Github user holdenk commented on a diff in the pull request:
https://github.com/apache/spark/pull/6339#discussion_r35838221
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/feature/StringIndexer.scala ---
@@ -151,4 +152,91 @@ class StringIndexerModel private[ml] (
val copied = new StringIndexerModel(uid, labels)
copyValues(copied, extra)
}
+
+ /**
+ * Return a model to perform the inverse transformation.
+ * Note: By default we keep the original columns during this
transformation, so the inverse
+ * should only be used on new columns such as predicted labels.
+ */
+ def invert(inputCol: String, outputCol: String): StringIndexerInverse = {
+ new StringIndexerInverse()
+ .setInputCol(inputCol)
+ .setOutputCol(outputCol)
+ .setLabels(labels)
+ }
+}
+
+/**
+ * :: Experimental ::
+ * Transform a provided column back to the original input types using
either the metadata
+ * on the input column, or if provided using the labels supplied by the
user.
+ * Note: By default we keep the original columns during this
transformation,
+ * so the inverse should only be used on new columns such as predicted
labels.
+ */
+@Experimental
+class StringIndexerInverse private[ml] (
+ override val uid: String) extends Transformer
+ with HasInputCol with HasOutputCol {
+
+ def this(labels: Option[Array[String]] = None) =
+ this(Identifiable.randomUID("strIdxInv"))
+
+ /** @group setParam */
+ def setInputCol(value: String): this.type = set(inputCol, value)
+
+ /** @group setParam */
+ def setOutputCol(value: String): this.type = set(outputCol, value)
+
+ /** @group setParam */
+ def setLabels(value: Array[String]): this.type = set(labels, value)
+
+ /**
+ * Param for array of labels.
+ * @group param
+ */
+ final val labels: StringArrayParam = new StringArrayParam(this,
"labels", "array of labels")
+
+ setDefault(labels, null)
--- End diff --
That seems less nice, with null we can just wrap it in option use
getOrElse. Whats the reason for an empty array? (ok either way just want to
make sure I understand).
---
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]
