-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils.py
159 lines (125 loc) · 5.14 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
import os
import torch
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.manifold import TSNE
from sklearn.decomposition import PCA
def load_data(args):
"""Load traffic and create corresponding links."""
if args.dataset == "abilene":
base = "./Dataset/Abilene"
tm_file = os.path.join(base, "Abilene_TM.csv")
rm_file = os.path.join(base, "abilene_rm.csv")
div_num = 10 ** 9
mult_num = 288
train_day = 16 * 7
test_day = 1 * 7
if args.dataset == "geant":
base = "./Dataset/GEANT"
tm_file = os.path.join(base, "GEANT_TM.csv")
rm_file = os.path.join(base, "geant_rm.csv")
div_num = 10 ** 7
mult_num = 96
train_day = 11 * 7
test_day = 1 * 7
data = pd.read_csv(tm_file, header=None)
data.drop(data.columns[-1], axis=1, inplace=True)
data_tensor = torch.from_numpy(data.values / div_num)
data_tensor = data_tensor.float()
rm = pd.read_csv(rm_file, header=None)
rm.drop(rm.columns[-1], axis=1, inplace=True)
rm_tensor = torch.from_numpy(rm.values)
rm_tensor = rm_tensor.float()
train_size = int(train_day * mult_num)
test_size = int(test_day * mult_num)
train_id = np.arange(train_size)
test_id = np.arange(test_size) + train_size
train_flow = data_tensor[train_id]
test_flow = data_tensor[test_id]
test_link = test_flow @ rm_tensor
return train_flow, test_flow, test_link, rm_tensor
class EarlyStopping:
"""Early stops the training if link loss doesn't decrease after a given patience."""
def __init__(self, var_tensor, patience=50, verbose=False):
"""
Args:
patience (int): How long to wait after last time validation loss improved.
Default: 50
verbose (bool): If True, prints a message for each validation loss improvement.
Default: False
"""
self.patience = patience
self.verbose = verbose
self.counter = 0
self.best_score = None
self.best_tensor = var_tensor
self.early_stop = False
self.opt_loss_min = np.Inf
def __call__(self, opt_loss, var_tensor):
score = -opt_loss
if self.best_score is None:
self.best_score = score
self.save(opt_loss, var_tensor)
elif score < self.best_score:
self.counter += 1
print(f'EarlyStopping counter: {self.counter} out of {self.patience}.')
if self.counter >= self.patience:
self.early_stop = True
else:
self.best_score = score
self.save(opt_loss, var_tensor)
self.counter = 0
def save(self, opt_loss, var_tensor):
self.best_tensor = var_tensor
if self.verbose:
print(f'Loss decreased ({self.opt_loss_min:.6f} --> {opt_loss:.6f}).')
self.opt_loss_min = opt_loss
def visualization(ori_data, generated_data, analysis):
"""Using PCA or tSNE for generated and original data visualization.
Args:
- ori_data: original data
- generated_data: generated synthetic data
- analysis: tsne or pca
"""
# Analysis sample size (for faster computation)
anal_sample_no = min([5000, ori_data.shape[0]])
idx = np.random.permutation(ori_data.shape[0])[:anal_sample_no]
prep_data = ori_data[idx]
prep_data_hat = generated_data[idx]
# Visualization parameter
colors = ["red" for i in range(anal_sample_no)] + ["blue" for i in range(anal_sample_no)]
if analysis == 'pca':
# PCA Analysis
pca = PCA(n_components=2)
pca.fit(prep_data)
pca_results = pca.transform(prep_data)
pca_hat_results = pca.transform(prep_data_hat)
# Plotting
f, ax = plt.subplots(1)
plt.scatter(pca_results[:, 0], pca_results[:, 1],
c=colors[:anal_sample_no], alpha=0.2, label="Original")
plt.scatter(pca_hat_results[:, 0], pca_hat_results[:, 1],
c=colors[anal_sample_no:], alpha=0.2, label="Synthetic")
ax.legend()
plt.title('PCA plot')
plt.xlabel('x-pca')
plt.ylabel('y_pca')
plt.show()
elif analysis == 'tsne':
# Do t-SNE Analysis together
prep_data_final = np.concatenate((prep_data, prep_data_hat), axis=0)
# TSNE anlaysis
tsne = TSNE(n_components=2, verbose=1, perplexity=40, n_iter=300)
tsne_results = tsne.fit_transform(prep_data_final)
# Plotting
f, ax = plt.subplots(1)
plt.scatter(tsne_results[:anal_sample_no, 0], tsne_results[:anal_sample_no, 1],
c=colors[:anal_sample_no], alpha=0.2, label="Original")
plt.scatter(tsne_results[anal_sample_no:, 0], tsne_results[anal_sample_no:, 1],
c=colors[anal_sample_no:], alpha=0.2, label="Synthetic")
ax.legend()
plt.title('t-SNE plot')
plt.xlabel('x-tsne')
plt.ylabel('y_tsne')
plt.show()