def run():
assert os.path.exists('data/ratings.csv'), \
'File not exists in path, run preprocess.py before this.'
print('Start..')
start = time.time()
if not os.path.exists('data/lfm_items.dict'):
Corpus.pre_process()
if not os.path.exists('data/lfm.model'):
LFM().train()
movies = LFM().predict(user_id=1)
for movie in movies:
print(movie)
print('Cost time: %f' % (time.time() - start))
def run(input_user_id=1, input_top_n=10):
assert os.path.exists('data/ratings.csv'), \
'File not exists in path, run preprocess.py before this.'
print('Start..')
print('用户ID：%s' % input_user_id)
start = time.time()
if not os.path.exists('data/lfm_items.dict'):
Corpus.pre_process()
if not os.path.exists('data/lfm.model'):
LFM().train()
movies = LFM().predict(user_id=input_user_id, top_n=input_top_n)
for movie in movies:
print(movie)
print('用时: %f' % (time.time() - start))
def run():
assert os.path.exists('data/user/user_follows.csv'), \
'File not exists in path, run preprocess.py before this.'
print('Start..')
start = time.time()
if not os.path.exists('data/lfm_model/lfm_items.dict'):
Corpus.pre_process()
if not os.path.exists('data/lfm_model/lfm.model'):
LFM().train()
uids = [2005098, 6083125955935600644, 6099765861231362052]
lfm = LFM()
for uid in uids:
movies = lfm.predict(user_id=uid)
for movie in movies:
print(movie)
print('Cost time: %f' % (time.time() - start))
def run():
assert os.path.exists('data/ratings.csv'), FILE_NOT_EXIT
print('Start..')
start = time.time()
if not os.path.exists('data/lfm_items.dict'):
print('Start to Generate lfm_items.dict.')
Corpus.pre_process()
print('Generate lfm_items.dict ok.')
if not os.path.exists('data/lfm.model'):
print('Start to Generate lfm.model.')
LFM().train()
print('Generate lfm.model ok.')
movies = LFM().predict(user_id=1)
for movie in movies:
print(movie)
print('Cost time: %f' % (time.time() - start))
def run(user_id):
'''
调试lfm模型
:param user_id:
:return:
'''
assert os.path.exists('data/ratings.csv'), \
'File not exists in path, run preprocess.py before this.'
print('Start..')
start = time.time()
if not os.path.exists('data/lfm_items.dict'):
Corpus.pre_process()
if not os.path.exists('data/lfm.model'):
LFM().train()
movies = LFM().predict(user_id)
movies_list = list(map(lambda x:x[0], movies))
print(movies_list)
print('End.')
print('Cost Time:', time.time()-start)
def predict(user_id):
cars = LFM().predict(user_id)
return cars