运行生成文件
效果展示
plot_epoch_073_generated（图片文件）
# -*- coding: utf-8 -*- """ Train an Auxiliary Classifier Generative Adversarial Network (ACGAN) on the MNIST dataset. See https://arxiv.org/abs/1610.09585 for more details. 基于MINIST数据集训练辅助分类器ACGAN（生成对抗网络） 详情见：https://arxiv.org/abs/1610.09585 You should start to see reasonable images after ~5 epochs, and good images by ~15 epochs. You should use a GPU, as the convolution-heavy operations are very slow on the CPU. Prefer the TensorFlow backend if you plan on iterating, as the compilation time can be a blocker using Theano. 5周期后，开始看到合理的图像，大约15周期后，看到合理的图像。 应该使用GPU，因为卷积运算在CPU上的速度非常慢。如果计划迭代，则首选TensorFlow后端， 因为使用Theano编译时速度慢。 Timings: Hardware | Backend | Time / Epoch ------------------------------------------- CPU | TF | 3 hrs Titan X (maxwell) | TF | 4 min Titan X (maxwell) | TH | 7 min Consult https://github.com/lukedeo/keras-acgan for more information and example output 更多信息和示例输出，请参阅https://github.com/lukedeo/keras-acgan """ from __future__ import print_function from collections import defaultdict try: import cPickle as pickle except ImportError: import pickle from PIL import Image from six.moves import range from keras.datasets import mnist from keras import layers from keras.layers import Input, Dense, Reshape, Flatten, Embedding, Dropout from keras.layers import BatchNormalization from keras.layers.advanced_activations import LeakyReLU from keras.layers.convolutional import Conv2DTranspose, Conv2D from keras.models import Sequential, Model from keras.optimizers import Adam from keras.utils.generic_utils import Progbar import numpy as np np.random.seed(1337) num_classes = 10 def build_generator(latent_size): # we will map a pair of (z, L), where z is a latent vector and L is a # label drawn from P_c, to image space (..., 28, 28, 1) #映射一对(z, L)，其中z是一个隐矢，L是P_c绘制的标签。，到图像空间（…，28, 28, 1） cnn = Sequential() cnn.add(Dense(3 * 3 * 384, input_dim=latent_size, activation='relu')) cnn.add(Reshape((3, 3, 384))) # upsample to (7, 7, ...) # 上采样到（7, 7，…） cnn.add(Conv2DTranspose(192, 5, strides=1, padding='valid', activation='relu', kernel_initializer='glorot_normal')) cnn.add(BatchNormalization()) # upsample to (14, 14, ...) # 上采样到(14, 14, ...) cnn.add(Conv2DTranspose(96, 5, strides=2, padding='same', activation='relu', kernel_initializer='glorot_normal')) cnn.add(BatchNormalization()) # upsample to (28, 28, ...) # 上采样到(28, 28, ...) cnn.add(Conv2DTranspose(1, 5, strides=2, padding='same', activation='tanh', kernel_initializer='glorot_normal')) # this is the z space commonly referred to in GAN papers # GAN文献中常用的Z空间。 latent = Input(shape=(latent_size, )) # this will be our label # 标签 image_class = Input(shape=(1,), dtype='int32') cls = Flatten()(Embedding(num_classes, latent_size, embeddings_initializer='glorot_normal')(image_class)) # hadamard product between z-space and a class conditional embedding # z-space与条件嵌入的Hadamard积 h = layers.multiply([latent, cls]) fake_image = cnn(h) return Model([latent, image_class], fake_image) def build_discriminator(): # build a relatively standard conv net, with LeakyReLUs as suggested in # the reference paper # 根据论文，建立一个相对标准的conv网络（卷积网络） cnn = Sequential() cnn.add(Conv2D(32, 3, padding='same', strides=2, input_shape=(28, 28, 1))) cnn.add(LeakyReLU(0.2)) cnn.add(Dropout(0.3)) cnn.add(Conv2D(64, 3, padding='same', strides=1)) cnn.add(LeakyReLU(0.2)) cnn.add(Dropout(0.3)) cnn.add(Conv2D(128, 3, padding='same', strides=2)) cnn.add(LeakyReLU(0.2)) cnn.add(Dropout(0.3)) cnn.add(Conv2D(256, 3, padding='same', strides=1)) cnn.add(LeakyReLU(0.2)) cnn.add(Dropout(0.3)) cnn.add(Flatten()) image = Input(shape=(28, 28, 1)) features = cnn(image) # first output (name=generation) is whether or not the discriminator # thinks the image that is being shown is fake, and the second output # (name=auxiliary) is the class that the discriminator thinks the image # belongs to. # 第一输出（name=generation）是鉴别器是否认为所显示的图像是假的，而第二输出（名称＝辅助）是鉴别器认为图像属于的类。 fake = Dense(1, activation='sigmoid', name='generation')(features) aux = Dense(num_classes, activation='softmax', name='auxiliary')(features) return Model(image, [fake, aux]) if __name__ == '__main__': # batch and latent size taken from the paper # 来自论文的批次和尺寸 epochs = 100 batch_size = 100 latent_size = 100 # Adam parameters suggested in https://arxiv.org/abs/1511.06434 # Adm参数详见：https://arxiv.org/abs/1511.06434 adam_lr = 0.0002 adam_beta_1 = 0.5 # build the discriminator # 创建判别器 print('Discriminator model:') discriminator = build_discriminator() discriminator.compile( optimizer=Adam(lr=adam_lr, beta_1=adam_beta_1), loss=['binary_crossentropy', 'sparse_categorical_crossentropy'] ) discriminator.summary() # build the generator # 创建生成器 generator = build_generator(latent_size) latent = Input(shape=(latent_size, )) image_class = Input(shape=(1,), dtype='int32') # get a fake image # 获取赝品图片 fake = generator([latent, image_class]) # we only want to be able to train generation for the combined model # 为组合模型训练生成器。 discriminator.trainable = False fake, aux = discriminator(fake) combined = Model([latent, image_class], [fake, aux]) print('Combined model:') combined.compile( optimizer=Adam(lr=adam_lr, beta_1=adam_beta_1), loss=['binary_crossentropy', 'sparse_categorical_crossentropy'] ) combined.summary() # get our mnist data, and force it to be of shape (..., 28, 28, 1) with # range [-1, 1] # 获取MNIST数据集，并强制调整它的形状为（…，28, 28, 1）。取值范围[-1, 1] (x_train, y_train), (x_test, y_test) = mnist.load_data() x_train = (x_train.astype(np.float32) - 127.5) / 127.5 # 图像像素取值范围[0,255]，以其1/2作为基准 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(1, epochs + 1): print('Epoch {}/{}'.format(epoch, epochs)) num_batches = int(x_train.shape[0] / batch_size) progress_bar = Progbar(target=num_batches) # we don't want the discriminator to also maximize the classification # accuracy of the auxiliary classifier on generated images, so we # don't train discriminator to produce class labels for generated # images (see https://openreview.net/forum?id=rJXTf9Bxg). # 我们不希望鉴别器也最大化辅助分类器在生成图像上的分类精度，所以我们不训练鉴别器来产生生成图像的类标签 # （详见：https://openreview.net/forum?id=rJXTf9Bxg） # To preserve sum of sample weights for the auxiliary classifier, # we assign sample weight of 2 to the real images. # 为了保持辅助分类器的样本权重之和，分配2个相同的实际图像的样本权重。 disc_sample_weight = [np.ones(2 * batch_size), np.concatenate((np.ones(batch_size) * 2, np.zeros(batch_size)))] epoch_gen_loss = [] epoch_disc_loss = [] for index in range(num_batches): # generate a new batch of noise # 产生新一批噪声 noise = np.random.uniform(-1, 1, (batch_size, latent_size)) # get a batch of real images # 获取一批真实图像 image_batch = x_train[index * batch_size:(index + 1) * batch_size] label_batch = y_train[index * batch_size:(index + 1) * batch_size] # sample some labels from p_c # 从p_c取样标签 sampled_labels = np.random.randint(0, num_classes, batch_size) # generate a batch of fake images, using the generated labels as a # conditioner. We reshape the sampled labels to be # (batch_size, 1) so that we can feed them into the embedding # layer as a length one sequence # 生成一批假图像，使用生成的标签作为调节器。我们将采样标签整形为（batch_size, 1）， # 这样我们就可以把它们作为长度一个序列输入到嵌入层中。 generated_images = generator.predict( [noise, sampled_labels.reshape((-1, 1))], verbose=0) x = np.concatenate((image_batch, generated_images)) # use one-sided soft real/fake labels # 使用单面软真/假标签 # Salimans et al., 2016 # https://arxiv.org/pdf/1606.03498.pdf (Section 3.4) soft_zero, soft_one = 0, 0.95 y = np.array([soft_one] * batch_size + [soft_zero] * batch_size) aux_y = np.concatenate((label_batch, sampled_labels), axis=0) # see if the discriminator can figure itself out... # 看看鉴别器是否能算出它自己… epoch_disc_loss.append(discriminator.train_on_batch( x, [y, aux_y], sample_weight=disc_sample_weight)) # make new noise. we generate 2 * batch size here such that we have # the generator optimize over an identical number of images as the # discriminator # 制造新的噪音。产生 * batch大小在这里，这样我们有生成器优化在相同数量的图像作为鉴别器 noise = np.random.uniform(-1, 1, (2 * batch_size, latent_size)) sampled_labels = np.random.randint(0, num_classes, 2 * batch_size) # we want to train the generator to trick the discriminator # For the generator, we want all the {fake, not-fake} labels to say # not-fake # 训练生成器来欺骗鉴别器，希望鉴别器对所有的{假，不假}标签都说不是假的。 trick = np.ones(2 * batch_size) * soft_one epoch_gen_loss.append(combined.train_on_batch( [noise, sampled_labels.reshape((-1, 1))], [trick, sampled_labels])) progress_bar.update(index + 1) print('Testing for epoch {}:'.format(epoch)) # evaluate the testing loss here # 评估测试损失 # generate a new batch of noise # 产生新一批噪声 noise = np.random.uniform(-1, 1, (num_test, latent_size)) # sample some labels from p_c and generate images from them # 从p_c中采样一些标签并根据标签生成图像 sampled_labels = np.random.randint(0, num_classes, 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) # see if the discriminator can figure itself out... # 看看鉴别器是否能算出… discriminator_test_loss = discriminator.evaluate( x, [y, aux_y], verbose=False) discriminator_train_loss = np.mean(np.array(epoch_disc_loss), axis=0) # make new noise # 制造新噪音 noise = np.random.uniform(-1, 1, (2 * num_test, latent_size)) sampled_labels = np.random.randint(0, num_classes, 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) # generate an epoch report on performance # 生成有关性能的周期性报告 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:<15.4f} | {3:<5.4f}' 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])) # save weights every epoch # 保存每个周期的权重 generator.save_weights( 'params_generator_epoch_{0:03d}.hdf5'.format(epoch), True) discriminator.save_weights( 'params_discriminator_epoch_{0:03d}.hdf5'.format(epoch), True) # generate some digits to display # 生成一些数字显示 num_rows = 40 noise = np.tile(np.random.uniform(-1, 1, (num_rows, latent_size)), (num_classes, 1)) sampled_labels = np.array([ [i] * num_rows for i in range(num_classes) ]).reshape(-1, 1) # get a batch to display # 获得批量显示 generated_images = generator.predict( [noise, sampled_labels], verbose=0) # prepare real images sorted by class label # 准备按类标签排序的真实图像 real_labels = y_train[(epoch - 1) * num_rows * num_classes: epoch * num_rows * num_classes] indices = np.argsort(real_labels, axis=0) real_images = x_train[(epoch - 1) * num_rows * num_classes: epoch * num_rows * num_classes][indices] # display generated images, white separator, real images # 显示生成的图像，白色分离器，真实图像 img = np.concatenate( (generated_images, np.repeat(np.ones_like(x_train[:1]), num_rows, axis=0), real_images)) # arrange them into a grid # 将它们排列成网格 img = (np.concatenate([r.reshape(-1, 28) for r in np.split(img, 2 * num_classes + 1) ], axis=-1) * 127.5 + 127.5).astype(np.uint8) Image.fromarray(img).save( 'plot_epoch_{0:03d}_generated.png'.format(epoch)) with open('acgan-history.pkl', 'wb') as f: pickle.dump({'train': train_history, 'test': test_history}, f)C:\ProgramData\Anaconda3\python.exe E:/keras-master/examples/mnist_acgan.py Using TensorFlow backend. Discriminator model: __________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ================================================================================================== input_1 (InputLayer) (None, 28, 28, 1) 0 __________________________________________________________________________________________________ sequential_1 (Sequential) (None, 12544) 387840 input_1[0][0] __________________________________________________________________________________________________ generation (Dense) (None, 1) 12545 sequential_1[1][0] __________________________________________________________________________________________________ auxiliary (Dense) (None, 10) 125450 sequential_1[1][0] ================================================================================================== Total params: 525,835 Trainable params: 525,835 Non-trainable params: 0 __________________________________________________________________________________________________ Combined model: __________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ================================================================================================== input_4 (InputLayer) (None, 100) 0 __________________________________________________________________________________________________ input_5 (InputLayer) (None, 1) 0 __________________________________________________________________________________________________ model_2 (Model) (None, 28, 28, 1) 2657897 input_4[0][0] input_5[0][0] __________________________________________________________________________________________________ model_1 (Model) [(None, 1), (None, 1 525835 model_2[1][0] ================================================================================================== Total params: 3,183,732 Trainable params: 2,657,321 Non-trainable params: 526,411 __________________________________________________________________________________________________ Epoch 1/100 1/600 [..............................] - ETA: 45:16 2/600 [..............................] - ETA: 27:37 3/600 [..............................] - ETA: 19:05 4/600 [..............................] - ETA: 14:50 5/600 [..............................] - ETA: 12:17 6/600 [..............................] - ETA: 10:33 7/600 [..............................] - ETA: 9:20 8/600 [..............................] - ETA: 8:25 9/600 [..............................] - ETA: 7:42 10/600 [..............................] - ETA: 7:08 11/600 [..............................] - ETA: 6:41 12/600 [..............................] - ETA: 6:17 13/600 [..............................] - ETA: 5:57 14/600 [..............................] - ETA: 5:40 15/600 [..............................] - ETA: 5:24 16/600 [..............................] - ETA: 5:13 17/600 [..............................] - ETA: 5:02 18/600 [..............................] - ETA: 4:51 19/600 [..............................] - ETA: 4:42 20/600 [>.............................] - ETA: 4:34 21/600 [>.............................] - ETA: 4:26 22/600 [>.............................] - ETA: 4:19 23/600 [>.............................] - ETA: 4:12 24/600 [>.............................] - ETA: 4:07 25/600 [>.............................] - ETA: 4:01 26/600 [>.............................] - ETA: 3:56 27/600 [>.............................] - ETA: 3:51 28/600 [>.............................] - ETA: 3:51 29/600 [>.............................] - ETA: 3:47 30/600 [>.............................] - ETA: 3:43 31/600 [>.............................] - ETA: 3:39 32/600 [>.............................] - ETA: 3:37 33/600 [>.............................] - ETA: 3:34 34/600 [>.............................] - ETA: 3:31 35/600 [>.............................] - ETA: 3:28 36/600 [>.............................] - ETA: 3:25 37/600 [>.............................] - ETA: 3:23 38/600 [>.............................] - ETA: 3:20 39/600 [>.............................] - ETA: 3:18 40/600 [=>............................] - ETA: 3:16 41/600 [=>............................] - ETA: 3:14 42/600 [=>............................] - ETA: 3:12 43/600 [=>............................] - ETA: 3:11 44/600 [=>............................] - ETA: 3:09 45/600 [=>............................] - ETA: 3:07 46/600 [=>............................] - ETA: 3:06 47/600 [=>............................] - ETA: 3:04 48/600 [=>............................] - ETA: 3:02 49/600 [=>............................] - ETA: 3:01 50/600 [=>............................] - ETA: 2:59 51/600 [=>............................] - ETA: 2:58 52/600 [=>............................] - ETA: 2:56 53/600 [=>............................] - ETA: 2:55 54/600 [=>............................] - ETA: 2:53 55/600 [=>............................] - ETA: 2:52 56/600 [=>............................] - ETA: 2:51 57/600 [=>............................] - ETA: 2:49 58/600 [=>............................] - ETA: 2:48 59/600 [=>............................] - ETA: 2:47 60/600 [==>...........................] - ETA: 2:46 61/600 [==>...........................] - ETA: 2:44 62/600 [==>...........................] - ETA: 2:43 63/600 [==>...........................] - ETA: 2:42 64/600 [==>...........................] - ETA: 2:41 65/600 [==>...........................] - ETA: 2:40 66/600 [==>...........................] - ETA: 2:39 67/600 [==>...........................] - ETA: 2:38 68/600 [==>...........................] - ETA: 2:37 69/600 [==>...........................] - ETA: 2:36 70/600 [==>...........................] - ETA: 2:35 71/600 [==>...........................] - ETA: 2:35 72/600 [==>...........................] - ETA: 2:34 73/600 [==>...........................] - ETA: 2:33 74/600 [==>...........................] - ETA: 2:32 75/600 [==>...........................] - ETA: 2:31 76/600 [==>...........................] - ETA: 2:30 77/600 [==>...........................] - ETA: 2:29 78/600 [==>...........................] - ETA: 2:29 79/600 [==>...........................] - ETA: 2:28 80/600 [===>..........................] - ETA: 2:27 81/600 [===>..........................] - ETA: 2:26 82/600 [===>..........................] - ETA: 2:26 83/600 [===>..........................] - ETA: 2:25 84/600 [===>..........................] - ETA: 2:24 85/600 [===>..........................] - ETA: 2:24 86/600 [===>..........................] - ETA: 2:23 87/600 [===>..........................] - ETA: 2:22 88/600 [===>..........................] - ETA: 2:22 89/600 [===>..........................] - ETA: 2:21 90/600 [===>..........................] - ETA: 2:20 91/600 [===>..........................] - ETA: 2:20 92/600 [===>..........................] - ETA: 2:19 93/600 [===>..........................] - ETA: 2:18 94/600 [===>..........................] - ETA: 2:18 95/600 [===>..........................] - ETA: 2:17 96/600 [===>..........................] - ETA: 2:17 97/600 [===>..........................] - ETA: 2:16 98/600 [===>..........................] - ETA: 2:16 99/600 [===>..........................] - ETA: 2:15 100/600 [====>.........................] - ETA: 2:15 101/600 [====>.........................] - ETA: 2:14 102/600 [====>.........................] - ETA: 2:13 103/600 [====>.........................] - ETA: 2:13 104/600 [====>.........................] - ETA: 2:13 105/600 [====>.........................] - ETA: 2:12 106/600 [====>.........................] - ETA: 2:11 107/600 [====>.........................] - ETA: 2:11 108/600 [====>.........................] - ETA: 2:10 109/600 [====>.........................] - ETA: 2:10 110/600 [====>.........................] - ETA: 2:09 111/600 [====>.........................] - ETA: 2:09 112/600 [====>.........................] - ETA: 2:08 113/600 [====>.........................] - ETA: 2:08 114/600 [====>.........................] - ETA: 2:07 115/600 [====>.........................] - ETA: 2:07 116/600 [====>.........................] - ETA: 2:06 117/600 [====>.........................] - ETA: 2:06 118/600 [====>.........................] - ETA: 2:06 119/600 [====>.........................] - ETA: 2:05 120/600 [=====>........................] - ETA: 2:05 121/600 [=====>........................] - ETA: 2:04 122/600 [=====>........................] - ETA: 2:04 123/600 [=====>........................] - ETA: 2:03 124/600 [=====>........................] - ETA: 2:03 125/600 [=====>........................] - ETA: 2:02 126/600 [=====>........................] - ETA: 2:02 127/600 [=====>........................] - ETA: 2:02 128/600 [=====>........................] - ETA: 2:01 129/600 [=====>........................] - ETA: 2:01 130/600 [=====>........................] - ETA: 2:00 131/600 [=====>........................] - ETA: 2:00 132/600 [=====>........................] - ETA: 2:00 133/600 [=====>........................] - ETA: 1:59 134/600 [=====>........................] - ETA: 1:59 135/600 [=====>........................] - ETA: 1:58 136/600 [=====>........................] - ETA: 1:58 137/600 [=====>........................] - ETA: 1:58 138/600 [=====>........................] - ETA: 1:57 139/600 [=====>........................] - ETA: 1:57 140/600 [======>.......................] - ETA: 1:57 141/600 [======>.......................] - ETA: 1:56 142/600 [======>.......................] - ETA: 1:56 143/600 [======>.......................] - ETA: 1:55 144/600 [======>.......................] - ETA: 1:55 145/600 [======>.......................] - ETA: 1:55 146/600 [======>.......................] - ETA: 1:54 147/600 [======>.......................] - ETA: 1:54 148/600 [======>.......................] - ETA: 1:54 149/600 [======>.......................] - ETA: 1:53 150/600 [======>.......................] - ETA: 1:54 151/600 [======>.......................] - ETA: 1:53 152/600 [======>.......................] - ETA: 1:53 153/600 [======>.......................] - ETA: 1:54 154/600 [======>.......................] - ETA: 1:54 155/600 [======>.......................] - ETA: 1:53 156/600 [======>.......................] - ETA: 1:53 157/600 [======>.......................] - ETA: 1:52 158/600 [======>.......................] - ETA: 1:52 159/600 [======>.......................] - ETA: 1:52 160/600 [=======>......................] - ETA: 1:51 161/600 [=======>......................] - ETA: 1:51 162/600 [=======>......................] - ETA: 1:51 163/600 [=======>......................] - ETA: 1:50 164/600 [=======>......................] - ETA: 1:50 165/600 [=======>......................] - ETA: 1:50 166/600 [=======>......................] - ETA: 1:49 167/600 [=======>......................] - ETA: 1:49 168/600 [=======>......................] - ETA: 1:48 169/600 [=======>......................] - ETA: 1:48 170/600 [=======>......................] - ETA: 1:48 171/600 [=======>......................] - ETA: 1:47 172/600 [=======>......................] - ETA: 1:47 173/600 [=======>......................] - ETA: 1:47 174/600 [=======>......................] - ETA: 1:46 175/600 [=======>......................] - ETA: 1:46 176/600 [=======>......................] - ETA: 1:46 177/600 [=======>......................] - ETA: 1:45 178/600 [=======>......................] - ETA: 1:45 179/600 [=======>......................] - ETA: 1:45 180/600 [========>.....................] - ETA: 1:44 181/600 [========>.....................] - ETA: 1:44 182/600 [========>.....................] - ETA: 1:44 183/600 [========>.....................] - ETA: 1:43 184/600 [========>.....................] - ETA: 1:43 185/600 [========>.....................] - ETA: 1:43 186/600 [========>.....................] - ETA: 1:42 187/600 [========>.....................] - ETA: 1:42 188/600 [========>.....................] - ETA: 1:42 189/600 [========>.....................] - ETA: 1:41 190/600 [========>.....................] - ETA: 1:41 191/600 [========>.....................] - ETA: 1:41 192/600 [========>.....................] - ETA: 1:40 193/600 [========>.....................] - ETA: 1:40 194/600 [========>.....................] - ETA: 1:40 195/600 [========>.....................] - ETA: 1:39 196/600 [========>.....................] - ETA: 1:39 197/600 [========>.....................] - ETA: 1:39 198/600 [========>.....................] - ETA: 1:38 199/600 [========>.....................] - ETA: 1:38 200/600 [=========>....................] - ETA: 1:38 201/600 [=========>....................] - ETA: 1:37 202/600 [=========>....................] - ETA: 1:37 203/600 [=========>....................] - ETA: 1:37 204/600 [=========>....................] - ETA: 1:37 205/600 [=========>....................] - ETA: 1:36 206/600 [=========>....................] - ETA: 1:36 207/600 [=========>....................] - ETA: 1:36 208/600 [=========>....................] - ETA: 1:35 209/600 [=========>....................] - ETA: 1:35 210/600 [=========>....................] - ETA: 1:35 211/600 [=========>....................] - ETA: 1:34 212/600 [=========>....................] - ETA: 1:34 213/600 [=========>....................] - ETA: 1:34 214/600 [=========>....................] - ETA: 1:34 215/600 [=========>....................] - ETA: 1:33 216/600 [=========>....................] - ETA: 1:33 217/600 [=========>....................] - ETA: 1:33 218/600 [=========>....................] - ETA: 1:32 219/600 [=========>....................] - ETA: 1:32 220/600 [==========>...................] - ETA: 1:32 221/600 [==========>...................] - ETA: 1:31 222/600 [==========>...................] - ETA: 1:31 223/600 [==========>...................] - ETA: 1:31 224/600 [==========>...................] - ETA: 1:31 225/600 [==========>...................] - ETA: 1:30 226/600 [==========>...................] - ETA: 1:30 227/600 [==========>...................] - ETA: 1:30 228/600 [==========>...................] - ETA: 1:30 229/600 [==========>...................] - ETA: 1:29 230/600 [==========>...................] - ETA: 1:29 231/600 [==========>...................] - ETA: 1:29 232/600 [==========>...................] - ETA: 1:28 233/600 [==========>...................] - ETA: 1:28 234/600 [==========>...................] - ETA: 1:28 235/600 [==========>...................] - ETA: 1:28 236/600 [==========>...................] - ETA: 1:27 237/600 [==========>...................] - ETA: 1:27 238/600 [==========>...................] - ETA: 1:27 239/600 [==========>...................] - ETA: 1:26 240/600 [===========>..................] - ETA: 1:26 241/600 [===========>..................] - ETA: 1:26 242/600 [===========>..................] - ETA: 1:26 243/600 [===========>..................] - ETA: 1:25 244/600 [===========>..................] - ETA: 1:25 245/600 [===========>..................] - ETA: 1:25 246/600 [===========>..................] - ETA: 1:24 247/600 [===========>..................] - ETA: 1:24 248/600 [===========>..................] - ETA: 1:24 249/600 [===========>..................] - ETA: 1:24 250/600 [===========>..................] - ETA: 1:23 251/600 [===========>..................] - ETA: 1:23 252/600 [===========>..................] - ETA: 1:23 253/600 [===========>..................] - ETA: 1:22 254/600 [===========>..................] - ETA: 1:22 255/600 [===========>..................] - ETA: 1:22 256/600 [===========>..................] - ETA: 1:22 257/600 [===========>..................] - ETA: 1:21 258/600 [===========>..................] - ETA: 1:21 259/600 [===========>..................] - ETA: 1:21 260/600 [============>.................] - ETA: 1:20 261/600 [============>.................] - ETA: 1:20 262/600 [============>.................] - ETA: 1:20 263/600 [============>.................] - ETA: 1:20 264/600 [============>.................] - ETA: 1:19 265/600 [============>.................] - ETA: 1:19 266/600 [============>.................] - ETA: 1:19 267/600 [============>.................] - ETA: 1:19 268/600 [============>.................] - ETA: 1:18 269/600 [============>.................] - ETA: 1:18 270/600 [============>.................] - ETA: 1:18 271/600 [============>.................] - ETA: 1:18 272/600 [============>.................] - ETA: 1:17 273/600 [============>.................] - ETA: 1:17 274/600 [============>.................] - ETA: 1:17 275/600 [============>.................] - ETA: 1:17 276/600 [============>.................] - ETA: 1:16 277/600 [============>.................] - ETA: 1:16 278/600 [============>.................] - ETA: 1:16 279/600 [============>.................] - ETA: 1:15 280/600 [=============>................] - ETA: 1:15 281/600 [=============>................] - ETA: 1:15 282/600 [=============>................] - ETA: 1:15 283/600 [=============>................] - ETA: 1:14 284/600 [=============>................] - ETA: 1:14 285/600 [=============>................] - ETA: 1:14 286/600 [=============>................] - ETA: 1:14 287/600 [=============>................] - ETA: 1:13 288/600 [=============>................] - ETA: 1:13 289/600 [=============>................] - ETA: 1:13 290/600 [=============>................] - ETA: 1:13 291/600 [=============>................] - ETA: 1:12 292/600 [=============>................] - ETA: 1:12 293/600 [=============>................] - ETA: 1:12 294/600 [=============>................] - ETA: 1:12 