def fit(self, ratings):
import hpfrec
users = pd.Index(ratings.user.unique())
items = pd.Index(ratings.item.unique())
if 'rating' in ratings.columns:
count = ratings.rating.values.copy()
else:
_logger.info('no ratings found, assuming 1.0')
count = 1.0
hpfdf = pd.DataFrame({
'UserId': users.get_indexer(ratings.user),
'ItemId': items.get_indexer(ratings.item),
'Count': count
})
hpf = hpfrec.HPF(self.features, reindex=False, **self._kwargs)
_logger.info('fitting HPF model with %d features', self.features)
hpf.fit(hpfdf)
self.user_index_ = users
self.item_index_ = items
self.user_features_ = hpf.Theta
self.item_features_ = hpf.Beta
return self
def run_recommender(train_df, latent_factor):
"""Start the recommender."""
recommender = hpfrec.HPF(k=latent_factor, random_seed=123,
allow_inconsistent_math=True, ncores=24)
recommender.step_size = None
logger.warning("Model is training, Don't interrupt.")
recommender.fit(train_df)
return recommender
def run_recommender(train_df, latent_factors): # pragma: no cover
"""Start the recommender."""
recommender = hpfrec.HPF(k=latent_factors, random_seed=123,
ncores=-1, stop_crit='train-llk', verbose=True,
reindex=False, stop_thr=0.000001, maxiter=3000)
recommender.step_size = None
_logger.warning("Model is training, Don't interrupt.")
recommender.fit(train_df)
return recommender
def train():
print('Training HPF model')
recommender = hpfrec.HPF(k=latent_factors,
random_seed=123,
ncores=-1,
stop_crit='train-llk',
verbose=True,
reindex=False,
stop_thr=0.000001,
maxiter=3000)
recommender.step_size = None
_logger.warning("Model is training, Don't interrupt.")
recommender.fit(train_df)
return recommender
