Minor Improvement in ProtoSmartBeta

README.md

@@ -40,6 +40,8 @@ where DPS denotes dividends per share. Existing holdings are sold before purchas
 
 ## Implementation
 ### Environment Setup
+**Recommendation:** We recommend using Python 3.12 for this project. While other versions may work, Python 3.12 has been tested and verified to work smoothly with all dependencies. Some dependencies (e.g., matplotlib) may have compatibility issues with Python 3.13.
+
 1. Set up python virtual environment
 ```bash
 python -m venv venv
@@ -101,11 +103,6 @@ python evaluation.py
 [TODO: change the name of optimization folder to out-of-sample-backtesting or something like that]: #
 The script will get value from `parameter/optimized_parameter.json` to execute. The results are stored in the `result/optimization` folder.
 
-## In-sample Backtesting
-Running the in-sample backtesting by execute the command:
-```bash
-python backtesting.py
-```
 ### Evaluation Metrics
 - Backtesting results are stored in the `result/backtest/` folder. 
 - Used metrics to compare with VNINDEX are: 

optimization.py

@@ -3,6 +3,7 @@
 """
 
 import logging
+import json
 import optuna
 from optuna.samplers import TPESampler
 from config.config import OPTIMIZATION_CONFIG
@@ -83,3 +84,15 @@ def objective(trial):
     study.optimize(
         objective, n_trials=OPTIMIZATION_CONFIG["no_trials"], callbacks=[optunaCallBack]
     )
+
+    # Save best parameters to optimized_parameter.json
+    best_trial = study.best_trial
+    best_params = {
+        "pe": [0, best_trial.params["peub"]],
+        "dy": [best_trial.params["dylb"], 1e6],
+    }
+    with open("parameter/optimized_parameter.json", "w", encoding="utf-8") as f:
+        json.dump(best_params, f, indent=4)
+    print(f"\nBest trial: {best_trial.number}")
+    print(f"Best Sharpe ratio: {best_trial.value}")
+    print(f"Best parameters saved to parameter/optimized_parameter.json")