python 线性回归源码

# -*- coding: utf-8 -*-
"""
Created on Fri Jan 12 17:21:11 2018

@author: Administrator
"""

# -*- coding: utf-8 -*-
"""
Created on Thu Jan 11 11:09:36 2018

@author: Administrator
"""

import pymongo
import matplotlib.pyplot as plt
import pandas
import random


client136 = pymongo.MongoClient('192.168.0.136',27017)
db136 = client136.fangjia
dsf136 = db136.average_statistics


ct='上海'
rl='徐汇'
lk='嘉创华狮国际公寓'
query1 = {"city":ct,'region':rl,'name':lk}

fields={ 'city':1,'region':1,'name':1 ,
        'level':1,'fangjia_price':1,'s_date':1}

lct= list(dsf136.find(query1,fields))
df=pandas.DataFrame(lct).dropna()


import numpy as np

def matrix_lstsqr(x, y):
    X = np.vstack([x, np.ones(len(x))]).T
    return (np.linalg.inv(X.T.dot(X)).dot(X.T)).dot(y)

def classic_lstsqr(x_list, y_list):
    N = len(x_list)
    x_avg = sum(x_list)/N
    y_avg = sum(y_list)/N
    var_x, cov_xy = 0, 0
    for x,y in zip(x_list, y_list):
        temp = x - x_avg
        var_x += temp**2
        cov_xy += temp * (y - y_avg)
    slope = cov_xy / var_x
    y_interc = y_avg - slope*x_avg
    return (slope, y_interc)
""" 
x=list(df.index)
y=list(df['fangjia_price'])
#slope, intercept = matrix_lstsqr(x, y)
slope, intercept = classic_lstsqr(x, y)  # 或用classic_lstsqr

print('slope=',slope, 'intercept=',intercept)


"""
#######################################################################
random.seed(12345)

x = [x_i*random.randrange(8,12)/10 for x_i in range(500)]
y = [y_i*random.randrange(8,12)/10 for y_i in range(100,600)]
slope, intercept = matrix_lstsqr(x, y)  # 或用classic_lstsqr
#######################################################################


line_x = [round(min(x)) - 1, round(max(x)) + 1]
line_y = [slope*x_i + intercept for x_i in line_x]

plt.figure(figsize=(8,8))
plt.scatter(x,y)
plt.plot(line_x, line_y, color='red', lw='2')

plt.ylabel('y')
plt.xlabel('x')
plt.title('Linear regression via least squares fit')

#ftext = 'y = ax + b = {:.3f} + {:.3f}x'.format(slope, intercept)
#plt.figtext(.15,.8, ftext, fontsize=11, ha='left')

plt.show()