{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3d87ca8d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "aa\n"
     ]
    }
   ],
   "source": [
    "print('aa')\n",
    "# /usr/local/bin/python -m jupyter notebook --no-browser --port=8888\n",
    "# /usr/local/bin/python\n",
    "# 后台启动 Jupyter Server，不打开浏览器，指定端口\n",
    "# /usr/local/bin/python -m jupyter server --no-browser --port=8888 \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6255246a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/zhaoxiangsi_train.json',\"r\", encoding=\"utf-8\") as f:\n",
    "    data = json.load(f)\n",
    "number_1 = 0\n",
    "number_0 = 0\n",
    "print(data[0].keys())\n",
    "print(data[0])\n",
    "print(data[1])\n",
    "print(data[2])\n",
    "number = 0\n",
    "# for i in data:\n",
    "    # if i['label'] == '0':\n",
    "    #     number += 1\n",
    "    #     print(i)\n",
    "    #     break\n",
    "    # if i['rn'] == '037a6b4e192533b2e0dc9ab3fb692806':\n",
    "    #     number += 1\n",
    "        # if i['label'] == '0':\n",
    "        #     number_0 += 1\n",
    "        #     print(i)\n",
    "        # if i['label'] == '1':\n",
    "        #     number_1 += 1\n",
    "        \n",
    "print(number_0)\n",
    "print(number_1)\n",
    "print(number)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0e5a8159",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "2 sample: 6 6 10\n",
      "3 sample: 9 9 4\n",
      "4 sample: 5 5 10\n",
      "5 sample: 10 10 6\n",
      "6 sample: 10 10 4\n",
      "7 sample: 2 2 14\n",
      "8 sample: 10 10 2\n",
      "9 sample: 10 10 6\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/zhaoxiangsi_train.json',\"r\", encoding=\"utf-8\") as f:\n",
    "    data = json.load(f)\n",
    "for i in range(0,10):\n",
    "    prompt_list = data[i]['pos_prompts'].split('##')\n",
    "    pos_list = data[i]['pos_item_tfsids'].split('##')\n",
    "    neg_list = data[i]['neg_item_tfsids'].split('##')\n",
    "    if len(prompt_list) == len(pos_list) and len(prompt_list) == len(neg_list):\n",
    "        print(i)\n",
    "    else:\n",
    "        print(f\"{i} sample: {len(prompt_list)} {len(pos_list)} {len(neg_list)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9e6501bd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "193145\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/bianji_train.json',\"r\", encoding=\"utf-8\") as f:\n",
    "    data = json.load(f)\n",
    "print(len(data))\n",
    "# with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/cot_output_new/dapei_train_cot.json',\"r\", encoding=\"utf-8\") as f:\n",
    "#     data2 = json.load(f)\n",
    "# print(len(data2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c55c3414",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rn: 10_copy1\n",
      "query_tfsid: O1CN01b1XDLi1ng65mWbXUm_!!4611686018427384350-0-imgsearch.jpg\n",
      "query_bbox: 50 185 357 565\n",
      "instruction: 要个带推杆的儿童三轮车\n",
      "cot: 粉色儿童三轮车，配备黑色手推杆，紫色座椅带有浅绿色箭头图案，车身侧面设储物篮，适合1-3岁宝宝骑行。\n",
      "prompt_index: 0\n",
      "pos_item_tfsid: O1CN01e6SdyJ1zEPYnOJHvb_!!4611686018427385642-0-fleamarket.jpg\n",
      "pos_item_title: 儿童三轮车脚踏车小孩轻便脚蹬自行车1-3-6岁宝宝手推滑行车\n",
      "pos_item_bbox: 16 66 841 1184\n",
      "neg_item_tfsid: O1CN01uY8GlQ1FZ34pNSBLB_!!2206509150500.jpg\n",
      "neg_item_title: 儿童平衡滑步车1-3岁宝宝小孩学步6手推车三合一三轮溜娃神器骑行\n",
      "neg_item_bbox: 50 49 764 653\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/cot_output_new/dapei_train_cot.json',\"r\", encoding=\"utf-8\") as f:\n",
    "    data2 = json.load(f)\n",
    "print(len(data2))\n",
    "# data = data2[0]\n",
    "# for i in data.keys():\n",
    "#     print(f\"{i}: {data[i]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5a649d5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import json\n",
    "with open('/data/oss_bucket_0/liqi/dataset/embedding_pf/bianji_test.json',\"r\", encoding=\"utf-8\") as f:\n",
    "    data2 = json.load(f)\n",
    "new_data = data2[:200]\n",
    "json.dump(new_data,open(\"./bianji_test_200.json\", \"w\", encoding=\"utf-8\"),ensure_ascii=False,indent=2)\n",
    "# data = data2[0]\n",
    "# for i in data.keys():\n",
    "#     print(f\"{i}: {data[i]}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.13.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}