initial release

README.md

@@ -1,3 +1,43 @@
 ---
-license: apache-2.0
+language:
+- "lzh"
+tags:
+- "classical chinese"
+- "literary chinese"
+- "ancient chinese"
+- "token-classification"
+- "pos"
+- "dependency-parsing"
+datasets:
+- "universal_dependencies"
+license: "apache-2.0"
+pipeline_tag: "token-classification"
+widget:
+- text: "子曰學而時習之不亦説乎有朋自遠方來不亦樂乎人不知而不慍不亦君子乎"
 ---
+
+# bert-ancient-chinese-base-upos
+
+## Model Description
+
+This is a BERT model pre-trained on Classical Chinese texts for POS-tagging and dependency-parsing, derived from [bert-ancient-chinese](https://huggingface.co/Jihuai/bert-ancient-chinese). Every word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech) and [FEATS](https://universaldependencies.org/u/feat/).
+
+## How to Use
+
+```py
+from transformers import AutoTokenizer,AutoModelForTokenClassification
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/bert-ancient-chinese-base-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/bert-ancient-chinese-base-upos")
+```
+
+or
+
+```py
+import esupar
+nlp=esupar.load("KoichiYasuoka/bert-ancient-chinese-base-upos")
+```
+
+## See Also
+
+[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa/DeBERTa models
+

