model improved

config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "KoichiYasuoka/roberta-large-japanese-aozora-char",
   "architectures": [
     "RobertaForTokenClassification"
   ],
@@ -11,110 +10,108 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 1024,
   "id2label": {
-    "0": "PRON",
-    "1": "I-PART",
-    "2": "B-NOUN",
-    "3": "VERB",
-    "4": "ADJ",
-    "5": "I-SYM",
-    "6": "B-VERB",
-    "7": "AUX",
-    "8": "PROPN",
-    "9": "PUNCT",
-    "10": "I-PRON",
-    "11": "I-VERB",
-    "12": "I-X",
-    "13": "B-ADV",
-    "14": "B-ADP",
-    "15": "B-DET",
-    "16": "B-CCONJ",
-    "17": "B-X",
+    "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-NUM",
+    "13": "B-PART",
+    "14": "B-PRON",
+    "15": "B-PROPN",
+    "16": "B-PUNCT",
+    "17": "B-SCONJ",
     "18": "B-SYM",
-    "19": "CCONJ",
-    "20": "PART",
-    "21": "B-PUNCT",
-    "22": "I-NOUN",
-    "23": "I-INTJ",
-    "24": "INTJ",
-    "25": "I-ADJ",
-    "26": "I-NUM",
-    "27": "B-INTJ",
-    "28": "I-AUX",
-    "29": "I-CCONJ",
-    "30": "NUM",
-    "31": "SCONJ",
-    "32": "I-PUNCT",
-    "33": "I-SCONJ",
-    "34": "B-PART",
-    "35": "B-ADJ",
-    "36": "I-PROPN",
-    "37": "B-NUM",
-    "38": "X",
-    "39": "B-AUX",
-    "40": "B-SCONJ",
-    "41": "ADP",
-    "42": "B-PROPN",
-    "43": "NOUN",
-    "44": "I-DET",
-    "45": "SYM",
-    "46": "I-ADP",
-    "47": "I-ADV",
-    "48": "B-PRON",
-    "49": "ADV"
+    "19": "B-VERB",
+    "20": "B-X",
+    "21": "CCONJ",
+    "22": "I-ADJ",
+    "23": "I-ADP",
+    "24": "I-ADV",
+    "25": "I-AUX",
+    "26": "I-CCONJ",
+    "27": "I-DET",
+    "28": "I-INTJ",
+    "29": "I-NOUN",
+    "30": "I-NUM",
+    "31": "I-PART",
+    "32": "I-PRON",
+    "33": "I-PROPN",
+    "34": "I-PUNCT",
+    "35": "I-SCONJ",
+    "36": "I-SYM",
+    "37": "I-VERB",
+    "38": "I-X",
+    "39": "NOUN",
+    "40": "NUM",
+    "41": "PART",
+    "42": "PRON",
+    "43": "PROPN",
+    "44": "PUNCT",
+    "45": "SCONJ",
+    "46": "SYM",
+    "47": "VERB",
+    "48": "X"
   },
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "label2id": {
-    "ADJ": 4,
-    "ADP": 41,
-    "ADV": 49,
-    "AUX": 7,
-    "B-ADJ": 35,
-    "B-ADP": 14,
-    "B-ADV": 13,
-    "B-AUX": 39,
-    "B-CCONJ": 16,
-    "B-DET": 15,
-    "B-INTJ": 27,
-    "B-NOUN": 2,
-    "B-NUM": 37,
-    "B-PART": 34,
-    "B-PRON": 48,
-    "B-PROPN": 42,
-    "B-PUNCT": 21,
-    "B-SCONJ": 40,
+    "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-NUM": 12,
+    "B-PART": 13,
+    "B-PRON": 14,
+    "B-PROPN": 15,
+    "B-PUNCT": 16,
+    "B-SCONJ": 17,
     "B-SYM": 18,
-    "B-VERB": 6,
-    "B-X": 17,
-    "CCONJ": 19,
-    "I-ADJ": 25,
-    "I-ADP": 46,
-    "I-ADV": 47,
-    "I-AUX": 28,
-    "I-CCONJ": 29,
-    "I-DET": 44,
-    "I-INTJ": 23,
-    "I-NOUN": 22,
-    "I-NUM": 26,
-    "I-PART": 1,
-    "I-PRON": 10,
-    "I-PROPN": 36,
-    "I-PUNCT": 32,
-    "I-SCONJ": 33,
-    "I-SYM": 5,
-    "I-VERB": 11,
-    "I-X": 12,
-    "INTJ": 24,
-    "NOUN": 43,
-    "NUM": 30,
-    "PART": 20,
-    "PRON": 0,
-    "PROPN": 8,
-    "PUNCT": 9,
-    "SCONJ": 31,
-    "SYM": 45,
-    "VERB": 3,
-    "X": 38
+    "B-VERB": 19,
+    "B-X": 20,
+    "CCONJ": 21,
+    "I-ADJ": 22,
+    "I-ADP": 23,
+    "I-ADV": 24,
+    "I-AUX": 25,
+    "I-CCONJ": 26,
+    "I-DET": 27,
+    "I-INTJ": 28,
+    "I-NOUN": 29,
+    "I-NUM": 30,
+    "I-PART": 31,
+    "I-PRON": 32,
+    "I-PROPN": 33,
+    "I-PUNCT": 34,
+    "I-SCONJ": 35,
+    "I-SYM": 36,
+    "I-VERB": 37,
+    "I-X": 38,
+    "NOUN": 39,
+    "NUM": 40,
+    "PART": 41,
+    "PRON": 42,
+    "PROPN": 43,
+    "PUNCT": 44,
+    "SCONJ": 45,
+    "SYM": 46,
+    "VERB": 47,
+    "X": 48
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
@@ -125,7 +122,7 @@
   "position_embedding_type": "absolute",
   "tokenizer_class": "BertTokenizerFast",
   "torch_dtype": "float32",
-  "transformers_version": "4.11.3",
+  "transformers_version": "4.19.2",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 9415

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da2e0d83b40f3976aa7d2a8fd47596b0789446b3ad1c28c3ec2d761cf377c007
-size 1250288662
+oid sha256:72a82baa53d0f94a7ae47b5933238aa15c369913006dc75dedec8e75c55bb7c8
+size 1250252273

supar.model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bdcefe8e6d943edda200dab52b7e195f6030312242854185240662f72290c842
-size 1303629217
+oid sha256:836d88dbb4f11fd2abfe580f94809b656a335f544a17ca15ba4ef6107577240d
+size 1302635621

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "never_split": ["[PAD]", "[UNK]", "[CLS]", "[SEP]", "[MASK]"], "model_max_length": 512, "do_basic_tokenize": true, "special_tokens_map_file": "/home/yasuoka/.cache/huggingface/transformers/f6e99603eee497be86830d375f662df1256064f3228eb468f7a8d4b546585e34.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "name_or_path": "KoichiYasuoka/roberta-large-japanese-aozora-char", "tokenizer_class": "BertTokenizer"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "never_split": ["[PAD]", "[UNK]", "[CLS]", "[SEP]", "[MASK]"], "model_max_length": 512, "do_basic_tokenize": true, "tokenizer_class": "BertTokenizerFast"}