Update app.py

app.py

@@ -1,62 +1,62 @@
-import gradio as gr
-import torch
-from transformers import AutoTokenizer, AutoModelForSequenceClassification
-import torch.nn.functional as F
-
-# ==============================
-# 1. Load Model
-# ==============================
-MODEL_ID = "anhgf/visec-phobert-sentiment-vi"
-DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
-model = AutoModelForSequenceClassification.from_pretrained(MODEL_ID)
-model.to(DEVICE)
-model.eval()
-
-LABELS = ["negative", "positive"]
-
-# ==============================
-# 2. Inference Function
-# ==============================
-def classify_sentiment(text):
-    if not text or text.strip() == "":
-        return {"label": "empty input", "probabilities": {}}
-
-    # Tokenize
-    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-    inputs = {k: v.to(DEVICE) for k, v in inputs.items()}
-
-    # Model forward
-    with torch.no_grad():
-        logits = model(**inputs).logits
-        softmax = F.softmax(logits, dim=-1)[0].cpu().tolist()
-
-    # Because model only has 2 classes: [neg, pos]
-    neg_p = softmax[0]
-    pos_p = softmax[1]
-
-    # Predict label
-    label = "negative" if neg_p > pos_p else "positive"
-
-    return {
-        "label": label,
-        "probabilities": {
-            "negative": float(neg_p),
-            "positive": float(pos_p)
-        }
-    }
-
-# ==============================
-# 3. Gradio UI
-# ==============================
-app = gr.Interface(
-    fn=classify_sentiment,
-    inputs=gr.Textbox(lines=3, label="Nhập văn bản tiếng Việt"),
-    outputs=gr.JSON(label="Kết quả phân tích cảm xúc (POS / NEG)"),
-    title="Vietnamese Sentiment Classification (PhoBERT)",
-    description="Model này chỉ có 2 lớp: Positive và Negative."
-)
-
-if __name__ == "__main__":
-    app.launch()
+import gradio as gr
+import torch
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch.nn.functional as F
+
+# ==============================
+# 1. Load Model
+# ==============================
+MODEL_ID = "anhgf/visec-phobert-sentiment-vi"
+DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+model = AutoModelForSequenceClassification.from_pretrained(MODEL_ID)
+model.to(DEVICE)
+model.eval()
+
+LABELS = ["negative", "positive"]
+
+# ==============================
+# 2. Inference Function
+# ==============================
+def classify_sentiment(text):
+    if not text or text.strip() == "":
+        return {"label": "empty input", "probabilities": {}}
+
+    # Tokenize
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
+    inputs = {k: v.to(DEVICE) for k, v in inputs.items()}
+
+    # Model forward
+    with torch.no_grad():
+        logits = model(**inputs).logits
+        softmax = F.softmax(logits, dim=-1)[0].cpu().tolist()
+
+    # Because model only has 2 classes: [neg, pos]
+    neg_p = softmax[0]
+    pos_p = softmax[1]
+
+    # Predict label
+    label = "negative" if neg_p > pos_p else "positive"
+
+    return {
+        "label": label,
+        # "probabilities": {
+        #     "negative": float(neg_p),
+        #     "positive": float(pos_p)
+        # }
+    }
+
+# ==============================
+# 3. Gradio UI
+# ==============================
+app = gr.Interface(
+    fn=classify_sentiment,
+    inputs=gr.Textbox(lines=3, label="Nhập văn bản tiếng Việt"),
+    outputs=gr.JSON(label="Kết quả phân tích cảm xúc (POS / NEG)"),
+    title="Vietnamese Sentiment Classification (PhoBERT)",
+    description="Model này chỉ có 2 lớp: Positive và Negative."
+)
+
+if __name__ == "__main__":
+    app.launch()