Update README.md
Browse files
README.md
CHANGED
|
@@ -3,167 +3,9 @@ license: apache-2.0
|
|
| 3 |
pipeline_tag: image-to-text
|
| 4 |
---
|
| 5 |
|
| 6 |
-
Model Architecture:
|
| 7 |
-
|
| 8 |
-
The mychen76/invoice-and-receipts_donut_v1 (LLM) is a finetuned for convert Invoice or Receipt Image to XML or Json data strucutre task. this experimental model is based on Donut model.
|
| 9 |
-
|
| 10 |
-
Motivation:
|
| 11 |
-
|
| 12 |
-
Remove OCR engine, use only LLM model to convert an invoice or receipt json object could reduce the conversion step and reduce resource utilization and deployment dependencies. Result, better performance.
|
| 13 |
-
|
| 14 |
-
Model Usage:
|
| 15 |
-
|
| 16 |
-
Take following an invoice receipt image and get an output Json or xml like this:
|
| 17 |
|
| 18 |
***JSON OUTPUT***
|
| 19 |
```json
|
| 20 |
|
| 21 |
-
{
|
| 22 |
-
'header': {
|
| 23 |
-
'invoice_no': '13194726',
|
| 24 |
-
'invoice_date': '05/29/2021',
|
| 25 |
-
'seller': 'Hopkins and Sons 62283 Flores Tunnel North Luis, IA 69983',
|
| 26 |
-
'client': 'Sims PLC USS Kramer FPO AA 81651',
|
| 27 |
-
'seller_tax_id': '952-73-7223',
|
| 28 |
-
'client_tax_id': '995-88-9495',
|
| 29 |
-
'iban': 'GB31LZX520242755934691'
|
| 30 |
-
},
|
| 31 |
-
'items': [
|
| 32 |
-
{
|
| 33 |
-
'item_desc': 'Beach Lunch Lounge Striped Shirt Dress Large Navy Blue White Long Sleeve Casual',
|
| 34 |
-
'item_qty': '1,00',
|
| 35 |
-
'item_net_price': '16,99',
|
| 36 |
-
'item_net_worth': '16,99',
|
| 37 |
-
'item_vat': '10%',
|
| 38 |
-
'item_gross_worth': '18,69'
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
'item_desc': 'Jams World Hawaiian 0 Dress Rayon SZ.L',
|
| 42 |
-
'item_qty': '5,00',
|
| 43 |
-
'item_net_price': '65,00',
|
| 44 |
-
'item_net_worth': '325,00',
|
| 45 |
-
'item_vat': '10%',
|
| 46 |
-
'item_gross_worth': '357,50'
|
| 47 |
-
},
|
| 48 |
-
{
|
| 49 |
-
'item_desc': 'LuLaRoe Nicole Dress Size Large 26',
|
| 50 |
-
'item_qty': '2,00',
|
| 51 |
-
'item_net_price': '1,99',
|
| 52 |
-
'item_net_worth': '3,98',
|
| 53 |
-
'item_vat': '10%',
|
| 54 |
-
'item_gross_worth': '4,38'
|
| 55 |
-
},
|
| 56 |
-
{
|
| 57 |
-
'item_desc': 'phynny Was Medium Linen Wrap Dress Dessert Rose Embroidered Bohemian',
|
| 58 |
-
'item_qty': '2,00',
|
| 59 |
-
'item_net_price': '89,99',
|
| 60 |
-
'item_net_worth': '179,98',
|
| 61 |
-
'item_vat': '10%',
|
| 62 |
-
'item_gross_worth': '197,98'
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
'item_desc': "Eileen Fisher Women's Long Sleeve Fleece Lined Front Pockets Dress XS Gray",
|
| 66 |
-
'item_qty': '2,00',
|
| 67 |
-
'item_net_price': '15,99',
|
| 68 |
-
'item_net_worth': '31,98',
|
| 69 |
-
'item_vat': '10%',
|
| 70 |
-
'item_gross_worth': '35,18'
|
| 71 |
-
},
|
| 72 |
-
{
|
| 73 |
-
'item_desc': "Hanna Anderson Women's L Large Coral Short Sleeve Casual Fall Tee Shirt Dress",
|
| 74 |
-
'item_qty': '1,00',
|
| 75 |
-
'item_net_price': '24,00',
|
| 76 |
-
'item_net_worth': '24,00',
|
| 77 |
-
'item_vat': '10%',
|
| 78 |
-
'item_gross_worth': '26,40'
|
| 79 |
-
}
|
| 80 |
-
],
|
| 81 |
-
'summary': {'total_net_worth': '$581,93', 'total_vat': '$58,19', 'total_gross_worth': '$ 640,12'}
|
| 82 |
-
}
|
| 83 |
-
|
| 84 |
-
```
|
| 85 |
-
|
| 86 |
-
***XML OUTPUT***
|
| 87 |
-
|
| 88 |
-
```xml
|
| 89 |
-
<s_header>
|
| 90 |
-
<s_invoice_no> 13194726</s_invoice_no>
|
| 91 |
-
<s_invoice_date> 05/29/2021</s_invoice_date>
