main.py

from sampleDURW import sample
from generateStats import graphSampleStatistics, graphStatistics
import argparse
import os
import numpy as np
import sys
import matplotlib.pyplot as plt
import networkx as nx
import operator
from datetime import datetime
startTime = datetime.now()

def findNextNum(index, array):
    for i in range(index, len(array)):
        if not array[i] is 0:
            return array[i]



def main():
    # Argument parsing for various options
    parser = argparse.ArgumentParser(
        description="Main Sampling and Statistics Driver Program - Supply relative or absolute path where the Graph(s) reside")
    parser.add_argument('-d', '--inDir', type=str, required=True,
                        help='Name of directory where input graphs reside')
    parser.add_argument('-oG', '--outFileG', type=bool, default=False,
                        help='Whether or not output gpickles of graphs are wanted (May be very taxing with huge graphs)')
    parser.add_argument('-oP', '--outFileP', type=bool, default=False,
                        help='Whether or not output images of graphs are wanted (May be very taxing with huge graphs)')
    parser.add_argument('-it', '--iternum', type=int, default=200000,
                        help='Number of nodes sampled - Default is 200000')
    parser.add_argument('-bw', '--bweight', type=float, default=1,
                        help='Weight of edges to determine frequency of random jumps (1:less -->  inf:more, w>0) - default is 1')
    parser.add_argument('-incr', '--increment', type=int, default=1,
                        help='Amount to increment w by for each iteration of sampling')
    parser.add_argument('-amt', '--amount', type=int, default=1,
                        help='Amount of iterations for different samplings based on increment amount')
    parser.add_argument('-ws', '--wset', nargs='+', type=float, default=[False],
                        help='Array of w values to test')
    args = parser.parse_args()

    # loop indir
    # perform sampleDURW on each graph, for number of times specified by amount, starting at bw, incrementing by increment each time
    # number of sampling rounds for underlying sampling function
    path = args.inDir
    if(os.path.isdir(path)):
        if not os.path.isabs(path):
            path = os.path.abspath(path)
        count = 0
        for file in os.listdir(path):
            end = 0
            if args.wset[0] is False:
                end = args.amount
            else:
                end = len(args.wset)
            inFile = os.path.dirname(os.path.abspath(
                    __file__)) + "\\testGraphs\\" + file
            # Creates a Directed Graph to load the data into
            G1 = nx.DiGraph()
            print("Beginning Processing of input file")
            sys.stdout.flush()
            # Parses each line and adds an edge.
            # Since the underlying structure is a dictionary
            # there are no duplicates created
            # Creates the initial directed graph, Gd, from the given data.
            with open(inFile) as f:
                for line in f:
                    # Edge format is <nodeID> <nodeID>
                    edge = line.split()
                    # Skips Comments
                    if edge[0] == '#':
                        continue
                    node1 = int(edge[0])
                    node2 = int(edge[1])
                    # Add edges as they are parsed
                    G1.add_edge(node1, node2, weight=1)
                    G1.add_edge(node2, node1, weight=1)

            print("Done Processing Input File")
            sys.stdout.flush()
            outdegree = G1.out_degree()
            max1 = max(outdegree, key=lambda k: outdegree[k])
            sys.stdout.flush()
            outdegreedistr = graphStatistics(G1, file.split('.')[0])
            nmse = []
            degrees = []
            for i in range(0, end):
                if args.wset[0] is False:
                    if count == 0:
                        weight = args.bweight
                        count += 1
                    else:
                        weight = weight + args.increment
                else:
                    weight = args.wset[count]
                    count += 1

                graphs = sample(G1, args.outFileG, args.outFileP,
                                args.iternum, weight, count, inFile)
                print('I made it after the sampling')
                sys.stdout.flush()
                n,d = graphSampleStatistics(G1, graphs[0], graphs[
                                      1], file.split('.')[0], weight, outdegreedistr, max1)
                nmse.append(n)
                degrees.append(d)
                print('Graphed sample statistics')
                sys.stdout.flush()
            count = 0
            avg1 = np.mean(nmse, axis=0)
            avg2 = np.mean(degrees, axis=0)
            plt.figure()
            plt.yscale('log')
            plt.xscale('log')
            for i in range(len(avg1)):
                if avg1[i] == 0:
                    nextT = findNextNum(i, avg1)
                    avg1[i] = nextT
            plt.plot(avg2, avg1, 'ro-')
            plt.xlabel('Degree')
            plt.ylabel('NMSE')
            title = 'NMSE vs OutDegree for WikiTalk'.format(inFile)
            plt.title(title)
            outGraph = 'stats/final.jpg'.format(file.split('.')[0], weight)
            plt.savefig(outGraph)
            plt.close()
            print(datetime.now() - startTime)
    else:
        print("Input Directory Doesn't Exist - Insert New Directory")


if __name__ == '__main__':
    main()