Share via

Classification interface

  • Reference
Package:
@azure/arm-machinelearning

Classification task in AutoML Table vertical.

Extends

TableVertical,AutoMLVertical

Properties

positiveLabel

Positive label for binary metrics calculation.
primaryMetric

Primary metric for the task.
taskType

Polymorphic discriminator, which specifies the different types this object can be
trainingSettings

Inputs for training phase for an AutoML Job.

Inherited Properties

cvSplitColumnNames

Columns to use for CVSplit data.
featurizationSettings

Featurization inputs needed for AutoML job.
limitSettings

Execution constraints for AutoMLJob.
logVerbosity

Log verbosity for the job.
nCrossValidations

Number of cross validation folds to be applied on training dataset when validation dataset is not provided.
targetColumnName

Target column name: This is prediction values column. Also known as label column name in context of classification tasks.
testData

Test data input.
testDataSize

The fraction of test dataset that needs to be set aside for validation purpose. Values between (0.0 , 1.0) Applied when validation dataset is not provided.
trainingData

[Required] Training data input.
validationData

Validation data inputs.
validationDataSize

The fraction of training dataset that needs to be set aside for validation purpose. Values between (0.0 , 1.0) Applied when validation dataset is not provided.
weightColumnName

The name of the sample weight column. Automated ML supports a weighted column as an input, causing rows in the data to be weighted up or down.

Property Details

positiveLabel

Positive label for binary metrics calculation.

positiveLabel?: string

Property Value

string

primaryMetric

Primary metric for the task.

primaryMetric?: string

Property Value

string

taskType

Polymorphic discriminator, which specifies the different types this object can be

taskType: "Classification"

Property Value

"Classification"

trainingSettings

Inputs for training phase for an AutoML Job.

trainingSettings?: ClassificationTrainingSettings

Property Value

ClassificationTrainingSettings

Inherited Property Details

cvSplitColumnNames

Columns to use for CVSplit data.

cvSplitColumnNames?: string[]

Property Value

string[]

Inherited From TableVertical.cvSplitColumnNames

featurizationSettings

Featurization inputs needed for AutoML job.

featurizationSettings?: TableVerticalFeaturizationSettings

Property Value

TableVerticalFeaturizationSettings

Inherited From TableVertical.featurizationSettings

limitSettings

Execution constraints for AutoMLJob.

limitSettings?: TableVerticalLimitSettings

Property Value

TableVerticalLimitSettings

Inherited From TableVertical.limitSettings

logVerbosity

Log verbosity for the job.

logVerbosity?: string

Property Value

string

Inherited From AutoMLVertical.logVerbosity

nCrossValidations

Number of cross validation folds to be applied on training dataset when validation dataset is not provided.

nCrossValidations?: NCrossValidationsUnion

Property Value

NCrossValidationsUnion

Inherited From TableVertical.nCrossValidations

targetColumnName

Target column name: This is prediction values column. Also known as label column name in context of classification tasks.

targetColumnName?: string

Property Value

string

Inherited From AutoMLVertical.targetColumnName

testData

Test data input.

testData?: MLTableJobInput

Property Value

MLTableJobInput

Inherited From TableVertical.testData

testDataSize

The fraction of test dataset that needs to be set aside for validation purpose. Values between (0.0 , 1.0) Applied when validation dataset is not provided.

testDataSize?: number

Property Value

number

Inherited From TableVertical.testDataSize

trainingData

[Required] Training data input.

trainingData: MLTableJobInput

Property Value

MLTableJobInput

Inherited From AutoMLVertical.trainingData

validationData

Validation data inputs.

validationData?: MLTableJobInput

Property Value

MLTableJobInput

Inherited From TableVertical.validationData

validationDataSize

The fraction of training dataset that needs to be set aside for validation purpose. Values between (0.0 , 1.0) Applied when validation dataset is not provided.

validationDataSize?: number

Property Value

number

Inherited From TableVertical.validationDataSize

weightColumnName

The name of the sample weight column. Automated ML supports a weighted column as an input, causing rows in the data to be weighted up or down.

weightColumnName?: string

Property Value

string

Inherited From TableVertical.weightColumnName