_vel: 0.1435, task5.loss_heatmap: 1.2116, loss: 19.5861, grad_norm: 13.7163 2023-03-13 04:41:16,917 - mmdet - INFO - Iter [6550/10536] lr: 2.000e-04, eta: 2:43:07, time: 2.473, data_time: 0.056, memory: 32270, loss_depth: 8.4785, task0.loss_xy: 0.1003, task0.loss_z: 0.0711, task0.loss_whl: 0.0573, task0.loss_yaw: 0.1584, task0.loss_vel: 0.2780, task0.loss_heatmap: 1.0461, task1.loss_xy: 0.1063, task1.loss_z: 0.0842, task1.loss_whl: 0.1015, task1.loss_yaw: 0.1915, task1.loss_vel: 0.2270, task1.loss_heatmap: 1.3903, task2.loss_xy: 0.1066, task2.loss_z: 0.0883, task2.loss_whl: 0.1004, task2.loss_yaw: 0.2097, task2.loss_vel: 0.2465, task2.loss_heatmap: 1.2454, task3.loss_xy: 0.1068, task3.loss_z: 0.0496, task3.loss_whl: 0.1179, task3.loss_yaw: 0.2273, task3.loss_vel: 0.0189, task3.loss_heatmap: 0.9127, task4.loss_xy: 0.0946, task4.loss_z: 0.0600, task4.loss_whl: 0.0991, task4.loss_yaw: 0.2693, task4.loss_vel: 0.2080, task4.loss_heatmap: 0.9723, task5.loss_xy: 0.1091, task5.loss_z: 0.0679, task5.loss_whl: 0.1321, task5.loss_yaw: 0.2692, task5.loss_vel: 0.1668, task5.loss_heatmap: 1.2353, loss: 19.4043, grad_norm: 12.7794 2023-03-13 04:43:18,802 - mmdet - INFO - Iter [6600/10536] lr: 2.000e-04, eta: 2:41:04, time: 2.438, data_time: 0.056, memory: 32270, loss_depth: 8.6033, task0.loss_xy: 0.1005, task0.loss_z: 0.0701, task0.loss_whl: 0.0560, task0.loss_yaw: 0.1613, task0.loss_vel: 0.2284, task0.loss_heatmap: 1.0358, task1.loss_xy: 0.1089, task1.loss_z: 0.0914, task1.loss_whl: 0.1001, task1.loss_yaw: 0.1947, task1.loss_vel: 0.1906, task1.loss_heatmap: 1.4747, task2.loss_xy: 0.1099, task2.loss_z: 0.0937, task2.loss_whl: 0.1046, task2.loss_yaw: 0.2373, task2.loss_vel: 0.1603, task2.loss_heatmap: 1.1498, task3.loss_xy: 0.1069, task3.loss_z: 0.0558, task3.loss_whl: 0.1176, task3.loss_yaw: 0.2505, task3.loss_vel: 0.0140, task3.loss_heatmap: 0.8988, task4.loss_xy: 0.0969, task4.loss_z: 0.0536, task4.loss_whl: 0.0965, task4.loss_yaw: 0.2707, task4.loss_vel: 0.1733, task4.loss_heatmap: 0.9107, task5.loss_xy: 0.1088, task5.loss_z: 0.0630, task5.loss_whl: 0.1301, task5.loss_yaw: 0.2647, task5.loss_vel: 0.1471, task5.loss_heatmap: 1.2080, loss: 19.2382, grad_norm: 11.2700 2023-03-13 04:45:23,393 - mmdet - INFO - Iter [6650/10536] lr: 2.000e-04, eta: 2:39:03, time: 2.492, data_time: 0.058, memory: 32270, loss_depth: 8.5466, task0.loss_xy: 0.1013, task0.loss_z: 0.0696, task0.loss_whl: 0.0581, task0.loss_yaw: 0.1598, task0.loss_vel: 0.2379, task0.loss_heatmap: 1.0539, task1.loss_xy: 0.1045, task1.loss_z: 0.0802, task1.loss_whl: 0.0929, task1.loss_yaw: 0.1880, task1.loss_vel: 0.2147, task1.loss_heatmap: 1.3939, task2.loss_xy: 0.1071, task2.loss_z: 0.0762, task2.loss_whl: 0.0931, task2.loss_yaw: 0.2197, task2.loss_vel: 0.2251, task2.loss_heatmap: 1.1421, task3.loss_xy: 0.1050, task3.loss_z: 0.0451, task3.loss_whl: 0.1193, task3.loss_yaw: 0.2236, task3.loss_vel: 0.0172, task3.loss_heatmap: 0.8156, task4.loss_xy: 0.0986, task4.loss_z: 0.0528, task4.loss_whl: 0.0920, task4.loss_yaw: 0.2663, task4.loss_vel: 0.1811, task4.loss_heatmap: 0.8122, task5.loss_xy: 0.1102, task5.loss_z: 0.0625, task5.loss_whl: 0.1353, task5.loss_yaw: 0.2597, task5.loss_vel: 0.1539, task5.loss_heatmap: 1.1923, loss: 18.9078, grad_norm: 10.9844 2023-03-13 04:47:25,625 - mmdet - INFO - Iter [6700/10536] lr: 2.000e-04, eta: 2:36:59, time: 2.445, data_time: 0.056, memory: 32270, loss_depth: 8.7239, task0.loss_xy: 0.1014, task0.loss_z: 0.0724, task0.loss_whl: 0.0563, task0.loss_yaw: 0.1664, task0.loss_vel: 0.2545, task0.loss_heatmap: 1.0697, task1.loss_xy: 0.1061, task1.loss_z: 0.0894, task1.loss_whl: 0.1052, task1.loss_yaw: 0.1955, task1.loss_vel: 0.1771, task1.loss_heatmap: 1.4300, task2.loss_xy: 0.1066, task2.loss_z: 0.0754, task2.loss_whl: 0.0920, task2.loss_yaw: 0.2196, task2.loss_vel: 0.2821, task2.loss_heatmap: 1.1539, task3.loss_xy: 0.1050, task3.loss_z: 0.0488, task3.loss_whl: 0.1058, task3.loss_yaw: 0.2323, task3.loss_vel: 0.0190, task3.loss_heatmap: 0.9183, task4.loss_xy: 0.0920, task4.loss_z: 0.0558, task4.loss_whl: 0.1035, task4.loss_yaw: 0.2457, task4.loss_vel: 0.3178, task4.loss_heatmap: 0.9713, task5.loss_xy: 0.1096, task5.loss_z: 0.0630, task5.loss_whl: 0.1282, task5.loss_yaw: 0.2682, task5.loss_vel: 0.1616, task5.loss_heatmap: 1.2012, loss: 19.6246, grad_norm: 12.4994 2023-03-13 04:49:27,793 - mmdet - INFO - Iter [6750/10536] lr: 2.000e-04, eta: 2:34:56, time: 2.443, data_time: 0.055, memory: 32270, loss_depth: 8.5537, task0.loss_xy: 0.1003, task0.loss_z: 0.0679, task0.loss_whl: 0.0559, task0.loss_yaw: 0.1644, task0.loss_vel: 0.2587, task0.loss_heatmap: 1.0377, task1.loss_xy: 0.1074, task1.loss_z: 0.0852, task1.loss_whl: 0.0974, task1.loss_yaw: 0.1922, task1.loss_vel: 0.2038, task1.loss_heatmap: 1.4213, task2.loss_xy: 0.1130, task2.loss_z: 0.0907, task2.loss_whl: 0.1102, task2.loss_yaw: 0.2313, task2.loss_vel: 0.3473, task2.loss_heatmap: 1.2873, task3.loss_xy: 0.1033, task3.loss_z: 0.0518, task3.loss_whl: 0.1209, task3.loss_yaw: 0.2493, task3.loss_vel: 0.0174, task3.loss_heatmap: 0.8506, task4.loss_xy: 0.0912, task4.loss_z: 0.0605, task4.loss_whl: 0.1039, task4.loss_yaw: 0.2517, task4.loss_vel: 0.2648, task4.loss_heatmap: 0.8894, task5.loss_xy: 0.1091, task5.loss_z: 0.0683, task5.loss_whl: 0.1267, task5.loss_yaw: 0.2653, task5.loss_vel: 0.1571, task5.loss_heatmap: 1.2057, loss: 19.5127, grad_norm: 11.3267 2023-03-13 04:51:29,888 - mmdet - INFO - Iter [6800/10536] lr: 2.000e-04, eta: 2:32:53, time: 2.442, data_time: 0.056, memory: 32270, loss_depth: 8.5299, task0.loss_xy: 0.1005, task0.loss_z: 0.0729, task0.loss_whl: 0.0562, task0.loss_yaw: 0.1613, task0.loss_vel: 0.2320, task0.loss_heatmap: 1.0411, task1.loss_xy: 0.1077, task1.loss_z: 0.0881, task1.loss_whl: 0.1052, task1.loss_yaw: 0.1883, task1.loss_vel: 0.2339, task1.loss_heatmap: 1.4442, task2.loss_xy: 0.1112, task2.loss_z: 0.0810, task2.loss_whl: 0.1018, task2.loss_yaw: 0.2362, task2.loss_vel: 0.2309, task2.loss_heatmap: 1.2852, task3.loss_xy: 0.1065, task3.loss_z: 0.0642, task3.loss_whl: 0.1146, task3.loss_yaw: 0.2421, task3.loss_vel: 0.0197, task3.loss_heatmap: 0.9054, task4.loss_xy: 0.0987, task4.loss_z: 0.0628, task4.loss_whl: 0.0985, task4.loss_yaw: 0.2679, task4.loss_vel: 0.2219, task4.loss_heatmap: 0.9848, task5.loss_xy: 0.1103, task5.loss_z: 0.0692, task5.loss_whl: 0.1271, task5.loss_yaw: 0.2670, task5.loss_vel: 0.1645, task5.loss_heatmap: 1.2127, loss: 19.5455, grad_norm: 10.6057 2023-03-13 04:53:30,711 - mmdet - INFO - Iter [6850/10536] lr: 2.000e-04, eta: 2:30:48, time: 2.416, data_time: 0.057, memory: 32270, loss_depth: 8.5586, task0.loss_xy: 0.0999, task0.loss_z: 0.0683, task0.loss_whl: 0.0566, task0.loss_yaw: 0.1628, task0.loss_vel: 0.2342, task0.loss_heatmap: 1.0251, task1.loss_xy: 0.1058, task1.loss_z: 0.0794, task1.loss_whl: 0.0943, task1.loss_yaw: 0.1780, task1.loss_vel: 0.2188, task1.loss_heatmap: 1.3075, task2.loss_xy: 0.1078, task2.loss_z: 0.0823, task2.loss_whl: 0.0985, task2.loss_yaw: 0.2119, task2.loss_vel: 0.2943, task2.loss_heatmap: 1.1955, task3.loss_xy: 0.1056, task3.loss_z: 0.0523, task3.loss_whl: 0.1147, task3.loss_yaw: 0.2267, task3.loss_vel: 0.0140, task3.loss_heatmap: 0.8368, task4.loss_xy: 0.0954, task4.loss_z: 0.0659, task4.loss_whl: 0.0975, task4.loss_yaw: 0.2806, task4.loss_vel: 0.1907, task4.loss_heatmap: 1.0102, task5.loss_xy: 0.1090, task5.loss_z: 0.0678, task5.loss_whl: 0.1272, task5.loss_yaw: 0.2687, task5.loss_vel: 0.1529, task5.loss_heatmap: 1.2089, loss: 19.2046, grad_norm: 12.1447 2023-03-13 04:55:33,252 - mmdet - INFO - Iter [6900/10536] lr: 2.000e-04, eta: 2:28:45, time: 2.451, data_time: 0.058, memory: 32270, loss_depth: 8.5416, task0.loss_xy: 0.1000, task0.loss_z: 0.0651, task0.loss_whl: 0.0543, task0.loss_yaw: 0.1590, task0.loss_vel: 0.2580, task0.loss_heatmap: 1.0249, task1.loss_xy: 0.1045, task1.loss_z: 0.0825, task1.loss_whl: 0.0931, task1.loss_yaw: 0.1828, task1.loss_vel: 0.2379, task1.loss_heatmap: 1.3201, task2.loss_xy: 0.1094, task2.loss_z: 0.0796, task2.loss_whl: 0.0978, task2.loss_yaw: 0.2317, task2.loss_vel: 0.3195, task2.loss_heatmap: 1.2326, task3.loss_xy: 0.1030, task3.loss_z: 0.0443, task3.loss_whl: 0.1116, task3.loss_yaw: 0.2447, task3.loss_vel: 0.0144, task3.loss_heatmap: 0.7654, task4.loss_xy: 0.0951, task4.loss_z: 0.0550, task4.loss_whl: 0.1131, task4.loss_yaw: 0.2558, task4.loss_vel: 0.2690, task4.loss_heatmap: 0.9418, task5.loss_xy: 0.1100, task5.loss_z: 0.0627, task5.loss_whl: 0.1322, task5.loss_yaw: 0.2683, task5.loss_vel: 0.1574, task5.loss_heatmap: 1.1997, loss: 19.2377, grad_norm: 11.3644 2023-03-13 04:57:35,033 - mmdet - INFO - Iter [6950/10536] lr: 2.000e-04, eta: 2:26:42, time: 2.436, data_time: 0.058, memory: 32270, loss_depth: 8.5459, task0.loss_xy: 0.1009, task0.loss_z: 0.0706, task0.loss_whl: 0.0553, task0.loss_yaw: 0.1619, task0.loss_vel: 0.2389, task0.loss_heatmap: 1.0605, task1.loss_xy: 0.1067, task1.loss_z: 0.0862, task1.loss_whl: 0.0975, task1.loss_yaw: 0.1882, task1.loss_vel: 0.1906, task1.loss_heatmap: 1.4137, task2.loss_xy: 0.1077, task2.loss_z: 0.0832, task2.loss_whl: 0.1043, task2.loss_yaw: 0.2269, task2.loss_vel: 0.2277, task2.loss_heatmap: 1.2142, task3.loss_xy: 0.1064, task3.loss_z: 0.0525, task3.loss_whl: 0.1177, task3.loss_yaw: 0.2654, task3.loss_vel: 0.0188, task3.loss_heatmap: 0.8931, task4.loss_xy: 0.1010, task4.loss_z: 0.0575, task4.loss_whl: 0.0987, task4.loss_yaw: 0.2724, task4.loss_vel: 0.2119, task4.loss_heatmap: 0.9778, task5.loss_xy: 0.1096, task5.loss_z: 0.0655, task5.loss_whl: 0.1265, task5.loss_yaw: 0.2662, task5.loss_vel: 0.1656, task5.loss_heatmap: 1.2219, loss: 19.4096, grad_norm: 11.1463 2023-03-13 04:59:39,536 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 04:59:39,536 - mmdet - INFO - Iter [7000/10536] lr: 2.000e-04, eta: 2:24:40, time: 2.490, data_time: 0.056, memory: 32270, loss_depth: 8.5300, task0.loss_xy: 0.1016, task0.loss_z: 0.0761, task0.loss_whl: 0.0576, task0.loss_yaw: 0.1615, task0.loss_vel: 0.2200, task0.loss_heatmap: 1.0684, task1.loss_xy: 0.1053, task1.loss_z: 0.0873, task1.loss_whl: 0.0990, task1.loss_yaw: 0.1846, task1.loss_vel: 0.1673, task1.loss_heatmap: 1.3914, task2.loss_xy: 0.1097, task2.loss_z: 0.0857, task2.loss_whl: 0.1002, task2.loss_yaw: 0.2314, task2.loss_vel: 0.2600, task2.loss_heatmap: 1.3344, task3.loss_xy: 0.1051, task3.loss_z: 0.0500, task3.loss_whl: 0.1100, task3.loss_yaw: 0.2322, task3.loss_vel: 0.0147, task3.loss_heatmap: 0.7987, task4.loss_xy: 0.0938, task4.loss_z: 0.0570, task4.loss_whl: 0.0931, task4.loss_yaw: 0.2637, task4.loss_vel: 0.1684, task4.loss_heatmap: 0.8724, task5.loss_xy: 0.1103, task5.loss_z: 0.0691, task5.loss_whl: 0.1297, task5.loss_yaw: 0.2716, task5.loss_vel: 0.1441, task5.loss_heatmap: 1.1835, loss: 19.1391, grad_norm: 10.5533 2023-03-13 05:00:38,043 - mmdet - INFO - Saving checkpoint at 7024 iterations 2023-03-13 05:01:42,161 - mmdet - INFO - Iter [7050/10536] lr: 2.000e-04, eta: 2:22:38, time: 2.453, data_time: 0.057, memory: 32270, loss_depth: 8.6144, task0.loss_xy: 0.0997, task0.loss_z: 0.0696, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1613, task0.loss_vel: 0.2630, task0.loss_heatmap: 1.0602, task1.loss_xy: 0.1037, task1.loss_z: 0.0812, task1.loss_whl: 0.0970, task1.loss_yaw: 0.1772, task1.loss_vel: 0.2462, task1.loss_heatmap: 1.3353, task2.loss_xy: 0.1060, task2.loss_z: 0.0839, task2.loss_whl: 0.0928, task2.loss_yaw: 0.2081, task2.loss_vel: 0.2973, task2.loss_heatmap: 1.1278, task3.loss_xy: 0.1057, task3.loss_z: 0.0571, task3.loss_whl: 0.1201, task3.loss_yaw: 0.2550, task3.loss_vel: 0.0201, task3.loss_heatmap: 0.9889, task4.loss_xy: 0.0964, task4.loss_z: 0.0552, task4.loss_whl: 0.0959, task4.loss_yaw: 0.2556, task4.loss_vel: 0.3174, task4.loss_heatmap: 0.9952, task5.loss_xy: 0.1109, task5.loss_z: 0.0740, task5.loss_whl: 0.1271, task5.loss_yaw: 0.2671, task5.loss_vel: 0.1675, task5.loss_heatmap: 1.2708, loss: 19.6618, grad_norm: 13.7279 2023-03-13 05:03:46,526 - mmdet - INFO - Iter [7100/10536] lr: 2.000e-04, eta: 2:20:36, time: 2.487, data_time: 0.056, memory: 32270, loss_depth: 8.5110, task0.loss_xy: 0.0983, task0.loss_z: 0.0632, task0.loss_whl: 0.0573, task0.loss_yaw: 0.1553, task0.loss_vel: 0.2148, task0.loss_heatmap: 0.9946, task1.loss_xy: 0.1039, task1.loss_z: 0.0794, task1.loss_whl: 0.0967, task1.loss_yaw: 0.1832, task1.loss_vel: 0.1849, task1.loss_heatmap: 1.3270, task2.loss_xy: 0.1111, task2.loss_z: 0.0781, task2.loss_whl: 0.0900, task2.loss_yaw: 0.2211, task2.loss_vel: 0.2278, task2.loss_heatmap: 1.2000, task3.loss_xy: 0.1089, task3.loss_z: 0.0481, task3.loss_whl: 0.1138, task3.loss_yaw: 0.2441, task3.loss_vel: 0.0147, task3.loss_heatmap: 1.0234, task4.loss_xy: 0.0957, task4.loss_z: 0.0560, task4.loss_whl: 0.0908, task4.loss_yaw: 0.2498, task4.loss_vel: 0.2158, task4.loss_heatmap: 0.8781, task5.loss_xy: 0.1086, task5.loss_z: 0.0674, task5.loss_whl: 0.1339, task5.loss_yaw: 0.2695, task5.loss_vel: 0.1424, task5.loss_heatmap: 1.1680, loss: 19.0265, grad_norm: 11.1363 2023-03-13 05:05:47,113 - mmdet - INFO - Iter [7150/10536] lr: 2.000e-04, eta: 2:18:32, time: 2.412, data_time: 0.056, memory: 32270, loss_depth: 8.4848, task0.loss_xy: 0.1010, task0.loss_z: 0.0684, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1597, task0.loss_vel: 0.2454, task0.loss_heatmap: 1.0555, task1.loss_xy: 0.1064, task1.loss_z: 0.0798, task1.loss_whl: 0.0956, task1.loss_yaw: 0.1814, task1.loss_vel: 0.2149, task1.loss_heatmap: 1.3631, task2.loss_xy: 0.1059, task2.loss_z: 0.0778, task2.loss_whl: 0.0918, task2.loss_yaw: 0.2241, task2.loss_vel: 0.1895, task2.loss_heatmap: 1.1710, task3.loss_xy: 0.1041, task3.loss_z: 0.0540, task3.loss_whl: 0.1162, task3.loss_yaw: 0.2060, task3.loss_vel: 0.0179, task3.loss_heatmap: 0.8975, task4.loss_xy: 0.0933, task4.loss_z: 0.0589, task4.loss_whl: 0.1017, task4.loss_yaw: 0.2656, task4.loss_vel: 0.1657, task4.loss_heatmap: 0.8630, task5.loss_xy: 0.1092, task5.loss_z: 0.0677, task5.loss_whl: 0.1277, task5.loss_yaw: 0.2633, task5.loss_vel: 0.1572, task5.loss_heatmap: 1.2042, loss: 18.9451, grad_norm: 10.9387 2023-03-13 05:07:48,210 - mmdet - INFO - Iter [7200/10536] lr: 2.000e-04, eta: 2:16:28, time: 2.422, data_time: 0.057, memory: 32270, loss_depth: 8.6529, task0.loss_xy: 0.1019, task0.loss_z: 0.0734, task0.loss_whl: 0.0546, task0.loss_yaw: 0.1611, task0.loss_vel: 0.2377, task0.loss_heatmap: 1.0900, task1.loss_xy: 0.1057, task1.loss_z: 0.0884, task1.loss_whl: 0.1021, task1.loss_yaw: 0.1782, task1.loss_vel: 0.1730, task1.loss_heatmap: 1.3266, task2.loss_xy: 0.1063, task2.loss_z: 0.0772, task2.loss_whl: 0.0992, task2.loss_yaw: 0.2130, task2.loss_vel: 0.2297, task2.loss_heatmap: 1.1758, task3.loss_xy: 0.1045, task3.loss_z: 0.0490, task3.loss_whl: 0.1083, task3.loss_yaw: 0.2522, task3.loss_vel: 0.0173, task3.loss_heatmap: 0.8352, task4.loss_xy: 0.0971, task4.loss_z: 0.0518, task4.loss_whl: 0.1025, task4.loss_yaw: 0.2651, task4.loss_vel: 0.1809, task4.loss_heatmap: 0.8478, task5.loss_xy: 0.1106, task5.loss_z: 0.0667, task5.loss_whl: 0.1256, task5.loss_yaw: 0.2661, task5.loss_vel: 0.1704, task5.loss_heatmap: 1.1991, loss: 19.0971, grad_norm: 10.7892 2023-03-13 05:09:48,883 - mmdet - INFO - Iter [7250/10536] lr: 2.000e-04, eta: 2:14:24, time: 2.413, data_time: 0.059, memory: 32270, loss_depth: 8.6083, task0.loss_xy: 0.1003, task0.loss_z: 0.0676, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1561, task0.loss_vel: 0.2385, task0.loss_heatmap: 1.0326, task1.loss_xy: 0.1050, task1.loss_z: 0.0796, task1.loss_whl: 0.0991, task1.loss_yaw: 0.1769, task1.loss_vel: 0.1722, task1.loss_heatmap: 1.3088, task2.loss_xy: 0.1089, task2.loss_z: 0.0764, task2.loss_whl: 0.0933, task2.loss_yaw: 0.2138, task2.loss_vel: 0.2395, task2.loss_heatmap: 1.2047, task3.loss_xy: 0.1085, task3.loss_z: 0.0462, task3.loss_whl: 0.1148, task3.loss_yaw: 0.2462, task3.loss_vel: 0.0162, task3.loss_heatmap: 0.9518, task4.loss_xy: 0.0949, task4.loss_z: 0.0567, task4.loss_whl: 0.0996, task4.loss_yaw: 0.2589, task4.loss_vel: 0.2335, task4.loss_heatmap: 0.9932, task5.loss_xy: 0.1107, task5.loss_z: 0.0638, task5.loss_whl: 0.1214, task5.loss_yaw: 0.2621, task5.loss_vel: 0.1668, task5.loss_heatmap: 1.2221, loss: 19.3045, grad_norm: 12.1370 2023-03-13 05:11:50,372 - mmdet - INFO - Iter [7300/10536] lr: 2.000e-04, eta: 2:12:20, time: 2.430, data_time: 0.057, memory: 32270, loss_depth: 8.5413, task0.loss_xy: 0.1010, task0.loss_z: 0.0671, task0.loss_whl: 0.0558, task0.loss_yaw: 0.1574, task0.loss_vel: 0.2682, task0.loss_heatmap: 1.0655, task1.loss_xy: 0.1046, task1.loss_z: 0.0880, task1.loss_whl: 0.0968, task1.loss_yaw: 0.1823, task1.loss_vel: 0.2135, task1.loss_heatmap: 1.4023, task2.loss_xy: 0.1112, task2.loss_z: 0.0839, task2.loss_whl: 0.1005, task2.loss_yaw: 0.2225, task2.loss_vel: 0.2890, task2.loss_heatmap: 1.2973, task3.loss_xy: 0.1018, task3.loss_z: 0.0461, task3.loss_whl: 0.1225, task3.loss_yaw: 0.2509, task3.loss_vel: 0.0153, task3.loss_heatmap: 0.7810, task4.loss_xy: 0.1001, task4.loss_z: 0.0496, task4.loss_whl: 0.0939, task4.loss_yaw: 0.2473, task4.loss_vel: 0.3578, task4.loss_heatmap: 1.0176, task5.loss_xy: 0.1092, task5.loss_z: 0.0652, task5.loss_whl: 0.1307, task5.loss_yaw: 0.2629, task5.loss_vel: 0.1453, task5.loss_heatmap: 1.2414, loss: 19.5868, grad_norm: 13.3316 2023-03-13 05:13:58,250 - mmdet - INFO - Iter [7350/10536] lr: 2.000e-04, eta: 2:10:21, time: 2.558, data_time: 0.058, memory: 32270, loss_depth: 8.5304, task0.loss_xy: 0.0991, task0.loss_z: 0.0669, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1536, task0.loss_vel: 0.2366, task0.loss_heatmap: 1.0222, task1.loss_xy: 0.1059, task1.loss_z: 0.0832, task1.loss_whl: 0.1005, task1.loss_yaw: 0.1860, task1.loss_vel: 0.1863, task1.loss_heatmap: 1.3899, task2.loss_xy: 0.1067, task2.loss_z: 0.0823, task2.loss_whl: 0.0958, task2.loss_yaw: 0.1972, task2.loss_vel: 0.2674, task2.loss_heatmap: 1.1482, task3.loss_xy: 0.1036, task3.loss_z: 0.0457, task3.loss_whl: 0.1049, task3.loss_yaw: 0.2393, task3.loss_vel: 0.0129, task3.loss_heatmap: 0.7818, task4.loss_xy: 0.0968, task4.loss_z: 0.0634, task4.loss_whl: 0.0926, task4.loss_yaw: 0.2704, task4.loss_vel: 0.2657, task4.loss_heatmap: 0.8878, task5.loss_xy: 0.1085, task5.loss_z: 0.0620, task5.loss_whl: 0.1292, task5.loss_yaw: 0.2624, task5.loss_vel: 0.1583, task5.loss_heatmap: 1.1585, loss: 18.9581, grad_norm: 11.5562 2023-03-13 05:16:02,658 - mmdet - INFO - Iter [7400/10536] lr: 2.000e-04, eta: 2:08:19, time: 2.488, data_time: 0.058, memory: 32270, loss_depth: 8.4071, task0.loss_xy: 0.1002, task0.loss_z: 0.0705, task0.loss_whl: 0.0567, task0.loss_yaw: 0.1591, task0.loss_vel: 0.2517, task0.loss_heatmap: 1.0691, task1.loss_xy: 0.1063, task1.loss_z: 0.0834, task1.loss_whl: 0.0927, task1.loss_yaw: 0.1804, task1.loss_vel: 0.2152, task1.loss_heatmap: 1.4214, task2.loss_xy: 0.1061, task2.loss_z: 0.0780, task2.loss_whl: 0.1101, task2.loss_yaw: 0.2044, task2.loss_vel: 0.2751, task2.loss_heatmap: 1.1225, task3.loss_xy: 0.1050, task3.loss_z: 0.0516, task3.loss_whl: 0.1125, task3.loss_yaw: 0.2168, task3.loss_vel: 0.0222, task3.loss_heatmap: 0.8521, task4.loss_xy: 0.0950, task4.loss_z: 0.0519, task4.loss_whl: 0.1088, task4.loss_yaw: 0.2613, task4.loss_vel: 0.2227, task4.loss_heatmap: 0.8505, task5.loss_xy: 0.1103, task5.loss_z: 0.0644, task5.loss_whl: 0.1283, task5.loss_yaw: 0.2671, task5.loss_vel: 0.1575, task5.loss_heatmap: 1.1561, loss: 18.9440, grad_norm: 10.8175 2023-03-13 05:18:04,853 - mmdet - INFO - Iter [7450/10536] lr: 2.000e-04, eta: 2:06:16, time: 2.444, data_time: 0.057, memory: 32270, loss_depth: 8.5697, task0.loss_xy: 0.0992, task0.loss_z: 0.0667, task0.loss_whl: 0.0567, task0.loss_yaw: 0.1484, task0.loss_vel: 0.2303, task0.loss_heatmap: 1.0020, task1.loss_xy: 0.1063, task1.loss_z: 0.0828, task1.loss_whl: 0.1016, task1.loss_yaw: 0.1826, task1.loss_vel: 0.2193, task1.loss_heatmap: 1.4029, task2.loss_xy: 0.1080, task2.loss_z: 0.0875, task2.loss_whl: 0.1032, task2.loss_yaw: 0.2071, task2.loss_vel: 0.2898, task2.loss_heatmap: 1.1913, task3.loss_xy: 0.1016, task3.loss_z: 0.0515, task3.loss_whl: 0.1139, task3.loss_yaw: 0.2701, task3.loss_vel: 0.0173, task3.loss_heatmap: 0.8974, task4.loss_xy: 0.0945, task4.loss_z: 0.0582, task4.loss_whl: 0.1022, task4.loss_yaw: 0.2681, task4.loss_vel: 0.1502, task4.loss_heatmap: 0.8828, task5.loss_xy: 0.1086, task5.loss_z: 0.0650, task5.loss_whl: 0.1298, task5.loss_yaw: 0.2633, task5.loss_vel: 0.1482, task5.loss_heatmap: 1.1823, loss: 19.1602, grad_norm: 12.0233 2023-03-13 05:20:08,093 - mmdet - INFO - Iter [7500/10536] lr: 2.000e-04, eta: 2:04:14, time: 2.465, data_time: 0.057, memory: 32270, loss_depth: 8.4979, task0.loss_xy: 0.1005, task0.loss_z: 0.0704, task0.loss_whl: 0.0562, task0.loss_yaw: 0.1537, task0.loss_vel: 0.2436, task0.loss_heatmap: 1.0299, task1.loss_xy: 0.1059, task1.loss_z: 0.0838, task1.loss_whl: 0.0961, task1.loss_yaw: 0.1733, task1.loss_vel: 0.2190, task1.loss_heatmap: 1.3196, task2.loss_xy: 0.1087, task2.loss_z: 0.0800, task2.loss_whl: 0.0977, task2.loss_yaw: 0.2073, task2.loss_vel: 0.3142, task2.loss_heatmap: 1.2055, task3.loss_xy: 0.1029, task3.loss_z: 0.0473, task3.loss_whl: 0.1092, task3.loss_yaw: 0.2516, task3.loss_vel: 0.0169, task3.loss_heatmap: 0.8325, task4.loss_xy: 0.0968, task4.loss_z: 0.0506, task4.loss_whl: 0.0994, task4.loss_yaw: 0.2420, task4.loss_vel: 0.3113, task4.loss_heatmap: 0.9339, task5.loss_xy: 0.1089, task5.loss_z: 0.0609, task5.loss_whl: 0.1298, task5.loss_yaw: 