Main model

.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

config.json

@@ -1,46 +1,202 @@
 {
-  "_name_or_path": "distilbert-base-uncased",
-  "activation": "gelu",
+  "_name_or_path": "bert-base-cased",
   "architectures": [
-    "DistilBertForTokenClassification"
+    "BertForTokenClassification"
   ],
-  "attention_dropout": 0.1,
-  "dim": 768,
-  "dropout": 0.1,
-  "hidden_dim": 3072,
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "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"
+    "0": "B-ANATOMICAL_STRUCTURE",
+    "1": "B-CHEMICAL",
+    "2": "Other",
+    "3": "B-MOLECULAR_FUNCTION",
+    "4": "I-LABORATORY_PROCEDURE",
+    "5": "B-PRODUCT",
+    "6": "B-CELL_COMPONENT",
+    "7": "B-BODY_SUBSTANCE",
+    "8": "I-CARDINAL",
+    "9": "B-LIVESTOCK",
+    "10": "B-CELL_OR_MOLECULAR_DYSFUNCTION",
+    "11": "I-CORONAVIRUS",
+    "12": "B-DATE",
+    "13": "B-LABORATORY_PROCEDURE",
+    "14": "I-RESEARCH_ACTIVITY",
+    "15": "I-CHEMICAL",
+    "16": "I-EXPERIMENTAL_MODEL_OF_DISEASE",
+    "17": "B-CELL_FUNCTION",
+    "18": "B-EVOLUTION",
+    "19": "I-SIGN_OR_SYMPTOM",
+    "20": "B-FOOD",
+    "21": "B-TIME",
+    "22": "B-PHYSICAL_SCIENCE",
+    "23": "B-CELL",
+    "24": "B-QUANTITY",
+    "25": "I-GROUP",
+    "26": "B-ORG",
+    "27": "B-GPE",
+    "28": "B-EDUCATIONAL_ACTIVITY",
+    "29": "B-DAILY_OR_RECREATIONAL_ACTIVITY",
+    "30": "I-DISEASE_OR_SYNDROME",
+    "31": "I-TIME",
+    "32": "B-BODY_PART_ORGAN_OR_ORGAN_COMPONENT",
+    "33": "I-PRODUCT",
+    "34": "B-EXPERIMENTAL_MODEL_OF_DISEASE",
+    "35": "B-NORP",
+    "36": "I-TISSUE",
+    "37": "I-GPE",
+    "38": "I-CELL",
+    "39": "I-EVOLUTION",
+    "40": "I-QUANTITY",
+    "41": "B-VIRAL_PROTEIN",
+    "42": "I-PERSON",
+    "43": "B-DISEASE_OR_SYNDROME",
+    "44": "I-CELL_COMPONENT",
+    "45": "I-WILDLIFE",
+    "46": "B-FAC",
+    "47": "B-THERAPEUTIC_OR_PREVENTIVE_PROCEDURE",
+    "48": "B-ORDINAL",
+    "49": "B-EUKARYOTE",
+    "50": "B-LABORATORY_OR_TEST_RESULT",
+    "51": "I-INJURY_OR_POISONING",
+    "52": "B-PERCENT",
+    "53": "B-TISSUE",
+    "54": "I-NORP",
+    "55": "I-GENE_OR_GENOME",
+    "56": "B-INJURY_OR_POISONING",
+    "57": "I-ORGAN_OR_TISSUE_FUNCTION",
+    "58": "B-SUBSTRATE",
+    "59": "I-DATE",
+    "60": "I-ORGANISM",
+    "61": "I-FAC",
+    "62": "B-GROUP",
+    "63": "I-MONEY",
+    "64": "I-PHYSICAL_SCIENCE",
+    "65": "I-ORG",
+    "66": "B-SIGN_OR_SYMPTOM",
+    "67": "I-PERCENT",
+    "68": "B-MONEY",
+    "69": "B-PERSON",
+    "70": "B-WILDLIFE",
+    "71": "B-INDIVIDUAL_BEHAVIOR",
+    "72": "I-THERAPEUTIC_OR_PREVENTIVE_PROCEDURE",
+    "73": "B-VIRUS",
+    "74": "B-ORGANISM",
+    "75": "I-EDUCATIONAL_ACTIVITY",
+    "76": "B-LOC",
+    "77": "B-RESEARCH_ACTIVITY",
+    "78": "B-SOCIAL_BEHAVIOR",
+    "79": "B-ORGAN_OR_TISSUE_FUNCTION",
+    "80": "I-DIAGNOSTIC_PROCEDURE",
+    "81": "B-CARDINAL",
+    "82": "B-GOVERNMENTAL_OR_REGULATORY_ACTIVITY",
+    "83": "B-GENE_OR_GENOME",
+    "84": "B-DIAGNOSTIC_PROCEDURE",
+    "85": "B-CORONAVIRUS"
   },
   "initializer_range": 0.02,
+  "intermediate_size": 3072,
   "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1,
-    "LABEL_2": 2,
-    "LABEL_3": 3,
-    "LABEL_4": 4,
-    "LABEL_5": 5,
-    "LABEL_6": 6,
-    "LABEL_7": 7,
-    "LABEL_8": 8
+    "B-ANATOMICAL_STRUCTURE": 0,
+    "B-BODY_PART_ORGAN_OR_ORGAN_COMPONENT": 32,
+    "B-BODY_SUBSTANCE": 7,
+    "B-CARDINAL": 81,
+    "B-CELL": 23,
+    "B-CELL_COMPONENT": 6,
+    "B-CELL_FUNCTION": 17,
+    "B-CELL_OR_MOLECULAR_DYSFUNCTION": 10,
+    "B-CHEMICAL": 1,
+    "B-CORONAVIRUS": 85,
+    "B-DAILY_OR_RECREATIONAL_ACTIVITY": 29,
+    "B-DATE": 12,
+    "B-DIAGNOSTIC_PROCEDURE": 84,
+    "B-DISEASE_OR_SYNDROME": 43,
+    "B-EDUCATIONAL_ACTIVITY": 28,
+    "B-EUKARYOTE": 49,
+    "B-EVOLUTION": 18,
+    "B-EXPERIMENTAL_MODEL_OF_DISEASE": 34,
+    "B-FAC": 46,
+    "B-FOOD": 20,
+    "B-GENE_OR_GENOME": 83,
+    "B-GOVERNMENTAL_OR_REGULATORY_ACTIVITY": 82,
+    "B-GPE": 27,
+    "B-GROUP": 62,
+    "B-INDIVIDUAL_BEHAVIOR": 71,
+    "B-INJURY_OR_POISONING": 56,
+    "B-LABORATORY_OR_TEST_RESULT": 50,
+    "B-LABORATORY_PROCEDURE": 13,
+    "B-LIVESTOCK": 9,
+    "B-LOC": 76,
+    "B-MOLECULAR_FUNCTION": 3,
+    "B-MONEY": 68,
+    "B-NORP": 35,
+    "B-ORDINAL": 48,
+    "B-ORG": 26,
+    "B-ORGANISM": 74,
+    "B-ORGAN_OR_TISSUE_FUNCTION": 79,
+    "B-PERCENT": 52,
+    "B-PERSON": 69,
+    "B-PHYSICAL_SCIENCE": 22,
+    "B-PRODUCT": 5,
+    "B-QUANTITY": 24,
+    "B-RESEARCH_ACTIVITY": 77,
+    "B-SIGN_OR_SYMPTOM": 66,
+    "B-SOCIAL_BEHAVIOR": 78,
+    "B-SUBSTRATE": 58,
+    "B-THERAPEUTIC_OR_PREVENTIVE_PROCEDURE": 47,
+    "B-TIME": 21,
+    "B-TISSUE": 53,
+    "B-VIRAL_PROTEIN": 41,
+    "B-VIRUS": 73,
+    "B-WILDLIFE": 70,
+    "I-CARDINAL": 8,
+    "I-CELL": 38,
+    "I-CELL_COMPONENT": 44,
+    "I-CHEMICAL": 15,
+    "I-CORONAVIRUS": 11,
+    "I-DATE": 59,
+    "I-DIAGNOSTIC_PROCEDURE": 80,
+    "I-DISEASE_OR_SYNDROME": 30,
+    "I-EDUCATIONAL_ACTIVITY": 75,
+    "I-EVOLUTION": 39,
+    "I-EXPERIMENTAL_MODEL_OF_DISEASE": 16,
+    "I-FAC": 61,
+    "I-GENE_OR_GENOME": 55,
+    "I-GPE": 37,
+    "I-GROUP": 25,
+    "I-INJURY_OR_POISONING": 51,
+    "I-LABORATORY_PROCEDURE": 4,
+    "I-MONEY": 63,
+    "I-NORP": 54,
+    "I-ORG": 65,
+    "I-ORGANISM": 60,
+    "I-ORGAN_OR_TISSUE_FUNCTION": 57,
+    "I-PERCENT": 67,
+    "I-PERSON": 42,
+    "I-PHYSICAL_SCIENCE": 64,
+    "I-PRODUCT": 33,
+    "I-QUANTITY": 40,
+    "I-RESEARCH_ACTIVITY": 14,
+    "I-SIGN_OR_SYMPTOM": 19,
+    "I-THERAPEUTIC_OR_PREVENTIVE_PROCEDURE": 72,
+    "I-TIME": 31,
+    "I-TISSUE": 36,
+    "I-WILDLIFE": 45,
+    "Other": 2
   },
+  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "distilbert",
-  "n_heads": 12,
-  "n_layers": 6,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
   "pad_token_id": 0,
-  "qa_dropout": 0.1,
-  "seq_classif_dropout": 0.2,
-  "sinusoidal_pos_embds": false,
-  "tie_weights_": true,
+  "position_embedding_type": "absolute",
   "torch_dtype": "float32",
   "transformers_version": "4.47.1",
-  "vocab_size": 30522
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
 }

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc590879cc9354f2dd062e6679283d7b88d795cf9781e4815e41993b934dc667
+size 431166592

special_tokens_map.json

@@ -1,7 +0,0 @@
-{
-  "cls_token": "[CLS]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": "[UNK]"
-}

tokenizer_config.json

@@ -43,7 +43,7 @@
   },
   "clean_up_tokenization_spaces": false,
   "cls_token": "[CLS]",
-  "do_lower_case": true,
+  "do_lower_case": false,
   "extra_special_tokens": {},
   "mask_token": "[MASK]",
   "model_max_length": 512,
@@ -51,6 +51,6 @@
   "sep_token": "[SEP]",
   "strip_accents": null,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "DistilBertTokenizer",
+  "tokenizer_class": "BertTokenizer",
   "unk_token": "[UNK]"
 }