demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Acknowledgement\n",
    "\n",
    "This notebook presents the practical work that was part of the group project \"Kuzushiji Classification\" in the Deep Learning course from Prof. Dr. Lippert (Hasso-Plattner-Institut, Germany) in 2022. \n",
    "\n",
    "We would like to thank our supervisor H. Shahriar for his support and assistance throughout the whole project development. \n",
    "\n",
    "Our project can also be found at GitHub: https://github.com/regnujAx/Deep-Learning-Kanji-Classification\n",
    "\n",
    "M. S. H. Chowdhury, A. Junger, M. Yakub, M. G. Rasul\n",
    "(Master students from University Potsdam and Hasso-Plattner-Institut, Germany)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Necessary Scripts and Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "5a_-MGcgeGfg"
   },
   "outputs": [],
   "source": [
    "# internal scripts\n",
    "import data_loading, test, train, transfer_learning\n",
    "\n",
    "# public libraries\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from PIL import Image\n",
    "\n",
    "from tensorflow import keras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Kuzushiji-Kanji Dataset Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: kkanji2\n",
      "label: U+241C6\n",
      "label: U+24FA3\n",
      "label: U+25DA1\n",
      "label: U+27752\n",
      "label: U+29780\n",
      "label: U+29DDA\n",
      "label: U+29E75\n",
      "label: U+4093\n",
      "label: U+4453\n",
      "label: U+4543\n",
      "label: U+4572\n",
      "label: U+4C61\n",
      "label: U+4C99\n",
      "label: U+4E00\n",
      "label: U+4E01\n",
      "label: U+4E03\n",
      "label: U+4E07\n",
      "label: U+4E08\n",
      "label: U+4E09\n",
      "label: U+4E0A\n",
      "label: U+4E0B\n",
      "label: U+4E0D\n",
      "label: U+4E0E\n",
      "label: U+4E11\n",
      "label: U+4E14\n",
      "label: U+4E16\n",
      "label: U+4E17\n",
      "label: U+4E18\n",
      "label: U+4E19\n",
      "label: U+4E1E\n",
      "label: U+4E21\n",
      "label: U+4E26\n",
      "label: U+4E2D\n",
      "label: U+4E32\n",
      "label: U+4E38\n",
      "label: U+4E39\n",
      "label: U+4E3B\n",
      "label: U+4E3C\n",
      "label: U+4E42\n",
      "label: U+4E43\n",
      "label: U+4E45\n",
      "label: U+4E4B\n",
      "label: U+4E4D\n",
      "label: U+4E4E\n",
      "label: U+4E4F\n",
      "label: U+4E57\n",
      "label: U+4E58\n",
      "label: U+4E59\n",
      "label: U+4E5D\n",
      "label: U+4E5E\n",
      "label: U+4E5F\n",
      "label: U+4E71\n",
      "label: U+4E73\n",
      "label: U+4E7E\n",
      "label: U+4E80\n",
      "label: U+4E82\n",
      "label: U+4E86\n",
      "label: U+4E88\n",
      "label: U+4E89\n",
      "label: U+4E8B\n",
      "label: U+4E8C\n",
      "label: U+4E8E\n",
      "label: U+4E91\n",
      "label: U+4E92\n",
      "label: U+4E94\n",
      "label: U+4E95\n",
      "label: U+4E98\n",
      "label: U+4E99\n",
      "label: U+4E9B\n",
      "label: U+4EA1\n",
      "label: U+4EA4\n",
      "label: U+4EA5\n",
      "label: U+4EA6\n",
      "label: U+4EA8\n",
      "label: U+4EAB\n",
      "label: U+4EAC\n",
      "label: U+4EAD\n",
      "label: U+4EAE\n",
      "label: U+4EBA\n",
      "label: U+4EC0\n",
      "label: U+4EC1\n",
      "label: U+4EC2\n",
      "label: U+4EC7\n",
      "label: U+4ECA\n",
      "label: U+4ECB\n",
      "label: U+4ECD\n",
      "label: U+4ECF\n",
      "label: U+4ED4\n",
      "label: U+4ED5\n",
      "label: U+4ED6\n",
      "label: U+4ED8\n",
      "label: U+4ED9\n",
      "label: U+4EDD\n",
      "label: U+4EE3\n",
      "label: U+4EE4\n",
      "label: U+4EE5\n",
      "label: U+4EED\n",
      "label: U+4EEE\n",
      "label: U+4EF0\n",
      "label: U+4EF2\n",
      "label: U+4EF6\n",
      "label: U+4EFB\n",
      "label: U+4F0A\n",
      "label: U+4F0D\n",
      "label: U+4F0F\n",
      "label: U+4F10\n",
      "label: U+4F11\n",
      "label: U+4F1A\n",
      "label: U+4F1D\n",
      "label: U+4F2F\n",
      "label: U+4F34\n",
      "label: U+4F38\n",
      "label: U+4F3A\n",
      "label: U+4F3C\n",
      "label: U+4F3D\n",
      "label: U+4F46\n",
      "label: U+4F47\n",
      "label: U+4F4D\n",
      "label: U+4F4E\n",
      "label: U+4F4F\n",
      "label: U+4F50\n",
      "label: U+4F53\n",
      "label: U+4F55\n",
      "label: U+4F59\n",
      "label: U+4F5B\n",
      "label: U+4F5C\n",
      "label: U+4F69\n",
      "label: U+4F73\n",
      "label: U+4F75\n",
      "label: U+4F76\n",
      "label: U+4F7F\n",
      "label: U+4F86\n",
      "label: U+4F8B\n",
      "label: U+4F8D\n",
      "label: U+4F98\n",
      "label: U+4F9B\n",
      "label: U+4F9D\n",
      "label: U+4FA0\n",
      "label: U+4FAD\n",
      "label: U+4FAE\n",
      "label: U+4FAF\n",
      "label: U+4FB2\n",
      "label: U+4FBF\n",
      "label: U+4FC2\n",
      "label: U+4FC3\n",
      "label: U+4FC4\n",
      "label: U+4FCA\n",
      "label: U+4FCE\n",
      "label: U+4FD7\n",
      "label: U+4FDD\n",
      "label: U+4FDF\n",
      "label: U+4FE1\n",
      "label: U+4FE4\n",
      "label: U+4FEE\n",
      "label: U+4FEF\n",
      "label: U+4FF3\n",
      "label: U+4FF5\n",
      "label: U+4FF8\n",
      "label: U+5009\n",
      "label: U+500B\n",
      "label: U+500D\n",
      "label: U+5011\n",
      "label: U+5012\n",
      "label: U+5016\n",
      "label: U+5018\n",
      "label: U+5019\n",
      "label: U+501A\n",
      "label: U+501F\n",
      "label: U+5023\n",
      "label: U+5026\n",
      "label: U+5029\n",
      "label: U+502B\n",
      "label: U+502D\n",
      "label: U+5036\n",
      "label: U+5047\n",
      "label: U+5048\n",
      "label: U+504F\n",
      "label: U+5056\n",
      "label: U+505C\n",
      "label: U+5065\n",
      "label: U+5074\n",
      "label: U+5076\n",
      "label: U+5078\n",
      "label: U+507D\n",
      "label: U+5080\n",
      "label: U+5085\n",
      "label: U+508D\n",
      "label: U+5098\n",
      "label: U+5099\n",
      "label: U+509A\n",
      "label: U+50AC\n",
      "label: U+50B3\n",
      "label: U+50B5\n",
      "label: U+50B7\n",
      "label: U+50BE\n",
      "label: U+50C5\n",
      "label: U+50C9\n",
      "label: U+50CD\n",
      "label: U+50CF\n",
      "label: U+50D5\n",
      "label: U+50DE\n",
      "label: U+50E5\n",
      "label: U+50E7\n",
      "label: U+50F5\n",
      "label: U+50F9\n",
      "label: U+50FB\n",
      "label: U+5100\n",
      "label: U+5104\n",
      "label: U+5109\n",
      "label: U+5112\n",
      "label: U+5118\n",
      "label: U+5121\n",
      "label: U+512A\n",
      "label: U+5141\n",
      "label: U+5143\n",
      "label: U+5144\n",
      "label: U+5145\n",
      "label: U+5146\n",
      "label: U+5147\n",
      "label: U+5148\n",
      "label: U+5149\n",
      "label: U+514A\n",
      "label: U+514B\n",
      "label: U+514C\n",
      "label: U+514D\n",
      "label: U+514E\n",
      "label: U+5150\n",
      "label: U+5152\n",
      "label: U+515C\n",
      "label: U+5165\n",
      "label: U+5168\n",
      "label: U+5169\n",
      "label: U+516A\n",
      "label: U+516B\n",
      "label: U+516C\n",
      "label: U+516D\n",
      "label: U+5171\n",
      "label: U+5175\n",
      "label: U+5176\n",
      "label: U+5177\n",
      "label: U+5178\n",
      "label: U+517C\n",
      "label: U+5185\n",
      "label: U+5186\n",
      "label: U+518A\n",
      "label: U+518C\n",
      "label: U+518D\n",
      "label: U+5199\n",
      "label: U+519D\n",
      "label: U+51A0\n",
      "label: U+51A5\n",
      "label: U+51A8\n",
      "label: U+51AC\n",
      "label: U+51B4\n",
      "label: U+51B6\n",
      "label: U+51B7\n",
      "label: U+51C4\n",
      "label: U+51C6\n",
      "label: U+51C9\n",
      "label: U+51CB\n",
      "label: U+51CC\n",
      "label: U+51CD\n",
      "label: U+51DD\n",
      "label: U+51E0\n",
      "label: U+51E1\n",
      "label: U+51E6\n",
      "label: U+51F6\n",
      "label: U+51F9\n",
      "label: U+51FA\n",
      "label: U+5200\n",
      "label: U+5203\n",
      "label: U+5204\n",
      "label: U+5206\n",
      "label: U+5207\n",
      "label: U+5208\n",
      "label: U+520A\n",
      "label: U+5211\n",
      "label: U+5215\n",
      "label: U+5217\n",
      "label: U+521D\n",
      "label: U+5224\n",
      "label: U+5225\n",
      "label: U+5229\n",
      "label: U+522A\n",
      "label: U+5230\n",
      "label: U+5236\n",
      "label: U+5239\n",
      "label: U+523A\n",
      "label: U+523B\n",
      "label: U+5243\n",
      "label: U+5247\n",
      "label: U+524A\n",
      "label: U+524B\n",
      "label: U+524D\n",
      "label: U+5256\n",
      "label: U+525B\n",
      "label: U+5263\n",
      "label: U+5265\n",
      "label: U+5269\n",
      "label: U+526A\n",
      "label: U+526F\n",
      "label: U+5271\n",
      "label: U+5272\n",
      "label: U+5275\n",
      "label: U+5287\n",
      "label: U+5289\n",
      "label: U+5294\n",
      "label: U+529B\n",
      "label: U+529F\n",
      "label: U+52A0\n",
      "label: U+52A3\n",
      "label: U+52A9\n",
      "label: U+52AA\n",
      "label: U+52AB\n",
      "label: U+52B1\n",
      "label: U+52B4\n",
      "label: U+52C3\n",
      "label: U+52C5\n",
      "label: U+52C7\n",
      "label: U+52C9\n",
      "label: U+52D5\n",
      "label: U+52D8\n",
      "label: U+52D9\n",
      "label: U+52DD\n",
      "label: U+52DE\n",
      "label: U+52DF\n",
      "label: U+52E2\n",
      "label: U+52E4\n",
      "label: U+52E6\n",
      "label: U+52E7\n",
      "label: U+52F8\n",
      "label: U+52FA\n",
      "label: U+52FE\n",
      "label: U+52FF\n",
      "label: U+5301\n",
      "label: U+5302\n",
      "label: U+5305\n",
      "label: U+5315\n",
      "label: U+5316\n",
      "label: U+5317\n",
      "label: U+5319\n",
      "label: U+531C\n",
      "label: U+5320\n",
      "label: U+5323\n",
      "label: U+5339\n",
      "label: U+533B\n",
      "label: U+5341\n",
      "label: U+5343\n",
      "label: U+5345\n",
      "label: U+5347\n",
      "label: U+5348\n",
      "label: U+534A\n",
      "label: U+5351\n",
      "label: U+5352\n",
      "label: U+5353\n",
      "label: U+5354\n",
      "label: U+5357\n",
      "label: U+5358\n",
      "label: U+535A\n",
      "label: U+535C\n",
      "label: U+5360\n",
      "label: U+5366\n",
      "label: U+536F\n",
      "label: U+5370\n",
      "label: U+5371\n",
      "label: U+5373\n",
      "label: U+5374\n",
      "label: U+5375\n",
      "label: U+5377\n",
      "label: U+5378\n",
      "label: U+537F\n",
      "label: U+5384\n",
      "label: U+5398\n",
      "label: U+539A\n",
      "label: U+539F\n",
      "label: U+53A0\n",
      "label: U+53A8\n",
      "label: U+53AD\n",
      "label: U+53B3\n",
      "label: U+53BB\n",
      "label: U+53C2\n",
      "label: U+53C3\n",
      "label: U+53C8\n",
      "label: U+53C9\n",
      "label: U+53CA\n",
      "label: U+53CB\n",
      "label: U+53CC\n",
