-
Notifications
You must be signed in to change notification settings - Fork 25
/
train.py
313 lines (243 loc) · 15.4 KB
/
train.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
import numpy as np
import sys
from sklearn.preprocessing import StandardScaler
import argparse
sys.path.append('./')
import torch
from torch.utils.data import Subset, TensorDataset, ConcatDataset
from torchvision import datasets, transforms
from distil.utils.models.resnet import ResNet18
from distil.active_learning_strategies import GLISTER, BADGE, EntropySampling, RandomSampling, LeastConfidenceSampling, \
MarginSampling, CoreSet, AdversarialBIM, AdversarialDeepFool, KMeansSampling, \
BALDDropout, FASS
from distil.utils.models.simple_net import TwoLayerNet
from distil.utils.train_helper import data_train
from distil.utils.config_helper import read_config_file
from distil.utils.utils import LabeledToUnlabeledDataset
import time
import pickle
class TrainClassifier:
def __init__(self, config_file):
self.config_file = config_file
self.config = read_config_file(config_file)
def getModel(self, model_config):
if model_config['architecture'] == 'resnet18':
if ('target_classes' in model_config) and ('channel' in model_config):
net = ResNet18(num_classes = model_config['target_classes'], channels = model_config['channel'])
elif 'target_classes' in model_config:
net = ResNet18(num_classes = model_config['target_classes'])
else:
net = ResNet18()
elif model_config['architecture'] == 'two_layer_net':
net = TwoLayerNet(model_config['input_dim'], model_config['target_classes'], model_config['hidden_units_1'])
return net
def libsvm_file_load(self, path,dim, save_data=False):
data = []
target = []
with open(path) as fp:
line = fp.readline()
while line:
temp = [i for i in line.strip().split(" ")]
target.append(int(float(temp[0]))) # Class Number. # Not assumed to be in (0, K-1)
temp_data = [0]*dim
for i in temp[1:]:
ind,val = i.split(':')
temp_data[int(ind)-1] = float(val)
data.append(temp_data)
line = fp.readline()
X_data = np.array(data,dtype=np.float32)
Y_label = np.array(target)
if save_data:
# Save the numpy files to the folder where they come from
data_np_path = path + '.data.npy'
target_np_path = path + '.label.npy'
np.save(data_np_path, X_data)
np.save(target_np_path, Y_label)
return (X_data, Y_label)
def getData(self, data_config):
# print(data_config)
if data_config['name'] == 'cifar10':
download_path = './downloaded_data/'
train_transform = transforms.Compose([transforms.RandomCrop(32, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))])
test_transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))])
train_dataset = datasets.CIFAR10(download_path, download=True, train=True, transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.CIFAR10(download_path, download=True, train=False, transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'mnist':
download_path = './downloaded_data/'
image_dim=28
train_transform = transforms.Compose([transforms.RandomCrop(image_dim, padding=4), transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
test_transform = transforms.Compose([transforms.Resize((image_dim, image_dim)), transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
train_dataset = datasets.MNIST(download_path, download=True, train=True, transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.MNIST(download_path, download=True, train=False, transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'fmnist':
download_path = './downloaded_data/'
train_transform = transforms.Compose([transforms.RandomCrop(28, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
test_transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))]) # Use mean/std of MNIST
train_dataset = datasets.FashionMNIST(download_path, download=True, train=True, transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.FashionMNIST(download_path, download=True, train=False, transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'svhn':
download_path = './downloaded_data/'
train_transform = transforms.Compose([transforms.RandomCrop(32, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))])
test_transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))]) # ImageNet mean/std
train_dataset = datasets.SVHN(download_path, download=True, split='train', transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.SVHN(download_path, download=True, split='test', transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'cifar100':
download_path = './downloaded_data/'
train_transform = transforms.Compose([transforms.RandomCrop(32, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize((0.5071, 0.4867, 0.4408), (0.2675, 0.2565, 0.2761))])
test_transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.5071, 0.4867, 0.4408), (0.2675, 0.2565, 0.2761))])
train_dataset = datasets.CIFAR100(download_path, download=True, train=True, transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.CIFAR100(download_path, download=True, train=False, transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'stl10':
download_path = './downloaded_data/'
train_transform = transforms.Compose([transforms.RandomCrop(96, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))])
test_transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))]) # ImageNet mean/std
train_dataset = datasets.STL10(download_path, download=True, train=True, transform=train_transform, target_transform=torch.tensor)
test_dataset = datasets.STL10(download_path, download=True, train=False, transform=test_transform, target_transform=torch.tensor)
elif data_config['name'] == 'satimage':
trn_file = '../datasets/satimage/satimage.scale.trn'
tst_file = '../datasets/satimage/satimage.scale.tst'
data_dims = 36
X, y = self.libsvm_file_load(trn_file, dim=data_dims)
X_test, y_test = self.libsvm_file_load(tst_file, dim=data_dims)
y -= 1 # First Class should be zero
y_test -= 1 # First Class should be zero
sc = StandardScaler()
X = sc.fit_transform(X)
X_test = sc.transform(X_test)
train_dataset = TensorDataset(torch.tensor(X), torch.tensor(y, dtype=torch.long))
test_dataset = TensorDataset(torch.tensor(X_test), torch.tensor(y_test, dtype=torch.long))
elif data_config['name'] == 'ijcnn1':
trn_file = '../datasets/ijcnn1/ijcnn1.trn'
tst_file = '../datasets/ijcnn1/ijcnn1.tst'
data_dims = 22
X, y = self.libsvm_file_load(trn_file, dim=data_dims)
X_test, y_test = self.libsvm_file_load(tst_file, dim=data_dims)
# The class labels are (-1,1). Make them to (0,1)
y[y < 0] = 0
y_test[y_test < 0] = 0
sc = StandardScaler()
X = sc.fit_transform(X)
X_test = sc.transform(X_test)
train_dataset = TensorDataset(torch.tensor(X), torch.tensor(y, dtype=torch.long))
test_dataset = TensorDataset(torch.tensor(X_test), torch.tensor(y_test, dtype=torch.long))
return train_dataset, test_dataset
def write_logs(self, logs, save_location, rd):
file_path = save_location
with open(file_path, 'a') as f:
f.write('---------------------\n')
f.write('Round '+str(rd)+'\n')
f.write('---------------------\n')
for key, val in logs.items():
if key == 'Training':
f.write(str(key)+ '\n')
for epoch in val:
f.write(str(epoch)+'\n')
else:
f.write(str(key) + ' - '+ str(val) +'\n')
def train_classifier(self):
net = self.getModel(self.config['model'])
full_train_dataset, test_dataset = self.getData(self.config['dataset'])
selected_strat = self.config['active_learning']['strategy']
budget = self.config['active_learning']['budget']
start = self.config['active_learning']['initial_points']
n_rounds = self.config['active_learning']['rounds']
nclasses = self.config['model']['target_classes']
strategy_args = self.config['active_learning']['strategy_args']
nSamps = len(full_train_dataset)
np.random.seed(42)
start_idxs = np.random.choice(nSamps, size=start, replace=False)
train_dataset = Subset(full_train_dataset, start_idxs)
unlabeled_dataset = Subset(full_train_dataset, list(set(range(len(full_train_dataset))) - set(start_idxs)))
if 'islogs' in self.config['train_parameters']:
islogs = self.config['train_parameters']['islogs']
save_location = self.config['train_parameters']['logs_location']
else:
islogs = False
dt = data_train(train_dataset, net, self.config['train_parameters'])
if selected_strat == 'badge':
strategy = BADGE(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'glister':
strategy = GLISTER(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args,validation_dataset=None,\
typeOf='Diversity',lam=10)
elif selected_strat == 'entropy_sampling':
strategy = EntropySampling(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'margin_sampling':
strategy = MarginSampling(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'least_confidence':
strategy = LeastConfidenceSampling(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'coreset':
strategy = CoreSet(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'fass':
strategy = FASS(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'random_sampling':
strategy = RandomSampling(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'bald_dropout':
strategy = BALDDropout(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'adversarial_bim':
strategy = AdversarialBIM(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'kmeans_sampling':
strategy = KMeansSampling(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
elif selected_strat == 'adversarial_deepfool':
strategy = AdversarialDeepFool(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset), net, nclasses, strategy_args)
else:
raise IOError('Enter Valid Strategy!')
if islogs:
clf, train_logs= dt.train()
else:
clf = dt.train()
strategy.update_model(clf)
acc = np.zeros(n_rounds)
acc[0] = dt.get_acc_on_set(test_dataset)
if islogs:
logs = {}
logs['Training Points'] = len(train_dataset)
logs['Test Accuracy'] = str(round(acc[0]*100, 2))
logs['Training'] = train_logs
self.write_logs(logs, save_location, 0)
print('***************************')
print('Starting Training..')
print('***************************')
##User Controlled Loop
for rd in range(1, n_rounds):
print('***************************')
print('Round', rd)
print('***************************')
logs = {}
t0 = time.time()
idx = strategy.select(budget)
t1 = time.time()
#Adding new points to training set
train_dataset = ConcatDataset([train_dataset, Subset(unlabeled_dataset, idx)])
remain_idx = list(set(range(len(unlabeled_dataset))) - set(idx))
unlabeled_dataset = Subset(unlabeled_dataset, remain_idx)
print('Total training points in this round', len(train_dataset))
strategy.update_data(train_dataset, LabeledToUnlabeledDataset(unlabeled_dataset))
dt.update_data(train_dataset)
if islogs:
clf, train_logs= dt.train()
else:
clf = dt.train()
t2 = time.time()
strategy.update_model(clf)
acc[rd] = dt.get_acc_on_set(test_dataset)
print('Testing Accuracy:', acc[rd])
if islogs:
logs['Training Points'] = len(train_dataset)
logs['Test Accuracy'] = str(round(acc[rd]*100, 2))
logs['Selection Time'] = str(t1 - t0)
logs['Trainining Time'] = str(t2 - t1)
logs['Training'] = train_logs
self.write_logs(logs, save_location, rd)
print('Training Completed!')
with open('./final_model.pkl', 'wb') as save_file:
pickle.dump(clf.state_dict(), save_file)
print('Model Saved!')
if __name__=='__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--config_path', required=True, help="Path to the config file")
args = parser.parse_args()
tc = TrainClassifier(args.config_path)
tc.train_classifier()
# tc = TrainClassifier('./configs/config_2lnet_satimage_randomsampling.json')
# # tc = TrainClassifier('./configs/config_cifar10_marginsampling.json')
# tc.train_classifier()