Hugging Face's logo

PuxAI
/
GenPII-framework-Data

Safetensors
Model card Files
xet
 Community
PuxAI commited on
Commit
d048256
·
verified ·
1 Parent(s): 9b7401d

Upload folder using huggingface_hub

Browse files
Files changed (13) hide show
  1. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/config.json +127 -0
  2. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/model.safetensors +3 -0
  3. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/optimizer.pt +3 -0
  4. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/rng_state.pth +3 -0
  5. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/scheduler.pt +3 -0
  6. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer.json +0 -0
  7. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer_config.json +23 -0
  8. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/trainer_state.json +33 -0
  9. gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/training_args.bin +3 -0
  10. gretel-pii-ready/SpanBased-CustomDeBERTa/config.json +127 -0
  11. gretel-pii-ready/SpanBased-CustomDeBERTa/pytorch_model.bin +3 -0
  12. gretel-pii-ready/SpanBased-CustomDeBERTa/tokenizer.json +0 -0
  13. gretel-pii-ready/SpanBased-CustomDeBERTa/tokenizer_config.json +23 -0
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/config.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaStandardSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": null,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "O",
14
+ "1": "ACCOUNT_NUMBER",
15
+ "2": "ADDRESS",
16
+ "3": "API_KEY",
17
+ "4": "BANK_ROUTING_NUMBER",
18
+ "5": "BIOMETRIC_IDENTIFIER",
19
+ "6": "CERTIFICATE_LICENSE_NUMBER",
20
+ "7": "CITY",
21
+ "8": "COMPANY_NAME",
22
+ "9": "COORDINATE",
23
+ "10": "COUNTRY",
24
+ "11": "CREDIT_CARD_NUMBER",
25
+ "12": "CUSTOMER_ID",
26
+ "13": "CVV",
27
+ "14": "DATE",
28
+ "15": "DATE_OF_BIRTH",
29
+ "16": "DATE_TIME",
30
+ "17": "DEVICE_IDENTIFIER",
31
+ "18": "EMAIL",
32
+ "19": "EMPLOYEE_ID",
33
+ "20": "FIRST_NAME",
34
+ "21": "HEALTH_PLAN_BENEFICIARY_NUMBER",
35
+ "22": "IPV4",
36
+ "23": "IPV6",
37
+ "24": "LAST_NAME",
38
+ "25": "LICENSE_PLATE",
39
+ "26": "MEDICAL_RECORD_NUMBER",
40
+ "27": "NAME",
41
+ "28": "NATIONAL_ID",
42
+ "29": "PASSWORD",
43
+ "30": "PHONE_NUMBER",
44
+ "31": "POSTCODE",
45
+ "32": "SSN",
46
+ "33": "STATE",
47
+ "34": "STREET_ADDRESS",
48
+ "35": "SWIFT_BIC",
49
+ "36": "TAX_ID",
50
+ "37": "TIME",
51
+ "38": "UNIQUE_IDENTIFIER",
52
+ "39": "URL",
53
+ "40": "USER_NAME",
54
+ "41": "VEHICLE_IDENTIFIER"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "label2id": {
59
+ "ACCOUNT_NUMBER": 1,
60
+ "ADDRESS": 2,
61
+ "API_KEY": 3,
62
+ "BANK_ROUTING_NUMBER": 4,
63
+ "BIOMETRIC_IDENTIFIER": 5,
64
+ "CERTIFICATE_LICENSE_NUMBER": 6,
65
+ "CITY": 7,
66
+ "COMPANY_NAME": 8,
67
+ "COORDINATE": 9,
68
+ "COUNTRY": 10,
69
+ "CREDIT_CARD_NUMBER": 11,
70
+ "CUSTOMER_ID": 12,
71
+ "CVV": 13,
72
+ "DATE": 14,
73
+ "DATE_OF_BIRTH": 15,
74
+ "DATE_TIME": 16,
75
+ "DEVICE_IDENTIFIER": 17,
76
+ "EMAIL": 18,
77
+ "EMPLOYEE_ID": 19,
78
+ "FIRST_NAME": 20,
79
+ "HEALTH_PLAN_BENEFICIARY_NUMBER": 21,
80
+ "IPV4": 22,
81
+ "IPV6": 23,
82
+ "LAST_NAME": 24,
83
+ "LICENSE_PLATE": 25,
84
+ "MEDICAL_RECORD_NUMBER": 26,
85
+ "NAME": 27,
86
+ "NATIONAL_ID": 28,
87
+ "O": 0,
88
+ "PASSWORD": 29,
89
+ "PHONE_NUMBER": 30,
90
+ "POSTCODE": 31,
91
+ "SSN": 32,
92
+ "STATE": 33,
93
+ "STREET_ADDRESS": 34,
94
+ "SWIFT_BIC": 35,
95
+ "TAX_ID": 36,
96
+ "TIME": 37,
97
+ "UNIQUE_IDENTIFIER": 38,
98
+ "URL": 39,
99
+ "USER_NAME": 40,
100
+ "VEHICLE_IDENTIFIER": 41
101
+ },
102
+ "layer_norm_eps": 1e-07,
103
+ "legacy": true,
104
+ "max_position_embeddings": 512,
105
+ "max_relative_positions": -1,
106
+ "model_type": "deberta-v2",
107
+ "norm_rel_ebd": "layer_norm",
108
+ "num_attention_heads": 12,
109
+ "num_hidden_layers": 12,
110
+ "pad_token_id": 0,
111
+ "pooler_dropout": 0,
112
+ "pooler_hidden_act": "gelu",
113
+ "pooler_hidden_size": 768,
114
+ "pos_att_type": [
115
+ "p2c",
116
+ "c2p"
117
+ ],
118
+ "position_biased_input": false,
119
+ "position_buckets": 256,
120
+ "relative_attention": true,
121
+ "share_att_key": true,
122
+ "tie_word_embeddings": true,
123
+ "transformers_version": "5.3.0",
124
+ "type_vocab_size": 0,
125
+ "use_cache": false,
126
+ "vocab_size": 128100
127
+ }
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af35eb2b023a34443c242ab7d104cb413ca0665cbd7bde846c3d60f5e8e0688f
3
+ size 735608808
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b6aa9aa730ba778f7e8ef613957ed917aaea3e8f612690605a9dd02301f09d8
3
+ size 1471339339
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d92fe139f614b1b540b84919c4925065c6e584b3192f8a5c5d36b8c769e2593a
3
+ size 14645
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b04611bea80597dee677a377375d057cf3fa3437fb4ed8809d4662285b8811b
3
+ size 1465
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/trainer_state.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 250,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [],
12
+ "logging_steps": 500,
13
+ "max_steps": 250,
14
+ "num_input_tokens_seen": 0,
15
+ "num_train_epochs": 1,
16
+ "save_steps": 500,
17
+ "stateful_callbacks": {
18
+ "TrainerControl": {
19
+ "args": {
20
+ "should_epoch_stop": false,
21
+ "should_evaluate": false,
22
+ "should_log": false,
23
+ "should_save": true,
24
+ "should_training_stop": true
25
+ },
26
+ "attributes": {}
27
+ }
28
+ },
29
+ "total_flos": 21916554756480.0,
30
+ "train_batch_size": 2,
31
+ "trial_name": null,
32
+ "trial_params": null
33
+ }
gretel-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37e1b6c470c69d56fa07915194fff99de4cb94a69c9cb2982281d428822e62a0
3
+ size 5201
gretel-pii-ready/SpanBased-CustomDeBERTa/config.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaStandardSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": null,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "O",
14
+ "1": "ACCOUNT_NUMBER",
15
+ "2": "ADDRESS",
16
+ "3": "API_KEY",
17
+ "4": "BANK_ROUTING_NUMBER",
18
+ "5": "BIOMETRIC_IDENTIFIER",
19
+ "6": "CERTIFICATE_LICENSE_NUMBER",
20
+ "7": "CITY",
21
+ "8": "COMPANY_NAME",
22
+ "9": "COORDINATE",
23
+ "10": "COUNTRY",
24
+ "11": "CREDIT_CARD_NUMBER",
25
+ "12": "CUSTOMER_ID",
26
+ "13": "CVV",
27
+ "14": "DATE",
28
+ "15": "DATE_OF_BIRTH",
29
+ "16": "DATE_TIME",
30
+ "17": "DEVICE_IDENTIFIER",
31
+ "18": "EMAIL",
32
+ "19": "EMPLOYEE_ID",
33
+ "20": "FIRST_NAME",
34
+ "21": "HEALTH_PLAN_BENEFICIARY_NUMBER",
35
+ "22": "IPV4",
36
+ "23": "IPV6",
37
+ "24": "LAST_NAME",
38
+ "25": "LICENSE_PLATE",
39
+ "26": "MEDICAL_RECORD_NUMBER",
40
+ "27": "NAME",
41
+ "28": "NATIONAL_ID",
42
+ "29": "PASSWORD",
43
+ "30": "PHONE_NUMBER",
44
+ "31": "POSTCODE",
45
+ "32": "SSN",
46
+ "33": "STATE",
47
+ "34": "STREET_ADDRESS",
48
+ "35": "SWIFT_BIC",
49
+ "36": "TAX_ID",
50
+ "37": "TIME",
51
+ "38": "UNIQUE_IDENTIFIER",
52
+ "39": "URL",
53
+ "40": "USER_NAME",
54
+ "41": "VEHICLE_IDENTIFIER"
55
+ },
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 3072,
58
+ "label2id": {
59
+ "ACCOUNT_NUMBER": 1,
60
+ "ADDRESS": 2,
61
+ "API_KEY": 3,
62
+ "BANK_ROUTING_NUMBER": 4,
63
+ "BIOMETRIC_IDENTIFIER": 5,
64
+ "CERTIFICATE_LICENSE_NUMBER": 6,
65
+ "CITY": 7,
66
+ "COMPANY_NAME": 8,
67
+ "COORDINATE": 9,
68
+ "COUNTRY": 10,
69
+ "CREDIT_CARD_NUMBER": 11,
70
+ "CUSTOMER_ID": 12,
71
+ "CVV": 13,
72
+ "DATE": 14,
73
+ "DATE_OF_BIRTH": 15,
74
+ "DATE_TIME": 16,
75
+ "DEVICE_IDENTIFIER": 17,
76
+ "EMAIL": 18,
77
+ "EMPLOYEE_ID": 19,
78
+ "FIRST_NAME": 20,
79
+ "HEALTH_PLAN_BENEFICIARY_NUMBER": 21,
80
+ "IPV4": 22,
81
+ "IPV6": 23,
82
+ "LAST_NAME": 24,
83
+ "LICENSE_PLATE": 25,
84
+ "MEDICAL_RECORD_NUMBER": 26,
85
+ "NAME": 27,
86
+ "NATIONAL_ID": 28,
87
+ "O": 0,
88
+ "PASSWORD": 29,
89
+ "PHONE_NUMBER": 30,
90
+ "POSTCODE": 31,
91
+ "SSN": 32,
92
+ "STATE": 33,
93
+ "STREET_ADDRESS": 34,
94
+ "SWIFT_BIC": 35,
95
+ "TAX_ID": 36,
96
+ "TIME": 37,
97
+ "UNIQUE_IDENTIFIER": 38,
98
+ "URL": 39,
99
+ "USER_NAME": 40,
100
+ "VEHICLE_IDENTIFIER": 41
101
+ },
102
+ "layer_norm_eps": 1e-07,
103
+ "legacy": true,
104
+ "max_position_embeddings": 512,
105
+ "max_relative_positions": -1,
106
+ "model_type": "deberta-v2",
107
+ "norm_rel_ebd": "layer_norm",
108
+ "num_attention_heads": 12,
109
+ "num_hidden_layers": 12,
110
+ "pad_token_id": 0,
111
+ "pooler_dropout": 0,
112
+ "pooler_hidden_act": "gelu",
113
+ "pooler_hidden_size": 768,
114
+ "pos_att_type": [
115
+ "p2c",
116
+ "c2p"
117
+ ],
118
+ "position_biased_input": false,
119
+ "position_buckets": 256,
120
+ "relative_attention": true,
121
+ "share_att_key": true,
122
+ "tie_word_embeddings": true,
123
+ "transformers_version": "5.3.0",
124
+ "type_vocab_size": 0,
125
+ "use_cache": false,
126
+ "vocab_size": 128100
127
+ }
gretel-pii-ready/SpanBased-CustomDeBERTa/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bd8f692a73c78b3361c3a599f432b60d42409727ebd66575be512ead2a74d01
3
+ size 735669891
gretel-pii-ready/SpanBased-CustomDeBERTa/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gretel-pii-ready/SpanBased-CustomDeBERTa/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }