The Environment I built to study Reinforcement Learning + Pokemon Showdown

Been a while since I ran this.

Think it is limited to showdown a specific checkout
checkout 529fdaaa685d8aa695fba01a80cccb3c1d761f7c

Post clone commands

After cloning, build the Pokemon showdown repo. Only needs to be done once
cd Pokemon-Showdown/
node build –force

I believe python3 agent.py is what you want.
I will look through this later and let you know how to run it. Probably will clean it up too.

pokemon-showdown-rl-environment is free for most use cases and the source is available. The source code is published
under the Server Side Public License (SSPL) v1.


View Github