1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
|
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.spark.ml.r
import org.apache.hadoop.fs.Path
import org.json4s._
import org.json4s.JsonDSL._
import org.json4s.jackson.JsonMethods._
import org.apache.spark.ml.{Pipeline, PipelineModel}
import org.apache.spark.ml.attribute.AttributeGroup
import org.apache.spark.ml.feature.RFormula
import org.apache.spark.ml.regression._
import org.apache.spark.ml.util._
import org.apache.spark.sql._
private[r] class GeneralizedLinearRegressionWrapper private (
val pipeline: PipelineModel,
val rFeatures: Array[String],
val rCoefficients: Array[Double],
val rDispersion: Double,
val rNullDeviance: Double,
val rDeviance: Double,
val rResidualDegreeOfFreedomNull: Long,
val rResidualDegreeOfFreedom: Long,
val rAic: Double,
val rNumIterations: Int,
val isLoaded: Boolean = false) extends MLWritable {
private val glm: GeneralizedLinearRegressionModel =
pipeline.stages(1).asInstanceOf[GeneralizedLinearRegressionModel]
lazy val rDevianceResiduals: DataFrame = glm.summary.residuals()
lazy val rFamily: String = glm.getFamily
def residuals(residualsType: String): DataFrame = glm.summary.residuals(residualsType)
def transform(dataset: Dataset[_]): DataFrame = {
pipeline.transform(dataset).drop(glm.getFeaturesCol)
}
override def write: MLWriter =
new GeneralizedLinearRegressionWrapper.GeneralizedLinearRegressionWrapperWriter(this)
}
private[r] object GeneralizedLinearRegressionWrapper
extends MLReadable[GeneralizedLinearRegressionWrapper] {
def fit(
formula: String,
data: DataFrame,
family: String,
link: String,
tol: Double,
maxIter: Int,
weightCol: String): GeneralizedLinearRegressionWrapper = {
val rFormula = new RFormula()
.setFormula(formula)
val rFormulaModel = rFormula.fit(data)
// get labels and feature names from output schema
val schema = rFormulaModel.transform(data).schema
val featureAttrs = AttributeGroup.fromStructField(schema(rFormula.getFeaturesCol))
.attributes.get
val features = featureAttrs.map(_.name.get)
// assemble and fit the pipeline
val glr = new GeneralizedLinearRegression()
.setFamily(family)
.setLink(link)
.setFitIntercept(rFormula.hasIntercept)
.setTol(tol)
.setMaxIter(maxIter)
.setWeightCol(weightCol)
val pipeline = new Pipeline()
.setStages(Array(rFormulaModel, glr))
.fit(data)
val glm: GeneralizedLinearRegressionModel =
pipeline.stages(1).asInstanceOf[GeneralizedLinearRegressionModel]
val summary = glm.summary
val rFeatures: Array[String] = if (glm.getFitIntercept) {
Array("(Intercept)") ++ features
} else {
features
}
val rCoefficientStandardErrors = if (glm.getFitIntercept) {
Array(summary.coefficientStandardErrors.last) ++
summary.coefficientStandardErrors.dropRight(1)
} else {
summary.coefficientStandardErrors
}
val rTValues = if (glm.getFitIntercept) {
Array(summary.tValues.last) ++ summary.tValues.dropRight(1)
} else {
summary.tValues
}
val rPValues = if (glm.getFitIntercept) {
Array(summary.pValues.last) ++ summary.pValues.dropRight(1)
} else {
summary.pValues
}
val rCoefficients: Array[Double] = if (glm.getFitIntercept) {
Array(glm.intercept) ++ glm.coefficients.toArray ++
rCoefficientStandardErrors ++ rTValues ++ rPValues
} else {
glm.coefficients.toArray ++ rCoefficientStandardErrors ++ rTValues ++ rPValues
}
val rDispersion: Double = summary.dispersion
val rNullDeviance: Double = summary.nullDeviance
val rDeviance: Double = summary.deviance
val rResidualDegreeOfFreedomNull: Long = summary.residualDegreeOfFreedomNull
val rResidualDegreeOfFreedom: Long = summary.residualDegreeOfFreedom
val rAic: Double = summary.aic
val rNumIterations: Int = summary.numIterations
new GeneralizedLinearRegressionWrapper(pipeline, rFeatures, rCoefficients, rDispersion,
rNullDeviance, rDeviance, rResidualDegreeOfFreedomNull, rResidualDegreeOfFreedom,
rAic, rNumIterations)
}
override def read: MLReader[GeneralizedLinearRegressionWrapper] =
new GeneralizedLinearRegressionWrapperReader
override def load(path: String): GeneralizedLinearRegressionWrapper = super.load(path)
class GeneralizedLinearRegressionWrapperWriter(instance: GeneralizedLinearRegressionWrapper)
extends MLWriter {
override protected def saveImpl(path: String): Unit = {
val rMetadataPath = new Path(path, "rMetadata").toString
val pipelinePath = new Path(path, "pipeline").toString
val rMetadata = ("class" -> instance.getClass.getName) ~
("rFeatures" -> instance.rFeatures.toSeq) ~
("rCoefficients" -> instance.rCoefficients.toSeq) ~
("rDispersion" -> instance.rDispersion) ~
("rNullDeviance" -> instance.rNullDeviance) ~
("rDeviance" -> instance.rDeviance) ~
("rResidualDegreeOfFreedomNull" -> instance.rResidualDegreeOfFreedomNull) ~
("rResidualDegreeOfFreedom" -> instance.rResidualDegreeOfFreedom) ~
("rAic" -> instance.rAic) ~
("rNumIterations" -> instance.rNumIterations)
val rMetadataJson: String = compact(render(rMetadata))
sc.parallelize(Seq(rMetadataJson), 1).saveAsTextFile(rMetadataPath)
instance.pipeline.save(pipelinePath)
}
}
class GeneralizedLinearRegressionWrapperReader
extends MLReader[GeneralizedLinearRegressionWrapper] {
override def load(path: String): GeneralizedLinearRegressionWrapper = {
implicit val format = DefaultFormats
val rMetadataPath = new Path(path, "rMetadata").toString
val pipelinePath = new Path(path, "pipeline").toString
val rMetadataStr = sc.textFile(rMetadataPath, 1).first()
val rMetadata = parse(rMetadataStr)
val rFeatures = (rMetadata \ "rFeatures").extract[Array[String]]
val rCoefficients = (rMetadata \ "rCoefficients").extract[Array[Double]]
val rDispersion = (rMetadata \ "rDispersion").extract[Double]
val rNullDeviance = (rMetadata \ "rNullDeviance").extract[Double]
val rDeviance = (rMetadata \ "rDeviance").extract[Double]
val rResidualDegreeOfFreedomNull = (rMetadata \ "rResidualDegreeOfFreedomNull").extract[Long]
val rResidualDegreeOfFreedom = (rMetadata \ "rResidualDegreeOfFreedom").extract[Long]
val rAic = (rMetadata \ "rAic").extract[Double]
val rNumIterations = (rMetadata \ "rNumIterations").extract[Int]
val pipeline = PipelineModel.load(pipelinePath)
new GeneralizedLinearRegressionWrapper(pipeline, rFeatures, rCoefficients, rDispersion,
rNullDeviance, rDeviance, rResidualDegreeOfFreedomNull, rResidualDegreeOfFreedom,
rAic, rNumIterations, isLoaded = true)
}
}
}
|
