Although we can’t reliably compare its future-predicting abilities to a crystal ball, the Google Prediction API unlocks a powerful mechanism to use machine learning in your applications.
The Prediction API allows developers to train their own predictive models, taking advantage of Google’s world-class machine learning algorithms. It can be used for all sorts of classification and recommendation problems from spam detection to message routing decisions. In the latest release, the Prediction API has added more detailed debugging information on trained models and a new App Engine sample, which illustrates how to use the Google Prediction API for the Java and Python runtimes.
To help App Engine developers get started with the prediction API, we’ve published an article and walkthrough detailing how to create and manage predictive models in App Engine apps with simple authentication using OAuth2 and service accounts. Check out the walkthrough and let us know what you think on the group. Happy coding!
- Posted by the App Engine Team