深度学习用于计算机视觉(1)

深度学习用于计算机视觉从零开始使用GPU训练模型

import tensorflow as tf
from keras import layers
from keras import models
import numpy as np
import matplotlib.pyplot as plt

Using TensorFlow backend.

使用GPU加速训练

import os
from tensorflow.python.client import device_lib
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "99"

if __name__ == "__main__":
    print(device_lib.list_local_devices())

[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 10969254508978968812
, name: "/device:GPU:0"
device_type: "GPU"
memory_limit: 3156659404
locality {
  bus_id: 1
  links {
  }
}
incarnation: 17995621738846306774
physical_device_desc: "device: 0, name: GeForce GTX 1050 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1"
]

1、使用卷积神经网络训练MNIST模型

from keras.datasets import mnist
from keras.utils import to_categorical
#(train_images, train_labels), (test_images, test_labels) = mnist.load_data()

def load_data(filename):
    f = np.load('mnist.npz')
    return (f['x_train'], f['y_train']), (f['x_test'], f['y_test'])
            
(train_images, train_labels), (test_images, test_labels) = load_data('mnist.npz')
train_images = train_images.reshape((60000,28,28,1))
train_images = train_images.astype('float32')/255
train_labels = to_categorical(train_labels)

test_images = test_images.reshape((10000,28,28,1))
test_images = test_images.astype('float32')/255
test_labels = to_categorical(test_labels)

# 搭建模型
model = models.Sequential()
model.add(layers.Conv2D(32,(3,3), activation='relu', input_shape=(28,28,1))) ##(3*3+1)*32=320
model.add(layers.MaxPool2D((2,2)))

model.add(layers.Conv2D(64, (3,3), activation='relu'))  ## 3*3*32*64+64
model.add(layers.MaxPool2D((2, 2)))

model.add(layers.Conv2D(64, (3,3), activation='relu')) ## 3*3*64*64+64
model.add(layers.Flatten())

model.add(layers.Dense(64, activation='relu'))
model.add(layers.Dense(10,activation='softmax')) ## 64*10 +10


#编译模型
model.compile(optimizer='rmsprop',
             loss='categorical_crossentropy',
             metrics=['accuracy'])

#训练模型
model.fit(train_images, train_labels, epochs=5, batch_size=64)

Epoch 1/5
60000/60000 [==============================] - 15s 245us/step - loss: 0.1650 - acc: 0.9486
Epoch 2/5
60000/60000 [==============================] - 7s 125us/step - loss: 0.0476 - acc: 0.9853
Epoch 3/5
60000/60000 [==============================] - 7s 123us/step - loss: 0.0335 - acc: 0.9899
Epoch 4/5
60000/60000 [==============================] - 8s 139us/step - loss: 0.0256 - acc: 0.9924
Epoch 5/5
60000/60000 [==============================] - 7s 119us/step - loss: 0.0198 - acc: 0.9938

model.summary()

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_1 (Conv2D)            (None, 26, 26, 32)        320       
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 13, 13, 32)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 11, 11, 64)        18496     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 5, 5, 64)          0         
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 3, 3, 64)          36928     
_________________________________________________________________
flatten_1 (Flatten)          (None, 576)               0         
_________________________________________________________________
dense_1 (Dense)              (None, 64)                36928     
_________________________________________________________________
dense_2 (Dense)              (None, 10)                650       
=================================================================
Total params: 93,322
Trainable params: 93,322
Non-trainable params: 0
_________________________________________________________________

test_loss, test_acc = model.evaluate(test_images, test_labels)
print(test_loss, '   ', test_acc)

10000/10000 [==============================] - 1s 85us/step
0.029478431314995213     0.9922

2、从头开始训练一个猫狗分类器

1、准备数据

import os , shutil

train_dir = r"F:\Data_Set\train_dir"
validation_dir = r"F:\Data_Set\validation_dir"
test_dir = r"F:\Data_Set\test_dir"

train_cats_dir = r"F:\Data_Set\train_dir\train_cats_dir"
train_dogs_dir = r"F:\Data_Set\train_dir\train_dogs_dir"

validation_cats_dir = r"F:\Data_Set\validation_dir\validation_cats_dir"
validation_dogs_dir = r"F:\Data_Set\validation_dir\validation_dogs_dir"

test_cats_dir = r"F:\Data_Set\test_dir\test_cats_dir"
test_dogs_dir = r"F:\Data_Set\test_dir\test_dogs_dir"

# 查看各个文件夹内的文件数量
print("猫的训练样本总共有：{}张图像".format(len(os.listdir(train_cats_dir))))
print("狗的训练样本总共有：{}张图像".format(len(os.listdir(train_dogs_dir))))
print("猫的验证样本总共有：{}张图像".format(len(os.listdir(validation_cats_dir))))
print("狗的验证样本总共有：{}张图像".format(len(os.listdir(validation_cats_dir))))
print("猫的测试样本总共有：{}张图像".format(len(os.listdir(test_cats_dir))))
print("狗的测试样本总共有：{}张图像".format(len(os.listdir(test_cats_dir))))

猫的训练样本总共有：1000张图像
狗的训练样本总共有：1000张图像
猫的验证样本总共有：500张图像
狗的验证样本总共有：500张图像
猫的测试样本总共有：500张图像
狗的测试样本总共有：500张图像

2、构建模型

model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(150, 150, 3)))#3*3*32*3+32
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(64,(3,3), activation='relu'))#3*3*64*32+64
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(128, (3,3), activation='relu'))
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(128, (3,3), activation='relu'))
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Flatten())
model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

model.summary()

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_9 (Conv2D)            (None, 148, 148, 32)      896       
_________________________________________________________________
max_pooling2d_9 (MaxPooling2 (None, 74, 74, 32)        0         
_________________________________________________________________
conv2d_10 (Conv2D)           (None, 72, 72, 64)        18496     
_________________________________________________________________
max_pooling2d_10 (MaxPooling (None, 36, 36, 64)        0         
_________________________________________________________________
conv2d_11 (Conv2D)           (None, 34, 34, 128)       73856     
_________________________________________________________________
max_pooling2d_11 (MaxPooling (None, 17, 17, 128)       0         
_________________________________________________________________
conv2d_12 (Conv2D)           (None, 15, 15, 128)       147584    
_________________________________________________________________
max_pooling2d_12 (MaxPooling (None, 7, 7, 128)         0         
_________________________________________________________________
flatten_3 (Flatten)          (None, 6272)              0         
_________________________________________________________________
dense_5 (Dense)              (None, 512)               3211776   
_________________________________________________________________
dense_6 (Dense)              (None, 1)                 513       
=================================================================
Total params: 3,453,121
Trainable params: 3,453,121
Non-trainable params: 0
_________________________________________________________________

3、模型编译

from keras import optimizers
model.compile(optimizer=optimizers.RMSprop(lr=1e-4),
             loss='binary_crossentropy',
             metrics=['acc'])

4、数据预处理

• 数据是以JPEG文件格式保存在硬盘中，所以需要将数据预处理，主要步骤如下：
  • 读取图像文件
  • 将JPEG的文件编码解码为RGB像素网格
  • 将这些像素网格转化为浮点数张量
  • 将0-255的像素值压缩到0-1的区间
• 上述步骤可以使用keras自带的图像预处理工具完成

from keras.preprocessing.image import ImageDataGenerator

train_datagen = ImageDataGenerator(rescale=1./255)#将所有图像像素值缩放
test_datagen = ImageDataGenerator(rescale=1./255)

train_generator = train_datagen.flow_from_directory(train_dir,
                                                   target_size=(150,150),#将目标文件调整为150*150
                                                   batch_size=20,
                                                   class_mode='binary')
validation_generator = test_datagen.flow_from_directory(validation_dir,
                                                       target_size=(150,150),
                                                       batch_size=20,
                                                       class_mode='binary')

Found 2000 images belonging to 2 classes.
Found 1000 images belonging to 2 classes.

#查看一下生成器的结果
for data_batch, labels_batch in train_generator:
    print("data batch shape: ", data_batch.shape)
    print('labels batch shape: ', labels_batch.shape)
    break

data batch shape:  (20, 150, 150, 3)
labels batch shape:  (20,)

5、训练模型

• 使用fit_generator()方法来训练

history = model.fit_generator(train_generator,
                             steps_per_epoch=100,
                             epochs=20,
                             validation_data=validation_generator,
                             validation_steps=50)

Epoch 1/20
100/100 [==============================] - 8s 85ms/step - loss: 0.6895 - acc: 0.5405 - val_loss: 0.6759 - val_acc: 0.5220
 .
 .
 .
Epoch 18/20
100/100 [==============================] - 8s 80ms/step - loss: 0.2282 - acc: 0.9120 - val_loss: 0.6205 - val_acc: 0.7240
Epoch 19/20
100/100 [==============================] - 9s 89ms/step - loss: 0.2059 - acc: 0.9165 - val_loss: 0.7097 - val_acc: 0.7170
Epoch 20/20
100/100 [==============================] - 8s 80ms/step - loss: 0.1814 - acc: 0.9370 - val_loss: 0.6499 - val_acc: 0.7400

# 保存模型
model.save_weights("./model_weights/cats_and_dogs_small_1.h5")

6、绘制训练过程中的损失和精度曲线

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

epoches = range(1, len(acc) + 1)

plt.plot(epoches, acc, 'b-', label='Training acc')
plt.plot(epoches, val_acc, 'r--',label='Validation acc')
plt.title("Training and validation accuracy")
plt.legend()
plt.savefig('./Result_imgs/Training and validation accuracy_1')
plt.show()

plt.plot(epoches, loss, 'b-', label='Training loss')
plt.plot(epoches, val_loss, 'r--',label='Validation loss')
plt.title("Training and validation loss")
plt.legend()
plt.savefig('./Result_imgs/Training and validation loss_1')
plt.show()

val_acc[15]

0.7390000033378601

• 从上图中可以看出，大概在第5轮迭代开始便产生了过拟合。原因之一便是训练的数据太少了

7、使用数据增强技术

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'                            
                            )

from keras.preprocessing import image
fnames = [os.path.join(train_cats_dir, fname) for fname in os.listdir(train_cats_dir)]
img_path = fnames[3]
img = image.load_img(img_path, target_size=(150,150)) #读取图像并调整大小
x = image.img_to_array(img)
print(x.shape)
x = x.reshape((1,)+x.shape)# 将其转变为（1，150，150，3）
print(x.shape)

i = 0
for batch in datagen.flow(x, batch_size=1):
    plt.figure(i)
    imgplot = plt.imshow(image.array_to_img(batch[0]))
    plt.savefig('./Result_imgs/image augmentation_{}'.format(i))
    i += 1
    if i % 4 ==0:
        break
    
plt.show()

(150, 150, 3)
(1, 150, 150, 3)

8、使用dropout技术

#重新定义一个包含dropout的模型

model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(150, 150, 3)))#3*3*32*3+32
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(64,(3,3), activation='relu'))#3*3*64*32+64
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(128, (3,3), activation='relu'))
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Conv2D(128, (3,3), activation='relu'))
model.add(layers.MaxPooling2D((2,2)))

model.add(layers.Flatten())
model.add(layers.Dropout(0.5))#####################################

model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

#编译模型
model.compile(optimizer=optimizers.RMSprop(lr=1e-4),
             loss='binary_crossentropy',
             metrics=['acc'])


#准备数据，使用数据增强生成器
# 对训练数据增强
datagen = ImageDataGenerator(rotation_range=40,
                             rescale=1./255,
                             width_shift_range=0.2,
                             height_shift_range=0.2,
                             shear_range=0.2,
                             zoom_range=0.2,
                             horizontal_flip=True,
                             fill_mode= 'nearest'                            
                            )


test_datagen = ImageDataGenerator(rescale=1./255)## 不可以对测试数据增强

train_generator = train_datagen.flow_from_directory(train_dir,
                                                   target_size=(150,150),#将目标文件调整为150*150
                                                   batch_size=32,
                                                   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)

Found 2000 images belonging to 2 classes.
Found 1000 images belonging to 2 classes.
Epoch 1/100
100/100 [==============================] - 12s 117ms/step - loss: 0.6922 - acc: 0.5328 - val_loss: 0.6781 - val_acc: 0.6150
Epoch 2/100
100/100 [==============================] - 11s 105ms/step - loss: 0.6627 - acc: 0.6053 - val_loss: 0.6445 - val_acc: 0.6250
.
.
.
Epoch 99/100
100/100 [==============================] - 11s 114ms/step - loss: 0.0202 - acc: 0.9928 - val_loss: 1.2302 - val_acc: 0.7670
Epoch 100/100
100/100 [==============================] - 11s 111ms/step - loss: 0.0166 - acc: 0.9947 - val_loss: 1.3127 - val_acc: 0.7780

model.save("./model_weights/cats_and_dogs_samll_2.h5")

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

epoches = range(1, len(acc) + 1)

plt.plot(epoches, acc, 'b-', label='Training acc')
plt.plot(epoches, val_acc, 'r--',label='Validation acc')
plt.title("Training and validation accuracy")
plt.legend()
plt.savefig('./Result_imgs/Training and validation accuracy_2')
plt.show()

plt.plot(epoches, loss, 'b-', label='Training loss')
plt.plot(epoches, val_loss, 'r--',label='Validation loss')
plt.title("Training and validation loss")
plt.legend()
plt.savefig('./Result_imgs/Training and validation loss_2')
plt.show()

• 可见模型在20次迭代后产生了过拟合，性能从 【0.73】提升到了【0.76】左右