The gqcnn Python package is for training and analysis of Grasp Quality Convolutional Neural Networks (GQ-CNNs). It is part of the ongoing Dexterity-Network (Dex-Net) project created and maintained by the AUTOLAB at UC Berkeley.

Please see the docs for installation and usage instructions.

If you use any part of this code in a publication, please cite the appropriate Dex-Net publication.

This repo has updated the codebase of GQCNN to newer versions and tested on Ubuntu20.04:

python 3.11

CUDA 12.2

tensorflow 2.15.0.post1

to set up the environment, follow the README in docker/updated_gpu

The link provided in the official website is obsolete, use the link below:

https://drive.google.com/file/d/1fbC0sGtVEUmAy7WPT_J-50IuIInMR9oO/view

or you can download it from huggingface using:

wget --content-disposition https://huggingface.co/WoodenHeart0214/gqcnn/resolve/main/model_zoo.zip?download=true && \
    unzip model_zoo.zip && \
    rm model_zoo.zip && \
    mv model_zoo models

. venv/bin/activate

make gqcnn_single_object_docker