DonutInvoiceCzechV3 / tokenizer_config.json
TomasFAV's picture
End of training
b0135c3 verified
{
"backend": "tokenizers",
"bos_token": "<s>",
"cls_token": "<s>",
"eos_token": "</s>",
"extra_special_tokens": [
"<s_cust_tax_id>",
"<s_issue_date>",
"</s_total>",
"</s_issue_date>",
"</s_const_symbol>",
"</s>",
"</s_taxable_supply_date>",
"<s_invoice_number>",
"<s_variable_symbol>",
"</s_payment_type>",
"</s_bank_account_number>",
"<s_supp_register_id>",
"</s_bic>",
"<s_bank_account_number>",
"<s_total>",
"</s_supp_register_id>",
"<s_payment_type>",
"</s_iban>",
"<s_due_date>",
"</s_cust_register_id>",
"</s_invoice_number>",
"<s_bic>",
"<s_supp_tax_id>",
"<s_taxable_supply_date>",
"</s_due_date>",
"</s_cust_tax_id>",
"<s_cord-v2>",
"</s_supp_tax_id>",
"</s_variable_symbol>",
"<s_cust_register_id>",
"<s_const_symbol>",
"<s_iban>"
],
"from_slow": true,
"is_local": false,
"mask_token": "<mask>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"processor_class": "DonutProcessor",
"sep_token": "</s>",
"sp_model_kwargs": {},
"tokenizer_class": "TokenizersBackend",
"unk_token": "<unk>"
}