Skip to content

Latest commit

 

History

History
20 lines (17 loc) · 434 Bytes

README.md

File metadata and controls

20 lines (17 loc) · 434 Bytes

gym-kuka-mujoco

An OpenAI gym environment for the Kuka arm.

Installation

Just the Environment

pip install -e .

The Environment and Baselines Code for Learning

pip install tensorflow
pip install -r requirements.txt

Example

Train on the KukaEnv with the baselines PPO implementation using

python examples/run_baselines.py --alg=ppo2 --env=KukaMujoco-v0 --network=mlp --num_timesteps=2e7 --play