Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/7099#discussion_r33641103
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/regression/LinearRegression.scala ---
@@ -193,11 +199,36 @@ class LinearRegression(override val uid: String)
val intercept = if ($(fitIntercept)) yMean - dot(weights,
Vectors.dense(featuresMean)) else 0.0
if (handlePersistence) instances.unpersist()
+ val summary = generateTrainingResults(instances, lossHistory.result(),
weights, intercept)
+
// TODO: Converts to sparse format based on the storage, but may base
on the scoring speed.
- copyValues(new LinearRegressionModel(uid, weights.compressed,
intercept))
+ copyValues(new LinearRegressionModel(uid, weights.compressed,
intercept, summary))
+ }
+
+ private def generateTrainingResults(
--- End diff --
I'd prefer this method be in LinearRegressionTrainingResults, either as a
constructor, or as a static factory method.
Also, you might want to pass in the LinearRegressionModel so that you can
use its transform() method. Rather than having LinearRegressionModel take
summary in the constructor, it can start with it set to null; later on, you can
set it using a setter method.
---
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]
