Update app.py

app.py

@@ -16,12 +16,14 @@ def generate_qa_pairs(text):
     tokenizer = AutoTokenizer.from_pretrained("valhalla/t5-base-qg-hl")
     model = AutoModelForSeq2SeqLM.from_pretrained("valhalla/t5-base-qg-hl")
     
-    input_text = "highlight: " + text
-    input_ids = tokenizer.encode(input_text, return_tensors="pt")
-    outputs = model.generate(input_ids)
-    questions = tokenizer.decode(outputs[0])
+    inputs = tokenizer.encode("generate questions: " + text, return_tensors="pt", max_length=512, truncation=True)
+    outputs = model.generate(inputs, max_length=512, num_beams=4, early_stopping=True)
+    questions = [tokenizer.decode(output, skip_special_tokens=True) for output in outputs]
     
-    return questions
+    # O modelo retorna apenas as perguntas, então precisamos criar respostas fictícias para o exemplo
+    qas = [{"question": question, "answer": "answer", "answer_start": 0} for question in questions]
+    
+    return qas
 
 # Função para converter os pares de QA no formato SQuAD
 def convert_to_squad_format(qas, context):
@@ -33,8 +35,8 @@ def convert_to_squad_format(qas, context):
             "question": qa['question'],
             "id": str(i),
             "answers": {
-                "answer_start": [qa['answer']['start']],
-                "text": [qa['answer']['text']]
+                "answer_start": [qa['answer_start']],
+                "text": [qa['answer']]
             }
         }
         squad_data.append(entry)