spark技术分享ValidatorParams
| Parameter | Default Value | Description |
|---|---|---|
|
(undefined) |
Estimator for best model selection |
|
|
(undefined) |
Param maps for the estimator |
|
|
(undefined) |
Evaluator to select hyper-parameters that maximize the validated metric |
logTuningParams Method
|
1 2 3 4 5 |
logTuningParams(instrumentation: Instrumentation[_]): Unit |
logTuningParams…FIXME
|
Note
|
logTuningParams is used when…FIXME
|
loadImpl Method
|
1 2 3 4 5 6 7 8 |
loadImpl[M]( path: String, sc: SparkContext, expectedClassName: String): (Metadata, Estimator[M], Evaluator, Array[ParamMap]) |
loadImpl…FIXME
|
Note
|
loadImpl is used when…FIXME
|
transformSchemaImpl Method
|
1 2 3 4 5 |
transformSchemaImpl(schema: StructType): StructType |
transformSchemaImpl…FIXME
|
Note
|
transformSchemaImpl is used when CrossValidator and TrainValidationSplit validate and transform schema.
|
