[Case Study] Multilayer Perceptron Classification using Python

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. from __future__ import print_function
   
2. 
   
3. from pyspark import SparkContext
   
4. from pyspark.sql import SQLContext
   
5. # $example on$
   
6. from pyspark.ml.classification import MultilayerPerceptronClassifier
   
7. from pyspark.ml.evaluation import MulticlassClassificationEvaluator
   
8. # $example off$
   
9. 
   
10. if __name__ == "__main__":
    
11. 
    
12.     sc = SparkContext(appName="multilayer_perceptron_classification_example")
    
13.     sqlContext = SQLContext(sc)
    
14. 
    
15.     # $example on$
    
16.     # Load training data
    
17.     data = sqlContext.read.format("libsvm")\
    
18.         .load("data/mllib/sample_multiclass_classification_data.txt")
    
19.     # Split the data into train and test
    
20.     splits = data.randomSplit([0.6, 0.4], 1234)
    
21.     train = splits[0]
    
22.     test = splits[1]
    
23.     # specify layers for the neural network:
    
24.     # input layer of size 4 (features), two intermediate of size 5 and 4
    
25.     # and output of size 3 (classes)
    
26.     layers = [4, 5, 4, 3]
    
27.     # create the trainer and set its parameters
    
28.     trainer = MultilayerPerceptronClassifier(maxIter=100, layers=layers, blockSize=128, seed=1234)
    
29.     # train the model
    
30.     model = trainer.fit(train)
    
31.     # compute precision on the test set
    
32.     result = model.transform(test)
    
33.     predictionAndLabels = result.select("prediction", "label")
    
34.     evaluator = MulticlassClassificationEvaluator(metricName="precision")
    
35.     print("Precision:" + str(evaluator.evaluate(predictionAndLabels)))
    
36.     # $example off$
    
37. 
    
38.     sc.stop()

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：multilayer Case study python cation multi training example future import