Skip to main content

Package to understand ML Models

Project description

ML Insights

Package to understand Supervised ML Models. This package has been tested with Scikit-Learn and XGBoost library. It should work with any machine learning library that has a predict and predict_proba methods for regression and classification estimators.

There are currently two main sets of capabilities of this package. The first is around understanding “black-box” models via the “Model X-Ray”. The second is for probability calibration.

For understanding black-box models, the main entry point is the ModelXRay class. Instantiate it with the model and data. The data can be what the model was trained with, but intended to be used for out of bag or test data to see how the model performs when one feature is changed, holding everything else constant.

For probability calibration, the main class is the SplineCalibratedClassifierCV. Using this class you can train your base model, and the corrective calibration function with just a couple of lines of code. See the examples by following the link below.

Python

Python 2.7 and 3.4+

Disclaimer

We have tested this tool to the best of our ability, but understand that it may have bugs. It was developed on Python 3.5, so should work better with Python 3 than 2. Use at your own risk, but feel free to report any bugs to our github. <https://github.com/numeristical/introspective>

Installation

$ pip install ml_insights

Usage

>>> import ml_insights as mli
>>> xray = mli.ModelXRay(model, data)
>>> rfm = RandomForestClassifier(n_estimators = 500, class_weight='balanced_subsample')
>>> rfm_cv = mli.SplineCalibratedClassifierCV(rfm)
>>> rfm_cv.fit(X_train,y_train)
>>> test_res_calib_cv = rfm_cv.predict_proba(X_test)[:,1]
>>> log_loss(y_test,test_res_calib_cv)

Source

Find the latest version on github: https://github.com/numeristical/introspective

Feel free to fork and contribute!

License

Free software: MIT license

Developed By

  • Brian Lucena
  • Ramesh Sampath

References

Alex Goldstein, Adam Kapelner, Justin Bleich, and Emil Pitkin. 2014. Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation. Journal of Computational and Graphical Statistics (March 2014)

Project details


Download files

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

Filename, size & hash SHA256 hash help File type Python version Upload date
ml_insights-0.0.17-py2.py3-none-any.whl (19.0 kB) Copy SHA256 hash SHA256 Wheel py2.py3
ml_insights-0.0.17.tar.gz (14.0 kB) Copy SHA256 hash SHA256 Source None

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page