Reputation: 21
Am having a CSV file as input, In which one of the column has date format as Year-Month. I need to change the column format as Year-Month-Date Format, which implies month end date. Am using Python 3. Further , Also am using the modified column in aggregation function and group by function.
Eg: 2020-01
2020-02
2020-03
2020-04
Expected Result :
2020-01-31
2020-02-29 ( Considering leap Year )
2020-03-31
2020-04-30
so on...
Upvotes: 0
Views: 353
Reputation: 21
import pandas as pd
import datetime as dt
import calendar as cal
# making data frame from csv file
df = pd.read_csv("Table.csv")
# (month_year)->Column Name
df['month_last_dates'] = [datetime.datetime(date.year, date.month,
calendar.monthrange(date.year, date.month)[1]) for date in month_year]
# defining aggregation process for each column
aggregations={ 'Metric':sum }
# Group by and aggregate
print( df.groupby(['col1', 'month_last_dates','col2']).agg(aggregations) )
Upvotes: 0
Reputation: 9059
from calendar import monthrange
from datetime import date
def month_end(year, month):
return date(year=year, month=month, day=monthrange(year, month)[1]).isoformat()
>>> month_end(2020, 2)
'2020-02-29'
Upvotes: 1