0.2632, task5.loss_vel: 0.1522, task5.loss_heatmap: 1.2021, loss: 19.1744, grad_norm: 12.9594 2023-03-13 05:22:11,814 - mmdet - INFO - Iter [7550/10536] lr: 2.000e-04, eta: 2:02:11, time: 2.474, data_time: 0.060, memory: 32270, loss_depth: 8.4326, task0.loss_xy: 0.1000, task0.loss_z: 0.0647, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1511, task0.loss_vel: 0.1932, task0.loss_heatmap: 0.9956, task1.loss_xy: 0.1046, task1.loss_z: 0.0788, task1.loss_whl: 0.0923, task1.loss_yaw: 0.1657, task1.loss_vel: 0.1884, task1.loss_heatmap: 1.3009, task2.loss_xy: 0.1076, task2.loss_z: 0.0761, task2.loss_whl: 0.0979, task2.loss_yaw: 0.2036, task2.loss_vel: 0.2792, task2.loss_heatmap: 1.0958, task3.loss_xy: 0.1035, task3.loss_z: 0.0468, task3.loss_whl: 0.1272, task3.loss_yaw: 0.2534, task3.loss_vel: 0.0154, task3.loss_heatmap: 0.8865, task4.loss_xy: 0.0894, task4.loss_z: 0.0487, task4.loss_whl: 0.0945, task4.loss_yaw: 0.2569, task4.loss_vel: 0.1317, task4.loss_heatmap: 0.7178, task5.loss_xy: 0.1103, task5.loss_z: 0.0652, task5.loss_whl: 0.1313, task5.loss_yaw: 0.2654, task5.loss_vel: 0.1635, task5.loss_heatmap: 1.2027, loss: 18.4938, grad_norm: 12.4236 2023-03-13 05:24:12,297 - mmdet - INFO - Iter [7600/10536] lr: 2.000e-04, eta: 2:00:07, time: 2.410, data_time: 0.058, memory: 32270, loss_depth: 8.5055, task0.loss_xy: 0.0994, task0.loss_z: 0.0687, task0.loss_whl: 0.0529, task0.loss_yaw: 0.1515, task0.loss_vel: 0.2448, task0.loss_heatmap: 1.0392, task1.loss_xy: 0.1051, task1.loss_z: 0.0778, task1.loss_whl: 0.0930, task1.loss_yaw: 0.1740, task1.loss_vel: 0.2296, task1.loss_heatmap: 1.3399, task2.loss_xy: 0.1052, task2.loss_z: 0.0785, task2.loss_whl: 0.0885, task2.loss_yaw: 0.1872, task2.loss_vel: 0.2658, task2.loss_heatmap: 1.0940, task3.loss_xy: 0.1017, task3.loss_z: 0.0514, task3.loss_whl: 0.1200, task3.loss_yaw: 0.2649, task3.loss_vel: 0.0204, task3.loss_heatmap: 0.8115, task4.loss_xy: 0.0970, task4.loss_z: 0.0568, task4.loss_whl: 0.0899, task4.loss_yaw: 0.2487, task4.loss_vel: 0.2769, task4.loss_heatmap: 0.9513, task5.loss_xy: 0.1104, task5.loss_z: 0.0676, task5.loss_whl: 0.1318, task5.loss_yaw: 0.2605, task5.loss_vel: 0.1576, task5.loss_heatmap: 1.2461, loss: 19.0654, grad_norm: 11.1646 2023-03-13 05:26:13,608 - mmdet - INFO - Iter [7650/10536] lr: 2.000e-04, eta: 1:58:04, time: 2.426, data_time: 0.059, memory: 32270, loss_depth: 8.3700, task0.loss_xy: 0.0980, task0.loss_z: 0.0638, task0.loss_whl: 0.0545, task0.loss_yaw: 0.1476, task0.loss_vel: 0.2155, task0.loss_heatmap: 0.9907, task1.loss_xy: 0.1041, task1.loss_z: 0.0767, task1.loss_whl: 0.0990, task1.loss_yaw: 0.1711, task1.loss_vel: 0.1749, task1.loss_heatmap: 1.2965, task2.loss_xy: 0.1039, task2.loss_z: 0.0703, task2.loss_whl: 0.0881, task2.loss_yaw: 0.1813, task2.loss_vel: 0.2711, task2.loss_heatmap: 1.0091, task3.loss_xy: 0.1015, task3.loss_z: 0.0437, task3.loss_whl: 0.0984, task3.loss_yaw: 0.2370, task3.loss_vel: 0.0148, task3.loss_heatmap: 0.8375, task4.loss_xy: 0.0942, task4.loss_z: 0.0498, task4.loss_whl: 0.1015, task4.loss_yaw: 0.2601, task4.loss_vel: 0.2104, task4.loss_heatmap: 0.8366, task5.loss_xy: 0.1095, task5.loss_z: 0.0596, task5.loss_whl: 0.1278, task5.loss_yaw: 0.2620, task5.loss_vel: 0.1503, task5.loss_heatmap: 1.1566, loss: 18.3376, grad_norm: 12.1061 2023-03-13 05:28:17,106 - mmdet - INFO - Iter [7700/10536] lr: 2.000e-04, eta: 1:56:01, time: 2.470, data_time: 0.058, memory: 32270, loss_depth: 8.5690, task0.loss_xy: 0.1008, task0.loss_z: 0.0746, task0.loss_whl: 0.0572, task0.loss_yaw: 0.1587, task0.loss_vel: 0.2268, task0.loss_heatmap: 1.0927, task1.loss_xy: 0.1050, task1.loss_z: 0.0861, task1.loss_whl: 0.0958, task1.loss_yaw: 0.1779, task1.loss_vel: 0.1953, task1.loss_heatmap: 1.3679, task2.loss_xy: 0.1086, task2.loss_z: 0.0797, task2.loss_whl: 0.0870, task2.loss_yaw: 0.1924, task2.loss_vel: 0.2968, task2.loss_heatmap: 1.1504, task3.loss_xy: 0.1042, task3.loss_z: 0.0491, task3.loss_whl: 0.1022, task3.loss_yaw: 0.2236, task3.loss_vel: 0.0161, task3.loss_heatmap: 0.8444, task4.loss_xy: 0.0938, task4.loss_z: 0.0619, task4.loss_whl: 0.0937, task4.loss_yaw: 0.2589, task4.loss_vel: 0.2207, task4.loss_heatmap: 0.8502, task5.loss_xy: 0.1080, task5.loss_z: 0.0639, task5.loss_whl: 0.1329, task5.loss_yaw: 0.2633, task5.loss_vel: 0.1432, task5.loss_heatmap: 1.1622, loss: 19.0152, grad_norm: 12.4144 2023-03-13 05:30:19,756 - mmdet - INFO - Iter [7750/10536] lr: 2.000e-04, eta: 1:53:59, time: 2.453, data_time: 0.057, memory: 32270, loss_depth: 8.5516, task0.loss_xy: 0.0993, task0.loss_z: 0.0688, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1536, task0.loss_vel: 0.2468, task0.loss_heatmap: 1.0371, task1.loss_xy: 0.1049, task1.loss_z: 0.0797, task1.loss_whl: 0.0979, task1.loss_yaw: 0.1730, task1.loss_vel: 0.2274, task1.loss_heatmap: 1.3162, task2.loss_xy: 0.1083, task2.loss_z: 0.0792, task2.loss_whl: 0.1007, task2.loss_yaw: 0.1831, task2.loss_vel: 0.2842, task2.loss_heatmap: 1.1286, task3.loss_xy: 0.1039, task3.loss_z: 0.0487, task3.loss_whl: 0.1019, task3.loss_yaw: 0.2472, task3.loss_vel: 0.0200, task3.loss_heatmap: 0.7968, task4.loss_xy: 0.0972, task4.loss_z: 0.0566, task4.loss_whl: 0.1011, task4.loss_yaw: 0.2690, task4.loss_vel: 0.1590, task4.loss_heatmap: 0.8475, task5.loss_xy: 0.1092, task5.loss_z: 0.0648, task5.loss_whl: 0.1241, task5.loss_yaw: 0.2666, task5.loss_vel: 0.1501, task5.loss_heatmap: 1.1817, loss: 18.8414, grad_norm: 11.9280 2023-03-13 05:32:23,296 - mmdet - INFO - Iter [7800/10536] lr: 2.000e-04, eta: 1:51:56, time: 2.471, data_time: 0.058, memory: 32270, loss_depth: 8.5283, task0.loss_xy: 0.1009, task0.loss_z: 0.0681, task0.loss_whl: 0.0569, task0.loss_yaw: 0.1495, task0.loss_vel: 0.2345, task0.loss_heatmap: 1.0481, task1.loss_xy: 0.1058, task1.loss_z: 0.0867, task1.loss_whl: 0.1013, task1.loss_yaw: 0.1767, task1.loss_vel: 0.2004, task1.loss_heatmap: 1.3654, task2.loss_xy: 0.1089, task2.loss_z: 0.0878, task2.loss_whl: 0.1059, task2.loss_yaw: 0.2055, task2.loss_vel: 0.2286, task2.loss_heatmap: 1.2216, task3.loss_xy: 0.1061, task3.loss_z: 0.0508, task3.loss_whl: 0.1103, task3.loss_yaw: 0.2236, task3.loss_vel: 0.0207, task3.loss_heatmap: 0.8126, task4.loss_xy: 0.0949, task4.loss_z: 0.0588, task4.loss_whl: 0.0939, task4.loss_yaw: 0.2625, task4.loss_vel: 0.2167, task4.loss_heatmap: 0.8516, task5.loss_xy: 0.1100, task5.loss_z: 0.0672, task5.loss_whl: 0.1219, task5.loss_yaw: 0.2598, task5.loss_vel: 0.1624, task5.loss_heatmap: 1.1992, loss: 19.0041, grad_norm: 10.9623 2023-03-13 05:34:25,024 - mmdet - INFO - Iter [7850/10536] lr: 2.000e-04, eta: 1:49:53, time: 2.435, data_time: 0.056, memory: 32270, loss_depth: 8.4797, task0.loss_xy: 0.0993, task0.loss_z: 0.0626, task0.loss_whl: 0.0566, task0.loss_yaw: 0.1471, task0.loss_vel: 0.2406, task0.loss_heatmap: 1.0132, task1.loss_xy: 0.1060, task1.loss_z: 0.0816, task1.loss_whl: 0.1005, task1.loss_yaw: 0.1680, task1.loss_vel: 0.1951, task1.loss_heatmap: 1.3419, task2.loss_xy: 0.1070, task2.loss_z: 0.0767, task2.loss_whl: 0.0919, task2.loss_yaw: 0.1968, task2.loss_vel: 0.2623, task2.loss_heatmap: 1.1381, task3.loss_xy: 0.1020, task3.loss_z: 0.0472, task3.loss_whl: 0.1236, task3.loss_yaw: 0.2377, task3.loss_vel: 0.0188, task3.loss_heatmap: 0.7861, task4.loss_xy: 0.0912, task4.loss_z: 0.0530, task4.loss_whl: 0.0965, task4.loss_yaw: 0.2537, task4.loss_vel: 0.1544, task4.loss_heatmap: 0.7548, task5.loss_xy: 0.1094, task5.loss_z: 0.0641, task5.loss_whl: 0.1240, task5.loss_yaw: 0.2634, task5.loss_vel: 0.1515, task5.loss_heatmap: 1.1505, loss: 18.5470, grad_norm: 10.7719 2023-03-13 05:36:29,152 - mmdet - INFO - Iter [7900/10536] lr: 2.000e-04, eta: 1:47:51, time: 2.483, data_time: 0.059, memory: 32270, loss_depth: 8.4734, task0.loss_xy: 0.1000, task0.loss_z: 0.0656, task0.loss_whl: 0.0558, task0.loss_yaw: 0.1519, task0.loss_vel: 0.2194, task0.loss_heatmap: 1.0226, task1.loss_xy: 0.1051, task1.loss_z: 0.0819, task1.loss_whl: 0.0989, task1.loss_yaw: 0.1771, task1.loss_vel: 0.1494, task1.loss_heatmap: 1.3615, task2.loss_xy: 0.1092, task2.loss_z: 0.0756, task2.loss_whl: 0.0988, task2.loss_yaw: 0.1997, task2.loss_vel: 0.2093, task2.loss_heatmap: 1.1166, task3.loss_xy: 0.1058, task3.loss_z: 0.0462, task3.loss_whl: 0.1166, task3.loss_yaw: 0.2161, task3.loss_vel: 0.0152, task3.loss_heatmap: 0.8544, task4.loss_xy: 0.0942, task4.loss_z: 0.0552, task4.loss_whl: 0.1036, task4.loss_yaw: 0.2493, task4.loss_vel: 0.2238, task4.loss_heatmap: 1.0211, task5.loss_xy: 0.1090, task5.loss_z: 0.0614, task5.loss_whl: 0.1299, task5.loss_yaw: 0.2652, task5.loss_vel: 0.1450, task5.loss_heatmap: 1.1613, loss: 18.8451, grad_norm: 12.1484 2023-03-13 05:38:32,539 - mmdet - INFO - Iter [7950/10536] lr: 2.000e-04, eta: 1:45:49, time: 2.468, data_time: 0.057, memory: 32270, loss_depth: 8.4430, task0.loss_xy: 0.0967, task0.loss_z: 0.0653, task0.loss_whl: 0.0547, task0.loss_yaw: 0.1395, task0.loss_vel: 0.2622, task0.loss_heatmap: 0.9838, task1.loss_xy: 0.1028, task1.loss_z: 0.0794, task1.loss_whl: 0.0946, task1.loss_yaw: 0.1724, task1.loss_vel: 0.1991, task1.loss_heatmap: 1.2719, task2.loss_xy: 0.1025, task2.loss_z: 0.0734, task2.loss_whl: 0.1005, task2.loss_yaw: 0.1785, task2.loss_vel: 0.2675, task2.loss_heatmap: 0.9504, task3.loss_xy: 0.1021, task3.loss_z: 0.0437, task3.loss_whl: 0.0993, task3.loss_yaw: 0.2129, task3.loss_vel: 0.0175, task3.loss_heatmap: 0.7164, task4.loss_xy: 0.0940, task4.loss_z: 0.0560, task4.loss_whl: 0.0961, task4.loss_yaw: 0.2477, task4.loss_vel: 0.3210, task4.loss_heatmap: 0.9064, task5.loss_xy: 0.1099, task5.loss_z: 0.0611, task5.loss_whl: 0.1276, task5.loss_yaw: 0.2667, task5.loss_vel: 0.1534, task5.loss_heatmap: 1.1700, loss: 18.4399, grad_norm: 11.8132 2023-03-13 05:40:37,510 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 05:40:37,510 - mmdet - INFO - Iter [8000/10536] lr: 2.000e-04, eta: 1:43:47, time: 2.499, data_time: 0.057, memory: 32270, loss_depth: 8.3343, task0.loss_xy: 0.1013, task0.loss_z: 0.0649, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1539, task0.loss_vel: 0.2322, task0.loss_heatmap: 1.0533, task1.loss_xy: 0.1061, task1.loss_z: 0.0841, task1.loss_whl: 0.0975, task1.loss_yaw: 0.1793, task1.loss_vel: 0.1775, task1.loss_heatmap: 1.3780, task2.loss_xy: 0.1098, task2.loss_z: 0.0839, task2.loss_whl: 0.0985, task2.loss_yaw: 0.2141, task2.loss_vel: 0.2245, task2.loss_heatmap: 1.2295, task3.loss_xy: 0.1041, task3.loss_z: 0.0453, task3.loss_whl: 0.1318, task3.loss_yaw: 0.2425, task3.loss_vel: 0.0193, task3.loss_heatmap: 0.9555, task4.loss_xy: 0.0947, task4.loss_z: 0.0555, task4.loss_whl: 0.1077, task4.loss_yaw: 0.2514, task4.loss_vel: 0.1624, task4.loss_heatmap: 0.8462, task5.loss_xy: 0.1084, task5.loss_z: 0.0582, task5.loss_whl: 0.1322, task5.loss_yaw: 0.2634, task5.loss_vel: 0.1429, task5.loss_heatmap: 1.1282, loss: 18.8279, grad_norm: 12.4871 2023-03-13 05:42:37,783 - mmdet - INFO - Iter [8050/10536] lr: 2.000e-04, eta: 1:41:43, time: 2.405, data_time: 0.057, memory: 32270, loss_depth: 8.5552, task0.loss_xy: 0.1012, task0.loss_z: 0.0695, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1488, task0.loss_vel: 0.2003, task0.loss_heatmap: 1.0408, task1.loss_xy: 0.1036, task1.loss_z: 0.0854, task1.loss_whl: 0.0942, task1.loss_yaw: 0.1677, task1.loss_vel: 0.2149, task1.loss_heatmap: 1.3499, task2.loss_xy: 0.1083, task2.loss_z: 0.0735, task2.loss_whl: 0.0870, task2.loss_yaw: 0.1790, task2.loss_vel: 0.2433, task2.loss_heatmap: 1.1050, task3.loss_xy: 0.1037, task3.loss_z: 0.0482, task3.loss_whl: 0.1094, task3.loss_yaw: 0.2224, task3.loss_vel: 0.0173, task3.loss_heatmap: 0.7957, task4.loss_xy: 0.0934, task4.loss_z: 0.0566, task4.loss_whl: 0.0985, task4.loss_yaw: 0.2558, task4.loss_vel: 0.1800, task4.loss_heatmap: 0.8833, task5.loss_xy: 0.1097, task5.loss_z: 0.0694, task5.loss_whl: 0.1251, task5.loss_yaw: 0.2631, task5.loss_vel: 0.1610, task5.loss_heatmap: 1.2285, loss: 18.8052, grad_norm: 12.6622 2023-03-13 05:44:40,303 - mmdet - INFO - Iter [8100/10536] lr: 2.000e-04, eta: 1:39:40, time: 2.450, data_time: 0.059, memory: 32270, loss_depth: 8.4675, task0.loss_xy: 0.0986, task0.loss_z: 0.0648, task0.loss_whl: 0.0566, task0.loss_yaw: 0.1408, task0.loss_vel: 0.2348, task0.loss_heatmap: 1.0007, task1.loss_xy: 0.1058, task1.loss_z: 0.0834, task1.loss_whl: 0.0967, task1.loss_yaw: 0.1712, task1.loss_vel: 0.2075, task1.loss_heatmap: 1.3177, task2.loss_xy: 0.1083, task2.loss_z: 0.0808, task2.loss_whl: 0.0891, task2.loss_yaw: 0.1909, task2.loss_vel: 0.2446, task2.loss_heatmap: 1.1244, task3.loss_xy: 0.1042, task3.loss_z: 0.0492, task3.loss_whl: 0.1037, task3.loss_yaw: 0.2603, task3.loss_vel: 0.0192, task3.loss_heatmap: 0.8396, task4.loss_xy: 0.0950, task4.loss_z: 0.0569, task4.loss_whl: 0.0981, task4.loss_yaw: 0.2564, task4.loss_vel: 0.1918, task4.loss_heatmap: 0.8266, task5.loss_xy: 0.1097, task5.loss_z: 0.0663, task5.loss_whl: 0.1303, task5.loss_yaw: 0.2643, task5.loss_vel: 0.1415, task5.loss_heatmap: 1.1975, loss: 18.6950, grad_norm: 11.0062 2023-03-13 05:46:41,442 - mmdet - INFO - Iter [8150/10536] lr: 2.000e-04, eta: 1:37:37, time: 2.423, data_time: 0.059, memory: 32270, loss_depth: 8.4173, task0.loss_xy: 0.0996, task0.loss_z: 0.0689, task0.loss_whl: 0.0563, task0.loss_yaw: 0.1478, task0.loss_vel: 0.2337, task0.loss_heatmap: 1.0485, task1.loss_xy: 0.1032, task1.loss_z: 0.0782, task1.loss_whl: 0.0951, task1.loss_yaw: 0.1609, task1.loss_vel: 0.1974, task1.loss_heatmap: 1.3450, task2.loss_xy: 0.1062, task2.loss_z: 0.0791, task2.loss_whl: 0.0966, task2.loss_yaw: 0.1726, task2.loss_vel: 0.2767, task2.loss_heatmap: 1.0714, task3.loss_xy: 0.1049, task3.loss_z: 0.0508, task3.loss_whl: 0.1354, task3.loss_yaw: 0.2579, task3.loss_vel: 0.0193, task3.loss_heatmap: 0.8977, task4.loss_xy: 0.0959, task4.loss_z: 0.0541, task4.loss_whl: 0.0999, task4.loss_yaw: 0.2517, task4.loss_vel: 0.2507, task4.loss_heatmap: 0.8832, task5.loss_xy: 0.1086, task5.loss_z: 0.0632, task5.loss_whl: 0.1261, task5.loss_yaw: 0.2585, task5.loss_vel: 0.1578, task5.loss_heatmap: 1.2265, loss: 18.8967, grad_norm: 11.3395 2023-03-13 05:48:40,484 - mmdet - INFO - Iter [8200/10536] lr: 2.000e-04, eta: 1:35:32, time: 2.381, data_time: 0.058, memory: 32270, loss_depth: 8.6195, task0.loss_xy: 0.0988, task0.loss_z: 0.0672, task0.loss_whl: 0.0548, task0.loss_yaw: 0.1481, task0.loss_vel: 0.2597, task0.loss_heatmap: 1.0421, task1.loss_xy: 0.1035, task1.loss_z: 0.0807, task1.loss_whl: 0.0899, task1.loss_yaw: 0.1683, task1.loss_vel: 0.2112, task1.loss_heatmap: 1.3191, task2.loss_xy: 0.1078, task2.loss_z: 0.0746, task2.loss_whl: 0.0833, task2.loss_yaw: 0.1513, task2.loss_vel: 0.3485, task2.loss_heatmap: 1.0507, task3.loss_xy: 0.1051, task3.loss_z: 0.0467, task3.loss_whl: 0.1134, task3.loss_yaw: 0.2144, task3.loss_vel: 0.0242, task3.loss_heatmap: 0.8441, task4.loss_xy: 0.0904, task4.loss_z: 0.0526, task4.loss_whl: 0.0984, task4.loss_yaw: 0.2439, task4.loss_vel: 0.2506, task4.loss_heatmap: 0.8109, task5.loss_xy: 0.1104, task5.loss_z: 0.0663, task5.loss_whl: 0.1230, task5.loss_yaw: 0.2620, task5.loss_vel: 0.1731, task5.loss_heatmap: 1.1967, loss: 18.9050, grad_norm: 11.3430 2023-03-13 05:50:40,626 - mmdet - INFO - Iter [8250/10536] lr: 2.000e-04, eta: 1:33:29, time: 2.403, data_time: 0.057, memory: 32270, loss_depth: 8.4646, task0.loss_xy: 0.0986, task0.loss_z: 0.0647, task0.loss_whl: 0.0550, task0.loss_yaw: 0.1482, task0.loss_vel: 0.2195, task0.loss_heatmap: 1.0060, task1.loss_xy: 0.1063, task1.loss_z: 0.0787, task1.loss_whl: 0.1025, task1.loss_yaw: 0.1683, task1.loss_vel: 0.1641, task1.loss_heatmap: 1.3286, task2.loss_xy: 0.1097, task2.loss_z: 0.0873, task2.loss_whl: 0.0994, task2.loss_yaw: 0.2003, task2.loss_vel: 0.1835, task2.loss_heatmap: 1.1587, task3.loss_xy: 0.1023, task3.loss_z: 0.0556, task3.loss_whl: 0.1040, task3.loss_yaw: 0.2384, task3.loss_vel: 0.0152, task3.loss_heatmap: 0.8644, task4.loss_xy: 0.0943, task4.loss_z: 0.0546, task4.loss_whl: 0.0956, task4.loss_yaw: 0.2554, task4.loss_vel: 0.1494, task4.loss_heatmap: 0.7945, task5.loss_xy: 0.1076, task5.loss_z: 0.0599, task5.loss_whl: 0.1267, task5.loss_yaw: 0.2600, task5.loss_vel: 0.1466, task5.loss_heatmap: 1.1134, loss: 18.4821, grad_norm: 10.2590 2023-03-13 05:52:42,777 - mmdet - INFO - Iter [8300/10536] lr: 2.000e-04, eta: 1:31:26, time: 2.443, data_time: 0.056, memory: 32270, loss_depth: 8.3860, task0.loss_xy: 0.0991, task0.loss_z: 0.0634, task0.loss_whl: 0.0553, task0.loss_yaw: 0.1473, task0.loss_vel: 0.2027, task0.loss_heatmap: 0.9914, task1.loss_xy: 0.1038, task1.loss_z: 0.0783, task1.loss_whl: 0.0953, task1.loss_yaw: 0.1631, task1.loss_vel: 0.1709, task1.loss_heatmap: 1.2918, task2.loss_xy: 0.1081, task2.loss_z: 0.0867, task2.loss_whl: 0.1142, task2.loss_yaw: 0.1867, task2.loss_vel: 0.1914, task2.loss_heatmap: 1.1631, task3.loss_xy: 0.0992, task3.loss_z: 0.0461, task3.loss_whl: 0.1113, task3.loss_yaw: 0.2509, task3.loss_vel: 0.0141, task3.loss_heatmap: 0.8147, task4.loss_xy: 0.0907, task4.loss_z: 0.0539, task4.loss_whl: 0.0930, task4.loss_yaw: 0.2698, task4.loss_vel: 0.1736, task4.loss_heatmap: 0.8262, task5.loss_xy: 0.1085, task5.loss_z: 0.0580, task5.loss_whl: 0.1257, task5.loss_yaw: 0.2605, task5.loss_vel: 0.1500, task5.loss_heatmap: 1.1410, loss: 18.3855, grad_norm: 11.2700 2023-03-13 05:54:45,461 - mmdet - INFO - Iter [8350/10536] lr: 2.000e-04, eta: 1:29:23, time: 2.454, data_time: 0.058, memory: 32270, loss_depth: 8.3322, task0.loss_xy: 0.0993, task0.loss_z: 0.0649, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1478, task0.loss_vel: 0.2075, task0.loss_heatmap: 0.9980, task1.loss_xy: 0.1053, task1.loss_z: 0.0780, task1.loss_whl: 0.0989, task1.loss_yaw: 0.1619, task1.loss_vel: 0.1881, task1.loss_heatmap: 1.2821, task2.loss_xy: 0.1048, task2.loss_z: 0.0754, task2.loss_whl: 0.0953, task2.loss_yaw: 0.1849, task2.loss_vel: 0.2449, task2.loss_heatmap: 1.0444, task3.loss_xy: 0.1036, task3.loss_z: 0.0424, task3.loss_whl: 0.1183, task3.loss_yaw: 0.2200, task3.loss_vel: 0.0183, task3.loss_heatmap: 0.7698, task4.loss_xy: 0.0933, task4.loss_z: 0.0538, task4.loss_whl: 0.1050, task4.loss_yaw: 0.2662, task4.loss_vel: 0.1877, task4.loss_heatmap: 0.8405, task5.loss_xy: 0.1089, task5.loss_z: 0.0593, task5.loss_whl: 0.1265, task5.loss_yaw: 0.2634, task5.loss_vel: 0.1486, task5.loss_heatmap: 1.1408, loss: 18.2365, grad_norm: 11.0144 2023-03-13 05:56:48,599 - mmdet - INFO - Iter [8400/10536] lr: 2.000e-04, eta: 1:27:20, time: 2.463, data_time: 0.057, memory: 32270, loss_depth: 8.3906, task0.loss_xy: 0.0992, task0.loss_z: 0.0656, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1446, task0.loss_vel: 0.2046, task0.loss_heatmap: 1.0094, task1.loss_xy: 0.1033, task1.loss_z: 0.0797, task1.loss_whl: 0.0915, task1.loss_yaw: 0.1661, task1.loss_vel: 0.1644, task1.loss_heatmap: 1.2966, task2.loss_xy: 0.1075, task2.loss_z: 0.0762, task2.loss_whl: 0.0914, task2.loss_yaw: 0.1670, task2.loss_vel: 0.2078, task2.loss_heatmap: 1.0486, task3.loss_xy: 0.1030, task3.loss_z: 0.0450, task3.loss_whl: 0.1058, task3.loss_yaw: 0.2487, task3.loss_vel: 0.0150, task3.loss_heatmap: 0.8839, task4.loss_xy: 0.0938, task4.loss_z: 0.0571, task4.loss_whl: 0.1023, task4.loss_yaw: 0.2534, task4.loss_vel: 0.1940, task4.loss_heatmap: 0.8456, task5.loss_xy: 0.1086, task5.loss_z: 0.0608, task5.loss_whl: 0.1250, task5.loss_yaw: 0.2630, task5.loss_vel: 0.1483, task5.loss_heatmap: 1.1651, loss: 18.3878, grad_norm: 11.5673 2023-03-13 05:58:51,112 - mmdet - INFO - Iter [8450/10536] lr: 2.000e-04, eta: 1:25:18, time: 2.450, data_time: 0.057, memory: 32270, loss_depth: 8.5157, task0.loss_xy: 0.1012, task0.loss_z: 0.0655, task0.loss_whl: 0.0558, task0.loss_yaw: 0.1493, task0.loss_vel: 0.2263, task0.loss_heatmap: 1.0403, task1.loss_xy: 0.1019, task1.loss_z: 0.0762, task1.loss_whl: 0.0951, task1.loss_yaw: 0.1613, task1.loss_vel: 0.1599, task1.loss_heatmap: 1.2946, task2.loss_xy: 0.1002, task2.loss_z: 0.0667, task2.loss_whl: 0.0929, task2.loss_yaw: 0.1646, task2.loss_vel: 0.1951, task2.loss_heatmap: 0.9788, task3.loss_xy: 0.1047, task3.loss_z: 0.0442, task3.loss_whl: 0.1132, task3.loss_yaw: 0.2240, task3.loss_vel: 0.0131, task3.loss_heatmap: 0.8246, task4.loss_xy: 0.0877, task4.loss_z: 0.0500, task4.loss_whl: 0.0969, task4.loss_yaw: 0.2563, task4.loss_vel: 0.2290, task4.loss_heatmap: 0.7747, task5.loss_xy: 0.1091, task5.loss_z: 0.0601, task5.loss_whl: 0.1272, task5.loss_yaw: 0.2590, task5.loss_vel: 0.1575, task5.loss_heatmap: 1.1593, loss: 18.3320, grad_norm: 12.4597 2023-03-13 06:00:53,682 - mmdet - INFO - Iter [8500/10536] lr: 2.000e-04, eta: 1:23:15, time: 2.451, data_time: 0.058, memory: 32270, loss_depth: 8.3682, task0.loss_xy: 0.0989, task0.loss_z: 0.0638, task0.loss_whl: 0.0558, task0.loss_yaw: 0.1416, task0.loss_vel: 0.2204, task0.loss_heatmap: 1.0140, task1.loss_xy: 0.1044, task1.loss_z: 0.0751, task1.loss_whl: 0.0946, task1.loss_yaw: 0.1597, task1.loss_vel: 0.1776, task1.loss_heatmap: 1.2467, task2.loss_xy: 0.1054, task2.loss_z: 0.0824, task2.loss_whl: 0.0859, task2.loss_yaw: 0.1793, task2.loss_vel: 0.3031, task2.loss_heatmap: 1.1424, task3.loss_xy: 0.1049, task3.loss_z: 0.0420, task3.loss_whl: 0.1240, task3.loss_yaw: 0.2197, task3.loss_vel: 0.0183, task3.loss_heatmap: 0.7862, task4.loss_xy: 0.0943, task4.loss_z: 0.0534, task4.loss_whl: 0.0969, task4.loss_yaw: 0.2522, task4.loss_vel: 0.2129, task4.loss_heatmap: 0.8286, task5.loss_xy: 0.1079, task5.loss_z: 0.0616, task5.loss_whl: 0.1248, task5.loss_yaw: 0.2573, task5.loss_vel: 0.1439, task5.loss_heatmap: 1.1209, loss: 