search

Atrus619 / DeckOfCards

Environment for simulating card games to train reinforcement learning agents through self-play to outperform human players
1 stars 0 forks source link

Add meld to game #7

Open Atrus619 opened 4 years ago

Atrus619 commented 4 years ago

Will need to be careful how it is handled with state/action pairs

Atrus619 commented 4 years ago

When a player wins a trick, we take output from both heads, in that determine the bot's meld and following trick play with one forward pass.