# -*- coding: utf-8 -*-
# Filename : dataSet.py
import numpy as np
import operator
import Untils
import Kohonen
from numpy import *
import matplotlib.pyplot as plt
# 加载坐标数据文件
dataSet = Untils.loadDataSet("dataset.txt")
dataMat = mat(dataSet)
# print dataMat
normDataset = Kohonen.mapMinMax(dataMat)
# print normDataset
# 生成int随机数，不包含高值
# print random.randint(0,30)
# 计算向量中最小值的索引值
xx = mat([1, 9])
w1 = mat([[1, 2, 3, 4], [5, 6, 7, 8]])
minIndx = Kohonen.distM(xx, w1).argmin()
# 计算距离
jdpx = mat([[0, 0], [0, 1], [1, 0], [1, 1]])
d1 = ceil(minIndx / 4)
d2 = mod(minIndx, 4)
mydist = Kohonen.distM(mat([d1, d2]), jdpx.transpose())
# print mydist
# -*- coding:utf-8 -*-
# Filename : testBoltzmann01.py
import operator
import copy
import Untils
import Boltzmann
from numpy import *
import matplotlib.pyplot as plt
dataSet = Untils.loadDataSet("dataSet25.txt")
cityPosition = mat(dataSet)
m, n = shape(cityPosition)
bestx, di = Boltzmann.boltzmann(cityPosition, MAX_ITER=1000, T0=100)
# 优化前城市图,路径图
Untils.drawScatter(cityPosition, flag=False)
Untils.drawPath(list(range(m)), cityPosition)
# 显示优化后城市图,路径图
Untils.drawScatter(cityPosition, flag=False)
Untils.drawPath(bestx, cityPosition, color='b')
# 绘制误差趋势线
x0 = list(range(len(di)))
Untils.TrendLine(x0, di)
# -*- coding: GBK -*-
# Filename :gradDecent.py
from numpy import *
import operator
import Untils
import matplotlib.pyplot as plt
# BP神经网络
# 数据集: 列1:截距 1;列2:x坐标; 列3:y坐标
dataMat,classLabels = Untils.loadDataSet("student.txt")
dataMat = mat(dataMat)
classMat= mat(classLabels)
# 数据归一化
dataMat = Untils.normalize(dataMat)
# 绘制数据集坐标散点图
Untils.drawClassScatter(dataMat,classLabels,False)
# m行数 n列数
m,n = shape(dataMat)
labelMat = classMat.transpose()
# 步长
alpha = 0.001
# 迭代次数
maxCycles = 500
#构成线性分割线 y=a*x+b: b:weights[0]; a:weights[1]/weights[2]
weights = ones((n,1))
# 计算回归系数 weights
# -*- coding:utf-8 -*-
# Filename : testBoltzmann01.py
import operator
import copy
import Untils
import Boltzmann
from numpy import *
import matplotlib.pyplot as plt
dataSet = Untils.loadDataSet("cities.txt")
cityPosition = mat(dataSet)
m,n = shape(cityPosition)
pn = m
# 将城市的坐标矩阵转换为邻接矩阵（城市间距离矩阵）
dist = Boltzmann.distM(cityPosition,cityPosition.transpose())
# 初始化
MAX_ITER = 2000 # 1000-2000
MAX_M = m;
Lambda = 0.97;
T0 = 1000; # 100-1000
# 构造一个初始可行解
x0 = arange(m)
random.shuffle(x0)
#
T = T0;
iteration = 0;
x = x0; # 路径变量
xx = x0.tolist(); # 每个路径
di = []
# -*- coding: utf-8 -*-
# Filename : testKohonen.py
import numpy as np
import operator
import Untils
import Kohonen
from numpy import *
import matplotlib.pyplot as plt
# 加载坐标数据文件
dataSet = Untils.loadDataSet("dataset.txt");
dataMat = mat(dataSet)
dm,dn = shape(dataMat)
# 归一化数据
normDataset = Kohonen.mapMinMax(dataMat)
# 参数
# 学习率
rate1max=0.8 #0.8
rate1min=0.05;
# 学习半径
r1max=3;
r1min=0.8 #0.8
## 网络构建
Inum=2;
M=2;
N=2;
K=M*N; #Kohonen总节点数
# Kohonen层节点排序
# -*- coding:utf-8 -*-
# Filename : testBoltzmann01.py
import operator
import copy
import Untils
import Boltzmann
from numpy import *
import matplotlib.pyplot as plt
dataSet = Untils.loadDataSet("cities.txt")
cityPosition = mat(dataSet)
m, n = shape(cityPosition)
pn = m
# 将城市的坐标矩阵转换为邻接矩阵（城市间距离矩阵）
dist = Boltzmann.distM(cityPosition, cityPosition.transpose())
# 初始化
MAX_ITER = 2000 # 1000-2000
MAX_M = m
Lambda = 0.97
T0 = 1000
# 100-1000
# 构造一个初始可行解
x0 = arange(m)
random.shuffle(x0)
#
T = T0
iteration = 0
x = x0
# 路径变量
# -*- coding:utf-8 -*-
# Filename : testBoltzmann01.py
import operator
import copy
import Untils
import Boltzmann
from numpy import *
import matplotlib.pyplot as plt
dataSet = Untils.loadDataSet("dataSet25.txt")
cityPosition = mat(dataSet)
m,n = shape(cityPosition)
bestx,di = Boltzmann.boltzmann(cityPosition,MAX_ITER = 1000,T0 = 100)
# 优化前城市图,路径图
Untils.drawScatter(cityPosition,flag=False)
Untils.drawPath(range(m),cityPosition)
# 显示优化后城市图,路径图
Untils.drawScatter(cityPosition,flag=False)
Untils.drawPath(bestx,cityPosition,color='b')
# 绘制误差趋势线
x0 = range(len(di));
Untils.TrendLine(x0,di)
# -*- coding: GBK -*-
# Filename : 01dataSet.py
import numpy as np
import operator
import Untils
import BackPropgation
from numpy import *
import matplotlib.pyplot as plt
dataMat,classLabels = Untils.loadDataSet("student.txt")
# 绘制图形：二维散点,无分类
# Untils.drawScatter(dataMat)
# 绘制图形：二维散点,有分类,适合训练集
# Untils.drawClassScatter(mat(dataMat),classLabels)
# 合并两个多维的matrix，并返回合并后的Matrix
# 输入参数有先后顺序
# [m,n]=shape(dataMat)
# classMat = transpose(mat(classLabels))
# matMerge = Untils.mergMatrix(mat(dataMat),classMat)
# 元素乘法
# a = mat([1,1,1]) ;b = mat([2,2,2])
# print multiply(a,b)
# 测试BackPropgation.dlogsig(hp,tau)
# A = mat([0,1,2]);
# print "A*(1-A)",multiply(A,(1-A))