Hugging Face's logo

Elamranilyas
/
CORDV2BOX

Image-Text-to-Text
Transformers
PyTorch
vision-encoder-decoder
Model card Files
xet
 Community
1
Elamranilyas commited on
Commit
cb97ca1
·
1 Parent(s): f0949e2

Training done

Browse files
Files changed (6) hide show
  1. added_tokens.json +116 -0
  2. preprocessor_config.json +26 -0
  3. sentencepiece.bpe.model +3 -0
  4. special_tokens_map.json +19 -0
  5. tokenizer.json +0 -0
  6. tokenizer_config.json +21 -0
added_tokens.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s_cashprice>": 57550,
3
+ "</s_category>": 57625,
4
+ "</s_changeprice>": 57552,
5
+ "</s_cnt>": 57530,
6
+ "</s_creditcardprice>": 57564,
7
+ "</s_discount_price>": 57558,
8
+ "</s_discountprice>": 57568,
9
+ "</s_dontcare>": 57635,
10
+ "</s_emoneyprice>": 57570,
11
+ "</s_etc>": 57542,
12
+ "</s_group_id>": 57627,
13
+ "</s_gt_parse>": 57581,
14
+ "</s_height>": 57595,
15
+ "</s_image_id>": 57589,
16
+ "</s_image_size>": 57591,
17
+ "</s_is_key>": 57619,
18
+ "</s_itemsubtotal>": 57578,
19
+ "</s_menu>": 57526,
20
+ "</s_menuqty_cnt>": 57556,
21
+ "</s_menutype_cnt>": 57554,
22
+ "</s_meta>": 57583,
23
+ "</s_nm>": 57528,
24
+ "</s_num>": 57566,
25
+ "</s_othersvc_price>": 57574,
26
+ "</s_price>": 57532,
27
+ "</s_quad>": 57601,
28
+ "</s_repeating_symbol>": 57633,
29
+ "</s_roi>": 57631,
30
+ "</s_row_id>": 57621,
31
+ "</s_service_price>": 57538,
32
+ "</s_split>": 57587,
33
+ "</s_sub>": 57548,
34
+ "</s_sub_group_id>": 57629,
35
+ "</s_sub_total>": 57534,
36
+ "</s_subtotal_price>": 57536,
37
+ "</s_tax_price>": 57540,
38
+ "</s_text>": 57623,
39
+ "</s_total>": 57544,
40
+ "</s_total_etc>": 57562,
41
+ "</s_total_price>": 57546,
42
+ "</s_unitprice>": 57560,
43
+ "</s_valid_line>": 57597,
44
+ "</s_vatyn>": 57576,
45
+ "</s_version>": 57585,
46
+ "</s_void_menu>": 57572,
47
+ "</s_width>": 57593,
48
+ "</s_words>": 57599,
49
+ "</s_x1>": 57611,
50
+ "</s_x2>": 57603,
51
+ "</s_x3>": 57607,
52
+ "</s_x4>": 57615,
53
+ "</s_y1>": 57613,
54
+ "</s_y2>": 57617,
55
+ "</s_y3>": 57605,
56
+ "</s_y4>": 57609,
57
+ "<s_cashprice>": 57549,
58
+ "<s_category>": 57624,
59
+ "<s_changeprice>": 57551,
60
+ "<s_cnt>": 57529,
61
+ "<s_cord-v2>": 57579,
62
+ "<s_creditcardprice>": 57563,
63
+ "<s_discount_price>": 57557,
64
+ "<s_discountprice>": 57567,
65
+ "<s_dontcare>": 57634,
66
+ "<s_emoneyprice>": 57569,
67
+ "<s_etc>": 57541,
68
+ "<s_group_id>": 57626,
69
+ "<s_gt_parse>": 57580,
70
+ "<s_height>": 57594,
71
+ "<s_iitcdip>": 57523,
72
+ "<s_image_id>": 57588,
73
+ "<s_image_size>": 57590,
74
+ "<s_is_key>": 57618,
75
+ "<s_itemsubtotal>": 57577,
76
+ "<s_menu>": 57525,
77
+ "<s_menuqty_cnt>": 57555,
78
+ "<s_menutype_cnt>": 57553,
79
+ "<s_meta>": 57582,
80
+ "<s_nm>": 57527,
81
+ "<s_num>": 57565,
82
+ "<s_othersvc_price>": 57573,
83
+ "<s_price>": 57531,
84
+ "<s_quad>": 57600,
85
+ "<s_repeating_symbol>": 57632,
86
+ "<s_roi>": 57630,
87
+ "<s_row_id>": 57620,
88
+ "<s_service_price>": 57537,
89
+ "<s_split>": 57586,
90
+ "<s_sub>": 57547,
91
+ "<s_sub_group_id>": 57628,
92
+ "<s_sub_total>": 57533,
93
+ "<s_subtotal_price>": 57535,
94
+ "<s_synthdog>": 57524,
95
+ "<s_tax_price>": 57539,
96
+ "<s_text>": 57622,
97
+ "<s_total>": 57543,
98
+ "<s_total_etc>": 57561,
99
+ "<s_total_price>": 57545,
100
+ "<s_unitprice>": 57559,
101
+ "<s_valid_line>": 57596,
102
+ "<s_vatyn>": 57575,
103
+ "<s_version>": 57584,
104
+ "<s_void_menu>": 57571,
105
+ "<s_width>": 57592,
106
+ "<s_words>": 57598,
107
+ "<s_x1>": 57610,
108
+ "<s_x2>": 57602,
109
+ "<s_x3>": 57606,
110
+ "<s_x4>": 57614,
111
+ "<s_y1>": 57612,
112
+ "<s_y2>": 57616,
113
+ "<s_y3>": 57604,
114
+ "<s_y4>": 57608,
115
+ "<sep/>": 57522
116
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_align_long_axis": false,
3
+ "do_normalize": true,
4
+ "do_pad": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "do_thumbnail": true,
8
+ "image_mean": [
9
+ 0.5,
10
+ 0.5,
11
+ 0.5
12
+ ],
13
+ "image_processor_type": "DonutImageProcessor",
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "DonutProcessor",
20
+ "resample": 2,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "size": [
23
+ 960,
24
+ 1280
25
+ ]
26
+ }
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
3
+ size 1296245
special_tokens_map.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s_iitcdip>",
4
+ "<s_synthdog>"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "cls_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "mask_token": {
10
+ "content": "<mask>",
11
+ "lstrip": true,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "unk_token": "<unk>"
19
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "<s>",
5
+ "eos_token": "</s>",
6
+ "mask_token": {
7
+ "__type": "AddedToken",
8
+ "content": "<mask>",
9
+ "lstrip": true,
10
+ "normalized": true,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "model_max_length": 1000000000000000019884624838656,
15
+ "pad_token": "<pad>",
16
+ "processor_class": "DonutProcessor",
17
+ "sep_token": "</s>",
18
+ "sp_model_kwargs": {},
19
+ "tokenizer_class": "XLMRobertaTokenizer",
20
+ "unk_token": "<unk>"
21
+ }