-
Notifications
You must be signed in to change notification settings - Fork 3
/
temp.py
370 lines (328 loc) · 13.4 KB
/
temp.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
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
# 代码测试的代码,没必要看
import torch
from torch import nn
import torch.nn.functional as F
from torch.autograd import Variable
import torch
import time, threading, sys, socket, json
from functions import my_functions
from model.ResNet import get_ResNet_model, get_inference_model
from functions.my_functions import Train_model, Eval_model
from torchsummary import summary
from functions.my_functions import Knowledge_distillation
from functions.branch_functions import Train_BranchyNet_Asynchronous, Train_BranchyNet_Synchronization, Eval_BranchyNet, \
get_Exit_Threshold, Train_BranchyNet_Asynchronous_KD, Train_BranchyNet_Asynchronous_KD_Back, Train_BranchyNet_Asynchronous_Back, \
get_threshold, get_voting_softmax
from datasets import get_cifar_100
# 复制网络层参数
def copy_layer_param(old_model, new_model, layer_count):
old_state_dict = old_model.state_dict()
new_state_dict = new_model.state_dict()
# 遍历赋值
for i ,data in enumerate(old_model.named_parameters()):
# for i ,data in enumerate(new_model.named_parameters()):
name, param = data
if (i >= layer_count[0] and i <= layer_count[1]):
new_state_dict[name] = old_state_dict[name]
new_model.load_state_dict(new_state_dict)
# def Eval_Branch_model(exit_threshold = [0.95,0.97,0.98], voting_weight = [0.30, 0.30, 0.40]):
# Branch_model性能测试
def Eval_Branch_model():
exit_threshold = [0.96, 0.00, 0.98]
voting_weight = [0.30, 0.30, 0.40]
test_num = 9990
model_num = 3
softmax = nn.Softmax(dim=1)
Project_dir = my_functions.get_project_dir()
main_model, branch_model = get_ResNet_model.get_model(num_classes=100)
model_0, model_1, model_2, model_3, model_4 = get_inference_model.get_model(num_classes=100)
for i in range(model_num):
# print (i)
dir = Project_dir + "/model/ResNet/KD/ResNet_As_model_" + str(i) + "_checkpoint.tar"
checkpoint = torch.load(dir, map_location=lambda storage, loc: storage)
branch_model[i].load_state_dict(checkpoint["state_dict"])
if i == 0:
model_0.load_state_dict(checkpoint["state_dict"])
model_1.load_state_dict(checkpoint["state_dict"])
elif i == 1:
model_2.load_state_dict(checkpoint["state_dict"])
model_3.load_state_dict(checkpoint["state_dict"])
else:
model_4.load_state_dict(checkpoint["state_dict"])
# 初始化
# copy_layer_param(branch_model[0], model_0, [0, 29])
# copy_layer_param(branch_model[0], model_1, [30, 48])
# copy_layer_param(branch_model[1], model_2, [30, 44])
# copy_layer_param(branch_model[1], model_3, [45, 57])
# copy_layer_param(branch_model[2], model_4, [45, 61])
model_0.eval()
model_1.eval()
model_2.eval()
model_3.eval()
model_4.eval()
# 将model设置为eval模式,防止模型参数变化
for i in range(model_num):
branch_model[i].eval()
if torch.cuda.is_available():
model_0 = model_0.cuda()
model_1 = model_1.cuda()
model_2 = model_2.cuda()
model_3 = model_3.cuda()
model_4 = model_4.cuda()
branch_model[0] = branch_model[0].cuda()
branch_model[1] = branch_model[1].cuda()
branch_model[2] = branch_model[2].cuda()
model_list = []
model_list.append(model_0)
model_list.append(model_1)
model_list.append(model_2)
model_list.append(model_3)
model_list.append(model_4)
train_loader, test_loader = get_cifar_100.get_data(train_batch_size=128, test_batch_size=1)
# 测试完整网络
# for i in range(model_num):
# print ("Model: ", i)
# start_time = time.time()
# for num, data in enumerate(test_loader, 0):
# if num < test_num:
# inputs, labels = data
# inputs, labels = Variable(inputs), Variable(labels)
# out = branch_model[i](inputs)
# else:
# break
# end_time = time.time()
# print("time : %2.2d s" % (end_time - start_time))
# 分段测试
# 中间运行记录
total = 0
