Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / kwai/DouZero issues and pull requests
#83 - 卡牌编码问题
Issue -
State: closed - Opened by lucky666liguan 4 days ago
#82 - 能否分享一下sl训练的代码? 不需要分享人类的训练数据!
Issue -
State: open - Opened by AlphaYe 6 days ago
- 1 comment
#81 - 在Windows系统上使用GPU进行模拟时,遇到 operation not supported 错误
Issue -
State: open - Opened by linkerlin 3 months ago
#80 - [readme suggestion] A gif is not good for demonstration, an embedded video would be much better
Issue -
State: open - Opened by zzz6519003 3 months ago
#79 - C#可以直接调用训练好后的模型吗?
Issue -
State: open - Opened by oookidoo 7 months ago
#78 - what's the output dimension of actor network?
Issue -
State: open - Opened by zkengz 7 months ago
#77 - 有没有大佬有学习好的斗地主的模型?
Issue -
State: open - Opened by QQ35081215 8 months ago
#76 - 请问如何使用baseline进行辅助训练,而不是单纯的evaluate
Issue -
State: open - Opened by cxk555 10 months ago
#75 - 如何查是否过拟合或者欠拟合呢
Issue -
State: open - Opened by XliOK 11 months ago
#74 - 当我使用5张A100 80G训练时,如何提高训练速度或者应该如何修改符合机器的参数
Issue -
State: open - Opened by yanzm4 12 months ago
- 2 comments
#73 - 有人大佬实现了癞子玩法了吗?
Issue -
State: open - Opened by lixiaozhui 12 months ago
- 1 comment
#72 - Consideration of suit of 3 landlord cards
Issue -
State: open - Opened by zizhang-qiu 12 months ago
#71 - 想询问一下baseline里面的ai训练了多久,我训练了几个礼拜了,表现感觉不如baseline
Issue -
State: closed - Opened by GGG3313 about 1 year ago
- 2 comments
#70 - EOFError: Ran out of input
Issue -
State: open - Opened by lingyunQAQ about 1 year ago
- 2 comments
#69 - 在colab在跑训练,时间到了可以从上一次继续么
Issue -
State: closed - Opened by lingyunQAQ about 1 year ago
- 1 comment
#68 - 勾吧pretrain models都没有 一训练全是0
Issue -
State: closed - Opened by SchoolBug over 1 year ago
- 2 comments
#67 - actor网络在收集数据时使用了model前向传播,leaner对网络参数再进行跟新,这两者对同一共享内存更新是否会有冲突?
Issue -
State: open - Opened by mengxi7 over 1 year ago
- 1 comment
#66 - 另外,训练过程中注意CPU和显存 我是actor用CPU,training用GPU,CPU一定要留余地,同时thread不要加太高 否则容易爆显存
Issue -
State: open - Opened by FYNIXqwq over 1 year ago
- 1 comment
#65 - 一点想法 关于不洗牌模式的训练
Issue -
State: open - Opened by FYNIXqwq over 1 year ago
- 9 comments
#64 - 关于深度学习与强化学习
Issue -
State: closed - Opened by Zz-Quan over 1 year ago
- 2 comments
#62 - 欢乐斗地主实测胜率很低 算法还有升级的办法吗?
Issue -
State: open - Opened by xinyujituan almost 2 years ago
- 2 comments
#61 - Added Resnet_01
Pull Request -
State: open - Opened by JXu97 almost 2 years ago
- 1 comment
#60 - colab上的训练代码一运行,ram就炸了
Issue -
State: open - Opened by jdy18 almost 2 years ago
- 3 comments
#59 - [fix] bug 修复4带两对,带了个炸弹的情况。
Pull Request -
State: closed - Opened by tinyzqh almost 2 years ago
#58 - Apple M1芯片支持模拟和训练吗
Issue -
State: closed - Opened by zhaowenzhi almost 2 years ago
- 1 comment
#57 - 多显卡瓶颈在第0卡上
Issue -
State: closed - Opened by SvenNJ almost 2 years ago
#56 - 咨询下LSTM策略调整
Issue -
State: closed - Opened by SvenNJ almost 2 years ago
#55 - 策略可能性
Issue -
State: closed - Opened by SvenNJ almost 2 years ago
#54 - 执行loss.backward()几个小时报错out of memory
Issue -
State: closed - Opened by SvenNJ almost 2 years ago
- 7 comments
#53 - 策略调整可行性
Issue -
State: closed - Opened by SvenNJ almost 2 years ago
- 2 comments
#52 - 求助,用CPU无法训练问题!!!
