add ner model configs

ner/en/roberta.base/config.json

@@ -1,7 +1,7 @@
 {
-  "_name_or_path": "jinmang2/tmp",
+  "_name_or_path": "jinmang2/dooly-hub/ner/en/roberta.base",
   "architectures": [
-    "RobertaForCharNER"
+    "RobertaForSequenceTagging"
   ],
   "attention_probs_dropout_prob": 0.1,
   "bos_token_id": 0,
@@ -11,80 +11,80 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "LABEL_0",
-    "1": "LABEL_1",
-    "2": "LABEL_2",
-    "3": "LABEL_3",
-    "4": "LABEL_4",
-    "5": "LABEL_5",
-    "6": "LABEL_6",
-    "7": "LABEL_7",
-    "8": "LABEL_8",
-    "9": "LABEL_9",
-    "10": "LABEL_10",
-    "11": "LABEL_11",
-    "12": "LABEL_12",
-    "13": "LABEL_13",
-    "14": "LABEL_14",
-    "15": "LABEL_15",
-    "16": "LABEL_16",
-    "17": "LABEL_17",
-    "18": "LABEL_18",
-    "19": "LABEL_19",
-    "20": "LABEL_20",
-    "21": "LABEL_21",
-    "22": "LABEL_22",
-    "23": "LABEL_23",
-    "24": "LABEL_24",
-    "25": "LABEL_25",
-    "26": "LABEL_26",
-    "27": "LABEL_27",
-    "28": "LABEL_28",
-    "29": "LABEL_29",
-    "30": "LABEL_30",
-    "31": "LABEL_31",
-    "32": "LABEL_32",
-    "33": "LABEL_33",
-    "34": "LABEL_34"
+    "0": "O",
+    "1": "I-ORG",
+    "2": "I-PERSON",
+    "3": "I-DATE",
+    "4": "I-GPE",
+    "5": "B-ORG",
+    "6": "B-GPE",
+    "7": "B-PERSON",
+    "8": "B-DATE",
+    "9": "I-MONEY",
+    "10": "B-CARDINAL",
+    "11": "I-PERCENT",
+    "12": "B-NORP",
+    "13": "I-CARDINAL",
+    "14": "B-MONEY",
+    "15": "B-PERCENT",
+    "16": "I-TIME",
+    "17": "I-LOC",
+    "18": "I-FAC",
+    "19": "I-QUANTITY",
+    "20": "I-NORP",
+    "21": "I-EVENT",
+    "22": "B-ORDINAL",
+    "23": "I-PRODUCT",
+    "24": "B-LOC",
+    "25": "B-TIME",
+    "26": "I-LAW",
+    "27": "B-QUANTITY",
+    "28": "B-FAC",
+    "29": "B-PRODUCT",
+    "30": "B-EVENT",
+    "31": "I-ORDINAL",
+    "32": "B-LAW",
+    "33": "B-LANGUAGE",
+    "34": "I-LANGUAGE"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1,
-    "LABEL_10": 10,
-    "LABEL_11": 11,
-    "LABEL_12": 12,
-    "LABEL_13": 13,
-    "LABEL_14": 14,
-    "LABEL_15": 15,
-    "LABEL_16": 16,
-    "LABEL_17": 17,
-    "LABEL_18": 18,
-    "LABEL_19": 19,
-    "LABEL_2": 2,
-    "LABEL_20": 20,
-    "LABEL_21": 21,
-    "LABEL_22": 22,
-    "LABEL_23": 23,
-    "LABEL_24": 24,
-    "LABEL_25": 25,
-    "LABEL_26": 26,
-    "LABEL_27": 27,
-    "LABEL_28": 28,
-    "LABEL_29": 29,
-    "LABEL_3": 3,
-    "LABEL_30": 30,
-    "LABEL_31": 31,
-    "LABEL_32": 32,
-    "LABEL_33": 33,
-    "LABEL_34": 34,
-    "LABEL_4": 4,
-    "LABEL_5": 5,
-    "LABEL_6": 6,
-    "LABEL_7": 7,
-    "LABEL_8": 8,
-    "LABEL_9": 9
+    "O": 0,
+    "I-ORG": 1,
+    "I-PERSON": 2,
+    "I-DATE": 3,
+    "I-GPE": 4,
+    "B-ORG": 5,
+    "B-GPE": 6,
+    "B-PERSON": 7,
+    "B-DATE": 8,
+    "I-MONEY": 9,
+    "B-CARDINAL": 10,
+    "I-PERCENT": 11,
+    "B-NORP": 12,
+    "I-CARDINAL": 13,
+    "B-MONEY": 14,
+    "B-PERCENT": 15,
+    "I-TIME": 16,
+    "I-LOC": 17,
+    "I-FAC": 18,
+    "I-QUANTITY": 19,
+    "I-NORP": 20,
+    "I-EVENT": 21,
+    "B-ORDINAL": 22,
+    "I-PRODUCT": 23,
+    "B-LOC": 24,
+    "B-TIME": 25,
+    "I-LAW": 26,
+    "B-QUANTITY": 27,
+    "B-FAC": 28,
+    "B-PRODUCT": 29,
+    "B-EVENT": 30,
+    "I-ORDINAL": 31,
+    "B-LAW": 32,
+    "B-LANGUAGE": 33,
+    "I-LANGUAGE": 34
   },
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
@@ -94,7 +94,7 @@
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.15.0",
+  "transformers_version": "4.16.2",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 50265

ner/ja/jaberta.base/config.json

@@ -0,0 +1,69 @@
+{
+  "_name_or_path": "jinmang2/dooly-hub/ner/ja/jaberta.base",
+  "architectures": [
+    "RobertaForSequenceTagging"
+  ],
+  "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": "O",
+    "1": "B-LOCATION",
+    "2": "I-LOCATION",
+    "3": "I-ARTIFACT",
+    "4": "I-ORGANIZATION",
+    "5": "I-DATE",
+    "6": "B-DATE",
+    "7": "I-PERSON",
+    "8": "B-ORGANIZATION",
+    "9": "B-ARTIFACT",
+    "10": "B-PERSON",
+    "11": "I-OPTIONAL",
+    "12": "B-OPTIONAL",
+    "13": "I-MONEY",
+    "14": "I-PERCENT",
+    "15": "B-MONEY",
+    "16": "I-TIME",
+    "17": "B-PERCENT",
+    "18": "B-TIME"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "O": 0,
+    "B-LOCATION": 1,
+    "I-LOCATION": 2,
+    "I-ARTIFACT": 3,
+    "I-ORGANIZATION": 4,
+    "I-DATE": 5,
+    "B-DATE": 6,
+    "I-PERSON": 7,
+    "B-ORGANIZATION": 8,
+    "B-ARTIFACT": 9,
+    "B-PERSON": 10,
+    "I-OPTIONAL": 11,
+    "B-OPTIONAL": 12,
+    "I-MONEY": 13,
+    "I-PERCENT": 14,
+    "B-MONEY": 15,
+    "I-TIME": 16,
+    "B-PERCENT": 17,
+    "B-TIME": 18
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

ner/ko/charbert.base/config.json

@@ -1,7 +1,7 @@
 {
-  "_name_or_path": "jinmang2/dooly-hub/ner/ko",
+  "_name_or_path": "jinmang2/dooly-hub/ner/ko/charbert.base",
   "architectures": [
-    "RobertaForCharNER"
+    "RobertaForSequenceTagging"
   ],
   "attention_probs_dropout_prob": 0.1,
   "bos_token_id": 0,

ner/zh/zhberta.base/config.json

@@ -0,0 +1,101 @@
+{
+  "_name_or_path": "jinmang2/dooly-hub/ner/zh/zhberta.base",
+  "architectures": [
+    "RobertaForSequenceTagging"
+  ],
+  "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": "O",
+    "1": "I-ORG",
+    "2": "I-PERSON",
+    "3": "I-GPE",
+    "4": "I-DATE",
+    "5": "B-GPE",
+    "6": "B-PERSON",
+    "7": "B-ORG",
+    "8": "B-DATE",
+    "9": "B-CARDINAL",
+    "10": "I-MONEY",
+    "11": "I-CARDINAL",
+    "12": "I-EVENT",
+    "13": "I-TIME",
+    "14": "I-FAC",
+    "15": "I-LOC",
+    "16": "I-PERCENT",
+    "17": "I-NORP",
+    "18": "I-QUANTITY",
+    "19": "B-LOC",
+    "20": "B-NORP",
+    "21": "B-TIME",
+    "22": "I-LAW",
+    "23": "B-FAC",
+    "24": "B-MONEY",
+    "25": "I-ORDINAL",
+    "26": "B-ORDINAL",
+    "27": "B-EVENT",
+    "28": "I-PRODUCT",
+    "29": "B-QUANTITY",
+    "30": "B-PERCENT",
+    "31": "I-LANGUAGE",
+    "32": "B-PRODUCT",
+    "33": "B-LANGUAGE",
+    "34": "B-LAW"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "O": 0,
+    "I-ORG": 1,
+    "I-PERSON": 2,
+    "I-GPE": 3,
+    "I-DATE": 4,
+    "B-GPE": 5,
+    "B-PERSON": 6,
+    "B-ORG": 7,
+    "B-DATE": 8,
+    "B-CARDINAL": 9,
+    "I-MONEY": 10,
+    "I-CARDINAL": 11,
+    "I-EVENT": 12,
+    "I-TIME": 13,
+    "I-FAC": 14,
+    "I-LOC": 15,
+    "I-PERCENT": 16,
+    "I-NORP": 17,
+    "I-QUANTITY": 18,
+    "B-LOC": 19,
+    "B-NORP": 20,
+    "B-TIME": 21,
+    "I-LAW": 22,
+    "B-FAC": 23,
+    "B-MONEY": 24,
+    "I-ORDINAL": 25,
+    "B-ORDINAL": 26,
+    "B-EVENT": 27,
+    "I-PRODUCT": 28,
+    "B-QUANTITY": 29,
+    "B-PERCENT": 30,
+    "I-LANGUAGE": 31,
+    "B-PRODUCT": 32,
+    "B-LANGUAGE": 33,
+    "B-LAW": 34
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 21128
+}