Training in progress epoch 0

.gitignore

@@ -0,0 +1,2 @@
+step_*
+epoch_*

config.json

@@ -0,0 +1,202 @@
+{
+  "arcface_margin": 0.2,
+  "arcface_scale": 30.0,
+  "architectures": [
+    "AngularBertForTokenClassification"
+  ],
+  "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": "O",
+    "1": "B-medical_record_number",
+    "2": "I-medical_record_number",
+    "3": "B-date_of_birth",
+    "4": "I-date_of_birth",
+    "5": "B-ssn",
+    "6": "I-ssn",
+    "7": "B-date",
+    "8": "I-date",
+    "9": "B-first_name",
+    "10": "I-first_name",
+    "11": "B-email",
+    "12": "I-email",
+    "13": "B-last_name",
+    "14": "I-last_name",
+    "15": "B-customer_id",
+    "16": "I-customer_id",
+    "17": "B-employee_id",
+    "18": "I-employee_id",
+    "19": "B-name",
+    "20": "I-name",
+    "21": "B-street_address",
+    "22": "I-street_address",
+    "23": "B-phone_number",
+    "24": "I-phone_number",
+    "25": "B-ipv4",
+    "26": "I-ipv4",
+    "27": "B-credit_card_number",
+    "28": "I-credit_card_number",
+    "29": "B-license_plate",
+    "30": "I-license_plate",
+    "31": "B-address",
+    "32": "I-address",
+    "33": "B-user_name",
+    "34": "I-user_name",
+    "35": "B-device_identifier",
+    "36": "I-device_identifier",
+    "37": "B-bank_routing_number",
+    "38": "I-bank_routing_number",
+    "39": "B-date_time",
+    "40": "I-date_time",
+    "41": "B-company_name",
+    "42": "I-company_name",
+    "43": "B-unique_identifier",
+    "44": "I-unique_identifier",
+    "45": "B-biometric_identifier",
+    "46": "I-biometric_identifier",
+    "47": "B-account_number",
+    "48": "I-account_number",
+    "49": "B-city",
+    "50": "I-city",
+    "51": "B-certificate_license_number",
+    "52": "I-certificate_license_number",
+    "53": "B-time",
+    "54": "I-time",
+    "55": "B-postcode",
+    "56": "I-postcode",
+    "57": "B-vehicle_identifier",
+    "58": "I-vehicle_identifier",
+    "59": "B-coordinate",
+    "60": "I-coordinate",
+    "61": "B-country",
+    "62": "I-country",
+    "63": "B-api_key",
+    "64": "I-api_key",
+    "65": "B-ipv6",
+    "66": "I-ipv6",
+    "67": "B-password",
+    "68": "I-password",
+    "69": "B-health_plan_beneficiary_number",
+    "70": "I-health_plan_beneficiary_number",
+    "71": "B-national_id",
+    "72": "I-national_id",
+    "73": "B-tax_id",
+    "74": "I-tax_id",
+    "75": "B-url",
+    "76": "I-url",
+    "77": "B-state",
+    "78": "I-state",
+    "79": "B-swift_bic",
+    "80": "I-swift_bic",
+    "81": "B-cvv",
+    "82": "I-cvv",
+    "83": "B-pin",
+    "84": "I-pin"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-account_number": 47,
+    "B-address": 31,
+    "B-api_key": 63,
+    "B-bank_routing_number": 37,
+    "B-biometric_identifier": 45,
+    "B-certificate_license_number": 51,
+    "B-city": 49,
+    "B-company_name": 41,
+    "B-coordinate": 59,
+    "B-country": 61,
+    "B-credit_card_number": 27,
+    "B-customer_id": 15,
+    "B-cvv": 81,
+    "B-date": 7,
+    "B-date_of_birth": 3,
+    "B-date_time": 39,
+    "B-device_identifier": 35,
+    "B-email": 11,
+    "B-employee_id": 17,
+    "B-first_name": 9,
+    "B-health_plan_beneficiary_number": 69,
+    "B-ipv4": 25,
+    "B-ipv6": 65,
+    "B-last_name": 13,
+    "B-license_plate": 29,
+    "B-medical_record_number": 1,
+    "B-name": 19,
+    "B-national_id": 71,
+    "B-password": 67,
+    "B-phone_number": 23,
+    "B-pin": 83,
+    "B-postcode": 55,
+    "B-ssn": 5,
+    "B-state": 77,
+    "B-street_address": 21,
+    "B-swift_bic": 79,
+    "B-tax_id": 73,
+    "B-time": 53,
+    "B-unique_identifier": 43,
+    "B-url": 75,
+    "B-user_name": 33,
+    "B-vehicle_identifier": 57,
+    "I-account_number": 48,
+    "I-address": 32,
+    "I-api_key": 64,
+    "I-bank_routing_number": 38,
+    "I-biometric_identifier": 46,
+    "I-certificate_license_number": 52,
+    "I-city": 50,
+    "I-company_name": 42,
+    "I-coordinate": 60,
+    "I-country": 62,
+    "I-credit_card_number": 28,
+    "I-customer_id": 16,
+    "I-cvv": 82,
+    "I-date": 8,
+    "I-date_of_birth": 4,
+    "I-date_time": 40,
+    "I-device_identifier": 36,
+    "I-email": 12,
+    "I-employee_id": 18,
+    "I-first_name": 10,
+    "I-health_plan_beneficiary_number": 70,
+    "I-ipv4": 26,
+    "I-ipv6": 66,
+    "I-last_name": 14,
+    "I-license_plate": 30,
+    "I-medical_record_number": 2,
+    "I-name": 20,
+    "I-national_id": 72,
+    "I-password": 68,
+    "I-phone_number": 24,
+    "I-pin": 84,
+    "I-postcode": 56,
+    "I-ssn": 6,
+    "I-state": 78,
+    "I-street_address": 22,
+    "I-swift_bic": 80,
+    "I-tax_id": 74,
+    "I-time": 54,
+    "I-unique_identifier": 44,
+    "I-url": 76,
+    "I-user_name": 34,
+    "I-vehicle_identifier": 58,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "loss_type": "chebyshev_arcface",
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea49f82cf3f15fc32d969653dfb2eb37fe1c9ff400ce72293297dc5b5b072eab
+size 435851404

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}