validphys.tests package

Subpackages

Submodules

validphys.tests.conftest module

validphys.tests.test_alpha_s_bundle_pdf module

validphys.tests.test_arclengths module

validphys.tests.test_calcutils module

validphys.tests.test_closuretest module

test_closuretest.py

contains some unit tests for closure test estimators

class validphys.tests.test_closuretest.TestResult(central_value, rawdata=None)[source]

Bases: object

class for testing base level estimators which expect a results object

validphys.tests.test_closuretest.test_bias_function()[source]

validphys.tests.test_closuretest.test_variance_function()[source]

validphys.tests.test_commondataparser module

validphys.tests.test_core module

validphys.tests.test_covmatreg module

validphys.tests.test_covmats module

validphys.tests.test_cuts module

validphys.tests.test_datafiles module

validphys.tests.test_effexponents module

validphys.tests.test_filter_rules module

validphys.tests.test_fitdata module

validphys.tests.test_fitveto module

validphys.tests.test_loader module

validphys.tests.test_mc2hessian module

validphys.tests.test_metaexps module

test_metaexps

Test that the experiments key defined in the commondata meta data, which is subsequently used for grouping makes sense.

validphys.tests.test_metaexps.test_no_systematic_overlaps()[source]

Take every available dataset and check that there are no overlapping systematics when the grouping is by metadata experiments.

This is important because we make the assumption that the total covariance matrix is block diagonal in metadata experiment, and it is therefore used as an optimisation in several places.

validphys.tests.test_multiclosure module

validphys.tests.test_overfit_metric module

validphys.tests.test_plots module

validphys.tests.test_postfit module

validphys.tests.test_pseudodata module

validphys.tests.test_pyfkdata module

validphys.tests.test_pythonmakereplica module

validphys.tests.test_regressions module

validphys.tests.test_results module

Tests for functions in the validphys.results file.

validphys.tests.test_results.test_groups_central_values_no_table(data_internal_cuts_config)[source]: Check if the output of groups_central_values_no_table agrees with the replica 0 value calculated by calling group_result_table_no_table. group_result_table_no_table also computes the predictions for all other replicas.

validphys.tests.test_sumrules module

validphys.tests.test_tableloader module

validphys.tests.test_tableloader.test_extrasum_slice()[source]

validphys.tests.test_tableloader.test_min_combination()[source]

validphys.tests.test_theorydbutils module

validphys.tests.test_totalchi2 module

test_totalchi2.py

test that the action which calculates the total chi2 produces sensible results for both MC and hessian pdfs

validphys.tests.test_totalchi2.test_abs_chi2_data(single_data_internal_cuts_config)[source]: Test abs_chi2_data with a normal dataset

validphys.tests.test_totalchi2.test_abs_chi2_data_singlepoint(single_data_single_point_internal_cuts_config)[source]: Test abs_chi2_data with the corner case of a single datapoint dataset

validphys.tests.test_totalchi2.test_hessian_total_chi2(hessian_data_internal_cuts_config)[source]

testing total chi2 for hessian pdf

In particular check that the sum across experiments is handled correctly

and that calculating the total chi2 from the flat list of datasets gives the same answer as using total_chi2_data

validphys.tests.test_totalchi2.test_mc_total_chi2(data_internal_cuts_config)[source]

Testing total chi2 for mc pdf