Edit model card

Usage

def qa(doc, q):
  doc = doc.replace('\n',' ')
  q = q.replace('\n',' ')
  q_pr = f'<SC6>Опираясь на информацию: {doc}\n ответь на вопрос: \"{q}\".\n Ответ: '
  data_inp = tokenizer(q_pr, return_tensors="pt").to('cuda:0')
  return data_inp

def generate(doc, q):
  t = qa(doc, q)
  output_ids = model.generate(
      **t,  do_sample=False, temperature=0.0, max_new_tokens=512, repetition_penalty=1, no_repeat_ngram_size=8
  )[0]
  out = tokenizer.decode(output_ids.tolist(), skip_special_tokens=True)
  out = out.replace("<extra_id_0>","")
  ans_sqs = sent_tokenize(out, language="russian")
  ans = ' '.join(ans_sqs[:3])
  return ans.split('Ответ:')[0].split('Вопрос:')[0]
Downloads last month
0
Safetensors
Model size
1.74B params
Tensor type
F32
·
Inference Examples
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.