Handle boolean and categorical features for time series#3083
Handle boolean and categorical features for time series#3083freddyaboulton merged 6 commits intomainfrom
Conversation
Codecov Report
@@ Coverage Diff @@
## main #3083 +/- ##
=======================================
+ Coverage 99.8% 99.8% +0.1%
=======================================
Files 312 313 +1
Lines 30437 30468 +31
=======================================
+ Hits 30347 30378 +31
Misses 90 90
Continue to review full report at Codecov.
|
| } | ||
| ) | ||
|
|
||
| X.ww.init(logical_types={"col_1": "categorical"}) |
There was a problem hiding this comment.
These logical types were treated as Unknown before.
| ProblemTypes.TIME_SERIES_REGRESSION, | ||
| ], | ||
| ) | ||
| def test_can_run_automl_for_time_series_with_categorical_and_boolean_features( |
There was a problem hiding this comment.
18 second run time seems reasonable to me
cc32008 to
33116f3
Compare
jeremyliweishih
left a comment
There was a problem hiding this comment.
This looks great to me! Just have a question about potentially adding a test case for the numeric OHE case you added.
| @@ -144,9 +144,9 @@ def test_null_values_in_dataframe(): | |||
| "col_4": [2, 0, 1, 3, 0, 1, 2], | |||
There was a problem hiding this comment.
is there a test case where encoder_error = OneHotEncoder(handle_missing="error") and it fits on a data frame with a numeric column with a null value?
There was a problem hiding this comment.
Tested through the time series integration tests but agree it makes sense to add a test here!
33116f3 to
04d003f
Compare
Pull Request Description
Fixes #3077, Fixes #3082
After creating the pull request: in order to pass the release_notes_updated check you will need to update the "Future Release" section of
docs/source/release_notes.rstto include this pull request by adding :pr:123.