AFCI offline kit - Part 3: Python ML & signal-processing environment (Windows wheels)
Offline development kit for an arc-fault detection / electrical-fire prevention system (current sensor -> STM32F407 ADC+DMA -> FFT/STFT features -> quantized DNN -> trip). Target offline PC: Windows 10/11 x64, CPU-only Python training.
Contents: Python 3.12 ML / signal-processing environment (Windows x64 wheels, CPU-only)
182 wheels resolved together (all versions compatible) for CPython 3.12 / win_amd64. Install Python 3.12.10 from Part 2 first, then run install_offline.bat (it first runs join_parts.ps1 to rebuild the split TensorFlow/PyTorch wheels, then creates %USERPROFILE%\afci-venv, installs everything with --no-index, then runs check_env.py).
Manual route: run join_parts.ps1 first (TensorFlow and PyTorch wheels are split into 100 MB pieces).
Manual install: pip install --no-index --find-links wheels_scientific_jupyter --find-links wheels_deep_learning -r lock_all.txt
| Folder | Main packages | Use in the AFCI project |
|---|---|---|
wheels_scientific_jupyter/ (146 wheels) |
numpy 2.5, scipy 1.18, pandas, pyarrow, matplotlib, seaborn, scikit-learn, imbalanced-learn, xgboost, lightgbm, optuna, PyWavelets, h5py, PyTables, PyYAML, tqdm, openpyxl, pyserial, cmsisdsp 1.10 (Python binding of CMSIS-DSP), JupyterLab 4.6, Notebook, ipykernel, ipywidgets | Waveform loading (CSV/HDF5), FFT/STFT/wavelet features, baselines (SVM, RF, XGBoost), metrics (ROC, F1, confusion matrix), reading ADC data from the STM32 over UART, checking that PC features match the MCU's CMSIS-DSP output |
wheels_deep_learning/ (36 wheels) |
torch 2.15 (CPU), torchaudio, torchinfo, torchmetrics, tensorflow 2.20 (CPU) + keras 3.15 + tensorboard, onnx 1.23, onnxruntime 1.31, onnxscript, tf2onnx, netron | Train 1D-CNN / CNN-LSTM / attention models, INT8 quantization, export to ONNX/TFLite for STM32 Edge AI / X-CUBE-AI / TFLite Micro, view model graphs (netron) |
Files: req_sci.txt, req_dl.txt (top-level wishes), lock_all.txt (exact pinned versions), install_offline.bat, check_env.py.
Licenses: each package keeps its own open-source license (BSD/MIT/Apache-2.0 mostly).
Rebuilding split files
Files over 100 MB were split into *.part001, *.part002, ... pieces. After downloading the whole repo:
- Windows:
powershell -ExecutionPolicy Bypass -File .\join_parts.ps1 - Linux/macOS:
bash join_parts.sh
Both scripts rebuild the files and check them against SHA256SUMS.txt.
Files (187 files, 761.2 MB total)
| File | Size | Split into |
|---|---|---|
check_env.py |
507 B | - |
install_offline.bat |
823 B | - |
lock_all.txt |
3.2 KB | - |
req_dl.txt |
94 B | - |
req_sci.txt |
224 B | - |
wheels_deep_learning/absl_py-2.5.1-py3-none-any.whl |
134.4 KB | - |
wheels_deep_learning/astunparse-1.6.3-py2.py3-none-any.whl |
12.4 KB | - |
wheels_deep_learning/filelock-4.2.7-py3-none-any.whl |
160.5 KB | - |
wheels_deep_learning/flatbuffers-25.12.19-py2.py3-none-any.whl |
26.0 KB | - |
wheels_deep_learning/fsspec-2026.9.0-py3-none-any.whl |
216.5 KB | - |
wheels_deep_learning/gast-0.7.0-py3-none-any.whl |
22.4 KB | - |
wheels_deep_learning/google_pasta-0.2.0-py3-none-any.whl |
56.1 KB | - |
wheels_deep_learning/grpcio-1.85.1-cp312-cp312-win_amd64.whl |
5.3 MB | - |
wheels_deep_learning/keras-3.15.1-py3-none-any.whl |
2.3 MB | - |
wheels_deep_learning/libclang-18.1.1-py2.py3-none-win_amd64.whl |
25.2 MB | - |
wheels_deep_learning/lightning_utilities-0.15.3-py3-none-any.whl |
31.2 KB | - |
wheels_deep_learning/markdown-3.11.2-py3-none-any.whl |
111.6 KB | - |
wheels_deep_learning/ml_dtypes-0.6.0-cp312-cp312-win_amd64.whl |
429.0 KB | - |
wheels_deep_learning/mpmath-1.3.0-py3-none-any.whl |
523.6 KB | - |
wheels_deep_learning/namex-0.1.0-py3-none-any.whl |
5.8 KB | - |
wheels_deep_learning/netron-9.3.4-py3-none-any.whl |
3.5 MB | - |
wheels_deep_learning/networkx-3.7-py3-none-any.whl |
2.0 MB | - |
wheels_deep_learning/onnx-1.23.2-cp312-abi3-win_amd64.whl |
7.5 MB | - |
wheels_deep_learning/onnx_ir-1.1.0-py3-none-any.whl |
187.7 KB | - |
wheels_deep_learning/onnxruntime-1.31.1-cp312-cp312-win_amd64.whl |
14.2 MB | - |
wheels_deep_learning/onnxscript-0.7.2-py3-none-any.whl |
736.5 KB | - |
wheels_deep_learning/opt_einsum-3.4.0-py3-none-any.whl |
70.2 KB | - |
wheels_deep_learning/optree-0.20.0-cp312-cp312-win_amd64.whl |
553.7 KB | - |
wheels_deep_learning/protobuf-7.36.2-cp310-abi3-win_amd64.whl |
445.8 KB | - |
wheels_deep_learning/sympy-1.14.0-py3-none-any.whl |
6.0 MB | - |
wheels_deep_learning/tensorboard-2.20.0-py3-none-any.whl |
5.3 MB | - |
wheels_deep_learning/tensorboard_data_server-0.7.2-py3-none-any.whl |
2.3 KB | - |
wheels_deep_learning/tensorflow-2.20.0-cp312-cp312-win_amd64.whl |
316.5 MB | 4 |
wheels_deep_learning/termcolor-3.3.0-py3-none-any.whl |
7.6 KB | - |
wheels_deep_learning/tf2onnx-1.17.0-py3-none-any.whl |
819.5 KB | - |
wheels_deep_learning/torch-2.15.0-cp312-cp312-win_amd64.whl |
120.9 MB | 2 |
wheels_deep_learning/torchaudio-2.11.1-py3-none-win_amd64.whl |
326.5 KB | - |
wheels_deep_learning/torchinfo-1.8.0-py3-none-any.whl |
22.8 KB | - |
wheels_deep_learning/torchmetrics-1.9.0-py3-none-any.whl |
960.3 KB | - |
wheels_deep_learning/werkzeug-3.1.9-py3-none-any.whl |
223.3 KB | - |
wheels_deep_learning/wrapt-2.5.2-cp312-cp312-win_amd64.whl |
103.7 KB | - |
wheels_scientific_jupyter/alembic-1.20.1-py3-none-any.whl |
264.0 KB | - |
wheels_scientific_jupyter/annotated_types-0.8.0-py3-none-any.whl |
13.1 KB | - |
wheels_scientific_jupyter/anyio-4.16.0-py3-none-any.whl |
131.4 KB | - |
wheels_scientific_jupyter/argon2_cffi-25.1.0-py3-none-any.whl |
14.3 KB | - |
wheels_scientific_jupyter/argon2_cffi_bindings-26.1.0-cp310-abi3-win_amd64.whl |
