Clarification on JSON Lines Dataset for Multi-Task Fine-Tuning of Florence-2

#78
by mariaac - opened

Hi everyone,

I came across the notebook discussing how to fine-tune Florence-2 for Object Detection, and I have a question regarding the structure of the JSON Lines dataset when fine-tuning for multiple tasks.

Specifically, how should the dataset be formatted if I want to fine-tune for more than one task?

Should the prefix field be a list of task string IDs, while the suffix field contains a list of strings that represent the answers for each task? For example, would the following structure be correct?

{
  "prefix": ["<OD>", "<OCR>"],
  "suffix": [
    "ace of hearts<loc_345><loc_315><loc_582><loc_721>2 of hearts<loc_709><loc_115><loc_888><loc_509>3 of hearts<loc_529><loc_228><loc_735><loc_613>4 of hearts<loc_98><loc_421><loc_415><loc_845>",
    "answer_for_ocr"
  ]
}

Additionally, is there a guide available on how to format datasets for each task?

I appreciate any guidance on this!

Thank you!

Hi @merve :) Could you help me out with this?

@mariaac looking into it

Hi @Andyrasika ,

What about finetuning for OCR with region?
Should the dataset be formatted in a similar way to the OD dataset?

{"image": "path/to/image.png", "prefix": "<OCR_WITH_REGION>", "suffix": "US<loc_895><loc_505><loc_965><loc_555>Hugging Face<loc_647><loc_164><loc_747><loc_216>"}

I haven't found any example for finetuning for this task

Sign up or log in to comment