ahayler
ahayler
Hey @Louis-Leee, Thank you for the answer! I think you might have misunderstood my post. I generated the light green voxels to explain, which of the voxels are problematic. By...
Thank you for the suggestion! Sadly I already accounted for that!
@npurson In the mobilestereonet prediction.py script you can find on top: ``` Mathmatical Formula to convert the disparity to depth: depth = baseline * focal / disparity For KITTI the...
For good measure, I tried to run this script on a different compute node with `torch.cuda.synchronize` added before and after the `fit` and `predict_proba` step. Sadly, this neither fixes the...
## A temporary solution I have tried to further diagnose the segmentation fault. For this, I used `gdb` and ran the above script a few times. The backtraces all look...
Hey @priorphil, Thanks for getting back! I agree with you regarding adding `torch.cuda.empty_cache()` to code base. This should be, at best, a temporary workaround. Do you have any idea why...