Skip to content
Snippets Groups Projects

Compare revisions

Changes are shown as if the source revision was being merged into the target revision. Learn more about comparing revisions.

Source

Select target project
No results found
Select Git revision
Loading items

Target

Select target project
0 results Searching
Select Git revision
Loading items
Show changes
150 files
+ 51863
− 14
Compare changes
  • Side-by-side
  • Inline

Files

.idea/.gitignore

0 → 100644
+3 −0
Original line number Diff line number Diff line 
# Default ignored files

/shelf/

/workspace.xml

.idea/.name

0 → 100644
+1 −0
Original line number Diff line number Diff line 
ANN_CLIQUES.py
 
 No newline at end of file

.idea/encodings.xml

0 → 100644
+7 −0
Original line number Diff line number Diff line 
<?xml version="1.0" encoding="UTF-8"?>

<project version="4">

  <component name="Encoding">

    <file url="file://$PROJECT_DIR$/node2vec/src/graph/test.npy" charset="windows-1252" />

    <file url="file://$PROJECT_DIR$/node2vec/src/graph/test_Cliques.npy" charset="UTF-16" />

  </component>

</project>
 
 No newline at end of file

.idea/inspectionProfiles/profiles_settings.xml

0 → 100644
+6 −0
Original line number Diff line number Diff line 
<component name="InspectionProjectProfileManager">

  <settings>

    <option name="USE_PROJECT_PROFILE" value="false" />

    <version value="1.0" />

  </settings>

</component>
 
 No newline at end of file

.idea/misc.xml

0 → 100644
+4 −0
Original line number Diff line number Diff line 
<?xml version="1.0" encoding="UTF-8"?>

<project version="4">

  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.7 (pythonProject2)" project-jdk-type="Python SDK" />

</project>
 
 No newline at end of file

.idea/modules.xml

0 → 100644
+8 −0
Original line number Diff line number Diff line 
<?xml version="1.0" encoding="UTF-8"?>

<project version="4">

  <component name="ProjectModuleManager">

    <modules>

      <module fileurl="file://$PROJECT_DIR$/.idea/walid.iml" filepath="$PROJECT_DIR$/.idea/walid.iml" />

    </modules>

  </component>

</project>
 
 No newline at end of file

.idea/vcs.xml

0 → 100644
+6 −0
Original line number Diff line number Diff line 
<?xml version="1.0" encoding="UTF-8"?>

<project version="4">

  <component name="VcsDirectoryMappings">

    <mapping directory="$PROJECT_DIR$" vcs="Git" />

  </component>

</project>
 
 No newline at end of file

.idea/walid.iml

0 → 100644
+8 −0
Original line number Diff line number Diff line 
<?xml version="1.0" encoding="UTF-8"?>

<module type="PYTHON_MODULE" version="4">

  <component name="NewModuleRootManager">

    <content url="file://$MODULE_DIR$" />

    <orderEntry type="jdk" jdkName="Python 3.7 (pythonProject2)" jdkType="Python SDK" />

    <orderEntry type="sourceFolder" forTests="false" />

  </component>

</module>
 
 No newline at end of file

ANN.py

0 → 100644
+112 −0
Original line number Diff line number Diff line 
import random

import numpy as np,os

import numba as nb

import time

import networkx as nx



def delta(A,B,U,m):

    for i in nb.prange(len(U)):

        z = np.zeros(len(U))

        x = np.zeros(len(U))

        c = np.asarray([i+1], dtype=np.str)

        if( np.count_nonzero(V == 1) > 0):

            if c[0] in O.nodes():

                for j in range(len(U)):

                    m = np.asarray([i+1 , j+1], dtype=np.str)

                    if V[j] != 0 and (m[1] in O.nodes() and O.number_of_edges(m[0], m[1]) != 0):

                        x[i] = x[i]

                    else:

                        x[i] = x[i] + V[j]

        U[i] = U[i] + (-x[i] + B * h(i,x[i]))

def h(i,r):

    if r + V[i] == 0:

        return 1

    else:

        return 0

def output(X):

    for i in range(len(X)):

        if (X[i] > 0):

            V[i] = 1

        else:

            V[i] = 0



def CHANGE(A):

    N = []

    E = A

    R = O.edges

    x = list(O.nodes())

    for i in range(len(U)):

        if U[i] > 0:

            print("true")

            N.append(i+1)

    if len(N) > 0:

        for k in x:

            for v in x:

                if v in x and k in x and O.number_of_edges(k, v) > 0:

                    O.remove_edge(k, v)

        A = O.edges



    print("new len A",len(A))

    return A

def Remplire(i):

    x = lab[i]

    for i in range(len(U)):

        if U[i] >= 0 and x[i] > 0:

            Ufin[i] = U[i]



def Arrange(x,i):

    t=0

    y=lab[i]

    for i in range(len(x)):

        if y[i] == 1:

                x[i] = B

        else:

                x[i] = random.uniform(-400.5,-0.5)



lab = np.load("data/clique_2/labels.npy",allow_pickle=True)

dat = np.load("data/clique_2/sam.npy",allow_pickle=True)

start = time.time()

outputs = []



for i in range(len(lab)):



    print(dat[i])

    O = nx.Graph()

    O.add_edges_from(dat[i], nodetype=int)

    m = np.array(O.nodes)

    size = O.number_of_nodes()

    print("====== Increasing embedding step =======")

    adj = np.count_nonzero(lab[i] == 1)

    size = len(lab[i])



    Ufin = np.random.uniform(-19,-1,size)*0

    x = 1

    U = np.random.uniform(-19,-1,size)

    V = np.random.randint(1,size=size)

    B = (adj / (size * (len(dat[i]) * 2 / (size * (size - 1))))) * 20

    while len(dat[0]) > 0:

        x = x+1

        U = np.random.uniform(-19,-1,size)

        delta(dat,B,U,m)

        output(U)

        dat[i] = CHANGE(dat[i])

        Remplire(i)



        O = nx.Graph()

        O.add_edges_from(dat[i])

        m = np.array(O.nodes)

    out = np.asarray(Ufin)

    Arrange(Ufin,i)

    output(Ufin)

    outputs.append(out)

    print("les resultats")

    print(np.count_nonzero(Ufin > 0))

    print(np.count_nonzero(V == 1))

    print(np.count_nonzero(lab[i] == 1))

end = time.time()

print("====== End of increasing ======")

print("Time", end-start)



out = np.asarray(outputs)

print(out.shape)

np.save(os.path.join("INoutput_data.npy"), out)  # generation des outputs"""

ANN_CLIQUES.py

0 → 100644
+132 −0
Original line number Diff line number Diff line 
import random

import numpy as np,os

import numba as nb

import time

import networkx as nx



def delta(A,B,U,m,V,O):

    for i in nb.prange(len(U)):

        z = np.zeros(len(U))

        x = np.zeros(len(U))

        c = np.asarray([i+1], dtype=np.str)

        if( np.count_nonzero(V == 1) > 0):

            if c[0] in O.nodes():

                for j in range(len(U)):

                    m = np.asarray([i+1 , j+1], dtype=np.str)

                    if V[j] != 0 and (m[1] in O.nodes() and O.number_of_edges(m[0], m[1]) != 0):

                        x[i] = x[i]

                    else:

                        x[i] = x[i] + V[j]

        U[i] = U[i] + (-x[i] + B * h(i,x[i],V))

def h(i,r,V):

    if r + V[i] == 0:

        return 1

    else:

        return 0

def output(X,V):

    for i in range(len(X)):

        if (X[i] > 0):

            V[i] = 1

        else:

            V[i] = 0



def CHANGE(A,O,U):

    N = []

    E = A

    R = O.edges

    x = list(O.nodes())

    for i in range(len(U)):

        if U[i] > 0:

            N.append(i+1)



    if len(N) > 0:

        for k in x:

            for v in x:

                if v in x and k in x and O.number_of_edges(k,v) > 0:

                    O.remove_edge(k, v)

        A = O.edges





    return A

def Remplire(U,Ufin,lab):

    for i in range(len(U)):

        if U[i] >= 0 and lab[i] > 0:

            Ufin[i] = U[i]

"""

        else:

            if lab[i] == 0:

                Ufin[i] = random.uniform(-400.5, -0.5)

            else:

                Ufin[i] = random.uniform(0.5, 400.5)

"""

def Arrange(lab, x, B, V):

    t=0

    y=0

    for i in range(len(x)):

        if lab[i] == 1:

                x[i] = B

                V[i] = 1

        else:

                x[i] = -B

                V[i] = 0





def PatternFinding(dat,lab):



    O = nx.Graph(dat)

    m = np.array(O.nodes)

    size = O.number_of_nodes()

    print("====== Increasing embedding step =======")

    adj = np.count_nonzero(lab == 1)

    size = len(lab)

    for i in range(1):

        Ufin = np.random.uniform(-1, 0, size) * 0

        #print("ufin",Ufin)

        #print(len(dat) * 2 / ((size-1) * (size - 1)))

        x = 1

        U = np.random.uniform(-1, 0, size)

        V = np.random.randint(1, size=size)

        B = (adj / (size * (len(list(O.edges)) * 2 / (size * (size - 1)))))

        #print("B",B)

        Arrange(lab,Ufin,B,V)

        #print(np.count_nonzero(V == 1))

        #print(np.count_nonzero(lab == 1))

        """

        while len(dat) > 0:

            x = x + 1

            U = np.random.uniform(-19, -1, size)

            delta(dat, B, U, m, V, O)

            output(U, V)

            # print(np.count_nonzero(U >= 0))

            # print(np.count_nonzero(lab == 1))

            dat = CHANGE(dat, O, U)

            print("hna")

            Remplire(U,Ufin,lab)

            # print("size",np.count_nonzero(Ufin >= 0),np.count_nonzero(U >= 0))

            # print(len(dat))

            O = nx.Graph(dat)

            #O.add_edges_from(dat)

            m = np.array(O.nodes)

        out = np.asarray(Ufin)

        Arrange(lab, Ufin, B)

        output(Ufin, V)

        outputs.append(out)

        print(np.count_nonzero(Ufin > 0))

        print(np.count_nonzero(V == 1))

        print(np.count_nonzero(lab == 1))1

        """

    #end = time.time()

    #print("====== End of increasing ======")

    #print("Time", end - start)

    out = np.asarray(Ufin)



    #out = np.asarray(outputs)

    # print(outputs)

    # print(lab)

    np.save(os.path.join("INoutput_data_val.npy"), out)  # generation des outputs"""



#lab = np.load("node2vec/src/graph/labfin.npy",allow_pickle=True)

#dat = np.load("node2vec/src/graph/sam.npy",allow_pickle=True)

#print(lab)

#print(type(dat))

#PatternFinding(dat,lab)
 
 No newline at end of file

Bip_selection.py

0 → 100644
+35 −0
Original line number Diff line number Diff line 


import random,numpy as np,os



def Voisin(x,k):

    if len(k) > 0:

        for i in k:

            if set(x).issubset(set(i)):

                return False

        return True

    return True



V = np.load("node2vec/src/graph/test_Bip.npy", allow_pickle=True)

V = list(V)

k = []

T = []

fo = open("example1.model", "w")

stri = "bc "

compteur = 0



for i in range(len(V)):

    print(V[i+compteur])

    x = V[i+compteur]

    k = x[:2]

    if Voisin(x,T):

        for j in range(V.index(x)+1 ,len(V)):

            y = V[j]

            compteur = compteur + 1

            if x[2:] == y[2:] and x[0] == y[0]:

                k.append(y[1])

            else:

                break

        if len(k) > 1:

            strt = ' '.join(map(str, k))

            stry = ' '.join(map(str, x[2:]))

            fo.write(stri + strt + ',' + stry + "\n")

Bipartie_etiquetage.py

0 → 100644
+26 −0
Original line number Diff line number Diff line 
import networkx as nx



def ensemble(v,K):

    for j in range(len(K)):

        print(type(v),type(K[j]),K[j])

        if B.number_of_edges(v, K[j]) == 1:

            return False

    return True

fh = open("C:/Users/LENOVO/Desktop/karate.edgelist", "rb")

B = nx.read_edgelist(fh,nodetype=int)# Add edges only between nodes of opposite node sets

nx.draw(B,with_labels=True)

K = list(B.nodes)

V = []

V.append([K[0]])

print(K)

print(V,V[0])

for i in range(len(K)):

    print(K[i],V)

    add = False

    for j in range(len(V)):

       if ensemble(K[i],V[j]) == True:

           V[j].append(K[i])

           add = True

    if add == False:

        V.append([K[i]])

Combine.py

0 → 100644
+35 −0
Original line number Diff line number Diff line 
import numpy as np,os

import time



nodes = []

Embedd2 = np.load("INoutput_data_val.npy",allow_pickle=True)

Embedd = np.load("data_val/transformed_0.npy",allow_pickle=True)

print(Embedd.shape)

print(Embedd2.shape)

"""

for i in range(len(Embedd)):

    liste = []

