Update app.py

app.py

@@ -1,6 +1,6 @@
-import streamlit as st
-from datetime import datetime, timedelta
+import os
 import yfinance as yf
+from datetime import datetime, timedelta
 import numpy as np
 import pandas as pd
 from sklearn.decomposition import PCA
@@ -9,6 +9,25 @@ from sklearn.ensemble import RandomForestClassifier
 import shap
 import matplotlib.pyplot as plt
 
+class MP3Playlist:
+    """Handles MP3 playlist functionality by listing and displaying .mp3 files in the app directory."""
+    def __init__(self, root_dir="."):
+        self.root_dir = root_dir
+        self.mp3_files = []
+
+    def scan_for_mp3(self):
+        """Scans the root directory for .mp3 files."""
+        self.mp3_files = [file for file in os.listdir(self.root_dir) if file.endswith(".mp3")]
+
+    def display_playlist(self):
+        """Displays the MP3 playlist."""
+        if not self.mp3_files:
+            print("No MP3 files found in the root directory.")
+        else:
+            print("\n=== MP3 Playlist ===")
+            for i, file in enumerate(self.mp3_files, start=1):
+                print(f"{i}. {file}")
+
 class DataFetcher:
     """Fetches historical financial data using yfinance."""
     def __init__(self, ticker, nb_days):
@@ -21,8 +40,8 @@ class DataFetcher:
         end_date = datetime.now().replace(hour=0, minute=0, second=0, microsecond=0)
         start_date = end_date - timedelta(days=self.nb_days)
         end_date = end_date + timedelta(days=1)
+        print(f"Fetching data from {start_date} to {end_date}")
         self.data = yf.download(self.ticker, start=start_date, end=end_date, interval="1h")
-        return self.data
 
 class FinancialDataProcessor:
     """Processes financial data to calculate returns, scenarios, and probabilities."""
@@ -111,9 +130,9 @@ class StrategyBuilder:
     def analyze_feature_importance(self, shap_values, feature_names):
         """Analyzes feature importance based on SHAP values."""
         if isinstance(shap_values, list):
-            shap_values = shap_values[1]
+            shap_values = shap_values[1]  # Use second class for binary classification
 
-        if len(shap_values.shape) == 3:
+        if len(shap_values.shape) == 3:  # Handle multi-output cases
             shap_values = shap_values[:, :, 1]
 
         mean_abs_shap = np.mean(np.abs(shap_values), axis=0)
@@ -126,6 +145,8 @@ class StrategyBuilder:
             'Mean_Abs_SHAP': mean_abs_shap
         }).sort_values(by='Mean_Abs_SHAP', ascending=False)
 
+        print("Feature Importance based on Shapley Values:")
+        print(feature_importance)
         return feature_importance
 
     def refine_thresholds(self, feature_importance, buy_threshold=0.5, sell_threshold=0.5):
@@ -170,13 +191,18 @@ class Backtester:
 
         return portfolio_values, position, entry_price
 
-def run_analysis():
-    """Runs the complete trading analysis."""
+def main():
     try:
-        fetcher = DataFetcher(ticker="BTC-USD", nb_days=50)
-        data = fetcher.fetch_data()
+        # MP3 Playlist Feature
+        playlist = MP3Playlist()
+        playlist.scan_for_mp3()
+        playlist.display_playlist()
 
-        processor = FinancialDataProcessor(data)
+        # Financial Data Processing and Strategy
+        fetcher = DataFetcher(ticker="BTC-USD", nb_days=200)
+        fetcher.fetch_data()
+
+        processor = FinancialDataProcessor(fetcher.data)
         processed_data = processor.calculate_returns().calculate_probabilities()
 
         pca_transformer = PCATransformer(n_components=1)
@@ -194,223 +220,30 @@ def run_analysis():
         portfolio_values, final_position, entry_price = backtester.backtest(buy_threshold, sell_threshold)
 
