tensorflow实现MNIST手写数字识别

1.导入模块

import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np

2.加载数据集

mnist = tf.keras.datasets.mnist
(x_train,y_train),(x_test,y_test) = mnist.load_data()
x_train, x_test= x_train/255.0,x_test/255.0

        MNIST是Keras内置的数据集

        Windows操作系统下在C:\Users\user_name\.keras\datasets文件夹中查看。

2.1查看数据集

x_train.shape
x_test.shape
y_train.shape
y_train.shape

3.构建模型

model = tf.keras.Sequential()
model.add(tf.keras.layers.Flatten(input_shape=(28, 28)))  
model.add(tf.keras.layers.Dense(128, activation='relu'))  
model.add(tf.keras.layers.Dense(10, activation='softmax'))
model.summary()

 4.编译模型

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

 5.训练

model.fit(x_train, y_train,batch_size=64, epochs=5)

 

6.评估模型

model.evaluate(x_test,  y_test, verbose=2)

 

 7.测试

plt.axis("off")
plt.imshow(x_test[0])
plt.show()

 

X_test =  x_test.reshape(10000,28*28)
X_test_batch = np.expand_dims(X_test[0], axis=0)
np.argmax(model.predict(X_test_batch))