This is a pytorch implementation of Hindsight Experience Replay.

• Openai Baselines

• python 3.5.2
• openai-gym
• mujoco-1.50.1.56
• pytorch-1.0.0
• mpi4py

• support GPU acceleration - although I have added GPU support, but I still not recommend if you don't have a powerful machine.
• add multi-env per MPI.
• add the plot and demo of the FetchSlide-v1.

If you want to use GPU, just add the flag --cuda (Not Recommended, Better Use CPU).

1. train the FetchReach-v1:

mpirun -np 1 python -u train.py --env-name='FetchReach-v1' --n-cycles=10 2>&1 | tee reach.log

2. train the FetchPush-v1:

mpirun -np 8 python -u train.py --env-name='FetchPush-v1' 2>&1 | tee push.log

3. train the FetchPickAndPlace-v1:

mpirun -np 16 python -u train.py --env-name='FetchPickAndPlace-v1' 2>&1 | tee pick.log

4. train the FetchSlide-v1:

mpirun -np 8 python -u train.py --env-name='FetchSlide-v1' --n-epochs=200 2>&1 | tee slide.log

python demo.py --env-name=<environment name>

Please download them from the Google Driver, then put the saved_models under the current folder.

Tips: when you watch the demo, you can press TAB to switch the camera in the mujoco.