Python Pandas Dataframe.at function

Question

using

a = df.at('x', 'y')

should this return the value of the cell in the row containing 'x' and the column header 'y'?

should it matter that the row containing 'x' is not either the name of the row or in the first column?

If it doesn't matter can anyone suggest what might be the problem here.

Here is some sample data

     Account   Asset Symbol SecType Currency  Position  Avg cost
0  DU2870371  USDCAD    USD    CASH      CAD -232475.0  1.286397
2  DU2870371  AUDUSD    AUD    CASH      USD  285611.0  0.756396

I have then saved this to a csv. In another script I have loaded it from CSV and using

a = df.at['USDCAD', 'Position']

I would expect it to return the value -232475.0 but instead is says KeyError: 'USDCAD'

Answer 1

Connvert column Asset to index for select by labels in index (from Asset column) and columns names:

df = df.set_index('Asset')
a = df.at['USDCAD', 'Position']

Also is possible use boolean indexing , but output is empty, one or more values in Series :

s = df.loc[df['Asset'].eq('USDCAD'), 'Position']

So for first value of Series is possible use:

a = s.iat[0]