(*)Simplify interpolate_column weight calculations #227
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## dev/gfdl #227 +/- ##
============================================
+ Coverage 35.93% 37.23% +1.29%
============================================
Files 260 263 +3
Lines 72002 73060 +1058
Branches 13590 13605 +15
============================================
+ Hits 25872 27201 +1329
+ Misses 41072 40841 -231
+ Partials 5058 5018 -40
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
Set the interpolation weights inside of interpolate_column to explicitly be the complement of one another, thereby saving an extra division at each point and reducing the number of variables that need to be stored, in preparation for the creation of a separate subroutine to find interface positions. This commit is mathematically equivalent to what was there before, and the extensive unit testing of interpolate_column is still passing, but it changes the value of some interpolated interface diagnostics at the level of roundoff (but not the MOM6 solutions themselves, as they do not depend on interpolate_column yet).
Hallberg-NOAA
force-pushed
the
interpolate_column_reweight
branch
from
67d9625 to
c0af21e
Compare
adcroft
approved these changes
Nov 1, 2022
There was a problem hiding this comment.
As explained in the PR description, this changes the output of variables on interfaces but does not change the solution. Testing at https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines/17322 . Assuming this passes regressions, I approve this change.
|
The testing at https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines/17322 has passed the regressions. |
|
Gaea regression: https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines/17322 ✔️ |
This was referenced Apr 6, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Set the interpolation weights inside of interpolate_column to explicitly be the complement of one another, thereby saving an extra division at each point and reducing the number of variables that need to be stored, in preparation for the creation of a separate subroutine to find interface positions. This commit is mathematically equivalent to what was there before, and the extensive unit testing of interpolate_column is still passing, but it changes the value of some interpolated interface diagnostics at the level of roundoff (but not the MOM6 solutions themselves, as they do not depend on interpolate_column yet).
Some of the C.I. tests for regressions of diagnostics are expected to fail to reproduce the previous values, but they do agree to many (13 or more) decimal places, apart from near-zero spatial means of these diagnostics (where the mean is many orders of magnitude smaller than the values that are being averaged) that can differ at leading order. All of the differences are well within the range of what would be expected from changes at roundoff in the interpolated interface diagnostics themselves.