anyrl-py
anyrl-py copied to clipboard
unixpickle
A reinforcement learning framework
Hello I was wondering if anyone could help me with exporting a model after training so it can continue to train afterwards. I'm new to tensorflow and using colab so...
In [`anyrl/models/feedforward_ac.py`](../blob/531dd920e77f1b77d63d52bd56aad0807bfdccd8/anyrl/models/feedforward_ac.py#L56), the function `step` may assign zero-values in `feed_dict`. This may cause [`self.session.run((self.actor_out, self.critic_out), feed_dict)`](../blob/531dd920e77f1b77d63d52bd56aad0807bfdccd8/anyrl/models/feedforward_ac.py#L58) to return `act` and `val` as arrays containing `nan`'s that may eventually throw an...
This way, you can use a Roller in conjunction with DQN (instead of needing to use a Player).
In particular, it would be nice to turn the existing docstrings, classes, and functions into a browsable web-friendly format.
Right now, there's three different types a model state can be: `NoneType`, an array, or a tuple of arrays. In the future, it would be nice to unify this. I...
[Distributional Reinforcement Learning with Quantile Regression](https://arxiv.org/abs/1710.10044) This supposedly improves the performance of distributional DQN, so I'd like to add it as an option for distributional Q-networks.