Update app.py

app.py

@@ -38,7 +38,7 @@ response_cache = {
     "what is better individual stocks or etfs?": (
         "Here’s a comparison of individual stocks vs. ETFs:\n"
         "1. **Individual Stocks**: High returns possible (e.g., Apple up 80% in 2020) but riskier due to lack of diversification. Require active research.\n"
-        "2. **ETFs**: Diversify risk by tracking indices (e.g., SPY, S&P 500, ~12% avg. return 2015–2025). Lower fees and less research needed.\n"
+        "2. **ETFs**: Diversify risk by tracking indices (e.g., SPY, S&P 500, ~12% avg. return 2015–2024). Lower fees and less research needed.\n"
         "3. **Recommendation**: Beginners should start with ETFs; experienced investors may add stocks.\n"
         "Consult a financial planner."
     ),
@@ -81,21 +81,21 @@ response_cache = {
         "Consult a financial planner."
     ),
     "do you have a list of companies you recommend?": (
-        "I can’t recommend specific companies without data. Try ETFs like SPY (S&P 500, ~12% avg. return 2015–2025) or QQQ (tech). "
+        "I can’t recommend specific companies without data. Try ETFs like SPY (S&P 500, ~12% avg. return 2015–2024) or QQQ (tech). "
         "Research stocks like Apple (AAPL, ~80% return in 2020) or Johnson & Johnson on Yahoo Finance.\n"
         "Consult a financial planner."
     ),
     "how do i start investing in stocks?": (
         "Learn from Investopedia. Set goals and assess risk. Open a brokerage account (Fidelity, Vanguard) "
-        "and start with ETFs (e.g., SPY, ~10% avg. return). Consult a financial planner."
+        "and start with ETFs (e.g., SPY, ~12% avg. return 2015–2024). Consult a financial planner."
     ),
     "what's the difference between stocks and bonds?": (
-        "Stocks are company ownership with high risk and growth potential (e.g., S&P 500 ~10% avg. return). Bonds are loans to companies/governments "
+        "Stocks are company ownership with high risk and growth potential (e.g., S&P 500 ~12% avg. return 2015–2024). Bonds are loans to companies/governments "
         "with lower risk and steady interest. Diversify for balance."
     ),
     "how much should i invest?": (
         "Invest what you can afford after expenses and an emergency fund. Start with $100-$500 monthly "
-        "in ETFs like SPY (~10% avg. return). Consult a financial planner."
+        "in ETFs like SPY (~12% avg. return 2015–2024). Consult a financial planner."
     ),
     "what is dollar-cost averaging?": (
         "Dollar-cost averaging is investing a fixed amount regularly (e.g., $100 monthly) in ETFs, "
@@ -105,7 +105,7 @@ response_cache = {
         "Here are investing ideas:\n"
         "1. Open a brokerage account (e.g., Fidelity) if 18 or older.\n"
         "2. Deposit $100 or what you can afford.\n"
-        "3. Buy a researched ETF (e.g., SPY, ~10% avg. return) or index fund.\n"
+        "3. Buy a researched ETF (e.g., SPY, ~12% avg. return 2015–2024) or index fund.\n"
         "4. Check regularly and enable dividend reinvesting.\n"
         "5. Use dollar-cost averaging (e.g., monthly buys).\n"
         "Consult a financial planner."
@@ -114,7 +114,7 @@ response_cache = {
         "Here are investing tips:\n"
         "1. Educate yourself with Investopedia or books.\n"
         "2. Open a brokerage account (e.g., Vanguard).\n"
-        "3. Start small with ETFs like SPY (~10% avg. return).\n"
+        "3. Start small with ETFs like SPY (~12% avg. return 2015–2024).\n"
         "4. Invest regularly using dollar-cost averaging.\n"
         "5. Diversify to manage risk.\n"
         "Consult a financial planner."
@@ -124,7 +124,7 @@ response_cache = {
         "1. Study basics on Investopedia.\n"
         "2. Open a brokerage account (e.g., Fidelity).\n"
         "3. Deposit $100 or more after securing savings.\n"
-        "4. Buy an ETF like SPY (~10% avg. return) after research.\n"
+        "4. Buy an ETF like SPY (~12% avg. return 2015–2024) after research.\n"
         "5. Invest monthly with dollar-cost averaging.\n"
         "Consult a financial planner."
