Commit Graph

  • a787f73cf6 add random preprocess, modify the uniform sample from training data rtz19970824 2017-12-23 16:27:09 +08:00
  • 4589fcf521 add random preprocess, modify the uniform sample from training data rtz19970824 2017-12-23 16:27:09 +08:00
  • c50ee8f029 minor fixed for mcts, check finish for go rtz19970824 2017-12-23 15:58:06 +08:00
  • 3f238864fb minor fixed for mcts, check finish for go rtz19970824 2017-12-23 15:58:06 +08:00
  • d01f8cd210 Merge remote-tracking branch 'origin/master' JialianLee 2017-12-23 15:49:39 +08:00
  • 430b78abf5 Merge remote-tracking branch 'origin/master' JialianLee 2017-12-23 15:49:39 +08:00
  • 282b496e49 Modification for reversi.py JialianLee 2017-12-23 15:43:45 +08:00
  • 84208a7ac9 Modification for reversi.py JialianLee 2017-12-23 15:43:45 +08:00
  • 86bf94fde1 Merge remote-tracking branch 'origin/master' haoshengzou 2017-12-23 15:36:44 +08:00
  • 8c13d8ebe6 Merge remote-tracking branch 'origin/master' haoshengzou 2017-12-23 15:36:44 +08:00
  • 2addef41d2 fix imports to support both python2 and python3. move contents from __init__.py to leave for work after major development. haoshengzou 2017-12-23 15:36:10 +08:00
  • 04048b7873 fix imports to support both python2 and python3. move contents from __init__.py to leave for work after major development. haoshengzou 2017-12-23 15:36:10 +08:00
  • fe54e4732d fix imports to support both python2 and python3. move contents from __init__.py to leave for work after major development. haoshengzou 2017-12-23 15:34:44 +08:00
  • 951eed60ed fix imports to support both python2 and python3. move contents from __init__.py to leave for work after major development. haoshengzou 2017-12-23 15:34:44 +08:00
  • deea09b2b7 minor fixed rtz19970824 2017-12-23 14:45:07 +08:00
  • b96fa9448b minor fixed rtz19970824 2017-12-23 14:45:07 +08:00
  • 0c3ebacc75 Merge branch 'master' of github.com:sproblvem/tianshou Dong Yan 2017-12-23 13:05:38 +08:00
  • d0e150f090 Merge branch 'master' of github.com:sproblvem/tianshou Dong Yan 2017-12-23 13:05:38 +08:00
  • e63338ab01 connect reversi with game Dong Yan 2017-12-23 13:05:25 +08:00
  • b2ef770415 connect reversi with game Dong Yan 2017-12-23 13:05:25 +08:00
  • 1eb46774c2 small modification JialianLee 2017-12-23 09:47:08 +08:00
  • 032ea46b7b small modification JialianLee 2017-12-23 09:47:08 +08:00
  • 3b534064bd fix virtual loss bug mcgrady00h 2017-12-23 02:48:53 +08:00
  • 5b044c9a0c merge rtz19970824 2017-12-22 17:17:50 +08:00
  • d19559857f merge rtz19970824 2017-12-22 17:17:50 +08:00
  • d8c0eae6a3 implement a stochastic sample training method rtz19970824 2017-12-22 17:16:44 +08:00
  • 67ba76a04d implement a stochastic sample training method rtz19970824 2017-12-22 17:16:44 +08:00
  • 51b8e7fc18 move the unit test of is_eye into go.py Dong Yan 2017-12-22 15:44:44 +08:00
  • c5e33af841 move the unit test of is_eye into go.py Dong Yan 2017-12-22 15:44:44 +08:00
  • 7964064242 Modification for reversi JialianLee 2017-12-22 15:26:47 +08:00
  • 511f64b3d6 Modification for reversi JialianLee 2017-12-22 15:26:47 +08:00
  • a062c7610c Merge branch 'master' of https://github.com/sproblvem/tianshou rtz19970824 2017-12-22 13:47:39 +08:00
