Skip to content

vincent8264/yoda_captioning_model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Yoda captioning model

Master Yoda tell you what's in the image!

Try it at
https://yodacaptioner.up.railway.app/

More details about the model can be found at:
https://huggingface.co/vkao8264/blip-yoda-captioning

Repo description

This repo contains the scripts to generate the captions used to fine-tune the BLIP image-to-text model, and the training script.
For the web app itself, the repo is here

In order to run the scripts, you will need to have the COCO captions dataset on your machine

captions_generator.py generates 30000 Yoda-style captions using phi3 LLM imported from lmphi.py
blip_training.py then pairs the captions with images and fine-tunes the BLIP model

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages