The AutoML TrainingPipeline
resource orchestrates tasks associated
with training an AutoML model. This resource always executes the
training task, and optionally may also export data from a Vertex AI
Dataset
which becomes the training input, upload the Model to
Vertex AI, and evaluate the Model. For information about
AutoML training in Vertex AI, see the
AutoML training documentation. For information
about Google Cloud Pipeline Components related to datasets, see
Dataset components.
The Google Cloud Pipeline Components SDK includes the following operators related to AutoML models and workflows:
Operators related to AutoML forecasting
Operators related to AutoML Tabular models
-
CvTrainerOp
-
EnsembleOp
-
FinalizerOp
-
InfraValidatorOp
-
SplitMaterializedDataOp
-
Stage1TunerOp
-
StatsAndExampleGenOp
-
TrainingConfiguratorAndValidatorOp
-
TransformOp
Operators related to AutoML model
resource creation
-
AutoMLForecastingTrainingJobRunOp
-
AutoMLImageTrainingJobRunOp
-
AutoMLTabularTrainingJobRunOp
-
AutoMLTextTrainingJobRunOp
-
AutoMLVideoTrainingJobRunOp
Learn more about training and using your own AutoML models.
API reference
For AutoML component reference, see the Google Cloud Pipeline Components SDK reference for AutoML components.
For Vertex AI API reference, see the following API reference pages:
Tutorials
- Learn how to use the Google Cloud pipeline components to train an image classification model using Vertex AI AutoML.
- Learn how to use the Google Cloud pipeline components to train a classification model using tabular data and Vertex AI AutoML.
- Learn how to use the Google Cloud pipeline components to train a linear regression model using tabular data and Vertex AI AutoML.
- Learn how to use the Google Cloud pipeline components to train a text classification model using Vertex AI AutoML.
- Learn how to use the Google Cloud pipeline components to upload and deploy a model.
Version history and release notes
To learn more about the version history and changes to the Google Cloud Pipeline Components SDK, see the Google Cloud Pipeline Components SDK Release Notes.
Technical support contacts
If you have any questions, reach out to [email protected].