18.3693, grad_norm: 12.1255 2023-03-13 06:02:56,400 - mmdet - INFO - Iter [8550/10536] lr: 2.000e-04, eta: 1:21:12, time: 2.455, data_time: 0.057, memory: 32270, loss_depth: 8.4936, task0.loss_xy: 0.0996, task0.loss_z: 0.0672, task0.loss_whl: 0.0558, task0.loss_yaw: 0.1468, task0.loss_vel: 0.2002, task0.loss_heatmap: 1.0256, task1.loss_xy: 0.1041, task1.loss_z: 0.0781, task1.loss_whl: 0.0930, task1.loss_yaw: 0.1657, task1.loss_vel: 0.1562, task1.loss_heatmap: 1.2909, task2.loss_xy: 0.1037, task2.loss_z: 0.0738, task2.loss_whl: 0.0957, task2.loss_yaw: 0.1697, task2.loss_vel: 0.2761, task2.loss_heatmap: 1.0385, task3.loss_xy: 0.1031, task3.loss_z: 0.0532, task3.loss_whl: 0.1074, task3.loss_yaw: 0.2622, task3.loss_vel: 0.0217, task3.loss_heatmap: 0.8947, task4.loss_xy: 0.0948, task4.loss_z: 0.0528, task4.loss_whl: 0.0962, task4.loss_yaw: 0.2564, task4.loss_vel: 0.1979, task4.loss_heatmap: 0.8283, task5.loss_xy: 0.1101, task5.loss_z: 0.0655, task5.loss_whl: 0.1266, task5.loss_yaw: 0.2629, task5.loss_vel: 0.1515, task5.loss_heatmap: 1.2128, loss: 18.6325, grad_norm: 13.3198 2023-03-13 06:04:59,136 - mmdet - INFO - Iter [8600/10536] lr: 2.000e-04, eta: 1:19:10, time: 2.455, data_time: 0.058, memory: 32270, loss_depth: 8.2952, task0.loss_xy: 0.0980, task0.loss_z: 0.0622, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1384, task0.loss_vel: 0.2261, task0.loss_heatmap: 0.9910, task1.loss_xy: 0.1050, task1.loss_z: 0.0796, task1.loss_whl: 0.0983, task1.loss_yaw: 0.1667, task1.loss_vel: 0.1539, task1.loss_heatmap: 1.2796, task2.loss_xy: 0.1059, task2.loss_z: 0.0783, task2.loss_whl: 0.1004, task2.loss_yaw: 0.1783, task2.loss_vel: 0.2446, task2.loss_heatmap: 1.1355, task3.loss_xy: 0.0997, task3.loss_z: 0.0432, task3.loss_whl: 0.1090, task3.loss_yaw: 0.2078, task3.loss_vel: 0.0203, task3.loss_heatmap: 0.7580, task4.loss_xy: 0.0961, task4.loss_z: 0.0570, task4.loss_whl: 0.0905, task4.loss_yaw: 0.2501, task4.loss_vel: 0.1900, task4.loss_heatmap: 0.8986, task5.loss_xy: 0.1084, task5.loss_z: 0.0596, task5.loss_whl: 0.1326, task5.loss_yaw: 0.2684, task5.loss_vel: 0.1412, task5.loss_heatmap: 1.1213, loss: 18.2447, grad_norm: 11.6074 2023-03-13 06:06:59,116 - mmdet - INFO - Iter [8650/10536] lr: 2.000e-04, eta: 1:17:06, time: 2.400, data_time: 0.058, memory: 32270, loss_depth: 8.5963, task0.loss_xy: 0.0998, task0.loss_z: 0.0688, task0.loss_whl: 0.0559, task0.loss_yaw: 0.1443, task0.loss_vel: 0.2594, task0.loss_heatmap: 1.0320, task1.loss_xy: 0.1017, task1.loss_z: 0.0767, task1.loss_whl: 0.1037, task1.loss_yaw: 0.1607, task1.loss_vel: 0.2409, task1.loss_heatmap: 1.2747, task2.loss_xy: 0.1057, task2.loss_z: 0.0794, task2.loss_whl: 0.0885, task2.loss_yaw: 0.1586, task2.loss_vel: 0.2988, task2.loss_heatmap: 1.0834, task3.loss_xy: 0.1018, task3.loss_z: 0.0508, task3.loss_whl: 0.1178, task3.loss_yaw: 0.2447, task3.loss_vel: 0.0134, task3.loss_heatmap: 0.9067, task4.loss_xy: 0.0929, task4.loss_z: 0.0553, task4.loss_whl: 0.0862, task4.loss_yaw: 0.2493, task4.loss_vel: 0.2139, task4.loss_heatmap: 0.7516, task5.loss_xy: 0.1088, task5.loss_z: 0.0637, task5.loss_whl: 0.1264, task5.loss_yaw: 0.2626, task5.loss_vel: 0.1425, task5.loss_heatmap: 1.1970, loss: 18.8147, grad_norm: 11.9010 2023-03-13 06:08:59,264 - mmdet - INFO - Iter [8700/10536] lr: 2.000e-04, eta: 1:15:03, time: 2.403, data_time: 0.059, memory: 32270, loss_depth: 8.3627, task0.loss_xy: 0.1007, task0.loss_z: 0.0668, task0.loss_whl: 0.0556, task0.loss_yaw: 0.1449, task0.loss_vel: 0.2231, task0.loss_heatmap: 1.0494, task1.loss_xy: 0.1066, task1.loss_z: 0.0807, task1.loss_whl: 0.0923, task1.loss_yaw: 0.1566, task1.loss_vel: 0.2060, task1.loss_heatmap: 1.3843, task2.loss_xy: 0.1075, task2.loss_z: 0.0755, task2.loss_whl: 0.0975, task2.loss_yaw: 0.1745, task2.loss_vel: 0.2247, task2.loss_heatmap: 1.0575, task3.loss_xy: 0.1002, task3.loss_z: 0.0448, task3.loss_whl: 0.1154, task3.loss_yaw: 0.2350, task3.loss_vel: 0.0170, task3.loss_heatmap: 0.8672, task4.loss_xy: 0.0930, task4.loss_z: 0.0499, task4.loss_whl: 0.1046, task4.loss_yaw: 0.2485, task4.loss_vel: 0.1750, task4.loss_heatmap: 0.8155, task5.loss_xy: 0.1100, task5.loss_z: 0.0651, task5.loss_whl: 0.1254, task5.loss_yaw: 0.2603, task5.loss_vel: 0.1647, task5.loss_heatmap: 1.2246, loss: 18.5829, grad_norm: 11.7457 2023-03-13 06:11:06,631 - mmdet - INFO - Iter [8750/10536] lr: 2.000e-04, eta: 1:13:01, time: 2.547, data_time: 0.059, memory: 32270, loss_depth: 8.3885, task0.loss_xy: 0.0992, task0.loss_z: 0.0643, task0.loss_whl: 0.0575, task0.loss_yaw: 0.1425, task0.loss_vel: 0.2215, task0.loss_heatmap: 1.0153, task1.loss_xy: 0.1057, task1.loss_z: 0.0782, task1.loss_whl: 0.0980, task1.loss_yaw: 0.1640, task1.loss_vel: 0.1991, task1.loss_heatmap: 1.3295, task2.loss_xy: 0.1099, task2.loss_z: 0.0856, task2.loss_whl: 0.0989, task2.loss_yaw: 0.1745, task2.loss_vel: 0.2556, task2.loss_heatmap: 1.2345, task3.loss_xy: 0.1026, task3.loss_z: 0.0483, task3.loss_whl: 0.1239, task3.loss_yaw: 0.2608, task3.loss_vel: 0.0156, task3.loss_heatmap: 0.8785, task4.loss_xy: 0.0922, task4.loss_z: 0.0502, task4.loss_whl: 0.0906, task4.loss_yaw: 0.2340, task4.loss_vel: 0.2110, task4.loss_heatmap: 0.8411, task5.loss_xy: 0.1100, task5.loss_z: 0.0634, task5.loss_whl: 0.1229, task5.loss_yaw: 0.2530, task5.loss_vel: 0.1661, task5.loss_heatmap: 1.1492, loss: 18.7360, grad_norm: 11.0066 2023-03-13 06:12:19,463 - mmdet - INFO - Saving checkpoint at 8780 iterations 2023-03-13 06:13:11,087 - mmdet - INFO - Iter [8800/10536] lr: 2.000e-04, eta: 1:10:59, time: 2.489, data_time: 0.059, memory: 32270, loss_depth: 8.3597, task0.loss_xy: 0.0990, task0.loss_z: 0.0645, task0.loss_whl: 0.0554, task0.loss_yaw: 0.1456, task0.loss_vel: 0.1904, task0.loss_heatmap: 0.9939, task1.loss_xy: 0.1050, task1.loss_z: 0.0796, task1.loss_whl: 0.0976, task1.loss_yaw: 0.1619, task1.loss_vel: 0.2074, task1.loss_heatmap: 1.3623, task2.loss_xy: 0.1108, task2.loss_z: 0.0841, task2.loss_whl: 0.0950, task2.loss_yaw: 0.1832, task2.loss_vel: 0.2130, task2.loss_heatmap: 1.1869, task3.loss_xy: 0.1019, task3.loss_z: 0.0486, task3.loss_whl: 0.1115, task3.loss_yaw: 0.2739, task3.loss_vel: 0.0199, task3.loss_heatmap: 0.8567, task4.loss_xy: 0.0940, task4.loss_z: 0.0513, task4.loss_whl: 0.0963, task4.loss_yaw: 0.2479, task4.loss_vel: 0.1632, task4.loss_heatmap: 0.7856, task5.loss_xy: 0.1096, task5.loss_z: 0.0622, task5.loss_whl: 0.1328, task5.loss_yaw: 0.2624, task5.loss_vel: 0.1476, task5.loss_heatmap: 1.1170, loss: 18.4778, grad_norm: 10.9060 2023-03-13 06:15:14,804 - mmdet - INFO - Iter [8850/10536] lr: 2.000e-04, eta: 1:08:57, time: 2.474, data_time: 0.059, memory: 32270, loss_depth: 8.2763, task0.loss_xy: 0.0986, task0.loss_z: 0.0630, task0.loss_whl: 0.0546, task0.loss_yaw: 0.1368, task0.loss_vel: 0.2479, task0.loss_heatmap: 0.9922, task1.loss_xy: 0.1027, task1.loss_z: 0.0769, task1.loss_whl: 0.0975, task1.loss_yaw: 0.1581, task1.loss_vel: 0.1604, task1.loss_heatmap: 1.2368, task2.loss_xy: 0.1102, task2.loss_z: 0.0817, task2.loss_whl: 0.0942, task2.loss_yaw: 0.1770, task2.loss_vel: 0.2127, task2.loss_heatmap: 1.1527, task3.loss_xy: 0.1039, task3.loss_z: 0.0510, task3.loss_whl: 0.1017, task3.loss_yaw: 0.2299, task3.loss_vel: 0.0161, task3.loss_heatmap: 0.8474, task4.loss_xy: 0.0937, task4.loss_z: 0.0563, task4.loss_whl: 0.0881, task4.loss_yaw: 0.2536, task4.loss_vel: 0.2130, task4.loss_heatmap: 0.8369, task5.loss_xy: 0.1082, task5.loss_z: 0.0609, task5.loss_whl: 0.1245, task5.loss_yaw: 0.2612, task5.loss_vel: 0.1457, task5.loss_heatmap: 1.1428, loss: 18.2654, grad_norm: 10.1609 2023-03-13 06:17:17,301 - mmdet - INFO - Iter [8900/10536] lr: 2.000e-04, eta: 1:06:54, time: 2.450, data_time: 0.058, memory: 32270, loss_depth: 8.2727, task0.loss_xy: 0.0979, task0.loss_z: 0.0646, task0.loss_whl: 0.0562, task0.loss_yaw: 0.1305, task0.loss_vel: 0.2545, task0.loss_heatmap: 0.9763, task1.loss_xy: 0.1054, task1.loss_z: 0.0818, task1.loss_whl: 0.0979, task1.loss_yaw: 0.1581, task1.loss_vel: 0.1907, task1.loss_heatmap: 1.2654, task2.loss_xy: 0.1094, task2.loss_z: 0.0750, task2.loss_whl: 0.1011, task2.loss_yaw: 0.1907, task2.loss_vel: 0.1796, task2.loss_heatmap: 1.1194, task3.loss_xy: 0.1008, task3.loss_z: 0.0445, task3.loss_whl: 0.1136, task3.loss_yaw: 0.2417, task3.loss_vel: 0.0235, task3.loss_heatmap: 0.7025, task4.loss_xy: 0.0940, task4.loss_z: 0.0555, task4.loss_whl: 0.0946, task4.loss_yaw: 0.2451, task4.loss_vel: 0.2335, task4.loss_heatmap: 0.8349, task5.loss_xy: 0.1085, task5.loss_z: 0.0620, task5.loss_whl: 0.1328, task5.loss_yaw: 0.2624, task5.loss_vel: 0.1573, task5.loss_heatmap: 1.1661, loss: 18.2004, grad_norm: 9.8749 2023-03-13 06:19:20,051 - mmdet - INFO - Iter [8950/10536] lr: 2.000e-04, eta: 1:04:51, time: 2.455, data_time: 0.058, memory: 32270, loss_depth: 8.3889, task0.loss_xy: 0.0989, task0.loss_z: 0.0661, task0.loss_whl: 0.0564, task0.loss_yaw: 0.1360, task0.loss_vel: 0.2083, task0.loss_heatmap: 0.9962, task1.loss_xy: 0.1028, task1.loss_z: 0.0773, task1.loss_whl: 0.0939, task1.loss_yaw: 0.1584, task1.loss_vel: 0.1854, task1.loss_heatmap: 1.2866, task2.loss_xy: 0.1064, task2.loss_z: 0.0737, task2.loss_whl: 0.0871, task2.loss_yaw: 0.1567, task2.loss_vel: 0.2450, task2.loss_heatmap: 1.0521, task3.loss_xy: 0.0982, task3.loss_z: 0.0469, task3.loss_whl: 0.1090, task3.loss_yaw: 0.2125, task3.loss_vel: 0.0174, task3.loss_heatmap: 0.6783, task4.loss_xy: 0.0891, task4.loss_z: 0.0524, task4.loss_whl: 0.0965, task4.loss_yaw: 0.2297, task4.loss_vel: 0.1960, task4.loss_heatmap: 0.8343, task5.loss_xy: 0.1088, task5.loss_z: 0.0601, task5.loss_whl: 0.1232, task5.loss_yaw: 0.2572, task5.loss_vel: 0.1505, task5.loss_heatmap: 1.1462, loss: 18.0827, grad_norm: 11.2237 2023-03-13 06:21:20,402 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 06:21:20,403 - mmdet - INFO - Iter [9000/10536] lr: 2.000e-04, eta: 1:02:48, time: 2.407, data_time: 0.057, memory: 32270, loss_depth: 8.2860, task0.loss_xy: 0.0977, task0.loss_z: 0.0598, task0.loss_whl: 0.0574, task0.loss_yaw: 0.1344, task0.loss_vel: 0.1840, task0.loss_heatmap: 0.9502, task1.loss_xy: 0.1042, task1.loss_z: 0.0748, task1.loss_whl: 0.0981, task1.loss_yaw: 0.1563, task1.loss_vel: 0.2072, task1.loss_heatmap: 1.3154, task2.loss_xy: 0.1056, task2.loss_z: 0.0841, task2.loss_whl: 0.1132, task2.loss_yaw: 0.1771, task2.loss_vel: 0.2465, task2.loss_heatmap: 1.0612, task3.loss_xy: 0.1045, task3.loss_z: 0.0449, task3.loss_whl: 0.1061, task3.loss_yaw: 0.2265, task3.loss_vel: 0.0175, task3.loss_heatmap: 0.9249, task4.loss_xy: 0.0940, task4.loss_z: 0.0539, task4.loss_whl: 0.1086, task4.loss_yaw: 0.2578, task4.loss_vel: 0.1617, task4.loss_heatmap: 0.8412, task5.loss_xy: 0.1103, task5.loss_z: 0.0609, task5.loss_whl: 0.1239, task5.loss_yaw: 