    #print(Embedd)

    for j in range(len(Embedd2[i])):

        nx = np.append(Embedd[i][j],Embedd2[i][j])

        liste.append(nx)

    nodes.append(liste)

    print(i)

node = np.asarray(nodes)

print(node.shape)

np.save(("data_val.npy"),node) #generation des outputs

"""

liste = []

start = time.time()



for i in range(len(Embedd)):

    nx = np.append(Embedd[i],Embedd2[0][i])

    liste.append(nx)

    #print(nx)

    #print(i)

node = np.asarray(liste)

end = time.time()

print(node.shape)

print("Time",end - start)

#print(node)

np.save(("data_tr.npy"),node) #generation des outputs"""
 
 No newline at end of file

Generate.py

0 → 100644
+118 −0
Original line number Diff line number Diff line 
from networkx.generators import community

from networkx.generators import random_graphs

from networkx.algorithms import clique

import networkx as nx

import random,numpy as np,os

import copy

import matplotlib.pyplot as plt



def generate_clique(nb,size,total_size):

    sub=community.caveman_graph(nb,size)#Generation de de graphes en forme de cliques

    G=random_graphs.fast_gnp_random_graph(total_size,0.1)#Generation de graphes aleatoires avec 0.1% de liens

    G=nx.compose(G,sub) #fusion des deux graphes, obtention d'un graphe aleatoire avec nb cliques

    node_mapping = dict(zip(G.nodes(), sorted(G.nodes(), key=lambda k: random.random())))#creation du mapping

    G_new = nx.relabel_nodes(G, node_mapping)#application du mapping

    cliques=list(clique.find_cliques(G_new))

    cliques=np.asarray(([y for x in cliques for y in x  if len(x)>=4]))

    nodes_cliques = np.unique(cliques)

    x = len(nodes_cliques)

    #print("nodes_cliques",x)

    output=np.zeros(total_size)

    output[nodes_cliques]=1

    return G_new,output,x,nodes_cliques,size,nb



def generate_without_clique(total_size):#generation de graphes aleatoires sans cliques

    while True:

        G=random_graphs.fast_gnp_random_graph(total_size,0.04)

        cliques=list(clique.find_cliques(G))

        cliques=[x for x in cliques if len(x)>=6]

        if len(cliques)==0:

            break

    return G, np.zeros(total_size)





def to_input_shape(G):# remplissage du fichier .edgelist format noeud -> noeud-voisin

    tab=[]

    for a,b in G.edges():

        tab.append([a,b])

    return tab



BASE_PATH = "data"

DIR ="clique_1"



if(not os.path.exists(BASE_PATH)):

    os.mkdir(BASE_PATH)

PATH = os.path.join(BASE_PATH,DIR)



if(not os.path.exists(PATH)):

    os.mkdir(PATH)



total_size = 100

max_size_clique = 10

max_clique_count = 10

outputs = []

Gr_size = 1000

graph = []

data = []

lab = []

nodes = []

input = []

sz = []

B = [None]*total_size

x = 0

for id in range(Gr_size):

    G,labels,y,z,s,ng = generate_clique(random.randint(4,max_clique_count),random.randint(4,max_size_clique),total_size)



    tab = to_input_shape(G)

    graph.append(tab)

    A = nx.adjacency_matrix(G, nodelist=range(total_size), weight='weight')

    A.setdiag(A.diagonal() * 2)

    A = A.todense()

    B = copy.deepcopy(A)

    for i in range(len(B)):

        if i not in z:

            B[i] = 0

    outputs.append(y)

    lab.append(labels)

    data.append(B)

    T = nx.edges(G)

    T = np.asarray(T)

    E = T

    for i in range(len(E)):

        x = E[i,0]

        c = E[i,1]

        if (x not in z) and (c not in z):

            w = -1

            t = np.argwhere(T == (x, y))

            d = np.argwhere(T == (c, x))

            t = np.concatenate((t, d))



            for r in range(len(t)):

                for k in range(len(t)):

                    if (t[r, 0] == t[k, 0]) and r != k and w != t[r, 0]:

                        w = t[r, 0]

                        #print(w)

            P = np.delete(T,w,axis=0)

            T=P

    print("id",id)

    sz.append(T)





np.save(os.path.join(PATH, "size.npy"), np.asarray(sz)) ###########################

#np.save(os.path.join(PATH, "data.npy"), np.asarray(graph)) ############################

#np.save(os.path.join(PATH, "data2.npy"), np.asarray(data)) ##########################

#print("out",sz[0])

#print("out",graph[0])

#print("out",data[0])



output = np.asarray(outputs)

#np.save(os.path.join(PATH,"output.npy"),output) #generation des outputs #######################

#print("out",output[0])

labs = np.asarray(lab)

np.save(os.path.join(PATH,"labels2.npy"),labs) #generation des outputs ##########################

#print("labs",labs[0])

#print(s)

print(len(sz[0]))



#nx.draw(G,with_labels=True)

#plt.show()
 
 No newline at end of file

Generate_biparti.py

0 → 100644
+102 −0
Original line number Diff line number Diff line 


import networkx as nx

import matplotlib.pyplot as plt

from networkx.generators import random_graphs

import random,numpy as np,os

import copy



def generate_clique(nb,size,total_size):

    sub = nx.complete_bipartite_graph(nb, size)

    G=random_graphs.fast_gnp_random_graph(total_size,0.01)#Generation de graphes aleatoires avec 0.1% de liens

    GS=nx.compose(G,sub) #fusion des deux graphes, obtention d'un graphe aleatoire avec nb cliques

    node_mapping = dict(zip(GS.nodes(), sorted(GS.nodes(), key=lambda k: random.random())))#creation du mapping

    G_new = nx.relabel_nodes(GS, node_mapping)#application du mapping

    A = nx.adjacency_matrix(G_new,nodelist=sorted(G.nodes()),weight='weight')

    A.setdiag(A.diagonal() * 2)

    A = A.todense()

    for i in range(len(A)):

        if (np.count_nonzero(A[i] == 1) > 4):

            Bipartie.append(i)

    output=np.zeros(total_size)

    output[Bipartie]=1

    return G_new,output,len(Bipartie),Bipartie,(size+nb),A



def to_input_shape(G):# remplissage du fichier .edgelist format noeud -> noeud-voisin

    tab=[]

    for a,b in G.edges():

        tab.append([a,b])

    return tab



BASE_PATH = "data"

DIR ="Bipartie"



if(not os.path.exists(BASE_PATH)):

    os.mkdir(BASE_PATH)

PATH = os.path.join(BASE_PATH,DIR)



if(not os.path.exists(PATH)):

    os.mkdir(PATH)



total_size = 100

max_size_clique = 30

max_clique_count = 30

outputs = []

Gr_size = 1

graph = []

data = []

lab = []

nodes = []

input = []

sz = []

B = [None]*total_size

x = 0

for id in range(Gr_size):

    Bipartie = []

    G,labels,y,z,s,A = generate_clique(random.randint(5,max_clique_count),random.randint(5,max_size_clique),total_size)

    tab = to_input_shape(G)

    graph.append(tab)

    B = copy.deepcopy(A)

    input.append(A)

    for i in range(len(B)):

        if i not in z:

            B[i] = 0

    outputs.append(y)

    lab.append(labels)

    data.append(B)

    T = nx.edges(G)

    T = np.asarray(T)

    E = T



    for i in range(len(E)):

        x = E[i, 0]

        c = E[i, 1]

        if (x not in z) and (c not in z):

            w = -1

            t = np.argwhere(T == (x, c))

            d = np.argwhere(T == (c, x))

            t = np.concatenate((t, d))

            for r in range(len(t)):

                for k in range(len(t)):

                    if (t[r, 0] == t[k, 0]) and r != k and w != t[r, 0]:

                        w = t[r, 0]

                        print("w", w)

            P = np.delete(T, w, axis=0)

            print(len(P), E[i])

            T = P



    sz.append(T)





output = np.asarray(outputs)

labs = np.asarray(lab)

node = np.asarray(input)





nx.draw(G, with_labels=True)

plt.show()

np.save(os.path.join(PATH, "size.npy"), np.asarray(sz[0])) ###########################

np.save(os.path.join(PATH, "data.npy"), np.asarray(graph)) ############################

np.save(os.path.join(PATH, "data2.npy"), np.asarray(data)) ##########################

np.save(os.path.join(PATH,"output.npy"),output) #generation des outputs #######################

np.save(os.path.join(PATH,"labels2.npy"),labs) #generation des outputs ##########################

np.save(os.path.join(PATH,"nodes.npy"),node) #generation des outputs

Generate_chaines.py

0 → 100644
+128 −0
Original line number Diff line number Diff line 
import networkx as nx

import matplotlib.pyplot as plt

from networkx.generators import random_graphs

import random,numpy as np,os

import copy



def find_all_paths(graph, start, end, path=[]):

    path = path + [start]

    if start == end:

        return [path]

    paths = []

    for node in graph[start]:

        if node not in path:

            newpaths = find_all_paths(graph, node, end, path)

            for newpath in newpaths:

                paths.append(newpath)

    return paths

def Chaines():

    G = random_graphs.fast_gnp_random_graph(100, 0.01)

    nx.draw(G,with_labels=True)

    plt.show()



    chaine = nx.chain_decomposition(G,1)

    y = []

    for i in range(100):

        print("hnaya")

        for j in range(100):

            if i != j:

                x = find_all_paths( G, i, j)

                if(len(x) > 0):

                    y.append(x)

    tab = []

    R = []

    for i in range(len(y)):

        if len(y[i]) > 1:

            x = y[i]

            for j in range(len(y[i])):

                for z in range(len(y[i])):

                    if set(x[j]).issubset(set(x[z])) and len(x[z]) > len(x[j]):

                        tab.append(j)

                    else:

                        if set(x[z]).issubset(set(x[j])) and len(x[z]) < len(x[j]):

                            tab.append(z)

            for k in range(len(x)):

                if k not in tab:

                    R.append(x[k])

            tab = []

    print(R)

    return G,R

def generate_clique(nb,size,total_size):

    Chaine = []

    G,ch = Chaines()

    A = nx.adjacency_matrix(G,nodelist=sorted(G.nodes()),weight='weight')

    A.setdiag(A.diagonal() * 2)

    A = A.todense()

    for i in range(len(ch)):

        x = ch[i]

        for j in range(len(x)):

            if x[j] not in Chaine:

                Chaine.append(x[j])

    print("hay la chaine",Chaine)

    output=np.zeros(total_size)

    output[Chaine]=1

    return G,output,len(Chaine),Chaine,(size+nb),A



def to_input_shape(G):# remplissage du fichier .edgelist format noeud -> noeud-voisin

    tab=[]

    for a,b in G.edges():

        tab.append([a,b])

    return tab



BASE_PATH = "data"

DIR ="Bipartie"



if(not os.path.exists(BASE_PATH)):

    os.mkdir(BASE_PATH)

PATH = os.path.join(BASE_PATH,DIR)



if(not os.path.exists(PATH)):

    os.mkdir(PATH)



total_size = 100

max_size_clique = 30

max_clique_count = 30

outputs = []

Gr_size = 1000

graph = []

data = []

lab = []

nodes = []

input = []

sz = []

B = [None]*total_size

x = 0

for id in range(1):

    Bipartie = []

    G,labels,y,z,s,A = generate_clique(random.randint(5,max_clique_count),random.randint(5,max_size_clique),total_size)

    tab = to_input_shape(G)

    graph.append(tab)

    B = copy.deepcopy(A)

    input.append(A)

    for i in range(len(B)):

        if i not in z:

            B[i] = 0

    outputs.append(y)

    lab.append(labels)

    data.append(B)

    sz.append(s)

    print(id)



output = np.asarray(outputs)

labs = np.asarray(lab)

node = np.asarray(input)

print("sz",sz[0])

print("graphe",graph[0])

print("matrix",data[0])

print("out",output[0])

print("labs",labs[0])

print("nodes",node[0])



"""

np.save(os.path.join(PATH, "size.npy"), np.asarray(sz)) ###########################

np.save(os.path.join(PATH, "data.npy"), np.asarray(graph)) ############################

np.save(os.path.join(PATH, "data2.npy"), np.asarray(data)) ##########################

np.save(os.path.join(PATH,"output.npy"),output) #generation des outputs #######################

np.save(os.path.join(PATH,"labels2.npy"),labs) #generation des outputs ##########################

np.save(os.path.join(PATH,"nodes.npy"),node) #generation des outputs

"""

Generate_stars.py

0 → 100644
+124 −0
Original line number Diff line number Diff line 
from networkx.generators import community

from networkx.generators import random_graphs

from networkx.algorithms import clique

import networkx as nx

import random,numpy as np,os

import copy

import matplotlib.pyplot as plt



def generate_clique(nb,size,total_size):

    j = 0

    sub=nx.Graph()#Generation de de graphes en forme de cliques

    for i in range(nb):

        nx.add_star(sub,[j,j+1,j+2,j+3,j+4,j+5])

        j = j + 6

    #nx.draw(sub)

    #plt.show()

    G=random_graphs.fast_gnp_random_graph(total_size,0.000000001)#Generation de graphes aleatoires avec 0.1% de liens

    G=nx.compose(G,sub) #fusion des deux graphes, obtention d'un graphe aleatoire avec nb cliques

    #nx.draw(G)

    #plt.show()

    node_mapping = dict(zip(G.nodes(), sorted(G.nodes(), key=lambda k: random.random())))#creation du mapping

    G_new = nx.relabel_nodes(G, node_mapping)#application du mapping

    A = nx.adjacency_matrix(G_new, nodelist=range(total_size), weight='weight')

    A.setdiag(A.diagonal() * 2)

    A = A.todense()

    B = copy.deepcopy(A)

    output=np.zeros(total_size)

    sortie=np.zeros(total_size)

    k = []

    for i in range(len(B)):

        if (np.count_nonzero(A[i] == 1) < 5):

            B[i] = 0

        else:

            sortie[i] = 1

            k.append(i)

            for j in range(len(B)):

                if B[i,j] == 1:

                    sortie[j] = 1

                    k.append(j)

    print("k",len(k),k)

    return G_new,sortie,4,B,5,A,k



def to_input_shape(G):# remplissage du fichier .edgelist format noeud -> noeud-voisin

    tab=[]

    for a,b in G.edges():

        tab.append([a,b])

    return tab



BASE_PATH = "data"

DIR ="star"



if(not os.path.exists(BASE_PATH)):

    os.mkdir(BASE_PATH)

PATH = os.path.join(BASE_PATH,DIR)



if(not os.path.exists(PATH)):

    os.mkdir(PATH)



total_size = 100

max_star_clique = 20

max_star_count = 12

outputs = []

Gr_size = 100

graph = []

data = []

lab = []

nodes = []

input = []

sz = []

x = 0

for id in range(Gr_size):

    G,labels,y,B,s,A,o = generate_clique(random.randint(4,max_star_count),random.randint(4,max_star_clique),total_size)

    #G,labels,y,z,s = generate_clique(,4,total_size)

    tab = to_input_shape(G)

    graph.append(tab)

    outputs.append(y)

    lab.append(labels)

    input.append(A)

    data.append(B)

    T = nx.edges(G)

    T = np.asarray(T)

    print("hay len ya t7a7na",len(T),T)

    E = T

    print("hay len ya t7a7na",len(T))



    for i in range(len(E)):

        x = E[i,0]

        c = E[i,1]

        if (x not in o) and (c not in o):

            w = -1

            t = np.argwhere(T == (x, c))

            d = np.argwhere(T == (c, x))

            t = np.concatenate((t, d))

            print("madkhelch")

            for r in range(len(t)):

                for k in range(len(t)):

                    if (t[r, 0] == t[k, 0]) and r != k and w != t[r, 0]:

                        w = t[r, 0]

                        print("w",w)

            P = np.delete(T,w,axis=0)

            print(len(P),E[i])

            T=P

    print("hay len ya t7a7na",len(T))



    sz.append(T)

    print(T)

    print(y)

    print(id)





print("graphe",len(sz[0]),len(sz))

print("matrix",np.count_nonzero(data[0]==1))

np.save(os.path.join(PATH, "size.npy"), np.asarray(sz[0])) ###########################

np.save(os.path.join(PATH, "data.npy"), np.asarray(graph)) ############################

np.save(os.path.join(PATH, "data2.npy"), np.asarray(data)) ##########################

output = np.asarray(outputs)

np.save(os.path.join(PATH,"output.npy"),output) #generation des outputs #######################

print("out",output[0])

labs = np.asarray(lab)

np.save(os.path.join(PATH,"labels2.npy"),labs) #generation des outputs ##########################

print("labs",np.count_nonzero(labs[0]==1))

node = np.asarray(input)

np.save(os.path.join(PATH,"nodes.npy"),node) #generation des outputs

print("nodes",np.count_nonzero(node[0]==1))
 
 No newline at end of file

GraphEmbedding-master/.gitignore

0 → 100644
+114 −0
Original line number Diff line number Diff line 
# Byte-compiled / optimized / DLL files

__pycache__/

*.py[cod]

*$py.class



# C extensions

*.so



# Distribution / packaging

.Python

build/

develop-eggs/

dist/

downloads/

eggs/

.eggs/

lib/

lib64/

parts/

sdist/

var/

wheels/

*.egg-info/

.installed.cfg

*.egg

MANIFEST



# PyInstaller

#  Usually these files are written by a python script from a template

#  before PyInstaller builds the exe, so as to inject date/other infos into it.

*.manifest

*.spec



# Installer logs

pip-log.txt

pip-delete-this-directory.txt



# Unit test / coverage reports

htmlcov/

.tox/

.nox/

.coverage

.coverage.*

.cache

nosetests.xml

coverage.xml

*.cover

.hypothesis/

.pytest_cache/



# Translations

*.mo

*.pot



# Django stuff:

*.log

local_settings.py

db.sqlite3



# Flask stuff:

instance/

.webassets-cache



# Scrapy stuff:

.scrapy



# Sphinx documentation

docs/_build/



# PyBuilder

target/



# Jupyter Notebook

.ipynb_checkpoints



# IPython

profile_default/

ipython_config.py



# pyenv

.python-version



# celery beat schedule file

celerybeat-schedule



# SageMath parsed files

*.sage.py



# Environments

.env

.venv

env/

venv/

ENV/

env.bak/

venv.bak/



# Spyder project settings

.spyderproject

.spyproject



# Rope project settings

.ropeproject



# mkdocs documentation

/site



# mypy

.mypy_cache/

.dmypy.json

dmypy.json



# Pyre type checker

.pyre/

GraphEmbedding-master/LICENSE

0 → 100644
+21 −0
Original line number Diff line number Diff line 
MIT License



Copyright (c) 2019 Weichen Shen



Permission is hereby granted, free of charge, to any person obtaining a copy

of this software and associated documentation files (the "Software"), to deal

in the Software without restriction, including without limitation the rights

to use, copy, modify, merge, publish, distribute, sublicense, and/or sell

copies of the Software, and to permit persons to whom the Software is

furnished to do so, subject to the following conditions:



The above copyright notice and this permission notice shall be included in all

copies or substantial portions of the Software.



THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE

AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE

SOFTWARE.
 
 No newline at end of file

GraphEmbedding-master/README.md

0 → 100644
+107 −0
Original line number Diff line number Diff line 
# GraphEmbedding



# Method





|   Model   | Paper                                                                                                                      | Note                                                                                        |

| :-------: | :------------------------------------------------------------------------------------------------------------------------- | :------------------------------------------------------------------------------------------ |

| DeepWalk  | [KDD 2014][DeepWalk: Online Learning of Social Representations](http://www.perozzi.net/publications/14_kdd_deepwalk.pdf)   | [【Graph Embedding】DeepWalk：算法原理，实现和应用](https://zhuanlan.zhihu.com/p/56380812)  |

|   LINE    | [WWW 2015][LINE: Large-scale Information Network Embedding](https://arxiv.org/pdf/1503.03578.pdf)                          | [【Graph Embedding】LINE：算法原理，实现和应用](https://zhuanlan.zhihu.com/p/56478167)      |

| Node2Vec  | [KDD 2016][node2vec: Scalable Feature Learning for Networks](https://www.kdd.org/kdd2016/papers/files/rfp0218-groverA.pdf) | [【Graph Embedding】Node2Vec：算法原理，实现和应用](https://zhuanlan.zhihu.com/p/56542707)  |

|   SDNE    | [KDD 2016][Structural Deep Network Embedding](https://www.kdd.org/kdd2016/papers/files/rfp0191-wangAemb.pdf)               | [【Graph Embedding】SDNE：算法原理，实现和应用](https://zhuanlan.zhihu.com/p/56637181)      |

| Struc2Vec | [KDD 2017][struc2vec: Learning Node Representations from Structural Identity](https://arxiv.org/pdf/1704.03165.pdf)        | [【Graph Embedding】Struc2Vec：算法原理，实现和应用](https://zhuanlan.zhihu.com/p/56733145) |





# How to run examples

1. clone the repo and make sure you have installed `tensorflow` or `tensorflow-gpu` on your local machine. 

2. run following commands

```bash

python setup.py install

cd examples

python deepwalk_wiki.py

```



## DisscussionGroup & Related Projects



<html>

    <table style="margin-left: 20px; margin-right: auto;">

        <tr>

            <td>

                公众号：<b>浅梦的学习笔记</b><br><br>

                <a href="https://github.com/shenweichen/GraphEmbedding">

  <img align="center" src="./pics/code.png" />

</a>

            </td>

            <td>

                微信：<b>deepctrbot</b><br><br>

 <a href="https://github.com/shenweichen/GraphEmbedding">

  <img align="center" src="./pics/deepctrbot.png" />

</a>

            </td>

            <td>

<ul>

<li><a href="https://github.com/shenweichen/AlgoNotes">AlgoNotes</a></li>

<li><a href="https://github.com/shenweichen/DeepCTR">DeepCTR</a></li>

<li><a href="https://github.com/shenweichen/DeepMatch">DeepMatch</a></li>

<li><a href="https://github.com/shenweichen/DeepCTR-Torch">DeepCTR-Torch</a></li>

</ul>

            </td>

        </tr>

    </table>

</html>



# Usage

The design and implementation follows simple principles(**graph in,embedding out**) as much as possible.

## Input format

we use `networkx`to create graphs.The input of networkx graph is as follows:

`node1 node2 <edge_weight>`



![](./pics/edge_list.png)

## DeepWalk



```python

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])# Read graph



model = DeepWalk(G,walk_length=10,num_walks=80,workers=1)#init model

model.train(window_size=5,iter=3)# train model

embeddings = model.get_embeddings()# get embedding vectors

```



## LINE



```python

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph



model = LINE(G,embedding_size=128,order='second') #init model,order can be ['first','second','all']

model.train(batch_size=1024,epochs=50,verbose=2)# train model

embeddings = model.get_embeddings()# get embedding vectors

```

## Node2Vec

```python

G=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',

                        create_using = nx.DiGraph(), nodetype = None, data = [('weight', int)])#read graph



model = Node2Vec(G, walk_length = 10, num_walks = 80,p = 0.25, q = 4, workers = 1)#init model

model.train(window_size = 5, iter = 3)# train model

embeddings = model.get_embeddings()# get embedding vectors

```

## SDNE



```python

G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph



model = SDNE(G,hidden_size=[256,128]) #init model

model.train(batch_size=3000,epochs=40,verbose=2)# train model

embeddings = model.get_embeddings()# get embedding vectors

```



## Struc2Vec





```python

G = nx.read_edgelist('../data/flight/brazil-airports.edgelist',create_using=nx.DiGraph(),nodetype=None,data=[('weight',int)])#read graph



model = model = Struc2Vec(G, 10, 80, workers=4, verbose=40, ) #init model

model.train(window_size = 5, iter = 3)# train model

embeddings = model.get_embeddings()# get embedding vectors

```

GraphEmbedding-master/data/flight/brazil-airports.edgelist

0 → 100644
+1074 −0
Original line number Diff line number Diff line 
7 77

29 50

3 35

9 84

25 82

6 28

64 51

108 74

19 4

2 84

7 25

56 25

24 66

23 26

30 51

10 61

50 84

65 84

53 53

9 0

30 15

45 35

10 7

87 52

106 19

70 53

58 130

6 98

103 80

7 19

36 40

25 15

15 4

4 50

2 27

7 98

9 123

1 54

58 65

4 5

96 77

9 127

50 27

1 120

55 2

79 1

3 40

36 67

75 15

67 7

0 67

61 69

9 39

6 41

36 41

49 71

31 51

46 2

49 67

9 9

57 7

6 67

36 5

34 3

58 4

30 69

80 25

61 25

10 66

15 71

36 29

97 25

2 18

69 58

71 71

9 124

51 5

54 54

43 91

38 51

29 4

40 51

30 27

4 42

34 70

7 15

22 45

61 94

3 55

18 5

0 84

25 70

4 48

117 71

15 70

9 50

66 37

3 68

54 64

58 15

106 5

45 47

10 19

25 25

49 52

63 66

0 5

24 2

98 25

83 87

27 31

21 77

54 21

38 40

2 7

74 30

29 29

3 22

77 3

10 71

8 8

6 1

30 30

87 87

9 121

7 4

65 5

5 20

3 60

24 87

6 27

54 33

112 24

129 9

61 49

5 71

74 75

75 75

6 61

36 37

2 117

108 108

125 125

87 49

9 29

80 80

6 87

36 3

1 103

27 52

127 42

61 5

41 71

74 63

69 69

25 4

36 8

78 25

2 30

0 50

74 25

3 31

1 51

58 68

30 5

40 5

2 56

50 30

9 90

66 29

87 74

42 43

15 84

7 27

50 50

29 42

18 17

32 50

70 70

6 52

36 44

6 21

8 7

50 82

122 126

61 36

56 40

36 10

10 5

65 65

63 50

61 30

0 19

94 117

97 6

2 25

94 94

74 0

98 5

74 74

3 4

18 84

95 4

25 55

67 120

54 25

28 4

77 15

51 36

40 31

4 45

1 46

3 48

9 37

24 75

36 43

2 71

9 128

22 2

37 42

61 61

27 117

42 39

25 101

3 65

69 45

2 97

15 97

54 27

80 27

45 0

63 15

36 25

45 52

36 50

15 15

63 1

7 49

0 0

50 41

3 13

51 7

15 53

28 3

122 124

31 61

5 24

101 27

2 36

36 38

65 2

79 54

3 57

27 84

24 82

25 68

47 25

36 82

43 9

41 41

0 2

79 92

33 26

58 120

50 102

25 110

33 84

3 70

36 56

60 45

2 120

67 58

7 61

68 61

24 30

5 68

6 82

45 45

25 31

37 91

0 7

74 50

9 82

51 9

25 1

15 50

21 75

2 5

7 22

83 83

3 24

77 1

10 69

6 7

123 123

88 7

8 51

2 63

31 5

79 63

3 62

77 27

92 21

25 77

6 25

1 84

84 99

63 67

6 83

94 7

127 127

23 25

0 71

3 79

36 39

1 67

8 2

120 50

44 39

89 89

62 4

45 2

72 5

54 97

53 87

1 31

49 5

97 3

93 93

7 97

74 7

66 91

79 30

3 1

49 49

102 2

85 74

31 42

5 27

79 4

3 39

101 23

96 51

120 84

36 0

58 53

0 64

53 48

29 40

51 68

6 42

45 5

1 68

2 74

8 9

50 80

64 21

9 4

24 108

66 42

60 25

108 3

30 70

61 28

76 84

7 64

54 63

25 19

21 93

53 45

6 120

77 51

25 53

15 46

21 71

50 5

67 51

113 51

9 65

0 30

46 71

10 42

8 42

7 18

3 50

29 39

24 77

25 65

81 108

4 53

1 93

2 69

15 69

7 40

22 0

29 91

33 24

25 107

6 71

3 67

54 77

10 10

24 25

3 105

45 50

58 77

68 15

58 108

80 51

7 87

24 63

53 5

3 15

10 38

75 4

58 84

28 29

76 7

5 13

79 10

3 21

9 74

6 2

30 31

41 5

29 58

3 59

7 130

24 84

6 36

60 49

50 66

94 2

9 41

74 76

122 66

3 72

36 58

93 55

7 63

75 63

2 61

23 15

10 31

64 64

26 26

0 25

36 71

1 4

66 74

103 21

2 3

0 63

29 25

25 41

6 5

21 51

65 130

65 27

84 87

40 9

56 50

5 16

113 7

25 83

6 31

64 50

36 91

19 5

67 31

7 30

63 77

113 25

94 5

18 18

23 27

6 49

21 7

2 113

58 18

0 97

79 77

122 122

77 77

121 122

71 27

45 36

130 84

23 23

7 69

42 91

62 30

99 76

69 65

15 5

97 1

7 99

24 51

3 3

51 61

25 50

58 64

46 46

95 69

22 51

3 41

67 2

25 84

6 22

45 105

58 51

76 33

22 21

64 1

68 7

66 51

6 40

30 53

45 3

32 24

80 2

9 10

25 96

81 3

119 66

10 1

58 7

21 113

22 97

10 94

7 66

32 58

38 4

96 27

2 21

39 65

24 58

9 125

51 2

2 15

9 126

33 75

46 55

79 15

9 71

10 95

25 93

6 9

4 41

7 106

84 67

58 58

41 2

46 25

3 52

18 4

24 79

25 71

34 35

8 50

7 42

46 3

94 25

18 30

74 67

66 36

6 69

3 69

84 23

67 57

106 4

24 27

66 82

3 107

45 48

25 26

0 4

24 1

123 125

1 1

64 97

41 39

28 31

120 120

0 58

5 11

77 4

1 71

54 84

6 0

7 5

65 6

5 21

79 50

3 61

77 30

24 86

15 1

23 6

10 41

67 24

22 25

7 80

94 0

18 25

42 42

47 3

70 2

36 36

10 51

58 9

45 21

18 51

10 120

10 29

41 28

8 25

0 27

57 51

98 19

25 5

98 2

2 1

74 24

6 130

58 71

30 4

4 31

84 111

50 97

63 97

59 48

9 91

57 49

6 29

21 27

36 69

0 77

0 75

29 43

127 123

65 51

4 67

10 58

32 3

79 79

9 7

5 73

38 41

36 4

36 9

10 4

71 84

52 58

30 67

69 1

9 63

7 71

24 15

5 83

71 51

120 3

51 25

79 71

125 89

7 93

24 53

98 4

3 5

9 122

51 63

25 48

75 51

21 58

90 129

2 50

0 46

61 106

1 47

57 58

43 123

41 31

50 51

69 67

3 49

59 60

83 27

6 46

36 42

38 106

58 27

7 47

37 43

98 98

25 102

6 64

3 94

29 53

10 15

2 96

58 5

70 45

42 64

48 51

43 125

119 82

9 1

87 5

0 15

66 0

74 3

3 10

75 3

21 67

88 25

28 2

45 15

51 42

75 25

6 15

64 2

17 5

31 29

24 81

25 69

40 43

123 89

74 71

50 71

79 93

94 31

24 4

3 71

79 23

5 69

45 46

51 84

15 21

39 56

22 80

7 83

74 53

6 119

15 51

75 77

28 25

71 4

6 51

79 22

3 25

77 2

51 51

6 6

23 3

84 84

76 25

7 7

71 30

3 63

6 24

45 115

108 87

10 82

27 130

42 40

6 50

45 13

119 10

0 23

37 39

89 90

6 84

3 82

1 102

15 18

30 94

61 4

24 10

5 104

77 33

8 65

27 33

19 61

2 31

0 51

96 96

3 30

1 50

21 63

54 7

2 57

50 31

45 116

79 5

9 89

25 87

6 19

21 25

88 51

75 67

0 79

5 38

57 25

6 53

45 6

66 127

58 30

7 48

9 5

25 99

6 79

10 2

15 27

4 4

45 1

81 32

75 54

24 9

127 125

50 60

74 1

41 69

126 66

3 7

66 123

99 27

30 13

28 5

2 48

85 84

31 8

5 5

79 2

9 66

51 39

6 10

19 41

36 119

2 42

4 69

3 51

25 66

10 98

19 19

2 68

58 25

7 41

56 9

7 84

33 25

25 100

90 90

3 64

45 25

31 71

24 24

62 5

45 51

45 27

0 1

51 120

53 26

27 27

25 63

58 87

6 18

66 4

5 14

70 84

3 18

97 84

82 82

45 78

66 66

31 31

55 50

3 56

27 87

24 83

66 6

6 39

21 21

4 51

60 48

84 53

46 7

59 45

25 109

6 57

3 73

60 46

67 71

28 71

9 25

5 67

7 102

36 7

10 30

25 30

114 25

68 28

70 60

0 24

74 51

70 48

7 120

1 5

80 79

2 2

7 107

48 24

3 27

51 53

6 4

30 1

2 60

76 27

7 1

53 55

92 22

66 1

6 30

19 2

7 31

97 92

94 4

83 84

51 69

23 24

40 58

10 63

7 53

37 37

59 24

69 27

6 74

3 84

10 25

2 106

0 31

74 58

96 3

53 28

49 4

61 45

7 96

24 50

79 25

3 0

10 77

127 89

42 5

4 27

79 7

3 38

125 121

36 65

58 50

79 45

21 1

76 76

60 60

58 28

7 50

97 79

80 1

79 75

57 1

45 84

89 124

10 0

58 6

39 40

63 63

124 123

23 7

48 48

54 76

1 23

122 89

74 15

3 9

77 50

15 41

4 2

28 7

84 32

22 53

91 91

58 31

51 41

6 8

53 27

4 40

2 40

55 55

3 53

9 42

25 64

40 40

21 10

1 92

2 66

22 1

55 45

61 58

6 68

3 66

69 71

58 1

64 84

42 58

90 66

45 49

25 27

22 93

50 53

36 31

75 7

4 84

28 30

2 9

5 12

79 21

53 2

27 51

25 39

6 3

4 23

65 7

66 58

79 51

3 58

77 31

24 85

25 73

38 69

45 118

10 40

2 93

67 25

50 67

46 5

74 77

3 29

119 9

52 5

97 51

78 79

27 70

9 31

0 21

53 71

10 28

66 81

67 67

41 42

26 27

41 65

24 7

6 38

1 27

29 37

29 5

78 27

45 7

74 27

111 3

25 40

49 27

21 50

54 0

28 51

2 58

GraphEmbedding-master/data/flight/labels-brazil-airports.txt

0 → 100644
+132 −0
Original line number Diff line number Diff line 
node label

0 0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 1

9 0

10 0

11 3

12 3

13 2

14 3

15 0

16 3

17 3

18 1

19 1

20 3

21 0

22 1

23 1

24 0

25 0

26 1

27 0

28 0

29 1

30 0

31 0

32 2

33 1

34 3

35 3

36 0

37 2

38 1

39 1

40 1

41 0

42 1

43 2

44 3

45 1

46 2

47 2

48 1

49 1

50 0

51 0

52 1

53 0

54 1

55 2

56 2

57 2

58 0

59 2

60 1

61 0

62 2

63 1

64 0

65 1

66 0

67 3

68 3

69 0

70 1

71 0

72 3

73 3

74 0

75 1

76 2

77 0

78 2

79 0

80 1

81 2

82 1

83 2

84 1

85 2

86 2

87 1

88 3

89 2

90 2

91 2

92 2

93 1

94 1

95 3

96 1

97 1

98 1

99 2

100 2

101 2

102 2

103 2

104 3

105 3

106 2

107 3

108 3

109 3

110 3

111 3

112 3

113 2

114 3

115 3

116 3

117 3

118 3

119 2

120 3

121 3

122 3

123 2

124 3

125 2

126 3

127 1

128 3

129 3

130 2

GraphEmbedding-master/data/flight/labels-europe-airports.txt

0 → 100644
+400 −0
Original line number Diff line number Diff line 
node label

0 1

1 1

2 2

3 1

4 0

5 0

6 0

7 0

8 0

9 0

10 0

11 0

12 0

13 1

14 0

15 0

16 0

17 1

18 0

19 0

20 0

21 0

22 0

23 0

24 1

25 1

26 0

27 0

28 3

29 0

30 1

31 0

32 1

33 0

34 0

35 0

36 0

37 0

38 0

39 0

40 0

41 0

42 3

43 0

44 0

45 0

46 1

47 0

48 3

49 1

50 0

51 0

52 1

53 0

54 0

55 0

56 0

57 0

58 0

59 1

60 0

61 0

62 0

63 0

64 0

65 0

66 0

67 0

68 1

69 1

70 0

71 0

72 0

73 0

74 0

75 0

76 1

77 0

78 0

79 0

80 0

81 2

82 0

83 0

84 1

85 2

86 0

87 0

88 0

89 0

90 0

91 1

92 1

93 0

94 0

95 1

96 0

97 0

98 1

99 0

100 0

101 0

102 1

103 0

104 0

105 0

106 0

107 2

108 3

109 3

110 1

111 2

112 3

113 2

114 1

115 3

116 2

117 1

118 1

119 0

120 1

121 1

122 2

123 3

124 3

125 1

126 3

127 3

128 3

129 1

130 1

131 1

132 1

133 1

134 1

135 2

136 1

137 2

138 1

139 0

140 1

141 0

142 0

143 0

144 2

145 3

146 0

147 1

148 2

149 2

150 3

151 1

152 2

153 1

154 1

155 0

156 3

157 3

158 1

159 1

160 2

161 2

162 1

163 0

164 3

165 1

166 1

167 0

168 1

169 1

170 0

171 0

172 0

173 1

174 1

175 0

176 2

177 1

178 0

179 1

180 3

181 2

182 1

183 0

184 1

185 1

186 2

187 1

188 2

189 2

190 2

191 1

192 2

193 2

194 1

195 1

196 3

197 2

198 2

199 2

200 2

201 2

202 1

203 2

204 2

205 3

206 3

207 2

208 1

209 3

210 3

211 3

212 3

213 2

214 1

215 1

216 2

217 2

218 2

219 3

220 3

221 0

222 3

223 2

224 2

225 3

226 3

227 3

228 1

229 2

230 2

231 1

232 3

233 2

234 3

235 3

236 3

237 1

238 3

239 2

240 0

241 1

242 1

243 1

244 0

245 2

246 1

247 2

248 1

249 1

250 3

251 2

252 1

253 3

254 2

255 3

256 3

257 2

258 3

259 3

260 2

261 3

262 3

263 3

264 2

265 2

266 2

267 2

268 3

269 3

270 3

271 2

272 1

273 2

274 3

275 2

276 3

277 2

278 1

279 3

280 1

281 2

282 2

283 3

284 2

285 1

286 3

287 3

288 3

289 3

290 1

291 1

292 0

293 3

294 2

295 2

296 3

297 2

298 3

299 1

300 3

301 3

302 2

303 2

304 3

305 2

306 3

307 2

308 1

309 2

310 1

311 3

312 2

313 2

314 1

315 3

316 3

317 3

318 1

319 3

320 2

321 3

322 3

323 3

324 3

325 2

326 2

327 2

328 3

329 3

330 3

331 2

332 3

333 2

334 2

335 3

336 3

337 3

338 3

339 3

340 3

341 3

342 3

