I have created the following Dataframe (which in reality is 1000+ rows and 20+ columns):
d = {'col1': [0, 0, 4, 6], 'col2': [3, 4, 0, 0], 'col3': [0, 10, 0, 0], 'END': [0, 0, 0, 0]}
df = pd.DataFrame(data=d)
print(df)
Out:
col1 col2 col3 END
0 0 3 0 0
1 0 4 10 0
2 4 0 0 0
3 6 0 0 0
Now I want to create a loop for every column by index to do the following: check if the current item in the column is greater than 0 and if yes, if the item in the column next to the right is equal to 0. If both are true, then subtract 1 from the item in the column and move to the next column and repeat.
So far my best try was with a while loop to find the correct fields:
Counter = len(list(df))
i = 0
while 0 < Counter:
if df.iloc[:,i] > 0 and df.iloc[:,i+1] == 0:
df.iloc[:,i] - 1
i = i +1
This code however raises a value error.
My desired result would look like this:
col1 col2 col3 END
0 0 2 0 0
1 0 4 9 0
2 3 0 0 0
3 5 0 0 0
This should be pretty fast:
df[df.ne(0) & df.shift(-1, axis=1).eq(0)] -= 1
Output:
>>> df
col1 col2 col3 END
0 0 2 0 0
1 0 4 9 0
2 3 0 0 0
3 5 0 0 0
Another option
df.sub(df.shift(-1, axis=1).eq(0)).clip(0)
col1 col2 col3 END
0 0 2 0 0
1 0 4 9 0
2 3 0 0 0
3 5 0 0 0