Update app.py

app.py

@@ -1,48 +1,51 @@
+import os
 import gradio as gr
 import joblib
-import torch
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
 
-# ✅ Use folder name directly (no ./)
-model_dir = "campaign-bert-model"
+# ✅ Detect environment and set paths
+if os.path.exists("/home/user/app"):
+    model_dir = "campaign-bert-model"  # Hugging Face Space
+    label_path = "label_encoder.pkl"
+else:
+    model_dir = r"C:\Users\gopic\campaign_bert_model\campaign_bert_model\campaign-bert-model"
+    label_path = r"C:\Users\gopic\campaign_bert_model\campaign_bert_model\label_encoder.pkl"
 
+# ✅ Load tokenizer, model, and label encoder
 try:
     tokenizer = AutoTokenizer.from_pretrained(model_dir, use_fast=False)
     model = AutoModelForSequenceClassification.from_pretrained(model_dir)
-    label_encoder = joblib.load("label_encoder.pkl")
-    print("✅ Model, Tokenizer, and Label Encoder loaded.")
+    label_encoder = joblib.load(label_path)
 except Exception as e:
     raise RuntimeError(f"❌ Failed to load model or tokenizer: {e}")
 
-# ✅ Prediction logic
+# ✅ Define prediction function
 def predict(client_interest, sentiment):
     try:
         text = f"{client_interest} Sentiment: {sentiment}"
         inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-        outputs = model(**inputs)
-        predicted_id = torch.argmax(outputs.logits, dim=1).item()
-        label = label_encoder.inverse_transform([predicted_id])[0]
-        confidence = torch.softmax(outputs.logits, dim=1)[0][predicted_id].item()
-        return f"🎯 Template: **{label}**\n📊 Confidence: **{confidence:.2%}**"
-    except Exception as err:
-        return f"❌ Prediction failed: {err}"
-
-# ✅ UI with styling
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("## 🎯 Campaign Personalizer")
-    gr.Markdown("Provide interest and sentiment to get the best marketing template.")
-
+        with torch.no_grad():
+            outputs = model(**inputs)
+            logits = outputs.logits
+            prediction = torch.argmax(logits, dim=1).item()
+            template = label_encoder.inverse_transform([prediction])[0]
+            confidence = torch.softmax(logits, dim=1)[0][prediction].item()
+            return f"📄 Suggested Template: {template}\n🔍 Confidence: {confidence:.2%}"
+    except Exception as e:
+        return f"❌ Error in prediction: {e}"
+
+# ✅ Build Gradio UI
+with gr.Blocks(css=".gradio-container {background: #f0f4f8;}") as demo:
+    gr.Markdown("<h1 style='text-align: center; color: #1f2937;'>📢 Campaign Personalizer</h1>")
     with gr.Row():
-        with gr.Column():
-            interest = gr.Textbox(label="📝 Client Interest")
-            sentiment = gr.Textbox(label="😊 Sentiment")
-            submit = gr.Button("🔍 Predict")
-        with gr.Column():
-            result = gr.Markdown(label="🎯 Output")
+        client_interest = gr.Textbox(label="Client Interest", placeholder="e.g. Term Insurance for family", lines=2)
+        sentiment = gr.Textbox(label="Sentiment", placeholder="e.g. Positive", lines=2)
+    predict_btn = gr.Button("🔍 Predict Template")
+    output = gr.Textbox(label="Prediction Result")
 
-    submit.click(predict, inputs=[interest, sentiment], outputs=result)
+    predict_btn.click(fn=predict, inputs=[client_interest, sentiment], outputs=output)
 
-# ✅ Launch app
+# ✅ Launch
 if __name__ == "__main__":
     demo.launch()
-