Update app.py

app.py

@@ -45,9 +45,12 @@ def predict(input, history=[]):
     if is_question: 
         sql_encoding = sql_tokenizer(table=table, query=input + sql_tokenizer.eos_token, return_tensors="pt")
         sql_outputs = sql_model.generate(**sql_encoding)
-        response = sql_tokenizer.batch_decode(sql_outputs, skip_special_tokens=True)
+        sql_response = sql_tokenizer.batch_decode(sql_outputs, skip_special_tokens=True)
 
-        history.append(response)
+        # Convert the SQL model's response to token IDs and add to the history
+        sql_response_ids = tokenizer.encode(sql_response + tokenizer.eos_token, return_tensors='pt')
+        history.extend(sql_response_ids.squeeze().tolist())
+        
         '''
         bot_input_ids = torch.cat([torch.LongTensor(history), sql_encoding], dim=-1)
         history = sql_model.generate(bot_input_ids, max_length=1000, pad_token_id=sql_tokenizer.eos_token_id).tolist()