You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Unfortunately we haven't released any of the agents trained for the paper. They use internal libraries and are hard to open source.
pysc2/bin/replay_actions.py is an example of how to collect stats from a large number of replays, so you could modify that to generate the specific dataset you care about.
pysc2/agents/ has some trivial example agents for how to interact with the environment. Take a look at #34 for people working on an a3c agent.
In the sc2le.pdf:
DeepMind:
But I can't find the agents about supervised learning from replays
The text was updated successfully, but these errors were encountered: