2.序贯模型
序贯模型(Sequential Model)是多个网络层的线性堆叠,也就是”一条路走到黑”。但是不适合网络之间穿插情况或者多输入多输出情况。
序贯模型有两种方式构建网络
2.1简单的汉堡式叠法
可以通过Sequential模型传递一个layer的list来构造序贯模型。
#两种方式1.汉堡生成
model = Sequential([
Dense(32,input_dim=784),
Activation('relu'),
Dense(10),
Activation('softmax')
])
#完成网络构建
2.2加菜式构建模型
或者通过.add()方法一个个将layer加入到模型中
#2.手动加菜
model = Sequential()
model.add(Dense(32,input_shape=(784,)))
model.add(Activation('relu'))
#完成网络构建
用序贯模型跑一个小demo基于多层感知器的softmax多分类
from keras.models import Sequential
#Sequential声明使用序贯模型
from keras.layers import Dense, Dropout, Activation
#导入层Dropout随机失活 Dense全连接层、Activation激活函数
from keras.optimizers import SGD
#SGD随机梯度下降
import keras
import numpy as np
x_train = np.random.random((1000, 20))
#随机训练样本,样本维度20维
y_train = keras.utils.to_categorical(np.random.randint(10, size=(1000, 1)), num_classes=10)
#映射成10个类别
x_test = np.random.random((100, 20))
y_test = keras.utils.to_categorical(np.random.randint(10, size=(100, 1)), num_classes=10)
model = Sequential()
#建立模型
#添加网络层
model.add(Dense(64, activation='relu', input_dim=20))
# Dense(64)神经元有64个
#input_dim必须与样本维度对应
#activation激活函数
model.add(Dropout(0.5))#随机失活
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.8, nesterov=True)#制定优化器参数学习率0.01,动量0.8
model.compile(loss='categorical_crossentropy',
optimizer=sgd,
metrics=['accuracy'])
model.fit(x_train, y_train,
epochs=20,
batch_size=128)
score = model.evaluate(x_test, y_test, batch_size=128)
#结果acc停留在0.1060左右,跑了N次结果都停留在10%左右
用序贯模型构建CNN
import numpy as np
import keras
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
#flatten作用是方便后面接全连接层
from keras.layers import Conv2D, MaxPooling2D
#Conv2D构建2D卷积层,maxpooling2D构建池化层
from keras.optimizers import SGD
准备数据
x_train = np.random.random((100, 100, 100, 3))
#100张图片宽高100,3彩色图RGB三个颜色通道
y_train = keras.utils.to_categorical(np.random.randint(10, size=(100, 1)), num_classes=10)
x_test = np.random.random((20, 100, 100, 3))
y_test = keras.utils.to_categorical(np.random.randint(10, size=(20, 1)), num_classes=10)
#模型构建
model = Sequential()
model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(100, 100, 3)))
model.add(Conv2D(32, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
#开始训练测试
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd)
model.fit(x_train, y_train, batch_size=32, epochs=10)
score = model.evaluate(x_test, y_test, batch_size=32)