TableVertical interface

Abstract class for AutoML tasks that use table dataset as input - such as Classification/Regression/Forecasting.

Properties

cvSplitColumnNames

Columns to use for CVSplit data.

featurizationSettings

Featurization inputs needed for AutoML job.

limitSettings

Execution constraints for AutoMLJob.

nCrossValidations

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

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.

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

cvSplitColumnNames

Columns to use for CVSplit data.

cvSplitColumnNames?: string[]

Property Value

string[]

featurizationSettings

Featurization inputs needed for AutoML job.

featurizationSettings?: TableVerticalFeaturizationSettings

Property Value

limitSettings

Execution constraints for AutoMLJob.

limitSettings?: TableVerticalLimitSettings

Property Value

nCrossValidations

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

nCrossValidations?: NCrossValidationsUnion

Property Value

testData

Test data input.

testData?: MLTableJobInput

Property Value

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

validationData

Validation data inputs.

validationData?: MLTableJobInput

Property Value

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

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