This is starting to get somewhat difficult compare to what I first envisioned this to be but thats ok
Need to implement node terminality with self.is_terminal.
This is gained though getting the legal moves (for the current player ofc) which will be winning moves once they are found.
We first check if we have winning moves else then we get the predicted policy and value.
^ this is very expensive
This is starting to get somewhat difficult compare to what I first envisioned this to be but thats ok Need to implement node terminality with self.is_terminal. This is gained though getting the legal moves (for the current player ofc) which will be winning moves once they are found. We first check if we have winning moves else then we get the predicted policy and value. ^ this is very expensive