def load_stat21x_trans(filename, label_type='int'):
with open(filename) as f:
for line in f:
if "++++++++++" in line or "==========" in line:
continue
items = line.strip().split('\t')
if label_type == 'int':
yield (my_util.str_to_sparse(items[0]), int(items[1]))
elif label_type == 'strings':
yield (my_util.str_to_sparse(items[0]), items[1].split(' '))
def load_stat21x_tx(filename):
with open(filename) as f:
for line in f:
items = line.strip().split('\t')
yield (my_util.str_to_sparse(items[0]), items[1])
def load_stat21x_slides(filename):
with open(filename) as f:
for line in f:
items = line.strip().split('_')
yield my_util.str_to_sparse(items[1])