def save_train_dev_data(gold_docs, split, train_file, dev_file):
# shuffle the docs
random.seed(27)
random.shuffle(gold_docs)
# split the gold data into training and evaluation
num_training_tasks = round(len(gold_docs) * split / 100)
train_docs = gold_docs[:num_training_tasks]
dev_docs = gold_docs[num_training_tasks:]
print("{} training entities".format(str(entity_count(train_docs))))
print("{} dev entities".format(str(entity_count(dev_docs))))
srsly.write_json(train_file, [docs_to_json(train_docs)])
srsly.write_json(dev_file, [docs_to_json(dev_docs)])
def format_data_to_jsonl(data, file_path, print_label=False):
result = []
labels = set()
i = 0
data = tqdm.tqdm(data, leave=False)
with file_path.open("w", encoding="utf-8") as f:
for d in data:
text = d['text']
ents = []
label_data = d["label"]
for l, label_l in label_data.items():
labels.update([l])
label_ent_array = []
for text_labeled, ent_arrays in label_l.items():
start_char, end_char = ent_arrays[0]
label_ent_array.append((start_char, end_char + 1, l))
ents.append(label_ent_array[0])
if True == diff_contain_overlapping(ents):
i = i + 1
doc = nlp(text)
tags = biluo_tags_from_offsets(doc, ents)
doc.ents = spans_from_biluo_tags(doc, tags)
line = docs_to_json([doc])
f.write(json_dumps(line) + "\n")
msg.good(f"Finished {file_path} :: {i} rows")
if print_label:
msg.info(f"{labels}")
def match_names_for_text(self, characters, text, results_dir, filename=None, tests_variant=False, displacy_option=False, save_ratios=False, save_doc=False):
if tests_variant:
train_data = []
matches_table = prepare_list_for_ratios(characters)
for sentence in text:
matches_table_row, data_for_sentence, _ = self.recognize_person_entities(sentence, characters)
train_data.append(data_for_sentence[0])
matches_table.extend(matches_table_row[1:])
else:
matches_table, train_data, doc = self.recognize_person_entities(text, characters)
if filename is not None:
if save_doc:
json_data = gold.docs_to_json(doc)
with open(results_dir + "\\docs\\" + filename, 'w') as result:
json.dump(json_data, result)
if save_ratios:
write_list_to_file(results_dir + "\\ratios\\" + filename, matches_table)
if tests_variant:
with open(results_dir + filename, 'w', encoding='utf8') as result:
json.dump(train_data, result, ensure_ascii=False)
else:
with open(results_dir + filename, 'w') as result:
json.dump(train_data, result)
if displacy_option:
displacy.serve(doc, style="ent")
def convert_to_spacy_format(data_path, output_path):
labels = get_labels(args.data)
with open(output_path, "w") as f_o:
json_format = []
for i, doc in enumerate(yield_docs(args.data, labels)):
print(i)
json_format.append(docs_to_json(doc, id=i))
f_o.write(json.dumps(json_format))
def test_roundtrip_docs_to_json():
text = "I flew to Silicon Valley via London."
cats = {"TRAVEL": 1.0, "BAKING": 0.0}
nlp = English()
doc = nlp(text)
doc.cats = cats
doc[0].is_sent_start = True
for i in range(1, len(doc)):
doc[i].is_sent_start = False
with make_tempdir() as tmpdir:
json_file = tmpdir / "roundtrip.json"
srsly.write_json(json_file, [docs_to_json(doc)])
goldcorpus = GoldCorpus(str(json_file), str(json_file))
reloaded_doc, goldparse = next(goldcorpus.train_docs(nlp))
assert len(doc) == goldcorpus.count_train()
assert text == reloaded_doc.text
assert "TRAVEL" in goldparse.cats
assert "BAKING" in goldparse.cats
assert cats["TRAVEL"] == goldparse.cats["TRAVEL"]
assert cats["BAKING"] == goldparse.cats["BAKING"]
def test_roundtrip_docs_to_json():
text = "I flew to Silicon Valley via London."
tags = ["PRP", "VBD", "IN", "NNP", "NNP", "IN", "NNP", "."]