25.1 KB | - |
wheels_scientific_jupyter/arrow-1.4.0-py3-none-any.whl |
67.2 KB | - |
wheels_scientific_jupyter/asttokens-3.0.2-py3-none-any.whl |
28.0 KB | - |
wheels_scientific_jupyter/async_lru-2.4.0-py3-none-any.whl |
9.8 KB | - |
wheels_scientific_jupyter/attrs-26.1.0-py3-none-any.whl |
66.0 KB | - |
wheels_scientific_jupyter/babel-2.18.0-py3-none-any.whl |
9.7 MB | - |
wheels_scientific_jupyter/beautifulsoup4-4.15.0-py3-none-any.whl |
107.3 KB | - |
wheels_scientific_jupyter/bleach-6.4.0-py3-none-any.whl |
161.2 KB | - |
wheels_scientific_jupyter/blosc2-4.14.1-cp311-abi3-win_amd64.whl |
4.5 MB | - |
wheels_scientific_jupyter/certifi-2026.10.23-py3-none-any.whl |
131.9 KB | - |
wheels_scientific_jupyter/cffi-2.1.1-cp312-cp312-win_amd64.whl |
181.6 KB | - |
wheels_scientific_jupyter/charset_normalizer-3.5.2-cp312-cp312-win_amd64.whl |
202.6 KB | - |
wheels_scientific_jupyter/cloudpickle-3.1.2-py3-none-any.whl |
21.7 KB | - |
wheels_scientific_jupyter/cmsisdsp-1.10.1-cp312-cp312-win_amd64.whl |
2.8 MB | - |
wheels_scientific_jupyter/colorama-0.4.6-py2.py3-none-any.whl |
24.7 KB | - |
wheels_scientific_jupyter/colorlog-6.12.0-py3-none-any.whl |
12.0 KB | - |
wheels_scientific_jupyter/comm-0.2.3-py3-none-any.whl |
7.1 KB | - |
wheels_scientific_jupyter/contourpy-1.4.0-cp312-cp312-win_amd64.whl |
228.6 KB | - |
wheels_scientific_jupyter/cycler-0.12.1-py3-none-any.whl |
8.1 KB | - |
wheels_scientific_jupyter/debugpy-1.8.22-cp312-cp312-win_amd64.whl |
5.1 MB | - |
wheels_scientific_jupyter/defusedxml-0.7.1-py2.py3-none-any.whl |
25.0 KB | - |
wheels_scientific_jupyter/et_xmlfile-2.0.0-py3-none-any.whl |
17.6 KB | - |
wheels_scientific_jupyter/executing-2.3.0-py3-none-any.whl |
28.7 KB | - |
wheels_scientific_jupyter/fastjsonschema-2.22.3-py3-none-any.whl |
27.8 KB | - |
wheels_scientific_jupyter/fonttools-4.68.0-cp312-cp312-win_amd64.whl |
2.4 MB | - |
wheels_scientific_jupyter/fqdn-1.6.0-py3-none-any.whl |
12.3 KB | - |
wheels_scientific_jupyter/h11-0.16.0-py3-none-any.whl |
36.6 KB | - |
wheels_scientific_jupyter/h2-4.4.1-py3-none-any.whl |
61.2 KB | - |
wheels_scientific_jupyter/h5py-3.16.0-cp312-cp312-win_amd64.whl |
3.0 MB | - |
wheels_scientific_jupyter/hpack-4.2.0-py3-none-any.whl |
33.4 KB | - |
wheels_scientific_jupyter/httpcore-1.0.9-py3-none-any.whl |
76.9 KB | - |
wheels_scientific_jupyter/httpx-0.28.1-py3-none-any.whl |
71.8 KB | - |
wheels_scientific_jupyter/hyperframe-6.1.0-py3-none-any.whl |
12.7 KB | - |
wheels_scientific_jupyter/idna-3.20-py3-none-any.whl |
68.0 KB | - |
wheels_scientific_jupyter/imbalanced_learn-0.14.2-py3-none-any.whl |
230.5 KB | - |
wheels_scientific_jupyter/ipykernel-7.4.0-py3-none-any.whl |
120.3 KB | - |
wheels_scientific_jupyter/ipython-9.18.1-py3-none-any.whl |
634.6 KB | - |
wheels_scientific_jupyter/ipython_pygments_lexers-1.1.1-py3-none-any.whl |
