153 Commits

Author SHA1 Message Date
rtz19970824
9824cf8bef merge 2017-12-22 13:31:41 +08:00
rtz19970824
d151f71ee3 modify the training config 2017-12-22 13:30:48 +08:00
rtz19970824
62a241e763 no restrict on saving checkpoints 2017-12-22 13:05:01 +08:00
rtz19970824
e75883a5fb debug the training process, initialize a nameserver if no nameserver exists 2017-12-22 13:04:02 +08:00
JialianLee
7f1191ef02 An initial version for Reversi 2017-12-22 01:57:48 +08:00
Haosheng Zou
b32418d11a Merge remote-tracking branch 'origin/master' 2017-12-22 00:24:06 +08:00
Haosheng Zou
6611d948dd add value_function (critic). value_function and policy not finished yet. 2017-12-22 00:22:23 +08:00
rtz19970824
a61c1f136a multi-instance support 2017-12-22 00:04:51 +08:00
rtz19970824
a20255249c modify for multi instance 2017-12-21 23:55:31 +08:00
rtz19970824
ff2ebd49c1 merge conflict 2017-12-21 23:36:57 +08:00
rtz19970824
c11eccbc90 implement the training process 2017-12-21 23:30:24 +08:00
Dong Yan
c3e9e55b24 eliminate all references of Game class in Go class 2017-12-21 22:48:53 +08:00
rtz19970824
2dad8e4020 implement data collection and part of training 2017-12-21 21:01:25 +08:00
Wenbo Hu
1e2567c174 fixing bug pass parameterg 2017-12-21 19:31:51 +08:00
Wenbo Hu
336cede197 repair komi. add todo for forbid pass: 2017-12-20 22:57:58 +08:00
Wenbo Hu
40909fa994 forbid pass, if we have other choices 2017-12-20 22:10:47 +08:00
Wenbo Hu
0ab38743aa minor revision. 2017-12-20 21:52:30 +08:00
Wenbo Hu
8875ad1bf7 minor revision 2017-12-20 21:40:03 +08:00
Wenbo Hu
818da800e2 simulator process a valid set, instead of a single action 2017-12-20 21:35:35 +08:00
Wenbo Hu
12f45d9dc6 checkpoint 2017-12-20 20:12:08 +08:00
rtz19970824
112fd07b13 modify the mcts, refactor the network 2017-12-20 16:43:42 +08:00
Dong Yan
db40994e11 merge Go and GoEnv finallygit status! 2017-12-20 01:14:05 +08:00
Dong Yan
0456e0c15e final version before merge Go and GoEnv 2017-12-20 00:43:31 +08:00
Dong Yan
afc5dbac5a rearrange the sequence of functions of Go and GoEnv before merging 2017-12-20 00:16:24 +08:00
Dong Yan
f8a70183b6 fix the copy bug in check_global_isomorphous; refactor code to eliminate side effect 2017-12-19 22:57:38 +08:00
Dong Yan
83f9e19fa5 merge flatten and deflatten, rename variable for clarity 2017-12-19 16:51:50 +08:00
rtz19970824
fae273f219 start a random player if checkpoint path is not specified 2017-12-19 15:39:31 +08:00
rtz19970824
d7b3b6aba9 deflatten debug 2017-12-19 15:09:46 +08:00
Dong Yan
e168df5609 fix bug in check_global_isomorphous and refactor _is_suicide again 2017-12-19 12:00:17 +08:00
Dong Yan
72a9f4823c rename variable for clarity 2017-12-19 11:16:17 +08:00
Dong Yan
1a164d4d7d rewrite _is_qi in a more understandable way 2017-12-19 00:47:21 +08:00
Dong Yan
243bbaff64 Merge branch 'master' of github.com:sproblvem/tianshou 2017-12-19 00:16:33 +08:00
Dong Yan
fb511aa76d delete unused parameter of _find_block, and using _find_group to replace _find_block 2017-12-19 00:16:21 +08:00
Tongzheng Ren
0e1287b5cb update gitignore 2017-12-18 23:34:32 +08:00
Tongzheng Ren
27c1017259 add a detailed Chinese google coding style for convenience 2017-12-18 23:32:41 +08:00
宋世虹
d220f7f2a8 add comments and todos 2017-12-17 13:28:21 +08:00
宋世虹
3624cc9036 finished very naive dqn: changed the interface of replay buffer by adding collect and next_batch, but still need refactoring; added implementation of dqn.py, but still need to consider the interface to make it more extensive; slightly refactored the code style of the codebase; more comments and todos will be in the next commit 2017-12-17 12:52:00 +08:00
Dong Yan
31199c7d0d 0. code refactor, try to merge Go and GoEnv 2017-12-16 23:29:11 +08:00
Dong Yan
01c0c2483a check if the network weights exists for every player 2017-12-16 14:55:19 +08:00
Dong Yan
d115c586d4 start the player server in a more robost way. 2017-12-16 14:33:31 +08:00
Dong Yan
4fc50c5f1b merge class strategy with class game. Next, merge Go with GoEnv 2017-12-15 22:19:44 +08:00
rtz19970824
d0bdccc25a assign TODO to Haosheng and Tongzheng 2017-12-15 14:27:04 +08:00
rtz19970824
cb9540b91c Merge branch 'master' of https://github.com/sproblvem/tianshou 2017-12-15 14:24:15 +08:00
rtz19970824
e5bf7a9270 implement dqn loss and dpg loss, add TODO for separate actor and critic 2017-12-15 14:24:08 +08:00
Haosheng Zou
92deae9f8d minor fix 2017-12-14 19:46:38 +08:00
Haosheng Zou
039c8140e2 add dqn.py to write 2017-12-13 22:43:45 +08:00
Haosheng Zou
7ab211b63c preliminary design of dqn_example, dqn interface. identify the assign of networks 2017-12-13 20:47:45 +08:00
Wenbo Hu
657422a4ed
Merge pull request #1 from sproblvem/add_rules
Add rules
2017-12-13 14:35:39 +08:00
Wenbo Hu
3f3d7b56f5 minor indent fix 2017-12-12 23:16:50 +08:00
Wenbo Hu
d52ee30259 add nearby stones 2017-12-12 23:13:31 +08:00