基于python的ELM（极限学习机）分类及回归实现（附带自己数据链接及推导）

ELM主程序来源：参考@QuantumEntanglement
经过了小小的修改，十分感谢该博主
自己理解推导过程：从最小二乘法到极限学习机

1.分类

数据说明

程序

# -*- coding: utf-8 -*-
"""
Created on Sun Mar 29 10:59:58 2020

@author: 小小飞在路上
"""
import numpy as np
from sklearn.preprocessing import OneHotEncoder
from sklearn.model_selection import train_test_split  #数据集的分割函数
from sklearn.preprocessing import StandardScaler      #数据预处理
from sklearn import metrics
import pandas as pd
from sklearn.utils import shuffle

class RELM_HiddenLayer:

    """
        正则化的极限学习机
        :param x: 初始化学习机时的训练集属性X
        :param num: 学习机隐层节点数
        :param C: 正则化系数的倒数
    """

    def __init__(self, x, num, C=10):
        row = x.shape[0]
        columns = x.shape[1]
        rnd = np.random.RandomState()
        # 权重w
        self.w = rnd.uniform(-1, 1, (columns, num))
        # 偏置b
        self.b = np.zeros([row, num], dtype=float)
        for i in range(num):
            rand_b = rnd.uniform(-0.4, 0.4)
            for j in range(row):
                self.b[j, i] = rand_b
        self.H0 = np.matrix(self.softplus(np.dot(x, self.w) + self.b))
        self.C = C
        self.P = (self.H0.H * self.H0 + len(x) / self.C).I 
        #.T:转置矩阵,.H:共轭转置,.I:逆矩阵

    @staticmethod
    def sigmoid(x):
        """
            激活函数sigmoid
            :param x: 训练集中的X
            :return: 激活值
        """
        return 1.0 / (1 + np.exp(-x))

    @staticmethod
    def softplus(x):
        """
            激活函数 softplus
            :param x: 训练集中的X
            :return: 激活值
        """
        return np.log(1 + np.exp(x))

    @staticmethod
    def tanh(x):
        """
            激活函数tanh
            :param x: 训练集中的X
            :return: 激活值
        """
        return (np.exp(x) - np.exp(-x))/(np.exp(x) + np.exp(-x))

    # 分类问题 训练
    def classifisor_train(self, T):
        """
            初始化了学习机后需要传入对应标签T
            :param T: 对应属性X的标签T
            :return: 隐层输出权值beta
        """
        if len(T.shape) > 1:
            pass
        else:
            self.en_one = OneHotEncoder()
            T = self.en_one.fit_transform(T.reshape(-1, 1)).toarray()
            pass
        all_m = np.dot(self.P, self.H0.H)
        self.beta = np.dot(all_m, T)
        return self.beta

    # 分类问题 测试
    def classifisor_test(self, test_x):
        """
            传入待预测的属性X并进行预测获得预测值
            :param test_x:被预测标签的属性X
            :return: 被预测标签的预测值T
        """
        b_row = test_x.shape[0]
        h = self.softplus(np.dot(test_x, self.w) + self.b[:b_row, :])
        result = np.dot(h, self.beta)
        result =np.argmax(result,axis=1)
        return result
    
# In[]
#数据读取及划分
url = 'C:/Users/weixifei/Desktop/TensorFlow程序/data1.csv'
data = pd. read_csv(url, sep=',',header=None)
data=np.array(data)
data=shuffle(data)
X_data=data[:,:23]
Y=data[:,23]
labels=np.asarray(pd.get_dummies(Y),dtype=np.int8)

num_train=0.3
X_train,X_,Y_train,Y_=train_test_split(X_data,labels,test_size=num_train,random_state=20)
X_test,X_vld,Y_test,Y_vld=train_test_split(X_,Y_,test_size=0.1,random_state=20)

# In[]
#数据标准化处理
stdsc = StandardScaler() 
X_train=stdsc.fit_transform(X_train)
X_test=stdsc.fit_transform(X_test)
X_vld=stdsc.fit_transform(X_vld)
Y_true=np.argmax(Y_test,axis=1)

# In[]
#不同隐藏层结果对比
result=[] 
for j in range(1,300,50):
    a = RELM_HiddenLayer(X_train,j)
    a.classifisor_train(Y_train)
    predict = a.classifisor_test(X_test)
    acc=metrics.precision_score(predict,Y_true, average='macro')
#    result.append(pre)
    print('hidden- %d,acc：%f'%(j,acc))

结果

2.回归

程序

"""
Created on Tue Apr 21 22:37:14 2020

@author: 小小飞在路上
"""
import numpy as np
import matplotlib.pyplot as plt
plt.rcParams['figure.figsize'] = (8, 5)
 
class RELM_HiddenLayer:

    """
        正则化的极限学习机
        :param x: 初始化学习机时的训练集属性X
        :param num: 学习机隐层节点数
        :param C: 正则化系数的倒数
    """

    def __init__(self, x, num, C=10):
        row = x.shape[0]
        columns = x.shape[1]
        rnd = np.random.RandomState()
        # 权重w
        self.w = rnd.uniform(-1, 1, (columns, num))
        # 偏置b
        self.b = np.zeros([row, num], dtype=float)
        for i in range(num):
            rand_b = rnd.uniform(-0.4, 0.4)
            for j in range(row):
                self.b[j, i] = rand_b
        self.H0 = np.matrix(self.sigmoid(np.dot(x, self.w) + self.b))
        self.C = C
        self.P = (self.H0.H * self.H0 + len(x) / self.C).I

    @staticmethod
    def sigmoid(x):
        """
            激活函数sigmoid
            :param x: 训练集中的X
            :return: 激活值
        """
        return 1.0 / (1 + np.exp(-x))
  
    # 回归问题 训练
    def regressor_train(self, T):
        """
            初始化了学习机后需要传入对应标签T
            :param T: 对应属性X的标签T
            :return: 隐层输出权值beta
        """
        all_m = np.dot(self.P, self.H0.H)
        self.beta = np.dot(all_m, T)
        return self.beta

    # 回归问题 测试
    def regressor_test(self, test_x):
        """
            传入待预测的属性X并进行预测获得预测值
            :param test_x:特征
            :return: 预测值
        """
        b_row = test_x.shape[0]
        h = self.sigmoid(np.dot(test_x, self.w) + self.b[:b_row, :])
        result = np.dot(h, self.beta)
        return result

#产生数据集
x=np.linspace(0,20,200)
noise=np.random.normal(0,0.08,200)
y=np.sin(x)+np.cos(0.5*x)+noise
#转化成二维形式
x=np.array(x).reshape(-1,1)
y=np.array(y).reshape(-1,1)


j=0
#绘制原始散点图
plt.plot(x,y,'or')

#不同隐藏层线条设置不同的颜色
color=['g','b','y','c','m']

#比较不同隐藏层拟合效果
for i in range(5,30,5):
    my_EML = RELM_HiddenLayer(x,i)
    my_EML.regressor_train(y)
    x_test = np.linspace(0,20,200).reshape(-1,1)
    y_test = my_EML.regressor_test(x_test)
    plt.plot(x_test,y_test,color[j])
    plt.title('EML_regress')
    plt.xlabel('x')
    plt.ylabel('y')
    j+=1
#增加图例
plt.legend([['original'],['hidden_5'],['hidden_10'],['hidden_15'],['hidden_20'],['hidden_25']],loc='upper right')
plt.show()

结果

分类数据同上一博文数据：数据链接博文
参考：ELM程序参考来源
再次感谢该博主！！！