Thomas Wolf

Still an issue in the current version 4.0, I will try to fix this.

Yes, nice! I think we should have probably have both framework and environment name indicated given that some environments are available through different frameworks (with usually subtil differences), e.g. `atari-py`...

I would rather reimplement these metrics than depend on other libs. My heuristic: - if the metric is something stable we don't expect to change (Jaccard score): I would reimplement....

Thanks for sharing your thoughts. This is a very important discussion. Also one of the first items on our mid-term roadmap (we will try to clean it and share it...

Hi, `python -m spacy download en` should do the trick. cf https://spacy.io/usage/models

Maybe you should install the latest version of `pytorch_pretrained_bert`? `pip install pytorch_pretrained_bert --upgrade`

Oh, how did you get those?

They are dropped. This is done several times in the model actually, also here: https://github.com/huggingface/pytorch-pretrained-BigGAN/blob/1e18aed2dff75db51428f13b940c38b923eb4a3d/pytorch_pretrained_biggan/model.py#L192-L194 If you read the latest version of the BigGAN paper, you will see it part...

Well I'm not very familiar with Andy's implementation but I see a channel dropping part here: https://github.com/ajbrock/BigGAN-PyTorch/blob/ba3d05754120e9d3b68313ec7b0f9833fc5ee8bc/BigGANdeep.py#L54-L56 I'm not sure Andy's implementation can load the `--deep` models, which is what...

Hi, which pre-trained model are you referring to?