Hugging Face's logo

Mozilla
/
tinybert-address-autofill

Text Classification
Transformers.js
ONNX
multilingual
bert
autofill
field-classification
tinybert
browser
Model card Files
xet
 Community
1
rolf-mozilla commited on
Commit
3bb0886
·
verified ·
1 Parent(s): 056ef4b

Upload 5 files

Browse files
Files changed (3) hide show
  1. config.json +168 -165
  2. special_tokens_map.json +37 -37
  3. tokenizer_config.json +63 -62
config.json CHANGED
@@ -1,165 +1,168 @@
1
- {
2
- "architectures": [
3
- "BertForSequenceClassification"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "cell": {},
7
- "classifier_dropout": null,
8
- "dtype": "float32",
9
- "emb_size": 312,
10
- "hidden_act": "gelu",
11
- "hidden_dropout_prob": 0.1,
12
- "hidden_size": 312,
13
- "id2label": {
14
- "1": "other",
15
- "2": "given-name",
16
- "3": "family-name",
17
- "4": "name",
18
- "5": "additional-name",
19
- "6": "phonetic-given-name",
20
- "7": "phonetic-family-name",
21
- "8": "phonetic-name",
22
- "9": "honorific-prefix",
23
- "10": "honorific-suffix",
24
- "11": "nickname",
25
- "12": "street-address",
26
- "13": "address-lookup",
27
- "14": "address-line1",
28
- "15": "address-line2",
29
- "16": "address-line3",
30
- "17": "address-level1",
31
- "18": "address-level2",
32
- "19": "address-level3",
33
- "20": "address-level4",
34
- "21": "street",
35
- "22": "address-streetname",
36
- "23": "address-housenumber",
37
- "24": "address-extra-housesuffix",
38
- "25": "postal-code",
39
- "26": "postal-code-lookup",
40
- "27": "postal-code-and-city",
41
- "28": "postal-code-or-suburb",
42
- "29": "country",
43
- "30": "country-name",
44
- "31": "tel",
45
- "32": "tel-country-code",
46
- "33": "tel-national",
47
- "34": "tel-area-code",
48
- "35": "tel-local",
49
- "36": "tel-local-prefix",
50
- "37": "tel-local-suffix",
51
- "38": "tel-extension",
52
- "39": "organization",
53
- "40": "organization-title",
54
- "41": "bday",
55
- "42": "bday-day",
56
- "43": "bday-month",
57
- "44": "bday-year",
58
- "45": "email",
59
- "46": "apartment",
60
- "47": "floor",
61
- "48": "stair",
62
- "49": "building",
63
- "50": "block",
64
- "51": "address-extra",
65
- "52": "cc-name",
66
- "53": "cc-given-name",
67
- "54": "cc-additional-name",
68
- "55": "cc-family-name",
69
- "56": "cc-number",
70
- "57": "cc-exp",
71
- "58": "cc-exp-month",
72
- "59": "cc-exp-year",
73
- "60": "cc-csc",
74
- "61": "cc-type",
75
- "62": "sex",
76
- "63": "id-number",
77
- "64": "vat-number",
78
- "65": "reference-point",
79
- "66": "loginname"
80
- },
81
- "initializer_range": 0.02,
82
- "intermediate_size": 1200,
83
- "label2id": {
84
- "additional-name": 5,
85
- "address-extra": 51,
86
- "address-extra-housesuffix": 24,
87
- "address-housenumber": 23,
88
- "address-level1": 17,
89
- "address-level2": 18,
90
- "address-level3": 19,
91
- "address-level4": 20,
92
- "address-line1": 14,
93
- "address-line2": 15,
94
- "address-line3": 16,
95
- "address-lookup": 13,
96
- "address-streetname": 22,
97
- "apartment": 46,
98
- "bday": 41,
99
- "bday-day": 42,
100
- "bday-month": 43,
101
- "bday-year": 44,
102
- "block": 50,
103
- "building": 49,
104
- "cc-additional-name": 54,
105
- "cc-csc": 60,
106
- "cc-exp": 57,
107
- "cc-exp-month": 58,
108
- "cc-exp-year": 59,
109
- "cc-family-name": 55,
110
- "cc-given-name": 53,
111
- "cc-name": 52,
112
- "cc-number": 56,
113
- "cc-type": 61,
114
- "country": 29,
115
- "country-name": 30,
116
- "email": 45,
117
- "family-name": 3,
118
- "floor": 47,
119
- "given-name": 2,
120
- "honorific-prefix": 9,
121
- "honorific-suffix": 10,
122
- "id-number": 63,
123
- "loginname": 66,
124
- "name": 4,
125
- "nickname": 11,
126
- "organization": 39,
127
- "organization-title": 40,
128
- "other": 1,
129
- "phonetic-family-name": 7,
130
- "phonetic-given-name": 6,
131
- "phonetic-name": 8,
132
- "postal-code": 25,
133
- "postal-code-and-city": 27,
134
- "postal-code-lookup": 26,
135
- "postal-code-or-suburb": 28,
136
- "reference-point": 65,
137
- "sex": 62,
138
- "stair": 48,
139
- "street": 21,
140
- "street-address": 12,
141
- "tel": 31,
142
- "tel-area-code": 34,
143
- "tel-country-code": 32,
144
- "tel-extension": 38,
145
- "tel-local": 35,
146
- "tel-local-prefix": 36,
147
- "tel-local-suffix": 37,
148
- "tel-national": 33,
149
- "vat-number": 64
150
- },
151
- "layer_norm_eps": 1e-12,
152
- "max_position_embeddings": 512,
153
- "model_type": "bert",
154
- "num_attention_heads": 12,
155
- "num_hidden_layers": 4,
156
- "pad_token_id": 0,
157
- "position_embedding_type": "absolute",
158
- "pre_trained": "",
159
- "problem_type": "single_label_classification",
160
- "structure": [],
161
- "transformers_version": "4.57.6",
162
- "type_vocab_size": 2,
163
- "use_cache": true,
164
- "vocab_size": 30522
165
- }
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "/Users/Rrando/Documents/GitHub/ml-form-autofill/outputs/autofill-tiny-supported-argo-autofillflow-jt4qd",
4
+ "architectures": [
5
+ "BertForSequenceClassification"
6
+ ],
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "cell": {},
9
+ "classifier_dropout": null,
10
+ "dtype": "float32",
11
+ "emb_size": 312,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 312,
15
+ "id2label": {
16
+ "1": "other",
17
+ "2": "given-name",
18
+ "3": "family-name",
19
+ "4": "name",
20
+ "5": "additional-name",
21
+ "6": "phonetic-given-name",
22
+ "7": "phonetic-family-name",
23
+ "8": "phonetic-name",
24
+ "9": "honorific-prefix",
25
+ "10": "honorific-suffix",
26
+ "11": "nickname",
27
+ "12": "street-address",
28
+ "13": "address-lookup",
29
+ "14": "address-line1",
30
+ "15": "address-line2",
31
+ "16": "address-line3",
32
+ "17": "address-level1",
33
+ "18": "address-level2",
34
+ "19": "address-level3",
35
+ "20": "address-level4",
36
+ "21": "street",
37
+ "22": "address-streetname",
38
+ "23": "address-housenumber",
39
+ "24": "address-extra-housesuffix",
40
+ "25": "postal-code",
41
+ "26": "postal-code-lookup",
42
+ "27": "postal-code-and-city",
43
+ "28": "postal-code-or-suburb",
44
+ "29": "country",
45
+ "30": "country-name",
46
+ "31": "tel",
47
+ "32": "tel-country-code",
48
+ "33": "tel-national",
49
+ "34": "tel-area-code",
50
+ "35": "tel-local",
51
+ "36": "tel-local-prefix",
52
+ "37": "tel-local-suffix",
53
+ "38": "tel-extension",
54
+ "39": "organization",
55
+ "40": "organization-title",
56
+ "41": "bday",
57
+ "42": "bday-day",
58
+ "43": "bday-month",
59
+ "44": "bday-year",
60
+ "45": "email",
61
+ "46": "apartment",
62
+ "47": "floor",
63
+ "48": "stair",
64
+ "49": "building",
65
+ "50": "block",
66
+ "51": "address-extra",
67
+ "52": "cc-name",
68
+ "53": "cc-given-name",
69
+ "54": "cc-additional-name",
70
+ "55": "cc-family-name",
71
+ "56": "cc-number",
72
+ "57": "cc-exp",
73
+ "58": "cc-exp-month",
74
+ "59": "cc-exp-year",
75
+ "60": "cc-csc",
76
+ "61": "cc-type",
77
+ "62": "sex",
78
+ "63": "id-number",
79
+ "64": "vat-number",
80
+ "65": "reference-point",
81
+ "66": "loginname"
82
+ },
83
+ "initializer_range": 0.02,
84
+ "intermediate_size": 1200,
85
+ "label2id": {
86
+ "additional-name": 5,
87
+ "address-extra": 51,
88
+ "address-extra-housesuffix": 24,
89
+ "address-housenumber": 23,
90
+ "address-level1": 17,
91
+ "address-level2": 18,
92
+ "address-level3": 19,
93
+ "address-level4": 20,
94
+ "address-line1": 14,
95
+ "address-line2": 15,
96
+ "address-line3": 16,
97
+ "address-lookup": 13,
98
+ "address-streetname": 22,
99
+ "apartment": 46,
100
+ "bday": 41,
101
+ "bday-day": 42,
102
+ "bday-month": 43,
103
+ "bday-year": 44,
104
+ "block": 50,
105
+ "building": 49,
106
+ "cc-additional-name": 54,
107
+ "cc-csc": 60,
108
+ "cc-exp": 57,
109
+ "cc-exp-month": 58,
110
+ "cc-exp-year": 59,
111
+ "cc-family-name": 55,
112
+ "cc-given-name": 53,
113
+ "cc-name": 52,
114
+ "cc-number": 56,
115
+ "cc-type": 61,
116
+ "country": 29,
117
+ "country-name": 30,
118
+ "email": 45,
119
+ "family-name": 3,
120
+ "floor": 47,
121
+ "given-name": 2,
122
+ "honorific-prefix": 9,
123
+ "honorific-suffix": 10,
124
+ "id-number": 63,
125
+ "loginname": 66,
126
+ "name": 4,
127
+ "nickname": 11,
128
+ "organization": 39,
129
+ "organization-title": 40,
130
+ "other": 1,
131
+ "phonetic-family-name": 7,
132
+ "phonetic-given-name": 6,
133
+ "phonetic-name": 8,
134
+ "postal-code": 25,
135
+ "postal-code-and-city": 27,
136
+ "postal-code-lookup": 26,
137
+ "postal-code-or-suburb": 28,
138
+ "reference-point": 65,
139
+ "sex": 62,
140
+ "stair": 48,
141
+ "street": 21,
142
+ "street-address": 12,
143
+ "tel": 31,
144
+ "tel-area-code": 34,
145
+ "tel-country-code": 32,
146
+ "tel-extension": 38,
147
+ "tel-local": 35,
148
+ "tel-local-prefix": 36,
149
+ "tel-local-suffix": 37,
150
+ "tel-national": 33,
151
+ "vat-number": 64
152
+ },
153
+ "layer_norm_eps": 1e-12,
154
+ "max_position_embeddings": 512,
155
+ "model_type": "bert",
156
+ "num_attention_heads": 12,
157
+ "num_hidden_layers": 4,
158
+ "pad_token_id": 0,
159
+ "position_embedding_type": "absolute",
160
+ "pre_trained": "",
161
+ "problem_type": "single_label_classification",
162
+ "structure": [],
163
+ "torch_dtype": "float32",
164
+ "transformers_version": "4.49.0",
165
+ "type_vocab_size": 2,
166
+ "use_cache": false,
167
+ "vocab_size": 30522
168
+ }
special_tokens_map.json CHANGED
@@ -1,37 +1,37 @@
1
- {
2
- "cls_token": {
3
- "content": "[CLS]",
4
- "lstrip": false,
5
- "normalized": false,
6
- "rstrip": false,
7
- "single_word": false
8
- },
9
- "mask_token": {
10
- "content": "[MASK]",
11
- "lstrip": false,
12
- "normalized": false,
13
- "rstrip": false,
14
- "single_word": false
15
- },
16
- "pad_token": {
17
- "content": "[PAD]",
18
- "lstrip": false,
19
- "normalized": false,
20
- "rstrip": false,
21
- "single_word": false
22
- },
23
- "sep_token": {
24
- "content": "[SEP]",
25
- "lstrip": false,
26
- "normalized": false,
27
- "rstrip": false,
28
- "single_word": false
29
- },
30
- "unk_token": {
31
- "content": "[UNK]",
32
- "lstrip": false,
33
- "normalized": false,
34
- "rstrip": false,
35
- "single_word": false
36
- }
37
- }
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer_config.json CHANGED
@@ -1,62 +1,63 @@
1
- {
2
- "added_tokens_decoder": {
3
- "0": {
4
- "content": "[PAD]",
5
- "lstrip": false,
6
- "normalized": false,
7
- "rstrip": false,
8
- "single_word": false,
9
- "special": true
10
- },
11
- "100": {
12
- "content": "[UNK]",
13
- "lstrip": false,
14
- "normalized": false,
15
- "rstrip": false,
16
- "single_word": false,
17
- "special": true
18
- },
19
- "101": {
20
- "content": "[CLS]",
21
- "lstrip": false,
22
- "normalized": false,
23
- "rstrip": false,
24
- "single_word": false,
25
- "special": true
26
- },
27
- "102": {
28
- "content": "[SEP]",
29
- "lstrip": false,
30
- "normalized": false,
31
- "rstrip": false,
32
- "single_word": false,
33
- "special": true
34
- },
35
- "103": {
36
- "content": "[MASK]",
37
- "lstrip": false,
38
- "normalized": false,
39
- "rstrip": false,
40
- "single_word": false,
41
- "special": true
42
- }
43
- },
44
- "clean_up_tokenization_spaces": true,
45
- "cls_token": "[CLS]",
46
- "do_basic_tokenize": true,
47
- "do_lower_case": true,
48
- "extra_special_tokens": {},
49
- "mask_token": "[MASK]",
50
- "max_length": 512,
51
- "model_max_length": 1000000000000000019884624838656,
52
- "never_split": null,
53
- "pad_token": "[PAD]",
54
- "sep_token": "[SEP]",
55
- "stride": 0,
56
- "strip_accents": null,
57
- "tokenize_chinese_chars": true,
58
- "tokenizer_class": "BertTokenizer",
59
- "truncation_side": "right",
60
- "truncation_strategy": "longest_first",
61
- "unk_token": "[UNK]"
62
- }
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "backend": "tokenizers",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "is_local": false,
50
+ "local_files_only": false,
51
+ "mask_token": "[MASK]",
52
+ "max_length": 512,
53
+ "model_max_length": 1000000000000000019884624838656,
54
+ "pad_token": "[PAD]",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }