initial release

README.md

@@ -1,3 +1,41 @@
----
-license: cc-by-sa-4.0
----
+---
+language:
+- "uk"
+tags:
+- "ukrainian"
+- "token-classification"
+- "pos"
+- "ubertext"
+- "dependency-parsing"
+datasets:
+- "universal_dependencies"
+license: "cc-by-sa-4.0"
+pipeline_tag: "token-classification"
+---
+
+# roberta-base-ukrainian-upos
+
+## Model Description
+
+This is a RoBERTa model pre-trained on Thai Wikipedia texts for POS-tagging and dependency-parsing, derived from [roberta-base-ukrainian](https://huggingface.co/KoichiYasuoka/roberta-base-ukrainian). Every 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/roberta-base-ukrainian-upos")
+model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/roberta-base-ukrainian-upos")
+```
+
+or
+
+```
+import esupar
+nlp=esupar.load("KoichiYasuoka/roberta-base-ukrainian-upos")
+```
+
+## See Also
+
+[esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa models
+

config.json

@@ -0,0 +1,223 @@
+{
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "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-INTJ",
+    "11": "B-NOUN",
+    "12": "B-NOUN+NUM",
+    "13": "B-NUM",
+    "14": "B-NUM+NOUN",
+    "15": "B-PART",
+    "16": "B-PRON",
+    "17": "B-PROPN",
+    "18": "B-PUNCT",
+    "19": "B-SCONJ",
+    "20": "B-SYM",
+    "21": "B-VERB",
+    "22": "B-VERB+ADV",
+    "23": "B-VERB+PRON",
+    "24": "B-X",
+    "25": "CCONJ",
+    "26": "DET",
+    "27": "I-ADJ",
+    "28": "I-ADP",
+    "29": "I-ADV",
+    "30": "I-AUX",
+    "31": "I-CCONJ",
+    "32": "I-DET",
+    "33": "I-INTJ",
+    "34": "I-NOUN",
+    "35": "I-NOUN+NUM",
+    "36": "I-NUM",
+    "37": "I-NUM+NOUN",
+    "38": "I-PART",
+    "39": "I-PRON",
+    "40": "I-PROPN",
+    "41": "I-PUNCT",
+    "42": "I-SCONJ",
+    "43": "I-SYM",
+    "44": "I-VERB",
+    "45": "I-VERB+ADV",
+    "46": "I-VERB+PRON",
+    "47": "I-X",
+    "48": "INTJ",
+    "49": "NOUN",
+    "50": "NOUN+NUM",
+    "51": "NUM",
+    "52": "NUM+NOUN",
+    "53": "PART",
+    "54": "PRON",
+    "55": "PROPN",
+    "56": "PUNCT",
+    "57": "SCONJ",
+    "58": "SYM",
+    "59": "VERB",
+    "60": "VERB+ADV",
+    "61": "VERB+PRON",
+    "62": "X"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "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-INTJ": 10,
+    "B-NOUN": 11,
+    "B-NOUN+NUM": 12,
+    "B-NUM": 13,
+    "B-NUM+NOUN": 14,
+    "B-PART": 15,
+    "B-PRON": 16,
+    "B-PROPN": 17,
+    "B-PUNCT": 18,
+    "B-SCONJ": 19,
+    "B-SYM": 20,
+    "B-VERB": 21,
+    "B-VERB+ADV": 22,
+    "B-VERB+PRON": 23,
+    "B-X": 24,
+    "CCONJ": 25,
+    "DET": 26,
+    "I-ADJ": 27,
+    "I-ADP": 28,
+    "I-ADV": 29,
+    "I-AUX": 30,
+    "I-CCONJ": 31,
+    "I-DET": 32,
+    "I-INTJ": 33,
+    "I-NOUN": 34,
+    "I-NOUN+NUM": 35,
+    "I-NUM": 36,
+    "I-NUM+NOUN": 37,
+    "I-PART": 38,
+    "I-PRON": 39,
+    "I-PROPN": 40,
+    "I-PUNCT": 41,
+    "I-SCONJ": 42,
+    "I-SYM": 43,
+    "I-VERB": 44,
+    "I-VERB+ADV": 45,
+    "I-VERB+PRON": 46,
+    "I-X": 47,
+    "INTJ": 48,
