Update app.py

app.py

@@ -1,95 +1,124 @@
 import gradio as gr
 import openai
-import os
 
 # Set OpenAI API Key
-openai.api_key = os.getenv("TRY_NEW_THINGS")
+openai.api_key = "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"
 openai.api_base = "https://api.groq.com/openai/v1"
 
-# Function to get response from GROQ API based on category keywords
-def get_groq_response(message, category):
-    system_message = ""
-    if category == "Stress Management":
-        system_message = "Provide soothing advice and tips to help the user manage stress. Be calm and empathetic."
-    elif category == "Career Advice":
-        system_message = "Provide professional and constructive career advice. Be encouraging and helpful."
-    elif category == "General":
-        system_message = "Provide general conversation. Be friendly and easygoing."
-    elif category == "Friendly Buddy":
-        system_message = "Respond as a supportive and fun friend. Be informal and light-hearted."
+# Dictionary to store categorized chats
+saved_chats = {
+    "Stress Management": [],
+    "Career Advice": [],
+    "General": [],
+    "Suggestions": []
+}
 
+# Function to get response from GROQ API
+def get_groq_response(message):
     try:
         response = openai.ChatCompletion.create(
             model="llama-3.1-70b-versatile",
             messages=[
-                {"role": "system", "content": system_message},
-                {"role": "user", "content": message}
+                {"role": "user", "content": message},
+                {"role": "system", "content": "You will talk like a Motivational Speaker to help people come out of stress."}
             ]
         )
         return response.choices[0].message["content"]
     except Exception as e:
         return f"Error: {str(e)}"
 
-# Function to recognize the category based on keywords
-def recognize_category(user_input):
-    stress_keywords = ["stress", "anxiety", "overwhelmed", "burnout", "calm"]
-    career_keywords = ["career", "job", "work", "interview", "skills"]
-    general_keywords = ["hello", "hi", "how", "what", "why", "chat", "talk"]
-    friendly_keywords = ["friend", "buddy", "hangout", "fun", "relax"]
-
-    if any(keyword in user_input.lower() for keyword in stress_keywords):
+# Function to classify messages based on the topic
+def classify_message(user_message, bot_response):
+    if "stress" in user_message.lower():
+        saved_chats["Stress Management"].append((user_message, bot_response))
         return "Stress Management"
-    elif any(keyword in user_input.lower() for keyword in career_keywords):
+    elif "career" in user_message.lower():
+        saved_chats["Career Advice"].append((user_message, bot_response))
         return "Career Advice"
-    elif any(keyword in user_input.lower() for keyword in general_keywords):
-        return "General"
-    elif any(keyword in user_input.lower() for keyword in friendly_keywords):
-        return "Friendly Buddy"
+    elif "suggestions" in user_message.lower():
+        saved_chats["Suggestions"].append((user_message, bot_response))
+        return "Suggestions"
     else:
-        return "General"  # Default category
-
-# Function to process input and generate category-specific responses
-def process_input(user_input, history):
-    category = recognize_category(user_input)
-    bot_response = get_groq_response(user_input, category)
-    history.append((category, user_input, bot_response))
+        saved_chats["General"].append((user_message, bot_response))
+        return "General"
 
-    # Group responses by category
-    stress_responses = [f"**User**: {u}\n**Bot**: {b}" for c, u, b in history if c == "Stress Management"]
-    career_responses = [f"**User**: {u}\n**Bot**: {b}" for c, u, b in history if c == "Career Advice"]
-    general_responses = [f"**User**: {u}\n**Bot**: {b}" for c, u, b in history if c == "General"]
-    friendly_responses = [f"**User**: {u}\n**Bot**: {b}" for c, u, b in history if c == "Friendly Buddy"]
+# Chatbot function
+def chatbot(user_input, history=[]):
+    bot_response = get_groq_response(user_input)
+    topic = classify_message(user_input, bot_response)
+    history.append((f"({topic}) You: {user_input}", f"Motivator Bot: {bot_response}"))
+    return history, saved_chats
 
-    # Return grouped responses for display
+# Function to display saved chats
+def display_saved_chats():
+    def format_chats(category):
+        return "\n".join([f"**You**: {u}\n**Bot**: {b}" for u, b in saved_chats[category]]) or "No messages yet."
+    
     return (
-        "\n\n".join(stress_responses) or "No responses yet.",
-        "\n\n".join(career_responses) or "No responses yet.",
-        "\n\n".join(general_responses) or "No responses yet.",
-        "\n\n".join(friendly_responses) or "No responses yet.",
-        history
+        format_chats("Stress Management"),
+        format_chats("Career Advice"),
+        format_chats("General"),
+        format_chats("Suggestions")
     )
 
-# Gradio UI with Blocks
-with gr.Blocks() as demo:
-    gr.Markdown("## Meet Your Personal Assistant: A Caring Chatbot")
-    gr.Markdown("This chatbot helps with stress management, career advice, general queries, or just being a friendly buddy!")
+# Gradio Interface setup
+chat_interface = gr.Blocks(css="""
+body {
+    font-family: 'Poppins', sans-serif;
+    background: linear-gradient(45deg, #ff9a9e, #fad0c4, #fbc2eb, #a1c4fd, #c2e9fb);
+    background-size: 400% 400%;
+    animation: gradientBG 10s ease infinite;
+    margin: 0;
+    padding: 0;
+    color: #333;
+}
 
-    with gr.Row():
-        user_input = gr.Textbox(label="Your Message", placeholder="Type your message here...")
-        submit_button = gr.Button("Submit")
+@keyframes gradientBG {
+    0% { background-position: 0% 50%; }
+    50% { background-position: 100% 50%; }
+    100% { background-position: 0% 50%; }
+}
 
+header, footer {
+    text-align: center;
+    background: linear-gradient(90deg, #ff758c, #ff7eb3);
+    color: white;
+    padding: 1rem;
+    border-radius: 15px;
+    margin-bottom: 1rem;
+    box-shadow: 0px 4px 15px rgba(0, 0, 0, 0.2);
+}
+""")
+
+with chat_interface:
+    with gr.Row():
+        gr.Markdown("<h1 style='text-align:center;'>🌈 Vibrant Motivational Chatbot</h1>")
     with gr.Row():
-        stress_column = gr.Textbox(label="Stress Management", interactive=False)
-        career_column = gr.Textbox(label="Career Advice", interactive=False)
-        general_column = gr.Textbox(label="General", interactive=False)
-        friendly_column = gr.Textbox(label="Friendly Buddy", interactive=False)
+        user_input = gr.Textbox(label="Your Message", placeholder="Type something...")
+        send_button = gr.Button("Send")
+    with gr.Row():
+        chatbot_output = gr.Chatbot(label="Chat History")
 
-    history = gr.State([])
+    with gr.Row():
+        with gr.Column():
+            stress_display = gr.Textbox(label="Stress Management", interactive=False, lines=10)
+        with gr.Column():
+            career_display = gr.Textbox(label="Career Advice", interactive=False, lines=10)
+        with gr.Column():
+            general_display = gr.Textbox(label="General", interactive=False, lines=10)
+        with gr.Column():
+            suggestions_display = gr.Textbox(label="Suggestions", interactive=False, lines=10)
+    
+    def handle_interaction(user_input, history):
+        if not user_input.strip():
+            return history, *display_saved_chats()
+        updated_history, _ = chatbot(user_input, history)
+        return updated_history, *display_saved_chats()
 
-    submit_button.click(
-        fn=process_input,
-        inputs=[user_input, history],
-        outputs=[stress_column, career_column, general_column, friendly_column, history]
+    send_button.click(
+        fn=handle_interaction,
+        inputs=[user_input, chatbot_output],
+        outputs=[chatbot_output, stress_display, career_display, general_display, suggestions_display]
     )
 
-demo.launch()
+chat_interface.launch()