Learning about Interpretable Machine Learning Methods.
-
Updated
Mar 13, 2021 - Jupyter Notebook
Learning about Interpretable Machine Learning Methods.
Decompose Thermo Gravimetrical Analysis (TGA) curves into simpler logistic curves representing mass-change events with a chemical interpretation. All of the analysis is performed with the TensorFlow library for the creation of a NN-analogous model and optimization.
Permutation feature importance
Pytorch example of path-explain using Pytorch
Course project for CS726 @ IIT Bombay.
A curated list of awesome machine learning interpretability resources.
Source code for the paper "Interpretable models from distributed data via merging of decision trees" (Artur Andrzejak, Felix Langner, Silvestre Zabala, CIDM 2013).
an end to end anomaly intrusion base on deep learn
Supplementary programmes for DeRDaVa: Deletion-Robust Data Valuation for Machine Learning.
Investigating Machine Learning explainability in credit risk models by utilising LIME and DiCE methods
JAX-based Model Explanation and Interpretation Library
A toolkit for interpreting and analyzing neural networks (vision)
Interpretability of deep representation learning models
The code for the paper "Nearest neighbour approaches for Emotion Detection in Tweets" by O. Kaminska, Ch. Cornelis and V. Hoste, presented at WASSA workshop (AECL 2021) as a poster.
Experimental tree boosted varying coefficient model
Explainable Data Decompositions with Desc (AAAI 2020)
Awesome papers on Interpretable Machine Learning
This repository provides a summarization of recent empirical studies/human studies that measure human understanding with machine explanations in human-AI interactions.
ML implementations in Multi-scale model for lignin biosynthesis in Populus Trichocarpa
Add a description, image, and links to the interpretable-machine-learning topic page so that developers can more easily learn about it.
To associate your repository with the interpretable-machine-learning topic, visit your repo's landing page and select "manage topics."