ZHANG Zhi

You can follow [these steps](https://zhuanlan.zhihu.com/p/30807580) to run it. Avisingh599 is right. You need to expend your swap~

We have provided the PyTorch version of the code and pre-trained model in https://github.com/tczhangzhi/VisionTransformer-Pytorch. If the authors have no other plans, we are glad to make PR to this repo....

Good job, I suppose this PR should be merged!

Of course, if you do `torch.cuda.set_device(local_rank)` it is okay to use `model.cuda()` instead of `model.cuda(args.gpu)`. The official repo kept the old fashion of best practice based on previous versions. And...

I finally encountered a situation where the GPU memory was not cleared after using ctrl+c to kill the process today. It happened when I used a custom CUDA module. If...

It annoys me a lot, too. I will push some cases about debugging, maybe days later. As for other suggestions, if u can do a hard job, unit tests help...

Sorry for the late reply. TorchEEG's `dataset` automatically generates intermediate IO (used to speed up subsequent dataset reads), including `info.csv`, based on the original raw dataset. Please pass in parameters...

Thank you for your enthusiastic response, and for the kind words about our work (๑•̀ㅂ•́)و✧. I look forward to communicating with you. MoABB is indeed one of my favorite tools...

I sent an email to discuss other details (灬ºωº灬). My email is [email protected].

Thank you for your report. It makes sense. If you are willing, please feel free to open a pull request.