search

JeffWang987 / OpenOccupancy

[ICCV 2023] OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perception
Apache License 2.0
551 stars 49 forks source link

test error #12

Closed shenxiaowrj closed 1 year ago

shenxiaowrj commented 1 year ago

I regenerated the ground_truth with the latest version of the code. And i use the python ./tools/test.py ./projects/configs/baselines/CAM-R50_img1600_128x128x10.py ./work_dirs/CAM-R50_img1600_128x128x10/latest.pth --deterministic --eval bbox to test the c-baseline model. The weights I used were obtained after training two epochs. The test exceeded the set maximum test dataset length. 711679b78e19efcadfc126f86e6c84b How can I solve this problem? Thanks.

JeffWang987 commented 1 year ago

It is a small updating bug of the ProgressBar, which does not influence the evaluation performance. We have fixed it, please pull the latest code.

shenxiaowrj commented 1 year ago

It is a small updating bug of the ProgressBar, which does not influence the evaluation performance. We have fixed it, please pull the latest code.

Thanks! The previous issue was solved. But the testing phase still didn't go smoothly. It encountered the following issues.

6019/6019, 0.8 task/s, elapsed: 7264s, ETA: 0sTraceback (most recent call last): File "./tools/test.py", line 278, in main() File "./tools/test.py", line 275, in main print(dataset.evaluate(outputs, **eval_kwargs)) File "/media/re/2384a6b4-4dae-400d-ad72-9b7044491b55/data/OpenOccupancy/projects/occ_plugin/datasets/nuscenes_occ_dataset.py", line 130, in evaluate ious = cm_to_ious(evaluation_semantic) File "/media/re/2384a6b4-4dae-400d-ad72-9b7044491b55/data/OpenOccupancy/projects/occ_plugin/utils/formating.py", line 8, in cm_to_ious tp = cm[i, i] IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed 7294cd5fc8f0d06605da38a75b8ce60

JeffWang987 commented 1 year ago

The problem is fixed, it is caused by repeated evaluation. You don't have to re-eval your model, as the last print info in the figure is your final performance. To avoid similar errors, please pull the latest code.

shenxiaowrj commented 1 year ago

Thanks!

dzp17 commented 3 months ago

It is a small updating bug of the ProgressBar, which does not influence the evaluation performance. We have fixed it, please pull the latest code.

Thanks! The previous issue was solved. But the testing phase still didn't go smoothly. It encountered the following issues.

6019/6019, 0.8 task/s, elapsed: 7264s, ETA: 0sTraceback (most recent call last): File "./tools/test.py", line 278, in main() File "./tools/test.py", line 275, in main print(dataset.evaluate(outputs, **eval_kwargs)) File "/media/re/2384a6b4-4dae-400d-ad72-9b7044491b55/data/OpenOccupancy/projects/occ_plugin/datasets/nuscenes_occ_dataset.py", line 130, in evaluate ious = cm_to_ious(evaluation_semantic) File "/media/re/2384a6b4-4dae-400d-ad72-9b7044491b55/data/OpenOccupancy/projects/occ_plugin/utils/formating.py", line 8, in cm_to_ious tp = cm[i, i] IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed 7294cd5fc8f0d06605da38a75b8ce60

Hi,I encountered the same error as you, have you solved it?