Update app.py

app.py

@@ -2,48 +2,76 @@ import os
 import gradio as gr
 import requests
 
-# Set up Hugging Face Inference API
+# Use Mistral via Hugging Face Inference API (requires token)
 API_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-7B-Instruct-v0.1"
 headers = {"Authorization": f"Bearer {os.getenv('HUGGINGFACE_TOKEN')}"}
 
+# Example prompts
+preset_prompts = [
+    "I finally got the promotion, but I feel guilty because my best friend got laid off.",
+    "Moving to a new city is exciting, but leaving my family breaks my heart.",
+    "I passed the test, but my friend failed — and I don’t know how to feel.",
+    "They applauded me on stage, but all I could think about was how lonely I felt.",
+    "I’m happy for her, but I wish I had that too.",
+]
+
+# Call Mistral Inference API with error handling
 def call_model(prompt):
     payload = {"inputs": prompt}
-    response = requests.post(API_URL, headers=headers, json=payload)
-    
     try:
-        return response.json()[0]["generated_text"]
-    except (KeyError, IndexError, TypeError):
-        return "⚠️ Error: Could not generate a valid response. Check API status or token."
+        response = requests.post(API_URL, headers=headers, json=payload, timeout=45)
+
+        print("Status:", response.status_code)
+        print("Raw:", response.text)
+
+        if response.status_code != 200:
+            return f"⚠️ Error: API returned {response.status_code} - model may be loading or access may be restricted."
+
+        data = response.json()
+        if isinstance(data, list) and "generated_text" in data[0]:
+            return data[0]["generated_text"]
+        elif isinstance(data, dict) and "error" in data:
+            return f"⚠️ API Error: {data['error']}"
+        else:
+            return "⚠️ Unexpected response format from model."
+
+    except Exception as e:
+        return f"⚠️ Request failed: {str(e)}"
 
-# Emotion disambiguation pipeline
+# Emotion analysis logic
 def emotion_annotator(text):
-    # Step 1: List candidate emotions
     prompt1 = f"""You are an expert in human emotions. Based on the sentence below, list all possible emotions the person might be feeling.
 
 Sentence: "{text}"
 Emotions:"""
     candidates = call_model(prompt1)
 
-    # Step 2: Choose most likely one
-    prompt2 = f"""You are an emotion disambiguation expert. Given the sentence "{text}" and the following candidate emotions: {candidates}, pick the most likely one and explain why.
+    prompt2 = f"""Given the sentence "{text}" and the following candidate emotions: {candidates}, pick the most likely emotion and explain why.
 
 Format:
 Most likely emotion: <emotion>
-Reason: <why you chose it>"""
+Reason: <why>"""
     final = call_model(prompt2)
 
     return candidates.strip(), final.strip()
 
 # Gradio UI
-demo = gr.Interface(
-    fn=emotion_annotator,
-    inputs=gr.Textbox(lines=2, placeholder="e.g., I’m proud but I feel like I let them down."),
-    outputs=[
-        gr.Textbox(label="Candidate Emotions"),
-        gr.Textbox(label="Most Likely Emotion + Explanation")
-    ],
-    title="🧠 Emotion Annotator AI (via Hugging Face Inference API)",
-    description="Uses Mistral 7B Instruct on Hugging Face to analyze and disambiguate mixed emotions from text input. No model download needed!"
-)
+with gr.Blocks() as demo:
+    gr.Markdown("## 🧠 Emotion Annotator AI (Mistral 7B via Hugging Face API)")
+    gr.Markdown("Disambiguates mixed emotions using `mistralai/Mistral-7B-Instruct-v0.1`. Token required.")
+
+    with gr.Row():
+        text_input = gr.Textbox(label="Enter your sentence", placeholder="e.g., I’m proud but I feel like I let them down.", lines=2)
+        dropdown = gr.Dropdown(preset_prompts, label="💬 Choose an example")
+
+    run_button = gr.Button("Submit")
+
+    with gr.Row():
+        candidate_output = gr.Textbox(label="🧠 Candidate Emotions")
+        final_output = gr.Textbox(label="🎯 Most Likely Emotion + Explanation")
+
+    # Auto-fill from dropdown
+    dropdown.change(fn=lambda x: x, inputs=dropdown, outputs=text_input)
+    run_button.click(fn=emotion_annotator, inputs=text_input, outputs=[candidate_output, final_output])
 
 demo.launch()