training custom envs

[dzako]

hello, thanks for very nice library with baseline RL implementations, which doesn't require to load tensorflow :) I have a question how to load a custom environment ? a wrapper is required ?

[fabiopardo]

Hi! Thanks a lot :)

I recommend to add environments like this:

1. Create a Python package with your set of environments, placing an __init__.py file at the root and registering the environments in Gym (for example like this) or dm_control (for example like this).
2. When using tonic.train, simply import your environment package in the "header" field.

I have just updated the paper, adding a small section about this at the end of Section 7.