microsoft

/

layoutlmv2-large-uncased

like 8

Hi everyone,

I hope you all are doing well, and good in health. I am working on LayoutLMv2 and LayoutLMv3, and for this reason, I have been working on these topics for the last 2 months, our main objective is to extract the key feature from the Invoices using the LayoutLM models where I have the already labeled dataset for this, but here I am using the pre-trained model try to train on the custom dataset. and then after training, I want to store the weights of that model, and then load these weights to test the model. currently, I am facing some difficulties in that phase, and I am confused can we use that commercially or not if not why? if yes or there is a possible way to do that then what are the steps for that?

Thanks, Hassan javed