-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathutil.py
179 lines (156 loc) · 5.95 KB
/
util.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
166
167
168
169
170
171
172
173
174
175
176
177
178
179
import torch.nn as nn
from models.ae import Linear01
import numpy
import os
import sys
import time
import math
class BinOp():
def __init__(self, model):
# count the number of Conv2d and Linear
count_targets = 0
for m in model.modules():
if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear) or isinstance(m, Linear01):
count_targets = count_targets + 1
# start_range = 1
# end_range = count_targets-2
# self.bin_range = numpy.linspace(start_range,
# end_range, end_range-start_range+1)\
# .astype('int').tolist()
self.num_of_params = count_targets
self.saved_params = []
self.target_modules = []
# index = -1
for m in model.modules():
if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear) or isinstance(m, Linear01):
# index = index + 1
# if index in self.bin_range:
tmp = m.weight.data.clone()
self.saved_params.append(tmp)
self.target_modules.append(m.weight)
return
def binarization(self):
self.meancenterConvParams()
self.clampConvParams()
self.save_params()
self.binarizeConvParams()
def meancenterConvParams(self):
for index in range(self.num_of_params):
s = self.target_modules[index].data.size()
negMean = self.target_modules[index].data.mean(1, keepdim=True).\
mul(-1).expand_as(self.target_modules[index].data)
self.target_modules[index].data = self.target_modules[index].data.add(negMean)
def clampConvParams(self):
for index in range(self.num_of_params):
self.target_modules[index].data = \
self.target_modules[index].data.clamp(-1.0, 1.0)
def save_params(self):
for index in range(self.num_of_params):
self.saved_params[index].copy_(self.target_modules[index].data)
def binarizeConvParams(self):
for index in range(self.num_of_params):
# n = self.target_modules[index].data[0].nelement()
# s = self.target_modules[index].data.size()
# if len(s) == 4:
# m = self.target_modules[index].data.norm(1, 3, keepdim=True)\
# .sum(2, keepdim=True).sum(1, keepdim=True).div(n)
# elif len(s) == 2:
# m = self.target_modules[index].data.norm(1, 1, keepdim=True).div(n)
# self.target_modules[index].data = \
# self.target_modules[index].data.sign().mul(m.expand(s))
self.target_modules[index].data = self.target_modules[index].data.sign()
def restore(self):
for index in range(self.num_of_params):
self.target_modules[index].data.copy_(self.saved_params[index])
def updateBinaryGradWeight(self):
for index in range(self.num_of_params):
weight = self.target_modules[index].data
n = weight[0].nelement()
s = weight.size()
if len(s) == 4:
m = weight.norm(1, 3, keepdim=True)\
.sum(2, keepdim=True).sum(1, keepdim=True).div(n).expand(s)
elif len(s) == 2:
m = weight.norm(1, 1, keepdim=True).div(n).expand(s)
m[weight.lt(-1.0)] = 0
m[weight.gt(1.0)] = 0
m = m.mul(self.target_modules[index].grad.data)
m_add = weight.sign().mul(self.target_modules[index].grad.data)
if len(s) == 4:
m_add = m_add.sum(3, keepdim=True)\
.sum(2, keepdim=True).sum(1, keepdim=True).div(n).expand(s)
elif len(s) == 2:
m_add = m_add.sum(1, keepdim=True).div(n).expand(s)
m_add = m_add.mul(weight.sign())
self.target_modules[index].grad.data = m.add(m_add).mul(1.0-1.0/s[1]).mul(n)
# _, term_width = os.popen('stty size', 'r').read().split()
# term_width = int(term_width)
term_width = 202
TOTAL_BAR_LENGTH = 65.
last_time = time.time()
begin_time = last_time
def progress_bar(current, total, msg=None):
global last_time, begin_time
if current == 0:
begin_time = time.time() # Reset for new bar.
cur_len = int(TOTAL_BAR_LENGTH*current/total)
rest_len = int(TOTAL_BAR_LENGTH - cur_len) - 1
sys.stdout.write(' [')
for i in range(cur_len):
sys.stdout.write('=')
sys.stdout.write('>')
for i in range(rest_len):
sys.stdout.write('.')
sys.stdout.write(']')
cur_time = time.time()
step_time = cur_time - last_time
last_time = cur_time
tot_time = cur_time - begin_time
L = []
L.append(' Step: %s' % format_time(step_time))
L.append(' | Tot: %s' % format_time(tot_time))
if msg:
L.append(' | ' + msg)
msg = ''.join(L)
sys.stdout.write(msg)
for i in range(term_width-int(TOTAL_BAR_LENGTH)-len(msg)-3):
sys.stdout.write(' ')
# Go back to the center of the bar.
for i in range(term_width-int(TOTAL_BAR_LENGTH/2)+2):
sys.stdout.write('\b')
sys.stdout.write(' %d/%d ' % (current+1, total))
if current < total-1:
sys.stdout.write('\r')
else:
sys.stdout.write('\n')
sys.stdout.flush()
def format_time(seconds):
days = int(seconds / 3600/24)
seconds = seconds - days*3600*24
hours = int(seconds / 3600)
seconds = seconds - hours*3600
minutes = int(seconds / 60)
seconds = seconds - minutes*60
secondsf = int(seconds)
seconds = seconds - secondsf
millis = int(seconds*1000)
f = ''
i = 1
if days > 0:
f += str(days) + 'D'
i += 1
if hours > 0 and i <= 2:
f += str(hours) + 'h'
i += 1
if minutes > 0 and i <= 2:
f += str(minutes) + 'm'
i += 1
if secondsf > 0 and i <= 2:
f += str(secondsf) + 's'
i += 1
if millis > 0 and i <= 2:
f += str(millis) + 'ms'
i += 1
if f == '':
f = '0ms'
return f