bp神经网络

import math
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt



data_tr = pd.read_csv('3.3 data_tr.txt')  # 训练集样本
data_te = pd.read_csv('3.3 data_te.txt')  # 测试集样本
n = len(data_tr)
yita = 0.1  # 学习率

out_in = np.array([0.0, 0, 0, 0, -1])   # 输出层的输入，即隐层的输出
w_mid = np.zeros([3,4])  # 隐层神经元的权值&阈值
w_out = np.zeros([5])     # 输出层神经元的权值阈值

delta_w_out = np.zeros([5])      # 输出层权值、阈值的修正量
delta_w_mid = np.zeros([3,4])   # 中间层权值、阈值的修正量
Err = []

def set_chinese(): # 设置字体
    import matplotlib
    matplotlib.rcParams['font.sans-serif'] = ['SimHei']
    matplotlib.rcParams['axes.unicode_minus'] = False

def sigmoid(x):    # 激活函数
    return 1/(1+math.exp(-x))

for j in range(500):
    error = []
    for it in range(n):
        net_in = np.array([data_tr.iloc[it, 0], data_tr.iloc[it, 1], -1])  # 网络输入
        real = data_tr.iloc[it, 2]
        for i in range(4):
            out_in[i] = sigmoid(sum(net_in * w_mid[:, i]))  # 从输入到隐层的传输过程
        res = sigmoid(sum(out_in * w_out))   # 模型预测值
        error.append(abs(real-res))#误差
        delta_w_out = yita*res*(1-res)*(real-res)*out_in  # 输出层权值的修正量
        delta_w_out[4] = -yita*res*(1-res)*(real-res)     # 输出层阈值的修正量
        w_out = w_out + delta_w_out   # 更新，加上修正量

        for i in range(4):
            delta_w_mid[:, i] = yita*out_in[i]*(1-out_in[i])*w_out[i]*res*(1-res)*(real-res)*net_in   # 中间层神经元的权值修正量
            delta_w_mid[2, i] = -yita*out_in[i]*(1-out_in[i])*w_out[i]*res*(1-res)*(real-res)         # 中间层神经元的阈值修正量,第2行是阈值
        w_mid = w_mid + delta_w_mid   # 更新，加上修正量
    Err.append(np.mean(error))


set_chinese() # 显示中文
fig = plt.figure()
plt.plot(Err)#训练集上每一轮的平均误差
plt.title("每次训练平均误差")
plt.xlabel("训练次数")
plt.ylabel("误差")
plt.savefig("每次训练误差.jpg")
plt.show()



def ceshi():
    net_in = np.array([data_te.iloc[it, 0], data_te.iloc[it, 1], -1])  # 网络输入
    for i in range(4):
        out_in[i] = sigmoid(sum(net_in * w_mid[:, i]))  # 从输入到隐层的传输过程
    res = sigmoid(sum(out_in * w_out))  # 模型预测值
    return res

if __name__ == '__main__':
    for it in range(len(data_te)):
        print('第', it + 1, '个测试值：', ceshi(),"真实值",data_te.iloc[it,2])

 

运行结果：

 

 

 

预测结果：