1. Auto Encoder

解析：

# -*- coding: utf-8 -*-
"""
Using an auto encoder on MNIST handwritten digits.
"""
from __future__ import division, print_function, absolute_import

import numpy as np
import matplotlib.pyplot as plt
import tflearn

# Data loading and preprocessing
import tflearn.datasets.mnist as mnist

X, Y, testX, testY = mnist.load_data(one_hot=True)

# Building the encoder
encoder = tflearn.input_data(shape=[None, 784])
encoder = tflearn.fully_connected(encoder, 256)
encoder = tflearn.fully_connected(encoder, 64)

# Building the decoder
decoder = tflearn.fully_connected(encoder, 256)
decoder = tflearn.fully_connected(decoder, 784, activation='sigmoid')

# Regression, with mean square error
net = tflearn.regression(decoder, optimizer='adam', learning_rate=0.001,
                         loss='mean_square', metric=None)

# Training the auto encoder
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X, X, n_epoch=20, validation_set=(testX, testX),
          run_id="auto_encoder", batch_size=256)

# Encoding X[0] for test
print("\nTest encoding of X[0]:")
# New model, re-using the same session, for weights sharing
encoding_model = tflearn.DNN(encoder, session=model.session)
print(encoding_model.predict([X[0]]))

# Testing the image reconstruction on new data (test set)
print("\nVisualizing results after being encoded and decoded:")
testX = tflearn.data_utils.shuffle(testX)[0]
# Applying encode and decode over test set
encode_decode = model.predict(testX)
# Compare original images with their reconstructions
f, a = plt.subplots(2, 10, figsize=(10, 2))
for i in range(10):
    temp = [[ii, ii, ii] for ii in list(testX[i])]
    a[0][i].imshow(np.reshape(temp, (28, 28, 3)))
    temp = [[ii, ii, ii] for ii in list(encode_decode[i])]
    a[1][i].imshow(np.reshape(temp, (28, 28, 3)))
f.show()
plt.draw()
plt.waitforbuttonpress()
 

2. Variational Auto Encoder

解析：

# -*- coding: utf-8 -*-
"""
Using a variational auto-encoder to generate digits images from noise.
MNIST handwritten digits are used as training examples.
"""
from __future__ import division, print_function, absolute_import

import numpy as np
import matplotlib.pyplot as plt
from scipy.stats import norm
import tensorflow as tf

import tflearn

# Data loading and preprocessing
import tflearn.datasets.mnist as mnist

X, Y, testX, testY = mnist.load_data(one_hot=True)

# Params
original_dim = 784  # MNIST images are 28x28 pixels
hidden_dim = 256
latent_dim = 2

# Building the encoder
encoder = tflearn.input_data(shape=[None, 784], name='input_images')
encoder = tflearn.fully_connected(encoder, hidden_dim, activation='relu')
z_mean = tflearn.fully_connected(encoder, latent_dim)
z_std = tflearn.fully_connected(encoder, latent_dim)

# Sampler: Normal (gaussian) random distribution
eps = tf.random_normal(tf.shape(z_std), dtype=tf.float32, mean=0., stddev=1.0,
                       name='epsilon')
z = z_mean + tf.exp(z_std / 2) * eps

# Building the decoder (with scope to re-use these layers later)
decoder = tflearn.fully_connected(z, hidden_dim, activation='relu',
                                  scope='decoder_h')
decoder = tflearn.fully_connected(decoder, original_dim, activation='sigmoid',
                                  scope='decoder_out')


# Define VAE Loss
def vae_loss(x_reconstructed, x_true):
    # Reconstruction loss
    encode_decode_loss = x_true * tf.log(1e-10 + x_reconstructed) \
                         + (1 - x_true) * tf.log(1e-10 + 1 - x_reconstructed)
    encode_decode_loss = -tf.reduce_sum(encode_decode_loss, 1)
    # KL Divergence loss
    kl_div_loss = 1 + z_std - tf.square(z_mean) - tf.exp(z_std)
    kl_div_loss = -0.5 * tf.reduce_sum(kl_div_loss, 1)
    return tf.reduce_mean(encode_decode_loss + kl_div_loss)


net = tflearn.regression(decoder, optimizer='rmsprop', learning_rate=0.001,
                         loss=vae_loss, metric=None, name='target_images')

# We will need 2 models, one for training that will learn the latent
# representation, and one that can take random normal noise as input and
# use the decoder part of the network to generate an image

# Train the VAE
training_model = tflearn.DNN(net, tensorboard_verbose=0)
training_model.fit({'input_images': X}, {'target_images': X}, n_epoch=100,
                   validation_set=(testX, testX), batch_size=256, run_id="vae")