Issue -
State: open - Opened by BESTTENG about 2 years ago
- 1 comment
#51 - 抢地主功能什么时候开放
Issue -
State: closed - Opened by SvenNJ about 2 years ago
- 1 comment
#50 - 关于models.py 几个问题
Issue -
State: closed - Opened by SvenNJ about 2 years ago
- 7 comments
#49 - 为什么同样是4个GPU我的训练时候的FPS很低呢,基本都在2000左右
Issue -
State: open - Opened by mfxiaosheng about 2 years ago
- 7 comments
#48 - 显存很快就满了
Issue -
State: open - Opened by Julong-Zheng over 2 years ago
- 8 comments
#47 - 没有找到main文件
Issue -
State: open - Opened by amadeus11111 over 2 years ago
- 2 comments
#46 - 关于evaluation阶段
Issue -
State: open - Opened by AlexWanghaoming over 2 years ago
- 1 comment
#45 - 新手求问free_queue和full_queue分别是什么意思?这个就是actor和learner之间的缓存区吗?
Issue -
State: closed - Opened by czqaq over 2 years ago
- 1 comment
#44 - 使用人类数据训练SL模型
Issue -
State: open - Opened by ctycode over 2 years ago
- 3 comments
#43 - AI会把大小王作为四带二的二带出去,这是明显掉胜率的走法
Issue -
State: open - Opened by changjiuxiong over 2 years ago
- 4 comments
#42 - 我自己train了一下,为什么出来的结果不像你们训练出的那样智能?
Issue -
State: open - Opened by profection over 2 years ago
- 15 comments
#41 - can you share the SL training code?
Issue -
State: open - Opened by jokerlc over 2 years ago
- 6 comments
#40 - 请教下如果是其它规则的斗地主要如果修改?
Issue -
State: closed - Opened by JakeWoki over 2 years ago
- 2 comments
#39 - Bs added comments
Pull Request -
State: closed - Opened by bensenescu almost 3 years ago
- 2 comments
#38 - i cannot train by cuda 11.3 win10 gtx1060,how can i do?
Issue -
State: open - Opened by qq782645210 almost 3 years ago
- 2 comments
#37 - C++ implementation
Issue -
State: open - Opened by yffbit almost 3 years ago
- 1 comment
#36 - Would the bidding phase be open sourced?
Issue -
State: closed - Opened by weichaoguo almost 3 years ago
- 2 comments
#35 - chore: check if action is legal in GameEnv
Pull Request -
State: closed - Opened by karoka almost 3 years ago
#34 - 出牌未校验是否大过上家
Issue -
State: closed - Opened by changjiuxiong almost 3 years ago
- 3 comments
#33 - reuse "_cards2array" function
Pull Request -
State: closed - Opened by yffbit about 3 years ago
- 1 comment
#32 - 有没有必要设置size变量
Issue -
State: closed - Opened by yffbit about 3 years ago
- 3 comments
#31 - Why is the first GPU over-consumed?
Issue -
State: closed - Opened by xlnwel about 3 years ago
- 4 comments
Labels: environment
#30 - 配置需求
Issue -
State: closed - Opened by aacai about 3 years ago
- 1 comment
#29 - train.py 运行 After 0 (L:0 U:0 D:0) frames: @ 0.0 fps (avg@ 0.0 fps) (L:0.0 U:0.0 D:0.0)
Issue -
State: closed - Opened by Roywaller about 3 years ago
- 4 comments
#28 - Add CPU support for actor process
Pull Request -
State: closed - Opened by Vincentzyx about 3 years ago
#27 - 新版本在Windows下仍然无法Train?