295/600 [=============>................] - ETA: 1:12 296/600 [=============>................] - ETA: 1:11 297/600 [=============>................] - ETA: 1:11 298/600 [=============>................] - ETA: 1:11 299/600 [=============>................] - ETA: 1:11 300/600 [==============>...............] - ETA: 1:10 301/600 [==============>...............] - ETA: 1:10 302/600 [==============>...............] - ETA: 1:10 303/600 [==============>...............] - ETA: 1:09 304/600 [==============>...............] - ETA: 1:09 305/600 [==============>...............] - ETA: 1:09 306/600 [==============>...............] - ETA: 1:09 307/600 [==============>...............] - ETA: 1:09 308/600 [==============>...............] - ETA: 1:08 309/600 [==============>...............] - ETA: 1:08 310/600 [==============>...............] - ETA: 1:08 311/600 [==============>...............] - ETA: 1:08 312/600 [==============>...............] - ETA: 1:07 313/600 [==============>...............] - ETA: 1:07 314/600 [==============>...............] - ETA: 1:07 315/600 [==============>...............] - ETA: 1:06 316/600 [==============>...............] - ETA: 1:06 317/600 [==============>...............] - ETA: 1:06 318/600 [==============>...............] - ETA: 1:06 319/600 [==============>...............] - ETA: 1:05 320/600 [===============>..............] - ETA: 1:05 321/600 [===============>..............] - ETA: 1:05 322/600 [===============>..............] - ETA: 1:05 323/600 [===============>..............] - ETA: 1:04 324/600 [===============>..............] - ETA: 1:04 325/600 [===============>..............] - ETA: 1:04 326/600 [===============>..............] - ETA: 1:04 327/600 [===============>..............] - ETA: 1:04 328/600 [===============>..............] - ETA: 1:04 329/600 [===============>..............] - ETA: 1:03 330/600 [===============>..............] - ETA: 1:03 331/600 [===============>..............] - ETA: 1:03 332/600 [===============>..............] - ETA: 1:03 333/600 [===============>..............] - ETA: 1:02 334/600 [===============>..............] - ETA: 1:02 335/600 [===============>..............] - ETA: 1:02 336/600 [===============>..............] - ETA: 1:02 337/600 [===============>..............] - ETA: 1:01 338/600 [===============>..............] - ETA: 1:01 339/600 [===============>..............] - ETA: 1:01 340/600 [================>.............] - ETA: 1:01 341/600 [================>.............] - ETA: 1:00 342/600 [================>.............] - ETA: 1:00 343/600 [================>.............] - ETA: 1:00 344/600 [================>.............] - ETA: 1:00 345/600 [================>.............] - ETA: 59s 346/600 [================>.............] - ETA: 59s 347/600 [================>.............] - ETA: 59s 348/600 [================>.............] - ETA: 59s 349/600 [================>.............] - ETA: 58s 350/600 [================>.............] - ETA: 58s 351/600 [================>.............] - ETA: 58s 352/600 [================>.............] - ETA: 58s 353/600 [================>.............] - ETA: 57s 354/600 [================>.............] - ETA: 57s 355/600 [================>.............] - ETA: 57s 356/600 [================>.............] - ETA: 57s 357/600 [================>.............] - ETA: 56s 358/600 [================>.............] - ETA: 56s 359/600 [================>.............] - ETA: 56s 360/600 [=================>............] - ETA: 56s 361/600 [=================>............] - ETA: 55s 362/600 [=================>............] - ETA: 55s 363/600 [=================>............] - ETA: 55s 364/600 [=================>............] - ETA: 55s 365/600 [=================>............] - ETA: 54s 366/600 [=================>............] - ETA: 54s 367/600 [=================>............] - ETA: 54s 368/600 [=================>............] - ETA: 54s 369/600 [=================>............] - ETA: 54s 370/600 [=================>............] - ETA: 53s 371/600 [=================>............] - ETA: 53s 372/600 [=================>............] - ETA: 53s 373/600 [=================>............] - ETA: 53s 374/600 [=================>............] - ETA: 52s 375/600 [=================>............] - ETA: 52s 376/600 [=================>............] - ETA: 52s 377/600 [=================>............] - ETA: 52s 378/600 [=================>............] - ETA: 51s 379/600 [=================>............] - ETA: 51s 380/600 [==================>...........] - ETA: 51s 381/600 [==================>...........] - ETA: 51s 382/600 [==================>...........] - ETA: 50s 383/600 [==================>...........] - ETA: 50s 384/600 [==================>...........] - ETA: 50s 385/600 [==================>...........] - ETA: 50s 386/600 [==================>...........] - ETA: 49s 387/600 [==================>...........] - ETA: 49s 388/600 [==================>...........] - ETA: 49s 389/600 [==================>...........] - ETA: 49s 390/600 [==================>...........] - ETA: 48s 391/600 [==================>...........] - ETA: 48s 392/600 [==================>...........] - ETA: 48s 393/600 [==================>...........] - ETA: 48s 394/600 [==================>...........] - ETA: 47s 395/600 [==================>...........] - ETA: 47s 396/600 [==================>...........] - ETA: 47s 397/600 [==================>...........] - ETA: 47s 398/600 [==================>...........] - ETA: 46s 399/600 [==================>...........] - ETA: 46s 400/600 [===================>..........] - ETA: 46s 401/600 [===================>..........] - ETA: 46s 402/600 [===================>..........] - ETA: 45s 403/600 [===================>..........] - ETA: 45s 404/600 [===================>..........] - ETA: 45s 405/600 [===================>..........] - ETA: 45s 406/600 [===================>..........] - ETA: 44s 407/600 [===================>..........] - ETA: 44s 408/600 [===================>..........] - ETA: 44s 409/600 [===================>..........] - ETA: 44s 410/600 [===================>..........] - ETA: 43s 411/600 [===================>..........] - ETA: 43s 412/600 [===================>..........] - ETA: 43s 413/600 [===================>..........] - ETA: 43s 414/600 [===================>..........] - ETA: 43s 415/600 [===================>..........] - ETA: 42s 416/600 [===================>..........] - ETA: 42s 417/600 [===================>..........] - ETA: 42s 418/600 [===================>..........] - ETA: 42s 419/600 [===================>..........] - ETA: 41s 420/600 [====================>.........] - ETA: 41s 421/600 [====================>.........] - ETA: 41s 422/600 [====================>.........] - ETA: 41s 423/600 [====================>.........] - ETA: 40s 424/600 [====================>.........] - ETA: 40s 425/600 [====================>.........] - ETA: 40s 426/600 [====================>.........] - ETA: 40s 427/600 [====================>.........] - ETA: 39s 428/600 [====================>.........] - ETA: 39s 429/600 [====================>.........] - ETA: 39s 430/600 [====================>.........] - ETA: 39s 431/600 [====================>.........] - ETA: 38s 432/600 [====================>.........] - ETA: 38s 433/600 [====================>.........] - ETA: 38s 434/600 [====================>.........] - ETA: 38s 435/600 [====================>.........] - ETA: 37s 436/600 [====================>.........] - ETA: 37s 437/600 [====================>.........] - ETA: 37s 438/600 [====================>.........] - ETA: 37s 439/600 [====================>.........] - ETA: 37s 440/600 [=====================>........] - ETA: 36s 441/600 [=====================>........] - ETA: 36s 442/600 [=====================>........] - ETA: 36s 443/600 [=====================>........] - ETA: 36s 444/600 [=====================>........] - ETA: 35s 445/600 [=====================>........] - ETA: 35s 446/600 [=====================>........] - ETA: 35s 447/600 [=====================>........] - ETA: 35s 448/600 [=====================>........] - ETA: 34s 449/600 [=====================>........] - ETA: 34s 450/600 [=====================>........] - ETA: 34s 451/600 [=====================>........] - ETA: 34s 452/600 [=====================>........] - ETA: 33s 453/600 [=====================>........] - ETA: 33s 454/600 [=====================>........] - ETA: 33s 455/600 [=====================>........] - ETA: 33s 456/600 [=====================>........] - ETA: 33s 457/600 [=====================>........] - ETA: 32s 458/600 [=====================>........] - ETA: 32s 459/600 [=====================>........] - ETA: 32s 460/600 [======================>.......] - ETA: 32s 461/600 [======================>.......] - ETA: 31s 462/600 [======================>.......] - ETA: 31s 463/600 [======================>.......] - ETA: 31s 464/600 [======================>.......] - ETA: 31s 465/600 [======================>.......] - ETA: 30s 466/600 [======================>.......] - ETA: 30s 467/600 [======================>.......] - ETA: 30s 468/600 [======================>.......] - ETA: 30s 469/600 [======================>.......] - ETA: 29s 470/600 [======================>.......] - ETA: 29s 471/600 [======================>.......] - ETA: 29s 472/600 [======================>.......] - ETA: 29s 473/600 [======================>.......] - ETA: 29s 474/600 [======================>.......] - ETA: 28s 475/600 [======================>.......] - ETA: 28s 476/600 [======================>.......] - ETA: 28s 477/600 [======================>.......] - ETA: 28s 478/600 [======================>.......] - ETA: 27s 479/600 [======================>.......] - ETA: 27s 480/600 [=======================>......] - ETA: 27s 481/600 [=======================>......] - ETA: 27s 482/600 [=======================>......] - ETA: 26s 483/600 [=======================>......] - ETA: 26s 484/600 [=======================>......] - ETA: 26s 485/600 [=======================>......] - ETA: 26s 486/600 [=======================>......] - ETA: 26s 487/600 [=======================>......] - ETA: 25s 488/600 [=======================>......] - ETA: 25s 489/600 [=======================>......] - ETA: 25s 490/600 [=======================>......] - ETA: 25s 491/600 [=======================>......] - ETA: 24s 492/600 [=======================>......] - ETA: 24s 493/600 [=======================>......] - ETA: 24s 494/600 [=======================>......] - ETA: 24s 495/600 [=======================>......] - ETA: 23s 496/600 [=======================>......] - ETA: 23s 497/600 [=======================>......] - ETA: 23s 498/600 [=======================>......] - ETA: 23s 499/600 [=======================>......] - ETA: 23s 500/600 [========================>.....] - ETA: 22s 501/600 [========================>.....] - ETA: 22s 502/600 [========================>.....] - ETA: 22s 503/600 [========================>.....] - ETA: 22s 504/600 [========================>.....] - ETA: 21s 505/600 [========================>.....] - ETA: 21s 506/600 [========================>.....] - ETA: 21s 507/600 [========================>.....] - ETA: 21s 508/600 [========================>.....] - ETA: 20s 509/600 [========================>.....] - ETA: 20s 510/600 [========================>.....] - ETA: 20s 511/600 [========================>.....] - ETA: 