sense-x / hop Goto Github PK

View Code? Open in Web Editor NEW

174.0 8.0 13.0 3.92 MB

[ICCV 2023] Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction

License: Apache License 2.0

Python 98.77% Shell 0.68% Dockerfile 0.15% C++ 0.11% Cuda 0.14% MATLAB 0.15%

hop's People

Stargazers

Watchers

Forkers

hulaifeng youyuehanxiao naurril gaowei724 zhearing houxin-j wang-jh18-svm tschunlintz ioevan qshuiqing moordo91 hirotoppi whuhxb

hop's Issues

训练中得到0map

代码没有做过更改，配置文件只把samples_per_gpu从8变成16.
depth loss 一直是上升的，lr没有变过, 其他的loss不下降

log如下：

{"mmdet_version": "2.24.0", "mmseg_version": "0.24.0", "mmdet3d_version": "1.0.0rc4", "config": "point_cloud_range = [-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]\nclass_names = [\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier',\n 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n]\ndataset_type = 'NuScenesDataset'\ndata_root = 'data/nuscenes/'\ninput_modality = dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False)\nfile_client_args = dict(backend='disk')\ntrain_pipeline = [\n dict(\n type='PrepareImageInputs',\n is_train=True,\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n align_adj_bbox=True,\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ]),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='PointToMultiViewDepth',\n downsample=1,\n grid_config=dict(\n x=[-51.2, 51.2, 0.8],\n y=[-51.2, 51.2, 0.8],\n z=[-5, 3, 8],\n depth=[1.0, 60.0, 0.5])),\n dict(\n type='ObjectRangeFilter',\n point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]),\n dict(\n type='ObjectNameFilter',\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ]),\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ]),\n dict(\n type='Collect3D',\n keys=['img_inputs', 'gt_bboxes_3d', 'gt_labels_3d', 'gt_depth'])\n]\ntest_pipeline = [\n dict(\n type='PrepareImageInputs',\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n align_adj_bbox=True,\n is_train=False),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='MultiScaleFlipAug3D',\n img_scale=(1333, 800),\n pts_scale_ratio=1,\n flip=False,\n transforms=[\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points', 'img_inputs'])\n ])\n]\neval_pipeline = [\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadPointsFromMultiSweeps',\n sweeps_num=10,\n file_client_args=dict(backend='disk')),\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'trailer', 'bus', 'construction_vehicle',\n 'bicycle', 'motorcycle', 'pedestrian', 'traffic_cone', 'barrier'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points'])\n]\ndata = dict(\n samples_per_gpu=16,\n workers_per_gpu=4,\n train=dict(\n type='NuScenesDataset',\n data_root='data/nuscenes/',\n ann_file='data/nuscenes/bevdetv2-nuscenes_infos_train.pkl',\n pipeline=[\n dict(\n type='PrepareImageInputs',\n is_train=True,\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n align_adj_bbox=True,\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ]),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='PointToMultiViewDepth',\n downsample=1,\n grid_config=dict(\n x=[-51.2, 51.2, 0.8],\n y=[-51.2, 51.2, 0.8],\n z=[-5, 3, 8],\n depth=[1.0, 60.0, 0.5])),\n dict(\n type='ObjectRangeFilter',\n point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]),\n dict(\n type='ObjectNameFilter',\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ]),\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ]),\n dict(\n type='Collect3D',\n keys=[\n 'img_inputs', 'gt_bboxes_3d', 'gt_labels_3d', 'gt_depth'\n ])\n ],\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n modality=dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False),\n test_mode=False,\n box_type_3d='LiDAR',\n use_valid_flag=True,\n img_info_prototype='bevdet4d',\n multi_adj_frame_id_cfg=(1, 9, 1)),\n val=dict(\n type='NuScenesDataset',\n data_root='data/nuscenes/',\n ann_file='data/nuscenes/bevdetv2-nuscenes_infos_val.pkl',\n pipeline=[\n dict(\n type='PrepareImageInputs',\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ],\n align_adj_bbox=True,\n is_train=False),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='MultiScaleFlipAug3D',\n img_scale=(1333, 800),\n pts_scale_ratio=1,\n flip=False,\n transforms=[\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus',\n 'trailer', 'barrier', 'motorcycle', 'bicycle',\n 'pedestrian', 'traffic_cone'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points', 'img_inputs'])\n ])\n ],\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n modality=dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False),\n test_mode=True,\n box_type_3d='LiDAR',\n img_info_prototype='bevdet4d',\n multi_adj_frame_id_cfg=(1, 9, 1)),\n test=dict(\n type='NuScenesDataset',\n data_root='data/nuscenes/',\n ann_file='data/nuscenes/bevdetv2-nuscenes_infos_val.pkl',\n pipeline=[\n dict(\n type='PrepareImageInputs',\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ],\n align_adj_bbox=True,\n is_train=False),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='MultiScaleFlipAug3D',\n img_scale=(1333, 800),\n pts_scale_ratio=1,\n flip=False,\n transforms=[\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus',\n 'trailer', 'barrier', 'motorcycle', 'bicycle',\n 'pedestrian', 'traffic_cone'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points', 'img_inputs'])\n ])\n ],\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n modality=dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False),\n test_mode=True,\n box_type_3d='LiDAR',\n img_info_prototype='bevdet4d',\n multi_adj_frame_id_cfg=(1, 9, 1)))\nevaluation = dict(\n interval=1,\n pipeline=[\n dict(\n type='PrepareImageInputs',\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ],\n align_adj_bbox=True,\n is_train=False),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='MultiScaleFlipAug3D',\n img_scale=(1333, 800),\n pts_scale_ratio=1,\n flip=False,\n transforms=[\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus',\n 'trailer', 'barrier', 'motorcycle', 'bicycle',\n 'pedestrian', 'traffic_cone'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points', 'img_inputs'])\n ])\n ])\ncheckpoint_config = dict(interval=3)\nlog_config = dict(\n interval=50,\n hooks=[dict(type='TextLoggerHook'),\n dict(type='TensorboardLoggerHook')])\ndist_params = dict(backend='nccl')\nlog_level = 'INFO'\nwork_dir = './work_dirs/hop_bevdet4d-r50-depth'\nload_from = None\nresume_from = None\nworkflow = [('train', 1)]\nopencv_num_threads = 0\nmp_start_method = 'fork'\nplugin = True\nplugin_dir = 'mmdet3d_plugin/'\ndata_config = dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT', 'CAM_BACK_LEFT',\n 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0)\ngrid_config = dict(\n x=[-51.2, 51.2, 0.8],\n y=[-51.2, 51.2, 0.8],\n z=[-5, 3, 8],\n depth=[1.0, 60.0, 0.5])\nvoxel_size = [0.1, 0.1, 0.2]\nnumC_Trans = 80\nmulti_adj_frame_id_cfg = (1, 9, 1)\nmodel = dict(\n type='HoPBEVDepth4D',\n align_after_view_transfromation=False,\n num_adj=8,\n with_prev=True,\n with_hop=True,\n img_backbone=dict(\n pretrained='torchvision://resnet50',\n type='ResNet',\n depth=50,\n num_stages=4,\n out_indices=(2, 3),\n frozen_stages=-1,\n norm_cfg=dict(type='BN', requires_grad=True),\n norm_eval=False,\n with_cp=True,\n style='pytorch'),\n img_neck=dict(\n type='CustomFPN',\n in_channels=[1024, 2048],\n out_channels=512,\n num_outs=1,\n start_level=0,\n out_ids=[0]),\n img_view_transformer=dict(\n type='LSSViewTransformerBEVDepth',\n grid_config=dict(\n x=[-51.2, 51.2, 0.8],\n y=[-51.2, 51.2, 0.8],\n z=[-5, 3, 8],\n depth=[1.0, 60.0, 0.5]),\n input_size=(256, 704),\n in_channels=512,\n out_channels=80,\n depthnet_cfg=dict(use_dcn=False),\n downsample=16),\n history_decoder=dict(\n type='BiTemporalPredictor',\n in_channels=80,\n out_channels=256,\n embed_dims=160,\n num_adj=7,\n reduction=4,\n bev_h=128,\n bev_w=128,\n decoder_short=dict(\n type='TemporalDecoder',\n num_layers=2,\n transformerlayers=dict(\n type='BEVFormerLayer',\n attn_cfgs=[\n dict(\n type='TemporalCrossAttention',\n embed_dims=160,\n num_heads=5,\n num_levels=1,\n num_bev_queue=2,\n dropout=0.0)\n ],\n ffn_cfgs=dict(\n type='FFN',\n embed_dims=160,\n feedforward_channels=512,\n num_fcs=2,\n ffn_drop=0.0,\n act_cfg=dict(type='ReLU', inplace=True)),\n feedforward_channels=512,\n ffn_dropout=0.0,\n operation_order=('self_attn', 'norm', 'ffn', 'norm'))),\n decoder_long=dict(\n type='TemporalDecoder',\n num_layers=2,\n transformerlayers=dict(\n type='BEVFormerLayer',\n attn_cfgs=[\n dict(\n type='TemporalCrossAttention',\n embed_dims=40,\n num_heads=2,\n num_levels=1,\n num_bev_queue=8,\n dropout=0.0)\n ],\n ffn_cfgs=dict(\n type='FFN',\n embed_dims=40,\n feedforward_channels=128,\n num_fcs=2,\n ffn_drop=0.0,\n act_cfg=dict(type='ReLU', inplace=True)),\n feedforward_channels=128,\n ffn_dropout=0.0,\n operation_order=('self_attn', 'norm', 'ffn', 'norm')))),\n img_bev_encoder_backbone=dict(\n type='CustomResNet', numC_input=720, num_channels=[160, 320, 640]),\n img_bev_encoder_neck=dict(\n type='FPN_LSS', in_channels=800, out_channels=256),\n pre_process=dict(\n type='CustomResNet',\n numC_input=80,\n num_layer=[2],\n num_channels=[80],\n stride=[1],\n backbone_output_ids=[0]),\n pts_bbox_head=dict(\n type='CenterHead',\n in_channels=256,\n tasks=[\n dict(num_class=1, class_names=['car']),\n dict(num_class=2, class_names=['truck', 'construction_vehicle']),\n dict(num_class=2, class_names=['bus', 'trailer']),\n dict(num_class=1, class_names=['barrier']),\n dict(num_class=2, class_names=['motorcycle', 'bicycle']),\n dict(num_class=2, class_names=['pedestrian', 'traffic_cone'])\n ],\n common_heads=dict(\n reg=(2, 2), height=(1, 2), dim=(3, 2), rot=(2, 2), vel=(2, 2)),\n share_conv_channel=64,\n bbox_coder=dict(\n type='CenterPointBBoxCoder',\n pc_range=[-51.2, -51.2],\n post_center_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],\n max_num=500,\n score_threshold=0.1,\n out_size_factor=8,\n voxel_size=[0.1, 0.1],\n code_size=9),\n separate_head=dict(\n type='SeparateHead', init_bias=-2.19, final_kernel=3),\n loss_cls=dict(type='GaussianFocalLoss', reduction='mean'),\n loss_bbox=dict(type='L1Loss', reduction='mean', loss_weight=0.25),\n norm_bbox=True),\n aux_bbox_head=[\n dict(\n type='CenterHead',\n in_channels=256,\n norm_cfg=dict(type='SyncBN', requires_grad=True),\n tasks=[\n dict(num_class=1, class_names=['car']),\n dict(\n num_class=2, class_names=['truck',\n 'construction_vehicle']),\n dict(num_class=2, class_names=['bus', 'trailer']),\n dict(num_class=1, class_names=['barrier']),\n dict(num_class=2, class_names=['motorcycle', 'bicycle']),\n dict(num_class=2, class_names=['pedestrian', 'traffic_cone'])\n ],\n common_heads=dict(\n reg=(2, 2), height=(1, 2), dim=(3, 2), rot=(2, 2), vel=(2, 2)),\n share_conv_channel=64,\n bbox_coder=dict(\n type='CenterPointBBoxCoder',\n pc_range=[-51.2, -51.2],\n post_center_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],\n max_num=500,\n score_threshold=0.1,\n out_size_factor=8,\n voxel_size=[0.1, 0.1],\n code_size=9),\n separate_head=dict(\n type='SeparateHead',\n init_bias=-2.19,\n final_kernel=3,\n norm_cfg=dict(type='SyncBN', requires_grad=True)),\n loss_cls=dict(\n type='GaussianFocalLoss', reduction='mean', loss_weight=0.5),\n loss_bbox=dict(type='L1Loss', reduction='mean', loss_weight=0.125),\n norm_bbox=True)\n ],\n aux_train_cfg=[\n dict(\n point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0],\n grid_size=[1024, 1024, 40],\n voxel_size=[0.1, 0.1, 0.2],\n out_size_factor=8,\n dense_reg=1,\n gaussian_overlap=0.1,\n max_objs=500,\n min_radius=2,\n code_weights=[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0])\n ],\n aux_test_cfg=[\n dict(\n pc_range=[-51.2, -51.2],\n post_center_limit_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],\n max_per_img=500,\n max_pool_nms=False,\n min_radius=[4, 12, 10, 1, 0.85, 0.175],\n score_threshold=0.1,\n out_size_factor=8,\n voxel_size=[0.1, 0.1],\n pre_max_size=1000,\n post_max_size=83,\n nms_type=[\n 'rotate', 'rotate', 'rotate', 'circle', 'rotate', 'rotate'\n ],\n nms_thr=[0.2, 0.2, 0.2, 0.2, 0.2, 0.5],\n nms_rescale_factor=[\n 1.0, [0.7, 0.7], [0.4, 0.55], 1.1, [1.0, 1.0], [4.5, 9.0]\n ])\n ],\n train_cfg=dict(\n pts=dict(\n point_cloud_range=[-51.2, -51.2, -5.0, 51.2, 51.2, 3.0],\n grid_size=[1024, 1024, 