ArmiNouri

Recreating DocVQA results for LayoutLMv2

11

[Related issue on the unilm repo](https://github.com/microsoft/unilm/issues/501). I'm trying to recreate the results reported in the [LayoutLMv2 paper, Table 6, row 7](https://arxiv.org/pdf/2012.14740.pdf). Following [this example](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/LayoutLMv2/DocVQA/Fine_tuning_LayoutLMv2ForQuestionAnswering_on_DocVQA.ipynb), I've fine-tuned the base model with...

# epochs for finetuning LayoutLMv2 on DocVQA

4

I'm trying to recreate the results reported in the [LayoutLMv2 paper, Table 6, row 7](https://arxiv.org/pdf/2012.14740.pdf). Following [this example](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/LayoutLMv2/DocVQA/Fine_tuning_LayoutLMv2ForQuestionAnswering_on_DocVQA.ipynb), I've fine-tuned the base model with DocVQA train set for 20 epochs....