mercredi 18 avril 2018

Creating a histogram in python

I am trying to implement a PageRank algorithm in python. I have created a function called randomwalk that takes as input the set edges, a teleport probability a and a positive integer iters and performs the random walk. Starting from any page, the function randomly follows links from one page to the next, teleporting to a completely random page with probability a at each iteration. It stores all visited states.

Now I want to create a histogram of the frequency by which each page was visited. This histogram is what the randomwalk function should return.

My code is below:

import matplotlib.pyplot as plt

edges =[[0,1], [1,1], [2,0], [2,2], [2,3], [3,3], [3,4], [4,6], [5,5], [6,6], [6,3]]

def randomWalk(edges, a ,iters):
    pages = list({edge[0] for edge in edges})
    dict_edges = {}
    for edge_from, edge_to in edges:
        if edge_from not in dict_edges:
            dict_edges[edge_from] = [edge_to]
        else:
            dict_edges[edge_from].append(edge_to)
    current_page = random.choice(pages)
    visit_counts_dictionary = {page:0 for page in pages}
    visit_counts_dictionary[current_page] +=1
    for _ in range(iters):
        if random.uniform(0,1) < a:
            current_page = random.choice(pages)
            visit_counts_dictionary[current_page] += 1
        else:
            current_page = random.choice(dict_edges[current_page])
            visit_counts_dictionary[current_page] += 1
    plt.hist(visit_counts_dictionary)
    plt.show()

print(randomWalk(edges, 0, 10))

An empty histogram appears, with the following error:

ValueError: x must be 1D or 2D 




Aucun commentaire:

Enregistrer un commentaire