Relax relative tolerance from 1e-8 to 1e-6

marcpaterno commented 6 days ago

Description

The CI testing is currently failing because of 3 tests which require a numerical answer to match with a tolerance of 1e-8. This PR will "fix" the problem by relaxing the tolerance to 1e-6.

Type of change

Please delete the bullet items below that do not apply to this pull request.

Bug fix (non-breaking change which fixes an issue)
Checklist:

The following checklist will make sure that you are following the code style and guidelines of the project as described in the contributing page.

[x] I have run bash pre-commit-check and fixed any issues
[x] I have added tests that prove my fix is effective or that my feature works
[x] I have made corresponding changes to the documentation
[x] I have 100% test coverage for my changes (please check this after the CI system has verified the coverage)

codecov[bot] commented 6 days ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 99.9%. Comparing base (e229d35) to head (4ff37c9).

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #472 +/- ## ====================================== Coverage 99.9% 99.9% ====================================== Files 54 54 Lines 4142 4142 ====================================== Hits 4139 4139 Misses 3 3 ```

marcpaterno commented 6 days ago

It was a change in CCL that caused the change in calculated results.

The purpose of these tests is to ensure no regressions in the theory calculations. Thus this is the wrong fix; we should update the expected values, and keep the tight tolerances.

LSSTDESC / firecrown