Spark ML机器学习库评估指标示例

2024-05-28 02:58

本文主要是介绍Spark ML机器学习库评估指标示例,希望对大家解决编程问题提供一定的参考价值,需要的开发者们随着小编来一起学习吧!

本文主要对 Spark ML库下模型评估指标的讲解,以下代码均以Jupyter Notebook进行讲解,Spark版本为2.4.5。模型评估指标位于包org.apache.spark.ml.evaluation下。

模型评估指标是指测试集的评估指标,而不是训练集的评估指标

1、回归评估指标

RegressionEvaluator

Evaluator for regression, which expects two input columns: prediction and label.

评估指标支持以下几种:

val metricName: Param[String]

  • "rmse" (default): root mean squared error
  • "mse": mean squared error
  • "r2": R2 metric
  • "mae": mean absolute error

Examples

# import dependencies
import org.apache.spark.ml.regression.LinearRegression
import org.apache.spark.ml.evaluation.RegressionEvaluator// Load training data
val data = spark.read.format("libsvm").load("/data1/software/spark/data/mllib/sample_linear_regression_data.txt")val lr = new LinearRegression().setMaxIter(10).setRegParam(0.3).setElasticNetParam(0.8)// Fit the model
val lrModel = lr.fit(training)// Summarize the model over the training set and print out some metrics
val trainingSummary = lrModel.summary
println(s"Train MSE: ${trainingSummary.meanSquaredError}")
println(s"Train RMSE: ${trainingSummary.rootMeanSquaredError}")
println(s"Train MAE: ${trainingSummary.meanAbsoluteError}")
println(s"Train r2: ${trainingSummary.r2}")val predictions = lrModel.transform(test)// 计算精度
val evaluator = new RegressionEvaluator().setLabelCol("label").setPredictionCol("prediction").setMetricName("mse")
val accuracy = evaluator.evaluate(predictions)
print(s"Test MSE: ${accuracy}")

输出:

Train MSE: 101.57870147367461
Train RMSE: 10.078625971513905
Train MAE: 8.108865602095849
Train r2: 0.039467152584195975Test MSE: 114.28454406581636

2、分类评估指标

2.1 BinaryClassificationEvaluator

Evaluator for binary classification, which expects two input columns: rawPrediction and label. The rawPrediction column can be of type double (binary 0/1 prediction, or probability of label 1) or of type vector (length-2 vector of raw predictions, scores, or label probabilities).

评估指标支持以下几种:

val metricName: Param[String]
param for metric name in evaluation (supports "areaUnderROC" (default), "areaUnderPR")

Examples

import org.apache.spark.ml.classification.LogisticRegression
import org.apache.spark.ml.evaluation.BinaryClassificationEvaluator
import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator// Load training data
val data = spark.read.format("libsvm").load("/data1/software/spark/data/mllib/sample_libsvm_data.txt")val Array(train, test) = data.randomSplit(Array(0.8, 0.2))val lr = new LogisticRegression().setMaxIter(10).setRegParam(0.3).setElasticNetParam(0.8)// Fit the model
val lrModel = lr.fit(train)// Summarize the model over the training set and print out some metrics
val trainSummary = lrModel.summary
println(s"Train accuracy: ${trainSummary.accuracy}")
println(s"Train weightedPrecision: ${trainSummary.weightedPrecision}")
println(s"Train weightedRecall: ${trainSummary.weightedRecall}")
println(s"Train weightedFMeasure: ${trainSummary.weightedFMeasure}")val predictions = lrModel.transform(test)
predictions.show(5)// 模型评估
val evaluator = new BinaryClassificationEvaluator().setLabelCol("label").setRawPredictionCol("rawPrediction").setMetricName("areaUnderROC")
val auc = evaluator.evaluate(predictions)
print(s"Test AUC: ${auc}")val mulEvaluator = new MulticlassClassificationEvaluator().setLabelCol("label").setPredictionCol("prediction").setMetricName("weightedPrecision")
val precision = evaluator.evaluate(predictions)
print(s"Test weightedPrecision: ${precision}")

输出结果:

Train accuracy: 0.9873417721518988
Train weightedPrecision: 0.9876110961486668
Train weightedRecall: 0.9873417721518987
Train weightedFMeasure: 0.9873124561568825+-----+--------------------+--------------------+--------------------+----------+
|label|            features|       rawPrediction|         probability|prediction|
+-----+--------------------+--------------------+--------------------+----------+
|  0.0|(692,[122,123,148...|[0.29746771419036...|[0.57382336211209...|       0.0|
|  0.0|(692,[125,126,127...|[0.42262389447949...|[0.60411095396791...|       0.0|
|  0.0|(692,[126,127,128...|[0.74220898710237...|[0.67747871191347...|       0.0|
|  0.0|(692,[126,127,128...|[0.77729372618481...|[0.68509655708828...|       0.0|
|  0.0|(692,[127,128,129...|[0.70928896866149...|[0.67024402884354...|       0.0|
+-----+--------------------+--------------------+--------------------+----------+Test AUC: 1.0Test weightedPrecision: 1.0

2.2 MulticlassClassificationEvaluator

Evaluator for multiclass classification, which expects two input columns: prediction and label.

注:既然适用于多分类,当然适用于上面的二分类

评估指标支持如下几种:

val metricName: Param[String]
param for metric name in evaluation (supports "f1" (default), "weightedPrecision", "weightedRecall", "accuracy")

Examples

import org.apache.spark.ml.Pipeline
import org.apache.spark.ml.classification.DecisionTreeClassificationModel
import org.apache.spark.ml.classification.DecisionTreeClassifier
import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator
import org.apache.spark.ml.feature.{IndexToString, StringIndexer, VectorIndexer}// Load the data stored in LIBSVM format as a DataFrame.
val data = spark.read.format("libsvm").load("/data1/software/spark/data/mllib/sample_libsvm_data.txt")// Index labels, adding metadata to the label column.
// Fit on whole dataset to include all labels in index.
val labelIndexer = new StringIndexer().setInputCol("label").setOutputCol("indexedLabel").fit(data)
// Automatically identify categorical features, and index them.
val featureIndexer = new VectorIndexer().setInputCol("features").setOutputCol("indexedFeatures").setMaxCategories(4) // features with > 4 distinct values are treated as continuous..fit(data)// Split the data into training and test sets (30% held out for testing).
val Array(trainingData, testData) = data.randomSplit(Array(0.7, 0.3))// Train a DecisionTree model.
val dt = new DecisionTreeClassifier().setLabelCol("indexedLabel").setFeaturesCol("indexedFeatures")// Convert indexed labels back to original labels.
val labelConverter = new IndexToString().setInputCol("prediction").setOutputCol("predictedLabel").setLabels(labelIndexer.labels)// Chain indexers and tree in a Pipeline.
val pipeline = new Pipeline().setStages(Array(labelIndexer, featureIndexer, dt, labelConverter))// Train model. This also runs the indexers.
val model = pipeline.fit(trainingData)// Make predictions.
val predictions = model.transform(testData)// Select example rows to display.
predictions.select("predictedLabel", "label", "features").show(5)// Select (prediction, true label) and compute test error.
val evaluator = new MulticlassClassificationEvaluator().setLabelCol("indexedLabel").setPredictionCol("prediction").setMetricName("accuracy")
val accuracy = evaluator.evaluate(predictions)
println(s"Test Error = ${(1.0 - accuracy)}")

输出结果:

+--------------+-----+--------------------+
|predictedLabel|label|            features|
+--------------+-----+--------------------+
|           0.0|  0.0|(692,[95,96,97,12...|
|           0.0|  0.0|(692,[122,123,124...|
|           0.0|  0.0|(692,[122,123,148...|
|           0.0|  0.0|(692,[126,127,128...|
|           0.0|  0.0|(692,[126,127,128...|
+--------------+-----+--------------------+
only showing top 5 rowsTest Error = 0.040000000000000036

欢迎关注微信公众号

这篇关于Spark ML机器学习库评估指标示例的文章就介绍到这儿,希望我们推荐的文章对编程师们有所帮助!



http://www.chinasem.cn/article/1009351

相关文章

Python中logging模块用法示例总结

《Python中logging模块用法示例总结》在Python中logging模块是一个强大的日志记录工具,它允许用户将程序运行期间产生的日志信息输出到控制台或者写入到文件中,:本文主要介绍Pyt... 目录前言一. 基本使用1. 五种日志等级2.  设置报告等级3. 自定义格式4. C语言风格的格式化方法

Spring 中的切面与事务结合使用完整示例

《Spring中的切面与事务结合使用完整示例》本文给大家介绍Spring中的切面与事务结合使用完整示例,本文通过实例代码给大家介绍的非常详细,对大家的学习或工作具有一定的参考借鉴价值,需要的朋友参考... 目录 一、前置知识:Spring AOP 与 事务的关系 事务本质上就是一个“切面”二、核心组件三、完

MySQL8 密码强度评估与配置详解

《MySQL8密码强度评估与配置详解》MySQL8默认启用密码强度插件,实施MEDIUM策略(长度8、含数字/字母/特殊字符),支持动态调整与配置文件设置,推荐使用STRONG策略并定期更新密码以提... 目录一、mysql 8 密码强度评估机制1.核心插件:validate_password2.密码策略级

sky-take-out项目中Redis的使用示例详解

《sky-take-out项目中Redis的使用示例详解》SpringCache是Spring的缓存抽象层,通过注解简化缓存管理,支持Redis等提供者,适用于方法结果缓存、更新和删除操作,但无法实现... 目录Spring Cache主要特性核心注解1.@Cacheable2.@CachePut3.@Ca

QT Creator配置Kit的实现示例

《QTCreator配置Kit的实现示例》本文主要介绍了使用Qt5.12.12与VS2022时,因MSVC编译器版本不匹配及WindowsSDK缺失导致配置错误的问题解决,感兴趣的可以了解一下... 目录0、背景:qt5.12.12+vs2022一、症状:二、原因:(可以跳过,直奔后面的解决方法)三、解决方

MySQL中On duplicate key update的实现示例

《MySQL中Onduplicatekeyupdate的实现示例》ONDUPLICATEKEYUPDATE是一种MySQL的语法,它在插入新数据时,如果遇到唯一键冲突,则会执行更新操作,而不是抛... 目录1/ ON DUPLICATE KEY UPDATE的简介2/ ON DUPLICATE KEY UP

Python中Json和其他类型相互转换的实现示例

《Python中Json和其他类型相互转换的实现示例》本文介绍了在Python中使用json模块实现json数据与dict、object之间的高效转换,包括loads(),load(),dumps()... 项目中经常会用到json格式转为object对象、dict字典格式等。在此做个记录,方便后续用到该方

MySQL分库分表的实践示例

《MySQL分库分表的实践示例》MySQL分库分表适用于数据量大或并发压力高的场景,核心技术包括水平/垂直分片和分库,需应对分布式事务、跨库查询等挑战,通过中间件和解决方案实现,最佳实践为合理策略、备... 目录一、分库分表的触发条件1.1 数据量阈值1.2 并发压力二、分库分表的核心技术模块2.1 水平分

SpringBoot请求参数传递与接收示例详解

《SpringBoot请求参数传递与接收示例详解》本文给大家介绍SpringBoot请求参数传递与接收示例详解,本文通过实例代码给大家介绍的非常详细,对大家的学习或工作具有一定的参考借鉴价值,需要的朋... 目录I. 基础参数传递i.查询参数(Query Parameters)ii.路径参数(Path Va

RabbitMQ 延时队列插件安装与使用示例详解(基于 Delayed Message Plugin)

《RabbitMQ延时队列插件安装与使用示例详解(基于DelayedMessagePlugin)》本文详解RabbitMQ通过安装rabbitmq_delayed_message_exchan... 目录 一、什么是 RabbitMQ 延时队列? 二、安装前准备✅ RabbitMQ 环境要求 三、安装延时队