Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / junxiaosong/AlphaZero_Gomoku issues and pull requests

#139 - 怎么训练其它棋盘大小的模型

Issue - State: open - Opened by Startshark 4 months ago

#138 - typo: rename mcts_porb to mcts_prob

Pull Request - State: closed - Opened by RealHurrison 5 months ago

#136 - Update policy_value_net_numpy.py

Pull Request - State: closed - Opened by ya0guang 6 months ago

#132 - Keras implementation is not RL

Issue - State: open - Opened by itschenxi over 1 year ago

#131 - 算法并非自己对战

Issue - State: closed - Opened by itschenxi almost 2 years ago

#130 - 关于pytorch训练无法收敛的问题

Issue - State: open - Opened by hxb123622 almost 2 years ago - 6 comments

#129 - 各位训练了多久，能下赢自己？

Issue - State: open - Opened by lvsh2012 almost 2 years ago

#128 - Added tensorflow V2 model

Pull Request - State: open - Opened by wangjia184 about 2 years ago

#127 - Added support of tensorflow v2

Pull Request - State: closed - Opened by wangjia184 about 2 years ago

#126 - Siauhoo

Pull Request - State: open - Opened by siauhoo over 2 years ago

#125 - 为什么要在'play_data'中反转输入特征？

Issue - State: open - Opened by Michi-123 over 2 years ago

#124 - 请问如何使用自己训练的权重

Issue - State: open - Opened by zsqdx over 2 years ago

#123 - python GUI + 代码分析

Issue - State: open - Opened by YoujiaZhang almost 3 years ago

#122 - Add res30 PolicyValueNet and gobang

Pull Request - State: open - Opened by minkefusiji over 3 years ago

#121 - roll out in mcts_pure

Issue - State: closed - Opened by yunanyan over 3 years ago

#120 - Daoz/add resnet for tf

Pull Request - State: closed - Opened by wangershi over 3 years ago

#119 - 非常想请教一个问题，希望大佬回复

Issue - State: closed - Opened by Lufffya over 3 years ago - 4 comments

#118 - 请问这个网络是多少*多少的大小

Issue - State: open - Opened by MsvcPytorch over 3 years ago - 1 comment

#117 - Disable TensorFlow v2 behavior

Pull Request - State: open - Opened by elmines over 3 years ago

#116 - 用Pytorch训练保存模型，在human_play中加载模型报错

Issue - State: open - Opened by bxdxstu over 3 years ago - 2 comments

#115 - 关于训练中断

Issue - State: open - Opened by racerlucas over 3 years ago

#114 - 修改了游戏规则后训练中的问题

Issue - State: closed - Opened by AnterCreeper almost 4 years ago

#113 - alpha zero是如何避免在不可行的位置落子的

Issue - State: open - Opened by ZhangXi20181002 almost 4 years ago - 2 comments

#112 - Differences between model amd model2

Issue - State: open - Opened by lironT74 about 4 years ago

#111 - 关于训练过程的疑问

Issue - State: open - Opened by dx-feng about 4 years ago

#110 - Potential error

Issue - State: open - Opened by Fernadoo about 4 years ago - 1 comment

#109 - pytorch是否使用了GPU的问题

Issue - State: open - Opened by Okiii-lh over 4 years ago - 3 comments

#108 - mcts_pure.py 中的 get_action() 函数 update_with_move(-1) 是否有误？

Issue - State: closed - Opened by honeyshine75 over 4 years ago - 2 comments

#107 - 增加multiprocess支持

Pull Request - State: closed - Opened by dofish over 4 years ago - 1 comment

#106 - Do you have model of 15x15 board?

Issue - State: open - Opened by herogan2017 over 4 years ago - 1 comment

#104 - 看完代码觉得这个Implemention有问题，欢迎指正

Issue - State: closed - Opened by ylf11235 over 4 years ago - 1 comment

#102 - 如何用 GPU 得到 self-play 数据呢？

Issue - State: open - Opened by cloxnu over 4 years ago

#101 - TensorFlow input_states有些疑问

Issue - State: open - Opened by MaybeChow1991 over 4 years ago - 1 comment

#100 - 关于CNN卷积核mask的问题

Issue - State: closed - Opened by huyp182 over 4 years ago

#100 - 关于CNN卷积核mask的问题

Issue - State: closed - Opened by huyp182 over 4 years ago

#96 - Keras训练出来的模型无法加载

Issue - State: open - Opened by LLYYKK over 4 years ago - 1 comment

#95 - values are the same for different inputs

Issue - State: open - Opened by yueyilia almost 5 years ago

#92 - 关于AMD CPU利用率的问题

Issue - State: open - Opened by kouzui98 about 5 years ago

#91 - 关于策略网络和价值网络

Issue - State: open - Opened by zhang-qiang-github about 5 years ago

#90 - ORENIST.data　ModuleNotFoundError: No module named 'numpy.core.multiarray\r'

Issue - State: open - Opened by Go-Tak over 5 years ago - 1 comment

#89 - Why the temperature parameter is always 1?

Issue - State: closed - Opened by terryzhao127 over 5 years ago

#88 - 使用项目中TensorFlow训练的时候出现memory error怎么解决的啊

Issue - State: open - Opened by liquorxm over 5 years ago - 2 comments

#87 - "play_data = list(play_data)[:]" vs "play_data = list(play_data)"

Issue - State: open - Opened by IanFan over 5 years ago - 2 comments

#85 - What do "explained_var_old" and "explained_var_new" imply?

