openai/train-procgen
Code for the paper "Leveraging Procedural Generation to Benchmark Reinforcement Learning"
PythonMIT
Issues
- 0
🦋
#22 opened by Blue108108 - 1
Cannot train with GPU
#18 opened by HadiSDev - 10
cannot reproduce results on Coinrun.
#6 opened by sjtuytc - 4
- 3
strange result from the maze environment
#14 opened by JINKEHE - 1
Reproducing + validating via graph
#16 opened by max-reuter - 2
Training on custom environments + passing extra command line args for training
#15 opened by max-reuter - 2
- 9
Achieving reported training performance
#11 opened by hfeniser - 4
- 2
Plotting the episode reward
#12 opened by dotchen - 1
- 2
Specifications for training/testing in easy mode
#9 opened by mklissa - 2
- 2
- 1
Dependency on baseline
#3 opened by kaixin96 - 1
no frame stacking with procgen environments
#2 opened by 1jsingh - 1
mpich
#1 opened by christopher-hesse