tungts1101
tungts1101
Does GraphicFuzz now support shaders with texture?
Is there any update on this issue? **Update**: I have checked the current code again, and they have called `dist.all_reduce()` before comparing it with `best_acc`, so I think everything is...
I got the same error. My OS is windows 11. Here is what I got with `pip show torch` ``` Name: torch Version: 2.3.0+cu118 Summary: Tensors and Dynamic neural networks...
It seems like Windows doesn't support NCCL backend. Does it mean that I can only run `llama3` on linux based machine?
I have tried again with my Ubuntu 22.04 installed under WSL. The `nccl` error has disappeared but I still get this error when trying to run the example ``` E0510...
> Could you please provide the complete error message and your hardware specs, along with the code you tried to run? > > NCCL isn't supported on Windows. If you...
I think the root cause is the hardware doesn't meet the minimum requirement to run the llama-7B model.
When I used only the backbone `torch.hub.load("facebookresearch/dinov2", 'dinov2_vits14')` and my own classification head `torch.nn.Linear(in_features=1920, out_features=1000, bias=True)`, I got an error `RuntimeError: mat1 and mat2 shapes cannot be multiplied (1x384 and...
@legel Doing what you said will work. The part that I do not understand is the linear head of the classification model has input features of 1920 and output features...
Thank you so much for your response. Have your team conducted the experiment without retraining the backbone but only the decoding part?