Skip to content

Latest commit

 

History

History
7 lines (5 loc) · 227 Bytes

README.md

File metadata and controls

7 lines (5 loc) · 227 Bytes

Deterministic-Policy-Gradient

Reinforcement Learning methods for Continuos Control

You will need gym and tensorflow installed on your pc to run the code.

To train the net, launch the file and press 0. For testing, press 1.