Initial fine-tuned model upload
Browse files- config.json +146 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- preprocessor_config.json +26 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +87 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,146 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LayoutLMv3ForTokenClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"coordinate_size": 128,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"has_relative_attention_bias": true,
|
| 11 |
+
"has_spatial_attention_bias": true,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "O",
|
| 17 |
+
"1": "B-vendor_name",
|
| 18 |
+
"2": "I-vendor_name",
|
| 19 |
+
"3": "B-vendor_address",
|
| 20 |
+
"4": "I-vendor_address",
|
| 21 |
+
"5": "B-vendor_phone_number",
|
| 22 |
+
"6": "I-vendor_phone_number",
|
| 23 |
+
"7": "B-date",
|
| 24 |
+
"8": "I-date",
|
| 25 |
+
"9": "B-time",
|
| 26 |
+
"10": "I-time",
|
| 27 |
+
"11": "B-receipt_id",
|
| 28 |
+
"12": "I-receipt_id",
|
| 29 |
+
"13": "B-currency",
|
| 30 |
+
"14": "I-currency",
|
| 31 |
+
"15": "B-total_amount",
|
| 32 |
+
"16": "I-total_amount",
|
| 33 |
+
"17": "B-subtotal_amount",
|
| 34 |
+
"18": "I-subtotal_amount",
|
| 35 |
+
"19": "B-tax_amount",
|
| 36 |
+
"20": "I-tax_amount",
|
| 37 |
+
"21": "B-service_charge_amount",
|
| 38 |
+
"22": "I-service_charge_amount",
|
| 39 |
+
"23": "B-discount_amount",
|
| 40 |
+
"24": "I-discount_amount",
|
| 41 |
+
"25": "B-tip_amount",
|
| 42 |
+
"26": "I-tip_amount",
|
| 43 |
+
"27": "B-cash_paid_amount",
|
| 44 |
+
"28": "I-cash_paid_amount",
|
| 45 |
+
"29": "B-change_amount",
|
| 46 |
+
"30": "I-change_amount",
|
| 47 |
+
"31": "B-credit_card_amount",
|
| 48 |
+
"32": "I-credit_card_amount",
|
| 49 |
+
"33": "B-e_money_amount",
|
| 50 |
+
"34": "I-e_money_amount",
|
| 51 |
+
"35": "B-payment_method",
|
| 52 |
+
"36": "I-payment_method",
|
| 53 |
+
"37": "B-line_item_name",
|
| 54 |
+
"38": "I-line_item_name",
|
| 55 |
+
"39": "B-line_item_quantity",
|
| 56 |
+
"40": "I-line_item_quantity",
|
| 57 |
+
"41": "B-line_item_unit_price",
|
| 58 |
+
"42": "I-line_item_unit_price",
|
| 59 |
+
"43": "B-line_item_total_price",
|
| 60 |
+
"44": "I-line_item_total_price",
|
| 61 |
+
"45": "B-line_item_discount_amount",
|
| 62 |
+
"46": "I-line_item_discount_amount",
|
| 63 |
+
"47": "B-line_item_vat_status",
|
| 64 |
+
"48": "I-line_item_vat_status",
|
| 65 |
+
"49": "B-other",
|
| 66 |
+
"50": "I-other"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"input_size": 224,
|
| 70 |
+
"intermediate_size": 3072,
|
| 71 |
+
"label2id": {
|
| 72 |
+
"B-cash_paid_amount": 27,
|
| 73 |
+
"B-change_amount": 29,
|
| 74 |
+
"B-credit_card_amount": 31,
|
| 75 |
+
"B-currency": 13,
|
| 76 |
+
"B-date": 7,
|
| 77 |
+
"B-discount_amount": 23,
|
| 78 |
+
"B-e_money_amount": 33,
|
| 79 |
+
"B-line_item_discount_amount": 45,
|
| 80 |
+
"B-line_item_name": 37,
|
| 81 |
+
"B-line_item_quantity": 39,
|
| 82 |
+
"B-line_item_total_price": 43,
|
| 83 |
+
"B-line_item_unit_price": 41,
|
| 84 |
+
"B-line_item_vat_status": 47,
|
| 85 |
+
"B-other": 49,
|
| 86 |
+
"B-payment_method": 35,
|
| 87 |
+
"B-receipt_id": 11,
|
| 88 |
+
"B-service_charge_amount": 21,
|
| 89 |
+
"B-subtotal_amount": 17,
|
| 90 |
+
