CODEWITHSUNDEEP
pythonpandasdataframemachine-learningdataset
lowlypalace
lowlypalace

Reputation: 157

Loc Pandas DataFrames on Date Range

I have a Pandas DataFrame with a 'DateTime' column. I need to filter out all rows in the DataFrame that have this date and add - 7 and + seven days to the target date.

It should be something like this: test = dataset.loc[-7:"4/17/2017":+7].hp1.values

What is the best way to achieve this?

Upvotes: 1

Views: 1107

Answers (1)

jezrael
jezrael

Reputation: 863481

Idea is create datetimes above and below by Timedelta and then filter with start and end datetimes, also for filter by column name add it to DataFrame.loc:

#sample data
dataset = pd.DataFrame({'hp1':range(20)}, index=pd.date_range('04-09-2017', periods=20))
#print(dataset)

d = "4/17/2017"
date = pd.to_datetime(d)

td = pd.Timedelta(7, unit='d')
#alternative
#td = pd.offsets.DateOffset(days=7)

start = date - td
end = date + td

test = dataset.loc[start:end, 'hp1']

print (test)
2017-04-10     1
2017-04-11     2
2017-04-12     3
2017-04-13     4
2017-04-14     5
2017-04-15     6
2017-04-16     7
2017-04-17     8
2017-04-18     9
2017-04-19    10
2017-04-20    11
2017-04-21    12
2017-04-22    13
2017-04-23    14
2017-04-24    15
Freq: D, Name: hp1, dtype: int64

Upvotes: 1

Related Questions