Issue - State: closed - Opened by bacTlink over 5 years ago - 1 comment

#84 - Update game.py

Pull Request - State: closed - Opened by autodataming over 5 years ago

#83 - fix index error when using higher pytorch version

Pull Request - State: closed - Opened by mrmitzh over 5 years ago

#82 - Fail to train policy network

Issue - State: closed - Opened by mrmitzh over 5 years ago - 1 comment

#81 - I win , computer lose,hahaha

Issue - State: open - Opened by AI-liu over 5 years ago - 1 comment

#80 - 支持

Issue - State: closed - Opened by Fangqingg over 5 years ago

#79 - pure MCTS FYI

Issue - State: open - Opened by zhenfelix over 5 years ago

#78 - 如何把现在的Net网络改成Resnet网络

Issue - State: closed - Opened by liunian321 over 5 years ago - 4 comments

#77 - MCTS权重更新问题

Issue - State: closed - Opened by sky88088 almost 6 years ago - 4 comments

#76 - tensorflow

Issue - State: open - Opened by PegasusLjj almost 6 years ago - 5 comments

#75 - 关于写禁手的疑问

Issue - State: open - Opened by tristantian almost 6 years ago - 5 comments

#73 - 关于添加Dirichlet Noise的疑问

Issue - State: open - Opened by bhpfelix almost 6 years ago - 4 comments

#71 - 关于探索度的问题

Issue - State: open - Opened by initial-h almost 6 years ago - 2 comments

#70 - 对本项目与另一个相似项目在大棋盘训练效率上差异的疑问

Issue - State: open - Opened by QvQQ almost 6 years ago - 4 comments

#69 - _playout方法中的update_recursive为什么要传入 -leaf_value?

Issue - State: open - Opened by Leonardo711 about 6 years ago - 1 comment

#68 - 关于五子棋训练的问题 ,以及explain_var_old的意义

Issue - State: open - Opened by tfliu about 6 years ago - 2 comments

#67 - tensorflow 版本的网络action_fc的激活函数是log_softmax？

Issue - State: closed - Opened by initial-h about 6 years ago - 1 comment

#66 - 关于human_play中编译报错

Issue - State: open - Opened by Curry30h about 6 years ago

#65 - 新手提问，如要运行，是否需要所有文件

Issue - State: closed - Opened by Curry30h about 6 years ago

#64 - 请问如何用tensorflow加载模型文件

Issue - State: open - Opened by moonsin about 6 years ago - 1 comment

#63 - 请问想增大网络input的维度应该在哪修改呢？

Issue - State: closed - Opened by survivingME about 6 years ago - 2 comments

#62 - 关于评估方法的疑问

Issue - State: open - Opened by Joker2770 about 6 years ago - 2 comments

#61 - 一些个人建议

Issue - State: open - Opened by Joker2770 about 6 years ago - 2 comments

#60 - use human play data.

Issue - State: open - Opened by anxingle about 6 years ago - 2 comments

#59 - 关于pure_MCTS的问题

Issue - State: open - Opened by moonsin about 6 years ago - 3 comments

#57 - 训练超级慢，有什么办法可以提高效率？

Issue - State: open - Opened by xbhuang1994 about 6 years ago - 6 comments

#56 - 这个公式是不是写错了

Issue - State: open - Opened by tianv about 6 years ago - 1 comment

#55 - Develop

Pull Request - State: closed - Opened by shbfeng about 6 years ago

#54 - 关于模型的走棋策略的问题

Issue - State: open - Opened by Egolas about 6 years ago - 5 comments

#53 - 现有的两个8*8*5的model是经过多少次训练得来的？

Issue - State: open - Opened by nothingeasy about 6 years ago - 1 comment

#52 - 对于 -leaf_value 的问题

Issue - State: closed - Opened by gmftbyGMFTBY about 6 years ago - 1 comment

#50 - 五子棋 比赛规则 不了解一下吗？？？

Issue - State: open - Opened by icetech233 over 6 years ago - 1 comment

#49 - 关于KL散度控制学习率的问题

Issue - State: open - Opened by rommeldhy over 6 years ago - 4 comments

#48 - 用tensorflow 训练1000局仍然无法收敛

Issue - State: open - Opened by yuan9778 over 6 years ago - 3 comments

#47 - 如果只有价值网络，没有策略网络可以吗？

Issue - State: open - Opened by huyp182 over 6 years ago - 1 comment

#46 - 关于c_puct的问题

Issue - State: open - Opened by rommeldhy over 6 years ago - 1 comment

#45 - 请问 如何提升使用GPU训练时的效率

Issue - State: closed - Opened by TheColdVoid over 6 years ago - 1 comment

#44 - 关于MCTS中计算Q值的方法的区别

Issue - State: open - Opened by JeffSoong over 6 years ago - 2 comments

#43 - 一个关于对局数据的问题

Issue - State: closed - Opened by GeneZC over 6 years ago - 2 comments
Labels: question

#42 - Can I change the board size?

Issue - State: open - Opened by SuminHan over 6 years ago - 1 comment

#41 - game.py的一點小問題

Issue - State: closed - Opened by karta0807913 over 6 years ago - 1 comment

#40 - 大棋盘的一点问题

Issue - State: open - Opened by gmftbyGMFTBY over 6 years ago - 7 comments