         last_row = processed_data.iloc[-1]
-
-        # Display results
-        col1, col2 = st.columns(2)
-        
-        with col1:
-            st.subheader("Current Position")
-            st.metric("Portfolio Value", f"${portfolio_values[-1]:.2f}")
-            st.metric("Position", final_position or "No position")
-            if final_position:
-                st.metric("Entry Price", f"${entry_price:.2f}")
-            st.metric("Latest Close", f"${last_row['Close']:.2f}")
-
-        with col2:
-            st.subheader("Decision Metrics")
-            st.metric("Buy%", f"{last_row['Buy%']:.4f}")
-            st.metric("Sell%", f"{last_row['Sell%']:.4f}")
-            st.metric("Optimal Buy%", f"{last_row['PCA_Buy%']:.4f}")
-            st.metric("Optimal Sell%", f"{last_row['PCA_Sell%']:.4f}")
-
-        return True
+        print("\n=== Final Position and Market Values ===")
+        print(f"Final Portfolio Value: ${portfolio_values[-1]:.2f}")
+        print(f"Current Position: {final_position or 'No position'}")
+        if final_position:
+            print(f"Entry Price: ${entry_price:.2f}")
+        print(f"Latest Close Price: ${last_row['Close']:.2f}")
+        print("\nDecision Metrics:")
+        print(f"Buy%: {last_row['Buy%']:.4f} | Sell%: {last_row['Sell%']:.4f}")
+        print(f"PCA_Buy%: {last_row['PCA_Buy%']:.4f} | PCA_Sell%: {last_row['PCA_Sell%']:.4f}")
+        print(f"Applied Thresholds - Buy: {buy_threshold:.2f}, Sell: {sell_threshold:.2f}")
+
+        plt.figure(figsize=(12, 6))
+        plt.plot(processed_data.index[1:], portfolio_values, label='Portfolio Value', color='blue')
+        plt.title('Portfolio Value Over Time (Backtest)')
+        plt.xlabel('Date')
+        plt.ylabel('Portfolio Value ($)')
+        plt.legend()
+        plt.grid()
+        plt.show()
 
