195 lines
5.0 KiB
Plaintext
Executable File
195 lines
5.0 KiB
Plaintext
Executable File
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import os \n",
|
|
"import glob \n",
|
|
"from tqdm import tqdm \n",
|
|
"import cv2 \n",
|
|
"import shutil\n",
|
|
"from pathlib import Path"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"DATA_DIR = \"/mnt/ssd1T/hoanglv/Projects/KIE/DATA/SDSAP_Invoice/raw/IMGS\"\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def copy_only_first_page(data_dir, out_dir, skip_types=['Receipt_taxi','Receipt_food_Cam', 'Receipt_food_Scan']):\n",
|
|
" paths = sorted(glob.glob(data_dir + \"/*/*\"))\n",
|
|
" print(\"Total paths: \", len(paths))\n",
|
|
" out_dir = Path(out_dir)\n",
|
|
" for path in paths:\n",
|
|
" type_doc = Path(path).parent.name\n",
|
|
" out_dir_full = out_dir / type_doc\n",
|
|
" if not out_dir_full.exists():\n",
|
|
" out_dir_full.mkdir(parents=True)\n",
|
|
" if type_doc in skip_types:\n",
|
|
" shutil.copy(path, str(out_dir_full))\n",
|
|
" else:\n",
|
|
" if \"_1.jpg\" in path:\n",
|
|
" shutil.copy(path, out_dir_full)\n",
|
|
" prefix_name = \"_\".join(path.split(\"_\")[:-1]) + \"_1.jpg\"\n",
|
|
" print(prefix_name)\n",
|
|
" if Path(prefix_name).exists():\n",
|
|
" continue\n",
|
|
" else:\n",
|
|
" shutil.copy(path, out_dir_full)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"classes = [\n",
|
|
" # id invoice\n",
|
|
" 'No_key', # số hóa đơn\n",
|
|
" 'No_value', \n",
|
|
" 'Form_key', # mẫu số hóa đơn\n",
|
|
" 'Form_value', \n",
|
|
" 'Serial_key', # số kí hiệu hoá đơn\n",
|
|
" 'Serial_value', \n",
|
|
" 'Date_value', \n",
|
|
"\n",
|
|
" # seller info\n",
|
|
" 'Seller_company_name_key', \n",
|
|
" 'Seller_company_name_value', \n",
|
|
" 'Seller_tax_code_key', \n",
|
|
" 'Seller_tax_code_value', \n",
|
|
" 'Seller_address_value',\n",
|
|
" 'Seller_address_key', \n",
|
|
" 'Seller_tel_key',\n",
|
|
" 'Seller_tel_value', \n",
|
|
" \n",
|
|
" # buyer info\n",
|
|
" 'Buyer_personal_name_key',\n",
|
|
" 'Buyer_personal_name_value', \n",
|
|
" 'Buyer_company_name_key', \n",
|
|
" 'Buyer_company_name_value', \n",
|
|
" 'Buyer_tax_code_key', \n",
|
|
" 'Buyer_tax_code_value', \n",
|
|
" 'Buyer_address_key', \n",
|
|
" 'Buyer_address_value', \n",
|
|
" 'Buyer_address_key',\n",
|
|
" 'Buyer_address_value',\n",
|
|
"\n",
|
|
" # money info\n",
|
|
" 'Tax_amount_key', \n",
|
|
" 'Tax_amount_value', \n",
|
|
" 'Total_key', \n",
|
|
" 'Total_value', \n",
|
|
" 'Total_in_words_key', \n",
|
|
" 'Total_in_words_value',\n",
|
|
" \n",
|
|
" 'Other', \n",
|
|
"]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"classes = [x.lower() for x in classes]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"['no_key',\n",
|
|
" 'no_value',\n",
|
|
" 'form_key',\n",
|
|
" 'form_value',\n",
|
|
" 'serial_key',\n",
|
|
" 'serial_value',\n",
|
|
" 'date_value',\n",
|
|
" 'seller_company_name_key',\n",
|
|
" 'seller_company_name_value',\n",
|
|
" 'seller_tax_code_key',\n",
|
|
" 'seller_tax_code_value',\n",
|
|
" 'seller_address_value',\n",
|
|
" 'seller_address_key',\n",
|
|
" 'seller_tel_key',\n",
|
|
" 'seller_tel_value',\n",
|
|
" 'buyer_personal_name_key',\n",
|
|
" 'buyer_personal_name_value',\n",
|
|
" 'buyer_company_name_key',\n",
|
|
" 'buyer_company_name_value',\n",
|
|
" 'buyer_tax_code_key',\n",
|
|
" 'buyer_tax_code_value',\n",
|
|
" 'buyer_address_key',\n",
|
|
" 'buyer_address_value',\n",
|
|
" 'buyer_address_key',\n",
|
|
" 'buyer_address_value',\n",
|
|
" 'tax_amount_key',\n",
|
|
" 'tax_amount_value',\n",
|
|
" 'total_key',\n",
|
|
" 'total_value',\n",
|
|
" 'total_in_words_key',\n",
|
|
" 'total_in_words_value',\n",
|
|
" 'other']"
|
|
]
|
|
},
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"classes"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "py38_hoanglv",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.8.16"
|
|
},
|
|
"orig_nbformat": 4
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|