卷积神经网络（CNN）实现mnist手写数字识别第一周学习-CSDN博客

🍨 本文为🔗365天深度学习训练营内部限免文章（版权归 K同学啊 所有）
🍦 参考文章地址： 🔗深度学习100例-循环神经网络（RNN）心脏病预测 | 第46天
🍖 作者：K同学啊 | 接辅导、程序定制

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.mnist.load_data()

Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz
11490434/11490434 [==============================] - 35s 3us/step

train_images, test_images = train_images / 255.0, test_images / 255.0

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

((60000, 28, 28), (10000, 28, 28), (60000,), (10000,))

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(train_labels[i])
plt.show()

在这里插入图片描述

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

((60000, 28, 28, 1), (10000, 28, 28, 1), (60000,), (10000,))

model = models.Sequential([
    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),#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.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()

Model: "sequential"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 conv2d (Conv2D)             (None, 26, 26, 32)        320       
                                                                 
 max_pooling2d (MaxPooling2D  (None, 13, 13, 32)       0         
 )                                                               
                                                                 
 conv2d_1 (Conv2D)           (None, 11, 11, 64)        18496     
                                                                 
 max_pooling2d_1 (MaxPooling  (None, 5, 5, 64)         0         
 2D)                                                             
                                                                 
 flatten (Flatten)           (None, 1600)              0         
                                                                 
 dense (Dense)               (None, 64)                102464    
                                                                 
 dense_1 (Dense)             (None, 10)                650       
                                                                 
=================================================================
Total params: 121,930
Trainable params: 121,930
Non-trainable params: 0
_________________________________________________________________

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

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

Epoch 1/10
1875/1875 [==============================] - 30s 16ms/step - loss: 0.1356 - accuracy: 0.9585 - val_loss: 0.0465 - val_accuracy: 0.9850
Epoch 2/10
1875/1875 [==============================] - 33s 18ms/step - loss: 0.0452 - accuracy: 0.9861 - val_loss: 0.0421 - val_accuracy: 0.9870
Epoch 3/10
1875/1875 [==============================] - 37s 20ms/step - loss: 0.0315 - accuracy: 0.9898 - val_loss: 0.0278 - val_accuracy: 0.9905
Epoch 4/10
1875/1875 [==============================] - 34s 18ms/step - loss: 0.0232 - accuracy: 0.9926 - val_loss: 0.0307 - val_accuracy: 0.9904
Epoch 5/10
1875/1875 [==============================] - 31s 17ms/step - loss: 0.0179 - accuracy: 0.9942 - val_loss: 0.0257 - val_accuracy: 0.9912
Epoch 6/10
1875/1875 [==============================] - 31s 16ms/step - loss: 0.0136 - accuracy: 0.9956 - val_loss: 0.0296 - val_accuracy: 0.9900
Epoch 7/10
1875/1875 [==============================] - 31s 17ms/step - loss: 0.0097 - accuracy: 0.9964 - val_loss: 0.0259 - val_accuracy: 0.9932
Epoch 8/10
1875/1875 [==============================] - 32s 17ms/step - loss: 0.0090 - accuracy: 0.9973 - val_loss: 0.0277 - val_accuracy: 0.9927
Epoch 9/10
1875/1875 [==============================] - 30s 16ms/step - loss: 0.0076 - accuracy: 0.9973 - val_loss: 0.0309 - val_accuracy: 0.9916
Epoch 10/10
1875/1875 [==============================] - 30s 16ms/step - loss: 0.0060 - accuracy: 0.9980 - val_loss: 0.0328 - val_accuracy: 0.9915

plt.imshow(test_images[1])

<matplotlib.image.AxesImage at 0x28e9c726160>

在这里插入图片描述

pre = model.predict(test_images) # 对所有测试图片进行预测
pre[1] # 输出第一张图片的预测结果

313/313 [==============================] - 1s 4ms/step

array([  4.8554344,   4.144166 ,  26.35997  , -10.464082 ,  -8.790907 ,
       -15.091766 ,   1.9568326, -10.758024 ,  -1.3119435, -12.41055  ],
      dtype=float32)