40],\n voxel_size=[0.1, 0.1, 0.2],\n out_size_factor=8,\n dense_reg=1,\n gaussian_overlap=0.1,\n max_objs=500,\n min_radius=2,\n code_weights=[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0])),\n test_cfg=dict(\n pts=dict(\n pc_range=[-51.2, -51.2],\n post_center_limit_range=[-61.2, -61.2, -10.0, 61.2, 61.2, 10.0],\n max_per_img=500,\n max_pool_nms=False,\n min_radius=[4, 12, 10, 1, 0.85, 0.175],\n score_threshold=0.1,\n out_size_factor=8,\n voxel_size=[0.1, 0.1],\n pre_max_size=1000,\n post_max_size=83,\n nms_type=[\n 'rotate', 'rotate', 'rotate', 'circle', 'rotate', 'rotate'\n ],\n nms_thr=[0.2, 0.2, 0.2, 0.2, 0.2, 0.5],\n nms_rescale_factor=[\n 1.0, [0.7, 0.7], [0.4, 0.55], 1.1, [1.0, 1.0], [4.5, 9.0]\n ])))\nbda_aug_conf = dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5)\nshare_data_config = dict(\n type='NuScenesDataset',\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier',\n 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n modality=dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False),\n img_info_prototype='bevdet4d',\n multi_adj_frame_id_cfg=(1, 9, 1))\ntest_data_config = dict(\n pipeline=[\n dict(\n type='PrepareImageInputs',\n data_config=dict(\n cams=[\n 'CAM_FRONT_LEFT', 'CAM_FRONT', 'CAM_FRONT_RIGHT',\n 'CAM_BACK_LEFT', 'CAM_BACK', 'CAM_BACK_RIGHT'\n ],\n Ncams=6,\n input_size=(256, 704),\n src_size=(900, 1600),\n resize=(-0.06, 0.11),\n rot=(-5.4, 5.4),\n flip=True,\n crop_h=(0.0, 0.0),\n resize_test=0.0),\n sequential=True,\n add_adj_bbox=True,\n file_client_args=dict(backend='disk')),\n dict(\n type='LoadAnnotationsBEVDepth',\n bda_aug_conf=dict(\n rot_lim=(-22.5, 22.5),\n scale_lim=(0.95, 1.05),\n flip_dx_ratio=0.5,\n flip_dy_ratio=0.5),\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer',\n 'barrier', 'motorcycle', 'bicycle', 'pedestrian',\n 'traffic_cone'\n ],\n align_adj_bbox=True,\n is_train=False),\n dict(\n type='LoadPointsFromFile',\n coord_type='LIDAR',\n load_dim=5,\n use_dim=5,\n file_client_args=dict(backend='disk')),\n dict(\n type='MultiScaleFlipAug3D',\n img_scale=(1333, 800),\n pts_scale_ratio=1,\n flip=False,\n transforms=[\n dict(\n type='DefaultFormatBundle3D',\n class_names=[\n 'car', 'truck', 'construction_vehicle', 'bus',\n 'trailer', 'barrier', 'motorcycle', 'bicycle',\n 'pedestrian', 'traffic_cone'\n ],\n with_label=False),\n dict(type='Collect3D', keys=['points', 'img_inputs'])\n ])\n ],\n ann_file='data/nuscenes/bevdetv2-nuscenes_infos_val.pkl',\n type='NuScenesDataset',\n classes=[\n 'car', 'truck', 'construction_vehicle', 'bus', 'trailer', 'barrier',\n 'motorcycle', 'bicycle', 'pedestrian', 'traffic_cone'\n ],\n modality=dict(\n use_lidar=False,\n use_camera=True,\n use_radar=False,\n use_map=False,\n use_external=False),\n img_info_prototype='bevdet4d',\n multi_adj_frame_id_cfg=(1, 9, 1))\nkey = 'test'\noptimizer = dict(type='AdamW', lr=0.0002, weight_decay=0.01)\noptimizer_config = dict(grad_clip=dict(max_norm=5, norm_type=2))\nlr_config = dict(\n policy='step',\n warmup='linear',\n warmup_iters=200,\n warmup_ratio=0.001,\n step=[24])\nrunner = dict(type='EpochBasedRunner', max_epochs=24)\ncustom_hooks = [\n dict(type='MEGVIIEMAHook', init_updates=10560, priority='NORMAL'),\n dict(type='SequentialControlHook', temporal_start_epoch=3)\n]\ngpu_ids = range(0, 8)\n", "CLASSES": ["car", "truck", "construction_vehicle", "bus", "trailer", "barrier", "motorcycle", "bicycle", "pedestrian", "traffic_cone"], "PALETTE": null, "env_info": "sys.platform: linux\nPython: 3.6.9 |Anaconda, Inc.| (default, Jul 30 2019, 19:07:31) [GCC 7.3.0]\nCUDA available: True\nGPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB\nCUDA_HOME: /usr/local/cuda\nNVCC: Cuda compilation tools, release 11.1, V11.1.105\nGCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0\nPyTorch: 1.8.1+cu111\nPyTorch compiling details: PyTorch built with:\n - GCC 7.3\n - C++ Version: 201402\n - Intel(R) Math Kernel Library Version 2020.0.2 Product Build 20200624 for Intel(R) 64 architecture applications\n - Intel(R) MKL-DNN v1.7.0 (Git Hash 7aed236906b1f7a05c0917e5257a1af05e9ff683)\n - OpenMP 201511 (a.k.a. OpenMP 4.5)\n - NNPACK is enabled\n - CPU capability usage: AVX2\n - CUDA Runtime 11.1\n - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86\n - CuDNN 8.0.5\n - Magma 2.5.2\n - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -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=1.8.1, 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, \n\nTorchVision: 0.9.1+cu102\nOpenCV: 3.4.13\nMMCV: 1.5.2\nMMCV Compiler: GCC 7.3\nMMCV CUDA Compiler: 11.1\nMMDetection: 2.24.0\nMMSegmentation: 0.24.0\nMMDetection3D: 1.0.0rc4+f130c13\nspconv2.0: False", "seed": 0, "exp_name": "hop_bevdet4d-r50-depth.py", "epoch": 9, "iter": 1980, "mmcv_version": "1.5.2", "time": "Tue Sep 5 22:20:41 2023", "hook_msgs": {"last_ckpt": "/cpfs01/user/huyihui/workspace/git_hop/main/hop/work_dirs/hop_bevdet4d-r50-depth/epoch_6.pth"}}
{"mode": "train", "epoch": 10, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 1.06588, "loss_depth": 29.20106, "task0.loss_xy": 0.12535, "task0.loss_z": 0.09718, "task0.loss_whl": 0.06164, "task0.loss_yaw": 0.29002, "task0.loss_vel": 0.44122, "task0.loss_heatmap": 2.65973, "task1.loss_xy": 0.12507, "task1.loss_z": 0.13239, "task1.loss_whl": 0.16616, "task1.loss_yaw": 0.29428, "task1.loss_vel": 0.34454, "task1.loss_heatmap": 3.58604, "task2.loss_xy": 0.12618, "task2.loss_z": 0.11281, "task2.loss_whl": 0.14884, "task2.loss_yaw": 0.29956, "task2.loss_vel": 0.44593, "task2.loss_heatmap": 4.08571, "task3.loss_xy": 0.12501, "task3.loss_z": 0.07148, "task3.loss_whl": 0.14659, "task3.loss_yaw": 0.28544, "task3.loss_vel": 0.0176, "task3.loss_heatmap": 2.95073, "task4.loss_xy": 0.1237, "task4.loss_z": 0.0939, "task4.loss_whl": 0.1298, "task4.loss_yaw": 0.30564, "task4.loss_vel": 0.38752, "task4.loss_heatmap": 4.38224, "task5.loss_xy": 0.12547, "task5.loss_z": 0.10509, "task5.loss_whl": 0.18806, "task5.loss_yaw": 0.30404, "task5.loss_vel": 0.19591, "task5.loss_heatmap": 2.94863, "task0.loss_xy0": 0.06305, "task0.loss_z0": 0.04899, "task0.loss_whl0": 0.03238, "task0.loss_yaw0": 0.14945, "task0.loss_vel0": 0.22227, "task0.loss_heatmap0": 1.35899, "task1.loss_xy0": 0.06308, "task1.loss_z0": 0.06672, "task1.loss_whl0": 0.0846, "task1.loss_yaw0": 0.15008, "task1.loss_vel0": 0.17253, "task1.loss_heatmap0": 1.85646, "task2.loss_xy0": 0.06415, "task2.loss_z0": 0.05748, "task2.loss_whl0": 0.0769, "task2.loss_yaw0": 0.15194, "task2.loss_vel0": 0.22418, "task2.loss_heatmap0": 2.40914, "task3.loss_xy0": 0.06269, "task3.loss_z0": 0.03523, "task3.loss_whl0": 0.07414, "task3.loss_yaw0": 0.1504, "task3.loss_vel0": 0.00966, "task3.loss_heatmap0": 1.52649, "task4.loss_xy0": 0.06182, "task4.loss_z0": 0.04702, "task4.loss_whl0": 0.06647, "task4.loss_yaw0": 0.15303, "task4.loss_vel0": 0.19121, "task4.loss_heatmap0": 2.5763, "task5.loss_xy0": 0.06284, "task5.loss_z0": 0.05285, "task5.loss_whl0": 0.09515, "task5.loss_yaw0": 0.15235, "task5.loss_vel0": 0.09864, "task5.loss_heatmap0": 1.52453, "loss": 69.82376, "grad_norm": 112852.33844, "time": 22.81787}
{"mode": "train", "epoch": 10, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.42875, "loss_depth": 30.63627, "task0.loss_xy": 0.1254, "task0.loss_z": 0.09809, "task0.loss_whl": 0.06149, "task0.loss_yaw": 0.29035, "task0.loss_vel": 0.43825, "task0.loss_heatmap": 2.65602, "task1.loss_xy": 0.12437, "task1.loss_z": 0.13632, "task1.loss_whl": 0.16584, "task1.loss_yaw": 0.29543, "task1.loss_vel": 0.33263, "task1.loss_heatmap": 3.59543, "task2.loss_xy": 0.12456, "task2.loss_z": 0.11289, "task2.loss_whl": 0.14541, "task2.loss_yaw": 0.30003, "task2.loss_vel": 0.43844, "task2.loss_heatmap": 4.07027, "task3.loss_xy": 0.12581, "task3.loss_z": 0.07143, "task3.loss_whl": 0.14617, "task3.loss_yaw": 0.28796, "task3.loss_vel": 0.01907, "task3.loss_heatmap": 2.97094, "task4.loss_xy": 0.125, "task4.loss_z": 0.09732, "task4.loss_whl": 0.13421, "task4.loss_yaw": 0.30435, "task4.loss_vel": 0.38484, "task4.loss_heatmap": 4.38646, "task5.loss_xy": 0.12539, "task5.loss_z": 0.10672, "task5.loss_whl": 0.18783, "task5.loss_yaw": 0.30453, "task5.loss_vel": 0.19689, "task5.loss_heatmap": 2.97659, "task0.loss_xy0": 0.06298, "task0.loss_z0": 0.04953, "task0.loss_whl0": 0.03235, "task0.loss_yaw0": 0.14942, "task0.loss_vel0": 0.22123, "task0.loss_heatmap0": 1.35685, "task1.loss_xy0": 0.06278, "task1.loss_z0": 0.06833, "task1.loss_whl0": 0.08412, "task1.loss_yaw0": 0.1502, "task1.loss_vel0": 0.16875, "task1.loss_heatmap0": 1.85443, "task2.loss_xy0": 0.06248, "task2.loss_z0": 0.05693, "task2.loss_whl0": 0.07539, "task2.loss_yaw0": 0.15179, "task2.loss_vel0": 0.2161, "task2.loss_heatmap0": 2.36523, "task3.loss_xy0": 0.06296, "task3.loss_z0": 0.03559, "task3.loss_whl0": 0.07411, "task3.loss_yaw0": 0.15104, "task3.loss_vel0": 0.00992, "task3.loss_heatmap0": 1.53767, "task4.loss_xy0": 0.06366, "task4.loss_z0": 0.04916, "task4.loss_whl0": 0.06807, "task4.loss_yaw0": 0.15283, "task4.loss_vel0": 0.19221, "task4.loss_heatmap0": 2.54586, "task5.loss_xy0": 0.06304, "task5.loss_z0": 0.05362, "task5.loss_whl0": 0.09517, "task5.loss_yaw0": 0.1526, "task5.loss_vel0": 0.09876, "task5.loss_heatmap0": 1.53591, "loss": 71.23008, "grad_norm": 139758.81504, "time": 21.16823}
{"mode": "train", "epoch": 10, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.44597, "loss_depth": 31.81456, "task0.loss_xy": 0.12501, "task0.loss_z": 0.09564, "task0.loss_whl": 0.06101, "task0.loss_yaw": 0.2901, "task0.loss_vel": 0.44684, "task0.loss_heatmap": 2.65653, "task1.loss_xy": 0.12612, "task1.loss_z": 0.13488, "task1.loss_whl": 0.16436, "task1.loss_yaw": 0.29417, "task1.loss_vel": 0.3439, "task1.loss_heatmap": 3.60723, "task2.loss_xy": 0.12717, "task2.loss_z": 0.1152, "task2.loss_whl": 0.14513, "task2.loss_yaw": 0.29852, "task2.loss_vel": 0.48208, "task2.loss_heatmap": 4.09951, "task3.loss_xy": 0.12513, "task3.loss_z": 0.07378, "task3.loss_whl": 0.14778, "task3.loss_yaw": 0.28643, "task3.loss_vel": 0.01816, "task3.loss_heatmap": 2.94256, "task4.loss_xy": 0.12631, "task4.loss_z": 0.08853, "task4.loss_whl": 0.13313, "task4.loss_yaw": 0.30434, "task4.loss_vel": 0.36831, "task4.loss_heatmap": 4.36568, "task5.loss_xy": 0.12524, "task5.loss_z": 0.10372, "task5.loss_whl": 0.1892, "task5.loss_yaw": 0.30417, "task5.loss_vel": 0.19394, "task5.loss_heatmap": 2.95937, "task0.loss_xy0": 0.06313, "task0.loss_z0": 0.04842, "task0.loss_whl0": 0.03424, "task0.loss_yaw0": 0.14955, "task0.loss_vel0": 0.22367, "task0.loss_heatmap0": 1.363, "task1.loss_xy0": 0.06375, "task1.loss_z0": 0.06807, "task1.loss_whl0": 0.08476, "task1.loss_yaw0": 0.15002, "task1.loss_vel0": 0.17191, "task1.loss_heatmap0": 1.89431, "task2.loss_xy0": 0.06392, "task2.loss_z0": 0.06009, "task2.loss_whl0": 0.07501, "task2.loss_yaw0": 0.15132, "task2.loss_vel0": 0.24118, "task2.loss_heatmap0": 2.44047, "task3.loss_xy0": 0.06299, "task3.loss_z0": 0.03685, "task3.loss_whl0": 0.07451, "task3.loss_yaw0": 0.15, "task3.loss_vel0": 0.00934, "task3.loss_heatmap0": 1.54008, "task4.loss_xy0": 0.06366, "task4.loss_z0": 0.04476, "task4.loss_whl0": 0.06749, "task4.loss_yaw0": 0.15229, "task4.loss_vel0": 0.18509, "task4.loss_heatmap0": 2.61775, "task5.loss_xy0": 0.0631, "task5.loss_z0": 0.05209, "task5.loss_whl0": 0.09607, "task5.loss_yaw0": 0.15229, "task5.loss_vel0": 0.09767, "task5.loss_heatmap0": 1.53957, "loss": 72.63617, "grad_norm": 172188.78039, "time": 20.92063}
{"mode": "train", "epoch": 10, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.44559, "loss_depth": 32.8679, "task0.loss_xy": 0.12527, "task0.loss_z": 0.09897, "task0.loss_whl": 0.06103, "task0.loss_yaw": 0.29, "task0.loss_vel": 0.46177, "task0.loss_heatmap": 2.66511, "task1.loss_xy": 0.12546, "task1.loss_z": 0.13506, "task1.loss_whl": 0.16637, "task1.loss_yaw": 0.29439, "task1.loss_vel": 0.35382, "task1.loss_heatmap": 3.60485, "task2.loss_xy": 0.12426, "task2.loss_z": 0.1152, "task2.loss_whl": 0.14544, "task2.loss_yaw": 0.29761, "task2.loss_vel": 0.472, "task2.loss_heatmap": 4.09434, "task3.loss_xy": 0.12581, "task3.loss_z": 0.07605, "task3.loss_whl": 0.15076, "task3.loss_yaw": 0.28626, "task3.loss_vel": 0.01739, "task3.loss_heatmap": 2.9543, "task4.loss_xy": 0.12645, "task4.loss_z": 0.09267, "task4.loss_whl": 0.13654, "task4.loss_yaw": 0.30452, "task4.loss_vel": 0.41018, "task4.loss_heatmap": 4.40907, "task5.loss_xy": 0.12535, "task5.loss_z": 0.10354, "task5.loss_whl": 0.18696, "task5.loss_yaw": 0.30506, "task5.loss_vel": 0.19738, "task5.loss_heatmap": 2.94435, "task0.loss_xy0": 0.06294, "task0.loss_z0": 0.04981, "task0.loss_whl0": 0.03222, "task0.loss_yaw0": 0.14968, "task0.loss_vel0": 0.23059, "task0.loss_heatmap0": 1.36367, "task1.loss_xy0": 0.06355, "task1.loss_z0": 0.06796, "task1.loss_whl0": 0.08461, "task1.loss_yaw0": 0.15061, "task1.loss_vel0": 0.17624, "task1.loss_heatmap0": 1.85256, "task2.loss_xy0": 0.06339, "task2.loss_z0": 0.05898, "task2.loss_whl0": 0.07569, "task2.loss_yaw0": 0.1515, "task2.loss_vel0": 0.23213, "task2.loss_heatmap0": 2.3359, "task3.loss_xy0": 0.06321, "task3.loss_z0": 0.03757, "task3.loss_whl0": 0.07625, "task3.loss_yaw0": 0.15057, "task3.loss_vel0": 0.00925, "task3.loss_heatmap0": 1.52256, "task4.loss_xy0": 0.06286, "task4.loss_z0": 0.0473, "task4.loss_whl0": 0.06947, "task4.loss_yaw0": 0.15257, "task4.loss_vel0": 0.19995, "task4.loss_heatmap0": 2.49849, "task5.loss_xy0": 0.06299, "task5.loss_z0": 0.05203, "task5.loss_whl0": 0.09474, "task5.loss_yaw0": 0.15289, "task5.loss_vel0": 0.09936, "task5.loss_heatmap0": 1.51675, "loss": 73.52232, "grad_norm": 89177.68613, "time": 21.03991}
{"mode": "val", "epoch": 10, "iter": 753, "lr": 0.0002, "pts_bbox_NuScenes/car_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/car_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/car_trans_err": 1.0, "pts_bbox_NuScenes/car_scale_err": 1.0, "pts_bbox_NuScenes/car_orient_err": 1.0, "pts_bbox_NuScenes/car_vel_err": 1.0, "pts_bbox_NuScenes/car_attr_err": 1.0, "pts_bbox_NuScenes/mATE": 1.0, "pts_bbox_NuScenes/mASE": 1.0, "pts_bbox_NuScenes/mAOE": 1.0, "pts_bbox_NuScenes/mAVE": 1.0, "pts_bbox_NuScenes/mAAE": 1.0, "pts_bbox_NuScenes/truck_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/truck_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/truck_trans_err": 1.0, "pts_bbox_NuScenes/truck_scale_err": 1.0, "pts_bbox_NuScenes/truck_orient_err": 1.0, "pts_bbox_NuScenes/truck_vel_err": 1.0, "pts_bbox_NuScenes/truck_attr_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_trans_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_scale_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_orient_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_vel_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_attr_err": 1.0, "pts_bbox_NuScenes/bus_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bus_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bus_trans_err": 1.0, "pts_bbox_NuScenes/bus_scale_err": 1.0, "pts_bbox_NuScenes/bus_orient_err": 1.0, "pts_bbox_NuScenes/bus_vel_err": 1.0, "pts_bbox_NuScenes/bus_attr_err": 1.0, "pts_bbox_NuScenes/trailer_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/trailer_trans_err": 1.0, "pts_bbox_NuScenes/trailer_scale_err": 1.0, "pts_bbox_NuScenes/trailer_orient_err": 1.0, "pts_bbox_NuScenes/trailer_vel_err": 1.0, "pts_bbox_NuScenes/trailer_attr_err": 1.0, "pts_bbox_NuScenes/barrier_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/barrier_trans_err": 1.0, "pts_bbox_NuScenes/barrier_scale_err": 1.0, "pts_bbox_NuScenes/barrier_orient_err": 1.0, "pts_bbox_NuScenes/barrier_vel_err": NaN, "pts_bbox_NuScenes/barrier_attr_err": NaN, "pts_bbox_NuScenes/motorcycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/motorcycle_trans_err": 1.0, "pts_bbox_NuScenes/motorcycle_scale_err": 1.0, "pts_bbox_NuScenes/motorcycle_orient_err": 1.0, "pts_bbox_NuScenes/motorcycle_vel_err": 1.0, "pts_bbox_NuScenes/motorcycle_attr_err": 1.0, "pts_bbox_NuScenes/bicycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bicycle_trans_err": 1.0, "pts_bbox_NuScenes/bicycle_scale_err": 1.0, "pts_bbox_NuScenes/bicycle_orient_err": 1.0, "pts_bbox_NuScenes/bicycle_vel_err": 1.0, "pts_bbox_NuScenes/bicycle_attr_err": 1.0, "pts_bbox_NuScenes/pedestrian_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/pedestrian_trans_err": 1.0, "pts_bbox_NuScenes/pedestrian_scale_err": 1.0, "pts_bbox_NuScenes/pedestrian_orient_err": 1.0, "pts_bbox_NuScenes/pedestrian_vel_err": 1.0, "pts_bbox_NuScenes/pedestrian_attr_err": 1.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/traffic_cone_trans_err": 1.0, "pts_bbox_NuScenes/traffic_cone_scale_err": 1.0, "pts_bbox_NuScenes/traffic_cone_orient_err": NaN, "pts_bbox_NuScenes/traffic_cone_vel_err": NaN, "pts_bbox_NuScenes/traffic_cone_attr_err": NaN, "pts_bbox_NuScenes/NDS": 0.0, "pts_bbox_NuScenes/mAP": 0.0}
{"mode": "train", "epoch": 11, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 0.98777, "loss_depth": 36.09954, "task0.loss_xy": 0.12515, "task0.loss_z": 0.0976, "task0.loss_whl": 0.06229, "task0.loss_yaw": 0.2897, "task0.loss_vel": 0.43929, "task0.loss_heatmap": 2.65704, "task1.loss_xy": 0.12515, "task1.loss_z": 0.13478, "task1.loss_whl": 0.16589, "task1.loss_yaw": 0.29311, "task1.loss_vel": 0.32181, "task1.loss_heatmap": 3.57989, "task2.loss_xy": 0.12508, "task2.loss_z": 0.11444, "task2.loss_whl": 0.14804, "task2.loss_yaw": 0.29562, "task2.loss_vel": 0.47412, "task2.loss_heatmap": 4.07949, "task3.loss_xy": 0.12571, "task3.loss_z": 0.07364, "task3.loss_whl": 0.14649, "task3.loss_yaw": 0.2864, "task3.loss_vel": 0.0174, "task3.loss_heatmap": 2.95947, "task4.loss_xy": 0.12532, "task4.loss_z": 0.09515, "task4.loss_whl": 0.13033, "task4.loss_yaw": 0.30286, "task4.loss_vel": 0.38373, "task4.loss_heatmap": 4.33773, "task5.loss_xy": 0.12544, "task5.loss_z": 0.10434, "task5.loss_whl": 0.18936, "task5.loss_yaw": 0.30456, "task5.loss_vel": 0.19448, "task5.loss_heatmap": 2.95509, "task0.loss_xy0": 0.0629, "task0.loss_z0": 0.04932, "task0.loss_whl0": 0.03212, "task0.loss_yaw0": 0.14935, "task0.loss_vel0": 0.21913, "task0.loss_heatmap0": 1.3567, "task1.loss_xy0": 0.06299, "task1.loss_z0": 0.06764, "task1.loss_whl0": 0.08412, "task1.loss_yaw0": 0.15029, "task1.loss_vel0": 0.16559, "task1.loss_heatmap0": 1.82724, "task2.loss_xy0": 0.06331, "task2.loss_z0": 0.05821, "task2.loss_whl0": 0.07629, "task2.loss_yaw0": 0.15105, "task2.loss_vel0": 0.23761, "task2.loss_heatmap0": 2.29912, "task3.loss_xy0": 0.063, "task3.loss_z0": 0.03686, "task3.loss_whl0": 0.07413, "task3.loss_yaw0": 0.15029, "task3.loss_vel0": 0.00968, "task3.loss_heatmap0": 1.52376, "task4.loss_xy0": 0.06276, "task4.loss_z0": 0.04716, "task4.loss_whl0": 0.06612, "task4.loss_yaw0": 0.15173, "task4.loss_vel0": 0.193, "task4.loss_heatmap0": 2.4309, "task5.loss_xy0": 0.0629, "task5.loss_z0": 0.05217, "task5.loss_whl0": 0.09526, "task5.loss_yaw0": 0.15248, "task5.loss_vel0": 0.09844, "task5.loss_heatmap0": 1.51806, "loss": 76.38724, "grad_norm": 102667.74645, "time": 21.27335}
{"mode": "train", "epoch": 11, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.37779, "loss_depth": 34.25549, "task0.loss_xy": 0.12539, "task0.loss_z": 0.09812, "task0.loss_whl": 0.06207, "task0.loss_yaw": 0.2903, "task0.loss_vel": 0.44401, "task0.loss_heatmap": 2.66231, "task1.loss_xy": 0.12568, "task1.loss_z": 0.13614, "task1.loss_whl": 0.16561, "task1.loss_yaw": 0.29352, "task1.loss_vel": 0.34373, "task1.loss_heatmap": 3.59603, "task2.loss_xy": 0.1264, "task2.loss_z": 0.11416, "task2.loss_whl": 0.14549, "task2.loss_yaw": 0.29902, "task2.loss_vel": 0.4672, "task2.loss_heatmap": 4.05724, "task3.loss_xy": 0.12536, "task3.loss_z": 0.07649, "task3.loss_whl": 0.14731, "task3.loss_yaw": 0.28443, "task3.loss_vel": 0.01836, "task3.loss_heatmap": 2.95564, "task4.loss_xy": 0.12648, "task4.loss_z": 0.09318, "task4.loss_whl": 0.13251, "task4.loss_yaw": 0.30488, "task4.loss_vel": 0.37379, "task4.loss_heatmap": 4.35971, "task5.loss_xy": 0.12517, "task5.loss_z": 0.10594, "task5.loss_whl": 0.18944, "task5.loss_yaw": 0.3048, "task5.loss_vel": 0.19566, "task5.loss_heatmap": 2.96156, "task0.loss_xy0": 0.06296, "task0.loss_z0": 0.04974, "task0.loss_whl0": 0.03272, "task0.loss_yaw0": 0.14961, "task0.loss_vel0": 0.22256, "task0.loss_heatmap0": 1.35898, "task1.loss_xy0": 0.06317, "task1.loss_z0": 0.06848, "task1.loss_whl0": 0.08391, "task1.loss_yaw0": 0.1504, "task1.loss_vel0": 0.16954, "task1.loss_heatmap0": 1.84215, "task2.loss_xy0": 0.06341, "task2.loss_z0": 0.05868, "task2.loss_whl0": 0.07467, "task2.loss_yaw0": 0.15192, "task2.loss_vel0": 0.23117, "task2.loss_heatmap0": 2.27454, "task3.loss_xy0": 0.06263, "task3.loss_z0": 0.03732, "task3.loss_whl0": 0.07476, "task3.loss_yaw0": 0.15114, "task3.loss_vel0": 0.00962, "task3.loss_heatmap0": 1.52854, "task4.loss_xy0": 0.0645, "task4.loss_z0": 0.04796, "task4.loss_whl0": 0.06685, "task4.loss_yaw0": 0.15288, "task4.loss_vel0": 0.19095, "task4.loss_heatmap0": 2.45317, "task5.loss_xy0": 0.06295, "task5.loss_z0": 0.0535, "task5.loss_whl0": 0.09578, "task5.loss_yaw0": 0.15267, "task5.loss_vel0": 0.09808, "task5.loss_heatmap0": 1.52367, "loss": 74.62418, "grad_norm": 105082.78656, "time": 20.78264}
{"mode": "train", "epoch": 11, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.39227, "loss_depth": 34.36546, "task0.loss_xy": 0.12507, "task0.loss_z": 0.09643, "task0.loss_whl": 0.06133, "task0.loss_yaw": 0.28943, "task0.loss_vel": 0.44952, "task0.loss_heatmap": 2.65213, "task1.loss_xy": 0.12584, "task1.loss_z": 0.13252, "task1.loss_whl": 0.16555, "task1.loss_yaw": 0.2937, "task1.loss_vel": 0.34428, "task1.loss_heatmap": 3.57812, "task2.loss_xy": 0.12586, "task2.loss_z": 0.11315, "task2.loss_whl": 0.14722, "task2.loss_yaw": 0.29768, "task2.loss_vel": 0.44794, "task2.loss_heatmap": 4.07587, "task3.loss_xy": 0.12566, "task3.loss_z": 0.07261, "task3.loss_whl": 0.15025, "task3.loss_yaw": 0.28678, "task3.loss_vel": 0.01887, "task3.loss_heatmap": 2.94518, "task4.loss_xy": 0.12743, "task4.loss_z": 0.08994, "task4.loss_whl": 0.13586, "task4.loss_yaw": 0.30476, "task4.loss_vel": 0.38156, "task4.loss_heatmap": 4.30861, "task5.loss_xy": 0.12524, "task5.loss_z": 0.10375, "task5.loss_whl": 0.18628, "task5.loss_yaw": 0.30494, "task5.loss_vel": 0.19632, "task5.loss_heatmap": 2.94577, "task0.loss_xy0": 0.06298, "task0.loss_z0": 0.04848, "task0.loss_whl0": 0.03225, "task0.loss_yaw0": 0.1497, "task0.loss_vel0": 0.2255, "task0.loss_heatmap0": 1.35483, "task1.loss_xy0": 0.06344, "task1.loss_z0": 0.06625, "task1.loss_whl0": 0.08429, "task1.loss_yaw0": 0.15019, "task1.loss_vel0": 0.17159, "task1.loss_heatmap0": 1.82644, "task2.loss_xy0": 0.06326, "task2.loss_z0": 0.05723, "task2.loss_whl0": 0.07579, "task2.loss_yaw0": 0.15162, "task2.loss_vel0": 0.22241, "task2.loss_heatmap0": 2.28197, "task3.loss_xy0": 0.0632, "task3.loss_z0": 0.03614, "task3.loss_whl0": 0.07552, "task3.loss_yaw0": 0.15019, "task3.loss_vel0": 0.00975, "task3.loss_heatmap0": 1.51473, "task4.loss_xy0": 0.06389, "task4.loss_z0": 0.04595, "task4.loss_whl0": 0.06919, "task4.loss_yaw0": 0.15286, "task4.loss_vel0": 0.19405, "task4.loss_heatmap0": 2.39402, "task5.loss_xy0": 0.063, "task5.loss_z0": 0.05208, "task5.loss_whl0": 0.09437, "task5.loss_yaw0": 0.15248, "task5.loss_vel0": 0.09847, "task5.loss_heatmap0": 1.51172, "loss": 74.52674, "grad_norm": 132278.6125, "time": 20.72689}
{"mode": "train", "epoch": 11, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.38001, "loss_depth": 35.73579, "task0.loss_xy": 0.12496, "task0.loss_z": 0.09792, "task0.loss_whl": 0.06124, "task0.loss_yaw": 0.28998, "task0.loss_vel": 0.45295, "task0.loss_heatmap": 2.6561, "task1.loss_xy": 0.12596, "task1.loss_z": 0.13488, "task1.loss_whl": 0.16565, "task1.loss_yaw": 0.29425, "task1.loss_vel": 0.35118, "task1.loss_heatmap": 3.5873, "task2.loss_xy": 0.12489, "task2.loss_z": 0.1147, "task2.loss_whl": 0.14672, "task2.loss_yaw": 0.29702, "task2.loss_vel": 0.44213, "task2.loss_heatmap": 4.05901, "task3.loss_xy": 0.12556, "task3.loss_z": 0.07158, "task3.loss_whl": 0.14959, "task3.loss_yaw": 0.28436, "task3.loss_vel": 0.01774, "task3.loss_heatmap": 2.93933, "task4.loss_xy": 0.1263, "task4.loss_z": 0.09196, "task4.loss_whl": 0.13165, "task4.loss_yaw": 0.30331, "task4.loss_vel": 0.39211, "task4.loss_heatmap": 4.32136, "task5.loss_xy": 0.12531, "task5.loss_z": 0.10458, "task5.loss_whl": 0.18659, "task5.loss_yaw": 0.30493, "task5.loss_vel": 0.19588, "task5.loss_heatmap": 2.94535, "task0.loss_xy0": 0.06291, "task0.loss_z0": 0.04931, "task0.loss_whl0": 0.03152, "task0.loss_yaw0": 0.14975, "task0.loss_vel0": 0.22788, "task0.loss_heatmap0": 1.35463, "task1.loss_xy0": 0.06353, "task1.loss_z0": 0.0676, "task1.loss_whl0": 0.08381, "task1.loss_yaw0": 0.15078, "task1.loss_vel0": 0.17679, "task1.loss_heatmap0": 1.82756, "task2.loss_xy0": 0.06322, "task2.loss_z0": 0.05835, "task2.loss_whl0": 0.07483, "task2.loss_yaw0": 0.15172, "task2.loss_vel0": 0.21861, "task2.loss_heatmap0": 2.24594, "task3.loss_xy0": 0.0628, "task3.loss_z0": 0.03557, "task3.loss_whl0": 0.07558, "task3.loss_yaw0": 0.15005, "task3.loss_vel0": 0.00907, "task3.loss_heatmap0": 1.51481, "task4.loss_xy0": 0.06316, "task4.loss_z0": 0.04617, "task4.loss_whl0": 0.06678, "task4.loss_yaw0": 0.15254, "task4.loss_vel0": 0.19033, "task4.loss_heatmap0": 2.37796, "task5.loss_xy0": 0.06278, "task5.loss_z0": 0.05232, "task5.loss_whl0": 0.09455, "task5.loss_yaw0": 0.15255, "task5.loss_vel0": 0.09858, "task5.loss_heatmap0": 1.51266, "loss": 75.8571, "grad_norm": 141621.39117, "time": 20.75901}
{"mode": "val", "epoch": 11, "iter": 753, "lr": 0.0002, "pts_bbox_NuScenes/car_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/car_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/car_trans_err": 1.0, "pts_bbox_NuScenes/car_scale_err": 1.0, "pts_bbox_NuScenes/car_orient_err": 1.0, "pts_bbox_NuScenes/car_vel_err": 1.0, "pts_bbox_NuScenes/car_attr_err": 1.0, "pts_bbox_NuScenes/mATE": 1.0, "pts_bbox_NuScenes/mASE": 1.0, "pts_bbox_NuScenes/mAOE": 1.0, "pts_bbox_NuScenes/mAVE": 1.0, "pts_bbox_NuScenes/mAAE": 1.0, "pts_bbox_NuScenes/truck_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/truck_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/truck_trans_err": 1.0, "pts_bbox_NuScenes/truck_scale_err": 1.0, "pts_bbox_NuScenes/truck_orient_err": 1.0, "pts_bbox_NuScenes/truck_vel_err": 1.0, "pts_bbox_NuScenes/truck_attr_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_trans_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_scale_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_orient_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_vel_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_attr_err": 1.0, "pts_bbox_NuScenes/bus_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bus_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bus_trans_err": 1.0, "pts_bbox_NuScenes/bus_scale_err": 1.0, "pts_bbox_NuScenes/bus_orient_err": 1.0, "pts_bbox_NuScenes/bus_vel_err": 1.0, "pts_bbox_NuScenes/bus_attr_err": 1.0, "pts_bbox_NuScenes/trailer_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/trailer_trans_err": 1.0, "pts_bbox_NuScenes/trailer_scale_err": 1.0, "pts_bbox_NuScenes/trailer_orient_err": 1.0, "pts_bbox_NuScenes/trailer_vel_err": 1.0, "pts_bbox_NuScenes/trailer_attr_err": 1.0, "pts_bbox_NuScenes/barrier_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/barrier_trans_err": 1.0, "pts_bbox_NuScenes/barrier_scale_err": 1.0, "pts_bbox_NuScenes/barrier_orient_err": 1.0, "pts_bbox_NuScenes/barrier_vel_err": NaN, "pts_bbox_NuScenes/barrier_attr_err": NaN, "pts_bbox_NuScenes/motorcycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/motorcycle_trans_err": 1.0, "pts_bbox_NuScenes/motorcycle_scale_err": 1.0, "pts_bbox_NuScenes/motorcycle_orient_err": 1.0, "pts_bbox_NuScenes/motorcycle_vel_err": 1.0, "pts_bbox_NuScenes/motorcycle_attr_err": 1.0, "pts_bbox_NuScenes/bicycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bicycle_trans_err": 1.0, "pts_bbox_NuScenes/bicycle_scale_err": 1.0, "pts_bbox_NuScenes/bicycle_orient_err": 1.0, "pts_bbox_NuScenes/bicycle_vel_err": 1.0, "pts_bbox_NuScenes/bicycle_attr_err": 1.0, "pts_bbox_NuScenes/pedestrian_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/pedestrian_trans_err": 1.0, "pts_bbox_NuScenes/pedestrian_scale_err": 1.0, "pts_bbox_NuScenes/pedestrian_orient_err": 1.0, "pts_bbox_NuScenes/pedestrian_vel_err": 1.0, "pts_bbox_NuScenes/pedestrian_attr_err": 1.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/traffic_cone_trans_err": 1.0, "pts_bbox_NuScenes/traffic_cone_scale_err": 1.0, "pts_bbox_NuScenes/traffic_cone_orient_err": NaN, "pts_bbox_NuScenes/traffic_cone_vel_err": NaN, "pts_bbox_NuScenes/traffic_cone_attr_err": NaN, "pts_bbox_NuScenes/NDS": 0.0, "pts_bbox_NuScenes/mAP": 0.0}
{"mode": "train", "epoch": 12, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 0.996, "loss_depth": 36.10105, "task0.loss_xy": 0.12501, "task0.loss_z": 0.09698, "task0.loss_whl": 0.06073, "task0.loss_yaw": 0.28919, "task0.loss_vel": 0.449, "task0.loss_heatmap": 2.65513, "task1.loss_xy": 0.1255, "task1.loss_z": 0.13391, "task1.loss_whl": 0.16809, "task1.loss_yaw": 0.29456, "task1.loss_vel": 0.33946, "task1.loss_heatmap": 3.58415, "task2.loss_xy": 0.12736, "task2.loss_z": 0.1195, "task2.loss_whl": 0.14654, "task2.loss_yaw": 0.29803, "task2.loss_vel": 0.47631, "task2.loss_heatmap": 4.0867, "task3.loss_xy": 0.12476, "task3.loss_z": 0.07257, "task3.loss_whl": 0.14758, "task3.loss_yaw": 0.28561, "task3.loss_vel": 0.01784, "task3.loss_heatmap": 2.93208, "task4.loss_xy": 0.12614, "task4.loss_z": 0.09557, "task4.loss_whl": 0.13295, "task4.loss_yaw": 0.30381, "task4.loss_vel": 0.39549, "task4.loss_heatmap": 4.31606, "task5.loss_xy": 0.1251, "task5.loss_z": 0.10408, "task5.loss_whl": 0.18807, "task5.loss_yaw": 0.3046, "task5.loss_vel": 0.19325, "task5.loss_heatmap": 2.94376, "task0.loss_xy0": 0.06296, "task0.loss_z0": 0.04909, "task0.loss_whl0": 0.03253, "task0.loss_yaw0": 0.14978, "task0.loss_vel0": 0.22624, "task0.loss_heatmap0": 1.35988, "task1.loss_xy0": 0.06369, "task1.loss_z0": 0.06731, "task1.loss_whl0": 0.08533, "task1.loss_yaw0": 0.15071, "task1.loss_vel0": 0.17169, "task1.loss_heatmap0": 1.84719, "task2.loss_xy0": 0.06371, "task2.loss_z0": 0.06077, "task2.loss_whl0": 0.07623, "task2.loss_yaw0": 0.15218, "task2.loss_vel0": 0.23445, "task2.loss_heatmap0": 2.30819, "task3.loss_xy0": 0.06259, "task3.loss_z0": 0.03598, "task3.loss_whl0": 0.07448, "task3.loss_yaw0": 0.15005, "task3.loss_vel0": 0.00948, "task3.loss_heatmap0": 1.51335, "task4.loss_xy0": 0.06293, "task4.loss_z0": 0.04839, "task4.loss_whl0": 0.06798, "task4.loss_yaw0": 0.15257, "task4.loss_vel0": 0.19771, "task4.loss_heatmap0": 2.42148, "task5.loss_xy0": 0.06291, "task5.loss_z0": 0.05231, "task5.loss_whl0": 0.09547, "task5.loss_yaw0": 0.15259, "task5.loss_vel0": 0.09701, "task5.loss_heatmap0": 1.51706, "loss": 76.42277, "grad_norm": 139970.51527, "time": 21.37935}
{"mode": "train", "epoch": 12, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.37892, "loss_depth": 39.78092, "task0.loss_xy": 0.12547, "task0.loss_z": 0.09647, "task0.loss_whl": 0.06109, "task0.loss_yaw": 0.28952, "task0.loss_vel": 0.44648, "task0.loss_heatmap": 2.65152, "task1.loss_xy": 0.12557, "task1.loss_z": 0.13549, "task1.loss_whl": 0.16489, "task1.loss_yaw": 0.29259, "task1.loss_vel": 0.33926, "task1.loss_heatmap": 3.5784, "task2.loss_xy": 0.1258, "task2.loss_z": 0.11271, "task2.loss_whl": 0.1481, "task2.loss_yaw": 0.29756, "task2.loss_vel": 0.46034, "task2.loss_heatmap": 4.05131, "task3.loss_xy": 0.12535, "task3.loss_z": 0.0754, "task3.loss_whl": 0.14369, "task3.loss_yaw": 0.285, "task3.loss_vel": 0.01838, "task3.loss_heatmap": 2.95747, "task4.loss_xy": 0.12554, "task4.loss_z": 0.08899, "task4.loss_whl": 0.13112, "task4.loss_yaw": 0.30448, "task4.loss_vel": 0.38277, "task4.loss_heatmap": 4.253, "task5.loss_xy": 0.12532, "task5.loss_z": 0.10342, "task5.loss_whl": 0.18673, "task5.loss_yaw": 0.30463, "task5.loss_vel": 0.19953, "task5.loss_heatmap": 2.94497, "task0.loss_xy0": 0.06296, "task0.loss_z0": 0.04872, "task0.loss_whl0": 0.03122, "task0.loss_yaw0": 0.14989, "task0.loss_vel0": 0.22303, "task0.loss_heatmap0": 1.35507, "task1.loss_xy0": 0.06346, "task1.loss_z0": 0.06802, "task1.loss_whl0": 0.0836, "task1.loss_yaw0": 0.15042, "task1.loss_vel0": 0.17047, "task1.loss_heatmap0": 1.82094, "task2.loss_xy0": 0.06328, "task2.loss_z0": 0.05761, "task2.loss_whl0": 0.07512, "task2.loss_yaw0": 0.15195, "task2.loss_vel0": 0.23242, "task2.loss_heatmap0": 2.21717, "task3.loss_xy0": 0.06285, "task3.loss_z0": 0.03778, "task3.loss_whl0": 0.07241, "task3.loss_yaw0": 0.15037, "task3.loss_vel0": 0.00972, "task3.loss_heatmap0": 1.51813, "task4.loss_xy0": 0.06328, "task4.loss_z0": 0.04528, "task4.loss_whl0": 0.06737, "task4.loss_yaw0": 0.15269, "task4.loss_vel0": 0.19149, "task4.loss_heatmap0": 2.3083, "task5.loss_xy0": 0.06292, "task5.loss_z0": 0.05236, "task5.loss_whl0": 0.09441, "task5.loss_yaw0": 0.15268, "task5.loss_vel0": 0.10036, "task5.loss_heatmap0": 1.5121, "loss": 79.71912, "grad_norm": 120825.95078, "time": 20.79492}
{"mode": "train", "epoch": 12, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.38591, "loss_depth": 38.34557, "task0.loss_xy": 0.12553, "task0.loss_z": 0.0986, "task0.loss_whl": 0.0615, "task0.loss_yaw": 0.28912, "task0.loss_vel": 0.45171, "task0.loss_heatmap": 2.65313, "task1.loss_xy": 0.12619, "task1.loss_z": 0.13351, "task1.loss_whl": 0.16578, "task1.loss_yaw": 0.29279, "task1.loss_vel": 0.35269, "task1.loss_heatmap": 3.56907, "task2.loss_xy": 0.12477, "task2.loss_z": 0.11412, "task2.loss_whl": 0.14518, "task2.loss_yaw": 0.2949, "task2.loss_vel": 0.46327, "task2.loss_heatmap": 4.06396, "task3.loss_xy": 0.12567, "task3.loss_z": 0.07342, "task3.loss_whl": 0.14931, "task3.loss_yaw": 0.28479, "task3.loss_vel": 0.01857, "task3.loss_heatmap": 2.94587, "task4.loss_xy": 0.12632, "task4.loss_z": 0.09334, "task4.loss_whl": 0.13557, "task4.loss_yaw": 0.30445, "task4.loss_vel": 0.40041, "task4.loss_heatmap": 4.27597, "task5.loss_xy": 0.12547, "task5.loss_z": 0.10497, "task5.loss_whl": 0.19064, "task5.loss_yaw": 0.3048, "task5.loss_vel": 0.19613, "task5.loss_heatmap": 2.93794, "task0.loss_xy0": 0.06306, "task0.loss_z0": 0.04981, "task0.loss_whl0": 0.03185, "task0.loss_yaw0": 0.14992, "task0.loss_vel0": 0.22653, "task0.loss_heatmap0": 1.35728, "task1.loss_xy0": 0.06357, "task1.loss_z0": 0.06692, "task1.loss_whl0": 0.08354, "task1.loss_yaw0": 0.15039, "task1.loss_vel0": 0.17648, "task1.loss_heatmap0": 1.82415, "task2.loss_xy0": 0.06281, "task2.loss_z0": 0.05833, "task2.loss_whl0": 0.07404, "task2.loss_yaw0": 0.15175, "task2.loss_vel0": 0.22754, "task2.loss_heatmap0": 2.2612, "task3.loss_xy0": 0.06303, "task3.loss_z0": 0.03666, "task3.loss_whl0": 0.07575, "task3.loss_yaw0": 0.15033, "task3.loss_vel0": 0.01004, "task3.loss_heatmap0": 1.52305, "task4.loss_xy0": 0.06312, "task4.loss_z0": 0.04691, "task4.loss_whl0": 0.06832, "task4.loss_yaw0": 0.15311, "task4.loss_vel0": 0.20085, "task4.loss_heatmap0": 2.35107, "task5.loss_xy0": 0.06276, "task5.loss_z0": 0.05307, "task5.loss_whl0": 0.09656, "task5.loss_yaw0": 0.15258, "task5.loss_vel0": 0.09844, "task5.loss_heatmap0": 1.513, "loss": 78.46285, "grad_norm": 133941.42793, "time": 20.9949}
{"mode": "train", "epoch": 12, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.379, "loss_depth": 38.63896, "task0.loss_xy": 0.12504, "task0.loss_z": 0.09662, "task0.loss_whl": 0.06116, "task0.loss_yaw": 0.28729, "task0.loss_vel": 0.44511, "task0.loss_heatmap": 2.64928, "task1.loss_xy": 0.12502, "task1.loss_z": 0.13313, "task1.loss_whl": 0.16287, "task1.loss_yaw": 0.29225, "task1.loss_vel": 0.34326, "task1.loss_heatmap": 3.56121, "task2.loss_xy": 0.12479, "task2.loss_z": 0.11561, "task2.loss_whl": 0.14673, "task2.loss_yaw": 0.29711, "task2.loss_vel": 0.4542, "task2.loss_heatmap": 4.05444, "task3.loss_xy": 0.12516, "task3.loss_z": 0.07234, "task3.loss_whl": 0.1487, "task3.loss_yaw": 0.28449, "task3.loss_vel": 0.01834, "task3.loss_heatmap": 2.9447, "task4.loss_xy": 0.12421, "task4.loss_z": 0.09425, "task4.loss_whl": 0.13024, "task4.loss_yaw": 0.30431, "task4.loss_vel": 0.37476, "task4.loss_heatmap": 4.29107, "task5.loss_xy": 0.12562, "task5.loss_z": 0.105, "task5.loss_whl": 0.18676, "task5.loss_yaw": 0.30479, "task5.loss_vel": 0.19628, "task5.loss_heatmap": 2.94535, "task0.loss_xy0": 0.06287, "task0.loss_z0": 0.04884, "task0.loss_whl0": 0.03195, "task0.loss_yaw0": 0.14935, "task0.loss_vel0": 0.22367, "task0.loss_heatmap0": 1.35595, "task1.loss_xy0": 0.06328, "task1.loss_z0": 0.06713, "task1.loss_whl0": 0.08291, "task1.loss_yaw0": 0.15025, "task1.loss_vel0": 0.17126, "task1.loss_heatmap0": 1.81805, "task2.loss_xy0": 0.06286, "task2.loss_z0": 0.0585, "task2.loss_whl0": 0.0751, "task2.loss_yaw0": 0.15234, "task2.loss_vel0": 0.22621, "task2.loss_heatmap0": 2.22019, "task3.loss_xy0": 0.06282, "task3.loss_z0": 0.03584, "task3.loss_whl0": 0.07464, "task3.loss_yaw0": 0.15013, "task3.loss_vel0": 0.00923, "task3.loss_heatmap0": 1.51444, "task4.loss_xy0": 0.06267, "task4.loss_z0": 0.04817, "task4.loss_whl0": 0.06651, "task4.loss_yaw0": 0.15279, "task4.loss_vel0": 0.18521, "task4.loss_heatmap0": 2.34252, "task5.loss_xy0": 0.06296, "task5.loss_z0": 0.05242, "task5.loss_whl0": 0.09496, "task5.loss_yaw0": 0.1526, "task5.loss_vel0": 0.09859, "task5.loss_heatmap0": 1.51484, "loss": 78.59253, "grad_norm": 164143.59039, "time": 20.92337}
