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
|
/*
* 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.ann
import org.apache.spark.SparkFunSuite
import org.apache.spark.mllib.linalg.Vectors
import org.apache.spark.mllib.util.MLlibTestSparkContext
import org.apache.spark.mllib.util.TestingUtils._
class ANNSuite extends SparkFunSuite with MLlibTestSparkContext {
// TODO: test for weights comparison with Weka MLP
test("ANN with Sigmoid learns XOR function with LBFGS optimizer") {
val inputs = Array(
Array(0.0, 0.0),
Array(0.0, 1.0),
Array(1.0, 0.0),
Array(1.0, 1.0)
)
val outputs = Array(0.0, 1.0, 1.0, 0.0)
val data = inputs.zip(outputs).map { case (features, label) =>
(Vectors.dense(features), Vectors.dense(label))
}
val rddData = sc.parallelize(data, 1)
val hiddenLayersTopology = Array(5)
val dataSample = rddData.first()
val layerSizes = dataSample._1.size +: hiddenLayersTopology :+ dataSample._2.size
val topology = FeedForwardTopology.multiLayerPerceptron(layerSizes, false)
val initialWeights = FeedForwardModel(topology, 23124).weights()
val trainer = new FeedForwardTrainer(topology, 2, 1)
trainer.setWeights(initialWeights)
trainer.LBFGSOptimizer.setNumIterations(20)
val model = trainer.train(rddData)
val predictionAndLabels = rddData.map { case (input, label) =>
(model.predict(input)(0), label(0))
}.collect()
predictionAndLabels.foreach { case (p, l) =>
assert(math.round(p) === l)
}
}
test("ANN with SoftMax learns XOR function with 2-bit output and batch GD optimizer") {
val inputs = Array(
Array(0.0, 0.0),
Array(0.0, 1.0),
Array(1.0, 0.0),
Array(1.0, 1.0)
)
val outputs = Array(
Array(1.0, 0.0),
Array(0.0, 1.0),
Array(0.0, 1.0),
Array(1.0, 0.0)
)
val data = inputs.zip(outputs).map { case (features, label) =>
(Vectors.dense(features), Vectors.dense(label))
}
val rddData = sc.parallelize(data, 1)
val hiddenLayersTopology = Array(5)
val dataSample = rddData.first()
val layerSizes = dataSample._1.size +: hiddenLayersTopology :+ dataSample._2.size
val topology = FeedForwardTopology.multiLayerPerceptron(layerSizes, false)
val initialWeights = FeedForwardModel(topology, 23124).weights()
val trainer = new FeedForwardTrainer(topology, 2, 2)
trainer.SGDOptimizer.setNumIterations(2000)
trainer.setWeights(initialWeights)
val model = trainer.train(rddData)
val predictionAndLabels = rddData.map { case (input, label) =>
(model.predict(input), label)
}.collect()
predictionAndLabels.foreach { case (p, l) =>
assert(p ~== l absTol 0.5)
}
}
}
|
