vae/vae_cifar10.py at master · lchingen/vae · GitHub

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
import numpy as np
import tensorflow as tf
from tensorflow.layers import conv2d, conv2d_transpose
from tensorflow.layers import batch_normalization as BN
from tensorflow.layers import flatten
from tensorflow.layers import dense
from tensorflow.nn import relu, leaky_relu, sigmoid

from config import *

def ACT(inputs, act_fn):
    if act_fn == 'relu':
        act = relu(inputs)
    elif act_fn == 'lrelu':
        act = leaky_relu(inputs)
    elif act_fn == 'sigmoid':
        act = sigmoid(inputs)
    return act


def CONV(inputs, filters, kernel_size, strides, padding, is_transpose):
    if is_transpose:
        conv = conv2d_transpose(inputs=inputs,
                                filters=filters,
                                kernel_size=kernel_size,
                                strides=strides,
                                padding=padding)
    else:
        conv = conv2d(inputs=inputs,
                      filters=filters,
                      kernel_size=kernel_size,
                      strides=strides,
                      padding=padding)
    return conv


def CONV_BN_ACT(inputs, filters, kernel_size, strides, padding, act_fn, is_training, is_transpose):
    conv = CONV(inputs=inputs,
                filters=filters,
                kernel_size=kernel_size,
                strides=strides,
                padding=padding,
                is_transpose=is_transpose)

    norm = BN(inputs=conv, training=is_training)
    act  = ACT(inputs=norm, act_fn=act_fn)
    return act


def Encoder(inputs, is_training):
    with tf.variable_scope('Encoder'):
        conv_0 = CONV_BN_ACT(inputs=inputs,
                             filters=32,
                             kernel_size=[3,3],
                             strides=[1,1],
                             padding='same',
                             act_fn='lrelu',
                             is_training=is_training,
                             is_transpose=False)

        conv_1 = CONV_BN_ACT(inputs=conv_0,
                             filters=64,
                             kernel_size=[3,3],
                             strides=[2,2],
                             padding='same',
                             act_fn='lrelu',
                             is_training=is_training,
                             is_transpose=False)

        conv_2 = CONV_BN_ACT(inputs=conv_1,
                             filters=128,
                             kernel_size=[3,3],
                             strides=[2,2],
                             padding='same',
                             act_fn='lrelu',
                             is_training=is_training,
                             is_transpose=False)

        flat = flatten(conv_2)

        z_mean = dense(inputs=flat, units=z_dim)
        z_log_var = dense(inputs=flat, units=z_dim)
        return z_mean, z_log_var


def Decoder(z, is_training):
    with tf.variable_scope('Decoder'):
        upsample = dense(inputs=z,
                         units=8192,
                         activation=relu)

        reshaped = tf.reshape(upsample, [-1,8,8,128])

        tconv_0 = CONV_BN_ACT(inputs=reshaped,
                              filters=64,
                              kernel_size=[3,3],
                              strides=[2,2],
                              padding='same',
                              act_fn='lrelu',
                              is_training=is_training,
                              is_transpose=True)

        tconv_1 = CONV_BN_ACT(inputs=tconv_0,
                              filters=32,
                              kernel_size=[3,3],
                              strides=[2,2],
                              padding='same',
                              act_fn='lrelu',
                              is_training=is_training,
                              is_transpose=True)

        tconv_2 = CONV_BN_ACT(inputs=tconv_1,
                              filters=3,
                              kernel_size=[3,3],
                              strides=[1,1],
                              padding='same',
                              act_fn='sigmoid',
                              is_training=is_training,
                              is_transpose=True)
        return tconv_2


def Vae(inputs, is_training=True):
    # Encode
    z_mean, z_log_var = Encoder(inputs, is_training)
    # Sample
    epsilon = tf.random_normal(tf.shape(z_mean))
    z = z_mean + tf.exp(z_log_var) * epsilon if (
        is_training) else z_mean
    # Decode
    y = Decoder(z, is_training)
    return y, z_mean, z_log_var