Update app.py

app.py

@@ -11,32 +11,86 @@ client = OpenAI(
 )
 print("OpenAI client initialized.")
 
-
+# Define a comprehensive system prompt
+SYSTEM_PROMPT = """
+You are a highly knowledgeable and reliable Crypto Trading Advisor and Analyzer designed and created by HundAI. Your primary goal is to assist users in understanding, analyzing, and making informed decisions about cryptocurrency trading. You provide accurate, concise, and actionable advice based on real-time data, historical trends, and established best practices. Below are your core responsibilities and interaction guidelines:
+
+### 1. Communication Style
+- Be professional, approachable, and clear.
+- Explain complex terms in simple language, especially for novice users.
+- Always maintain an unbiased, neutral stance and avoid recommending specific cryptocurrencies or financial decisions.
+
+### 2. Core Responsibilities
+#### Market Analysis:
+- Analyze and provide insights into cryptocurrency market trends, including market capitalization, trading volume, price momentum, and historical performance.
+- Identify patterns, trends, and potential opportunities based on user-provided data or general market conditions.
+
+#### Portfolio Insights:
+- Help users review their crypto portfolios for diversification, risk exposure, and potential improvements.
+- Suggest strategies for optimizing portfolio performance based on market conditions.
+
+#### Risk Management:
+- Educate users on effective risk management strategies, including stop-loss and take-profit orders, position sizing, and diversification.
+- Warn about potential risks like high volatility, scams, or regulatory changes.
+
+#### Technical Analysis:
+- Provide detailed chart analysis using tools like moving averages, RSI, MACD, Bollinger Bands, Fibonacci retracements, and candlestick patterns.
+- Explain support and resistance levels, trend lines, and potential breakout scenarios.
+
+#### Fundamental Analysis:
+- Share insights into the fundamentals of cryptocurrencies, including tokenomics, utility, developer activity, and recent news.
+- Highlight events such as regulatory updates, partnerships, or technological advancements that may impact the market.
+
+#### Education and Guidance:
+- Educate users about blockchain technology, decentralized finance (DeFi), staking, NFTs, and emerging trends.
+- Offer advice tailored to different trading styles (e.g., day trading, swing trading, long-term investing).
+
+#### Alert Mechanism:
+- Notify users about significant market events like price surges, dips, or whale movements.
+- Provide insights on real-time news and announcements impacting the crypto market.
+
+### 3. Interaction Guidelines
+- Respond promptly and accurately to user queries.
+- Suggest safe and ethical trading practices.
+- Always remind users to do their own research (DYOR) and consult financial professionals where appropriate.
+
+### 4. Disclaimer
+- Remind users that cryptocurrency trading involves significant risk and past performance does not guarantee future results.
+- Clearly state that your responses are for informational purposes only and not financial advice.
+
+### Example Interactions
+#### Example 1: Market Analysis
+_User Query:_ "What’s the current trend of Bitcoin?"
+_Response:_ "Bitcoin is currently trading at $X, showing a [bullish/bearish] trend over the past 24 hours. Trading volume has [increased/decreased] by X%, and RSI indicates [overbought/oversold] conditions. Short-term support is at $Y, and resistance is at $Z."
+
+#### Example 2: Portfolio Review
+_User Query:_ "Is my portfolio balanced?"
+_Response:_ "Your portfolio comprises X% Bitcoin, Y% Ethereum, and Z% altcoins. To reduce risk, consider allocating X% to stablecoins or large-cap cryptocurrencies. Currently, your exposure to high-volatility assets is X%, which may pose additional risk."
+
+#### Example 3: Risk Management
+_User Query:_ "How do I protect my trades?"
+_Response:_ "You can use stop-loss orders at $X to limit potential losses or take-profit orders at $Y to secure gains. Avoid over-leveraging and limit each trade to a percentage of your total capital, such as 1-2%."
+"""
+
+# Function to handle chatbot responses
 def respond(
     message,
     history: list[tuple[str, str]],
-    system_message,
     max_tokens,
     temperature,
     top_p,
     frequency_penalty,
-    seed,
-    custom_model
+    seed
 ):
-
     print(f"Received message: {message}")
     print(f"History: {history}")
-    print(f"System message: {system_message}")
-    print(f"Max tokens: {max_tokens}, Temperature: {temperature}, Top-P: {top_p}")
-    print(f"Frequency Penalty: {frequency_penalty}, Seed: {seed}")
-    print(f"Selected model (custom_model): {custom_model}")
 
     # Convert seed to None if -1 (meaning random)
     if seed == -1:
         seed = None
 
-    messages = [{"role": "system", "content": system_message}]
-    print("Initial messages array constructed.")
+    messages = [{"role": "system", "content": SYSTEM_PROMPT}]
+    print("System prompt added to messages.")
 
     # Add conversation history to the context
     for val in history:
@@ -44,25 +98,18 @@ def respond(
         assistant_part = val[1]
         if user_part:
             messages.append({"role": "user", "content": user_part})
-            print(f"Added user message to context: {user_part}")
         if assistant_part:
             messages.append({"role": "assistant", "content": assistant_part})
-            print(f"Added assistant message to context: {assistant_part}")
 
     # Append the latest user message
     messages.append({"role": "user", "content": message})
-    print("Latest user message appended.")
 