0.2594, task5.loss_vel: 0.1594, task5.loss_heatmap: 1.1812, loss: 18.3499, grad_norm: 10.8400 2023-03-13 06:23:22,807 - mmdet - INFO - Iter [9050/10536] lr: 2.000e-04, eta: 1:00:45, time: 2.448, data_time: 0.058, memory: 32270, loss_depth: 8.3718, task0.loss_xy: 0.1000, task0.loss_z: 0.0642, task0.loss_whl: 0.0543, task0.loss_yaw: 0.1416, task0.loss_vel: 0.2046, task0.loss_heatmap: 1.0245, task1.loss_xy: 0.1051, task1.loss_z: 0.0771, task1.loss_whl: 0.0976, task1.loss_yaw: 0.1544, task1.loss_vel: 0.1759, task1.loss_heatmap: 1.2857, task2.loss_xy: 0.1071, task2.loss_z: 0.0775, task2.loss_whl: 0.0970, task2.loss_yaw: 0.1666, task2.loss_vel: 0.2683, task2.loss_heatmap: 1.1382, task3.loss_xy: 0.1030, task3.loss_z: 0.0421, task3.loss_whl: 0.1262, task3.loss_yaw: 0.2149, task3.loss_vel: 0.0131, task3.loss_heatmap: 0.8722, task4.loss_xy: 0.0896, task4.loss_z: 0.0472, task4.loss_whl: 0.0935, task4.loss_yaw: 0.2499, task4.loss_vel: 0.1978, task4.loss_heatmap: 0.8431, task5.loss_xy: 0.1090, task5.loss_z: 0.0588, task5.loss_whl: 0.1242, task5.loss_yaw: 0.2601, task5.loss_vel: 0.1449, task5.loss_heatmap: 1.0762, loss: 18.3771, grad_norm: 11.2815 2023-03-13 06:25:24,260 - mmdet - INFO - Iter [9100/10536] lr: 2.000e-04, eta: 0:58:43, time: 2.429, data_time: 0.058, memory: 32270, loss_depth: 8.5213, task0.loss_xy: 0.0992, task0.loss_z: 0.0678, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1423, task0.loss_vel: 0.2158, task0.loss_heatmap: 1.0140, task1.loss_xy: 0.1024, task1.loss_z: 0.0827, task1.loss_whl: 0.1000, task1.loss_yaw: 0.1549, task1.loss_vel: 0.1576, task1.loss_heatmap: 1.2543, task2.loss_xy: 0.1041, task2.loss_z: 0.0729, task2.loss_whl: 0.0921, task2.loss_yaw: 0.1681, task2.loss_vel: 0.2326, task2.loss_heatmap: 1.0647, task3.loss_xy: 0.1029, task3.loss_z: 0.0462, task3.loss_whl: 0.1074, task3.loss_yaw: 0.2615, task3.loss_vel: 0.0144, task3.loss_heatmap: 0.7994, task4.loss_xy: 0.0950, task4.loss_z: 0.0534, task4.loss_whl: 0.0993, task4.loss_yaw: 0.2584, task4.loss_vel: 0.1936, task4.loss_heatmap: 0.8846, task5.loss_xy: 0.1088, task5.loss_z: 0.0609, task5.loss_whl: 0.1296, task5.loss_yaw: 0.2611, task5.loss_vel: 0.1502, task5.loss_heatmap: 1.1537, loss: 18.4832, grad_norm: 11.2929 2023-03-13 06:27:25,114 - mmdet - INFO - Iter [9150/10536] lr: 2.000e-04, eta: 0:56:39, time: 2.417, data_time: 0.058, memory: 32270, loss_depth: 8.6104, task0.loss_xy: 0.1007, task0.loss_z: 0.0656, task0.loss_whl: 0.0561, task0.loss_yaw: 0.1406, task0.loss_vel: 0.2190, task0.loss_heatmap: 1.0647, task1.loss_xy: 0.1048, task1.loss_z: 0.0765, task1.loss_whl: 0.0922, task1.loss_yaw: 0.1553, task1.loss_vel: 0.1913, task1.loss_heatmap: 1.3183, task2.loss_xy: 0.1058, task2.loss_z: 0.0742, task2.loss_whl: 0.0954, task2.loss_yaw: 0.1550, task2.loss_vel: 0.2940, task2.loss_heatmap: 1.0943, task3.loss_xy: 0.1016, task3.loss_z: 0.0458, task3.loss_whl: 0.1000, task3.loss_yaw: 0.2393, task3.loss_vel: 0.0166, task3.loss_heatmap: 0.8295, task4.loss_xy: 0.0983, task4.loss_z: 0.0526, task4.loss_whl: 0.1030, task4.loss_yaw: 0.2486, task4.loss_vel: 0.2265, task4.loss_heatmap: 0.9858, task5.loss_xy: 0.1086, task5.loss_z: 0.0648, task5.loss_whl: 0.1287, task5.loss_yaw: 0.2606, task5.loss_vel: 0.1549, task5.loss_heatmap: 1.2131, loss: 18.9922, grad_norm: 13.3447 2023-03-13 06:29:28,605 - mmdet - INFO - Iter [9200/10536] lr: 2.000e-04, eta: 0:54:37, time: 2.470, data_time: 0.058, memory: 32270, loss_depth: 8.2976, task0.loss_xy: 0.0986, task0.loss_z: 0.0657, task0.loss_whl: 0.0578, task0.loss_yaw: 0.1373, task0.loss_vel: 0.2250, task0.loss_heatmap: 1.0099, task1.loss_xy: 0.1014, task1.loss_z: 0.0805, task1.loss_whl: 0.0892, task1.loss_yaw: 0.1495, task1.loss_vel: 0.1766, task1.loss_heatmap: 1.2450, task2.loss_xy: 0.1038, task2.loss_z: 0.0715, task2.loss_whl: 0.0882, task2.loss_yaw: 0.1528, task2.loss_vel: 0.2195, task2.loss_heatmap: 1.0304, task3.loss_xy: 0.1021, task3.loss_z: 0.0478, task3.loss_whl: 0.1137, task3.loss_yaw: 0.2114, task3.loss_vel: 0.0130, task3.loss_heatmap: 0.8115, task4.loss_xy: 0.0916, task4.loss_z: 0.0525, task4.loss_whl: 0.1035, task4.loss_yaw: 0.2597, task4.loss_vel: 0.1459, task4.loss_heatmap: 0.8876, task5.loss_xy: 0.1095, task5.loss_z: 0.0635, task5.loss_whl: 0.1248, task5.loss_yaw: 0.2603, task5.loss_vel: 0.1559, task5.loss_heatmap: 1.1879, loss: 18.1425, grad_norm: 11.7843 2023-03-13 06:31:30,844 - mmdet - INFO - Iter [9250/10536] lr: 2.000e-04, eta: 0:52:34, time: 2.445, data_time: 0.058, memory: 32270, loss_depth: 8.4971, task0.loss_xy: 0.1005, task0.loss_z: 0.0704, task0.loss_whl: 0.0559, task0.loss_yaw: 0.1382, task0.loss_vel: 0.2280, task0.loss_heatmap: 1.0414, task1.loss_xy: 0.1038, task1.loss_z: 0.0774, task1.loss_whl: 0.0905, task1.loss_yaw: 0.1551, task1.loss_vel: 0.1487, task1.loss_heatmap: 1.2686, task2.loss_xy: 0.1020, task2.loss_z: 0.0704, task2.loss_whl: 0.0887, task2.loss_yaw: 0.1425, task2.loss_vel: 0.2475, task2.loss_heatmap: 0.8709, task3.loss_xy: 0.1024, task3.loss_z: 0.0451, task3.loss_whl: 0.1116, task3.loss_yaw: 0.2561, task3.loss_vel: 0.0152, task3.loss_heatmap: 0.8347, task4.loss_xy: 0.0949, task4.loss_z: 0.0573, task4.loss_whl: 0.0917, task4.loss_yaw: 0.2343, task4.loss_vel: 0.1744, task4.loss_heatmap: 0.8385, task5.loss_xy: 0.1080, task5.loss_z: 0.0640, task5.loss_whl: 0.1243, task5.loss_yaw: 0.2590, task5.loss_vel: 0.1386, task5.loss_heatmap: 1.1351, loss: 18.1827, grad_norm: 11.0727 2023-03-13 06:33:32,892 - mmdet - INFO - Iter [9300/10536] lr: 2.000e-04, eta: 0:50:31, time: 2.441, data_time: 0.058, memory: 32270, loss_depth: 8.2939, task0.loss_xy: 0.0992, task0.loss_z: 0.0623, task0.loss_whl: 0.0563, task0.loss_yaw: 0.1334, task0.loss_vel: 0.2485, task0.loss_heatmap: 1.0117, task1.loss_xy: 0.1033, task1.loss_z: 0.0698, task1.loss_whl: 0.0960, task1.loss_yaw: 0.1504, task1.loss_vel: 0.1976, task1.loss_heatmap: 1.2480, task2.loss_xy: 0.1066, task2.loss_z: 0.0727, task2.loss_whl: 0.0915, task2.loss_yaw: 0.1581, task2.loss_vel: 0.2113, task2.loss_heatmap: 1.0716, task3.loss_xy: 0.1014, task3.loss_z: 0.0432, task3.loss_whl: 0.1038, task3.loss_yaw: 0.2170, task3.loss_vel: 0.0168, task3.loss_heatmap: 0.7072, task4.loss_xy: 0.0929, task4.loss_z: 0.0500, task4.loss_whl: 0.0897, task4.loss_yaw: 0.2360, task4.loss_vel: 0.2393, task4.loss_heatmap: 0.8022, task5.loss_xy: 0.1070, task5.loss_z: 0.0622, task5.loss_whl: 0.1296, task5.loss_yaw: 0.2563, task5.loss_vel: 0.1362, task5.loss_heatmap: 1.1590, loss: 18.0321, grad_norm: 10.6911 2023-03-13 06:35:33,824 - mmdet - INFO - Iter [9350/10536] lr: 2.000e-04, eta: 0:48:28, time: 2.419, data_time: 0.058, memory: 32270, loss_depth: 8.3452, task0.loss_xy: 0.0983, task0.loss_z: 0.0626, task0.loss_whl: 0.0560, task0.loss_yaw: 0.1349, task0.loss_vel: 0.2080, task0.loss_heatmap: 0.9844, task1.loss_xy: 0.1025, task1.loss_z: 0.0721, task1.loss_whl: 0.0927, task1.loss_yaw: 0.1511, task1.loss_vel: 0.1569, task1.loss_heatmap: 1.2067, task2.loss_xy: 0.1080, task2.loss_z: 0.0821, task2.loss_whl: 0.0883, task2.loss_yaw: 0.1760, task2.loss_vel: 0.2016, task2.loss_heatmap: 1.1264, task3.loss_xy: 0.1061, task3.loss_z: 0.0547, task3.loss_whl: 0.1184, task3.loss_yaw: 0.2349, task3.loss_vel: 0.0221, task3.loss_heatmap: 0.9039, task4.loss_xy: 0.0915, task4.loss_z: 0.0470, task4.loss_whl: 0.1000, task4.loss_yaw: 0.2472, task4.loss_vel: 0.1576, task4.loss_heatmap: 0.7242, task5.loss_xy: 0.1095, task5.loss_z: 0.0626, task5.loss_whl: 0.1240, task5.loss_yaw: 0.2526, task5.loss_vel: 0.1615, task5.loss_heatmap: 1.1851, loss: 18.1567, grad_norm: 11.0483 2023-03-13 06:37:37,066 - mmdet - INFO - Iter [9400/10536] lr: 2.000e-04, eta: 0:46:26, time: 2.465, data_time: 0.058, memory: 32270, loss_depth: 8.2838, task0.loss_xy: 0.0990, task0.loss_z: 0.0640, task0.loss_whl: 0.0555, task0.loss_yaw: 0.1352, task0.loss_vel: 0.2292, task0.loss_heatmap: 1.0075, task1.loss_xy: 0.1046, task1.loss_z: 0.0762, task1.loss_whl: 0.0939, task1.loss_yaw: 0.1481, task1.loss_vel: 0.1871, task1.loss_heatmap: 1.2691, task2.loss_xy: 0.1086, task2.loss_z: 0.0798, task2.loss_whl: 0.0983, task2.loss_yaw: 0.1750, task2.loss_vel: 0.2232, task2.loss_heatmap: 1.0722, task3.loss_xy: 0.1043, task3.loss_z: 0.0472, task3.loss_whl: 0.1001, task3.loss_yaw: 0.2279, task3.loss_vel: 0.0199, task3.loss_heatmap: 0.8506, task4.loss_xy: 0.0927, task4.loss_z: 0.0516, task4.loss_whl: 0.0955, task4.loss_yaw: 0.2251, task4.loss_vel: 0.2455, task4.loss_heatmap: 0.7905, task5.loss_xy: 0.1089, task5.loss_z: 0.0614, task5.loss_whl: 0.1251, task5.loss_yaw: 0.2576, task5.loss_vel: 0.1575, task5.loss_heatmap: 1.1313, loss: 18.2032, grad_norm: 10.1647 2023-03-13 06:39:39,995 - mmdet - INFO - Iter [9450/10536] lr: 2.000e-04, eta: 0:44:23, time: 2.459, data_time: 0.057, memory: 32270, loss_depth: 8.2710, task0.loss_xy: 0.0989, task0.loss_z: 0.0625, task0.loss_whl: 0.0554, task0.loss_yaw: 0.1334, task0.loss_vel: 0.1756, task0.loss_heatmap: 0.9645, task1.loss_xy: 0.1047, task1.loss_z: 0.0784, task1.loss_whl: 0.0991, task1.loss_yaw: 0.1510, task1.loss_vel: 0.1838, task1.loss_heatmap: 1.2762, task2.loss_xy: 0.1067, task2.loss_z: 0.0758, task2.loss_whl: 0.1042, task2.loss_yaw: 0.1560, task2.loss_vel: 0.2424, task2.loss_heatmap: 1.0976, task3.loss_xy: 0.1058, task3.loss_z: 0.0472, task3.loss_whl: 0.1009, task3.loss_yaw: 0.2130, task3.loss_vel: 0.0153, task3.loss_heatmap: 0.8132, task4.loss_xy: 0.0915, task4.loss_z: 0.0497, task4.loss_whl: 0.0938, task4.loss_yaw: 0.2501, task4.loss_vel: 0.1503, task4.loss_heatmap: 0.8374, task5.loss_xy: 0.1074, task5.loss_z: 0.0624, task5.loss_whl: 0.1252, task5.loss_yaw: 0.2635, task5.loss_vel: 0.1376, task5.loss_heatmap: 1.1225, loss: 18.0239, grad_norm: 10.4632 2023-03-13 06:41:41,458 - mmdet - INFO - Iter [9500/10536] lr: 2.000e-04, eta: 0:42:21, time: 2.429, data_time: 0.059, memory: 32270, loss_depth: 8.2529, task0.loss_xy: 0.0988, task0.loss_z: 0.0637, task0.loss_whl: 0.0554, task0.loss_yaw: 0.1326, task0.loss_vel: 0.2450, task0.loss_heatmap: 1.0165, task1.loss_xy: 0.1034, task1.loss_z: 0.0760, task1.loss_whl: 0.0931, task1.loss_yaw: 0.1475, task1.loss_vel: 0.1706, task1.loss_heatmap: 1.2216, task2.loss_xy: 0.1042, task2.loss_z: 0.0706, task2.loss_whl: 0.0931, task2.loss_yaw: 0.1531, task2.loss_vel: 0.2034, task2.loss_heatmap: 0.9833, task3.loss_xy: 0.1012, task3.loss_z: 0.0428, task3.loss_whl: 0.1049, task3.loss_yaw: 0.2463, task3.loss_vel: 0.0215, task3.loss_heatmap: 0.8900, task4.loss_xy: 0.0918, task4.loss_z: 0.0499, task4.loss_whl: 0.0935, task4.loss_yaw: 0.2502, task4.loss_vel: 0.1589, task4.loss_heatmap: 0.6889, task5.loss_xy: 0.1083, task5.loss_z: 0.0604, task5.loss_whl: 0.1246, task5.loss_yaw: 0.2619, task5.loss_vel: 0.1439, task5.loss_heatmap: 1.1145, loss: 17.8384, grad_norm: 10.6633 2023-03-13 06:43:40,601 - mmdet - INFO - Iter [9550/10536] lr: 2.000e-04, eta: 0:40:17, time: 2.383, data_time: 0.058, memory: 32270, loss_depth: 8.4657, task0.loss_xy: 0.0996, task0.loss_z: 0.0665, task0.loss_whl: 0.0557, task0.loss_yaw: 0.1361, task0.loss_vel: 0.2207, task0.loss_heatmap: 1.0346, task1.loss_xy: 0.1032, task1.loss_z: 0.0793, task1.loss_whl: 0.0870, task1.loss_yaw: 0.1513, task1.loss_vel: 0.1417, task1.loss_heatmap: 1.2431, task2.loss_xy: 0.1085, task2.loss_z: 0.0813, task2.loss_whl: 0.0950, task2.loss_yaw: 0.1733, task2.loss_vel: 0.1753, task2.loss_heatmap: 1.0762, task3.loss_xy: 0.1027, task3.loss_z: 0.0486, task3.loss_whl: 0.1266, task3.loss_yaw: 0.2560, task3.loss_vel: 0.0162, task3.loss_heatmap: 0.9191, task4.loss_xy: 0.0946, task4.loss_z: 0.0562, task4.loss_whl: 0.0963, task4.loss_yaw: 0.2451, task4.loss_vel: 0.1841, task4.loss_heatmap: 0.8305, task5.loss_xy: 0.1107, task5.loss_z: 0.0626, task5.loss_whl: 0.1230, task5.loss_yaw: 0.2588, task5.loss_vel: 0.1549, task5.loss_heatmap: 1.1372, loss: 18.4172, grad_norm: 10.5853 2023-03-13 06:45:43,517 - mmdet - INFO - Iter [9600/10536] lr: 2.000e-04, eta: 0:38:15, time: 2.458, data_time: 0.059, memory: 32270, loss_depth: 8.3251, task0.loss_xy: 0.0985, task0.loss_z: 0.0632, task0.loss_whl: 0.0552, task0.loss_yaw: 0.1334, task0.loss_vel: 0.1970, task0.loss_heatmap: 0.9851, task1.loss_xy: 0.1011, task1.loss_z: 0.0767, task1.loss_whl: 0.0926, task1.loss_yaw: 0.1466, task1.loss_vel: 0.1750, task1.loss_heatmap: 1.2348, task2.loss_xy: 0.1052, task2.loss_z: 0.0736, task2.loss_whl: 0.0955, task2.loss_yaw: 0.1444, task2.loss_vel: 0.2099, task2.loss_heatmap: 1.0064, task3.loss_xy: 0.1000, task3.loss_z: 0.0440, task3.loss_whl: 0.1133, task3.loss_yaw: 0.2227, task3.loss_vel: 0.0167, task3.loss_heatmap: 0.7516, task4.loss_xy: 0.0961, task4.loss_z: 0.0561, task4.loss_whl: 0.1066, task4.loss_yaw: 0.2457, task4.loss_vel: 0.1641, task4.loss_heatmap: 0.8912, task5.loss_xy: 0.1091, task5.loss_z: 0.0611, task5.loss_whl: 0.1254, task5.loss_yaw: 0.2607, task5.loss_vel: 0.1576, task5.loss_heatmap: 1.1468, loss: 17.9879, grad_norm: 10.9118 2023-03-13 06:47:46,625 - mmdet - INFO - Iter [9650/10536] lr: 2.000e-04, eta: 0:36:12, time: 2.462, data_time: 0.056, memory: 32270, loss_depth: 8.3220, task0.loss_xy: 0.0979, task0.loss_z: 0.0619, task0.loss_whl: 0.0560, task0.loss_yaw: 0.1324, task0.loss_vel: 0.1894, task0.loss_heatmap: 1.0010, task1.loss_xy: 0.1056, task1.loss_z: 0.0786, task1.loss_whl: 0.0989, task1.loss_yaw: 0.1520, task1.loss_vel: 0.1835, task1.loss_heatmap: 1.3630, task2.loss_xy: 0.1060, task2.loss_z: 0.0740, task2.loss_whl: 0.0899, task2.loss_yaw: 0.1571, task2.loss_vel: 0.1881, task2.loss_heatmap: 1.0116, task3.loss_xy: 0.1003, task3.loss_z: 0.0368, task3.loss_whl: 0.1068, task3.loss_yaw: 0.2032, task3.loss_vel: 0.0152, task3.loss_heatmap: 0.7853, task4.loss_xy: 0.0936, task4.loss_z: 0.0526, task4.loss_whl: 0.0916, task4.loss_yaw: 0.2733, task4.loss_vel: 0.1525, task4.loss_heatmap: 0.8293, task5.loss_xy: 0.1070, task5.loss_z: 0.0559, task5.loss_whl: 0.1358, task5.loss_yaw: 0.2599, task5.loss_vel: 0.1532, task5.loss_heatmap: 1.1265, loss: 18.0476, grad_norm: 10.4410 2023-03-13 06:49:50,315 - mmdet - INFO - Iter [9700/10536] lr: 2.000e-04, eta: 0:34:10, time: 2.474, data_time: 0.059, memory: 32270, loss_depth: 8.2092, task0.loss_xy: 0.0983, task0.loss_z: 0.0604, task0.loss_whl: 0.0567, task0.loss_yaw: 0.1298, task0.loss_vel: 0.1880, task0.loss_heatmap: 0.9945, task1.loss_xy: 0.1055, task1.loss_z: 0.0758, task1.loss_whl: 0.0987, task1.loss_yaw: 0.1488, task1.loss_vel: 0.1692, task1.loss_heatmap: 1.3100, task2.loss_xy: 0.1088, task2.loss_z: 0.0781, task2.loss_whl: 0.0917, task2.loss_yaw: 0.1704, task2.loss_vel: 0.2451, task2.loss_heatmap: 1.1757, task3.loss_xy: 0.0981, task3.loss_z: 0.0423, task3.loss_whl: 0.0998, task3.loss_yaw: 0.2049, task3.loss_vel: 0.0158, task3.loss_heatmap: 0.7003, task4.loss_xy: 0.0944, task4.loss_z: 0.0542, task4.loss_whl: 0.1071, task4.loss_yaw: 0.2487, task4.loss_vel: 0.1521, task4.loss_heatmap: 0.7834, task5.loss_xy: 0.1094, task5.loss_z: 0.0590, task5.loss_whl: 0.1241, task5.loss_yaw: 0.2524, task5.loss_vel: 0.1652, task5.loss_heatmap: 1.1991, loss: 18.0250, grad_norm: 10.2922 2023-03-13 06:51:50,968 - mmdet - INFO - Iter [9750/10536] lr: 2.000e-04, eta: 0:32:07, time: 2.413, data_time: 0.057, memory: 32270, loss_depth: 8.2306, task0.loss_xy: 0.0987, task0.loss_z: 0.0646, task0.loss_whl: 0.0551, task0.loss_yaw: 0.1318, task0.loss_vel: 0.2065, task0.loss_heatmap: 0.9960, task1.loss_xy: 0.1009, task1.loss_z: 0.0749, task1.loss_whl: 0.0933, task1.loss_yaw: 0.1472, task1.loss_vel: 0.1938, task1.loss_heatmap: 1.1846, task2.loss_xy: 0.1021, task2.loss_z: 0.0653, task2.loss_whl: 0.0876, task2.loss_yaw: 0.1283, task2.loss_vel: 0.2698, task2.loss_heatmap: 0.9701, task3.loss_xy: 0.1017, task3.loss_z: 0.0441, task3.loss_whl: 0.0936, task3.loss_yaw: 0.2468, task3.loss_vel: 0.0175, task3.loss_heatmap: 0.7123, task4.loss_xy: 0.0943, task4.loss_z: 0.0492, task4.loss_whl: 0.0935, task4.loss_yaw: 0.2575, task4.loss_vel: 0.1743, task4.loss_heatmap: 0.7715, task5.loss_xy: 0.1088, task5.loss_z: 0.0593, task5.loss_whl: 0.1231, task5.loss_yaw: 0.2601, task5.loss_vel: 0.1515, task5.loss_heatmap: 1.1008, loss: 17.6610, grad_norm: 11.2878 2023-03-13 06:53:52,270 - mmdet - INFO - Iter [9800/10536] lr: 2.000e-04, eta: 0:30:04, time: 2.426, data_time: 0.057, memory: 32270, loss_depth: 8.2476, task0.loss_xy: 0.0984, task0.loss_z: 0.0621, task0.loss_whl: 0.0550, task0.loss_yaw: 0.1331, task0.loss_vel: 0.1938, task0.loss_heatmap: 1.0074, task1.loss_xy: 0.1006, task1.loss_z: 0.0711, task1.loss_whl: 0.0877, task1.loss_yaw: 0.1322, task1.loss_vel: 0.1728, task1.loss_heatmap: 1.1930, task2.loss_xy: 0.1013, task2.loss_z: 0.0718, task2.loss_whl: 0.0857, task2.loss_yaw: 0.1243, task2.loss_vel: 0.2162, task2.loss_heatmap: 0.8894, task3.loss_xy: 0.1022, task3.loss_z: 0.0432, task3.loss_whl: 0.1036, task3.loss_yaw: 0.2333, task3.loss_vel: 0.0139, task3.loss_heatmap: 0.8164, task4.loss_xy: 0.0912, task4.loss_z: 0.0474, task4.loss_whl: 0.0821, task4.loss_yaw: 0.2431, task4.loss_vel: 0.1689, task4.loss_heatmap: 0.7520, task5.loss_xy: 0.1087, task5.loss_z: 0.0581, task5.loss_whl: 0.1247, task5.loss_yaw: 0.2550, task5.loss_vel: 0.1438, task5.loss_heatmap: 1.1196, loss: 17.5509, grad_norm: 12.4251 2023-03-13 06:55:54,260 - mmdet - INFO - Iter [9850/10536] lr: 2.000e-04, eta: 0:28:02, time: 2.440, data_time: 0.057, memory: 32270, loss_depth: 8.2526, task0.loss_xy: 0.0974, task0.loss_z: 0.0609, task0.loss_whl: 0.0556, task0.loss_yaw: 0.1284, task0.loss_vel: 0.2091, task0.loss_heatmap: 0.9595, task1.loss_xy: 0.1045, task1.loss_z: 0.0729, task1.loss_whl: 0.0915, task1.loss_yaw: 0.1552, task1.loss_vel: 0.1510, task1.loss_heatmap: 1.2774, task2.loss_xy: 0.1037, task2.loss_z: 0.0800, task2.loss_whl: 0.1010, task2.loss_yaw: 0.1668, task2.loss_vel: 0.2172, task2.loss_heatmap: 1.0008, task3.loss_xy: 0.0969, task3.loss_z: 0.0410, task3.loss_whl: 0.1066, task3.loss_yaw: 0.1924, task3.loss_vel: 0.0166, task3.loss_heatmap: 0.8795, task4.loss_xy: 0.0933, task4.loss_z: 0.0500, task4.loss_whl: 0.0848, task4.loss_yaw: 0.2336, task4.loss_vel: 0.1882, task4.loss_heatmap: 0.7395, task5.loss_xy: 0.1107, task5.loss_z: 0.0598, task5.loss_whl: 0.1261, task5.loss_yaw: 0.2592, task5.loss_vel: 0.1603, task5.loss_heatmap: 1.1288, loss: 17.8528, grad_norm: 10.8005 2023-03-13 06:58:01,108 - mmdet - INFO - Iter [9900/10536] lr: 2.000e-04, eta: 0:25:59, time: 2.537, data_time: 0.058, memory: 32270, loss_depth: 8.2665, task0.loss_xy: 0.0984, task0.loss_z: 0.0618, task0.loss_whl: 0.0549, task0.loss_yaw: 0.1268, task0.loss_vel: 0.2319, task0.loss_heatmap: 0.9888, task1.loss_xy: 0.1017, task1.loss_z: 0.0771, task1.loss_whl: 0.0948, task1.loss_yaw: 0.1523, task1.loss_vel: 0.1736, task1.loss_heatmap: 1.2274, task2.loss_xy: 0.1073, task2.loss_z: 0.0751, task2.loss_whl: 0.1022, task2.loss_yaw: 0.1618, task2.loss_vel: 0.2562, task2.loss_heatmap: 1.0948, task3.loss_xy: 0.1021, task3.loss_z: 0.0477, task3.loss_whl: 0.1117, task3.loss_yaw: 0.1870, task3.loss_vel: 0.0167, task3.loss_heatmap: 0.7548, task4.loss_xy: 0.0934, task4.loss_z: 0.0541, task4.loss_whl: 0.0929, task4.loss_yaw: 0.2353, task4.loss_vel: 0.2228, task4.loss_heatmap: 0.8864, task5.loss_xy: 0.1086, task5.loss_z: 0.0616, task5.loss_whl: 0.1211, task5.loss_yaw: 0.2547, task5.loss_vel: 0.1480, task5.loss_heatmap: 1.0807, loss: 18.0326, grad_norm: 10.8950 2023-03-13 07:00:03,221 - mmdet - INFO - Iter [9950/10536] lr: 2.000e-04, eta: 0:23:57, time: 2.442, data_time: 0.060, memory: 32270, loss_depth: 8.3504, task0.loss_xy: 0.0982, task0.loss_z: 0.0598, task0.loss_whl: 0.0545, task0.loss_yaw: 0.1286, task0.loss_vel: 0.1946, task0.loss_heatmap: 0.9964, task1.loss_xy: 0.1025, task1.loss_z: 0.0746, task1.loss_whl: 0.0909, task1.loss_yaw: 0.1465, task1.loss_vel: 0.1482, task1.loss_heatmap: 1.1964, task2.loss_xy: 0.1059, task2.loss_z: 0.0666, task2.loss_whl: 0.0877, task2.loss_yaw: 0.1725, task2.loss_vel: 0.1576, task2.loss_heatmap: 1.0397, task3.loss_xy: 0.0998, task3.loss_z: 0.0429, task3.loss_whl: 0.1174, task3.loss_yaw: 0.2090, task3.loss_vel: 0.0160, task3.loss_heatmap: 0.7672, task4.loss_xy: 0.0924, task4.loss_z: 0.0529, task4.loss_whl: 0.1006, task4.loss_yaw: 0.2312, task4.loss_vel: 0.1566, task4.loss_heatmap: 0.8240, task5.loss_xy: 0.1072, task5.loss_z: 0.0567, task5.loss_whl: 0.1327, task5.loss_yaw: 0.2584, task5.loss_vel: 0.1382, task5.loss_heatmap: 1.1107, loss: 17.7854, grad_norm: 10.7283 2023-03-13 07:02:07,765 - mmdet - INFO - Exp name: r50-fp16_phase2.py 2023-03-13 07:02:07,766 - mmdet - INFO - Iter [10000/10536] lr: 2.000e-04, eta: 0:21:54, time: 2.491, data_time: 0.060, memory: 32270, loss_depth: 8.3513, task0.loss_xy: 0.0990, task0.loss_z: 0.0664, task0.loss_whl: 0.0548, task0.loss_yaw: 0.1307, task0.loss_vel: 0.1980, task0.loss_heatmap: 0.9896, task1.loss_xy: 0.1046, task1.loss_z: 0.0739, task1.loss_whl: 0.0969, task1.loss_yaw: 0.1459, task1.loss_vel: 0.1670, task1.loss_heatmap: 1.2510, task2.loss_xy: 0.1052, task2.loss_z: 0.0715, task2.loss_whl: 0.0913, task2.loss_yaw: 0.1549, task2.loss_vel: 0.1619, task2.loss_heatmap: 0.9668, task3.loss_xy: 0.0994, task3.loss_z: 0.0464, task3.loss_whl: 0.1024, task3.loss_yaw: 0.2447, task3.loss_vel: 0.0164, task3.loss_heatmap: 0.7304, task4.loss_xy: 0.0910, task4.loss_z: 0.0515, task4.loss_whl: 0.0968, task4.loss_yaw: 0.2464, task4.loss_vel: 0.2196, task4.loss_heatmap: 0.7918, task5.loss_xy: 0.1082, task5.loss_z: 0.0601, task5.loss_whl: 0.1289, task5.loss_yaw: 0.2576, task5.loss_vel: 0.1373, task5.loss_heatmap: 1.0909, loss: 17.8006, grad_norm: 10.2323 