HyperView

Interactive geometric workbench for embedding space analysis.

HyperView turns image datasets into live embedding workspaces. Load samples, compute embeddings, inspect Euclidean, hyperbolic, or spherical layouts, select clusters and outliers, and keep media, labels, layouts, selections, panels, and tools in one local workspace.

Open a live workspace in your browser — no install, no backend

Six prepared workspaces, each exported with hyperview export and served as static files: ABO Catalog · Fashion Products · Precision Regions · Logo Search · GeoSpatial · Visual Safety

Install

uv tool install --upgrade hyperview
hyperview skill install

For a project-local GitHub Copilot skill:

hyperview skill install --scope project --agent github-copilot --yes

Python

import hyperview as hv

dataset = hv.Dataset("cifar100")
dataset.add_from_huggingface("uoft-cs/cifar100", split="train", max_samples=1000)
dataset.compute_embeddings(model="openai/clip-vit-base-patch32")
dataset.compute_visualization()

hv.launch(dataset)

CLI

hyperview workspace create imagenette-demo \
  --dataset imagenette_clip_20260411 \
  --activate

hyperview serve \
  --workspace imagenette-demo \
  --dataset imagenette_clip_20260411

Agents and scripts can control the running workspace:

hyperview ui layout set --workspace imagenette-demo --layout-key <layout-key>
hyperview ui selection set --workspace imagenette-demo --ids sample-1,sample-8
hyperview extension add .hyperview/extensions/selection-profile --workspace imagenette-demo
hyperview tools run selection_profile.summarize --workspace imagenette-demo

Export a prepared workspace as a Static Space, a read-only bundle that can be served by Cloudflare Workers Static Assets without Python or a container:

hyperview export imagenette-demo --out dist/imagenette-demo
cd dist/imagenette-demo && npx wrangler deploy --config wrangler.jsonc

What It Does

• Ingests image data from Hugging Face datasets or local folders.
• Computes embeddings with built-in or custom providers.
• Projects spaces into Euclidean, Poincare/hyperbolic, and spherical layouts.
• Links scatter points to thumbnails, labels, selections, and nearest neighbors.
• Adds dataset-specific Python tools and native frontend panels.
• Gives coding agents a CLI control plane for the same workspace humans inspect.

Docs

• Datasets
• Static Spaces
• Google Colab
• Contributing
• Testing and linting

Why Geometry?

Embedding failures often hide in the shape of the space: collapsed classes, weak separation, hierarchy, long-tail samples, and boundary cases. HyperView lets you inspect those structures through multiple geometric views instead of a single fixed projection.

Related Projects

• hyperview-spaces: Source for the demos above; copy a folder and point it at your own dataset.
• hyper-scatter: WebGL scatterplot engine for Euclidean and Poincare views.
• hyper-models: Non-Euclidean model zoo and ONNX exports.
• hyper-lrp: Attribution tools for inspecting model evidence.

Community

Join the Hyper3Labs Discord for demos, setup help, and project discussion.

License

MIT License. See LICENSE.

Citation

If you use HyperView in research, please cite:

@software{hyperview2026,
  author  = {{Hyper3Labs}},
  title   = {HyperView: An Interactive Geometric Workbench for Embedding Space Analysis},
  year    = {2026},
  version = {1.0.0},
  url     = {https://github.com/Hyper3Labs/HyperView}
}