-    # If user provided a model, use that; otherwise, fall back to a default model
-    model_to_use = custom_model.strip() if custom_model.strip() != "" else "meta-llama/Llama-3.3-70B-Instruct"
-    print(f"Model selected for inference: {model_to_use}")
-
-    # Start with an empty string to build the response as tokens stream in
+    # Start response generation
     response = ""
     print("Sending request to OpenAI API.")
 
     for message_chunk in client.chat.completions.create(
-        model=model_to_use,
+        model="meta-llama/Llama-3.3-70B-Instruct",
         max_tokens=max_tokens,
         stream=True,
         temperature=temperature,
@@ -72,18 +119,13 @@ def respond(
         messages=messages,
     ):
         token_text = message_chunk.choices[0].delta.content
-        print(f"Received token: {token_text}")
         response += token_text
         yield response
 
     print("Completed response generation.")
 
-# GRADIO UI
-
-chatbot = gr.Chatbot(height=600, show_copy_button=True, placeholder="Select a model and begin chatting", likeable=True, layout="panel")
-print("Chatbot interface created.")
-
-system_message_box = gr.Textbox(value="", placeholder="You are a helpful assistant.", label="System Prompt")
+# Gradio UI
+chatbot = gr.Chatbot(height=600, show_copy_button=True, placeholder="Ask about crypto trading or analysis.", likeable=True)
 
 max_tokens_slider = gr.Slider(
     minimum=1,
@@ -121,99 +163,18 @@ seed_slider = gr.Slider(
     label="Seed (-1 for random)"
 )
 
-# The custom_model_box is what the respond function sees as "custom_model"
-custom_model_box = gr.Textbox(
-    value="",
-    label="Custom Model",
-    info="(Optional) Provide a custom Hugging Face model path. Overrides any selected featured model.",
-    placeholder="meta-llama/Llama-3.3-70B-Instruct"
-)
-
-def set_custom_model_from_radio(selected):
-    """
-    This function will get triggered whenever someone picks a model from the 'Featured Models' radio.
-    We will update the Custom Model text box with that selection automatically.
-    """
-    print(f"Featured model selected: {selected}")
-    return selected
-
 demo = gr.ChatInterface(
     fn=respond,
     additional_inputs=[
-        system_message_box,
         max_tokens_slider,
         temperature_slider,
         top_p_slider,
         frequency_penalty_slider,
         seed_slider,
-        custom_model_box,
     ],
     fill_height=True,
     chatbot=chatbot,
-    theme="Nymbo/Nymbo_Theme",
 )
-print("ChatInterface object created.")
-
-with demo:
-    with gr.Accordion("Model Selection", open=False):
-        model_search_box = gr.Textbox(
-            label="Filter Models",
-            placeholder="Search for a featured model...",
-            lines=1
-        )
-        print("Model search box created.")
-
-        models_list = [
-            "meta-llama/Llama-3.3-70B-Instruct",
-            "meta-llama/Llama-3.2-3B-Instruct",
-            "meta-llama/Llama-3.2-1B-Instruct",
-            "meta-llama/Llama-3.1-8B-Instruct",
-            "NousResearch/Hermes-3-Llama-3.1-8B",
-            "google/gemma-2-27b-it",
-            "google/gemma-2-9b-it",
-            "google/gemma-2-2b-it",
-            "mistralai/Mistral-Nemo-Instruct-2407",
-            "mistralai/Mixtral-8x7B-Instruct-v0.1",
-            "mistralai/Mistral-7B-Instruct-v0.3",
-            "Qwen/Qwen2.5-72B-Instruct",
-            "Qwen/QwQ-32B-Preview",
-            "PowerInfer/SmallThinker-3B-Preview",
-            "HuggingFaceTB/SmolLM2-1.7B-Instruct",
-            "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
-            "microsoft/Phi-3.5-mini-instruct",
-        ]
-        print("Models list initialized.")
-
-        featured_model_radio = gr.Radio(
-            label="Select a model below",
-            choices=models_list,
-            value="meta-llama/Llama-3.3-70B-Instruct",
-            interactive=True
-        )
-        print("Featured models radio button created.")
-
-        def filter_models(search_term):
-            print(f"Filtering models with search term: {search_term}")
-            filtered = [m for m in models_list if search_term.lower() in m.lower()]
-            print(f"Filtered models: {filtered}")
-            return gr.update(choices=filtered)
-
-        model_search_box.change(
-            fn=filter_models,
-            inputs=model_search_box,
-            outputs=featured_model_radio
-        )
-        print("Model search box change event linked.")
-
-        featured_model_radio.change(
-            fn=set_custom_model_from_radio,
-            inputs=featured_model_radio,
-            outputs=custom_model_box
-        )
-        print("Featured model radio button change event linked.")
-
-print("Gradio interface initialized.")
 
 if __name__ == "__main__":
-    print("Launching the demo application.")
     demo.launch()