model-ocr-fp-train-banghid-v1 / tokenizer_config.json
emisilab's picture
Upload 12 files
9aa8ce5 verified
raw
history blame
7.92 kB
{
"added_tokens_decoder": {
"0": {
"content": "<s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<pad>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "</s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"3": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57521": {
"content": "<mask>",
"lstrip": true,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"57522": {
"content": "<sep/>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"57523": {
"content": "<s_iitcdip>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57524": {
"content": "<s_synthdog>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57525": {
"content": "<s_referensi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57526": {
"content": "</s_referensi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57527": {
"content": "<s_ppn>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57528": {
"content": "</s_ppn>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57529": {
"content": "<s_hargaTotal>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57530": {
"content": "</s_hargaTotal>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57531": {
"content": "<s_dpp>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57532": {
"content": "</s_dpp>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57533": {
"content": "<s_diskon>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57534": {
"content": "</s_diskon>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57535": {
"content": "<s_TanggalFakturPajak>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57536": {
"content": "</s_TanggalFakturPajak>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57537": {
"content": "<s_PPnBM>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57538": {
"content": "</s_PPnBM>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57539": {
"content": "<s_NoNPWPlawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57540": {
"content": "</s_NoNPWPlawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57541": {
"content": "<s_NoFakturPajak>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57542": {
"content": "</s_NoFakturPajak>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57543": {
"content": "<s_NamaPenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57544": {
"content": "</s_NamaPenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57545": {
"content": "<s_NPWPpenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57546": {
"content": "</s_NPWPpenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57547": {
"content": "<s_NPWPlawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57548": {
"content": "</s_NPWPlawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57549": {
"content": "<s_AlamatPenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57550": {
"content": "</s_AlamatPenjual>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57551": {
"content": "<s_AlamatLawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"57552": {
"content": "</s_AlamatLawanTransaksi>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"additional_special_tokens": [
"<s_referensi>",
"</s_referensi>",
"<s_ppn>",
"</s_ppn>",
"<s_hargaTotal>",
"</s_hargaTotal>",
"<s_dpp>",
"</s_dpp>",
"<s_diskon>",
"</s_diskon>",
"<s_TanggalFakturPajak>",
"</s_TanggalFakturPajak>",
"<s_PPnBM>",
"</s_PPnBM>",
"<s_NoNPWPlawanTransaksi>",
"</s_NoNPWPlawanTransaksi>",
"<s_NoFakturPajak>",
"</s_NoFakturPajak>",
"<s_NamaPenjual>",
"</s_NamaPenjual>",
"<s_NPWPpenjual>",
"</s_NPWPpenjual>",
"<s_NPWPlawanTransaksi>",
"</s_NPWPlawanTransaksi>",
"<s_AlamatPenjual>",
"</s_AlamatPenjual>",
"<s_AlamatLawanTransaksi>",
"</s_AlamatLawanTransaksi>",
"<s>",
"</s>"
],
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"cls_token": "<s>",
"eos_token": "</s>",
"mask_token": "<mask>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"processor_class": "DonutProcessor",
"sep_token": "</s>",
"sp_model_kwargs": {},
"tokenizer_class": "XLMRobertaTokenizer",
"unk_token": "<unk>"
}