【theano】03-layer层定义

 

layer()是一个对象

 

#to define the layer like this:

'''
l1 = Layer(inputs,in_size=1,out_size=10,activation_function='relu')
#输入层定义
l2 = Layer(l1.outputs,10,1,None)    
#隐藏层,输入是l1的输出，in_size=10,out_size=1，无激励函数
'''

 

coding:

class Layer(object):    
def __init__(self,inputs,in_size,out_size,activation_function=None):
        self.W = theano.shared(np.random.normal(0,1,(in_size,out_size)))
        #初始化weights使用random.normal乱序随机效果比全0更好
        self.b = theano.shared(np.zeros((out_size,))+0.1)
        #初始化biases    
        self.Wx_plus_b = T.dot(inputs,self.W)+self.b
        #y=Wx+b
        #激励一下
        self.activation_function = activation_function
        if activation_function is None:
            self.outpus = self.Wx_plus_b
        else:
            self.outputs = activation_function(self.Wx_plus_b)

 

还有更多的function可以继续def定义...