drop_first attribute added in encode features #647
Conversation
|
|
Codecov Report
@@ Coverage Diff @@
## master #647 +/- ##
=========================================
Coverage ? 97.42%
=========================================
Files ? 118
Lines ? 9560
Branches ? 0
=========================================
Hits ? 9314
Misses ? 246
Partials ? 0
Continue to review full report at Codecov.
|
|
Thank you for the contribution, @ayushpatidar. A couple things
Once you are passing all the github checks, we can review. |
| @@ -23,6 +23,9 @@ def encode_features(feature_matrix, features, top_n=10, include_unknown=True, | |||
| defaults to encode all necessary features. | |||
| inplace (bool): Encode feature_matrix in place. Defaults to False. | |||
| verbose (str): Print progress info. | |||
| drop_first (bool): Whether to get l-1 dummies out of l categorical | |||
| unique = val_counts.head(top_n).index.tolist() | ||
| select_n = top_n | ||
| if drop_first: | ||
| if len(val_counts) != 1 or top_n != 1: |
There was a problem hiding this comment.
what if we remove the conditional and just do these two lines?
select_n = min(len(val_counts), top_n)
select_n = max(select_n - 1, 1) # make sure at least 1 category is selected
There was a problem hiding this comment.
@kmax12
But we have to keep outer conditional statement ( if drop_first ).
There was a problem hiding this comment.
yep, sorry, I was just referring to inner conditional
| es.entity_from_dataframe(entity_id='a', dataframe=df, index='index', make_index=True) | ||
| features, feature_defs = dfs(entityset=es, target_entity='a') | ||
| features_enc, feature_defs_enc = encode_features(features, feature_defs, | ||
| drop_first=True, top_n=10, include_unknown=False) |
There was a problem hiding this comment.
no need to include top_n in this test case
Pull Request Description
drop_first attribute is added as param in encode_feature() function.
This feature will help user to control on redundant feature in dataframe
Fixes #635
After creating the pull request: in order to pass the changelog_updated check you will need to update the "Future Release" section of
docs/source/changelog.rstto include this pull request.