Update app.py

app.py

@@ -79,7 +79,6 @@ def clean_text(text: str) -> str:
 
     return text
 
-# Vocabulary class unchanged...
 class Vocabulary:
     def __init__(self):
         self.word2id = dict()
@@ -187,11 +186,10 @@ bidirectional = False
 dropout = 0.3 
 pad_idx = vocab["<pad>"]
 unk_idx = vocab["<unk>"]
-n_classes = 3  # positive, neutral, negative
+n_classes = 3 
 
 label_map = {0: 'tiêu cực', 1: 'bình thường', 2: 'tích cực'}
 
-# Ensure model and its weights moved to correct device
 def load_model(path: str):
     model = RNN(input_dim, embedding_dim, hidden_dim, n_layers, bidirectional, dropout, pad_idx, n_classes) 
     model.load_state_dict(torch.load(path, map_location=device))
@@ -201,7 +199,6 @@ def load_model(path: str):
 
 model = load_model(model_path)
 
-# Prediction helper
 def predict_sentiment(model, sentence, vocab, label_mapping=None):
     tensor = vocab.corpus_to_tensor([sentence])[0]
     length = torch.LongTensor([tensor.size(0)]).to(device)
@@ -216,11 +213,9 @@ def predict_sentiment(model, sentence, vocab, label_mapping=None):
 def process_input(text_input, file):
     comments = []
 
-    # Xử lý văn bản nhập trực tiếp
     if text_input:
         comments += [line.strip() for line in text_input.splitlines() if line.strip()]
 
-    # Xử lý tệp tải lên
     if file is not None:
         file_name = file.name
         
@@ -234,14 +229,12 @@ def process_input(text_input, file):
             comments += [line.strip() for line in content.splitlines() if line.strip()] 
         
         elif file_name.endswith('.csv'):
-            df = pd.read_csv(file, encoding='utf-8')
-            comments += df.iloc[:, 0].dropna().astype(str).tolist()
+            df = pd.read_csv(file, header=None, names=['Comment'], encoding='utf-8')
+            comments = df['Comment'].dropna().astype(str).tolist()
 
-    # Kiểm tra nếu không có bình luận nào được cung cấp
     if not comments:
         return pd.DataFrame(columns=["Comment", "Dự đoán", "Xác suất"])
 
-    # Dự đoán cảm xúc cho từng bình luận
     results = []
     for comment in comments:
         label, probability = predict_sentiment(model, comment, vocab, label_map)