title | titleSuffix | description | services | ms.service | ms.subservice | ms.topic | ms.author | author | ms.reviewer | ms.date |
---|---|---|---|---|---|---|---|---|---|---|
AutoML Text Multi-label Classification |
Azure Machine Learning |
Learn how to use the AutoML Text Multi-label Classification component in Azure Machine Learning to create a classifier using ML Table data. |
machine-learning |
machine-learning |
core |
reference |
rasavage |
rsavage2 |
ssalgadodev |
12/1/2022 |
This article describes a component in Azure Machine Learning designer.
Use this component to create a machine learning model that is based on the AutoML Text Multi-label Classification.
Multi-label text classification is for use cases where each example may be assigned more than one label, as opposed to single-label multiclass text classification where every example is labeled with the single most probable class.
This component trains an NLP classification model on text data. Text classification is a supervised learning task and requires a labeled dataset that includes a label column with a value for all rows.
This model requires a training and a validation dataset. The datasets must be in ML Table format.
-
Add the AutoML Text Multi-label Classification component to your pipeline.
-
Specify the Target Column you want the model to output
-
Specify the Primary Metric you want AutoML to use to measure your model's success.
-
(Optional) Select the language your dataset consists of. Visit this link for a full list of supported languages.
-
(Optional) You are able to configure Hyperparameters. Visit this link for a full list of configurable Hyperparameters
-
(Optional) Job Sweep settings are configurable. Visit this link to learn more about each configurable parameter.
-
(Optional) Job Limit settings are configurable. Visit this link to learn more about these settings.
See the set of components available to Azure Machine Learning.