forked from bigscity-buaa/NASR
-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathutils.py
171 lines (150 loc) · 6.36 KB
/
utils.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
import numpy as np
import networkx as nx
#testGraph = [
#[0,0,0,0,1,0,0,0],
#[0,0,0,1,0,0,0,0],
#[0,0,0,0,0,1,0,0],
#[0,0,1,0,0,0,0,0],
#[0,0,0,1,0,1,1,0],
#[0,0,0,0,0,0,1,1],
#[0,0,0,0,0,0,0,1],
#[0,0,0,0,0,0,0,0]
# ]
#adj = np.matrix(testGraph)
def knbrs(G, start, k):
all_node = []
nbrs = set([start])
all_node.extend(list(nbrs))
for l in range(k):
try:
nbrs = set((nbr for n in nbrs for nbr in G[n]))
except:
nbrs = set([])
all_node.extend(list(nbrs))
for i in range(len(all_node)):
if all_node[i] >= 15200:
all_node[i] = i
return_node = []
for i in all_node:
if i not in return_node:
return_node.append(i)
return return_node
def generate_one_graph(road_embedding, G, road_num, cutoff=10, src=0):
node_set = set()
src_node = []
src_emb = []
src_node = knbrs(G, src, cutoff)
while len(src_node) < road_num + 15:
src_node.append(len(src_node))
# if(len(src_node)<10):
# print(len(src_node))
src_emb = [road_embedding[n] for n in src_node]
src_adj = nx.adjacency_matrix(G.subgraph(src_node)).todense()
src_adj = np.array(src_adj[:road_num, :road_num])
adj = np.eye(road_num) #self-loop
adj[:src_adj.shape[0],:src_adj.shape[1]] += src_adj
adj = (adj - 1) * (9999999.0)
mask = [0 for i in range(road_num)]
mask[0] = 1
return adj, src_emb[:road_num], mask
def generate_sub_graph(road_embedding, G, inv_G, road_num, cutoff=10, src=0, des=7):
# input:
# road_embedding feature matrix of roads [road_num * feature_dim]
# road_network road adj matrix [road_num * road_num]
# road_num the size of sub graph
# cutoff max length of path between src and des
# src source node (int)
# def destination node (int)
# return:
# src_adj the adj matrix of subgraph
# src_emb the corresponding emb matrix of subgraph
# des_adj
# des_emb
node_set = set()
# des_G = nx.DiGraph()
# src_G = nx.DiGraph()
src_node = []
des_node = []
src_emb = []
des_emb = []
# print(np.matrix(road_network).shape)
# adj = np.matrix(road_network)[:15500, :15500]
# G = nx.from_numpy_matrix(adj, create_using=nx.DiGraph())
# inv_G = nx.from_numpy_matrix(adj.T, create_using=nx.DiGraph())
# paths_between_generator = nx.all_simple_paths(G,source=src,target=des,cutoff=cutoff)
# for path in reversed(list(paths_between_generator)):
# print(path)
# sub_G.add_path(path)
# for item in path:
# if not item in node_set:
# sub_emb.append(road_embedding[item])
# node_set.append(item)
# if len(sub_emb) >= road_num:
# break
src_node = knbrs(G, src, cutoff)
des_node = knbrs(inv_G, des, cutoff)
while len(src_node) < road_num + 15:
src_node.append(len(src_node))
while len(des_node) < road_num + 15:
des_node.append(len(des_node))
src_emb = [road_embedding[n] for n in src_node]
des_emb = [road_embedding[n] for n in des_node]
des_adj = nx.adjacency_matrix(inv_G.subgraph(des_node)).todense()
src_adj = nx.adjacency_matrix(G.subgraph(src_node)).todense()
src_adj = np.array(src_adj[:road_num, :road_num])