343 2

344 2

345 2

346 2

347 2

348 3

349 1

350 2

351 2

352 3

353 2

354 0

355 1

356 2

357 3

358 3

359 3

360 1

361 1

362 2

363 1

364 2

365 1

366 1

367 3

368 2

369 3

370 2

371 2

372 3

373 2

374 1

375 1

376 1

377 2

378 1

379 2

380 2

381 3

382 3

383 3

384 3

385 3

386 3

387 3

388 3

389 2

390 1

391 2

392 2

393 3

394 1

395 1

396 3

397 2

398 2

GraphEmbedding-master/data/flight/labels-usa-airports.txt

0 → 100644
+1191 −0
Original line number Diff line number Diff line 
node label

10241 1

10243 2

10245 0

16390 1

10247 1

12297 2

16727 3

10257 0

12307 2

16729 3

12314 3

12315 2

10268 1

10272 1

12321 1

12323 0

12324 3

12294 2

10278 2

10279 0

11612 0

16428 1

16429 2

12335 1

12339 0

12638 2

12343 1

16441 3

10298 1

10299 0

12348 2

14397 2

10304 1

15030 3

16454 3

12641 2

10313 2

11834 2

12363 1

14412 3

12365 1

16463 3

10322 2

14691 3

10324 2

10325 2

12375 1

14424 3

10329 2

10333 1

14430 3

16479 2

15376 0

14435 3

12388 3

12389 0

12390 2

12391 0

12392 3

12394 3

16743 3

10348 3

12397 0

14447 3

16744 3

12402 0

14457 0

12412 1

16746 3

16515 3

10372 0

14470 1

15041 0

16520 3

14474 1

14475 2

10380 3

14477 2

16527 3

10385 3

12436 3

14485 1

14487 0

14488 1

12441 0

14492 0

10397 0

16543 3

12448 0

12451 0

10405 1

12455 1

10408 0

10409 1

10414 1

15389 0

14512 0

14736 2

10419 3

15154 2

10423 0

14520 1

12207 1

14524 0

14259 1

12478 0

10431 0

12320 2

10434 0

15051 1

12484 3

14533 2

14534 1

16584 2

11980 0

14539 1

12492 1

12494 1

14543 1

14254 0

10275 0

14551 2

14553 2

16603 2

10460 3

14557 3

12511 1

10466 0

12078 3

10469 0

12519 1

14716 0

14570 0

12523 0

14572 2

14574 0

15741 0

14576 0

14577 2

12206 0

15011 2

14582 1

16634 3

14588 0

15061 2

12544 0

16643 3

12549 1

11697 0

16647 1

12553 0

16651 3

11637 0

16654 3

10511 3

14262 0

14617 3

14618 2

12571 2

16668 3

11057 0

12574 2

12576 3

10529 0

14627 1

14628 2

14630 2

10535 2

16680 3

14633 0

16682 3

14635 0

10540 1

12591 2

13704 2

15411 0

14046 2

12598 3

10551 0

14648 1

15412 0

16698 3

16737 3

10557 1

10558 1

10559 1

14656 2

10561 0

10562 2

16709 3

16710 3

12615 1

10569 2

16715 3

12621 2

12622 3

14672 2

10577 0

14674 0

16723 3

16724 3

10581 0

14679 0

16728 3

12633 2

16730 3

14683 0

14685 0

10590 0

10221 1

16736 3

14689 0

16738 3

16739 3

16741 3

16742 3

10599 0

14696 0

12649 1

14698 0

12652 1

12653 2

14704 0

14706 3

14709 0

14711 0

14712 2

10617 3

10620 0

14718 1

12671 2

14400 3

10627 0

12676 2

10630 2

10631 1

15497 1

14730 0

10640 1

14738 1

10643 1

10990 0

10647 1

14744 3

14745 1

14747 0

15325 2

10654 1

14751 3

12704 1

12705 2

12708 1

12709 1

14761 0

10666 0

10994 0

10670 2

12719 1

12720 2

12721 1

14771 0

10676 0

14750 3

11337 0

12728 1

10996 2

14778 3

10683 1

10685 0

14783 0

12737 2

10693 0

12743 2

14792 3

12745 2

14794 0

12748 2

14797 1

11341 3

14802 1

12755 2

14804 1

14805 1

12758 0

10713 0

10715 3

14812 3

14814 0

10661 0

14816 2

10721 0

14819 1

12772 2

12773 2

12774 3

10728 1

10731 1

14828 0

12370 2

12782 2

14831 0

12784 2

12785 2

10739 1

10744 1

14842 0

10747 0

10754 0

14853 2

16725 2

12807 1

14856 1

10327 2

12815 1

10770 2

12819 0

14869 0

12822 1

10775 1

10778 3

10779 1

10781 0

10783 1

14880 2

10785 0

14769 3

10792 0

12841 1

12844 1

14893 0

14895 1

10800 0

14897 2

14898 2

12851 1

12853 1

10589 1

12855 1

14905 0

12635 2

14908 0

10815 2

10817 3

12866 2

10819 1

12868 2

10821 0

12870 3

14919 1

14922 3

12878 1

13795 0

12882 2

12883 2

12884 0

10838 3

12888 1

12889 0

14938 3

12891 0

12892 0

15023 0

14942 1

14943 1

14944 3

10849 0

12898 0

12899 1

12902 1

14952 0

10857 3

14955 0

15841 1

14960 0

12915 0

10868 0

12917 0

13076 0

10874 0

10687 3

14972 3

10347 3

12932 1

12054 3

12934 3

14785 3

14986 0

14987 2

10349 1

12944 2

12945 0

11615 1

12947 3

14996 2

12951 0

12953 0

12954 0

12955 3

12956 3

10910 2

12847 1

15008 1

11376 2

15010 3

14107 0

15012 2

12965 2

10918 1

15016 0

11719 1

15020 3

10925 1

10926 1

10927 3

15024 0

13768 1

10930 1

15027 0

10184 1

10933 2

12982 0

12983 2

16731 3

12992 0

10945 3

10946 3

15043 2

15045 2

10950 1

14113 1

15048 1

15049 1

13002 1

13003 1

11042 0

15054 3

11617 0

13008 3

10961 1

10965 2

10967 1

13016 2

15069 3

15070 0

13024 2

15074 1

10980 0

13029 0

10982 3

13032 1

13034 0

10140 0

15085 1

15086 2

15090 3

15091 2

13044 3

15093 3

13046 3

10999 3

15096 0

11002 3

11003 0

13052 3

14295 2

13055 2

15108 3

11013 1

11278 0

11619 2

13072 2

15370 0

11027 1

15124 1

13077 2

14468 1

11036 2

13087 2

11041 2

15138 1

10139 1

10938 2

11049 0

11050 3

11053 2

15153 1

11058 3

13109 1

13111 2

15160 2

13871 0

11066 0

13115 3

13117 3

11007 2

13121 1

15171 3

11076 1

13125 3

12769 1

15177 2

16345 2

12087 2

11085 1

11086 3

13139 0

11092 1

11097 0

11098 2

13151 3

11106 3

11067 0

11109 0

13158 0

11111 0

15215 0

15855 1

11122 0

11123 1

13873 0

11126 1

13176 1

15167 1

13182 1

11135 3

13184 0

15235 3

15236 2

13192 3

10732 0

11146 0

13195 3

13196 1

15245 1

11150 0

15248 2

15249 0

15250 2

13203 2

13204 0

14489 0

15257 1

13211 1

15511 3

11165 3

13219 3

13221 2

11174 3

11176 3

10396 1

13226 3

13127 0

15278 2

14493 2

13232 0

15282 1

11188 3

13241 1

11423 0

13244 0

11197 1

13246 3

15295 0

11200 2

10400 3

11203 0

13255 2

15304 0

13259 3

11214 1

13264 0

14150 1

15323 0

13277 0

11230 3

11233 1

13282 1

14843 0

10065 2

13891 0

11241 2

13290 0

15339 2

13292 3

11245 2

14559 2

13295 2

15344 3

13297 3

11252 0

13303 0

11259 0

15356 0

11267 0

10141 1

13484 1

11274 0

12119 1

11764 1

15374 1

11280 2

11282 3

11283 1

15380 0

15381 1

11288 1

15385 2

13487 0

11292 0

13341 1

13342 0

13344 1

10416 1

11298 0

13347 0

13348 1

11833 1

15401 1

14855 2

11308 0

15406 1

13360 0

13361 3

11315 1

13061 0

11317 2

13367 0

13369 1

15422 3

11445 1

13377 0

15427 2

13381 2

12129 1

11336 1

15433 3

13495 0

13388 1

13389 3

11789 2

15440 3

10424 3

16540 2

15203 3

13397 2

15446 2

15447 2

15448 2

14863 2

11450 3

15454 1

10085 2

15458 3

14182 2

14222 1

13415 2

13418 2

13422 0

11375 2

13424 2

13502 0

15478 1

15481 2

13434 1

11388 3

11391 2

11392 1

13504 1

13200 1

11778 0

11399 1

11401 1

13450 3

14871 2

14530 2

15502 1

15504 2

11411 1

13461 1

13463 3

13464 2

13467 1

11421 2

16581 2

14875 1

13476 0

15525 3

11430 3

11431 2

11433 0

15532 2

13485 0

13486 0

14877 1

11441 1

13490 1

15539 3

13493 3

11447 1

15546 3

12831 1

15548 3

11453 2

15550 3

14197 2

11456 2

13505 3

15554 1

13507 2

11463 2

16588 2

11468 1

11470 0

11471 1

13520 1

15569 1

15570 1

14542 2

11479 2

11481 0

15579 1

11484 2

15581 1

15582 1

11487 2

13536 2

13537 3

11492 1

13541 0

13543 1

13546 3

11503 0

15601 2

11506 3

13555 3

10135 0

11510 2

15607 0

11512 2

15231 2

13564 2

11517 2

13525 3

15232 3

15618 2

11525 1

13574 2

10613 1

15624 0

13577 0

15626 0

13579 1

15628 1

11535 1

12250 1

11537 0

15634 3

13587 3

11540 0

11140 0

11823 0

11550 1

13599 2

15650 3

11555 0

11559 2

15656 2

11563 0

10478 2

13619 3

11577 0

10463 3

13194 1

13630 3

10615 2

11588 1

12854 2

15798 1

11592 3

16553 3

12173 0

11603 0

15700 2

13198 0

13655 2

13540 3

15706 3

15707 2

15708 2

15709 1

14565 2

15713 3

11618 0

13667 3

12857 2

11624 0

15721 1

13674 2

11627 3

13676 2

15725 2

11630 0

11633 3

13682 3

15731 2

13685 3

11638 0

11641 0

11643 2

13693 2

11646 2

11648 0

11299 3

11650 3

12663 2

13700 3

11656 3

10476 2

15754 2

11659 3

11842 2

13710 1

11663 3

12184 0

13714 2

13715 2

11669 3

14314 0

11673 2

11675 2

10317 3

15774 3

12869 3

13398 1

11684 1

13736 3

15785 1

11695 0

15793 2

15794 1

11699 0

11028 2

13756 2

11711 2

11714 3

14467 3

13767 1

11720 2

11721 0

13770 1

11511 2

13772 3

11725 1

11726 1

11728 2

15825 2

15939 3

11732 3

13781 3

13785 3

13788 3

15838 3

13689 2

14371 3

15842 3

15843 2

13796 0

14303 0

15847 3

11493 2

11756 3

15853 2

13806 2

15599 2

15856 2

13809 2

11762 1

11344 2

15861 1

15862 1

15273 3

11415 3

15163 3

10154 0

11775 0

16498 2

13826 2

15275 3

13829 1

13830 0

15879 3

13832 1

15881 3

15882 3

15885 2

11790 3

15887 1

13827 3

13841 3

13230 0

14100 0

15897 1

13851 0

11867 1

11813 3

13862 1

13863 2

13865 3

13791 3

15919 0

11824 3

11825 3

11827 2

11828 3

14601 3

13881 3

15930 3

10267 3

13884 2

15934 2

11840 3

12896 0

13890 3

11531 1

11844 1

11845 1

13894 1

14945 1

12124 1

13898 2

13899 3

13731 2

15534 3

11856 3

14102 1

11859 2

11331 3

11193 0

11865 1

15057 2

12559 1

11869 0

11870 3

11871 3

15971 3

15973 3

11879 1

10011 1

14268 2

13930 0

13931 0

11413 0

13933 0

13934 2

15293 1

13936 3

11198 3

13942 1

15991 1

11898 0

11899 2

13949 2

11905 1

13459 0

11883 3

11908 1

13958 2

11495 2

11884 0

13964 1

13763 3

11885 3

11921 0

13970 0

11924 2

15982 3

11545 2

11931 1

13983 1

13984 2

11938 3

13987 2

11941 1

11947 3

12332 3

11952 1

13256 0

14004 2

12264 0

14006 1

11538 3

12916 2

14013 2

14646 3

14965 3

12235 2

11973 0

11975 3

14024 3

14025 0

14027 0

14028 2

11982 0

11986 0

15310 3

11992 2

12771 2

11994 3

16091 1

11996 0

11997 1

13105 1

14288 1

14050 2

12003 0

16101 0

14054 1

12007 0

14057 0

12583 2

12012 1

14062 2

12016 0

12018 3

16681 3

12244 1

12028 3

16665 3

12245 1

14081 1

14082 0

12587 3

10923 2

16133 0

12039 3

14088 1

15472 2

14091 3

14092 0

15063 3

14098 0

11977 0

10005 2

10006 3

12055 2

10543 3

14108 0

14109 1

10015 3

10016 2

10017 3

11227 1

12068 3

10886 3

11995 0

14119 2

14120 1

10204 1

14122 0

14125 1

10030 3

12079 3

10033 3

14130 1

14131 1

10039 3

10040 3

10041 3

10042 2

12255 1

12094 0

10050 3

10052 3

10053 3

11367 2

12103 3

10056 3

12599 2

12108 2

10064 3

16696 3

14307 0

14695 3

14167 2

16218 0

10076 2

12222 2

12127 3

14992 1

12131 2

12132 3

13535 2

12134 3

14993 2

12263 3

12141 2

13629 3

14193 0

12147 3

12149 2

10361 1

12155 3

12156 0

16567 1

13771 2

14112 0

16704 3

14044 3

14218 2

12171 1

12610 1

12174 1

12175 1

12177 1

11587 1

14229 2

14231 1

10136 0

12185 2

14234 1

12187 2

14236 2

14237 1

12190 1

12191 0

