uber/manifold
A model-agnostic visual debugging tool for machine learning
When you're trying to understand why your machine learning models are making mistakes, Manifold helps you visually pinpoint exactly which data points are causing issues and why. You provide your model's predictions, the actual correct answers, and the input data features. Manifold then shows you interactive visualizations that highlight underperforming data segments and the specific input features contributing to those errors. This tool is for data scientists, machine learning engineers, and researchers who need to debug and improve their ML models beyond just looking at overall performance scores.
1,672 stars. No commits in the last 6 months.
Use this if you need to visually analyze why your machine learning model performs poorly on certain data subsets and identify the input features that correlate with those errors.
Not ideal if you're looking for an automated model training solution or a tool to deploy models into production.
Stars
1,672
Forks
116
Language
JavaScript
License
Apache-2.0
Category
Last pushed
Feb 05, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/uber/manifold"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ndif-team/nnsight
The nnsight package enables interpreting and manipulating the internals of deep learned models.
huggingface/dataset-viewer
Backend that powers the dataset viewer on Hugging Face dataset pages through a public API.
LYH-YF/MWPToolkit
MWPToolkit is an open-source framework for math word problem(MWP) solvers.
epfml/disco
DISCO is a code-free and installation-free browser platform that allows any non-technical user...
ndif-team/ndif
The NDIF server, which performs deep inference and serves nnsight requests remotely