     ),
@@ -132,7 +132,7 @@ response_cache = {
         "Here’s investing advice:\n"
         "1. Learn basics from Investopedia.\n"
         "2. Open a brokerage account (e.g., Vanguard).\n"
-        "3. Start with $100 in an ETF like SPY (~10% avg. return).\n"
+        "3. Start with $100 in an ETF like SPY (~12% avg. return 2015–2024).\n"
         "4. Use dollar-cost averaging for regular investments.\n"
         "5. Monitor and diversify your portfolio.\n"
         "Consult a financial planner."
@@ -142,13 +142,13 @@ response_cache = {
         "1. Educate yourself using Investopedia.\n"
         "2. Open a brokerage account (e.g., Fidelity).\n"
         "3. Deposit an initial $100 after savings.\n"
-        "4. Buy an ETF like SPY (~10% avg. return) after research.\n"
+        "4. Buy an ETF like SPY (~12% avg. return 2015–2024) after research.\n"
         "5. Use dollar-cost averaging monthly.\n"
         "Consult a financial planner."
     ),
     "what is the s&p 500 index fund average growth rate?": (
         "The S&P 500 index fund’s average annual return is approximately 10–12% over the long term (1927–2025), including dividends, based on historical data. "
-        "For example, from 2015 to 2025, it averaged ~12% annually. Returns vary yearly due to market conditions. Consult a financial planner."
+        "For example, from 2015 to 2024, it averaged ~12.2% annually. Returns vary yearly due to market conditions. Consult a financial planner."
     ),
     "what was the s&p 500 return in 2020?": (
         "The S&P 500 returned approximately 16.3% in 2020, including dividends, driven by recovery from the COVID-19 market crash."
@@ -174,6 +174,9 @@ response_cache = {
     ),
     "what was the 5-year average annual growth rate of the s&p 500 from 2016?": (
         "The S&P 500’s average annual growth rate from 2016 to 2020 was approximately 13.6%, including dividends, driven by strong market recovery."
+    ),
+    "what is the average return rate of the s&p 500 in the past 10 years?": (
+        "The S&P 500’s average annual return rate from 2015 to 2024 was approximately 12.2%, including dividends, based on historical data."
     )
 }
 
@@ -200,267 +203,4 @@ try:
             torch_dtype=torch.float16,
             low_cpu_mem_usage=True
         ).to(device)
-    logger.info(f"Successfully loaded model: {model_name}")
-except Exception as e:
-    logger.error(f"Error loading model/tokenizer: {e}")
-    raise RuntimeError(f"Failed to load model: {str(e)}")
-
-# Parse period from user input
-def parse_period(query):
-    # Match specific year ranges (e.g., "2000 to 2008", "2011–2016")
-    match = re.search(r'(\d{4})\s*(?:to|-|–)\s*(\d{4})', query, re.IGNORECASE)
-    if match:
-        start_year, end_year = map(int, match.groups())
-        return start_year, end_year, None
-    # Match duration-based queries (e.g., "1-year from 2020", "3-year growth rate")
-    match = re.search(r'(\d+)-year.*from\s*(\d{4})', query, re.IGNORECASE)
-    if match:
-        duration, start_year = map(int, match.groups())
-        end_year = start_year + duration - 1
-        return start_year, end_year, duration
-    # Match general duration queries (e.g., "1-year growth rate")
-    match = re.search(r'(\d+)-year.*growth\s*rate', query, re.IGNORECASE)
-    if match:
-        duration = int(match.group(1))
-        max_year = df['Date'].dt.year.max() if df is not None else 2025
-        start_year = max_year - duration + 1
-        end_year = max_year
-        return start_year, end_year, duration
-    return None, None, None
-
-# Calculate average growth rate
-def calculate_growth_rate(start_year, end_year, duration=None):
-    if df is None or start_year is None or end_year is None:
-        return None, "Data not available or invalid period."
-    df_period = df[(df['Date'].dt.year >= start_year) & (df['Date'].dt.year <= end_year)]
-    if df_period.empty:
-        return None, f"No data available for {start_year} to {end_year}."
-    avg_return = df_period['Return'].mean()
-    if duration:
-        response = f"The S&P 500’s {duration}-year average annual growth rate from {start_year} to {end_year} was approximately {avg_return:.1f}%, including dividends."
