mmlspark.recommendation package¶

Submodules¶

mmlspark.recommendation.RankingAdapter module¶

class mmlspark.recommendation.RankingAdapter.RankingAdapter(itemCol=None, k=10, labelCol='label', minRatingsPerItem=1, minRatingsPerUser=1, mode='allUsers', ratingCol=None, recommender=None, userCol=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.wrapper.JavaEstimator

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – The name of the label column (default: label)
minRatingsPerItem (int) – min ratings for items > 0 (default: 1)
minRatingsPerUser (int) – min ratings for users > 0 (default: 1)
mode (str) – recommendation mode (default: allUsers)
ratingCol (str) – Column of ratings
recommender (object) – estimator for selection
userCol (str) – Column of users

getItemCol()[source]¶

Returns: Column of items
Return type: str

static getJavaPackage()[source]¶: Returns package name String.

getK()[source]¶

Returns: number of items (default: 10)
Return type: int

getLabelCol()[source]¶

Returns: The name of the label column (default: label)
Return type: str

getMinRatingsPerItem()[source]¶

Returns: min ratings for items > 0 (default: 1)
Return type: int

getMinRatingsPerUser()[source]¶

Returns: min ratings for users > 0 (default: 1)
Return type: int

getMode()[source]¶

Returns: recommendation mode (default: allUsers)
Return type: str

getRatingCol()[source]¶

Returns: Column of ratings
Return type: str

getRecommender()[source]¶

Returns: estimator for selection
Return type: object

getUserCol()[source]¶

Returns: Column of users
Return type: str

classmethod read()[source]¶: Returns an MLReader instance for this class.

setItemCol(value)[source]¶

Parameters: itemCol (str) – Column of items

setK(value)[source]¶

Parameters: k (int) – number of items (default: 10)

setLabelCol(value)[source]¶

Parameters: labelCol (str) – The name of the label column (default: label)

setMinRatingsPerItem(value)[source]¶

Parameters: minRatingsPerItem (int) – min ratings for items > 0 (default: 1)

setMinRatingsPerUser(value)[source]¶

Parameters: minRatingsPerUser (int) – min ratings for users > 0 (default: 1)

setMode(value)[source]¶

Parameters: mode (str) – recommendation mode (default: allUsers)

setParams(itemCol=None, k=10, labelCol='label', minRatingsPerItem=1, minRatingsPerUser=1, mode='allUsers', ratingCol=None, recommender=None, userCol=None)[source]¶

Set the (keyword only) parameters

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – The name of the label column (default: label)
minRatingsPerItem (int) – min ratings for items > 0 (default: 1)
minRatingsPerUser (int) – min ratings for users > 0 (default: 1)
mode (str) – recommendation mode (default: allUsers)
ratingCol (str) – Column of ratings
recommender (object) – estimator for selection
userCol (str) – Column of users

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – Column of ratings

setRecommender(value)[source]¶

Parameters: recommender (object) – estimator for selection

setUserCol(value)[source]¶

Parameters: userCol (str) – Column of users

class mmlspark.recommendation.RankingAdapter.RankingAdapterModel(java_model=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.wrapper.JavaModel, pyspark.ml.util.JavaMLWritable, pyspark.ml.util.JavaMLReadable

Model fitted by RankingAdapter.

This class is left empty on purpose. All necessary methods are exposed through inheritance.

static getJavaPackage()[source]¶: Returns package name String.

classmethod read()[source]¶: Returns an MLReader instance for this class.

mmlspark.recommendation.RankingAdapterModel module¶

class mmlspark.recommendation.RankingAdapterModel.RankingAdapterModel(itemCol=None, k=10, labelCol='label', minRatingsPerItem=1, minRatingsPerUser=1, mode='allUsers', ratingCol=None, recommender=None, recommenderModel=None, userCol=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.wrapper.JavaTransformer