{"mode": "val", "epoch": 12, "iter": 753, "lr": 0.0002, "pts_bbox_NuScenes/car_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/car_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/car_trans_err": 1.2787, "pts_bbox_NuScenes/car_scale_err": 0.1976, "pts_bbox_NuScenes/car_orient_err": 0.6896, "pts_bbox_NuScenes/car_vel_err": 1.9859, "pts_bbox_NuScenes/car_attr_err": 0.7601, "pts_bbox_NuScenes/mATE": 1.0279, "pts_bbox_NuScenes/mASE": 0.9198, "pts_bbox_NuScenes/mAOE": 0.9655, "pts_bbox_NuScenes/mAVE": 1.1232, "pts_bbox_NuScenes/mAAE": 0.97, "pts_bbox_NuScenes/truck_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/truck_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/truck_trans_err": 1.0, "pts_bbox_NuScenes/truck_scale_err": 1.0, "pts_bbox_NuScenes/truck_orient_err": 1.0, "pts_bbox_NuScenes/truck_vel_err": 1.0, "pts_bbox_NuScenes/truck_attr_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_trans_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_scale_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_orient_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_vel_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_attr_err": 1.0, "pts_bbox_NuScenes/bus_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bus_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bus_trans_err": 1.0, "pts_bbox_NuScenes/bus_scale_err": 1.0, "pts_bbox_NuScenes/bus_orient_err": 1.0, "pts_bbox_NuScenes/bus_vel_err": 1.0, "pts_bbox_NuScenes/bus_attr_err": 1.0, "pts_bbox_NuScenes/trailer_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/trailer_trans_err": 1.0, "pts_bbox_NuScenes/trailer_scale_err": 1.0, "pts_bbox_NuScenes/trailer_orient_err": 1.0, "pts_bbox_NuScenes/trailer_vel_err": 1.0, "pts_bbox_NuScenes/trailer_attr_err": 1.0, "pts_bbox_NuScenes/barrier_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/barrier_trans_err": 1.0, "pts_bbox_NuScenes/barrier_scale_err": 1.0, "pts_bbox_NuScenes/barrier_orient_err": 1.0, "pts_bbox_NuScenes/barrier_vel_err": NaN, "pts_bbox_NuScenes/barrier_attr_err": NaN, "pts_bbox_NuScenes/motorcycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/motorcycle_trans_err": 1.0, "pts_bbox_NuScenes/motorcycle_scale_err": 1.0, "pts_bbox_NuScenes/motorcycle_orient_err": 1.0, "pts_bbox_NuScenes/motorcycle_vel_err": 1.0, "pts_bbox_NuScenes/motorcycle_attr_err": 1.0, "pts_bbox_NuScenes/bicycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bicycle_trans_err": 1.0, "pts_bbox_NuScenes/bicycle_scale_err": 1.0, "pts_bbox_NuScenes/bicycle_orient_err": 1.0, "pts_bbox_NuScenes/bicycle_vel_err": 1.0, "pts_bbox_NuScenes/bicycle_attr_err": 1.0, "pts_bbox_NuScenes/pedestrian_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/pedestrian_trans_err": 1.0, "pts_bbox_NuScenes/pedestrian_scale_err": 1.0, "pts_bbox_NuScenes/pedestrian_orient_err": 1.0, "pts_bbox_NuScenes/pedestrian_vel_err": 1.0, "pts_bbox_NuScenes/pedestrian_attr_err": 1.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/traffic_cone_trans_err": 1.0, "pts_bbox_NuScenes/traffic_cone_scale_err": 1.0, "pts_bbox_NuScenes/traffic_cone_orient_err": NaN, "pts_bbox_NuScenes/traffic_cone_vel_err": NaN, "pts_bbox_NuScenes/traffic_cone_attr_err": NaN, "pts_bbox_NuScenes/NDS": 0.01447, "pts_bbox_NuScenes/mAP": 0.0}
{"mode": "train", "epoch": 13, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 0.99281, "loss_depth": 37.36668, "task0.loss_xy": 0.12504, "task0.loss_z": 0.09728, "task0.loss_whl": 0.06089, "task0.loss_yaw": 0.2875, "task0.loss_vel": 0.433, "task0.loss_heatmap": 2.63874, "task1.loss_xy": 0.12474, "task1.loss_z": 0.13661, "task1.loss_whl": 0.16483, "task1.loss_yaw": 0.2911, "task1.loss_vel": 0.33862, "task1.loss_heatmap": 3.55371, "task2.loss_xy": 0.12533, "task2.loss_z": 0.11293, "task2.loss_whl": 0.14769, "task2.loss_yaw": 0.2961, "task2.loss_vel": 0.45573, "task2.loss_heatmap": 4.03338, "task3.loss_xy": 0.12565, "task3.loss_z": 0.07395, "task3.loss_whl": 0.14783, "task3.loss_yaw": 0.28178, "task3.loss_vel": 0.01789, "task3.loss_heatmap": 2.93707, "task4.loss_xy": 0.12453, "task4.loss_z": 0.09289, "task4.loss_whl": 0.13118, "task4.loss_yaw": 0.30363, "task4.loss_vel": 0.37539, "task4.loss_heatmap": 4.20944, "task5.loss_xy": 0.12526, "task5.loss_z": 0.10679, "task5.loss_whl": 0.18684, "task5.loss_yaw": 0.30446, "task5.loss_vel": 0.19285, "task5.loss_heatmap": 2.94023, "task0.loss_xy0": 0.06279, "task0.loss_z0": 0.04904, "task0.loss_whl0": 0.03102, "task0.loss_yaw0": 0.14935, "task0.loss_vel0": 0.21674, "task0.loss_heatmap0": 1.34919, "task1.loss_xy0": 0.06303, "task1.loss_z0": 0.06847, "task1.loss_whl0": 0.08323, "task1.loss_yaw0": 0.15007, "task1.loss_vel0": 0.16941, "task1.loss_heatmap0": 1.80745, "task2.loss_xy0": 0.06334, "task2.loss_z0": 0.05718, "task2.loss_whl0": 0.07486, "task2.loss_yaw0": 0.15186, "task2.loss_vel0": 0.22729, "task2.loss_heatmap0": 2.19014, "task3.loss_xy0": 0.06288, "task3.loss_z0": 0.03671, "task3.loss_whl0": 0.07466, "task3.loss_yaw0": 0.15007, "task3.loss_vel0": 0.00945, "task3.loss_heatmap0": 1.50982, "task4.loss_xy0": 0.06311, "task4.loss_z0": 0.04625, "task4.loss_whl0": 0.06652, "task4.loss_yaw0": 0.15275, "task4.loss_vel0": 0.18654, "task4.loss_heatmap0": 2.25864, "task5.loss_xy0": 0.06274, "task5.loss_z0": 0.05373, "task5.loss_whl0": 0.09494, "task5.loss_yaw0": 0.15249, "task5.loss_vel0": 0.09672, "task5.loss_heatmap0": 1.50568, "loss": 77.0157, "grad_norm": 165960.23625, "time": 21.46178}
{"mode": "train", "epoch": 13, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.36927, "loss_depth": 39.60399, "task0.loss_xy": 0.12495, "task0.loss_z": 0.09786, "task0.loss_whl": 0.06119, "task0.loss_yaw": 0.28661, "task0.loss_vel": 0.45547, "task0.loss_heatmap": 2.64917, "task1.loss_xy": 0.12542, "task1.loss_z": 0.13331, "task1.loss_whl": 0.16387, "task1.loss_yaw": 0.29164, "task1.loss_vel": 0.3534, "task1.loss_heatmap": 3.5725, "task2.loss_xy": 0.12575, "task2.loss_z": 0.11853, "task2.loss_whl": 0.1443, "task2.loss_yaw": 0.29308, "task2.loss_vel": 0.49016, "task2.loss_heatmap": 4.09894, "task3.loss_xy": 0.12498, "task3.loss_z": 0.0736, "task3.loss_whl": 0.14925, "task3.loss_yaw": 0.28302, "task3.loss_vel": 0.01762, "task3.loss_heatmap": 2.92614, "task4.loss_xy": 0.12525, "task4.loss_z": 0.09309, "task4.loss_whl": 0.13126, "task4.loss_yaw": 0.30202, "task4.loss_vel": 0.41069, "task4.loss_heatmap": 4.26038, "task5.loss_xy": 0.12462, "task5.loss_z": 0.10361, "task5.loss_whl": 0.1869, "task5.loss_yaw": 0.30491, "task5.loss_vel": 0.19414, "task5.loss_heatmap": 2.92617, "task0.loss_xy0": 0.06259, "task0.loss_z0": 0.04947, "task0.loss_whl0": 0.03106, "task0.loss_yaw0": 0.14936, "task0.loss_vel0": 0.22866, "task0.loss_heatmap0": 1.35425, "task1.loss_xy0": 0.06331, "task1.loss_z0": 0.06722, "task1.loss_whl0": 0.08262, "task1.loss_yaw0": 0.15038, "task1.loss_vel0": 0.17808, "task1.loss_heatmap0": 1.81306, "task2.loss_xy0": 0.06293, "task2.loss_z0": 0.05947, "task2.loss_whl0": 0.07267, "task2.loss_yaw0": 0.15152, "task2.loss_vel0": 0.24372, "task2.loss_heatmap0": 2.22674, "task3.loss_xy0": 0.06267, "task3.loss_z0": 0.0367, "task3.loss_whl0": 0.07492, "task3.loss_yaw0": 0.15017, "task3.loss_vel0": 0.00898, "task3.loss_heatmap0": 1.50159, "task4.loss_xy0": 0.06343, "task4.loss_z0": 0.04779, "task4.loss_whl0": 0.06673, "task4.loss_yaw0": 0.15263, "task4.loss_vel0": 0.20661, "task4.loss_heatmap0": 2.28839, "task5.loss_xy0": 0.06273, "task5.loss_z0": 0.05223, "task5.loss_whl0": 0.09463, "task5.loss_yaw0": 0.15269, "task5.loss_vel0": 0.09748, "task5.loss_heatmap0": 1.49808, "loss": 79.59336, "grad_norm": 202517.58242, "time": 21.33384}
{"mode": "train", "epoch": 13, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.37283, "loss_depth": 41.04097, "task0.loss_xy": 0.12567, "task0.loss_z": 0.09654, "task0.loss_whl": 0.06164, "task0.loss_yaw": 0.28642, "task0.loss_vel": 0.44867, "task0.loss_heatmap": 2.64858, "task1.loss_xy": 0.12529, "task1.loss_z": 0.135, "task1.loss_whl": 0.16593, "task1.loss_yaw": 0.29201, "task1.loss_vel": 0.3221, "task1.loss_heatmap": 3.54616, "task2.loss_xy": 0.12474, "task2.loss_z": 0.11462, "task2.loss_whl": 0.14895, "task2.loss_yaw": 0.2955, "task2.loss_vel": 0.44178, "task2.loss_heatmap": 4.03513, "task3.loss_xy": 0.12554, "task3.loss_z": 0.07508, "task3.loss_whl": 0.15021, "task3.loss_yaw": 0.28732, "task3.loss_vel": 0.01944, "task3.loss_heatmap": 2.93852, "task4.loss_xy": 0.12619, "task4.loss_z": 0.08977, "task4.loss_whl": 0.13546, "task4.loss_yaw": 0.30346, "task4.loss_vel": 0.38409, "task4.loss_heatmap": 4.23582, "task5.loss_xy": 0.12514, "task5.loss_z": 0.1041, "task5.loss_whl": 0.18791, "task5.loss_yaw": 0.30429, "task5.loss_vel": 0.19804, "task5.loss_heatmap": 2.94364, "task0.loss_xy0": 0.06295, "task0.loss_z0": 0.04849, "task0.loss_whl0": 0.03125, "task0.loss_yaw0": 0.1492, "task0.loss_vel0": 0.22369, "task0.loss_heatmap0": 1.35063, "task1.loss_xy0": 0.06293, "task1.loss_z0": 0.06812, "task1.loss_whl0": 0.08344, "task1.loss_yaw0": 0.15006, "task1.loss_vel0": 0.16236, "task1.loss_heatmap0": 1.79257, "task2.loss_xy0": 0.06323, "task2.loss_z0": 0.05823, "task2.loss_whl0": 0.07439, "task2.loss_yaw0": 0.1511, "task2.loss_vel0": 0.22086, "task2.loss_heatmap0": 2.16884, "task3.loss_xy0": 0.06293, "task3.loss_z0": 0.03756, "task3.loss_whl0": 0.07547, "task3.loss_yaw0": 0.14973, "task3.loss_vel0": 0.00996, "task3.loss_heatmap0": 1.50209, "task4.loss_xy0": 0.06314, "task4.loss_z0": 0.04597, "task4.loss_whl0": 0.06905, "task4.loss_yaw0": 0.15264, "task4.loss_vel0": 0.19336, "task4.loss_heatmap0": 2.25915, "task5.loss_xy0": 0.0627, "task5.loss_z0": 0.05264, "task5.loss_whl0": 0.09476, "task5.loss_yaw0": 0.15243, "task5.loss_vel0": 0.09947, "task5.loss_heatmap0": 1.50157, "loss": 80.69668, "grad_norm": 161916.16258, "time": 21.26432}
{"mode": "train", "epoch": 13, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.38289, "loss_depth": 40.74849, "task0.loss_xy": 0.12521, "task0.loss_z": 0.09731, "task0.loss_whl": 0.06148, "task0.loss_yaw": 0.2867, "task0.loss_vel": 0.45252, "task0.loss_heatmap": 2.64602, "task1.loss_xy": 0.12599, "task1.loss_z": 0.13194, "task1.loss_whl": 0.16612, "task1.loss_yaw": 0.29166, "task1.loss_vel": 0.3473, "task1.loss_heatmap": 3.56019, "task2.loss_xy": 0.12674, "task2.loss_z": 0.11274, "task2.loss_whl": 0.14561, "task2.loss_yaw": 0.2936, "task2.loss_vel": 0.46658, "task2.loss_heatmap": 4.05051, "task3.loss_xy": 0.12505, "task3.loss_z": 0.06995, "task3.loss_whl": 0.14387, "task3.loss_yaw": 0.2797, "task3.loss_vel": 0.01806, "task3.loss_heatmap": 2.94678, "task4.loss_xy": 0.12622, "task4.loss_z": 0.0954, "task4.loss_whl": 0.13184, "task4.loss_yaw": 0.30364, "task4.loss_vel": 0.38005, "task4.loss_heatmap": 4.26632, "task5.loss_xy": 0.12552, "task5.loss_z": 0.10247, "task5.loss_whl": 0.18719, "task5.loss_yaw": 0.30491, "task5.loss_vel": 0.19507, "task5.loss_heatmap": 2.93325, "task0.loss_xy0": 0.06291, "task0.loss_z0": 0.04936, "task0.loss_whl0": 0.03107, "task0.loss_yaw0": 0.14925, "task0.loss_vel0": 0.22724, "task0.loss_heatmap0": 1.34928, "task1.loss_xy0": 0.06325, "task1.loss_z0": 0.06593, "task1.loss_whl0": 0.08381, "task1.loss_yaw0": 0.15047, "task1.loss_vel0": 0.17445, "task1.loss_heatmap0": 1.80126, "task2.loss_xy0": 0.06355, "task2.loss_z0": 0.05725, "task2.loss_whl0": 0.07316, "task2.loss_yaw0": 0.15177, "task2.loss_vel0": 0.23343, "task2.loss_heatmap0": 2.17356, "task3.loss_xy0": 0.06277, "task3.loss_z0": 0.03449, "task3.loss_whl0": 0.07265, "task3.loss_yaw0": 0.14972, "task3.loss_vel0": 0.00913, "task3.loss_heatmap0": 1.50546, "task4.loss_xy0": 0.06353, "task4.loss_z0": 0.0476, "task4.loss_whl0": 0.06712, "task4.loss_yaw0": 0.15273, "task4.loss_vel0": 0.18849, "task4.loss_heatmap0": 2.26108, "task5.loss_xy0": 0.06295, "task5.loss_z0": 0.0513, "task5.loss_whl0": 0.09512, "task5.loss_yaw0": 0.15276, "task5.loss_vel0": 0.09777, "task5.loss_heatmap0": 1.49716, "loss": 80.5048, "grad_norm": 164685.57617, "time": 21.14536}
