def tag_files():
path = "untagged/"
tokenizer = nltk.data.load('tokenizers/punkt/PY3/english.pickle')
files = FileReadingFuncts.get_files(path)
for file in files:
data = FileReadingFuncts.read_file(path, file)
#tags the times
entities_to_tag = TimeFunct.get_end_time_examples(data)
data = tag_entities(data, entities_to_tag)
#tags the named entitites
data = tag_named_entities(data)
#tags the sentences
sentences = SentenceTaggingFuncts.get_sentences(data)
entities_to_tag = get_in_dict(sentences, "sentence")
data = tag_entities(data, entities_to_tag)
#tags the paragraphs
paragraphs = SentenceTaggingFuncts.get_paragraphs(data)
entities_to_tag = get_in_dict(paragraphs, "paragraph")
data = tag_entities(data, entities_to_tag)
FileWritingFuncts.writeTaggedFile(data, file)
def get_all_tags(path, file, tag_name):
#reads in the file:
data = FileReadingFuncts.read_file(path, file)
#finds all the matches
tagged_matches = TagExtractingFuncts.find_tag_matches(tag_name, data)
#removes the tags from every match
untagged_matches = []
for match in tagged_matches:
match = match.replace(".", "")
untagged_matches.append(TagExtractingFuncts.get_rid_of_tags(match))
return untagged_matches
def createsFiles(tag_names):
#reads in the tags
path = "training/"
files = FileReadingFuncts.get_files(path)
matches = getblank2d(len(tag_names))
for file in files:
data = FileReadingFuncts.read_file(path, file)
for i in range(0, len(tag_names)):
newMatches = TagExtractingFuncts.find_tag_matches(tag_names[i], data)
matches[i] = matches[i] + (newMatches)
#writes the tags to the files
for i in range(0, len(tag_names)):
outputFile = "tagFiles/{}.txt".format(tag_names[i])
writeFile(set(matches[i]), outputFile)