from keras.datasets import mnist
#下载手写数字数据库,分为训练集和测试集
(X_train, Y_train), (X_test, Y_test) = mnist.load_data()
x_val = X_train[10000:12000]
y_val = Y_train[10000:12000]
x_train = X_train[:10000]
y_train = Y_train[:10000]
import tensorflow as tf
from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
# 创建一个Sequential模型
model = Sequential()
# 添加第一个卷积层
model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)))
# 添加池化层
model.add(MaxPooling2D((2, 2)))
# 添加第二个卷积层
model.add(Conv2D(64, (3, 3), activation='relu'))
# 再次添加池化层
model.add(MaxPooling2D((2, 2)))
# 将前面的层展平为一维向量
model.add(Flatten())
# 添加全连接层
model.add(Dense(64, activation='relu'))
# 输出层
model.add(Dense(10, activation="softmax"))
# 编译模型
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
# 打印模型结构
model.summary()
#训练模型
train_history = model.fit(x_train, y_train, validation_data=(x_val, y_val), epochs=20, batch_size=32, verbose=2)
#测试模型
test_loss, test_acc = model.evaluate(X_test, Y_test)
print('Test accuracy:', test_acc)
CNN实现手写数字集分类
最新推荐文章于 2024-10-20 21:05:16 发布