Python实现的线性回归算法示例【附csv文件下载】

本文实例讲述了python实现的线性回归算法。分享给大家供大家参考，具体如下：

用python实现线性回归

using python to implement line regression algorithm

小菜鸟记录学习过程

代码：

#encoding:utf-8
"""
  author:   njulpy
  version:   1.0
  data:   2018/04/09
  project： using python to implement lineregression algorithm
"""
import numpy as np
import pandas as pd
from numpy.linalg import inv
from numpy import dot
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
from sklearn import linear_model
# 最小二乘法
def lms(x_train,y_train,x_test):
  theta_n = dot(dot(inv(dot(x_train.t, x_train)), x_train.t), y_train) # theta = (x'x)^(-1)x'y
  #print(theta_n)
  y_pre = dot(x_test,theta_n)
  mse = np.average((y_test-y_pre)**2)
  #print(len(y_pre))
  #print(mse)
  return theta_n,y_pre,mse
#梯度下降算法
def train(x_train, y_train, num, alpha,m, n):
  beta = np.ones(n)
  for i in range(num):
    h = np.dot(x_train, beta)       # 计算预测值
    error = h - y_train.t         # 计算预测值与训练集的差值
    delt = 2*alpha * np.dot(error, x_train)/m # 计算参数的梯度变化值
    beta = beta - delt
    #print('error', error)
  return beta
if __name__ == "__main__":
  iris = pd.read_csv('iris.csv')
  iris['bias'] = float(1)
  x = iris[['sepal.width', 'petal.length', 'petal.width', 'bias']]
  y = iris['sepal.length']
  x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=5)
  t = np.arange(len(x_test))
  m, n = np.shape(x_train)
  # leastsquare
  theta_n, y_pre, mse = lms(x_train, y_train, x_test)
  # plt.plot(t, y_test, label='test')
  # plt.plot(t, y_pre, label='predict')
  # plt.show()
  # gradientdescent
  beta = train(x_train, y_train, 1000, 0.001, m, n)
  y_predict = np.dot(x_test, beta.t)
  # plt.plot(t, y_predict)
  # plt.plot(t, y_test)
  # plt.show()
  # sklearn
  regr = linear_model.linearregression()
  regr.fit(x_train, y_train)
  y_p = regr.predict(x_test)
  print(regr.coef_,theta_n,beta)
  l1,=plt.plot(t, y_predict)
  l2,=plt.plot(t, y_p)
  l3,=plt.plot(t, y_pre)
  l4,=plt.plot(t, y_test)
  plt.legend(handles=[l1, l2,l3,l4 ], labels=['gradientdescent', 'sklearn','leastsquare','true'], loc='best')
  plt.show()

输出结果

sklearn: [ 0.65368836  0.70955523 -0.54193454  0.        ]
 leastsquare: [ 0.65368836  0.70955523 -0.54193454  1.84603897]
 gradientdescent: [ 0.98359285  0.29325906  0.60084232  1.006859  ]

附：上述示例中的iris.csv文件。

更多关于python相关内容感兴趣的读者可查看本站专题：《python数学运算技巧总结》、《python数据结构与算法教程》、《python函数使用技巧总结》、《python字符串操作技巧汇总》及《python入门与进阶经典教程》

希望本文所述对大家python程序设计有所帮助。