20s 512/600 [========================>.....] - ETA: 20s 513/600 [========================>.....] - ETA: 19s 514/600 [========================>.....] - ETA: 19s 515/600 [========================>.....] - ETA: 19s 516/600 [========================>.....] - ETA: 19s 517/600 [========================>.....] - ETA: 18s 518/600 [========================>.....] - ETA: 18s 519/600 [========================>.....] - ETA: 18s 520/600 [=========================>....] - ETA: 18s 521/600 [=========================>....] - ETA: 17s 522/600 [=========================>....] - ETA: 17s 523/600 [=========================>....] - ETA: 17s 524/600 [=========================>....] - ETA: 17s 525/600 [=========================>....] - ETA: 17s 526/600 [=========================>....] - ETA: 16s 527/600 [=========================>....] - ETA: 16s 528/600 [=========================>....] - ETA: 16s 529/600 [=========================>....] - ETA: 16s 530/600 [=========================>....] - ETA: 15s 531/600 [=========================>....] - ETA: 15s 532/600 [=========================>....] - ETA: 15s 533/600 [=========================>....] - ETA: 15s 534/600 [=========================>....] - ETA: 14s 535/600 [=========================>....] - ETA: 14s 536/600 [=========================>....] - ETA: 14s 537/600 [=========================>....] - ETA: 14s 538/600 [=========================>....] - ETA: 14s 539/600 [=========================>....] - ETA: 13s 540/600 [==========================>...] - ETA: 13s 541/600 [==========================>...] - ETA: 13s 542/600 [==========================>...] - ETA: 13s 543/600 [==========================>...] - ETA: 12s 544/600 [==========================>...] - ETA: 12s 545/600 [==========================>...] - ETA: 12s 546/600 [==========================>...] - ETA: 12s 547/600 [==========================>...] - ETA: 12s 548/600 [==========================>...] - ETA: 11s 549/600 [==========================>...] - ETA: 11s 550/600 [==========================>...] - ETA: 11s 551/600 [==========================>...] - ETA: 11s 552/600 [==========================>...] - ETA: 10s 553/600 [==========================>...] - ETA: 10s 554/600 [==========================>...] - ETA: 10s 555/600 [==========================>...] - ETA: 10s 556/600 [==========================>...] - ETA: 9s 557/600 [==========================>...] - ETA: 9s 558/600 [==========================>...] - ETA: 9s 559/600 [==========================>...] - ETA: 9s 560/600 [===========================>..] - ETA: 9s 561/600 [===========================>..] - ETA: 8s 562/600 [===========================>..] - ETA: 8s 563/600 [===========================>..] - ETA: 8s 564/600 [===========================>..] - ETA: 8s 565/600 [===========================>..] - ETA: 7s 566/600 [===========================>..] - ETA: 7s 567/600 [===========================>..] - ETA: 7s 568/600 [===========================>..] - ETA: 7s 569/600 [===========================>..] - ETA: 7s 570/600 [===========================>..] - ETA: 6s 571/600 [===========================>..] - ETA: 6s 572/600 [===========================>..] - ETA: 6s 573/600 [===========================>..] - ETA: 6s 574/600 [===========================>..] - ETA: 5s 575/600 [===========================>..] - ETA: 5s 576/600 [===========================>..] - ETA: 5s 577/600 [===========================>..] - ETA: 5s 578/600 [===========================>..] - ETA: 4s 579/600 [===========================>..] - ETA: 4s 580/600 [============================>.] - ETA: 4s 581/600 [============================>.] - ETA: 4s 582/600 [============================>.] - ETA: 4s 583/600 [============================>.] - ETA: 3s 584/600 [============================>.] - ETA: 3s 585/600 [============================>.] - ETA: 3s 586/600 [============================>.] - ETA: 3s 587/600 [============================>.] - ETA: 2s 588/600 [============================>.] - ETA: 2s 589/600 [============================>.] - ETA: 2s 590/600 [============================>.] - ETA: 2s 591/600 [============================>.] - ETA: 2s 592/600 [============================>.] - ETA: 1s 593/600 [============================>.] - ETA: 1s 594/600 [============================>.] - ETA: 1s 595/600 [============================>.] - ETA: 1s 596/600 [============================>.] - ETA: 0s 597/600 [============================>.] - ETA: 0s 598/600 [============================>.] - ETA: 0s 599/600 [============================>.] - ETA: 0s 600/600 [==============================] - 135s 225ms/step Testing for epoch 1: component | loss | generation_loss | auxiliary_loss ----------------------------------------------------------------- generator (train) | 1.40 | 0.9750 | 0.4249 generator (test) | 1.18 | 1.1657 | 0.0161 discriminator (train) | 1.55 | 0.6922 | 0.8536 discriminator (test) | 0.59 | 0.4920 | 0.0960 Epoch 2/100 1/600 [..............................] - ETA: 2:14 2/600 [..............................] - ETA: 2:09 599/600 [============================>.] - ETA: 0s 600/600 [==============================] - 124s 206ms/step Testing for epoch 100: component | loss | generation_loss | auxiliary_loss ----------------------------------------------------------------- generator (train) | 0.76 | 0.7628 | 0.0005 generator (test) | 0.82 | 0.8151 | 0.0000 discriminator (train) | 0.71 | 0.6932 | 0.0145 discriminator (test) | 0.70 | 0.6942 | 0.0106 Process finished with exit code 0Keras详细介绍
中文：http://keras-cn.readthedocs.io/en/latest/
实例下载
https://github.com/keras-team/keras
https://github.com/keras-team/keras/tree/master/examples