Test support (numpy.testing) — NumPy v2.2 Manual (original) (raw)

Common test support for all numpy test scripts.

This single module should provide all the common functionality for numpy tests in a single location, so that test scripts can just import it and work right away. For background, see the Testing guidelines

Asserts#

Asserts (not recommended)#

It is recommended to use one of assert_allclose,assert_array_almost_equal_nulp or assert_array_max_ulp instead of these functions for more consistent floating point comparisons.

Decorators#

Test running#

Testing custom array containers (numpy.testing.overrides)#

These functions can be useful when testing custom array container implementations which make use of __array_ufunc__/__array_function__.

Guidelines#

• Testing guidelines
  • Introduction
  • Testing NumPy
    * Running tests from inside Python
    * Running tests from the command line
    * Running doctests
    * Other methods of running tests
  • Writing your own tests
    * Using C code in tests
    * build_and_import_extension
    * Labeling tests
    * Easier setup and teardown functions / methods
    * Parametric tests
    * Doctests
    * tests/
    * __init__.py and setup.py
  • Tips & Tricks
    * Known failures & skipping tests
    * Tests on random data
    * Documentation for numpy.test
    * test