13296 0

10146 1

12195 1

12197 0

14321 0

12779 1

11522 3

10155 0

14252 0

10157 0

10158 0

14255 2

14256 1

13947 3

12211 2

10165 1

12214 1

16311 2

12217 0

10170 0

10171 2

10172 3

14270 3

12223 0

14273 2

14275 1

12228 2

12232 3

10185 0

14282 2

14283 3

11953 0

16336 3

13501 2

10194 1

16339 2

16340 2

14670 3

10198 2

13433 0

11961 2

16346 2

12252 2

12253 1

10917 1

10208 0

16720 3

14306 1

13307 2

14320 2

16721 3

10216 2

12265 0

12266 0

16363 3

12268 1

16722 3

11702 3

10224 3

10225 1

12278 0

12280 0

14332 3

10237 2

16353 2

GraphEmbedding-master/examples/__init__.py

0 → 100644
+9 −0
Original line number Diff line number Diff line 
from .models import *

from .deepwalk import DeepWalk

from .node2vec import Node2Vec

from .line import LINE

from .sdne import SDNE

from .struc2vec import Struc2Vec





__all__ = ["DeepWalk", "Node2Vec", "LINE", "SDNE", "Struc2Vec"]
 
 No newline at end of file

GraphEmbedding-master/examples/alias.py

0 → 100644
+54 −0
Original line number Diff line number Diff line 
import numpy as np





def create_alias_table(area_ratio):

    """



    :param area_ratio: sum(area_ratio)=1

    :return: accept,alias

    """

    l = len(area_ratio)

    accept, alias = [0] * l, [0] * l

    small, large = [], []

    area_ratio_ = np.array(area_ratio) * l

    for i, prob in enumerate(area_ratio_):

        if prob < 1.0:

            small.append(i)

        else:

            large.append(i)



    while small and large:

        small_idx, large_idx = small.pop(), large.pop()

        accept[small_idx] = area_ratio_[small_idx]

        alias[small_idx] = large_idx

        area_ratio_[large_idx] = area_ratio_[large_idx] - \

            (1 - area_ratio_[small_idx])

        if area_ratio_[large_idx] < 1.0:

            small.append(large_idx)

        else:

            large.append(large_idx)



    while large:

        large_idx = large.pop()

        accept[large_idx] = 1

    while small:

        small_idx = small.pop()

        accept[small_idx] = 1



    return accept, alias





def alias_sample(accept, alias):

    """



    :param accept:

    :param alias:

    :return: data_train index

    """

    N = len(accept)

    i = int(np.random.random()*N)

    r = np.random.random()

    if r < accept[i]:

        return i

    else:

        return alias[i]

GraphEmbedding-master/examples/classify.py

0 → 100644
+85 −0
Original line number Diff line number Diff line 
from __future__ import print_function





import numpy

from sklearn.metrics import f1_score, accuracy_score

from sklearn.multiclass import OneVsRestClassifier

from sklearn.preprocessing import MultiLabelBinarizer





class TopKRanker(OneVsRestClassifier):

    def predict(self, X, top_k_list):

        probs = numpy.asarray(super(TopKRanker, self).predict_proba(X))

        all_labels = []

        for i, k in enumerate(top_k_list):

            probs_ = probs[i, :]

            labels = self.classes_[probs_.argsort()[-k:]].tolist()

            probs_[:] = 0

            probs_[labels] = 1

            all_labels.append(probs_)

        return numpy.asarray(all_labels)





class Classifier(object):



    def __init__(self, embeddings, clf):

        self.embeddings = embeddings

        self.clf = TopKRanker(clf)

        self.binarizer = MultiLabelBinarizer(sparse_output=True)



    def train(self, X, Y, Y_all):

        self.binarizer.fit(Y_all)

        X_train = [self.embeddings[x] for x in X]

        Y = self.binarizer.transform(Y)

        self.clf.fit(X_train, Y)



    def evaluate(self, X, Y):

        top_k_list = [len(l) for l in Y]

        Y_ = self.predict(X, top_k_list)

        Y = self.binarizer.transform(Y)

        averages = ["micro", "macro", "samples", "weighted"]

        results = {}

        for average in averages:

            results[average] = f1_score(Y, Y_, average=average)

        results['acc'] = accuracy_score(Y,Y_)

        print('-------------------')

        print(results)

        return results

        print('-------------------')



    def predict(self, X, top_k_list):

        X_ = numpy.asarray([self.embeddings[x] for x in X])

        Y = self.clf.predict(X_, top_k_list=top_k_list)

        return Y



    def split_train_evaluate(self, X, Y, train_precent, seed=0):

        state = numpy.random.get_state()



        training_size = int(train_precent * len(X))

        numpy.random.seed(seed)

        shuffle_indices = numpy.random.permutation(numpy.arange(len(X)))

        X_train = [X[shuffle_indices[i]] for i in range(training_size)]

        Y_train = [Y[shuffle_indices[i]] for i in range(training_size)]

        X_test = [X[shuffle_indices[i]] for i in range(training_size, len(X))]

        Y_test = [Y[shuffle_indices[i]] for i in range(training_size, len(X))]



        self.train(X_train, Y_train, Y)

        numpy.random.set_state(state)

        return self.evaluate(X_test, Y_test)





def read_node_label(filename, skip_head=False):

    fin = open(filename, 'r')

    X = []

    Y = []

    while 1:

        if skip_head:

            fin.readline()

        l = fin.readline()

        if l == '':

            break

        vec = l.strip().split(' ')

        X.append(vec[0])

        Y.append(vec[1:])

    fin.close()

    return X, Y

GraphEmbedding-master/examples/deepwalk.py

0 → 100644
+63 −0
Original line number Diff line number Diff line 
# -*- coding:utf-8 -*-



"""







Author:



    Weichen Shen,wcshen1994@163.com







Reference:



    [1] Perozzi B, Al-Rfou R, Skiena S. Deepwalk: Online learning of social representations[C]//Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2014: 701-710.(http://www.perozzi.net/publications/14_kdd_deepwalk.pdf)







"""

import walker

from gensim.models import Word2Vec

import pandas as pd





class DeepWalk:

    def __init__(self, graph, walk_length, num_walks, workers=16):



        self.graph = graph

        self.w2v_model = None

        self._embeddings = {}

        self.walker = walker.RandomWalker(

            graph, p=1, q=1, )

        self.sentences = self.walker.simulate_walks(

            num_walks=num_walks, walk_length=walk_length, workers=workers, verbose=1)



    def train(self, embed_size=6, window_size=5, workers=16, iter=4, **kwargs):



        kwargs["sentences"] = self.sentences

        kwargs["min_count"] = kwargs.get("min_count", 0)

        kwargs["size"] = embed_size

        kwargs["sg"] = 1  # skip gram

        kwargs["hs"] = 1  # deepwalk use Hierarchical Softmax

        kwargs["workers"] = workers

        kwargs["window"] = window_size

        kwargs["iter"] = iter



        print("Learning embedding vectors...")

        model = Word2Vec(**kwargs)

        print("Learning embedding vectors done!")



        self.w2v_model = model

        return model



    def get_embeddings(self,):

        if self.w2v_model is None:

            print("model not train")

            return {}



        self._embeddings = {}

        for word in self.graph.nodes():

            self._embeddings[word] = self.w2v_model.wv[word]



        return self._embeddings

GraphEmbedding-master/examples/deepwalk_wiki.py

0 → 100644
+106 −0
Original line number Diff line number Diff line 


import numpy as np

import time

from sklearn.linear_model import LogisticRegression

import classify

import deepwalk

import matplotlib.pyplot as plt

import networkx as nx

from sklearn.manifold import TSNE

from gensim.models import Word2Vec

from multiprocessing import Pool

from itertools import repeat



def evaluate_embeddings(embeddings):

    X, Y = classify.read_node_label('../data/wiki/wiki_labels.txt')

    tr_frac = 0.8

    print("Training classifier using {:.2f}% nodes...".format(

        tr_frac * 100))

    clf = classify.Classifier(embeddings=embeddings, clf=LogisticRegression())

    clf.split_train_evaluate(X, Y, tr_frac)



def read(arr):

    G = nx.Graph()

    for a in arr:

        print(a[0])

        if not G.has_node(a[0]):

            G.add_node(a[0])

        if not G.has_node(a[1]):

            G.add_node(a[1])

        G.add_edge(a[0],a[1],weight=1)

    return G



def plot_embeddings(embeddings,):

    X, Y = classify.read_node_label('../data/wiki/wiki_labels.txt')



    emb_list = []

    for k in X:

        emb_list.append(embeddings[k])

    emb_list = np.array(emb_list)

    print(emb_list.shape)



    model = TSNE(n_components=2)

    node_pos = model.fit_transform(emb_list)



    color_idx = {}

    for i in range(len(X)):

        color_idx.setdefault(Y[i][0], [])

        color_idx[Y[i][0]].append(i)



    for c, idx in color_idx.items():

        plt.scatter(node_pos[idx, 0], node_pos[idx, 1], label=c)

    plt.legend()

    plt.show()

def Affect(z,i):

    #for i in z.wv.index2entity:

        # key = i + 1

        # if key in z.wv.index2entity:

        print(i)

        print(z.index2entity.index(i))

        # print(z.wv.vectors[z.wv.index2entity.index(key)])

        x = z.index2entity.index(i)

        return z.vectors[x]

def read_all():

    data = np.load('graph/test_Bip.npy',allow_pickle=True)

    id=0

    results = [] #a retirer

    print(len(data),len(data[0]),data[0])

    for x in data:

        w = time.time()

        G=read(x)

        #print("A",len(G))

        model = deepwalk.DeepWalk(G, walk_length=5, num_walks=4, workers=1)

        z = model.train(window_size=5, iter=3)

        print(len(z.wv.vectors))

        result = np.zeros((116835, int(6)))



        for i in z.wv.index2entity:

                x = z.wv.index2entity.index(i)

                result[i] = z.wv.vectors[x]

        results.append(result)

        name = str('graph/data_train.npy')

        name = name[:name.index('.')]

        np.save(name+"\\transformed_"+str(id),results)

        print(id, "DONE")

        id += 1

    np.save('graph/test_train.npy', results)

    return model

if __name__ == "__main__":

    '''G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',

                         create_using=nx.DiGraph(), nodetype=None, data=[('weight', int)])



    model = deepwalk.DeepWalk(G, walk_length=15, num_walks=80, workers=1)

    z = model.train(window_size=5, iter=3)

    print(len(z.wv.vectors))

    result = np.zeros((2405, int(64)))

    for i in range(len(z.wv.vectors)):

        key = str(i + 1)

        if key in z.wv.index2entity:

            result[i] = z.wv.vectors[z.wv.index2entity.index(key)]

    result = np.asarray(result)

    print("result shape",result.shape)'''

    model = read_all()

