initial release

README.md

@@ -0,0 +1,47 @@
+---
+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-large-japanese-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-large-japanese-aozora](https://huggingface.co/KoichiYasuoka/deberta-large-japanese-aozora). Every short-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-large-japanese-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/deberta-large-japanese-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-large-japanese-upos")
+print(nlp("国境の長いトンネルを抜けると雪国であった。"))
+```
+
+## See Also
+
+[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa/DeBERTa models
+

config.json

@@ -0,0 +1,397 @@
+{
+  "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": 1024,
+  "id2label": {
+    "0": "ADJ",
+    "1": "ADP",
+    "2": "ADV",
+    "3": "AUX",
+    "4": "AUX+AUX",
+    "5": "B-ADJ",
+    "6": "B-ADJ+AUX",
+    "7": "B-ADP",
+    "8": "B-ADV",
+    "9": "B-ADV+ADV",
+    "10": "B-AUX",
+    "11": "B-CCONJ",
+    "12": "B-DET",
+    "13": "B-INTJ",
+    "14": "B-NOUN",
+    "15": "B-NOUN+VERB",
+    "16": "B-NUM",
+    "17": "B-PART",
+    "18": "B-PRON",
+    "19": "B-PRON+ADP",
+    "20": "B-PROPN",
+    "21": "B-VERB",
+    "22": "B-VERB+AUX",
+    "23": "B-X",
+    "24": "CCONJ",
+    "25": "DET",
+    "26": "I-ADJ",
+    "27": "I-ADJ+AUX",
+    "28": "I-ADP",
+    "29": "I-ADV",
+    "30": "I-ADV+ADV",
+    "31": "I-AUX",
+    "32": "I-CCONJ",
+    "33": "I-DET",
+    "34": "I-INTJ",
+    "35": "I-NOUN",
+    "36": "I-NOUN+VERB",
+    "37": "I-NUM",
+    "38": "I-PART",
+    "39": "I-PRON",
+    "40": "I-PRON+ADP",
+    "41": "I-PROPN",
+    "42": "I-VERB",
+    "43": "I-VERB+AUX",
+    "44": "I-X",
+    "45": "INTJ",
+    "46": "NOUN",
+    "47": "NOUN+AUX",
+    "48": "NOUN+NOUN",
+    "49": "NUM",
+    "50": "NUM+NOUN",
+    "51": "NUM+PART",
+    "52": "PART",
+    "53": "PRON",
+    "54": "PROPN",
+    "55": "PUNCT",
+    "56": "SCONJ",
+    "57": "SYM",
+    "58": "VERB",
+    "59": "VERB+SCONJ"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "ADJ": 0,
+    "ADP": 1,
+    "ADV": 2,
+    "AUX": 3,
+    "AUX+AUX": 4,
+    "B-ADJ": 5,
+    "B-ADJ+AUX": 6,
+    "B-ADP": 7,
+    "B-ADV": 8,
+    "B-ADV+ADV": 9,
+    "B-AUX": 10,
+    "B-CCONJ": 11,
+    "B-DET": 12,
+    "B-INTJ": 13,
+    "B-NOUN": 14,
+    "B-NOUN+VERB": 15,
+    "B-NUM": 16,
+    "B-PART": 17,
+    "B-PRON": 18,
+    "B-PRON+ADP": 19,
+    "B-PROPN": 20,
+    "B-VERB": 21,
+    "B-VERB+AUX": 22,
+    "B-X": 23,
+    "CCONJ": 24,
+    "DET": 25,
+    "I-ADJ": 26,
+    "I-ADJ+AUX": 27,
+    "I-ADP": 28,
+    "I-ADV": 29,
+    "I-ADV+ADV": 30,
+    "I-AUX": 31,
+    "I-CCONJ": 32,
+    "I-DET": 33,
+    "I-INTJ": 34,
+    "I-NOUN": 35,
+    "I-NOUN+VERB": 36,
+    "I-NUM": 37,
+    "I-PART": 38,
+    "I-PRON": 39,
+    "I-PRON+ADP": 40,
+    "I-PROPN": 41,
+    "I-VERB": 42,
+    "I-VERB+AUX": 43,
+    "I-X": 44,
+    "INTJ": 45,
+    "NOUN": 46,
+    "NOUN+AUX": 47,
+    "NOUN+NOUN": 48,
+    "NUM": 49,
+    "NUM+NOUN": 50,
+    "NUM+PART": 51,
+    "PART": 52,
+    "PRON": 53,
+    "PROPN": 54,
+    "PUNCT": 55,
+    "SCONJ": 56,
+    "SYM": 57,
+    "VERB": 58,
+    "VERB+SCONJ": 59
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 1024,
+  "pos_att_type": null,
+  "position_biased_input": true,
+  "relative_attention": false,
+  "task_specific_params": {
+    "upos_multiword": {
+      "ADJ+AUX": {
+        "\u5e73\u304b\u306a\u308a": [
+          "\u5e73\u304b",
+          "\u306a\u308a"
+        ]
+      },
+      "ADV+ADV": {
+        "\u672a\u66fe\u3066": [
+          "\u672a",
+          "\u66fe\u3066"
+        ]
+      },
+      "AUX+AUX": {
+        "\u305b\u3057\u3081": [
+          "\u305b",
+          "\u3057\u3081"
+        ],
+        "\u306a\u3089\u305a": [
+          "\u306a\u3089",
+          "\u305a"
+        ]
+      },
+      "NOUN+AUX": {
+        "\u7570\u306a\u3089": [
+          "\u7570",
+          "\u306a\u3089"
+        ]
+      },
+      "NOUN+NOUN": {
+        "\u8fb2\u5de5": [
+          "\u8fb2",
+          "\u5de5"
+        ]
+      },
+      "NOUN+VERB": {
+        "\u4e16\u51fa\u3057": [
+          "\u4e16\u51fa",
+          "\u3057"
+        ],
+        "\u58d3\u4f0f\u3057": [
+          "\u58d3\u4f0f",
+          "\u3057"
+        ],
+        "\u653e\u8a31\u3057": [
+          "\u653e\u8a31",
+          "\u3057"
+        ],
+        "\u767e\u51fa\u3057": [
+          "\u767e\u51fa",
+          "\u3057"
+        ],
+        "\u7a81\u8d77\u3057": [
+          "\u7a81\u8d77",
+          "\u3057"
+        ],
+        "\u8655\u51b3\u3057": [
+          "\u8655\u51b3",
+          "\u3057"
+        ],
+        "\u8f29\u51fa\u3057": [
+          "\u8f29\u51fa",
+          "\u3057"
+        ],
+        "\u9192\u8d77\u3057": [
+          "\u9192\u8d77",
+          "\u3057"
+        ]
+      },
+      "NUM+NOUN": {
+        "\u4e00\u4eba": [
+          "\u4e00",
+          "\u4eba"
+        ]
+      },
+      "NUM+PART": {
+        "\u4e00\u4eba": [
+          "\u4e00",
+          "\u4eba"
+        ]
+      },
+      "PRON+ADP": {
+        "\u4f55\u3093\u305e": [
+          "\u4f55\u3093",
+          "\u305e"
+        ]
+      },
+      "VERB+AUX": {
+        "\u3059\u3079\u304b\u3089": [
+          "\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u3059\u3079\u304d": [
+          "\u3059",
+          "\u3079\u304d"
+        ],
+        "\u306a\u3059\u3079\u304b\u3089": [
+          "\u306a\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u4e00\u81f4\u305b": [
+          "\u4e00\u81f4",
+          "\u305b"
+        ],
+        "\u4ed8\u3059\u3079\u3051": [
+          "\u4ed8\u3059",
+          "\u3079\u3051"
+        ],
+        "\u4f3c\u3059\u3079\u304b\u3089": [
+          "\u4f3c\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u4fc3\u3055\u3093": [
+          "\u4fc3\u3055",
+          "\u3093"
+        ],
+        "\u5047\u3059\u3079\u304b\u3089": [
+          "\u5047\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u51b3\u3059\u3079\u3057": [
+          "\u51b3\u3059",
+          "\u3079\u3057"
+        ],
+        "\u5c5e\u305b\u3057\u3081": [
+          "\u5c5e\u305b",
+          "\u3057\u3081"
+        ],
+        "\u6210\u3059\u3079\u304b\u3089": [
+          "\u6210\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u653e\u6d41\u3057": [
+          "\u653e\u6d41",
+          "\u3057"
+        ],
+        "\u6575\u3059\u3079\u304b\u3089": [
