kengz / SLM-Lab

Modular Deep Reinforcement Learning framework in PyTorch. Companion library of the book "Foundations of Deep Reinforcement Learning".
https://slm-lab.gitbook.io/slm-lab/
MIT License
1.25k stars 264 forks source link

Revert "use vec eval" #386

Closed kengz closed 5 years ago

kengz commented 5 years ago

This reverts commit b1c7473eac274310a1a50022be0cfc941af62afa.

eval reward did not compute properly