issues
search
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
Dueling DQN
#412
Closed
lgraesser
closed
5 years ago
lgraesser
commented
5 years ago
Feature / Fix
Fix ModuleList syntax in DuelingConvNet
Add Dueling DQN specs
Fix reward accumulation bug if env not done before reset (the bug was introduced in #407 so it didn't impact any of the benchmark or older runs)
Feature / Fix