ai.h2o.sparkling.ml.params.H2OAutoMLStoppingCriteriaParams.scala Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of sparkling-water-ml_2.12 Show documentation
Sparkling Water ML Pipelines
The newest version!
/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package ai.h2o.sparkling.ml.params

import ai.h2o.automl.AutoMLBuildSpec.AutoMLStoppingCriteria
import ai.h2o.sparkling.H2OFrame
import hex.ScoreKeeper.StoppingMetric

trait H2OAutoMLStoppingCriteriaParams
  extends H2OAlgoParamsBase {


  //
  // Parameter definitions
  //
  protected val seed = longParam(
    name = "seed",
    doc = """Seed for random number generator; set to a value other than -1 for reproducibility.""")

  protected val maxModels = intParam(
    name = "maxModels",
    doc = """Maximum number of models to build (optional). Always set this parameter to ensure AutoML reproducibility: all models are then trained until convergence and none is constrained by a time budget.""")

  protected val maxRuntimeSecs = doubleParam(
    name = "maxRuntimeSecs",
    doc = """This argument specifies the maximum time that the AutoML process will run for. If both max_runtime_secs and max_models are specified, then the AutoML run will stop as soon as it hits either of these limits. If neither max_runtime_secs nor max_models are specified, then max_runtime_secs defaults to 3600 seconds (1 hour).""")

  protected val maxRuntimeSecsPerModel = doubleParam(
    name = "maxRuntimeSecsPerModel",
    doc = """Maximum time to spend on each individual model (optional). Note that models constrained by a time budget are not guaranteed reproducible.""")

  protected val stoppingRounds = intParam(
    name = "stoppingRounds",
    doc = """Early stopping based on convergence of stopping_metric. Stop if simple moving average of length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0 to disable).""")

  protected val stoppingMetric = stringParam(
    name = "stoppingMetric",
    doc = """Metric to use for early stopping (AUTO: logloss for classification, deviance for regression). Possible values are ``"AUTO"``, ``"deviance"``, ``"logloss"``, ``"MSE"``, ``"RMSE"``, ``"MAE"``, ``"RMSLE"``, ``"AUC"``, ``"AUCPR"``, ``"lift_top_group"``, ``"misclassification"``, ``"mean_per_class_error"``, ``"anomaly_score"``, ``"AUUC"``, ``"ATE"``, ``"ATT"``, ``"ATC"``, ``"qini"``, ``"custom"``, ``"custom_increasing"``.""")

  protected val stoppingTolerance = doubleParam(
    name = "stoppingTolerance",
    doc = """Relative tolerance for metric-based stopping criterion (stop if relative improvement is not at least this much).""")

  //
  // Default values
  //
  setDefault(
    seed -> -1L,
    maxModels -> 0,
    maxRuntimeSecs -> 0.0,
    maxRuntimeSecsPerModel -> 0.0,
    stoppingRounds -> 3,
    stoppingMetric -> StoppingMetric.AUTO.name(),
    stoppingTolerance -> -1.0)

  //
  // Getters
  //
  def getSeed(): Long = $(seed)

  def getMaxModels(): Int = $(maxModels)

  def getMaxRuntimeSecs(): Double = $(maxRuntimeSecs)

  def getMaxRuntimeSecsPerModel(): Double = $(maxRuntimeSecsPerModel)

  def getStoppingRounds(): Int = $(stoppingRounds)

  def getStoppingMetric(): String = $(stoppingMetric)

  def getStoppingTolerance(): Double = $(stoppingTolerance)

  //
  // Setters
  //
  def setSeed(value: Long): this.type = {
    set(seed, value)
  }
           
  def setMaxModels(value: Int): this.type = {
    set(maxModels, value)
  }
           
  def setMaxRuntimeSecs(value: Double): this.type = {
    set(maxRuntimeSecs, value)
  }
           
  def setMaxRuntimeSecsPerModel(value: Double): this.type = {
    set(maxRuntimeSecsPerModel, value)
  }
           
  def setStoppingRounds(value: Int): this.type = {
    set(stoppingRounds, value)
  }
           
  def setStoppingMetric(value: String): this.type = {
    val validated = EnumParamValidator.getValidatedEnumValue[StoppingMetric](value)
    set(stoppingMetric, validated)
  }
           
  def setStoppingTolerance(value: Double): this.type = {
    set(stoppingTolerance, value)
  }
           

  override private[sparkling] def getH2OAlgorithmParams(trainingFrame: H2OFrame): Map[String, Any] = {
    super.getH2OAlgorithmParams(trainingFrame) ++ getH2OAutoMLStoppingCriteriaParams(trainingFrame)
  }

  private[sparkling] def getH2OAutoMLStoppingCriteriaParams(trainingFrame: H2OFrame): Map[String, Any] = {
      Map(
        "seed" -> getSeed(),
        "max_models" -> getMaxModels(),
        "max_runtime_secs" -> getMaxRuntimeSecs(),
        "max_runtime_secs_per_model" -> getMaxRuntimeSecsPerModel(),
        "stopping_rounds" -> getStoppingRounds(),
        "stopping_metric" -> getStoppingMetric(),
        "stopping_tolerance" -> getStoppingTolerance())
  }

  override private[sparkling] def getSWtoH2OParamNameMap(): Map[String, String] = {
    super.getSWtoH2OParamNameMap() ++
      Map(
        "seed" -> "seed",
        "maxModels" -> "max_models",
        "maxRuntimeSecs" -> "max_runtime_secs",
        "maxRuntimeSecsPerModel" -> "max_runtime_secs_per_model",
        "stoppingRounds" -> "stopping_rounds",
        "stoppingMetric" -> "stopping_metric",
        "stoppingTolerance" -> "stopping_tolerance")
  }
      
}