search

PaddlePaddle / Paddle3D

A 3D computer vision development toolkit based on PaddlePaddle. It supports point-cloud object detection, segmentation, and monocular 3D object detection models.
Apache License 2.0
573 stars 141 forks source link

The evaluation results of the centerpoint model are all 0 but inference was fine #480

Open tongfengqi opened 1 day ago

tongfengqi commented 1 day ago

I trained the centerpoint model using the official KITTI dataset and evaluated the inference without any issues. Then I converted my dataset to the KITTI dataset format. Training was fine, inference was also fine, and the visualization of inference results was very accurate, but the evaluation results were all 0. May I ask what the possible reason may be?

我使用官方KITTI数据集训练centerpoint模型，评估推理没问题。然后把自己的数据集转成KITTI数据集的格式，训练没问题，推理也没问题，而且推理的结果可视化很正确，但是评估结果都是0。请问可能是什么原因?

This is my training log

2024-11-21 20:24:24,608 -     INFO - [TRAIN] epoch=160/160, iter=34690/34720 , total_loss=1.114057, lr=0.000000, batch_cost: 0.260677 sec, ips: 18.923797 images/s | ETA 00:00:06, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:24,608 -     INFO - [TRAIN] epoch=160/160, iter=34690/34720 , total_loss=1.114057, lr=0.000000, batch_cost: 0.260677 sec, ips: 18.923797 images/s | ETA 00:00:06, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:26,717 -     INFO - [TRAIN] epoch=160/160, iter=34700/34720 , total_loss=1.094697, lr=0.000000, batch_cost: 0.260662 sec, ips: 18.981339 images/s | ETA 00:00:04, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:26,717 -     INFO - [TRAIN] epoch=160/160, iter=34700/34720 , total_loss=1.094697, lr=0.000000, batch_cost: 0.260662 sec, ips: 18.981339 images/s | ETA 00:00:04, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:28,832 -     INFO - [TRAIN] epoch=160/160, iter=34710/34720 , total_loss=1.071305, lr=0.000000, batch_cost: 0.260648 sec, ips: 18.917163 images/s | ETA 00:00:02, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:28,832 -     INFO - [TRAIN] epoch=160/160, iter=34710/34720 , total_loss=1.071305, lr=0.000000, batch_cost: 0.260648 sec, ips: 18.917163 images/s | ETA 00:00:02, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:30,950 -     INFO - [TRAIN] epoch=160/160, iter=34720/34720 , total_loss=1.087590, lr=0.000000, batch_cost: 0.260634 sec, ips: 18.895549 images/s | ETA 00:00:00, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:30,950 -     INFO - [TRAIN] epoch=160/160, iter=34720/34720 , total_loss=1.087590, lr=0.000000, batch_cost: 0.260634 sec, ips: 18.895549 images/s | ETA 00:00:00, max_mem_reserved: 24085 MB, max_mem_allocated: 10682 MB
2024-11-21 20:24:30,964 -     INFO - Pop model from ./output_kitti_custom_1121/epoch_135
2024-11-21 20:24:30,964 -     INFO - Pop model from ./output_kitti_custom_1121/epoch_135
2024-11-21 20:24:31,232 -     INFO - Push model to checkpoint ./output_kitti_custom_1121/epoch_160
2024-11-21 20:24:31,232 -     INFO - Push model to checkpoint ./output_kitti_custom_1121/epoch_160
2024-11-21 20:24:33,258 -     INFO - Training is complete.
2024-11-21 20:24:33,258 -     INFO - Training is complete.

this is my inference result:

