Vijay Vignesh comments

Results 11 comments of


                                            Vijay Vignesh

mujoco_py Does not Recognize GCC@10

Is it still not supported on M1?

Add support for multi sample item in optimize and yielding from the _getitem_ of the StreamingDataset

Hey @deependujha, @tchaton, @bhimrazy Is this still available to take up?

Add support for multi sample item in optimize and yielding from the _getitem_ of the StreamingDataset

Is there a reason why you suggested modifying the index.json? The solution I had in my mind is: 1. Optimize the dataset normally. 2. Add is_multisample boolean as a class...

Segmentation IOU compute Ignore some tagged values that don't need to be recorded (such as 255)

@Borda I can work on this patch if it's still open. Can you assign me this to me?

Adding multisample feature along with testcases

@tchaton @deependujha @bhimrazy Can you verify the approach once? I can then make changes to the README.

Deprecation Warning in raised when calling `gather_all_tensors`

I was able to reproduce this. I can raise a PR to fix this.

Upper Face Dynamic Deviation (FDD) Metrics for 3D Talking Heads Evaluation

Hi, I can take up this feature, if it's not being worked on already.

`GeneralizedDiceScore` yields 0 scores when using `per_class=True` for samples where class is not present

Hi @Borda I can try looking into it.

Provide BinaryAUROC support for Masked/Sparse Labels

Hi @robertreaney Please let me know if you're still planning to work on this. If not, I can pick it up.

hausdorff_distance() throws IndexError for certain input shapes

Hi @pyscorcher I just confirmed the bug. It basically happens in `line 351` in `/torchmetrics/functional/segmentation/utils.py`. The correct calculation should be `z[i1 * w + j1] = mindis` instead of `z[i1...