# tensorspace-team/tensorspace

Neural network 3D visualization framework, build interactive and intuitive model in browsers, support pre-trained deep learning models from TensorFlow, Keras, TensorFlow.js

Repository: https://github.com/tensorspace-team/tensorspace
Canonical: https://ross.abutalabs.com/products/tensorspace
Homepage: https://tensorspace.org
Language: JavaScript
License: Apache-2.0
License Family: permissive
Topics: deep-learning, tensorflow, keras, tfjs, 3d, visualization, threejs, machine-learning, nerual-network
Last push: 2022-12-05T22:21:36+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2964, "days_push": 1367, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5191, forks 450 (observed 2026-08-28T04:09:11.960419+00:00)

## What it is
TensorSpace is a neural network 3D visualization framework built on TensorFlow.js, Three.js, and Tween.js. It provides Keras-like APIs to build deep learning layer topologies in the browser and render interactive 3D visualizations of pre-trained models from TensorFlow, Keras, and TensorFlow.js.

## Use cases
- visualize a neural network model structure in 3D in the browser
- show intermediate layer outputs of a deep learning model interactively
- create an interactive LeNet or YOLOv2 visualization for teaching
- load a pre-trained Keras or TensorFlow model and explore it visually
- build an educational demo explaining how a CNN predicts results
- embed an interactive deep learning model viewer in a web page

## When to choose
- you want interactive, browser-based 3D visualization of a neural network's structure and intermediate inferences
- your model comes from TensorFlow, Keras, or TensorFlow.js and can be preprocessed with the TensorSpace converter
- you are building educational or demo content about deep learning models

## When to avoid
- you need general-purpose plotting or charting rather than neural network visualization
- you need a maintained, actively developed tool - the last release was in 2022
- you work with frameworks other than TensorFlow/Keras/TF.js, or need server-side or non-browser visualization
- you need to train models or run heavy inference - TensorSpace is for visualization only

## Facets
- artifact type: framework
- maturity: maintenance
- function: data-visualization, deep-learning, machine-learning, graphics
- domain: deep-learning, machine-learning, data-visualization, web-development, frontend
- platform: browser, cross-platform
- tags: neural-networks, 3d-visualization, tensorflow, keras, tensorflowjs, threejs, model-visualization, interactive, nodejs

## Member repositories
- tensorspace-team/tensorspace (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:11.960419+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:01:45.322201+00:00, confidence not recorded.
  - readme: https://github.com/tensorspace-team/tensorspace (fetched 2026-08-28T04:09:11.960419+00:00, sha 0177ce641e98)
  - homepage: https://tensorspace.org (fetched 2026-08-29T08:55:47.576485+00:00, sha 5d75408d8ae9)
  - site_page: https://tensorspace.org/html/docs/startIntro.html (fetched 2026-08-29T08:55:47.583856+00:00, sha 03607487fd89)
  - site_page: https://tensorspace.org/html/docs/startHello.html (fetched 2026-08-29T08:55:47.586113+00:00, sha 49d865777e18)
  - registry_npm: https://registry.npmjs.org/tensorspace (fetched 2026-08-29T08:55:47.588336+00:00, sha bc776279342b)
- Data as of 2026-08-30T08:39:29.467469+00:00.