deberta-base-ainu-upos / config.json
KoichiYasuoka's picture
model improved
1337e98
raw
history blame
11.5 kB
{
"architectures": [
"DebertaV2ForTokenClassification"
],
"attention_probs_dropout_prob": 0.1,
"bos_token_id": 0,
"eos_token_id": 2,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"id2label": {
"0": "ADP",
"1": "ADV",
"2": "AUX",
"3": "AUX+NOUN",
"4": "B-ADP",
"5": "B-ADP+VERB+NOUN",
"6": "B-ADV",
"7": "B-AUX",
"8": "B-AUX+PART",
"9": "B-CCONJ",
"10": "B-DET",
"11": "B-DET+NOUN",
"12": "B-DET+VERB",
"13": "B-INFR.EV",
"14": "B-INTJ",
"15": "B-NOUN",
"16": "B-NOUN+ADP",
"17": "B-NOUN+ADP+NOUN",
"18": "B-NOUN+ADV",
"19": "B-NOUN+NOUN",
"20": "B-NOUN+VERB",
"21": "B-NUM",
"22": "B-NUM+NOUN",
"23": "B-PART",
"24": "B-PART+NOUN",
"25": "B-PART+VERB",
"26": "B-PRON",
"27": "B-PROPN",
"28": "B-PUNCT",
"29": "B-SCONJ",
"30": "B-SCONJ+ADV",
"31": "B-VERB",
"32": "B-VERB+AUX",
"33": "B-VERB+NOUN",
"34": "B-VERB+PART",
"35": "B-VERB+SCONJ",
"36": "B-VERT",
"37": "B-X",
"38": "CCONJ",
"39": "DET",
"40": "DET+NOUN",
"41": "DET+SCONJ+VERB",
"42": "I-ADP",
"43": "I-ADP+VERB+NOUN",
"44": "I-ADV",
"45": "I-AUX",
"46": "I-AUX+PART",
"47": "I-CCONJ",
"48": "I-DET",
"49": "I-DET+NOUN",
"50": "I-DET+VERB",
"51": "I-INFR.EV",
"52": "I-INTJ",
"53": "I-NOUN",
"54": "I-NOUN+ADP",
"55": "I-NOUN+ADP+NOUN",
"56": "I-NOUN+ADV",
"57": "I-NOUN+NOUN",
"58": "I-NOUN+VERB",
"59": "I-NUM",
"60": "I-NUM+NOUN",
"61": "I-PART",
"62": "I-PART+NOUN",
"63": "I-PART+VERB",
"64": "I-PRON",
"65": "I-PROPN",
"66": "I-PUNCT",
"67": "I-SCONJ",
"68": "I-SCONJ+ADV",
"69": "I-VERB",
"70": "I-VERB+AUX",
"71": "I-VERB+NOUN",
"72": "I-VERB+PART",
"73": "I-VERB+SCONJ",
"74": "I-VERT",
"75": "I-X",
"76": "INTJ",
"77": "NOUN",
"78": "NOUN+ADP",
"79": "NOUN+NOUN",
"80": "NOUN+VERB",
"81": "NUM",
"82": "NUM+VERB+NOUN",
"83": "PART",
"84": "PART+NOUN",
"85": "PART+VERB",
"86": "PROPN",
"87": "PUNCT",
"88": "SCONJ",
"89": "SYM",
"90": "VERB",
"91": "VERB+AUX",
"92": "VERB+NOUN",
"93": "VERB+PART",
"94": "VERB+VERB",
"95": "VERT",
"96": "X"
},
"initializer_range": 0.02,
"intermediate_size": 3072,
"label2id": {
"ADP": 0,
"ADV": 1,
"AUX": 2,
"AUX+NOUN": 3,
"B-ADP": 4,
"B-ADP+VERB+NOUN": 5,
"B-ADV": 6,
"B-AUX": 7,
"B-AUX+PART": 8,
"B-CCONJ": 9,
"B-DET": 10,
"B-DET+NOUN": 11,
"B-DET+VERB": 12,
"B-INFR.EV": 13,
"B-INTJ": 14,
"B-NOUN": 15,
"B-NOUN+ADP": 16,
"B-NOUN+ADP+NOUN": 17,
"B-NOUN+ADV": 18,
"B-NOUN+NOUN": 19,
"B-NOUN+VERB": 20,