+    "NOUN": 49,
+    "NOUN+NUM": 50,
+    "NUM": 51,
+    "NUM+NOUN": 52,
+    "PART": 53,
+    "PRON": 54,
+    "PROPN": 55,
+    "PUNCT": 56,
+    "SCONJ": 57,
+    "SYM": 58,
+    "VERB": 59,
+    "VERB+ADV": 60,
+    "VERB+PRON": 61,
+    "X": 62
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "task_specific_params": {
+    "upos_multiword": {
+      "NUM+NOUN": {
+        "\u043f\u0456\u0432'\u044f\u0440\u0434\u0430": [
+          "\u043f\u0456\u0432",
+          "\u044f\u0440\u0434\u0430"
+        ],
+        "\u043f\u0456\u0432\u0440\u043e\u043a\u0443": [
+          "\u043f\u0456\u0432",
+          "\u0440\u043e\u043a\u0443"
+        ],
+        "\u043f\u0456\u0432\u2019\u044f\u0433\u043d\u044f\u0442\u0438": [
+          "\u043f\u0456\u0432",
+          "\u044f\u0433\u043d\u044f\u0442\u0438"
+        ],
+        "\u043f\u0456\u0432\u2019\u044f\u0437\u0438\u043a\u0430": [
+          "\u043f\u0456\u0432",
+          "\u044f\u0437\u0438\u043a\u0430"
+        ],
+        "\u043f\u0456\u0432\u2019\u044f\u0449\u0438\u043a\u0430": [
+          "\u043f\u0456\u0432",
+          "\u044f\u0449\u0438\u043a\u0430"
+        ]
+      },
+      "VERB+ADV": {
+        "\u043d\u0456\u0434\u0435": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u0434\u0435"
+        ],
+        "\u043d\u0456\u0437\u0432\u0456\u0434\u043a\u0438": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u0437\u0432\u0456\u0434\u043a\u0438"
+        ],
+        "\u043d\u0456\u043a\u043e\u043b\u0438": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u043a\u043e\u043b\u0438"
+        ],
+        "\u043d\u0456\u043a\u0443\u0434\u0438": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u043a\u0443\u0434\u0438"
+        ],
+        "\u043d\u0456\u044f\u043a": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u044f\u043a"
+        ]
+      },
+      "VERB+PRON": {
+        "\u043d\u0456\u043a\u0438\u043c": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u043a\u0438\u043c"
+        ],
+        "\u043d\u0456\u043a\u043e\u0433\u043e": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u043a\u043e\u0433\u043e"
+        ],
+        "\u043d\u0456\u043a\u043e\u043c\u0443": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u043a\u043e\u043c\u0443"
+        ],
+        "\u043d\u0456\u0447\u0438\u043c": [
+          "\u043d\u0435\u043c\u0430\u0454",
+          "\u0447\u0438\u043c"
+        ]
+      }
+    }
+  },
+  "tokenizer_class": "BertTokenizerFast",
+  "torch_dtype": "float32",
+  "transformers_version": "4.14.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:735ce76697fea88bd807ed7cf843d11a691c2fb81c9637a5f42f9f6848ceb7ac
+size 434241393

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 sha256:16e19fd58bfb82b801655327a1b4ded01417a453e643091ea6baef06bf5ad6b4
+size 487153701

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": false, "lowercase": false, "never_split": ["[CLS]", "[PAD]", "[SEP]", "[UNK]", "[MASK]"], "do_basic_tokenize": true, "model_max_length": 512, "tokenizer_class": "BertTokenizerFast"}