# ashnkumar/sketch-code

Keras model to generate HTML code from hand-drawn website mockups. Implements an image captioning architecture to drawn source images.

Repository: https://github.com/ashnkumar/sketch-code
Canonical: https://ross.abutalabs.com/products/sketch-code
Language: Python
License: MIT
License Family: permissive
Topics: keras, tensorflow, image-processing, deep-learning, augmentation, html-generation, image-captioning
Last push: 2026-08-06T20:19:49+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 96, release rhythm 35, longevity 100
- inputs: {"age_days": 3148, "days_push": 27, "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 5143, forks 678 (observed 2026-08-28T04:09:11.176066+00:00)

## What it is
A Keras/TensorFlow deep learning model that converts hand-drawn website wireframes into HTML code using an image captioning architecture (CNN encoder, LSTM decoder). Built in 2018 as a proof-of-concept based on pix2code, it is no longer maintained and targets TensorFlow 1.x.

## Use cases
- convert hand-drawn website mockups to html
- generate html from wireframe sketches
- image captioning model for code generation
- experiment with sketch-to-code deep learning
- reproduce pix2code-style html generation

## When to choose
- studying historical image-captioning approaches to code generation
- working with wireframes closely resembling its synthetic training dataset
- learning how CNN+LSTM captioning architectures were adapted for HTML output

## When to avoid
- production sketch-to-code conversion on real-world drawings
- modern TensorFlow/Keras environments (targets TF 1.x)
- projects needing maintained or generalizable models - modern vision-language models handle this zero-shot

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, image-processing, nlp
- domain: deep-learning, machine-learning, web-development, computer-vision
- platform: python
- tags: image-captioning, html-generation, keras, tensorflow, wireframe-to-code, proof-of-concept, unmaintained

## Member repositories
- ashnkumar/sketch-code (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:11.176066+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 2026-08-29T18:01:59.115693+00:00, confidence not recorded.
  - readme: https://github.com/ashnkumar/sketch-code (fetched 2026-08-28T04:09:11.176066+00:00, sha 7fcb46c98752)
- Data as of 2026-08-30T08:39:29.467469+00:00.