Lokesh Veeramacheneni

You should be able to write your own metric calculation, for that you just need the class label and predicted class label then calculate TP, TN, FP, FN. From this...

Changing this parameter will show the effect in the prediction performance. Lower number of points have faster inference but poor segmentation performance and vice versa. From what I understood this...

Are you using GPU/CPU?. With Semantic KITTI on CPU it took around 4 hours per epoch but in GPU it was 30 min. check whether the code is using GPU...

@ARITRA2296 Did you install tensorflow or tensorflow-gpu. tensorflow-gpu should pick the GPU automatically. You can use something like this is https://stackoverflow.com/questions/38559755/how-to-get-current-available-gpus-in-tensorflow. To check your GPUs in the code.

Is your problem solved, because I am also facing same issue with semantic KITTI dataset. If yes, it would be helpful to share the solution.

I kind of solved it but the issue is with new classes the network doesn't optimize. I changed the semantic KITTI classes from 19 to 17 and also changed the...

I think the problem is with division if your denominator is zero then leads to infinity which leads to NaN. I would try adding a small constant of 1e-6 to...

Hi Unfortuntely no, I didn't solve it.

If you want to use SemanticKITTI dataloader, this shouldn't be a issue because SemanticKITTI has no color. But if you want to use Semantic3D or S3DIS dataloader, then I think...

This error is because in newer version of parafac it returns norms and factors, so change your statement to last, first, vertical, horizontal = parafac(W, rank=rank, init='random')[1].Then this return four...