Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/7099#discussion_r33641128
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/regression/LinearRegression.scala ---
@@ -208,20 +239,61 @@ class LinearRegression(override val uid: String)
class LinearRegressionModel private[ml] (
override val uid: String,
val weights: Vector,
- val intercept: Double)
+ val intercept: Double,
+ val trainingSummary: LinearRegressionTrainingResults)
extends RegressionModel[Vector, LinearRegressionModel]
with LinearRegressionParams {
+ /**
+ * Evaluates the model on a test-set.
+ * @param testset Test dataset to evaluate model on.
+ * @return
+ */
+ def evaluate(testset: DataFrame): LinearRegressionResults = {
+ val predictionAndObservations = testset
+ .select($(labelCol), $(featuresCol))
+ .map { case Row(label: Double, features: Vector) =>
(predict(features), label) }
+
+ new LinearRegressionResults(predictionAndObservations)
+ }
+
override protected def predict(features: Vector): Double = {
dot(features, weights) + intercept
}
override def copy(extra: ParamMap): LinearRegressionModel = {
- copyValues(new LinearRegressionModel(uid, weights, intercept), extra)
+ copyValues(new LinearRegressionModel(uid, weights, intercept,
trainingSummary), extra)
}
}
/**
+ * :: Experimental ::
+ * Linear regression training results.
+ * @param predictionAndObservations dataframe with columns predictions (0)
and observations(0).
+ * @param totalIterations total number of training iterations.
+ * @param objectiveTrace objective function value at each iteration.
+ */
+@Experimental
+class LinearRegressionTrainingResults(
+ @transient predictionAndObservations: RDD[(Double, Double)],
--- End diff --
No need to mark this transient, I believe, since it's not stored as a val
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
