initial commit

app.py

@@ -6,38 +6,32 @@ import yfinance as yf
 from sklearn.metrics import mean_absolute_error, mean_squared_error
 from prophet.plot import plot_plotly, plot_components_plotly
 
-# List of ticker symbols
 ticker_symbols = st.secrets["TICKER_SYMBOLS"].split(",")
 
-# Function to fetch stock data from Yahoo Finance
 def fetch_stock_data(ticker_symbol, start_date, end_date):
-    ticker_symbol = ticker_symbol +".NS"
+    ticker_symbol = ticker_symbol +st.secrets["TICKER_FLAG"]
     stock_data = yf.download(ticker_symbol, start=start_date, end=end_date)
     df = stock_data[['Adj Close']].reset_index()
     df = df.rename(columns={'Date': 'ds', 'Adj Close': 'y'})
     # df.to_csv(f"{ticker_symbol}.csv")
     return df
 
-# Function to train the Prophet model
 def train_prophet_model(df):
     model = Prophet()
     model.fit(df)
     return model
 
-# Function to make the forecast
 def make_forecast(model, periods):
     future = model.make_future_dataframe(periods=periods)
     forecast = model.predict(future)
     return forecast
 
-# Function to calculate performance metrics
 def calculate_performance_metrics(actual, predicted):
     mae = mean_absolute_error(actual, predicted)
     mse = mean_squared_error(actual, predicted)
     rmse = np.sqrt(mse)
     return {'MAE': mae, 'MSE': mse, 'RMSE': rmse}
 
-# Function to determine sentiment
 def determine_sentiment(actual, predicted):
     if actual > predicted:
         sentiment = 'Negative'
@@ -48,19 +42,15 @@ def determine_sentiment(actual, predicted):
     return sentiment
 
 
-# Streamlit app
 def main():
     st.title('Stock Prediction on NSE Stocks')
 
-    # Set up the layout
     st.sidebar.header('User Input Parameters')
     ticker_symbol = st.sidebar.selectbox('Enter Ticker Symbol', options=ticker_symbols, index=0)
 
-    # Dropdown for training period selection
     training_period = st.sidebar.selectbox('Select Training Period', 
                                             options=['1 week', '1 month', '1 year', '10 years'])
 
-    # Calculate start date and end date based on training period
     if training_period == '1 week':
         start_date = pd.to_datetime('today') - pd.DateOffset(weeks=1)
     elif training_period == '1 month':
@@ -72,15 +62,12 @@ def main():
 
     end_date = pd.to_datetime('today')
 
-    # Fetching the data for the selected training period
     df = fetch_stock_data(ticker_symbol, start_date, end_date)
 
-    # Dropdown for forecast horizon selection
     forecast_horizon = st.sidebar.selectbox('Forecast Horizon', 
                                             options=['Next day', 'Next week', 'Next month'],
                                             format_func=lambda x: x.capitalize())
     
-    # Convert the selected horizon to days
     horizon_mapping = {'Next day': 1, 'Next week': 7, 'Next month': 30}
     forecast_days = horizon_mapping[forecast_horizon]
 
@@ -97,8 +84,6 @@ def main():
         """)
         st.subheader(f'Forecast Summary for {ticker_symbol}')
         latest_forecast = forecast_reversed.iloc[0]
-
-        # Last Stock Price details with sentiment indicator
         actual_last_price = df["y"].iloc[-1]
         predicted_last_price = latest_forecast['yhat']
         sentiment = determine_sentiment(actual_last_price, predicted_last_price)
@@ -111,7 +96,6 @@ def main():
         else:
             st.info(f'Overall predication indicates neutral sentiment.')        
 
-        # Prediction details
         st.markdown(f"""
             **Prediction for {forecast_horizon.lower()}:**
             
@@ -128,8 +112,6 @@ def main():
         st.write(forecast_reversed)
 
         
-        # Calculate performance metrics
-        # Function to determine if performance metrics are in a good range
         def evaluate_performance_metrics(metrics):
             evaluation = {}
             evaluation['MAE'] = 'Good' if metrics['MAE'] < 0.05 * (df['y'].max() - df['y'].min()) else 'Not Good'
@@ -137,21 +119,17 @@ def main():
             evaluation['RMSE'] = 'Good' if metrics['RMSE'] < 0.1 * (df['y'].max() - df['y'].min()) else 'Not Good'
             return evaluation
 
-        # Calculate performance metrics
         actual = df['y']
         predicted = forecast['yhat'][:len(df)]
         metrics = calculate_performance_metrics(actual, predicted)
 
-        # Evaluate performance metrics
         evaluation = evaluate_performance_metrics(metrics)
 
         metrics = calculate_performance_metrics(actual, predicted)
         MAE =metrics['MAE']
         MSE = metrics['MSE']
         RMSE = metrics['RMSE']        
-             
 
-        # Display evaluation
         st.subheader('Performance Evaluation')
         st.write('The metrics below provide a quantitative measure of the model’s accuracy:')
         maecolor = "green" if evaluation["MAE"] == "Good" else "red"
@@ -168,7 +146,6 @@ def main():
         st.markdown("(The square root of MSE, which is more interpretable in the same units as the target variable.)")
         
         
-        
 
 
 # Run the main function