Skip to main content

Microsoft Interpret Community SDK for Python

This package is tested with Python 3.10, 3.11, 3.12, 3.13 and 3.14

The Interpret Community SDK builds on Interpret, an open source python package from Microsoft Research for training interpretable models, and helps to explain blackbox systems by adding additional extensions from the community to interpret ML models.

Interpret-Community is an experimental repository that hosts a wide range of community developed machine learning interpretability techniques. This repository makes it easy for anyone involved in the development of a machine learning system to improve transparency around their machine learning models. Data scientists, machine learning engineers, and researchers can easily add their own interpretability techniques via the set of extension hooks built into the peer repository, Interpret, and expand this repository to include their custom-made interpretability techniques.

Highlights of the package include:

  • The TabularExplainer can be used to give local and global feature importances
  • The best explainer is automatically chosen for the user based on the model
  • Local feature importances are for each evaluation row
  • Global feature importances summarize the most importance features at the model-level
  • The API supports both dense (numpy or pandas) and sparse (scipy) datasets
  • There are utilities provided to convert engineered explanations, based on preprocessed data before training a model, to raw explanations on the original dataset
  • For more advanced users, individual explainers can be used
  • The KernelExplainer, GPUKernelExplainer, PFIExplainer and MimicExplainer are for BlackBox models
  • The MimicExplainer is faster but less accurate than the KernelExplainer, and supports various surrogate model types
  • The TreeExplainer is for tree-based models
  • The LinearExplainer is for linear models
  • The DeepExplainer is for DNN tensorflow or pytorch models
  • The PFIExplainer can quickly compute global importance values
  • LIMEExplainer builds local linear approximations of the model's behavior by perturbing each instance
  • GPUKernelExplainer is GPU-accelerated implementation of SHAP's KernelExplainer as a part of RAPIDS's cuML library, and is optimized for GPU models, like those in cuML. It can be used with CPU-based estimators too.
  • An adapter to convert any feature importance values to an interpret-community style explanation

Please see the github website for the documentation and sample notebooks: https://github.com/interpretml/interpret-community

Auto-generated sphinx API documentation can be found here: https://interpret-community.readthedocs.io/en/latest/index.html

More information on the ExplanationDashboard can be found here: https://github.com/microsoft/responsible-ai-toolbox

Metadata

Release files for interpret-community 0.34.0

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

Source distribution (sdist)

Source distribution for interpret-community 0.34.0
File Size Uploaded Provenance Actions
interpret_community-0.34.0.tar.gz 91.7 kB Without provenance Details for interpret_community-0.34.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for interpret-community 0.34.0
File Interpreter ABI Platform Actions
interpret_community-0.34.0-py3-none-any.whl Python 3 none any Details for interpret_community-0.34.0-py3-none-any.whl

Total release size: 223.5 kB

Release files / interpret_community-0.34.0.tar.gz

Download URL interpret_community-0.34.0.tar.gz
Size 91.7 kB
Tags Source
SHA-256 checksum
How to use checksums
84b54c4b93368ece10f1a7ce86469db76e22b313c562ebe41bd305fa632cb37e
BLAKE2b-256 checksum
How to use checksums
5411b5a2742e550ad470ed8a3c211fd882cc76c75070cdc1e84b4909ee4dbbdd
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.16

Release files / interpret_community-0.34.0-py3-none-any.whl

Download URL interpret_community-0.34.0-py3-none-any.whl
Size 131.9 kB
Tags Python 3
SHA-256 checksum
How to use checksums
c169933029db053a0e843a46c2a969772cb3ba9dde457852fcf07e021f5f9559
BLAKE2b-256 checksum
How to use checksums
6eacb5cc8a298bc29487dfdc757a8b72af08c1f332876d7de6f5f08ae2b77078
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.16

Security for interpret-community 0.34.0

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.

No provenance found

None of the files in this release were uploaded with provenance metadata.

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.

Release history Release notifications | RSS feed

This release

0.34.0 This release

2 release files No provenance

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