{"mode": "val", "epoch": 13, "iter": 753, "lr": 0.0002, "pts_bbox_NuScenes/car_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/car_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/car_trans_err": 1.2915, "pts_bbox_NuScenes/car_scale_err": 0.1992, "pts_bbox_NuScenes/car_orient_err": 0.7637, "pts_bbox_NuScenes/car_vel_err": 1.6765, "pts_bbox_NuScenes/car_attr_err": 0.6607, "pts_bbox_NuScenes/mATE": 1.0292, "pts_bbox_NuScenes/mASE": 0.9199, "pts_bbox_NuScenes/mAOE": 0.9737, "pts_bbox_NuScenes/mAVE": 1.0846, "pts_bbox_NuScenes/mAAE": 0.9576, "pts_bbox_NuScenes/truck_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/truck_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/truck_trans_err": 1.0, "pts_bbox_NuScenes/truck_scale_err": 1.0, "pts_bbox_NuScenes/truck_orient_err": 1.0, "pts_bbox_NuScenes/truck_vel_err": 1.0, "pts_bbox_NuScenes/truck_attr_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_trans_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_scale_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_orient_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_vel_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_attr_err": 1.0, "pts_bbox_NuScenes/bus_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bus_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bus_trans_err": 1.0, "pts_bbox_NuScenes/bus_scale_err": 1.0, "pts_bbox_NuScenes/bus_orient_err": 1.0, "pts_bbox_NuScenes/bus_vel_err": 1.0, "pts_bbox_NuScenes/bus_attr_err": 1.0, "pts_bbox_NuScenes/trailer_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/trailer_trans_err": 1.0, "pts_bbox_NuScenes/trailer_scale_err": 1.0, "pts_bbox_NuScenes/trailer_orient_err": 1.0, "pts_bbox_NuScenes/trailer_vel_err": 1.0, "pts_bbox_NuScenes/trailer_attr_err": 1.0, "pts_bbox_NuScenes/barrier_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/barrier_trans_err": 1.0, "pts_bbox_NuScenes/barrier_scale_err": 1.0, "pts_bbox_NuScenes/barrier_orient_err": 1.0, "pts_bbox_NuScenes/barrier_vel_err": NaN, "pts_bbox_NuScenes/barrier_attr_err": NaN, "pts_bbox_NuScenes/motorcycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/motorcycle_trans_err": 1.0, "pts_bbox_NuScenes/motorcycle_scale_err": 1.0, "pts_bbox_NuScenes/motorcycle_orient_err": 1.0, "pts_bbox_NuScenes/motorcycle_vel_err": 1.0, "pts_bbox_NuScenes/motorcycle_attr_err": 1.0, "pts_bbox_NuScenes/bicycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bicycle_trans_err": 1.0, "pts_bbox_NuScenes/bicycle_scale_err": 1.0, "pts_bbox_NuScenes/bicycle_orient_err": 1.0, "pts_bbox_NuScenes/bicycle_vel_err": 1.0, "pts_bbox_NuScenes/bicycle_attr_err": 1.0, "pts_bbox_NuScenes/pedestrian_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/pedestrian_trans_err": 1.0, "pts_bbox_NuScenes/pedestrian_scale_err": 1.0, "pts_bbox_NuScenes/pedestrian_orient_err": 1.0, "pts_bbox_NuScenes/pedestrian_vel_err": 1.0, "pts_bbox_NuScenes/pedestrian_attr_err": 1.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/traffic_cone_trans_err": 1.0, "pts_bbox_NuScenes/traffic_cone_scale_err": 1.0, "pts_bbox_NuScenes/traffic_cone_orient_err": NaN, "pts_bbox_NuScenes/traffic_cone_vel_err": NaN, "pts_bbox_NuScenes/traffic_cone_attr_err": NaN, "pts_bbox_NuScenes/NDS": 0.01487, "pts_bbox_NuScenes/mAP": 0.0}
{"mode": "train", "epoch": 14, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 1.0319, "loss_depth": 46.47417, "task0.loss_xy": 0.12549, "task0.loss_z": 0.09938, "task0.loss_whl": 0.06176, "task0.loss_yaw": 0.28587, "task0.loss_vel": 0.44775, "task0.loss_heatmap": 2.65396, "task1.loss_xy": 0.12521, "task1.loss_z": 0.13601, "task1.loss_whl": 0.16526, "task1.loss_yaw": 0.29105, "task1.loss_vel": 0.3376, "task1.loss_heatmap": 3.5538, "task2.loss_xy": 0.12627, "task2.loss_z": 0.11302, "task2.loss_whl": 0.14698, "task2.loss_yaw": 0.29181, "task2.loss_vel": 0.45428, "task2.loss_heatmap": 4.04557, "task3.loss_xy": 0.12593, "task3.loss_z": 0.07389, "task3.loss_whl": 0.14791, "task3.loss_yaw": 0.28189, "task3.loss_vel": 0.01771, "task3.loss_heatmap": 2.94198, "task4.loss_xy": 0.12652, "task4.loss_z": 0.09196, "task4.loss_whl": 0.1294, "task4.loss_yaw": 0.30417, "task4.loss_vel": 0.39151, "task4.loss_heatmap": 4.21568, "task5.loss_xy": 0.1252, "task5.loss_z": 0.1051, "task5.loss_whl": 0.18879, "task5.loss_yaw": 0.30419, "task5.loss_vel": 0.19589, "task5.loss_heatmap": 2.93927, "task0.loss_xy0": 0.06313, "task0.loss_z0": 0.05022, "task0.loss_whl0": 0.03129, "task0.loss_yaw0": 0.14908, "task0.loss_vel0": 0.22476, "task0.loss_heatmap0": 1.35392, "task1.loss_xy0": 0.06317, "task1.loss_z0": 0.06784, "task1.loss_whl0": 0.08349, "task1.loss_yaw0": 0.14985, "task1.loss_vel0": 0.17037, "task1.loss_heatmap0": 1.80052, "task2.loss_xy0": 0.06404, "task2.loss_z0": 0.05784, "task2.loss_whl0": 0.07449, "task2.loss_yaw0": 0.15124, "task2.loss_vel0": 0.22741, "task2.loss_heatmap0": 2.1634, "task3.loss_xy0": 0.06312, "task3.loss_z0": 0.03681, "task3.loss_whl0": 0.07483, "task3.loss_yaw0": 0.14982, "task3.loss_vel0": 0.00915, "task3.loss_heatmap0": 1.50611, "task4.loss_xy0": 0.06338, "task4.loss_z0": 0.04593, "task4.loss_whl0": 0.06613, "task4.loss_yaw0": 0.15312, "task4.loss_vel0": 0.19752, "task4.loss_heatmap0": 2.23259, "task5.loss_xy0": 0.06264, "task5.loss_z0": 0.05283, "task5.loss_whl0": 0.09573, "task5.loss_yaw0": 0.1523, "task5.loss_vel0": 0.09822, "task5.loss_heatmap0": 1.50135, "loss": 86.14987, "grad_norm": 104923.42766, "time": 21.37059}
{"mode": "train", "epoch": 14, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.42377, "loss_depth": 43.87746, "task0.loss_xy": 0.12514, "task0.loss_z": 0.09692, "task0.loss_whl": 0.06152, "task0.loss_yaw": 0.28591, "task0.loss_vel": 0.44886, "task0.loss_heatmap": 2.64542, "task1.loss_xy": 0.12599, "task1.loss_z": 0.1321, "task1.loss_whl": 0.16638, "task1.loss_yaw": 0.29062, "task1.loss_vel": 0.33299, "task1.loss_heatmap": 3.54567, "task2.loss_xy": 0.12705, "task2.loss_z": 0.1158, "task2.loss_whl": 0.14829, "task2.loss_yaw": 0.29335, "task2.loss_vel": 0.46565, "task2.loss_heatmap": 4.03343, "task3.loss_xy": 0.12502, "task3.loss_z": 0.07319, "task3.loss_whl": 0.1472, "task3.loss_yaw": 0.28261, "task3.loss_vel": 0.01844, "task3.loss_heatmap": 2.94442, "task4.loss_xy": 0.12653, "task4.loss_z": 0.09265, "task4.loss_whl": 0.13222, "task4.loss_yaw": 0.30183, "task4.loss_vel": 0.37925, "task4.loss_heatmap": 4.24881, "task5.loss_xy": 0.12554, "task5.loss_z": 0.10495, "task5.loss_whl": 0.18501, "task5.loss_yaw": 0.30429, "task5.loss_vel": 0.19385, "task5.loss_heatmap": 2.94448, "task0.loss_xy0": 0.06298, "task0.loss_z0": 0.04906, "task0.loss_whl0": 0.03104, "task0.loss_yaw0": 0.14915, "task0.loss_vel0": 0.22536, "task0.loss_heatmap0": 1.35107, "task1.loss_xy0": 0.06314, "task1.loss_z0": 0.06662, "task1.loss_whl0": 0.08375, "task1.loss_yaw0": 0.15029, "task1.loss_vel0": 0.16544, "task1.loss_heatmap0": 1.80113, "task2.loss_xy0": 0.06385, "task2.loss_z0": 0.05815, "task2.loss_whl0": 0.07382, "task2.loss_yaw0": 0.15136, "task2.loss_vel0": 0.22953, "task2.loss_heatmap0": 2.16371, "task3.loss_xy0": 0.06274, "task3.loss_z0": 0.03642, "task3.loss_whl0": 0.07483, "task3.loss_yaw0": 0.14995, "task3.loss_vel0": 0.00976, "task3.loss_heatmap0": 1.50259, "task4.loss_xy0": 0.06378, "task4.loss_z0": 0.04687, "task4.loss_whl0": 0.06694, "task4.loss_yaw0": 0.15321, "task4.loss_vel0": 0.18906, "task4.loss_heatmap0": 2.2531, "task5.loss_xy0": 0.06287, "task5.loss_z0": 0.05274, "task5.loss_whl0": 0.09356, "task5.loss_yaw0": 0.15249, "task5.loss_vel0": 0.09795, "task5.loss_heatmap0": 1.50067, "loss": 83.5578, "grad_norm": 102913.36957, "time": 20.83831}
{"mode": "train", "epoch": 14, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.42885, "loss_depth": 46.76983, "task0.loss_xy": 0.12493, "task0.loss_z": 0.09542, "task0.loss_whl": 0.06086, "task0.loss_yaw": 0.28516, "task0.loss_vel": 0.44736, "task0.loss_heatmap": 2.63827, "task1.loss_xy": 0.12542, "task1.loss_z": 0.13558, "task1.loss_whl": 0.16561, "task1.loss_yaw": 0.28968, "task1.loss_vel": 0.34993, "task1.loss_heatmap": 3.55286, "task2.loss_xy": 0.12703, "task2.loss_z": 0.10929, "task2.loss_whl": 0.14389, "task2.loss_yaw": 0.29239, "task2.loss_vel": 0.47141, "task2.loss_heatmap": 4.04555, "task3.loss_xy": 0.12607, "task3.loss_z": 0.07214, "task3.loss_whl": 0.14797, "task3.loss_yaw": 0.28616, "task3.loss_vel": 0.01829, "task3.loss_heatmap": 2.92073, "task4.loss_xy": 0.1256, "task4.loss_z": 0.0932, "task4.loss_whl": 0.13423, "task4.loss_yaw": 0.30107, "task4.loss_vel": 0.37759, "task4.loss_heatmap": 4.18742, "task5.loss_xy": 0.1253, "task5.loss_z": 0.10342, "task5.loss_whl": 0.18588, "task5.loss_yaw": 0.30384, "task5.loss_vel": 0.19855, "task5.loss_heatmap": 2.9323, "task0.loss_xy0": 0.06283, "task0.loss_z0": 0.04801, "task0.loss_whl0": 0.03112, "task0.loss_yaw0": 0.14907, "task0.loss_vel0": 0.22471, "task0.loss_heatmap0": 1.34931, "task1.loss_xy0": 0.06321, "task1.loss_z0": 0.06821, "task1.loss_whl0": 0.08426, "task1.loss_yaw0": 0.15038, "task1.loss_vel0": 0.17719, "task1.loss_heatmap0": 1.80202, "task2.loss_xy0": 0.06333, "task2.loss_z0": 0.05525, "task2.loss_whl0": 0.07267, "task2.loss_yaw0": 0.15169, "task2.loss_vel0": 0.23483, "task2.loss_heatmap0": 2.16324, "task3.loss_xy0": 0.06325, "task3.loss_z0": 0.03586, "task3.loss_whl0": 0.07549, "task3.loss_yaw0": 0.15052, "task3.loss_vel0": 0.00902, "task3.loss_heatmap0": 1.49424, "task4.loss_xy0": 0.06237, "task4.loss_z0": 0.04699, "task4.loss_whl0": 0.0685, "task4.loss_yaw0": 0.15235, "task4.loss_vel0": 0.18755, "task4.loss_heatmap0": 2.20698, "task5.loss_xy0": 0.06292, "task5.loss_z0": 0.05229, "task5.loss_whl0": 0.09416, "task5.loss_yaw0": 0.15228, "task5.loss_vel0": 0.09989, "task5.loss_heatmap0": 1.49701, "loss": 86.33321, "grad_norm": 174789.22547, "time": 20.9751}
{"mode": "train", "epoch": 14, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.43101, "loss_depth": 49.63238, "task0.loss_xy": 0.12522, "task0.loss_z": 0.09738, "task0.loss_whl": 0.06113, "task0.loss_yaw": 0.28427, "task0.loss_vel": 0.44764, "task0.loss_heatmap": 2.63712, "task1.loss_xy": 0.12459, "task1.loss_z": 0.13493, "task1.loss_whl": 0.1649, "task1.loss_yaw": 0.28918, "task1.loss_vel": 0.34509, "task1.loss_heatmap": 3.53735, "task2.loss_xy": 0.12631, "task2.loss_z": 0.11641, "task2.loss_whl": 0.14485, "task2.loss_yaw": 0.29163, "task2.loss_vel": 0.45845, "task2.loss_heatmap": 4.03136, "task3.loss_xy": 0.12507, "task3.loss_z": 0.07371, "task3.loss_whl": 0.14656, "task3.loss_yaw": 0.28012, "task3.loss_vel": 0.01887, "task3.loss_heatmap": 2.93659, "task4.loss_xy": 0.12571, "task4.loss_z": 0.09444, "task4.loss_whl": 0.13256, "task4.loss_yaw": 0.29916, "task4.loss_vel": 0.39557, "task4.loss_heatmap": 4.19188, "task5.loss_xy": 0.12505, "task5.loss_z": 0.10524, "task5.loss_whl": 0.18804, "task5.loss_yaw": 0.30418, "task5.loss_vel": 0.19445, "task5.loss_heatmap": 2.92849, "task0.loss_xy0": 0.06282, "task0.loss_z0": 0.04898, "task0.loss_whl0": 0.03097, "task0.loss_yaw0": 0.14887, "task0.loss_vel0": 0.22376, "task0.loss_heatmap0": 1.3465, "task1.loss_xy0": 0.06301, "task1.loss_z0": 0.06754, "task1.loss_whl0": 0.08303, "task1.loss_yaw0": 0.15036, "task1.loss_vel0": 0.17427, "task1.loss_heatmap0": 1.79269, "task2.loss_xy0": 0.06373, "task2.loss_z0": 0.05894, "task2.loss_whl0": 0.07335, "task2.loss_yaw0": 0.15151, "task2.loss_vel0": 0.2304, "task2.loss_heatmap0": 2.13893, "task3.loss_xy0": 0.0628, "task3.loss_z0": 0.03637, "task3.loss_whl0": 0.07465, "task3.loss_yaw0": 0.14991, "task3.loss_vel0": 0.00956, "task3.loss_heatmap0": 1.49823, "task4.loss_xy0": 0.06251, "task4.loss_z0": 0.04706, "task4.loss_whl0": 0.06665, "task4.loss_yaw0": 0.15212, "task4.loss_vel0": 0.19576, "task4.loss_heatmap0": 2.20993, "task5.loss_xy0": 0.06263, "task5.loss_z0": 0.05307, "task5.loss_whl0": 0.0954, "task5.loss_yaw0": 0.15243, "task5.loss_vel0": 0.09734, "task5.loss_heatmap0": 1.49235, "loss": 89.14433, "grad_norm": 183207.74695, "time": 20.8469}
