Merging DataFrame by an amount of rows

Question

let's suppose I have this kind of data frame:

            Count
Trip Date        
2014-01-01   5370
2014-01-02   8374
2014-01-03   1121
2014-01-04   2246
2014-01-05   2626
...           ...
2019-11-26  63410
2019-11-27  51121
2019-11-28  16090
2019-11-29  23095
2019-11-30  26389

How can I get a new df that looks like this:

             Count
Week number        
1             C1
2             C2
3             C3
4             C4
5             C4
...           ...
n-1         C(n-1)
n             Cn

where Ci is a sum of Count for each 7 days?

Shoopyman · Accepted Answer · 2020-01-10T23:22:53.950

1

First you need to set your index to the date.

df.set_index(df["date"],inplace=True)

Then you just do day week or month like this

df.resample('D').sum()
df.resample('M').sum()
df.resample('Y').sum()

https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.resample.html

edited Jan 10 '20 at 23:22

answered Jan 10 '20 at 22:51

Shoopyman

53
5

Thanks! The problem is that when I do it in this way nothing happens, but when I type `df = df['count']...` I get a series as an output with no column `count`. – alladinsane Jan 10 '20 at 23:04
Is the dtype some kind of int for that col? – Shoopyman Jan 10 '20 at 23:11
1

It appears I was combining old and new syntax. What about df without the ['count'] – Shoopyman Jan 10 '20 at 23:23
I mean how can I get a dataframe again after resampling? – alladinsane Jan 10 '20 at 23:23
Now it works perfectly. Thanks! – alladinsane Jan 10 '20 at 23:25

Merging DataFrame by an amount of rows

1 Answers1