# Build an image generator (re-using the decoding layers)
# Input data is a normal (gaussian) random distribution (with dim = latent_dim)
input_noise = tflearn.input_data(shape=[None, latent_dim], name='input_noise')
decoder = tflearn.fully_connected(input_noise, hidden_dim, activation='relu',
                                  scope='decoder_h', reuse=True)
decoder = tflearn.fully_connected(decoder, original_dim, activation='sigmoid',
                                  scope='decoder_out', reuse=True)
generator_model = tflearn.DNN(decoder, session=training_model.session)

# Building a manifold of generated digits
n = 25  # Figure row size
figure = np.zeros((28 * n, 28 * n))
# Random normal distributions to feed network with
x_axis = norm.ppf(np.linspace(0., 1., n))
y_axis = norm.ppf(np.linspace(0., 1., n))

for i, x in enumerate(x_axis):
    for j, y in enumerate(y_axis):
        samples = np.array([[x, y]])
        x_reconstructed = generator_model.predict({'input_noise': samples})
        digit = np.array(x_reconstructed[0]).reshape(28, 28)
        figure[i * 28: (i + 1) * 28, j * 28: (j + 1) * 28] = digit

plt.figure(figsize=(10, 10))
plt.imshow(figure, cmap='Greys_r')
plt.show()
 

3. GAN (Generative Adversarial Networks)

解析：

# -*- coding: utf-8 -*-
"""
Use a generative adversarial network (GAN) to generate digit images from a
noise distribution.
"""

from __future__ import division, print_function, absolute_import

import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
import tflearn

# Data loading and preprocessing
import tflearn.datasets.mnist as mnist

X, Y, testX, testY = mnist.load_data()

image_dim = 784  # 28*28 pixels
z_dim = 200  # Noise data points
total_samples = len(X)


# Generator
def generator(x, reuse=False):
    with tf.variable_scope('Generator', reuse=reuse):
        x = tflearn.fully_connected(x, 256, activation='relu')
        x = tflearn.fully_connected(x, image_dim, activation='sigmoid')
        return x


# Discriminator
def discriminator(x, reuse=False):
    with tf.variable_scope('Discriminator', reuse=reuse):
        x = tflearn.fully_connected(x, 256, activation='relu')
        x = tflearn.fully_connected(x, 1, activation='sigmoid')
        return x


# Build Networks
gen_input = tflearn.input_data(shape=[None, z_dim], name='input_noise')
disc_input = tflearn.input_data(shape=[None, 784], name='disc_input')

gen_sample = generator(gen_input)
disc_real = discriminator(disc_input)
disc_fake = discriminator(gen_sample, reuse=True)

# Define Loss
disc_loss = -tf.reduce_mean(tf.log(disc_real) + tf.log(1. - disc_fake))
gen_loss = -tf.reduce_mean(tf.log(disc_fake))

# Build Training Ops for both Generator and Discriminator.
# Each network optimization should only update its own variable, thus we need
# to retrieve each network variables (with get_layer_variables_by_scope) and set
# 'placeholder=None' because we do not need to feed any target.
gen_vars = tflearn.get_layer_variables_by_scope('Generator')
gen_model = tflearn.regression(gen_sample, placeholder=None, optimizer='adam',
                               loss=gen_loss, trainable_vars=gen_vars,
                               batch_size=64, name='target_gen', op_name='GEN')
disc_vars = tflearn.get_layer_variables_by_scope('Discriminator')
disc_model = tflearn.regression(disc_real, placeholder=None, optimizer='adam',
                                loss=disc_loss, trainable_vars=disc_vars,
                                batch_size=64, name='target_disc', op_name='DISC')
# Define GAN model, that output the generated images.
gan = tflearn.DNN(gen_model)

# Training
# Generate noise to feed to the generator
z = np.random.uniform(-1., 1., size=[total_samples, z_dim])
# Start training, feed both noise and real images.
gan.fit(X_inputs={gen_input: z, disc_input: X},
        Y_targets=None,
        n_epoch=100)

# Generate images from noise, using the generator network.
f, a = plt.subplots(2, 10, figsize=(10, 4))
for i in range(10):
    for j in range(2):
        # Noise input.
        z = np.random.uniform(-1., 1., size=[1, z_dim])
        # Generate image from noise. Extend to 3 channels for matplot figure.
        temp = [[ii, ii, ii] for ii in list(gan.predict([z])[0])]
        a[j][i].imshow(np.reshape(temp, (28, 28, 3)))
f.show()
plt.draw()
plt.waitforbuttonpress()
 

4. DCGAN (Deep Convolutional Generative Adversarial Networks)

解析：

# -*- coding: utf-8 -*-
"""
Use a deep convolutional generative adversarial network (DCGAN) to generate
digit images from a noise distribution.
"""

from __future__ import division, print_function, absolute_import

import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
import tflearn

# Data loading and preprocessing
import tflearn.datasets.mnist as mnist

X, Y, testX, testY = mnist.load_data()
X = np.reshape(X, newshape=[-1, 28, 28, 1])

z_dim = 200  # Noise data points
total_samples = len(X)


