How can Python Pandas fill in an empty column and not count its sum?-CodePudding

I am solving a problem in which there is a column with NaN values that need to be filled in by the number of rows, starting from one. Like this:

       j  П, k=1  П, k=2
0      1    10.0    40.0
1      2    20.0    50.0
2      3    30.0    60.0
Sum    NaN  60.0   150.0

I try to do this in two ways: the first one works out incorrectly - it gives all threes instead of 1,2,3, the second option does not work at all.

In addition, the first option counts the amount in this column, and I don't need it there. What should I do?

Code:

import numpy as np
import pandas as pd

m=4
df1 = pd.DataFrame(data = {'j': [np.nan, np.nan, np.nan], 'П, k=1': [10, 20, 30], 'П, k=2': [40, 50, 60]})
df1.loc['Sum'] = df1.sum()

# THE FIRST OPTION:
# for j in range(1, m):
#    df1['j'] = j

# THE SECOND OPTION:
bc = [x for x in range(1, m)]
print(bc)
df1['j'] = bc

print(df1)

CodePudding user response：

You need to select first three rows in your second option

df1.loc[df1.index[:-1], 'j'] = bc

       j  П, k=1  П, k=2
0    1.0    10.0    40.0
1    2.0    20.0    50.0
2    3.0    30.0    60.0
Sum  0.0    60.0   150.0

In your first option, with df1['j'] = j you are overriding j column in every loop. You may consider df1.loc[j, 'j'] = j.