你当前正在访问 Microsoft Azure Global Edition 技术文档网站。 如果需要访问由世纪互联运营的 Microsoft Azure 中国技术文档网站,请访问 https://docs.azure.cn

ArmMachineLearningModelFactory.TextClassificationMultilabel 方法

定义

初始化 TextClassificationMultilabel 的新实例。

public static Azure.ResourceManager.MachineLearning.Models.TextClassificationMultilabel TextClassificationMultilabel (Azure.ResourceManager.MachineLearning.Models.MachineLearningLogVerbosity? logVerbosity = default, string targetColumnName = default, Azure.ResourceManager.MachineLearning.Models.MachineLearningTableJobInput trainingData = default, Azure.ResourceManager.MachineLearning.Models.ClassificationMultilabelPrimaryMetric? primaryMetric = default, string featurizationDatasetLanguage = default, Azure.ResourceManager.MachineLearning.Models.NlpFixedParameters fixedParameters = default, Azure.ResourceManager.MachineLearning.Models.NlpVerticalLimitSettings limitSettings = default, System.Collections.Generic.IEnumerable<Azure.ResourceManager.MachineLearning.Models.NlpParameterSubspace> searchSpace = default, Azure.ResourceManager.MachineLearning.Models.NlpSweepSettings sweepSettings = default, Azure.ResourceManager.MachineLearning.Models.MachineLearningTableJobInput validationData = default);
static member TextClassificationMultilabel : Nullable<Azure.ResourceManager.MachineLearning.Models.MachineLearningLogVerbosity> * string * Azure.ResourceManager.MachineLearning.Models.MachineLearningTableJobInput * Nullable<Azure.ResourceManager.MachineLearning.Models.ClassificationMultilabelPrimaryMetric> * string * Azure.ResourceManager.MachineLearning.Models.NlpFixedParameters * Azure.ResourceManager.MachineLearning.Models.NlpVerticalLimitSettings * seq<Azure.ResourceManager.MachineLearning.Models.NlpParameterSubspace> * Azure.ResourceManager.MachineLearning.Models.NlpSweepSettings * Azure.ResourceManager.MachineLearning.Models.MachineLearningTableJobInput -> Azure.ResourceManager.MachineLearning.Models.TextClassificationMultilabel
Public Shared Function TextClassificationMultilabel (Optional logVerbosity As Nullable(Of MachineLearningLogVerbosity) = Nothing, Optional targetColumnName As String = Nothing, Optional trainingData As MachineLearningTableJobInput = Nothing, Optional primaryMetric As Nullable(Of ClassificationMultilabelPrimaryMetric) = Nothing, Optional featurizationDatasetLanguage As String = Nothing, Optional fixedParameters As NlpFixedParameters = Nothing, Optional limitSettings As NlpVerticalLimitSettings = Nothing, Optional searchSpace As IEnumerable(Of NlpParameterSubspace) = Nothing, Optional sweepSettings As NlpSweepSettings = Nothing, Optional validationData As MachineLearningTableJobInput = Nothing) As TextClassificationMultilabel

参数

logVerbosity
Nullable<MachineLearningLogVerbosity>

记录作业的详细程度。

targetColumnName
String

目标列名称:这是预测值列。 在分类任务上下文中也称为标签列名称。

trainingData
MachineLearningTableJobInput

[必需]训练数据输入。

primaryMetric
Nullable<ClassificationMultilabelPrimaryMetric>

文本分类多标签任务的主要指标。 目前,仅支持将“准确性”作为主要指标,因此用户无需显式设置它。

featurizationDatasetLanguage
String

AutoML 作业所需的特征化输入。

fixedParameters
NlpFixedParameters

在整个训练过程中保持不变的模型/训练参数。

limitSettings
NlpVerticalLimitSettings

AutoMLJob 的执行约束。

searchSpace
IEnumerable<NlpParameterSubspace>

搜索空间以对模型及其超参数的不同组合进行采样。

sweepSettings
NlpSweepSettings

用于模型扫描和超参数优化的设置。

validationData
MachineLearningTableJobInput

验证数据输入。

返回

用于模拟的新 TextClassificationMultilabel 实例。

适用于