# julrog/nn_vis

A project for processing neural networks and rendering to gain insights on the architecture and parameters of a model through a decluttered representation.

Repository: https://github.com/julrog/nn_vis
Canonical: https://ross.abutalabs.com/products/nn_vis
Language: Python
License: MIT
License Family: permissive
Topics: neural-networks, visualization, rendering-engine, opengl, neural-network, virtual-reality, vr
Last push: 2024-01-06T15:03:12+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": 2374, "days_push": 970, "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 1174, forks 190 (observed 2026-08-28T04:03:51.986942+00:00)

## What it is
A Python tool that processes trained neural networks and renders them as decluttered 3D visualizations of their architecture and parameters, using importance estimation, edge bundling, and ray tracing. It includes a desktop visualization tool with optional VR support and a Docker image.

## Use cases
- visualize neural network architecture in 3d
- inspect parameters of a trained model visually
- understand which parts of a network are important
- explore deep learning models in virtual reality
- create decluttered renderings of large neural networks
- compare neural network architectures visually

## When to choose
- you want an interactive 3D/VR visualization of a trained PyTorch-style model's architecture and learned importance
- you need publication-quality renderings of network structure
- you want to explore model internals beyond static diagrams

## When to avoid
- you need training dashboards or scalar metrics like loss curves (use TensorBoard or W&B)
- you need a lightweight in-notebook plotting library
- you need support for arbitrary model formats without preprocessing

## Facets
- artifact type: application
- maturity: maintenance
- function: data-visualization, machine-learning, graphics, developer-tools
- domain: deep-learning, data-visualization, machine-learning, graphics
- platform: python, cross-platform
- tags: neural-networks, opengl, virtual-reality, 3d-visualization, model-interpretability, rendering-engine, docker

## Member repositories
- julrog/nn_vis (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:51.986942+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 2026-08-30T06:28:17.825913+00:00, confidence not recorded.
  - readme: https://github.com/julrog/nn_vis (fetched 2026-08-28T04:03:51.986942+00:00, sha 351f184b2f14)
- Data as of 2026-08-30T08:39:29.467469+00:00.