/DDPG_pytorch

A simple implement of DDPG using PyTorch.

Primary LanguagePython

DDPG_pytorch

This is a simple implement of Deep Deterministic Policy Gradient（DDPG） using PyTorch.

To do list

  • Seperating the training process from the testing process.
  • Testing in other environments.

How to run

  • Clone repository :
$ git clone https://github.com/xiaopeng-whu/DDPG_pytorch.git 
$ cd DDPG_pytorch

  • Training : results of one environment and its training curves:

    • Pendulum-v0 $ python main.py -x 1

References

DDPG-implementation