heads = [1, 1, 1, 4, 2, 1, 5, 1]
deps = [
"nsubj", "ROOT", "prep", "compound", "pobj", "prep", "pobj", "punct"
]
biluo_tags = ["O", "O", "O", "B-LOC", "L-LOC", "O", "U-GPE", "O"]
cats = {"TRAVEL": 1.0, "BAKING": 0.0}
nlp = English()
doc = nlp(text)
for i in range(len(tags)):
doc[i].tag_ = tags[i]
doc[i].dep_ = deps[i]
doc[i].head = doc[heads[i]]
doc.ents = spans_from_biluo_tags(doc, biluo_tags)
doc.cats = cats
doc.is_tagged = True
doc.is_parsed = True
# roundtrip to JSON
with make_tempdir() as tmpdir:
json_file = tmpdir / "roundtrip.json"
srsly.write_json(json_file, [docs_to_json(doc)])
goldcorpus = GoldCorpus(str(json_file), str(json_file))
reloaded_doc, goldparse = next(goldcorpus.train_docs(nlp))
assert len(doc) == goldcorpus.count_train()
assert text == reloaded_doc.text
assert tags == goldparse.tags
assert deps == goldparse.labels
assert heads == goldparse.heads
assert biluo_tags == goldparse.ner
assert "TRAVEL" in goldparse.cats
assert "BAKING" in goldparse.cats
assert cats["TRAVEL"] == goldparse.cats["TRAVEL"]
assert cats["BAKING"] == goldparse.cats["BAKING"]
# roundtrip to JSONL train dicts
with make_tempdir() as tmpdir:
jsonl_file = tmpdir / "roundtrip.jsonl"
srsly.write_jsonl(jsonl_file, [docs_to_json(doc)])
goldcorpus = GoldCorpus(str(jsonl_file), str(jsonl_file))
reloaded_doc, goldparse = next(goldcorpus.train_docs(nlp))
assert len(doc) == goldcorpus.count_train()
assert text == reloaded_doc.text
assert tags == goldparse.tags
assert deps == goldparse.labels
assert heads == goldparse.heads
assert biluo_tags == goldparse.ner
assert "TRAVEL" in goldparse.cats
assert "BAKING" in goldparse.cats
assert cats["TRAVEL"] == goldparse.cats["TRAVEL"]
assert cats["BAKING"] == goldparse.cats["BAKING"]
# roundtrip to JSONL tuples
with make_tempdir() as tmpdir:
jsonl_file = tmpdir / "roundtrip.jsonl"
# write to JSONL train dicts
srsly.write_jsonl(jsonl_file, [docs_to_json(doc)])
goldcorpus = GoldCorpus(str(jsonl_file), str(jsonl_file))
# load and rewrite as JSONL tuples
srsly.write_jsonl(jsonl_file, goldcorpus.train_tuples)
goldcorpus = GoldCorpus(str(jsonl_file), str(jsonl_file))
reloaded_doc, goldparse = next(goldcorpus.train_docs(nlp))
assert len(doc) == goldcorpus.count_train()
assert text == reloaded_doc.text
assert tags == goldparse.tags
assert deps == goldparse.labels
assert heads == goldparse.heads
assert biluo_tags == goldparse.ner
assert "TRAVEL" in goldparse.cats
assert "BAKING" in goldparse.cats
assert cats["TRAVEL"] == goldparse.cats["TRAVEL"]
assert cats["BAKING"] == goldparse.cats["BAKING"]
texts = []
texts.append(my_text)
# naming of files
fname = 'example'
txt_file = fname + '.txt'
json_file = fname + '.json'
# f = open(txt_file) # will scrape strings from here
# lines = f.readlines() # list of srings from the txt file
docs = [] # initialize a list to be populated wih nlp doc objects
# for line in lines:
# print(line[:]) # display the sentence from that line
for text in texts:
doc = nlp(text) # convert string into a spacy doc object using nlp
# doc = nlp(line) # convert string into a spacy doc object using nlp
docs.append(doc) # add new doc to the list of docs
json_data = docs_to_json(docs) # convert doc into a json file
# import json
# with open('json/' + json_file, 'w+') as outfile:
# json.dump(json_data, outfile)
import srsly
# srsly.write_json('json/' + json_file, [spacy.gold.docs_to_json(docs)])
srsly.write_json(json_file, [spacy.gold.docs_to_json(docs)])
import spacy
import srsly
import json
from spacy.gold import docs_to_json, biluo_tags_from_offsets, spans_from_biluo_tags
nlp = spacy.load('en_core_web_lg')
for i in range(114):
train_data = json.load(
open(
f"/home/marco/Scrivania/tirocinio-unicredit/news/final_attempt/training_data/sector/cli/train_placeholder/{i}.json"
))
docs = []
c = 0
for kgid, text, annot in train_data:
c += 1
print(c)
doc = nlp(text)
tags = biluo_tags_from_offsets(doc, annot['entities'])
entities = spans_from_biluo_tags(doc, tags)
doc.ents = entities
docs.append(doc)
srsly.write_json(
f"/home/marco/Scrivania/tirocinio-unicredit/news/final_attempt/training_data/sector/cli/train_placeholder/gold/{i}.json",
[docs_to_json(docs)])
