Home

Awesome

From Play to Policy: Conditional Behavior Generation from Uncurated Robot Data

[Paper] [Project Website] [Data]

Zichen Jeff Cui, Yibin Wang, Nur Muhammad (Mahi) Shafiullah, and Lerrel Pinto, New York University

This repo contains code for reproducing sim environment experiments, and the real-world robotic experiment gym environment, and data collection tools. Datasets for the simulated environments will be uploaded to this OSF link.

Getting started

The following assumes our current working directory is the root folder of this project repository; tested on Ubuntu 20.04 LTS (amd64).

Setting up the project environments

Getting the training datasets

Datasets used for training will be uploaded to this OSF link.

Reproducing experiments

The following assumes our current working directory is the root folder of this project repository.

To reproduce the experiment results, the overall steps are:

  1. Activate the conda environment with
    conda activate cbet
    
  2. Train with python3 train.py. A model snapshot will be saved to ./exp_local/...;
  3. In the corresponding environment config, set the load_dir to the absolute path of the snapshot directory above;
  4. Eval with python3 run_on_env.py.

See below for detailed steps for each environment.

CARLA

Franka kitchen

Block push

Speeding up evaluation