The MLflow Java client tests require that MLflow is on the PATH (to start a local server), so it is recommended to run them from within a development conda environment.

To build a deployable JAR and run tests:

mvn package

Run

To run a simple sample.

java -cp target/mlflow-java-client-0.4.2.jar \
  com.databricks.mlflow.client.samples.QuickStartDriver http://localhost:5001

JSON Serialization

MLflow Java client uses Protobuf 3.6.0 to serialize the JSON payload.

service.proto - Protobuf definition of data objects.
com.databricks.api.proto.mlflow.Service.java - Generated Java classes of all data objects.
generate_protos.py - One time script to generate Service.java. If service.proto changes you will need to re-run this script.
Javadoc can be generated by running mvn javadoc:javadoc. The output will be in target/site/apidocs/index.html. Here is the javadoc for Service.java.

Java Client API

See ApiClient.java and Service.java domain objects.

Run getRun(String runId)
RunInfo createRun()
RunInfo createRun(String experimentId)
RunInfo createRun(String experimentId, String appName)
RunInfo createRun(CreateRun request)
List<RunInfo> listRunInfos(String experimentId)


List<Experiment> searchExperiments()
GetExperiment.Response getExperiment(String experimentId)
Optional<Experiment> getExperimentByName(String experimentName)
long createExperiment(String experimentName)

void logParam(String runId, String key, String value)
void logMetric(String runId, String key, float value)
void setTerminated(String runId)
void setTerminated(String runId, RunStatus status)
void setTerminated(String runId, RunStatus status, long endTime)
ListArtifacts.Response listArtifacts(String runId, String path)

Usage

Java Usage

For a simple example see QuickStartDriver.java. For full examples of API coverage see the tests such as MlflowClientTest.java.

package org.mlflow.tracking.samples;

import java.util.List;
import java.util.Optional;

import org.apache.log4j.Level;
import org.apache.log4j.LogManager;

import org.mlflow.api.proto.Service.*;
import org.mlflow.tracking.MlflowClient;

/**
 * This is an example application which uses the MLflow Tracking API to create and manage
 * experiments and runs.
 */
public class QuickStartDriver {
  public static void main(String[] args) throws Exception {
    (new QuickStartDriver()).process(args);
  }

  void process(String[] args) throws Exception {
    MlflowClient client;
    if (args.length < 1) {
      client = new MlflowClient();
    } else {
      client = new MlflowClient(args[0]);
    }

    boolean verbose = args.length >= 2 && "true".equals(args[1]);
    if (verbose) {
      LogManager.getLogger("org.mlflow.client").setLevel(Level.DEBUG);
    }

    System.out.println("====== createExperiment");
    String expName = "Exp_" + System.currentTimeMillis();
    String expId = client.createExperiment(expName);
    System.out.println("createExperiment: expId=" + expId);

    System.out.println("====== getExperiment");
    GetExperiment.Response exp = client.getExperiment(expId);
    System.out.println("getExperiment: " + exp);

    System.out.println("====== searchExperiments");
    List<Experiment> exps = client.searchExperiments();
    System.out.println("#experiments: " + exps.size());
    exps.forEach(e -> System.out.println("  Exp: " + e));

    createRun(client, expId);

    System.out.println("====== getExperiment again");
    GetExperiment.Response exp2 = client.getExperiment(expId);
    System.out.println("getExperiment: " + exp2);

    System.out.println("====== getExperiment by name");
    Optional<Experiment> exp3 = client.getExperimentByName(expName);
    System.out.println("getExperimentByName: " + exp3);
  }

  void createRun(MlflowClient client, String expId) {
    System.out.println("====== createRun");

    // Create run
    String sourceFile = "MyFile.java";
    RunInfo runCreated = client.createRun(expId, sourceFile);
    System.out.println("CreateRun: " + runCreated);
    String runId = runCreated.getRunUuid();

    // Log parameters
    client.logParam(runId, "min_samples_leaf", "2");
    client.logParam(runId, "max_depth", "3");

    // Log metrics
    client.logMetric(runId, "auc", 2.12F);
    client.logMetric(runId, "accuracy_score", 3.12F);
    client.logMetric(runId, "zero_one_loss", 4.12F);

    // Update finished run
    client.setTerminated(runId, RunStatus.FINISHED);

    // Get run details
    Run run = client.getRun(runId);
    System.out.println("GetRun: " + run);
    client.close();
  }
}