Tensorflow2.0学习笔记------猫狗分类(数据增强解决过拟合)

通过MNIST和CIFAR10两个数据集的练习之后，对整个模型的搭建、训练以及测试已经比较熟练了。猫狗分类同样也是比较经典的cv问题，本文再进一步强化之前知识点的同时，学习使用tf.keras.preprocessing.image.ImageDataGenerator来处理数据，并且尝试使用数据增强的方法来解决过拟合问题。

Step1：国际惯例加载tensorflow

import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Conv2D, Flatten, Dropout, MaxPooling2D
from tensorflow.keras.preprocessing.image import ImageDataGenerator

import os
import numpy as np
import matplotlib.pyplot as plt

Step2：加载数据集

数据集来自kaggle

_URL = 'https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip'

path_to_zip = tf.keras.utils.get_file('cats_and_dogs.zip', origin=_URL, extract=True)

PATH = os.path.join(os.path.dirname(path_to_zip), 'cats_and_dogs_filtered')

Step3：查看训练集内容

为变量分配相应的路径

train_dir = os.path.join(PATH, 'train')
validation_dir = os.path.join(PATH, 'validation')

train_cats_dir = os.path.join(train_dir, 'cats')  
train_dogs_dir = os.path.join(train_dir, 'dogs') 
validation_cats_dir = os.path.join(validation_dir, 'cats') 
validation_dogs_dir = os.path.join(validation_dir, 'dogs')  

查看数据集中的图片数量

num_cats_tr = len(os.listdir(train_cats_dir))
num_dogs_tr = len(os.listdir(train_dogs_dir))

num_cats_val = len(os.listdir(validation_cats_dir))
num_dogs_val = len(os.listdir(validation_dogs_dir))

total_train = num_cats_tr + num_dogs_tr
total_val = num_cats_val + num_dogs_val

print('total training cat images:', num_cats_tr)
print('total training dog images:', num_dogs_tr)

print('total validation cat images:', num_cats_val)
print('total validation dog images:', num_dogs_val)
print("--")
print("Total training images:", total_train)
print("Total validation images:", total_val)

Step4：使用ImageDataGenerator处理数据

先设置一下参数

batch_size = 128
epochs = 15
IMG_HEIGHT = 150
IMG_WIDTH = 150

# 归一化处理
train_image_generator = ImageDataGenerator(rescale=1./255) 
validation_image_generator = ImageDataGenerator(rescale=1./255) 

#
train_data_gen = train_image_generator.flow_from_directory(batch_size=batch_size,
                                                           directory=train_dir,
                                                           shuffle=True,
                                                           target_size=(IMG_HEIGHT, IMG_WIDTH),
                                                           class_mode='binary')

val_data_gen = validation_image_generator.flow_from_directory(batch_size=batch_size,
                                                              directory=validation_dir,
                                                              target_size=(IMG_HEIGHT, IMG_WIDTH),
                                                              class_mode='binary')

Step5：搭建模型并设置训练流程

同样还是使用卷积网络

model = Sequential([
    Conv2D(16, 3, padding='same', activation='relu', input_shape=(IMG_HEIGHT, IMG_WIDTH ,3)),
    MaxPooling2D(),
    Conv2D(32, 3, padding='same', activation='relu'),
    MaxPooling2D(),
    Conv2D(64, 3, padding='same', activation='relu'),
    MaxPooling2D(),
    Flatten(),
    Dense(512, activation='relu'),
    Dense(1)
])

model.compile(optimizer='adam',
              loss=tf.keras.losses.BinaryCrossentropy(from_logits=True),
              metrics=['accuracy'])

显示网络瞧瞧

model.summary()

Step6：启动训练并可视化训练结果

history = model.fit_generator(
    train_data_gen,
    steps_per_epoch=total_train // batch_size,
    epochs=epochs,
    validation_data=val_data_gen,
    validation_steps=total_val // batch_size
)

acc = history.history['accuracy']
val_acc = history.history['val_accuracy']

loss=history.history['loss']
val_loss=history.history['val_loss']

epochs_range = range(epochs)

plt.figure(figsize=(15, 5))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, acc, label='Training Accuracy')
plt.plot(epochs_range, val_acc, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, loss, label='Training Loss')
plt.plot(epochs_range, val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()