Installation

QSARmil is built on three core packages: RDKit (conformer generation and 3D descriptors), MolFeat (additional 3D descriptors), and click (the command-line interface).

Two companion packages are installed automatically as dependencies and power the modeling side of the pipeline:

• milearn (distributed on PyPI as mikit-learn) — the multi-instance learning algorithms, see Multi-instance learning.

• QSARcons (distributed on PyPI as qsarcons) — consensus model selection, see Consensus modelling.

You do not need to install either of them separately.

Option 1 — one-line install from scratch with conda

conda env create -f environment.yml && conda activate qsarmil

Option 2 — install into an existing environment

pip install qsarmil