Commit Graph

34 Commits

Author SHA1 Message Date
Tomasz Sobczyk
11b28ad3b5 Don't treat unknown options in learn as file names. Add targetfile to specify individual files. 2020-10-22 20:01:55 +09:00
Tomasz Sobczyk
8f3e64a6d5 move sfen reader to separate file 2020-10-22 10:42:28 +09:00
Tomasz Sobczyk
ff06d1e0ad Rewrite learner to be based on stockfish's thread pool. Reduce coupling along the way 2020-10-21 18:17:34 +09:00
Tomasz Sobczyk
146a6b056e PascalCase -> snake_case for consistency with the rest of the codebase. 2020-10-19 18:37:23 +09:00
Tomasz Sobczyk
69ea3d30b2 Move the extra new line to after check health. 2020-10-19 08:29:51 +09:00
Tomasz Sobczyk
c93f8732bf Force Use NNUE to pure when learning. 2020-10-17 08:44:38 +09:00
Tomasz Sobczyk
5db46d0c82 Verify whether there is a network being used during training. 2020-10-17 08:44:38 +09:00
Tomasz Sobczyk
e503cc4ea8 Add one more empty line between progress reports. 2020-10-17 00:13:50 +09:00
Tomasz Sobczyk
5856237e3f Rename hirate to startpos 2020-10-16 09:07:02 +09:00
Tomasz Sobczyk
904adb9a32 Indentation consistency in learn folder 2020-10-15 22:11:31 +09:00
Tomasz Sobczyk
880d23af1c Move sfen input/output streams to sfen_stream.h 2020-10-15 20:37:03 +09:00
Tomasz Sobczyk
14f83ad7b9 Move public search/qsearch interface from namespace Learner to namespace Search 2020-10-15 20:37:03 +09:00
Tomasz Sobczyk
0494adeb2c Move nnue evaluation stuff from evaluate.h to nnue/evaluate_nnue.h 2020-10-15 20:37:03 +09:00
Tomasz Sobczyk
4a2bf16b30 Add option "auto_lr_drop" that specifies the amount of positions from previous lr drop after which to reduce lr by newbob_decay. 2020-10-12 23:56:26 +09:00
Joost VandeVondele
adddf339bb Output sfens/second in the trainer, to track performance more easily 2020-10-08 08:29:42 +09:00
Tomasz Sobczyk
8830209125 Change some learn parameter defaults. 2020-10-07 23:09:33 +09:00
Tomasz Sobczyk
2e57f3fa22 Switch to set recommended learn UCI options 2020-10-07 23:09:33 +09:00
noobpwnftw
b44d539c94 Fix a bug that LR is not correctly scaled when initial LR is not 1.0 2020-09-29 16:18:42 +09:00
noobpwnftw
5e8a49f7f2 Restore lambda and gradient function post-merge and minor fixes.
bench: 3788313
2020-09-26 12:55:02 +09:00
noobpwnftw
9827411b7c Merge remote-tracking branch 'remotes/nodchip/master' into trainer 2020-09-24 21:45:28 +08:00
noobpwnftw
5be8b573be Merge remote-tracking branch 'remotes/origin/master' into trainer 2020-09-23 19:02:27 +08:00
noobpwnftw
411adab149 Merge remote-tracking branch 'remotes/nodchip/master' into trainer 2020-09-23 18:29:30 +08:00
Joost VandeVondele
2931463d3a Revert earlier TB changes.
they were not correct. Unfortunately, also restores the race on RootInTB
2020-09-21 09:45:42 +09:00
Joost VandeVondele
da28ce3339 Add initialization also to learning patch
fixes https://github.com/nodchip/Stockfish/issues/160
2020-09-20 08:16:54 +09:00
Joost VandeVondele
61bc8d12d3 Fix some races in learning
declare a few variables atomic.

Other races remain...
2020-09-19 10:09:30 +09:00
Tomasz Sobczyk
184bde47dc Add "seed" option to gensfen and learn 2020-09-16 23:09:45 +09:00
nodchip
6ae09ba266 Fixed a bug that the root color is wrong. 2020-09-16 12:10:35 +09:00
Tomasz Sobczyk
0f270f7cbf Merge branch 'master' into no_eval_learn 2020-09-13 13:36:16 +02:00
nodchip
a94a076e39 Fixed a comment. 2020-09-13 16:35:52 +09:00
nodchip
1c84da9caa Fixed a bug that an assertion fails in the trainer. if the SkipLoading is false.
Fixes #128
2020-09-13 16:32:01 +09:00
Tomasz Sobczyk
d33e7a9b07 Remove conditional compilation on EVAL_LEARN 2020-09-12 16:19:24 +02:00
Joost VandeVondele
580b09381b Add a learning command to CI
fixes a small issue, with ponder

Probably the learning command can be improved a bit, so that despite the limited data, the code coverage is better.
2020-09-12 22:25:36 +09:00
Tomasz Sobczyk
98f24570ab Add src to include paths, remove non-standard ".." in includes in learn directory. 2020-09-11 21:53:17 +09:00
Tomasz Sobczyk
683c6146ce Move declarations around and split them. 2020-09-11 21:53:17 +09:00