I am trying to create a dataframe in Pandas from the AB column in my csv file. (AB is the 27th column).
I am using this line:
df = pd.read_csv(filename, error_bad_lines = False, usecols = [27])
... which is resulting in this error:
ValueError: Usecols do not match names.
I'm very new to Pandas, could someone point out what i'm doing wrong to me?
Here is a small demo:
CSV file (without header, i.e. there is NO column names):
1,2,3,4,5,6,7,8,9,10
11,12,13,14,15,16,17,18,19,20
We are going to read only 8-th column:
In [1]: fn = r'D:\temp\.data\1.csv'
In [2]: df = pd.read_csv(fn, header=None, usecols=[7], names=['col8'])
In [3]: df
Out[3]:
col8
0 8
1 18
PS pay attention at header=None, usecols=[7], names=['col8']
If you don't use header=None and names parameters, the first row will be used as a header:
In [6]: df = pd.read_csv(fn, usecols=[7])
In [7]: df
Out[7]:
8
0 18
In [8]: df.columns
Out[8]: Index(['8'], dtype='object')
and if we want to read only the last 10-th column:
In [9]: df = pd.read_csv(fn, usecols=[10])
... skipped ...
ValueError: Usecols do not match names.
because pandas counts columns starting from 0, so we have to do it this way:
In [12]: df = pd.read_csv(fn, usecols=[9], names=['col10'])
In [13]: df
Out[13]:
col10
0 10
1 20
usecols uses the column name in your csv file rather than the column number. in your case it should be usecols=['AB'] rather than usecols=[28] that is the reason of your error stating usecols do not match names.
usecols supports both positional column indexes or column names. From docs: All elements in this array must either be positional (i.e. integer indices into the document columns) or strings that correspond to column names provided either by the user in names or inferred from the document header row(s)
usecols=['AB']to avoid all that confusion.df = pd.read_csv(filename, error_bad_lines = False, usecols = ['AB'])the error is still the same.