/WGAIL

[ICML 2021] Learning to Weight Imperfect Demonstrations

Primary LanguagePythonMIT LicenseMIT

Learning to Weight Imperfect Demonstrations

This repository contains the PyTorch code for the paper "Learning to Weight Imperfect Demonstrations" in ICML 2021. Code for Atari experiments can be found in this repo.

Requirement

  • Python 3.7
  • torch 1.3.1
  • gym 0.15.7
  • mujoco 2.0.2.4
  • numpy 1.16.2

Execute

  • WGAIL
python wgail.py --env Ant-v2 --num-epochs 5000 --traj-size 1000 --stage 2
  • 2IWIL
python 2iwil.py --env Ant-v2 --num-epochs 5000 --traj-size 1000 --stage 2
  • GAIL
python gail.py --env Ant-v2 --num-epochs 5000 --traj-size 1000 --stage 2
  • T-REX
python trpo_irl.py --env Ant-v2 --num-epochs 5000 --reward-path 'reward_model/ant_reward_stage2.pth' --stage 2

The re-implementation of T-REX/D-REX can be found in SAIL.

Acknowledegement

We would like to thank the authors of 2IWIL/IC-GAIL. Our code structure is largely based on their source code.