Can kosmos-2 be finetuning with paires of Chinese text and image?

microsoft / unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

https://aka.ms/GeneralAI

MIT License

19.08k stars 2.43k forks source link

Can kosmos-2 be finetuning with paires of Chinese text and image? #1548

Open KDD2018 opened 1 month ago

KDD2018 commented 1 month ago

I am interested in Kosmos-2 and appreciate it. I want finetune kosmos-2 on my dataset which included images and Chinese texts, but I couldn't do it successfully. So I tried to call Kosmos2Processor on a text, I found it could not be decoded.