MAPPO Implementation - #721

Open

Harshul-18 wants to merge 52 commits into

assume-framework:mainfrom

Harshul-18:with_PPO_and_DDPG

Commits on Jan 9, 2026

DONE: Added DDPG, PPO in multi-agent environment in /reinforcement_learning module.
Harshul-18
committed
1fec8d4
DONE: Added DDPG, PPO in multi-agent environment in /reinforcement_learning module.
Harshul-18
committed
8385449

Commits on Jan 12, 2026

Commits on Jan 14, 2026

Commits on Feb 16, 2026

DONE: Added DDPG, PPO in multi-agent environment in /reinforcement_learning module.
Harshul-18
committed
8d39963
DONE: Added DDPG, PPO in multi-agent environment in /reinforcement_learning module.
Harshul-18
committed
7cc7035
UPDATED ppo-input-pipeline, code-documentation DELETED rollout_buffer.py ADDED RolloutBuffer-in-buffer.py
Harshul-18
committed
fc2f6e0
FIX: initial values_data assignment
Harshul-18
committed
78d8033
updated the structure and segmented the parameters for off-policy and on-policy algorithms
Harshul-18
committed
36292ac
fixed independent working of on-policy algorithms
Harshul-18
committed
a464155
Improved the parameters flow independently and fixed different buffer definitions to defining a single buffer argument
Harshul-18
committed
d5521ab
Updated base_algorithm.py documentation.
Harshul-18
committed
fddaf2c
Updated maddpg.py documentation.
Harshul-18
committed
239fb43
Merge branch 'with_PPO_and_DDPG' of https://github.com/Harshul-18/assume into with_PPO_and_DDPG
Harshul-18
committed
0c78a24
temporary fix
Harshul-18
committed
19c975a
Updated the example_02a config file to test the scenario
Harshul-18
committed
7727cb9

Commits on Feb 28, 2026

Commits on Apr 1, 2026

fixed the PPO implementation, the policy update logic, corrected the ordering for the agent
Harshul-18
committed
171b5e9

Commits on Apr 9, 2026

added 'get_distribution' method and splitted the forward method into _extract_features and forward to avoid redundancy in LSTMActorPPO
Harshul-18
committed
47349da

Commits on Apr 23, 2026

Merge branch 'main' into with_PPO_and_DDPG
Harshul-18
committed
937bc05

Commits on Apr 24, 2026

integrated MAPPO pipeline in other files
Harshul-18
committed
563ea05

Commits on Apr 25, 2026

Fix Path import bug in learning_role.py — fixing the runtime error for continue_learning
Harshul-18
committed
cdfc3d6

Commits on Apr 26, 2026

Fixed assume import logic API
Harshul-18
committed
90ca112
Fixed test_matd3.py to use the nested off_policy config structure
Harshul-18
committed
c8a2059
Added Rollout Buffer test file (test_rl_rolloutbuffer.py)
Harshul-18
committed
89eb665
Added MADDPG test cases (test_maddpg.py file)
Harshul-18
committed
24364ef
Updated the docs and completed the MADDPG implementation
Harshul-18
committed
a1442a8
Merge branch 'with_PPO_and_DDPG_merge' into with_PPO_and_DDPG
Harshul-18
committed
d34c4d9

Commits on Apr 28, 2026

Added MAPPO test file (test_mappo.py)
Harshul-18
committed
0ba82b7
Moved get_action from learning_strategies to RLAlgorithm
Harshul-18
committed
eb6522f
Merge remote-tracking branch 'upstream/main' into with_PPO_and_DDPG
Harshul-18
committed
28cb3a0

Commits on May 7, 2026

Fixed the test_learning_role.py by updating the OffPolicyConfig parameters passing structure
Harshul-18
committed
d2f8f9d

Commits on May 8, 2026

Commits on May 16, 2026

Fixed update_policy's strategies listing, temporarily fixed circular import issue, fixed convert_to_tensors function calling in buffer.py
Harshul-18
committed
b0b8457

Commits on May 17, 2026

Removed sorting from MAPPO pipeline, Removed adding default values instead warning generation, fixed last_idx/last_dones in buffer handling (mappo.py), Reverted back to the main/upstream branch's t…
Harshul-18
committed
eb3990a

Commits on Jul 5, 2026

Fix buffer setup and MAPPO policy loading.
Harshul-18
committed
bcc450b