initial release

README.md

@@ -0,0 +1,26 @@
+---
+language:
+- "zh"
+tags:
+- "chinese"
+- "token-classification"
+- "pos"
+- "wikipedia"
+license: "apache-2.0"
+pipeline_tag: "token-classification"
+---
+
+# chinese-roberta-large-upos
+
+## Model Description
+
+This is a BERT model pre-trained on Chinese Wikipedia texts (both simplified and traditional) for POS-tagging, derived from [chinese-roberta-wwm-ext-large](https://huggingface.co/hfl/chinese-roberta-wwm-ext-large). Every word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech).
+
+## How to Use
+
+```py
+from transformers import AutoTokenizer,AutoModelForTokenClassification
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/chinese-roberta-large-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/chinese-roberta-large-upos")
+```
+

config.json

@@ -0,0 +1,128 @@
+{
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "directionality": "bidi",
+  "eos_token_id": 2,
+  "finetuning_task": "ner",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "ADJ",
+    "1": "ADP",
+    "2": "ADV",
+    "3": "AUX",
+    "4": "B-ADJ",
+    "5": "B-ADP",
+    "6": "B-ADV",
+    "7": "B-AUX",
+    "8": "B-CCONJ",
+    "9": "B-DET",
+    "10": "B-NOUN",
+    "11": "B-NUM",
+    "12": "B-PART",
+    "13": "B-PRON",
+    "14": "B-PROPN",
+    "15": "B-PUNCT",
+    "16": "B-SYM",
+    "17": "B-VERB",
+    "18": "B-X",
+    "19": "CCONJ",
+    "20": "DET",
+    "21": "I-ADJ",
+    "22": "I-ADP",
+    "23": "I-ADV",
+    "24": "I-AUX",
+    "25": "I-CCONJ",
+    "26": "I-DET",
+    "27": "I-NOUN",
+    "28": "I-NUM",
+    "29": "I-PART",
+    "30": "I-PRON",
+    "31": "I-PROPN",
+    "32": "I-PUNCT",
+    "33": "I-SYM",
+    "34": "I-VERB",
+    "35": "I-X",
+    "36": "NOUN",
+    "37": "NUM",
+    "38": "PART",
+    "39": "PRON",
+    "40": "PROPN",
+    "41": "PUNCT",
+    "42": "SYM",
+    "43": "VERB",
+    "44": "X"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "ADJ": 0,
+    "ADP": 1,
+    "ADV": 2,
+    "AUX": 3,
+    "B-ADJ": 4,
+    "B-ADP": 5,
+    "B-ADV": 6,
+    "B-AUX": 7,
+    "B-CCONJ": 8,
+    "B-DET": 9,
+    "B-NOUN": 10,
+    "B-NUM": 11,
+    "B-PART": 12,
+    "B-PRON": 13,
+    "B-PROPN": 14,
+    "B-PUNCT": 15,
+    "B-SYM": 16,
+    "B-VERB": 17,
+    "B-X": 18,
+    "CCONJ": 19,
+    "DET": 20,
+    "I-ADJ": 21,
+    "I-ADP": 22,
+    "I-ADV": 23,
+    "I-AUX": 24,
+    "I-CCONJ": 25,
+    "I-DET": 26,
+    "I-NOUN": 27,
+    "I-NUM": 28,
+    "I-PART": 29,
+    "I-PRON": 30,
+    "I-PROPN": 31,
+    "I-PUNCT": 32,
+    "I-SYM": 33,
+    "I-VERB": 34,
+    "I-X": 35,
+    "NOUN": 36,
+    "NUM": 37,
+    "PART": 38,
+    "PRON": 39,
+    "PROPN": 40,
+    "PUNCT": 41,
+    "SYM": 42,
+    "VERB": 43,
+    "X": 44
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "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",
+  "torch_dtype": "float32",
+  "transformers_version": "4.9.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 21128
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f040a4aa16f618a52787f20459e476a4aaebd981ade4bb04db77db2adb241f5
+size 1298231217

special_tokens_map.json

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

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, "tokenizer_class": "BertTokenizer"}