run_time = []
exit_count = []
pre_right_count = []
voting_count = 0
voting_time = 0.0
voting_right_count = 0
# 初始化
for i in range(model_num):
run_time.append(0.0)
exit_count.append(0)
pre_right_count.append(0)
temp_start_time = time.time()
# 开始测试BranchyNet
for key, data in enumerate(test_loader):
# if key == test_num:
# break
# 阈值和中间输出值
Threshold = []
output_list = []
output_temp = []
for i in range(model_num):
Threshold.append(0.0)
output_list.append(0.0)
output_temp.append(0.0)
# start_time = time.time()
inputs, labels = data
# 使用cuda,并且有cuda资源
if torch.cuda.is_available():
inputs = inputs.cuda()
labels = labels.cuda()
# a = softmax(branch_model[0](inputs))
# temp = torch.max(a.data, 1)
# print(temp[1], a)
#
# bb = model_0(inputs)
# print (bb)
# b = softmax(model_1(bb))
# temp = torch.max(b.data, 1)
# print(temp[1], b)
# 计数
total += labels.size(0)
# 按照退出点依次计算
start_time = time.time()
for i in range(model_num):
# 计算中间结果和退出点输出
output_temp[i] = model_list[2*i](inputs)
if i < model_num-1:
output = model_list[2*i + 1](output_temp[i])
else:
# 最后一个出口,直接计算
output = output_temp[i]
output_list[i] = output
inputs = output_temp[i]
# 计算阈值
Threshold[i] = get_threshold(output)
# 退出条件判断
if Threshold[i] > exit_threshold[i]:
# 满足退出条件
end_time = time.time()
exit_count[i] += 1
softmax = nn.Softmax(dim=1)
# _, predicted = torch.max(softmax(output.data), 1)
_, predicted = torch.max(output.data, 1)
pre_right_count[i] += (predicted == labels).sum().item()
run_time[i] += (end_time - start_time)
break
elif i == (model_num - 1):
# 如果到主网络依然不满足退出条件
# 需要多个网络投票,决出最终结果,新的决策方式
# print ("Voting Network")
end_time = time.time()
voting_count += 1
# 此函数可能存在问题,需后期调试##########################################
voting_softmax = get_voting_softmax(output_list, voting_weight)
_, predicted = torch.max(voting_softmax.data, 1)
# if torch.equal(labels, predicted):
# # print ("相等")
# voting_right_count += 1
voting_right_count += (predicted == labels).sum().item()
# other_prediction_right += (predicted == labels).sum().item()
voting_time += (end_time - start_time)
break
else:
continue
temp_end_time = time.time()
# 打印测试信息
print ('总共测试 %d条数据' % total)
for i in range(model_num):
if exit_count[i]==0:
print ("Exit point: %d,Count: %d" % (i, exit_count[i]))
else:
print("Exit point: %d,Count: %d, Pre_acc: %2.2f" % ( i, exit_count[i], pre_right_count[i] / exit_count[i] * 100), "%",
" Threshold: %.3f, one_used_time:%.3fms, Exit Percentage: %2.2f" % (exit_threshold[i], run_time[i] / exit_count[i] * 1000, exit_count[i] / total * 100), "%")
if voting_count==0:
print ("Voting count: %d," % voting_count)
else:
print ("Voting count: %d, Pre_acc: %2.2f" % (voting_count, voting_right_count / voting_count * 100), "%","one_used_time:%.3fms, Percentage: %2.2f" % (voting_time / voting_count * 1000, voting_count / total * 100), "%")
total_acc = (sum(pre_right_count) + voting_right_count)/total*100
one_used_time = (sum(run_time) + voting_time)/total*1000
print ("Total Pre_acc:%2.2f" % (total_acc), "%", "one_used_time:%.3fms" % (one_used_time))
total_time = (temp_end_time - temp_start_time)/total*1000
print("Total Pre_acc:%2.2f" % (total_acc), "%", "one_used_time:%.3fms" % (total_time) )
print ("Eval BranchyNet 结束")
def test():
model_1, model_2, model_3, model_4, model_5 = get_inference_model.get_model(num_classes=100)
# for i, data in enumerate(model_5.named_parameters()):
# name, param = data
# print (i, name, param.size())
model_num = 3
Project_dir = my_functions.get_project_dir()
main_model, branch_model = get_ResNet_model.get_model(num_classes=100)
for i in range(model_num):
# print (i)
dir = Project_dir + "/model/ResNet/KD/ResNet_As_model_" + str(i) + "_checkpoint.tar"
