suinleelab/attributionpriors
Tools for training explainable models using attribution priors.
This project helps data scientists and machine learning engineers build more understandable and accurate models by directly influencing how the model learns to make decisions. It takes your existing model training data and outputs a model that not only performs well but also clearly shows which features were most important for its predictions. This is for anyone who needs to explain why their AI model made a particular decision, especially in fields like image analysis, genomics, or business forecasting.
125 stars. No commits in the last 6 months. Available on PyPI.
Use this if you need to train a machine learning model where understanding *why* it makes a prediction is as crucial as the prediction itself, or if you want to incorporate specific domain knowledge into how your model attributes importance to different input features.
Not ideal if your primary goal is simply to achieve the highest possible prediction accuracy without any concern for model interpretability or feature attribution.
Stars
125
Forks
8
Language
Jupyter Notebook
License
MIT
Last pushed
Mar 19, 2021
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/suinleelab/attributionpriors"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
obss/sahi
Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
tensorflow/tcav
Code for the TCAV ML interpretability project
MAIF/shapash
🔅 Shapash: User-friendly Explainability and Interpretability to Develop Reliable and Transparent...
TeamHG-Memex/eli5
A library for debugging/inspecting machine learning classifiers and explaining their predictions
csinva/imodels
Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling...