Datasets:

MixTex

/

Pseudo-Latex-ZhEn-1

like 4

thanks

ctex default, the fake data generation code is a bit ugly and is generated by chatgpt3.5. I will organize it and upload it, or you can directly give the method introduced in the article to AI to let it regenerate. If you can collect some handwriting fonts and formula fonts, and use the fonts to compile PDF from tex to make a data set and share it in open source, I will be very grateful.

ctex 默认，伪造数据生成代码有点丑陋是chatgpt3.5生成的，我会整理一下上传，或者可以直接把文章中介绍的方法给AI让他重新生成， 如果你能够收集一些手写字体和公式字体，根据字体从tex编译pdf制作数据集并开源分享，我将感激不尽。