Installation Instructions¶
The dex-net package can be installed as a standalone Python module or a ROS package. We suggest installing as a standalone Python module unless you want to use dex-net to attempt to plan grasps for a physical robot.
Installation has been tested on Ubuntu 12.04 LTS, Ubuntu 14.04 LTS, and Ubuntu 16.04 LTS. We do not recommend using dex-net with anaconda, as there are known issues with the installation of the visualization tools in conda envrionments.
1. Clone the repository¶
Change directories to the desired install location of dex-net. If you are installing as a ROS package, this should be /path/to/your/catkin_ws
Clone or download the project from Github.
$ git clone https://github.com/BerkeleyAutomation/dex-net.git
2. Run the installation script¶
Run the dex-net installation helper script:
$ sudo sh install.sh {cpu|gpu} {python|ros}
The brackets indicate optional arguments to switch installation methods.
The first argument specifies the version:
- cpu: no TensorFlow GPU support
- gpu: TensorFlow GPU support for GQ-CNN training
The second argument specifies the installation mode:
- python: Python-only installation. No ROS services will be installed for dex-net or any of the Berkeley AUTOLAB modules.
- ros: Installation as a ROS package. Enables ROS services for the Berkeley AUTOLAB modules and supports dex-net ROS nodes and services that may be developed in the future.
3. Test the installation¶
To test your installation, run
$ python setup.py test
We highly recommend testing before using the module.
4. Try it out!¶
Go the dex-net Command Line Interface (CLI) example to see the basic functionality.
Issues¶
If you are having issues with the installation script, then you should try to install manually by following our extended installation instructions, which include individual commands, optional dependencies, and workarounds for some known issues.
Please raise installation issues on the Github Issues.
Documentation¶
Building¶
The API documentation is available on the dex-net website.
You can build dex-net’s documentation from scratch with a few extra dependencies – specifically, sphinx and a few plugins. This is important for developers only.
Go to the docs directory and run make with the appropriate target.
For example,
$ cd docs/
$ make html
will generate a set of web pages. Any documentation files generated in this manner can be found in docs/build.
Deploying Documentation¶
To deploy documentation to the Github Pages site for the repository, simply push any changes to the documentation source to master and then run
$ . gh_deploy.sh
from the docs folder. This script will automatically checkout the
gh-pages branch, build the documentation from source, and push it
to Github.