Google Cloud ML Engine

Getting started

To use Google Cloud ML Engine, you need to deploy your model first! Deployment might depend on the type of model you are using:

• TensorFlow.
• scikit-learn or XGBoost.

Usage

Include the following dependency:

Maven

<dependency>
  <groupId>com.spotify</groupId>
  <artifactId>zoltar-mlengine</artifactId>
  <version>0.4.0</version>
</dependency>

Gradle

dependencies {
  compile group: 'com.spotify', name: 'zoltar-mlengine', version: '0.4.0'
}

sbt

libraryDependencies += "com.spotify" % "zoltar-mlengine" % "0.4.0"

Replace your existing ModelLoader with MlEngineLoader

final MlEngineLoader mlEngineLoader =
    Models.mlEngine(projectId, modelId, versionId);

Example

Follow this example to create a Predictor that uses Google Cloud ML Engine deployed models.