in seed/util/preprocess.py [0:0]
def remove_short_sentences(text_summaries, thres=10):
remove_indices = []
for idx, text in enumerate(text_summaries):
num_words = len(text.split())
if num_words < thres:
remove_indices.append(idx)
text_summaries = [i for j, i in enumerate(text_summaries) if j not in remove_indices]
return text_summaries