How to use microsoft/xdoc-base with Transformers:
# Load model directly from transformers import AutoTokenizer, Layoutlmv1ForMaskedLM_roberta tokenizer = AutoTokenizer.from_pretrained("microsoft/xdoc-base") model = Layoutlmv1ForMaskedLM_roberta.from_pretrained("microsoft/xdoc-base")
336f749
1
{"model_max_length": 512, "bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}