CODEWITHSUNDEEP
pythonnumpypandas
Skorpeo
Skorpeo

Reputation: 2562

Pandas cumulative/elementwise

I am trying to figure out if there is a simple way to solve the following problem with pandas. Let's say I have four containers, A, B, C, D and each has a specific volume. Let's say I now get an amount of liquid that I want to fill these containers with. How can I come up with a "series" that shows the amount that each container got filled with. Essentially, I want the liquid to fill in container A, then container B, etc until there is no more liquid left. Not all containers have to fill up and there is never more liquid than there are containers. Here is the problem:

Container  Volume (L)  FILLED                            
 A           6           0
 B           7           0
 C           8           0
 D           9           0


Case A:

Liquid = 10L

Result:                  
Container  Volume (L) FILLED                             
 A           6           6
 B           7           4 
 C           8           0
 D           9           0

Case B:
Liquid = 20L
Result:
Container  Volume (L) FILLED                             
 A           6           6
 B           7           7 
 C           8           7
 D           9           0

I have played around with pandas.cumsum() and pandas.clip_upper() but it takes me a few steps to get to where I need and was hoping there is a slick way to do it. Like I can do for case B something like:

filled_a = df["Volume"].cumsum().clip_upper(20) and continue mangling until I get to my intended result

Upvotes: 1

Views: 106

Answers (1)

satomacoto
satomacoto

Reputation: 11963

a = df["Volume"].cumsum().clip_upper(20)
b = a.shift().fillna(0)
filled_a = a - b

Upvotes: 1

Related Questions