def user_id_to_relation_block(self, user_ids):
uid_unique, index = np.unique(user_ids, return_inverse=True)
X = self.user_encoders.encode_df(
self.user_unique.reindex(uid_unique).reset_index(),
right_tables=[self.df_train],
)
return RelationBlock(index, X)
X_date_train, X_date_test = (None, None)
# setup grouping
feature_group_sizes.extend(user_encoder.encoder_shapes)
feature_group_sizes.extend(movie_encoder.encoder_shapes)
# Create RelationBlock.
train_blocks: List[RelationBlock] = []
test_blocks: List[RelationBlock] = []
for source, target in [(df_train, train_blocks), (df_test, test_blocks)]:
unique_users, user_map = np.unique(source.user_id, return_inverse=True)
target.append(
RelationBlock(
user_map,
user_encoder.encode_df(
pd.DataFrame(dict(user_id=unique_users)),
[implicit_data_source],
),
))
unique_movies, movie_map = np.unique(source.movie_id,
return_inverse=True)
target.append(
RelationBlock(
movie_map,
movie_encoder.encode_df(
pd.DataFrame(dict(movie_id=unique_movies)),
[implicit_data_source],
),
))
trace_path = "rmse_variational_fold_{0}.csv".format(FOLD_INDEX)
row.append(index)
col.append(user_to_internal[uid])
return sps.hstack([
X,
sps.csr_matrix(
(data, (row, col)),
shape=(len(movie_ids), len(user_to_internal)),
),
])
# Create RelationBlock.
train_blocks: List[RelationBlock] = []
test_blocks: List[RelationBlock] = []
for source, target in [(df_train, train_blocks), (df_test, test_blocks)]:
unique_users, user_map = np.unique(source.user_id, return_inverse=True)
target.append(RelationBlock(user_map, augment_user_id(unique_users)))
unique_movies, movie_map = np.unique(source.movie_id,
return_inverse=True)
target.append(RelationBlock(movie_map,
augment_movie_id(unique_movies)))
trace_path = "rmse_{0}_fold_{1}.csv".format(ALGORITHM, FOLD_INDEX)
callback: LibFMLikeCallbackBase
fm: Union[MyFMGibbsRegressor, MyFMOrderedProbit]
if ALGORITHM == "regression":
fm = myfm.MyFMRegressor(rank=DIMENSION)
callback = RegressionCallback(
n_iter=ITERATION,
X_test=X_date_test,
y_test=df_test.rating.values,
def movie_id_to_relation_block(self, movie_ids):
mid_unique, index = np.unique(movie_ids, return_inverse=True)
X = self.movie_encoders.encode_df(
self.movie_unique.reindex(mid_unique).reset_index(),
[self.df_train])
return RelationBlock(index, X)
if not use_ii:
continue
watched_users = movie_vs_watched.get(movie_id, [])
normalizer = 1 / max(len(watched_users), 1) ** 0.5
for uid in watched_users:
X[index, user_to_internal[uid] +
len(movie_to_internal)] = normalizer
return X.tocsr()
# Create RelationBlock.
train_blocks = []
test_blocks = []
for source, target in [(df_train, train_blocks), (df_test, test_blocks)]:
unique_users, user_map = np.unique(source.user_id, return_inverse=True)
target.append(
RelationBlock(user_map, augment_user_id(unique_users))
)
unique_movies, movie_map = np.unique(
source.movie_id, return_inverse=True)
target.append(
RelationBlock(movie_map, augment_movie_id(unique_movies))
)
trace_path="rmse_{0}_fold_{1}.csv".format(ALGORITHM, FOLD_INDEX)
if ALGORITHM == "regression":
fm = myfm.MyFMRegressor(rank=DIMENSION)
callback = RegressionCallback(
ITERATION, X_date_test, df_test.rating.values, X_rel_test=test_blocks,
clip_min=0.5, clip_max=5.0, trace_path=trace_path
)
else: