Predict.java

package water.api;

import hex.glm.GLMModel;
import water.*;
import water.fvec.Frame;
import water.fvec.Vec;
import water.util.RString;

public class Predict extends Request2 {
  static final int API_WEAVER = 1; // This file has auto-gen'd doc & json fields
  static public DocGen.FieldDoc[] DOC_FIELDS; // Initialized from Auto-Gen code.

  @API(help = "Model", required = true, filter = Default.class)
  public Model model; // Type to Model when retired OldModel

  @API(help = "Data frame", required = true, filter = Default.class)
  public Frame data;

  @API(help = "Prediction", filter = Default.class)
  public Key prediction;

  public static String link(Key k, String content) {
    RString rs = new RString("<a href='Predict.query?model=%$key'>%content</a>");
    rs.replace("key", k.toString());
    rs.replace("content", content);
    return rs.toString();
  }

  @Override protected Response serve() {
    Frame fr = null;
    try {
      if( model == null )
        throw new IllegalArgumentException("Model is required to perform validation!");
      // Create a new random key
      if ( prediction == null )
        prediction = Key.make("__Prediction_" + Key.make());
      fr = new Frame(prediction,new String[0],new Vec[0]).delete_and_lock(null);
      fr = model.score(data);
      fr = new Frame(prediction,fr._names,fr.vecs()); // Jam in the frame key
      return Inspect2.redirect(this, prediction.toString());
    } catch( Throwable t ) {
      return Response.error(t);
    } finally {
      if( fr != null ) fr.unlock(null);
    }
  }
}