值错误：无法将字符串转换为浮点数：''

Question

我尝试转换company_size 列中的值，并尝试了不同的方法，但仍然得到 ValueError。有人有什么建议吗？

代码如下：

handle_df.loc[handle_df['company_size'].isin(['I prefer not to answer',"I don't know"]),['company_size']]=np.NaN
handle_df.loc[handle_df['company_size'] == 'Fortune 1000 (1,000+)' ,['company_size']]='1000'
handle_df.loc[handle_df['company_size'] == '1/25/2013' ,['company_size']]='1-25'
handle_df.loc[handle_df['company_size'] == '1/5/2014' ,['company_size']]='1-5'
handle_df.loc[handle_df['company_size'] == '6/15/2014' ,['company_size']]='6-15'

handle_df.loc[handle_df['company_size'].isin(['Student',"Other (not working, consultant, etc.)"]),['company_size']]='1'

handle_df['company_size'] = handle_df['company_size'].map(lambda x:str(x).replace(" to ",'-'))

handle_df = process_data_range(handle_df,'company_size',dropna=False)

完整的错误回溯

ValueError: could not convert string to float: ''
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-22-84fdd9e45198> in <module>
      9 handle_df['company_size'] = handle_df['company_size'].map(lambda x:str(x).replace(" to ",'-'))
     10 
---> 11 handle_df = process_data_range(handle_df,'company_size',dropna=False)
     12 

<ipython-input-11-510bc7e7040d> in process_data_range(df, feature, dropna)
     28     # for na value,I mark them as 'NaN'
     29     change_to_midpoint = lambda x: np.average([float(i) for i in x.split("-")]) if "-" in x else np.NaN if x == "" else float(x)
---> 30     df[feature] = df[feature].map(change_to_midpoint)
     31 
     32     if dropna:

~\anaconda3\lib\site-packages\pandas\core\series.py in map(self, arg, na_action)
   3907         January     History     Final exam      A
   3908         February    Geography   Final exam      B
-> 3909         March       History     Coursework      A
   3910         April       Geography   Coursework      C
   3911         dtype: object

~\anaconda3\lib\site-packages\pandas\core\base.py in _map_values(self, mapper, na_action)
    935         **bins**
    936 
--> 937         Bins can be useful for going from a continuous variable to a
    938         categorical variable; instead of counting unique
    939         apparitions of values, divide the index in the specified

~\anaconda3\lib\site-packages\pandas\_libs\lib.pyx in pandas._libs.lib.map_infer()

<ipython-input-11-510bc7e7040d> in <lambda>(x)
     27     # if age range ,change it to the midpoint
     28     # for na value,I mark them as 'NaN'
---> 29     change_to_midpoint = lambda x: np.average([float(i) for i in x.split("-")]) if "-" in x else np.NaN if x == "" else float(x)
     30     df[feature] = df[feature].map(change_to_midpoint)
     31 

<ipython-input-11-510bc7e7040d> in <listcomp>(.0)
     27     # if age range ,change it to the midpoint
     28     # for na value,I mark them as 'NaN'
---> 29     change_to_midpoint = lambda x: np.average([float(i) for i in x.split("-")]) if "-" in x else np.NaN if x == "" else float(x)
     30     df[feature] = df[feature].map(change_to_midpoint)
     31 

ValueError: could not convert string to float: ''

Answer 1

检查您是否有像“-”、“nnn-”或“-nnn”这样的数据，其中 nnn 是任意数字。正如 0x5453 所说，您正在尝试对来自数据的内容执行 float('')。