Spark ML Pipeline: Classification
2016-03-16 16:14:04 11 举报Apache Spark ML Pipeline Classes
作者其他创作
大纲/内容
read raw data
fits
is a
HashingTF
+setInputCol(String)+setOutputCol(String)+setNumFeatures(Int)+transform()+params()
Transformer
BinaryClassificationEvaluator
org.apache.spark.ml.feature
属性
CrossValidatorModel
org.apache.spark.ml.classification
Tokenizer
+getOutputCol(): String+setInputCol(String)+setOutputCol(String)+transform()+params()
org.apache.spark.ml
Estimator
+fit(DataFrame)
DataFrame
org.apache.spark.ml.tuning
ClassificationModel
+explainParams(): String+extractParamMap(): ParamMap+getParam(String): Param[Any]+params: Array[Param[_]]+validateParams()
Model
+parent: Estimator[M]
org.apache.spark.ml.evaluation
Evaluator
output prediction
PredictionModel
org.apache.spark.sql
Pipeline
+setStages(Array[PipelineStage])
PipelineStage
LogisticRegressionModel
org.apache.spark.ml.param
ParamGridBuilder
+addGrid(): ParamGridBuilder.this.type+build(): Array[ParamMap]
LogisticRegression
+setMaxIter(Int)+setRegParam(Double)
ProbabilisticClassificationModel
CrossValidator
+setEstimator(Estimator[_])+setEstimatorParamMaps(Array[ParamMap])+setEvaluator(Evaluator)+setNumFolds(Int)
PipelineModel
0 条评论
下一页
