T2周：彩色图片分类

>- **🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/0dvHCaOoFnW8SCp3JpzKxg) 中的学习记录博客**
>- **🍖 原作者：[K同学啊](https://mtyjkh.blog.csdn.net/)**

我的环境

语言环境：Python 3.9.13

编译器：jupyter notebook

深度学习环境：Tensorflow-gpu 2.7.0

一、前期准备

1.设置GPU（没有GPU的可以忽略）

#设置GPU
import tensorflow as tf
gpus = tf.config.list_physical_devices("GPU")

2.导入数据

#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.cifar10.load_data()

3.归一化

#归一化
# 将像素的值标准化至0到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

((50000, 32, 32, 3), (10000, 32, 32, 3), (50000, 1), (10000, 1))

4.可视化

# 可视化。
class_names = ['airpalne', 'automobile', 'bird','cat','deer','dog','frog','horse','ship', 'truck']

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][0]])
plt.show()

二、构建CNN网络

构建CNN网络模型

#构建CNN网络模型
model = models.Sequential([
    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)),  #卷积层1，卷积核3*3
    layers.MaxPooling2D((2, 2)),                   #池化层1，2*2采样
    layers.Conv2D(64, (3, 3), activation='relu'),  #卷积层2，卷积核3*3
    layers.MaxPooling2D((2, 2)),                   #池化层2，2*2采样
    layers.Conv2D(64, (3, 3), activation='relu'),  #卷积层3，卷积核3*3
                  
    layers.Flatten(),                            
    layers.Dense(64, activation='relu'),   #Flatten层，连接卷积层与全连接层
    layers.Dense(10)
])

model.summary() #打印网络结构

Model: "sequential_1"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 conv2d_3 (Conv2D)           (None, 30, 30, 32)        896       
                                                                 
 max_pooling2d_2 (MaxPooling  (None, 15, 15, 32)       0         
 2D)                                                             
                                                                 
 conv2d_4 (Conv2D)           (None, 13, 13, 64)        18496     
                                                                 
 max_pooling2d_3 (MaxPooling  (None, 6, 6, 64)         0         
 2D)                                                             
                                                                 
 conv2d_5 (Conv2D)           (None, 4, 4, 64)          36928     
                                                                 
 flatten_1 (Flatten)         (None, 1024)              0         
                                                                 
 dense_2 (Dense)             (None, 64)                65600     
                                                                 
 dense_3 (Dense)             (None, 10)                650       
                                                                 
=================================================================
Total params: 122,570
Trainable params: 122,570
Non-trainable params: 0
_________________________________________________________________

三、编译

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

四、训练模型

#训练模型
history = model.fit(train_images, train_labels, epochs=10,
                    validation_data=(test_images, test_labels))

Epoch 1/10
1563/1563 [==============================] - 4s 2ms/step - loss: 1.5697 - accuracy: 0.4254 - val_loss: 1.2839 - val_accuracy: 0.5417
Epoch 2/10
1563/1563 [==============================] - 4s 2ms/step - loss: 1.2034 - accuracy: 0.5723 - val_loss: 1.1407 - val_accuracy: 0.5947
Epoch 3/10
1563/1563 [==============================] - 4s 2ms/step - loss: 1.0472 - accuracy: 0.6318 - val_loss: 1.0527 - val_accuracy: 0.6282
Epoch 4/10
1563/1563 [==============================] - 4s 3ms/step - loss: 0.9459 - accuracy: 0.6668 - val_loss: 0.9622 - val_accuracy: 0.6634
Epoch 5/10
1563/1563 [==============================] - 4s 3ms/step - loss: 0.8667 - accuracy: 0.6952 - val_loss: 0.9198 - val_accuracy: 0.6762
Epoch 6/10
1563/1563 [==============================] - 4s 2ms/step - loss: 0.8090 - accuracy: 0.7161 - val_loss: 0.8993 - val_accuracy: 0.6831
Epoch 7/10
1563/1563 [==============================] - 4s 3ms/step - loss: 0.7539 - accuracy: 0.7365 - val_loss: 0.8768 - val_accuracy: 0.6982
Epoch 8/10
1563/1563 [==============================] - 4s 2ms/step - loss: 0.7082 - accuracy: 0.7507 - val_loss: 0.8688 - val_accuracy: 0.7049
Epoch 9/10
1563/1563 [==============================] - 4s 3ms/step - loss: 0.6712 - accuracy: 0.7647 - val_loss: 0.8618 - val_accuracy: 0.7039
Epoch 10/10
1563/1563 [==============================] - 4s 3ms/step - loss: 0.6319 - accuracy: 0.7779 - val_loss: 0.8827 - val_accuracy: 0.7008

五、预测模型

#预测模型
plt.imshow(test_images[1])

<matplotlib.image.AxesImage at 0x14adecf0ac0>

import numpy as np

pre = model.predict(test_images)
print(class_names[np.argmax(pre[1])])

ship

六、模型评估

#模型评估
import matplotlib.pyplot as plt

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

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

313/313 - 0s - loss: 0.8827 - accuracy: 0.7008 - 444ms/epoch - 1ms/step

print(test_acc)

0.7008000016212463