Update app.py

app.py

@@ -7,8 +7,8 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 import torch
 import yfinance as yf  # For dynamic data fetching
 
-# Hardcoded common symbols for fuzzy matching (expand as needed)
-available_symbols = ['TSLA', 'MSFT', 'NVDA', 'GOOG', 'AMZN', 'SPY']
+# Expanded list of common symbols for fuzzy matching (added more tickers for better coverage)
+available_symbols = ['TSLA', 'MSFT', 'NVDA', 'GOOG', 'AMZN', 'SPY', 'AAPL', 'META', 'NFLX', 'INTC', 'AMD', 'IBM', 'ORCL', 'CSCO', 'JPM', 'BAC', 'WFC', 'V', 'MA', 'XOM', 'CVX', 'PFE', 'JNJ', 'MRK']
 
 # Financial calculation functions
 def parse_period(query):
@@ -70,31 +70,39 @@ model = AutoModelForCausalLM.from_pretrained(
 def generate_response(user_query, enable_thinking=False):
     # Parse query for symbol and period
     symbol_match = re.search(r'\b([A-Z]{1,5})\b', user_query.upper())
-    symbol = find_closest_symbol(symbol_match.group(1)) if symbol_match else 'SPY'  # Default to SPY
-    period = parse_period(user_query)
-    end_date = datetime.now()
-    start_date = end_date - period
-
-    # Calculate growth rate if applicable
-    growth_rate = calculate_growth_rate(start_date, end_date, symbol)
-    if growth_rate is not None:
-        summary = f"The CAGR for {symbol} over the period is {growth_rate:.2f}%."
-    else:
-        summary = f"No data available for {symbol} in the specified period."
-
-    # Handle investment projection
-    investment_match = re.search(r'\$(\d+)', user_query)
-    if investment_match:
-        principal = float(investment_match.group(1))
-        years = period.days / 365.25
-        projected = calculate_investment(principal, years)
-        summary += f" Projecting ${principal} at 7% return over {years:.1f} years: ${projected:.2f}."
-
-    # Prepare prompt for model
-    system_prompt = "You are a knowledgeable financial advisor. Provide accurate, concise advice based on the data provided. Disclaimer: This is not professional advice."
+    symbol = find_closest_symbol(symbol_match.group(1)) if symbol_match else None  # No default; skip if no match
+    summary = ""  # Initialize empty to handle general queries
+
+    if symbol:  # Only perform calculations if a valid symbol is detected
+        period = parse_period(user_query)
+        end_date = datetime.now()
+        start_date = end_date - period
+
+        # Calculate growth rate if applicable
+        growth_rate = calculate_growth_rate(start_date, end_date, symbol)
+        if growth_rate is not None:
+            summary = f"The CAGR for {symbol} over the period is {growth_rate:.2f}%."
+        else:
+            summary = f"No data available for {symbol} in the specified period."
+
+        # Handle investment projection
+        investment_match = re.search(r'\$(\d+)', user_query)
+        if investment_match:
+            principal = float(investment_match.group(1))
+            years = period.days / 365.25
+            projected = calculate_investment(principal, years)
+            summary += f" Projecting ${principal} at 7% return over {years:.1f} years: ${projected:.2f}."
+
+    # Prepare prompt for model with enhanced instructions for natural chatbot behavior
+    system_prompt = (
+        "You are FinChat, a knowledgeable financial advisor. Always respond in a friendly, professional manner like a helpful chatbot. "
+        "For greetings such as 'Hi' or 'Hello', reply warmly, e.g., 'Hi! I'm FinChat, your financial advisor. What can I help you with today regarding stocks, investments, or advice?' "
+        "Provide accurate, concise advice based on any provided data. If no specific data is available, offer general financial insights or ask for clarification. "
+        "Disclaimer: This is not professional financial advice; consult experts for decisions."
+    )
     messages = [
         {"role": "system", "content": system_prompt},
-        {"role": "user", "content": f"{summary} {user_query}"}
+        {"role": "user", "content": f"{summary} {user_query}" if summary else user_query}
     ]
 
     text = tokenizer.apply_chat_template(