Github user tomerk commented on a diff in the pull request: https://github.com/apache/spark/pull/3637#discussion_r22752339 --- Diff: mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala --- @@ -80,69 +50,157 @@ class LogisticRegression extends Estimator[LogisticRegressionModel] with Logisti def setRegParam(value: Double): this.type = set(regParam, value) def setMaxIter(value: Int): this.type = set(maxIter, value) - def setLabelCol(value: String): this.type = set(labelCol, value) def setThreshold(value: Double): this.type = set(threshold, value) - def setFeaturesCol(value: String): this.type = set(featuresCol, value) - def setScoreCol(value: String): this.type = set(scoreCol, value) - def setPredictionCol(value: String): this.type = set(predictionCol, value) override def fit(dataset: SchemaRDD, paramMap: ParamMap): LogisticRegressionModel = { + // Check schema transformSchema(dataset.schema, paramMap, logging = true) - import dataset.sqlContext._ + + // Extract columns from data. If dataset is persisted, do not persist oldDataset. + val oldDataset = extractLabeledPoints(dataset, paramMap) val map = this.paramMap ++ paramMap - val instances = dataset.select(map(labelCol).attr, map(featuresCol).attr) - .map { case Row(label: Double, features: Vector) => - LabeledPoint(label, features) - }.persist(StorageLevel.MEMORY_AND_DISK) + val handlePersistence = dataset.getStorageLevel == StorageLevel.NONE + if (handlePersistence) { + oldDataset.persist(StorageLevel.MEMORY_AND_DISK) + } + + // Train model val lr = new LogisticRegressionWithLBFGS lr.optimizer .setRegParam(map(regParam)) .setNumIterations(map(maxIter)) - val lrm = new LogisticRegressionModel(this, map, lr.run(instances).weights) - instances.unpersist() + val oldModel = lr.run(oldDataset) + val lrm = new LogisticRegressionModel(this, map, oldModel.weights, oldModel.intercept) + + if (handlePersistence) { + oldDataset.unpersist() + } + // copy model params Params.inheritValues(map, this, lrm) lrm } - private[ml] override def transformSchema(schema: StructType, paramMap: ParamMap): StructType = { - validateAndTransformSchema(schema, paramMap, fitting = true) - } + override protected def featuresDataType: DataType = new VectorUDT } + /** * :: AlphaComponent :: + * * Model produced by [[LogisticRegression]]. */ @AlphaComponent class LogisticRegressionModel private[ml] ( override val parent: LogisticRegression, --- End diff -- Why do models need to have a reference to the Estimator that produced them?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org