Update app.py

app.py

@@ -1,45 +1,40 @@
 import gradio as gr
-from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
-import json
-
-# Load models
-chatbot_model = "microsoft/DialoGPT-medium"
-def chatbot_response(message, session_id="default"):
-    if session_id not in chat_histories:
-        chat_histories[session_id] = []
-    
-    # Generate response
-    input_ids = tokenizer.encode(message + tokenizer.eos_token, return_tensors="pt")
-    output = model.generate(input_ids, max_length=200, pad_token_id=tokenizer.eos_token_id)
-    response = tokenizer.decode(output[:, input_ids.shape[-1]:][0], skip_special_tokens=True)
+import requests
+
+# Define the sentiment analysis function that communicates with the Hugging Face model API
+def analyze_sentiment(message: str):
+    try:
+        # Send request to the Hugging Face Space for sentiment analysis
+        response = requests.post(
+            'https://safwansajad-emotion-detection-gpt.hf.space/predict',
+            json={'text': message},
+            headers={'Content-Type': 'application/json'}
+        )
+        
+        # Extract the sentiment and score from the response
+        data = response.json()
+        
+        # Return the label and score in the format expected by your app
+        if 'emotion' in data and 'score' in data:
+            return [{"label": data['emotion'], "score": data['score']}]
+        else:
+            return [{"label": "Unknown", "score": 0}]
     
-    # Detect emotion
-    emotion_result = emotion_pipeline(message)
-    emotion = emotion_result[0]["label"]
-    score = float(emotion_result[0]["score"])
-    
-    # Store history
-    chat_histories[session_id].append((message, response))
-    return response, emotion, score
-
-# Gradio Interface (Primary for Spaces)
-with gr.Blocks() as demo:
-    gr.Markdown("# 🤖 Mental Health Chatbot")
-    with gr.Row():
-
-    btn.click(respond, [msg, chatbot, session_id], [msg, chatbot, emotion_out, score_out])
-    msg.submit(respond, [msg, chatbot, session_id], [msg, chatbot, emotion_out, score_out])
-    clear_btn.click(lambda s_id: ([], "", 0.0) if s_id in chat_histories else ([], "", 0.0), 
-                   inputs=[session_id], 
-                   outputs=[chatbot, emotion_out, score_out])
-
-
-
-
-
-
-
-
-# For Hugging Face Spaces, Gradio must be the main interface
+    except Exception as e:
+        print(f"Error during sentiment analysis: {e}")
+        return [{"label": "Error", "score": 0}]
+
+# Set up Gradio interface
+iface = gr.Interface(
+    fn=analyze_sentiment,  # Function to be called
+    inputs=gr.Textbox(label="Enter your message", placeholder="How are you feeling today?", lines=2),  # User input
+    outputs=gr.JSON(),  # Output in JSON format (label and score)
+    live=True,  # Enables live input processing
+    title="Sentiment Analysis with SerenityAI",  # Title of the interface
+    description="Enter a message and get feedback about your emotional state. Your feelings matter!",
+    theme="huggingface",  # Optionally set the theme to Hugging Face's style
+)
+
+# Launch the Gradio app
 if __name__ == "__main__":
-    demo.launch()
+    iface.launch(share=True)  # share=True gives you a public link