fix: resolve import issues for GitHub Actions tests

- Convert all relative imports to use dot notation (from .module import)
- Add __init__.py to make src a proper Python package
- Update GitHub Actions workflow to run tests from src directory
- Fix ModuleNotFoundError issues in CI environment
- Ensure all modules can be imported correctly

Author: IsmailMoudden

.github/workflows/test.yml

@@ -36,20 +36,22 @@ jobs:
     
     - name: Test optimization modules
       run: |
-        python -c "from src.markowitz import markowitz_weights; print('✅ Markowitz module import successful!')"
-        python -c "from src.risk_parity import risk_parity_weights; print('✅ Risk parity module import successful!')"
-        python -c "from src.monte_carlo import monte_carlo_simulation; print('✅ Monte Carlo module import successful!')"
-        python -c "from src.black_litterman import black_litterman_weights; print('✅ Black-Litterman module import successful!')"
-        python -c "from src.ml_predictor import ml_predictor; print('✅ ML predictor module import successful!')"
-        python -c "from src.hybrid_model import hybrid_weights; print('✅ Hybrid model module import successful!')"
-        python -c "from src.custom_metrics_opt import custom_metrics_weights; print('✅ Custom metrics module import successful!')"
-        python -c "from src.walkforward_backtest import walkforward_backtest; print('✅ Walkforward backtest module import successful!')"
+        cd src
+        python -c "from markowitz import markowitz_weights; print('✅ Markowitz module import successful!')"
+        python -c "from risk_parity import risk_parity_weights; print('✅ Risk parity module import successful!')"
+        python -c "from monte_carlo import monte_carlo_simulation; print('✅ Monte Carlo module import successful!')"
+        python -c "from black_litterman import black_litterman_weights; print('✅ Black-Litterman module import successful!')"
+        python -c "from ml_predictor import ml_predictor; print('✅ ML predictor module import successful!')"
+        python -c "from hybrid_model import hybrid_weights; print('✅ Hybrid model module import successful!')"
+        python -c "from custom_metrics_opt import custom_metrics_weights; print('✅ Custom metrics module import successful!')"
+        python -c "from walkforward_backtest import walkforward_backtest; print('✅ Walkforward backtest module import successful!')"
     
     - name: Test basic functionality
       run: |
+        cd src
         python -c "
         import numpy as np
-        from src.risk_parity import risk_parity_weights
+        from risk_parity import risk_parity_weights
         
         # Test basic functionality
         cov_matrix = np.array([[0.01, 0.005], [0.005, 0.01]])

src/__init__.py

@@ -0,0 +1,40 @@
+"""
+SiraEdge Portfolio Optimization Module
+
+This package contains implementations of various portfolio optimization models:
+- Markowitz (Mean-Variance) optimization
+- Risk Parity optimization
+- Monte Carlo simulation
+- Black-Litterman model
+- Machine Learning predictor
+- Hybrid model
+- Custom metrics optimization
+- Walk-forward backtesting
+
+All modules are designed to work together for comprehensive portfolio analysis.
+"""
+
+__version__ = "1.0.0"
+__author__ = "Ismail Moudden"
+__email__ = "[email protected]"
+
+# Import main functions for easy access
+from .markowitz import markowitz_weights
+from .risk_parity import risk_parity_weights
+from .monte_carlo import monte_carlo_simulation
+from .black_litterman import black_litterman_weights
+from .ml_predictor import ml_predictor
+from .hybrid_model import hybrid_weights
+from .custom_metrics_opt import custom_metrics_weights
+from .walkforward_backtest import walkforward_backtest
+
+__all__ = [
+    "markowitz_weights",
+    "risk_parity_weights", 
+    "monte_carlo_simulation",
+    "black_litterman_weights",
+    "ml_predictor",
+    "hybrid_weights",
+    "custom_metrics_weights",
+    "walkforward_backtest"
+]

src/black_litterman.py

@@ -2,7 +2,7 @@
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
-from data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def run_black_litterman(tickers=None, start="2020-01-01", end="2023-12-31") -> pd.Series:

src/custom_metrics_opt.py

@@ -3,7 +3,7 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 from scipy.optimize import minimize
-from data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def sharpe_ratio(returns: pd.Series, rf: float = 0.0) -> float:

src/hybrid_model.py

@@ -2,8 +2,8 @@
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
-from data_utils import download_prices, compute_indicators, ensure_figures_dir, DEFAULT_TICKERS
-from risk_parity import risk_parity_weights
+from .data_utils import download_prices, compute_indicators, ensure_figures_dir, DEFAULT_TICKERS
+from .risk_parity import risk_parity_weights
 from sklearn.linear_model import Ridge
 from sklearn.preprocessing import StandardScaler
 from sklearn.pipeline import Pipeline

src/markowitz.py

@@ -2,7 +2,7 @@
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
-from data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def random_portfolios(mu: np.ndarray, cov: np.ndarray, n: int = 20000, seed: int = 42):

src/ml_predictor.py

@@ -7,7 +7,7 @@
 from sklearn.preprocessing import StandardScaler
 from sklearn.pipeline import Pipeline
 from sklearn.model_selection import TimeSeriesSplit
-from data_utils import download_prices, compute_indicators, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, compute_indicators, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def prepare_ml_dataset(start="2020-01-01", end="2023-12-31", tickers=None) -> Tuple[pd.DataFrame, pd.Series, pd.Series]:

src/ml_training_report.py

@@ -10,7 +10,7 @@
 from sklearn.pipeline import Pipeline
 from sklearn.metrics import r2_score, mean_absolute_error
 
-from data_utils import (
+from .data_utils import (
     download_prices,
     compute_indicators,
     align_features_and_targets,

src/monte_carlo.py

@@ -2,7 +2,7 @@
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
-from data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def simulate_portfolios(returns: pd.DataFrame, n_portfolios: int = 10000, seed: int = 42):

src/risk_parity.py

@@ -3,7 +3,7 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 from scipy.optimize import minimize
-from data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
+from .data_utils import download_prices, ensure_figures_dir, DEFAULT_TICKERS
 
 
 def risk_contribution(weights: np.ndarray, cov: np.ndarray) -> np.ndarray: