具有多个标签的逻辑回归 PySpark MLlib 问题

Question

我正在尝试创建一个 LogisticRegression 模型 (LogisticRegressionWithSGD)，但它得到一个错误

org.apache.spark.SparkException: Input validation failed.

如果我给它二进制输入（0,1 而不是 0,1,2）它确实会成功。

示例输入：

parsed_data = [LabeledPoint(0.0, [4.6,3.6,1.0,0.2]),
LabeledPoint(0.0, [5.7,4.4,1.5,0.4]),
LabeledPoint(1.0, [6.7,3.1,4.4,1.4]),
LabeledPoint(0.0, [4.8,3.4,1.6,0.2]),
LabeledPoint(2.0, [4.4,3.2,1.3,0.2])]

代码： model = LogisticRegressionWithSGD.train(parsed_data)

Spark 中的逻辑回归模型是否应该仅用于二元分类？

Answer 1

虽然从文档中不清楚（您必须深入了解source code 才能实现它），LogisticRegressionWithSGD 仅适用于二进制数据；对于多项回归，您应该使用LogisticRegressionWithLBFGS:

 from pyspark.mllib.classification import LogisticRegressionWithLBFGS, LogisticRegressionModel, LogisticRegressionWithSGD
 from pyspark.mllib.regression import LabeledPoint
 parsed_data = [LabeledPoint(0.0, [4.6,3.6,1.0,0.2]),
                LabeledPoint(0.0, [5.7,4.4,1.5,0.4]),
                LabeledPoint(1.0, [6.7,3.1,4.4,1.4]),
                LabeledPoint(0.0, [4.8,3.4,1.6,0.2]),
                LabeledPoint(2.0, [4.4,3.2,1.3,0.2])]     

 model = LogisticRegressionWithSGD.train(sc.parallelize(parsed_data)) # gives error:
 # org.apache.spark.SparkException: Input validation failed.

 model = LogisticRegressionWithLBFGS.train(sc.parallelize(parsed_data), numClasses=3)  # works OK