Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / marlbenchmark/on-policy issues and pull requests
#116 - not resetting the environment
Issue -
State: open - Opened by JBY-Faith about 1 month ago
#115 - algorithms in MPE don't have policy.actor and policy.critic
Issue -
State: open - Opened by h1657 about 1 month ago
#114 - Update on-policy to work with pettingzoo environment
Pull Request -
State: closed - Opened by conor-wallace about 2 months ago
#113 - fix: r_mappo & mat buffer factor unpack
Pull Request -
State: closed - Opened by HosnLS 4 months ago
#112 - MAPPO simple spread not converging
Issue -
State: open - Opened by zhangmingcheng28 4 months ago
- 2 comments
#111 - Master
Pull Request -
State: closed - Opened by Malhaar-k 4 months ago
#110 - Can't the mat algorithm be called yet?
Issue -
State: closed - Opened by BUCK111111 5 months ago
- 1 comment
#109 - Cannot reproduce MPE simple_speaker_listener
Issue -
State: open - Opened by AlbertoSinigaglia 6 months ago
- 2 comments
#108 - 求助FP和AS状态!!
Issue -
State: closed - Opened by 1999Zzh 6 months ago
- 1 comment
#107 - fixed bug in initializing ValueNorm
Pull Request -
State: closed - Opened by Nikunj-Gupta 7 months ago
#106 - shared_buffer.py is an outdated file
Issue -
State: closed - Opened by sixie10 7 months ago
#105 - Does the state of the grid world need to be normalized?
Issue -
State: closed - Opened by saiyuhang123 7 months ago
- 1 comment
#104 - Why share_obs in shared/mpe_runner.py just repeats obs when self.use_centralized_V set to true
Issue -
State: closed - Opened by Jaroan 8 months ago
- 1 comment
#103 - 无法解析导入“wandb”
Issue -
State: open - Opened by jun-running 8 months ago
- 1 comment
#102 - Action mask!
Issue -
State: open - Opened by shushushulian 9 months ago
- 1 comment
#101 - Shape of buffered log_probs
Issue -
State: open - Opened by Maxtoq 9 months ago
#100 - 'on-policy\onpolicy\scripts\train_smac_scripts\train_smac_8m_vs_9m.sh' training script file bug fix
Pull Request -
State: closed - Opened by JensenLZX 9 months ago
#99 - Model...
Issue -
State: closed - Opened by bitbjt 10 months ago
- 1 comment
#98 - Error when run ./train_mpe_spread.sh
Issue -
State: open - Opened by ChuangZhang1999 10 months ago
- 2 comments
#97 - When I tried to train the code for smacv2, I encountered this error message——AssertionError: check recurrent policy!
Issue -
State: closed - Opened by chenzihan1 10 months ago
- 1 comment
#95 - Update Overcooked
Pull Request -
State: closed - Opened by Aidilele 12 months ago
#94 - missing "onpolicy.runner.separated.hanabi_runner_forward"
Issue -
State: closed - Opened by AlbertoSinigaglia 12 months ago
- 1 comment
#93 - question about reply buffer size in MAPPO
Issue -
State: open - Opened by Gloriabhsfer 12 months ago
#92 - 请问我该如何保存replay?
Issue -
State: open - Opened by BUPT-zeld151 about 1 year ago
#91 - 模型性能问题
Issue -
State: closed - Opened by bitbjt about 1 year ago
- 3 comments
#90 - HAPPO为何与mappo的policy文件完全相同
Issue -
State: closed - Opened by strivebfq about 1 year ago
#89 - Runtime Error: Tensors on Different Devices (cuda:0 and cpu)
Issue -
State: closed - Opened by itstyren about 1 year ago
- 1 comment
#88 - add if stmt to handle Policy constructor calling
Pull Request -
State: closed - Opened by colourfulspring about 1 year ago
#87 - __init__() got multiple values for argument 'device'
Issue -
State: open - Opened by colourfulspring about 1 year ago
- 2 comments
#86 - fix smacv2 dones bugs and a type bug
Pull Request -
State: closed - Opened by jason-huang03 about 1 year ago
#85 - Doubts about the results of rmappo/mappo in SMAC
Issue -
State: closed - Opened by linshi9658 about 1 year ago
- 1 comment
#84 - render_mpe produces much worse performance than training or evaluation
Issue -
State: closed - Opened by wenshuaizhao about 1 year ago
- 2 comments
#83 - Questions on the episode length of 1000 on gfootball env instead of a maximum env limit of 400
Issue -
State: open - Opened by DeeDive about 1 year ago
- 1 comment
#82 - add other marl algorithms
Pull Request -
State: closed - Opened by Sapio-S over 1 year ago
#81 - Fixed render errors
Pull Request -
State: closed - Opened by XHN-1 over 1 year ago
#80 - Update shared_buffer.py
Pull Request -
State: closed - Opened by HosnLS over 1 year ago
#79 - reder问题
Issue -
State: closed - Opened by sdabj over 1 year ago
- 1 comment
#78 - How to caculate rewards in MARL environments?
