Skip to content

Pass device to pipeline and fix regexp warnings #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
koolkhel wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Pass device to pipeline and fix regexp warnings #2

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c895e84
Pass given device to the pipeline
koolkhel Nov 24, 2024
95676a0
fix re warnings
koolkhel Nov 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions runorm/runorm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ def __init__(self):
self.rule_normalizer = RuleNormalizer()
self.numbers_normalizer = Numbers2Words()
self.re_tokens = re.compile(r"(?:[.,!?]|[а-яА-Я]\S*|-?\d\S*(?:\.\d+)?|[^а-яА-Я\d\s-]+)\s*")
self.re_normalization = re.compile(r"[^a-zA-Z0-9\sа-яА-ЯёЁ.,!?:;""''(){}\[\]«»„“”-]")
self.re_normalization = re.compile(r"[^a-zA-Z0-9\sа-яА-ЯёЁ.,!?:;""''(){}[]«»„“”-]")
self.paths = {
"tagger": "RUNorm/RUNorm-tagger",
"kirillizator": "RUNorm/RUNorm-kirillizator",
Expand All @@ -44,7 +44,7 @@ def load(self, model_size="small", device="cpu", workdir=None):
self.angl_model = T5ForConditionalGeneration.from_pretrained(self.paths["kirillizator"], cache_dir=self.workdir)
self.tagger_model = BertForTokenClassification.from_pretrained(self.paths["tagger"], cache_dir=self.workdir)
self.tagger_tokenizer = AutoTokenizer.from_pretrained(self.paths["tagger"], cache_dir=self.workdir)
self.tagger = pipeline("ner", model=self.tagger_model, tokenizer=self.tagger_tokenizer, aggregation_strategy="average")
self.tagger = pipeline("ner", model=self.tagger_model, tokenizer=self.tagger_tokenizer, aggregation_strategy="average", device=device)
self.abbr_model.to(device)
self.angl_model.to(device)
self.abbr_model.eval()
Expand Down Expand Up @@ -107,7 +107,7 @@ def construct_prompt(self, text, angl_mode=False):
etid = 0
token_to_add = ""
for token in self.process_sentence(text) + [""]:
if not re.search("[a-zA-Z\d]", token):
if not re.search(r"[a-zA-Z\d]", token):
if token_to_add:
end_match = re.search(r"(.+?)(\W*)$", token_to_add, re.M).groups()
if self.is_english(end_match[0].strip()):
Expand Down Expand Up @@ -292,4 +292,4 @@ def norm(self, message):
out = out + " " + final_answer

#elapsed_time = time.time() - start
return out.strip()
return out.strip()