Train an agent with PARL to solve the CartPole problem, a classical benchmark in RL.
- parl>=1.4
- paddlepaddle>=2.0.0
- gym
python train.py
The agent can get around 200 points in a few seconds.
QuickStart
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
parent directory.. | ||||