Fix mapping set inversion #555
Merged
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
When inverting a data frame, the reordering of columns post-inversion should not rely on the columns that existed in the original data frame (pre-inversion), because it is not guaranted that all columns in the original data frame will exist in the inverted data frame: if the original frame contained, e.g., a `subject_label` column but did not contain a `object_label` column, then the inverted frame will contain an `object_label` column but will _not_ contain a `subject_label` column. Instead, we use the dedicated sort_df_rows_columns function to perform the reordering. We sort the columns only, as there should be no need at this stage to sort the rows.
We add a new test file containing an example of an "asymmetric" or "imbalanced" set: that is, a set where not all "subject_*" columns have an equivalent "object_*" column and vice versa. The set is derived from the "basic.tsv" set by: (1) removing most of the set metadata (not needed for the test); (2) removing most of the rows, leaving just a handful (more than enough); (3) removing most of the columns. In the columns we keep, we specifically keep "subject_label" but NOT "object_label" (thereby creating the "asymmetry"), and conversely we keep "object_source" but NOT "subject_source" (another asymmetry). We then add a new test case in tests/test_utils.py in which we try inverting the asymmetric test (checking that it does not trigger the issue mapping-commons#554) and check that the inverted object labels correspond to the original subject labels.
matentzn
approved these changes
Oct 24, 2024
| self.assertEqual(len(inverted_df), len(msdf.df)) | ||
| original_subject_labels = msdf.df["subject_label"].values | ||
| inverted_object_labels = inverted_df["object_label"].values | ||
| self.assertNotIn(False, original_subject_labels == inverted_object_labels) |
There was a problem hiding this comment.
I would have tried to use something like assertSequenceEqual for this but I zero priority as the current looks correct
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.
This is a putative fix to #554.
It implements the fix suggested in this comment, and adds a test case to check that inverting a set containing “imbalanced” columns (e.g., with a
subject_labelcolumn but without aobject_labelcolumn) does not trigger the #554 bug.