search

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

fix: initialize max_episode_return to negative infinity instead of zero for each algorithm #1024

Closed SimonDuToit closed 7 months ago

CLAassistant commented 7 months ago

CLA assistant check
All committers have signed the CLA.

SimonDuToit commented 7 months ago

I've removed the float conversions and added it to the advanced usage file.