Colab notebooks

This page links to some Colaboratory notebooks hosted in GitHub that walk you through some common AutoML Tables usage scenarios.

How to use notebooks

To use the Colaboratory notebooks, you copy the notebook to your own Google Drive and open it with Colaboratory (or Colab). You can run each step, or cell, and see its results. To run a cell, use Shift+Enter. Colab automatically displays the return value of the last line in each cell. For more information about running notebooks in Colab, see the Colab welcome page.

Runtime environment

For easiest setup, you can run a Colab notebook on a hosted runtime in the Cloud. When you do this, the hosted VM times out after 90 minutes of inactivity. Because some steps in the AutoML Tables process can take a few hours, your session will timeout while you wait for those steps to complete. When you restart after the timeout, you must repeat the initialization and authentication steps, and then continue the notebook from where you left off. You might need to copy the values of some variables, such as the dataset name, from the printed output of previous cells.

Alternatively, you can run the notebook in a local runtime environment. For instructions, see Local runtimes.

If your session gets disconnected before the 90-minute timeout (for example, if you close your laptop), click RECONNECT and resume the session.

Before you begin

Before you can run a AutoML Tables notebook, you must enable AutoML Tables for your Google Cloud project as described in Before you begin.

After you finish

Make sure you undeploy any models you deploy for your notebooks when you are done to avoid model deployment charges.

AutoML Tables notebooks

  • Getting Started Notebook

    Train a binary classification model to predict whether a person's income is above or below a threshold.

  • Purchase Prediction

    Train a binary classification model to perform purchase predictions.

  • Result Slicing

    Use open source tools to slice and analyze results from a classification model.

  • Music Recommendation

    Train a binary classification model to predict user-song similarity and produce recommendations.