Skip to main content

A Python interface to AZURE2 that makes it easy to sample R-matrix parameters.

Project description

BRICK

Bayesian R-Matrix Inference Code Kit

BRICK is a Python package that serves as an interface to AZURE2 and readily permits the sampling of R-matrix parameters.

It accompanies AZURE2. The primary goal is to allow the user to deploy Markov Chain Monte Carlo (MCMC) to sample parameters that are typically optimized in a χ2-minimization analysis.

Requirements

AZURE2 must be installed and available at the command line via AZURE2. Currently, command-line execution is not available on Windows or macOS.

NumPy and Matplotlib must be available in order to run the test script in test directory.

emcee is the MCMC sampler that is used in the test scripts. BRICK is intentionally designed such that other samplers can be used with little effort.

Overview

The classes defined in this module are:

  1. AZR
  2. Parameter
  3. Level
  4. Output
  5. Segment
  6. Data

AZR

Handles communication with AZURE2 and its output.

Parameter

Defines a sampled or "free" parameter.

Level

Defines an R-matrix level (a line in the <levels> section of the .azr file).

Output

Data structure for accessing output data. (I got tired of consulting the extremely well-documented manual for the output file format.)

Segment

Data structure to organize the information contained in line of the <segmentsData> section of the .azr file.

Data

Data structure that holds a list of Segments and provides some convenient functions for applying actions to all of them.

Example

In the test directory there is a Python script (test.py) that predicts the 12C(p,gamma) cross section and compares it to the Vogl data.

Note that the script uses NumPy and Matplotlib.

Installation

The simplest way to install brick is to use pip via:

pip install brick-james

Alternatively, one may clone the repository on GitHub.

Tests

BRICK includes a suite of basic funcationality tests in the tests/12Cpg directory. The tests perform calculations of the 12C(p,gamma) reaction with two data sets. To run them, within the tests/12Cpg directory, use

python -m unittests -v tests.py

Use

Once installed, brick allows the user to access the relevant classes and functions by:

import brick

azr_object = brick.azr.AZR('input.azr')

More instructive test are forthcoming.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

brick-james-0.0.6.tar.gz (13.4 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page