-
Notifications
You must be signed in to change notification settings - Fork 0
/
keras_vae.py
179 lines (152 loc) · 6.41 KB
/
keras_vae.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
from __future__ import print_function
import numpy as np
import matplotlib.pyplot as plt
from scipy.stats import norm
from keras.layers import Input, Dense, Lambda, Flatten, Reshape
from keras.layers import Conv2D, Conv2DTranspose
from keras.models import Model
from keras import backend as K
from keras import metrics
from keras.datasets import mnist
# input image dimensions
img_rows, img_cols, img_chns = 28, 28, 1
# number of convolutional filters to use
filters = 64
# convolution kernel size
num_conv = 3
batch_size = 100
if K.image_data_format() == 'channels_first':
original_img_size = (img_chns, img_rows, img_cols)
else:
original_img_size = (img_rows, img_cols, img_chns)
latent_dim = 2
intermediate_dim = 128
epsilon_std = 1.0
epochs = 5
x = Input(shape=original_img_size)
conv_1 = Conv2D(img_chns,
kernel_size=(2, 2),
padding='same', activation='relu')(x)
conv_2 = Conv2D(filters,
kernel_size=(2, 2),
padding='same', activation='relu',
strides=(2, 2))(conv_1)
conv_3 = Conv2D(filters,
kernel_size=num_conv,
padding='same', activation='relu',
strides=1)(conv_2)
conv_4 = Conv2D(filters,
kernel_size=num_conv,
padding='same', activation='relu',
strides=1)(conv_3)
flat = Flatten()(conv_4)
hidden = Dense(intermediate_dim, activation='relu')(flat)
z_mean = Dense(latent_dim)(hidden)
z_log_var = Dense(latent_dim)(hidden)
def sampling(args):
z_mean, z_log_var = args
epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim),
mean=0., stddev=epsilon_std)
return z_mean + K.exp(z_log_var) * epsilon
# note that "output_shape" isn't necessary with the TensorFlow backend
# so you could write `Lambda(sampling)([z_mean, z_log_var])`
z = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_var])
# we instantiate these layers separately so as to reuse them later
decoder_hid = Dense(intermediate_dim, activation='relu')
decoder_upsample = Dense(filters * 14 * 14, activation='relu')
if K.image_data_format() == 'channels_first':
output_shape = (batch_size, filters, 14, 14)
else:
output_shape = (batch_size, 14, 14, filters)
decoder_reshape = Reshape(output_shape[1:])
decoder_deconv_1 = Conv2DTranspose(filters,
kernel_size=num_conv,
padding='same',
strides=1,
activation='relu')
decoder_deconv_2 = Conv2DTranspose(filters,
kernel_size=num_conv,
padding='same',
strides=1,
activation='relu')
if K.image_data_format() == 'channels_first':
output_shape = (batch_size, filters, 29, 29)
else:
output_shape = (batch_size, 29, 29, filters)
decoder_deconv_3_upsamp = Conv2DTranspose(filters,
kernel_size=(3, 3),
strides=(2, 2),
padding='valid',
activation='relu')
decoder_mean_squash = Conv2D(img_chns,
kernel_size=2,
padding='valid',
activation='sigmoid')
hid_decoded = decoder_hid(z)
up_decoded = decoder_upsample(hid_decoded)
reshape_decoded = decoder_reshape(up_decoded)
deconv_1_decoded = decoder_deconv_1(reshape_decoded)
deconv_2_decoded = decoder_deconv_2(deconv_1_decoded)
x_decoded_relu = decoder_deconv_3_upsamp(deconv_2_decoded)
x_decoded_mean_squash = decoder_mean_squash(x_decoded_relu)
# instantiate VAE model
vae = Model(x, x_decoded_mean_squash)
# Compute VAE loss
xent_loss = img_rows * img_cols * metrics.binary_crossentropy(
K.flatten(x),
K.flatten(x_decoded_mean_squash))
kl_loss = - 0.5 * K.sum(1 + z_log_var - K.square(z_mean) - K.exp(z_log_var), axis=-1)
vae_loss = K.mean(xent_loss + kl_loss)
vae.add_loss(vae_loss)
vae.compile(optimizer='rmsprop')
vae.summary()
# train the VAE on MNIST digits
(x_train, _), (x_test, y_test) = mnist.load_data()
x_train = x_train.astype('float32') / 255.
x_train = x_train.reshape((x_train.shape[0],) + original_img_size)
x_test = x_test.astype('float32') / 255.
x_test = x_test.reshape((x_test.shape[0],) + original_img_size)
print('x_train.shape:', x_train.shape)
print(x_train)
print(x_train.shape)
# vae.fit(x_train,
# shuffle=True,
# epochs=epochs,
# batch_size=batch_size)
# # build a model to project inputs on the latent space
# encoder = Model(x, z_mean)
# # display a 2D plot of the digit classes in the latent space
# x_test_encoded = encoder.predict(x_test, batch_size=batch_size)
# plt.figure(figsize=(6, 6))
# plt.scatter(x_test_encoded[:, 0], x_test_encoded[:, 1], c=y_test)
# plt.colorbar()
# plt.show()
# # build a digit generator that can sample from the learned distribution
# decoder_input = Input(shape=(latent_dim,))
# _hid_decoded = decoder_hid(decoder_input)
# _up_decoded = decoder_upsample(_hid_decoded)
# _reshape_decoded = decoder_reshape(_up_decoded)
# _deconv_1_decoded = decoder_deconv_1(_reshape_decoded)
# _deconv_2_decoded = decoder_deconv_2(_deconv_1_decoded)
# _x_decoded_relu = decoder_deconv_3_upsamp(_deconv_2_decoded)
# _x_decoded_mean_squash = decoder_mean_squash(_x_decoded_relu)
# generator = Model(decoder_input, _x_decoded_mean_squash)
# # display a 2D manifold of the digits
# n = 15 # figure with 15x15 digits
# digit_size = 28
# figure = np.zeros((digit_size * n, digit_size * n))
# # linearly spaced coordinates on the unit square were transformed through the inverse CDF (ppf) of the Gaussian
# # to produce values of the latent variables z, since the prior of the latent space is Gaussian
# grid_x = norm.ppf(np.linspace(0.05, 0.95, n))
# grid_y = norm.ppf(np.linspace(0.05, 0.95, n))
# for i, yi in enumerate(grid_x):
# for j, xi in enumerate(grid_y):
# z_sample = np.array([[xi, yi]])
# z_sample = np.tile(z_sample, batch_size).reshape(batch_size, 2)
# x_decoded = generator.predict(z_sample, batch_size=batch_size)
# digit = x_decoded[0].reshape(digit_size, digit_size)
# figure[i * digit_size: (i + 1) * digit_size,
# j * digit_size: (j + 1) * digit_size] = digit
# plt.figure(figsize=(10, 10))
# plt.imshow(figure, cmap='Greys_r')
# plt.show()