Question

我的函数遇到以下回溯错误：

TypeError: unsupported operand type(s) for &: 'str' and 'str'

这是我的代码：

def age():
    thirties_df = (df.loc[df['age'] <= 39]) & (df.loc[df['age'] >= 30])
    fourties_df = (df.loc[df['age'] <= 49]) & (df.loc[df['age'] >= 40])
    fifties_df = (df.loc[df['age'] <= 59]) & (df.loc[df['age'] >= 50])
    sixties_df = (df.loc[df['age'] <= 69]) & (df.loc[df['age'] >= 60])
    seventies_df = (df.loc[df['age'] <= 79]) & (df.loc[df['age'] >= 70])
    eighties_df = (df.loc[df['age'] <= 89]) & (df.loc[df['age'] >= 80])
    for i in thirties_df, fourties_df, fifties_df, sixties_df, 
    \ seventies_df, eighties_df:

Answer 1

错误与在实例化这些变量时使用可疑的'＆'字符有关。（这两个字符串之间的'＆'毫无意义，但会产生错误）

另外，放and也没有意义（我想您是在尝试放这个）在字符串之间，因为最后一个赋值用作该变量的值。看到这个：-

>>> var = 'str1' and 'str2'
>>> var
'str2'

Answer 2

没有 df 很难说，但看起来你的年龄列是字符串。括号错位：df['age'] <= 39 是一系列 dtype bool。并且 2 Series of bool 的 & 本身就是一个 Series of bool，您可以使用它在 DataFrame 上进行选择 ^{_{（如果长度合适）}}

df = pd.DataFrame(
    map(str, np.random.randint(5, 90, size=1000)),
    columns=('age',)
)  # what I think your df roughly looks like.
df['age'] = df['age'].astype('int')  # make sure age is numeric
# building thirties_df step by step
younger_than_forty = df['age'] <= 39
older_than_thirty = df['age'] >= 30
thirties_df = df[younger_than_forty & older_than_thirty]
# or all at once and using loc like in your question
thirties_df = df.loc[(df['age'] <= 39) & (df['age'] >= 30)]
# or using between
thirties_df = df[df['age'].between(30, 39)]

# because
assert all(df['age'].between(30, 39) == younger_than_forty & older_than_thirty)

# so age could become something like
def age(df):
    min_age = 30
    max_age = 80
    buckets = (
        df[df["age"].between(n, n + 9)]
        for n in range(min_age, max_age + 1, 10)
    )
    for i in buckets:
        print(i)

TypeError：＆不支持的操作数类型：“ str”和“ str”

2 个答案: