-
Notifications
You must be signed in to change notification settings - Fork 0
/
make_graph_distr.py
122 lines (104 loc) · 3.8 KB
/
make_graph_distr.py
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
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
import matplotlib.colors as mcolors
import networkit as nk
import matplotlib.ticker as mtick
import pandas as pd
import csv
from pandas.io.formats import style
from sklearn.preprocessing import MinMaxScaler
from sklearn.preprocessing import StandardScaler
import seaborn as sns
cname = {1:"USDT", 2:"MGC", 3:"LINK", 4:"WETH", 5:"EOS", 6:"BAT", 7:"OMG", 8:"CPCT", 9:"TRX", 10:"SHIB"}
fo = open('./risultati_analisi/media_var_degree.csv','w')
csv_writer = csv.writer(fo)
csv_writer.writerow(['contratto','media','varianza'])
list_patch = []
ax = None
#bdf = pd.DataFrame(columns=['USDT','MGC','LINK','WETH','EOS','BAT','OMG','CPCT','TRX','SHIB'])
ldf = []
for color,n in zip(mcolors.TABLEAU_COLORS,range(1,11)):
fname = "./edgelist/edgelist_"+str(n)+".csv"
reader = nk.graphio.EdgeListReader(',',1,'#',directed=True,continuous=False)
try:
g = reader.read(fname)
except:
print("File not exist")
exit()
gu = nk.graphtools.toUndirected(g)
dd = nk.centrality.DegreeCentrality(gu).run()
list_res = dd.ranking()
#GROUPBY(VALORE) -> OCCORRENZE DI OGNI VALORE
df = pd.DataFrame(list_res,columns=['id_node','degree'])
#print(df)
dfg = df.groupby(['degree']).size().reset_index(name='counts')
#dfg.loc[-1] = [0, 0] # adding a row
#dfg.index = dfg.index + 1 # shifting index
#dfg.sort_index(inplace=True)
#media
#media = df['degree'].mean()
#var = df['degree'].var()
#csv_writer.writerow([n,media,var])
#NORMALIZZARE - MAX
#nt = gu.numberOfNodes()
#list_occ_norm = [((int(occ)/nt)*100) for occ in list_occ]
#list_gradi_sorted = sorted(list_gradi)
#max_degree = list_gradi_sorted[-1]
##print(max_degree)
#list_gradi_norm = [((int(gradi)/int(max_degree))*100) for gradi in list_gradi_sorted]
#RECUPERO LE DUE LISTE DA PLOTTARE
#list_occ_norm = dfg_norm[1].tolist()
#list_gradi_norm = dfg_norm[0].tolist()
#print (list_occ_norm)
#print(list_gradi_norm)
#PLOTTO
#plt.plot(list_gradi_norm,list_occ_norm)
#plt.xscale("log")
#plt.yscale("log")
#NORMALIZZO
#scaler = MinMaxScaler()
#dfg = pd.DataFrame(scaler.fit_transform(dfg),columns=['degree','counts'])
#dfg['degree'] = (dfg['degree'] - dfg['degree'].min()) / (dfg['degree'].max() - dfg['degree'].min())
#print(dfg)
#CDF
#pdf
dfg['pdf'] = dfg['counts'] / sum(dfg['counts'])
#print(dfg)
#cdf
dfg['cdf'] = dfg['pdf'].cumsum()
dfg = dfg.reset_index()
#dfg['degree'] = dfg['degree']+1
print(dfg)
ax = dfg.plot(x = 'degree', y = 'cdf',ax=ax)
#PLOT GRAFICO NORMALIZZATO
#ax = dfg_norm.plot(x='degree',y='counts',kind='line',ax=ax)
#BOXPLOT
#bdf[cname[n]] = df['degree']
#bx = dfg.boxplot()
#ldf.append(dfg_norm.assign(Location=cname[n]))
#SCRIVI IL NOME CONTRATTO AL POSTO DEL NUMERO
patch = mpatches.Patch(color=color, label=cname[n])
list_patch.append(patch)
#print(map_degree)
#print(dfg)
#print(list_gradi_norm)
#print(list_occ_norm)
#print("Numero nodi : "+str(g.numberOfNodes()))
#i=10078
#print("degree di "+str(i)+" : "+str(list_res[i]))
#PLOTTO LEGENDA E SALVO FILE
#plt.yscale('log')
plt.xscale('log')
#cdf = pd.concat(ldf)
#print (cdf)
#ax = sns.boxplot(x="Location", y="degree", data=cdf) #dfg.boxplot(column=['USDT','MGC','LINK','WETH','EOS','BAT','OMG','CPCT','TRX','SHIB'])
plt.xticks(fontsize=12,weight='bold')
plt.yticks(fontsize=12,weight='bold')
plt.ylabel('FREQUENCY', fontsize=18,weight='bold')
plt.xlabel('DEGREE (normalized)',fontsize=18,weight='bold')
plt.title('NODES DEGREE DISTRIBUTION',fontsize=18,weight='bold')
f = plt.figure(num=1)
f.set_figheight(10)
f.set_figwidth(10)
plt.legend(fontsize=15,handles=list_patch)
plt.savefig('./risultati_analisi/cdf_gradi_norm.png')