sungyubkim/gex
Official code implementation of "GEX: A flexible method for approximating influence via Geometric Ensemble" (NeurIPS 2023)
This project helps machine learning researchers understand which specific training data points most influence the predictions of a neural network. You input your pre-trained neural network and training/test datasets, and it outputs a measure of 'influence' for each data point, indicating its impact. This tool is designed for ML researchers or data scientists who need to diagnose and fix issues like bias or unexpected model behavior.
No commits in the last 6 months.
Use this if you are a machine learning researcher who needs to quantify the impact of individual training data points on your model's predictions, especially when dealing with potential data bias or noise.
Not ideal if you are looking for a general-purpose explainability tool for end-users, or if your primary goal is real-time inference optimization.
Stars
14
Forks
2
Language
Python
License
MIT
Last pushed
Jan 03, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sungyubkim/gex"
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...