-
Notifications
You must be signed in to change notification settings - Fork 5
/
distill.py
375 lines (322 loc) · 20.5 KB
/
distill.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
371
372
373
374
375
import os
import time
import copy
import argparse
import numpy as np
import torch
import torch.nn as nn
from torchvision.utils import save_image
from utils import get_loops, get_dataset, get_network, get_eval_pool, evaluate_synset, get_daparam, \
match_loss, get_time, TensorDataset, epoch, DiffAugment, ParamDiffAug
import os
import logging
import random
import torch.nn as nn
def build_logger(work_dir, cfgname):
assert cfgname is not None
log_file = cfgname + '.log'
log_path = os.path.join(work_dir, log_file)
logger = logging.getLogger(cfgname)
logger.setLevel(logging.INFO)
# formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s')
formatter = logging.Formatter('%(asctime)s: %(message)s', datefmt='%Y-%m-%d %H:%M:%S')
handler1 = logging.FileHandler(log_path)
handler1.setFormatter(formatter)
logger.addHandler(handler1)
handler2 = logging.StreamHandler()
handler2.setFormatter(formatter)
logger.addHandler(handler2)
logger.propagate = False
return logger
def adjust_learning_rate(optimizer, epoch, init_lr):
"""Decay the learning rate based on schedule"""
lr = init_lr
for milestone in [1200, 1600, 1800]:
lr *= 0.5 if epoch >= milestone else 1.
for param_group in optimizer.param_groups:
param_group['lr'] = lr
def criterion_middle(real_feature, syn_feature):
MSE_Loss = nn.MSELoss(reduction='sum')
shape_real = real_feature.shape
real_feature = torch.mean(real_feature.view(10, shape_real[0] // 10, *shape_real[1:]), dim=1)
shape_syn = syn_feature.shape
syn_feature = torch.mean(syn_feature.view(10, shape_syn[0] // 10, *shape_syn[1:]), dim=1)
return MSE_Loss(real_feature, syn_feature)
def main():
parser = argparse.ArgumentParser(description='Parameter Processing')
parser.add_argument('--method', type=str, default='DC', help='DC/DSA')
parser.add_argument('--dataset', type=str, default='CIFAR10', help='dataset')
parser.add_argument('--model', type=str, default='ConvNet', help='model')
parser.add_argument('--ipc', type=int, default=1, help='image(s) per class')
parser.add_argument('--eval_mode', type=str, default='S',
help='eval_mode') # S: the same to training model, M: multi architectures, W: net width, D: net depth, A: activation function, P: pooling layer, N: normalization layer,
parser.add_argument('--num_exp', type=int, default=1, help='the number of experiments')
parser.add_argument('--num_eval', type=int, default=20, help='the number of evaluating randomly initialized models')
parser.add_argument('--epoch_eval_train', type=int, default=100, help='epochs to train a model with synthetic data')
parser.add_argument('--Iteration', type=int, default=2000, help='training iterations')
parser.add_argument('--lr_img', type=float, default=0.1, help='learning rate for updating synthetic images')
parser.add_argument('--lr_net', type=float, default=0.01, help='learning rate for updating network parameters')
parser.add_argument('--batch_real', type=int, default=256, help='batch size for real data')
parser.add_argument('--batch_train', type=int, default=256, help='batch size for training networks')
parser.add_argument('--init', type=str, default='noise',
help='noise/real: initialize synthetic images from random noise or randomly sampled real images.')
