生成式对搞互连网富含多个浮动模型(generative

生成式对抗网络(gennerative adversarial
network,GAN),Google二〇一五年建议网络模型。灵感自二人博艺的零和博艺,近来最火的非监督深度学习。GAN之父,伊恩J.Goodfellow,公众承认人工智能顶尖专家。

原理。
生成式对搞网络满含叁个变迁模型(generative
model,G)和一个鉴定区别模型(discriminative model,D)。伊恩 J.Goodfellow、JeanPouget-Abadie、Mehdi Mirza、Bing Xu、大卫 Warde-Farley、Sherjil
Ozair、亚伦 Courville、Yoshua Bengio杂文,《Generative Adversarial
Network》,https://arxiv.org/abs/1406.2661
生成式对抗互联网布局:
噪音数据->生成模型->假图片—|
|->推断模型->真/假
打乱练习多少->磨炼集->真图片-|
生成式对抗网络重大化解什么从磨炼样本中学习出新样本。生成模型负担磨炼出样本的遍及,要是训练样本是图表就变化相似的图形,如若陶冶样本是小说名子就转变相似的作品名子。决断模型是贰个二分类器,用来推断输入样本是实在数据依旧教练调换的样本。
生成式对抗网络优化,是三个二元极小十分大博艺(minimax two-player
game)难题。使生成模型输出在输入给判定模型时,推断模型秀难判别是实际数据照旧虚似数据。演习好的生成模型,能把贰个噪音向量转化成和操练集类似的样书。Argustus
Odena、Christopher Olah、Jonathon Shlens散文《Coditional Image Synthesis
with Auxiliary Classifier GANs》。
赞助分类器生成式对抗网络(auxiliary classifier GAN,AC-GAN)落成。

生成式对抗网络利用。生成数字,生中年人脸图像。

生成式对抗互联网完成。https://github.com/fchollet/keras/blob/master/examples/mnist\_acgan.py

Augustus Odena、Chistopher Olah和Jonathon Shlens 论文《Conditional Image
Synthesis With Auxiliary Classifier GANs》。
通过噪声,让变化模型G生成虚假数据,和真正数据一同送到判定模型D,判断模型一方面输出数据真/假,一方面输出图片分类。
第一定义生成模型,目标是生成一对(z,L)数据,z是噪声向量,L是(1,28,28)的图像空间。

def build_generator(latent_size):
cnn = Sequential()
cnn.add(Dense(1024, input_dim=latent_size, activation=’relu’))
cnn.add(Dense(128 * 7 * 7, activation=’relu’))
cnn.add(Reshape((128, 7, 7)))
#上采集样品,图你尺寸变为 14X14
cnn.add(UpSampling2D(size=(2,2)))
cnn.add(Convolution2D(256, 5, 5, border_mode=’same’, activation=’relu’,
init=’glorot_normal’))
#上采集样品,图像尺寸变为28X28
cnn.add(UpSampling2D(size=(2,2)))
cnn.add(Convolution2D(128, 5, 5, border_mode=’same’, activation=’relu’,
init=’glorot_normal’))
#规约到1个通道
cnn.add(Convolution2D(1, 2, 2, border_mode=’same’, activation=’tanh’,
init=’glorot_normal’))
#浮动模型输入层,特征向量
latent = Input(shape=(latent_size, ))
#变迁模型输入层,标识
image_class = Input(shape=(1,), dtype=’int32′)
cls = Flatten()(Embedding(10, latent_size,
init=’glorot_normal’)(image_class))
h = merge([latent, cls], mode=’mul’)
fake_image = cnn(h) #出口虚假图片
return Model(input=[latent, image_class], output=fake_image)
概念判定模型,输入(1,28,28)图片,输出五个值,贰个是识别模型以为那张图片是还是不是是虚假图片,另三个是甄别模型以为这第图片所属分类。

def build_discriminator();
#使用激活函数Leaky ReLU来替换标准的卷积神经网络中的激活函数
cnn = Wequential()
cnn.add(Convolution2D(32, 3, 3, border_mode=’same’, subsample=(2, 2),
input_shape=(1, 28, 28)))
cnn.add(LeakyReLU())
cnn.add(Dropout(0.3))
cnn.add(Convolution2D(64, 3, 3, border_mode=’same’, subsample=(1,
1)))
cnn.add(LeakyReLU())
cnn.add(Dropout(0.3))
cnn.add(Convolution2D(128, 3, 3, border_mode=’same’, subsample=(1,
1)))
cnn.add(LeakyReLU())
cnn.add(Dropout(0.3))
cnn.add(Convolution2D(256, 3, 3, border_mode=’same’, subsample=(1,
1)))
cnn.add(LeakyReLU())
cnn.add(Dropout(0.3))
cnn.add(Flatten())
image = Input(shape=(1, 28, 28))
features = cnn(image)
#有三个出口
#出口真假值,范围在0~1
fake = Dense(1, activation=’sigmoid’,name=’generation’)(features)
#帮扶分类器,输出图片分类
aux = Dense(10, activation=’softmax’, name=’auxiliary’)(features)
return Model(input=image, output=[fake, aux])
锻练进度,50轮(epoch),把权重保存,每轮把假冒伪劣数据生成图处保存,观望虚假数据演化进程。

if __name__ ==’__main__’:
#概念超参数
nb_epochs = 50
batch_size = 100
latent_size = 100
#优化器学习率
adam_lr = 0.0002
adam_beta_l = 0.5
#创设决断网络
discriminator = build_discriminator()
discriminator.compile(optimizer=adam(lr=adam_lr,
beta_l=adam_beta_l), loss=’binary_crossentropy’)
latent = Input(shape=(lastent_size, ))
image_class = Input(shape-(1, ), dtype=’int32′)
#变动组合模型
discriminator.trainable = False
fake, aux = discriminator(fake)
combined = Model(input=[latent, image_class], output=[fake, aux])
combined.compile(optimizer=Adam(lr=adam_lr, beta_l=adam_beta_1),
loss=[‘binary_crossentropy’, ‘sparse_categorical_crossentropy’])
#将mnist数据转载为(…,1,28,28)维度,取值范围为[-1,1]
(X_train,y_train),(X_test,y_test) = mnist.load_data()
X_train = (X_train.astype(np.float32) – 127.5) / 127.5
X_train = np.expand_dims(X_train, axis=1)
X_test = (X_test.astype(np.float32) – 127.5) / 127.5
X_test = np.expand_dims(X_test, axis=1)
num_train, num_test = X_train.shape[0], X_test.shape[0]
train_history = defaultdict(list)
test_history = defaultdict(list)
for epoch in range(epochs):
print(‘Epoch {} of {}’.format(epoch + 1, epochs))
num_batches = int(X_train.shape[0] / batch_size)
progress_bar = Progbar(target=num_batches)
epoch_gen_loss = []
epoch_disc_loss = []
for index in range(num_batches):
progress_bar.update(index)
#产生叁个批次的噪音数据
noise = np.random.uniform(-1, 1, (batch_size, latent_size))
# 获取一个批次的真实数据
image_batch = X_train[index * batch_size:(index + 1) *
batch_size]
label_batch = y_train[index * batch_size:(index + 1) *
batch_size]
# 生成一些噪音标志
sampled_labels = np.random.randint(0, 10, batch_size)
# 产生叁个批次的仿真图片
generated_images = generator.predict(
[noise, sampled_labels.reshape((-1, 1))], verbose=0)
X = np.concatenate((image_batch, generated_images))
y = np.array([1] * batch_size + [0] * batch_size)
aux_y = np.concatenate((label_batch, sampled_labels), axis=0)
epoch_disc_loss.append(discriminator.train_on_batch(X, [y,
aux_y]))
# 产生四个批次噪声和标识
noise = np.random.uniform(-1, 1, (2 * batch_size, latent_size))
sampled_labels = np.random.randint(0, 10, 2 * batch_size)
# 练习转换模型来掩人耳目决断模型,输出真/假都设为真
trick = np.ones(2 * batch_size)
epoch_gen_loss.append(combined.train_on_batch(
[noise, sampled_labels.reshape((-1, 1))],
[trick, sampled_labels]))
