filaPro / oneformer3d

[CVPR2024] OneFormer3D: One Transformer for Unified Point Cloud Segmentation
Other
351 stars 32 forks source link

Error encountered during testing script execution #39

Closed BaptFontaine closed 7 months ago

BaptFontaine commented 8 months ago

I am encountering an error when attempting to run the testing script after successfully training the model on the S3DIS dataset. The error message I'm receiving is as follows: Traceback (most recent call last): File "tools/test.py", line 149, in main() File "tools/test.py", line 145, in main runner.test() File "/home/baptiste/miniconda3/envs/OF3D/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1823, in test metrics = self.test_loop.run() # type: ignore File "/home/baptiste/miniconda3/envs/OF3D/lib/python3.8/site-packages/mmengine/runner/loops.py", line 443, in run self.run_iter(idx, data_batch) File "/home/baptiste/miniconda3/envs/OF3D/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/home/baptiste/miniconda3/envs/OF3D/lib/python3.8/site-packages/mmengine/runner/loops.py", line 464, in run_iter self.runner.call_hook( File "/home/baptiste/miniconda3/envs/OF3D/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1841, in call_hook raise TypeError(f'{e} in {hook}') from None TypeError: add_datasample() got an unexpected keyword argument 'vis_task' in <mmdet3d.engine.hooks.visualization_hook.Det3DVisualizationHook object at 0x7fa091151640>

This error occurs when executing the testing script with the following command: python tools/test.py configs/oneformer3d_1xb2_s3dis-area-5.py work_dirs/oneformer3d_1xb2_s3dis-area-5/epoch_512.pth --show-dir results/oneformer3d_1xb2_s3dis-area-5 --task lidar_det

I seek assistance in either resolving this error or finding an alternative method to save the results generated by the testing script. Any guidance or suggestions would be appreciated.

oneformer3d-contributor commented 8 months ago

Yes, these code is not ready from mmdetecion3d visualization, mainly because we predict three types of segmentation masks simultaniously. You can simply save the input and the output from predict function and then visualize them.

jiachen0212 commented 8 months ago

Yes, these code is not ready from mmdetecion3d visualization, mainly because we predict three types of segmentation masks simultaniously. You can simply save the input and the output from predict function and then visualize them.

will the code add mmdetecion3d visualization? we also need to predict the orientation of 3D objects~

oneformer3d-contributor commented 8 months ago

Unfortunately no plans for adding object detection or visualization to this repo.