Robin Meersman
|
43de6a9670
|
merge: merged dev into branch
|
2026-05-04 22:23:58 +02:00 |
|
RobinMeersman
|
61064ca70e
|
Merge branch 'dev' into feat/message_passing
|
2026-05-04 21:44:06 +02:00 |
|
Robin Meersman
|
791dcf231c
|
other(ppo training loop): code cleanup + enabled debug logging of model parameters
|
2026-05-04 21:29:27 +02:00 |
|
Robin Meersman
|
dd2612e2ec
|
fix(ppo training): more shape-related bug fixes + moved computation of #arms and segments outside of jit compiled function
|
2026-05-04 21:26:34 +02:00 |
|
Robin Meersman
|
a43f3c0172
|
fix(ppo training): fixed shape error while clipping action
|
2026-05-04 19:36:59 +02:00 |
|
Robin Meersman
|
bb14507fc7
|
fix(message passing): forgot vmapping over params :thumbs_up:
|
2026-05-02 23:34:52 +02:00 |
|
Robin Meersman
|
3cb3c765cc
|
feat(message passing): fixed tracing bug, todo: dense layer input dimension incorrect
|
2026-05-02 23:27:31 +02:00 |
|
Robin Meersman
|
027b13b30f
|
feat(message passing): used config value for message passing repetitions
|
2026-05-02 23:15:55 +02:00 |
|
Robin Meersman
|
6114e36e27
|
feat(PPOTrainer): decentralized message passing babyyyy
|
2026-05-02 21:50:42 +02:00 |
|
|
|
70ef9d4058
|
Merge pull request #45 from SELab-3-2026/fix/input-space
|
2026-05-02 20:30:59 +02:00 |
|
JibrilExe
|
9398592e34
|
merged
|
2026-05-02 11:12:01 +02:00 |
|
JibrilExe
|
2b46b3f5d2
|
fix: only flatten actions for environ
|
2026-05-02 11:08:48 +02:00 |
|
JibrilExe
|
f77766ba08
|
fix:logging shapes
|
2026-05-02 09:38:22 +02:00 |
|
Robin Meersman
|
9f7c73b009
|
other: backup of current progress
|
2026-05-01 18:15:09 +02:00 |
|
JibrilExe
|
bceb7fe4cf
|
logging: now we can see shapes in ppo update
|
2026-05-01 11:52:57 +02:00 |
|
JibrilExe
|
951f682723
|
fix: latest attempt at fexing exploding mem leak
|
2026-05-01 11:36:25 +02:00 |
|
JibrilExe
|
e867dbab49
|
fix: dubble reshape bad
|
2026-05-01 10:43:27 +02:00 |
|
JibrilExe
|
22c43f05b1
|
fix: critic init now fixed?
|
2026-05-01 10:40:02 +02:00 |
|
|
|
304a8e9c43
|
refactor: clean tools
|
2026-04-30 22:18:07 +02:00 |
|
|
|
86a53ee9f7
|
fix: robot direction to target
|
2026-04-30 21:28:29 +02:00 |
|
|
|
be78fb15bd
|
feat: allow model metadata cli override
|
2026-04-30 20:19:42 +02:00 |
|
Robin Meersman
|
1530e7d210
|
good luck cedric 🫡
|
2026-04-30 19:32:20 +02:00 |
|
Robin Meersman
|
f8eb43004f
|
other: backup, finding bug in conversion from dict to array
|
2026-04-30 15:58:46 +02:00 |
|
Robin Meersman
|
e85fc09ed6
|
feat(PPOTrainer): vectorized init agent state, partially added message passer param to rollout, ...
|
2026-04-29 11:23:01 +02:00 |
|
Cedric
|
bc3408b59a
|
feat: init agents array
|
2026-04-29 08:14:13 +00:00 |
|
Robin Meersman
|
db5d4c3dc2
|
fix(PPOTrainer): fixed duplicate definition of conversion function, tracing error related to obs_mode, added adjacent matrix argument
|
2026-04-28 22:23:46 +02:00 |
|
|
|
ba8184b034
|
test: fix outdated imports etc.
|
2026-04-28 18:03:53 +02:00 |
|
|
|
2a34b84f83
|
chore: min/max of inputs
|
2026-04-28 17:38:02 +02:00 |
|
|
|
49f5874035
|
chore: remove distance to target from model input
|
2026-04-28 17:27:08 +02:00 |
|
|
|
33fbc4c96f
|
feat: tool to construct XML
|
2026-04-28 16:42:46 +02:00 |
|
|
|
5716b73eb9
|
docs: restructure API docs
|
2026-04-28 16:25:05 +02:00 |
|
|
|
a1b4df0921
|
test(simulate): evaluation
|
2026-04-28 14:20:27 +02:00 |
|
|
|
aabf1bc186
|
feat(simulate): configurable camera id
|
2026-04-28 14:00:12 +02:00 |
|
|
|
d6146850fb
|
feat(simulate): capture video
|
2026-04-28 13:55:36 +02:00 |
|
|
|
9d1e2c9bff
|
refactor: evaluation subpackage
|
2026-04-28 13:39:01 +02:00 |
|
|
|
8f2a5d25ed
|
feat(simulate): metadata config loading
|
2026-04-28 13:17:24 +02:00 |
|
|
|
a3a1f3643f
|
chore: observation pipeline in simulate
|
2026-04-28 11:40:28 +02:00 |
|
|
|
036edf8aab
|
chore: observation pipeline in training
|
2026-04-28 11:15:16 +02:00 |
|
|
|
0032496073
|
docs: describe normalization
|
2026-04-28 10:44:55 +02:00 |
|
|
|
a1db544e76
|
feat: extract observations pipeline
|
2026-04-28 10:43:48 +02:00 |
|
Cedric
|
a005c0ccad
|
feat: upgraded obs_to_array, now takes care of splitting for agents
|
2026-04-28 08:20:00 +00:00 |
|
Cedric
|
39945e8821
|
feat: helper to visualize adjacency in a file
|
2026-04-28 07:27:08 +00:00 |
|
Cedric
|
2777d4f6da
|
feat: adjacency builder
|
2026-04-28 06:46:26 +00:00 |
|
Cedric
|
8830289992
|
feat: made obs_dict handle morphology
|
2026-04-28 05:44:26 +00:00 |
|
|
|
b29818a144
|
docs: Clarify input space in more detail
|
2026-04-25 17:05:46 +02:00 |
|
Jona Reynaert
|
509ad491fb
|
Merge pull request #21 from SELab-3-2026/simulate-results
Model Simulation
|
2026-04-23 10:09:42 +02:00 |
|
Jona Reynaert
|
94e50f3d83
|
fix: used jax type to remove warning
|
2026-04-22 18:51:08 +02:00 |
|
Jona Reynaert
|
395b04d9a8
|
feat: adapted simulate to trained config
|
2026-04-22 18:47:35 +02:00 |
|
Jona Reynaert
|
c4447976ab
|
Merge branch 'dev' into simulate-results
|
2026-04-19 16:54:47 +02:00 |
|
RobinMeersman
|
ea59c821a3
|
feat: improved the used reward function to better the training results
|
2026-04-18 15:52:33 +02:00 |
|