      "label: U+53CD\n",
      "label: U+53CE\n",
      "label: U+53D4\n",
      "label: U+53D6\n",
      "label: U+53D7\n",
      "label: U+53D9\n",
      "label: U+53DB\n",
      "label: U+53DF\n",
      "label: U+53E1\n",
      "label: U+53E2\n",
      "label: U+53E3\n",
      "label: U+53E4\n",
      "label: U+53E5\n",
      "label: U+53EA\n",
      "label: U+53EB\n",
      "label: U+53EC\n",
      "label: U+53EF\n",
      "label: U+53F0\n",
      "label: U+53F2\n",
      "label: U+53F3\n",
      "label: U+53F6\n",
      "label: U+53F7\n",
      "label: U+53F8\n",
      "label: U+5403\n",
      "label: U+5404\n",
      "label: U+5408\n",
      "label: U+5409\n",
      "label: U+540A\n",
      "label: U+540C\n",
      "label: U+540D\n",
      "label: U+540E\n",
      "label: U+540F\n",
      "label: U+5410\n",
      "label: U+5411\n",
      "label: U+541B\n",
      "label: U+541D\n",
      "label: U+541F\n",
      "label: U+5426\n",
      "label: U+542B\n",
      "label: U+542E\n",
      "label: U+5436\n",
      "label: U+5438\n",
      "label: U+5439\n",
      "label: U+543B\n",
      "label: U+543C\n",
      "label: U+543E\n",
      "label: U+5442\n",
      "label: U+5446\n",
      "label: U+5449\n",
      "label: U+544A\n",
      "label: U+5451\n",
      "label: U+545F\n",
      "label: U+5468\n",
      "label: U+546A\n",
      "label: U+5473\n",
      "label: U+5475\n",
      "label: U+547C\n",
      "label: U+547D\n",
      "label: U+5480\n",
      "label: U+5484\n",
      "label: U+548C\n",
      "label: U+548E\n",
      "label: U+5492\n",
      "label: U+54A5\n",
      "label: U+54A8\n",
      "label: U+54AB\n",
      "label: U+54B2\n",
      "label: U+54B3\n",
      "label: U+54B8\n",
      "label: U+54BD\n",
      "label: U+54C0\n",
      "label: U+54C1\n",
      "label: U+54C9\n",
      "label: U+54E1\n",
      "label: U+54E2\n",
      "label: U+54E5\n",
      "label: U+54ED\n",
      "label: U+54F2\n",
      "label: U+54FD\n",
      "label: U+5504\n",
      "label: U+5507\n",
      "label: U+5510\n",
      "label: U+5516\n",
      "label: U+552F\n",
      "label: U+5531\n",
      "label: U+5544\n",
      "label: U+5546\n",
      "label: U+554F\n",
      "label: U+5553\n",
      "label: U+5557\n",
      "label: U+555C\n",
      "label: U+557B\n",
      "label: U+557C\n",
      "label: U+5583\n",
      "label: U+5584\n",
      "label: U+5589\n",
      "label: U+558B\n",
      "label: U+5598\n",
      "label: U+559C\n",
      "label: U+559D\n",
      "label: U+559E\n",
      "label: U+559F\n",
      "label: U+55A7\n",
      "label: U+55A9\n",
      "label: U+55AA\n",
      "label: U+55AB\n",
      "label: U+55B0\n",
      "label: U+55B6\n",
      "label: U+55C5\n",
      "label: U+55C7\n",
      "label: U+55D4\n",
      "label: U+55DA\n",
      "label: U+55DC\n",
      "label: U+55DF\n",
      "label: U+55E3\n",
      "label: U+55FD\n",
      "label: U+5608\n",
      "label: U+5609\n",
      "label: U+5617\n",
      "label: U+5618\n",
      "label: U+5629\n",
      "label: U+562F\n",
      "label: U+5632\n",
      "label: U+5638\n",
      "label: U+5640\n",
      "label: U+5642\n",
      "label: U+564C\n",
      "label: U+5668\n",
      "label: U+566A\n",
      "label: U+566C\n",
      "label: U+567A\n",
      "label: U+56A2\n",
      "label: U+56AE\n",
      "label: U+56B4\n",
      "label: U+56C0\n",
      "label: U+56C2\n",
      "label: U+56C3\n",
      "label: U+56C9\n",
      "label: U+56D8\n",
      "label: U+56DB\n",
      "label: U+56DE\n",
      "label: U+56E0\n",
      "label: U+56E3\n",
      "label: U+56F0\n",
      "label: U+56F2\n",
      "label: U+56F3\n",
      "label: U+56FA\n",
      "label: U+56FD\n",
      "label: U+570B\n",
      "label: U+570D\n",
      "label: U+5711\n",
      "label: U+5712\n",
      "label: U+5713\n",
      "label: U+5716\n",
      "label: U+5718\n",
      "label: U+571F\n",
      "label: U+5728\n",
      "label: U+572D\n",
      "label: U+5730\n",
      "label: U+5742\n",
      "label: U+5747\n",
      "label: U+574A\n",
      "label: U+574F\n",
      "label: U+5750\n",
      "label: U+5751\n",
      "label: U+5761\n",
      "label: U+576A\n",
      "label: U+5782\n",
      "label: U+57A2\n",
      "label: U+57A3\n",
      "label: U+57C3\n",
      "label: U+57CB\n",
      "label: U+57CE\n",
      "label: U+57D2\n",
      "label: U+57D3\n",
      "label: U+57DC\n",
      "label: U+57DF\n",
      "label: U+57F5\n",
      "label: U+57F7\n",
      "label: U+57FA\n",
      "label: U+5800\n",
      "label: U+5802\n",
      "label: U+5805\n",
      "label: U+5815\n",
      "label: U+5824\n",
      "label: U+582A\n",
      "label: U+582F\n",
      "label: U+5830\n",
      "label: U+5831\n",
      "label: U+5834\n",
      "label: U+5835\n",
      "label: U+583A\n",
      "label: U+5840\n",
      "label: U+584A\n",
      "label: U+5852\n",
      "label: U+5854\n",
      "label: U+5857\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+585A\n",
      "label: U+585E\n",
      "label: U+5869\n",
      "label: U+5872\n",
      "label: U+5875\n",
      "label: U+5883\n",
      "label: U+5893\n",
      "label: U+5897\n",
      "label: U+58A8\n",
      "label: U+58AE\n",
      "label: U+58BE\n",
      "label: U+58C1\n",
      "label: U+58C7\n",
      "label: U+58CC\n",
      "label: U+58D2\n",
      "label: U+58DC\n",
      "label: U+58DF\n",
      "label: U+58EB\n",
      "label: U+58EC\n",
      "label: U+58EE\n",
      "label: U+58EF\n",
      "label: U+58F0\n",
