Hyper3Labs/HyperView
HyperView curates datasets and provides model introspection in hyperbolic and Euclidean geometries.
HyperView helps machine learning practitioners visually analyze image datasets and understand how their models 'see' and group images. It takes image data, computes its underlying structure, and outputs interactive visualizations showing relationships between images in both traditional and advanced geometric layouts. This tool is ideal for data scientists and ML engineers working with image classification, clustering, or content moderation.
Available on PyPI.
Use this if you need to visually explore complex image datasets and inspect model embeddings to identify data biases, misclassifications, or hidden patterns.
Not ideal if you are looking for a tool to train models or perform data augmentation; it focuses solely on visualization and introspection.
Stars
16
Forks
1
Language
Python
License
MIT
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Dependencies
12
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/Hyper3Labs/HyperView"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cosmosgl/graph
GPU-accelerated force graph layout and rendering
Clay-foundation/model
The Clay Foundation Model - An open source AI model and interface for Earth
nomic-ai/nomic
Nomic Developer API SDK
omoindrot/tensorflow-triplet-loss
Implementation of triplet loss in TensorFlow
sashakolpakov/dire-jax
DImensionality REduction in JAX