Issue -
State: closed - Opened by BUPT-zeld151 over 1 year ago
- 1 comment
#77 - Modify action space creation (MPE)
Pull Request -
State: closed - Opened by XHN-1 over 1 year ago
#76 - Update README.md on installation guideline
Pull Request -
State: closed - Opened by DeeDive over 1 year ago
#75 - add smacv2 scripts
Pull Request -
State: closed - Opened by Sapio-S over 1 year ago
#74 - No smac_runner in the separated file
Issue -
State: closed - Opened by Precola over 1 year ago
- 1 comment
#73 - How to speed up training on hanabi
Issue -
State: closed - Opened by yanxue7 over 1 year ago
- 1 comment
#72 - bug in football_runner.py
Issue -
State: closed - Opened by Random8756 over 1 year ago
- 1 comment
#71 - the value for the available_actions
Issue -
State: open - Opened by EddyJason over 1 year ago
- 3 comments
#70 - The running for the simple_speaker_listener
Issue -
State: closed - Opened by EddyJason over 1 year ago
- 1 comment
#69 - Wrong with the reward in the simple_speaker_listener.py
Issue -
State: closed - Opened by EddyJason over 1 year ago
- 1 comment
#68 - MatplotlibDeprecationWarning
Issue -
State: closed - Opened by EddyJason over 1 year ago
- 1 comment
#67 - terminal error-render_mpe.sh
Issue -
State: closed - Opened by EddyJason over 1 year ago
- 5 comments
#66 - gfootball win_rate calculation
Issue -
State: closed - Opened by wenshuaizhao over 1 year ago
- 3 comments
#65 - [Environment] gfootball verision
Issue -
State: closed - Opened by LXXXXR almost 2 years ago
- 4 comments
#64 - continues action space(Box)
Issue -
State: closed - Opened by SatriaChen almost 2 years ago
- 1 comment
#63 - Update README.md
Pull Request -
State: closed - Opened by jxwuyi almost 2 years ago
#62 - Time consumption of Hanabi
Issue -
State: closed - Opened by 0xWelt almost 2 years ago
- 2 comments
#61 - Update distributions.py
Pull Request -
State: closed - Opened by SamueleBolotta almost 2 years ago
#60 - .
Issue -
State: closed - Opened by papercode2022 almost 2 years ago
#59 - Reproducing IPPO results.
Issue -
State: closed - Opened by RuanJohn about 2 years ago
- 1 comment
#58 - An error occurs when I run rmappo on football
Issue -
State: closed - Opened by cugbbaiyun about 2 years ago
- 3 comments
#57 - Can you open-source MASAC code base?
Issue -
State: closed - Opened by kailashg26 about 2 years ago
- 1 comment
#56 - M: shared runner. Bugfix.
Pull Request -
State: closed - Opened by HosnLS about 2 years ago
#55 - training MAPPO and RMAPPO with MPE gives an error!
Issue -
State: closed - Opened by kailashg26 about 2 years ago
- 2 comments
#54 - Clarification on Environment Names
Issue -
State: closed - Opened by aravindvenu7 about 2 years ago
- 2 comments
#53 - M: base runner. Add ValueNorm save/restore.