-    else:
-        response = f"The S&P 500’s average annual growth rate from {start_year} to {end_year} was approximately {avg_return:.1f}%, including dividends."
-    return avg_return, response
-
-# Parse investment return query
-def parse_investment_query(query):
-    match = re.search(r'\$(\d+).*\s(\d+)\s*years?.*\bs&p\s*500', query, re.IGNORECASE)
-    if match:
-        amount = float(match.group(1))
-        years = int(match.group(2))
-        return amount, years
-    return None, None
-
-# Calculate future value
-def calculate_future_value(amount, years):
-    if df is None or amount is None or years is None:
-        return None, "Data not available or invalid input."
-    avg_annual_return = 10.0  # Historical S&P 500 average (1927–2025)
-    future_value = amount * (1 + avg_annual_return / 100) ** years
-    return future_value, (
-        f"Assuming a 10% average annual return, a ${amount:,.0f} investment in the S&P 500 would grow to approximately ${future_value:,.0f} "
-        f"in {years} years with annual compounding. This is based on the historical average return of 10–12% (1927–2025). "
-        "Future returns vary and are not guaranteed. Consult a financial planner."
-    )
-
-# Define chat function
-def chat_with_model(user_input, history=None, is_processing=False):
-    try:
-        start_time = time.time()
-        logger.info(f"Processing user input: {user_input}")
-        is_processing = True
-        logger.info("Showing loading animation")
-
-        # Normalize and check cache
-        cache_key = user_input.lower().strip()
-        cache_keys = list(response_cache.keys())
-        closest_key = cache_key if cache_key in response_cache else get_closest_cache_key(cache_key, cache_keys)
-        if closest_key:
-            logger.info(f"Cache hit for: {closest_key}")
-            response = response_cache[closest_key]
-            logger.info(f"Chatbot response: {response}")
-            history = history or []
-            history.append({"role": "user", "content": user_input})
-            history.append({"role": "assistant", "content": response})
-            end_time = time.time()
-            logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-            return response, history, False, ""
-
-        # Check for investment return query
-        amount, years = parse_investment_query(user_input)
-        if amount and years:
-            future_value, response = calculate_future_value(amount, years)
-            if future_value is not None:
-                response_cache[cache_key] = response
-                logger.info(f"Investment query: ${amount} for {years} years, added to cache")
-                logger.info(f"Chatbot response: {response}")
-                history = history or []
-                history.append({"role": "user", "content": user_input})
-                history.append({"role": "assistant", "content": response})
-                end_time = time.time()
-                logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-                return response, history, False, ""
-
-        # Check for period-specific query
-        start_year, end_year, duration = parse_period(user_input)
-        if start_year and end_year:
-            avg_return, response = calculate_growth_rate(start_year, end_year, duration)
-            if avg_return is not None:
-                response_cache[cache_key] = response
-                logger.info(f"Dynamic period query: {start_year}–{end_year}, added to cache")
-                logger.info(f"Chatbot response: {response}")
-                history = history or []
-                history.append({"role": "user", "content": user_input})
-                history.append({"role": "assistant", "content": response})
-                end_time = time.time()
-                logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-                return response, history, False, ""
-
-        # Skip model for short prompts
-        if len(user_input.strip()) <= 5:
-            logger.info("Short prompt, returning default response")
-            response = "Hello! I'm FinChat, your financial advisor. Ask about investing!"
-            logger.info(f"Chatbot response: {response}")
-            history = history or []
-            history.append({"role": "user", "content": user_input})
-            history.append({"role": "assistant", "content": response})
-            end_time = time.time()
-            logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-            return response, history, False, ""
-
-        # Construct prompt
-        full_prompt = prompt_prefix + user_input + "\nA:"
-        try:
-            inputs = tokenizer(full_prompt, return_tensors="pt", truncation=True, max_length=512).to(device)
-        except Exception as e:
-            logger.error(f"Error tokenizing input: {e}")
-            response = f"Error: Failed to process input: {str(e)}"
-            logger.info(f"Chatbot response: {response}")
-            history = history or []
-            history.append({"role": "user", "content": user_input})
-            history.append({"role": "assistant", "content": response})
-            end_time = time.time()
-            logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-            return response, history, False, ""
-
-        # Generate response
-        with torch.inference_mode():
-            logger.info("Generating response with model")
-            gen_start_time = time.time()
-            outputs = model.generate(
-                **inputs,
-                max_new_tokens=50,
-                min_length=20,
-                do_sample=False,
-                repetition_penalty=2.0,
-                pad_token_id=tokenizer.eos_token_id
-            )
-            gen_end_time = time.time()
-            logger.info(f"Generation time: {gen_end_time - gen_start_time:.2f} seconds")
-        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        response = response[len(full_prompt):].strip() if response.startswith(full_prompt) else response
-        logger.info(f"Chatbot response: {response}")
-
-        # Update cache
-        response_cache[cache_key] = response
-        logger.info("Cache miss, added to in-memory cache")
-
-        # Update history
-        history = history or []
-        history.append({"role": "user", "content": user_input})
-        history.append({"role": "assistant", "content": response})
-        torch.cuda.empty_cache()
-        end_time = time.time()
-        logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-        return response, history, False, ""
-
-    except Exception as e:
-        logger.error(f"Error generating response: {e}")
-        response = f"Error: {str(e)}"
-        logger.info(f"Chatbot response: {response}")
-        history = history or []
-        history.append({"role": "user", "content": user_input})
-        history.append({"role": "assistant", "content": response})
-        end_time = time.time()
-        logger.info(f"Response time: {end_time - start_time:.2f} seconds")
-        return response, history, False, ""
-
-# Save cache on exit
-def save_cache():
-    try:
-        with open(cache_file, 'w') as f:
-            json.dump(response_cache, f, indent=2)
-        logger.info("Saved cache to cache.json")
-    except Exception as e:
-        logger.warning(f"Failed to save cache.json: {e}")
-
-# Create Gradio interface with loading animation
-logger.info("Initializing Gradio interface")
-try:
-    with gr.Blocks(
-        title="FinChat: An LLM based on distilgpt2 model",
-        css="""
-        .loader {
-            border: 5px solid #f3f3f3;
-            border-top: 5px solid #3498db;
-            border-radius: 50%;
-            width: 30px;
-            height: 30px;
-            animation: spin 1s linear infinite;
-            margin: 10px auto;
-            display: block;
-        }
-        @keyframes spin {
-            0% { transform: rotate(0deg); }
-            100% { transform: rotate(360deg); }
-        }
-        .hidden { display: none; }
-        """
-    ) as interface:
-        gr.Markdown(
-            """
-            # FinChat: An LLM based on distilgpt2 model
-            FinChat provides financial advice using the lightweight distilgpt2 model, optimized for fast, detailed responses.
-            Ask about investing strategies, ETFs, stocks, or budgeting to get started!
-            """
-        )
-        chatbot = gr.Chatbot(type="messages")
-        msg = gr.Textbox(label="Your message")
-        submit = gr.Button("Send")
-        clear = gr.Button("Clear")
-        loading = gr.HTML('<div class="loader hidden"></div>', label="Loading")
-        is_processing = gr.State(value=False)
-
-        def submit_message(user_input, history, is_processing):
-            response, updated_history, new_processing, clear_input = chat_with_model(user_input, history, is_processing)
-            loader_html = '<div class="loader"></div>' if new_processing else '<div class="loader hidden"></div>'
-            return clear_input, updated_history, loader_html, new_processing
-
-        submit.click(
-            fn=submit_message,
-            inputs=[msg, chatbot, is_processing],
-            outputs=[msg, chatbot, loading, is_processing]
-        )
-        clear.click(
-            fn=lambda: ("", [], '<div class="loader hidden"></div>', False),
-            outputs=[msg, chatbot, loading, is_processing]
-        )
-    logger.info("Gradio interface initialized successfully")
-except Exception as e:
-    logger.error(f"Error initializing Gradio interface: {e}")
-    raise
-
-# Launch interface (conditional for Spaces)
-if __name__ == "__main__" and not os.getenv("HF_SPACE"):
-    logger.info("Launching Gradio interface locally")
-    try:
-        interface.launch(share=False, debug=True)
-    except Exception as e:
-        logger.error(f"Error launching interface: {e}")
-        raise
-    finally:
-        save_cache()
-else:
-    logger.info("Running in Hugging Face Spaces, interface defined but not launched")
-    import atexit
-    atexit.register(save_cache)
+    logger.info(f"Successfully