utils.py

import torch


def get_text(title: str, abstract: str):
  if abstract and title:
    text = abstract + ' ' + title
  elif title:
    text = title
  else:
    text = None
    
  return text

def get_labels(text, model, tokenizer, count_labels=8):
  tokens = tokeinizer(text, return_tensors='pt')
  outputs = model(**tokens)
  probs = torch.nn.Softmax(dim=count_labels)(outputs.logits)
  
  labels = ['Computer_science', 'Economics',
       'Electrical_Engineering_and_Systems_Science', 'Mathematics',
       'Physics', 'Quantitative_Biology', 'Quantitative_Finance',
       'Statistics']
       
  sort_lst = sorted([(prob, label) for prob, label in zip(probs.detach().numpy()[0], labels)], key=lambda x: -x[0])
  cumsum = 0
  result_labels = []
  for pair in sort_lst:
      cumsum += pair[0]
      if cumsum > 0.95 and len(result_labels) >= 1:
          return result_labels
      result_labels.append(pair[1])