      "label: U+58F1\n",
      "label: U+58F2\n",
      "label: U+58F3\n",
      "label: U+58F7\n",
      "label: U+58F9\n",
      "label: U+58FA\n",
      "label: U+58FC\n",
      "label: U+58FD\n",
      "label: U+5909\n",
      "label: U+590F\n",
      "label: U+5915\n",
      "label: U+5916\n",
      "label: U+5919\n",
      "label: U+591A\n",
      "label: U+591C\n",
      "label: U+5922\n",
      "label: U+5925\n",
      "label: U+5927\n",
      "label: U+5929\n",
      "label: U+592A\n",
      "label: U+592B\n",
      "label: U+592E\n",
      "label: U+5931\n",
      "label: U+5937\n",
      "label: U+593E\n",
      "label: U+5947\n",
      "label: U+5948\n",
      "label: U+5949\n",
      "label: U+594F\n",
      "label: U+5951\n",
      "label: U+5954\n",
      "label: U+5955\n",
      "label: U+595A\n",
      "label: U+5960\n",
      "label: U+5962\n",
      "label: U+5965\n",
      "label: U+5967\n",
      "label: U+596A\n",
      "label: U+596E\n",
      "label: U+5973\n",
      "label: U+5974\n",
      "label: U+5978\n",
      "label: U+597D\n",
      "label: U+5982\n",
      "label: U+5983\n",
      "label: U+5984\n",
      "label: U+598A\n",
      "label: U+5993\n",
      "label: U+5996\n",
      "label: U+5999\n",
      "label: U+59A3\n",
      "label: U+59A8\n",
      "label: U+59AC\n",
      "label: U+59B9\n",
      "label: U+59BB\n",
      "label: U+59BE\n",
      "label: U+59C9\n",
      "label: U+59CA\n",
      "label: U+59CB\n",
      "label: U+59D1\n",
      "label: U+59D3\n",
      "label: U+59D4\n",
      "label: U+59DA\n",
      "label: U+59DC\n",
      "label: U+59E6\n",
      "label: U+59E8\n",
      "label: U+59EA\n",
      "label: U+59EB\n",
      "label: U+59FF\n",
      "label: U+5A01\n",
      "label: U+5A11\n",
      "label: U+5A18\n",
      "label: U+5A1F\n",
      "label: U+5A23\n",
      "label: U+5A25\n",
      "label: U+5A2F\n",
      "label: U+5A35\n",
      "label: U+5A36\n",
      "label: U+5A3C\n",
      "label: U+5A46\n",
      "label: U+5A5A\n",
      "label: U+5A62\n",
      "label: U+5A66\n",
      "label: U+5A6C\n",
      "label: U+5A7F\n",
      "label: U+5A92\n",
      "label: U+5A9A\n",
      "label: U+5A9B\n",
      "label: U+5AB1\n",
      "label: U+5AB3\n",
      "label: U+5ABD\n",
      "label: U+5AC1\n",
      "label: U+5AC9\n",
      "label: U+5ACC\n",
      "label: U+5AD7\n",
      "label: U+5AE1\n",
      "label: U+5AE9\n",
      "label: U+5AF0\n",
      "label: U+5B09\n",
      "label: U+5B0B\n",
      "label: U+5B22\n",
      "label: U+5B30\n",
      "label: U+5B32\n",
      "label: U+5B36\n",
      "label: U+5B40\n",
      "label: U+5B41\n",
      "label: U+5B50\n",
      "label: U+5B51\n",
      "label: U+5B54\n",
      "label: U+5B57\n",
      "label: U+5B58\n",
      "label: U+5B5C\n",
      "label: U+5B5D\n",
      "label: U+5B5F\n",
      "label: U+5B63\n",
      "label: U+5B64\n",
      "label: U+5B66\n",
      "label: U+5B6B\n",
      "label: U+5B70\n",
      "label: U+5B78\n",
      "label: U+5B85\n",
      "label: U+5B87\n",
      "label: U+5B88\n",
      "label: U+5B89\n",
      "label: U+5B8B\n",
      "label: U+5B8C\n",
      "label: U+5B95\n",
      "label: U+5B97\n",
      "label: U+5B98\n",
      "label: U+5B99\n",
      "label: U+5B9A\n",
      "label: U+5B9B\n",
      "label: U+5B9C\n",
      "label: U+5B9D\n",
      "label: U+5B9F\n",
      "label: U+5BA2\n",
      "label: U+5BA3\n",
      "label: U+5BA4\n",
      "label: U+5BA5\n",
      "label: U+5BA6\n",
      "label: U+5BAE\n",
      "label: U+5BB0\n",
      "label: U+5BB3\n",
      "label: U+5BB4\n",
      "label: U+5BB5\n",
      "label: U+5BB6\n",
      "label: U+5BB8\n",
      "label: U+5BB9\n",
      "label: U+5BBF\n",
      "label: U+5BC2\n",
      "label: U+5BC4\n",
      "label: U+5BC5\n",
      "label: U+5BC6\n",
      "label: U+5BC7\n",
      "label: U+5BCC\n",
      "label: U+5BD0\n",
      "label: U+5BD2\n",
      "label: U+5BD3\n",
      "label: U+5BD4\n",
      "label: U+5BDB\n",
      "label: U+5BDD\n",
      "label: U+5BDF\n",
      "label: U+5BE1\n",
      "label: U+5BE2\n",
      "label: U+5BE4\n",
      "label: U+5BE6\n",
      "label: U+5BE7\n",
      "label: U+5BE8\n",
      "label: U+5BE9\n",
      "label: U+5BF3\n",
      "label: U+5BF5\n",
      "label: U+5BF6\n",
      "label: U+5BF8\n",
      "label: U+5BFA\n",
      "label: U+5BFE\n",
      "label: U+5BFF\n",
      "label: U+5C01\n",
      "label: U+5C02\n",
      "label: U+5C04\n",
      "label: U+5C05\n",
      "label: U+5C06\n",
      "label: U+5C07\n",
      "label: U+5C08\n",
      "label: U+5C09\n",
      "label: U+5C0A\n",
      "label: U+5C0B\n",
      "label: U+5C0D\n",
      "label: U+5C0E\n",
      "label: U+5C0F\n",
      "label: U+5C11\n",
      "label: U+5C13\n",
      "label: U+5C16\n",
      "label: U+5C1A\n",
      "label: U+5C24\n",
      "label: U+5C28\n",
      "label: U+5C2D\n",
      "label: U+5C31\n",
      "label: U+5C3A\n",
      "label: U+5C3B\n",
      "label: U+5C3C\n",
      "label: U+5C3D\n",
      "label: U+5C3E\n",
      "label: U+5C3F\n",
      "label: U+5C40\n",
      "label: U+5C41\n",
      "label: U+5C45\n",
      "label: U+5C48\n",
      "label: U+5C4A\n",
      "label: U+5C4B\n",
      "label: U+5C4D\n",
      "label: U+5C4E\n",
      "label: U+5C4F\n",
      "label: U+5C50\n",
      "label: U+5C51\n",
      "label: U+5C53\n",
      "label: U+5C55\n",
      "label: U+5C5E\n",
      "label: U+5C60\n",
      "label: U+5C61\n",
