albertosei commited on
Commit
cb04d08
·
verified ·
1 Parent(s): 8bd69a8

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,138 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "albertosei/layout_ubiai",
3
+ "architectures": [
4
+ "LayoutLMv3ForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "coordinate_size": 128,
10
+ "eos_token_id": 2,
11
+ "has_relative_attention_bias": true,
12
+ "has_spatial_attention_bias": true,
13
+ "hidden_act": "gelu",
14
+ "hidden_dropout_prob": 0.1,
15
+ "hidden_size": 768,
16
+ "id2label": {
17
+ "0": "O",
18
+ "1": "B-BALANCE_DUE",
19
+ "2": "B-BANK_NAME",
20
+ "3": "B-CLIENT_ADDRESS",
21
+ "4": "B-CLIENT_EMAIL",
22
+ "5": "B-CLIENT_NAME",
23
+ "6": "B-CLIENT_PHONE",
24
+ "7": "B-CLIENT_TAX_ID",
25
+ "8": "B-CURRENCY",
26
+ "9": "B-DUE_DATE",
27
+ "10": "B-IBAN",
28
+ "11": "B-INVOICE_DATE",
29
+ "12": "B-INVOICE_NUMBER",
30
+ "13": "B-ITEM_DESC",
31
+ "14": "B-ITEM_NO",
32
+ "15": "B-LINE_DISCOUNT",
33
+ "16": "B-LINE_GROSS_WORTH",
34
+ "17": "B-LINE_NET_WORTH",
35
+ "18": "B-LINE_UNIT_PRICE",
36
+ "19": "B-LINE_VAT",
37
+ "20": "B-ORDER_NUMBER",
38
+ "21": "B-QTY",
39
+ "22": "B-REFERENCE_NUMBER",
40
+ "23": "B-SELLER_ADDRESS",
41
+ "24": "B-SELLER_EMAIL",
42
+ "25": "B-SELLER_NAME",
43
+ "26": "B-SELLER_PHONE",
44
+ "27": "B-SELLER_TAX_ID",
45
+ "28": "B-SUBTOTAL",
46
+ "29": "B-TOTAL_GROSS_WORTH",
47
+ "30": "B-TOTAL_NET_WORTH",
48
+ "31": "B-TOTAL_VAT",
49
+ "32": "I-BANK_NAME",
50
+ "33": "I-CLIENT_ADDRESS",
51
+ "34": "I-CLIENT_EMAIL",
52
+ "35": "I-CLIENT_NAME",
53
+ "36": "I-CLIENT_PHONE",
54
+ "37": "I-INVOICE_NUMBER",
55
+ "38": "I-ITEM_DESC",
56
+ "39": "I-LINE_NET_WORTH",
57
+ "40": "I-LINE_UNIT_PRICE",
58
+ "41": "I-SELLER_ADDRESS",
59
+ "42": "I-SELLER_EMAIL",
60
+ "43": "I-SELLER_NAME",
61
+ "44": "I-SELLER_PHONE",
62
+ "45": "I-SELLER_TAX_ID",
63
+ "46": "I-TOTAL_NET_WORTH"
64
+ },
65
+ "initializer_range": 0.02,
66
+ "input_size": 224,
67
+ "intermediate_size": 3072,
68
+ "label2id": {
69
+ "B-BALANCE_DUE": 1,
70
+ "B-BANK_NAME": 2,
71
+ "B-CLIENT_ADDRESS": 3,
72
+ "B-CLIENT_EMAIL": 4,
73
+ "B-CLIENT_NAME": 5,
74
+ "B-CLIENT_PHONE": 6,
75
+ "B-CLIENT_TAX_ID": 7,
76
+ "B-CURRENCY": 8,
77
+ "B-DUE_DATE": 9,
78
+ "B-IBAN": 10,
79
+ "B-INVOICE_DATE": 11,
80
+ "B-INVOICE_NUMBER": 12,
81
+ "B-ITEM_DESC": 13,
82
+ "B-ITEM_NO": 14,
83
+ "B-LINE_DISCOUNT": 15,
84
+ "B-LINE_GROSS_WORTH": 16,
85
+ "B-LINE_NET_WORTH": 17,
86
+ "B-LINE_UNIT_PRICE": 18,
87
+ "B-LINE_VAT": 19,
88
+ "B-ORDER_NUMBER": 20,
89
+ "B-QTY": 21,
90
+ "B-REFERENCE_NUMBER": 22,
91
+ "B-SELLER_ADDRESS": 23,
92
+ "B-SELLER_EMAIL": 24,
93
+ "B-SELLER_NAME": 25,
94
+ "B-SELLER_PHONE": 26,
95
+ "B-SELLER_TAX_ID": 27,
96
+ "B-SUBTOTAL": 28,
97
+ "B-TOTAL_GROSS_WORTH": 29,
98
+ "B-TOTAL_NET_WORTH": 30,
99
+ "B-TOTAL_VAT": 31,
100
+ "I-BANK_NAME": 32,
101
+ "I-CLIENT_ADDRESS": 33,
102
+ "I-CLIENT_EMAIL": 34,
103
+ "I-CLIENT_NAME": 35,
104
+ "I-CLIENT_PHONE": 36,
105
+ "I-INVOICE_NUMBER": 37,
106
+ "I-ITEM_DESC": 38,
107
+ "I-LINE_NET_WORTH": 39,
108
+ "I-LINE_UNIT_PRICE": 40,
109
+ "I-SELLER_ADDRESS": 41,
110
+ "I-SELLER_EMAIL": 42,
111
+ "I-SELLER_NAME": 43,
112
+ "I-SELLER_PHONE": 44,
113
+ "I-SELLER_TAX_ID": 45,
114
+ "I-TOTAL_NET_WORTH": 46,
115
+ "O": 0
116
+ },
117
+ "layer_norm_eps": 1e-05,
118
+ "max_2d_position_embeddings": 1024,
119
+ "max_position_embeddings": 514,
120
+ "max_rel_2d_pos": 256,
121
+ "max_rel_pos": 128,
122
+ "model_type": "layoutlmv3",
123
+ "num_attention_heads": 12,
124
+ "num_channels": 3,
125
+ "num_hidden_layers": 12,
126
+ "pad_token_id": 1,
127
+ "patch_size": 16,
128
+ "rel_2d_pos_bins": 64,
129
+ "rel_pos_bins": 32,
130
+ "second_input_size": 112,
131
+ "shape_size": 128,
132
+ "text_embed": true,
133
+ "torch_dtype": "float32",
134
+ "transformers_version": "4.35.2",
135
+ "type_vocab_size": 1,
136
+ "visual_embed": true,
137
+ "vocab_size": 50265
138
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83f23ea33a0c173de6e7689acd8f6620ac574b5f1e745b292767626bb21c00b4
3
+ size 503841172
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "apply_ocr": false,
46
+ "bos_token": "<s>",
47
+ "clean_up_tokenization_spaces": true,
48
+ "cls_token": "<s>",
49
+ "cls_token_box": [
50
+ 0,
51
+ 0,
52
+ 0,
53
+ 0
54
+ ],
55
+ "eos_token": "</s>",
56
+ "errors": "replace",
57
+ "mask_token": "<mask>",
58
+ "model_max_length": 512,
59
+ "only_label_first_subword": true,
60
+ "pad_token": "<pad>",
61
+ "pad_token_box": [
62
+ 0,
63
+ 0,
64
+ 0,
65
+ 0
66
+ ],
67
+ "pad_token_label": -100,
68
+ "processor_class": "LayoutLMv3Processor",
69
+ "sep_token": "</s>",
70
+ "sep_token_box": [
71
+ 0,
72
+ 0,
73
+ 0,
74
+ 0
75
+ ],
76
+ "tokenizer_class": "LayoutLMv3Tokenizer",
77
+ "trim_offsets": true,
78
+ "unk_token": "<unk>"
79
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0a1427bcb7fc329ca3d7f281fd276711080bfb04d2bc59f51c455cb640f4800
3
+ size 4536
vocab.json ADDED
The diff for this file is too large to render. See raw diff