Reinforcement-Learning-for-Smart-HVAC-Control-Using-CityLearn/install.bat at main · HanSun103/Reinforcement-Learning-for-Smart-HVAC-Control-Using-CityLearn · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
@echo off
REM ==========================================================================
REM install.bat — CityLearn HVAC RL Project (Windows, Anaconda)
REM ==========================================================================
REM
REM IMPORTANT: Run from an ANACONDA PROMPT (not regular PowerShell).
REM   Start → search "Anaconda Prompt" → open it
REM   Then: cd to the project folder and type:  install.bat
REM
REM This creates a FRESH conda environment called "citylearn-rl".
REM Using a dedicated env avoids DLL conflicts with the base environment.
REM ==========================================================================

setlocal enabledelayedexpansion

echo.
echo ============================================================
echo  CityLearn HVAC RL — Setup (Anaconda Prompt required)
echo ============================================================
echo.

REM --------------------------------------------------------------------------
REM Step 1: Create a fresh conda environment
REM --------------------------------------------------------------------------
REM A dedicated env isolates all packages and avoids DLL/TBB/MKL conflicts
REM that affect pip-installed PyTorch in the base Anaconda environment.
REM --------------------------------------------------------------------------
echo [1/6] Creating conda environment "citylearn-rl" (Python 3.11)...
call conda create -n citylearn-rl python=3.11 -y
if %ERRORLEVEL% NEQ 0 (
    echo ERROR: Could not create conda environment. Is Anaconda Prompt open?
    pause & exit /b 1
)

REM --------------------------------------------------------------------------
REM Step 2: Activate the environment
REM --------------------------------------------------------------------------
echo.
echo [2/6] Activating conda environment "citylearn-rl"...
call conda activate citylearn-rl
if %ERRORLEVEL% NEQ 0 (
    echo ERROR: Could not activate the environment.
    pause & exit /b 1
)

REM --------------------------------------------------------------------------
REM Step 3: PyTorch (CPU-only) via conda
REM --------------------------------------------------------------------------
REM conda-installed PyTorch bundles the correct MKL/TBB/DLL dependencies.
REM pip-installed PyTorch frequently causes "DLL init failed" in conda envs.
REM --------------------------------------------------------------------------
echo.
echo [3/6] Installing PyTorch 2.3.1 (CPU-only) via conda...
echo       This may take 5-10 minutes. Do not close this window.
call conda install pytorch=2.3.1 cpuonly -c pytorch -y
if %ERRORLEVEL% NEQ 0 (
    echo ERROR: PyTorch install failed. Try running manually:
    echo   conda install pytorch=2.3.1 cpuonly -c pytorch -y
    pause & exit /b 1
)

REM --------------------------------------------------------------------------
REM Step 4: CityLearn (without openstudio)
REM --------------------------------------------------------------------------
REM CityLearn 2.5.0 declares openstudio as a dependency, but no PyPI wheel
REM exists for Python 3.11 on Windows. We install with --no-deps and add
REM only the packages we actually need.
REM --------------------------------------------------------------------------
echo.
echo [4/6] Installing CityLearn 2.5.0 (skipping openstudio)...
pip install citylearn==2.5.0 --no-deps
if %ERRORLEVEL% NEQ 0 (
    echo ERROR: CityLearn install failed.
    pause & exit /b 1
)

REM --------------------------------------------------------------------------
REM Step 5: CityLearn minimal deps + Stable Baselines3
REM --------------------------------------------------------------------------
REM gymnasium is pinned to 0.28.1 (CityLearn's maximum supported version).
REM SB3 is pinned to 2.2.1 (last version compatible with gymnasium 0.28).
REM SB3 is installed with --no-deps so it cannot auto-upgrade gymnasium.
REM platformdirs is a direct CityLearn runtime dep (missed by --no-deps).
REM --------------------------------------------------------------------------
echo.
echo [5/6] Installing gymnasium, SB3, and CityLearn dependencies...
pip install "gymnasium==0.28.1" "scikit-learn==1.2.2" simplejson pyyaml platformdirs
pip install "stable-baselines3==2.2.1" --no-deps
pip install cloudpickle
if %ERRORLEVEL% NEQ 0 (
    echo ERROR: Package install failed.
    pause & exit /b 1
)

REM --------------------------------------------------------------------------
REM Step 6: Utilities
REM --------------------------------------------------------------------------
REM numpy is capped at <2.0.0 — CityLearn 2.5.0 is not compatible with numpy 2.x.
REM matplotlib/pandas sometimes auto-install numpy 2.x; we pin it here after.
REM --------------------------------------------------------------------------
echo.
echo [6/6] Installing matplotlib, pandas, numpy...
pip install "matplotlib>=3.7.0" "pandas>=2.0.0"
pip install "numpy>=1.24.0,<2.0.0"

REM --------------------------------------------------------------------------
REM Verify all key packages load correctly
REM --------------------------------------------------------------------------
echo.
echo ============================================================
echo  Verifying installation...
echo ============================================================
python -c "import torch;            print('  torch    :', torch.__version__)"
python -c "import gymnasium;        print('  gymnasium:', gymnasium.__version__)"
python -c "import stable_baselines3;print('  sb3      :', stable_baselines3.__version__)"
python -c "import citylearn;        print('  citylearn:', citylearn.__version__)"
python -c "import simplejson;       print('  simplejson: OK')"
python -c "import sklearn;          print('  sklearn  :', sklearn.__version__)"

echo.
echo ============================================================
echo  Environment "citylearn-rl" is ready!
echo.
echo  NEXT STEPS:
echo.
echo  1. In Cursor/VS Code: press Ctrl+Shift+P, choose
echo       "Python: Select Interpreter"
echo     and pick:  citylearn-rl (conda)
echo.
echo  2. Run the quick smoke test:
echo       conda activate citylearn-rl
echo       python src/quick_test.py
echo.
echo  3. Train the model:
echo       python src/train.py
echo ============================================================
echo.
pause