class FBSM_Rating_Cython(BaseItemCBFRecommender,
BaseItemSimilarityMatrixRecommender,
Incremental_Training_Early_Stopping):
RECOMMENDER_NAME = "FBSM_Rating_Cython"
INIT_TYPE_VALUES = ["random", "one", "BM25", "TF-IDF"]
def __init__(self, URM_train, ICM):
super(FBSM_Rating_Cython, self).__init__(URM_train)
self.n_items_icm, self.n_features = ICM.shape
self.ICM = check_matrix(ICM, 'csr')
def fit(self,
topK=300,
epochs=30,
n_factors=2,
learning_rate=1e-5,
precompute_user_feature_count=False,
initialization_mode_D="random",
positive_only_D=True,
positive_only_V=True,
l2_reg_D=0.01,
l2_reg_V=0.01,
non_negative_weights=False,
verbose=False,
sgd_mode='adam',
gamma=0.9,
beta_1=0.9,
beta_2=0.999,
**earlystopping_kwargs):
if initialization_mode_D not in self.INIT_TYPE_VALUES:
raise ValueError(
"Value for 'initialization_mode_D' not recognized. Acceptable values are {}, provided was '{}'"
.format(self.INIT_TYPE_VALUES, initialization_mode_D))
from FeatureWeighting.Cython.FBSM_Rating_Cython_SGD import FBSM_Rating_Cython_SGD
self.n_factors = n_factors
self.learning_rate = learning_rate
self.l2_reg_D = l2_reg_D
self.l2_reg_V = l2_reg_V
self.topK = topK
self.epochs = epochs
self.verbose = verbose
# For mean_init use Xavier Initialization
if self.n_factors != 0:
std_init = 1 / self.n_features / self.n_factors
else:
std_init = 0
mean_init = 0
weights_initialization_D = None
if initialization_mode_D == "random":
weights_initialization_D = np.random.normal(
0.001, 0.1, self.n_features).astype(np.float64)
elif initialization_mode_D == "one":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
elif initialization_mode_D == "zero":
weights_initialization_D = np.zeros(self.n_features,
dtype=np.float64)
elif initialization_mode_D == "BM25":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
self.ICM = self.ICM.astype(np.float32)
self.ICM = okapi_BM_25(self.ICM)
elif initialization_mode_D == "TF-IDF":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
self.ICM = self.ICM.astype(np.float32)
self.ICM = TF_IDF(self.ICM)
else:
raise ValueError(
"CFW_D_Similarity_Cython: 'init_type' not recognized")
self.FBSM_Rating = FBSM_Rating_Cython_SGD(
self.URM_train,
self.ICM,
n_factors=self.n_factors,
precompute_user_feature_count=precompute_user_feature_count,
learning_rate=self.learning_rate,
l2_reg_D=self.l2_reg_D,
l2_reg_V=self.l2_reg_V,
weights_initialization_D=weights_initialization_D,
weights_initialization_V=None,
positive_only_D=positive_only_D,
positive_only_V=positive_only_V,
verbose=self.verbose,
sgd_mode=sgd_mode,
gamma=gamma,
beta_1=beta_1,
beta_2=beta_2,
mean_init=mean_init,
std_init=std_init)
if self.verbose:
print(self.RECOMMENDER_NAME + ": Initialization completed")
self._train_with_early_stopping(epochs,
algorithm_name=self.RECOMMENDER_NAME,
**earlystopping_kwargs)
self.compute_W_sparse(model_to_use="best")
sys.stdout.flush()
def _prepare_model_for_validation(self):
self.D_incremental = self.FBSM_Rating.get_D()
self.V_incremental = self.FBSM_Rating.get_V()
self.compute_W_sparse(model_to_use="last")
def _update_best_model(self):
self.D_best = self.D_incremental.copy()
self.V_best = self.V_incremental.copy()
def _run_epoch(self, num_epoch):
self.loss = self.FBSM_Rating.fit()
def set_ICM_and_recompute_W(self, ICM_new, recompute_w=True):
self.ICM = ICM_new.copy()
if recompute_w:
self.compute_W_sparse(use_D=True, use_V=True, model_to_use="best")
def compute_W_sparse(self, use_D=True, use_V=True, model_to_use="best"):
assert model_to_use in [
"last", "best"
], "{}: compute_W_sparse, 'model_to_use' parameter not recognized".format(
self.RECOMMENDER_NAME)
if self.verbose:
print("FBSM_Rating_Cython: Building similarity matrix...")
start_time = time.time()
start_time_print_batch = start_time
# Diagonal
if use_D:
if model_to_use == "last":
D = self.D_incremental
else:
D = self.D_best
similarity = Compute_Similarity(self.ICM.T,
shrink=0,
topK=self.topK,
normalize=False,
row_weights=D)
self.W_sparse = similarity.compute_similarity()
else:
self.W_sparse = sps.csr_matrix((self.n_items, self.n_items))
if use_V:
if model_to_use == "last":
V = self.V_incremental
else:
V = self.V_best
# V * V.T
W1 = self.ICM.dot(V.T)
#self.W_sparse += W1.dot(W1.T)
# Use array as it reduces memory requirements compared to lists
dataBlock = 10000000
values = np.zeros(dataBlock, dtype=np.float32)
rows = np.zeros(dataBlock, dtype=np.int32)
cols = np.zeros(dataBlock, dtype=np.int32)
numCells = 0
for numItem in range(self.n_items):
V_weights = W1[numItem, :].dot(W1.T)
V_weights[numItem] = 0.0
relevant_items_partition = (
-V_weights).argpartition(self.topK - 1)[0:self.topK]
relevant_items_partition_sorting = np.argsort(
-V_weights[relevant_items_partition])
top_k_idx = relevant_items_partition[
relevant_items_partition_sorting]
# Incrementally build sparse matrix, do not add zeros
notZerosMask = V_weights[top_k_idx] != 0.0
numNotZeros = np.sum(notZerosMask)
values_to_add = V_weights[top_k_idx][notZerosMask]
rows_to_add = top_k_idx[notZerosMask]
cols_to_add = np.ones(numNotZeros) * numItem
for index in range(len(values_to_add)):
if numCells == len(rows):
rows = np.concatenate(
(rows, np.zeros(dataBlock, dtype=np.int32)))
cols = np.concatenate(
(cols, np.zeros(dataBlock, dtype=np.int32)))
values = np.concatenate(
(values, np.zeros(dataBlock, dtype=np.float32)))
rows[numCells] = rows_to_add[index]
cols[numCells] = cols_to_add[index]
values[numCells] = values_to_add[index]
numCells += 1
if self.verbose and (time.time() - start_time_print_batch >= 30
or numItem == self.n_items - 1):
columnPerSec = numItem / (time.time() - start_time)
print(
"Weighted similarity column {} ( {:2.0f} % ), {:.2f} column/sec, elapsed time {:.2f} min"
.format(numItem, numItem / self.n_items * 100,
columnPerSec, (time.time() - start_time) / 60))
sys.stdout.flush()
sys.stderr.flush()
start_time_print_batch = time.time()
V_weights = sps.csr_matrix(
(values[:numCells], (rows[:numCells], cols[:numCells])),
shape=(self.n_items, self.n_items),
dtype=np.float32)
self.W_sparse += V_weights
self.W_sparse = check_matrix(self.W_sparse, format='csr')
if self.verbose:
print("FBSM_Rating_Cython: Building similarity matrix... complete")
def save_model(self, folder_path, file_name=None):
if file_name is None:
file_name = self.RECOMMENDER_NAME
print("{}: Saving model in file '{}'".format(self.RECOMMENDER_NAME,
folder_path + file_name))
data_dict_to_save = {
"D_best": self.D_best,
"V_best": self.V_best,
"topK": self.topK,
"W_sparse": self.W_sparse
}
dataIO = DataIO(folder_path=folder_path)
dataIO.save_data(file_name=file_name,
data_dict_to_save=data_dict_to_save)
print("{}: Saving complete".format(self.RECOMMENDER_NAME))
class CFW_DVV_Similarity_Cython(BaseItemCBFRecommender,
BaseItemSimilarityMatrixRecommender,
Incremental_Training_Early_Stopping):
RECOMMENDER_NAME = "CFW_DVV_Similarity_Cython"
INIT_TYPE_VALUES = ["random", "one", "BM25", "TF-IDF"]
def __init__(self, URM_train, ICM, S_matrix_target):
super(CFW_DVV_Similarity_Cython, self).__init__(URM_train)
self.ICM = check_matrix(ICM, 'csr')
self.n_features = self.ICM.shape[1]
self.S_matrix_target = check_matrix(S_matrix_target, 'csr')
def fit(self,
n_factors=1,
learning_rate=0.01,
l2_reg_D=0.0,
l2_reg_V=0.0,
epochs=50,
topK=300,
positive_only_weights=True,
precompute_common_features=False,
add_zeros_quota=0.0,
initialization_mode_D="random",
positive_only_D=True,
positive_only_V=True,
verbose=True,
sgd_mode='adagrad',
gamma=0.9,
beta_1=0.9,
beta_2=0.999,
**earlystopping_kwargs):
from FeatureWeighting.Cython.CFW_DVV_Similarity_Cython_SGD import CFW_DVV_Similarity_Cython_SGD
if initialization_mode_D not in self.INIT_TYPE_VALUES:
raise ValueError(
"Value for 'initialization_mode_D' not recognized. Acceptable values are {}, provided was '{}'"
.format(self.INIT_TYPE_VALUES, initialization_mode_D))
# if initialization_mode_V not in self.INIT_TYPE_VALUES:
# raise ValueError("Value for 'initialization_mode_V' not recognized. Acceptable values are {}, provided was '{}'".format(self.INIT_TYPE_VALUES, initialization_mode_V))
self.n_factors = n_factors
self.learning_rate = learning_rate
self.add_zeros_quota = add_zeros_quota
self.precompute_common_features = precompute_common_features
self.l2_reg_D = l2_reg_D
self.l2_reg_V = l2_reg_V
self.epochs = epochs
self.topK = topK
self.verbose = verbose
self._generate_train_data()
# For mean_init use Xavier Initialization
if self.n_factors != 0:
std_init = 1 / self.n_features / self.n_factors
else:
std_init = 0
mean_init = 0
weights_initialization_D = None
if initialization_mode_D == "random":
weights_initialization_D = np.random.normal(
0.001, 0.1, self.n_features).astype(np.float64)
elif initialization_mode_D == "one":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
elif initialization_mode_D == "zero":
weights_initialization_D = np.zeros(self.n_features,
dtype=np.float64)
elif initialization_mode_D == "BM25":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
self.ICM = self.ICM.astype(np.float32)
self.ICM = okapi_BM_25(self.ICM)
elif initialization_mode_D == "TF-IDF":
weights_initialization_D = np.ones(self.n_features,
dtype=np.float64)
self.ICM = self.ICM.astype(np.float32)
self.ICM = TF_IDF(self.ICM)
else:
raise ValueError(
"CFW_D_Similarity_Cython: 'init_type' not recognized")
self.CFW_DVV_Cython = CFW_DVV_Similarity_Cython_SGD(
self.row_list,
self.col_list,
self.data_list,
self.ICM,
n_factors=self.n_factors,
precompute_common_features=precompute_common_features,
weights_initialization_D=weights_initialization_D,
weights_initialization_V=None,
learning_rate=self.learning_rate,
add_zeros_quota=add_zeros_quota,
positive_only_D=positive_only_D,
positive_only_V=positive_only_V,
l2_reg_D=self.l2_reg_D,
l2_reg_V=self.l2_reg_V,
sgd_mode=sgd_mode,
verbose=self.verbose,
gamma=gamma,
beta_1=beta_1,
beta_2=beta_2,
mean_init=mean_init,
std_init=std_init)
if self.verbose:
print(self.RECOMMENDER_NAME + ": Initialization completed")
self._train_with_early_stopping(epochs,
algorithm_name=self.RECOMMENDER_NAME,
**earlystopping_kwargs)
self.compute_W_sparse(model_to_use="best")
sys.stdout.flush()
def _prepare_model_for_validation(self):
self.D_incremental = self.CFW_DVV_Cython.get_D()
self.V_incremental = self.CFW_DVV_Cython.get_V()
self.compute_W_sparse(model_to_use="last")
def _update_best_model(self):
self.D_best = self.D_incremental.copy()
self.V_best = self.V_incremental.copy()
def _run_epoch(self, num_epoch):
self.loss = self.CFW_DVV_Cython.fit()
def set_ICM_and_recompute_W(self, ICM_new, recompute_w=True):
self.ICM = ICM_new.copy()
if recompute_w:
self.compute_W_sparse()
def _generate_train_data(self):
if self.verbose:
print(self.RECOMMENDER_NAME + ": Generating train data...")
self.S_matrix_target.eliminate_zeros()
numSamples = self.S_matrix_target.nnz
zeros_to_add = int(numSamples * self.add_zeros_quota)
self.S_matrix_target = self.S_matrix_target.tocoo()
if zeros_to_add != 0.0:
self.row_list = np.concatenate((np.array(self.S_matrix_target.row,
dtype=np.int32),
np.zeros(zeros_to_add,
dtype=np.int32)))
self.col_list = np.concatenate((np.array(self.S_matrix_target.col,
dtype=np.int32),
np.zeros(zeros_to_add,
dtype=np.int32)))
self.data_list = np.concatenate(
(np.array(self.S_matrix_target.data, dtype=np.float64),
np.zeros(zeros_to_add, dtype=np.float64)))
else:
self.row_list = np.array(self.S_matrix_target.row, dtype=np.int32)
self.col_list = np.array(self.S_matrix_target.col, dtype=np.int32)
self.data_list = np.array(self.S_matrix_target.data,
dtype=np.float64)
self._add_zeros_in_train_data_row_wise()
# Initialize samples
self.n_samples = len(self.data_list)
def _add_zeros_in_train_data_row_wise(self):
"""
This function uses a set of tuples to ensure the zero elements to be added are not already existent
:return:
"""
if self.verbose:
print(self.RECOMMENDER_NAME + ": Adding zeros in train data...")
self.S_matrix_target = check_matrix(self.S_matrix_target, "csr")
numSamples = self.S_matrix_target.nnz
n_items = self.S_matrix_target.shape[0]
zeros_to_add_global = int(numSamples * self.add_zeros_quota)
zeros_added_global = 0
if zeros_to_add_global + numSamples >= n_items**2:
raise ValueError(
self.RECOMMENDER_NAME +
": Too many zeros to add, not enough unique coordinates in matrix"
)
zeros_to_add_per_item = int(zeros_to_add_global / self.n_items)
while zeros_added_global < zeros_to_add_global:
for current_item_row in range(self.n_items):
start_pos = self.S_matrix_target.indptr[current_item_row]
end_pos = self.S_matrix_target.indptr[current_item_row + 1]
nonzero_coordinates = set(
self.S_matrix_target.indices[start_pos:end_pos])
zeros_added_per_item = 0
while zeros_added_per_item < zeros_to_add_per_item and zeros_added_global < zeros_to_add_global:
new_coordinate = np.random.randint(0, n_items)
if new_coordinate not in nonzero_coordinates:
nonzero_coordinates.add(new_coordinate)
self.row_list[numSamples +
zeros_added_global] = current_item_row
self.col_list[numSamples +
zeros_added_global] = new_coordinate
self.data_list[numSamples + zeros_added_global] = 0.0
zeros_added_per_item += 1
zeros_added_global += 1
if self.verbose:
print("Added: {} zeros. Average per item is: {} ".format(
zeros_added_global, zeros_to_add_per_item))
print(self.RECOMMENDER_NAME +
": Added zeros, data points are {}".format(
len(self.data_list)))
def compute_W_sparse(self, use_D=True, use_V=True, model_to_use="best"):
assert model_to_use in [
"last", "best"
], "{}: compute_W_sparse, 'model_to_use' parameter not recognized".format(
self.RECOMMENDER_NAME)
if self.verbose:
print(self.RECOMMENDER_NAME + ": Building similarity matrix...")
start_time = time.time()
start_time_print_batch = start_time
# Diagonal
if use_D:
if model_to_use == "last":
D = self.D_incremental
else:
D = self.D_best
similarity = Compute_Similarity(self.ICM.T,
shrink=0,
topK=self.topK,
normalize=False,
row_weights=D)
self.W_sparse = similarity.compute_similarity()
else:
self.W_sparse = sps.csr_matrix((self.n_items, self.n_items))
if use_V:
if model_to_use == "last":
V = self.V_incremental
else:
V = self.V_best
# V * V.T
W1 = self.ICM.dot(V.T)
#self.W_sparse += W1.dot(W1.T)
# Use array as it reduces memory requirements compared to lists
dataBlock = 10000000
values = np.zeros(dataBlock, dtype=np.float32)
rows = np.zeros(dataBlock, dtype=np.int32)
cols = np.zeros(dataBlock, dtype=np.int32)
numCells = 0
for numItem in range(self.n_items):
V_weights = W1[numItem, :].dot(W1.T)
V_weights[numItem] = 0.0
relevant_items_partition = (
-V_weights).argpartition(self.topK - 1)[0:self.topK]
relevant_items_partition_sorting = np.argsort(
-V_weights[relevant_items_partition])
top_k_idx = relevant_items_partition[
relevant_items_partition_sorting]
# Incrementally build sparse matrix, do not add zeros
notZerosMask = V_weights[top_k_idx] != 0.0
numNotZeros = np.sum(notZerosMask)
values_to_add = V_weights[top_k_idx][notZerosMask]
rows_to_add = top_k_idx[notZerosMask]
cols_to_add = np.ones(numNotZeros) * numItem
for index in range(len(values_to_add)):
if numCells == len(rows):
rows = np.concatenate(
(rows, np.zeros(dataBlock, dtype=np.float32)))
cols = np.concatenate(
(cols, np.zeros(dataBlock, dtype=np.int32)))
values = np.concatenate(
(values, np.zeros(dataBlock, dtype=np.int32)))
rows[numCells] = rows_to_add[index]
cols[numCells] = cols_to_add[index]
values[numCells] = values_to_add[index]
numCells += 1
if self.verbose and (time.time() - start_time_print_batch >= 30
or numItem == self.n_items - 1):
columnPerSec = numItem / (time.time() - start_time)
print(
"{}: Weighted similarity column {} ( {:2.0f} % ), {:.2f} column/sec, elapsed time {:.2f} min"
.format(self.RECOMMENDER_NAME, numItem,
numItem / self.n_items * 100, columnPerSec,
(time.time() - start_time) / 60))
sys.stdout.flush()
sys.stderr.flush()
start_time_print_batch = time.time()
V_weights = sps.csr_matrix(
(values[:numCells], (rows[:numCells], cols[:numCells])),
shape=(self.n_items, self.n_items),
dtype=np.float32)
self.W_sparse += V_weights
self.W_sparse = check_matrix(self.W_sparse, format='csr')
if self.verbose:
print(self.RECOMMENDER_NAME +
": Building similarity matrix... complete")
def runCompilationScript(self):
# Run compile script setting the working directory to ensure the compiled file are contained in the
# appropriate subfolder and not the project root
file_subfolder = "FeatureWeighting/Cython"
file_to_compile_list = ['CFW_DVV_Similarity_Cython_SGD.pyx']
run_compile_subprocess(file_subfolder, file_to_compile_list)
print("{}: Compiled module {} in subfolder: {}".format(
self.RECOMMENDER_NAME, file_to_compile_list, file_subfolder))
# Command to run compilation script
# python compile_script.py CFW_DVV_Similarity_Cython_SGD.pyx build_ext --inplace
# Command to generate html report
# cython -a CFW_DVV_Similarity_Cython_SGD.pyx
def save_model(self, folder_path, file_name=None):
if file_name is None:
file_name = self.RECOMMENDER_NAME
print("{}: Saving model in file '{}'".format(self.RECOMMENDER_NAME,
folder_path + file_name))
data_dict_to_save = {
"D_best": self.D_best,
"V_best": self.V_best,
"topK": self.topK,
"W_sparse": self.W_sparse
}
dataIO = DataIO(folder_path=folder_path)
dataIO.save_data(file_name=file_name,
data_dict_to_save=data_dict_to_save)
print("{}: Saving complete".format(self.RECOMMENDER_NAME))