How to write mathematical combination formula coding in python

Question

I have a data set as the following:

input file:

id  addr
301 1
301 2
301 3
301 4
302 6
302 7
302 8
302 9
302 1
303 14
303 15
303 2
304 16
304 17
304 1

and I need Python code to print out all the possible pair combinations of addr values with common id. There are millions of id and corresponding addr value records in the main test file. So, the code should be able to read columns from a text file.The output will be as follows (only showing for 301 and 302, the rest will continue the pattern):

1   2
1   3
1   4   
2   3
2   4
3   4   
6   7
6   8
6   9
7   8
7   9   
8   9
1   6
1   7
1   8
1   9
2   6
2   7
2   8
2   9
3   6  
3   7
3   8
3   9
4   6
4   7
4   8
4   9
1   15
2   15
3   15
......
1   16
2   16
......
15  16   

So far I have done the following, but I do not have any idea how to code the pair combination part. I am new in Python, so will appreciate if someone can help me do the coding with a little bit of explanation.

# coding: utf-8

# sample tested in python 3.6

import sys
from pip._vendor.pyparsing import empty

if len(sys.argv) < 2:
    sys.stderr.write("Usage: {0} filename\n".format(sys.argv[0]))
    sys.exit()

fn = sys.argv[1]
sys.stderr.write("reading " + fn + "...\n")

# Initialize empty set 
s = {}
line= 0
fin = open(fn,"r")
for line in fin:
    line = line.rstrip()
    f = line.split("\t")
    line +=1
    if line is 1:
        txid_prev = line 
        addr = line 
        s= addr
        continue
    txid=line
    txid_prev=line
    if txid is txid_prev:
        s.push(addr)
    else:
        # connect all pairs in s
        # print all pairs as edges
        s=addr
    txid_prev=txid
if s is not empty:
    # connect and print all edges   

Answer 1

How about something like this:

import pandas as pd
import io
import itertools

file="""id addr
301 1
301 2
301 3
301 4
302 6
302 7
302 8
302 9
302 1"""

df= pd.read_csv(io.StringIO(file), sep=" ")

for key,value in df.set_index("addr").groupby("id").groups.items():
    print(key)
    for item in list(itertools.combinations(value.values, 2)):
        print("{} {}".format(*item))

Prints:

301
1 2
1 3
1 4
2 3
2 4
3 4
302
6 7
6 8
6 9
6 1
7 8
7 9
7 1
8 9
8 1
9 1

---

Alternatively we can put the values in a Dictionary:

a = {} 

for id_,addr in df.values.tolist():
    a.setdefault(str(id_),[]).append(addr)

output = {key:list(itertools.combinations(value, 2)) for key,value in a.items()}


def return_combos(dict_, keys):
    values = []
    for i in keys:
        values.append(a[i])
    values = list(set([i for item in values for i in item]))
    return {','.join(keys):list(itertools.combinations(values, 2))}


output2 = return_combos(a, ["301","302"])

output prints:

{'301': [(1, 2), (1, 3), (1, 4), (2, 3), (2, 4), (3, 4)],
 '302': [(6, 7),
  (6, 8),
  (6, 9),
  (6, 1),
  (7, 8),
  (7, 9),
  (7, 1),
  (8, 9),
  (8, 1),
  (9, 1)]} 

meanwhile output2 outputs:

{'301,302': [(1, 2),
  (1, 3),
  (1, 4),
  (1, 6),
  (1, 7),
  (1, 8),
  (1, 9),
  (2, 3),
  (2, 4),
  (2, 6),
  (2, 7),
  (2, 8),
  (2, 9),
  (3, 4),
  (3, 6),
  (3, 7),
  (3, 8),
  (3, 9),
  (4, 6),
  (4, 7),
  (4, 8),
  (4, 9),
  (6, 7),
  (6, 8),
  (6, 9),
  (7, 8),
  (7, 9),
  (8, 9)]}

update2 or 3: Is this the desired output??

import pandas as pd
import io
import itertools
from collections import OrderedDict

file="""id addr
301 1
301 2
301 3
301 4
302 6
302 7
302 8
302 9
302 1
303 14
303 12"""

df= pd.read_csv(io.StringIO(file), sep=" ")

b = OrderedDict()

for id_,addr in df.values.tolist():
    b.setdefault(str(id_),[]).append((id_,addr))

pairs = [(list(b.keys())[i],list(b.keys())[i+1]) for i in range(len(list(b.keys()))-1)]

output = {}
for pair in pairs:
    output[pair] = [[(i[0][0],i[1][0]),i[0][1],i[1][1]] for i in list(itertools.combinations(b[pair[0]]+b[pair[1]], 2))]

output    

{('301', '302'): [[(301, 301), 1, 2],
  [(301, 301), 1, 3],
  [(301, 301), 1, 4],
  [(301, 302), 1, 6],
  [(301, 302), 1, 7],
  [(301, 302), 1, 8],
  [(301, 302), 1, 9],
  [(301, 302), 1, 1],
  [(301, 301), 2, 3],
  [(301, 301), 2, 4],
  [(301, 302), 2, 6],
  [(301, 302), 2, 7],
  [(301, 302), 2, 8],
  [(301, 302), 2, 9],
  [(301, 302), 2, 1],
  [(301, 301), 3, 4],
  [(301, 302), 3, 6],
  [(301, 302), 3, 7],
  [(301, 302), 3, 8],
  [(301, 302), 3, 9],
  [(301, 302), 3, 1],
  [(301, 302), 4, 6],
  [(301, 302), 4, 7],
  [(301, 302), 4, 8],
  [(301, 302), 4, 9],
  [(301, 302), 4, 1],
  [(302, 302), 6, 7],
  [(302, 302), 6, 8],
  [(302, 302), 6, 9],
  [(302, 302), 6, 1],
  [(302, 302), 7, 8],
  [(302, 302), 7, 9],
  [(302, 302), 7, 1],
  [(302, 302), 8, 9],
  [(302, 302), 8, 1],
  [(302, 302), 9, 1]],
 ('302', '303'): [[(302, 302), 6, 7],
  [(302, 302), 6, 8],
  [(302, 302), 6, 9],
  [(302, 302), 6, 1],
  [(302, 303), 6, 14],
  [(302, 303), 6, 12],
  [(302, 302), 7, 8],
  [(302, 302), 7, 9],
  [(302, 302), 7, 1],
  [(302, 303), 7, 14],
  [(302, 303), 7, 12],
  [(302, 302), 8, 9],
  [(302, 302), 8, 1],
  [(302, 303), 8, 14],
  [(302, 303), 8, 12],
  [(302, 302), 9, 1],
  [(302, 303), 9, 14],
  [(302, 303), 9, 12],
  [(302, 303), 1, 14],
  [(302, 303), 1, 12],
  [(303, 303), 14, 12]]}

Answer 2

How about this (if you don't care about duplicates)? [Live Demo]

import io
import pandas as pd

data =  io.StringIO("""id addr\n301 1\n301 2\n301 3\n301 4\n302 6\n302 7\n302 8\n302 9\n302 1\n303 14\n303 15\n303 2\n304 16\n304 17\n304 1""")

df = pd.read_table(data, sep=' ', index_col=0)
print(df.join(df, rsuffix="_2"))

Output:

     addr  addr_2
id               
301     1       1
301     1       2
301     1       3
301     1       4
301     2       1
301     2       2
301     2       3
301     2       4
301     3       1
...

Answer 3

This problem can be split up into two parts. First, construct a dictionary that maps identifiers to lists of addresses. Second, generate the length-2 combinations of each of these lists.

from collections import defaultdict
from itertools import combinations

# get lines from your file
f = open('input_file.txt')
lines = f.readlines()
f.close()

# build mapping from file
iden_mapping = defaultdict(list)
for row in lines[1:]:
    iden, addr = row.split()
    iden_mapping[iden].append(addr)

# generate combination from address lists
for iden in sorted(iden_mapping):
    for c in combinations(iden_mapping[iden], 2):
        print(c)