Depth study of 100 cases - convolution neural network (CNN) to realize the clothing image classification

活动地址：CSDN21天学习挑战赛

深度学习100例——卷积神经网络(CNN)服装图像分类

1. 前期准备工作

我的环境：

1.1 设置GPU

import tensorflow as tf
gpus = tf.config.list_physical_devices("GPU") # tf.config.list_physical_devices# 获得当前主机上某种特定运算设备类型（如 GPU 或 CPU ）的列表

if gpus:
    gpu0 = gpus[0] #如果有多个GPU,仅使用第0个GPU
    tf.config.experimental.set_memory_growth(gpu0, True) #设置GPU显存用量按需使用
    tf.config.set_visible_devices([gpu0],"GPU") # 设置可见设备列表

There is only one current machineGPU

1.2 导入数据

import tensorflow as tf
from tensorflow.keras import datasets, layers, models
import matplotlib.pyplot as plt

(train_images, train_labels), (test_images, test_labels) = datasets.fashion_mnist.load_data()

查看一下数据

(train_images, train_labels) → 训练集

(test_images, test_labels) → 测试集

和mnist数据集类似,60000个训练样本,10000个测试样本,且70000张图片均是28 x 28的像素图.

The pixel value of each image is between 0 - 255 之间,标签是整数数组,介于 0 - 9 之间.

Label correspondence table：

标签	类	标签	类
0	T恤	5	凉鞋
1	裤子	6	衬衫
2	套头衫	7	运动鞋
3	连衣裙	8	包
4	外套	9	短靴

1.3 数据归一化

→ Normalize the pixel's value to 0 - 1的区间内.

train_images, test_images = train_images / 255.0, test_images / 255.0
train_images.shape,test_images.shape,train_labels.shape,test_labels.shape

归一化前：

归一化后：

1.4 调整图片格式

train_images = train_images.reshape((60000, 28, 28, 1))
test_images = test_images.reshape((10000, 28, 28, 1))

train_images.shape,test_images.shape,train_labels.shape,test_labels.shape

数据维度转换,增加通道数

1.5 Visually view data images

class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',
               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']

plt.figure(figsize=(20,10))
for i in range(20):
    plt.subplot(5,10,i+1)
    plt.xticks([])
    plt.yticks([])
    plt.grid(False)
    plt.imshow(train_images[i], cmap=plt.cm.binary)
    plt.xlabel(class_names[train_labels[i]])
plt.show()

同样只展示0-19Indexed sample image.

2. 构建CNN网络

2.1 构建网络

model = models.Sequential([
    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)), #卷积层1,卷积核3 x 3
    layers.MaxPooling2D((2, 2)),                   #池化层1,2 x 2采样
    layers.Conv2D(64, (3, 3), activation='relu'),  #卷积层2,卷积核3 x 3
    layers.MaxPooling2D((2, 2)),                   #池化层2,2 x 2采样
    layers.Conv2D(64, (3, 3), activation='relu'),  #卷积层3,卷积核3 x 3
    
    layers.Flatten(),                      #Flatten层,连接卷积层与全连接层
    layers.Dense(64, activation='relu'),   #全连接层,特征进一步提取
    layers.Dense(10)                       #输出层,输出预期结果
])

model.summary()  # 打印网络结构

2.2 模型说明

各层的作用：

• 输入层：将数据输入到训练网络
• 卷积层：使用卷积核提取图片的特征
• 池化层：进行下采样,用更高层的抽象表示图像特征
• Flatten层：将多维的输入一维化,常用在卷积层到全连接层的过渡
• 全连接层：特征提取器
• 输出层：输出结果

3. 编译模型

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

compile()方法用于设置训练时,使用的优化器optimizer、损失函数loss、准确率评测标准metrics

SparseCategoricalCrossentropy → 交叉熵损失函数,当from_logits参数为True时,会使用softmax将预测y转换为概率.

4.训练模型

history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

5. 模型预测

The predicted result of the model is an inclusion10个数字的数组,represents a model pair10of each of the different garments 置信度.

pre = model.predict(test_images)

6. 模型评估

plt.plot(history.history['accuracy'], label='accuracy')
plt.plot(history.history['val_accuracy'], label = 'val_accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.ylim([0.5, 1])
plt.legend(loc='lower right')
plt.show()

test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)