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – The name of the label column (default: label)
minRatingsPerItem (int) – min ratings for items > 0 (default: 1)
minRatingsPerUser (int) – min ratings for users > 0 (default: 1)
mode (str) – recommendation mode (default: allUsers)
ratingCol (str) – Column of ratings
recommender (object) – estimator for selection
recommenderModel (object) – recommenderModel
userCol (str) – Column of users

getItemCol()[source]¶

Returns: Column of items
Return type: str

static getJavaPackage()[source]¶: Returns package name String.

getK()[source]¶

Returns: number of items (default: 10)
Return type: int

getLabelCol()[source]¶

Returns: The name of the label column (default: label)
Return type: str

getMinRatingsPerItem()[source]¶

Returns: min ratings for items > 0 (default: 1)
Return type: int

getMinRatingsPerUser()[source]¶

Returns: min ratings for users > 0 (default: 1)
Return type: int

getMode()[source]¶

Returns: recommendation mode (default: allUsers)
Return type: str

getRatingCol()[source]¶

Returns: Column of ratings
Return type: str

getRecommender()[source]¶

Returns: estimator for selection
Return type: object

getRecommenderModel()[source]¶

Returns: recommenderModel
Return type: object

getUserCol()[source]¶

Returns: Column of users
Return type: str

classmethod read()[source]¶: Returns an MLReader instance for this class.

setItemCol(value)[source]¶

Parameters: itemCol (str) – Column of items

setK(value)[source]¶

Parameters: k (int) – number of items (default: 10)

setLabelCol(value)[source]¶

Parameters: labelCol (str) – The name of the label column (default: label)

setMinRatingsPerItem(value)[source]¶

Parameters: minRatingsPerItem (int) – min ratings for items > 0 (default: 1)

setMinRatingsPerUser(value)[source]¶

Parameters: minRatingsPerUser (int) – min ratings for users > 0 (default: 1)

setMode(value)[source]¶

Parameters: mode (str) – recommendation mode (default: allUsers)

setParams(itemCol=None, k=10, labelCol='label', minRatingsPerItem=1, minRatingsPerUser=1, mode='allUsers', ratingCol=None, recommender=None, recommenderModel=None, userCol=None)[source]¶

Set the (keyword only) parameters

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – The name of the label column (default: label)
minRatingsPerItem (int) – min ratings for items > 0 (default: 1)
minRatingsPerUser (int) – min ratings for users > 0 (default: 1)
mode (str) – recommendation mode (default: allUsers)
ratingCol (str) – Column of ratings
recommender (object) – estimator for selection
recommenderModel (object) – recommenderModel
userCol (str) – Column of users

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – Column of ratings

setRecommender(value)[source]¶

Parameters: recommender (object) – estimator for selection

setRecommenderModel(value)[source]¶

Parameters: recommenderModel (object) – recommenderModel

setUserCol(value)[source]¶

Parameters: userCol (str) – Column of users

mmlspark.recommendation.RankingEvaluator module¶

class mmlspark.recommendation.RankingEvaluator.RankingEvaluator(itemCol=None, k=10, labelCol='label', metricName='ndcgAt', nItems=-1, predictionCol='prediction', ratingCol=None, userCol=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.evaluation.JavaEvaluator

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – label column name (default: label)
metricName (str) – metric name in evaluation (ndcgAt|map|precisionAtk|recallAtK|diversityAtK|maxDiversity|mrr|fcp) (default: ndcgAt)
nItems (long) – number of items (default: -1)
predictionCol (str) – prediction column name (default: prediction)
ratingCol (str) – Column of ratings
userCol (str) – Column of users

getItemCol()[source]¶

Returns: Column of items
Return type: str

static getJavaPackage()[source]¶: Returns package name String.

getK()[source]¶

Returns: number of items (default: 10)
Return type: int

getLabelCol()[source]¶

Returns: label column name (default: label)
Return type: str

getMetricName()[source]¶

Returns: metric name in evaluation (ndcgAt|map|precisionAtk|recallAtK|diversityAtK|maxDiversity|mrr|fcp) (default: ndcgAt)
Return type: str

getNItems()[source]¶

Returns: number of items (default: -1)
Return type: long

getPredictionCol()[source]¶

Returns: prediction column name (default: prediction)
Return type: str

getRatingCol()[source]¶

Returns: Column of ratings
Return type: str

getUserCol()[source]¶

Returns: Column of users
Return type: str

classmethod read()[source]¶: Returns an MLReader instance for this class.

setItemCol(value)[source]¶

Parameters: itemCol (str) – Column of items

setK(value)[source]¶

Parameters: k (int) – number of items (default: 10)

setLabelCol(value)[source]¶

Parameters: labelCol (str) – label column name (default: label)

setMetricName(value)[source]¶

Parameters: metricName (str) – metric name in evaluation (ndcgAt|map|precisionAtk|recallAtK|diversityAtK|maxDiversity|mrr|fcp) (default: ndcgAt)

setNItems(value)[source]¶

Parameters: nItems (long) – number of items (default: -1)

setParams(itemCol=None, k=10, labelCol='label', metricName='ndcgAt', nItems=-1, predictionCol='prediction', ratingCol=None, userCol=None)[source]¶

Set the (keyword only) parameters

Parameters

itemCol (str) – Column of items
k (int) – number of items (default: 10)
labelCol (str) – label column name (default: label)
metricName (str) – metric name in evaluation (ndcgAt|map|precisionAtk|recallAtK|diversityAtK|maxDiversity|mrr|fcp) (default: ndcgAt)
nItems (long) – number of items (default: -1)
predictionCol (str) – prediction column name (default: prediction)
ratingCol (str) – Column of ratings
userCol (str) – Column of users

setPredictionCol(value)[source]¶

Parameters: predictionCol (str) – prediction column name (default: prediction)

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – Column of ratings

setUserCol(value)[source]¶

Parameters: userCol (str) – Column of users

mmlspark.recommendation.RankingTrainValidationSplit module¶

class mmlspark.recommendation.RankingTrainValidationSplit.RankingTrainValidationSplit(estimator=None, estimatorParamMaps=None, evaluator=None, seed=None)[source]¶

Bases: pyspark.ml.base.Estimator, pyspark.ml.tuning.ValidatorParams

copy(extra=None)[source]¶

Creates a copy of this instance with a randomly generated uid and some extra com.microsoft.ml.spark.core.serialize.params. This copies creates a deep copy of the embedded paramMap, and copies the embedded and extra parameters over.

Parameters: extra – Extra parameters to copy to the new instance
Returns: Copy of this instance

getItemCol()[source]¶

Returns: column name for item ids. Ids must be within the integer value range. (default: item)
Return type: str

getRatingCol()[source]¶

Returns: column name for ratings (default: rating)
Return type: str

getTrainRatio()[source]¶: Gets the value of trainRatio or its default value.

getUserCol()[source]¶

Returns: column name for user ids. Ids must be within the integer value range. (default: user)
Return type: str

itemCol = Param(parent='undefined', name='itemCol', doc='itemCol: column name for item ids. Ids must be within the integer value range. (default: item)')¶

ratingCol = Param(parent='undefined', name='ratingCol', doc='ratingCol: column name for ratings (default: rating)')¶

setItemCol(value)[source]¶

Parameters: itemCol (str) – column name for item ids. Ids must be within the integer value range. (default: item)

setParams(estimator=None, estimatorParamMaps=None, evaluator=None, seed=None)[source]¶: setParams(self, estimator=None, estimatorParamMaps=None, evaluator=None, numFolds=3, seed=None): Sets com.microsoft.ml.spark.core.serialize.params for cross validator.

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – column name for ratings (default: rating)

setTrainRatio(value)[source]¶: Sets the value of trainRatio.

setUserCol(value)[source]¶

Parameters: userCol (str) – column name for user ids. Ids must be within the integer value range. (default: user)

trainRatio = Param(parent='undefined', name='trainRatio', doc='Param for ratio between train and validation data. Must be between 0 and 1.')¶

userCol = Param(parent='undefined', name='userCol', doc='userCol: column name for user ids. Ids must be within the integer value range. (default: user)')¶

mmlspark.recommendation.RankingTrainValidationSplitModel module¶

class mmlspark.recommendation.RankingTrainValidationSplitModel.RankingTrainValidationSplitModel(bestModel=None, validationMetrics=[])[source]¶

Bases: mmlspark.recommendation._RankingTrainValidationSplitModel._RankingTrainValidationSplitModel, pyspark.ml.tuning.ValidatorParams

bestModel = None¶: best model from cross validation

copy(extra=None)[source]¶

Creates a copy of this instance with a randomly generated uid and some extra com.microsoft.ml.spark.core.serialize.params. This copies the underlying bestModel, creates a deep copy of the embedded paramMap, and copies the embedded and extra parameters over. And, this creates a shallow copy of the validationMetrics.

Parameters: extra – Extra parameters to copy to the new instance
Returns: Copy of this instance

recommendForAllItems(numItems)[source]¶

recommendForAllUsers(numItems)[source]¶

validationMetrics = None¶: evaluated validation metrics

mmlspark.recommendation.RecommendationIndexer module¶

class mmlspark.recommendation.RecommendationIndexer.RecommendationIndexer(itemInputCol=None, itemOutputCol=None, ratingCol=None, userInputCol=None, userOutputCol=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.wrapper.JavaEstimator

Parameters

itemInputCol (str) – Item Input Col
itemOutputCol (str) – Item Output Col
ratingCol (str) – Rating Col
userInputCol (str) – User Input Col
userOutputCol (str) – User Output Col

getItemInputCol()[source]¶

Returns: Item Input Col
Return type: str

getItemOutputCol()[source]¶

Returns: Item Output Col
Return type: str

static getJavaPackage()[source]¶: Returns package name String.

getRatingCol()[source]¶

Returns: Rating Col
Return type: str

getUserInputCol()[source]¶

Returns: User Input Col
Return type: str

getUserOutputCol()[source]¶

Returns: User Output Col
Return type: str

classmethod read()[source]¶: Returns an MLReader instance for this class.

setItemInputCol(value)[source]¶

Parameters: itemInputCol (str) – Item Input Col

setItemOutputCol(value)[source]¶

Parameters: itemOutputCol (str) – Item Output Col

setParams(itemInputCol=None, itemOutputCol=None, ratingCol=None, userInputCol=None, userOutputCol=None)[source]¶

Set the (keyword only) parameters

Parameters

itemInputCol (str) – Item Input Col
itemOutputCol (str) – Item Output Col
ratingCol (str) – Rating Col
userInputCol (str) – User Input Col
userOutputCol (str) – User Output Col

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – Rating Col

setUserInputCol(value)[source]¶

Parameters: userInputCol (str) – User Input Col

setUserOutputCol(value)[source]¶

Parameters: userOutputCol (str) – User Output Col

class mmlspark.recommendation.RecommendationIndexer.RecommendationIndexerModel(java_model=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.wrapper.JavaModel, pyspark.ml.util.JavaMLWritable, pyspark.ml.util.JavaMLReadable

Model fitted by RecommendationIndexer.

This class is left empty on purpose. All necessary methods are exposed through inheritance.

static getJavaPackage()[source]¶: Returns package name String.

classmethod read()[source]¶: Returns an MLReader instance for this class.

mmlspark.recommendation.RecommendationIndexerModel module¶

class mmlspark.recommendation.RecommendationIndexerModel.RecommendationIndexerModel(itemIndexModel=None, itemInputCol=None, itemOutputCol=None, ratingCol=None, userIndexModel=None, userInputCol=None, userOutputCol=None)[source]¶

Bases: mmlspark.core.schema.Utils.ComplexParamsMixin, pyspark.ml.util.JavaMLReadable, pyspark.ml.util.JavaMLWritable, pyspark.ml.wrapper.JavaTransformer

