Numpy ValueError:无法将字符串转换为浮点：'；2019278'；-潜在的非法字符或格式？

我在从对象到“float64”的格式转换中遇到问题。我不能执行任何矩阵运算，直到这种转换发生，一般来说，这是一个主要的烦恼，因为我陷入杂草

它可能与在原始格式的数组中包含空格或空格和撇号有关

原始数组如下所示：

array([[' 2,019,278 ', ' 14,569,743 ', ' 14,116,057 ', ' 30,705,078 '],
   [' 514,049 ', ' 3,301,330 ', ' 1,775,624 ', ' 5,591,003 '],
   [' 40,364 ', ' 283,894 ', ' 138,342 ', ' 462,600 '],
   [' 77,849 ', ' 528,504 ', ' 665,829 ', ' 1,272,182 '],
   [' 4,534 ', ' 39,282 ', ' 23,902 ', ' 67,718 '],
   [' 182,313 ', ' 795,102 ', ' 369,981 ', ' 1,347,396 '],
   [' 256,867 ', ' 694,895 ', ' 240,025 ', ' 1,191,787 '],
   [' 1,527,829 ', ' 12,690,612 ', ' 12,968,625 ', ' 27,187,066 '],
   [' 771,142 ', ' 2,937,748 ', ' 1,612,082 ', ' 5,320,972 ']],
  dtype=object)

a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
a = a.astype('float64')

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-18-ba9570a13fdd> in <module>
      1 a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
      2 #b = pd.DataFrame.to_numpy(df_By_Race_Total, dtype=None, copy=False)
----> 3 a = a.astype('float64')
      4 #b = b.astype('float64')
      5 #b = np.reciprocal(b)

ValueError: could not convert string to float: ' 2,019,278 '

我的代码如下所示：

array([[' 2,019,278 ', ' 14,569,743 ', ' 14,116,057 ', ' 30,705,078 '],
   [' 514,049 ', ' 3,301,330 ', ' 1,775,624 ', ' 5,591,003 '],
   [' 40,364 ', ' 283,894 ', ' 138,342 ', ' 462,600 '],
   [' 77,849 ', ' 528,504 ', ' 665,829 ', ' 1,272,182 '],
   [' 4,534 ', ' 39,282 ', ' 23,902 ', ' 67,718 '],
   [' 182,313 ', ' 795,102 ', ' 369,981 ', ' 1,347,396 '],
   [' 256,867 ', ' 694,895 ', ' 240,025 ', ' 1,191,787 '],
   [' 1,527,829 ', ' 12,690,612 ', ' 12,968,625 ', ' 27,187,066 '],
   [' 771,142 ', ' 2,937,748 ', ' 1,612,082 ', ' 5,320,972 ']],
  dtype=object)

a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
a = a.astype('float64')

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-18-ba9570a13fdd> in <module>
      1 a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
      2 #b = pd.DataFrame.to_numpy(df_By_Race_Total, dtype=None, copy=False)
----> 3 a = a.astype('float64')
      4 #b = b.astype('float64')
      5 #b = np.reciprocal(b)

ValueError: could not convert string to float: ' 2,019,278 '

这是我收到的错误：

array([[' 2,019,278 ', ' 14,569,743 ', ' 14,116,057 ', ' 30,705,078 '],
   [' 514,049 ', ' 3,301,330 ', ' 1,775,624 ', ' 5,591,003 '],
   [' 40,364 ', ' 283,894 ', ' 138,342 ', ' 462,600 '],
   [' 77,849 ', ' 528,504 ', ' 665,829 ', ' 1,272,182 '],
   [' 4,534 ', ' 39,282 ', ' 23,902 ', ' 67,718 '],
   [' 182,313 ', ' 795,102 ', ' 369,981 ', ' 1,347,396 '],
   [' 256,867 ', ' 694,895 ', ' 240,025 ', ' 1,191,787 '],
   [' 1,527,829 ', ' 12,690,612 ', ' 12,968,625 ', ' 27,187,066 '],
   [' 771,142 ', ' 2,937,748 ', ' 1,612,082 ', ' 5,320,972 ']],
  dtype=object)

a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
a = a.astype('float64')

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-18-ba9570a13fdd> in <module>
      1 a = pd.DataFrame.to_numpy(df_By_Race_Disability, dtype=None, copy=False)
      2 #b = pd.DataFrame.to_numpy(df_By_Race_Total, dtype=None, copy=False)
----> 3 a = a.astype('float64')
      4 #b = b.astype('float64')
      5 #b = np.reciprocal(b)

ValueError: could not convert string to float: ' 2,019,278 '

---------------------------------------------------------------------------
ValueError回溯（最近一次调用上次）
在里面
1 a=pd.DataFrame.to\u numpy（df\u By\u Race\u Disability，dtype=None，copy=False）
2#b=pd.DataFrame.to_numpy（df_By_Race_Total，dtype=None，copy=False）
---->3 a=a.astype（'float64'）
4#b=b.astype（'float64'）
5#b=np.倒数（b）
ValueError:无法将字符串转换为浮点：“2019278”

首先执行以下操作：

df_temp = df_By_Race_Disability.applymap(lambda x: x.replace(',', ''))

然后你可以继续你的代码

a = pd.DataFrame.to_numpy(df_temp, dtype=None, copy=False)
a = a.astype('float64')

首先要做到：

df_temp = df_By_Race_Disability.applymap(lambda x: x.replace(',', ''))

然后你可以继续你的代码

a = pd.DataFrame.to_numpy(df_temp, dtype=None, copy=False)
a = a.astype('float64')

---

字符串中的逗号

，

造成问题，请尝试使用

.apply（）

替换逗号，然后转换字符串中的逗号

，

造成问题，尝试使用

.apply（）

替换逗号，然后转换