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
| from keras.datasets import mnist
from keras.models import Model, Sequential
from keras.layers import Dense, Flatten, BatchNormalization, Reshape, Input, \
LeakyReLU, Embedding, multiply, Activation, Conv2DTranspose, Conv2D, Dropout
from keras.optimizers import Adam
from keras.utils import np_utils
import numpy as np
import matplotlib.pyplot as plt
class CGAN:
def __init__(self):
self.img_row = 28
self.img_col = 28
self.channel = 1
self.img_shape = (self.img_row, self.img_col, self.channel)
self.latent_dim = 100
self.num_class = 10
self.buildGAN()
def buildGenerator(self):
model = Sequential()
model.add(Dense(input_dim=self.latent_dim, units=7 * 7 * 256))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation('relu'))
model.add(Reshape((7, 7, 256)))
model.add(Conv2DTranspose(filters=128, kernel_size=3, strides=2, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation('relu'))
model.add(Conv2DTranspose(filters=64, kernel_size=3, strides=2, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation('relu'))
model.add(Conv2DTranspose(filters=32, kernel_size=3, strides=1, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation('relu'))
model.add(Conv2DTranspose(filters=1, kernel_size=3, strides=1, padding='same'))
model.add(Activation('tanh'))
noise = Input(shape=(self.latent_dim,))
label = Input(shape=(1,))
labelEmbeded = Flatten()(Embedding(self.num_class, self.latent_dim)(label))
input = multiply([noise, labelEmbeded])
img = model(input)
return Model([noise, label], img)
def buildDiscriminator(self):
model = Sequential()
model.add(Conv2D(input_shape=self.img_shape, filters=64, kernel_size=3, strides=2, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(0.2))
model.add(Dropout(0.4))
model.add(Conv2D(filters=128, kernel_size=3, strides=2, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(0.2))
model.add(Dropout(0.4))
model.add(Conv2D(filters=256, kernel_size=3, strides=2, padding='same'))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(0.2))
model.add(Dropout(0.4))
model.add(Flatten())
img = Input(shape=self.img_shape)
x = model(img)
label_pred = Dense(10, activation='softmax')(x)
validity = Dense(1, activation='sigmoid')(x)
discriminator = Model(img, [validity, label_pred])
discriminator.compile(optimizer=Adam(2e-4),
loss=['binary_crossentropy', 'categorical_crossentropy'])
return discriminator
def buildGAN(self):
self.generator = self.buildGenerator()
self.discriminator = self.buildDiscriminator()
noise = Input(shape=(self.latent_dim,))
label = Input(shape=(1,))
img = self.generator([noise, label])
self.discriminator.trainable = False
validity, label_pred = self.discriminator(img)
self.combined = Model([noise, label], [validity, label_pred])
self.combined.compile(optimizer=Adam(2e-4),
loss=['binary_crossentropy', 'categorical_crossentropy'])
def trainModel(self, epochs, batch_size=64):
(X_train, Y_train), (X_test, Y_test) = mnist.load_data()
X_train = X_train / 127.5 - 1.
X_train = np.expand_dims(X_train, axis=3)
valid = np.ones((batch_size, 1))
fake = np.zeros((batch_size, 1))
for epoch in range(epochs):
epoch += 1
idx = np.random.randint(0, X_train.shape[0], batch_size)
orgImgs = X_train[idx]
label = Y_train[idx]
label_onehot = np_utils.to_categorical(label, num_classes=10)
noise = np.random.normal(0, 1, (batch_size, self.latent_dim))
sampleLabel = np.random.randint(0, 10, batch_size).reshape(-1, 1)
sampleLabel_onehot = np_utils.to_categorical(sampleLabel, num_classes=10)
genImgs = self.generator.predict([noise, sampleLabel])
D_loss_real = self.discriminator.train_on_batch(orgImgs, [valid, label_onehot])
D_loss_fake = self.discriminator.train_on_batch(genImgs, [fake, sampleLabel_onehot])
D_loss = 0.5 * np.add(D_loss_real, D_loss_fake)
noise = np.random.normal(0, 1, (batch_size, self.latent_dim))
sampleLabel = np.random.randint(0, 10, batch_size).reshape(-1, 1)
sampleLabel_onehot = np_utils.to_categorical(sampleLabel, num_classes=10)
G_loss = self.combined.train_on_batch([noise, sampleLabel], [valid, sampleLabel_onehot])
print("epoch {} --- D_loss: {:.4f}, G loss: {:.4f}".format(
epoch, np.mean(D_loss), np.mean(G_loss)))
if epoch % 400 == 0:
self.saveImage(epoch)
def saveImage(self, epoch):
r, c = 3, 10
noise = np.random.normal(0, 1, (r * c, self.latent_dim))
label = np.arange(0, 10).reshape(-1, 1)
for i in range(2):
label = np.vstack((label, np.arange(0, 10).reshape(-1, 1)))
genImgs = self.generator.predict([noise, label])
genImgs = 0.5 * genImgs + 0.5
fig, axs = plt.subplots(r, c)
cnt = 0
for i in range(r):
for j in range(c):
axs[i, j].imshow(genImgs[cnt, :, :, 0], cmap='gray')
axs[i, j].axis('off')
cnt += 1
fig.savefig('generated\\%d.png' % epoch)
plt.close()
def main():
cgan = CGAN()
cgan.trainModel(epochs=10000)
if __name__ == '__main__':
main()
|