parser.add_argument('--dsa_strategy', type=str, default='None', help='differentiable Siamese augmentation strategy')
parser.add_argument('--data_path', type=str, default='data', help='dataset path')
parser.add_argument('--save_path', type=str, default='oi_cifar10_ipc10_watcher_5_v3', help='path to save results')
parser.add_argument('--dis_metric', type=str, default='ours', help='distance metric')
parser.add_argument('--fourth_weight', type=float, default=0.1, help='batch size for training networks')
parser.add_argument('--third_weight', type=float, default=0.1, help='batch size for training networks')
parser.add_argument('--second_weight', type=float, default=1.0, help='batch size for training networks')
parser.add_argument('--first_weight', type=float, default=1.0, help='batch size for training networks')
parser.add_argument('--inner_weight', type=float, default=0.01, help='batch size for training networks')
parser.add_argument('--lambda_1', type=float, default=0.04, help='break outlooper')
parser.add_argument('--lambda_2', type=float, default=0.03, help='break innerlooper')
parser.add_argument('--gpu_id', type=str, default='0', help='dataset path')
args = parser.parse_args()
logger = build_logger('.', cfgname=str(args.lambda_1) + "_" + str(args.lambda_2) + "_" + str(
args.inner_weight) + '_' + str(args.fourth_weight) + '_' + str(args.third_weight) + '_' + str(
args.second_weight) + '_' + str(args.first_weight) + 'oi_cifar10_dsa_ipc50_watcher_5_v3')
os.environ["CUDA_VISIBLE_DEVICES"] = args.gpu_id
args.outer_loop, args.inner_loop = get_loops(args.ipc)
# import pdb; pdb.set_trace()
args.save_path = str(args.lambda_1) + "_" + str(args.lambda_2) + "_" + 'oi_cifar10_ipc10_watcher_5_v3'
args.device = 'cuda' if torch.cuda.is_available() else 'cpu'
args.dsa_param = ParamDiffAug()
args.dsa = True if args.method == 'DSA' else False
if not os.path.exists(args.data_path):
os.mkdir(args.data_path)
if not os.path.exists(args.save_path):
os.mkdir(args.save_path)
eval_it_pool = np.arange(0, args.Iteration + 1, 100).tolist() if args.eval_mode == 'S' else [
args.Iteration] # The list of iterations when we evaluate models and record results.
channel, im_size, num_classes, class_names, mean, std, dst_train, dst_test, testloader = get_dataset(args.dataset,
args.data_path)
model_eval_pool = get_eval_pool(args.eval_mode, args.model, args.model)
accs_all_exps = dict() # record performances of all experiments
for key in model_eval_pool:
accs_all_exps[key] = []
data_save = []
for exp in range(args.num_exp):
logger.info('================== Exp %d ==================' % exp)
logger.info('Hyper-parameters: \n', args.__dict__)
print('Evaluation model pool: ', model_eval_pool)
''' organize the real dataset '''
images_all = []
labels_all = []
indices_class = [[] for c in range(num_classes)]
images_all = [torch.unsqueeze(dst_train[i][0], dim=0) for i in range(len(dst_train))]
labels_all = [dst_train[i][1] for i in range(len(dst_train))]
for i, lab in enumerate(labels_all):
indices_class[lab].append(i)
images_all = torch.cat(images_all, dim=0)
labels_all = torch.tensor(labels_all, dtype=torch.long, device=args.device)
for c in range(num_classes):
logger.info('class c = %d: %d real images' % (c, len(indices_class[c])))
def get_images(c, n): # get random n images from class c
# import pdb; pdb.set_trace()
idx_shuffle = np.random.permutation(indices_class[c])[:n]
return images_all[idx_shuffle].to(args.device)
for ch in range(channel):
logger.info('real images channel %d, mean = %.4f, std = %.4f' % (
ch, torch.mean(images_all[:, ch]), torch.std(images_all[:, ch])))
''' initialize the synthetic data '''
image_syn = torch.randn(size=(num_classes * args.ipc, channel, im_size[0], im_size[1]), dtype=torch.float,
requires_grad=True, device=args.device)
label_syn = torch.tensor([np.ones(args.ipc) * i for i in range(num_classes)], dtype=torch.int,
requires_grad=False, device=args.device).view(-1) # [0,0,0, 1,1,1, ..., 9,9,9]
if args.init == 'real':
logger.info('initialize synthetic data from random real images')
for c in range(num_classes):
image_syn.data[c * args.ipc:(c + 1) * args.ipc] = get_images(c, args.ipc).detach().data
else:
logger.info('initialize synthetic data from random noise')
''' training '''
optimizer_img = torch.optim.SGD([image_syn, ], lr=args.lr_img, momentum=0.5) # optimizer_img for synthetic data
optimizer_img.zero_grad()
criterion = nn.CrossEntropyLoss().to(args.device)
criterion_sum = nn.CrossEntropyLoss(reduction='sum').to(args.device)
logger.info('%s training begins' % get_time())
for it in range(args.Iteration + 1):
adjust_learning_rate(optimizer_img, it, args.lr_img)
''' Evaluate synthetic data '''
if it in eval_it_pool:
for model_eval in model_eval_pool:
logger.info(
'-------------------------\nEvaluation\nmodel_train = %s, model_eval = %s, iteration = %d' % (
args.model, model_eval, it))
if args.dsa:
args.epoch_eval_train = 1000
args.dc_aug_param = None
logger.info('DSA augmentation strategy: \n' + args.dsa_strategy)
logger.info('DSA augmentation parameters: \n' + str(args.dsa_param.__dict__))
else:
# This augmentation parameter set is only for DC method. It will be muted when args.dsa is True.
args.dc_aug_param = get_daparam(args.dataset, args.model, model_eval,
args.ipc)
logger.info('DC augmentation parameters: \n' + str(args.dc_aug_param))
if args.dsa or args.dc_aug_param['strategy'] != 'none':
args.epoch_eval_train = 1000 # Training with data augmentation needs more epochs.