# Generator
def generator(x, reuse=False):
    with tf.variable_scope('Generator', reuse=reuse):
        x = tflearn.fully_connected(x, n_units=7 * 7 * 128)
        x = tflearn.batch_normalization(x)
        x = tf.nn.tanh(x)
        x = tf.reshape(x, shape=[-1, 7, 7, 128])
        x = tflearn.upsample_2d(x, 2)
        x = tflearn.conv_2d(x, 64, 5, activation='tanh')
        x = tflearn.upsample_2d(x, 2)
        x = tflearn.conv_2d(x, 1, 5, activation='sigmoid')
        return x


# Discriminator
def discriminator(x, reuse=False):
    with tf.variable_scope('Discriminator', reuse=reuse):
        x = tflearn.conv_2d(x, 64, 5, activation='tanh')
        x = tflearn.avg_pool_2d(x, 2)
        x = tflearn.conv_2d(x, 128, 5, activation='tanh')
        x = tflearn.avg_pool_2d(x, 2)
        x = tflearn.fully_connected(x, 1024, activation='tanh')
        x = tflearn.fully_connected(x, 2)
        x = tf.nn.softmax(x)
        return x


# Input Data
gen_input = tflearn.input_data(shape=[None, z_dim], name='input_gen_noise')
input_disc_noise = tflearn.input_data(shape=[None, z_dim], name='input_disc_noise')
input_disc_real = tflearn.input_data(shape=[None, 28, 28, 1], name='input_disc_real')

# Build Discriminator
disc_fake = discriminator(generator(input_disc_noise))
disc_real = discriminator(input_disc_real, reuse=True)
disc_net = tf.concat([disc_fake, disc_real], axis=0)
# Build Stacked Generator/Discriminator
gen_net = generator(gen_input, reuse=True)
stacked_gan_net = discriminator(gen_net, reuse=True)

# Build Training Ops for both Generator and Discriminator.
# Each network optimization should only update its own variable, thus we need
# to retrieve each network variables (with get_layer_variables_by_scope).
disc_vars = tflearn.get_layer_variables_by_scope('Discriminator')
# We need 2 target placeholders, for both the real and fake image target.
disc_target = tflearn.multi_target_data(['target_disc_fake', 'target_disc_real'],
                                        shape=[None, 2])
disc_model = tflearn.regression(disc_net, optimizer='adam',
                                placeholder=disc_target,
                                loss='categorical_crossentropy',
                                trainable_vars=disc_vars,
                                batch_size=64, name='target_disc',
                                op_name='DISC')

gen_vars = tflearn.get_layer_variables_by_scope('Generator')
gan_model = tflearn.regression(stacked_gan_net, optimizer='adam',
                               loss='categorical_crossentropy',
                               trainable_vars=gen_vars,
                               batch_size=64, name='target_gen',
                               op_name='GEN')

# Define GAN model, that output the generated images.
gan = tflearn.DNN(gan_model)

# Training
# Prepare input data to feed to the discriminator
disc_noise = np.random.uniform(-1., 1., size=[total_samples, z_dim])
# Prepare target data to feed to the discriminator (0: fake image, 1: real image)
y_disc_fake = np.zeros(shape=[total_samples])
y_disc_real = np.ones(shape=[total_samples])
y_disc_fake = tflearn.data_utils.to_categorical(y_disc_fake)
y_disc_real = tflearn.data_utils.to_categorical(y_disc_real)

# Prepare input data to feed to the stacked generator/discriminator
gen_noise = np.random.uniform(-1., 1., size=[total_samples, z_dim])
# Prepare target data to feed to the discriminator
# Generator tries to fool the discriminator, thus target is 1 (e.g. real images)
y_gen = np.ones(shape=[total_samples])
y_gen = tflearn.data_utils.to_categorical(y_gen)

# Start training, feed both noise and real images.
gan.fit(X_inputs={'input_gen_noise': gen_noise,
                  'input_disc_noise': disc_noise,
                  'input_disc_real': X},
        Y_targets={'target_gen': y_gen,
                   'target_disc_fake': y_disc_fake,
                   'target_disc_real': y_disc_real},
        n_epoch=10)

# Create another model from the generator graph to generate some samples
# for testing (re-using same session to re-use the weights learnt).
gen = tflearn.DNN(gen_net, session=gan.session)

f, a = plt.subplots(4, 10, figsize=(10, 4))
for i in range(10):
    # Noise input.
    z = np.random.uniform(-1., 1., size=[4, z_dim])
    g = np.array(gen.predict({'input_gen_noise': z}))
    for j in range(4):
        # Generate image from noise. Extend to 3 channels for matplot figure.
        img = np.reshape(np.repeat(g[j][:, :, np.newaxis], 3, axis=2),
                         newshape=(28, 28, 3))
        a[j][i].imshow(img)

f.show()
plt.draw()
plt.waitforbuttonpress()

參考文獻：