Adding data to Pandas Dataframe from a CSV file causing Value Errors

Question

I am trying to add an int to an existing value in a Pandas DataFrame with

>>> df.ix['index 5','Total Dollars'] += 10

I get the error: ValueError: Must have equal len keys and value when setting with an iterable.

I think the error comes from the datatype as gotten from:

>>> print type(df.ix['index 5','Total Dollars'] 
int64 <class 'pandas.core.series.Series'>

The dataframe is populated via CSV file. I tried loading the database from another CSV file:

>>> print type(df.ix['index 5','Total Dollars']
int64

What could be causing this difference in type?

Answer 1

This looks like a bug for some earlier pandas versions, fixed at least with 0.16.2 if not earlier as discussed here and here.

With 0.17.1, this works fine:

df = pd.DataFrame(data=[5], columns=['Total Dollars'], index=['index 5'])

         Total Dollars
index 5              5

df.ix['index 5', 'Total Dollars'] += 10

         Total Dollars
index 5             15