else:
args.epoch_eval_train = 600
accs = []
for it_eval in range(args.num_eval):
# get a random model
net_eval = get_network(model_eval, channel, num_classes, im_size).to(
args.device)
# avoid any unaware modification
image_syn_eval, label_syn_eval = copy.deepcopy(image_syn.detach()), copy.deepcopy(
label_syn.detach())
_, acc_train, acc_test = evaluate_synset(it_eval, net_eval, image_syn_eval, label_syn_eval,
testloader, args)
accs.append(acc_test)
logger.info('Evaluate %d random %s, mean = %.4f std = %.4f\n-------------------------' % (
len(accs), model_eval, np.mean(accs), np.std(accs)))
# record the final results
if it == args.Iteration:
accs_all_exps[model_eval] += accs
''' visualize and save '''
save_name = os.path.join(args.save_path, 'vis_%s_%s_%s_%dipc_exp%d_iter%d.png' % (
args.method, args.dataset, args.model, args.ipc, exp, it))
image_syn_vis = copy.deepcopy(image_syn.detach().cpu())
for ch in range(channel):
image_syn_vis[:, ch] = image_syn_vis[:, ch] * std[ch] + mean[ch]
image_syn_vis[image_syn_vis < 0] = 0.0
image_syn_vis[image_syn_vis > 1] = 1.0
# Trying normalize = True/False may get better visual effects.
save_image(image_syn_vis, save_name,
nrow=args.ipc)
''' Train synthetic data '''
# get a random model
net = get_network(args.model, channel, num_classes, im_size).to(args.device)
net.train()
net_parameters = list(net.parameters())
optimizer_net = torch.optim.SGD(net.parameters(), lr=args.lr_net) # optimizer_img for synthetic data
optimizer_net.zero_grad()
loss_avg = 0
loss_kai = 0
loss_middle_item = 0
args.dc_aug_param = None # Mute the DC augmentation when training synthetic data.
# for ol in range(args.outer_loop):
acc_watcher = list()
pop_cnt = 0
acc_test = 0.0
while True:
syn_centers = []
real_feature_concat = []
real_feature_concat_mm = []
real_label_concat = []
img_real_gather = []
img_syn_gather = []
lab_real_gather = []
lab_syn_gather = []
loss = torch.tensor(0.0).to(args.device)
for c in range(num_classes):
img_real = get_images(c, args.batch_real)
lab_real = torch.ones((img_real.shape[0],), device=args.device, dtype=torch.long) * c
img_syn = image_syn[c * args.ipc:(c + 1) * args.ipc].reshape(
(args.ipc, channel, im_size[0], im_size[1]))
lab_syn = torch.ones((args.ipc,), device=args.device, dtype=torch.long) * c
if args.dsa:
seed = int(time.time() * 1000) % 100000
img_real = DiffAugment(img_real, args.dsa_strategy, seed=seed, param=args.dsa_param)
img_syn = DiffAugment(img_syn, args.dsa_strategy, seed=seed, param=args.dsa_param)
img_real_gather.append(img_real)
lab_real_gather.append(lab_real)
img_syn_gather.append(img_syn)
lab_syn_gather.append(lab_syn)
img_real_gather = torch.stack(img_real_gather, dim=0).reshape(args.batch_real * 10, 3, 32, 32)
img_syn_gather = torch.stack(img_syn_gather, dim=0).reshape(args.ipc * 10, 3, 32, 32)
lab_real_gather = torch.stack(lab_real_gather, dim=0).reshape(args.batch_real * 10)
lab_syn_gather = torch.stack(lab_syn_gather, dim=0).reshape(args.ipc * 10)
####forward#####
output_real, real_features = net(
img_real_gather)
output_syn, syn_features = net(
img_syn_gather)
loss_middle = args.fourth_weight * criterion_middle(real_features[-1], syn_features[-1]) + args.third_weight * criterion_middle(real_features[-2], syn_features[-2]) + args.second_weight * criterion_middle(real_features[-3], syn_features[-3]) + args.first_weight * criterion_middle(real_features[-4], syn_features[-4])
loss_real = criterion(output_real, lab_real_gather)
loss += loss_middle
loss += loss_real
