@echo off REM ============================================================ REM PoolCoach - Thi nghiem 12/07/2026 (double-click de chay) REM REM Run 1: PPO 300k, scratch penalty -0.5 (dang -0.3) REM -> kiem chung: scratch giam ma pot giu duoc khong? REM Run 2: SAC 300k, reward giu nguyen (-0.3) REM -> so sanh off-policy vs PPO tren cung reward REM REM Uoc luong: Run 1 ~5 phut; Run 2 (SAC, CPU) co the 30-60 phut. REM Ket qua: logs\\learning_curve.png + eval in ra cuoi moi run. REM ============================================================ if "%~1"=="_go" goto :inner cmd /k "%~f0" _go exit /b :inner set "VENV=D:\Khoa luan\poolcoach-env\Scripts\activate.bat" set "REPO=D:\Khoa luan\poolcoach-rl" if not exist "%VENV%" ( echo [loi] Khong tim thay venv: "%VENV%" goto :eof ) call "%VENV%" cd /d "%REPO%" echo. echo ===== Run 1/2: PPO 300k, scratch -0.5 ===== python scripts\train_sink.py --scratch-penalty 0.5 --run-name ppo_scr05_20260712 echo. echo ===== Run 2/2: SAC 300k, scratch -0.3 (mac dinh) ===== python scripts\train_sink.py --algo sac --run-name sac_scr03_20260712 echo. echo ============================================================ echo XONG CA 2 RUN. Xem: echo logs\ppo_scr05_20260712\learning_curve.png echo logs\sac_scr03_20260712\learning_curve.png echo ============================================================