Den4ikAI/ruBert-base-qa-ranker
Модель для оценки релевантности ответов на вопросы.
Использование
import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained('Den4ikAI/ruBert-base-qa-ranker')
model = AutoModelForSequenceClassification.from_pretrained('Den4ikAI/ruBert-base-qa-ranker')
inputs = tokenizer('[CLS]Что такое QR-код?[RESPONSE_TOKEN]QR-код - это тип матричного штрих-кода.', max_length=512, add_special_tokens=False, return_tensors='pt')
with torch.inference_mode():
logits = model(**inputs).logits
probas = torch.sigmoid(logits)[0].cpu().detach().numpy()
relevance, no_relevance = probas
print('Relevance: {}'.format(relevance))
Citation
@MISC{Den4ikAI/ruBert-base-qa-ranker,
author = {Denis Petrov},
title = {Russian QA relevancy model},
url = {https://huggingface.co/Den4ikAI/ruBert-base-qa-ranker},
year = 2023
}
- Downloads last month
- 35
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.