Skip to content
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.

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

Multiagent simplerl #5066

Merged
merged 15 commits into from
Mar 10, 2021
Merged

Conversation

andrewcoh
Copy link
Contributor

Proposed change(s)

Describe the changes made in this PR.

Useful links (Github issues, JIRA tickets, ML-Agents forum threads etc.)

Types of change(s)

  • Bug fix
  • New feature
  • Code refactor
  • Breaking change
  • Documentation update
  • Other (please describe)

Checklist

  • Added tests that prove my fix is effective or that my feature works
  • Updated the changelog (if applicable)
  • Updated the documentation (if applicable)
  • Updated the migration guide (if applicable)

Other comments

pass

def set_actions(self, behavior_name, action):
# im so sorry
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lolwut

@@ -314,6 +314,185 @@ def _make_batched_step(
return (decision_step, terminal_step)


class MultiAgentEnvironment(BaseEnv):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I understand nothing this class does. Please add comments

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, it is a pretty horrible thing


# tests in this file won't be tested on GPU machine
pytestmark = pytest.mark.check_environment_trains


@pytest.mark.parametrize("action_sizes", [(0, 1), (1, 0)])
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can this be tested for combinations of rank 1, 2 and 3 observations and with and LSTM config? (To make sure it does not crash at least)

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added some tests for LSTM, variable length obs, and visual

self.dones[name_and_num] = False
self.envs[name_and_num].reset()
# HACK
self.behavior_spec = self.envs[name_and_num].behavior_spec
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This dos not need to be in the loop

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Moved

@ervteng ervteng merged commit a7d2a65 into develop-coma2-trainer Mar 10, 2021
@delete-merged-branch delete-merged-branch bot deleted the develop-multiagent-simplerl branch March 10, 2021 20:57
@github-actions github-actions bot locked as resolved and limited conversation to collaborators Mar 11, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants