ImageClassificationMultilabel Class

Definition

Namespace:

Azure.ResourceManager.MachineLearning.Models

Assembly:

Azure.ResourceManager.MachineLearning.dll

Package:

Azure.ResourceManager.MachineLearning v1.1.1

Source:

ImageClassificationMultilabel.cs

Image Classification Multilabel. Multi-label image classification is used when an image could have one or more labels from a set of labels - e.g. an image could be labeled with both 'cat' and 'dog'.

public class ImageClassificationMultilabel : Azure.ResourceManager.MachineLearning.Models.AutoMLVertical

type ImageClassificationMultilabel = class
    inherit AutoMLVertical

Public Class ImageClassificationMultilabel
Inherits AutoMLVertical

Inheritance

Object

AutoMLVertical

ImageClassificationMultilabel

Constructors

ImageClassificationMultilabel(MachineLearningTableJobInput, ImageLimitSettings)

Initializes a new instance of ImageClassificationMultilabel.

Properties

LimitSettings	[Required] Limit settings for the AutoML job.
LogVerbosity	Log verbosity for the job. (Inherited from AutoMLVertical)
ModelSettings	Settings used for training the model.
PrimaryMetric	Primary metric to optimize for this task.
SearchSpace	Search space for sampling different combinations of models and their hyperparameters.
SweepSettings	Model sweeping and hyperparameter sweeping related settings.
TargetColumnName	Target column name: This is prediction values column. Also known as label column name in context of classification tasks. (Inherited from AutoMLVertical)
TrainingData	[Required] Training data input. (Inherited from AutoMLVertical)
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.