def run_model(text):
'''
run the model on given text
'''
model_path = settings.model_path
dictionary_path = settings.dictionary_path
pp = Predict(model_path, dictionary_path)
results = pp.run(text)
return results
def main(argv):
# get base & to_diff filenames
base_file, diff_file = get_filenames(argv)
# get doc-readers for the 2 files...
base_doc = DOCReader(base_file)
diff_doc = DOCReader(diff_file)
# process the sections into hash
base_doc_sections_map = {}
diff_doc_sections_map = {}
for para in base_doc.sections:
para = para.strip()
k = (para[:11]).strip()
base_doc_sections_map[k] = para
for para in diff_doc.sections:
para = para.strip()
k = (para[:11]).strip()
diff_doc_sections_map[k] = para
# IMP BITs...
# for k in base_doc_sections_map.keys():
# # print(base_doc_sections_map[k])
# # print("key:{}; baseSection:{}; diffSection:{}".format(k,base_doc_sections_map[k], diff_doc_sections_map[k]))
# logging.debug("key:{}; \nbaseSection:{}; \ndiffSection:{}".format(k,base_doc_sections_map[k], diff_doc_sections_map[k]))
# zip 2 files sections
zippedClauses = list(
zip(base_doc_sections_map.keys(), base_doc_sections_map.values(),
diff_doc_sections_map.values()))
# call the predictor
pred = Predict(zippedClauses)
results = pred.run()
for res in results:
print(res)
# -*- coding: utf-8 -*-
from cli import Cli
from train import Train
from predict import Predict
import os
import shutil
if __name__ == '__main__':
if not os.path.isdir('output'):
os.mkdir('output')
# reset
if os.path.isdir('temp'):
shutil.rmtree('temp')
os.mkdir('temp')
# create cli
args = Cli.create_parser().parse_args()
if args.subparser_name == 'train':
t = Train(int(args.model), args.lang)
t.run()
else:
p = Predict(int(args.model))
p.run()