https://github.com/tillahoffmann/pytest-bootstrap

https://github.com/tillahoffmann/pytest-bootstrap

bootstrap pytest scientific-computing

Last synced: 27 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/tillahoffmann/pytest-bootstrap
• Owner: tillahoffmann
• Created: 2022-01-27T17:57:02.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2022-03-04T21:06:24.000Z (over 2 years ago)
• Last Synced: 2024-04-25T10:43:40.834Z (7 months ago)
• Topics: bootstrap, pytest, scientific-computing
• Language: Python
• Homepage:
• Size: 33.2 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.rst

Awesome Lists containing this project

README

        
🧪 pytest-bootstrap

===================

.. image:: https://github.com/tillahoffmann/pytest-bootstrap/actions/workflows/main.yml/badge.svg

  :target: https://github.com/tillahoffmann/pytest-bootstrap/actions/workflows/main.yml

  :alt: Build status.

.. image:: https://readthedocs.org/projects/pytest-bootstrap/badge/?version=latest

  :target: https://pytest-bootstrap.readthedocs.io/en/latest/?badge=latest

  :alt: Documentation status.

.. image:: https://img.shields.io/pypi/v/pytest-bootstrap

  :target: https://pypi.org/project/pytest-bootstrap/

  :alt: PyPI package version.

Scientific software development often relies on stochasticity, e.g. for `Monte Carlo `_ integration or simulating the `Ising model `_. Testing non-deterministic code is difficult. This package offers a bootstrap test to validate stochastic algorithms, including multiple hypothesis correction for vector statistics. It can be installed by running :code:`pip install pytest-bootstrap`.

Example

-------

Suppose we want to implement the expected value of `log-normal distribution `_ with location parameter :math:`\mu` and scale parameter :math:`\sigma`.

.. doctest::

  >>> import numpy as np

  >>>

  >>> def lognormal_expectation(mu, sigma):

  ...   return np.exp(mu + sigma ** 2 / 2)

  >>>

  >>> def lognormal_expectation_wrong(mu, sigma):

  ...   return np.exp(mu + sigma ** 2)

We can validate our implementation by simulating from a lognormal distribution and comparing with the bootstrapped mean.

.. doctest::

  >>> from pytest_bootstrap import bootstrap_test

  >>>

  >>> mu = -1

  >>> sigma = 1

  >>> reference = lognormal_expectation(mu, sigma)

  >>> x = np.exp(np.random.normal(mu, sigma, 1000))

  >>> result = bootstrap_test(x, np.mean, reference)

This returns a summary of the test, such as the bootstrapped statistics.

.. doctest::

  >>> result.keys()

  dict_keys(['alpha', 'alpha_corrected', 'reference', 'lower', 'upper',

             'z_score', 'median', 'iqr', 'tol', 'statistics'])

Comparing with our incorrect implementation reveals the bug.

.. doctest::

  >>> reference_wrong = lognormal_expectation_wrong(mu, sigma)

  >>> result = bootstrap_test(x, np.mean, reference_wrong)

  Traceback (most recent call last):

    ...

  pytest_bootstrap.BootstrapTestError: the reference value 1.0 lies outside

    the 1 - (alpha = 0.01) interval ...

Visualising the bootstrapped distribution using :func:`pytest_bootstrap.result_hist` can help identify discrepancies between the bootstrapped statistics and the theoretical reference value. Note that you need to install matplotlib separately or install pytest-bootstrap using :code:`pip install pytest-bootstrap[plot]`.

.. plot:: examples/lognormal.py

  :caption: Histogram of bootstrapped means reveals the erroneous implementation of the log-normal mean.

A comprehensive set of examples can be found in `the tests `_.

Interface

---------

.. automodule:: pytest_bootstrap

  :members: