https://github.com/andosa/treeinterpreter

https://github.com/andosa/treeinterpreter

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/andosa/treeinterpreter
• Owner: andosa
• License: bsd-3-clause
• Created: 2015-08-02T20:26:21.000Z (over 9 years ago)
• Default Branch: master
• Last Pushed: 2023-07-18T10:50:27.000Z (over 1 year ago)
• Last Synced: 2024-11-07T10:18:57.915Z (3 months ago)
• Language: Python
• Size: 27.3 KB
• Stars: 745
• Watchers: 25
• Forks: 140
• Open Issues: 26
• Metadata Files:
  • Readme: README.rst
  • Changelog: HISTORY.rst
  • License: LICENSE

Awesome Lists containing this project

• awesome-production-machine-learning - TreeInterpreter - Package for interpreting scikit-learn's decision tree and random forest predictions. Allows decomposing each prediction into bias and feature contribution components as described in http://blog.datadive.net/interpreting-random-forests/. (Explaining Black Box Models and Datasets)
• Awesome-AIML-Data-Ops - TreeInterpreter - Package for interpreting scikit-learn's decision tree and random forest predictions. Allows decomposing each prediction into bias and feature contribution components as described in http://blog.datadive.net/interpreting-random-forests/. (Explaining Black Box Models and Datasets)
• awesome-production-machine-learning - TreeInterpreter - Package for interpreting scikit-learn's decision tree and random forest predictions. Allows decomposing each prediction into bias and feature contribution components as described [here](http://blog.datadive.net/interpreting-random-forests). (Explainability and Fairness)
• Awesome-explainable-AI - https://github.com/andosa/treeinterpreter - learn,  (Python Libraries(sort in alphabeta order) / Evaluation methods)
• awesome-python-machine-learning-resources - GitHub - 82% open · ⏱️ 28.02.2021): (模型的可解释性)

README

        
===============================

TreeInterpreter

===============================

Package for interpreting scikit-learn's decision tree and random forest predictions.

Allows decomposing each prediction into bias and feature contribution components as described in http://blog.datadive.net/interpreting-random-forests/. For a dataset with ``n`` features, each prediction on the dataset is decomposed  as ``prediction = bias + feature_1_contribution + ... + feature_n_contribution``.

It works on scikit-learn's

* DecisionTreeRegressor

* DecisionTreeClassifier

* ExtraTreeRegressor

* ExtraTreeClassifier

* RandomForestRegressor

* RandomForestClassifier

* ExtraTreesRegressor

* ExtraTreesClassifier

Free software: BSD license

Dependencies

------------

- scikit-learn 0.17+

Installation

------------

The easiest way to install the package is via ``pip``::

    $ pip install treeinterpreter

Usage

-----

::

 from treeinterpreter import treeinterpreter as ti

 # fit a scikit-learn's regressor model

 rf = RandomForestRegressor()

 rf.fit(trainX, trainY)

 

 prediction, bias, contributions = ti.predict(rf, testX)

 

Prediction is the sum of bias and feature contributions::

 

 assert(numpy.allclose(prediction, bias + np.sum(contributions, axis=1)))

 assert(numpy.allclose(rf.predict(testX), bias + np.sum(contributions, axis=1)))

More usage examples at http://blog.datadive.net/random-forest-interpretation-with-scikit-learn/.