+          "\u6575\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u6575\u3059\u3079\u3051": [
+          "\u6575\u3059",
+          "\u3079\u3051"
+        ],
+        "\u66f8\u3059\u3079\u304b\u3089": [
+          "\u66f8\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u66f8\u3059\u3079\u304d": [
+          "\u66f8\u3059",
+          "\u3079\u304d"
+        ],
+        "\u670d\u3059\u3079\u304d": [
+          "\u670d\u3059",
+          "\u3079\u304d"
+        ],
+        "\u6bba\u3055\u3093": [
+          "\u6bba\u3055",
+          "\u3093"
+        ],
+        "\u6df7\u5408\u305b": [
+          "\u6df7\u5408",
+          "\u305b"
+        ],
+        "\u7232\u3055\u3093": [
+          "\u7232\u3055",
+          "\u3093"
+        ],
+        "\u7232\u3059\u3079\u304b\u3089": [
+          "\u7232\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u7232\u3059\u3079\u304d": [
+          "\u7232\u3059",
+          "\u3079\u304d"
+        ],
+        "\u7232\u3059\u3079\u3057": [
+          "\u7232\u3059",
+          "\u3079\u3057"
+        ],
+        "\u72af\u3059\u3079\u304d": [
+          "\u72af\u3059",
+          "\u3079\u304d"
+        ],
+        "\u751f\u305c\u3093": [
+          "\u751f\u305c",
+          "\u3093"
+        ],
+        "\u76ee\u3059\u3079\u304d": [
+          "\u76ee\u3059",
+          "\u3079\u304d"
+        ],
+        "\u7f6e\u304b\u3056\u308a": [
+          "\u7f6e\u304b",
+          "\u3056\u308a"
+        ],
+        "\u7f70\u3059\u3079\u304b\u3089": [
+          "\u7f70\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u7f70\u3059\u3079\u3057": [
+          "\u7f70\u3059",
+          "\u3079\u3057"
+        ],
+        "\u8207\u3059\u3079\u304b\u3089": [
+          "\u8207\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u8655\u3059\u3079\u304b\u3089": [
+          "\u8655\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u89e3\u3059\u3079\u304b\u3089": [
+          "\u89e3\u3059",
+          "\u3079\u304b\u3089"
+        ],
+        "\u8aa6\u3059\u3079\u304f": [
+          "\u8aa6\u3059",
+          "\u3079\u304f"
+        ],
+        "\u8d77\u3055\u3093": [
+          "\u8d77\u3055",
+          "\u3093"
+        ],
+        "\u8d77\u3059\u3079\u304d": [
+          "\u8d77\u3059",
+          "\u3079\u304d"
+        ],
+        "\u8d77\u3059\u3079\u3057": [
+          "\u8d77\u3059",
+          "\u3079\u3057"
+        ]
+      },
+      "VERB+SCONJ": {
+        "\u7acb\u3066": [
+          "\u7acb",
+          "\u3066"
+        ]
+      }
+    }
+  },
+  "tokenizer_class": "DebertaV2TokenizerFast",
+  "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 sha256:6e9c374541f717f9eabb6064a95a89766bb56a75bb226f7c8d4f45453338f77b
+size 1342797683

special_tokens_map.json

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

spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01ba4719c80b6fe911b091a7c05124b64eeece964e09c058ef8f9805daca546b
+size 1

supar.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08261816ec7da11116a145547b46d89ea5e58b8fbcb3732cb99613ac6f38bc53
+size 1391495275

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "split_by_punct": true, "keep_accents": true, "model_max_length": 512, "tokenizer_class": "DebertaV2TokenizerFast"}