initial release

README.md

@@ -0,0 +1,49 @@
+---
+language:
+- "ja"
+tags:
+- "japanese"
+- "token-classification"
+- "pos"
+- "dependency-parsing"
+datasets:
+- "universal_dependencies"
+license: "cc-by-sa-4.0"
+pipeline_tag: "token-classification"
+widget:
+- text: "国境の長いトンネルを抜けると雪国であった。"
+---
+
+# deberta-base-japanese-unidic-luw-upos
+
+## Model Description
+
+This is a DeBERTa(V2) model pre-trained on 青空文庫 texts for POS-tagging and dependency-parsing, derived from [deberta-base-japanese-unidic](https://huggingface.co/KoichiYasuoka/deberta-base-japanese-unidic). Every long-unit-word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech).
+
+## How to Use
+
+```py
+import torch
+from transformers import AutoTokenizer,AutoModelForTokenClassification
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/deberta-base-japanese-unidic-luw-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/deberta-base-japanese-unidic-luw-upos")
+s="国境の長いトンネルを抜けると雪国であった。"
+t=tokenizer.tokenize(s)
+p=[model.config.id2label[q] for q in torch.argmax(model(tokenizer.encode(s,return_tensors="pt"))["logits"],dim=2)[0].tolist()[1:-1]]
+print(list(zip(t,p)))
+```
+
+or
+
+```py
+import esupar
+nlp=esupar.load("KoichiYasuoka/deberta-base-japanese-unidic-luw-upos")
+print(nlp("国境の長いトンネルを抜けると雪国であった。"))
+```
+
+[fugashi](https://pypi.org/project/fugashi), [unidic-lite](https://pypi.org/project/unidic-lite) and [pytokenizations](https://pypi.org/project/pytokenizations) are required.
+
+## See Also
+
+[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa/DeBERTa models
+

config.json

@@ -0,0 +1,299 @@
+{
+  "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": "ADJ",
+    "1": "ADJ+VERB",
+    "2": "ADP",
+    "3": "ADP+ADP",
+    "4": "ADP+ADP+VERB",
+    "5": "ADP+VERB",
+    "6": "ADV",
+    "7": "AUX",
+    "8": "B-ADJ",
+    "9": "B-ADJ+VERB",
+    "10": "B-ADP",
+    "11": "B-ADP+ADJ",
+    "12": "B-ADP+NOUN+ADP",
+    "13": "B-ADV",
+    "14": "B-AUX",
+    "15": "B-AUX+AUX",
+    "16": "B-AUX+NOUN",
+    "17": "B-CCONJ",
+    "18": "B-INTJ",
+    "19": "B-NOUN",
+    "20": "B-NOUN+ADP",
+    "21": "B-NOUN+NOUN",
+    "22": "B-NUM",
+    "23": "B-PART",
+    "24": "B-PRON",
+    "25": "B-PROPN",
+    "26": "B-PROPN+ADP",
+    "27": "B-PUNCT",
+    "28": "B-SCONJ",
+    "29": "B-SYM",
+    "30": "B-VERB",
+    "31": "B-VERB+AUX",
+    "32": "B-VERB+SCONJ",
+    "33": "B-X",
+    "34": "CCONJ",
+    "35": "DET",
+    "36": "DET+NOUN",
+    "37": "I-ADJ",
+    "38": "I-ADJ+VERB",
+    "39": "I-ADP",
+    "40": "I-ADP+ADJ",
+    "41": "I-ADP+NOUN+ADP",
+    "42": "I-ADV",
+    "43": "I-AUX",
+    "44": "I-AUX+AUX",
+    "45": "I-AUX+NOUN",
+    "46": "I-CCONJ",
+    "47": "I-INTJ",
+    "48": "I-NOUN",
+    "49": "I-NOUN+ADP",
+    "50": "I-NOUN+NOUN",
+    "51": "I-NUM",
+    "52": "I-PART",
+    "53": "I-PRON",
+    "54": "I-PROPN",
+    "55": "I-PROPN+ADP",
+    "56": "I-PUNCT",
+    "57": "I-SCONJ",
+    "58": "I-SYM",
+    "59": "I-VERB",
+    "60": "I-VERB+AUX",
+    "61": "I-VERB+SCONJ",
+    "62": "I-X",
+    "63": "INTJ",
+    "64": "NOUN",
+    "65": "NUM",
+    "66": "PART",
+    "67": "PRON",
+    "68": "PROPN",
+    "69": "PUNCT",
+    "70": "SCONJ",
+    "71": "SYM",
+    "72": "VERB",
+    "73": "VERB+AUX",
+    "74": "X"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "ADJ": 0,
+    "ADJ+VERB": 1,
+    "ADP": 2,
+    "ADP+ADP": 3,
+    "ADP+ADP+VERB": 4,
+    "ADP+VERB": 5,
+    "ADV": 6,
+    "AUX": 7,
+    "B-ADJ": 8,
+    "B-ADJ+VERB": 9,
+    "B-ADP": 10,
+    "B-ADP+ADJ": 11,
+    "B-ADP+NOUN+ADP": 12,
+    "B-ADV": 13,
+    "B-AUX": 14,
+    "B-AUX+AUX": 15,
+    "B-AUX+NOUN": 16,
+    "B-CCONJ": 17,
+    "B-INTJ": 18,
+    "B-NOUN": 19,
+    "B-NOUN+ADP": 20,
+    "B-NOUN+NOUN": 21,
+    "B-NUM": 22,
+    "B-PART": 23,
+    "B-PRON": 24,
+    "B-PROPN": 25,
+    "B-PROPN+ADP": 26,
+    "B-PUNCT": 27,
+    "B-SCONJ": 28,
+    "B-SYM": 29,
+    "B-VERB": 30,
+    "B-VERB+AUX": 31,
+    "B-VERB+SCONJ": 32,
+    "B-X": 33,
+    "CCONJ": 34,
+    "DET": 35,
+    "DET+NOUN": 36,
+    "I-ADJ": 37,
+    "I-ADJ+VERB": 38,
+    "I-ADP": 39,
+    "I-ADP+ADJ": 40,
+    "I-ADP+NOUN+ADP": 41,
+    "I-ADV": 42,
+    "I-AUX": 43,
+    "I-AUX+AUX": 44,
+    "I-AUX+NOUN": 45,
+    "I-CCONJ": 46,
+    "I-INTJ": 47,
+    "I-NOUN": 48,
+    "I-NOUN+ADP": 49,
+    "I-NOUN+NOUN": 50,
+    "I-NUM": 51,
+    "I-PART": 52,
+    "I-PRON": 53,
+    "I-PROPN": 54,
+    "I-PROPN+ADP": 55,
+    "I-PUNCT": 56,
+    "I-SCONJ": 57,
+    "I-SYM": 58,
+    "I-VERB": 59,
+    "I-VERB+AUX": 60,
+    "I-VERB+SCONJ": 61,
+    "I-X": 62,
+    "INTJ": 63,
+    "NOUN": 64,
+    "NUM": 65,
+    "PART": 66,
+    "PRON": 67,
+    "PROPN": 68,
+    "PUNCT": 69,
+    "SCONJ": 70,
+    "SYM": 71,
+    "VERB": 72,
+    "VERB+AUX": 73,
+    "X": 74
+  },
+  "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": null,
+  "position_biased_input": true,
+  "relative_attention": false,
+  "task_specific_params": {
+    "upos_multiword": {
+      "ADJ+VERB": {
+        "\u7121\u304f\u3057": [
+          "\u7121\u304f",
+          "\u3057"
+        ],
+        "\u7a0b\u306a\u304f\u3057": [
+          "\u7a0b\u306a\u304f",
+          "\u3057"
+        ]
+      },
+      "ADP+ADJ": {
+        "\u306f\u3089\u30fb\u3080\u3046\u3093\u4f5c\u54c1\u5171\u901a": [
+          "\u306f",
+          "\u3089\u30fb\u3080\u3046\u3093\u4f5c\u54c1\u5171\u901a"
+        ]
+      },
+      "ADP+ADP": {
+        "\u3068\u3082": [
+          "\u3068",
+          "\u3082"
+        ]
+      },
+      "ADP+ADP+VERB": {
+        "\u3068\u3082\u3057": [
+          "\u3068",
+          "\u3082",
+          "\u3057"
+        ]
+      },
+      "ADP+NOUN+ADP": {
+        "\u306e\u307f\u305d\u306e": [
+          "\u306e",
+          "\u307f\u305d",
+          "\u306e"
+        ]
+      },
+      "ADP+VERB": {
+        "\u3067\u304d": [
+          "\u3067",
+          "\u304d"
+        ],
+        "\u3067\u3057": [
+          "\u3067",
+          "\u3057"
+        ],
+        "\u306f\u3057": [
+          "\u306f",
+          "\u3057"
+        ]
+      },
+      "AUX+AUX": {
+        "\u30c1\u30e3\u30c3\u30bf": [
+          "\u30c1\u30e3\u30c3",
+          "\u30bf"
+        ]
+      },
+      "AUX+NOUN": {
+        "\u306a\u304a\u5024\u6bb5": [
+          "\u306a",
+          "\u304a\u5024\u6bb5"
+        ],
+        "\u306a\u304a\u5e03\u65bd": [
+          "\u306a",
+          "\u304a\u5e03\u65bd"
+        ]
+      },
+      "DET+NOUN": {
+        "\u3053\u306e\u9803": [
+          "\u3053\u306e",
+          "\u9803"
+        ]
+      },
+      "NOUN+ADP": {
+        "\u3046\u305d\u306e": [
+          "\u3046\u305d",
+          "\u306e"
+        ]
+      },
+      "NOUN+NOUN": {
+        "\u9ce5\u53d6\u770c\u6559\u80b2\u59d4\u54e1\u4f1a\u793e\u4f1a\u6559\u80b2\u59d4\u54e1": [
+          "\u9ce5\u53d6\u770c\u6559\u80b2\u59d4\u54e1\u4f1a",
+          "\u793e\u4f1a\u6559\u80b2\u59d4\u54e1"
+        ]
+      },
+      "PROPN+ADP": {
+        "\u3061\u307b\u306e": [
+          "\u3061\u307b",
+          "\u306e"
+        ],
+        "\u3088\u3090\u3053\u306e": [
+          "\u3088\u3090\u3053",
+          "\u306e"
+        ]
+      },
+      "VERB+AUX": {
+        "\u305f\u3063\u305f": [
+          "\u305f\u3063",
+          "\u305f"
+        ],
+        "\u5c0f\u5c4b\u639b\u3051\u3055\u305b": [
+          "\u5c0f\u5c4b\u639b\u3051\u3055",
+          "\u305b"
+        ]
+      },
+      "VERB+SCONJ": {
+        "\u8cb8\u5207\u3063\u3066": [
+          "\u8cb8\u5207\u3063",
+          "\u3066"
+        ]
+      }
+    }
+  },
+  "tokenizer_class": "BertJapaneseTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.19.2",
+  "type_vocab_size": 0,
+  "vocab_size": 32000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6f2f96ca2cb845c7d5bb259b9f437ecf4a20dd87253d0e7e98010d4d244cce3
+size 440402547

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 sha256:769e3144ddf44f3ab4b191aafa5aa8e5a695dc99cf1e8538da6aa4f8f267aa23
+size 488904747

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "do_lower_case": false, "do_word_tokenize": true, "do_subword_tokenize": true, "word_tokenizer_type": "mecab", "subword_tokenizer_type": "wordpiece", "never_split": ["[CLS]", "[PAD]", "[SEP]", "[UNK]", "[MASK]"], "mecab_kwargs": {"mecab_dic": "unidic_lite"}, "model_max_length": 512, "tokenizer_class": "BertJapaneseTokenizer"}