Upload app.py with huggingface_hub

app.py

@@ -1,143 +1,44 @@
 import gradio as gr
 from transformers import pipeline
-import torch
 
-# Load the NEW menu-aware model (v2)
-# This model doesn't have pipeline.py - it's just a fine-tuned DistilBERT
-model_id = "samurai9776/thought-classifier-menu-aware"
-
-print(f"Loading model: {model_id}")
-
-# Load without trust_remote_code (new model doesn't need it)
+# Load the clean v2 model
 classifier = pipeline(
     "text-classification", 
-    model=model_id,
-    device=0 if torch.cuda.is_available() else -1
+    model="samurai9776/thought-classifier-menu-aware"
 )
 
-print("✅ Model loaded successfully!")
-
 def classify_thought(ai_utterance, cx_utterance):
-    """Classify if the conversation is complete or incomplete"""
-    
     if not ai_utterance or not cx_utterance:
-        return "Please enter both utterances", 0, 0, ""
+        return "Please enter both utterances", 0, 0
     
-    # Combine utterances with [SEP] token
     text = f"{ai_utterance} [SEP] {cx_utterance}"
-    
-    # Get prediction
     results = classifier(text)
     
-    # Extract scores
     scores = {r['label']: r['score'] for r in results}
     complete_score = scores.get('COMPLETE', 0)
     incomplete_score = scores.get('INCOMPLETE', 0)
     
-    # Determine prediction
     prediction = "Complete ✓" if complete_score > incomplete_score else "Incomplete ⚠️"
-    
-    # Determine confidence
-    confidence = max(complete_score, incomplete_score)
-    confidence_text = f"{confidence:.1%} confidence"
-    
-    return prediction, complete_score, incomplete_score, confidence_text
+    return prediction, complete_score, incomplete_score
 
-# Create Gradio interface
-with gr.Blocks(
-    title="Thought Completion Classifier - Menu Aware v2",
-    theme=gr.themes.Soft()
-) as demo:
-    gr.Markdown("""
-    # 🤖 Thought Completion Classifier - Menu Aware v2
-    
-    Enhanced model with better context understanding and menu awareness.
-    
-    ### ✨ What's New in v2:
-    - ✅ Trained on 15,000+ examples (vs 900 in v1)
-    - ✅ Fixed 3,000+ mislabeled training examples
-    - ✅ Handles ambiguous terms like "chicken" (75 menu items!)
-    - ✅ Better context understanding
-    - ✅ ~92% accuracy
-    
-    ### 🎯 How it works:
-    The model understands context:
-    - "What sandwich?" + "Chicken" = **Complete** (context is clear)
-    - "Anything else?" + "Chicken" = **Incomplete** (ambiguous)
-    """)
+with gr.Blocks(title="Thought Classifier v2") as demo:
+    gr.Markdown("# 🤖 Thought Completion Classifier v2")
     
     with gr.Row():
         with gr.Column():
-            ai_input = gr.Textbox(
-                label="AI Utterance",
-                placeholder="e.g., What sandwich would you like?",
-                lines=2
-            )
-            cx_input = gr.Textbox(
-                label="Customer Utterance", 
-                placeholder="e.g., Chicken",
-                lines=2
-            )
-            
-            classify_btn = gr.Button("🔍 Classify", variant="primary", size="lg")
+            ai_input = gr.Textbox(label="AI Utterance", placeholder="What sandwich?")
+            cx_input = gr.Textbox(label="Customer Utterance", placeholder="Chicken")
+            classify_btn = gr.Button("Classify", variant="primary")
             
         with gr.Column():
-            prediction = gr.Textbox(
-                label="Prediction", 
-                interactive=False,
-                elem_classes=["prediction-output"]
-            )
-            with gr.Row():
-                complete_score = gr.Number(
-                    label="Complete Score", 
-                    precision=3, 
-                    interactive=False
-                )
-                incomplete_score = gr.Number(
-                    label="Incomplete Score", 
-                    precision=3, 
-                    interactive=False
-                )
-            confidence = gr.Textbox(
-                label="Confidence", 
-                interactive=False
-            )
-    
-    # Examples showing context awareness
-    gr.Markdown("### 📝 Try these examples to see context awareness:")
-    gr.Examples(
-        examples=[
-            ["What can I get for you?", "Chicken", "Should be INCOMPLETE (ambiguous)"],
-            ["What sandwich would you like?", "Chicken", "Should be COMPLETE (context clear)"],
-            ["Chicken or beef?", "Chicken", "Should be COMPLETE (direct answer)"],
-            ["Chicken or beef?", "Large", "Should be INCOMPLETE (doesn't answer question)"],
-            ["What size?", "Large", "Should be COMPLETE (answers size question)"],
-            ["Anything else?", "Large", "Should be INCOMPLETE (size without item)"],
-            ["Ready to order?", "Spicy Chicken Sandwich Combo", "Should be COMPLETE (full item)"],
-            ["Anything else?", "That's all", "Should be COMPLETE (clear ending)"],
-        ],
-        inputs=[ai_input, cx_input],
-        outputs=[prediction, complete_score, incomplete_score, confidence],
-        fn=classify_thought,
-        cache_examples=True,
-    )
+            prediction = gr.Textbox(label="Prediction")
+            complete_score = gr.Number(label="Complete Score")
+            incomplete_score = gr.Number(label="Incomplete Score")
     
     classify_btn.click(
-        fn=classify_thought,
+        classify_thought,
         inputs=[ai_input, cx_input],
-        outputs=[prediction, complete_score, incomplete_score, confidence]
+        outputs=[prediction, complete_score, incomplete_score]
     )
-    
-    gr.Markdown("""
-    ---
-    ### 📊 Model Comparison:
-    - **v1**: Rule-based + neural (samurai9776/thought-classifier)
-    - **v2**: Pure neural with better training (samurai9776/thought-classifier-menu-aware)
-    
-    ### 🔗 Links:
-    - [Model v2 (Current)](https://huggingface.co/samurai9776/thought-classifier-menu-aware)
-    - [Model v1 (Previous)](https://huggingface.co/samurai9776/thought-classifier)
-    """)
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()