roberta_wikiann_conll_finetuned_cv / tokenizer_config.json
mrfirdauss's picture
Upload 7 files
f044ff7 verified
raw
history blame contribute delete
No virus
2.68 kB
{
"add_prefix_space": true,
"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
},
"50264": {
"content": "<mask>",
"lstrip": true,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"cls_token": "<s>",
"eos_token": "</s>",
"errors": "replace",
"id2label": {
"0": "O",
"1": "B-NAME",
"2": "I-NAME",
"3": "B-NATION",
"4": "I-NATION",
"5": "B-EMAIL",
"6": "I-EMAIL",
"7": "B-URL",
"8": "I-URL",
"9": "B-CAMPUS",
"10": "I-CAMPUS",
"11": "B-MAJOR",
"12": "I-MAJOR",
"13": "B-COMPANY",
"14": "I-COMPANY",
"15": "B-DESIGNATION",
"16": "I-DESIGNATION",
"17": "B-GPA",
"18": "I-GPA",
"19": "B-PHONE NUMBER",
"20": "I-PHONE NUMBER",
"21": "B-ACHIEVEMENT",
"22": "I-ACHIEVEMENT",
"23": "B-EXPERIENCES DESC",
"24": "I-EXPERIENCES DESC",
"25": "B-SKILLS",
"26": "I-SKILLS",
"27": "B-PROJECTS",
"28": "I-PROJECTS"
},
"ignore_mismatched_sizes": true,
"label2id": {
"B-ACHIEVEMENT": 21,
"B-CAMPUS": 9,
"B-COMPANY": 13,
"B-DESIGNATION": 15,
"B-EMAIL": 5,
"B-EXPERIENCES DESC": 23,
"B-GPA": 17,
"B-MAJOR": 11,
"B-NAME": 1,
"B-NATION": 3,
"B-PHONE NUMBER": 19,
"B-PROJECTS": 27,
"B-SKILLS": 25,
"B-URL": 7,
"I-ACHIEVEMENT": 22,
"I-CAMPUS": 10,
"I-COMPANY": 14,
"I-DESIGNATION": 16,
"I-EMAIL": 6,
"I-EXPERIENCES DESC": 24,
"I-GPA": 18,
"I-MAJOR": 12,
"I-NAME": 2,
"I-NATION": 4,
"I-PHONE NUMBER": 20,
"I-PROJECTS": 28,
"I-SKILLS": 26,
"I-URL": 8,
"O": 0
},
"mask_token": "<mask>",
"max_length": 512,
"model_max_length": 512,
"num_labels": 29,
"pad_token": "<pad>",
"sep_token": "</s>",
"stride": 0,
"tokenizer_class": "RobertaTokenizer",
"trim_offsets": true,
"truncation_side": "right",
"truncation_strategy": "longest_first",
"unk_token": "<unk>"
}