      "label: U+5C65\n",
      "label: U+5C69\n",
      "label: U+5C6F\n",
      "label: U+5C71\n",
      "label: U+5C90\n",
      "label: U+5CA1\n",
      "label: U+5CA8\n",
      "label: U+5CA9\n",
      "label: U+5CB3\n",
      "label: U+5CB8\n",
      "label: U+5CE0\n",
      "label: U+5CE8\n",
      "label: U+5CEF\n",
      "label: U+5CF0\n",
      "label: U+5CF6\n",
      "label: U+5CFD\n",
      "label: U+5D07\n",
      "label: U+5D0E\n",
      "label: U+5D11\n",
      "label: U+5D14\n",
      "label: U+5D16\n",
      "label: U+5D19\n",
      "label: U+5D29\n",
      "label: U+5D50\n",
      "label: U+5D69\n",
      "label: U+5D6F\n",
      "label: U+5D8B\n",
      "label: U+5D8C\n",
      "label: U+5DAE\n",
      "label: U+5DBA\n",
      "label: U+5DBD\n",
      "label: U+5DCC\n",
      "label: U+5DD6\n",
      "label: U+5DDD\n",
      "label: U+5DDE\n",
      "label: U+5DE1\n",
      "label: U+5DE2\n",
      "label: U+5DE3\n",
      "label: U+5DE5\n",
      "label: U+5DE6\n",
      "label: U+5DE7\n",
      "label: U+5DE8\n",
      "label: U+5DEB\n",
      "label: U+5DEE\n",
      "label: U+5DF1\n",
      "label: U+5DF2\n",
      "label: U+5DF3\n",
      "label: U+5DF4\n",
      "label: U+5DF7\n",
      "label: U+5DFB\n",
      "label: U+5DFE\n",
      "label: U+5E02\n",
      "label: U+5E03\n",
      "label: U+5E06\n",
      "label: U+5E0B\n",
      "label: U+5E0C\n",
      "label: U+5E16\n",
      "label: U+5E1B\n",
      "label: U+5E1D\n",
      "label: U+5E25\n",
      "label: U+5E2B\n",
      "label: U+5E2D\n",
      "label: U+5E2F\n",
      "label: U+5E30\n",
      "label: U+5E33\n",
      "label: U+5E36\n",
      "label: U+5E37\n",
      "label: U+5E38\n",
      "label: U+5E3D\n",
      "label: U+5E45\n",
      "label: U+5E55\n",
      "label: U+5E5F\n",
      "label: U+5E61\n",
      "label: U+5E62\n",
      "label: U+5E63\n",
      "label: U+5E72\n",
      "label: U+5E73\n",
      "label: U+5E74\n",
      "label: U+5E76\n",
      "label: U+5E78\n",
      "label: U+5E7B\n",
      "label: U+5E7C\n",
      "label: U+5E7D\n",
      "label: U+5E7E\n",
      "label: U+5E7F\n",
      "label: U+5E83\n",
      "label: U+5E84\n",
      "label: U+5E87\n",
      "label: U+5E8A\n",
      "label: U+5E8F\n",
      "label: U+5E95\n",
      "label: U+5E96\n",
      "label: U+5E97\n",
      "label: U+5E9A\n",
      "label: U+5E9C\n",
      "label: U+5EA6\n",
      "label: U+5EA7\n",
      "label: U+5EAB\n",
      "label: U+5EAD\n",
      "label: U+5EB5\n",
      "label: U+5EB6\n",
      "label: U+5EB7\n",
      "label: U+5EB8\n",
      "label: U+5EC3\n",
      "label: U+5EC8\n",
      "label: U+5EC9\n",
      "label: U+5ECA\n",
      "label: U+5ED3\n",
      "label: U+5EDF\n",
      "label: U+5EE2\n",
      "label: U+5EE3\n",
      "label: U+5EE9\n",
      "label: U+5EF1\n",
      "label: U+5EF3\n",
      "label: U+5EF6\n",
      "label: U+5EF7\n",
      "label: U+5EFA\n",
      "label: U+5EFB\n",
      "label: U+5EFC\n",
      "label: U+5EFF\n",
      "label: U+5F01\n",
      "label: U+5F09\n",
      "label: U+5F0F\n",
      "label: U+5F10\n",
      "label: U+5F11\n",
      "label: U+5F13\n",
      "label: U+5F14\n",
      "label: U+5F15\n",
      "label: U+5F17\n",
      "label: U+5F18\n",
      "label: U+5F1F\n",
      "label: U+5F25\n",
      "label: U+5F27\n",
      "label: U+5F31\n",
      "label: U+5F35\n",
      "label: U+5F37\n",
      "label: U+5F3E\n",
      "label: U+5F48\n",
      "label: U+5F4C\n",
      "label: U+5F53\n",
      "label: U+5F59\n",
      "label: U+5F62\n",
      "label: U+5F66\n",
      "label: U+5F69\n",
      "label: U+5F6B\n",
      "label: U+5F70\n",
      "label: U+5F71\n",
      "label: U+5F73\n",
      "label: U+5F79\n",
      "label: U+5F7C\n",
      "label: U+5F80\n",
      "label: U+5F81\n",
      "label: U+5F82\n",
      "label: U+5F83\n",
      "label: U+5F85\n",
      "label: U+5F8A\n",
      "label: U+5F8B\n",
      "label: U+5F8C\n",
      "label: U+5F90\n",
      "label: U+5F91\n",
      "label: U+5F92\n",
      "label: U+5F93\n",
      "label: U+5F97\n",
      "label: U+5F98\n",
      "label: U+5F9E\n",
      "label: U+5FA0\n",
      "label: U+5FA1\n",
      "label: U+5FA8\n",
      "label: U+5FA9\n",
      "label: U+5FAE\n",
      "label: U+5FB3\n",
      "label: U+5FB9\n",
      "label: U+5FC3\n",
      "label: U+5FC5\n",
      "label: U+5FCC\n",
      "label: U+5FCD\n",
      "label: U+5FD7\n",
      "label: U+5FD8\n",
      "label: U+5FD9\n",
      "label: U+5FDC\n",
      "label: U+5FDD\n",
      "label: U+5FE0\n",
      "label: U+5FEB\n",
      "label: U+5FF5\n",
      "label: U+5FFD\n",
      "label: U+5FFF\n",
      "label: U+6012\n",
      "label: U+6015\n",
      "label: U+6016\n",
      "label: U+601C\n",
      "label: U+601D\n",
      "label: U+6020\n",
      "label: U+6021\n",
      "label: U+6025\n",
      "label: U+6027\n",
      "label: U+6028\n",
      "label: U+602A\n",
      "label: U+602F\n",
      "label: U+6043\n",
      "label: U+604B\n",
      "label: U+604D\n",
      "label: U+6050\n",
      "label: U+6052\n",
      "label: U+6056\n",
      "label: U+6059\n",
      "label: U+605A\n",
      "label: U+605F\n",
      "label: U+6060\n",
      "label: U+6063\n",
      "label: U+6065\n",
      "label: U+6068\n",
      "label: U+6069\n",
      "label: U+606D\n",
      "label: U+606F\n",
      "label: U+6070\n",
      "label: U+6075\n",
      "label: U+6089\n",
      "label: U+608B\n",
      "label: U+608C\n",
      "label: U+6092\n",
