# alexandre01/deepsvg

[NeurIPS 2020] Official code for the paper "DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation". Includes a PyTorch library for deep learning with SVG data.

Repository: https://github.com/alexandre01/deepsvg
Canonical: https://ross.abutalabs.com/products/deepsvg
Homepage: https://www.reshot.ai
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: deep-learning, pytorch, svg, svg-animations, transformer, library, python, deep-svg, machine-learning, autoencoder, sketches, sketch-rnn, svg-vae
Last push: 2024-08-26T21:18:03+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2233, "days_push": 737, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1164, forks 115 (observed 2026-08-28T04:03:49.836051+00:00)

## What it is
Official PyTorch code for the NeurIPS 2020 paper 'DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation'. It provides a library for deep learning with SVG data, training code, pretrained models, the SVG-Icons8 dataset, and a GUI demo for vector graphics animation.

## Use cases
- generate vector graphics with deep learning
- animate SVG icons with a generative model
- convert SVG files into differentiable PyTorch tensors
- train a hierarchical generative network on SVG data
- experiment with SVG autoencoders and VAEs
- use the SVG-Icons8 dataset for sketch or icon modeling

## When to choose
- you need a research-grade PyTorch library for SVG deep learning
- you want to reproduce or build on the DeepSVG paper
- you need a dataset of SVG icons for generative modeling
- you want to interpolate or animate between vector graphics

## When to avoid
- you need a production-ready SVG rendering or editing tool
- you want a maintained library with active support for new PyTorch versions
- you need general-purpose vector graphics software rather than ML research code

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, graphics, animation, data-generation
- domain: machine-learning, deep-learning, graphics, computer-vision
- platform: python
- tags: svg, pytorch, generative-model, autoencoder, transformer, vector-graphics, research-code, neurips-2020, dataset

## Member repositories
- alexandre01/deepsvg (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:49.836051+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 2026-08-30T06:29:35.534002+00:00, confidence not recorded.
  - readme: https://github.com/alexandre01/deepsvg (fetched 2026-08-28T04:03:49.836051+00:00, sha 8d83e71f90cd)
  - homepage: https://www.reshot.ai (fetched 2026-08-29T12:35:28.491965+00:00, sha c2d0374ce255)
- Data as of 2026-08-30T08:39:29.467469+00:00.