Nicolas Pielawski

Thanks @adriantrunzo! I could change the colors to my vim-deus theme! IMO it would be better if the `LS_COLORS` variable wouldn't have to be set at all (at least on...

Thanks for you input! I will then create a local backbone then, copying the ResNest arch. I won't be lagging behind master, good idea!

I looked into using DeepLabV3, but I can only use registered backbones [1], and BaseNet also forces pretrained=True [2] (and I don't have a pretrained model). I could bypass pretrained=True,...

Well, I would have preferred not branching the repo and modifying it if that makes sense.

Let's say you have 27 classes and your dataset is relatively balanced, that is p(C)=1/27. If you pick two random images, they will likely not have the same class, the...

The class analogy from my previous comment is not entirely correct with STEGO. They use a backbone to project the images to a latent space, so imagine that all the...

I am having the same problem. Looking at the code, the ground truth images can only be exported, otherwise they are dumped. In our case, this means, that they are...

I have the same issue with Python 3.10, also using Mamba. self.nargs is created here: https://github.com/openai/triton/blob/638e531a7158cd971d32e2a6bd5ca0f2309c8227/python/triton/runtime/autotuner.py#L120 self.nargs holds a lot of information that is destroyed in the if condition below....

I am trying to make one, but I am not sure how to. If I put a try catch around the full_nargs line and start an IPython embed session, I...

I think the issue is created by a race condition, when self.nargs is set to None [line 162](https://github.com/openai/triton/blob/638e531a7158cd971d32e2a6bd5ca0f2309c8227/python/triton/runtime/autotuner.py#L162), and another thread is accessing the same method at the same time....