7.9 KB | - |
wheels_scientific_jupyter/ipywidgets-8.1.9-py3-none-any.whl |
136.8 KB | - |
wheels_scientific_jupyter/isoduration-20.11.0-py3-none-any.whl |
11.1 KB | - |
wheels_scientific_jupyter/jedi-0.20.1-py2.py3-none-any.whl |
4.7 MB | - |
wheels_scientific_jupyter/jinja2-3.1.6-py3-none-any.whl |
131.7 KB | - |
wheels_scientific_jupyter/joblib-1.6.0-py3-none-any.whl |
298.9 KB | - |
wheels_scientific_jupyter/json5-0.17.3-py3-none-any.whl |
34.0 KB | - |
wheels_scientific_jupyter/jsonpointer-3.7-py3-none-any.whl |
9.1 KB | - |
wheels_scientific_jupyter/jsonschema-4.26.0-py3-none-any.whl |
88.5 KB | - |
wheels_scientific_jupyter/jsonschema_specifications-2025.9.1-py3-none-any.whl |
18.0 KB | - |
wheels_scientific_jupyter/jupyter_builder-1.2.4-py3-none-any.whl |
925.6 KB | - |
wheels_scientific_jupyter/jupyter_client-8.10.0-py3-none-any.whl |
107.6 KB | - |
wheels_scientific_jupyter/jupyter_core-5.9.1-py3-none-any.whl |
28.4 KB | - |
wheels_scientific_jupyter/jupyter_events-0.12.1-py3-none-any.whl |
19.1 KB | - |
wheels_scientific_jupyter/jupyter_lsp-2.3.1-py3-none-any.whl |
75.7 KB | - |
wheels_scientific_jupyter/jupyter_server-2.21.1-py3-none-any.whl |
385.4 KB | - |
wheels_scientific_jupyter/jupyter_server_terminals-0.5.4-py3-none-any.whl |
13.4 KB | - |
wheels_scientific_jupyter/jupyterlab-4.6.5-py3-none-any.whl |
16.5 MB | - |
wheels_scientific_jupyter/jupyterlab_pygments-0.3.0-py3-none-any.whl |
15.5 KB | - |
wheels_scientific_jupyter/jupyterlab_server-2.28.1-py3-none-any.whl |
58.8 KB | - |
wheels_scientific_jupyter/jupyterlab_widgets-3.0.17-py3-none-any.whl |
212.2 KB | - |
wheels_scientific_jupyter/kiwisolver-1.5.1-cp312-cp312-win_amd64.whl |
69.1 KB | - |
wheels_scientific_jupyter/lark-1.3.1-py3-none-any.whl |
110.5 KB | - |
wheels_scientific_jupyter/lightgbm-4.7.0-py3-none-win_amd64.whl |
1.3 MB | - |
wheels_scientific_jupyter/mako-1.4.3-py3-none-any.whl |
78.3 KB | - |
wheels_scientific_jupyter/markdown_it_py-4.2.0-py3-none-any.whl |
89.5 KB | - |
wheels_scientific_jupyter/markupsafe-3.0.4-cp312-cp312-win_amd64.whl |
14.0 KB | - |
wheels_scientific_jupyter/matplotlib-3.11.2-cp312-cp312-win_amd64.whl |
8.9 MB | - |
wheels_scientific_jupyter/matplotlib_inline-0.2.2-py3-none-any.whl |
9.3 KB | - |
wheels_scientific_jupyter/mdurl-0.1.2-py3-none-any.whl |
9.7 KB | - |
wheels_scientific_jupyter/mistune-3.4.0-py3-none-any.whl |
68.7 KB | - |
wheels_scientific_jupyter/msgpack-1.2.4-cp312-cp312-win_amd64.whl |
75.5 KB | - |
wheels_scientific_jupyter/narwhals-2.28.0-py3-none-any.whl |
479.1 KB | - |
wheels_scientific_jupyter/nbclient-0.11.0-py3-none-any.whl |
24.7 KB | - |
wheels_scientific_jupyter/nbconvert-7.17.2-py3-none-any.whl |
256.0 KB | - |
wheels_scientific_jupyter/nbformat-5.11.1-py3-none-any.whl |
78.0 KB | - |
wheels_scientific_jupyter/ndindex-1.10.1-cp312-cp312-win_amd64.whl |
