深度学习TF—12.Auto-Encoders（AE）与Varational Auto Encoder（VAE）实战

文章目录

一、Auto-Encoders（AE）实战
二、Varational Auto Encoder（VAE）实战

一、Auto-Encoders（AE）实战

导入模块

# 图片的重建
import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'# 使用GPU，设置内存自动增长
from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSessionconfig = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)import numpy as np
import random
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
from tensorflow import keras
from tensorflow.keras import Sequential,layers# 设置随机数种子
def seed_everying(SEED):os.environ['TF_DETERMINISTIC_OPS'] = '1'os.environ['PYTHONHASHSEED'] = str(SEED)random.seed(SEED)np.random.seed(SEED)tf.random.set_seed(SEED)seed_everying(42)assert tf.__version__.startswith('2.')

保存图片

# 定义Image保存函数,将多张Image保存到一张Image里面去
def save_images(imgs,name):news_im = Image.new('L',(280,280))index = 0# 将10张Image保存到一张Image里面去for i in range(0,280,28):for j in range(0,280,28):im = imgs[index]im = Image.fromarray(im,mode='L')news_im.paste(im,(i,j))index += 1news_im.save(name)

加载数据集并进行预处理

h_dim = 20  # 将原来的784维的数据降维到20维上
batchsz = 512
lr = 1e-3# 加载数据集
(x_train, y_train), (x_test, y_test) = keras.datasets.fashion_mnist.load_data()
x_train, x_test = x_train.astype(np.float32) / 255., x_test.astype(np.float32) / 255.
# 无监督学习，我们不需要使用label，即y_train、y_test不使用
train_db = tf.data.Dataset.from_tensor_slices((x_train))
train_db = train_db.shuffle(batchsz * 5).batch(batchsz)
test_db = tf.data.Dataset.from_tensor_slices((x_test))
test_db = test_db.batch(batchsz)print(x_train.shape, y_train.shape)
print(x_test.shape, y_test.shape)

搭建网络

# 建立模型
class AE(keras.Model):def __init__(self):super(AE, self).__init__()# Encoders-编码self.encoder = Sequential([layers.Dense(256, activation=tf.nn.relu),layers.Dense(128, activation=tf.nn.relu),layers.Dense(h_dim)])# Decoders-解码self.decoder = Sequential([layers.Dense(128, activation=tf.nn.relu),layers.Dense(256, activation=tf.nn.relu),layers.Dense(784)])def call(self, inputs, training=None):# [b,784] -> [b,10]h = self.encoder(inputs)# [b,10] -> [b,784]x_hat = self.decoder(h)return x_hat

全部代码：

# 图片的重建
import osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'# 使用GPU，设置内存自动增长
from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSessionconfig = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)import numpy as np
import random
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
from tensorflow import keras
from tensorflow.keras import Sequential, layers# 设置随机数种子
# 随机数种子
def seed_everying(SEED):os.environ['TF_DETERMINISTIC_OPS'] = '1'os.environ['PYTHONHASHSEED'] = str(SEED)random.seed(SEED)np.random.seed(SEED)tf.random.set_seed(SEED)seed_everying(42)assert tf.__version__.startswith('2.')# 定义Image保存函数,将多张Image保存到一张Image里面去
def save_images(imgs, name):news_im = Image.new('L', (280, 280))index = 0# 将10张Image保存到一张Image里面去for i in range(0, 280, 28):for j in range(0, 280, 28):im = imgs[index]im = Image.fromarray(im, mode='L')news_im.paste(im, (i, j))index += 1news_im.save(name)h_dim = 20  # 将原来的784维的数据降维到20维上
batchsz = 512
lr = 1e-3# 加载数据集
(x_train, y_train), (x_test, y_test) = keras.datasets.fashion_mnist.load_data()
x_train, x_test = x_train.astype(np.float32) / 255., x_test.astype(np.float32) / 255.
# 无监督学习，我们不需要使用label，即y_train、y_test不使用
train_db = tf.data.Dataset.from_tensor_slices((x_train))
train_db = train_db.shuffle(batchsz * 5).batch(batchsz)
test_db = tf.data.Dataset.from_tensor_slices((x_test))
test_db = test_db.batch(batchsz)print(x_train.shape, y_train.shape)
print(x_test.shape, y_test.shape)# 建立模型
class AE(keras.Model):def __init__(self):super(AE, self).__init__()# Encoders-编码self.encoder = Sequential([layers.Dense(256, activation=tf.nn.relu),layers.Dense(128, activation=tf.nn.relu),layers.Dense(h_dim)])# Decoders-解码self.decoder = Sequential([layers.Dense(128, activation=tf.nn.relu),layers.Dense(256, activation=tf.nn.relu),layers.Dense(784)])def call(self, inputs, training=None):# [b,784] -> [b,10]h = self.encoder(inputs)# [b,10] -> [b,784]x_hat = self.decoder(h)return x_hatmodel = AE()
model.build(input_shape=(None, 784))
# 查看网络结构
model.summary()optimizer = tf.optimizers.Adam(lr=lr)
# training
for epoch in range(50):for step, x in enumerate(train_db):# [b,28,28] -> [b,784]x = tf.reshape(x, [-1, 784])with tf.GradientTape() as tape:x_rec_logits = model.call(x)rec_loss = tf.losses.binary_crossentropy(x, x_rec_logits, from_logits=True)rec_loss = tf.reduce_mean(rec_loss)grads = tape.gradient(rec_loss, model.trainable_variables)optimizer.apply_gradients(zip(grads, model.trainable_variables))# 打印training进度if step % 100 == 0:print(epoch, step, float(rec_loss))# evaluationx = next(iter(test_db))# training时，将x打平了x = tf.reshape(x, [-1, 784])        logits = model.predict(x)x_hat = tf.sigmoid(logits)# [b,784] => [b,28,28]x_hat = tf.reshape(x_hat, [-1, 28, 28])# [b,28,28] => [2b,28,28]# x_concat = tf.concat([x, x_hat], axis=0)x_concat = x_hatx_concat = x_concat.numpy() * 255.x_concat = x_concat.astype(np.uint8)save_images(x_concat, 'ae_images/rec_epoch_%d.png' % epoch)

(60000, 28, 28) (60000,)
(10000, 28, 28) (10000,)
Model: "ae"
_________________________________________________________________
Layer (type)                 Output Shape              Param #
=================================================================
sequential (Sequential)      multiple                  235146
_________________________________________________________________
sequential_1 (Sequential)    multiple                  235920
=================================================================
Total params: 471,066
Trainable params: 471,066
Non-trainable params: 0
_________________________________________________________________
0 0 0.6929969787597656
0 100 0.3294386863708496
1 0 0.31267669796943665
1 100 0.30951932072639465
2 0 0.2991226315498352
2 100 0.30571430921554565
3 0 0.30124884843826294
3 100 0.29927968978881836
4 0 0.29633498191833496
4 100 0.29800713062286377
5 0 0.28601688146591187
...

每5epoch展示图片

二、Varational Auto Encoder（VAE）实战

导入模块

# 图片的重建
import osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'# 使用GPU，设置内存自动增长
from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSessionconfig = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)import numpy as np
import random
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
from tensorflow import keras
from tensorflow.keras import Sequential, layers# 设置随机数种子
def seed_everying(SEED):os.environ['TF_DETERMINISTIC_OPS'] = '1'os.environ['PYTHONHASHSEED'] = str(SEED)random.seed(SEED)np.random.seed(SEED)tf.random.set_seed(SEED)seed_everying(42)assert tf.__version__.startswith('2.')

保存图片

# 定义Image保存函数,将多张Image保存到一张Image里面去
def save_images(imgs,name):news_im = Image.new('L',(280,280))index = 0# 将10张Image保存到一张Image里面去for i in range(0,280,28):for j in range(0,280,28):im = imgs[index]im = Image.fromarray(im,mode='L')news_im.paste(im,(i,j))index += 1news_im.save(name)

加载数据并进行预处理

z_dim = 10
batchsz = 256
lr = 1e-3# 加载数据集
(x_train, y_train), (x_test, y_test) = keras.datasets.fashion_mnist.load_data()
x_train, x_test = x_train.astype(np.float32) / 255., x_test.astype(np.float32) / 255.
# 无监督学习，我们不需要使用label，即y_train、y_test不使用
train_db = tf.data.Dataset.from_tensor_slices((x_train))
train_db = train_db.shuffle(batchsz * 5).batch(batchsz)
test_db = tf.data.Dataset.from_tensor_slices((x_test))
test_db = test_db.batch(batchsz)print(x_train.shape, y_train.shape)
print(x_test.shape, y_test.shape)

损失函数定义

rec_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels = x,logits = x_hat)
rec_loss = tf.reduce_sum(rec_loss) / x.shape[0]# compute KL divergence (mu, var) ~N(0,1)
# p~N(mu,∨var) q~N(0,1) 故u2=0,σ2=1
kl_div = -0.5 * (2*log_var + 1 - mu ** 2 - tf.exp(log_var))
kl_div = tf.reduce_mean(kl_div) / x.shape[0]loss = rec_loss + 1. * kl_div

构建网络

class VAE(keras.Model):def __init__(self):super(VAE, self).__init__()# Encoder - 编码self.fc1 = layers.Dense(128)self.fc2 = layers.Dense(z_dim)  # get mean predictionself.fc3 = layers.Dense(z_dim)  # get var  prediction# Decoderself.fc4 = layers.Dense(128)self.fc5 = layers.Dense(784)# 定义编码函数def encoder(self, x):h = tf.nn.relu(self.fc1(x))# get meanmu = self.fc2(h)# get variancelog_var = self.fc3(h)return mu, log_var# 定义解码函数def decoder(self, z):out = tf.nn.relu(self.fc4(z))out = self.fc5(out)return out# 定义z函数def reparameterize(self, mu, log_var):eps = tf.random.normal(log_var.shape)# 标准差std = tf.exp(log_var) ** 0.5# std = tf.exp(log_var * 0.5)z = mu + std * epsreturn z# 定义前向传播def call(self, inputs, training=None):# [b,784] => [b,z_dim],[b,z_dim]mu, log_var = self.encoder(inputs)# reparameterization trickz = self.reparameterize(mu, log_var)x_hat = self.decoder(z)# 除了前向传播不同以外，还有一个约束，使得mu，var趋向于正态分布return x_hat, mu, log_var

全部代码

# 图片的重建
import osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'# 使用GPU，设置内存自动增长
from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSessionconfig = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)import numpy as np
import random
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
from tensorflow import keras
from tensorflow.keras import Sequential, layers# 设置随机数种子
def seed_everying(SEED):os.environ['TF_DETERMINISTIC_OPS'] = '1'os.environ['PYTHONHASHSEED'] = str(SEED)random.seed(SEED)np.random.seed(SEED)tf.random.set_seed(SEED)seed_everying(42)assert tf.__version__.startswith('2.')def save_images(imgs, name):new_im = Image.new('L', (280, 280))index = 0for i in range(0, 280, 28):for j in range(0, 280, 28):im = imgs[index]im = Image.fromarray(im, mode='L')new_im.paste(im, (i, j))index += 1new_im.save(name)z_dim = 10
batchsz = 256
lr = 1e-3# 加载数据集
(x_train, y_train), (x_test, y_test) = keras.datasets.fashion_mnist.load_data()
x_train, x_test = x_train.astype(np.float32) / 255., x_test.astype(np.float32) / 255.
# 无监督学习，我们不需要使用label，即y_train、y_test不使用
train_db = tf.data.Dataset.from_tensor_slices((x_train))
train_db = train_db.shuffle(batchsz * 5).batch(batchsz)
test_db = tf.data.Dataset.from_tensor_slices((x_test))
test_db = test_db.batch(batchsz)print(x_train.shape, y_train.shape)
print(x_test.shape, y_test.shape)class VAE(keras.Model):def __init__(self):super(VAE, self).__init__()# Encoder - 编码self.fc1 = layers.Dense(128)self.fc2 = layers.Dense(z_dim)  # get mean predictionself.fc3 = layers.Dense(z_dim)  # get var  prediction# Decoderself.fc4 = layers.Dense(128)self.fc5 = layers.Dense(784)# 定义编码函数def encoder(self, x):h = tf.nn.relu(self.fc1(x))# get meanmu = self.fc2(h)# get variancelog_var = self.fc3(h)return mu, log_var# 定义解码函数def decoder(self, z):out = tf.nn.relu(self.fc4(z))out = self.fc5(out)return out# 定义z函数def reparameterize(self, mu, log_var):eps = tf.random.normal(log_var.shape)# 标准差std = tf.exp(log_var) ** 0.5# std = tf.exp(log_var * 0.5)z = mu + std * epsreturn z# 定义前向传播def call(self, inputs, training=None):# [b,784] => [b,z_dim],[b,z_dim]mu, log_var = self.encoder(inputs)# reparameterization trickz = self.reparameterize(mu, log_var)x_hat = self.decoder(z)# 除了前向传播不同以外，还有一个约束，使得mu，var趋向于正态分布return x_hat, mu, log_varmodel = VAE()
model.build(input_shape=(4, 784))               # 这一个与往常构造的时候有区别
optimizer = tf.optimizers.Adam(lr=lr)for epoch in range(50):for step, x in enumerate(train_db):# [b,28,28] => [b,784]x = tf.reshape(x, [-1, 784])with tf.GradientTape() as tape:x_hat, mu, log_var = model(x)rec_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels = x,logits = x_hat)rec_loss = tf.reduce_sum(rec_loss) / x.shape[0]# compute KL divergence (mu, var) ~N(0,1)# p~N(mu,∨var) q~N(0,1) 故u2=0,σ2=1kl_div = -0.5 * (2*log_var + 1 - mu ** 2 - tf.exp(log_var))kl_div = tf.reduce_mean(kl_div) / x.shape[0]loss = rec_loss + 1. * kl_divgrads = tape.gradient(loss, model.trainable_variables)optimizer.apply_gradients(zip(grads, model.trainable_variables))if step % 100 == 0:print(epoch, step, 'kl div:', float(kl_div), 'rec loss:', float(loss))# evaluation-测试生成效果z = tf.random.normal((batchsz,z_dim))logits = model.decoder(z)x_hat = tf.sigmoid(logits)x_hat = tf.reshape(x_hat,[-1,28,28]).numpy()* 255.x_hat = x_hat.astype(np.uint8)save_images(x_hat,'vae_image/sampled_epoch%d.png'%epoch)# 重建图片x = next(iter(test_db))x = tf.reshape(x, [-1, 784])x_hat_logits, _, _ = model(x)x_hat = tf.sigmoid(x_hat_logits)x_hat = tf.reshape(x_hat, [-1, 28, 28]).numpy() * 255.x_hat = x_hat.astype(np.uint8)save_images(x_hat,'vae_image_1/sampled_epoch%d.png'%epoch)

(60000, 28, 28) (60000,)
(10000, 28, 28) (10000,)
0 0 kl div: 0.001522199483588338 rec loss: 545.1880493164062
0 100 kl div: 0.12963518500328064 rec loss: 276.12176513671875
0 200 kl div: 0.15164682269096375 rec loss: 247.87600708007812
1 0 kl div: 0.1401350051164627 rec loss: 252.12911987304688
1 100 kl div: 0.14920486509799957 rec loss: 241.8990478515625
1 200 kl div: 0.16304132342338562 rec loss: 233.77679443359375
2 0 kl div: 0.16015133261680603 rec loss: 238.06817626953125
2 100 kl div: 0.15632715821266174 rec loss: 240.16061401367188
2 200 kl div: 0.157673180103302 rec loss: 233.2672882080078
3 0 kl div: 0.1759733110666275 rec loss: 230.8567352294922
3 100 kl div: 0.16381677985191345 rec loss: 230.72335815429688
3 200 kl div: 0.15521030128002167 rec loss: 233.7858123779297
4 0 kl div: 0.16824732720851898 rec loss: 227.0357208251953
4 100 kl div: 0.16316646337509155 rec loss: 235.9604949951172
4 200 kl div: 0.1655232012271881 rec loss: 229.02232360839844
...

