README.md

Reproduce PPO with PARL

Based on PARL, the PPO algorithm of deep reinforcement learning has been reproduced, reaching the same level of indicators as the paper in mujoco benchmarks.

Paper: PPO in Proximal Policy Optimization Algorithms

Mujoco games introduction

Please see here to know more about Mujoco games.

Benchmark result

How to use

Dependencies:

• paddle>=2.0.0
• parl>=2.0.2
• gym==0.9.2
• mujoco-py==0.5.7

Start Training:

# To train an agent for Hopper-v1 game
python train.py

# For more customized arguments
# python train.py --help