  • d42a76f8f3 Merge branch 'master' of https://github.com/sproblvem/tianshou rtz19970824 2017-12-22 13:47:39 +08:00
  • e72fd52913 Merge branch 'master' of https://github.com/sproblvem/tianshou rtz19970824 2017-12-22 13:47:38 +08:00
  • ed96268454 Merge branch 'master' of https://github.com/sproblvem/tianshou rtz19970824 2017-12-22 13:47:38 +08:00
  • e8a10f189e print in the loading process rtz19970824 2017-12-22 13:47:27 +08:00
  • 8328153b86 print in the loading process rtz19970824 2017-12-22 13:47:27 +08:00
  • 97161f37ef faster the loading rtz19970824 2017-12-22 13:42:53 +08:00
  • a8509ba292 faster the loading rtz19970824 2017-12-22 13:42:53 +08:00
  • 9824cf8bef merge rtz19970824 2017-12-22 13:31:41 +08:00
  • 5f296ce009 merge rtz19970824 2017-12-22 13:31:41 +08:00
  • d151f71ee3 modify the training config rtz19970824 2017-12-22 13:30:48 +08:00
  • 6b3efd7fca modify the training config rtz19970824 2017-12-22 13:30:48 +08:00
  • 62a241e763 no restrict on saving checkpoints rtz19970824 2017-12-22 13:05:01 +08:00
  • d281ecc6e0 no restrict on saving checkpoints rtz19970824 2017-12-22 13:05:01 +08:00
  • e75883a5fb debug the training process, initialize a nameserver if no nameserver exists rtz19970824 2017-12-22 13:04:02 +08:00
  • 2b1285143c debug the training process, initialize a nameserver if no nameserver exists rtz19970824 2017-12-22 13:04:02 +08:00
  • 7f1191ef02 An initial version for Reversi JialianLee 2017-12-22 01:57:48 +08:00
  • 5c29dad263 An initial version for Reversi JialianLee 2017-12-22 01:57:48 +08:00
  • b32418d11a Merge remote-tracking branch 'origin/master' Haosheng Zou 2017-12-22 00:24:06 +08:00
  • 8ba16a8808 Merge remote-tracking branch 'origin/master' Haosheng Zou 2017-12-22 00:24:06 +08:00
  • 6611d948dd add value_function (critic). value_function and policy not finished yet. Haosheng Zou 2017-12-22 00:22:23 +08:00
  • 1cc5063007 add value_function (critic). value_function and policy not finished yet. Haosheng Zou 2017-12-22 00:22:23 +08:00
  • a61c1f136a multi-instance support rtz19970824 2017-12-22 00:04:51 +08:00
  • 6835ec62e1 multi-instance support rtz19970824 2017-12-22 00:04:51 +08:00
  • a20255249c modify for multi instance rtz19970824 2017-12-21 23:55:31 +08:00
  • 43f6527d8e modify for multi instance rtz19970824 2017-12-21 23:55:31 +08:00
  • ff2ebd49c1 merge conflict rtz19970824 2017-12-21 23:36:57 +08:00
  • 6bb34afba5 merge conflict rtz19970824 2017-12-21 23:36:57 +08:00
  • c11eccbc90 implement the training process rtz19970824 2017-12-21 23:30:24 +08:00
  • 9ad53de54f implement the training process rtz19970824 2017-12-21 23:30:24 +08:00
  • c3e9e55b24 eliminate all references of Game class in Go class Dong Yan 2017-12-21 22:48:53 +08:00
  • 2acb1aab07 eliminate all references of Game class in Go class Dong Yan 2017-12-21 22:48:53 +08:00
  • 2dad8e4020 implement data collection and part of training rtz19970824 2017-12-21 21:01:25 +08:00
  • eda7ed07a1 implement data collection and part of training rtz19970824 2017-12-21 21:01:25 +08:00
  • 1e2567c174 fixing bug pass parameterg Wenbo Hu 2017-12-21 19:31:51 +08:00
