Skip to content

Update README with new file direction removing the old one#83

Open
akshatgarg06 wants to merge 1 commit intomasterfrom
akshatgarg06-patch-2
Open

Update README with new file direction removing the old one#83
akshatgarg06 wants to merge 1 commit intomasterfrom
akshatgarg06-patch-2

Conversation

@akshatgarg06
Copy link
Copy Markdown
Collaborator

  • Removed the conversion script is soccer_sar_to_rl_dataset.py direction.
  • SAR2RL expects SAR outputs (events.jsonl) already exist.

@github-actions
Copy link
Copy Markdown

github-actions bot commented Mar 5, 2026

Init Test Results 📝

  • Status: success
  • OS: Linux
  • Python: 3.10

@github-actions
Copy link
Copy Markdown

github-actions bot commented Mar 5, 2026

Init Test Results 📝

  • Status: success
  • OS: Linux
  • Python: 3.8

@github-actions
Copy link
Copy Markdown

github-actions bot commented Mar 5, 2026

Init Test Results 📝

  • Status: success
  • OS: Linux
  • Python: 3.9

@akshatgarg06
Copy link
Copy Markdown
Collaborator Author

PR #82 and #83 are the updated versions of the previous PRs (making sure the README file is up-to-date)

@akshatgarg06 akshatgarg06 requested a review from kenjiro-mk March 5, 2026 07:33
@akshatgarg06
Copy link
Copy Markdown
Collaborator Author

@kenjiro-mk would this file require any changes, as it depends upon PR #82?
Thank you!


- DataStadium
- Statsbomb with Skillcorner Tracking Data

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Please add `robcup2d'.

Comment on lines 29 to +41

## SAR-to-RL Dataset Conversion (DQN / QMIX)
This section describes a SAR-to-RL dataset conversion step that formats SAR outputs (`events.jsonl`) into tensors used by
DQN and QMIX training. This is a preprocessing/data-format step, not a training algorithm.
The conversion script is `soccer_sar_to_rl_dataset.py`.

This produces a single shared multi-agent dataset with:
- `observation`: `(B, T, N, O)` (N=10 attackers)
- `action`: `(B, T, N)` (discrete action ids; default vocab size 16 with `PAD=15`)
- `reward`, `done`, `mask`: `(B, T)`
- `onball_mask`: `(B, T, N)` (for masking unavailable actions)

### Run via `SAR_data(...).preprocess_data()`
You can run SAR2RL through the same entry point as SAR
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Please add the SAR-to-RL Dataset Conversion (DQN / QMIX) to open-starlab/Document. After that, please add information about RoboCup2D to the Examples section, following the format of the others. Thank you.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants