-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathitem.py
148 lines (127 loc) · 4.03 KB
/
item.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
import matrix_creation
import pandas as pd
from sklearn.metrics import pairwise_distances
from sklearn.metrics.pairwise import cosine_similarity
import operator
import numpy as np
import math
# def csv_reader(filename):
# return pd.read_csv(filename,sep='\t',header=None)
# def create_pivot_matrix(train_data):
# return train_data.pivot(index='User', columns='Movie', values='Rating').reset_index(drop=True)
def mean_rating(user_movies_matrix,user):
counta = suma = 0
aa = user_movies_matrix[user].tolist()
for i in range(len(aa)):
suma+=aa[i]
if aa[i]>0:
counta+=1
meanadd = suma/counta
return meanadd
# def getr(summy,count):
# return ((summy/float(count))-0.1)
def mean_rating2(user_movies_matrix,user):
counta = suma = 0
aa = user_movies_matrix[user].tolist()
for i in range(len(aa)):
suma+=aa[i]
counta+=1
meanadd = suma/counta
return meanadd
# files = [1,2,3,4,5]
kvals = [10,20,30,40,50]
for k_index in range(len(kvals)):
print "\n"
print "K value: " + str(kvals[k_index])
avg_mae=0
train_matrix, test_matrix, train_grades, test_grades, train_data, test_data, train_matrix_shape0,train_matrix_shape1 = matrix_creation.get_data()
train_matrix = train_matrix.T
user_movies_matrix = train_matrix
matrix_val = cosine_similarity(train_matrix)
user_similarity_matrix = pd.DataFrame(matrix_val)
summy = count = rmsesum = 0
for ind in range(len(test_grades)):
try:
# if count>4000:
count+=1
user = test_grades[ind][1]
movie = test_grades[ind][0]
rating = test_grades[ind][2]
similar_users_list = user_similarity_matrix[user].tolist()
# print "1"
simdic = {}
for i in range(len(similar_users_list)):
simdic[i] = similar_users_list[i]
sorted_sim = sorted(simdic.items(), key=operator.itemgetter(1),reverse=True)
topk = []
sum_weights = 0
#----------------------co-rated users--------------------------------
# ks = kvals[k_index]
# inlop = 1
# while(ks>0):
# temp_user = sorted_sim[inlop][0]
# inlop+=1
# # print "count, inlop: "+str(count)+" "+str(inlop)
# # print temp_user,movie
# if user_movies_matrix[temp_user][movie]>0 and inlop<len(sorted_sim):
# topk.append(sorted_sim[inlop])
# sum_weights+=sorted_sim[inlop][1]
# ks-=1
# if count%1000==0:
# print count, temp_user, movie, len(topk)
# -----------------without cor-rated users----------------------------
for i in range(kvals[k_index]):
topk.append(sorted_sim[i+1])
sum_weights+=sorted_sim[i+1][1]
# print "2"
adjust_weights = []
ratings = []
for i in range(len(topk)):
uid = topk[i][0]
vid = topk[i][1]
adjust_weights.append(vid/sum_weights)
# ------------------base shift and scale shit normalization---------------
maxy = -9999999
miny = 9999999
ratt = np.array(user_movies_matrix[uid])
for i in range(len(ratt)):
if ratt[i]>maxy:
maxy = ratt[i]
if ratt[i]<miny:
miny = ratt[i]
normalized_rat = (user_movies_matrix[uid][movie]-miny)/(maxy-miny)
# normalized_rat = user_movies_matrix[uid][movie]
meanadd = mean_rating2(user_movies_matrix,uid)
ratings.append(normalized_rat-meanadd)
# print adjust_weights
# print ratings
# break
# print "3"
# zip_val = zip(ratings,adjust_weights)
# zip_val = []
p_rating = 0
for i in range(len(ratings)):
# zip_val.append((ratings[i],adjust_weights[i]))
p_rating+=ratings[i]*adjust_weights[i]
# p_rating = sum([x*y for x,y in zip_val])
# counta = suma = 0
# aa = user_movies_matrix[user].tolist()
# for i in range(len(aa)):
# suma+=aa[i]
# if aa[i]>0:
# counta+=1
# meanadd = suma/counta
meanadd = mean_rating(user_movies_matrix,user)
# meanadd = mean_item_rating(user_movies_matrix,user)
summy+=abs(meanadd+p_rating-rating)
rmsesum += abs(meanadd+p_rating-rating)*abs(meanadd+p_rating-rating)
# print "4"
# else:
# count+=1
except Exception,e:
# print count
# print "Exception: "+ str(e)
continue
# break
print summy/float(count), math.sqrt(rmsesum/count)
avg_mae+= summy/float(count)