Installation¶
Requirements¶
PoreMind requires Python 3.10 or later. Download the ZIP archive from the PoreMind GitHub repository, extract it, and open a terminal in the extracted project root—the directory containing pyproject.toml and src/.
Install the package¶
conda create -n poremind python=3.10 -y
conda activate poremind
pip install -e .
Optional dependencies¶
Install only the packages needed for the planned analysis:
pip install pyabf # ABF input
pip install torch # waveform deep learning
pip install captum tqdm # DL attribution
pip install umap-learn # UMAP
pip install xgboost # XGBoost models
Verify the installation¶
from poremind import __version__, create_analysis_object
print(__version__)
To build or preview this documentation locally, install its separate requirements with pip install -r docs/requirements-docs.txt, then run python scripts/prepare_docs.py and mkdocs serve from the project root.