小編給大家分享一下python中實(shí)現(xiàn)BP神經(jīng)網(wǎng)絡(luò)原理是什么，希望大家閱讀完這篇文章之后都有所收獲，下面讓我們一起去探討吧！
本文主要講如何不依賴TenserFlow等高級API實(shí)現(xiàn)一個(gè)簡單的神經(jīng)網(wǎng)絡(luò)來做分類，所有的代碼都在下面；在構(gòu)造的數(shù)據(jù)（通過程序構(gòu)造）上做了驗(yàn)證，經(jīng)過1個(gè)小時(shí)的訓(xùn)練分類的準(zhǔn)確率可以達(dá)到97%。
先來說說原理
網(wǎng)絡(luò)構(gòu)造
上面是一個(gè)簡單的三層網(wǎng)絡(luò)；輸入層包含節(jié)點(diǎn)X1 , X2；隱層包含H1，H2；輸出層包含O1。
輸入節(jié)點(diǎn)的數(shù)量要等于輸入數(shù)據(jù)的變量數(shù)目。
隱層節(jié)點(diǎn)的數(shù)量通過經(jīng)驗(yàn)來確定。
如果只是做分類，輸出層一般一個(gè)節(jié)點(diǎn)就夠了。
從輸入到輸出的過程
1.輸入節(jié)點(diǎn)的輸出等于輸入，X1節(jié)點(diǎn)輸入x1時(shí)，輸出還是x1.
2. 隱層和輸出層的輸入I為上層輸出的加權(quán)求和再加偏置，輸出為f(I) ， f為激活函數(shù)，可以取sigmoid。H1的輸出為 sigmoid(w1x1 + w2x2 + b)
誤差反向傳播的過程
Python實(shí)現(xiàn)
構(gòu)造測試數(shù)據(jù)
重慶分公司
重慶分公司
# -*- coding: utf-8 -*-
import numpy as np
from random import random as rdn
'''
說明：我們構(gòu)造1000條數(shù)據(jù)，每條數(shù)據(jù)有三個(gè)屬性(用a1 , a2 , a3表示)
a1 離散型 取值 1 到 10 ， 均勻分布
a2 離散型 取值 1 到 10 ， 均勻分布
a3 連續(xù)型 取值 1 到 100 ， 且符合正態(tài)分布
各屬性之間獨(dú)立。
共2個(gè)分類(0 , 1)，屬性值與類別之間的關(guān)系如下，
0 : a1 in [1 , 3] and a2 in [4 , 10] and a3 <= 50
1 : a1 in [1 , 3] and a2 in [4 , 10] and a3 > 50
0 : a1 in [1 , 3] and a2 in [1 , 3] and a3 > 30
1 : a1 in [1 , 3] and a2 in [1 , 3] and a3 <= 30
0 : a1 in [4 , 10] and a2 in [4 , 10] and a3 <= 50
1 : a1 in [4 , 10] and a2 in [4 , 10] and a3 > 50
0 : a1 in [4 , 10] and a2 in [1 , 3] and a3 > 30
1 : a1 in [4 , 10] and a2 in [1 , 3] and a3 <= 30
'''
def genData() :
#為a3生成符合正態(tài)分布的數(shù)據(jù)
a3_data = np.random.randn(1000) * 30 + 50
data = []
for i in range(1000) :
#生成a1
a1 = int(rdn()*10) + 1
if a1 > 10 :
a1 = 10
#生成a2
a2 = int(rdn()*10) + 1
if a2 > 10 :
a2 = 10
#取a3
a3 = a3_data[i]
#計(jì)算這條數(shù)據(jù)對應(yīng)的類別
c_id = 0
if a1 <= 3 and a2 >= 4 and a3 <= 50 :
c_id = 0
elif a1 <= 3 and a2 >= 4 and a3 > 50 :
c_id = 1
elif a1 <= 3 and a2 < 4 and a3 > 30 :
c_id = 0
elif a1 <= 3 and a2 < 4 and a3 <= 30 :
c_id = 1
elif a1 > 3 and a2 >= 4 and a3 <= 50 :
c_id = 0
elif a1 > 3 and a2 >= 4 and a3 > 50 :
c_id = 1
elif a1 > 3 and a2 < 4 and a3 > 30 :
c_id = 0
elif a1 > 3 and a2 < 4 and a3 <= 30 :
c_id = 1
else :
print('error')
#拼合成字串
str_line = str(i) + ',' + str(a1) + ',' + str(a2) + ',' + str(a3) + ',' + str(c_id)
data.append(str_line)
return '\n'.join(data)激活函數(shù)
# -*- coding: utf-8 -*- """ Created on Sun Dec 2 14:49:31 2018 @author: congpeiqing """ import numpy as np #sigmoid函數(shù)的導(dǎo)數(shù)為 f(x)*(1-f(x)) def sigmoid(x) : return 1/(1 + np.exp(-x))
網(wǎng)絡(luò)實(shí)現(xiàn)
# -*- coding: utf-8 -*-
"""
Created on Sun Dec 2 14:49:31 2018
@author: congpeiqing
"""
from activation_funcs import sigmoid
from random import random
class InputNode(object) :
def __init__(self , idx) :
self.idx = idx
self.output = None
def setInput(self , value) :
self.output = value
def getOutput(self) :
return self.output
def refreshParas(self , p1 , p2) :
pass
class Neurode(object) :
def __init__(self , layer_name , idx , input_nodes , activation_func = None , powers = None , bias = None) :
self.idx = idx
self.layer_name = layer_name
self.input_nodes = input_nodes
if activation_func is not None :
self.activation_func = activation_func
else :
#默認(rèn)取 sigmoid
self.activation_func = sigmoid
if powers is not None :
self.powers = powers
else :
self.powers = [random() for i in range(len(self.input_nodes))]
if bias is not None :
self.bias = bias
else :
self.bias = random()
self.output = None
def getOutput(self) :
self.output = self.activation_func(sum(map(lambda x : x[0].getOutput()*x[1] , zip(self.input_nodes, self.powers))) + self.bias)
return self.output
def refreshParas(self , err , learn_rate) :
err_add = self.output * (1 - self.output) * err
for i in range(len(self.input_nodes)) :
#調(diào)用子節(jié)點(diǎn)
self.input_nodes[i].refreshParas(self.powers[i] * err_add , learn_rate)
#調(diào)節(jié)參數(shù)
power_delta = learn_rate * err_add * self.input_nodes[i].output
self.powers[i] += power_delta
bias_delta = learn_rate * err_add
self.bias += bias_delta
class SimpleBP(object) :
def __init__(self , input_node_num , hidden_layer_node_num , trainning_data , test_data) :
self.input_node_num = input_node_num
self.input_nodes = [InputNode(i) for i in range(input_node_num)]
self.hidden_layer_nodes = [Neurode('H' , i , self.input_nodes) for i in range(hidden_layer_node_num)]
self.output_node = Neurode('O' , 0 , self.hidden_layer_nodes)
self.trainning_data = trainning_data
self.test_data = test_data
#逐條訓(xùn)練
def trainByItem(self) :
cnt = 0
while True :
cnt += 1
learn_rate = 1.0/cnt
sum_diff = 0.0
#對于每一條訓(xùn)練數(shù)據(jù)進(jìn)行一次訓(xùn)練過程
for item in self.trainning_data :
for i in range(self.input_node_num) :
self.input_nodes[i].setInput(item[i])
item_output = item[-1]
nn_output = self.output_node.getOutput()
#print('nn_output:' , nn_output)
diff = (item_output-nn_output)
sum_diff += abs(diff)
self.output_node.refreshParas(diff , learn_rate)
#print('refreshedParas')
#結(jié)束條件
print(round(sum_diff / len(self.trainning_data) , 4))
if sum_diff / len(self.trainning_data) < 0.1 :
break
def getAccuracy(self) :
cnt = 0
for item in self.test_data :
for i in range(self.input_node_num) :
self.input_nodes[i].setInput(item[i])
item_output = item[-1]
nn_output = self.output_node.getOutput()
if (nn_output > 0.5 and item_output > 0.5) or (nn_output < 0.5 and item_output < 0.5) :
cnt += 1
return cnt/(len(self.test_data) + 0.0)主調(diào)流程
# -*- coding: utf-8 -*-
"""
Created on Sun Dec 2 14:49:31 2018
@author: congpeiqing
"""
import os
from SimpleBP import SimpleBP
from GenData import genData
if not os.path.exists('data'):
os.makedirs('data')
#構(gòu)造訓(xùn)練和測試數(shù)據(jù)
data_file = open('data/trainning_data.dat' , 'w')
data_file.write(genData())
data_file.close()
data_file = open('data/test_data.dat' , 'w')
data_file.write(genData())
data_file.close()
#文件格式：rec_id,attr1_value,attr2_value,attr3_value,class_id
#讀取和解析訓(xùn)練數(shù)據(jù)
trainning_data_file = open('data/trainning_data.dat')
trainning_data = []
for line in trainning_data_file :
line = line.strip()
fld_list = line.split(',')
trainning_data.append(tuple([float(field) for field in fld_list[1:]]))
trainning_data_file.close()
#讀取和解析測試數(shù)據(jù)
test_data_file = open('data/test_data.dat')
test_data = []
for line in test_data_file :
line = line.strip()
fld_list = line.split(',')
test_data.append(tuple([float(field) for field in fld_list[1:]]))
test_data_file.close()
#構(gòu)造一個(gè)二分類網(wǎng)絡(luò) 輸入節(jié)點(diǎn)3個(gè)，隱層節(jié)點(diǎn)10個(gè)，輸出節(jié)點(diǎn)一個(gè)
simple_bp = SimpleBP(3 , 10 , trainning_data , test_data)
#訓(xùn)練網(wǎng)絡(luò)
simple_bp.trainByItem()
#測試分類準(zhǔn)確率
print('Accuracy : ' , simple_bp.getAccuracy())
#訓(xùn)練時(shí)長比較長，準(zhǔn)確率可以達(dá)到97%看完了這篇文章，相信你對“python中實(shí)現(xiàn)BP神經(jīng)網(wǎng)絡(luò)原理是什么”有了一定的了解，如果想了解更多相關(guān)知識，歡迎關(guān)注創(chuàng)新互聯(lián)成都網(wǎng)站設(shè)計(jì)公司行業(yè)資訊頻道，感謝各位的閱讀！
另外有需要云服務(wù)器可以了解下創(chuàng)新互聯(lián)scvps.cn，海內(nèi)外云服務(wù)器15元起步，三天無理由+7*72小時(shí)售后在線，公司持有idc許可證，提供“云服務(wù)器、裸金屬服務(wù)器、高防服務(wù)器、香港服務(wù)器、美國服務(wù)器、虛擬主機(jī)、免備案服務(wù)器”等云主機(jī)租用服務(wù)以及企業(yè)上云的綜合解決方案，具有“安全穩(wěn)定、簡單易用、服務(wù)可用性高、性價(jià)比高”等特點(diǎn)與優(yōu)勢，專為企業(yè)上云打造定制，能夠滿足用戶豐富、多元化的應(yīng)用場景需求。
