Kevin Ferguson

Hi @Zaxorn ! One important thing we did not note in the book is to **clear out the .npy files in between runs**. Otherwise, data from the previous run can...

Hi @Sarah112358, thank you for reporting this. I found the solution is to set `bufsize=0` on the call to Popen. See this diff for a fix: https://github.com/maxpumperla/deep_learning_and_the_game_of_go/commit/6afbe8c7ca048c946c2cb512559c9a437b6c24b6

Hello all, @DrVecctor was correct. The simplest fix is to make pass a legal move even after the game is over. This means the MCTS could continue to read out...

Hi aletote, you can use any function that is suitable for the DepthPrunedAgent: it should just take a GameState argument, and return a score indicating how good the state is...

Hey @maxpumperla, does the dlgo library actually need `gomill`? We included a sideport of the SGF parser that is updated for Python 3: https://github.com/maxpumperla/deep_learning_and_the_game_of_go/blob/master/code/dlgo/gosgf/sgf.py Maybe we can just drop `gomill`...

Good catch, thank you for this fix! Will review this tomorrow

Hi fuhaoda, just to re-iterate what Max said, since `node = self.select_child(node)` is called in a loop, you may call `select_child` an arbitrary number of times. So the _final_ value...

Hi @arisliang, I don't think you've missed anything. As you noted, when it's exhausted the legal moves of the current state, it starts to go deeper. However, after the first...

Hi Nate, I actually just ran into this same problem myself recently. This is an issue of the `channels_first`/`channels_last` options for indexing tensors (also known as NCHW / NHWC). (See...

Hello @Nkonovalenko, please see this writeup here: https://kferg.dev/posts/2021/deep-learning-and-the-game-of-go-training-results-from-chapter-7/ Hopefully that gets you unblocked!