-
Notifications
You must be signed in to change notification settings - Fork 3
/
Autoencoder.scala
79 lines (74 loc) · 3.08 KB
/
Autoencoder.scala
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
/*
* 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 scaladl
import java.io._
import org.apache.spark.ml.feature.MinMaxScaler
import org.apache.spark.ml.linalg.Vectors
import org.apache.spark.ml.scaladl.{Autoencoder, RegressionEvaluator}
import org.apache.spark.sql.SparkSession
import org.joda.time.{DateTime, DateTimeZone}
object Autoencoder {
def main(args: Array[String]): Unit = {
if(args.length != 5){
println("Dataset, train size, iterations, num_workers and num_cores should be pased.")
return
}
val spark = SparkSession.builder
.appName("ML Autoencoder")
.getOrCreate()
val dataset = "./dataset/" + args(0)
val data = spark.read.format("libsvm").load(dataset)
val scaler = new MinMaxScaler()
.setInputCol("features")
.setOutputCol("scaledFeatures")
// materialize data lazily persisted in memory
val scalerModel = scaler.fit(data)
// rescale each feature to range [min, max].
val scaledData = scalerModel.transform(data)
val processedData = scaledData.select("label", "scaledFeatures").toDF("label", "features")
val trainSize = args(1).toFloat
val testSize = (1.0 - trainSize)
val pseudoTestSize : Double = (testSize / 4.toDouble)
val split = processedData.randomSplit(Array(trainSize, testSize), 1234L)
val train = split(0)
val test = split(1)
val testSplitted = test.randomSplit(Array(pseudoTestSize, pseudoTestSize, pseudoTestSize, pseudoTestSize), 1234L)
data.unpersist()
test.unpersist()
val layers = Array[Int](220, 140, 60, 140, 220)
// create the trainer and set its parameters
val trainer = new Autoencoder()
.setLayers(layers)
.setBlockSize(256)
.setSeed(1234L)
.setMaxIter(args(2).toInt)
// train the model
println("Training on %d workers with %d cores each".format(args(3).toInt, args(4).toInt))
val model = trainer.startTrain(train)
println("Initialized regression evaluator")
val eval = new RegressionEvaluator()
var total : Double = 0.toDouble;
train.unpersist()
for (i <- 0 until testSplitted.length) {
val predicted = model.predictDataset(testSplitted(i));
total = total + eval.evaluate(predicted)
predicted.unpersist()
testSplitted(i).unpersist()
}
println("Final test Mean Squared Error (MSE): %f".format(total / testSplitted.length))
}
}