fauzansalim commited on
Commit
25c6ece
·
verified ·
1 Parent(s): b1ba992

Upload folder using huggingface_hub

Browse files
added_tokens.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s_address>": 57539,
3
+ "</s_attention>": 57573,
4
+ "</s_attn>": 57533,
5
+ "</s_base>": 57555,
6
+ "</s_city>": 57583,
7
+ "</s_company>": 57537,
8
+ "</s_credit_term>": 57571,
9
+ "</s_customer>": 57579,
10
+ "</s_customer_gfb_id>": 57569,
11
+ "</s_customer_number>": 57567,
12
+ "</s_date>": 57527,
13
+ "</s_desc>": 57543,
14
+ "</s_description>": 57589,
15
+ "</s_due_date>": 57565,
16
+ "</s_end_user>": 57531,
17
+ "</s_extended_value>": 57593,
18
+ "</s_header>": 57525,
19
+ "</s_invoice_date>": 57563,
20
+ "</s_invoice_number>": 57561,
21
+ "</s_items>": 57587,
22
+ "</s_line>": 57541,
23
+ "</s_name>": 57581,
24
+ "</s_no_bapt>": 57545,
25
+ "</s_npwp>": 57585,
26
+ "</s_number>": 57529,
27
+ "</s_pic>": 57535,
28
+ "</s_po_number>": 57575,
29
+ "</s_quantity>": 57591,
30
+ "</s_quotation_number>": 57577,
31
+ "</s_summary>": 57553,
32
+ "</s_termin>": 57547,
33
+ "</s_total>": 57559,
34
+ "</s_total_excl_vat>": 57601,
35
+ "</s_total_incl_vat>": 57607,
36
+ "</s_total_price>": 57551,
37
+ "</s_total_tax_based_value>": 57603,
38
+ "</s_total_vat_amount>": 57605,
39
+ "</s_totals>": 57599,
40
+ "</s_unit_price>": 57549,
41
+ "</s_vat>": 57557,
42
+ "</s_vat_amount>": 57597,
43
+ "</s_vat_percent>": 57595,
44
+ "<s_address>": 57540,
45
+ "<s_attention>": 57574,
46
+ "<s_attn>": 57534,
47
+ "<s_base>": 57556,
48
+ "<s_city>": 57584,
49
+ "<s_company>": 57538,
50
+ "<s_credit_term>": 57572,
51
+ "<s_customer>": 57580,
52
+ "<s_customer_gfb_id>": 57570,
53
+ "<s_customer_number>": 57568,
54
+ "<s_date>": 57528,
55
+ "<s_desc>": 57544,
56
+ "<s_description>": 57590,
57
+ "<s_due_date>": 57566,
58
+ "<s_end_user>": 57532,
59
+ "<s_extended_value>": 57594,
60
+ "<s_header>": 57526,
61
+ "<s_iitcdip>": 57523,
62
+ "<s_invoice_date>": 57564,
63
+ "<s_invoice_number>": 57562,
64
+ "<s_invoices-git>": 57609,
65
+ "<s_items>": 57588,
66
+ "<s_line>": 57542,
67
+ "<s_name>": 57582,
68
+ "<s_no_bapt>": 57546,
69
+ "<s_npwp>": 57586,
70
+ "<s_number>": 57530,
71
+ "<s_pic>": 57536,
72
+ "<s_po_number>": 57576,
73
+ "<s_quantity>": 57592,
74
+ "<s_quotation_number>": 57578,
75
+ "<s_summary>": 57554,
76
+ "<s_synthdog>": 57524,
77
+ "<s_termin>": 57548,
78
+ "<s_total>": 57560,
79
+ "<s_total_excl_vat>": 57602,
80
+ "<s_total_incl_vat>": 57608,
81
+ "<s_total_price>": 57552,
82
+ "<s_total_tax_based_value>": 57604,
83
+ "<s_total_vat_amount>": 57606,
84
+ "<s_totals>": 57600,
85
+ "<s_unit_price>": 57550,
86
+ "<s_vat>": 57558,
87
+ "<s_vat_amount>": 57598,
88
+ "<s_vat_percent>": 57596,
89
+ "<sep/>": 57522
90
+ }
artifacts.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cdef8666d3b31ddafe408903a47cd4e72f3788762940f22bece6084ae360f59
3
+ size 1609544016
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "naver-clova-ix/donut-base",
3
+ "align_long_axis": false,
4
+ "architectures": [
5
+ "DonutModel"
6
+ ],
7
+ "decoder_layer": 4,
8
+ "encoder_layer": [
9
+ 2,
10
+ 2,
11
+ 14,
12
+ 2
13
+ ],
14
+ "input_size": [
15
+ 1280,
16
+ 1600
17
+ ],
18
+ "max_length": 768,
19
+ "max_position_embeddings": 768,
20
+ "model_type": "donut",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.24.0",
23
+ "window_size": 10
24
+ }
config.yaml ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ resume_from_checkpoint_path: None
2
+ result_path: './result'
3
+ pretrained_model_name_or_path: 'naver-clova-ix/donut-base'
4
+ dataset_name_or_paths:
5
+ - 'fauzansalim/invoices-git'
6
+ sort_json_key: False
7
+ train_batch_sizes:
8
+ - 1
9
+ val_batch_sizes:
10
+ - 1
11
+ input_size:
12
+ - 1280
13
+ - 1600
14
+ max_length: 768
15
+ align_long_axis: False
16
+ num_nodes: 1
17
+ seed: 2022
18
+ lr: 3e-05
19
+ warmup_steps: 300
20
+ num_training_samples_per_epoch: 100
21
+ max_epochs: 20
22
+ max_steps: -1
23
+ num_workers: 8
24
+ val_check_interval: 1.0
25
+ check_val_every_n_epoch: 10
26
+ gradient_clip_val: 1.0
27
+ verbose: True
28
+ exp_name: 'train_sroie'
29
+ exp_version: '20250617_034905'
events.out.tfevents.1750132202.09db7e852e7f.971.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3227331a126d698d1488c0194e8f40c891e2050be0d1cd7d3455b5bd44b0561f
3
+ size 15428
hparams.yaml ADDED
@@ -0,0 +1 @@
 
 
1
+ {}
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6684296d85898654e621f789c9020a25d2d0deaaeb867b209f03105f0da0550e
3
+ size 893553918
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 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s_invoices-git>"
4
+ ],
5
+ "bos_token": "<s>",
6
+ "cls_token": "<s>",
7
+ "eos_token": "</s>",
8
+ "mask_token": {
9
+ "content": "<mask>",
10
+ "lstrip": true,
11
+ "normalized": true,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ },
15
+ "pad_token": "<pad>",
16
+ "sep_token": "</s>",
17
+ "unk_token": "<unk>"
18
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "from_slow": true,
6
+ "mask_token": {
7
+ "__type": "AddedToken",
8
+ "content": "<mask>",
9
+ "lstrip": true,
10
+ "normalized": true,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "name_or_path": "naver-clova-ix/donut-base",
15
+ "pad_token": "<pad>",
16
+ "processor_class": "DonutProcessor",
17
+ "sep_token": "</s>",
18
+ "sp_model_kwargs": {},
19
+ "special_tokens_map_file": null,
20
+ "tokenizer_class": "XLMRobertaTokenizer",
21
+ "unk_token": "<unk>"
22
+ }