ArrayAssembler#

class pyspark.ml.connect.feature.ArrayAssembler(*, inputCols=None, outputCol=None, featureSizes=None, handleInvalid='error')[source]#

A feature transformer that merges multiple input columns into an array type column.

Parameters

You need to set param `inputCols` for specifying input column names,
and set param `featureSizes` for specifying corresponding input column
feature size, for scalar type input column, corresponding feature size must be set to 1,
otherwise, set corresponding feature size to feature array length.
Output column is “array<double”> type and contains array of assembled features.
All elements in input feature columns must be convertible to double type.
You can set ‘handler_invalid’ param to specify how to handle invalid input value
(None or NaN), if it is set to ‘error’, error is thrown for invalid input value,
if it is set to ‘keep’, it returns relevant number of NaN in the output.
.. versionadded:: 4.0.0

Examples

>>> from pyspark.ml.connect.feature import ArrayAssembler
>>> import numpy as np
>>>
>>> spark_df = spark.createDataFrame(
...     [
...         ([2.0, 3.5, 1.5], 3.0, True, 1),
...         ([-3.0, np.nan, -2.5], 4.0, False, 2),
...     ],
...     schema=["f1", "f2", "f3", "f4"],
... )
>>> assembler = ArrayAssembler(
...     inputCols=["f1", "f2", "f3", "f4"],
...     outputCol="out",
...     featureSizes=[3, 1, 1, 1],
...     handleInvalid="keep",
... )
>>> assembler.transform(spark_df).select("out").show(truncate=False)

Methods

`clear`(param)	Clears a param from the param map if it has been explicitly set.
`copy`([extra])	Creates a copy of this instance with the same uid and some extra params.
`explainParam`(param)	Explains a single param and returns its name, doc, and optional default value and user-supplied value in a string.
`explainParams`()	Returns the documentation of all params with their optionally default values and user-supplied values.
`extractParamMap`([extra])	Extracts the embedded default param values and user-supplied values, and then merges them with extra values from input into a flat param map, where the latter value is used if there exist conflicts, i.e., with ordering: default param values < user-supplied values < extra.
`getFeatureSizes`()	Gets the value of featureSizes or its default value.
`getHandleInvalid`()	Gets the value of handleInvalid or its default value.
`getInputCols`()	Gets the value of inputCols or its default value.
`getOrDefault`(param)	Gets the value of a param in the user-supplied param map or its default value.
`getOutputCol`()	Gets the value of outputCol or its default value.
`getParam`(paramName)	Gets a param by its name.
`hasDefault`(param)	Checks whether a param has a default value.
`hasParam`(paramName)	Tests whether this instance contains a param with a given (string) name.
`isDefined`(param)	Checks whether a param is explicitly set by user or has a default value.
`isSet`(param)	Checks whether a param is explicitly set by user.
`load`(path)	Load Estimator / Transformer / Model / Evaluator from provided cloud storage path.
`loadFromLocal`(path)	Load Estimator / Transformer / Model / Evaluator from provided local path.
`save`(path, *[, overwrite])	Save Estimator / Transformer / Model / Evaluator to provided cloud storage path.
`saveToLocal`(path, *[, overwrite])	Save Estimator / Transformer / Model / Evaluator to provided local path.
`set`(param, value)	Sets a parameter in the embedded param map.
`transform`(dataset[, params])	Transforms the input dataset.

Attributes

`featureSizes`
`handleInvalid`
`inputCols`
`outputCol`
`params`	Returns all params ordered by name.

Methods Documentation

clear(param)#: Clears a param from the param map if it has been explicitly set.

copy(extra=None)#

Creates a copy of this instance with the same uid and some extra params. The default implementation creates a shallow copy using copy.copy(), and then copies the embedded and extra parameters over and returns the copy. Subclasses should override this method if the default approach is not sufficient.

Parameters

extradict, optional: Extra parameters to copy to the new instance

Returns

Params: Copy of this instance

explainParam(param)#: Explains a single param and returns its name, doc, and optional default value and user-supplied value in a string.

explainParams()#: Returns the documentation of all params with their optionally default values and user-supplied values.

extractParamMap(extra=None)#

Extracts the embedded default param values and user-supplied values, and then merges them with extra values from input into a flat param map, where the latter value is used if there exist conflicts, i.e., with ordering: default param values < user-supplied values < extra.

Parameters

extradict, optional: extra param values

Returns

dict: merged param map

getFeatureSizes()#: Gets the value of featureSizes or its default value.

getHandleInvalid()#: Gets the value of handleInvalid or its default value.

getInputCols()#: Gets the value of inputCols or its default value.

getOrDefault(param)#: Gets the value of a param in the user-supplied param map or its default value. Raises an error if neither is set.

getOutputCol()#: Gets the value of outputCol or its default value.

getParam(paramName)#: Gets a param by its name.

hasDefault(param)#: Checks whether a param has a default value.

hasParam(paramName)#: Tests whether this instance contains a param with a given (string) name.

isDefined(param)#: Checks whether a param is explicitly set by user or has a default value.

isSet(param)#: Checks whether a param is explicitly set by user.

classmethod load(path)#: Load Estimator / Transformer / Model / Evaluator from provided cloud storage path.

New in version 3.5.0.

classmethod loadFromLocal(path)#: Load Estimator / Transformer / Model / Evaluator from provided local path.

New in version 3.5.0.

save(path, *, overwrite=False)#: Save Estimator / Transformer / Model / Evaluator to provided cloud storage path.

New in version 3.5.0.

saveToLocal(path, *, overwrite=False)#: Save Estimator / Transformer / Model / Evaluator to provided local path.

New in version 3.5.0.

set(param, value)#: Sets a parameter in the embedded param map.

transform(dataset, params=None)#

Transforms the input dataset. The dataset can be either pandas dataframe or spark dataframe， if it is a spark DataFrame, the result of transformation is a new spark DataFrame that contains all existing columns and output columns with names, If it is a pandas DataFrame, the result of transformation is a shallow copy of the input pandas dataframe with output columns with names.

Note: Transformers does not allow output column having the same name with existing columns.

Parameters

datasetpyspark.sql.DataFrame or py:class:pandas.DataFrame: input dataset.
paramsdict, optional: an optional param map that overrides embedded params.

Returns

pyspark.sql.DataFrame or py:class:pandas.DataFrame: transformed dataset, the type of output dataframe is consistent with input dataframe.

Attributes Documentation

featureSizes = Param(parent='undefined', name='featureSizes', doc='input feature size list for input columns of vector assembler')#

handleInvalid = Param(parent='undefined', name='handleInvalid', doc="how to handle invalid entries. Options are 'error' (throw an error), or 'keep' (return relevant number of NaN in the output). Default value is 'error'")#

inputCols = Param(parent='undefined', name='inputCols', doc='input column names.')#

outputCol = Param(parent='undefined', name='outputCol', doc='output column name.')#

params#: Returns all params ordered by name. The default implementation uses dir() to get all attributes of type Param.

uid#: A unique id for the object.