ortools not installed, install it by "pip install ortools==9.1.9490" if you run BEVLaneDet model
2024-11-22 14:50:59,372 -  WARNING - No custom op centerpoint_postprocess found, try JIT build
Compiling user custom op, it will cost a few seconds.....
[2024-11-22 14:51:00,601] [    INFO] dist.py:985 - running build
[2024-11-22 14:51:00,602] [    INFO] dist.py:985 - running build_ext
2024-11-22 14:51:00,892 - INFO - using custom operator only
[2024-11-22 14:51:00,892] [    INFO] extension_utils.py:1489 - using custom operator only
2024-11-22 14:51:00,894 -     INFO - centerpoint_postprocess builded success!
2024-11-22 14:51:00,894 -  WARNING - No custom op voxelize found, try JIT build
Compiling user custom op, it will cost a few seconds.....
[2024-11-22 14:51:02,168] [    INFO] dist.py:985 - running build
[2024-11-22 14:51:02,168] [    INFO] dist.py:985 - running build_ext
W1122 14:51:02.452482 3178968 custom_operator.cc:968] Operator (centerpoint_postprocess) has been registered.
2024-11-22 14:51:02,458 - INFO - using custom operator only
[2024-11-22 14:51:02,458] [    INFO] extension_utils.py:1489 - using custom operator only
2024-11-22 14:51:02,459 -     INFO - voxelize builded success!
I1122 14:51:02.462194 3178968 helper.cc:56] The operator `hard_voxelize` has been registered. Therefore, we will not repeat the registration here.
I1122 14:51:02.462210 3178968 helper.cc:56] The operator `centerpoint_postprocess` has been registered. Therefore, we will not repeat the registration here.
--- Running analysis [ir_graph_build_pass]
I1122 14:51:03.580410 3178968 executor.cc:187] Old Executor is Running.
--- Running analysis [ir_analysis_pass]
--- Running IR pass [map_op_to_another_pass]
--- Running IR pass [is_test_pass]
--- Running IR pass [simplify_with_basic_ops_pass]
--- Running IR pass [delete_quant_dequant_linear_op_pass]
--- Running IR pass [delete_weight_dequant_linear_op_pass]
--- Running IR pass [constant_folding_pass]
I1122 14:51:03.649338 3178968 fuse_pass_base.cc:59] ---  detected 18 subgraphs
--- Running IR pass [silu_fuse_pass]
--- Running IR pass [conv_bn_fuse_pass]
I1122 14:51:03.665633 3178968 fuse_pass_base.cc:59] ---  detected 18 subgraphs
--- Running IR pass [conv_eltwiseadd_bn_fuse_pass]
W1122 14:51:03.672838 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672847 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
W1122 14:51:03.672852 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672855 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
W1122 14:51:03.672860 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672863 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
W1122 14:51:03.672870 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672873 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
W1122 14:51:03.672878 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672881 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
W1122 14:51:03.672885 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_eltwiseadd_bn_fuse_pass) failed!
W1122 14:51:03.672888 3178968 conv_bn_fuse_pass.cc:632] Pass in op compat failed.
--- Running IR pass [embedding_eltwise_layernorm_fuse_pass]
--- Running IR pass [multihead_matmul_fuse_pass_v2]
--- Running IR pass [vit_attention_fuse_pass]
--- Running IR pass [fused_multi_transformer_encoder_pass]
--- Running IR pass [fused_multi_transformer_decoder_pass]
--- Running IR pass [fused_multi_transformer_encoder_fuse_qkv_pass]
--- Running IR pass [fused_multi_transformer_decoder_fuse_qkv_pass]
--- Running IR pass [multi_devices_fused_multi_transformer_encoder_pass]
--- Running IR pass [multi_devices_fused_multi_transformer_encoder_fuse_qkv_pass]
--- Running IR pass [multi_devices_fused_multi_transformer_decoder_fuse_qkv_pass]
--- Running IR pass [fuse_multi_transformer_layer_pass]
--- Running IR pass [gpu_cpu_squeeze2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_reshape2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_flatten2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_map_matmul_v2_to_mul_pass]
I1122 14:51:03.901919 3178968 fuse_pass_base.cc:59] ---  detected 1 subgraphs
--- Running IR pass [gpu_cpu_map_matmul_v2_to_matmul_pass]
W1122 14:51:03.902436 3178968 gpu_cpu_map_matmul_to_mul_pass.cc:425] matmul op not support broadcast, please check inputs'shape. 
--- Running IR pass [matmul_scale_fuse_pass]
--- Running IR pass [multihead_matmul_fuse_pass_v3]
--- Running IR pass [gpu_cpu_map_matmul_to_mul_pass]
--- Running IR pass [fc_fuse_pass]
--- Running IR pass [fc_elementwise_layernorm_fuse_pass]
--- Running IR pass [conv_elementwise_add_act_fuse_pass]
I1122 14:51:03.922371 3178968 fuse_pass_base.cc:59] ---  detected 18 subgraphs
--- Running IR pass [conv_elementwise_add2_act_fuse_pass]
