CODEWITHSUNDEEP
pythonpandasdatetimecounterseries
Reub
Reub

Reputation: 785

Pandas Dataframe: group by year and month

I have a pandas dataframe with a column of 'date_created'

the column has a following format of YYYY-MM-DD. for example: 2017-10-05

I want to group it by MM/YYYY

I used this but it included DD

frame['date_created'].value_counts()

is there anyway to query it by using YYYY and MM

Upvotes: 2

Views: 7128

Answers (2)

rafaelc
rafaelc

Reputation: 59274

You can also group by month and count over 

frame.set_index("date_created").groupby(pd.Grouper(freq='M')).count()

Upvotes: 4

jpp
jpp

Reputation: 164633

I would create a new column, say "YearMonth" with day fixed to 1. Then group by this column:

import pandas as pd

df = pd.DataFrame({'Date': ['2017-10-05', '2017-10-20']})
df['YearMonth'] = pd.to_datetime(df['Date']).map(lambda dt: dt.replace(day=1))

res = df['YearMonth'].value_counts()

# 2017-10-01    2
# Name: YearMonth, dtype: int64

Upvotes: 5

Related Questions