# rmokady/CLIP_prefix_caption

Simple image captioning model

Repository: https://github.com/rmokady/CLIP_prefix_caption
Canonical: https://ross.abutalabs.com/products/clip_prefix_caption
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2024-06-09T13:42:21+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1812, "days_push": 815, "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 1423, forks 225 (observed 2026-08-28T04:04:41.169858+00:00)

## What it is
Official implementation of ClipCap, a CLIP-based image captioning model that maps CLIP image encodings to a GPT-2 prefix to generate captions. It requires only images and captions for training and achieves near state-of-the-art results with fast training.

## Use cases
- generate captions for images automatically
- caption photos without object annotations
- fine-tune an image captioning model on my own image-caption dataset
- describe images with a pretrained vision-language model
- reproduce ClipCap paper results on COCO or Conceptual Captions
- add image-to-text captioning to a Python project

## When to choose
- you need image captioning without extra object-annotation supervision
- you want a lightweight, fast-training captioning model
- you want to build on CLIP embeddings for image-to-text generation
- you want a Colab-ready inference notebook

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, computer-vision, image-processing
- domain: machine-learning, computer-vision, deep-learning
- platform: python
- tags: image-captioning, clip, gpt-2, clipcap, research-code, jupyter-notebook, natural-language-processing

## Member repositories
- rmokady/CLIP_prefix_caption (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:41.169858+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 2026-08-30T04:37:39.673913+00:00, confidence not recorded.
  - readme: https://github.com/rmokady/CLIP_prefix_caption (fetched 2026-08-28T04:04:41.169858+00:00, sha 403c2539e7fb)
- Data as of 2026-08-30T08:39:29.467469+00:00.