In [82]:

    

import numpy as np
import matplotlib.pyplot as plt
from scipy import interpolate


    

In [83]:

    

import MySQLdb


    

In [84]:

    

%matplotlib inline


    

In [149]:

    

# takes a 1-element list
def running_avg(data):
    avg = [0]*len(data)
    total = 0
    for k in range(len(data)):
        #avg[k] = np.mean(data[:k])
        total += data[k]
        avg[k] = float(total/(k+1))
    return avg


    

In [86]:

    

db = MySQLdb.connect(host="localhost", user="root", db = "home_kitchen")


    

In [87]:

    

cursor = db.cursor()


    

In [88]:

    

tablename = 'all_hk'


    

In [307]:

    

PID = 'B000GTR2F6'
PID = ' ' + PID


    

In [308]:

    

sql = "Select RTime, RScore From " +tablename + " Where PID = " + '"' + PID +'";'


    

In [309]:

    

cursor.execute(sql)


    

    
Out[309]:





1432L

In [310]:

    

data = cursor.fetchall()


    

In [311]:

    

data = sorted(data)


    

In [312]:

    

rating = zip(*data)[1]
time = zip(*data)[0]
avg =[]
normal_time = [0]*len(time)


    

In [313]:

    

for i in range(len(time)):
    normal_time[i]=(float(time[i])-788918400)/(365*24*60*60)


    

In [314]:

    

avg = running_avg(rating)


    

In [315]:

    

plt.scatter(normal_time, avg)


    

    
Out[315]:





<matplotlib.collections.PathCollection at 0x10da845d0>






    

In [316]:

    

avg = np.array(avg)


    

In [317]:

    

args = np.polyfit(normal_time, avg, 6)


    

In [318]:

    

x = normal_time
f = np.poly1d(args)


    

In [319]:

    

print f


    

    




            6           5         4        3         2
-0.0004657 x + 0.03986 x - 1.405 x + 26.1 x - 269.3 x + 1462 x - 3258

In [320]:

    

plt.plot(x, f(x), '-', x, avg, 'o')


    

    
Out[320]:





[<matplotlib.lines.Line2D at 0x10db5f710>,
 <matplotlib.lines.Line2D at 0x10db5f990>]






    

In [ ]: