A selection of 8 curated ml-interpretability Python libraries and frameworks ordered by stars.
Checkout the interactive version that you can filter and sort: https://www.awesomepython.org/
A game theoretic approach to explain the output of any machine learning model.
https://github.com/slundberg/shap
56 stars per week over 269 weeks
15,188 stars, 2,288 forks, 246 watches
created 2016-11-22, last commit 2021-12-04, main language Jupyter Notebook
deep-learning, explainability, gradient-boosting, interpretability, machine-learning, shap, shapley
Lime: Explaining the predictions of any machine learning classifier
https://github.com/marcotcr/lime
30 stars per week over 305 weeks
9,476 stars, 1,565 forks, 273 watches
created 2016-03-15, last commit 2021-07-29, main language JavaScript
Model interpretability and understanding for PyTorch
https://captum.ai
https://github.com/pytorch/captum
22 stars per week over 125 weeks
2,891 stars, 307 forks, 195 watches
created 2019-08-27, last commit 2022-01-19, main language Python
feature-attribution, feature-importance, interpretability, interpretable-ai, interpretable-ml
The Language Interpretability Tool: Interactively analyze NLP models for model understanding in an extensible and framework agnostic interface.
https://pair-code.github.io/lit
https://github.com/pair-code/lit
35 stars per week over 77 weeks
2,788 stars, 282 forks, 73 watches
created 2020-07-28, last commit 2021-12-21, main language Python
machine-learning, natural-language-processing, visualization
Algorithms for explaining machine learning models
https://docs.seldon.io/projects/alibi/en/latest/
https://github.com/SeldonIO/alibi
9.82 stars per week over 151 weeks
1,490 stars, 175 forks, 43 watches
created 2019-02-26, last commit 2022-01-18, main language Python
counterfactual, explanations, interpretability, machine-learning, xai
Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations directly in Jupyter notebooks explaining the behavior of Transformer-based language models (like GPT2, BERT, RoBERTA, T5, and T0).
https://ecco.readthedocs.io
https://github.com/jalammar/ecco
20 stars per week over 63 weeks
1,295 stars, 83 forks, 19 watches
created 2020-11-07, last commit 2022-01-18, main language Jupyter Notebook
explorables, language-models, natural-language-processing, nlp, pytorch, visualization
Quickly build Explainable AI dashboards that show the inner workings of so-called "blackbox" machine learning models.
http://explainerdashboard.readthedocs.io
https://github.com/oegedijk/explainerdashboard
8.28 stars per week over 116 weeks
965 stars, 122 forks, 14 watches
created 2019-10-30, last commit 2021-12-24, main language Python
dash, dashboard, data-scientists, explainer, inner-workings, interactive-dashboards, interactive-plots, model-predictions, permutation-importances, plotly, shap, shap-values, xai, xai-library
Model explainability that works seamlessly with 🤗 transformers. Explain your transformers model in just 2 lines of code.
https://github.com/cdpierse/transformers-interpret
6.23 stars per week over 86 weeks
539 stars, 46 forks, 16 watches
created 2020-05-27, last commit 2021-11-25, main language Python
captum, deep-learning, explainable-ai, interpretability, machine-learning, model-explainability, natural-language-processing, neural-network, nlp, transformers, transformers-model
This file was automatically generated on 2022-01-23.
To curate your own github list, simply clone and change the input csv file.
Inspired by:
https://github.com/vinta/awesome-python
https://github.com/trananhkma/fucking-awesome-python