tdpeuter/2026SEL3-project-Brittle_Star_Controller_Modularity
tdpeuter/2026SEL3-project-Brittle_Star_Controller_Modularity
Archived
1
Fork 0
Code Issues Releases Activity

feat(message passing): fixed tracing bug, todo: dense layer input dimension incorrect

Browse source
This commit is contained in:
Robin Meersman 2026-05-02 23:27:31 +02:00
parent 027b13b30f
commit 3cb3c765cc
1 changed files with 5 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

8
src/brittle_star_project/trainers/PPOTrainer.py
View file

@ -134,8 +134,8 @@ def _normalize_obs(obs, mean, var, eps=1e-8):
def _convert_obs_dict_to_array_morphology(obs_dict, morph_mode, segments_per_arm: jnp.ndarray):
num_segments = segments_per_arm.sum()
num_arms = jnp.where(segments_per_arm > 0, 1, 0).sum()
num_segments = int(segments_per_arm.sum())
num_arms = int(jnp.where(segments_per_arm > 0, 1, 0).sum())
@logged_jit
def _filter_and_flatten(o) -> jnp.ndarray:
@ -668,9 +668,11 @@ class PPOTrainer:
message_passer_params = {}
if self.morph_mode != MorphMode.CENTRALIZED:
assert self.message_passer is not None, "MessagePasser is None"
message_passer_params = jax.vmap(
lambda k: self.message_passer.init(
k, self.sensor.apply(single_sensor_param, sample_obs)
k, self.sensor.apply(single_sensor_param, sample_obs), self.adj
)
)(message_passer_keys)
self.logger.info(