如何通过特定列的最大分组值过滤python数据框

Question

So I am try to get a df that shows the latest Price based on a grouped dataframe. Below is a df of 9 records to help explain.

         Country   ISO Month  Price
0        Germany   DE  201311 7.99     
1        Germany   DE  201310 8.99
2        Germany   DE  201309 6.99
3   United States  US  201310 4.99
4   United States  US  201309 11.99
5   United States  US  201308 2.99
7         France   FR  201311 7.99
8         France   FR  201310 1.99
9         France   FR  201309 1.50

I want the df to return the price that relates to the maximum Month. Like to return the below view:

         Country   ISO Month  Price
0        Germany   DE  201311 7.99     
1   United States  US  201310 4.99
2         France   FR  201311 7.99

I tried doing a groupby with a max() and could return the maximum month by doing the below

df.groupby(['Country','ISO'])['Month'].max().reset_index()

but the above dropped the Price out that relates to that month, and if included the 'Price' in the groupby it obviously then return all the 3 values per country as they are different values.

Sorry I couldn't find the answer i was exactly looking for after 2 hours on the net.

Answer 1

Is the data stored in mysql? If the data is in a list like[(0,"Ger..","DE",201311,7.99),(....] then your goal could be achived by this code:

the_list.sort(key=lambda x:x[3], reverse=True)[:3]

returns the data from first three largest month.

Answer 2

Ok, using the sort and first in a groupby seemed to get the result I need.

df.sort('Month', ascending=False).groupby(['Country','ISO'], as_index=False).first()

would return:

         Country   ISO Month  Price
0        Germany   DE  201311 7.99     
1   United States  US  201310 4.99
2         France   FR  201311 7.99