154.5 KB | - |
wheels_scientific_jupyter/nest_asyncio2-1.7.4-py3-none-any.whl |
8.3 KB | - |
wheels_scientific_jupyter/notebook-7.6.4-py3-none-any.whl |
5.4 MB | - |
wheels_scientific_jupyter/notebook_shim-0.2.4-py3-none-any.whl |
13.0 KB | - |
wheels_scientific_jupyter/numexpr-2.14.2-cp312-cp312-win_amd64.whl |
153.9 KB | - |
wheels_scientific_jupyter/numpy-2.5.4-cp312-cp312-win_amd64.whl |
12.0 MB | - |
wheels_scientific_jupyter/openpyxl-3.1.5-py2.py3-none-any.whl |
245.0 KB | - |
wheels_scientific_jupyter/optuna-5.0.0-py3-none-any.whl |
432.0 KB | - |
wheels_scientific_jupyter/packaging-26.3-py3-none-any.whl |
126.9 KB | - |
wheels_scientific_jupyter/pandas-3.1.0-cp312-cp312-win_amd64.whl |
10.3 MB | - |
wheels_scientific_jupyter/pandocfilters-1.5.1-py2.py3-none-any.whl |
8.5 KB | - |
wheels_scientific_jupyter/parso-0.8.7-py2.py3-none-any.whl |
104.5 KB | - |
wheels_scientific_jupyter/pillow-13.0.0-cp312-cp312-win_amd64.whl |
6.9 MB | - |
wheels_scientific_jupyter/pip-26.2.2-py3-none-any.whl |
1.7 MB | - |
wheels_scientific_jupyter/platformdirs-4.14.3-py3-none-any.whl |
34.4 KB | - |
wheels_scientific_jupyter/prometheus_client-0.26.0-py3-none-any.whl |
63.0 KB | - |
wheels_scientific_jupyter/prompt_toolkit-3.0.53-py3-none-any.whl |
383.1 KB | - |
wheels_scientific_jupyter/psutil-7.2.2-cp37-abi3-win_amd64.whl |
134.5 KB | - |
wheels_scientific_jupyter/pure_eval-0.2.4-py3-none-any.whl |
11.6 KB | - |
wheels_scientific_jupyter/py_cpuinfo-9.0.0-py3-none-any.whl |
21.8 KB | - |
wheels_scientific_jupyter/pyarrow-26.0.0-cp312-cp312-win_amd64.whl |
27.2 MB | - |
wheels_scientific_jupyter/pycparser-3.12-py3-none-any.whl |
61.1 KB | - |
wheels_scientific_jupyter/pydantic-2.14.2-py3-none-any.whl |
478.9 KB | - |
wheels_scientific_jupyter/pydantic_core-2.50.2-cp312-cp312-win_amd64.whl |
2.1 MB | - |
wheels_scientific_jupyter/pygments-2.21.0-py3-none-any.whl |
1.2 MB | - |
wheels_scientific_jupyter/pyparsing-3.3.3-py3-none-any.whl |
123.5 KB | - |
wheels_scientific_jupyter/pyserial-3.5-py2.py3-none-any.whl |
88.5 KB | - |
wheels_scientific_jupyter/python_dateutil-4.0.0-py3-none-any.whl |
90.7 KB | - |
wheels_scientific_jupyter/python_json_logger-4.2.0-py3-none-any.whl |
14.6 KB | - |
wheels_scientific_jupyter/pywavelets-1.10.0-cp312-cp312-win_amd64.whl |
4.1 MB | - |
wheels_scientific_jupyter/pywinpty-3.1.2-cp312-cp312-win_amd64.whl |
2.0 MB | - |
wheels_scientific_jupyter/pyyaml-6.0.3-cp312-cp312-win_amd64.whl |
150.4 KB | - |
wheels_scientific_jupyter/pyzmq-27.2.0-cp312-abi3-win_amd64.whl |
613.9 KB | - |
wheels_scientific_jupyter/referencing-0.37.0-py3-none-any.whl |
26.1 KB | - |
wheels_scientific_jupyter/requests-2.34.2-py3-none-any.whl |
71.4 KB | - |
wheels_scientific_jupyter/rfc3339_validator-0.1.4-py2.py3-none-any.whl |
3.4 KB | - |
wheels_scientific_jupyter/rfc3986_validator-0.1.1-py2.py3-none-any.whl |
