Florence-2-OCRJP-remake4

This model is a fine-tuned version of microsoft/Florence-2-base-ft on EtashGuha/JapaneseDocQA dataset. It achieves the following results on the evaluation set:

• Loss: 0.76

Add new model layer OCR_JP for OCR for Japanese sentences.

Model description

Please use new task prompt "OCR_JP" Instead of "OCR", so you can get results of OCR task in Japanese words.

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

Training results

Framework versions

• Transformers 4.46.3
• Pytorch 2.3.0
• Datasets 2.20.0
• Tokenizers 0.20.3