      "label: U+6094\n",
      "label: U+6096\n",
      "label: U+609F\n",
      "label: U+60A3\n",
      "label: U+60A6\n",
      "label: U+60A7\n",
      "label: U+60A9\n",
      "label: U+60AA\n",
      "label: U+60B2\n",
      "label: U+60B5\n",
      "label: U+60B6\n",
      "label: U+60C5\n",
      "label: U+60C6\n",
      "label: U+60D1\n",
      "label: U+60DA\n",
      "label: U+60DC\n",
      "label: U+60DF\n",
      "label: U+60E0\n",
      "label: U+60E1\n",
      "label: U+60E3\n",
      "label: U+60F0\n",
      "label: U+60F1\n",
      "label: U+60F3\n",
      "label: U+60FA\n",
      "label: U+60FB\n",
      "label: U+6101\n",
      "label: U+6108\n",
      "label: U+610D\n",
      "label: U+610F\n",
      "label: U+611A\n",
      "label: U+611B\n",
      "label: U+611F\n",
      "label: U+6127\n",
      "label: U+613C\n",
      "label: U+6148\n",
      "label: U+614B\n",
      "label: U+614C\n",
      "label: U+614E\n",
      "label: U+6155\n",
      "label: U+6159\n",
      "label: U+615D\n",
      "label: U+6162\n",
      "label: U+6165\n",
      "label: U+6167\n",
      "label: U+6168\n",
      "label: U+616E\n",
      "label: U+6170\n",
      "label: U+6173\n",
      "label: U+6176\n",
      "label: U+617E\n",
      "label: U+6182\n",
      "label: U+6187\n",
      "label: U+618E\n",
      "label: U+6190\n",
      "label: U+6191\n",
      "label: U+6196\n",
      "label: U+619A\n",
      "label: U+61A4\n",
      "label: U+61B6\n",
      "label: U+61C7\n",
      "label: U+61C8\n",
      "label: U+61C9\n",
      "label: U+61D0\n",
      "label: U+61F2\n",
      "label: U+61F6\n",
      "label: U+61F7\n",
      "label: U+61F8\n",
      "label: U+61FA\n",
      "label: U+6200\n",
      "label: U+6208\n",
      "label: U+620A\n",
      "label: U+620C\n",
      "label: U+620E\n",
      "label: U+6210\n",
      "label: U+6211\n",
      "label: U+6212\n",
      "label: U+6216\n",
      "label: U+621B\n",
      "label: U+621D\n",
      "label: U+621F\n",
      "label: U+6226\n",
      "label: U+622F\n",
      "label: U+6230\n",
      "label: U+6232\n",
      "label: U+6234\n",
      "label: U+6236\n",
      "label: U+6238\n",
      "label: U+623B\n",
      "label: U+623E\n",
      "label: U+623F\n",
      "label: U+6240\n",
      "label: U+6241\n",
      "label: U+6247\n",
      "label: U+6248\n",
      "label: U+6249\n",
      "label: U+624B\n",
      "label: U+624D\n",
      "label: U+6253\n",
      "label: U+6255\n",
      "label: U+6258\n",
      "label: U+6268\n",
      "label: U+626E\n",
      "label: U+6271\n",
      "label: U+6276\n",
      "label: U+6279\n",
      "label: U+627F\n",
      "label: U+6280\n",
      "label: U+6284\n",
      "label: U+628A\n",
      "label: U+6291\n",
      "label: U+6293\n",
      "label: U+6294\n",
      "label: U+6295\n",
      "label: U+6298\n",
      "label: U+629B\n",
      "label: U+629C\n",
      "label: U+62AB\n",
      "label: U+62B1\n",
      "label: U+62B5\n",
      "label: U+62BC\n",
      "label: U+62BD\n",
      "label: U+62C2\n",
      "label: U+62C5\n",
      "label: U+62C7\n",
      "label: U+62CA\n",
      "label: U+62CC\n",
      "label: U+62CD\n",
      "label: U+62D0\n",
      "label: U+62D2\n",
      "label: U+62D3\n",
      "label: U+62D4\n",
      "label: U+62D7\n",
      "label: U+62D8\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+62D9\n",
      "label: U+62DB\n",
      "label: U+62DC\n",
      "label: U+62DD\n",
      "label: U+62E0\n",
      "label: U+62EC\n",
      "label: U+62ED\n",
      "label: U+62F3\n",
      "label: U+62F5\n",
      "label: U+62F6\n",
      "label: U+62FE\n",
      "label: U+62FF\n",
      "label: U+6301\n",
      "label: U+6307\n",
      "label: U+6309\n",
      "label: U+630C\n",
      "label: U+6311\n",
      "label: U+6319\n",
      "label: U+631F\n",
      "label: U+6328\n",
      "label: U+632B\n",
      "label: U+632F\n",
      "label: U+633A\n",
      "label: U+633D\n",
      "label: U+633E\n",
      "label: U+633F\n",
      "label: U+6349\n",
      "label: U+634F\n",
      "label: U+6355\n",
      "label: U+6367\n",
      "label: U+6368\n",
      "label: U+636E\n",
      "label: U+637B\n",
      "label: U+6383\n",
      "label: U+6388\n",
      "label: U+6389\n",
      "label: U+638C\n",
      "label: U+6392\n",
      "label: U+6398\n",
      "label: U+639B\n",
      "label: U+639F\n",
      "label: U+63A0\n",
      "label: U+63A1\n",
      "label: U+63A2\n",
      "label: U+63A5\n",
      "label: U+63A8\n",
      "label: U+63A9\n",
      "label: U+63AC\n",
      "label: U+63B2\n",
      "label: U+63B4\n",
      "label: U+63BB\n",
      "label: U+63BE\n",
      "label: U+63C3\n",
      "label: U+63C9\n",
      "label: U+63CF\n",
      "label: U+63D0\n",
      "label: U+63DA\n",
      "label: U+63DB\n",
      "label: U+63E1\n",
      "label: U+63E9\n",
      "label: U+63EE\n",
      "label: U+63F4\n",
      "label: U+63FA\n",
      "label: U+640D\n",
      "label: U+640F\n",
      "label: U+6414\n",
      "label: U+6416\n",
      "label: U+6417\n",
      "label: U+641C\n",
      "label: U+643A\n",
      "label: U+6442\n",
      "label: U+6458\n",
      "label: U+6460\n",
      "label: U+6467\n",
      "label: U+6469\n",
      "label: U+6478\n",
      "label: U+647A\n",
      "label: U+6483\n",
      "label: U+6492\n",
      "label: U+649E\n",
      "label: U+64A5\n",
      "label: U+64AB\n",
      "label: U+64AD\n",
      "label: U+64AE\n",
      "label: U+64B0\n",
