[SPARK-19734] OneHotEncoder __init__ uses dropLast but doc strings all say includeFirst

• Type: Documentation
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.5.2, 1.6.3, 2.0.2, 2.1.0
• Fix Version/s: 2.2.0
• Component/s: PySpark
• Labels: documentation easyfix
• Assignee: Mark Grover Assign to me
• Reporter: Corey
• Created: 25/Feb/17 00:45
• Updated: 02/Mar/17 07:00
• Resolved: 02/Mar/17 07:00

Description

The OneHotEncoder._init_ doc string in PySpark has an input keyword listed as includeFirst, whereas the code actually uses dropLast.

This especially confusing because the _init_ function accepts only keywords, and following the documentation on the web (https://spark.apache.org/docs/2.0.1/api/python/pyspark.ml.html#pyspark.ml.feature.OneHotEncoder) or of help in Python will result in the error:

TypeError: _init_() got an unexpected keyword argument 'includeFirst'

The error is immediately viewable in the source code:

    @keyword_only
    def __init__(self, dropLast=True, inputCol=None, outputCol=None):
        """
        __init__(self, includeFirst=True, inputCol=None, outputCol=None)
        """

Issue Links

• links to: [Github] Pull Request #17127 (markgrover)