    #embeddings = model.get_embeddings()

    #print(embeddings)

    #evaluate_embeddings(embeddings)

    #plot_embeddings(embeddings)

GraphEmbedding-master/examples/line.py

0 → 100644
+213 −0
Original line number Diff line number Diff line 
# -*- coding:utf-8 -*-



"""







Author:



    Weichen Shen,wcshen1994@163.com







Reference:



    [1] Tang J, Qu M, Wang M, et al. Line: Large-scale information network embedding[C]//Proceedings of the 24th International Conference on World Wide Web. International World Wide Web Conferences Steering Committee, 2015: 1067-1077.(https://arxiv.org/pdf/1503.03578.pdf)







"""

import math

import random



import numpy as np

import tensorflow as tf

from tensorflow.python.keras import backend as K

from tensorflow.python.keras.layers import Embedding, Input, Lambda

from tensorflow.python.keras.models import Model

import alias

import utils

import time





def line_loss(y_true, y_pred):

    return -K.mean(K.log(K.sigmoid(y_true*y_pred)))





def create_model(numNodes, embedding_size, order='second'):



    v_i = Input(shape=(1,))

    v_j = Input(shape=(1,))



    first_emb = Embedding(numNodes, embedding_size, name='first_emb')

    second_emb = Embedding(numNodes, embedding_size, name='second_emb')

    context_emb = Embedding(numNodes, embedding_size, name='context_emb')



    v_i_emb = first_emb(v_i)

    v_j_emb = first_emb(v_j)



    v_i_emb_second = second_emb(v_i)

    v_j_context_emb = context_emb(v_j)



    first = Lambda(lambda x: tf.reduce_sum(

        x[0]*x[1], axis=-1, keepdims=False), name='first_order')([v_i_emb, v_j_emb])

    second = Lambda(lambda x: tf.reduce_sum(

        x[0]*x[1], axis=-1, keepdims=False), name='second_order')([v_i_emb_second, v_j_context_emb])



    if order == 'first':

        output_list = [first]

    elif order == 'second':

        output_list = [second]

    else:

        output_list = [first, second]



    model = Model(inputs=[v_i, v_j], outputs=output_list)



    return model, {'first': first_emb, 'second': second_emb}





class LINE:

    def __init__(self, graph, embedding_size=8, negative_ratio=5, order='second',):

        """



        :param graph:

        :param embedding_size:

        :param negative_ratio:

        :param order: 'first','second','all'

        """

        if order not in ['first', 'second', 'all']:

            raise ValueError('mode must be fisrt,second,or all')



        self.graph = graph

        self.idx2node, self.node2idx = utils.preprocess_nxgraph(graph)

        self.use_alias = True



        self.rep_size = embedding_size

        self.order = order



        self._embeddings = {}

        self.negative_ratio = negative_ratio

        self.order = order



        self.node_size = graph.number_of_nodes()

        self.edge_size = graph.number_of_edges()

        self.samples_per_epoch = self.edge_size*(1+negative_ratio)



        self._gen_sampling_table()

        self.reset_model()



    def reset_training_config(self, batch_size, times):

        self.batch_size = batch_size

        self.steps_per_epoch = (

            (self.samples_per_epoch - 1) // self.batch_size + 1)*times



    def reset_model(self, opt='adam'):



        self.model, self.embedding_dict = create_model(

            self.node_size, self.rep_size, self.order)

        self.model.compile(opt, line_loss)

        self.batch_it = self.batch_iter(self.node2idx)



    def _gen_sampling_table(self):



        # create sampling table for vertex

        power = 0.75

        numNodes = self.node_size

        node_degree = np.zeros(numNodes)  # out degree

        node2idx = self.node2idx



        for edge in self.graph.edges():

            node_degree[node2idx[edge[0]]

                        ] += self.graph[edge[0]][edge[1]].get('weight', 1.0)



        total_sum = sum([math.pow(node_degree[i], power)

                         for i in range(numNodes)])

        norm_prob = [float(math.pow(node_degree[j], power)) /

                     total_sum for j in range(numNodes)]



        self.node_accept, self.node_alias = alias.create_alias_table(norm_prob)



        # create sampling table for edge

        numEdges = self.graph.number_of_edges()

        total_sum = sum([self.graph[edge[0]][edge[1]].get('weight', 1.0)

                         for edge in self.graph.edges()])

        norm_prob = [self.graph[edge[0]][edge[1]].get('weight', 1.0) *

                     numEdges / total_sum for edge in self.graph.edges()]



        self.edge_accept, self.edge_alias = alias.create_alias_table(norm_prob)



    def batch_iter(self, node2idx):



        edges = [(node2idx[x[0]], node2idx[x[1]]) for x in self.graph.edges()]



        data_size = self.graph.number_of_edges()

        shuffle_indices = np.random.permutation(np.arange(data_size))

        # positive or negative mod

        mod = 0

        mod_size = 1 + self.negative_ratio

        h = []

        t = []

        sign = 0

        count = 0

        start_index = 0

        end_index = min(start_index + self.batch_size, data_size)

        while True:

            if mod == 0:



                h = []

                t = []

                for i in range(start_index, end_index):

                    if random.random() >= self.edge_accept[shuffle_indices[i]]:

                        shuffle_indices[i] = self.edge_alias[shuffle_indices[i]]

                    cur_h = edges[shuffle_indices[i]][0]

                    cur_t = edges[shuffle_indices[i]][1]

                    h.append(cur_h)

                    t.append(cur_t)

                sign = np.ones(len(h))

            else:

                sign = np.ones(len(h))*-1

                t = []

                for i in range(len(h)):



                    t.append(alias.alias_sample(

                        self.node_accept, self.node_alias))



            if self.order == 'all':

                yield ([np.array(h), np.array(t)], [sign, sign])

            else:

                yield ([np.array(h), np.array(t)], [sign])

            mod += 1

            mod %= mod_size

            if mod == 0:

                start_index = end_index

                end_index = min(start_index + self.batch_size, data_size)



            if start_index >= data_size:

                count += 1

                mod = 0

                h = []

                shuffle_indices = np.random.permutation(np.arange(data_size))

                start_index = 0

                end_index = min(start_index + self.batch_size, data_size)



    def get_embeddings(self,):

        self._embeddings = {}

        result = np.zeros((100, int(64)))

        if self.order == 'first':

            embeddings = self.embedding_dict['first'].get_weights()[0]

        elif self.order == 'second':

            embeddings = self.embedding_dict['second'].get_weights()[0]

        else:

            embeddings = np.hstack((self.embedding_dict['first'].get_weights()[

                                   0], self.embedding_dict['second'].get_weights()[0]))

        idx2node = self.idx2node

        for i, embedding in enumerate(embeddings):

            self._embeddings[idx2node[i]] = embedding

            result[int(idx2node[i])] = embedding

        return self._embeddings, result



    def train(self, batch_size=1024, epochs=1, initial_epoch=0, verbose=1, times=1):

        self.reset_training_config(batch_size, times)

        hist = self.model.fit_generator(self.batch_it, epochs=epochs, initial_epoch=initial_epoch, steps_per_epoch=self.steps_per_epoch,

                                        verbose=verbose)

        return hist

GraphEmbedding-master/examples/line_wiki.py

0 → 100644
+80 −0
Original line number Diff line number Diff line 


import numpy as np

import classify

from sklearn.linear_model import LogisticRegression

import line

import matplotlib.pyplot as plt

import networkx as nx

from sklearn.manifold import TSNE





def evaluate_embeddings(embeddings):

    X, Y = classify.read_node_label('../data/wiki/wiki_labels.txt')

    tr_frac = 0.8

    print("Training classifier using {:.2f}% nodes...".format(

        tr_frac * 100))

    clf = classify.Classifier(embeddings=embeddings, clf=LogisticRegression())

    clf.split_train_evaluate(X, Y, tr_frac)





def plot_embeddings(embeddings,):

    X, Y = classify.read_node_label('../data/wiki/wiki_labels.txt')



    emb_list = []

    for k in X:

        emb_list.append(embeddings[k])

    emb_list = np.array(emb_list)



    model = TSNE(n_components=2)

    node_pos = model.fit_transform(emb_list)



    color_idx = {}

    for i in range(len(X)):

        color_idx.setdefault(Y[i][0], [])

        color_idx[Y[i][0]].append(i)



    for c, idx in color_idx.items():

        plt.scatter(node_pos[idx, 0], node_pos[idx, 1], label=c)

    plt.legend()

    plt.show()



def read(arr):

	G = nx.Graph()

	for a,b in arr:

		if not G.has_node(a):

			G.add_node(a)

		if not G.has_node(b):

			G.add_node(b)

		G.add_edge(a,b,weight=1)

	return G



def read_all():

    data = np.load('graph/data_val.npy',allow_pickle=True)

    id=0

    for x in data:

        G=read(x)

        model = line.LINE(G, embedding_size=64, order='second')

        model.train(batch_size=120, epochs=100, verbose=2)

        embeddings, result = model.get_embeddings()

        result = np.asarray(result)

        result = np.asarray(result)

        name = str('graph/data_val.npy')

        name = name[:name.index('.')]

        np.save(name+"\\transformed_"+str(id),result)

        print(id,"DONE")

        id+=1

    return model

if __name__ == "__main__":

    '''G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',

                         create_using=nx.DiGraph(), nodetype=None, data=[('weight', int)])



    model = line.LINE(G, embedding_size=64, order='second')

    model.train(batch_size=1024, epochs=50, verbose=2)

    print("hnaya",model)

    embeddings, result = model.get_embeddings()

    #print(embeddings)

    result = np.asarray(embeddings)

    #print(result)

    #evaluate_embeddings(embeddings)

    #plot_embeddings(embeddings)'''

    model = read_all()

GraphEmbedding-master/examples/node2vec.py

0 → 100644
+70 −0
Original line number Diff line number Diff line 
# -*- coding:utf-8 -*-



"""







Author:



    Weichen Shen,wcshen1994@163.com







Reference:



    [1] Grover A, Leskovec J. node2vec: Scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2016: 855-864.(https://www.kdd.org/kdd2016/papers/files/rfp0218-groverA.pdf)







"""



from gensim.models import Word2Vec

import pandas as pd



from ..walker import RandomWalker





class Node2Vec:



    def __init__(self, graph, walk_length, num_walks, p=1.0, q=1.0, workers=1, use_rejection_sampling=0):



        self.graph = graph

        self._embeddings = {}

        self.walker = RandomWalker(

            graph, p=p, q=q, use_rejection_sampling=use_rejection_sampling)



        print("Preprocess transition probs...")

        self.walker.preprocess_transition_probs()



        self.sentences = self.walker.simulate_walks(

            num_walks=num_walks, walk_length=walk_length, workers=workers, verbose=1)



    def train(self, embed_size=128, window_size=5, workers=3, iter=5, **kwargs):



        kwargs["sentences"] = self.sentences

        kwargs["min_count"] = kwargs.get("min_count", 0)

        kwargs["size"] = embed_size

        kwargs["sg"] = 1

        kwargs["hs"] = 0  # node2vec not use Hierarchical Softmax

        kwargs["workers"] = workers

        kwargs["window"] = window_size

        kwargs["iter"] = iter



        print("Learning embedding vectors...")

        model = Word2Vec(**kwargs)

        print("Learning embedding vectors done!")



        self.w2v_model = model



        return model



    def get_embeddings(self,):

        if self.w2v_model is None:

            print("model not train")

            return {}



        self._embeddings = {}

        for word in self.graph.nodes():

            self._embeddings[word] = self.w2v_model.wv[word]



        return self._embeddings

GraphEmbedding-master/examples/node2vec_flight.py

0 → 100644
+88 −0
Original line number Diff line number Diff line 
import numpy as np







from ge.classify import read_node_label,Classifier



from ge import Node2Vec



from sklearn.linear_model import LogisticRegression







import matplotlib.pyplot as plt



import networkx as nx



from sklearn.manifold import TSNE







def evaluate_embeddings(embeddings):



    X, Y = read_node_label('../data/flight/labels-brazil-airports.txt',skip_head=True)



    tr_frac = 0.8



    print("Training classifier using {:.2f}% nodes...".format(



        tr_frac * 100))



    clf = Classifier(embeddings=embeddings, clf=LogisticRegression())



    clf.split_train_evaluate(X, Y, tr_frac)











def plot_embeddings(embeddings,):



    X, Y = read_node_label('../data/flight/labels-brazil-airports.txt',skip_head=True)







    emb_list = []



    for k in X:



        emb_list.append(embeddings[k])



    emb_list = np.array(emb_list)







    model = TSNE(n_components=2)



    node_pos = model.fit_transform(emb_list)







    color_idx = {}



    for i in range(len(X)):



        color_idx.setdefault(Y[i][0], [])



        color_idx[Y[i][0]].append(i)







    for c, idx in color_idx.items():



        plt.scatter(node_pos[idx, 0], node_pos[idx, 1], label=c)  # c=node_colors)



    plt.legend()



    plt.show()



if __name__ == "__main__":

