PMACE: Projected Multi-Agent Consensus Equilibrium

The documentation for this package is available here:

https://ptycho-pmace.readthedocs.io/

Overview

PMACE is a framework for distributed reconstruction such as ptychographic image reconstruction.

This python package implements PMACE approach in python.

Conceptual rendering of the PMACE pipeline.

Installation

1. Download the source code:

   Move to a directory of your choice and run the following commands.

   git clone https://github.com/cabouman/ptycho_pmace.git
   cd ptycho_pmace

2. Installation:

   It is recommended that you install the package to a virtual environment. Install Anaconda and follow any of the two methods.

• 2.1. Easy installation: If you have Anaconda installed, run the following commands.

  cd dev_scripts
  source ./clean_install_all.sh
  cd ..

• 2.2. Manual installation: Note the pmace environment needs to be activated every time you use the package.

  • 2.2.1 Create a Virtual Environment:

    conda create --name pmace python=3.8
    conda activate pmace

  • 2.2.2 Install the dependencies:

    pip install -r requirements.txt

  • 2.2.3 Install the package:

    pip install .

  • 2.2.4 Install the documentation:

    cd docs/
    pip install -r requirements.txt
    make clean html
    cd ..

3. Validate installation:

   Once the installation is done, you can validate the installation by running the demo script.

   cd demo/
   pyhon demo_pmace.py

Related publications

For more information on the methods implemented in this code, please see:

paper link Zhai, Qiuchen, Brendt Wohlberg, Gregery T. Buzzard, and Charles A. Bouman. "Projected multi-agent consensus equilibrium for ptychographic image reconstruction." In 2021 55th Asilomar Conference on Signals, Systems, and Computers, pp. 1694-1698. IEEE, 2021, doi: 10.1109/IEEECONF53345.2021.9723357.

paper link Zhai, Qiuchen, Gregery T. Buzzard, Kevin Mertes, Brendt Wohlberg, and Charles A. Bouman. "Projected Multi-Agent Consensus Equilibrium (PMACE) for Distributed Reconstruction with Application to Ptychography." arXiv preprint arXiv:2303.15679 (2023).