c

com.johnsnowlabs.nlp.annotators.classifier.dl

MultiClassifierDLApproach

class MultiClassifierDLApproach extends AnnotatorApproach[MultiClassifierDLModel] with ParamsAndFeaturesWritable

Trains a MultiClassifierDL for Multi-label Text Classification.

MultiClassifierDL uses a Bidirectional GRU with a convolutional model that we have built inside TensorFlow and supports up to 100 classes.

For instantiated/pretrained models, see MultiClassifierDLModel.

The input to MultiClassifierDL are Sentence Embeddings such as the state-of-the-art UniversalSentenceEncoder, BertSentenceEmbeddings, or SentenceEmbeddings.

In machine learning, multi-label classification and the strongly related problem of multi-output classification are variants of the classification problem where multiple labels may be assigned to each instance. Multi-label classification is a generalization of multiclass classification, which is the single-label problem of categorizing instances into precisely one of more than two classes; in the multi-label problem there is no constraint on how many of the classes the instance can be assigned to. Formally, multi-label classification is the problem of finding a model that maps inputs x to binary vectors y (assigning a value of 0 or 1 for each element (label) in y).

Notes:

For extended examples of usage, see the Spark NLP Workshop and the MultiClassifierDLTestSpec.

Example

In this example, the training data has the form (Note: labels can be arbitrary)

mr,ref
"name[Alimentum], area[city centre], familyFriendly[no], near[Burger King]",Alimentum is an adult establish found in the city centre area near Burger King.
"name[Alimentum], area[city centre], familyFriendly[yes]",Alimentum is a family-friendly place in the city centre.
...

It needs some pre-processing first, so the labels are of type Array[String]. This can be done like so:

import spark.implicits._
import com.johnsnowlabs.nlp.annotators.classifier.dl.MultiClassifierDLApproach
import com.johnsnowlabs.nlp.base.DocumentAssembler
import com.johnsnowlabs.nlp.embeddings.UniversalSentenceEncoder
import org.apache.spark.ml.Pipeline
import org.apache.spark.sql.functions.{col, udf}

// Process training data to create text with associated array of labels
def splitAndTrim = udf { labels: String =>
  labels.split(", ").map(x=>x.trim)
}

val smallCorpus = spark.read
  .option("header", true)
  .option("inferSchema", true)
  .option("mode", "DROPMALFORMED")
  .csv("src/test/resources/classifier/e2e.csv")
  .withColumn("labels", splitAndTrim(col("mr")))
  .withColumn("text", col("ref"))
  .drop("mr")

smallCorpus.printSchema()
// root
// |-- ref: string (nullable = true)
// |-- labels: array (nullable = true)
// |    |-- element: string (containsNull = true)

// Then create pipeline for training
val documentAssembler = new DocumentAssembler()
  .setInputCol("text")
  .setOutputCol("document")
  .setCleanupMode("shrink")

val embeddings = UniversalSentenceEncoder.pretrained()
  .setInputCols("document")
  .setOutputCol("embeddings")

val docClassifier = new MultiClassifierDLApproach()
  .setInputCols("embeddings")
  .setOutputCol("category")
  .setLabelColumn("labels")
  .setBatchSize(128)
  .setMaxEpochs(10)
  .setLr(1e-3f)
  .setThreshold(0.5f)
  .setValidationSplit(0.1f)

val pipeline = new Pipeline()
  .setStages(
    Array(
      documentAssembler,
      embeddings,
      docClassifier
    )
  )

val pipelineModel = pipeline.fit(smallCorpus)
See also

Multi-label classification on Wikipedia

ClassifierDLApproach for single-class classification

SentimentDLApproach for sentiment analysis

Linear Supertypes
ParamsAndFeaturesWritable, HasFeatures, AnnotatorApproach[MultiClassifierDLModel], CanBeLazy, DefaultParamsWritable, MLWritable, HasOutputAnnotatorType, HasOutputAnnotationCol, HasInputAnnotationCols, Estimator[MultiClassifierDLModel], PipelineStage, Logging, Params, Serializable, Serializable, Identifiable, AnyRef, Any
Ordering
  1. Grouped
  2. Alphabetic
  3. By Inheritance
Inherited
  1. MultiClassifierDLApproach
  2. ParamsAndFeaturesWritable
  3. HasFeatures
  4. AnnotatorApproach
  5. CanBeLazy
  6. DefaultParamsWritable
  7. MLWritable
  8. HasOutputAnnotatorType
  9. HasOutputAnnotationCol
  10. HasInputAnnotationCols
  11. Estimator
  12. PipelineStage
  13. Logging
  14. Params
  15. Serializable
  16. Serializable
  17. Identifiable
  18. AnyRef
  19. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new MultiClassifierDLApproach()
  2. new MultiClassifierDLApproach(uid: String)

