Daniel Filan

Huh, I did not notice this when playing around with training CNN policies and rewards.

Is this related to tests failing on master? I notice that in the past few days there's usually a couple unit tests failing on linux, and right now it looks...

TODO get rid of scripts that download from aws

Having a touch of trouble testing due to #576.

Current status: - have removed experiment scripts that critically rely on aws - have retouched `experiments/README.md` accordingly - have edited experiment config files to match with what's in HuggingFace -...

BTW I kept `imit_benchmark.sh` because I think it makes sense as its own thing, even without the context of `download_models.sh`.

See https://github.com/HumanCompatibleAI/imitation/issues/556 for a related issue.

Oh just realized that error should be an f-string, will make a quick PR for that. [EDIT: said PR is #577]

The result of me running it locally on a fresh clone of imitation: ``` [15:56:30] daniel@sanrensei:~/imitation$ python -m imitation.scripts.eval_policy with seals_ant INFO - eval_policy - Running command 'eval_policy' INFO -...

It does seem to run on a remote server tho so now I'm confused. ``` (fresh_venv) daniel@rnn:~/imitation$ python -m imitation.scripts.eval_policy with seals_ant INFO - eval_policy - Running command 'eval_policy' INFO...