"B-NUM": 21,
"B-NUM+NOUN": 22,
"B-PART": 23,
"B-PART+NOUN": 24,
"B-PART+VERB": 25,
"B-PRON": 26,
"B-PROPN": 27,
"B-PUNCT": 28,
"B-SCONJ": 29,
"B-SCONJ+ADV": 30,
"B-VERB": 31,
"B-VERB+AUX": 32,
"B-VERB+NOUN": 33,
"B-VERB+PART": 34,
"B-VERB+SCONJ": 35,
"B-VERT": 36,
"B-X": 37,
"CCONJ": 38,
"DET": 39,
"DET+NOUN": 40,
"DET+SCONJ+VERB": 41,
"I-ADP": 42,
"I-ADP+VERB+NOUN": 43,
"I-ADV": 44,
"I-AUX": 45,
"I-AUX+PART": 46,
"I-CCONJ": 47,
"I-DET": 48,
"I-DET+NOUN": 49,
"I-DET+VERB": 50,
"I-INFR.EV": 51,
"I-INTJ": 52,
"I-NOUN": 53,
"I-NOUN+ADP": 54,
"I-NOUN+ADP+NOUN": 55,
"I-NOUN+ADV": 56,
"I-NOUN+NOUN": 57,
"I-NOUN+VERB": 58,
"I-NUM": 59,
"I-NUM+NOUN": 60,
"I-PART": 61,
"I-PART+NOUN": 62,
"I-PART+VERB": 63,
"I-PRON": 64,
"I-PROPN": 65,
"I-PUNCT": 66,
"I-SCONJ": 67,
"I-SCONJ+ADV": 68,
"I-VERB": 69,
"I-VERB+AUX": 70,
"I-VERB+NOUN": 71,
"I-VERB+PART": 72,
"I-VERB+SCONJ": 73,
"I-VERT": 74,
"I-X": 75,
"INTJ": 76,
"NOUN": 77,
"NOUN+ADP": 78,
"NOUN+NOUN": 79,
"NOUN+VERB": 80,
"NUM": 81,
"NUM+VERB+NOUN": 82,
"PART": 83,
"PART+NOUN": 84,
"PART+VERB": 85,
"PROPN": 86,
"PUNCT": 87,
"SCONJ": 88,
"SYM": 89,
"VERB": 90,
"VERB+AUX": 91,
"VERB+NOUN": 92,
"VERB+PART": 93,
"VERB+VERB": 94,
"VERT": 95,
"X": 96
},
"layer_norm_eps": 1e-07,
"max_position_embeddings": 512,
"max_relative_positions": -1,
"model_type": "deberta-v2",
"num_attention_heads": 12,
"num_hidden_layers": 12,
"pad_token_id": 1,
"pooler_dropout": 0,
"pooler_hidden_act": "gelu",
"pooler_hidden_size": 768,
"pos_att_type": [
"p2c",
"c2p"
],
"position_biased_input": false,
"relative_attention": true,
"task_specific_params": {
"upos_multiword": {
"ADP+VERB+NOUN": {
"tambe": [
"ta",
"m",
"be"
]
},
"AUX+NOUN": {
"nep": [
"ne",
"p"
]
},
"AUX+PART": {
"nangonna": [
"nangon",
"na"
],
"nankonna": [
"nankon",
"na"
]
},
"DET+NOUN": {
"Tamba": [
"Tam",
"ba"
],
"Tampa": [
"Tam",
"pa"
],
"oararke": [
"oar",
"arke"
],
"oararkehe": [
"oar",
"arkehe"
],
"tanto": [
"tan",
"to"
]
},
"DET+SCONJ+VERB": {
"Newaan": [
"Ne",
"wa",
"an"
],
"newaan": [
"ne",
"wa",
"an"
]
},
"DET+VERB": {
"iyorun": [
"iyor",
"un"
]
},
"NOUN+ADP": {
"Kunneiwano": [
"Kunnei",
"wano"
],
"Orowano": [
"Oro",
"wano"
],
"Pet-samaketa": [
"Pet-samake",
"ta"
],
"Shoita": [
"Shoi",
"ta"
],
"Soita": [
"Soi",
"ta"
],
"keseta": [
"kese",
"ta"
],
"kunneywano": [
"kunney",
"wano"
],
"orowa": [
"oro",
"wa"
],
"orowano": [
"oro",
"wano"
],
"otta": [
"ot",
"ta"
],
"samaketa": [
"samake",
"ta"
],
"shoita": [
"shoi",
"ta"
],
"soyta": [
"soy",
"ta"
],
"tomta": [
"tom",
"ta"
],
"tumukeheta": [
"tumukehe",
"ta"
]
},
"NOUN+ADP+NOUN": {
"rorunpurai": [
"ror",
"un",
"purai"
],
"rorunpuray": [
"ror",
"un",
"puray"
]
},
"NOUN+ADV": {
"Tambeta ne": [
"Tambe",
"ta ne"
]
},
"NOUN+NOUN": {
"Hinakoro": [
"Hinak",
"oro"
],
"Petetoko": [
"Pet",
"etoko"
],
"hekattar": [
"hekat",
"tar"
],
"hinakoro": [
"hinak",
"oro"
],
"inaanpe": [
"inaan",
"pe"
],
"inanpe": [
"inan",
"pe"
],
"iporohoka": [
"iporoho",
"ka"
],
"kamuinish": [
"kamui",
"nish"
],
"kamuynis": [
"kamuy",
"nis"
],
"petetok": [
"pet",
"etok"
],
"petetoko": [
"pet",
"etoko"
]
},
"NOUN+VERB": {
"Omakun": [
"Omak",
"un"
],
"Orepun": [
"Orep",
"un"
],
"Shiriki": [
"Shiri",
"ki"
],
"kotankor": [
"kotan",
"kor"
],
"makun": [
"mak",
"un"
],
"repun": [
"rep",
"un"
],
"rikunruke": [
"rik",
"unruke"
],
"siriki": [
"siri",
"ki"
],
"ukakushte": [
"uka",
"kushte"
],
"ukakuste": [
"uka",
"kuste"
],
"uraikik": [
"urai",
"kik"
]
},
"NUM+NOUN": {
"Wanto": [
"Wan",
"to"
],
"hotnepa": [
"hotne",
"pa"
],
"wanpe": [
"wan",
"pe"
],
"wanto": [
"wan",
"to"
]
},
"NUM+VERB+NOUN": {
"Shineanto": [
"Shine",
"an",
"to"
],
"sineanto": [
"sine",
"an",
"to"
]
},
"PART+NOUN": {
"=anpe": [
"=an",
"pe"
],
"shichorpok": [
"shi",
"chorpok"
]
},
"PART+VERB": {
"Chirushka": [
"Chi",
"rushka"
],
"ainu-wap": [
"a",
"inu-wap"
],
"akus": [
"a",
"kus"
],
"chiki": [
"chi",
"ki"
],
"chikik": [
"chi",
"kik"
],
"eram an": [
"e",
"ram an"
],
"karapa": [
"k",
"arapa"
],
"shiokote": [
"shi",
"okote"
]
},
"SCONJ+ADV": {
"koiramno": [
"ko",
"iramno"
]
},
"VERB+AUX": {
"poppeta ashinnangoro": [
"poppeta ashin",
"nangoro"
],
"poppetaasinnankor": [
"poppetaasin",
"nankor"
],
"sattek": [
"sat",
"tek"
]
},
"VERB+NOUN": {
"Hesepa": [
"Hese",
"pa"
],
"ambe": [
"am",
"be"
],
"anpe": [
"an",
"pe"
],
"ashbe": [
"ash",
"be"
],
"aspe": [
"as",
"pe"
],
"h\u00e9sep\u00e1ha": [
"h\u00e9se",
"p\u00e1ha"
],
"kari": [
"kar",
"i"
],
"ohasiri": [
"oha",
"siri"
],
"wenpuri": [
"wen",
"puri"
]
},
"VERB+PART": {
"kari": [
"kar",
"i"
],
"rokash": [
"rok",
"ash"
],
"sapash": [
"sap",
"ash"
]
},
"VERB+SCONJ": {
"anak un": [
"an",
"ak un"
],
"anakanakne": [
"an",
"akanakne"
],
"sakno": [
"sak",
"no"
]
},
"VERB+VERB": {
"ranran": [
"ran",
"ran"
]
}
}
},
"tokenizer_class": "DebertaV2TokenizerFast",
"torch_dtype": "float32",
"transformers_version": "4.22.1",
"type_vocab_size": 0,
"vocab_size": 5092
}