AI Platform (Unified) brings AutoML and AI Platform (Classic) together into a unified API, client library, and user interface. With AI Platform (Unified), both AutoML training and custom training are available options. Whichever option you choose for training, you can save models, deploy models and request predictions with AI Platform (Unified). Learn more.

Tutorials

Getting started

Resources

Training and tutorials

Try AI Platform tutorials, training courses, and Qwiklabs from Google Cloud.

Tutorial

Hello image data

If you're new to artificial intelligence (AI), start with image classification on AutoML. This tutorial shows you how to train a model using a pre-prepared dataset and AutoML's codeless interface. After training your model, use this interface to send prediction requests to the model.

Learn more  
Tutorial

Hello custom training

If you're planning to code your own training app, use this tutorial to see how you can bring your training code to AI Platform. The tutorial provides a sample app for an image classification model, and leads you through the process of running the app on AI Platform. After training your model, use a sample web app to send prediction requests to the model.

Learn more  
Tutorial

Hello structured data

If your data is in a database, a spreadsheet, or some other structured form, use this tutorial to explore and prepare your data on AI Platform, then train your model. The tutorial provides a dataset that you can experiment with, and shows you how to train a model with the AutoML codeless training service and request predictions using the Google Cloud Console.

Learn more  

Use cases

Explore use cases, reference architectures, whitepapers, best practices, and industry solutions.

Article

Considerations for Sensitive Data within Machine Learning Datasets

Discusses strategies for identifying and protecting sensitive data while training machine learning (ML) models.

security privacy

Learn more  
Architecture

Preparing ML-ready data for personalization

Provides a set of Dataflow pipelines that you can run to create machine learning (ML) datasets.

BigQuery Dataflow Cloud Storage Google Analytics 360 CRM

Learn more  

Code samples

Dive into coding with examples that demonstrate how to use and connect Google Cloud services.

github

Python client library samples

See samples that use the Python client library for AI Platform (Unified).

Python AI Platform (Unified)

Open GitHub  

github

Java client library samples

See samples that use the Java client library for AI Platform (Unified).

Java AI Platform (Unified)

Open GitHub  

github

Node.js client library samples

See samples that use the Node.js client library for AI Platform (Unified).

Node.js JavaScript AI Platform (Unified)

Open GitHub