Hugging Face's logo

riteshbehera857
/
layoutlmv3-finetuned-ubiai-dataset-invoice-6

Token Classification
Transformers
TensorBoard
Safetensors
layoutlmv3
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
riteshbehera857 commited on
Commit
b4392be
·
verified ·
1 Parent(s): 217fedd

layoutlmv3-model-ubiai-finetuned

Browse files
Files changed (33) hide show
  1. README.md +0 -0
  2. config.json +102 -0
  3. eval_split/data-00000-of-00001.arrow +3 -0
  4. eval_split/dataset_info.json +76 -0
  5. eval_split/state.json +13 -0
  6. final_model/config.json +102 -0
  7. final_model/merges.txt +0 -0
  8. final_model/model.safetensors +3 -0
  9. final_model/preprocessor_config.json +26 -0
  10. final_model/special_tokens_map.json +51 -0
  11. final_model/tokenizer.json +0 -0
  12. final_model/tokenizer_config.json +79 -0
  13. final_model/training_args.bin +3 -0
  14. final_model/vocab.json +0 -0
  15. logs/events.out.tfevents.1728368534.17f33622ae76.757.1 +3 -0
  16. merges.txt +0 -0
  17. model.safetensors +3 -0
  18. preprocessor_config.json +26 -0
  19. raw_data/dataset_dict.json +1 -0
  20. raw_data/test/data-00000-of-00001.arrow +3 -0
  21. raw_data/test/dataset_info.json +69 -0
  22. raw_data/test/state.json +13 -0
  23. raw_data/train/data-00000-of-00001.arrow +3 -0
  24. raw_data/train/dataset_info.json +69 -0
  25. raw_data/train/state.json +13 -0
  26. special_tokens_map.json +51 -0
  27. tokenizer.json +0 -0
  28. tokenizer_config.json +79 -0
  29. train_split/data-00000-of-00001.arrow +3 -0
  30. train_split/dataset_info.json +76 -0
  31. train_split/state.json +13 -0
  32. training_args.bin +3 -0
  33. vocab.json +0 -0
README.md ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/layoutlmv3-base",
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": "B-B-ADDRESS",
18
+ "1": "B-B-NAME",
19
+ "2": "B-INVOICE NO",
20
+ "3": "B-ORDER DATE",
21
+ "4": "B-S-ADDRESS",
22
+ "5": "B-S-NAME",
23
+ "6": "B-TOTAL GROSS",
24
+ "7": "E-B-ADDRESS",
25
+ "8": "E-B-NAME",
26
+ "9": "E-INVOICE NO",
27
+ "10": "E-ORDER DATE",
28
+ "11": "E-S-ADDRESS",
29
+ "12": "E-S-NAME",
30
+ "13": "E-TOTAL GROSS",
31
+ "14": "I-B-ADDRESS",
32
+ "15": "I-B-NAME",
33
+ "16": "I-INVOICE NO",
34
+ "17": "I-ORDER DATE",
35
+ "18": "I-S-ADDRESS",
36
+ "19": "I-S-NAME",
37
+ "20": "O",
38
+ "21": "S-B-ADDRESS",
39
+ "22": "S-B-NAME",
40
+ "23": "S-GST NO",
41
+ "24": "S-INVOICE NO",
42
+ "25": "S-ORDER DATE",
43
+ "26": "S-ORDER NO",
44
+ "27": "S-S-ADDRESS",
45
+ "28": "S-TOTAL GROSS"
46
+ },
47
+ "initializer_range": 0.02,
48
+ "input_size": 224,
49
+ "intermediate_size": 3072,
50
+ "label2id": {
51
+ "B-B-ADDRESS": 0,
52
+ "B-B-NAME": 1,
53
+ "B-INVOICE NO": 2,
54
+ "B-ORDER DATE": 3,
55
+ "B-S-ADDRESS": 4,
56
+ "B-S-NAME": 5,
57
+ "B-TOTAL GROSS": 6,
58
+ "E-B-ADDRESS": 7,
59
+ "E-B-NAME": 8,
60
+ "E-INVOICE NO": 9,
61
+ "E-ORDER DATE": 10,
62
+ "E-S-ADDRESS": 11,
63
+ "E-S-NAME": 12,
64
+ "E-TOTAL GROSS": 13,
65
+ "I-B-ADDRESS": 14,
66
+ "I-B-NAME": 15,
67