des_adj = np.array(des_adj[:road_num, :road_num])
s_adj = np.eye(road_num) #self-loop
s_adj[:src_adj.shape[0],:src_adj.shape[1]] += src_adj
s_adj = (s_adj - 1) * (9999999.0)
d_adj = np.eye(road_num) #self-loop
d_adj[:des_adj.shape[0],:des_adj.shape[1]] += des_adj
d_adj = (d_adj - 1) * (9999999.0)
mask = [0 for i in range(road_num)]
mask[0] = 1
return d_adj, s_adj, des_emb[:road_num], src_emb[:road_num], mask, mask, src_node[:road_num], des_node[:road_num]
def generate_batch(maskData, historyData, trainData, trainTimeData, trainUser):
for mask_bat, tra_bat, time_bat, user_bat in zip(maskData, trainData, trainTimeData, trainUser):
hour_bat = np.array(time_bat)[:, :, 0]
day_bat = np.array(time_bat)[:, :, 1]
his_bat = []
his_hour_bat = []
his_day_bat = []
his_mask_bat = []
for user in user_bat:
his_bat.append([item[0] for item in historyData[user]])
his_hour_bat.append([[time[0] for time in item[1]] for item in historyData[user]])
his_day_bat.append([[time[1] for time in item[1]] for item in historyData[user]])
his_mask_bat.append([item[2] for item in historyData[user]])
# mask_bat = np.sum(np.array(mask_bat)[:, 1:], 1)
# des = [tra[ind] for tra, ind in zip(tra_bat, mask_bat)]
yield np.array(tra_bat), np.array(hour_bat), np.array(day_bat), np.array(his_bat)[:, :, :-1], np.array(his_hour_bat)[:, :, :-1], np.array(his_day_bat)[:, :, :-1], np.array(his_mask_bat)[:, :, :-1]
# for i in range(2):
# yield np.array(des[i*50: (i+1)*50]), np.array(tra_bat)[i*50: (i+1)*50], np.array(mask_bat)[i*50: (i+1)*50], np.array(hour_bat)[i*50: (i+1)*50, :-1], np.array(day_bat)[i*50: (i+1)*50, :-1], np.array(his_bat)[i*50: (i+1)*50, :, :-1], np.array(his_hour_bat)[i*50: (i+1)*50, :, :-1], np.array(his_day_bat)[i*50: (i+1)*50, :, :-1], np.array(his_mask_bat)[i*50: (i+1)*50, :, :-1]
def generate_st_batch(historyData, trainData, trainTimeData, trainUser, trainGeo):
for tra_bat, time_bat, user_bat, geo_bat in zip(trainData, trainTimeData, trainUser, trainGeo):
time_bat = np.array(time_bat)[:, :, 0] * np.array(time_bat)[:, :, 1]
his_bat = []
his_time_bat = []
his_user_bat = []
his_mask_bat = []
for user in user_bat:
his_bat.append([item[0] for item in historyData[user][:5]])
his_time_bat.append([[np.array(time[0]) * np.array(time[1]) for time in item[1]] for item in historyData[user][:5]])
his_user_bat.append([[user for item_ in range(50)] for item in historyData[user][:5]])
his_mask_bat.append([[0.0 for item_ in range(50)] for item in historyData[user][:5]])
for i in range(len(his_bat)):
for j in range(len(his_bat[i])):
if len(his_bat[i][j]) > 50:
his_bat[i][j] = his_bat[i][j][:50]
his_time_bat[i][j] = his_time_bat[i][j][:50]
for k in range(len(his_bat[i][j])):
his_mask_bat[i][j][k] = 1.0
else:
for k in range(len(his_bat[i][j])):
his_mask_bat[i][j][k] = 1.0
while len(his_bat[i][j]) < 50:
his_bat[i][j].append(0)
while len(his_time_bat[i][j]) < 50:
his_time_bat[i][j].append(0)
yield np.array(tra_bat), np.array(time_bat), np.array(user_bat), np.array(geo_bat), np.array(his_bat), np.array(his_time_bat), np.array(his_user_bat), np.array(his_mask_bat)
#