Parameters

itemIndexModel (object) – itemIndexModel
itemInputCol (str) – Item Input Col
itemOutputCol (str) – Item Output Col
ratingCol (str) – Rating Col
userIndexModel (object) – userIndexModel
userInputCol (str) – User Input Col
userOutputCol (str) – User Output Col

getItemIndexModel()[source]¶

Returns: itemIndexModel
Return type: object

getItemInputCol()[source]¶

Returns: Item Input Col
Return type: str

getItemOutputCol()[source]¶

Returns: Item Output Col
Return type: str

static getJavaPackage()[source]¶: Returns package name String.

getRatingCol()[source]¶

Returns: Rating Col
Return type: str

getUserIndexModel()[source]¶

Returns: userIndexModel
Return type: object

getUserInputCol()[source]¶

Returns: User Input Col
Return type: str

getUserOutputCol()[source]¶

Returns: User Output Col
Return type: str

classmethod read()[source]¶: Returns an MLReader instance for this class.

setItemIndexModel(value)[source]¶

Parameters: itemIndexModel (object) – itemIndexModel

setItemInputCol(value)[source]¶

Parameters: itemInputCol (str) – Item Input Col

setItemOutputCol(value)[source]¶

Parameters: itemOutputCol (str) – Item Output Col

setParams(itemIndexModel=None, itemInputCol=None, itemOutputCol=None, ratingCol=None, userIndexModel=None, userInputCol=None, userOutputCol=None)[source]¶

Set the (keyword only) parameters

Parameters

itemIndexModel (object) – itemIndexModel
itemInputCol (str) – Item Input Col
itemOutputCol (str) – Item Output Col
ratingCol (str) – Rating Col
userIndexModel (object) – userIndexModel
userInputCol (str) – User Input Col
userOutputCol (str) – User Output Col

setRatingCol(value)[source]¶

Parameters: ratingCol (str) – Rating Col

setUserIndexModel(value)[source]¶

Parameters: userIndexModel (object) – userIndexModel

setUserInputCol(value)[source]¶

Parameters: userInputCol (str) – User Input Col

setUserOutputCol(value)[source]¶

Parameters: userOutputCol (str) – User Output Col

mmlspark.recommendation.SAR module¶

class mmlspark.recommendation.SAR.SAR(activityTimeFormat="yyyy/MM/dd'T'h:mm:ss", alpha=1.0, checkpointInterval=10, coldStartStrategy='nan', finalStorageLevel='MEMORY_AND_DISK', implicitPrefs=False, intermediateStorageLevel='MEMORY_AND_DISK', itemCol='item', maxIter=10, nonnegative=False, numItemBlocks=10, numUserBlocks=10, predictionCol='prediction', rank=10, ratingCol='rating', regParam=0.1, seed=-1219638142, similarityFunction='jaccard', startTime=None, startTimeFormat='EEE MMM dd HH:mm:ss Z yyyy', supportThreshold=4, timeCol='time', timeDecayCoeff=30, userCol='user')[source]¶: Bases: mmlspark.recommendation._SAR._SAR

mmlspark.recommendation.SARModel module¶

class mmlspark.recommendation.SARModel.SARModel(activityTimeFormat="yyyy/MM/dd'T'h:mm:ss", alpha=1.0, checkpointInterval=10, coldStartStrategy='nan', finalStorageLevel='MEMORY_AND_DISK', implicitPrefs=False, intermediateStorageLevel='MEMORY_AND_DISK', itemCol='item', itemDataFrame=None, maxIter=10, nonnegative=False, numItemBlocks=10, numUserBlocks=10, predictionCol='prediction', rank=10, ratingCol='rating', regParam=0.1, seed=-809975865, similarityFunction='jaccard', startTime=None, startTimeFormat='EEE MMM dd HH:mm:ss Z yyyy', supportThreshold=4, timeCol='time', timeDecayCoeff=30, userCol='user', userDataFrame=None)[source]¶

Bases: mmlspark.recommendation._SARModel._SARModel

recommendForAllUsers(numItems)[source]¶

Module contents¶

MicrosoftML is a library of Python classes to interface with the Microsoft scala APIs to utilize Apache Spark to create distibuted machine learning models.

MicrosoftML simplifies training and scoring classifiers and regressors, as well as facilitating the creation of models using the CNTK library, images, and text.