Transformer
is the contract in Spark MLlib for transformers that transform one dataset into another.
Transformer
is a PipelineStage and so…FIXME
package org.apache.spark.ml
abstract class Evaluator {
// only required methods that have no implementation
def transform(dataset: Dataset[_]): DataFrame
def copy(extra: ParamMap): Transformer
}
Method | Description |
---|---|
Used when… |
|
Used when… |