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

笔者在使用LogisticRegression模型进行预测时，报错
Traceback (most recent call last):
File “D:/软件（学习）/Python/MachineLearing/taitannike/train.py”, line 55, in
predicted_np = clf.predict(test_np)
File “D:\Python\Anaconda\lib\site-packages\sklearn\linear_model\base.py”, line 281, in predict
scores = self.decision_function(X)
File “D:\Python\Anaconda\lib\site-packages\sklearn\linear_model\base.py”, line 257, in decision_function
X = check_array(X, accept_sparse=‘csr’)
File “D:\Python\Anaconda\lib\site-packages\sklearn\utils\validation.py”, line 573, in check_array
allow_nan=force_all_finite == ‘allow-nan’)
File “D:\Python\Anaconda\lib\site-packages\sklearn\utils\validation.py”, line 56, in _assert_all_finite
raise ValueError(msg_err.format(type_err, X.dtype))
ValueError: Input contains NaN, infinity or a value too large for dtype(‘float64’).
Age False

问题：pandas在处理数据时出现以下错误

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

解决方法：

1、检查数据中是否有缺失值

例如，读取得到的原始数据如下
读取数据

data_test = pd.read_csv('test.csv')

检查数据中是否有缺失值

print(np.isnan(data_test).any())

Flase:表示对应特征的特征值中无缺失值
True：表示有缺失值

2、删除有缺失值的行

train.dropna(inplace=True)

然后再看数据中是否有缺失值
也可以根据需要对缺失值进行填充处理：
train.fillna(‘100’)