Skip to main content

Compute the statistical impact of features given a trained estimator

Project description

featureimpact let’s you compute the statistical impact of features given a trained estimator. The computation is based on the mean variation of the difference between perturbed and original predictions. The estimator must predict purely numerical values. All features must also consist of purely numerical values.

Example: ` from featureimpact import FeatureImpact fi = FeatureImpact() fi.make_quantiles(X_train) impact = fi.compute_impact(model, X_test) `

Note: In order to run the examples you’ll need scikit-learn installed in addition to this package and its regular dependencies.

The algorithm is described here: https://bloomen.github.io/pub/featureimpact.pdf

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

featureimpact-2.2.0.tar.gz (5.7 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page