Forecasting with AutoML

To see an example of how to create, train, and use an AutoML time-series forecasting model for batch prediction, run the "AutoML tabular forecasting model for batch prediction" Jupyter notebook in one of the following environments:

Open in Colab | Open in Colab Enterprise | Open in Vertex AI Workbench user-managed notebooks | View on GitHub

Forecasting models predict a sequence of values. For example, as a retailer, you might want to forecast daily demand of your products for the next 3 months so that you can appropriately stock product inventories in advance.

Workflow for creating a forecast model and making predictions

The process for creating a forecast model in Vertex AI is as follows:

Steps	Description
1. Prepare tabular training data for forecast models	Prepare your tabular training data for forecast model training.
2. Create a dataset for training forecast models	Create a new dataset and associate your prepared training data to it.
3. Train a forecast model	Train a forecast model in Vertex AI using your dataset.
4. Evaluate your model	Evaluate your newly trained forecast model for prediction accuracy.
5. Get predictions for a forecast model	Request batch predictions from your forecast model.

Forecasting with AutoML does not support online predictions. If you want to request online predictions from your forecast model, use Tabular Workflow for Forecasting.