Type Members

  1. type AnnotatorType = String
    Definition Classes
    HasOutputAnnotatorType

Value Members

  1. final def !=(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int
    Definition Classes
    AnyRef → Any
  3. final def $[T](param: Param[T]): T
    Attributes
    protected
    Definition Classes
    Params
  4. def $$[T](feature: StructFeature[T]): T
    Attributes
    protected
    Definition Classes
    HasFeatures
  5. def $$[K, V](feature: MapFeature[K, V]): Map[K, V]
    Attributes
    protected
    Definition Classes
    HasFeatures
  6. def $$[T](feature: SetFeature[T]): Set[T]
    Attributes
    protected
    Definition Classes
    HasFeatures
  7. def $$[T](feature: ArrayFeature[T]): Array[T]
    Attributes
    protected
    Definition Classes
    HasFeatures
  8. final def ==(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  9. def _fit(dataset: Dataset[_], recursiveStages: Option[PipelineModel]): MultiClassifierDLModel
    Attributes
    protected
    Definition Classes
    AnnotatorApproach
  10. final def asInstanceOf[T0]: T0
    Definition Classes
    Any
  11. val batchSize: IntParam

    Batch size (Default: 64)

  12. def beforeTraining(spark: SparkSession): Unit
  13. final def checkSchema(schema: StructType, inputAnnotatorType: String): Boolean
    Attributes
    protected
    Definition Classes
    HasInputAnnotationCols
  14. final def clear(param: Param[_]): MultiClassifierDLApproach.this.type
    Definition Classes
    Params
  15. def clone(): AnyRef
    Attributes
    protected[lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... ) @native()
  16. val configProtoBytes: IntArrayParam

    ConfigProto from tensorflow, serialized into byte array.

    ConfigProto from tensorflow, serialized into byte array. Get with config_proto.SerializeToString()

  17. final def copy(extra: ParamMap): Estimator[MultiClassifierDLModel]
    Definition Classes
    AnnotatorApproach → Estimator → PipelineStage → Params
  18. def copyValues[T <: Params](to: T, extra: ParamMap): T
    Attributes
    protected
    Definition Classes
    Params
  19. final def defaultCopy[T <: Params](extra: ParamMap): T
    Attributes
    protected
    Definition Classes
    Params
  20. val description: String

    Trains TensorFlow model for multi-class text classification

    Trains TensorFlow model for multi-class text classification

    Definition Classes
    MultiClassifierDLApproachAnnotatorApproach
  21. val enableOutputLogs: BooleanParam

    Whether to output to annotators log folder (Default: false)

  22. final def eq(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  23. def equals(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  24. def explainParam(param: Param[_]): String
    Definition Classes
    Params
  25. def explainParams(): String
    Definition Classes
    Params
  26. final def extractParamMap(): ParamMap
    Definition Classes
    Params
  27. final def extractParamMap(extra: ParamMap): ParamMap
    Definition Classes
    Params
  28. val features: ArrayBuffer[Feature[_, _, _]]
    Definition Classes
    HasFeatures
  29. def finalize(): Unit
    Attributes
    protected[lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  30. final def fit(dataset: Dataset[_]): MultiClassifierDLModel
    Definition Classes
    AnnotatorApproach → Estimator
  31. def fit(dataset: Dataset[_], paramMaps: Array[ParamMap]): Seq[MultiClassifierDLModel]
    Definition Classes
    Estimator
    Annotations
    @Since( "2.0.0" )
  32. def fit(dataset: Dataset[_], paramMap: ParamMap): MultiClassifierDLModel
    Definition Classes
    Estimator
    Annotations
    @Since( "2.0.0" )
  33. def fit(dataset: Dataset[_], firstParamPair: ParamPair[_], otherParamPairs: ParamPair[_]*): MultiClassifierDLModel
    Definition Classes
    Estimator
    Annotations
    @Since( "2.0.0" ) @varargs()
  34. def get[T](feature: StructFeature[T]): Option[T]
    Attributes
    protected
    Definition Classes
    HasFeatures
  35. def get[K, V](feature: MapFeature[K, V]): Option[Map[K, V]]
    Attributes
    protected
    Definition Classes
    HasFeatures
  36. def get[T](feature: SetFeature[T]): Option[Set[T]]
    Attributes
    protected
    Definition Classes
    HasFeatures
  37. def get[T](feature: ArrayFeature[T]): Option[Array[T]]
    Attributes
    protected
    Definition Classes
    HasFeatures
  38. final def get[T](param: Param[T]): Option[T]
    Definition Classes
    Params
  39. def getBatchSize: Int