print(‘\nTesting for epoch {}:’.format(epoch + 1))
# 评估测量试验集,发生三个新批次噪声数据
noise = np.random.uniform(-1, 1, (num_test, latent_size))
sampled_labels = np.random.randint(0, 10, num_test)
generated_images = generator.predict(
[noise, sampled_labels.reshape((-1, 1))], verbose=False)
X = np.concatenate((X_test, generated_images))
y = np.array([1] * num_test + [0] * num_test)
aux_y = np.concatenate((y_test, sampled_labels), axis=0)
# 剖断模型是还是不是能识别
discriminator_test_loss = discriminator.evaluate(
X, [y, aux_y], verbose=False)
discriminator_train_loss = np.mean(np.array(epoch_disc_loss),
axis=0)
# 创立五个批次新噪声数据
noise = np.random.uniform(-1, 1, (2 * num_test, latent_size))
sampled_labels = np.random.randint(0, 10, 2 * num_test)
trick = np.ones(2 * num_test)
generator_test_loss = combined.evaluate(
[noise, sampled_labels.reshape((-1, 1))],
[trick, sampled_labels], verbose=False)
generator_train_loss = np.mean(np.array(epoch_gen_loss), axis=0)
# 损失值等质量指标识录下来,并出口
train_history[‘generator’].append(generator_train_loss)
train_history[‘discriminator’].append(discriminator_train_loss)
test_history[‘generator’].append(generator_test_loss)
test_history[‘discriminator’].append(discriminator_test_loss)
print(‘{0:<22s} | {1:4s} | {2:15s} | {3:5s}’.format(
‘component’, *discriminator.metrics_names))
print(‘-‘ * 65)
ROW_FMT = ‘{0:<22s} | {1:<4.2f} | {2:<15.2f} | {3:<5.2f}’
print(ROW_FMT.format(‘generator (train)’,
*train_history[‘generator’][-1]))
print(ROW_FMT.format(‘generator (test)’,
*test_history[‘generator’][-1]))
print(ROW_FMT.format(‘discriminator (train)’,
*train_history[‘discriminator’][-1]))
print(ROW_FMT.format(‘discriminator (test)’,
*test_history[‘discriminator’][-1]))
# 各类epoch保存叁次权重
generator.save_weights(
‘params_generator_epoch_{0:03d}.hdf5’.format(epoch), True)
discriminator.save_weights(
‘params_discriminator_epoch_{0:03d}.hdf5’.format(epoch), True)
# 生成一些可视化虚假数字看演化进度
noise = np.random.uniform(-1, 1, (100, latent_size))
sampled_labels = np.array([
[i] * 10 for i in range(10)
]).reshape(-1, 1)
generated_images = generator.predict(
[noise, sampled_labels], verbose=0)
# 整理到一个方格
img = (np.concatenate([r.reshape(-1, 28)
for r in np.split(generated_images, 10)
], axis=-1) * 127.5 + 127.5).astype(np.uint8)
Image.fromarray(img).save(
‘plot_epoch_{0:03d}_generated.png’.format(epoch))
pickle.dump({‘train’: train_history, ‘test’: test_history},
open(‘acgan-history.pkl’, ‘wb’))

教练截止,创立3类文件。params_discriminator_epoch_{{epoch_number}}.hdf5,判定模型权重参数。params_generator_epoch_{{epoch_number}}.hdf5,生成模型权重参数。plot_epoch_{{epoch_number}}_generated.png

生成式对抗互连网创新。生成式对抗网络(generative adversarial
network,GAN)在无监督学习十一分实惠。常规生成式对抗互联网判断器使用Sigmoid交叉熵损失函数,学习进度梯度消失。Wasserstein生成式对抗互联网(Wasserstein
generative adversarial
network,WGAN),使用Wasserstein距离衡量,实际不是Jensen-Shannon散度(延森-Shannon
divergence,JSD)。使用最小二乘生成式对抗网络(least squares generative
adversarial network,LSGAN),判定模型用异常的小平方损失小函数(least squares
loss function)。塞BathTyne Nowozin、Botond Cseke、Ryota
汤姆ioka诗歌《f-GAN: Training Generative Neural 萨姆plers using
Variational Divergence Minimization》。

仿效资料:
《TensorFlow技艺解析与实战》

接待付费咨询(150元每小时),我的微信:qingxingfengzi

相关文章