alibi 0.3.2

Algorithms for monitoring and explaining machine learning models

Alibi is an open source Python library aimed at machine learning model inspection and interpretation. The initial focus on the library is on black-box, instance based model explanations.

• Documentation

Goals

• Provide high quality reference implementations of black-box ML model explanation algorithms
• Define a consistent API for interpretable ML methods
• Support multiple use cases (e.g. tabular, text and image data classification, regression)
• Implement the latest model explanation, concept drift, algorithmic bias detection and other ML model monitoring and interpretation methods

Installation

Alibi can be installed from PyPI:

pip install alibi

This will install alibi with all its dependencies:

  beautifulsoup4
  numpy
  Pillow
  pandas
  requests
  scikit-learn
  spacy
  scikit-image
  tensorflow

To run all the example notebooks, you may additionally run pip install alibi[examples] which will install the following:

  seaborn
  Keras

Supported algorithms

Black-box model explanaton

• Anchor explanations (Ribeiro et al., 2018)

  • Documentation
  • Examples: income prediction, Iris dataset, movie sentiment classification, ImageNet, fashion MNIST

• Contrastive Explanation Method (CEM, Dhurandhar et al., 2018)

  • Documentation
  • Examples: MNIST, Iris dataset

• Counterfactual Explanations (extension of Wachter et al., 2017)

  • Documentation
  • Examples: MNIST

• Counterfactual Explanations Guided by Prototypes (Van Looveren et al., 2019)

  • Documentation
  • Examples: MNIST, Boston housing dataset, Adult income (one-hot), Adult income (ordinal)

Model confidence metrics

• Trust Scores (Jiang et al., 2018)
  • Documentation
  • Examples: MNIST, Iris dataset
• Linearity Measure
  • Examples: Iris dataset, fashion MNIST

Example outputs

Anchor method applied to the InceptionV3 model trained on ImageNet:

Prediction: Persian Cat	Anchor explanation

Contrastive Explanation method applied to a CNN trained on MNIST:

Prediction: 4	Pertinent Negative: 9	Pertinent Positive: 4

Trust scores applied to a softmax classifier trained on MNIST: