ValueError: cannot convert float NaN to integer"

Question

在下面的代码中，我有数据框，我希望如果某个条件满足该条件恰好被 nan 行替换的行，但它不起作用，当我试图查看它是如何工作时逐一我收到错误

```{x_r = data.drop(['y'], axis="columns")
    x_nan = np.empty((len(x_r), len(x_r.columns)))
    x_nan[:] = np.nan
    x_2 = x_r.values
    for i in range(0, len(x_r)):
    if y[i] < q_:
    x_2[i, :] = x_nan[i, :]}```

然后是个人路线

```{x_2[0,0]=x_nan[0,0]}```

它给了我以下错误：

 ```{"exec(code_obj, self.user_global_ns, self.user_ns)
   File "<ipython-input-42-d9db7fa21a9c>", line 1, in <module>
   x_2[0,0]=x_nan[0,0]
   ValueError: cannot convert float NaN to integer"}```

Answer 1

那是因为nan被识别为float数组的特殊字符（一种特殊的float），显然你的x_2数组是int类型; nan 无法转换为 int 以适合您的数组。

解决方法是将数组转换为浮点数：

>>>import numpy as np
>>>a = np.array([[1,2,3], [4,5,6]])
>>>a
array([[1, 2, 3],
       [4, 5, 6]])

>>>a[0,0] = np.nan
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-43-8e6dceaece5a> in <module>
----> 1 a[0,0] = np.nan

ValueError: cannot convert float NaN to integer

>>>a = a.astype('float32')
>>>a[0,0] = np.nan
>>>a
array([[nan,  2.,  3.],
   [ 4.,  5.,  6.]], dtype=float32)

ValueError: cannot convert float NaN to integer"

ValueError: cannot convert float NaN to integer"

python

machine-learning

pycharm

supervised-learning