    Batch size (Default: 64)

  40. final def getClass(): Class[_]
    Definition Classes
    AnyRef → Any
    Annotations
    @native()
  41. def getConfigProtoBytes: Option[Array[Byte]]

    Tensorflow config Protobytes passed to the TF session

  42. final def getDefault[T](param: Param[T]): Option[T]
    Definition Classes
    Params
  43. def getEnableOutputLogs: Boolean

    Whether to output to annotators log folder (Default: false)

  44. def getInputCols: Array[String]

    returns

    input annotations columns currently used

    Definition Classes
    HasInputAnnotationCols
  45. def getLabelColumn: String

    Column with label per each document

  46. def getLazyAnnotator: Boolean
    Definition Classes
    CanBeLazy
  47. def getLr: Float

    Learning Rate (Default: 1e-3f)

  48. def getMaxEpochs: Int

    Maximum number of epochs to train (Default: 10)

  49. final def getOrDefault[T](param: Param[T]): T
    Definition Classes
    Params
  50. final def getOutputCol: String

    Gets annotation column name going to generate

    Gets annotation column name going to generate

    Definition Classes
    HasOutputAnnotationCol
  51. def getOutputLogsPath: String

    Folder path to save training logs (Default: "")

  52. def getParam(paramName: String): Param[Any]
    Definition Classes
    Params
  53. def getRandomSeed: Int

    Random seed

  54. def getShufflePerEpoch: Boolean

    Max sequence length to feed into TensorFlow

  55. def getThreshold: Float

    The minimum threshold for each label to be accepted (Default: 0.5f)

  56. def getValidationSplit: Float

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f).

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f). The value should be between 0.0 and 1.0 and by default it is 0.0 and off.

  57. final def hasDefault[T](param: Param[T]): Boolean
    Definition Classes
    Params
  58. def hasParam(paramName: String): Boolean
    Definition Classes
    Params
  59. def hashCode(): Int
    Definition Classes
    AnyRef → Any
    Annotations
    @native()
  60. def initializeLogIfNecessary(isInterpreter: Boolean, silent: Boolean): Boolean
    Attributes
    protected
    Definition Classes
    Logging
  61. def initializeLogIfNecessary(isInterpreter: Boolean): Unit
    Attributes
    protected
    Definition Classes
    Logging
  62. val inputAnnotatorTypes: Array[AnnotatorType]

    Input annotator type : SENTENCE_EMBEDDINGS

    Input annotator type : SENTENCE_EMBEDDINGS

    Definition Classes
    MultiClassifierDLApproachHasInputAnnotationCols
  63. final val inputCols: StringArrayParam

    columns that contain annotations necessary to run this annotator AnnotatorType is used both as input and output columns if not specified

    columns that contain annotations necessary to run this annotator AnnotatorType is used both as input and output columns if not specified

    Attributes
    protected
    Definition Classes
    HasInputAnnotationCols
  64. final def isDefined(param: Param[_]): Boolean
    Definition Classes
    Params
  65. final def isInstanceOf[T0]: Boolean
    Definition Classes
    Any
  66. final def isSet(param: Param[_]): Boolean
    Definition Classes
    Params
  67. def isTraceEnabled(): Boolean
    Attributes
    protected
    Definition Classes
    Logging
  68. val labelColumn: Param[String]

    Column with label per each document

  69. val lazyAnnotator: BooleanParam
    Definition Classes
    CanBeLazy
  70. def loadSavedModel(): TensorflowWrapper
  71. def log: Logger
    Attributes
    protected
    Definition Classes
    Logging
  72. def logDebug(msg: ⇒ String, throwable: Throwable): Unit
    Attributes
    protected
    Definition Classes
    Logging
  73. def logDebug(msg: ⇒ String): Unit
    Attributes
    protected
    Definition Classes
    Logging
  74. def logError(msg: ⇒ String, throwable: Throwable): Unit
    Attributes
    protected
    Definition Classes
    Logging
  75. def logError(msg: ⇒ String): Unit
    Attributes
    protected
    Definition Classes
    Logging
  76. def logInfo(msg: ⇒ String, throwable: Throwable): Unit
    Attributes
    protected
    Definition Classes
    Logging
  77. def logInfo(msg: ⇒ String): Unit
    Attributes
    protected
    Definition Classes
    Logging
  78. def logName: String
    Attributes
    protected
    Definition Classes
    Logging
  79. def logTrace(msg: ⇒ String, throwable: Throwable): Unit
    Attributes
    protected
    Definition Classes
    Logging
  80. def logTrace(msg: ⇒ String): Unit
    Attributes
    protected
    Definition Classes
    Logging
  81. def logWarning(msg: ⇒ String, throwable: Throwable): Unit
    Attributes
    protected
    Definition Classes
    Logging
  82. def logWarning(msg: ⇒ String): Unit
    Attributes
    protected
    Definition Classes
    Logging
  83. val lr: FloatParam

    Learning Rate (Default: 1e-3f)

  84. val maxEpochs: IntParam

    Maximum number of epochs to train (Default: 10)

  85. def msgHelper(schema: StructType): String
    Attributes
    protected
    Definition Classes
    HasInputAnnotationCols
  86. final def ne(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  87. final def notify(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native()
  88. final def notifyAll(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native()
  89. def onTrained(model: MultiClassifierDLModel, spark: SparkSession): Unit
    Definition Classes
    AnnotatorApproach
  90. def onWrite(path: String, spark: SparkSession): Unit
    Attributes
    protected
    Definition Classes
    ParamsAndFeaturesWritable
  91. val optionalInputAnnotatorTypes: Array[String]
    Definition Classes
    HasInputAnnotationCols
  92. val outputAnnotatorType: String

    Output annotator type : CATEGORY

    Output annotator type : CATEGORY

    Definition Classes
    MultiClassifierDLApproachHasOutputAnnotatorType
  93. final val outputCol: Param[String]
    Attributes
    protected
    Definition Classes
    HasOutputAnnotationCol
  94. val outputLogsPath: Param[String]

    Folder path to save training logs (Default: "")

  95. lazy val params: Array[Param[_]]
    Definition Classes
    Params
  96. val randomSeed: IntParam

    Random seed for shuffling the dataset (Default: 44)

  97. def save(path: String): Unit
    Definition Classes
    MLWritable
    Annotations
    @Since( "1.6.0" ) @throws( ... )
  98. def set[T](feature: StructFeature[T], value: T): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  99. def set[K, V](feature: MapFeature[K, V], value: Map[K, V]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  100. def set[T](feature: SetFeature[T], value: Set[T]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  101. def set[T](feature: ArrayFeature[T], value: Array[T]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  102. final def set(paramPair: ParamPair[_]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    Params
  103. final def set(param: String, value: Any): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    Params
  104. final def set[T](param: Param[T], value: T): MultiClassifierDLApproach.this.type
    Definition Classes
    Params
  105. def setBatchSize(batch: Int): MultiClassifierDLApproach.this.type

    Batch size (Default: 64)

  106. def setConfigProtoBytes(bytes: Array[Int]): MultiClassifierDLApproach.this.type

    Tensorflow config Protobytes passed to the TF session

  107. def setDefault[T](feature: StructFeature[T], value: () ⇒ T): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  108. def setDefault[K, V](feature: MapFeature[K, V], value: () ⇒ Map[K, V]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  109. def setDefault[T](feature: SetFeature[T], value: () ⇒ Set[T]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  110. def setDefault[T](feature: ArrayFeature[T], value: () ⇒ Array[T]): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    HasFeatures
  111. final def setDefault(paramPairs: ParamPair[_]*): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    Params
  112. final def setDefault[T](param: Param[T], value: T): MultiClassifierDLApproach.this.type
    Attributes
    protected
    Definition Classes
    Params
  113. def setEnableOutputLogs(enableOutputLogs: Boolean): MultiClassifierDLApproach.this.type

    Whether to output to annotators log folder (Default: false)

  114. final def setInputCols(value: String*): MultiClassifierDLApproach.this.type
    Definition Classes
    HasInputAnnotationCols
  115. final def setInputCols(value: Array[String]): MultiClassifierDLApproach.this.type

    Overrides required annotators column if different than default

    Overrides required annotators column if different than default

    Definition Classes
    HasInputAnnotationCols
  116. def setLabelColumn(column: String): MultiClassifierDLApproach.this.type

