在pyspark中使用logistic回归分析特征重要性

from pyspark.ml.classification import LogisticRegression
from pyspark.ml.feature import VectorAssembler

# 假设数据集已经加载到一个DataFrame中，包含特征列features和目标列label
df = spark.read.csv("data.csv", header=True, inferSchema=True)

# 将特征列合并为一个向量列
assembler = VectorAssembler(inputCols=df.columns[:-1], outputCol="features")
df = assembler.transform(df)

# 创建Logistic回归模型对象
lr = LogisticRegression(featuresCol="features", labelCol="label")

# 拟合模型
model = lr.fit(df)

# 获取特征重要性
importance = model.coefficients

# 将特征重要性与特征名称对应起来
feature_importance = list(zip(df.columns[:-1], importance))

# 按照特征重要性降序排序
feature_importance.sort(key=lambda x: abs(x[1]), reverse=True)