last_real_feature = torch.mean(real_features[0].view(10, int(real_features[0].shape[0] / num_classes), real_features[0].shape[1]), dim=1)
last_syn_feature = torch.mean(syn_features[0].view(10, int(syn_features[0].shape[0] / num_classes), syn_features[0].shape[1]), dim=1)
output = torch.mm(real_features[0], last_syn_feature.t())
last_real_feature = torch.mean(
last_real_feature.unsqueeze(0).reshape(10, int(last_real_feature.shape[0] / num_classes),
last_real_feature.shape[1]), dim=1)
loss_output = criterion_middle(last_syn_feature, last_real_feature) + args.inner_weight * criterion_sum(output, lab_real_gather)
loss += loss_output
loss.backward()
optimizer_img.step()
optimizer_img.zero_grad()
loss_avg += loss.item()
loss_kai += loss_output.item()
loss_middle_item += loss_middle.item()
############ for outloop testing ############
for c in range(num_classes):
img_real_test = get_images(c, 128)
lab_real_test = torch.ones((img_real_test.shape[0],), device=args.device, dtype=torch.long) * c
prob, _ = net(img_real_test)
acc_test += (lab_real_test == prob.max(dim=1)[1]).float().mean()
acc_test /= num_classes
acc_watcher.append(acc_test.detach().cpu())
pop_cnt += 1
if len(acc_watcher) == 10:
if max(acc_watcher) - min(acc_watcher) < args.lambda_1:
acc_watcher = list()
pop_cnt = 0
acc_test = 0.0
break
else:
acc_watcher.pop(0)
''' update network '''
image_syn_train, label_syn_train = copy.deepcopy(image_syn.detach()), copy.deepcopy(
label_syn.detach()) # avoid any unaware modification
dst_syn_train = TensorDataset(image_syn_train, label_syn_train)
trainloader = torch.utils.data.DataLoader(dst_syn_train, batch_size=args.batch_train, shuffle=True,
num_workers=0)
acc_inner_watcher = list()
acc_syn_inner_watcher = list()
pop_inner_cnt = 0
acc_inner_test = 0
# for il in range(args.inner_loop):
while (1):
inner_loss, inner_acc = epoch('train', trainloader, net, optimizer_net, criterion, args,
aug=True if args.dsa else False)
acc_syn_inner_watcher.append(inner_acc)
for c in range(num_classes):
img_real_test = get_images(c, 128)
lab_real_test = torch.ones((img_real_test.shape[0],), device=args.device, dtype=torch.long) * c
prob, _ = net(img_real_test)
acc_inner_test += (lab_real_test == prob.max(dim=1)[1]).float().mean()
acc_inner_test /= num_classes
acc_inner_watcher.append(acc_inner_test.detach().cpu())
pop_inner_cnt += 1
if len(acc_inner_watcher) == 10:
if max(acc_inner_watcher) - min(acc_inner_watcher) > args.lambda_2:
acc_inner_watcher = list()
acc_syn_inner_watcher = list()
pop_inner_cnt = 0
acc_inner_test = 0
break
else:
acc_inner_watcher.pop(0)
epoch('test', trainloader, net, optimizer_net, criterion, args, aug=True if args.dsa else False)
loss_avg /= (num_classes * args.outer_loop)
if it % 10 == 0:
logger.info('%s iter = %04d, loss = %.4f, loss_kai = %.4f, loss_middle = %.4f' % (
get_time(), it, loss_avg, loss_kai, loss_middle_item))
if it == args.Iteration: # only record the final results
data_save.append([copy.deepcopy(image_syn.detach().cpu()), copy.deepcopy(label_syn.detach().cpu())])
torch.save({'data': data_save, 'accs_all_exps': accs_all_exps, }, os.path.join(args.save_path,
'res_%s_%s_%s_%dipc.pt' % (
args.method,
args.dataset,
args.model,
args.ipc)))
logger.info('\n==================== Final Results ====================\n')
for key in model_eval_pool:
accs = accs_all_exps[key]
logger.info('Run %d experiments, train on %s, evaluate %d random %s, mean = %.2f%% std = %.2f%%' % (
args.num_exp, args.model, len(accs), key, np.mean(accs) * 100, np.std(accs) * 100))
if __name__ == '__main__':
main()