Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

MISTIC logo: hands interpreting a selected feature

MISTIC

Model Informed Feature Selection Through Importance and Contribution

MISTIC provides feature selection, boundary counterfactuals, and attribution for interpretable and explainable support vector machines.

Installation

Install the release from PyPI:

python -m pip install mistic-svm

The PyPI distribution is named mistic-svm because mistic is already used by an unrelated project. The Python import remains concise:

from mistic import MixedKernel, cvSet, kernelWrapper, paramSet, score_svc, svmSet

For development, install the repository and its development tools in editable mode:

python -m pip install -e ".[dev]"

The package requires Python 3.10 or newer. Example classification and regression workflows are available in mistic/examples.

MISTIC 0.2 beta

The 0.1 series remains the default stable release. Install the opt-in 0.2 beta with an exact version:

python -m pip install "mistic-svm==0.2.0b1"

The beta source and documentation are maintained on the 0.2 branch. Beta releases may introduce compatibility changes before 0.2 becomes stable; see the migration guide and changelog before upgrading.

Latin-hypercube parameter designs are available without additional dependencies:

from mistic import loguniform, parameterSpace

grid = parameterSpace(
    model={"C": loguniform(1e-3, 1e3)},
    kernel={"gamma": loguniform(1e-6, 1e1)},
).sample(n_trials=24, random_state=7)

All kernels use the composable MixedKernel API. A single kernel is a one-leaf mixture, while named leaves can be added, multiplied, powered, given tunable nonnegative weights, and assigned independent feature sets:

kernel = MixedKernel.weighted_sum(
    [
        kernelWrapper("linear", name="linear"),
        kernelWrapper("rbf", name="radial"),
    ],
    weight_parameters=["linear_weight", "radial_weight"],
)

See the mixed-kernel tutorial in the documentation for parameter naming, kernel-specific candidate pools, and unified forward/backward selection.

Synthetic-data validation studies and their generated results are kept in validation so they remain separate from the user-facing examples.

Documentation and tests

Read the complete installation guide, framework overview, tutorials, examples, and API reference at biomodsquad.org/interpretable-explainable-svms.

Build the documentation locally with sphinx-build -W docs docs/_build/html and run the test suite with pytest.

Repository

The canonical source repository is biomodsquad/interpretable-explainable-svms.

Metadata

Release files for mistic-svm 0.2.0b1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for mistic-svm 0.2.0b1
File Size Uploaded Provenance Actions
mistic_svm-0.2.0b1.tar.gz 162.2 kB With provenance Details for mistic_svm-0.2.0b1.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for mistic-svm 0.2.0b1
File Interpreter ABI Platform Actions
mistic_svm-0.2.0b1-py3-none-any.whl Python 3 none any Details for mistic_svm-0.2.0b1-py3-none-any.whl

Total release size: 309.4 kB

Release files / mistic_svm-0.2.0b1.tar.gz

Download URL mistic_svm-0.2.0b1.tar.gz
Size 162.2 kB
Tags Source
SHA-256 checksum
How to use checksums
6dc25fff5edd9367e10a9bed79ca93cbe151abfadedb01f3d36c66748becd1fc
BLAKE2b-256 checksum
How to use checksums
6882ba4ed6806ca58f1e5ed9dff4a60ba3f3b5064496be55dbd3047f85cbb07f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 30, 2026.

Transparency log

Release files / mistic_svm-0.2.0b1-py3-none-any.whl

Download URL mistic_svm-0.2.0b1-py3-none-any.whl
Size 147.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
ecf766bbfbffd12d7fb273a498ecd825d7e3e3f8fcca825999236ebb3ed5d6e8
BLAKE2b-256 checksum
How to use checksums
06bdc134833fef85926de7bbf9af03a0792f91d68025293ea948d123ecd52e82
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 30, 2026.

Transparency log

Security for mistic-svm 0.2.0b1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for mistic_svm-0.2.0b1.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 30, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page