Berkeley Photonics Generator

The Berkeley Photonics Generator (BPG) is a Python framework that enables you to generate and simulate silicon photonics layout. BPG leverages the BAG 2.0 framework, and has similar syntax and file structure. If you would like to cite this software in a publication, please use the following link:

Table of Contents

• Documentation
• Installation
• Usage
• Contributing
• Credits

Documentation

The most up to date BPG documentation can be found here

Installation

WARNING: Installation instructions are currently in flux.

We highly recommend you use an Anaconda environment with a Python version greater than 3.6. BPG generally will not function with Python versions less than 3.6, and requires packages with C/C++ dependencies that are most easily installed using Anaconda.

Once Anaconda is set up, please run the following commands to install packages with C/C++ dependencies:

conda install numpy
conda install rtree
conda install shapely

Then clone and install BAG with in any folder with:

git clone git@github.com:ucb-art/BAG_Framework.git
cd BAG_Framework
pip install .

Finally clone and install BPG in any folder with:

git clone git@github.com:BerkeleyPhotonicsGenerator/BPG.git
cd BPG
pip install .

BPG generally generates output layouts in the GDSII format. To view these layouts, we recommend you install and use the free open-source software package, Klayout.

Usage

1) In order to setup a brand new workspace, navigate to a clean folder and run bpg setup_workspace. This will copy in a directory with sample technology information, and a new file called sourceme.sh 2) Next run source sourceme.sh to set up all of the environment variables needed. This soruce file assumes that you are using a bash/sh/zsh shell. 3) At this point any layout generator can be run by executing python <PATH TO PYTHONFILE>

For more information check out the getting started guide

Contributing

We'd love your help in building and improving BPG! Please create an issue and contact Pavan at pvnbhargava@berkeley.edu before sending your first pull request. Please make sure to run the test suite (run_tests.py) in the Photonics_Dev repo prior to making your pull request.

Credits

Thanks to Sidney Buchbinder, Ruocheng Wang, and Brian Sun for building many of the core features in BPG!