config.json

@@ -0,0 +1,203 @@
+{
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "ADP",
+    "1": "ADP|Degree=Equ",
+    "2": "ADV",
+    "3": "ADV|AdvType=Cau",
+    "4": "ADV|AdvType=Deg|Degree=Cmp",
+    "5": "ADV|AdvType=Deg|Degree=Pos",
+    "6": "ADV|AdvType=Deg|Degree=Sup",
+    "7": "ADV|AdvType=Tim",
+    "8": "ADV|AdvType=Tim|Aspect=Perf",
+    "9": "ADV|AdvType=Tim|Tense=Fut",
+    "10": "ADV|AdvType=Tim|Tense=Past",
+    "11": "ADV|AdvType=Tim|Tense=Pres",
+    "12": "ADV|Degree=Equ|VerbForm=Conv",
+    "13": "ADV|Degree=Pos|VerbForm=Conv",
+    "14": "ADV|Polarity=Neg",
+    "15": "ADV|Polarity=Neg|VerbForm=Conv",
+    "16": "ADV|VerbForm=Conv",
+    "17": "AUX|Mood=Des",
+    "18": "AUX|Mood=Nec",
+    "19": "AUX|Mood=Pot",
+    "20": "AUX|VerbType=Cop",
+    "21": "AUX|Voice=Pass",
+    "22": "B-ADV|VerbForm=Conv",
+    "23": "B-NOUN",
+    "24": "B-NOUN|Case=Loc",
+    "25": "B-NOUN|Case=Tem",
+    "26": "B-NUM",
+    "27": "B-NUM|NumType=Ord",
+    "28": "B-PROPN|Case=Loc|NameType=Geo",
+    "29": "B-PROPN|Case=Loc|NameType=Nat",
+    "30": "B-PROPN|NameType=Giv",
+    "31": "B-PROPN|NameType=Prs",
+    "32": "B-PROPN|NameType=Sur",
+    "33": "B-VERB",
+    "34": "B-VERB|Degree=Equ",
+    "35": "B-VERB|Degree=Pos",
+    "36": "B-VERB|VerbForm=Part",
+    "37": "CCONJ",
+    "38": "I-ADV|VerbForm=Conv",
+    "39": "I-NOUN",
+    "40": "I-NOUN|Case=Loc",
+    "41": "I-NOUN|Case=Tem",
+    "42": "I-NUM",
+    "43": "I-NUM|NumType=Ord",
+    "44": "I-PROPN|Case=Loc|NameType=Geo",
+    "45": "I-PROPN|Case=Loc|NameType=Nat",
+    "46": "I-PROPN|NameType=Giv",
+    "47": "I-PROPN|NameType=Prs",
+    "48": "I-PROPN|NameType=Sur",
+    "49": "I-VERB",
+    "50": "I-VERB|Degree=Equ",
+    "51": "I-VERB|Degree=Pos",
+    "52": "I-VERB|VerbForm=Part",
+    "53": "INTJ",
+    "54": "NOUN",
+    "55": "NOUN|Case=Loc",
+    "56": "NOUN|Case=Tem",
+    "57": "NOUN|NounType=Clf",
+    "58": "NUM",
+    "59": "NUM|NumType=Ord",
+    "60": "PART",
+    "61": "PRON|Person=1|PronType=Prs",
+    "62": "PRON|Person=2|PronType=Prs",
+    "63": "PRON|Person=3|PronType=Prs",
+    "64": "PRON|PronType=Dem",
+    "65": "PRON|PronType=Int",
+    "66": "PRON|PronType=Prs",
+    "67": "PRON|PronType=Prs|Reflex=Yes",
+    "68": "PROPN|Case=Loc|NameType=Geo",
+    "69": "PROPN|Case=Loc|NameType=Nat",
+    "70": "PROPN|NameType=Giv",
+    "71": "PROPN|NameType=Prs",
+    "72": "PROPN|NameType=Sur",
+    "73": "SCONJ",
+    "74": "SYM",
+    "75": "VERB",
+    "76": "VERB|Degree=Equ",
+    "77": "VERB|Degree=Equ|VerbForm=Part",
+    "78": "VERB|Degree=Pos",
+    "79": "VERB|Degree=Pos|VerbForm=Part",
+    "80": "VERB|Polarity=Neg",
+    "81": "VERB|Polarity=Neg|VerbForm=Part",
+    "82": "VERB|VerbForm=Part"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "ADP": 0,
+    "ADP|Degree=Equ": 1,
+    "ADV": 2,
+    "ADV|AdvType=Cau": 3,
+    "ADV|AdvType=Deg|Degree=Cmp": 4,
+    "ADV|AdvType=Deg|Degree=Pos": 5,
+    "ADV|AdvType=Deg|Degree=Sup": 6,
+    "ADV|AdvType=Tim": 7,
+    "ADV|AdvType=Tim|Aspect=Perf": 8,
+    "ADV|AdvType=Tim|Tense=Fut": 9,
+    "ADV|AdvType=Tim|Tense=Past": 10,
+    "ADV|AdvType=Tim|Tense=Pres": 11,
+    "ADV|Degree=Equ|VerbForm=Conv": 12,
+    "ADV|Degree=Pos|VerbForm=Conv": 13,
+    "ADV|Polarity=Neg": 14,
+    "ADV|Polarity=Neg|VerbForm=Conv": 15,
+    "ADV|VerbForm=Conv": 16,
+    "AUX|Mood=Des": 17,
+    "AUX|Mood=Nec": 18,
+    "AUX|Mood=Pot": 19,
+    "AUX|VerbType=Cop": 20,
+    "AUX|Voice=Pass": 21,
+    "B-ADV|VerbForm=Conv": 22,
+    "B-NOUN": 23,
+    "B-NOUN|Case=Loc": 24,
+    "B-NOUN|Case=Tem": 25,
+    "B-NUM": 26,
+    "B-NUM|NumType=Ord": 27,
+    "B-PROPN|Case=Loc|NameType=Geo": 28,
+    "B-PROPN|Case=Loc|NameType=Nat": 29,
+    "B-PROPN|NameType=Giv": 30,
+    "B-PROPN|NameType=Prs": 31,
+    "B-PROPN|NameType=Sur": 32,
+    "B-VERB": 33,
+    "B-VERB|Degree=Equ": 34,
+    "B-VERB|Degree=Pos": 35,
+    "B-VERB|VerbForm=Part": 36,
+    "CCONJ": 37,
+    "I-ADV|VerbForm=Conv": 38,
+    "I-NOUN": 39,
+    "I-NOUN|Case=Loc": 40,
+    "I-NOUN|Case=Tem": 41,
+    "I-NUM": 42,
+    "I-NUM|NumType=Ord": 43,
+    "I-PROPN|Case=Loc|NameType=Geo": 44,
+    "I-PROPN|Case=Loc|NameType=Nat": 45,
+    "I-PROPN|NameType=Giv": 46,
+    "I-PROPN|NameType=Prs": 47,
+    "I-PROPN|NameType=Sur": 48,
+    "I-VERB": 49,
+    "I-VERB|Degree=Equ": 50,
+    "I-VERB|Degree=Pos": 51,
+    "I-VERB|VerbForm=Part": 52,
+    "INTJ": 53,
+    "NOUN": 54,
+    "NOUN|Case=Loc": 55,
+    "NOUN|Case=Tem": 56,
+    "NOUN|NounType=Clf": 57,
+    "NUM": 58,
+    "NUM|NumType=Ord": 59,
+    "PART": 60,
+    "PRON|Person=1|PronType=Prs": 61,
+    "PRON|Person=2|PronType=Prs": 62,
+    "PRON|Person=3|PronType=Prs": 63,
+    "PRON|PronType=Dem": 64,
+    "PRON|PronType=Int": 65,
+    "PRON|PronType=Prs": 66,
+    "PRON|PronType=Prs|Reflex=Yes": 67,
+    "PROPN|Case=Loc|NameType=Geo": 68,
+    "PROPN|Case=Loc|NameType=Nat": 69,
+    "PROPN|NameType=Giv": 70,
+    "PROPN|NameType=Prs": 71,
+    "PROPN|NameType=Sur": 72,
+    "SCONJ": 73,
+    "SYM": 74,
+    "VERB": 75,
+    "VERB|Degree=Equ": 76,
+    "VERB|Degree=Equ|VerbForm=Part": 77,
+    "VERB|Degree=Pos": 78,
+    "VERB|Degree=Pos|VerbForm=Part": 79,
+    "VERB|Polarity=Neg": 80,
+    "VERB|Polarity=Neg|VerbForm=Part": 81,
+    "VERB|VerbForm=Part": 82
+  },
+  "layer_norm_eps": 1e-12,
+  "lstm_dropout_prob": 0.5,
+  "lstm_embedding_size": 768,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.19.4",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 38208
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef7f89faff627b146dcee894d57c7b42eff4b9c39148913f29ac82a593e940ff
+size 459503729

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

supar.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01e1401f9d855b637616a6fb098ac9725b0174cdc6b5f8a9d65a4eeb97143aef
+size 509755365

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}