Fix installation bugs and testing bugs

[HelloWorldLTY]

Hi, thank you for your great work. When I download the pretrained models and testing note books, I find some small issues which might affect the reproductibility of the results, and I have fixed them.

1. It seems that we need huggingface_hub to download the pretrained models, without it, models cannot be successfully downloaded. I have fixed this issue by modifying environment.yml file.

2. In the gigatime_testing.ipynb notebook, the inference step is performed in GPU, however, the model is loaded in cpu originally. Your codes are

# Download the repo snapshot 
local_dir = snapshot_download(repo_id=repo_id)

weights_path = os.path.join(local_dir, "model.pth")
state_dict = torch.load(weights_path, map_location="cpu")
model.load_state_dict(state_dict)

So you load models only using CPU. I have modifed the codes and put model in GPU so we can make inference with GPU.

If you think these modifications can be helpful, please consider merging it. Thanks.