CODEWITHSUNDEEP
pythonpython-3.xpandasdataframe
DataPlankton
DataPlankton

Reputation: 145

Getting multiple min and max dates from a pandas data frame

I need to get the minimum and maximum dates of the two periods where column y is greater than 0.

Dataframe:

Date          Y
25/02/2019    0
24/02/2019    1
23/02/2019    1
22/02/2019    0
21/02/2019    1
20/02/2019    1
19/02/2019    1
18/02/2019    0
17/02/2019    0
16/02/2019    0

Perhaps return the results in a list. Using the table above I expect to get the following: 

[[23/02/2019, 24/02/2019],[21/02/2019, 19/02/2019]]

How the results are returned is not important. It's getting them. I understand how to get min and max dates if there was only one period where Y > 0 :

min(ua_dates['Date'])
max(ua_dates['Date'])

Trouble is this gives me 19/02/2019 and 24/02/2019 using the table above. Perhaps there is some sort of library out there?

Upvotes: 1

Views: 908

Answers (2)

BENY
BENY

Reputation: 323226

IIUC you need select all Y==1 out , then if they are not continue , you want them into different group, here we using cumsum 

ua_dates.loc[df.Y==1,'Date'].groupby(ua_dates.Y.eq(0).cumsum()).agg(['min','max']).values.tolist()
Out[713]: [['23/02/2019', '24/02/2019'], ['19/02/2019', '21/02/2019']]

Upvotes: 2

Polkaguy6000
Polkaguy6000

Reputation: 1208

You just need to mask your data:

 mask = ua_dates['Y'] > 0
 min(ua_dates[mask]['Date'])
 max(ua_dates[mask]['Date'])

Upvotes: 1

Related Questions