pandas for each date select minimum and maximum time row

Question

i have a dataset like this:

id  service         datetime
4   connectsInfo    2016-11-05 06:50:23
4   dayFirstFeed    2016-11-05 06:50:57
4   dayFirstFeed    2016-11-05 06:51:43
4   connectsInfo    2016-11-22 06:42:55
4   dayFirstFeed    2016-11-22 06:42:33
4   connectsInfo    2016-11-22 10:52:11

for each date, I want the min and max time rows

an approach that I have tried:

doc_4.loc[doc_4.groupby(df.request_time.dt.date, as_index=False).request_time.idxmin()]

Edit:

I want the result like this:

date        min                 max
2016-11-05  2016-11-05 06:50:23 2016-11-05 06:51:43
2016-11-22  2016-11-22 06:42:33 2016-11-22 10:52:11

`doc_4.groupby(df.request_time.dt.date, as_index=False).request_time.agg(['min','max'])` — jezrael, Jul 05 '18 at 10:52
@jezrael please check the question again, something is wrong with the code which you gave. — Sriram Arvind Lakshmanakumar, Jul 05 '18 at 11:52
Sorry, need `df` inside groupby also - `df = df.groupby(df.datetime.dt.date, as_index=False).datetime.agg(['min','max'])` — jezrael, Jul 05 '18 at 11:57
Now your code and results are mismatched with your inputs, e.g. what is `request_time`, where did `2016-11-22` data go? Can you [edit your question](https://stackoverflow.com/posts/51189456/edit) to make your input & output (real & desired) consistent. — jpp, Jul 05 '18 at 13:52
@jezrael , please post your code as an answer so that i can mark it. — Sriram Arvind Lakshmanakumar, Jul 06 '18 at 07:14

score 2 · Accepted Answer · answered Jul 06 '18 at 07:16

2

You can use:

df = df.groupby(df.datetime.dt.date, as_index=False).datetime.agg(['min','max'])

answered Jul 06 '18 at 07:16

jezrael

822,522
95
1,334
1,252

pandas for each date select minimum and maximum time row

1 Answers1