|
| 92 |
-
<s_seller> Hopkins and
|
| 93 |
-
Sons 62283 Flores Tunnel North Luis, IA 69983</s_seller>
|
| 94 |
-
<s_client> Sims PLC USS Kramer FPO AA
|
| 95 |
-
81651</s_client>
|
| 96 |
-
<s_seller_tax_id> 952-73-7223</s_seller_tax_id>
|
| 97 |
-
<s_client_tax_id>
|
| 98 |
-
995-88-9495</s_client_tax_id>
|
| 99 |
-
<s_iban> GB31LZX520242755934691</s_iban>
|
| 100 |
-
</s_header>
|
| 101 |
-
<s_items>
|
| 102 |
-
<s_item_desc> Beach Lunch
|
| 103 |
-
Lounge Striped Shirt Dress Large Navy Blue White Long Sleeve Casual</s_item_desc>
|
| 104 |
-
<s_item_qty>
|
| 105 |
-
1,00</s_item_qty>
|
| 106 |
-
<s_item_net_price> 16,99</s_item_net_price>
|
| 107 |
-
<s_item_net_worth> 16,99</s_item_net_worth>
|
| 108 |
-
<s_item_vat>
|
| 109 |
-
10%</s_item_vat>
|
| 110 |
-
<s_item_gross_worth> 18,69</s_item_gross_worth>
|
| 111 |
-
<sep/>
|
| 112 |
-
<s_item_desc> Jams World Hawaiian 0 Dress
|
| 113 |
-
Rayon SZ.L</s_item_desc>
|
| 114 |
-
<s_item_qty> 5,00</s_item_qty>
|
| 115 |
-
<s_item_net_price> 65,00</s_item_net_price>
|
| 116 |
-
<s_item_net_worth>
|
| 117 |
-
325,00</s_item_net_worth>
|
| 118 |
-
<s_item_vat> 10%</s_item_vat>
|
| 119 |
-
<s_item_gross_worth>
|
| 120 |
-
357,50</s_item_gross_worth>
|
| 121 |
-
<sep/>
|
| 122 |
-
<s_item_desc> LuLaRoe Nicole Dress Size Large 26</s_item_desc>
|
| 123 |
-
<s_item_qty>
|
| 124 |
-
2,00</s_item_qty>
|
| 125 |
-
<s_item_net_price> 1,99</s_item_net_price>
|
| 126 |
-
<s_item_net_worth> 3,98</s_item_net_worth>
|
| 127 |
-
<s_item_vat>
|
| 128 |
-
10%</s_item_vat>
|
| 129 |
-
<s_item_gross_worth> 4,38</s_item_gross_worth>
|
| 130 |
-
<sep/>
|
| 131 |
-
<s_item_desc> phynny Was Medium Linen Wrap
|
| 132 |
-
Dress Dessert Rose Embroidered Bohemian</s_item_desc>
|
| 133 |
-
<s_item_qty> 2,00</s_item_qty>
|
| 134 |
-
<s_item_net_price>
|
| 135 |
-
89,99</s_item_net_price>
|
| 136 |
-
<s_item_net_worth> 179,98</s_item_net_worth>
|
| 137 |
-
<s_item_vat>
|
| 138 |
-
10%</s_item_vat>
|
| 139 |
-
<s_item_gross_worth> 197,98</s_item_gross_worth>
|
| 140 |
-
<sep/>
|
| 141 |
-
<s_item_desc> Eileen Fisher Women's Long
|
| 142 |
-
Sleeve Fleece Lined Front Pockets Dress XS Gray</s_item_desc>
|
| 143 |
-
<s_item_qty> 2,00</s_item_qty>
|
| 144 |
-
<s_item_net_price>
|
| 145 |
-
15,99</s_item_net_price>
|
| 146 |
-
<s_item_net_worth> 31,98</s_item_net_worth>
|
| 147 |
-
<s_item_vat>
|
| 148 |
-
10%</s_item_vat>
|
| 149 |
-
<s_item_gross_worth> 35,18</s_item_gross_worth>
|
| 150 |
-
<sep/>
|
| 151 |
-
<s_item_desc> Hanna Anderson Women's L Large
|
| 152 |
-
Coral Short Sleeve Casual Fall Tee Shirt Dress</s_item_desc>
|
| 153 |
-
<s_item_qty> 1,00</s_item_qty>
|
| 154 |
-
<s_item_net_price>
|
| 155 |
-
24,00</s_item_net_price>
|
| 156 |
-
<s_item_net_worth> 24,00</s_item_net_worth>
|
| 157 |
-
<s_item_vat>
|
| 158 |
-
10%</s_item_vat>
|
| 159 |
-
<s_item_gross_worth> 26,40</s_item_gross_worth>
|
| 160 |
-
</s_items>
|
| 161 |
-
<s_summary>
|
| 162 |
-
<s_total_net_worth>
|
| 163 |
-
$581,93</s_total_net_worth>
|
| 164 |
-
<s_total_vat> $58,19</s_total_vat>
|
| 165 |
-
<s_total_gross_worth> $
|
| 166 |
-
640,12</s_total_gross_worth>
|
| 167 |
-
</s_summary>
|
| 168 |
|
| 169 |
```
|
|
|
|
| 3 |
pipeline_tag: image-to-text
|
| 4 |
---
|
| 5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
***JSON OUTPUT***
|
| 8 |
```json
|
| 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
```
|