Describes the types of cloud prediction offered by AI Platform, and what's required to use them.
Exporting models for prediction
Describes how to export machine learning models as artifacts that you can deploy to AI Platform Prediction.
Describes how to deploy your trained model to the cloud in order to use it to request predictions.
Custom prediction routines
Describes how you can customize what code runs when AI Platform Prediction responds to online prediction requests.
Choosing a machine type for online prediction
Describes how to use the virtual machine types and GPUS available for online prediction.
Getting online predictions
Describes how to request online predictions from models hosted with AI Platform.
Exporting a TensorFlow SavedModel for prediction
Describes how to export a TensorFlow SavedModel that you can deploy to AI Platform Prediction.
Getting batch predictions
Describes how to start batch prediction jobs on AI Platform.
Online versus batch prediction
Compares getting online and batch predictions from AI Platform Prediction.