目录
- 一、配置环境
- 二、准备数据集
- 三、基准模型
- 四、基准模型调整
- 五、使用VGG19实现猫狗分类
- 六、参考资料
一、配置环境
这里我用命令行创建虚拟环境
- 输入如下命令创建虚拟环境
- 如图,输入y
- 如图,说明安装成功
- 输入命令进入 cat_dog 虚拟环境
activate
conda activate cat_dog
- 进入环境后,下载所需包
pip install tensorflow -i https://pypi.douban.com/simple
pip install keras -i https://pypi.douban.com/simple
二、准备数据集
数据集的下载
kaggle网站的数据集下载地址:https://www.kaggle.com/lizhensheng/-2000
或百度网盘下载:https://pan.baidu.com/s/13hw4LK8ihR6-6-8mpjLKDA
提取码:dmp4
1.将下载的数据集解压
2.进入jupyter notebook,输入如下代码对图片进行分类
import tensorflow as tf
import keras
import os, shutil
# 原始目录所在的路径
original_dataset_dir = 'G:\\Cat_And_Dog\\kaggle\\train\\'
# 数据集分类后的目录
base_dir = 'G:\\Cat_And_Dog\\kaggle\\cats_and_dogs_small'
os.mkdir(base_dir)
# # 训练、验证、测试数据集的目录
train_dir = os.path.join(base_dir, 'train')
os.mkdir(train_dir)
validation_dir = os.path.join(base_dir, 'validation')
os.mkdir(validation_dir)
test_dir = os.path.join(base_dir, 'test')
os.mkdir(test_dir)
# 猫训练图片所在目录
train_cats_dir = os.path.join(train_dir, 'cats')
os.mkdir(train_cats_dir)
# 狗训练图片所在目录
train_dogs_dir = os.path.join(train_dir, 'dogs')
os.mkdir(train_dogs_dir)
# 猫验证图片所在目录
validation_cats_dir = os.path.join(validation_dir, 'cats')
os.mkdir(validation_cats_dir)
# 狗验证数据集所在目录
validation_dogs_dir = os.path.join(validation_dir, 'dogs')
os.mkdir(validation_dogs_dir)
# 猫测试数据集所在目录
test_cats_dir = os.path.join(test_dir, 'cats')
os.mkdir(test_cats_dir)
# 狗测试数据集所在目录
test_dogs_dir = os.path.join(test_dir, 'dogs')
os.mkdir(test_dogs_dir)
# 将前1000张猫图像复制到train_cats_dir
fnames = ['cat.{}.jpg'.format(i) for i in range(1000)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(train_cats_dir, fname)
shutil.copyfile(src, dst)
# 将下500张猫图像复制到validation_cats_dir
fnames = ['cat.{}.jpg'.format(i) for i in range(1000, 1500)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(validation_cats_dir, fname)
shutil.copyfile(src, dst)
# 将下500张猫图像复制到test_cats_dir
fnames = ['cat.{}.jpg'.format(i) for i in range(1500, 2000)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(test_cats_dir, fname)
shutil.copyfile(src, dst)
# 将前1000张狗图像复制到train_dogs_dir
fnames = ['dog.{}.jpg'.format(i) for i in range(1000)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(train_dogs_dir, fname)
shutil.copyfile(src, dst)
# 将下500张狗图像复制到validation_dogs_dir
fnames = ['dog.{}.jpg'.format(i) for i in range(1000, 1500)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(validation_dogs_dir, fname)
shutil.copyfile(src, dst)
# 将下500张狗图像复制到test_dogs_dir
fnames = ['dog.{}.jpg'.format(i) for i in range(1500, 2000)]
for fname in fnames:
src = os.path.join(original_dataset_dir, fname)
dst = os.path.join(test_dogs_dir, fname)
shutil.copyfile(src, dst)
得到新的分类文件夹
3.输入如下代码,统计图片数量
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)))
print('total test cat images:', len(os.listdir(test_cats_dir)))
print('total test dog images:', len(os.listdir(test_dogs_dir)))
三、基准模型
1.网络模型构建
#网络模型构建
from keras import layers
from keras import models
#keras的序贯模型
model = models.Sequential()
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(32, (3, 3), activation='relu',
input_shape=(150, 150, 3)))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核2*2,激活函数relu
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#flatten层,用于将多维的输入一维化,用于卷积层和全连接层的过渡
model.add(layers.Flatten())
#全连接,激活函数relu
model.add(layers.Dense(512, activation='relu'))
#全连接,激活函数sigmoid
model.add(layers.Dense(1, activation='sigmoid'))
#显示
model.summary()
2.配置训练方法
from keras import optimizers
model.compile(loss='binary_crossentropy',
optimizer=optimizers.RMSprop(lr=1e-4),
metrics=['acc'])
3.将文件中图像转换成所需格式
from keras.preprocessing.image import ImageDataGenerator
# 所有图像将按1/255重新缩放
train_datagen = ImageDataGenerator(rescale=1./255)
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
# 这是目标目录
train_dir,
# 所有图像将调整为150x150
target_size=(150, 150),
batch_size=20,
# 因为我们使用二元交叉熵损失,我们需要二元标签
class_mode='binary')
validation_generator = test_datagen.flow_from_directory(
validation_dir,
target_size=(150, 150),
batch_size=20,
class_mode='binary')
查看结果
#查看上面对于图片预处理的处理结果
for data_batch, labels_batch in train_generator:
print('data batch shape:', data_batch.shape)
print('labels batch shape:', labels_batch.shape)
break
4.训练模型并保存
#模型训练过程
history = model.fit_generator(
train_generator,
steps_per_epoch=100,
epochs=30,
validation_data=validation_generator,
validation_steps=50)
#保存训练得到的的模型
model.save('路径')
5.查看可视化结果
#对于模型进行评估,查看预测的准确性
import matplotlib.pyplot as plt
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(len(acc))
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()
四、基准模型调整
1.图形增强
from keras.preprocessing.image import ImageDataGenerator
datagen = ImageDataGenerator(
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')
2.网络模型增加一层dropout
#网络模型构建
from keras import layers
from keras import models
#keras的序贯模型
model = models.Sequential()
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(32, (3, 3), activation='relu',
input_shape=(150, 150, 3)))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核2*2,激活函数relu
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#卷积层,卷积核是3*3,激活函数relu
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
#最大池化层
model.add(layers.MaxPooling2D((2, 2)))
#flatten层,用于将多维的输入一维化,用于卷积层和全连接层的过渡
model.add(layers.Flatten())
#退出层
model.add(layers.Dropout(0.5))
#全连接,激活函数relu
model.add(layers.Dense(512, activation='relu'))
#全连接,激活函数sigmoid
model.add(layers.Dense(1, activation='sigmoid'))
#输出模型各层的参数状况
model.summary()
from keras import optimizers
model.compile(loss='binary_crossentropy',
optimizer=optimizers.RMSprop(lr=1e-4),
metrics=['acc'])
3.训练模型
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,)
# Note that the validation data should not be augmented!
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
# This is the target directory
train_dir,
# All images will be resized to 150x150
target_size=(150, 150),
batch_size=32,
# Since we use binary_crossentropy loss, we need binary labels
class_mode='binary')
validation_generator = test_datagen.flow_from_directory(
validation_dir,
target_size=(150, 150),
batch_size=32,
class_mode='binary')
history = model.fit_generator(
train_generator,
steps_per_epoch=100,
epochs=100,
validation_data=validation_generator,
validation_steps=50)
model.save('路径')
4.结果可视化
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(len(acc))
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()
在模型结构中加入一层Dropout后,调整并重新训练改为100个epoch。重新训练后的结果如图所示。可以看出,准确率由基准的67%提高到82%。
五、使用VGG19实现猫狗分类
1.初始化VGG19网络实例
from keras.applications import VGG19
conv_base = VGG19(weights = 'imagenet',include_top = False,input_shape=(150, 150, 3))
conv_base.summary()
2.将猫狗数据集传递给神经网络
import os
import numpy as np
from keras.preprocessing.image import ImageDataGenerator
# 数据集分类后的目录
base_dir = '路径'
train_dir = os.path.join(base_dir, 'train')
validation_dir = os.path.join(base_dir, 'validation')
test_dir = os.path.join(base_dir, 'test')
datagen = ImageDataGenerator(rescale = 1. / 255)
batch_size = 20
def extract_features(directory, sample_count):
features = np.zeros(shape = (sample_count, 4, 4, 512))
labels = np.zeros(shape = (sample_count))
generator = datagen.flow_from_directory(directory, target_size = (150, 150),
batch_size = batch_size,
class_mode = 'binary')
i = 0
for inputs_batch, labels_batch in generator:
#把图片输入VGG16卷积层,让它把图片信息抽取出来
features_batch = conv_base.predict(inputs_batch)
#feature_batch 是 4*4*512结构
features[i * batch_size : (i + 1)*batch_size] = features_batch
labels[i * batch_size : (i+1)*batch_size] = labels_batch
i += 1
if i * batch_size >= sample_count :
#for in 在generator上的循环是无止境的,因此我们必须主动break掉
break
return features , labels
#extract_features 返回数据格式为(samples, 4, 4, 512)
train_features, train_labels = extract_features(train_dir, 2000)
validation_features, validation_labels = extract_features(validation_dir, 1000)
test_features, test_labels = extract_features(test_dir, 1000)
3.将抽取的特征进行分类训练
from keras import models
from keras import layers
from keras import optimizers
#构造我们自己的网络层对输出数据进行分类
model = models.Sequential()
model.add(layers.Dense(256, activation='relu', input_dim = 4 * 4 * 512))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(1, activation = 'sigmoid'))
model.compile(optimizer=optimizers.RMSprop(lr = 2e-5), loss = 'binary_crossentropy', metrics = ['acc'])
history = model.fit(train_features, train_labels, epochs = 30, batch_size = 20,
validation_data = (validation_features, validation_labels))
4.可视化训练结果和校验结果
import matplotlib.pyplot as plt
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(1, len(acc) + 1)
plt.plot(epochs, acc, 'bo', label = 'Train_acc')
plt.plot(epochs, val_acc, 'b', label = 'Validation acc')
plt.title('Trainning and validation accuracy')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'bo', label = 'Training loss')
plt.plot(epochs, val_loss, 'b', label = 'Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()
六、参考资料
1.基于Tensorflow和Keras实现卷积神经网络CNN
2.从头开始训练CNN进行图像分类的完整过程(猫狗大战为例,使用Keras框架)