# PYME.localization.Test.test_localization_algorithms module¶

This file contains unit tests for the localization algorithms. The goal of these tests is to pick up on backwards incompatible changes (regressions) which will break things majorly, rather than small changes in localization precision. For this reason, the tests only look at the IQR of the localization error (i.e. ignore tails) and set fairly permissive bounds. Because we are using random data for the testing however, and the errors follow a distribution there is a finite chance that we will get the occasional test failure. If the test passes on a re-run, this is not a huge cause for concern.

PYME.localization.Test.test_localization_algorithms.test_AstigGaussFitFR()

Test the Astigmatic Gaussian fit by fitting some randomly generated events. The pass condition here is fairly loose, but should be sufficient to detect when the code has been broken

PYME.localization.Test.test_localization_algorithms.test_InterpFitR_PRI()

Test the 3D interpolated fit by fitting some randomly generated events. The pass condition here is fairly loose, but should be sufficient to detect when the code has been broken

PYME.localization.Test.test_localization_algorithms.test_InterpFitR_astigmatism()

Test the 3D interpolated fit by fitting some randomly generated events. The pass condition here is fairly loose, but should be sufficient to detect when the code has been broken

PYME.localization.Test.test_localization_algorithms.test_LatGaussFitFR()
PYME.localization.Test.test_localization_algorithms.test_SplitterFitInterpNR_astigmatism()

Test the 3D interpolated fit by fitting some randomly generated events. The pass condition here is fairly loose, but should be sufficient to detect when the code has been broken

PYME.localization.Test.test_localization_algorithms.test_numpy_records_view_bug()

Some versions of numpy have problems with the .view on complex dtypes. Check for this. (Known to affect SplitterFitFNR)