checkpoint = torch.load(dir, map_location=lambda storage, loc: storage)
branch_model[i].load_state_dict(checkpoint["state_dict"])
# 初始化
copy_layer_param(branch_model[0], model_1, [0, 29])
copy_layer_param(branch_model[0], model_2, [30, 48])
copy_layer_param(branch_model[1], model_3, [30, 44])
copy_layer_param(branch_model[1], model_4, [45, 57])
copy_layer_param(branch_model[2], model_5, [45, 61])
train_loader, test_loader = get_cifar_100.get_data(train_batch_size = 128, test_batch_size = 1)
num = 1
start_time = time.time()
model_1.eval()
model_2.eval()
for i in range(model_num):
branch_model[i].eval()
# for i, data in enumerate(model_2.named_parameters()):
# name, param = data
# print (i, name, param)
# for i, data in enumerate(branch_model[0].named_parameters()):
# name, param = data
# print (i, name, param)
for i, data in enumerate(test_loader, 0):
if i < num:
inputs, labels = data
inputs, labels = Variable(inputs), Variable(labels)
out = model_1(inputs)
# out = model_2(out)
out = model_3(out)
# print (out.size())
print (out)
out = model_4(out)
# out = model_5(out)
print(out)
new_out = branch_model[1](inputs)
print(new_out)
end_time = time.time()
print("time : %2.2d s" % (end_time - start_time))
start_time = time.time()
for i, data in enumerate(test_loader, 0):
if i < num:
inputs, labels = data
inputs, labels = Variable(inputs), Variable(labels)
out = branch_model[1](inputs)
# print (out)
end_time = time.time()
print("time : %2.2d s" % (end_time - start_time))
if __name__ == "__main__":
Eval_Branch_model()
# 创建网络连接
# sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
# sock.connect(('192.168.10.128', 5000))
# 直接调用socket通信
# model_1_out = 0
# model_3_out = 0
# col_com()
# 结束
# msg = "bye"
# sock.send(str.encode(msg))
# sock.close()
# main_model, branch_model = get_ResNet_model.get_model(num_classes=100)
# model = branch_model[2]
# model_0 = ResNet_0(BasicBlock, [2,2,2,2],num_classes=100)
# model_1 = ResNet_1(BasicBlock, [2,2,2,2],num_classes=100)
#
# dir = my_functions.get_project_dir() + "/model/ResNet/KD/ResNet_As_model_2_checkpoint.tar"
# checkpoint = torch.load(dir, map_location=lambda storage, loc: storage)
# model.load_state_dict(checkpoint["state_dict"])
#
# copy_layer_param(model, model_0, [0, 44])
# # print (model.state_dict())
# # print (model_0.state_dict())
# copy_layer_param(model, model_1, [45, 61])
# # print (model_1.state_dict())
#
# # print (model_0)
# # summary(model_0, input_size=(3, 32, 32))
# # summary(model_1, input_size=(256, 8, 8))
# train_loader, test_loader = get_cifar_100.get_data(train_batch_size=128, test_batch_size=1)
#
# num =100
# start_time = time.time()
# for i, data in enumerate(test_loader, 0):
# if i < num:
# # print (i)
# inputs, labels = data
# inputs, labels = Variable(inputs), Variable(labels)
#
# middle_output = model_0(inputs)
# divide_output = model_1(middle_output)
# # print(divide_output)
# else:
# break
# end_time = time.time()
# print("time : %.2d s" % (end_time - start_time) )
#
# start_time = time.time()
# for i, data in enumerate(test_loader, 0):
# if i < num:
# # print (i)
# inputs, labels = data
# inputs, labels = Variable(inputs), Variable(labels)
#
# output = model(inputs)
# # print("time : ", end_time - start_time)
# # print (output)
# else:
# break
# end_time = time.time()
# print("time : %.2d s" % (end_time - start_time))
# train_loader, test_loader = get_cifar_100.get_data(train_batch_size = 128, test_batch_size = 1)
# main_model, branch_model = get_ResNet_model.get_model(num_classes=100)
# model = branch_model[2]
# my_model = nn.Sequential(*list(model.children())[1:5])
# print (my_model)
# for idx, m in enumerate(model.modules()):
# key ,value = m
# print(idx, '->', key)