Python：使用取决于先前值的值填充数据框中的列

Question

i need to find a concise way to populate my dataframe with code like the following:

x = 1
df.at[x,'col2']=df.at[x,'col1'] + df.at[x-1,'col2']
x = 2
df.at[x,'col2']=df.at[x,'col1'] + df.at[x-1,'col2']
x = 3
df.at[x,'col2']=df.at[x,'col1'] + df.at[x-1,'col2']
x = 4
df.at[x,'col2']=df.at[x,'col1'] + df.at[x-1,'col2']

... and so on til the final rows. Each line of code depends on the result of the one above so I am not sure if it can use the rolling function? I have also tried .shift(1) but haven't been successful either. Basically I need to compute this above from x=1 to x=last column

Take the following as an example:

       col1         col2
0       0             0
1       1           value above (0) plus value from col1 (1) = 1
2       3           value above (1) plus value from col1 (3) = 4 

Can anyone help?

Answer 1

A quick solution, it could be made better by making it generic (this is specific to the question)

import pandas as pd
d = {'col1': [1, 2, 3], 'col2': [4, 5, 6]}
df = pd.DataFrame(data=d)
print(df,"\n\n")
x = len(df)
df.iloc[1:,1] = df.iloc[1:x,0].values + df.iloc[0:x-1,1].values

print(df)

I used.values to get a round the problem of the index giving NaNs and was helped by this post which ran into the same problems.

This returns,

   col1  col2
0     1     4
1     2     5
2     3     6 


   col1  col2
0     1     4
1     2     6
2     3     8