Made categorical dictionary more cohesive to overall structure #1454
Conversation
This file contains hidden or 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
… are marked as categorical. Previously string columns are treated as not categorical which transfers to numerical. We decided that encoded string columns are more similar to OHE as they are to numerical transformations. This change effects the metalearning part of `autosklearn` exclusively.
… are marked as categorical. Previously string columns are treated as not categorical which transfers to numerical. We decided that encoded string columns are more similar to OHE as they are to numerical transformations. This change effects the metalearning part of `autosklearn` exclusively.
# Conflicts: # autosklearn/smbo.py
…feat_type dictionary which stores more information and reduce it to categorical where it is really needed.
Codecov Report
@@ Coverage Diff @@
## development #1454 +/- ##
===============================================
- Coverage 84.31% 84.25% -0.06%
===============================================
Files 147 147
Lines 11284 11287 +3
Branches 1934 1939 +5
===============================================
- Hits 9514 9510 -4
- Misses 1256 1257 +1
- Partials 514 520 +6 |
eddiebergman
reviewed
May 2, 2022
Contributor
eddiebergman
left a comment
eddiebergman
left a comment
There was a problem hiding this comment.
I imagine this needs some kind of clear test to describe the intended behaviours as I see what's going on but not what's meant to come from it.
…longer the categorical one
mfeurer
reviewed
May 2, 2022
Collaborator
Author
|
Do some know what the issue is with the |
Contributor
|
Seems it was just a timeout, I woul just rerun it and hope it works. I manually checked the link and it seems to work fine so I dunno what's up |
mfeurer
reviewed
May 4, 2022
Contributor
mfeurer
left a comment
mfeurer
left a comment
There was a problem hiding this comment.
Alright, this look very nice overall. I believe these are a few places where the code can be simplified and it would be great if you could have a look at them.
test/test_metalearning/pyMetaLearn/test_meta_features_sparse.py
Outdated
Show resolved
Hide resolved
test/test_metalearning/pyMetaLearn/test_meta_features_sparse.py
Outdated
Show resolved
Hide resolved
mfeurer
reviewed
May 5, 2022
mfeurer
reviewed
May 5, 2022
mfeurer
approved these changes
May 9, 2022
eddiebergman
pushed a commit
that referenced
this pull request
Aug 18, 2022
* change treatment of string features. In file `smbo.py` string columns are marked as categorical. Previously string columns are treated as not categorical which transfers to numerical. We decided that encoded string columns are more similar to OHE as they are to numerical transformations. This change effects the metalearning part of `autosklearn` exclusively. * change treatment of string features. In file `smbo.py` string columns are marked as categorical. Previously string columns are treated as not categorical which transfers to numerical. We decided that encoded string columns are more similar to OHE as they are to numerical transformations. This change effects the metalearning part of `autosklearn` exclusively. * made categorical dictionary more cohesive. Use the previously create feat_type dictionary which stores more information and reduce it to categorical where it is really needed. * change test files since they now get the feat type dictionary and no longer the categorical one * change test files since they now get the feat type dictionary and no longer the categorical one * moved feat_type to categorical conv. to the helper functions * fixed minor issues * fixed minor issues * fixed minor issues * fixed bug in metalearning tests
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
In this PR we change the
categoricaldictionary to the 'feat_type' dictionary which stores all column name column type pairs.We do not changed the behavior of the code overall.
In the calculate metafeatures.py:1180 we still create the boolean dictionary which than is passed to the helper functions. This ensures that the behavior of the code does not change. Also more advanced changes include including strings to the meta features.