【吴恩达Tensorflow 2.0实践课】2.1 Augmentation

Tensorflow 2.0实践课】2.1 Augmentation"/>

【吴恩达Tensorflow 2.0实践课】2.1 Augmentation

目录

 

 

2.1.1 Training with th Cats & Dogs dataset

2.1.2 walking through the notebook

---

 

2.1.1 Training with th Cats & Dogs dataset

把图片放入不同文件夹中，TensorFlow能够自动添加标签

 

2.1.2 walking through the notebook

下载数据（来源Kaggle）

!wget --no-check-certificate \.zip \-O /tmp/cats_and_dogs_filtered.zip

-O 接重新命名的文件

解压文件

import os
import zipfilelocal_zip = '/tmp/cats_and_dogs_filtered.zip'zip_ref = zipfile.ZipFile(local_zip, 'r')zip_ref.extractall('/tmp')
zip_ref.close()

zipfile.ZipFile: 用于读写 ZIP 文件的类

解压到 /tmp/cats_and_dogs_filtered  文件夹中

定义目录

base_dir = '/tmp/cats_and_dogs_filtered'train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')# Directory with our training cat/dog pictures
train_cats_dir = os.path.join(train_dir, 'cats')
train_dogs_dir = os.path.join(train_dir, 'dogs')# Directory with our validation cat/dog pictures
validation_cats_dir = os.path.join(validation_dir, 'cats')
validation_dogs_dir = os.path.join(validation_dir, 'dogs')

 

打印文件名

train_cat_fnames = os.listdir( train_cats_dir )
train_dog_fnames = os.listdir( train_dogs_dir )print(train_cat_fnames[:10])
print(train_dog_fnames[:10])

 

打印训练/验证图片的个数

print('total training cat images :', len(os.listdir( train_cats_dir ) ))
print('total training dog images :', len(os.listdir( train_dogs_dir ) ))print('total validation cat images :', len(os.listdir( validation_cats_dir ) ))
print('total validation dog images :', len(os.listdir( validation_dogs_dir ) ))

显示结果：

total training cat images : 1000
total training dog images : 1000
total validation cat images : 500
total validation dog images : 500

 

显示部分图片

%matplotlib inline  
# 用 jupyter/Colab 需要加这一行
# 作用是当你调用matplotlib.pyplot的绘图函数plot()进行绘图的时候，或者生成一个figure画布的时候，可以直接在你的python console里面生成图像。import matplotlib.image as mpimg
import matplotlib.pyplot as plt# Parameters for our graph; we'll output images in a 4x4 configuration
nrows = 4
ncols = 4pic_index = 0 # Index for iterating over images# Set up matplotlib fig, and size it to fit 4x4 pics
fig = plt.gcf()
fig.set_size_inches(ncols*4, nrows*4)# 猫和狗，各自显示8张，所以索引+8
pic_index+=8# pic_index-8为0；    pic_index为8
# next_cat_pix为一个列表，./train/cat中前8个图片的完整地址
next_cat_pix = [os.path.join(train_cats_dir, fname) for fname in train_cat_fnames[ pic_index-8:pic_index] ]# next_dog_pix为一个列表，./train/dog中前8个图片的完整地址
next_dog_pix = [os.path.join(train_dogs_dir, fname) for fname in train_dog_fnames[ pic_index-8:pic_index]]# 一共16张图片，索引存入i中，路径存入img_path中
for i, img_path in enumerate(next_cat_pix + next_dog_pix):# Set up subplot; subplot indices start at 1# 第一幅图位置为(4, 4, 1)sp = plt.subplot(nrows, ncols, i + 1)sp.axis('Off') # Don't show axes (or gridlines)# 读取图片img = mpimg.imread(img_path)# 显示图片plt.imshow(img)plt.show()

显示结果：

 

创建模型

import tensorflow as tfmodel = tf.keras.models.Sequential([# Note the input shape is the desired size of the image 150x150 with 3 bytes colortf.keras.layers.Conv2D(16, (3 ,3), activation='relu', input_shape=(150, 150, 3)),tf.keras.layers.MaxPooling2D(2 ,2),tf.keras.layers.Conv2D(32, (3 ,3), activation='relu'),tf.keras.layers.MaxPooling2D(2 ,2),tf.keras.layers.Conv2D(64, (3 ,3), activation='relu'),tf.keras.layers.MaxPooling2D(2 ,2),# Flatten the results to feed into a DNNtf.keras.layers.Flatten(),# 512 neuron hidden layertf.keras.layers.Dense(512, activation='relu'),# Only 1 output neuron. It will contain a value from 0-1 where 0 for 1 class ('cats') and 1 for the other ('dogs')tf.keras.layers.Dense(1, activation='sigmoid')
])model.summary()

 显示结果：

设置优化器、损失函数

from tensorflow.keras.optimizers import RMSpropmodelpile(optimizer=RMSprop(lr=0.001),loss='binary_crossentropy',metrics = ['acc'])

数据预处理

from tensorflow.keras.preprocessing.image import ImageDataGenerator# All images will be rescaled by 1./255.
train_datagen = ImageDataGenerator( rescale = 1.0/255. )
test_datagen  = ImageDataGenerator( rescale = 1.0/255. )# --------------------
# Flow training images in batches of 20 using train_datagen generator
# --------------------
train_generator = train_datagen.flow_from_directory(train_dir,batch_size=20,class_mode='binary',target_size=(150, 150))     
# --------------------
# Flow validation images in batches of 20 using test_datagen generator
# --------------------
validation_generator =  test_datagen.flow_from_directory(validation_dir,batch_size=20,class_mode  = 'binary',target_size = (150, 150))

训练

history = model.fit_generator(train_generator,validation_data=validation_generator,steps_per_epoch=100,epochs=15,validation_steps=50,verbose=2)

可视化卷积过程

import numpy as np
import random
from   tensorflow.keras.preprocessing.image import img_to_array, load_img# Let's define a new Model that will take an image as input, and will output
# intermediate representations for all layers in the previous model after
# the first.
successive_outputs = [layer.output for layer in model.layers[1:]]#visualization_model = Model(img_input, successive_outputs)
visualization_model = tf.keras.models.Model(inputs = model.input, outputs = successive_outputs)# Let's prepare a random input image of a cat or dog from the training set.
cat_img_files = [os.path.join(train_cats_dir, f) for f in train_cat_fnames]
dog_img_files = [os.path.join(train_dogs_dir, f) for f in train_dog_fnames]img_path = random.choice(cat_img_files + dog_img_files)
img = load_img(img_path, target_size=(150, 150))  # this is a PIL imagex   = img_to_array(img)                           # Numpy array with shape (150, 150, 3)
x   = x.reshape((1,) + x.shape)                   # Numpy array with shape (1, 150, 150, 3)# Rescale by 1/255
x /= 255.0# Let's run our image through our network, thus obtaining all
# intermediate representations for this image.
successive_feature_maps = visualization_model.predict(x)# These are the names of the layers, so can have them as part of our plot
layer_names = [layer.name for layer in model.layers]# -----------------------------------------------------------------------
# Now let's display our representations
# -----------------------------------------------------------------------
for layer_name, feature_map in zip(layer_names, successive_feature_maps):if len(feature_map.shape) == 4:#-------------------------------------------# Just do this for the conv / maxpool layers, not the fully-connected layers#-------------------------------------------n_features = feature_map.shape[-1]  # number of features in the feature mapsize       = feature_map.shape[ 1]  # feature map shape (1, size, size, n_features)# We will tile our images in this matrixdisplay_grid = np.zeros((size, size * n_features))#-------------------------------------------------# Postprocess the feature to be visually palatable#-------------------------------------------------for i in range(n_features):x  = feature_map[0, :, :, i]x -= x.mean()x /= x.std ()x *=  64x += 128x  = np.clip(x, 0, 255).astype('uint8')display_grid[:, i * size : (i + 1) * size] = x # Tile each filter into a horizontal grid#-----------------# Display the grid#-----------------scale = 20. / n_featuresplt.figure( figsize=(scale * n_features, scale) )plt.title ( layer_name )plt.grid  ( False )plt.imshow( display_grid, aspect='auto', cmap='viridis' ) 

测试准确率和损失

#-----------------------------------------------------------
# Retrieve a list of list results on training and test data
# sets for each training epoch
#-----------------------------------------------------------
acc      = history.history[     'acc' ]
val_acc  = history.history[ 'val_acc' ]
loss     = history.history[    'loss' ]
val_loss = history.history['val_loss' ]epochs   = range(len(acc)) # Get number of epochs#------------------------------------------------
# Plot training and validation accuracy per epoch
#------------------------------------------------
plt.plot  ( epochs,     acc )
plt.plot  ( epochs, val_acc )
plt.title ('Training and validation accuracy')
plt.figure()#------------------------------------------------
# Plot training and validation loss per epoch
#------------------------------------------------
plt.plot  ( epochs,     loss )
plt.plot  ( epochs, val_loss )
plt.title ('Training and validation loss'   )

 

清理

import os, signalos.kill(     os.getpid() , signal.SIGKILL)

 

2.1.3 Coding Augmentation with ImageDataGenerator

图片旋转有助于提高识别

# update to do image augmentation
train_datagen = ImageDataGenerator(rescale=1.0/255,            # 归一化rotation_range=40,          # 0-40° 随机旋转图像width_shift_range=0.2,      # 移位，将图片在水平方向上，随机移位20%height_shift_range=0.2,     # 移位，将图片在垂直方向上，随机移位20%shear_range=0.2,            # 剪力，随机剪去图像的20%zoom_range=0.2,             # 随机缩放比例20%horizontal_flip=True        # 随机水平镜像fill_mode='nearest'         # 填充周围像素
)

图片剪力：

 

2.1.4 Demonstrating overfitting in cats vs. dogs

 

train_datagen = ImageDataGenerator(rescale=1./255)

过拟合

2.1.5 Adding augmentation to cats vs. dogs

train_datagen = ImageDataGenerator(rescale=1./255,rotation_range=40,width_shift_range=0.2,height_shift_range=0.2,shear_range=0.2,zoom_range=0.2,horizontal_flip=True,fill_mode='nearest')

 

验证数据集的损失波动较大

俺算的结果：

 

2.1.6 Exploring augmentation horses vs. humans