Variational Autoencoder 变分自动编码器

一步一步实现一个VAE

大部分来自Keras VAE的教程，不过没有使用mnist，而是用了cifar10的数据集

最简单的两个全连接层的Autoencoder

先贴个代码:

# this is the size of our encoded representations
encoding_dim = 32  # 32 floats -> compression of factor 24.5, assuming the input is 784 floats
# this is our input placeholder
input_img = Input(shape=(784,))
# "encoded" is the encoded representation of the input
encoded = Dense(encoding_dim, activation='relu')(input_img)
# "decoded" is the lossy reconstruction of the input
decoded = Dense(784, activation='sigmoid')(encoded)
# this model maps an input to its reconstruction
autoencoder = Model(input_img, decoded)
# this model maps an input to its encoded representation
autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')

这里就是把输入层放入一个全连接层，压缩成为一个32维的数组，而后再使用这个32维的数组用一个全连接层去还原原来的图片。
下面这段代码是用来显示效果，这里仍是用的mnist的数据集。

encoder = Model(input_img, encoded)
# create a placeholder for an encoded (32-dimensional) input
encoded_input = Input(shape=(encoding_dim,))
# retrieve the last layer of the autoencoder model
decoder_layer = autoencoder.layers[-1]
# create the decoder model
decoder = Model(encoded_input, decoder_layer(encoded_input))

结果以下：

使用多层全连接层的Autoencoder

input_img = Input(shape=(784,))
encoded = Dense(128, activation='relu')(input_img)
encoded = Dense(64, activation='relu')(encoded)
encoded = Dense(32, activation='relu')(encoded)

decoded = Dense(64, activation='relu')(encoded)
decoded = Dense(128, activation='relu')(decoded)
decoded = Dense(784, activation='sigmoid')(decoded)


autoencoder = Model(input_img, decoded)
autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')

这里使用三层全连接层作encoder，三层作decoder，仍是压缩到32维

效果好了一点点

深度卷积网络来构建的autoencoder

咱们在这里用了cifar10做为数据集

input_img = Input(shape=(32, 32, 3))
x = Conv2D(64, (3, 3), padding='same')(input_img)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = MaxPooling2D((2, 2), padding='same')(x)
x = Conv2D(32, (3, 3), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = MaxPooling2D((2, 2), padding='same')(x)
x = Conv2D(16, (3, 3), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
encoded = MaxPooling2D((2, 2), padding='same')(x)

x = Conv2D(16, (3, 3), padding='same')(encoded)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = UpSampling2D((2, 2))(x)
x = Conv2D(32, (3, 3), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = UpSampling2D((2, 2))(x)
x = Conv2D(64, (3, 3), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = UpSampling2D((2, 2))(x)
x = Conv2D(3, (3, 3), padding='same')(x)
x = BatchNormalization()(x)
decoded = Activation('sigmoid')(x)

autoencoder = Model(input_img, decoded)
autoencoder.compile(optimizer='adam', loss='binary_crossentropy')

输入为(32,32,3) 最后压缩到16维。