FLAIROx JaxMARL issues - Githubissues

FLAIROx / JaxMARL

Multi-Agent Reinforcement Learning with JAX

Apache License 2.0

395 stars 69 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Generalised STORM environment implementation.

#114 ali-shihab closed 1 week ago
6
fix bug in jaxnav's agent agent collisions for single agent case

#113 amacrutherford closed 2 weeks ago
0
Jaxnav docs & improvements

#112 amacrutherford closed 2 weeks ago
0
Checkpointing

#111 shray closed 2 weeks ago
1
SMACv2

#110 balloch closed 2 weeks ago
3
Question about refactoring baselines/IPPO/ippo_ff_overcooked.py

#109 LukavdBoogaard closed 1 week ago
1
AttributeError: 'Config' object has no attribute 'define_bool_state'

#108 paulzyzy closed 1 week ago
3
made coin game compatible with iql_rnn

#107 Dronie closed 1 week ago
3
Hanabi Legal Actions Bugfix

#106 tindiz closed 1 month ago
0
increment version before release

#105 amacrutherford closed 2 months ago
0
update of qlearning scripts and add pqn

#104 mttga closed 2 months ago
0
Hanabi Environment Fixes & Action Space Refactoring

#103 tindiz closed 2 months ago
0
Question regarding Overcooked V2 observations

#102 ConstantinRuhdorfer opened 2 months ago
2
Failed to Run Baseline iql.py in mpe_spread without parameters sharing

#101 iamlilAJ closed 2 months ago
1
Difference between “unit list” vs “conic” obs?

#100 Chulabhaya closed 2 months ago
1
JaxNav, base class update, version increment, scipy bound

#99 amacrutherford closed 3 months ago
5
Update smax_env.py

#98 syrkis closed 3 months ago
1
Non root Docker, Unpin Jax req, small IPPO qual of life improvements, MaBrax ReadMe note

#97 amacrutherford closed 3 months ago
1
Fix hint masking dimensions and restructure deck insertion.

#96 tindiz closed 2 months ago
3
Naive question about multiple seeds

#95 kfu02 opened 3 months ago
3
Redo

#94 kwantlin closed 3 months ago
0
Probable issue with implementation of VDN

#93 ElliotXinqiWang closed 3 months ago
1
typo

#92 amacrutherford closed 4 months ago
0
To what extent is it possible to use SMAX in code that was originally designed for SMAC v1/PyTorch?

#91 Chulabhaya closed 4 months ago
2
Fix action-direction mismatch

#90 lupuandr closed 4 months ago
0
Fix out of bounds

#89 benellis3 closed 4 months ago
0
Best way to create new SMAX scenarios?

#88 Chulabhaya closed 4 months ago
2
fix order of MAPPO reshape in MPE & Hanabi

#87 amacrutherford closed 4 months ago
3
nit; should be `critic_network.apply`?

#86 victor-qin closed 4 months ago
2
Dimensions of `world_state` and `reward` do not match

#85 victor-qin closed 4 months ago
1
Feature request: Support POLA

#84 cool-RR closed 3 months ago
5
Jax version

#83 alexunderch closed 3 months ago
9
Gru step size fix

#82 amacrutherford closed 5 months ago
0
Strange behaviour from agents in coin game environment

#81 Dronie closed 3 months ago
5
GRU_HIDDEN_DIM/FC_DIM_SIZE and NUM_STEPS still tied to each other

#80 satpreetsingh closed 5 months ago
1
add eval test generation for 4 cases

#79 collinfeng closed 5 months ago
2
added Jitable eval_reset for test cases generation

#78 collinfeng closed 5 months ago
0
fix bug with rnn policy reconstruction, remove dependence on num_steps, fix smax conic obs space size

#77 amacrutherford closed 5 months ago
2
Hanabi af

#76 collinfeng closed 5 months ago
0
Run tests in docker container, correct jax dependency

#75 amacrutherford closed 5 months ago
0
Baselines for STORM (PPO working with individual rewards instead of team rewards)

#74 Ueshima73 closed 5 months ago
8
IPPO trained model shows poor test performance

#73 satpreetsingh closed 5 months ago
2
IPPO inference + save animation

#72 satpreetsingh closed 5 months ago
7
Corrected Hanabi, new Dockerfile, python 3.10 and other fixes

#71 mttga closed 5 months ago
5
Unable to apply Q-learning baselines on envs with non-homogeneous agents

#70 zez2001 closed 5 months ago
3
Add different observation and action space

#69 benellis3 closed 6 months ago
0
Why are HIDDEN_SIZE and NUM_STEPS tied to 128?

#68 satpreetsingh closed 5 months ago
2
Refactor

#67 clarech712 closed 6 months ago
0
Unable to replicate performance with Q-Learning on SMAX

#66 corentinartaud closed 5 months ago
5
Suggestion of integrating HARL algorithms

#65 fmxFranky closed 6 months ago
1