genrl
genrl copied to clipboard
SforAiDl
A PyTorch reinforcement learning library for generalizable and reproducible algorithm implementations with an aim to improve accessibility in RL
Add documentation on the types of available env wrappers
@DarylRodrigo pointed out that the current explanations of Policy, Value functions are not good. One of our aims is to tackle Accessibility. Quoting him: "the intro into policy and value...
More explanations of bellman equations and their relation with MDPs and then Q-Learning, etc. At the moment, we have almost nothing wrt bellman equations! Thanks @DarylRodrigo
Current training is not reproducible. (Observation based on `deep.py` in `examples`). Something is messing up, not sure why seeding is not doing the job here. Could be that we're missing...
We currently have a lot of issues, a lot of which are either incomplete, not relevant for the time being, need to be done at a later period or we...
I have been playing around with the DCBTrainer and found some potential inconsistencies. 1) **StatlogData** example found [here](https://genrl.readthedocs.io/en/latest/usage/tutorials/bandit/contextual_overview.html) ``` from genrl.utils import StatlogDataBandit bandit = StatlogDataBandit(download=True) context = bandit.reset() from...
Should we shift to using [DM Env](https://github.com/deepmind/dm_env)? Should probably evaluate this as a potential option.
