[Case Study]Multilayer Perceptron Classification using Scala

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

1. /**
   
2. * An example for Multilayer Perceptron Classification.
   
3. */
   
4. object MultilayerPerceptronClassifierExample {
   
5. 
   
6.   def main(args: Array[String]): Unit = {
   
7.     val conf = new SparkConf().setAppName("MultilayerPerceptronClassifierExample")
   
8.     val sc = new SparkContext(conf)
   
9.     val sqlContext = new SQLContext(sc)
   
10. 
    
11.     // $example on$
    
12.     // Load the data stored in LIBSVM format as a DataFrame.
    
13.     val data = sqlContext.read.format("libsvm")
    
14.       .load("data/mllib/sample_multiclass_classification_data.txt")
    
15.     // Split the data into train and test
    
16.     val splits = data.randomSplit(Array(0.6, 0.4), seed = 1234L)
    
17.     val train = splits(0)
    
18.     val test = splits(1)
    
19.     // specify layers for the neural network:
    
20.     // input layer of size 4 (features), two intermediate of size 5 and 4
    
21.     // and output of size 3 (classes)
    
22.     val layers = Array[Int](4, 5, 4, 3)
    
23.     // create the trainer and set its parameters
    
24.     val trainer = new MultilayerPerceptronClassifier()
    
25.       .setLayers(layers)
    
26.       .setBlockSize(128)
    
27.       .setSeed(1234L)
    
28.       .setMaxIter(100)
    
29.     // train the model
    
30.     val model = trainer.fit(train)
    
31.     // compute precision on the test set
    
32.     val result = model.transform(test)
    
33.     val predictionAndLabels = result.select("prediction", "label")
    
34.     val evaluator = new MulticlassClassificationEvaluator()
    
35.       .setMetricName("precision")
    
36.     println("Precision:" + evaluator.evaluate(predictionAndLabels))
    
37.     // $example off$
    
38. 
    
39.     sc.stop()
    
40.   }
    
41. }
    
42. // scalastyle:off println

复制代码

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Case study multilayer cation SCALA study example train