Pzhvk/Captioneer

An in-depth comparison of three image captioning models built with PyTorch. This repository contains the full pipeline for preprocessing the Flickr30k dataset, training three distinct architectures (LSTM, LSTM w/ Attention, Transformer w/ DistilBERT), and performing a comprehensive evaluation using BLEU, METEOR, CIDEr, and SPICE.

16
/ 100
Experimental
No Package No Dependents
Maintenance 6 / 25
Adoption 1 / 25
Maturity 9 / 25
Community 0 / 25

How are scores calculated?

Stars

1

Forks

Language

Jupyter Notebook

License

MIT

Category

image-captioning

Last pushed

Oct 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Pzhvk/Captioneer"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.