This is the accompanying code repository for the paper "Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets" (arXiv paper). Here is the project website where you can find more information about how to use and contribute to the dataset.
In your .bashrc set the environment variables EXP for experiment data and DATA for trainingdata:
export EXP=<folder to store experiment results>
export DATA=<folder where the bridge dataset is stored>
Setup conda environment by running
conda create --name bridgedata python=3.6.8 pip
pip install -r requirements.txt
then in this directory run
pip install -e .
Clone the bridge data robot infrastructure repository, install the dependencies, and run
pip install -e widowx_envs
python bridgedata/train.py --path bridgedata_experiments/bc_fromscratch/conf.py
The example config file trains the "wipe plate with sponge task". You can change the training task and the training parameters in bridgedata_experiments/bc_fromscratch/conf.py.
python bridgedata/train.py --path bridgedata_experiments/task_id_conditioning/conf.py
The example config file trains a multi-task, task-id conditioned imitation learning policy on all of the tasks in toykitchen1.
Another example config file bridgedata_experiments/task_id_conditioning/conf_exclude_toykitchen1.py trains a multi-task policy on all of the environments except toykitchen1 (to evaluation transferability of policies).
python bridgedata/train.py --path bridgedata_experiments/random_mixing_task_id/conf.py
The example config file trains a multi-task, task-id conditioned imitation learning policy on all of the environments except real kitchen 1, and the wipe plate with sponge task. The dataset is re-balanced such that the wipe plate with sponge task takes up 10% of the training dataset.
Another example config file bridgedata_experiments/random_mixing_task_id/conf_toykitchen1.py rebalances the dataset such that trajectories in toy kitchen 1 takes up 30% of the training dataset.
This repository also provides an example script docker/azure/doodad_launch.py for launching jobs on cloud compute services like AWS, GCP or Azure with Doodad.