{"mode": "val", "epoch": 14, "iter": 753, "lr": 0.0002, "pts_bbox_NuScenes/car_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/car_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/car_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/car_trans_err": 1.2989, "pts_bbox_NuScenes/car_scale_err": 0.2005, "pts_bbox_NuScenes/car_orient_err": 0.7706, "pts_bbox_NuScenes/car_vel_err": 2.0431, "pts_bbox_NuScenes/car_attr_err": 0.6862, "pts_bbox_NuScenes/mATE": 1.0299, "pts_bbox_NuScenes/mASE": 0.9201, "pts_bbox_NuScenes/mAOE": 0.9745, "pts_bbox_NuScenes/mAVE": 1.1304, "pts_bbox_NuScenes/mAAE": 0.9608, "pts_bbox_NuScenes/truck_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/truck_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/truck_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/truck_trans_err": 1.0, "pts_bbox_NuScenes/truck_scale_err": 1.0, "pts_bbox_NuScenes/truck_orient_err": 1.0, "pts_bbox_NuScenes/truck_vel_err": 1.0, "pts_bbox_NuScenes/truck_attr_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/construction_vehicle_trans_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_scale_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_orient_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_vel_err": 1.0, "pts_bbox_NuScenes/construction_vehicle_attr_err": 1.0, "pts_bbox_NuScenes/bus_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bus_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bus_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bus_trans_err": 1.0, "pts_bbox_NuScenes/bus_scale_err": 1.0, "pts_bbox_NuScenes/bus_orient_err": 1.0, "pts_bbox_NuScenes/bus_vel_err": 1.0, "pts_bbox_NuScenes/bus_attr_err": 1.0, "pts_bbox_NuScenes/trailer_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/trailer_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/trailer_trans_err": 1.0, "pts_bbox_NuScenes/trailer_scale_err": 1.0, "pts_bbox_NuScenes/trailer_orient_err": 1.0, "pts_bbox_NuScenes/trailer_vel_err": 1.0, "pts_bbox_NuScenes/trailer_attr_err": 1.0, "pts_bbox_NuScenes/barrier_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/barrier_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/barrier_trans_err": 1.0, "pts_bbox_NuScenes/barrier_scale_err": 1.0, "pts_bbox_NuScenes/barrier_orient_err": 1.0, "pts_bbox_NuScenes/barrier_vel_err": NaN, "pts_bbox_NuScenes/barrier_attr_err": NaN, "pts_bbox_NuScenes/motorcycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/motorcycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/motorcycle_trans_err": 1.0, "pts_bbox_NuScenes/motorcycle_scale_err": 1.0, "pts_bbox_NuScenes/motorcycle_orient_err": 1.0, "pts_bbox_NuScenes/motorcycle_vel_err": 1.0, "pts_bbox_NuScenes/motorcycle_attr_err": 1.0, "pts_bbox_NuScenes/bicycle_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/bicycle_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/bicycle_trans_err": 1.0, "pts_bbox_NuScenes/bicycle_scale_err": 1.0, "pts_bbox_NuScenes/bicycle_orient_err": 1.0, "pts_bbox_NuScenes/bicycle_vel_err": 1.0, "pts_bbox_NuScenes/bicycle_attr_err": 1.0, "pts_bbox_NuScenes/pedestrian_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/pedestrian_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/pedestrian_trans_err": 1.0, "pts_bbox_NuScenes/pedestrian_scale_err": 1.0, "pts_bbox_NuScenes/pedestrian_orient_err": 1.0, "pts_bbox_NuScenes/pedestrian_vel_err": 1.0, "pts_bbox_NuScenes/pedestrian_attr_err": 1.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_0.5": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_1.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_2.0": 0.0, "pts_bbox_NuScenes/traffic_cone_AP_dist_4.0": 0.0, "pts_bbox_NuScenes/traffic_cone_trans_err": 1.0, "pts_bbox_NuScenes/traffic_cone_scale_err": 1.0, "pts_bbox_NuScenes/traffic_cone_orient_err": NaN, "pts_bbox_NuScenes/traffic_cone_vel_err": NaN, "pts_bbox_NuScenes/traffic_cone_attr_err": NaN, "pts_bbox_NuScenes/NDS": 0.01447, "pts_bbox_NuScenes/mAP": 0.0}
{"mode": "train", "epoch": 15, "iter": 50, "lr": 0.0002, "memory": 54498, "data_time": 1.02477, "loss_depth": 54.6088, "task0.loss_xy": 0.12515, "task0.loss_z": 0.09866, "task0.loss_whl": 0.06128, "task0.loss_yaw": 0.28339, "task0.loss_vel": 0.45146, "task0.loss_heatmap": 2.63582, "task1.loss_xy": 0.12469, "task1.loss_z": 0.13267, "task1.loss_whl": 0.16555, "task1.loss_yaw": 0.28951, "task1.loss_vel": 0.33729, "task1.loss_heatmap": 3.52397, "task2.loss_xy": 0.1265, "task2.loss_z": 0.11457, "task2.loss_whl": 0.14748, "task2.loss_yaw": 0.29232, "task2.loss_vel": 0.46032, "task2.loss_heatmap": 4.01335, "task3.loss_xy": 0.12579, "task3.loss_z": 0.07375, "task3.loss_whl": 0.14472, "task3.loss_yaw": 0.28166, "task3.loss_vel": 0.01893, "task3.loss_heatmap": 2.94339, "task4.loss_xy": 0.12553, "task4.loss_z": 0.09144, "task4.loss_whl": 0.13256, "task4.loss_yaw": 0.30148, "task4.loss_vel": 0.39313, "task4.loss_heatmap": 4.18695, "task5.loss_xy": 0.12531, "task5.loss_z": 0.10264, "task5.loss_whl": 0.18599, "task5.loss_yaw": 0.30374, "task5.loss_vel": 0.1953, "task5.loss_heatmap": 2.93539, "task0.loss_xy0": 0.06275, "task0.loss_z0": 0.04968, "task0.loss_whl0": 0.03114, "task0.loss_yaw0": 0.14879, "task0.loss_vel0": 0.22676, "task0.loss_heatmap0": 1.34767, "task1.loss_xy0": 0.06285, "task1.loss_z0": 0.06658, "task1.loss_whl0": 0.08332, "task1.loss_yaw0": 0.15021, "task1.loss_vel0": 0.16813, "task1.loss_heatmap0": 1.78939, "task2.loss_xy0": 0.06335, "task2.loss_z0": 0.05821, "task2.loss_whl0": 0.0749, "task2.loss_yaw0": 0.15165, "task2.loss_vel0": 0.22887, "task2.loss_heatmap0": 2.12667, "task3.loss_xy0": 0.06304, "task3.loss_z0": 0.03661, "task3.loss_whl0": 0.07529, "task3.loss_yaw0": 0.15001, "task3.loss_vel0": 0.00953, "task3.loss_heatmap0": 1.50015, "task4.loss_xy0": 0.06292, "task4.loss_z0": 0.04607, "task4.loss_whl0": 0.06782, "task4.loss_yaw0": 0.15266, "task4.loss_vel0": 0.19631, "task4.loss_heatmap0": 2.20481, "task5.loss_xy0": 0.06274, "task5.loss_z0": 0.05195, "task5.loss_whl0": 0.09469, "task5.loss_yaw0": 0.1523, "task5.loss_vel0": 0.09781, "task5.loss_heatmap0": 1.49649, "loss": 94.07259, "grad_norm": 199028.93719, "time": 22.18509}
{"mode": "train", "epoch": 15, "iter": 100, "lr": 0.0002, "memory": 54498, "data_time": 0.41768, "loss_depth": 67.97311, "task0.loss_xy": 0.12527, "task0.loss_z": 0.09809, "task0.loss_whl": 0.06126, "task0.loss_yaw": 0.28584, "task0.loss_vel": 0.44575, "task0.loss_heatmap": 2.63259, "task1.loss_xy": 0.12527, "task1.loss_z": 0.13756, "task1.loss_whl": 0.166, "task1.loss_yaw": 0.29001, "task1.loss_vel": 0.3477, "task1.loss_heatmap": 3.52754, "task2.loss_xy": 0.12599, "task2.loss_z": 0.11547, "task2.loss_whl": 0.14534, "task2.loss_yaw": 0.29207, "task2.loss_vel": 0.46956, "task2.loss_heatmap": 4.04174, "task3.loss_xy": 0.1257, "task3.loss_z": 0.07259, "task3.loss_whl": 0.14347, "task3.loss_yaw": 0.28305, "task3.loss_vel": 0.01711, "task3.loss_heatmap": 2.91441, "task4.loss_xy": 0.12681, "task4.loss_z": 0.09661, "task4.loss_whl": 0.13432, "task4.loss_yaw": 0.30095, "task4.loss_vel": 0.39139, "task4.loss_heatmap": 4.15589, "task5.loss_xy": 0.12501, "task5.loss_z": 0.10522, "task5.loss_whl": 0.18698, "task5.loss_yaw": 0.30566, "task5.loss_vel": 0.19623, "task5.loss_heatmap": 2.92844, "task0.loss_xy0": 0.06299, "task0.loss_z0": 0.04957, "task0.loss_whl0": 0.03202, "task0.loss_yaw0": 0.14938, "task0.loss_vel0": 0.22213, "task0.loss_heatmap0": 1.35218, "task1.loss_xy0": 0.06349, "task1.loss_z0": 0.06919, "task1.loss_whl0": 0.08491, "task1.loss_yaw0": 0.15029, "task1.loss_vel0": 0.17553, "task1.loss_heatmap0": 1.79849, "task2.loss_xy0": 0.0635, "task2.loss_z0": 0.05932, "task2.loss_whl0": 0.07449, "task2.loss_yaw0": 0.15139, "task2.loss_vel0": 0.23246, "task2.loss_heatmap0": 2.15886, "task3.loss_xy0": 0.06291, "task3.loss_z0": 0.03628, "task3.loss_whl0": 0.07538, "task3.loss_yaw0": 0.14988, "task3.loss_vel0": 0.00962, "task3.loss_heatmap0": 1.49805, "task4.loss_xy0": 0.06373, "task4.loss_z0": 0.04817, "task4.loss_whl0": 0.06852, "task4.loss_yaw0": 0.15309, "task4.loss_vel0": 0.19274, "task4.loss_heatmap0": 2.19779, "task5.loss_xy0": 0.06298, "task5.loss_z0": 0.05332, "task5.loss_whl0": 0.0954, "task5.loss_yaw0": 0.15325, "task5.loss_vel0": 0.098, "task5.loss_heatmap0": 1.49615, "loss": 107.48143, "grad_norm": 133588.77598, "time": 21.48395}
{"mode": "train", "epoch": 15, "iter": 150, "lr": 0.0002, "memory": 54498, "data_time": 0.40668, "loss_depth": 67.41367, "task0.loss_xy": 0.12525, "task0.loss_z": 0.09634, "task0.loss_whl": 0.06174, "task0.loss_yaw": 0.2852, "task0.loss_vel": 0.44555, "task0.loss_heatmap": 2.63838, "task1.loss_xy": 0.12558, "task1.loss_z": 0.13255, "task1.loss_whl": 0.16654, "task1.loss_yaw": 0.28958, "task1.loss_vel": 0.34102, "task1.loss_heatmap": 3.54341, "task2.loss_xy": 0.12546, "task2.loss_z": 0.11505, "task2.loss_whl": 0.14851, "task2.loss_yaw": 0.29172, "task2.loss_vel": 0.44854, "task2.loss_heatmap": 4.02985, "task3.loss_xy": 0.12546, "task3.loss_z": 0.07337, "task3.loss_whl": 0.13946, "task3.loss_yaw": 0.27921, "task3.loss_vel": 0.01843, "task3.loss_heatmap": 2.93502, "task4.loss_xy": 0.12616, "task4.loss_z": 0.09148, "task4.loss_whl": 0.13181, "task4.loss_yaw": 0.30074, "task4.loss_vel": 0.39242, "task4.loss_heatmap": 4.19586, "task5.loss_xy": 0.12496, "task5.loss_z": 0.10455, "task5.loss_whl": 0.18619, "task5.loss_yaw": 0.30412, "task5.loss_vel": 0.19472, "task5.loss_heatmap": 2.92822, "task0.loss_xy0": 0.06279, "task0.loss_z0": 0.04862, "task0.loss_whl0": 0.03094, "task0.loss_yaw0": 0.14868, "task0.loss_vel0": 0.22295, "task0.loss_heatmap0": 1.34908, "task1.loss_xy0": 0.06278, "task1.loss_z0": 0.06655, "task1.loss_whl0": 0.08374, "task1.loss_yaw0": 0.15017, "task1.loss_vel0": 0.17271, "task1.loss_heatmap0": 1.79802, "task2.loss_xy0": 0.06325, "task2.loss_z0": 0.05811, "task2.loss_whl0": 0.07453, "task2.loss_yaw0": 0.15153, "task2.loss_vel0": 0.22137, "task2.loss_heatmap0": 2.13673, "task3.loss_xy0": 0.06294, "task3.loss_z0": 0.03637, "task3.loss_whl0": 0.07379, "task3.loss_yaw0": 0.14926, "task3.loss_vel0": 0.00961, "task3.loss_heatmap0": 1.49371, "task4.loss_xy0": 0.06265, "task4.loss_z0": 0.04658, "task4.loss_whl0": 0.0666, "task4.loss_yaw0": 0.15243, "task4.loss_vel0": 0.19465, "task4.loss_heatmap0": 2.19577, "task5.loss_xy0": 0.06281, "task5.loss_z0": 0.05239, "task5.loss_whl0": 0.09381, "task5.loss_yaw0": 0.15247, "task5.loss_vel0": 0.09766, "task5.loss_heatmap0": 1.49384, "loss": 106.87601, "grad_norm": 138266.66062, "time": 21.11937}
{"mode": "train", "epoch": 15, "iter": 200, "lr": 0.0002, "memory": 54498, "data_time": 0.40838, "loss_depth": 70.17884, "task0.loss_xy": 0.12502, "task0.loss_z": 0.09688, "task0.loss_whl": 0.0611, "task0.loss_yaw": 0.2831, "task0.loss_vel": 0.44751, "task0.loss_heatmap": 2.63139, "task1.loss_xy": 0.12541, "task1.loss_z": 0.13558, "task1.loss_whl": 0.16616, "task1.loss_yaw": 0.28955, "task1.loss_vel": 0.34297, "task1.loss_heatmap": 3.51373, "task2.loss_xy": 0.12694, "task2.loss_z": 0.11269, "task2.loss_whl": 0.14576, "task2.loss_yaw": 0.28749, "task2.loss_vel": 0.47795, "task2.loss_heatmap": 4.00812, "task3.loss_xy": 0.12529, "task3.loss_z": 0.07237, "task3.loss_whl": 0.13758, "task3.loss_yaw": 0.27905, "task3.loss_vel": 0.01851, "task3.loss_heatmap": 2.93214, "task4.loss_xy": 0.12455, "task4.loss_z": 0.09211, "task4.loss_whl": 0.13253, "task4.loss_yaw": 0.30025, "task4.loss_vel": 0.37858, "task4.loss_heatmap": 4.17708, "task5.loss_xy": 0.12507, "task5.loss_z": 0.10353, "task5.loss_whl": 0.1882, "task5.loss_yaw": 0.30415, "task5.loss_vel": 0.19759, "task5.loss_heatmap": 2.93479, "task0.loss_xy0": 0.06299, "task0.loss_z0": 0.04876, "task0.loss_whl0": 0.03092, "task0.loss_yaw0": 0.14881, "task0.loss_vel0": 0.22438, "task0.loss_heatmap0": 1.34522, "task1.loss_xy0": 0.06317, "task1.loss_z0": 0.06805, "task1.loss_whl0": 0.08376, "task1.loss_yaw0": 0.15038, "task1.loss_vel0": 0.17188, "task1.loss_heatmap0": 1.78366, "task2.loss_xy0": 0.06315, "task2.loss_z0": 0.05739, "task2.loss_whl0": 0.0736, "task2.loss_yaw0": 0.15118, "task2.loss_vel0": 0.23893, "task2.loss_heatmap0": 2.11517, "task3.loss_xy0": 0.06282, "task3.loss_z0": 0.03594, "task3.loss_whl0": 0.0742, "task3.loss_yaw0": 0.14945, "task3.loss_vel0": 0.00954, "task3.loss_heatmap0": 1.49138, "task4.loss_xy0": 0.06282, "task4.loss_z0": 0.04696, "task4.loss_whl0": 0.0668, "task4.loss_yaw0": 0.15283, "task4.loss_vel0": 0.19825, "task4.loss_heatmap0": 2.18225, "task5.loss_xy0": 0.06299, "task5.loss_z0": 0.05227, "task5.loss_whl0": 0.09472, "task5.loss_yaw0": 0.15241, "task5.loss_vel0": 0.09974, "task5.loss_heatmap0": 1.49292, "loss": 109.54923, "grad_norm": 193172.12914, "time": 21.26379}

