[Bug] Pointpillars：how to set the dataset path about converting to onnx when run on KITTI dataset #2839
Comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
{{ message }}
Checklist
Describe the bug
Mine：python mmdeploy/tools/deploy.py mmdeploy/configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime_dynamic.py mmdetection3d/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py mmdetection3d/work_dir/hv_pointpillars_secfpn_6x8_160e_kitti-3d-car_20220331_134606-d42d15ed.pth mmdetection3d/data/kitti/testing/velodyne/000001.bin --work-dir mmdetection3d/mmdeploy_model/pointpillars
when i run tools/test.py like python mmdeploy/tools/test.py mmdeploy/configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime-fp16_dynamic.py mmdetection3d/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py --model mmdetection3d/mmdeploy_model/pointpillars/end2end.onnx --device cpu --show-dir mmdetection3d/mmdeploy_model/pointpillars, it will leads to Traceback (most recent call last):
File "/root/model/mmdeploy/tools/test.py", line 159, in
main()
File "/root/model/mmdeploy/tools/test.py", line 153, in main
runner.test()
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1823, in test
metrics = self.test_loop.run() # type: ignore
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/runner/loops.py", line 463, in run
self.run_iter(idx, data_batch)
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/runner/loops.py", line 492, in run_iter
self.runner.call_hook(
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/runner/runner.py", line 1839, in call_hook
getattr(hook, fn_name)(self, **kwargs)
File "/root/model/mmdetection3d/mmdet3d/engine/hooks/visualization_hook.py", line 228, in after_test_iter
self._visualizer.add_datasample(
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/dist/utils.py", line 427, in wrapper
return func(*args, **kwargs)
File "/root/model/mmdetection3d/mmdet3d/visualization/local_visualizer.py", line 1031, in add_datasample
pred_data_3d = self._draw_instances_3d(data_input,
File "/root/model/mmdetection3d/mmdet3d/visualization/local_visualizer.py", line 707, in _draw_instances_3d
self.draw_proj_bboxes_3d(
File "/root/miniconda3/envs/mmdet/lib/python3.10/site-packages/mmengine/dist/utils.py", line 427, in wrapper
return func(*args, **kwargs)
File "/root/model/mmdetection3d/mmdet3d/visualization/local_visualizer.py", line 533, in draw_proj_bboxes_3d
corners_2d = proj_bbox3d_to_img(bboxes_3d, input_meta)
File "/root/model/mmdetection3d/mmdet3d/visualization/vis_utils.py", line 105, in proj_lidar_bbox3d_to_img
lidar2img = copy.deepcopy(input_meta['lidar2img']).reshape(4, 4)
AttributeError: 'list' object has no attribute 'reshape'
Reproduction
python mmdeploy/tools/deploy.py mmdeploy/configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime_dynamic.py mmdetection3d/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py mmdetection3d/work_dir/hv_pointpillars_secfpn_6x8_160e_kitti-3d-car_20220331_134606-d42d15ed.pth mmdetection3d/data/kitti/testing/velodyne/000001.bin --work-dir mmdetection3d/mmdeploy_model/pointpillars
python mmdeploy/tools/test.py mmdeploy/configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime-fp16_dynamic.py mmdetection3d/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py --model mmdetection3d/mmdeploy_model/pointpillars/end2end.onnx --device cpu --show-dir mmdetection3d/mmdeploy_model/pointpillars
Environment
Error traceback
No response
The text was updated successfully, but these errors were encountered: