def to_dense(x):
feature_vector = set_str_to_idx(
x, metadata['str2idx'], preprocessing_parameters['tokenizer'])
set_vector = np.zeros((len(metadata['str2idx']), ))
set_vector[feature_vector] = 1
return set_vector.astype(np.bool)
def to_vector(set_str):
bag_vector = np.zeros((len(metadata["str2idx"]), ),
dtype=np.float32)
col_counter = Counter(
set_str_to_idx(set_str, metadata["str2idx"],
preprocessing_parameters["tokenizer"]))
bag_vector[list(col_counter.keys())] = list(col_counter.values())
return bag_vector
def feature_data(column, metadata, preprocessing_parameters):
bag_matrix = np.zeros((len(column), len(metadata['str2idx'])),
dtype=np.float32)
for i, set_str in enumerate(column):
col_counter = Counter(
set_str_to_idx(set_str, metadata['str2idx'],
preprocessing_parameters['tokenizer']))
bag_matrix[i,
list(col_counter.keys())] = list(col_counter.values())
return bag_matrix
def feature_data(column, metadata, preprocessing_parameters):
feature_vector = np.array(
column.map(
lambda x: set_str_to_idx(x, metadata[
'str2idx'], preprocessing_parameters['tokenizer'])))
set_matrix = np.zeros((len(column), len(metadata['str2idx'])), )
for i in range(len(column)):
set_matrix[i, feature_vector[i]] = 1
return set_matrix.astype(np.bool)
def feature_data(column, metadata, preprocessing_parameters):
bag_matrix = np.zeros((len(column), len(metadata['str2idx'])),
dtype=float)
for i in range(len(column)):
col_counter = Counter(
set_str_to_idx(column[i], metadata['str2idx'],
preprocessing_parameters['format']))
bag_matrix[i,
list(col_counter.keys())] = list(col_counter.values())
return bag_matrix
