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
|
/*
* 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.
*/
// scalastyle:off println
package org.apache.spark.examples.ml
import scala.collection.mutable
import scala.language.reflectiveCalls
import scopt.OptionParser
import org.apache.spark.examples.mllib.AbstractParams
import org.apache.spark.ml.{Pipeline, PipelineStage}
import org.apache.spark.ml.classification.{LogisticRegression, LogisticRegressionModel}
import org.apache.spark.ml.feature.StringIndexer
import org.apache.spark.sql.{DataFrame, SparkSession}
/**
* An example runner for logistic regression with elastic-net (mixing L1/L2) regularization.
* Run with
* {{{
* bin/run-example ml.LogisticRegressionExample [options]
* }}}
* A synthetic dataset can be found at `data/mllib/sample_libsvm_data.txt` which can be
* trained by
* {{{
* bin/run-example ml.LogisticRegressionExample --regParam 0.3 --elasticNetParam 0.8 \
* data/mllib/sample_libsvm_data.txt
* }}}
* If you use it as a template to create your own app, please use `spark-submit` to submit your app.
*/
object LogisticRegressionExample {
case class Params(
input: String = null,
testInput: String = "",
dataFormat: String = "libsvm",
regParam: Double = 0.0,
elasticNetParam: Double = 0.0,
maxIter: Int = 100,
fitIntercept: Boolean = true,
tol: Double = 1E-6,
fracTest: Double = 0.2) extends AbstractParams[Params]
def main(args: Array[String]) {
val defaultParams = Params()
val parser = new OptionParser[Params]("LogisticRegressionExample") {
head("LogisticRegressionExample: an example Logistic Regression with Elastic-Net app.")
opt[Double]("regParam")
.text(s"regularization parameter, default: ${defaultParams.regParam}")
.action((x, c) => c.copy(regParam = x))
opt[Double]("elasticNetParam")
.text(s"ElasticNet mixing parameter. For alpha = 0, the penalty is an L2 penalty. " +
s"For alpha = 1, it is an L1 penalty. For 0 < alpha < 1, the penalty is a combination of " +
s"L1 and L2, default: ${defaultParams.elasticNetParam}")
.action((x, c) => c.copy(elasticNetParam = x))
opt[Int]("maxIter")
.text(s"maximum number of iterations, default: ${defaultParams.maxIter}")
.action((x, c) => c.copy(maxIter = x))
opt[Boolean]("fitIntercept")
.text(s"whether to fit an intercept term, default: ${defaultParams.fitIntercept}")
.action((x, c) => c.copy(fitIntercept = x))
opt[Double]("tol")
.text(s"the convergence tolerance of iterations, Smaller value will lead " +
s"to higher accuracy with the cost of more iterations, default: ${defaultParams.tol}")
.action((x, c) => c.copy(tol = x))
opt[Double]("fracTest")
.text(s"fraction of data to hold out for testing. If given option testInput, " +
s"this option is ignored. default: ${defaultParams.fracTest}")
.action((x, c) => c.copy(fracTest = x))
opt[String]("testInput")
.text(s"input path to test dataset. If given, option fracTest is ignored." +
s" default: ${defaultParams.testInput}")
.action((x, c) => c.copy(testInput = x))
opt[String]("dataFormat")
.text("data format: libsvm (default), dense (deprecated in Spark v1.1)")
.action((x, c) => c.copy(dataFormat = x))
arg[String]("<input>")
.text("input path to labeled examples")
.required()
.action((x, c) => c.copy(input = x))
checkConfig { params =>
if (params.fracTest < 0 || params.fracTest >= 1) {
failure(s"fracTest ${params.fracTest} value incorrect; should be in [0,1).")
} else {
success
}
}
}
parser.parse(args, defaultParams) match {
case Some(params) => run(params)
case _ => sys.exit(1)
}
}
def run(params: Params): Unit = {
val spark = SparkSession
.builder
.appName(s"LogisticRegressionExample with $params")
.getOrCreate()
println(s"LogisticRegressionExample with parameters:\n$params")
// Load training and test data and cache it.
val (training: DataFrame, test: DataFrame) = DecisionTreeExample.loadDatasets(params.input,
params.dataFormat, params.testInput, "classification", params.fracTest)
// Set up Pipeline.
val stages = new mutable.ArrayBuffer[PipelineStage]()
val labelIndexer = new StringIndexer()
.setInputCol("label")
.setOutputCol("indexedLabel")
stages += labelIndexer
val lor = new LogisticRegression()
.setFeaturesCol("features")
.setLabelCol("indexedLabel")
.setRegParam(params.regParam)
.setElasticNetParam(params.elasticNetParam)
.setMaxIter(params.maxIter)
.setTol(params.tol)
.setFitIntercept(params.fitIntercept)
stages += lor
val pipeline = new Pipeline().setStages(stages.toArray)
// Fit the Pipeline.
val startTime = System.nanoTime()
val pipelineModel = pipeline.fit(training)
val elapsedTime = (System.nanoTime() - startTime) / 1e9
println(s"Training time: $elapsedTime seconds")
val lorModel = pipelineModel.stages.last.asInstanceOf[LogisticRegressionModel]
// Print the weights and intercept for logistic regression.
println(s"Weights: ${lorModel.coefficients} Intercept: ${lorModel.intercept}")
println("Training data results:")
DecisionTreeExample.evaluateClassificationModel(pipelineModel, training, "indexedLabel")
println("Test data results:")
DecisionTreeExample.evaluateClassificationModel(pipelineModel, test, "indexedLabel")
spark.stop()
}
}
// scalastyle:on println
|
