Lukas Rauch

Hey, any new insights into how to fix the problem? I just ran into the same issue on a dockerized ubuntu20.04 with cudnn11.7. Happy, if someone could share their latest...

Hey, I have the same problem. Did you find any solution? Form my understanding, the SemanticSegmentation pipeline should be equipped with Adam optimizer by default...

Thank you, @Gofinge. I should be able to stitch this together, for my data. In my example, `i` refers to the reflection intensity (scalar field, [n, 1] dim) vector. Is...

Finally, I had time to come back to this issue. Thanks for your answer. Unfortunately, I now have a new problem O.o I preprocessed my data to meet the required...

okay, it seems to work if I take the nuScenes example and iteratively adjust the two grid_size 's so that my GPU VRam is sufficient. However, the results in exp/model/results...

Excuse me, @adamvln. I just came back to this issue and realized that you addressed me directly with your question. Could you resolve the problem? If you renamed the file...

Sorry to bother you with my problems again. Thank you for your answer. I have been dealing with the problem for several days, so I thought it would be easier...

@Carl-Yang98, thanks for your interest in my issue. I haven't tested this configuration without the `SphereCrop` yet. The dataset I am training on is Point Cloud Data from a terrestrial...

Thank You @Gofinge. I think I can work with that 👍 To my understanding, what was the actual purpose of the code snipped you posted? I see that you replace...

the above script requires your dataset to already have normal features as input for the model. If this is not the case, you can try to estimate the normals in...