TETRIS-code/degree_bin_plot.py at master · skalyanb/TETRIS-code · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
import os
import sys
import time
import pandas as pd
import matplotlib.pyplot as plt
import statistics
import numpy as np


dirname = os.path.dirname
sys.path.append(dirname(dirname(os.path.realpath(__file__))))

def skip_rows_err (index):
    if index <10 or index > 10:
        return True
    return False

def plot_degree_bin (bin_dir, out_filename, title_info):
    # We will go into each directory in the bin_dir list
    # and then iterate over each file inside the directory.
    # For each file we take the median accuracy and the variance accuracy
    #
    median_err=[]
    variance_err=[]


    # Enumerate over each bin
    for i,data_dir in enumerate(bin_dir):
        # We are inside a bin; Go over each file and collect the median accuracy and variance.
        bin_median_error = []
        bin_variance_error = []
        for i, file in enumerate(os.listdir(data_dir)):
            path = os.path.join(data_dir, file)
            if os.path.isdir(path):
                # skip directories
                continue
            if not file.startswith('.'):
                data_file = data_dir + file
                df_median_error = pd.read_csv(data_file, sep=",",skiprows=lambda x: skip_rows_err(x),header=None, usecols=[1,3])
                bin_median_error.append(df_median_error.iloc[0,0])
                bin_variance_error.append(df_median_error.iloc[0,1])
        median_err.append(bin_median_error)
        variance_err.append(bin_variance_error)

    num_bin = len(bin_dir)
    group_len = len(median_err[0])

    # Set bar width
    barWidth = 0.2
    starting_pos = 1
    gap = 1

    x={}
    # Set bar position
    for i in range(num_bin):
        x[i] = [starting_pos+i*barWidth for i in range(group_len)]
        starting_pos = starting_pos + gap + barWidth*group_len


    fig, ax = plt.subplots()
    plt.subplots_adjust(bottom = 0.2, top = 0.8, right = 0.95, left = 0.1, hspace = 0.38)

    ax.bar(x[0], median_err[0], width=barWidth, color='b',  edgecolor= "black",label="Gender")
    ax.bar(x[1], median_err[1], width=barWidth, color='b',  edgecolor= "black",label="Type")
    ax.bar(x[2], median_err[2], width=barWidth, color='b',  edgecolor= "black",label="Type")
    ax.bar(x[3], median_err[3], width=barWidth, color='b',  edgecolor= "black",label="Type")
    #rects4 = ax.bar(x_4, median_err[4], width=barWidth, color='b',  edgecolor= "black",label="Type")

    y_min = 0
    y_max = 2
    ax.set_ylim(y_min,y_max)

    # y_ind = [i/2 for i in range(4)]
    # y_label = [str(i/2)+'%' for i in range(4)]
    # ax.set_yticks(y_ind)
    # ax.set_yticklabels(y_label)

    ax.set_title(title_info,fontsize=24)

    ax.set_ylabel('Median Relative Error (%)',fontsize=20)

    ind = [x[i][1] for i in range(num_bin)]
    ax.set_xticks(ind)
    ax.tick_params(axis="y", labelsize=18)
    ax.set_xticklabels(('deg$\in [1,10]$', 'deg$\in [10,10^2]$','deg$\in [10^2,10^3]$', 'deg$\in [10^3,10^4]$'),fontsize=18)
    ax.set_xlabel('Degree based buckets',fontsize=24)
    plt.show()

    timestr = time.strftime("%Y%m%d-%H%M%S")

    fig.savefig("output/plots/degree-bin/"+out_filename+"-"+timestr+".eps",format='eps',bbox_inches="tight")


if __name__ == "__main__":

    x_max = 1.1

    file_name = []
    title_info = []

    # f_name = "soc-flickr-und"
    # file_name.append(f_name)

    # title_info.append(f_name + ": 16M edges, 1.7M vertices")

    # f_name = "socfb-A-anon"
    # file_name.append(f_name)
    # title_info.append(f_name + ": 24M edges")
    # no_bin = 4
    # #
    f_name = "soc-orkut"
    file_name.append(f_name)
    title_info.append(f_name + ": 213M edges")

    # f_name = "soc-sinaweibo"
    # file_name.append(f_name)
    # title_info.append(f_name + ": 260M edges, 58M vertices")
    #
    # f_name = "soc-twitter-konect"
    # file_name.append(f_name)
    # title_info.append(f_name + ": 2.4B edges")
    no_bin = 4

    # f_name = "soc-friendster"
    # file_name.append(f_name)
    # title_info.append(f_name + ": 1.8B edges, 65M vertices")
    # #
    for i,file in enumerate(file_name):
        bin_directory = []
        # The main data dirextory
        data_dir = "output/plot_data/degree_bin_data/"+file+".edges/EstTriByRWandWghtedSampling/0.3/bin"
        for j in range(no_bin):
            bin_directory.append(data_dir+'_'+str(j)+'/')
        out_filename = file
        plot_degree_bin(bin_directory,out_filename,title_info[i])