Transformers-Tutorials icon indicating copy to clipboard operation
Transformers-Tutorials copied to clipboard

Published 20 hours ago • verified publisher icon NielsRogge

Tokenizer

Open dongxuewang-123 opened this issue 2 years ago • 1 comments

I am training layoutlmv2 token VQA model consist words more than 512. but somehow only 512 words are only feeding to the model. so, other words are not training. any way by which I can train the whole document?

dongxuewang-123 avatar Sep 10 '21 13:09 dongxuewang-123

See #23

NielsRogge avatar Sep 13 '21 10:09 NielsRogge