Chen Li

Hi, I'm able to reproduce the bug, and here's the full error report. I'm working on it. ```Bash graphnet [MainProcess] WARNING 2024-05-15 09:26:04 - has_icecube_package - `icecube` not available. Some...

The problem is with `torchvision 0.17.0+rocm5.7`, I re-installed [the cuda version](https://pytorch.org/get-started/previous-versions/#linux-and-windows-5) and now it works: ```Bash pip install torch==2.2.0 torchvision==0.17.0 torchaudio==2.2.0 --index-url https://download.pytorch.org/whl/cu118 ``` Please refer to [this issue](https://github.com/pytorch/pytorch/issues/122494). This...

Maybe the `requirements` folder should be modified somehow? I followed https://graphnet-team.github.io/graphnet/installation/install.html#quick-start (PyTorch 2.2.*, Linux, 11.8) and reproduced this bug. @RasmusOrsoe

Yes, this method `def no_weight_decay()` is [called in BEiT-2 when training](https://github.com/microsoft/unilm/blob/master/beit/run_class_finetuning.py#L474). However, I searched for it and it is not called in [the original 2nd place solution](https://github.com/search?q=repo%3ADrHB%2Ficecube-2nd-place%20no_weight_decay&type=code) or [fastai](https://github.com/search?q=repo%3Afastai%2Ffastai%20no_weight_decay&type=code). Also,...