Fix softmax masking for illegal moves in inference#14
Open
Conversation
Multiplying logits by a binary 0/1 mask sets illegal move logits to 0, but softmax(0) is not zero — it still contributes probability mass. Use masked_fill with -inf so softmax correctly assigns exactly 0 probability to illegal moves. Fixes the issue raised in PR #9. https://claude.ai/code/session_01ELpknikQ4vWB4q8hWhxoVG
Use canonical PyPI name 'PyYAML' instead of 'pyyaml' in both requirements.txt and pyproject.toml. https://claude.ai/code/session_01ELpknikQ4vWB4q8hWhxoVG
Lichess PGN files include WhiteTitle/BlackTitle headers set to "BOT" for bot accounts, including Maia itself. Skip any game where either player has the BOT title. Fixes #2 https://claude.ai/code/session_01ELpknikQ4vWB4q8hWhxoVG
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Multiplying logits by a binary 0/1 mask sets illegal move logits to 0,
but softmax(0) is not zero — it still contributes probability mass.
Use masked_fill with -inf so softmax correctly assigns exactly 0
probability to illegal moves.
Fixes the issue raised in PR #9.
https://claude.ai/code/session_01ELpknikQ4vWB4q8hWhxoVG