pandas.read_excel 列作为具有 NA 值的 int64

Question

I'm trying to import an excel spreadsheet onto pandas using:

df= pd.read_excel(excel_file)

This reads integer columns fine as int64, as long as they don't have missing or nan values. When they do have nan values it is read as float64.

I have tried using:

df= pd.read_excel(excel_file, converters={'column_x': np.int64, 'column_y': np.int64})

and

df= pd.read_excel(excel_file, dtype={'column_x': np.int64, 'column_y': np.int64})

I'd like to keep the missing values as nan but the other values as int. Is this possible?

Thanks

Answer 1

是的，在带有Nullable 整数数据类型的Pandas 0.24+ 中：

df= pd.read_excel(excel_file, dtype={'column_x': 'Int64', 'column_y': 'Int64'})