4.1 KB | - |
wheels_scientific_jupyter/rfc3987_syntax-1.1.0-py3-none-any.whl |
7.9 KB | - |
wheels_scientific_jupyter/rich-15.0.0-py3-none-any.whl |
303.4 KB | - |
wheels_scientific_jupyter/rpds_py-2026.9.1-cp312-cp312-win_amd64.whl |
217.7 KB | - |
wheels_scientific_jupyter/scikit_learn-1.9.1-cp312-cp312-win_amd64.whl |
7.9 MB | - |
wheels_scientific_jupyter/scipy-1.18.1-cp312-cp312-win_amd64.whl |
35.0 MB | - |
wheels_scientific_jupyter/seaborn-0.13.2-py3-none-any.whl |
288.0 KB | - |
wheels_scientific_jupyter/send2trash-2.1.0-py3-none-any.whl |
17.2 KB | - |
wheels_scientific_jupyter/setuptools-84.2.1-py3-none-any.whl |
855.7 KB | - |
wheels_scientific_jupyter/six-1.17.0-py2.py3-none-any.whl |
10.8 KB | - |
wheels_scientific_jupyter/sklearn_compat-0.1.6-py3-none-any.whl |
22.3 KB | - |
wheels_scientific_jupyter/soupsieve-3.0.4-py3-none-any.whl |
41.5 KB | - |
wheels_scientific_jupyter/sqlalchemy-2.1.6-cp312-cp312-win_amd64.whl |
2.3 MB | - |
wheels_scientific_jupyter/stack_data-0.6.3-py3-none-any.whl |
23.9 KB | - |
wheels_scientific_jupyter/tables-3.11.1-cp311-abi3-win_amd64.whl |
6.2 MB | - |
wheels_scientific_jupyter/terminado-0.18.1-py3-none-any.whl |
13.8 KB | - |
wheels_scientific_jupyter/threadpoolctl-3.7.0-py3-none-any.whl |
25.7 KB | - |
wheels_scientific_jupyter/tinycss2-1.5.1-py3-none-any.whl |
27.7 KB | - |
wheels_scientific_jupyter/tornado-6.6-cp311-abi3-win_amd64.whl |
518.4 KB | - |
wheels_scientific_jupyter/tqdm-4.70.1-py3-none-any.whl |
78.3 KB | - |
wheels_scientific_jupyter/traitlets-5.16.1-py3-none-any.whl |
84.2 KB | - |
wheels_scientific_jupyter/typing_extensions-4.16.0-py3-none-any.whl |
44.5 KB | - |
wheels_scientific_jupyter/typing_inspection-0.4.4-py3-none-any.whl |
14.4 KB | - |
wheels_scientific_jupyter/tzdata-2026.6-py2.py3-none-any.whl |
340.0 KB | - |
wheels_scientific_jupyter/uri_template-1.3.0-py3-none-any.whl |
10.9 KB | - |
wheels_scientific_jupyter/urllib3-2.8.0-py3-none-any.whl |
132.5 KB | - |
wheels_scientific_jupyter/wcwidth-0.9.4-cp310-abi3-win_amd64.whl |
583.8 KB | - |
wheels_scientific_jupyter/webcolors-25.10.0-py3-none-any.whl |
14.6 KB | - |
wheels_scientific_jupyter/webencodings-0.6.1-py3-none-any.whl |
8.5 KB | - |
wheels_scientific_jupyter/websocket_client-1.9.2-py3-none-any.whl |
93.5 KB | - |
wheels_scientific_jupyter/wheel-0.48.0-py3-none-any.whl |
32.5 KB | - |
wheels_scientific_jupyter/widgetsnbextension-4.0.16-py3-none-any.whl |
2.1 MB | - |
wheels_scientific_jupyter/xgboost-3.4.1-py3-none-win_amd64.whl |
46.7 MB | - |
Each file keeps the license of its original publisher (see the notes above).
Master index
This is one of 19 repos (AFCI_part1 … AFCI_part19). Overview of all parts: AFCI_part6 / 00_START_HERE_MASTER_INDEX.md · Installation: AFCI_part6 / 00_INSTALLATION_GUIDE.md
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support