PageRank for EDAN55

pagerank.py

#!/usr/bin/env python
# encoding: utf-8
import numpy as np
import sys

def parse(path):
    
    f = open(path, 'r')
    lines = f.readlines()
    #init empty
    n = int(lines[0])
    m = np.zeros((n, n))
    f = open(path, 'r')
    first = True
    for row in f:
        if first:
            first = False
        else:
            nrs = row.split()
            i = 0
            while i < len(nrs) - 1:
                m[int(nrs[i])][int(nrs[i+1])] = 1.0 / (len(nrs) / 2.0)
                i += 2
    return m


def pagerank_simple(A, iterations):  
    """ Simple pagerank """
    p = np.ones(len(A))
    Q = np.linalg.matrix_power(A, iterations)
    print p
    print Q
    print p * Q
    p = np.transpose(p * Q)
    for i in xrange(len(p)):
        print str(i) + " PageRank: " + str(p[i])

def pagerank(H, iterations):
    """ Simple pagerank """
    p = np.ones(len(H))  
    n = len(H)
    alpha = 0.85
    D = [[0 for y in range(n)] for x in range(n)]
    for i in xrange(len(D)):
        for j in xrange(len(D)):
            if sum(H[i]) == 0:
                D[i][j] = 0
            else:
                #bigger then deg(i) = 0 else 1/n 
                D[i][j] = 1.0 / len(H) 
    P = alpha * ( H + D ) + (1 - alpha) / n * np.ones((n, n))
    
    # CALCULATE pP HERE !

    for i in xrange(len(p)):
        print str(i) + " PageRank: " + str(p[i])

if __name__ == "__main__":
    if len(sys.argv) == 1: 
        print "Call like: python pagerank.py <nbr of iterations> <test case>"
        exit()
    
    iterations = int(sys.argv[1])
    if len(sys.argv) == 2:
        graph = parse('data/three.txt')
    else:
        graph = parse('data/' + sys.argv[2] +".txt")

    pagerank(graph, iterations)