每5epoch打印一次：

如果对您有帮助，麻烦点赞关注，这真的对我很重要！！！如果需要互关，请评论留言！

深度学习TF—12.Auto-Encoders（AE）与Varational Auto Encoder（VAE）实战相关推荐

深度学习（12）TensorFlow高阶操作一: 合并与分割
深度学习(12)TensorFlow高阶操作一: 合并与分割 1. concat 2. stack: create new dim 3. Dim mismatch 4. unstuck 5. spli ...
HALCON 20.11：深度学习笔记(12)---语义分割
HALCON 20.11:深度学习笔记(12)--- 语义分割 HALCON 20.11.0.0中,实现了深度学习方法. 本章解释了如何使用基于深度学习的语义分割,包括训练和推理阶段. 通过语义分割, ...
深度学习——（12）Knowledge distillation（Demo）
深度学习--(12)Knowledge distillation(Demo) 原本昨天晚上要写的,但是奈何手中有更紧迫的任务需要做,所以自己还没有实战,昨天看到了一个简单的demo,自己写了一部分注释 ...
深度学习之自编码器（2）Fashion MNIST图片重建实战
深度学习之自编码器(2)Fashion MNIST图片重建实战 1. Fashion MNIST数据集 2. 编码器 3. 解码器 4. 自编码器 5. 网络训练 6. 图片重建完整代码自编码器 ...
深度学习之循环神经网络（5）RNN情感分类问题实战
深度学习之循环神经网络(5)RNN情感分类问题实战 1. 数据集 2. 网络模型 3. 训练与测试完整代码运行结果现在利用基础的RNN网络来挑战情感分类问题.网络结构如下图所示,RNN网络共两 ...
深度学习时间序列预测：GRU算法构建多变量时间序列预测模型+代码实战
深度学习时间序列预测:GRU算法构建多变量时间序列预测模型+代码实战注意参考:深度学习多变量时间序列预测:GRU算法构建单变量时间序列预测模型+代码实战 GRU(Gate Recurrent Uni ...
深度学习（31）随机梯度下降九: Himmelblau函数优化实战
深度学习(31)随机梯度下降九: Himmelblau函数优化实战 1. Himmelblau函数 2. 函数优化实战 1. Himmelblau函数 Himmelblau函数是用来测试后话算法的常用 ...
深度学习模型处理多标签（multi_label）分类任务——keras实战
深度学习模型处理多标签(multi_label)分类任务--keras实战 https://zhuanlan.zhihu.com/p/107737824
机器学习深度学习高阶内容系列-kaggle广告点击欺诈识别实战
机器学习深度学习高阶内容系列-kaggle广告点击欺诈识别实战中国是全球最大的智能移动设备市场,每月有超过10亿台智能移动设备投入使用 .TalkingData是中国最大的独立大数据服务平台,覆盖全 ...
计算机视觉与深度学习（12）
原文地址:Image Completion with Deep Learning in TensorFlow by Brandon Amos 原文翻译与校对:@MOLLY && 寒小阳 ...

深度学习TF—12.Auto-Encoders（AE）与Varational Auto Encoder（VAE）实战

文章目录

一、Auto-Encoders（AE）实战

二、Varational Auto Encoder（VAE）实战

深度学习TF—12.Auto-Encoders（AE）与Varational Auto Encoder（VAE）实战相关推荐

最新文章

热门文章