Python-文件的列平均

Question

I have a file whose content looks like:

A B 2 4
C D 1 2
A D 3 4
A D 1 2
A B 4 7 and so on..

My objective is to get the final output as below:

A B 3 5.5
C D 1 2
A D 2 3

That is, for each unique combination of first two columns, the result should be column-wise average of other two columns of the file. I tried using loops and it is just increasing the complexity of the program. Is there any other way to achieve the objective.

Sample Code:

with open(r"C:\Users\priya\Desktop\test.txt") as f:
    content = f.readlines()
content = [x.split() for x in content] 
for i in range(len(content)):
    valueofa=[content[i][2]]
    valueofb=[content[i][3]]
    for j in xrange(i+1,len(content)):
        if content[i][0]==content[j][0] and content[i][1]==content[j][1]:
            valueofa.append(content[j][2])
            valueofb.append(content[j][3])

and I intended to take the average of both lists by index.

Answer 1

You can store each combination of letters as a tuple in a dictionary and then average at the end, eg:

d = {}
with open(r"C:\Users\priya\Desktop\test.txt") as f:
    for line in f:
        a, b, x, y = line.split()
        d.setdefault((a, b), []).append((int(x), int(y)))

for (a, b), v in d.items():
    xs, ys = zip(*v)
    print("{} {} {:g} {:g}".format(a, b, sum(xs)/len(v), sum(ys)/len(v)))

Output:

A B 3 5.5
C D 1 2
A D 2 3

Answer 2

If you can use pandas, it will much simpler:

import pandas as pd

df = pd.read_csv(r"C:\Users\priya\Desktop\test.txt", names=['A','B','C','D'])
df
    A   B   C   D
0   A   B   2   4
1   C   D   1   2
2   A   D   3   4
3   A   D   1   2
4   A   B   4   7

df.groupby(['A','B']).mean().reset_index()

    A   B    C   D
0   A   B   3.0 5.5
1   A   D   2.0 3.0
2   C   D   1.0 2.0