Skip to content

Fix trajectory ending handling by adding end reason to agent observations #392

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ondrej-lukas merged 1 commit into from
Jan 6, 2026
Merged

Fix trajectory ending handling by adding end reason to agent observations #392

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions AIDojoCoordinator/coordinator.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -637,12 +637,13 @@ async def _process_game_action(self, agent_addr: tuple, action:Action)->None:
await self._episode_rewards_condition.wait()
# append step to the trajectory if needed

async with self._agents_lock:
self._add_step_to_trajectory(agent_addr, action, self._agent_rewards[agent_addr], new_state,end_reason=None)
# add information to 'info' field if needed

info = {}
if self._agent_status[agent_addr] not in [AgentStatus.Playing, AgentStatus.PlayingWithTimeout]:
info["end_reason"] = str(self._agent_status[agent_addr])
async with self._agents_lock:
self._add_step_to_trajectory(agent_addr, action, self._agent_rewards[agent_addr], new_state,end_reason=info.get("end_reason", ""))
# add information to 'info' field if needed
new_observation = Observation(self._agent_states[agent_addr], self._agent_rewards[agent_addr], self._episode_ends[agent_addr], info=info)
self._agent_observations[agent_addr] = new_observation
output_message_dict = {
Expand Down