AntonotnaWang/HINT
[CVPR 2022] HINT: Hierarchical Neuron Concept Explainer
This tool helps machine learning researchers and practitioners understand how image classification models 'think' by identifying which parts of the neural network are responsible for detecting specific visual concepts, from concrete objects like 'dog' to abstract categories like 'animal'. You input a pre-trained image classification model and image datasets, and it outputs insights into neuron activation for various concepts, along with visualizations of how the model localizes these concepts within images.
No commits in the last 6 months.
Use this if you need to systematically and quantitatively analyze the internal workings of vision models, especially to debug their behavior or interpret their predictions by linking internal neurons to human-understandable concepts.
Not ideal if you are looking for a tool to improve model accuracy or for general model development and training, as its focus is on interpretability, not performance optimization.
Stars
20
Forks
—
Language
Jupyter Notebook
License
MIT
Last pushed
Apr 19, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/AntonotnaWang/HINT"
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...