Skip to content

Latest commit

 

History

History
22 lines (14 loc) · 820 Bytes

README.md

File metadata and controls

22 lines (14 loc) · 820 Bytes

PlanGAN

Code for our NeurIPS 2020 paper "PlanGAN"

Requirements:

  • Mujoco-py
  • PyTorch
  • NumPy
  • SkLearn
  • Joblib

To train an agent on FetchPickAndPlace run:

python train.py --env="fetch_push" --expt_name="FetchPush"

Pre-trained agents for FetchPush and FetchPickAndPlace are included in this repo. They can be evaluated with:

python evaluate.py --expt_name="FetchPickAndPlace" --num_trajectories=50

We also include a Jupyter notebook that allows you to visualise the imagined trajectories that the GANs generate (visualise_trajectories.ipynb).

Note that the exact hyperparameters used for the reported FetchPush and FetchPickAndPlace experiments are saved in JSON files located in the experiments folder.