movielens.py 6.1 KB
Newer Older
D
dangqingqing 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
Y
Yu Yang 已提交
14 15 16 17 18
"""
Movielens 1-M dataset.

TODO(yuyang18): Complete comments.
"""
D
dangqingqing 已提交
19

Y
Yu Yang 已提交
20
import zipfile
王益 已提交
21
from common import download
Y
Yu Yang 已提交
22 23 24 25
import re
import random
import functools

Y
Refine  
Yu Yang 已提交
26 27
__all__ = [
    'train', 'test', 'get_movie_title_dict', 'max_movie_id', 'max_user_id',
Y
Yu Yang 已提交
28
    'age_table', 'movie_categories', 'max_job_id', 'user_info', 'movie_info'
Y
Refine  
Yu Yang 已提交
29 30 31
]

age_table = [1, 18, 25, 35, 45, 50, 56]
Y
Yu Yang 已提交
32

Y
Yancey1989 已提交
33 34 35
URL = 'http://files.grouplens.org/datasets/movielens/ml-1m.zip'
MD5 = 'c4d9eecfca2ab87c1945afe126590906'

Y
Yu Yang 已提交
36 37 38 39 40 41 42 43 44 45 46 47 48

class MovieInfo(object):
    def __init__(self, index, categories, title):
        self.index = int(index)
        self.categories = categories
        self.title = title

    def value(self):
        return [
            self.index, [CATEGORIES_DICT[c] for c in self.categories],
            [MOVIE_TITLE_DICT[w.lower()] for w in self.title.split()]
        ]

Y
Yu Yang 已提交
49 50 51 52 53 54 55
    def __str__(self):
        return "<MovieInfo id(%d), title(%s), categories(%s)>" % (
            self.index, self.title, self.categories)

    def __repr__(self):
        return self.__str__()

Y
Yu Yang 已提交
56 57 58 59 60

class UserInfo(object):
    def __init__(self, index, gender, age, job_id):
        self.index = int(index)
        self.is_male = gender == 'M'
Y
Refine  
Yu Yang 已提交
61
        self.age = age_table.index(int(age))
Y
Yu Yang 已提交
62 63 64 65 66
        self.job_id = int(job_id)

    def value(self):
        return [self.index, 0 if self.is_male else 1, self.age, self.job_id]

Y
Yu Yang 已提交
67 68 69 70 71 72 73 74
    def __str__(self):
        return "<UserInfo id(%d), gender(%s), age(%d), job(%d)>" % (
            self.index, "M"
            if self.is_male else "F", age_table[self.age], self.job_id)

    def __repr__(self):
        return str(self)

Y
Yu Yang 已提交
75 76 77 78 79 80 81 82

MOVIE_INFO = None
MOVIE_TITLE_DICT = None
CATEGORIES_DICT = None
USER_INFO = None


def __initialize_meta_info__():
Y
Yancey1989 已提交
83
    fn = download(URL, "movielens", MD5)
Y
Yu Yang 已提交
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
    global MOVIE_INFO
    if MOVIE_INFO is None:
        pattern = re.compile(r'^(.*)\((\d+)\)$')
        with zipfile.ZipFile(file=fn) as package:
            for info in package.infolist():
                assert isinstance(info, zipfile.ZipInfo)
                MOVIE_INFO = dict()
                title_word_set = set()
                categories_set = set()
                with package.open('ml-1m/movies.dat') as movie_file:
                    for i, line in enumerate(movie_file):
                        movie_id, title, categories = line.strip().split('::')
                        categories = categories.split('|')
                        for c in categories:
                            categories_set.add(c)
                        title = pattern.match(title).group(1)
                        MOVIE_INFO[int(movie_id)] = MovieInfo(
                            index=movie_id, categories=categories, title=title)
                        for w in title.split():
                            title_word_set.add(w.lower())

                global MOVIE_TITLE_DICT
                MOVIE_TITLE_DICT = dict()
                for i, w in enumerate(title_word_set):
                    MOVIE_TITLE_DICT[w] = i

                global CATEGORIES_DICT
                CATEGORIES_DICT = dict()
                for i, c in enumerate(categories_set):
                    CATEGORIES_DICT[c] = i

                global USER_INFO
                USER_INFO = dict()
                with package.open('ml-1m/users.dat') as user_file:
                    for line in user_file:
                        uid, gender, age, job, _ = line.strip().split("::")
                        USER_INFO[int(uid)] = UserInfo(
                            index=uid, gender=gender, age=age, job_id=job)
    return fn


def __reader__(rand_seed=0, test_ratio=0.1, is_test=False):
    fn = __initialize_meta_info__()
    rand = random.Random(x=rand_seed)
    with zipfile.ZipFile(file=fn) as package:
        with package.open('ml-1m/ratings.dat') as rating:
            for line in rating:
                if (rand.random() < test_ratio) == is_test:
                    uid, mov_id, rating, _ = line.strip().split("::")
                    uid = int(uid)
                    mov_id = int(mov_id)
                    rating = float(rating) * 2 - 5.0

                    mov = MOVIE_INFO[mov_id]
                    usr = USER_INFO[uid]
                    yield usr.value() + mov.value() + [[rating]]


def __reader_creator__(**kwargs):
    return lambda: __reader__(**kwargs)


Y
Refine  
Yu Yang 已提交
146 147
train = functools.partial(__reader_creator__, is_test=False)
test = functools.partial(__reader_creator__, is_test=True)
Y
Yu Yang 已提交
148 149


Y
Yu Yang 已提交
150 151 152 153 154
def get_movie_title_dict():
    __initialize_meta_info__()
    return MOVIE_TITLE_DICT


Y
Refine  
Yu Yang 已提交
155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
def __max_index_info__(a, b):
    if a.index > b.index:
        return a
    else:
        return b


def max_movie_id():
    __initialize_meta_info__()
    return reduce(__max_index_info__, MOVIE_INFO.viewvalues()).index


def max_user_id():
    __initialize_meta_info__()
    return reduce(__max_index_info__, USER_INFO.viewvalues()).index


Y
Yu Yang 已提交
172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
def __max_job_id_impl__(a, b):
    if a.job_id > b.job_id:
        return a
    else:
        return b


def max_job_id():
    __initialize_meta_info__()
    return reduce(__max_job_id_impl__, USER_INFO.viewvalues()).job_id


def movie_categories():
    __initialize_meta_info__()
    return CATEGORIES_DICT


Y
Yu Yang 已提交
189 190 191 192 193 194 195 196 197 198
def user_info():
    __initialize_meta_info__()
    return USER_INFO


def movie_info():
    __initialize_meta_info__()
    return MOVIE_INFO


Y
Yu Yang 已提交
199
def unittest():
Y
Refine  
Yu Yang 已提交
200
    for train_count, _ in enumerate(train()()):
Y
Yu Yang 已提交
201
        pass
Y
Refine  
Yu Yang 已提交
202
    for test_count, _ in enumerate(test()()):
Y
Yu Yang 已提交
203 204 205 206 207
        pass

    print train_count, test_count


Y
Yancey1989 已提交
208 209 210 211
def download():
    paddle.v2.dataset.common.download(URL, "movielens", MD5)


Y
Yu Yang 已提交
212 213
if __name__ == '__main__':
    unittest()