  • ced63af18f fixing bug pass parameterg Wenbo Hu 2017-12-21 19:31:51 +08:00
  • 336cede197 repair komi. add todo for forbid pass: Wenbo Hu 2017-12-20 22:57:58 +08:00
  • 00d2aa86bf repair komi. add todo for forbid pass: Wenbo Hu 2017-12-20 22:57:58 +08:00
  • 40909fa994 forbid pass, if we have other choices Wenbo Hu 2017-12-20 22:10:47 +08:00
  • f0d59dab6c forbid pass, if we have other choices Wenbo Hu 2017-12-20 22:10:47 +08:00
  • 0ab38743aa minor revision. Wenbo Hu 2017-12-20 21:52:30 +08:00
  • e2c6b96e57 minor revision. Wenbo Hu 2017-12-20 21:52:30 +08:00
  • 8875ad1bf7 minor revision Wenbo Hu 2017-12-20 21:40:03 +08:00
  • cabbb21968 minor revision Wenbo Hu 2017-12-20 21:40:03 +08:00
  • 818da800e2 simulator process a valid set, instead of a single action Wenbo Hu 2017-12-20 21:35:35 +08:00
  • 48e95a21ea simulator process a valid set, instead of a single action Wenbo Hu 2017-12-20 21:35:35 +08:00
  • 12f45d9dc6 checkpoint Wenbo Hu 2017-12-20 20:12:08 +08:00
  • 50e306368f checkpoint Wenbo Hu 2017-12-20 20:12:08 +08:00
  • 112fd07b13 modify the mcts, refactor the network rtz19970824 2017-12-20 16:43:42 +08:00
  • 7fca90c61b modify the mcts, refactor the network rtz19970824 2017-12-20 16:43:42 +08:00
  • db40994e11 merge Go and GoEnv finallygit status! Dong Yan 2017-12-20 01:14:05 +08:00
  • c2b46c44e7 merge Go and GoEnv finallygit status! Dong Yan 2017-12-20 01:14:05 +08:00
  • 0456e0c15e final version before merge Go and GoEnv Dong Yan 2017-12-20 00:43:31 +08:00
  • d1af137686 final version before merge Go and GoEnv Dong Yan 2017-12-20 00:43:31 +08:00
  • afc5dbac5a rearrange the sequence of functions of Go and GoEnv before merging Dong Yan 2017-12-20 00:16:24 +08:00
  • 2a9d949510 rearrange the sequence of functions of Go and GoEnv before merging Dong Yan 2017-12-20 00:16:24 +08:00
  • f8a70183b6 fix the copy bug in check_global_isomorphous; refactor code to eliminate side effect Dong Yan 2017-12-19 22:57:38 +08:00
  • 232204d797 fix the copy bug in check_global_isomorphous; refactor code to eliminate side effect Dong Yan 2017-12-19 22:57:38 +08:00
  • 1f011a44ef add mcts virtual loss version (may have bugs) mcgrady00h 2017-12-19 17:04:55 +08:00
  • 83f9e19fa5 merge flatten and deflatten, rename variable for clarity Dong Yan 2017-12-19 16:51:50 +08:00
  • fc8114fe35 merge flatten and deflatten, rename variable for clarity Dong Yan 2017-12-19 16:51:50 +08:00
  • fae273f219 start a random player if checkpoint path is not specified rtz19970824 2017-12-19 15:39:31 +08:00
  • 4a2d8f0003 start a random player if checkpoint path is not specified rtz19970824 2017-12-19 15:39:31 +08:00
  • d7b3b6aba9 deflatten debug rtz19970824 2017-12-19 15:09:46 +08:00
  • 0991fef527 deflatten debug rtz19970824 2017-12-19 15:09:46 +08:00
  • e168df5609 fix bug in check_global_isomorphous and refactor _is_suicide again Dong Yan 2017-12-19 12:00:17 +08:00
  • 4440294c12 fix bug in check_global_isomorphous and refactor _is_suicide again Dong Yan 2017-12-19 12:00:17 +08:00
  • 72a9f4823c rename variable for clarity Dong Yan 2017-12-19 11:16:17 +08:00
  • 99a617a1f0 rename variable for clarity Dong Yan 2017-12-19 11:16:17 +08:00