search

instadeepai / Mava

🦁 A research-friendly codebase for fast experimentation of multi-agent reinforcement learning in JAX
Apache License 2.0
737 stars 90 forks source link

fix: bug fixes + add missing kwargs #995

Closed WiemKhlifi closed 9 months ago

WiemKhlifi commented 10 months ago

What?

Set the appropriate type for episode_return in the Checkpointer class, fix the ExperimentOutput type in the quickstart notebook, and add missing kwargs in the jumanji_make_env function.

Why?

The changes made have resolved the bugs identified.

WiemKhlifi commented 9 months ago

All checks are done 🙌