    Column with label per each document

  117. def setLazyAnnotator(value: Boolean): MultiClassifierDLApproach.this.type
    Definition Classes
    CanBeLazy
  118. def setLr(lr: Float): MultiClassifierDLApproach.this.type

    Learning Rate (Default: 1e-3f)

  119. def setMaxEpochs(epochs: Int): MultiClassifierDLApproach.this.type

    Maximum number of epochs to train (Default: 10)

  120. final def setOutputCol(value: String): MultiClassifierDLApproach.this.type

    Overrides annotation column name when transforming

    Overrides annotation column name when transforming

    Definition Classes
    HasOutputAnnotationCol
  121. def setOutputLogsPath(path: String): MultiClassifierDLApproach.this.type

    Folder path to save training logs (Default: "")

  122. def setRandomSeed(seed: Int): MultiClassifierDLApproach.this.type

    Random seed

  123. def setShufflePerEpoch(value: Boolean): MultiClassifierDLApproach.this.type

    shufflePerEpoch

  124. def setThreshold(threshold: Float): MultiClassifierDLApproach.this.type

    The minimum threshold for each label to be accepted (Default: 0.5f)

  125. def setValidationSplit(validationSplit: Float): MultiClassifierDLApproach.this.type

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f).

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f). The value should be between 0.0 and 1.0 and by default it is 0.0 and off.

  126. def setVerbose(verbose: Level): MultiClassifierDLApproach.this.type

    Level of verbosity during training (Default: Verbose.Silent.id)

  127. def setVerbose(verbose: Int): MultiClassifierDLApproach.this.type

    Level of verbosity during training (Default: Verbose.Silent.id)

  128. val shufflePerEpoch: BooleanParam

    Whether to shuffle the training data on each Epoch (Default: false)

  129. final def synchronized[T0](arg0: ⇒ T0): T0
    Definition Classes
    AnyRef
  130. val threshold: FloatParam

    The minimum threshold for each label to be accepted (Default: 0.5f)

  131. def toString(): String
    Definition Classes
    Identifiable → AnyRef → Any
  132. def train(dataset: Dataset[_], recursivePipeline: Option[PipelineModel]): MultiClassifierDLModel
  133. final def transformSchema(schema: StructType): StructType

    requirement for pipeline transformation validation.

    requirement for pipeline transformation validation. It is called on fit()

    Definition Classes
    AnnotatorApproach → PipelineStage
  134. def transformSchema(schema: StructType, logging: Boolean): StructType
    Attributes
    protected
    Definition Classes
    PipelineStage
    Annotations
    @DeveloperApi()
  135. val uid: String
    Definition Classes
    MultiClassifierDLApproach → Identifiable
  136. def validate(schema: StructType): Boolean

    takes a Dataset and checks to see if all the required annotation types are present.

    takes a Dataset and checks to see if all the required annotation types are present.

    schema

    to be validated

    returns

    True if all the required types are present, else false

    Attributes
    protected
    Definition Classes
    AnnotatorApproach
  137. val validationSplit: FloatParam

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f).

    Choose the proportion of training dataset to be validated against the model on each Epoch (Default: 0.0f). The value should be between 0.0 and 1.0 and by default it is 0.0 and off.

  138. val verbose: IntParam

    Level of verbosity during training (Default: Verbose.Silent.id)

  139. final def wait(): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  140. final def wait(arg0: Long, arg1: Int): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  141. final def wait(arg0: Long): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... ) @native()
  142. def write: MLWriter
    Definition Classes
    ParamsAndFeaturesWritable → DefaultParamsWritable → MLWritable

Inherited from ParamsAndFeaturesWritable

Inherited from HasFeatures

Inherited from CanBeLazy

Inherited from DefaultParamsWritable

Inherited from MLWritable

Inherited from HasOutputAnnotatorType

Inherited from HasOutputAnnotationCol

Inherited from HasInputAnnotationCols

Inherited from Estimator[MultiClassifierDLModel]

Inherited from PipelineStage

Inherited from Logging

Inherited from Params

Inherited from Serializable

Inherited from Serializable

Inherited from Identifiable

Inherited from AnyRef

Inherited from Any

Parameters

A list of (hyper-)parameter keys this annotator can take. Users can set and get the parameter values through setters and getters, respectively.

Annotator types

Required input and expected output annotator types

Members

Parameter setters

Parameter getters