Tensorflow学习笔记:CNN篇(4)——CIFAR-10数据集LeNet实现(Keras版)

Tensorflow学习笔记:CNN篇(4)——CIFAR-10数据集LeNet实现(Keras版)


完整代码

from keras.datasets import cifar10
from keras.models import Sequential, load_model
from keras.layers import Dense, Dropout, Activation, Flatten, Convolution2D, MaxPooling2D, ZeroPadding2D
from keras.utils import np_utils
import matplotlib.pyplot as plt
import sys

#参数设定
batch_size = 64
nb_classes = 10
nb_epoch = 50
activation_function = 'relu'
drop_out = 0.4

#输入图片维度
img_rows, img_cols = 32, 32
#RGB
img_channels = 3

#读档
(X_train, y_train), (X_test, y_test) = cifar10.load_data()


#图片预处理
X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255

Y_train = np_utils.to_categorical(y_train, nb_classes)
Y_test = np_utils.to_categorical(y_test, nb_classes)

#模型搭建
model = Sequential()
model.add(Convolution2D(32, 3, 3, activation=activation_function, input_shape=X_train.shape[1:]))
model.add(Convolution2D(32, 3, 3, activation=activation_function))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(drop_out))

model.add(Convolution2D(64, 3, 3, activation=activation_function))
model.add(Convolution2D(64, 3, 3, activation=activation_function))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(drop_out))

model.add(Flatten())
model.add(Dense(512, activation='relu'))
model.add(Dropout(drop_out))
model.add(Dense(nb_classes, activation='softmax'))

model.summary()

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

history = model.fit(X_train, Y_train, batch_size=batch_size,
                    nb_epoch=nb_epoch, validation_data=(X_test, Y_test), shuffle=True)


loss = history.history.get('loss')
acc = history.history.get('acc')
val_loss = history.history.get('val_loss')
val_acc = history.history.get('val_acc')

plt.figure(0)
plt.subplot(121)
plt.plot(range(len(loss)), loss, label="Training")
plt.plot(range(len(val_loss)), val_loss, label='Validation')
plt.title('Loss')
plt.legend(loc='upper left')
plt.subplot(122)
plt.plot(range(len(acc)), acc, label='Training')
plt.plot(range(len(val_acc)), val_acc, label='Validation')
plt.title('Accuracy')
plt.show()
# plt.savefig('00_firstModel.png', dpi=300, format='png')

运行结果

这里写图片描述

  • 3
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 1
    评论
评论 1
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值