      "label: U+64B2\n",
      "label: U+64C2\n",
      "label: U+64CD\n",
      "label: U+64D2\n",
      "label: U+64DA\n",
      "label: U+64E1\n",
      "label: U+64E7\n",
      "label: U+64EC\n",
      "label: U+64F2\n",
      "label: U+6518\n",
      "label: U+652C\n",
      "label: U+652F\n",
      "label: U+6536\n",
      "label: U+6537\n",
      "label: U+6539\n",
      "label: U+653E\n",
      "label: U+653F\n",
      "label: U+6545\n",
      "label: U+6548\n",
      "label: U+6551\n",
      "label: U+6556\n",
      "label: U+6559\n",
      "label: U+655D\n",
      "label: U+6562\n",
      "label: U+6563\n",
      "label: U+6566\n",
      "label: U+656C\n",
      "label: U+6570\n",
      "label: U+6572\n",
      "label: U+6574\n",
      "label: U+6575\n",
      "label: U+6577\n",
      "label: U+6578\n",
      "label: U+6583\n",
      "label: U+6587\n",
      "label: U+6589\n",
      "label: U+658E\n",
      "label: U+6590\n",
      "label: U+6591\n",
      "label: U+6597\n",
      "label: U+6599\n",
      "label: U+659B\n",
      "label: U+659C\n",
      "label: U+659F\n",
      "label: U+65A4\n",
      "label: U+65A7\n",
      "label: U+65AC\n",
      "label: U+65AD\n",
      "label: U+65AF\n",
      "label: U+65B0\n",
      "label: U+65B7\n",
      "label: U+65B9\n",
      "label: U+65BC\n",
      "label: U+65BD\n",
      "label: U+65C1\n",
      "label: U+65C5\n",
      "label: U+65CF\n",
      "label: U+65D7\n",
      "label: U+65E0\n",
      "label: U+65E2\n",
      "label: U+65E5\n",
      "label: U+65E6\n",
      "label: U+65E7\n",
      "label: U+65E8\n",
      "label: U+65E9\n",
      "label: U+65EC\n",
      "label: U+6602\n",
      "label: U+6606\n",
      "label: U+6607\n",
      "label: U+660C\n",
      "label: U+660E\n",
      "label: U+660F\n",
      "label: U+6613\n",
      "label: U+6614\n",
      "label: U+661F\n",
      "label: U+6620\n",
      "label: U+6625\n",
      "label: U+6627\n",
      "label: U+6628\n",
      "label: U+662D\n",
      "label: U+662F\n",
      "label: U+6634\n",
      "label: U+663C\n",
      "label: U+6642\n",
      "label: U+664B\n",
      "label: U+6652\n",
      "label: U+665D\n",
      "label: U+665E\n",
      "label: U+6666\n",
      "label: U+6668\n",
      "label: U+6669\n",
      "label: U+666E\n",
      "label: U+666F\n",
      "label: U+6674\n",
      "label: U+6676\n",
      "label: U+667A\n",
      "label: U+6681\n",
      "label: U+6684\n",
      "label: U+6687\n",
      "label: U+6691\n",
      "label: U+6696\n",
      "label: U+6697\n",
      "label: U+66A6\n",
      "label: U+66AB\n",
      "label: U+66AE\n",
      "label: U+66B4\n",
      "label: U+66B9\n",
      "label: U+66C9\n",
      "label: U+66D9\n",
      "label: U+66DC\n",
      "label: U+66DD\n",
      "label: U+66E0\n",
      "label: U+66F0\n",
      "label: U+66F2\n",
      "label: U+66F3\n",
      "label: U+66F4\n",
      "label: U+66F8\n",
      "label: U+66F9\n",
      "label: U+66FC\n",
      "label: U+66FD\n",
      "label: U+66FE\n",
      "label: U+66FF\n",
      "label: U+6700\n",
      "label: U+6703\n",
      "label: U+6708\n",
      "label: U+6709\n",
      "label: U+670B\n",
      "label: U+670D\n",
      "label: U+6714\n",
      "label: U+6715\n",
      "label: U+6717\n",
      "label: U+671B\n",
      "label: U+671D\n",
      "label: U+671F\n",
      "label: U+6726\n",
      "label: U+6727\n",
      "label: U+6728\n",
      "label: U+672A\n",
      "label: U+672B\n",
      "label: U+672C\n",
      "label: U+672D\n",
      "label: U+672E\n",
      "label: U+6731\n",
      "label: U+6734\n",
      "label: U+673A\n",
      "label: U+673D\n",
      "label: U+6749\n",
      "label: U+674E\n",
      "label: U+674F\n",
      "label: U+6750\n",
      "label: U+6751\n",
      "label: U+6753\n",
      "label: U+6756\n",
      "label: U+6759\n",
      "label: U+675C\n",
      "label: U+675E\n",
      "label: U+675F\n",
      "label: U+6760\n",
      "label: U+6761\n",
      "label: U+6762\n",
      "label: U+6765\n",
      "label: U+676D\n",
      "label: U+676F\n",
      "label: U+6771\n",
      "label: U+6772\n",
      "label: U+6773\n",
      "label: U+6775\n",
      "label: U+6777\n",
      "label: U+677C\n",
      "label: U+677E\n",
      "label: U+677F\n",
      "label: U+6787\n",
      "label: U+678B\n",
      "label: U+6795\n",
      "label: U+6797\n",
      "label: U+679A\n",
      "label: U+679C\n",
      "label: U+679D\n",
      "label: U+67A1\n",
      "label: U+67A9\n",
      "label: U+67AF\n",
      "label: U+67B2\n",
      "label: U+67B4\n",
      "label: U+67B6\n",
      "label: U+67B7\n",
      "label: U+67B8\n",
      "label: U+67C4\n",
      "label: U+67CA\n",
      "label: U+67CF\n",
      "label: U+67D0\n",
      "label: U+67D1\n",
      "label: U+67D3\n",
      "label: U+67D4\n",
      "label: U+67D8\n",
      "label: U+67DA\n",
      "label: U+67E4\n",
      "label: U+67E9\n",
      "label: U+67F1\n",
      "label: U+67F3\n",
      "label: U+67F4\n",
      "label: U+67F9\n",
      "label: U+67FB\n",
      "label: U+67FE\n",
      "label: U+67FF\n",
      "label: U+6801\n",
      "label: U+6803\n",
      "label: U+6804\n",
      "label: U+6816\n",
      "label: U+6817\n",
      "label: U+681D\n",
      "label: U+6821\n",
      "label: U+6822\n",
      "label: U+682A\n",
      "label: U+6838\n",
      "label: U+6839\n",
      "label: U+683C\n",
      "label: U+683D\n",
      "label: U+6840\n",
      "label: U+6841\n",
