instadeepai / Mava

🦁 A research-friendly codebase for fast experimentation of multi-agent reinforcement learning in JAX
Apache License 2.0
709 stars 83 forks source link

feat: system folders #1020

Closed sash-a closed 7 months ago

sash-a commented 7 months ago

What?

Seeing as we're soon to have many more systems and each system is likely to have a ff and rec and ma and i version, I think moving the systems into folders makes sense. We can then have a types.py inside those folders so that our main types.py doesn't get too bloated (since each system will have a different params type, different learner state etc.)