Upload 10 files
Browse files- added_tokens.json +54 -0
- artifacts.ckpt +3 -0
- config.json +24 -0
- config.yaml +29 -0
- events.out.tfevents.1747145499.15dc75f70484.40531.0 +3 -0
- hparams.yaml +1 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +18 -0
- tokenizer_config.json +23 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s_cashprice>": 57567,
|
| 3 |
+
"</s_changeprice>": 57569,
|
| 4 |
+
"</s_cnt>": 57533,
|
| 5 |
+
"</s_date>": 57527,
|
| 6 |
+
"</s_discount_price>": 57555,
|
| 7 |
+
"</s_discountprice>": 57561,
|
| 8 |
+
"</s_etc>": 57563,
|
| 9 |
+
"</s_itemsubtotal>": 57535,
|
| 10 |
+
"</s_menu>": 57529,
|
| 11 |
+
"</s_menuqty_cnt>": 57565,
|
| 12 |
+
"</s_merchant>": 57525,
|
| 13 |
+
"</s_nm>": 57531,
|
| 14 |
+
"</s_num>": 57549,
|
| 15 |
+
"</s_othersvc_price>": 57559,
|
| 16 |
+
"</s_price>": 57553,
|
| 17 |
+
"</s_recipient>": 57547,
|
| 18 |
+
"</s_service_price>": 57557,
|
| 19 |
+
"</s_subtotal>": 57537,
|
| 20 |
+
"</s_subtotal_price>": 57539,
|
| 21 |
+
"</s_tax_price>": 57541,
|
| 22 |
+
"</s_total>": 57543,
|
| 23 |
+
"</s_total_etc>": 57571,
|
| 24 |
+
"</s_total_price>": 57545,
|
| 25 |
+
"</s_unitprice>": 57551,
|
| 26 |
+
"<s_cashprice>": 57568,
|
| 27 |
+
"<s_changeprice>": 57570,
|
| 28 |
+
"<s_cnt>": 57534,
|
| 29 |
+
"<s_dataset_for_donut>": 57573,
|
| 30 |
+
"<s_date>": 57528,
|
| 31 |
+
"<s_discount_price>": 57556,
|
| 32 |
+
"<s_discountprice>": 57562,
|
| 33 |
+
"<s_etc>": 57564,
|
| 34 |
+
"<s_iitcdip>": 57523,
|
| 35 |
+
"<s_itemsubtotal>": 57536,
|
| 36 |
+
"<s_menu>": 57530,
|
| 37 |
+
"<s_menuqty_cnt>": 57566,
|
| 38 |
+
"<s_merchant>": 57526,
|
| 39 |
+
"<s_nm>": 57532,
|
| 40 |
+
"<s_num>": 57550,
|
| 41 |
+
"<s_othersvc_price>": 57560,
|
| 42 |
+
"<s_price>": 57554,
|
| 43 |
+
"<s_recipient>": 57548,
|
| 44 |
+
"<s_service_price>": 57558,
|
| 45 |
+
"<s_subtotal>": 57538,
|
| 46 |
+
"<s_subtotal_price>": 57540,
|
| 47 |
+
"<s_synthdog>": 57524,
|
| 48 |
+
"<s_tax_price>": 57542,
|
| 49 |
+
"<s_total>": 57544,
|
| 50 |
+
"<s_total_etc>": 57572,
|
| 51 |
+
"<s_total_price>": 57546,
|
| 52 |
+
"<s_unitprice>": 57552,
|
| 53 |
+
"<sep/>": 57522
|
| 54 |
+
}
|
artifacts.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a1facea4c5f2de7ed91f5791cbec4759f50475819e42dcbdcc8ce0ff54070e3
|
| 3 |
+
size 1609249232
|
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 |
+
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: '/content/drive/MyDrive/Final_Year_Project/FYP_data/donut_output'
|
| 3 |
+
pretrained_model_name_or_path: 'naver-clova-ix/donut-base'
|
| 4 |
+
dataset_name_or_paths:
|
| 5 |
+
- '/content/drive/MyDrive/Final_Year_Project/FYP_data/dataset_for_donut'
|
| 6 |
+
sort_json_key: False
|
| 7 |
+
train_batch_sizes:
|
| 8 |
+
- 6
|
| 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: 300
|
| 20 |
+
num_training_samples_per_epoch: 800
|
| 21 |
+
max_epochs: 30
|
| 22 |
+
max_steps: -1
|
| 23 |
+
num_workers: 8
|
| 24 |
+
val_check_interval: 1.0
|
| 25 |
+
check_val_every_n_epoch: 3
|
| 26 |
+
gradient_clip_val: 1.0
|
| 27 |
+
verbose: True
|
| 28 |
+
exp_name: 'train_cord'
|
| 29 |
+
exp_version: 'Training_3'
|
events.out.tfevents.1747145499.15dc75f70484.40531.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d721225ce9a3e814824ce3c8959d588b0102956e78052106a2c3924cadeeaf3
|
| 3 |
+
size 8200
|
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 sha256:9d802941ff129a3bf8f27b5b6d876399b9c5b4bf36f61d56116a860e51fb57a8
|
| 3 |
+
size 858526462
|
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_dataset_for_donut>"
|
| 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",
|
| 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 |
+
}
|