Does it work for 2 camera views?

Hello! thanks for your great work!
I wonder weather HoP can work in a harder setting like 2 camera views instead of 6 cameras in nuscean dataset.
Actually, my robot has only 2 front view cameras.
Thank you!

Will open sota config on nuscenes benchmark ?

Some Issues with Replication

(1) I used mmcv.imread to read the data, because directly using mmcv.load as in your code would result in an error, indicating unsupported color.
(2) Using the configuration hop_bevdet4d-r50-depth.py and reading the data with mmcv.imread shouldn't cause any issues. However, the replicated accuracy is only around 0.36+ mAP.

Can you please provide more details about your configuration and setup? It's possible that there might be some misconfigurations or parameters that need adjustment. Additionally, double-checking the dataset, model settings, and any preprocessing steps could help identify any potential sources of the lower accuracy.

typeError: Unsupported format:color

hi. i got on error, but i have no idea to fix it.
Traceback (most recent call last):
File "tools/train.py", line 288, in
main()
File "tools/train.py", line 277, in main
train_model(
File "/workspace/HoP/mmdet3d/apis/train.py", line 344, in train_model
train_detector(
File "/workspace/HoP/mmdet3d/apis/train.py", line 319, in train_detector
runner.run(data_loaders, cfg.workflow)
File "/opt/conda/lib/python3.8/site-packages/mmcv/runner/epoch_based_runner.py", line 130, in run
epoch_runner(data_loaders[i], **kwargs)
File "/opt/conda/lib/python3.8/site-packages/mmcv/runner/epoch_based_runner.py", line 47, in train
for i, data_batch in enumerate(self.data_loader):
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 517, in next
data = self._next_data()
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 1199, in _next_data
return self._process_data(data)
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 1225, in _process_data
data.reraise()
File "/opt/conda/lib/python3.8/site-packages/torch/_utils.py", line 429, in reraise
raise self.exc_type(msg)
TypeError: Caught TypeError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/_utils/worker.py", line 202, in _worker_loop
data = fetcher.fetch(index)
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/opt/conda/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/workspace/HoP/mmdet3d/datasets/custom_3d.py", line 435, in getitem
data = self.prepare_train_data(idx)
File "/workspace/HoP/mmdet3d/datasets/custom_3d.py", line 229, in prepare_train_data
example = self.pipeline(input_dict)
File "/workspace/HoP/mmdet3d/datasets/pipelines/compose.py", line 49, in call
data = t(data)
File "/workspace/HoP/mmdet3d/datasets/pipelines/loading.py", line 1130, in call
results['img_inputs'] = self.get_inputs(results)
File "/workspace/HoP/mmdet3d/datasets/pipelines/loading.py", line 1019, in get_inputs
img = self.load_image(filename)
File "/workspace/HoP/mmdet3d/datasets/pipelines/loading.py", line 1151, in load_image
img_array = load_fun(filename, color_type)
File "/opt/conda/lib/python3.8/site-packages/mmcv/fileio/io.py", line 51, in load
raise TypeError(f'Unsupported format: {file_format}')
TypeError: Unsupported format: color

When will release code for HoP on BEVFormer?

Hi, @TempleX98 @Sense-X @CaraJ7
Thanks for your great work!
I'd like to know any plan to release code for HoP on BEVFormer?

Why is HoP useful even thought it is not used in inferring？

We can know it is useful when training because HoP can extract feature with spatial and temporal information. However, why it is also useful when inferring without HoP? How can model get the information about spatial and temporal?

Bevformer+HoP config,please

hi ,can you share with me the bevformer+hop config? many thanks! email : [email protected]

Why does the channel reduction can prune the height info ?

The paper claims that "Accordingly, we first employ a channel reduction operation to the input set Brem to prune the height information and achieve better training efficiency". But why channel reduction can represent height information ? And how do you do channel reduction ?

Can Hop used in PETR model?

How to get the ground truth of history frames?

Good job! I am wondering how do you get the ground truth of history frames for nuscenes dataset?

Pretrained weights for NuScenes detection

Hi all, congrats on your amazing results. Is it possible to share pretrained weights on NuScenes? I would like to try the model without training it from scratch.

Temporal BEV feature align

Dear author,
Hello,
Great Works! When will the code be released?
Before the code is released, I am wondering how do you align the BEV features at different timestamps in your short-term/long-term temporal decoder. Can you give some clues?
Thank you for your time.

Ablation study of the bev feature at timestamp t-k

Hi, I am wondering do you compare the results of the reconstructed bev feature with the real bev feature at timestamp t-k?

When will the SOTA model for nuScenes detection task be posted in model_zoo?

Hi,

I'm looking for the model that is on the nuScenes detection leaderboard. Will you be posting it in the model Zoo? Or is it the same as one of the models already listed in the model zoo? I'm wondering if there is a difference since the mAPs listed in the model zoo is much lower than the leaderboard model.

Thank you.

Segmentation Fault while inference over R50 Models

Describe the bug
Hi HoP authors,
Thank you for your great work. I am trying to run the inference over the released model. However, the code crashes with the following message:

None
mmdet3d_plugin
/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/mmdet/models/backbones/resnet.py:401: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead
  warnings.warn('DeprecationWarning: pretrained is deprecated, '
/home/abhinav/project/HoP/mmdet3d_plugin/hop/modules/custom_base_transformer_layer.py:94: UserWarning: The arguments `feedforward_channels` in BaseTransformerLayer has been deprecated, now you should set `feedforward_channels` and other FFN related arguments to a dict named `ffn_cfgs`. 
  warnings.warn(
/home/abhinav/project/HoP/mmdet3d_plugin/hop/modules/custom_base_transformer_layer.py:94: UserWarning: The arguments `ffn_dropout` in BaseTransformerLayer has been deprecated, now you should set `ffn_drop` and other FFN related arguments to a dict named `ffn_cfgs`. 
  warnings.warn(
/home/abhinav/project/HoP/mmdet3d_plugin/hop/modules/custom_base_transformer_layer.py:94: UserWarning: The arguments `ffn_num_fcs` in BaseTransformerLayer has been deprecated, now you should set `num_fcs` and other FFN related arguments to a dict named `ffn_cfgs`. 
  warnings.warn(
/home/abhinav/project/HoP/mmdet3d_plugin/hop/modules/temporal_cross_attention.py:88: UserWarning: You'd better set embed_dims in MultiScaleDeformAttention to make the dimension of each attention head a power of 2 which is more efficient in our CUDA implementation.
  warnings.warn(
load checkpoint from local path: work_dirs/HoP_BEVDet_ep24_ema.pth
[                                                  ] 0/6019, elapsed: 0s, ETA:/home/abhinav/project/HoP/mmdet3d/datasets/pipelines/loading.py:1221: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  cam2lidar_rt = torch.tensor(extrinsic)
/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py:210: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
  ranks_depth = torch.range(
/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py:212: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
  ranks_feat = torch.range(
/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py:220: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
  batch_idx = torch.range(0, B - 1).reshape(B, 1). \
[                                                  ] 13/6019, 0.5 task/s, elapsed: 24s, ETA: 11072sFatal Python error: Segmentation fault

Current thread 0x00007f2e4b84a5c0 (most recent call first):
  File "/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py", line 212 in voxel_pooling_prepare_v2
  File "/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py", line 171 in voxel_pooling_v2
  File "/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py", line 279 in view_transform_core
  File "/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py", line 287 in view_transform
  File "/home/abhinav/project/HoP/mmdet3d/models/necks/view_transformer.py", line 699 in forward
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889 in _call_impl
  File "/home/abhinav/project/HoP/mmdet3d_plugin/hop/detectors/hop_bevdet.py", line 148 in prepare_bev_feat
  File "/home/abhinav/project/HoP/mmdet3d_plugin/hop/detectors/hop_bevdet.py", line 232 in extract_img_feat
  File "/home/abhinav/project/HoP/mmdet3d_plugin/hop/detectors/hop_bevdet.py", line 185 in extract_feat
  File "/home/abhinav/project/HoP/mmdet3d_plugin/hop/detectors/hop_bevdet.py", line 343 in simple_test
  File "/home/abhinav/project/HoP/mmdet3d/models/detectors/bevdet.py", line 157 in forward_test
  File "/home/abhinav/project/HoP/mmdet3d/models/detectors/base.py", line 62 in forward
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/mmcv/runner/fp16_utils.py", line 110 in new_func
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889 in _call_impl
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/torch/nn/parallel/data_parallel.py", line 165 in forward
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/mmcv/parallel/data_parallel.py", line 50 in forward
  File "/home/abhinav/anaconda3/envs/hop2/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889 in _call_impl
  File "/home/abhinav/project/HoP/mmdet3d/apis/test.py", line 41 in single_gpu_test
  File "tools/test.py", line 260 in main
  File "tools/test.py", line 290 in <module>
Segmentation fault (core dumped)

Reproduction

What command or script did you run?

python tools/test.py configs/hop_bevdet/hop_bevdet4d-r50-depth.py work_dirs/HoP_BEVDet_ep24_ema.pth --eval bbox --deterministic

Did you make any modifications on the code or config? Did you understand what you have modified?
No
What dataset did you use?
nuScenes Val dataset

Environment

Please run python mmdet3d/utils/collect_env.py to collect necessary environment information and paste it here.

sys.platform: linux
Python: 3.8.18 (default, Sep 11 2023, 13:40:15) [GCC 11.2.0]
CUDA available: True
GPU 0: NVIDIA GeForce GTX 1080 Ti
CUDA_HOME: /usr/local/cuda-11.1
NVCC: Cuda compilation tools, release 11.1, V11.1.74
GCC: gcc (Ubuntu 8.4.0-1ubuntu1~18.04) 8.4.0
PyTorch: 1.8.1+cu111
PyTorch compiling details: PyTorch built with:
  - GCC 7.3
  - C++ Version: 201402
  - Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v1.7.0 (Git Hash 7aed236906b1f7a05c0917e5257a1af05e9ff683)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - NNPACK is enabled
  - CPU capability usage: AVX2
  - CUDA Runtime 11.1
  - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86
  - CuDNN 8.0.5
  - Magma 2.5.2
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -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=1.8.1, 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: 0.9.1+cu111
OpenCV: 4.8.1
MMCV: 1.5.2
MMCV Compiler: GCC 7.3
MMCV CUDA Compiler: 11.1
MMDetection: 2.24.0
MMSegmentation: 0.30.0
MMDetection3D: 1.0.0rc4+9f6e882
spconv2.0: False

You may add addition that may be helpful for locating the problem, such as
- How you installed PyTorch [e.g., pip, conda, source]
  I installed PyTorch by pip
- Other environment variables that may be related (such as $PATH, $LD_LIBRARY_PATH, $PYTHONPATH, etc.)
Installation instructions:

conda create -n hop2 python=3.8 -y
conda install -c anaconda ipython
pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html
pip install openmim
mim install mmcv-full==1.5.2
mim install mmengine
pip install mmdet==2.24.0
pip install mmsegmentation==0.30.0
pip install -e .
pip install numba
pip install numpy==1.23.5 timm einops

It would be great if you could let us know how to reproduce your R50 nuScenes Val results.

Update
The BEVDepth baseline model is also crashing. The code for running BEVDepth4D model is:

python tools/test.py configs/hop_bevdet/bevdet4d-r50-depth.py work_dirs/BEVDet_ep24_ema.pth --eval bbox --deterministic

KeyError: 'ann_info'

Hello, I am trying to prepare my dataset by typing the cmd:

python tools/create_data.py --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag nuscenes

as suggested in the tutorial but i get this error:

[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 81/81, 7.8 task/s, elapsed: 10s, ETA: 0s Create GT Database of NuScenesDataset [ ] 0/323, elapsed: 0s, ETA:{'sample_idx': 'ca9a282c9e77460f8360f564131a8af5', 'pts_filename': './data/nuscenes/samples/LIDAR_TOP/n015-2018-07-24-11-22-45+0800__LIDAR_TOP__1532402927647951.pcd.bin', 'sweeps': [], 'timestamp': 1532402927.647951, 'img_fields': [], 'bbox3d_fields': [], 'pts_mask_fields': [], 'pts_seg_fields': [], 'bbox_fields': [], 'mask_fields': [], 'seg_fields': [], 'box_type_3d': <class 'mmdet3d.core.bbox.structures.lidar_box3d.LiDARInstance3DBoxes'>, 'box_mode_3d': <Box3DMode.LIDAR: 0>} Traceback (most recent call last): File "tools/create_data.py", line 267, in nuscenes_data_prep( File "tools/create_data.py", line 89, in nuscenes_data_prep create_groundtruth_database(dataset_name, root_path, info_prefix, File "HoP/tools/data_converter/create_gt_database.py", line 241, in create_groundtruth_database example = dataset.pipeline(input_dict) File "HoP/mmdet3d/datasets/pipelines/compose.py", line 49, in call data = t(data) File "HoP/mmdet3d/datasets/pipelines/loading.py", line 682, in call results = self._load_bboxes_3d(results) File "HoP/mmdet3d/datasets/pipelines/loading.py", line 577, in _load_bboxes_3d results['gt_bboxes_3d'] = results['ann_info']['gt_bboxes_3d'] KeyError: 'ann_info'

Has anyone faced this problem yet ? Thank you

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.