     except Exception as e:
-        st.error(f"An error occurred: {str(e)}")
-        return False
-
-# Page configuration
-st.set_page_config(page_title="BTC-USD Trading Bot", layout="wide")
-st.title("BTC-USD Trading Bot")
-
-# Run analysis automatically on page load
-run_analysis()
-
-# Auto-refresh every 5 minutes
-st.empty()
-st.markdown("*Data updates automatically every 5 minutes*")
-if st.button("Refresh Now"):
-    run_analysis()
-
-# Educational Slides Section
-st.header("Educational Slides")
-
-with st.expander("Slide 1: Introduction", expanded=False):
-    st.subheader("Real-Time Crypto Trading Bot Using Machine Learning and PCA")
-    st.markdown("""
-    **Subtitle:** Leveraging Financial Data Analysis for Optimal Trading Decisions
-    
-    **Overview:**  
-    This system integrates real-time financial data, machine learning, and principal component analysis (PCA) 
-    to automate trading decisions in cryptocurrency markets.
-    """)
-
-with st.expander("Slide 2: Objective", expanded=False):
-    st.subheader("Main Goals")
-    st.markdown("""
-    - Develop an automated trading system that optimizes buy and sell decisions based on historical financial data
-    - Use machine learning models, specifically Random Forest, to predict market movements
-    - Implement Principal Component Analysis (PCA) for dimensionality reduction and feature extraction
-    - Backtest the system and evaluate portfolio performance
-    """)
-
-with st.expander("Slide 3: Key Concepts", expanded=False):
-    st.subheader("Core Technologies and Methods")
-    st.markdown("""
-    - **Logarithmic Return:** A continuous-time return calculation used to model price changes in markets
-    - **Principal Component Analysis (PCA):** A dimensionality reduction technique to extract meaningful features
-    - **Machine Learning:** Using Random Forest to classify "Buy" and "Sell" scenarios
-    - **SHAP Values:** A method to interpret model outputs through feature contribution analysis
-    """)
-
-with st.expander("Slide 4: Data Collection and Preprocessing", expanded=False):
-    col1, col2 = st.columns(2)
-    with col1:
-        st.markdown("### Data Fetching")
-        st.markdown("""
-        - **Source:** Yahoo Finance (yfinance)
-        - **Data:** Historical cryptocurrency data
-        - **Interval:** Hourly data points
-        - **Period:** Last 50 days
-        """)
-    with col2:
-        st.markdown("### Preprocessing")
-        st.markdown("""
-        - Calculate logarithmic returns
-        - Classify scenarios (Buy/Sell)
-        - Adjust returns for sell scenarios
-        - Handle missing values and outliers
-        """)
-
-with st.expander("Slide 5: Principal Component Analysis (PCA)", expanded=False):
-    st.subheader("PCA Process")
-    st.markdown("""
-    1. **Data Standardization:**
-       ```
-       X' = (X - μ) / σ
-       ```
-    
-    2. **Covariance Matrix:**
-       ```
-       Cov(X) = 1/(n-1) Σ(Xi - μ)(Xi - μ)ᵀ
-       ```
-    
-    3. **Eigenvalue Decomposition:**
-       - Find principal components
-       - Sort by variance explained
-    
-    4. **Dimensionality Reduction:**
-       - Transform data to lower dimensions
-       - Preserve important features
-    """)
-
-with st.expander("Slide 6: Machine Learning Strategy", expanded=False):
-    col1, col2 = st.columns(2)
-    with col1:
-        st.markdown("### Model Architecture")
-        st.markdown("""
-        - Random Forest Classifier
-        - Feature selection from PCA
-        - Binary classification (Buy/Sell)
-        """)
-    with col2:
-        st.markdown("### Training Process")
-        st.markdown("""
-        - Cross-validation
-        - Hyperparameter tuning
-        - Performance metrics
-        - Model evaluation
-        """)
-
-with st.expander("Slide 7: SHAP Analysis", expanded=False):
-    st.subheader("Shapley Additive Explanations")
-    st.markdown("""
-    **SHAP Value Formula:**
-    ```
-    φᵢ(f) = 1/|N|! Σ [f(S ∪ {i}) - f(S)]
-    ```
-    
-    **Key Components:**
-    - Feature importance ranking
-    - Individual prediction explanations
-    - Global model interpretation
-    """)
-
-with st.expander("Slide 8: Strategy Refinement", expanded=False):
-    st.subheader("Dynamic Strategy Adjustment")
-    st.markdown("""
-    1. **Threshold Refinement:**
-       - Use SHAP values to identify key features
-       - Adjust thresholds based on importance
-    
-    2. **Signal Processing:**
-       - Buy signal strengthening
-       - Sell signal validation
-       - Risk management integration
-    """)
-
-with st.expander("Slide 9: Backtesting Framework", expanded=False):
-    st.markdown("""
-    ### Portfolio Value Calculation
-    ```
-    Portfolio Valueₜ₊₁ = Portfolio Valueₜ × (Pₜ₊₁/Pₜ)
-    ```
-    
-    ### Trading Logic
-    - Buy when probability > threshold
-    - Sell when probability > threshold
-    - Position management
-    
-    ### Performance Metrics
-    - Total Return
-    - Risk Metrics
-    - Sharpe Ratio
-    """)
-
-with st.expander("Slide 10: Real-Time Interface", expanded=False):
-    st.subheader("Streamlit Dashboard Features")
-    st.markdown("""
-    - Live trading signals
-    - Portfolio performance tracking
-    - Position monitoring
-    - Automatic updates
-    - Historical performance
-    """)
-
-with st.expander("Slide 11: Summary", expanded=False):
-    st.markdown("""
-    ### Key System Components
-    - Automated trading system
-    - ML & PCA integration
-    - SHAP-based interpretation
-    - Real-time analytics
-    - Performance tracking
-    """)
-
-with st.expander("Slide 12: Future Improvements", expanded=False):
-    st.markdown("""
-    ### Planned Enhancements
-    1. Additional data sources integration
-    2. Advanced optimization techniques
-    3. Live trading deployment
-    4. Enhanced risk management
-    5. Portfolio diversification
-    """)
-
-with st.expander("Slide 13: Q&A", expanded=False):
-    st.markdown("""
-    ### Questions & Discussion
-    
-    Thank you for exploring our trading system! For questions or suggestions:
-    - System architecture
-    - Implementation details
-    - Performance metrics
-    - Future developments
-    """)
-
-# Footer
-st.markdown("---")
-st.markdown("*Support Development & Info : ")
-st.markdown("Send your email in comment to your btc donation at ")
-st.markdown("1P9R71C6JYJxrPVEzMz4K3hoHYGRW39A9A")
-# Educational Slides Section dona
+        print(f"An error occurred: {e}")
+        import traceback
+        traceback.print_exc()
+
+if __name__ == "__main__":
+    main()