在该项目中，你将使用生成式对抗网络（Generative Adversarial Nets）来生成新的人脸图像。
该项目将使用以下数据集：
由于 CelebA 数据集比较复杂，而且这是你第一次使用 GANs。我们想让你先在 MNIST 数据集上测试你的 GANs 模型，以让你更快的评估所建立模型的性能。
如果你在使用 FloydHub, 请将 data_dir 设置为 "/input" 并使用 FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".
In [5]:
data_dir = './data'
# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper
helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)
In [6]:
show_n_images = 25
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')
Out[6]:
CelebFaces Attributes Dataset (CelebA) 是一个包含 20 多万张名人图片及相关图片说明的数据集。你将用此数据集生成人脸，不会用不到相关说明。你可以更改 show_n_images 探索此数据集。
In [7]:
show_n_images = 25
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))
Out[7]:
由于该项目的重点是建立 GANs 模型，我们将为你预处理数据。
经过数据预处理，MNIST 和 CelebA 数据集的值在 28×28 维度图像的 [-0.5, 0.5] 范围内。CelebA 数据集中的图像裁剪了非脸部的图像部分，然后调整到 28x28 维度。
MNIST 数据集中的图像是单通道的黑白图像，CelebA 数据集中的图像是 三通道的 RGB 彩色图像。
你将通过部署以下函数来建立 GANs 的主要组成部分:
model_inputsdiscriminatorgeneratormodel_lossmodel_opttrain检查你是否使用正确的 TensorFlow 版本，并获取 GPU 型号
In [8]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf
# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer. You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))
# Check for a GPU
if not tf.test.gpu_device_name():
warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
部署 model_inputs 函数以创建用于神经网络的 占位符 (TF Placeholders)。请创建以下占位符：
image_width，image_height 和 image_channels 设置为 rank 4。z_dim。返回占位符元组的形状为 (tensor of real input images, tensor of z data, learning rate)。
In [10]:
import problem_unittests as tests
def model_inputs(image_width, image_height, image_channels, z_dim):
"""
Create the model inputs
:param image_width: The input image width
:param image_height: The input image height
:param image_channels: The number of image channels
:param z_dim: The dimension of Z
:return: Tuple of (tensor of real input images, tensor of z data, learning rate)
"""
# TODO: Implement Function
x = tf.placeholder(dtype=tf.float32,shape=(None,image_width,image_height,image_channels),name="inputs_x")
z = tf.placeholder(dtype=tf.float32,shape=(None,z_dim),name="z_dim")
learning_rate = tf.placeholder(dtype=tf.float32,shape=(),name="learning_rate")
return x, z, learning_rate
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)
部署 discriminator 函数创建辨别器神经网络以辨别 images。该函数应能够重复使用神经网络中的各种变量。 在 tf.variable_scope 中使用 "discriminator" 的变量空间名来重复使用该函数中的变量。
该函数应返回形如 (tensor output of the discriminator, tensor logits of the discriminator) 的元组。
In [11]:
def discriminator(images, reuse=False):
"""
Create the discriminator network
:param image: Tensor of input image(s)
:param reuse: Boolean if the weights should be reused
:return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
"""
# TODO: Implement Function
alpha=0.2
with tf.variable_scope("discriminator",reuse=reuse):
conv_1 = tf.layers.conv2d(images, 64, 4, strides=2, padding='same')
conv_1_bn = tf.layers.batch_normalization(conv_1, training=True)
relu1 = tf.maximum(alpha * conv_1_bn, conv_1_bn)
# 14x14x64
conv_2 = tf.layers.conv2d(relu1, 128, 4, strides=2, padding='same')
conv_2_bn = tf.layers.batch_normalization(conv_2, training=True)
relu2 = tf.maximum(alpha * conv_2_bn, conv_2_bn)
# 7x7x128
conv_3 = tf.layers.conv2d(relu2, 256, 4, strides=2, padding='same')
conv_3_bn = tf.layers.batch_normalization(conv_3, training=True)
relu3 = tf.maximum(alpha * conv_3_bn, conv_3_bn)
# 4x4x256
flat = tf.reshape(relu3, (-1, 4*4*256))
logits = tf.layers.dense(flat, 1)
out = tf.sigmoid(logits)
return out, logits
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)
部署 generator 函数以使用 z 生成图像。该函数应能够重复使用神经网络中的各种变量。
在 tf.variable_scope 中使用 "generator" 的变量空间名来重复使用该函数中的变量。
该函数应返回所生成的 28 x 28 x out_channel_dim 维度图像。
In [12]:
def generator(z, out_channel_dim, is_train=True):
"""
Create the generator network
:param z: Input z
:param out_channel_dim: The number of channels in the output image
:param is_train: Boolean if generator is being used for training
:return: The tensor output of the generator
"""
# TODO: Implement Function
alpha=0.2
with tf.variable_scope('generator',reuse=not(is_train)):
# Reshape it to start the convolutional stack
fc = tf.layers.dense(z, 4*4*512)
x = tf.reshape(fc, (-1, 4, 4, 512))
x = tf.layers.batch_normalization(x, training=is_train)
relu_x = tf.maximum(alpha * x, x)
# 4x4x512 now
#conv2d_1_t = tf.nn.conv2d_transpose(relu_x, kernel, output_shape=[64,7,7,256], strides=[1,2,2,1], padding='SAME')
conv2d_1_t = tf.layers.conv2d_transpose(relu_x, 256,4, strides=1, padding='valid')
conv2d_1_t_bn = tf.layers.batch_normalization(conv2d_1_t, training=is_train)
relu1 = tf.maximum(alpha * conv2d_1_t_bn, conv2d_1_t_bn)
# 7x7x256 now
conv2d_2_t = tf.layers.conv2d_transpose(relu1, 128, 4, strides=2, padding='same')
conv2d_2_t_bn = tf.layers.batch_normalization(conv2d_2_t, training=is_train)
relu2 = tf.maximum(alpha * conv2d_2_t_bn, conv2d_2_t_bn)
# 14x14x128
# Output layer
logits = tf.layers.conv2d_transpose(relu2, out_channel_dim, 4, strides=2, padding='same')
# 28x28x3 now
out = tf.tanh(logits)
return out
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)
In [15]:
def model_loss(input_real, input_z, out_channel_dim):
"""
Get the loss for the discriminator and generator
:param input_real: Images from the real dataset
:param input_z: Z input
:param out_channel_dim: The number of channels in the output image
:return: A tuple of (discriminator loss, generator loss)
"""
# TODO: Implement Function
g_model = generator(input_z, out_channel_dim)
d_model_real, d_logits_real = discriminator(input_real)
d_model_fake, d_logits_f = discriminator(g_model, reuse=True)
d_loss_real = tf.reduce_mean(
tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_logits_real)*(1 - 0.1)))
d_loss_f = tf.reduce_mean(
tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_f, labels=tf.zeros_like(d_logits_f)))
g_loss = tf.reduce_mean(
tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_f, labels=tf.ones_like(d_logits_f)))
d_loss = d_loss_real + d_loss_f
return d_loss, g_loss
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)
部署 model_opt 函数实现对 GANs 的优化。使用 tf.trainable_variables 获取可训练的所有变量。通过变量空间名 discriminator 和 generator 来过滤变量。该函数应返回形如 (discriminator training operation, generator training operation) 的元组。
In [16]:
def model_opt(d_loss, g_loss, learning_rate, beta1):
"""
Get optimization operations
:param d_loss: Discriminator loss Tensor
:param g_loss: Generator loss Tensor
:param learning_rate: Learning Rate Placeholder
:param beta1: The exponential decay rate for the 1st moment in the optimizer
:return: A tuple of (discriminator training operation, generator training operation)
"""
# TODO: Implement Function
t_vars = tf.trainable_variables()
d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
g_vars = [var for var in t_vars if var.name.startswith('generator')]
# Optimize
with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)
return d_train_opt, g_train_opt
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)
In [18]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np
def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
"""
Show example output for the generator
:param sess: TensorFlow session
:param n_images: Number of Images to display
:param input_z: Input Z Tensor
:param out_channel_dim: The number of channels in the output image
:param image_mode: The mode to use for images ("RGB" or "L")
"""
cmap = None if image_mode == 'RGB' else 'gray'
z_dim = input_z.get_shape().as_list()[-1]
example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])
samples = sess.run(
generator(input_z, out_channel_dim, False),
feed_dict={input_z: example_z})
images_grid = helper.images_square_grid(samples, image_mode)
pyplot.imshow(images_grid, cmap=cmap)
pyplot.show()
部署 train 函数以建立并训练 GANs 模型。记得使用以下你已完成的函数：
model_inputs(image_width, image_height, image_channels, z_dim)model_loss(input_real, input_z, out_channel_dim)model_opt(d_loss, g_loss, learning_rate, beta1)使用 show_generator_output 函数显示 generator 在训练过程中的输出。
注意：在每个批次 (batch) 中运行 show_generator_output 函数会显著增加训练时间与该 notebook 的体积。推荐每 100 批次输出一次 generator 的输出。
In [19]:
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
"""
Train the GAN
:param epoch_count: Number of epochs
:param batch_size: Batch Size
:param z_dim: Z dimension
:param learning_rate: Learning Rate
:param beta1: The exponential decay rate for the 1st moment in the optimizer
:param get_batches: Function to get batches
:param data_shape: Shape of the data
:param data_image_mode: The image mode to use for images ("RGB" or "L")
"""
# TODO: Build Model
input_real,input_z, lr = model_inputs(*data_shape[1:], z_dim)
out_channel_dim = 1
if data_image_mode == "RGB" :
out_channel_dim = 3
d_loss, g_loss = model_loss(input_real,input_z,out_channel_dim)
d_train_opt, g_train_opt = model_opt(d_loss,g_loss,lr,beta1)
steps = 0
start = time.clock()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
for epoch_i in range(epoch_count):
for batch_images in get_batches(batch_size):
# TODO: Train Model
steps += 1
#print(batch_images)
batch_images = batch_images * 2
batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))
_ = sess.run(d_train_opt, feed_dict={input_real: batch_images, input_z: batch_z, lr :learning_rate})
_ = sess.run(g_train_opt, feed_dict={input_real: batch_images, input_z: batch_z, lr :learning_rate})
if steps % 100 == 0:
end = time.clock()
train_loss_d = d_loss.eval({input_real: batch_images, input_z: batch_z})
train_loss_g = g_loss.eval({input_z: batch_z})
print("Time: {:.4f}s ...Epoch {}/{}...setp:{}...".format((end - start),epoch_i+1, epoch_count,steps),
"Discriminator Loss: {:.4f}...".format(train_loss_d),
"Generator Loss: {:.4f}".format(train_loss_g))
start = time.clock()
if steps % 500 == 0:
show_generator_output(sess,4,input_z,out_channel_dim,data_image_mode)
show_generator_output(sess,4,input_z,out_channel_dim,data_image_mode)
In [ ]:
import time
batch_size = 32
z_dim = 100
learning_rate = 0.001
beta1 = 0.5
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2
mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
mnist_dataset.shape, mnist_dataset.image_mode)
In [ ]:
batch_size = None
z_dim = None
learning_rate = None
beta1 = None
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1
celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
celeba_dataset.shape, celeba_dataset.image_mode)