"B-tax_amount": 19,
|
| 91 |
+
"B-time": 9,
|
| 92 |
+
"B-tip_amount": 25,
|
| 93 |
+
"B-total_amount": 15,
|
| 94 |
+
"B-vendor_address": 3,
|
| 95 |
+
"B-vendor_name": 1,
|
| 96 |
+
"B-vendor_phone_number": 5,
|
| 97 |
+
"I-cash_paid_amount": 28,
|
| 98 |
+
"I-change_amount": 30,
|
| 99 |
+
"I-credit_card_amount": 32,
|
| 100 |
+
"I-currency": 14,
|
| 101 |
+
"I-date": 8,
|
| 102 |
+
"I-discount_amount": 24,
|
| 103 |
+
"I-e_money_amount": 34,
|
| 104 |
+
"I-line_item_discount_amount": 46,
|
| 105 |
+
"I-line_item_name": 38,
|
| 106 |
+
"I-line_item_quantity": 40,
|
| 107 |
+
"I-line_item_total_price": 44,
|
| 108 |
+
"I-line_item_unit_price": 42,
|
| 109 |
+
"I-line_item_vat_status": 48,
|
| 110 |
+
"I-other": 50,
|
| 111 |
+
"I-payment_method": 36,
|
| 112 |
+
"I-receipt_id": 12,
|
| 113 |
+
"I-service_charge_amount": 22,
|
| 114 |
+
"I-subtotal_amount": 18,
|
| 115 |
+
"I-tax_amount": 20,
|
| 116 |
+
"I-time": 10,
|
| 117 |
+
"I-tip_amount": 26,
|
| 118 |
+
"I-total_amount": 16,
|
| 119 |
+
"I-vendor_address": 4,
|
| 120 |
+
"I-vendor_name": 2,
|
| 121 |
+
"I-vendor_phone_number": 6,
|
| 122 |
+
"O": 0
|
| 123 |
+
},
|
| 124 |
+
"layer_norm_eps": 1e-05,
|
| 125 |
+
"max_2d_position_embeddings": 1024,
|
| 126 |
+
"max_position_embeddings": 514,
|
| 127 |
+
"max_rel_2d_pos": 256,
|
| 128 |
+
"max_rel_pos": 128,
|
| 129 |
+
"model_type": "layoutlmv3",
|
| 130 |
+
"num_attention_heads": 12,
|
| 131 |
+
"num_channels": 3,
|
| 132 |
+
"num_hidden_layers": 12,
|
| 133 |
+
"pad_token_id": 1,
|
| 134 |
+
"patch_size": 16,
|
| 135 |
+
"rel_2d_pos_bins": 64,
|
| 136 |
+
"rel_pos_bins": 32,
|
| 137 |
+
"second_input_size": 112,
|
| 138 |
+
"shape_size": 128,
|
| 139 |
+
"text_embed": true,
|
| 140 |
+
"torch_dtype": "float32",
|
| 141 |
+
"transformers_version": "4.53.3",
|
| 142 |
+
"type_vocab_size": 1,
|
| 143 |
+
"use_cache": false,
|
| 144 |
+
"visual_embed": true,
|
| 145 |
+
"vocab_size": 50265
|
| 146 |
+
}
|
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:f16a7d271a6782cc336d7c281593aee36225f9719009ca1b5cfe8e8bfc46c4ad
|
| 3 |
+
size 503853476
|
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 |
+
}
|
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 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 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 |
+
"extra_special_tokens": {},
|
| 58 |
+
"mask_token": "<mask>",
|
| 59 |
+
"max_length": 512,
|
| 60 |
+
"model_max_length": 512,
|
| 61 |
+
"only_label_first_subword": true,
|
| 62 |
+
"pad_to_multiple_of": null,
|
| 63 |
+
"pad_token": "<pad>",
|
| 64 |
+
"pad_token_box": [
|
| 65 |
+
0,
|
| 66 |
+
0,
|
| 67 |
+
0,
|
| 68 |
+
0
|
| 69 |
+
],
|
| 70 |
+
"pad_token_label": -100,
|
| 71 |
+
"pad_token_type_id": 0,
|
| 72 |
+
"padding_side": "right",
|
| 73 |
+
"processor_class": "LayoutLMv3Processor",
|
| 74 |
+
"sep_token": "</s>",
|
| 75 |
+
"sep_token_box": [
|
| 76 |
+
0,
|
| 77 |
+
0,
|
| 78 |
+
0,
|
| 79 |
+
0
|
| 80 |
+
],
|
| 81 |
+
"stride": 0,
|
| 82 |
+
"tokenizer_class": "LayoutLMv3Tokenizer",
|
| 83 |
+
"trim_offsets": true,
|
| 84 |
+
"truncation_side": "right",
|
| 85 |
+
"truncation_strategy": "longest_first",
|
| 86 |
+
"unk_token": "<unk>"
|
| 87 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|