python训练模型，java预测模型(sklearn2pmml)

超、凢脫俗 2021-09-29 09:46 880阅读 0赞

很多情况下，线上一般使用java，然后训练机器学习模型一般都是python，那么就存在一个问题，python训练的模型，
    java怎么去调用进行线上预测呢，下面以随机森林算法来介绍一下这个过程
    python脚本如下：
    
    #!/usr/bin/python
    # -*- coding:utf-8 -*-
    
    
    from sklearn import tree
    from sklearn2pmml.pipeline import PMMLPipeline
    from sklearn2pmml import sklearn2pmml
    from sklearn2pmml import make_pmml_pipeline    # 转换pkl文件为pmml_pipeline格式
    
    import os
    os.environ["PATH"] += os.pathsep + 'C:/Program Files/Java/jdk1.8.0_171/bin'
    
    X=[[1,2,3,1],[2,4,1,5],[7,8,3,6],[4,8,4,7],[2,5,6,9]]
    y=[0,1,0,2,1]
    # PMMLPipeline只是处理estimator不能处理transformer
    # 注意：有时候需要自定义函数加入到PMMLPipeline中，可以参考博客https://blog.csdn.net/weixin_38569817/article/details/87810658
    pipeline = PMMLPipeline([("classifier", tree.DecisionTreeClassifier(random_state=9))])
    pipeline.fit(X,y)
    
    sklearn2pmml(pipeline, ".\demo.pmml", with_repr = True)
    
    
    
    java依赖添加
    <dependency>
            <groupId>org.jpmml</groupId>
            <artifactId>pmml-evaluator</artifactId>
            <version>1.4.1</version>
        </dependency>
        <dependency>
            <groupId>org.jpmml</groupId>
            <artifactId>pmml-evaluator-extension</artifactId>
            <version>1.4.1</version>
        </dependency>
    java调用脚本如下：
    
    package xxx;
    
    import org.dmg.pmml.FieldName;
    import org.dmg.pmml.PMML;
    import org.jpmml.evaluator.*;
    import org.xml.sax.SAXException;
    
    import javax.xml.bind.JAXBException;
    import java.io.FileInputStream;
    import java.io.IOException;
    import java.io.InputStream;
    import java.util.HashMap;
    import java.util.LinkedHashMap;
    import java.util.List;
    import java.util.Map;
    
    public class PMMLDemo {
        private Evaluator loadPmml(){
            PMML pmml = new PMML();
            InputStream inputStream = null;
            try {
                inputStream = new FileInputStream("demo.pmml");      # 需要引用python保存下来的pmml文件
            } catch (IOException e) {
                e.printStackTrace();
            }
            if(inputStream == null){
                return null;
            }
            InputStream is = inputStream;
            try {
                pmml = org.jpmml.model.PMMLUtil.unmarshal(is);
            } catch (SAXException e1) {
                e1.printStackTrace();
            } catch (JAXBException e1) {
                e1.printStackTrace();
            }finally {
                //关闭输入流
                try {
                    is.close();
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
            ModelEvaluatorFactory modelEvaluatorFactory = ModelEvaluatorFactory.newInstance();
            Evaluator evaluator = modelEvaluatorFactory.newModelEvaluator(pmml);
            pmml = null;
            return evaluator;
        }
        private int predict(Evaluator evaluator,int a, int b, int c, int d) {
            Map<String, Integer> data = new HashMap<String, Integer>();
            data.put("x1", a);
            data.put("x2", b);
            data.put("x3", c);
            data.put("x4", d);
            List<InputField> inputFields = evaluator.getInputFields();
            //过模型的原始特征，从画像中获取数据，作为模型输入
            Map<FieldName, FieldValue> arguments = new LinkedHashMap<FieldName, FieldValue>();
            for (InputField inputField : inputFields) {
                FieldName inputFieldName = inputField.getName();
                Object rawValue = data.get(inputFieldName.getValue());
                FieldValue inputFieldValue = inputField.prepare(rawValue);
                arguments.put(inputFieldName, inputFieldValue);
            }
    
            Map<FieldName, ?> results = evaluator.evaluate(arguments);
            List<TargetField> targetFields = evaluator.getTargetFields();
    
            TargetField targetField = targetFields.get(0);
            FieldName targetFieldName = targetField.getName();
    
            Object targetFieldValue = results.get(targetFieldName);
            System.out.println("target: " + targetFieldName.getValue() + " value: " + targetFieldValue);
            int primitiveValue = -1;
            if (targetFieldValue instanceof Computable) {
                Computable computable = (Computable) targetFieldValue;
                primitiveValue = (Integer)computable.getResult();
            }
            System.out.println(a + " " + b + " " + c + " " + d + ":" + primitiveValue);
            return primitiveValue;
        }
        public static void main(String args[]){
            PMMLDemo demo = new PMMLDemo();
            Evaluator model = demo.loadPmml();
            demo.predict(model,1,8,99,1);           # 填充数据即可预测
            demo.predict(model,111,89,9,11);
    
        }
    }