doshan1250 commited on
Commit
34e840d
1 Parent(s): 6e1f083

first commit

Browse files
added_tokens.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s_address>": 57584,
3
+ "</s_cashprice>": 57549,
4
+ "</s_changeprice>": 57551,
5
+ "</s_cnt>": 57529,
6
+ "</s_company>": 57580,
7
+ "</s_creditcardprice>": 57563,
8
+ "</s_date>": 57582,
9
+ "</s_discount_price>": 57557,
10
+ "</s_discountprice>": 57567,
11
+ "</s_emoneyprice>": 57569,
12
+ "</s_etc>": 57541,
13
+ "</s_itemsubtotal>": 57577,
14
+ "</s_menu>": 57525,
15
+ "</s_menuqty_cnt>": 57555,
16
+ "</s_menutype_cnt>": 57553,
17
+ "</s_nm>": 57527,
18
+ "</s_num>": 57565,
19
+ "</s_othersvc_price>": 57573,
20
+ "</s_price>": 57531,
21
+ "</s_service_price>": 57537,
22
+ "</s_sub>": 57547,
23
+ "</s_sub_total>": 57533,
24
+ "</s_subtotal_price>": 57535,
25
+ "</s_tax_price>": 57539,
26
+ "</s_total>": 57543,
27
+ "</s_total_etc>": 57561,
28
+ "</s_total_price>": 57545,
29
+ "</s_unitprice>": 57559,
30
+ "</s_vatyn>": 57575,
31
+ "</s_void_menu>": 57571,
32
+ "<s_address>": 57585,
33
+ "<s_cashprice>": 57550,
34
+ "<s_changeprice>": 57552,
35
+ "<s_cnt>": 57530,
36
+ "<s_company>": 57581,
37
+ "<s_cord-v2>": 57579,
38
+ "<s_creditcardprice>": 57564,
39
+ "<s_date>": 57583,
40
+ "<s_discount_price>": 57558,
41
+ "<s_discountprice>": 57568,
42
+ "<s_emoneyprice>": 57570,
43
+ "<s_etc>": 57542,
44
+ "<s_iitcdip>": 57523,
45
+ "<s_itemsubtotal>": 57578,
46
+ "<s_menu>": 57526,
47
+ "<s_menuqty_cnt>": 57556,
48
+ "<s_menutype_cnt>": 57554,
49
+ "<s_nm>": 57528,
50
+ "<s_num>": 57566,
51
+ "<s_othersvc_price>": 57574,
52
+ "<s_preparedFinetuneData>": 57586,
53
+ "<s_price>": 57532,
54
+ "<s_service_price>": 57538,
55
+ "<s_sub>": 57548,
56
+ "<s_sub_total>": 57534,
57
+ "<s_subtotal_price>": 57536,
58
+ "<s_synthdog>": 57524,
59
+ "<s_tax_price>": 57540,
60
+ "<s_total>": 57544,
61
+ "<s_total_etc>": 57562,
62
+ "<s_total_price>": 57546,
63
+ "<s_unitprice>": 57560,
64
+ "<s_vatyn>": 57576,
65
+ "<s_void_menu>": 57572,
66
+ "<sep/>": 57522
67
+ }
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
+ 960
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: 'gdrive/MyDrive/result'
3
+ pretrained_model_name_or_path: 'naver-clova-ix/donut-base-finetuned-cord-v2'
4
+ dataset_name_or_paths:
5
+ - 'gdrive/MyDrive/preparedFinetuneData'
6
+ sort_json_key: False
7
+ train_batch_sizes:
8
+ - 1
9
+ val_batch_sizes:
10
+ - 1
11
+ input_size:
12
+ - 1280
13
+ - 960
14
+ max_length: 768
15
+ align_long_axis: False
16
+ num_nodes: 1
17
+ seed: 2022
18
+ lr: 3e-05
19
+ warmup_steps: 0
20
+ num_training_samples_per_epoch: 800
21
+ max_epochs: 30
22
+ max_steps: 110
23
+ num_workers: 2
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_cord_finetune'
29
+ exp_version: '20240225_054518'
events.out.tfevents.1708839928.ffc4795ce2c1.5936.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f25f5c507574d42bcf76afae689fc6d28a231868228aab401ec99fbed4e2f472
3
+ size 456
hparams.yaml ADDED
@@ -0,0 +1 @@
 
 
1
+ {}
model_checkpoint.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e2342a1e66b0ab85c3d2742ee24002bd3bb402cc1f745ed1cbbe5d16585d224
3
+ size 1609352862
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0259482434e45d16a63db6bc8eb0692bb778fa57a0e17cde673837c5c272d902
3
+ size 858582726
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_preparedFinetuneData>"
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
+ }