Replacing cell of a dataframe in a loop

Question

I was working on this question filter iteration using FOR. I wonder how to replace the last cell of year column in each csv file generated. Lets say I want to replace the last cell (of column year) of each cvs file by current year (2018). I did the following code:

for i, g in df.sort_values("year").groupby("Univers", sort=False):
    for y in g.iloc[-1, g.columns.get_loc('year')]:
        y = 2018
    g.to_csv('{}.xls'.format(i))

But I get the same column with any changes. Any ideas how to do this?

Tai · Answer 1 · 2018-01-03T23:13:26.030

The problem seems to be of two fold: first find the index of the last row and then replace it at (last_row_idx, "year").

Try this

for i, g in df.sort_values("year").groupby("Univers", sort=False):
    last_row_idx = g.tail(1).index.item()  # first task: find index
    g.at[last_row_idx, "year"] = 2018      # replace
    g.to_csv('{}.xls'.format(i))

Alternatively, one can also use g.set_value(last_row_idx, "year", 2018) to set value at a particular cell.