mesh2pcf

mesh2pcf is a Python wrapper for a high-performance C++ library that computes the two-point correlation function (2PCF) multipoles from gridded density fields. It supports both isotropic and anisotropic measurements and uses OpenMP for fast execution.

---

Features

• Computes monopole (ℓ=0), quadrupole (ℓ=2), and hexadecapole (ℓ=4) of the 2PCF
• Supports regular 3D gridded input (e.g., from density fields)
• Anisotropic correlation function with customizable μ-binning
• Multi-threaded execution using OpenMP
• C++ core with Python interface

---

Installation

Make sure you have the required dependencies:

• Python 3.6+
• NumPy
• GSL and OpenMP libraries (required by the C++ code)
• SWIG (for building from source)
• A C++ compiler supporting OpenMP

---

Installing from source

1. Clone the repository:

   git clone https://github.com/EdoardoMaragliano/Mesh2PCF.git
   cd mesh2pcf
   python setup.py build_ext --inplace
   python setup.py install

Usage

See the notebooks and scripts in the examples folder for basic usage and demonstrations of the package.

See the following for a quick example.

import os

os.environ['OMP_NUM_THREADS'] = '24'

import numpy as np
from mesh2pcf.Mesh2pcfCpp.Mesh2pcfCpp import Mesh2PCF, Mesh2PCFAniso, DoubleVector, DoubleVectorVector, DoubleGrid

mesh = np.load('real_space_field.npy')
mesh = np.array(mesh, dtype=np.float64)
c_mesh = DoubleGrid(mesh)

test_2pcf_r_mu = Mesh2PCFAniso(c_mesh, 1000)
test_2pcf_r_mu.compute_mesh2PCF(
    max_distance = 200.0,
    bin_size = 5.0,
    num_mu_bins = 200,
)
r_array = test_2pcf_r_mu.get_r_array()
xi0_r_mu_128 = test_2pcf_r_mu.get_monopole()
xi2_r_mu_128 = test_2pcf_r_mu.get_quadrupole()
xi4_r_mu_128 = test_2pcf_r_mu.get_hexadecapole()

output = np.array([r_array, xi0_r_mu_128, xi2_r_mu_128, xi4_r_mu_128])

np.save('xi_r_mu_128.npy', output)

Acknowledgments

I thank A. Veropalumbo for the OMP parallelization.