Deep Learning 学习笔记（二）：神经网络Python实现

最新推荐文章于 2020-12-07 08:22:21 发布

我好毒

最新推荐文章于 2020-12-07 08:22:21 发布

阅读量1.9k

点赞数 1

分类专栏：机器学习 python 神经网络文章标签：神经网络 python 机器学习

本文链接：https://blog.csdn.net/arsenicer/article/details/43565015

版权

多层神经网络的Python实现。

代码先贴上，编程的东西不解释。

基本理论参考下一篇：Deep Learning 学习笔记（三）：神经网络反向传播算法推导

代码中出现的SupervisedLearningModel、NNLayer和SoftmaxRegression，请参考上一篇笔记：Deep Learning 学习笔记（一）——softmax Regression

多层神经网络：

import numpy as np
from NNBase import NNLayer
from softmax import SoftmaxRegression
from dp.supervised import NNBase
from time import time
class MNN(NNBase.SupervisedLearningModel):
    '''
    classdocs
    '''
    def __init__(self, params):
        '''
        Constructor
        parameters:
            params     - the network configuration, dict
            params.inputSize        - dimension of input features
            params.outputSize        - number of output classes
            params.layerSizes        - an array, sizes of all layer, including all hidden layers and output layer
            params.Lambda            -scaling parameter for l2 weight regularization penalty
            params.activitionFunc    -which type of activation function to use in hidden layers
        
        '''         
        layerSizes =  params['layerSizes']
        self.numLayers = len(layerSizes)
        self.allLayers = []
        self.X=0
        
        #initialize all hidden layers
        inputSize = params['inputSize']
        for i in range(self.numLayers-1):
            layer = NNLayer(inputSize,layerSizes[i],params['Lambda&#