fix code

app.py

@@ -14,8 +14,8 @@ from transformers import TFAutoModel, AutoTokenizer
 from sklearn.model_selection import train_test_split
 
 # load the tokenizer and transformer model
-tokenizer = AutoTokenizer.from_pretrained("nlptown/flaubert_small_cased_sentiment",max_length=60) #xlm-roberta-base bert-base-multilingual-cased
-transformer_model = TFAutoModel.from_pretrained("nlptown/flaubert_small_cased_sentiment") #philschmid/tiny-bert-sst2-distilled
+tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-base",max_length=60) #xlm-roberta-base bert-base-multilingual-cased
+transformer_model = TFAutoModel.from_pretrained("xlm-roberta-base") #philschmid/tiny-bert-sst2-distilled
 max_seq_length = 32
 
 def create_model():
@@ -86,6 +86,7 @@ def predict(text):
   test_padded_sequences = pad_sequences(new_input_ids, maxlen=max_seq_length,padding='post',truncating='post',value=1) #post pre
   print(test_padded_sequences.shape)
   predicted_labels = model.predict(test_padded_sequences)
+  output = []
 
   for i in range(len(test_texts)):
     print(test_texts[i])
@@ -112,7 +113,18 @@ def predict(text):
     print(f'hour : {hour_label}')
     print(f'minute : {minute_label}')
     print('----')
-  return 'hello'
+    tmp = {
+      'valid' : valid,
+      'is_scene' : is_scene,
+      'has_num' : has_num,
+      'turn_on_off' : turn,
+      'device' : 'ไฟ',
+      'env' : env_label,
+      'hour' : hour,
+      'minute' : minute,
+    }
+  output.append(tmp)
+  return output
 
 iface = gr.Interface(
   fn=predict,