--- Running IR pass [conv_elementwise_add_fuse_pass]
W1122 14:51:03.926470 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926481 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926486 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926488 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926492 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926496 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926501 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926503 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926507 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926510 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926514 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926517 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926522 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926524 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926527 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926532 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926534 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926537 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926541 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926544 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
W1122 14:51:03.926548 3178968 op_compat_sensible_pass.cc:232]  Check the Attr(axis) of Op(elementwise_add) in pass(conv_elementwise_add_fuse_pass) failed!
W1122 14:51:03.926551 3178968 conv_elementwise_add_fuse_pass.cc:94] Pass in op compat failed.
--- Running IR pass [transpose_flatten_concat_fuse_pass]
--- Running IR pass [fused_conv2d_add_act_layout_transfer_pass]
--- Running IR pass [transfer_layout_elim_pass]
I1122 14:51:03.933919 3178968 transfer_layout_elim_pass.cc:346] move down 0 transfer_layout
I1122 14:51:03.933924 3178968 transfer_layout_elim_pass.cc:347] eliminate 0 pair of transfer_layout
--- Running IR pass [auto_mixed_precision_pass]
--- Running IR pass [identity_op_clean_pass]
I1122 14:51:03.938210 3178968 fuse_pass_base.cc:59] ---  detected 6 subgraphs
--- Running IR pass [inplace_op_var_pass]
I1122 14:51:03.938984 3178968 fuse_pass_base.cc:59] ---  detected 19 subgraphs
--- Running analysis [save_optimized_model_pass]
--- Running analysis [ir_params_sync_among_devices_pass]
I1122 14:51:03.939541 3178968 ir_params_sync_among_devices_pass.cc:53] Sync params from CPU to GPU
--- Running analysis [adjust_cudnn_workspace_size_pass]
--- Running analysis [inference_op_replace_pass]
--- Running analysis [memory_optimize_pass]
I1122 14:51:05.276466 3178968 memory_optimize_pass.cc:118] The persistable params in main graph are : 70.6412MB
I1122 14:51:05.278610 3178968 memory_optimize_pass.cc:246] Cluster name : _generated_var_7_slice_0  size: 4
I1122 14:51:05.278617 3178968 memory_optimize_pass.cc:246] Cluster name : conv2d_20.tmp_1  size: 428544
I1122 14:51:05.278623 3178968 memory_optimize_pass.cc:246] Cluster name : relu_11.tmp_0  size: 27426816
I1122 14:51:05.278626 3178968 memory_optimize_pass.cc:246] Cluster name : tmp_6  size: 4
I1122 14:51:05.278630 3178968 memory_optimize_pass.cc:246] Cluster name : relu_18.tmp_0  size: 27426816
I1122 14:51:05.278632 3178968 memory_optimize_pass.cc:246] Cluster name : transpose_8.tmp_0  size: 54853632
I1122 14:51:05.278635 3178968 memory_optimize_pass.cc:246] Cluster name : relu_4.tmp_0  size: 54853632
I1122 14:51:05.278636 3178968 memory_optimize_pass.cc:246] Cluster name : conv2d_22.tmp_1  size: 214272
I1122 14:51:05.278640 3178968 memory_optimize_pass.cc:246] Cluster name : concat_2.tmp_0  size: 82280448
--- Running analysis [ir_graph_to_program_pass]
I1122 14:51:05.302268 3178968 analysis_predictor.cc:1838] ======= optimize end =======
I1122 14:51:05.303556 3178968 naive_executor.cc:200] ---  skip [feed], feed -> data
I1122 14:51:05.305168 3178968 naive_executor.cc:200] ---  skip [save_infer_model/scale_0.tmp_0], fetch -> fetch
I1122 14:51:05.305179 3178968 naive_executor.cc:200] ---  skip [save_infer_model/scale_1.tmp_0], fetch -> fetch
I1122 14:51:05.305184 3178968 naive_executor.cc:200] ---  skip [save_infer_model/scale_2.tmp_0], fetch -> fetch
W1122 14:51:05.416098 3178968 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.0, Driver API Version: 12.0, Runtime API Version: 11.7
W1122 14:51:05.425606 3178968 gpu_resources.cc:164] device: 0, cuDNN Version: 8.5.
Score: 0.9749983549118042 Label: 0 Box(x_c, y_c, z_c, w, l, h, -rot): 39.909244537353516 -4.224442005157471 -1.3094401359558105 6.859255790710449 10.765372276306152 5.646881580352783 -2.280179977416992
Score: 0.9394095540046692 Label: 0 Box(x_c, y_c, z_c, w, l, h, -rot): 50.38455581665039 15.922264099121094 -0.9328028559684753 6.891967296600342 10.820916175842285 5.673035144805908 1.295194387435913

