lismin

Training data

2

Thank you for your work. Can you disclose how the training data was obtained, whether it was synthesized or scraped?

latex detection dataset

1

https://github.com/microsoft/ArxivFormula

Florence-2: How to add custom tokens during fine-tuning training?

1

This is the official way to add: ``` # processing_florence2.py tokens_to_add = {'additional_special_tokens': \ tokenizer.additional_special_tokens + \ ['', '', '', ''] + \ [f'' for x in range(1000)] + \...