Pymela - Python Matrix Element Analysis

A Python Application for analyzing correlation functions data to extract matrix elements from Lattice QCD measurements.

The package computes ratios of three- and two-point functions and performs fits to extract matrix elements that lead to Parton Distribution Functions (PDFs). Currently, plateau/constant fits on the plain ratio and linear fits on the summed ratio are supported. From the extracted matrix elements, the reduced Ioffe-time distributions are computed for the types of fits supported and they are stored in HDF5 files.

Contents

• Pymela: Contains modules and class definitions related to the operations supported by the application. Submodules:

  • io: Parse and check JSON input files; file conventions.
  • fit: Constant and linear fits.
  • tools: Tools and utilities, including a module for Jackknife sampling.

• Tests: Tests that parse an input JSON file and perform various operations. Currently supported tests and operations are:

  • tests/read_2pt_corr.py: Read two-point correlation functions in ASCII format and write the data in HDF5 format.
  • tests/read_3pt_corr.py: Read three-point correlation functions in ASCII format and write the data in HDF5 format.
  • tests/effective_energy.py: Read two-point correlation functions in ASCII format, compute Effective Energy and perform constant fits on the Effective Energy; write the data in HDF5 format.
  • tests/compute_ratio.py: Read two- and three-point correlation functions in ASCII format, compute ratios of three- and two-point functions and store the data in HDF5 format.
  • tests/fit_ratio.py: Read two- and three-point correlation functions in ASCII format, compute ratios of three- and two-point functions, perform fits on the ratio to extract matrix elements and store the data in HDF5 format.
  • tests/compute_rITD.py: Most comprehensive test. Read two- and three-point correlation functions in ASCII format, compute ratios of three- and two-point functions, perform fits on the ratio to extract matrix elements and compute reduced Ioffe-time distributions (rITD) from the matrix elements. Store all the data in HDF5 format.

Dependencies

The following packages are required:

• json
• numpy
• h5py
• scipy (scipy.optimize)

Author and Contact

• Christos Kallidonis - Jefferson Lab
• Web: https://christoskallidonis.com.
• Copyright (C) 2020. All rights reserved.