Introduce trained PSQT depending on king position #12
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Outdated suggestions cannot be applied.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.
This patch introduces PSQ tables dependent on the attacking/defending king position. Similarly to regular PSQ tables, these can be efficiently updated for most moves, and only need to be recomputed from scratch after a king move. The king's position is encoded in a mirrored fashion.
The current table has been trained from a dataset of roughly 8M games at depth 6 (generated using the branch
data_gen
).The default table is embedded in the binary using incbin.
STC: LLR: 2.94/2.94<0.00, 6.00> Elo diff: 70.84 [52.71, 89.35] (95%)
Games: 728 W: 271 L: 124 D: 333 Draw ratio: 45.7%
Pntl: [10, 57, 117, 135, 44]
LTC: LLR: 2.95/2.94<0.00, 6.00> Elo diff: 87.82 [69.86, 106.26] (95%)
Games: 608 W: 227 L: 76 D: 305 Draw ratio: 50.2%
Pntl: [2, 35, 113, 117, 36]
Closes #12
Bench: 2169197