2017-05-19 18:27:51 +00:00
|
|
|
# coding: utf8
|
|
|
|
from __future__ import unicode_literals
|
2017-10-02 14:50:26 +00:00
|
|
|
from cytoolz import partition_all, concat
|
2017-05-19 18:27:51 +00:00
|
|
|
|
|
|
|
from ...compat import json_dumps, path2str
|
|
|
|
from ...util import prints
|
2017-05-26 16:32:34 +00:00
|
|
|
from ...gold import iob_to_biluo
|
2017-05-19 18:27:51 +00:00
|
|
|
|
|
|
|
|
|
|
|
def iob2json(input_path, output_path, n_sents=10, *a, **k):
|
|
|
|
"""
|
|
|
|
Convert IOB files into JSON format for use with train cli.
|
|
|
|
"""
|
2017-05-28 13:11:39 +00:00
|
|
|
with input_path.open('r', encoding='utf8') as file_:
|
2017-10-02 14:50:26 +00:00
|
|
|
if n_sents:
|
|
|
|
lines = [' '.join(para) for para in partition_all(n_sents, file_)]
|
|
|
|
else:
|
|
|
|
lines = file_
|
|
|
|
sentences = read_iob(lines)
|
2017-05-19 18:27:51 +00:00
|
|
|
|
|
|
|
output_filename = input_path.parts[-1].replace(".iob", ".json")
|
|
|
|
output_file = output_path / output_filename
|
|
|
|
with output_file.open('w', encoding='utf-8') as f:
|
2017-10-02 14:50:26 +00:00
|
|
|
f.write(json_dumps(sentences))
|
|
|
|
prints("Created %d documents" % len(sentences),
|
2017-05-19 18:27:51 +00:00
|
|
|
title="Generated output file %s" % path2str(output_file))
|
|
|
|
|
|
|
|
|
2017-10-02 14:50:26 +00:00
|
|
|
def read_iob(raw_sents):
|
2017-05-19 18:27:51 +00:00
|
|
|
sentences = []
|
2017-10-02 14:50:26 +00:00
|
|
|
for line in raw_sents:
|
2017-05-19 18:27:51 +00:00
|
|
|
if not line.strip():
|
|
|
|
continue
|
2017-05-28 13:11:39 +00:00
|
|
|
tokens = [t.split('|') for t in line.split()]
|
|
|
|
if len(tokens[0]) == 3:
|
|
|
|
words, pos, iob = zip(*tokens)
|
|
|
|
else:
|
|
|
|
words, iob = zip(*tokens)
|
|
|
|
pos = ['-'] * len(words)
|
2017-05-26 16:32:34 +00:00
|
|
|
biluo = iob_to_biluo(iob)
|
2017-05-19 18:27:51 +00:00
|
|
|
sentences.append([
|
|
|
|
{'orth': w, 'tag': p, 'ner': ent}
|
2017-05-26 16:32:34 +00:00
|
|
|
for (w, p, ent) in zip(words, pos, biluo)
|
2017-05-19 18:27:51 +00:00
|
|
|
])
|
|
|
|
sentences = [{'tokens': sent} for sent in sentences]
|
|
|
|
paragraphs = [{'sentences': [sent]} for sent in sentences]
|
|
|
|
docs = [{'id': 0, 'paragraphs': [para]} for para in paragraphs]
|
|
|
|
return docs
|