I have a dataframe with a date column that is represented as ndarray object. I want to clusterize dataframe and I decided to convert each string (like '2009-01-04') in this column to int. How can I represent it as a recency in days considering the last date as today's date? I got some problems with "to_datetime" method.
index col1 col2 date
0 a 34 '2009-01-04'
1 a 21 '2009-01-05'
2 b 8 '2009-01-06'
Thank you