python-recsys：一款实现推荐系统的python库

来源：互联网发布：什么电台软件好听编辑：程序博客网时间：2024/05/19 21:16

recsys

python-recsys是一个用来实现推荐系统的python库。

安装

依赖项

python-recsys构建于Divisi2（基于语义网络的常识推理库）之上，使用了csc-pysparse（稀疏矩阵计算库），而Divisi2依赖于NumPy和Networkx库。另外python-recsys也依赖于SciPy库。

安装依赖库过程如下（以Ubuntu为例）：

Shell
1
2
3
4
5
6
7
8
sudo apt-get install python-scipy python-numpy
sudo apt-get install python-pip
sudo pip install csc-pysparse networkx divisi2
 
# If you don't have pip installed then do:
# sudo easy_install csc-pysparse
# sudo easy_install networkx
# sudo easy_install divisi2

先从github上下载安装文件，再安装python-recsys：

Shell
1
2
3
tarxvfzpython-recsys.tar.gz
cdpython-recsys
sudopythonsetup.pyinstall

示例

加载Movielens数据集:

Python
1
2
3
4
5
from recsys.algorithm.factorize import SVD
svd = SVD()
svd.load_data(filename='./data/movielens/ratings.dat',
            sep='::',
            format={'col':0, 'row':1, 'value':2, 'ids': int})

进行奇异值分解 (SVD), M=U Sigma V^t:

Python
1
2
3
4
5
6
7
k=100
svd.compute(k=k,
            min_values=10,
            pre_normalize=None,
            mean_center=True,
            post_normalize=True,
            savefile='/tmp/movielens')

得到两部电影的相似性:

Python
1
2
3
4
5
ITEMID1 = 1    # Toy Story (1995)
ITEMID2 = 2355 # A bug's life (1998)
 
svd.similarity(ITEMID1, ITEMID2)
# 0.67706936677315799

获得和电影Toy Story相似的电影:

Python
1
2
3
4
5
6
7
8
9
10
11
12
13
svd.similar(ITEMID1)
 
# Returns: <ITEMID, Cosine Similarity Value>
[(1,    0.99999999999999978),# Toy Story
(3114,0.87060391051018071),# Toy Story 2
(2355,0.67706936677315799),# A bug's life
(588,  0.5807351496754426),  # Aladdin
(595,  0.46031829709743477),# Beauty and the Beast
(1907,0.44589398718134365),# Mulan
(364,  0.42908159895574161),# The Lion King
(2081,0.42566581277820803),# The Little Mermaid
(3396,0.42474056361935913),# The Muppet Movie
(2761,0.40439361857585354)]# The Iron Giant

预测一个用户 (USERID) 将给一部电影 (ITEMID)的打分:

Python
1
2
3
4
5
6
7
8
9
10
MIN_RATING = 0.0
MAX_RATING = 5.0
ITEMID = 1
USERID = 1
 
svd.predict(ITEMID, USERID, MIN_RATING, MAX_RATING)
# Predicted value 5.0
 
svd.get_matrix().value(ITEMID, USERID)
# Real value 5.0

推荐 (没被用户打过分的) 电影给用户:

Python
1
2
3
4
5
6
7
8
9
10
11
12
13
svd.recommend(USERID,is_row=False)#cols are users and rows are items, thus we set is_row=False
 
# Returns: <ITEMID, Predicted Rating>
[(2905,5.2133848204673416),# Shaggy D.A., The
(318,  5.2052108435956033),# Shawshank Redemption, The
(2019,5.1037438278755474),# Seven Samurai (The Magnificent Seven)
(1178,5.0962756861447023),# Paths of Glory (1957)
(904,  5.0771405690055724),# Rear Window (1954)
(1250,5.0744156653222436),# Bridge on the River Kwai, The
(858,  5.0650911066862907),# Godfather, The
(922,  5.0605327279819408),# Sunset Blvd.
(1198,5.0554543765500419),# Raiders of the Lost Ark
(1148,5.0548789542105332)]# Wrong Trousers, The

哪些用户应该会看Toy Story (哪些没给Toy Story打过分的用户将给它一个高的打分?)？

Python
1
2
3
4
5
6
7
8
9
10
11
12
13
svd.recommend(ITEMID)
 
# Returns: <USERID, Predicted Rating>
[(283,  5.716264440514446),
 (3604, 5.6471765418323141),
 (5056, 5.6218800339214496),
 (446,  5.5707524860615738),
 (3902, 5.5494529168484652),
 (4634, 5.51643364021289),
 (3324, 5.5138903299082802),
 (4801, 5.4947999354188548),
 (1131, 5.4941438045650068),
 (2339, 5.4916048051511659)]

文档

从doc/source目录创建HTML文档：

1
2
cddoc
makehtml

HTML 将被创建在下面路径中:

1
doc/build/html/index.html

开源地址：https://github.com/ocelma/python-recsys

阅读全文

0 0