Pull Request -
State: closed - Opened by HosnLS about 2 years ago
#52 - Update mpe_runner.py
Pull Request -
State: closed - Opened by SamueleBolotta about 2 years ago
#51 - Update ValueNorm
Pull Request -
State: closed - Opened by HosnLS about 2 years ago
#50 - Update mpe_runner.py
Pull Request -
State: closed - Opened by SamueleBolotta over 2 years ago
#49 - cannot reproduce the performance of MPE
Issue -
State: closed - Opened by hccz95 over 2 years ago
- 3 comments
#48 - cannot reproduce the performance of MPE
Issue -
State: closed - Opened by hccz95 over 2 years ago
#47 - Being confused about The huber_loss
Issue -
State: closed - Opened by Atan03 over 2 years ago
- 1 comment
#46 - Cannot reproduce the MPE results
Issue -
State: closed - Opened by wenshuaizhao over 2 years ago
- 2 comments
#46 - Cannot reproduce the MPE results
Issue -
State: closed - Opened by wenshuaizhao over 2 years ago
- 2 comments
#45 - Error in setting `bad_transition`
Issue -
State: closed - Opened by xihuai18 over 2 years ago
- 1 comment
#44 - VecEnvWrapper is not defined in env_wrappers.py
Issue -
State: closed - Opened by SamueleBolotta over 2 years ago
- 3 comments
#43 - "Vec_env_wrapper" and "Share_vec_evn" not defined in "env_wrappers.py"
Issue -
State: closed - Opened by SamueleBolotta over 2 years ago
#42 - Using a global state
Issue -
State: closed - Opened by SamueleBolotta over 2 years ago
- 1 comment
#41 - The problem about actor loss
Issue -
State: closed - Opened by rayLrayL over 2 years ago
#40 - Differences between the starcraft environment and environment in SMAC
Issue -
State: closed - Opened by waterhorse1 over 2 years ago
- 2 comments
#39 - In the definition of FixedNormal class have some wrong code.
Issue -
State: closed - Opened by chillybird over 2 years ago
- 4 comments
#38 - Why use self.buffer[agent_id].after_update()?
Issue -
State: closed - Opened by LiZhYun over 2 years ago
- 6 comments
#37 - How to train MAPPO on MPE with the separated policy?
Issue -
State: closed - Opened by whz204168-lab over 2 years ago
- 2 comments
#36 - continuous action space
Issue -
State: closed - Opened by ollehhello over 2 years ago
- 11 comments
#35 - The wandb backend process has shutdown
Issue -
State: closed - Opened by bigssmart over 2 years ago
- 1 comment
#34 - A few questions about death masking
Issue -
State: closed - Opened by Denys88 almost 3 years ago
- 2 comments
#33 - About Hanabi Implementation Details
Issue -
State: closed - Opened by 0xJchen almost 3 years ago
- 1 comment
#32 - Metric problem about MAPPO
Issue -
State: closed - Opened by PaParaZz1 almost 3 years ago
- 2 comments
#31 - separate runner for SMAC
Issue -
State: closed - Opened by xubo92 almost 3 years ago
#30 - Centralized-V between IPPO and MAPPO
Issue -
State: closed - Opened by xubo92 almost 3 years ago
- 4 comments
#29 - Is seperated runner fully supported?
Issue -
State: closed - Opened by xubo92 almost 3 years ago
- 2 comments
#28 - Does the param 'n_rollout_threads' influence the final performance greatly?
Issue -
State: closed - Opened by chaobiubiu almost 3 years ago
- 1 comment
#27 - 在MAPPO中运行MPE的speaker_listener在7个智能体环境中的实验问题
Issue -
State: closed - Opened by AI-Hello almost 3 years ago
- 1 comment
#26 - permission denied
Issue -
State: closed - Opened by weixinguang almost 3 years ago
- 1 comment
#25 - About QMix(MG)
Issue -
State: closed - Opened by Henry668 about 3 years ago
- 5 comments
#24 - W & B errors about running train_mpe
Issue -
State: closed - Opened by AI-Hello about 3 years ago
- 1 comment
#23 - Is MAPPO similar or same with the Parameter sharing framework in one earlier work?
Issue -
State: closed - Opened by wenshuaizhao about 3 years ago
- 1 comment
#22 - QMIX performance in Hanabi
Issue -
State: closed - Opened by 51616 about 3 years ago
- 2 comments
#21 - Values of Hyperparameters
Issue -
State: closed - Opened by gbyuHub over 3 years ago
- 2 comments
#20 - how render
Issue -
State: closed - Opened by zhaichenhao1889 over 3 years ago
- 1 comment
#19 - TypeError: cannot assign 'torch.FloatTensor' as parameter 'stddev' (torch.nn.Parameter or None expected)
Issue -
State: closed - Opened by DosHacker over 3 years ago
- 1 comment
#18 - Update README.md
Pull Request -
State: closed - Opened by eugenevinitsky over 3 years ago