Issue -
State: closed - Opened by Vincentzyx about 3 years ago
- 1 comment
#26 - 使用训练模型的过程中,会出现普通炸大于王炸的情况
Issue -
State: closed - Opened by jiugewx about 3 years ago
- 5 comments
#25 - SL模型的训练数据是如何实现和加载的?
Issue -
State: closed - Opened by whiplash003 about 3 years ago
- 3 comments
Labels: question
#24 - Error when fighting with RLCard in landlord_up position
Issue -
State: closed - Opened by orange90 about 3 years ago
- 1 comment
#23 - how to avoid interrupt, when learner model loads weights to the actor model while it is running 'forward' in 'act' function
Issue -
State: open - Opened by huzhoudaxia about 3 years ago
- 1 comment
#22 - why use 'Lock' in "get_batch()" while using queue to communication between threads
Issue -
State: closed - Opened by huzhoudaxia about 3 years ago
- 3 comments
#21 - why use multiple threads to execute the function "batch_and_learn" in dmc.py.
Issue -
State: closed - Opened by huzhoudaxia about 3 years ago
- 2 comments
#20 - 为什么evaluate出错?
Issue -
State: closed - Opened by leolle over 3 years ago
- 6 comments
Labels: environment
#19 - Actions generate by model differ from online demo
Issue -
State: closed - Opened by boscotsang over 3 years ago
- 1 comment
#18 - 关于模型版本的问题
Issue -
State: closed - Opened by cookie-yang over 3 years ago
- 1 comment
Labels: question
#17 - env/env.py _get_obs_landlord函数的疑问
Issue -
State: closed - Opened by cookie-yang over 3 years ago
- 1 comment
#16 - 请问基于强化学习?数据怎么得到?
Issue -
State: closed - Opened by henbucuoshanghai over 3 years ago
- 1 comment
Labels: question
#15 - jim
Issue -
State: closed - Opened by Starshowss over 3 years ago
Labels: off topic
#14 - 弄
Issue -
State: closed - Opened by jianghushao over 3 years ago
Labels: off topic
#13 - 关于算法
Issue -
State: closed - Opened by budaLi over 3 years ago
- 3 comments
#12 - 斗得过人吗?
Issue -
State: closed - Opened by hyuan1991 over 3 years ago
- 1 comment
Labels: question
#11 - update readme; update demo gif link
Pull Request -
State: closed - Opened by hsywhu over 3 years ago
#10 - 2个问题
Issue -
State: closed - Opened by Fengzhijjin over 3 years ago
- 3 comments
Labels: question
#8 - Ch
Issue -
State: closed - Opened by natem122787 over 3 years ago
- 1 comment
Labels: off topic
#7 - 什么时候支持欢乐斗地主?
Issue -
State: closed - Opened by tikazyq over 3 years ago
- 1 comment
Labels: question
#6 - 能协助我斗地主吗
Issue -
State: closed - Opened by zhengzhen0417 over 3 years ago
- 18 comments
Labels: question
#5 - 这算游戏bug吗?
Issue -
State: closed - Opened by dfaofeng over 3 years ago
- 1 comment
#4 - doc: add personal websites
Pull Request -
State: closed - Opened by NOBLES5E over 3 years ago
#3 - 亲测,效果一般。对战水平低下,我这样一般人胜利都在50%以上
Issue -
State: closed - Opened by luyifanlu over 3 years ago
- 2 comments
#2 - 请问能实现像AlfaGo一样用于与人类对战吗?
Issue -
State: closed - Opened by NLPIG over 3 years ago
- 3 comments
Labels: question
#1 - lol~ , interesting job
Issue -
State: closed - Opened by GitHubChrischen over 3 years ago
- 1 comment