search

microsoft / unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
https://aka.ms/GeneralAI
MIT License
19.62k stars 2.5k forks source link

Pretrain Huggingface LayoutLM model with different language #253

Open Sharathmk99 opened 3 years ago

Sharathmk99 commented 3 years ago

Describe the bug LayoutLM

The problem arises when using: LayoutLM model is available in huggingface transformers now. Is it possible to pretrain the LayoutLM model with custom corpus(German language) with huggingface transformers now?

It will be really good if we can pretrain on own forms or invoice for different languages.

Looking forward for your response.

wolfshow commented 3 years ago

@Sharathmk99 We are working on the multilingual LayoutLM. Please stay tuned.

Sharathmk99 commented 3 years ago

@wolfshow any date when multilingual model will be available?

Bunoviske commented 3 years ago

Hey! Great work and intuition on the LayoutLM pre-training approach. Indeed, having a multilingual model would be also interesting to me. Any estimation on when will be released? Looking forward to it, thanks!

yellowishee commented 3 years ago

multilingual或者中文版的模型开源了的时候，踢我一下，谢谢！🙏

knitemblazor commented 3 years ago

everyone looking for multilingual support kindly go through this project derived from layoutlm https://github.com/knitemblazor/Multilingual_LayoutLM

nissansz commented 2 years ago

hi, where to get or pretrain the models for Japanese, Korean, etc.? steve8000818@gmail.com