关于python:attributeError:’str’对象没有属性’isNumeric’

AttributeError: 'str' object has no attribute 'isnumeric'

有点困惑,因为我是肯定的,我以前就有过这样的工作。

我创建了以下方法…

1
2
3
4
5
6
7
def p2f(x):
    if x.strip('%').isnumeric():
        return float(x.strip('%'))/100
    elif x in ['SUPP', 'NEW', 'LOWCOV', 'NA', '']:
        return 0.0
    else:
        return x

但是当我在导入的csv文件上运行它时,会产生以下错误:

1
AttributeError: 'str' object has no attribute 'isnumeric'

虽然我可以看到isnumeric是文档中str的一个属性…

https://pandas.pydata.org/pandas-docs/stable/generated/pandas.series.str.isnumeric.html?highlight=isnumeric pandas.series.str.isnumeric

除非我没有正确解释这些信息?


str.isnumeric()仅在python3上可用。错误表明您使用的是python2,其中只存在unicode.isnumeric()

您应该使用str.isdecimal(),或者更好的方法是使用异常处理:

1
2
3
4
5
def p2f(x):
    try:
        return float(x.strip('%'))/100
    except ValueError:
        return 0.0 if x in ('SUPP', 'NEW', 'LOWCOV', 'NA', '') else x

.isnumeric()float()不接受的BMP中430个unicode码点匹配,并且有.isdigit()返回true的码点,因此也不可转换。

您可以生成自己的表来检查:

1
2
3
4
5
6
7
8
9
for i in range(2 ** 16):
    c = chr(i)
    if c.isnumeric() or c.isdigit() or c.isdecimal():
        try:
            f = float(c)
        except ValueError:
            f = '<not convertible>'
        di, de, nu = ('\u2705' if test() else '\u274c' for test in (c.isdigit, c.isdecimal, c.isnumeric))
        print(f'{c!a:<6} {c}\tdigit: {di}   decimal: {de}   numeric: {nu}  float: {f}')

它产生如下输出:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
'0'    0    digit: ?   decimal: ?   numeric: ?  float: 0.0
'1'    1    digit: ?   decimal: ?   numeric: ?  float: 1.0
'2'    2    digit: ?   decimal: ?   numeric: ?  float: 2.0
'3'    3    digit: ?   decimal: ?   numeric: ?  float: 3.0
'4'    4    digit: ?   decimal: ?   numeric: ?  float: 4.0
'5'    5    digit: ?   decimal: ?   numeric: ?  float: 5.0
'6'    6    digit: ?   decimal: ?   numeric: ?  float: 6.0
'7'    7    digit: ?   decimal: ?   numeric: ?  float: 7.0
'8'    8    digit: ?   decimal: ?   numeric: ?  float: 8.0
'9'    9    digit: ?   decimal: ?   numeric: ?  float: 9.0
'\xb2' 2    digit: ?   decimal: ?   numeric: ?  float: <not convertible>
'\xb3' 3    digit: ?   decimal: ?   numeric: ?  float: <not convertible>
'\xb9' 1    digit: ?   decimal: ?   numeric: ?  float: <not convertible>
'\xbc' ?    digit: ?   decimal: ?   numeric: ?  float: <not convertible>
'\xbd' ?    digit: ?   decimal: ?   numeric: ?  float: <not convertible>
'\xbe' ?    digit: ?   decimal: ?   numeric: ?  float: <not convertible>

你会发现只有decimal列有所有不可转换代码点的十字。

如果您想在python2中使用isdecimal(),就必须先将字节串解码为unicode。