这是我的评估结果：

ortools not installed, install it by "pip install ortools==9.1.9490" if you run BEVLaneDet model
W1122 14:43:15.693286 3164589 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.0, Driver API Version: 12.0, Runtime API Version: 11.7
W1122 14:43:15.697922 3164589 gpu_resources.cc:164] device: 0, cuDNN Version: 8.5.
2024-11-22 14:43:19,199 -     INFO - Finish CenterHead Initialization
2024-11-22 14:43:19,207 -     INFO - Load 830 truck database infos
2024-11-22 14:43:19,207 -     INFO - After filtering min_num_points_in_box:
2024-11-22 14:43:19,207 -     INFO - Load 830 truck database infos
2024-11-22 14:43:19,209 -     INFO - After filtering ignored difficulty:
2024-11-22 14:43:19,209 -     INFO - Load 830 truck database infos
2024-11-22 14:43:19,245 -     INFO - There are 151/151 variables loaded into CenterPoint.
2024-11-22 14:43:19,822 -     INFO - evaluate on validate dataset
2024-11-22 14:43:19,822 -     INFO - evaluate on validate dataset
2024-11-22 14:43:20,579 -     INFO - [                                                  ] 0.00%
2024-11-22 14:43:20,581 -  WARNING - No custom op voxelize found, try JIT build         ] 0.00%
2024-11-22 14:43:20,581 -  WARNING - No custom op voxelize found, try JIT build
Compiling user custom op, it will cost a few seconds.....
[2024-11-22 14:43:21,911] [    INFO] dist.py:985 - running build
[2024-11-22 14:43:21,911] [    INFO] dist.py:985 - running build_ext
2024-11-22 14:43:22,129 - INFO - using custom operator only
[2024-11-22 14:43:22,129] [    INFO] extension_utils.py:1489 - using custom operator only
2024-11-22 14:43:22,130 -     INFO - voxelize builded success!
2024-11-22 14:43:22,130 -     INFO - voxelize builded success!
2024-11-22 14:43:25,320 -  WARNING - No custom op centerpoint_postprocess found, try JIT build
2024-11-22 14:43:25,320 -  WARNING - No custom op centerpoint_postprocess found, try JIT build
Compiling user custom op, it will cost a few seconds.....
[2024-11-22 14:43:27,049] [    INFO] dist.py:985 - running build
[2024-11-22 14:43:27,049] [    INFO] dist.py:985 - running build_ext
W1122 14:43:27.220949 3164589 custom_operator.cc:968] Operator (hard_voxelize) has been registered.
2024-11-22 14:43:27,227 - INFO - using custom operator only
[2024-11-22 14:43:27,227] [    INFO] extension_utils.py:1489 - using custom operator only
2024-11-22 14:43:27,229 -     INFO - centerpoint_postprocess builded success!
2024-11-22 14:43:27,229 -     INFO - centerpoint_postprocess builded success!
2024-11-22 14:43:27,232 -     INFO - [                                                  ] 0.46%
2024-11-22 14:43:27,395 -     INFO - [#                                                 ] 3.23%
2024-11-22 14:43:27,504 -     INFO - [##                                                ] 5.99%
2024-11-22 14:43:27,607 -     INFO - [####                                              ] 8.76%
2024-11-22 14:43:27,711 -     INFO - [#####                                             ] 11.52%
2024-11-22 14:43:27,829 -     INFO - [#######                                           ] 14.75%
2024-11-22 14:43:27,931 -     INFO - [########                                          ] 17.51%
2024-11-22 14:43:28,034 -     INFO - [##########                                        ] 20.28%
2024-11-22 14:43:28,136 -     INFO - [###########                                       ] 23.04%
2024-11-22 14:43:28,253 -     INFO - [#############                                     ] 26.27%
2024-11-22 14:43:28,356 -     INFO - [##############                                    ] 29.03%
2024-11-22 14:43:28,461 -     INFO - [###############                                   ] 31.80%
2024-11-22 14:43:28,566 -     INFO - [#################                                 ] 34.56%
2024-11-22 14:43:28,668 -     INFO - [##################                                ] 37.33%
2024-11-22 14:43:28,771 -     INFO - [###################                               ] 38.25%
2024-11-22 14:43:28,875 -     INFO - [####################                              ] 41.01%
2024-11-22 14:43:28,977 -     INFO - [#####################                             ] 43.78%
2024-11-22 14:43:29,078 -     INFO - [#######################                           ] 46.54%
2024-11-22 14:43:29,178 -     INFO - [########################                          ] 49.31%
2024-11-22 14:43:29,280 -     INFO - [##########################                        ] 52.07%
2024-11-22 14:43:29,384 -     INFO - [###########################                       ] 54.84%
2024-11-22 14:43:29,504 -     INFO - [#############################                     ] 58.06%
2024-11-22 14:43:29,608 -     INFO - [##############################                    ] 60.83%
2024-11-22 14:43:29,711 -     INFO - [###############################                   ] 63.59%
2024-11-22 14:43:29,812 -     INFO - [#################################                 ] 66.36%
2024-11-22 14:43:29,916 -     INFO - [##################################                ] 69.12%
2024-11-22 14:43:30,018 -     INFO - [###################################               ] 71.89%
2024-11-22 14:43:30,121 -     INFO - [#####################################             ] 74.65%
2024-11-22 14:43:30,224 -     INFO - [######################################            ] 77.42%
2024-11-22 14:43:30,330 -     INFO - [########################################          ] 80.18%
2024-11-22 14:43:30,431 -     INFO - [#########################################         ] 82.95%
2024-11-22 14:43:30,548 -     INFO - [###########################################       ] 86.18%
2024-11-22 14:43:30,650 -     INFO - [############################################      ] 88.94%
2024-11-22 14:43:30,751 -     INFO - [#############################################     ] 91.71%
2024-11-22 14:43:30,854 -     INFO - [###############################################   ] 94.47%
2024-11-22 14:43:30,958 -     INFO - [################################################  ] 97.24%
2024-11-22 14:43:31,094 -     INFO - [##################################################] 100.00%
2024-11-22 14:43:31,094 -     INFO - [##################################################] 100.00%
2024-11-22 14:43:31,094 -     INFO - 
/home/ma-user/work/Paddle3D/paddle3d/geometries/bbox.py:721: RuntimeWarning: invalid value encountered in divide
  point_2d_res = point_2d[..., :2] / point_2d[..., 2:3]
2024-11-22 14:43:42,965 -     INFO - truck:
2024-11-22 14:43:42,965 -     INFO - truck:
2024-11-22 14:43:42,966 -     INFO - BBOX AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BBOX AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BEV  AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BEV  AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - 3D   AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - 3D   AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BBOX AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BBOX AP_R40@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BEV  AP_R40@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - BEV  AP_R40@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - 3D   AP_R40@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - 3D   AP_R40@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,966 -     INFO - truck:
2024-11-22 14:43:42,966 -     INFO - truck:
2024-11-22 14:43:42,967 -     INFO - BBOX AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BBOX AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BEV  AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BEV  AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - 3D   AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - 3D   AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BBOX AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BBOX AP_R11@70%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BEV  AP_R11@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - BEV  AP_R11@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - 3D   AP_R11@50%: 0.00 0.00 0.00
2024-11-22 14:43:42,967 -     INFO - 3D   AP_R11@50%: 0.00 0.00 0.00
tongfengqi commented 1 day ago 
------------Environment Information-------------
platform:
    Linux-4.18.0-147.5.2.15.h1109.eulerosv2r10.x86_64-x86_64-with-glibc2.31
    gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
    Python - 3.9.11 (main, Mar 29 2022, 19:08:29)  [GCC 7.5.0]

Science Toolkits:
    cv2 - 4.5.5
    numpy - 1.23.5
    numba - 0.60.0
    pandas - 1.5.1
    pillow - 9.3.0
    skimage - 0.24.0

PaddlePaddle:
    paddle(gpu) - 2.6.1
    paddle3d - 1.0.0
    paddleseg - 2.8.0
    FLAGS_cudnn_deterministic - Not set.
    FLAGS_cudnn_exhaustive_search - Not set.

CUDA:
    cudnn - 8500
    nvcc - Build cuda_11.7.r11.7/compiler.31442593_0