+ "I-INVOICE NO": 16,
68
+ "I-ORDER DATE": 17,
69
+ "I-S-ADDRESS": 18,
70
+ "I-S-NAME": 19,
71
+ "O": 20,
72
+ "S-B-ADDRESS": 21,
73
+ "S-B-NAME": 22,
74
+ "S-GST NO": 23,
75
+ "S-INVOICE NO": 24,
76
+ "S-ORDER DATE": 25,
77
+ "S-ORDER NO": 26,
78
+ "S-S-ADDRESS": 27,
79
+ "S-TOTAL GROSS": 28
80
+ },
81
+ "layer_norm_eps": 1e-05,
82
+ "max_2d_position_embeddings": 1024,
83
+ "max_position_embeddings": 514,
84
+ "max_rel_2d_pos": 256,
85
+ "max_rel_pos": 128,
86
+ "model_type": "layoutlmv3",
87
+ "num_attention_heads": 12,
88
+ "num_channels": 3,
89
+ "num_hidden_layers": 12,
90
+ "pad_token_id": 1,
91
+ "patch_size": 16,
92
+ "rel_2d_pos_bins": 64,
93
+ "rel_pos_bins": 32,
94
+ "second_input_size": 112,
95
+ "shape_size": 128,
96
+ "text_embed": true,
97
+ "torch_dtype": "float32",
98
+ "transformers_version": "4.44.2",
99
+ "type_vocab_size": 1,
100
+ "visual_embed": true,
101
+ "vocab_size": 50265
102
+ }
eval_split/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6aaa6e406b7011d5c3cddc77bbce2241cb1464bbb8fedc9c8c82e1bb2023c0fe
3
+ size 6358320
eval_split/dataset_info.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "citation": "",
3
+ "description": "",
4
+ "features": {
5
+ "pixel_values": {
6
+ "shape": [
7
+ 3,
8
+ 224,
9
+ 224
10
+ ],
11
+ "dtype": "float32",
12
+ "_type": "Array3D"
13
+ },
14
+ "input_ids": {
15
+ "feature": {
16
+ "dtype": "int64",
17
+ "_type": "Value"
18
+ },
19
+ "_type": "Sequence"
20
+ },
21
+ "attention_mask": {
22
+ "feature": {
23
+ "dtype": "int64",
24
+ "_type": "Value"
25
+ },
26
+ "_type": "Sequence"
27
+ },
28
+ "bbox": {
29
+ "shape": [
30
+ 512,
31
+ 4
32
+ ],
33
+ "dtype": "int64",
34
+ "_type": "Array2D"
35
+ },
36
+ "labels": {
37
+ "feature": {
38
+ "names": [
39
+ "B-B-ADDRESS",
40
+ "B-B-NAME",
41
+ "B-INVOICE NO",
42
+ "B-ORDER DATE",
43
+ "B-S-ADDRESS",
44
+ "B-S-NAME",
45
+ "B-TOTAL GROSS",
46
+ "E-B-ADDRESS",
47
+ "E-B-NAME",
48
+ "E-INVOICE NO",
49
+ "E-ORDER DATE",
50
+ "E-S-ADDRESS",
51
+ "E-S-NAME",
52
+ "E-TOTAL GROSS",
53
+ "I-B-ADDRESS",
54
+ "I-B-NAME",
55
+ "I-INVOICE NO",
56
+ "I-ORDER DATE",
57
+ "I-S-ADDRESS",
58
+ "I-S-NAME",
59
+ "O",
60
+ "S-B-ADDRESS",
61
+ "S-B-NAME",
62
+ "S-GST NO",
63
+ "S-INVOICE NO",
64
+ "S-ORDER DATE",
65
+ "S-ORDER NO",
66
+ "S-S-ADDRESS",
67
+ "S-TOTAL GROSS"
68
+ ],
69
+ "_type": "ClassLabel"
70
+ },
71
+ "_type": "Sequence"
72
+ }
73
+ },
74
+ "homepage": "",
75
+ "license": ""
76
+ }
eval_split/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "9fb4f7eb82d3d0b0",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": "torch",
11
+ "_output_all_columns": false,
12
+ "_split": null
13
+ }
final_model/config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/layoutlmv3-base",
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": "B-B-ADDRESS",
18
+ "1": "B-B-NAME",
19
+ "2": "B-INVOICE NO",
20
+ "3": "B-ORDER DATE",
21
+ "4": "B-S-ADDRESS",
22
+ "5": "B-S-NAME",
23
+ "6": "B-TOTAL GROSS",
24
+ "7": "E-B-ADDRESS",
25
+ "8": "E-B-NAME",
26
+ "9": "E-INVOICE NO",
27
+ "10": "E-ORDER DATE",
28
+ "11": "E-S-ADDRESS",
29
+ "12": "E-S-NAME",
30
+ "13": "E-TOTAL GROSS",
31
+ "14": "I-B-ADDRESS",
32
+ "15": "I-B-NAME",
33
+ "16": "I-INVOICE NO",
34
+ "17": "I-ORDER DATE",
35
+ "18": "I-S-ADDRESS",
36
+ "19": "I-S-NAME",
37
+ "20": "O",
38
+ "21": "S-B-ADDRESS",
39
+ "22": "S-B-NAME",
40
+ "23": "S-GST NO",
41
+ "24": "S-INVOICE NO",
42
+ "25": "S-ORDER DATE",
43
+ "26": "S-ORDER NO",
44
+ "27": "S-S-ADDRESS",
45
+ "28": "S-TOTAL GROSS"
46
+ },
47
+ "initializer_range": 0.02,
48
+ "input_size": 224,
49
+ "intermediate_size": 3072,
50
+ "label2id": {
51
+ "B-B-ADDRESS": 0,
52
+ "B-B-NAME": 1,
53
+ "B-INVOICE NO": 2,
54
+ "B-ORDER DATE": 3,
55
+ "B-S-ADDRESS": 4,
56
+ "B-S-NAME": 5,
57
+ "B-TOTAL GROSS": 6,
58
+ "E-B-ADDRESS": 7,
59
+ "E-B-NAME": 8,
60
+ "E-INVOICE NO": 9,
61
+ "E-ORDER DATE": 10,
62
+ "E-S-ADDRESS": 11,
63
+ "E-S-NAME": 12,
64
+ "E-TOTAL GROSS": 13,
65
+ "I-B-ADDRESS": 14,
66
+ "I-B-NAME": 15,
67
+ "I-INVOICE NO": 16,
68
+ "I-ORDER DATE": 17,
69
+ "I-S-ADDRESS": 18,
70
+ "I-S-NAME": 19,
71
+ "O": 20,
72
+ "S-B-ADDRESS": 21,
73
+ "S-B-NAME": 22,
74
+ "S-GST NO": 23,
75
+ "S-INVOICE NO": 24,
76
+ "S-ORDER DATE": 25,
77
+ "S-ORDER NO": 26,
78
+ "S-S-ADDRESS": 27,
79
+ "S-TOTAL GROSS": 28
80
+ },
81
+ "layer_norm_eps": 1e-05,
82
+ "max_2d_position_embeddings": 1024,
83
+ "max_position_embeddings": 514,
84
+ "max_rel_2d_pos": 256,
85
+ "max_rel_pos": 128,
86
+ "model_type": "layoutlmv3",
87
+ "num_attention_heads": 12,
88
+ "num_channels": 3,
89
+ "num_hidden_layers": 12,
90
+ "pad_token_id": 1,
91
+ "patch_size": 16,
92
+ "rel_2d_pos_bins": 64,
93
+ "rel_pos_bins": 32,
94
+ "second_input_size": 112,
95
+ "shape_size": 128,
96
+ "text_embed": true,
97
+ "torch_dtype": "float32",
98
+ "transformers_version": "4.44.2",
99
+ "type_vocab_size": 1,
100
+ "visual_embed": true,
101
+ "vocab_size": 50265
102
+ }
final_model/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
final_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a307b5e5008e63f46a87f6a9da22ebc72b4e13beccc0663a06e909fc989af234
3
+ size 503785804
final_model/preprocessor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_ocr": false,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "LayoutLMv3ImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "ocr_lang": null,
18
+ "processor_class": "LayoutLMv3Processor",
19
+ "resample": 2,
20
+ "rescale_factor": 0.00392156862745098,
21
+ "size": {
22
+ "height": 224,
23
+ "width": 224
24
+ },
25
+ "tesseract_config": ""
26
+ }
final_model/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
+ }
final_model/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
final_model/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
+ }
final_model/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5aa391e0bf6eadba88c56f040bc5f393a6fd979a26984317839ea734e4efb1f
3
+ size 5240
final_model/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
logs/events.out.tfevents.1728368534.17f33622ae76.757.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c4aa4a4857e6caa70cd33df3f0022a77c4cafda0a2e34c1907c13257788a2eb
3
+ size 37186
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:a307b5e5008e63f46a87f6a9da22ebc72b4e13beccc0663a06e909fc989af234
3
+ size 503785804
preprocessor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_ocr": false,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "LayoutLMv3ImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "ocr_lang": null,
18
+ "processor_class": "LayoutLMv3Processor",
19
+ "resample": 2,
20
+ "rescale_factor": 0.00392156862745098,
21
+ "size": {
22
+ "height": 224,
23
+ "width": 224
24
+ },
25
+ "tesseract_config": ""
26
+ }
raw_data/dataset_dict.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"splits": ["train", "test"]}
raw_data/test/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5908503d5506f4fe2f400057e1339d89eed9444239b77a58afd9edf30eb7c01c
3
+ size 6532312
raw_data/test/dataset_info.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "citation": "",
3
+ "description": "",
4
+ "features": {
5
+ "id": {
6
+ "dtype": "string",
7
+ "_type": "Value"
8
+ },
9
+ "tokens": {
10
+ "feature": {
11
+ "dtype": "string",
12
+ "_type": "Value"
13
+ },
14
+ "_type": "Sequence"
15
+ },
16
+ "bboxes": {
17
+ "feature": {
18
+ "feature": {
19
+ "dtype": "int64",
20
+ "_type": "Value"
21
+ },
22
+ "_type": "Sequence"
23
+ },
24
+ "_type": "Sequence"
25
+ },
26
+ "ner_tags": {
27
+ "feature": {
28
+ "names": [
29
+ "B-B-ADDRESS",
30
+ "B-B-NAME",
31
+ "B-INVOICE NO",
32
+ "B-ORDER DATE",
33
+ "B-S-ADDRESS",
34
+ "B-S-NAME",
35
+ "B-TOTAL GROSS",
36
+ "E-B-ADDRESS",
37
+ "E-B-NAME",
38
+ "E-INVOICE NO",
39
+ "E-ORDER DATE",
40
+ "E-S-ADDRESS",
41
+ "E-S-NAME",
42
+ "E-TOTAL GROSS",
43
+ "I-B-ADDRESS",
44
+ "I-B-NAME",
45
+ "I-INVOICE NO",
46
+ "I-ORDER DATE",
47
+ "I-S-ADDRESS",
48
+ "I-S-NAME",
49
+ "O",
50
+ "S-B-ADDRESS",
51
+ "S-B-NAME",
52
+ "S-GST NO",
53
+ "S-INVOICE NO",
54
+ "S-ORDER DATE",
55
+ "S-ORDER NO",
56
+ "S-S-ADDRESS",
57
+ "S-TOTAL GROSS"
58
+ ],
59
+ "_type": "ClassLabel"
60
+ },
61
+ "_type": "Sequence"
62
+ },
63
+ "image": {
64
+ "_type": "Image"
65
+ }
66
+ },
67
+ "homepage": "",
68
+ "license": ""
69
+ }
raw_data/test/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "15ed5fda6ceb8a3f",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": null
13
+ }
raw_data/train/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1da60b1c0d26f846a92d8963fb32f4f87f0bab8a0bedee8d787a3165fa5adfa9
3
+ size 56378416
raw_data/train/dataset_info.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "citation": "",
3
+ "description": "",
4
+ "features": {
5
+ "id": {
6
+ "dtype": "string",
7
+ "_type": "Value"
8
+ },
9
+ "tokens": {
10
+ "feature": {
11
+ "dtype": "string",
12
+ "_type": "Value"
13
+ },
14
+ "_type": "Sequence"
15
+ },
16
+ "bboxes": {
17
+ "feature": {
18
+ "feature": {
19
+ "dtype": "int64",
20
+ "_type": "Value"
21
+ },
22
+ "_type": "Sequence"
23
+ },
24
+ "_type": "Sequence"
25
+ },
26
+ "ner_tags": {
27
+ "feature": {
28
+ "names": [
29
+ "B-B-ADDRESS",
30
+ "B-B-NAME",
31
+ "B-INVOICE NO",
32
+ "B-ORDER DATE",
33
+ "B-S-ADDRESS",
34
+ "B-S-NAME",
35
+ "B-TOTAL GROSS",
36
+ "E-B-ADDRESS",
37
+ "E-B-NAME",
38
+ "E-INVOICE NO",
39
+ "E-ORDER DATE",
40
+ "E-S-ADDRESS",
41
+ "E-S-NAME",
42
+ "E-TOTAL GROSS",
43
+ "I-B-ADDRESS",
44
+ "I-B-NAME",
45
+ "I-INVOICE NO",
46
+ "I-ORDER DATE",
47
+ "I-S-ADDRESS",
48
+ "I-S-NAME",
49
+ "O",
50
+ "S-B-ADDRESS",
51
+ "S-B-NAME",
52
+ "S-GST NO",
53
+ "S-INVOICE NO",
54
+ "S-ORDER DATE",
55
+ "S-ORDER NO",
56
+ "S-S-ADDRESS",
57
+ "S-TOTAL GROSS"
58
+ ],
59
+ "_type": "ClassLabel"
60
+ },
61
+ "_type": "Sequence"
62
+ },
63
+ "image": {
64
+ "_type": "Image"
65
+ }
66
+ },
67
+ "homepage": "",
68
+ "license": ""
69
+ }
raw_data/train/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "6af9ffbb94da8a7f",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": null
13
+ }
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
+ }
train_split/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fc36993bded307d30797ecdb0eb330a0e2b85772904930d8a1d9242ed3be3a0
3
+ size 57202480
train_split/dataset_info.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "citation": "",
3
+ "description": "",
4
+ "features": {
5
+ "pixel_values": {
6
+ "shape": [
7
+ 3,
8
+ 224,
9
+ 224
10
+ ],
11
+ "dtype": "float32",
12
+ "_type": "Array3D"
13
+ },
14
+ "input_ids": {
15
+ "feature": {
16
+ "dtype": "int64",
17
+ "_type": "Value"
18
+ },
19
+ "_type": "Sequence"
20
+ },
21
+ "attention_mask": {
22
+ "feature": {
23
+ "dtype": "int64",
24
+ "_type": "Value"
25
+ },
26
+ "_type": "Sequence"
27
+ },
28
+ "bbox": {
29
+ "shape": [
30
+ 512,
31
+ 4
32
+ ],
33
+ "dtype": "int64",
34
+ "_type": "Array2D"
35
+ },
36
+ "labels": {
37
+ "feature": {
38
+ "names": [
39
+ "B-B-ADDRESS",
40
+ "B-B-NAME",
41
+ "B-INVOICE NO",
42
+ "B-ORDER DATE",
43
+ "B-S-ADDRESS",
44
+ "B-S-NAME",
45
+ "B-TOTAL GROSS",
46
+ "E-B-ADDRESS",
47
+ "E-B-NAME",
48
+ "E-INVOICE NO",
49
+ "E-ORDER DATE",
50
+ "E-S-ADDRESS",
51
+ "E-S-NAME",
52
+ "E-TOTAL GROSS",
53
+ "I-B-ADDRESS",
54
+ "I-B-NAME",
55
+ "I-INVOICE NO",
56
+ "I-ORDER DATE",
57
+ "I-S-ADDRESS",
58
+ "I-S-NAME",
59
+ "O",
60
+ "S-B-ADDRESS",
61
+ "S-B-NAME",
62
+ "S-GST NO",
63
+ "S-INVOICE NO",
64
+ "S-ORDER DATE",
65
+ "S-ORDER NO",
66
+ "S-S-ADDRESS",
67
+ "S-TOTAL GROSS"
68
+ ],
69
+ "_type": "ClassLabel"
70
+ },
71
+ "_type": "Sequence"
72
+ }
73
+ },
74
+ "homepage": "",
75
+ "license": ""
76
+ }
train_split/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "c7986c6f095016a8",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": "torch",
11
+ "_output_all_columns": false,
12
+ "_split": null
13
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5aa391e0bf6eadba88c56f040bc5f393a6fd979a26984317839ea734e4efb1f
3
+ size 5240
vocab.json ADDED
The diff for this file is too large to render. See raw diff