forked from akhilgoel1997/EECS553-DefenseGAN
-
Notifications
You must be signed in to change notification settings - Fork 0
/
fgsm_attacks.py
191 lines (149 loc) · 6.1 KB
/
fgsm_attacks.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
# !pip install cleverhans
# Tensorflow 2.12
import os
# from google.colab import drive
# drive.mount('/content/drive')
import numpy as np
import tensorflow as tf
import tensorflow_datasets as tfds
from easydict import EasyDict
from tensorflow.keras import Model
from tensorflow.keras.layers import Dense, Dropout, Flatten, Conv2D
from cleverhans.tf2.attacks.projected_gradient_descent import projected_gradient_descent
from cleverhans.tf2.attacks.fast_gradient_method import fast_gradient_method
import glob
# import imageio
import matplotlib.pyplot as plt
import os
import PIL
from tensorflow.keras import layers
import time
# from IPython import display
def make_generator_model():
model = tf.keras.Sequential()
model.add(layers.Dense(7*7*256, use_bias=False, input_shape=(100,)))
model.add(layers.BatchNormalization())
model.add(layers.LeakyReLU())
model.add(layers.Reshape((7, 7, 256)))
assert model.output_shape == (None, 7, 7, 256) # Note: None is the batch size
model.add(layers.Conv2DTranspose(128, (5, 5), strides=(1, 1), padding='same', use_bias=False))
assert model.output_shape == (None, 7, 7, 128)
model.add(layers.BatchNormalization())
model.add(layers.LeakyReLU())
model.add(layers.Conv2DTranspose(64, (5, 5), strides=(2, 2), padding='same', use_bias=False))
assert model.output_shape == (None, 14, 14, 64)
model.add(layers.BatchNormalization())
model.add(layers.LeakyReLU())
model.add(layers.Conv2DTranspose(1, (5, 5), strides=(2, 2), padding='same', use_bias=False, activation='tanh'))
assert model.output_shape == (None, 28, 28, 1)
return model
generator = make_generator_model()
noise = tf.random.normal([1, 100])
generated_image = generator(noise, training=False)
plt.imshow(generated_image[0, :, :, 0], cmap='gray')
checkpoint_dir = 'training_checkpoints'
checkpoint_prefix = os.path.join(checkpoint_dir, "ckpt")
checkpoint = tf.train.Checkpoint(generator=generator)
checkpoint.restore(tf.train.latest_checkpoint(checkpoint_dir))
noise = tf.random.normal([1, 100])
generated_image = generator(noise, training=False)
plt.imshow(generated_image[0, :, :, 0], cmap='gray')
R = 50
L = 200
def argminZ(image):
mainList = []
ZList = []
for i in range(R):
myloss = tf.keras.losses.MeanSquaredError()
myoptim = tf.keras.optimizers.SGD(1e-1)
Z = tf.Variable(tf.random.normal([image.shape[0], 100]), trainable=True)
with tf.GradientTape() as Z_tape:
generated_image = generator(Z, training=False)
LOSS = myloss(generated_image, image)
gradients = Z_tape.gradient(LOSS, [Z])
for j in range(L):
myoptim.apply_gradients(zip(gradients, [Z]))
mainList.append(myloss(generator(Z, training=False), image))
ZList.append(Z)
mainList = np.array(mainList)
final_z = ZList[np.argmin(mainList)]
return final_z
Xgen = None
class Net(Model):
def __init__(self):
super(Net, self).__init__()
self.conv1 = Conv2D(64, 8, strides=(2, 2), activation="relu", padding="same")
self.conv2 = Conv2D(128, 6, strides=(2, 2), activation="relu", padding="valid")
self.conv3 = Conv2D(128, 5, strides=(1, 1), activation="relu", padding="valid")
self.dropout = Dropout(0.25)
self.flatten = Flatten()
self.dense1 = Dense(128, activation="relu")
self.dense2 = Dense(10)
def call(self, x):
x = self.conv1(x)
x = self.conv2(x)
x = self.conv3(x)
x = self.dropout(x)
x = self.flatten(x)
x = self.dense1(x)
return self.dense2(x)
def ld_mnist(batch=128):
def convert_types(image, label):
image = tf.cast(image, tf.float32)
image /= 255
return image, label
dataset, info = tfds.load("mnist", with_info=True, as_supervised=True)
mnist_train, mnist_test = dataset["train"], dataset["test"]
mnist_train = mnist_train.map(convert_types).shuffle(10000).batch(batch)
mnist_test = mnist_test.map(convert_types).batch(batch)
return EasyDict(train=mnist_train, test=mnist_test)
def main():
global Xgen
data = ld_mnist()
model = Net()
loss_object = tf.losses.SparseCategoricalCrossentropy(from_logits=True)
optimizer = tf.optimizers.Adam(learning_rate=0.001)
train_loss = tf.metrics.Mean(name="train_loss")
test_acc_clean = tf.metrics.SparseCategoricalAccuracy()
test_acc_fgsm = tf.metrics.SparseCategoricalAccuracy()
test_acc_pgd = tf.metrics.SparseCategoricalAccuracy()
test_acc_defense_gan = tf.metrics.SparseCategoricalAccuracy()
@tf.function
def train_step(x, y):
with tf.GradientTape() as tape:
predictions = model(x)
loss = loss_object(y, predictions)
gradients = tape.gradient(loss, model.trainable_variables)
optimizer.apply_gradients(zip(gradients, model.trainable_variables))
train_loss(loss)
for epoch in range(8):
progress_bar_train = tf.keras.utils.Progbar(60000)
for (x, y) in data.train:
train_step(x, y)
progress_bar_train.add(x.shape[0], values=[("loss", train_loss.result())])
data = ld_mnist(1)
progress_bar_test = tf.keras.utils.Progbar(10000)
i = 0
for x, y in data.test:
i += 1
y_pred = model(x)
test_acc_clean(y, y_pred)
plt.imshow(x[0, :, :, 0], cmap='gray')
plt.show()
x_fgm = fast_gradient_method(model, x, 0.3, np.inf)
y_pred_fgm = model(x_fgm)
test_acc_fgsm(y, y_pred_fgm)
myloss = tf.keras.losses.MeanSquaredError()
loss = myloss(x_fgm, x)
Z = argminZ(x_fgm)
Xgen = generator(Z, training=False)
y_pred_defense_gan = model(Xgen)
test_acc_defense_gan(y, y_pred_defense_gan)
progress_bar_test.add(x.shape[0])
if i == 100:
break
print("test acc on clean examples (%): {:.3f}".format(test_acc_clean.result() * 100))
print("test acc on FGM adversarial examples (%): {:.3f}".format(test_acc_fgsm.result() * 100))
print("test acc on Defense GAN examples (%): {:.3f}".format(test_acc_defense_gan.result() * 100))
return x
x = main()