2023-03-13 07:04:10,233 - mmdet - INFO - Iter [10050/10536] lr: 2.000e-04, eta: 0:19:52, time: 2.449, data_time: 0.058, memory: 32270, loss_depth: 8.2495, task0.loss_xy: 0.0975, task0.loss_z: 0.0618, task0.loss_whl: 0.0559, task0.loss_yaw: 0.1196, task0.loss_vel: 0.2096, task0.loss_heatmap: 0.9699, task1.loss_xy: 0.1021, task1.loss_z: 0.0719, task1.loss_whl: 0.0889, task1.loss_yaw: 0.1405, task1.loss_vel: 0.1849, task1.loss_heatmap: 1.2316, task2.loss_xy: 0.1038, task2.loss_z: 0.0672, task2.loss_whl: 0.0895, task2.loss_yaw: 0.1497, task2.loss_vel: 0.1919, task2.loss_heatmap: 0.9400, task3.loss_xy: 0.1008, task3.loss_z: 0.0432, task3.loss_whl: 0.1188, task3.loss_yaw: 0.2372, task3.loss_vel: 0.0189, task3.loss_heatmap: 0.7056, task4.loss_xy: 0.0855, task4.loss_z: 0.0491, task4.loss_whl: 0.1011, task4.loss_yaw: 0.2378, task4.loss_vel: 0.1428, task4.loss_heatmap: 0.6784, task5.loss_xy: 0.1067, task5.loss_z: 0.0598, task5.loss_whl: 0.1270, task5.loss_yaw: 0.2606, task5.loss_vel: 0.1307, task5.loss_heatmap: 1.1287, loss: 17.4583, grad_norm: 11.3201 2023-03-13 07:06:10,137 - mmdet - INFO - Iter [10100/10536] lr: 2.000e-04, eta: 0:17:49, time: 2.398, data_time: 0.057, memory: 32270, loss_depth: 8.4525, task0.loss_xy: 0.0976, task0.loss_z: 0.0629, task0.loss_whl: 0.0539, task0.loss_yaw: 0.1255, task0.loss_vel: 0.2077, task0.loss_heatmap: 0.9691, task1.loss_xy: 0.1044, task1.loss_z: 0.0799, task1.loss_whl: 0.0922, task1.loss_yaw: 0.1428, task1.loss_vel: 0.1635, task1.loss_heatmap: 1.3098, task2.loss_xy: 0.1070, task2.loss_z: 0.0726, task2.loss_whl: 0.0854, task2.loss_yaw: 0.1459, task2.loss_vel: 0.1755, task2.loss_heatmap: 1.0174, task3.loss_xy: 0.1005, task3.loss_z: 0.0461, task3.loss_whl: 0.1112, task3.loss_yaw: 0.2342, task3.loss_vel: 0.0161, task3.loss_heatmap: 0.7587, task4.loss_xy: 0.0901, task4.loss_z: 0.0593, task4.loss_whl: 0.1091, task4.loss_yaw: 0.2342, task4.loss_vel: 0.1659, task4.loss_heatmap: 0.7491, task5.loss_xy: 0.1076, task5.loss_z: 0.0600, task5.loss_whl: 0.1271, task5.loss_yaw: 0.2546, task5.loss_vel: 0.1496, task5.loss_heatmap: 1.1789, loss: 18.0180, grad_norm: 11.0986 2023-03-13 07:08:12,502 - mmdet - INFO - Iter [10150/10536] lr: 2.000e-04, eta: 0:15:46, time: 2.447, data_time: 0.059, memory: 32270, loss_depth: 8.2719, task0.loss_xy: 0.0992, task0.loss_z: 0.0645, task0.loss_whl: 0.0534, task0.loss_yaw: 0.1242, task0.loss_vel: 0.2243, task0.loss_heatmap: 1.0156, task1.loss_xy: 0.1065, task1.loss_z: 0.0752, task1.loss_whl: 0.0895, task1.loss_yaw: 0.1496, task1.loss_vel: 0.1626, task1.loss_heatmap: 1.2976, task2.loss_xy: 0.1066, task2.loss_z: 0.0696, task2.loss_whl: 0.0939, task2.loss_yaw: 0.1584, task2.loss_vel: 0.1561, task2.loss_heatmap: 1.0760, task3.loss_xy: 0.0998, task3.loss_z: 0.0433, task3.loss_whl: 0.1049, task3.loss_yaw: 0.2408, task3.loss_vel: 0.0195, task3.loss_heatmap: 0.7068, task4.loss_xy: 0.0918, task4.loss_z: 0.0484, task4.loss_whl: 0.1030, task4.loss_yaw: 0.2204, task4.loss_vel: 0.2625, task4.loss_heatmap: 0.8653, task5.loss_xy: 0.1078, task5.loss_z: 0.0621, task5.loss_whl: 0.1256, task5.loss_yaw: 0.2585, task5.loss_vel: 0.1512, task5.loss_heatmap: 1.1492, loss: 18.0558, grad_norm: 11.5615 2023-03-13 07:10:12,264 - mmdet - INFO - Iter [10200/10536] lr: 2.000e-04, eta: 0:13:43, time: 2.395, data_time: 0.058, memory: 32270, loss_depth: 8.3283, task0.loss_xy: 0.0959, task0.loss_z: 0.0624, task0.loss_whl: 0.0546, task0.loss_yaw: 0.1233, task0.loss_vel: 0.2065, task0.loss_heatmap: 0.9506, task1.loss_xy: 0.1036, task1.loss_z: 0.0811, task1.loss_whl: 0.1000, task1.loss_yaw: 0.1542, task1.loss_vel: 0.1786, task1.loss_heatmap: 1.3103, task2.loss_xy: 0.1090, task2.loss_z: 0.0753, task2.loss_whl: 0.0970, task2.loss_yaw: 0.1617, task2.loss_vel: 0.2442, task2.loss_heatmap: 1.1162, task3.loss_xy: 0.1002, task3.loss_z: 0.0427, task3.loss_whl: 0.1073, task3.loss_yaw: 0.2406, task3.loss_vel: 0.0175, task3.loss_heatmap: 0.7431, task4.loss_xy: 0.0901, task4.loss_z: 0.0515, task4.loss_whl: 0.1060, task4.loss_yaw: 0.2592, task4.loss_vel: 0.1652, task4.loss_heatmap: 0.7271, task5.loss_xy: 0.1068, task5.loss_z: 0.0597, task5.loss_whl: 0.1275, task5.loss_yaw: 0.2562, task5.loss_vel: 0.1439, task5.loss_heatmap: 1.1572, loss: 18.0541, grad_norm: 11.1162 2023-03-13 07:12:13,716 - mmdet - INFO - Iter [10250/10536] lr: 2.000e-04, eta: 0:11:41, time: 2.429, data_time: 0.058, memory: 32270, loss_depth: 8.1659, task0.loss_xy: 0.0975, task0.loss_z: 0.0601, task0.loss_whl: 0.0546, task0.loss_yaw: 0.1270, task0.loss_vel: 0.2048, task0.loss_heatmap: 0.9776, task1.loss_xy: 0.1029, task1.loss_z: 0.0753, task1.loss_whl: 0.0926, task1.loss_yaw: 0.1332, task1.loss_vel: 0.1579, task1.loss_heatmap: 1.1900, task2.loss_xy: 0.1057, task2.loss_z: 0.0713, task2.loss_whl: 0.0879, task2.loss_yaw: 0.1483, task2.loss_vel: 0.1480, task2.loss_heatmap: 1.0341, task3.loss_xy: 0.1017, task3.loss_z: 0.0423, task3.loss_whl: 0.1100, task3.loss_yaw: 0.2314, task3.loss_vel: 0.0181, task3.loss_heatmap: 0.7266, task4.loss_xy: 0.0864, task4.loss_z: 0.0466, task4.loss_whl: 0.0895, task4.loss_yaw: 0.2376, task4.loss_vel: 0.1706, task4.loss_heatmap: 0.7554, task5.loss_xy: 0.1092, task5.loss_z: 0.0608, task5.loss_whl: 0.1270, task5.loss_yaw: 0.2607, task5.loss_vel: 0.1505, task5.loss_heatmap: 1.1586, loss: 17.5176, grad_norm: 9.7149 2023-03-13 07:14:15,965 - mmdet - INFO - Iter [10300/10536] lr: 2.000e-04, eta: 0:09:38, time: 2.445, data_time: 0.060, memory: 32270, loss_depth: 8.3178, task0.loss_xy: 0.0993, task0.loss_z: 0.0614, task0.loss_whl: 0.0549, task0.loss_yaw: 0.1242, task0.loss_vel: 0.2189, task0.loss_heatmap: 0.9931, task1.loss_xy: 0.0999, task1.loss_z: 0.0711, task1.loss_whl: 0.0976, task1.loss_yaw: 0.1457, task1.loss_vel: 0.1816, task1.loss_heatmap: 1.1437, task2.loss_xy: 0.1041, task2.loss_z: 0.0770, task2.loss_whl: 0.0859, task2.loss_yaw: 0.1258, task2.loss_vel: 0.2470, task2.loss_heatmap: 0.9481, task3.loss_xy: 0.1006, task3.loss_z: 0.0420, task3.loss_whl: 0.1014, task3.loss_yaw: 0.2198, task3.loss_vel: 0.0165, task3.loss_heatmap: 0.7197, task4.loss_xy: 0.0924, task4.loss_z: 0.0550, task4.loss_whl: 0.0969, task4.loss_yaw: 0.2297, task4.loss_vel: 0.1953, task4.loss_heatmap: 0.7984, task5.loss_xy: 0.1082, task5.loss_z: 0.0560, task5.loss_whl: 0.1195, task5.loss_yaw: 0.2579, task5.loss_vel: 0.1402, task5.loss_heatmap: 1.0825, loss: 17.6292, grad_norm: 9.8028 2023-03-13 07:16:16,591 - mmdet - INFO - Iter [10350/10536] lr: 2.000e-04, eta: 0:07:35, time: 2.413, data_time: 0.058, memory: 32270, loss_depth: 8.2784, task0.loss_xy: 0.0973, task0.loss_z: 0.0607, task0.loss_whl: 0.0565, task0.loss_yaw: 0.1205, task0.loss_vel: 0.1995, task0.loss_heatmap: 0.9589, task1.loss_xy: 0.1036, task1.loss_z: 0.0745, task1.loss_whl: 0.0843, task1.loss_yaw: 0.1334, task1.loss_vel: 0.1545, task1.loss_heatmap: 1.2420, task2.loss_xy: 0.1019, task2.loss_z: 0.0666, task2.loss_whl: 0.0855, task2.loss_yaw: 0.1258, task2.loss_vel: 0.1945, task2.loss_heatmap: 0.8993, task3.loss_xy: 0.1026, task3.loss_z: 0.0425, task3.loss_whl: 0.1261, task3.loss_yaw: 0.2129, task3.loss_vel: 0.0222, task3.loss_heatmap: 0.7727, task4.loss_xy: 0.0929, task4.loss_z: 0.0488, task4.loss_whl: 0.0896, task4.loss_yaw: 0.2368, task4.loss_vel: 0.1976, task4.loss_heatmap: 0.7773, task5.loss_xy: 0.1100, task5.loss_z: 0.0633, task5.loss_whl: 0.1237, task5.loss_yaw: 0.2555, task5.loss_vel: 0.1687, task5.loss_heatmap: 1.1337, loss: 17.6150, grad_norm: 10.6651 2023-03-13 07:18:20,254 - mmdet - INFO - Iter [10400/10536] lr: 2.000e-04, eta: 0:05:33, time: 2.473, data_time: 0.061, memory: 32270, loss_depth: 8.2263, task0.loss_xy: 0.0984, task0.loss_z: 0.0610, task0.loss_whl: 0.0565, task0.loss_yaw: 0.1292, task0.loss_vel: 0.1878, task0.loss_heatmap: 0.9867, task1.loss_xy: 0.1039, task1.loss_z: 0.0737, task1.loss_whl: 0.0971, task1.loss_yaw: 0.1412, task1.loss_vel: 0.1642, task1.loss_heatmap: 1.2391, task2.loss_xy: 0.1044, task2.loss_z: 0.0758, task2.loss_whl: 0.0921, task2.loss_yaw: 0.1364, task2.loss_vel: 0.2191, task2.loss_heatmap: 0.9672, task3.loss_xy: 0.1006, task3.loss_z: 0.0450, task3.loss_whl: 0.1017, task3.loss_yaw: 0.1896, task3.loss_vel: 0.0163, task3.loss_heatmap: 0.7748, task4.loss_xy: 0.0935, task4.loss_z: 0.0566, task4.loss_whl: 0.0914, task4.loss_yaw: 0.2516, task4.loss_vel: 0.1491, task4.loss_heatmap: 0.8303, task5.loss_xy: 0.1081, task5.loss_z: 0.0602, task5.loss_whl: 0.1241, task5.loss_yaw: 0.2558, task5.loss_vel: 0.1500, task5.loss_heatmap: 1.1254, loss: 17.6844, grad_norm: 11.2891 2023-03-13 07:20:23,283 - mmdet - INFO - Iter [10450/10536] lr: 2.000e-04, eta: 0:03:30, time: 2.461, data_time: 0.059, memory: 32270, loss_depth: 8.4010, task0.loss_xy: 0.0995, task0.loss_z: 0.0648, task0.loss_whl: 0.0574, task0.loss_yaw: 0.1342, task0.loss_vel: 0.1791, task0.loss_heatmap: 1.0015, task1.loss_xy: 0.1014, task1.loss_z: 0.0763, task1.loss_whl: 0.0888, task1.loss_yaw: 0.1384, task1.loss_vel: 0.1787, task1.loss_heatmap: 1.2437, task2.loss_xy: 0.1067, task2.loss_z: 0.0799, task2.loss_whl: 0.0941, task2.loss_yaw: 0.1481, task2.loss_vel: 0.2821, task2.loss_heatmap: 1.0732, task3.loss_xy: 0.0989, task3.loss_z: 0.0490, task3.loss_whl: 0.1047, task3.loss_yaw: 0.2093, task3.loss_vel: 0.0195, task3.loss_heatmap: 0.8988, task4.loss_xy: 0.0917, task4.loss_z: 0.0532, task4.loss_whl: 0.0996, task4.loss_yaw: 0.2575, task4.loss_vel: 0.1202, task4.loss_heatmap: 0.7086, task5.loss_xy: 0.1059, task5.loss_z: 0.0592, task5.loss_whl: 0.1311, task5.loss_yaw: 0.2607, task5.loss_vel: 0.1320, task5.loss_heatmap: 1.1382, loss: 18.0869, grad_norm: 12.9059 2023-03-13 07:22:27,107 - mmdet - INFO - Iter [10500/10536] lr: 2.000e-04, eta: 0:01:28, time: 2.476, data_time: 0.061, memory: 32270, loss_depth: 8.2086, task0.loss_xy: 0.0973, task0.loss_z: 0.0613, task0.loss_whl: 0.0554, task0.loss_yaw: 0.1254, task0.loss_vel: 0.1954, task0.loss_heatmap: 0.9592, task1.loss_xy: 0.1032, task1.loss_z: 0.0744, task1.loss_whl: 0.0975, task1.loss_yaw: 0.1445, task1.loss_vel: 0.1631, task1.loss_heatmap: 1.2213, task2.loss_xy: 0.1062, task2.loss_z: 0.0680, task2.loss_whl: 0.0903, task2.loss_yaw: 0.1410, task2.loss_vel: 0.2261, task2.loss_heatmap: 0.9939, task3.loss_xy: 0.0992, task3.loss_z: 0.0424, task3.loss_whl: 0.1141, task3.loss_yaw: 0.1817, task3.loss_vel: 0.0147, task3.loss_heatmap: 0.8159, task4.loss_xy: 0.0908, task4.loss_z: 0.0494, task4.loss_whl: 0.0976, task4.loss_yaw: 0.2351, task4.loss_vel: 0.1446, task4.loss_heatmap: 0.7452, task5.loss_xy: 0.1094, task5.loss_z: 0.0565, task5.loss_whl: 0.1209, task5.loss_yaw: 0.2549, task5.loss_vel: 0.1530, task5.loss_heatmap: 1.1166, loss: 17.5741, grad_norm: 11.1496 2023-03-13 07:23:56,168 - mmdet - INFO - Saving checkpoint at 10536 iterations