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
Thank you for the notice. Right now we are working with a tree policy instead of this approach. The policy follows more or less the same condition as UnitMCTS paper you shared. It will be added when we push a first version of the code.
https://github.com/clinfo/GraphGenerativeModel/blob/master/lib/models.py#L109
scores and or 1/node.visits should be re-considered
np.array([abs(node.performance) / node.visits for node in candidates])
The text was updated successfully, but these errors were encountered: