Altimis commited on
Commit
5001499
1 Parent(s): cdf47fc

Upload 9 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s_cashprice>": 57549,
3
+ "</s_changeprice>": 57551,
4
+ "</s_cnt>": 57529,
5
+ "</s_creditcardprice>": 57563,
6
+ "</s_customer_party_legal_id>": 57584,
7
+ "</s_discount_price>": 57557,
8
+ "</s_discountprice>": 57567,
9
+ "</s_document_currency_code>": 57588,
10
+ "</s_emoneyprice>": 57569,
11
+ "</s_etc>": 57541,
12
+ "</s_financial_account>": 57592,
13
+ "</s_invoice_number>": 57582,
14
+ "</s_issue_date>": 57580,
15
+ "</s_itemsubtotal>": 57577,
16
+ "</s_menu>": 57525,
17
+ "</s_menuqty_cnt>": 57555,
18
+ "</s_menutype_cnt>": 57553,
19
+ "</s_nm>": 57527,
20
+ "</s_num>": 57565,
21
+ "</s_order_reference>": 57598,
22
+ "</s_othersvc_price>": 57573,
23
+ "</s_payable_amount>": 57590,
24
+ "</s_price>": 57531,
25
+ "</s_service_price>": 57537,
26
+ "</s_sub>": 57547,
27
+ "</s_sub_total>": 57533,
28
+ "</s_subtotal_price>": 57535,
29
+ "</s_supplier_party_legal_id>": 57594,
30
+ "</s_tax_amount>": 57586,
31
+ "</s_tax_exclusive_amount>": 57596,
32
+ "</s_tax_price>": 57539,
33
+ "</s_total>": 57543,
34
+ "</s_total_etc>": 57561,
35
+ "</s_total_price>": 57545,
36
+ "</s_unitprice>": 57559,
37
+ "</s_vatyn>": 57575,
38
+ "</s_void_menu>": 57571,
39
+ "<s_cashprice>": 57550,
40
+ "<s_changeprice>": 57552,
41
+ "<s_cnt>": 57530,
42
+ "<s_cord-v2>": 57579,
43
+ "<s_creditcardprice>": 57564,
44
+ "<s_customer_party_legal_id>": 57585,
45
+ "<s_discount_price>": 57558,
46
+ "<s_discountprice>": 57568,
47
+ "<s_document_currency_code>": 57589,
48
+ "<s_donut_11_3_24>": 57600,
49
+ "<s_emoneyprice>": 57570,
50
+ "<s_etc>": 57542,
51
+ "<s_financial_account>": 57593,
52
+ "<s_iitcdip>": 57523,
53
+ "<s_invoice_number>": 57583,
54
+ "<s_issue_date>": 57581,
55
+ "<s_itemsubtotal>": 57578,
56
+ "<s_menu>": 57526,
57
+ "<s_menuqty_cnt>": 57556,
58
+ "<s_menutype_cnt>": 57554,
59
+ "<s_nm>": 57528,
60
+ "<s_num>": 57566,
61
+ "<s_order_reference>": 57599,
62
+ "<s_othersvc_price>": 57574,
63
+ "<s_payable_amount>": 57591,
64
+ "<s_price>": 57532,
65
+ "<s_service_price>": 57538,
66
+ "<s_sub>": 57548,
67
+ "<s_sub_total>": 57534,
68
+ "<s_subtotal_price>": 57536,
69
+ "<s_supplier_party_legal_id>": 57595,
70
+ "<s_synthdog>": 57524,
71
+ "<s_tax_amount>": 57587,
72
+ "<s_tax_exclusive_amount>": 57597,
73
+ "<s_tax_price>": 57540,
74
+ "<s_total>": 57544,
75
+ "<s_total_etc>": 57562,
76
+ "<s_total_price>": 57546,
77
+ "<s_unitprice>": 57560,
78
+ "<s_vatyn>": 57576,
79
+ "<s_void_menu>": 57572,
80
+ "<sep/>": 57522
81
+ }
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "naver-clova-ix/donut-base-finetuned-cord-v2",
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
+ 1920
17
+ ],
18
+ "max_length": 768,
19
+ "max_position_embeddings": 768,
20
+ "model_type": "donut",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.25.1",
23
+ "window_size": 10
24
+ }
config.yaml ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ resume_from_checkpoint_path: None
2
+ result_path: 'donut/result'
3
+ pretrained_model_name_or_path: 'naver-clova-ix/donut-base-finetuned-cord-v2'
4
+ dataset_name_or_paths:
5
+ - 'gdrive/MyDrive/konta-backup/konta_experiments/donut_11_3_24'
6
+ sort_json_key: False
7
+ train_batch_sizes:
8
+ - 3
9
+ val_batch_sizes:
10
+ - 1
11
+ input_size:
12
+ - 1280
13
+ - 1920
14
+ max_length: 768
15
+ align_long_axis: False
16
+ num_nodes: 1
17
+ seed: 2022
18
+ lr: 3e-05
19
+ warmup_steps: 800
20
+ num_training_samples_per_epoch: 800
21
+ max_epochs: 30
22
+ max_steps: -1
23
+ num_workers: 2
24
+ val_check_interval: 0.5
25
+ check_val_every_n_epoch: 3
26
+ gradient_clip_val: 1.0
27
+ verbose: True
28
+ exp_name: 'train_cord_finetune'
29
+ exp_version: '20240313_001018'
events.out.tfevents.1710288685.481e13bb0051.32960.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93e5d71dcaeded9d626c719eea034674942701c25892c97b5904bf1bee97790d
3
+ size 25856
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 sha256:798d895102649bdc712df1f3e0beb86e67046139860148160c47e7dd06843e37
3
+ size 910965894
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_donut_11_3_24>"
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 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "model_max_length": 1000000000000000019884624838656,
15
+ "name_or_path": "naver-clova-ix/donut-base-finetuned-cord-v2",
16
+ "pad_token": "<pad>",
17
+ "processor_class": "DonutProcessor",
18
+ "sep_token": "</s>",
19
+ "sp_model_kwargs": {},
20
+ "special_tokens_map_file": null,
21
+ "tokenizer_class": "XLMRobertaTokenizer",
22
+ "unk_token": "<unk>"
23
+ }