ERNIE-Layout-Pytorch / tokenizer_config.json
Norm's picture
Upload tokenizer_config.json
fac39e5
raw
history blame
No virus
432 Bytes
{
"do_tokenize_postprocess": false,
"sep_token": "[SEP]",
"cls_token": "[CLS]",
"unk_token": "[UNK]",
"pad_token": "[PAD]",
"mask_token": "[MASK]",
"do_lower_case": true,
"model_max_length": 512,
"vocab_file": "/home/ysocr/data/pretrain/ernie-layoutx-base-uncased/torch_version/vocab.txt",
"sentencepiece_model_file": "/home/ysocr/data/pretrain/ernie-layoutx-base-uncased/torch_version/sentencepiece.bpe.model"
}