时间戳

首页 > 解决方案 > 从决策树回归器拟合训练数据会导致崩溃

python - 从决策树回归器拟合训练数据会导致崩溃

问题描述

尝试在一些训练数据上实现决策树回归算法，但是当我调用 fit() 时出现错误。

    (trainingData, testData) = data.randomSplit([0.7, 0.3])

    vecAssembler = VectorAssembler(inputCols=["_1", "_2", "_3", "_4", "_5", "_6", "_7", "_8", "_9", "_10"], outputCol="features")

    dt = DecisionTreeRegressor(featuresCol="features", labelCol="_11")

    dt_model = dt.fit(trainingData)

产生错误

  File "spark.py", line 100, in <module>
    main()
  File "spark.py", line 87, in main
    dt_model = dt.fit(trainingData)
  File "/opt/spark/python/pyspark/ml/base.py", line 132, in fit
    return self._fit(dataset)
  File "/opt/spark/python/pyspark/ml/wrapper.py", line 295, in _fit
    java_model = self._fit_java(dataset)
  File "/opt/spark/python/pyspark/ml/wrapper.py", line 292, in _fit_java
    return self._java_obj.fit(dataset._jdf)
  File "/opt/spark/python/lib/py4j-0.10.7-src.zip/py4j/java_gateway.py", line 1257, in __call__
  File "/opt/spark/python/pyspark/sql/utils.py", line 79, in deco
    raise IllegalArgumentException(s.split(': ', 1)[1], stackTrace)
pyspark.sql.utils.IllegalArgumentException: u'requirement failed: Column features must be of type struct<type:tinyint,size:int,indices:array<int>,values:array<double>> but was actually struct<type:tinyint,size:int,indices:array<int>,values:array<double>>.'

但数据结构完全相同。

标签： pythonapache-sparkpyspark

解决方案

您缺少两个步骤。1. 转换部分，以及 2. 从转换后的数据中选择特征和标签。我假设数据只包含数字数据，即没有分类数据。我将写下一个通用的训练模型的流程pyspark.ml来帮助你。

from pyspark.ml.feature
from pyspark.ml.classification import DecisionTreeClassifier

#date processing part

vecAssembler = VectorAssembler(input_cols=['col_1','col_2',...,'col_10'],outputCol='features')

#you missed these two steps
trans_data = vecAssembler.transform(data)

final_data = trans_data.select('features','col_11') #your label column name is col_11

train_data, test_data = final_data.randomSplit([0.7,0.3])

#ml part

dt = DecisionTreeClassifier(featuresCol='features',labelCol='col_11')

dt_model = dt.fit(train_data)

dt_predictions = dt_model.transform(test_data)

#proceed with the model evaluation part after this

推荐阅读