    G = nx.read_edgelist('../data/flight/brazil-airports.edgelist', create_using=nx.DiGraph(), nodetype=None,

                         data=[('weight', int)])



    model = Node2Vec(G, 10, 80, workers=1, p=0.25, q=2, use_rejection_sampling=0)

    model.train()

    embeddings = model.get_embeddings()



    evaluate_embeddings(embeddings)

    plot_embeddings(embeddings)

GraphEmbedding-master/examples/node2vec_wiki.py

0 → 100644
+53 −0
Original line number Diff line number Diff line 


import numpy as np



from ge.classify import read_node_label, Classifier

from ge import Node2Vec

from sklearn.linear_model import LogisticRegression



import matplotlib.pyplot as plt

import networkx as nx

from sklearn.manifold import TSNE





def evaluate_embeddings(embeddings):

    X, Y = read_node_label('../data/wiki/wiki_labels.txt')

    tr_frac = 0.8

    print("Training classifier using {:.2f}% nodes...".format(

        tr_frac * 100))

    clf = Classifier(embeddings=embeddings, clf=LogisticRegression())

    clf.split_train_evaluate(X, Y, tr_frac)





def plot_embeddings(embeddings,):

    X, Y = read_node_label('../data/wiki/wiki_labels.txt')



    emb_list = []

    for k in X:

        emb_list.append(embeddings[k])

    emb_list = np.array(emb_list)



    model = TSNE(n_components=2)

    node_pos = model.fit_transform(emb_list)



    color_idx = {}

    for i in range(len(X)):

        color_idx.setdefault(Y[i][0], [])

        color_idx[Y[i][0]].append(i)



    for c, idx in color_idx.items():

        plt.scatter(node_pos[idx, 0], node_pos[idx, 1], label=c)

    plt.legend()

    plt.show()





if __name__ == "__main__":

    G=nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',

                         create_using = nx.DiGraph(), nodetype = None, data = [('weight', int)])

    model = Node2Vec(G, walk_length=10, num_walks=80,

                     p=0.25, q=4, workers=1, use_rejection_sampling=0)

    model.train(window_size = 5, iter = 3)

    embeddings=model.get_embeddings()



    evaluate_embeddings(embeddings)

    plot_embeddings(embeddings)

GraphEmbedding-master/examples/sdne.py

0 → 100644
+174 −0
Original line number Diff line number Diff line 
# -*- coding:utf-8 -*-



"""







Author:



    Weichen Shen,wcshen1994@163.com







Reference:



    [1] Wang D, Cui P, Zhu W. Structural deep network embedding[C]//Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2016: 1225-1234.(https://www.kdd.org/kdd2016/papers/files/rfp0191-wangAemb.pdf)







"""

import time



import numpy as np

import scipy.sparse as sp

import tensorflow as tf

from tensorflow.python.keras import backend as K

from tensorflow.python.keras.callbacks import History

from tensorflow.python.keras.layers import Dense, Input

from tensorflow.python.keras.models import Model

from tensorflow.python.keras.regularizers import l1_l2



from ..utils import preprocess_nxgraph





def l_2nd(beta):

    def loss_2nd(y_true, y_pred):

        b_ = np.ones_like(y_true)

        b_[y_true != 0] = beta

        x = K.square((y_true - y_pred) * b_)

        t = K.sum(x, axis=-1, )

        return K.mean(t)



    return loss_2nd





def l_1st(alpha):

    def loss_1st(y_true, y_pred):

        L = y_true

        Y = y_pred

        batch_size = tf.to_float(K.shape(L)[0])

        return alpha * 2 * tf.linalg.trace(tf.matmul(tf.matmul(Y, L, transpose_a=True), Y)) / batch_size



    return loss_1st





def create_model(node_size, hidden_size=[256, 128], l1=1e-5, l2=1e-4):

    A = Input(shape=(node_size,))

    L = Input(shape=(None,))

    fc = A

    for i in range(len(hidden_size)):

        if i == len(hidden_size) - 1:

            fc = Dense(hidden_size[i], activation='relu',

                       kernel_regularizer=l1_l2(l1, l2), name='1st')(fc)

        else:

            fc = Dense(hidden_size[i], activation='relu',

                       kernel_regularizer=l1_l2(l1, l2))(fc)

    Y = fc

    for i in reversed(range(len(hidden_size) - 1)):

        fc = Dense(hidden_size[i], activation='relu',

                   kernel_regularizer=l1_l2(l1, l2))(fc)



    A_ = Dense(node_size, 'relu', name='2nd')(fc)

    model = Model(inputs=[A, L], outputs=[A_, Y])

    emb = Model(inputs=A, outputs=Y)

    return model, emb





class SDNE(object):

    def __init__(self, graph, hidden_size=[32, 16], alpha=1e-6, beta=5., nu1=1e-5, nu2=1e-4, ):



        self.graph = graph

        # self.g.remove_edges_from(self.g.selfloop_edges())

        self.idx2node, self.node2idx = preprocess_nxgraph(self.graph)



        self.node_size = self.graph.number_of_nodes()

        self.hidden_size = hidden_size

        self.alpha = alpha

        self.beta = beta

        self.nu1 = nu1

        self.nu2 = nu2



        self.A, self.L = self._create_A_L(

            self.graph, self.node2idx)  # Adj Matrix,L Matrix

        self.reset_model()

        self.inputs = [self.A, self.L]

        self._embeddings = {}



    def reset_model(self, opt='adam'):



        self.model, self.emb_model = create_model(self.node_size, hidden_size=self.hidden_size, l1=self.nu1,

                                                  l2=self.nu2)

        self.model.compile(opt, [l_2nd(self.beta), l_1st(self.alpha)])

        self.get_embeddings()



    def train(self, batch_size=1024, epochs=1, initial_epoch=0, verbose=1):

        if batch_size >= self.node_size:

            if batch_size > self.node_size:

                print('batch_size({0}) > node_size({1}),set batch_size = {1}'.format(

                    batch_size, self.node_size))

                batch_size = self.node_size

            return self.model.fit([self.A.todense(), self.L.todense()], [self.A.todense(), self.L.todense()],

                                  batch_size=batch_size, epochs=epochs, initial_epoch=initial_epoch, verbose=verbose,

                                  shuffle=False, )

        else:

            steps_per_epoch = (self.node_size - 1) // batch_size + 1

            hist = History()

            hist.on_train_begin()

            logs = {}

            for epoch in range(initial_epoch, epochs):

                start_time = time.time()

                losses = np.zeros(3)

                for i in range(steps_per_epoch):

                    index = np.arange(

                        i * batch_size, min((i + 1) * batch_size, self.node_size))

                    A_train = self.A[index, :].todense()

                    L_mat_train = self.L[index][:, index].todense()

                    inp = [A_train, L_mat_train]

                    batch_losses = self.model.train_on_batch(inp, inp)

                    losses += batch_losses

                losses = losses / steps_per_epoch



                logs['loss'] = losses[0]

                logs['2nd_loss'] = losses[1]

                logs['1st_loss'] = losses[2]

                epoch_time = int(time.time() - start_time)

                hist.on_epoch_end(epoch, logs)

                if verbose > 0:

                    print('Epoch {0}/{1}'.format(epoch + 1, epochs))

                    print('{0}s - loss: {1: .4f} - 2nd_loss: {2: .4f} - 1st_loss: {3: .4f}'.format(

                        epoch_time, losses[0], losses[1], losses[2]))

            return hist



    def evaluate(self, ):

        return self.model.evaluate(x=self.inputs, y=self.inputs, batch_size=self.node_size)



    def get_embeddings(self):

        self._embeddings = {}

        embeddings = self.emb_model.predict(self.A.todense(), batch_size=self.node_size)

        look_back = self.idx2node

        for i, embedding in enumerate(embeddings):

            self._embeddings[look_back[i]] = embedding



        return self._embeddings



    def _create_A_L(self, graph, node2idx):

        node_size = graph.number_of_nodes()

        A_data = []

        A_row_index = []

        A_col_index = []



        for edge in graph.edges():

            v1, v2 = edge

            edge_weight = graph[v1][v2].get('weight', 1)



            A_data.append(edge_weight)

            A_row_index.append(node2idx[v1])

            A_col_index.append(node2idx[v2])



        A = sp.csr_matrix((A_data, (A_row_index, A_col_index)), shape=(node_size, node_size))

        A_ = sp.csr_matrix((A_data + A_data, (A_row_index + A_col_index, A_col_index + A_row_index)),

                           shape=(node_size, node_size))



        D = sp.diags(A_.sum(axis=1).flatten().tolist()[0])

        L = D - A_

        return A, L

GraphEmbedding-master/examples/sdne_wiki.py

0 → 100644
+54 −0
Original line number Diff line number Diff line 


import numpy as np



from ge.classify import read_node_label, Classifier

from ge import SDNE

from sklearn.linear_model import LogisticRegression



import matplotlib.pyplot as plt

import networkx as nx

from sklearn.manifold import TSNE





def evaluate_embeddings(embeddings):

    X, Y = read_node_label('../data/wiki/wiki_labels.txt')

    tr_frac = 0.8

    print("Training classifier using {:.2f}% nodes...".format(

        tr_frac * 100))

    clf = Classifier(embeddings=embeddings, clf=LogisticRegression())

    clf.split_train_evaluate(X, Y, tr_frac)





def plot_embeddings(embeddings,):

    X, Y = read_node_label('../data/wiki/wiki_labels.txt')



    emb_list = []

    for k in X:

        emb_list.append(embeddings[k])

    emb_list = np.array(emb_list)



    model = TSNE(n_components=2)

    node_pos = model.fit_transform(emb_list)



    color_idx = {}

    for i in range(len(X)):

        color_idx.setdefault(Y[i][0], [])

        color_idx[Y[i][0]].append(i)



    for c, idx in color_idx.items():

        plt.scatter(node_pos[idx, 0], node_pos[idx, 1],

                    label=c)  # c=node_colors)

    plt.legend()

    plt.show()





if __name__ == "__main__":

    G = nx.read_edgelist('../data/wiki/Wiki_edgelist.txt',

                         create_using=nx.DiGraph(), nodetype=None, data=[('weight', int)])



    model = SDNE(G, hidden_size=[256, 128],)

    model.train(batch_size=3000, epochs=40, verbose=2)

    embeddings = model.get_embeddings()



    evaluate_embeddings(embeddings)

    plot_embeddings(embeddings)

GraphEmbedding-master/ge/models/__init__.py

0 → 100644
+8 −0
Original line number Diff line number Diff line 
from .deepwalk import DeepWalk

from .node2vec import Node2Vec

from .line import LINE

from .sdne import SDNE

from .struc2vec import Struc2Vec





__all__ = ["DeepWalk", "Node2Vec", "LINE", "SDNE", "Struc2Vec"]

PatternRecognition/.gitignore

0 → 100644
+14 −0
Original line number Diff line number Diff line 
*.pyc

.DS_Store

target

bin

build

.gradle

*.iml

*.ipr

*.iws

*.log

.classpath

.project

.settings

.idea
 
 No newline at end of file

Quasi-clique_generate.py

0 → 100644
+98 −0

File added.

Preview size limit exceeded, changes collapsed.

README.md

+4 −14

File changed.

Preview size limit exceeded, changes collapsed.

Skit-Cluster.py

0 → 100644
+175 −0

File added.

Preview size limit exceeded, changes collapsed.

Train2.py

0 → 100644
+97 −0

File added.

Preview size limit exceeded, changes collapsed.

Voisins.py

0 → 100644
+157 −0

File added.

Preview size limit exceeded, changes collapsed.

example1.model

0 → 100644
+3 −0

File added.

Preview size limit exceeded, changes collapsed.

generation.py

0 → 100644
+67 −0

File added.

Preview size limit exceeded, changes collapsed.

graph_node_add.py

0 → 100644
+16 −0

File added.

Preview size limit exceeded, changes collapsed.

merge.py

0 → 100644
+27 −0

File added.

Preview size limit exceeded, changes collapsed.

merge_all.py

0 → 100644
+26 −0

File added.

Preview size limit exceeded, changes collapsed.

merge_embaded.py

0 → 100644
+18 −0

File added.

Preview size limit exceeded, changes collapsed.

model.h5

0 → 100644
+551 KiB

File added.

Preview size limit exceeded, changes collapsed.

View file

model2.h5

0 → 100644
+5.05 MiB

File added.

Preview size limit exceeded, changes collapsed.

View file

split_data.py

0 → 100644
+83 −0

File added.

Preview size limit exceeded, changes collapsed.

train.py

0 → 100644
+118 −0

File added.

Preview size limit exceeded, changes collapsed.

transformation_clus.py

0 → 100644
+119 −0

File added.

Preview size limit exceeded, changes collapsed.