Sum with substr in Python

[Scott]

Hi everyone,

I am a SAS user and it has a very useful function call substr. I cannot find an equivalent for Pandas or Python.

I am trying to sum a variable where another string variables first two letters = '08'

my attempt is below:

1	data.loc[data.Hs_code[0:1] == '08'].Imports_vfd.sum()

My error:TypeError: invalid type comparison

Any help is appreciated
Thanks

[ichabod801]

Note that SAS's substr(text, 1, 2) is equivalent to text[0:2] in Python, or more simply text[:2]. But since strings are sequences of characters in Python, and dataframe columns are sequences in pandas, You are really getting a slice of the column Hs_code.

The way around this is to use the str method of the column:

1	data.loc[data.Hs_code.str[:2] == '08'].Imports_vfd.sum()

[Scott]

I tried that and got this error:

AttributeError: Can only use .str accessor with string values, which use np.object_ dtype in pandas

[ThomasL]

Your column or series Hs_code has not the type object