[pondering] Should `mutate` handle sliding window operations?

[talegari]

First thoughts:

df.mutate({'rolling_3_sum': (np.sum, 'col_1')}, window = [-1, 1])

-1: start at one index before
1: end at one index after

---

df.mutate({'rolling_3_sum': lambda x: (x['col_1'] + x['col_2'])/2}, window = [-1, 1])

we should be having the entire dataframe context here.

---

https://davisvaughan.github.io/slider/
https://dbplyr.tidyverse.org/articles/translation-function.html#window-functions
https://pandas.pydata.org/docs/user_guide/window.html