[
https://issues.apache.org/jira/browse/SPARK-14023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15208308#comment-15208308
]
Sean Owen commented on SPARK-14023:
-----------------------------------
For now I'd just focus on standardizing the terminology. The only little
drawback to custom exceptions is a) tiny bit more code and b) you can't just
use {{require()}}. If it's just for descriptiveness, update the text. Custom
exceptions make sense when you need callers to selectively catch one but that's
not the case here.
> Make exceptions consistent regarding fields and columns
> -------------------------------------------------------
>
> Key: SPARK-14023
> URL: https://issues.apache.org/jira/browse/SPARK-14023
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Affects Versions: 2.0.0
> Reporter: Jacek Laskowski
> Priority: Trivial
>
> As you can see below, a column is called a field depending on where an
> exception is thrown. I think it should be "column" everywhere (since that's
> what has a type from a schema).
> {code}
> scala> lr
> res32: org.apache.spark.ml.regression.LinearRegression = linReg_d9bfe808e743
> scala> lr.fit(ds)
> java.lang.IllegalArgumentException: Field "features" does not exist.
> at
> org.apache.spark.sql.types.StructType$$anonfun$apply$1.apply(StructType.scala:214)
> at
> org.apache.spark.sql.types.StructType$$anonfun$apply$1.apply(StructType.scala:214)
> at scala.collection.MapLike$class.getOrElse(MapLike.scala:128)
> at scala.collection.AbstractMap.getOrElse(Map.scala:59)
> at org.apache.spark.sql.types.StructType.apply(StructType.scala:213)
> at
> org.apache.spark.ml.util.SchemaUtils$.checkColumnType(SchemaUtils.scala:40)
> at
> org.apache.spark.ml.PredictorParams$class.validateAndTransformSchema(Predictor.scala:50)
> at
> org.apache.spark.ml.Predictor.validateAndTransformSchema(Predictor.scala:71)
> at org.apache.spark.ml.Predictor.transformSchema(Predictor.scala:116)
> at org.apache.spark.ml.PipelineStage.transformSchema(Pipeline.scala:67)
> at org.apache.spark.ml.Predictor.fit(Predictor.scala:89)
> ... 51 elided
> scala> lr.fit(ds)
> java.lang.IllegalArgumentException: requirement failed: Column label must be
> of type DoubleType but was actually StringType.
> at scala.Predef$.require(Predef.scala:219)
> at
> org.apache.spark.ml.util.SchemaUtils$.checkColumnType(SchemaUtils.scala:42)
> at
> org.apache.spark.ml.PredictorParams$class.validateAndTransformSchema(Predictor.scala:53)
> at
> org.apache.spark.ml.Predictor.validateAndTransformSchema(Predictor.scala:71)
> at org.apache.spark.ml.Predictor.transformSchema(Predictor.scala:116)
> at org.apache.spark.ml.PipelineStage.transformSchema(Pipeline.scala:67)
> at org.apache.spark.ml.Predictor.fit(Predictor.scala:89)
> ... 51 elided
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
