This is the official repo for $\Sigma$-Agent [CoRL 2024]

Installation

1. Clone this repo

git clone https://github.com/TeleeMa/Sigma-Agent.git

2. Install conda environment and simulator

Following the RVT to install conda environment and CoppeliaSim.

3. Install RVT, PyRep, RLBench, YARR and PerAct Colab

cd Sigma-Agent
pip install -e .
pip install -e rvt/libs/PyRep 
pip install -e rvt/libs/RLBench 
pip install -e rvt/libs/YARR 
pip install -e rvt/libs/peract_colab

Training

python train.py --exp_cfg_path configs/sigma_agent.yaml --device [GPU NUM]

Evaluation

xvfb-run -a python eval.py --model-folder [PATH TO CKPT FOLDER]/sigma_agent --eval-datafolder [PATH TO RLBENCH/TEST] --tasks all --eval-episodes 25 --device 0 --headless --model-name [MODEL NAME]

Bibtex

If you find this useful, please cite the paper!

@article{ma2024contrastive,
 title={Contrastive Imitation Learning for Language-guided Multi-Task Robotic Manipulation},
 author={Ma, Teli and Zhou, Jiaming and Wang, Zifan and Qiu, Ronghe and Liang, Junwei},
 journal={arXiv preprint arXiv:2406.09738},
 year={2024}
}