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
|
/*
* 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.feature
import org.apache.spark.SparkFunSuite
import org.apache.spark.ml.util.DefaultReadWriteTest
import org.apache.spark.mllib.util.MLlibTestSparkContext
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.functions.udf
class QuantileDiscretizerSuite
extends SparkFunSuite with MLlibTestSparkContext with DefaultReadWriteTest {
test("Test observed number of buckets and their sizes match expected values") {
val spark = this.spark
import spark.implicits._
val datasetSize = 100000
val numBuckets = 5
val df = sc.parallelize(1.0 to datasetSize by 1.0).map(Tuple1.apply).toDF("input")
val discretizer = new QuantileDiscretizer()
.setInputCol("input")
.setOutputCol("result")
.setNumBuckets(numBuckets)
val result = discretizer.fit(df).transform(df)
val observedNumBuckets = result.select("result").distinct.count
assert(observedNumBuckets === numBuckets,
"Observed number of buckets does not equal expected number of buckets.")
val relativeError = discretizer.getRelativeError
val isGoodBucket = udf {
(size: Int) => math.abs( size - (datasetSize / numBuckets)) <= (relativeError * datasetSize)
}
val numGoodBuckets = result.groupBy("result").count.filter(isGoodBucket($"count")).count
assert(numGoodBuckets === numBuckets,
"Bucket sizes are not within expected relative error tolerance.")
}
test("Test on data with high proportion of duplicated values") {
val spark = this.spark
import spark.implicits._
val numBuckets = 5
val expectedNumBuckets = 3
val df = sc.parallelize(Array(1.0, 3.0, 2.0, 1.0, 1.0, 2.0, 3.0, 2.0, 2.0, 2.0, 1.0, 3.0))
.map(Tuple1.apply).toDF("input")
val discretizer = new QuantileDiscretizer()
.setInputCol("input")
.setOutputCol("result")
.setNumBuckets(numBuckets)
val result = discretizer.fit(df).transform(df)
val observedNumBuckets = result.select("result").distinct.count
assert(observedNumBuckets == expectedNumBuckets,
s"Observed number of buckets are not correct." +
s" Expected $expectedNumBuckets but found $observedNumBuckets")
}
test("Test transform on data with NaN value") {
val spark = this.spark
import spark.implicits._
val numBuckets = 3
val df = sc.parallelize(Array(1.0, 1.0, 1.0, Double.NaN))
.map(Tuple1.apply).toDF("input")
val discretizer = new QuantileDiscretizer()
.setInputCol("input")
.setOutputCol("result")
.setNumBuckets(numBuckets)
// Reserve extra one bucket for NaN
val expectedNumBuckets = discretizer.fit(df).getSplits.length - 1
val result = discretizer.fit(df).transform(df)
val observedNumBuckets = result.select("result").distinct.count
assert(observedNumBuckets == expectedNumBuckets,
s"Observed number of buckets are not correct." +
s" Expected $expectedNumBuckets but found $observedNumBuckets")
}
test("Test transform method on unseen data") {
val spark = this.spark
import spark.implicits._
val trainDF = sc.parallelize(1.0 to 100.0 by 1.0).map(Tuple1.apply).toDF("input")
val testDF = sc.parallelize(-10.0 to 110.0 by 1.0).map(Tuple1.apply).toDF("input")
val discretizer = new QuantileDiscretizer()
.setInputCol("input")
.setOutputCol("result")
.setNumBuckets(5)
val result = discretizer.fit(trainDF).transform(testDF)
val firstBucketSize = result.filter(result("result") === 0.0).count
val lastBucketSize = result.filter(result("result") === 4.0).count
assert(firstBucketSize === 30L,
s"Size of first bucket ${firstBucketSize} did not equal expected value of 30.")
assert(lastBucketSize === 31L,
s"Size of last bucket ${lastBucketSize} did not equal expected value of 31.")
}
test("read/write") {
val t = new QuantileDiscretizer()
.setInputCol("myInputCol")
.setOutputCol("myOutputCol")
.setNumBuckets(6)
testDefaultReadWrite(t)
}
test("Verify resulting model has parent") {
val spark = this.spark
import spark.implicits._
val df = sc.parallelize(1 to 100).map(Tuple1.apply).toDF("input")
val discretizer = new QuantileDiscretizer()
.setInputCol("input")
.setOutputCol("result")
.setNumBuckets(5)
val model = discretizer.fit(df)
assert(model.hasParent)
}
}
|
