synapse.ml.nn package

Submodules

synapse.ml.nn.ConditionalBallTree module

class synapse.ml.nn.ConditionalBallTree.ConditionalBallTree(keys, values, labels, leafSize, java_obj=None)[source]

Bases: object

findMaximumInnerProducts(queryPoint, conditioner, k)[source]: Find the best match to the queryPoint given the conditioner and k from self. :param queryPoint: array vector to use to query for NNs :param conditioner: set of labels that will subset or condition the NN query :param k: int representing the maximum number of neighbors to return :return: array of tuples representing the index of the match and its distance

static load(filename)[source]

save(filename)[source]

synapse.ml.nn.ConditionalKNN module

class synapse.ml.nn.ConditionalKNN.ConditionalKNN(java_obj=None, conditionerCol='conditioner', featuresCol='features', k=5, labelCol='labels', leafSize=50, outputCol='ConditionalKNN_2f37bc8ec7f9_output', valuesCol='values')[source]

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

Parameters

conditionerCol (object) – column holding identifiers for features that will be returned when queried
featuresCol (object) – The name of the features column
k (int) – number of matches to return
labelCol (object) – The name of the label column
leafSize (int) – max size of the leaves of the tree
outputCol (object) – The name of the output column
valuesCol (object) – column holding values for each feature (key) that will be returned when queried

conditionerCol = Param(parent='undefined', name='conditionerCol', doc='column holding identifiers for features that will be returned when queried')

featuresCol = Param(parent='undefined', name='featuresCol', doc='The name of the features column')

getConditionerCol()[source]

Returns: column holding identifiers for features that will be returned when queried
Return type: conditionerCol

getFeaturesCol()[source]

Returns: The name of the features column
Return type: featuresCol

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

getK()[source]

Returns: number of matches to return
Return type: k

getLabelCol()[source]

Returns: The name of the label column
Return type: labelCol

getLeafSize()[source]

Returns: max size of the leaves of the tree
Return type: leafSize

getOutputCol()[source]

Returns: The name of the output column
Return type: outputCol

getValuesCol()[source]

Returns: column holding values for each feature (key) that will be returned when queried
Return type: valuesCol

k = Param(parent='undefined', name='k', doc='number of matches to return')

labelCol = Param(parent='undefined', name='labelCol', doc='The name of the label column')

leafSize = Param(parent='undefined', name='leafSize', doc='max size of the leaves of the tree')

outputCol = Param(parent='undefined', name='outputCol', doc='The name of the output column')

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

setConditionerCol(value)[source]

Parameters: conditionerCol – column holding identifiers for features that will be returned when queried

setFeaturesCol(value)[source]

Parameters: featuresCol – The name of the features column

setK(value)[source]

Parameters: k – number of matches to return

setLabelCol(value)[source]

Parameters: labelCol – The name of the label column

setLeafSize(value)[source]

Parameters: leafSize – max size of the leaves of the tree

setOutputCol(value)[source]

Parameters: outputCol – The name of the output column

setParams(conditionerCol='conditioner', featuresCol='features', k=5, labelCol='labels', leafSize=50, outputCol='ConditionalKNN_2f37bc8ec7f9_output', valuesCol='values')[source]: Set the (keyword only) parameters

setValuesCol(value)[source]

Parameters: valuesCol – column holding values for each feature (key) that will be returned when queried

valuesCol = Param(parent='undefined', name='valuesCol', doc='column holding values for each feature (key) that will be returned when queried')

synapse.ml.nn.ConditionalKNNModel module

class synapse.ml.nn.ConditionalKNNModel.ConditionalKNNModel(java_obj=None, ballTree=None, conditionerCol=None, featuresCol=None, k=None, labelCol=None, leafSize=None, outputCol=None, valuesCol=None)[source]

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

Parameters

ballTree (object) – the ballTree model used for perfoming queries
conditionerCol (object) – column holding identifiers for features that will be returned when queried
featuresCol (object) – The name of the features column
k (int) – number of matches to return
labelCol (object) – The name of the label column
leafSize (int) – max size of the leaves of the tree
outputCol (object) – The name of the output column
valuesCol (object) – column holding values for each feature (key) that will be returned when queried

ballTree = Param(parent='undefined', name='ballTree', doc='the ballTree model used for perfoming queries')

conditionerCol = Param(parent='undefined', name='conditionerCol', doc='column holding identifiers for features that will be returned when queried')

featuresCol = Param(parent='undefined', name='featuresCol', doc='The name of the features column')

getBallTree()[source]

Returns: the ballTree model used for perfoming queries
Return type: ballTree

getConditionerCol()[source]

Returns: column holding identifiers for features that will be returned when queried
Return type: conditionerCol

getFeaturesCol()[source]

Returns: The name of the features column
Return type: featuresCol

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

getK()[source]

Returns: number of matches to return
Return type: k

getLabelCol()[source]

Returns: The name of the label column
Return type: labelCol

getLeafSize()[source]

Returns: max size of the leaves of the tree
Return type: leafSize

getOutputCol()[source]

Returns: The name of the output column
Return type: outputCol

getValuesCol()[source]

Returns: column holding values for each feature (key) that will be returned when queried
Return type: valuesCol

k = Param(parent='undefined', name='k', doc='number of matches to return')

labelCol = Param(parent='undefined', name='labelCol', doc='The name of the label column')

leafSize = Param(parent='undefined', name='leafSize', doc='max size of the leaves of the tree')

outputCol = Param(parent='undefined', name='outputCol', doc='The name of the output column')

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

setBallTree(value)[source]

Parameters: ballTree – the ballTree model used for perfoming queries

setConditionerCol(value)[source]

Parameters: conditionerCol – column holding identifiers for features that will be returned when queried

setFeaturesCol(value)[source]

Parameters: featuresCol – The name of the features column

setK(value)[source]

Parameters: k – number of matches to return

setLabelCol(value)[source]

Parameters: labelCol – The name of the label column

setLeafSize(value)[source]

Parameters: leafSize – max size of the leaves of the tree

setOutputCol(value)[source]

Parameters: outputCol – The name of the output column

setParams(ballTree=None, conditionerCol=None, featuresCol=None, k=None, labelCol=None, leafSize=None, outputCol=None, valuesCol=None)[source]: Set the (keyword only) parameters

setValuesCol(value)[source]

Parameters: valuesCol – column holding values for each feature (key) that will be returned when queried

valuesCol = Param(parent='undefined', name='valuesCol', doc='column holding values for each feature (key) that will be returned when queried')

synapse.ml.nn.KNN module

class synapse.ml.nn.KNN.KNN(java_obj=None, featuresCol='features', k=5, leafSize=50, outputCol='KNN_12971ac2f3db_output', valuesCol='values')[source]

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

Parameters

featuresCol (object) – The name of the features column
k (int) – number of matches to return
leafSize (int) – max size of the leaves of the tree
outputCol (object) – The name of the output column
valuesCol (object) – column holding values for each feature (key) that will be returned when queried

featuresCol = Param(parent='undefined', name='featuresCol', doc='The name of the features column')

getFeaturesCol()[source]

Returns: The name of the features column
Return type: featuresCol

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

getK()[source]

Returns: number of matches to return
Return type: k

getLeafSize()[source]

Returns: max size of the leaves of the tree
Return type: leafSize

getOutputCol()[source]

Returns: The name of the output column
Return type: outputCol

getValuesCol()[source]

Returns: column holding values for each feature (key) that will be returned when queried
Return type: valuesCol

k = Param(parent='undefined', name='k', doc='number of matches to return')

leafSize = Param(parent='undefined', name='leafSize', doc='max size of the leaves of the tree')

outputCol = Param(parent='undefined', name='outputCol', doc='The name of the output column')

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

setFeaturesCol(value)[source]

Parameters: featuresCol – The name of the features column

setK(value)[source]

Parameters: k – number of matches to return

setLeafSize(value)[source]

Parameters: leafSize – max size of the leaves of the tree

setOutputCol(value)[source]

Parameters: outputCol – The name of the output column

setParams(featuresCol='features', k=5, leafSize=50, outputCol='KNN_12971ac2f3db_output', valuesCol='values')[source]: Set the (keyword only) parameters

setValuesCol(value)[source]

Parameters: valuesCol – column holding values for each feature (key) that will be returned when queried

valuesCol = Param(parent='undefined', name='valuesCol', doc='column holding values for each feature (key) that will be returned when queried')

synapse.ml.nn.KNNModel module

class synapse.ml.nn.KNNModel.KNNModel(java_obj=None, ballTree=None, featuresCol=None, k=None, leafSize=None, outputCol=None, valuesCol=None)[source]

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

Parameters

ballTree (object) – the ballTree model used for performing queries
featuresCol (object) – The name of the features column
k (int) – number of matches to return
leafSize (int) – max size of the leaves of the tree
outputCol (object) – The name of the output column
valuesCol (object) – column holding values for each feature (key) that will be returned when queried

ballTree = Param(parent='undefined', name='ballTree', doc='the ballTree model used for performing queries')

featuresCol = Param(parent='undefined', name='featuresCol', doc='The name of the features column')

getBallTree()[source]

Returns: the ballTree model used for performing queries
Return type: ballTree

getFeaturesCol()[source]

Returns: The name of the features column
Return type: featuresCol

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

getK()[source]

Returns: number of matches to return
Return type: k

getLeafSize()[source]

Returns: max size of the leaves of the tree
Return type: leafSize

getOutputCol()[source]

Returns: The name of the output column
Return type: outputCol

getValuesCol()[source]

Returns: column holding values for each feature (key) that will be returned when queried
Return type: valuesCol

k = Param(parent='undefined', name='k', doc='number of matches to return')

leafSize = Param(parent='undefined', name='leafSize', doc='max size of the leaves of the tree')

outputCol = Param(parent='undefined', name='outputCol', doc='The name of the output column')

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

setBallTree(value)[source]

Parameters: ballTree – the ballTree model used for performing queries

setFeaturesCol(value)[source]

Parameters: featuresCol – The name of the features column

setK(value)[source]

Parameters: k – number of matches to return

setLeafSize(value)[source]

Parameters: leafSize – max size of the leaves of the tree

setOutputCol(value)[source]

Parameters: outputCol – The name of the output column

setParams(ballTree=None, featuresCol=None, k=None, leafSize=None, outputCol=None, valuesCol=None)[source]: Set the (keyword only) parameters

setValuesCol(value)[source]

Parameters: valuesCol – column holding values for each feature (key) that will be returned when queried

valuesCol = Param(parent='undefined', name='valuesCol', doc='column holding values for each feature (key) that will be returned when queried')

Module contents

SynapseML is an ecosystem of tools aimed towards expanding the distributed computing framework Apache Spark in several new directions. SynapseML adds many deep learning and data science tools to the Spark ecosystem, including seamless integration of Spark Machine Learning pipelines with Microsoft Cognitive Toolkit (CNTK), LightGBM and OpenCV. These tools enable powerful and highly-scalable predictive and analytical models for a variety of datasources.

SynapseML also brings new networking capabilities to the Spark Ecosystem. With the HTTP on Spark project, users can embed any web service into their SparkML models. In this vein, SynapseML provides easy to use SparkML transformers for a wide variety of Microsoft Cognitive Services. For production grade deployment, the Spark Serving project enables high throughput, sub-millisecond latency web services, backed by your Spark cluster.

SynapseML requires Scala 2.12, Spark 3.0+, and Python 3.6+.