srowen commented on a change in pull request #26413: [SPARK-16872][ML][PYSPARK]
Impl Gaussian Naive Bayes Classifier
URL: https://github.com/apache/spark/pull/26413#discussion_r344248622
##########
File path:
mllib/src/main/scala/org/apache/spark/ml/classification/NaiveBayes.scala
##########
@@ -396,15 +545,29 @@ object NaiveBayesModel extends
MLReadable[NaiveBayesModel] {
private val className = classOf[NaiveBayesModel].getName
override def load(path: String): NaiveBayesModel = {
+ implicit val format = DefaultFormats
val metadata = DefaultParamsReader.loadMetadata(path, sc, className)
+ val (major, minor) =
VersionUtils.majorMinorVersion(metadata.sparkVersion)
+ val modelTypeJson = metadata.getParamValue("modelType")
+ val modelType = Param.jsonDecode[String](compact(render(modelTypeJson)))
val dataPath = new Path(path, "data").toString
val data = sparkSession.read.parquet(dataPath)
val vecConverted = MLUtils.convertVectorColumnsToML(data, "pi")
- val Row(pi: Vector, theta: Matrix) =
MLUtils.convertMatrixColumnsToML(vecConverted, "theta")
- .select("pi", "theta")
- .head()
- val model = new NaiveBayesModel(metadata.uid, pi, theta)
+
+ val model = if (major.toInt < 3 || modelType != NaiveBayes.Gaussian) {
Review comment:
Do we use the Spark version elsewhere like this? rather than a version of
the model in the representation?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
