joycaption
llama-joycaption-beta-one-hf-llavallama-joycaption-alpha-two-hf-llavallama-joycaption-beta-one-hf-llava-mmproj-ggufllama-joycaption-beta-one-hf-llava-GGUFllama-joycaption-beta-one-hf-llava-FP8-Dynamicllama-joycaption-beta-one-hf-llava-nf4llama-joycaption-beta-onellama-joycaption-beta-one-hf-llava-i1-GGUF
joy-captioning-20250328b
Work In Progress
I'm still going back through my data to add in the URLs.
joy-captioning-20250408aThis is the dataset used to do the initial training for JoyCaption Beta One (https://huggingface.co/fancyfeast/llama-joycaption-beta-one-hf-llava), before post-training.
Contents
Most of the dataset focusses on descriptions and captions for images, with a smaller subset covering general VQA tasks.
Some of the questions and answers are human written, some are automated, some are machine written. The is_human column is True when the answer text is human written.
WARNING… See the full description on the dataset page: https://huggingface.co/datasets/fancyfeast/joy-captioning-20250408a.Yu_Lexuan_Archer_JoyCaption_512pxFeng_Long_Mao_Zijun_JoyCaption_1024pxjoycaption_humanMisuzu-Gundou_JoyCaption