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
{{ message }}
This repository has been archived by the owner on Jun 16, 2021. It is now read-only.
Thanks for developing this wonderful tools. But I have a question: when I let Mugo match against itself, the game of each time is exactly same. How to let Mugos play differently?
Thanks again.
The text was updated successfully, but these errors were encountered:
Try using 'randompolicy' as a option instead of 'policy'. Policy will take the most likely move, but randompolicy will select randomly, weighted by the NN's output probabilities.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hi there,
Thanks for developing this wonderful tools. But I have a question: when I let Mugo match against itself, the game of each time is exactly same. How to let Mugos play differently?
Thanks again.
The text was updated successfully, but these errors were encountered: