ValueError: Input contains NaN, infinity or a value too large for dtype(‘float64‘)

╰半夏微凉° 2022-11-07 04:23 353阅读 0赞

问题

刚开始学习 sklearn ，运行下面的代码时报错，

  1. from sklearn.feature_extraction import DictVectorizer
  2. from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer
  3. from  sklearn.preprocessing import  MinMaxScaler,StandardScaler ,Normalizer
  4. from sklearn.impute import SimpleImputer
  5. import  numpy as np
  7. import jieba
  9. def im():
  10.     """ 缺失值处理 :return: """
  11.     im  = SimpleImputer(missing_values='NaN',strategy='mean')
  12.     data = im.fit_transform([[1,2],[np.nan,3],[7,6]])
  13.     print(data)
  15. if __name__ == "__main__":   
  16.     im()

运行报错，

ValueError: Input contains NaN, infinity or a value too large for dtype(‘float64’)。具体如下：

  1. Traceback (most recent call last):
  2.   File "E:/pycharm_workspace/matplotlibDemo/feature.py", line 104, in <module>
  3.     im()
  4.   File "E:/pycharm_workspace/matplotlibDemo/feature.py", line 95, in im
  5.     data = im.fit_transform([[1,2],[np.nan,3],[7,6]])
  6.   File "D:\skl3\lib\site-packages\sklearn\base.py", line 699, in fit_transform
  7.     return self.fit(X, **fit_params).transform(X)
  8.   File "D:\skl3\lib\site-packages\sklearn\impute\_base.py", line 288, in fit
  9.     X = self._validate_input(X, in_fit=True)
  10.   File "D:\skl3\lib\site-packages\sklearn\impute\_base.py", line 262, in _validate_input
  11.     raise ve
  12.   File "D:\skl3\lib\site-packages\sklearn\impute\_base.py", line 255, in _validate_input
  13.     copy=self.copy)
  14.   File "D:\skl3\lib\site-packages\sklearn\base.py", line 421, in _validate_data
  15.     X = check_array(X, **check_params)
  16.   File "D:\skl3\lib\site-packages\sklearn\utils\validation.py", line 63, in inner_f
  17.     return f(*args, **kwargs)
  18.   File "D:\skl3\lib\site-packages\sklearn\utils\validation.py", line 664, in check_array
  19.     allow_nan=force_all_finite == 'allow-nan')
  20.   File "D:\skl3\lib\site-packages\sklearn\utils\validation.py", line 106, in _assert_all_finite
  21.     msg_dtype if msg_dtype is not None else X.dtype)
  23. ValueError: Input contains NaN, infinity or a value too large for dtype('float64').

Input contains NaN, infinity or a value too large for dtype('float64') 表示 Input 的值包含太长了。

解决方法

  1. from sklearn.feature_extraction import DictVectorizer
  2. from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer
  3. from  sklearn.preprocessing import  MinMaxScaler,StandardScaler ,Normalizer
  4. from sklearn.impute import SimpleImputer
  5. import  numpy as np
  7. import jieba
  9. def im():
  10.     """ 缺失值处理 :return: """
  11.     im  = SimpleImputer(missing_values=np.nan,strategy='most_frequent')
  12.     data = im.fit_transform([[1,2],[np.nan,3],[7,6]])
  13.     print(data)
  15. if __name__ == "__main__":   
  16.     im()

运行结果报错：

  1. [[1. 2.]
  2.  [1. 3.]
  3.  [7. 6.]]

发表评论

表情：
评论列表 （有 0 条评论，353人围观）

还没有评论，来说两句吧...

相关阅读