      "label: U+6842\n",
      "label: U+6843\n",
      "label: U+6848\n",
      "label: U+6850\n",
      "label: U+6851\n",
      "label: U+6853\n",
      "label: U+6854\n",
      "label: U+685C\n",
      "label: U+6867\n",
      "label: U+6874\n",
      "label: U+6876\n",
      "label: U+6881\n",
      "label: U+6885\n",
      "label: U+6893\n",
      "label: U+6894\n",
      "label: U+6897\n",
      "label: U+689D\n",
      "label: U+689F\n",
      "label: U+68A2\n",
      "label: U+68A7\n",
      "label: U+68A8\n",
      "label: U+68AD\n",
      "label: U+68AF\n",
      "label: U+68B0\n",
      "label: U+68B3\n",
      "label: U+68B5\n",
      "label: U+68B6\n",
      "label: U+68BC\n",
      "label: U+68C4\n",
      "label: U+68C9\n",
      "label: U+68CD\n",
      "label: U+68D0\n",
      "label: U+68D2\n",
      "label: U+68D7\n",
      "label: U+68D8\n",
      "label: U+68DA\n",
      "label: U+68DF\n",
      "label: U+68E0\n",
      "label: U+68E7\n",
      "label: U+68EE\n",
      "label: U+68F2\n",
      "label: U+68F9\n",
      "label: U+68FA\n",
      "label: U+6900\n",
      "label: U+6905\n",
      "label: U+690B\n",
      "label: U+690D\n",
      "label: U+690E\n",
      "label: U+6912\n",
      "label: U+6919\n",
      "label: U+691C\n",
      "label: U+6939\n",
      "label: U+693D\n",
      "label: U+693F\n",
      "label: U+694A\n",
      "label: U+6951\n",
      "label: U+6953\n",
      "label: U+6959\n",
      "label: U+695A\n",
      "label: U+695C\n",
      "label: U+6960\n",
      "label: U+6961\n",
      "label: U+696B\n",
      "label: U+696D\n",
      "label: U+696F\n",
      "label: U+6974\n",
      "label: U+6975\n",
      "label: U+697C\n",
      "label: U+697D\n",
      "label: U+6982\n",
      "label: U+698A\n",
      "label: U+698E\n",
      "label: U+6991\n",
      "label: U+6995\n",
      "label: U+699B\n",
      "label: U+69AA\n",
      "label: U+69AE\n",
      "label: U+69B4\n",
      "label: U+69BE\n",
      "label: U+69BF\n",
      "label: U+69C1\n",
      "label: U+69C3\n",
      "label: U+69C7\n",
      "label: U+69CB\n",
      "label: U+69CC\n",
      "label: U+69CD\n",
      "label: U+69CE\n",
      "label: U+69D6\n",
      "label: U+69D8\n",
      "label: U+69D9\n",
      "label: U+69EA\n",
      "label: U+69F9\n",
      "label: U+69FF\n",
      "label: U+6A02\n",
      "label: U+6A0B\n",
      "label: U+6A12\n",
      "label: U+6A13\n",
      "label: U+6A14\n",
      "label: U+6A19\n",
      "label: U+6A1F\n",
      "label: U+6A21\n",
      "label: U+6A23\n",
      "label: U+6A29\n",
      "label: U+6A2A\n",
      "label: U+6A39\n",
      "label: U+6A3D\n",
      "label: U+6A47\n",
      "label: U+6A4B\n",
      "label: U+6A58\n",
      "label: U+6A59\n",
      "label: U+6A5F\n",
      "label: U+6A61\n",
      "label: U+6A66\n",
      "label: U+6A80\n",
      "label: U+6A8B\n",
      "label: U+6A8E\n",
      "label: U+6A90\n",
      "label: U+6A97\n",
      "label: U+6A9C\n",
      "label: U+6ABB\n",
      "label: U+6AC2\n",
      "label: U+6AC3\n",
      "label: U+6AD3\n",
      "label: U+6ADA\n",
      "label: U+6ADB\n",
      "label: U+6AE8\n",
      "label: U+6AFB\n",
      "label: U+6B0A\n",
      "label: U+6B1D\n",
      "label: U+6B20\n",
      "label: U+6B21\n",
      "label: U+6B27\n",
      "label: U+6B2C\n",
      "label: U+6B32\n",
      "label: U+6B3A\n",
      "label: U+6B3D\n",
      "label: U+6B47\n",
      "label: U+6B4C\n",
      "label: U+6B4E\n",
      "label: U+6B53\n",
      "label: U+6B5F\n",
      "label: U+6B61\n",
      "label: U+6B62\n",
      "label: U+6B63\n",
      "label: U+6B64\n",
      "label: U+6B66\n",
      "label: U+6B69\n",
      "label: U+6B6F\n",
      "label: U+6B73\n",
      "label: U+6B74\n",
      "label: U+6B78\n",
      "label: U+6B7B\n",
      "label: U+6B8A\n",
      "label: U+6B8B\n",
      "label: U+6B96\n",
      "label: U+6B98\n",
      "label: U+6B9E\n",
      "label: U+6BAF\n",
      "label: U+6BB5\n",
      "label: U+6BBA\n",
      "label: U+6BBB\n",
      "label: U+6BBC\n",
      "label: U+6BBD\n",
      "label: U+6BBF\n",
      "label: U+6BC0\n",
      "label: U+6BC6\n",
      "label: U+6BCB\n",
      "label: U+6BCD\n",
      "label: U+6BCE\n",
      "label: U+6BD2\n",
      "label: U+6BD4\n",
      "label: U+6BD8\n",
      "label: U+6BDB\n",
      "label: U+6BDF\n",
      "label: U+6BEB\n",
      "label: U+6BEC\n",
      "label: U+6C08\n",
      "label: U+6C0F\n",
      "label: U+6C11\n",
      "label: U+6C17\n",
      "label: U+6C23\n",
      "label: U+6C34\n",
      "label: U+6C37\n",
      "label: U+6C38\n",
      "label: U+6C40\n",
      "label: U+6C41\n",
      "label: U+6C42\n",
      "label: U+6C50\n",
      "label: U+6C57\n",
      "label: U+6C5A\n",
      "label: U+6C5D\n",
      "label: U+6C5F\n",
      "label: U+6C60\n",
      "label: U+6C70\n",
      "label: U+6C72\n",
      "label: U+6C7A\n",
      "label: U+6C7E\n",
      "label: U+6C83\n",
      "label: U+6C88\n",
      "label: U+6C92\n",
      "label: U+6C93\n",
      "label: U+6C96\n",
      "label: U+6C99\n",
      "label: U+6CA1\n",
      "label: U+6CA2\n",
      "label: U+6CAB\n",
      "label: U+6CB3\n",
      "label: U+6CB8\n",
      "label: U+6CB9\n",
      "label: U+6CBB\n",
      "label: U+6CBC\n",
      "label: U+6CBD\n",
      "label: U+6CC1\n",
      "label: U+6CC4\n",
      "label: U+6CC9\n",
      "label: U+6CCA\n",
      "label: U+6CD4\n",
      "label: U+6CD5\n",
      "label: U+6CE1\n",
      "label: U+6CE2\n",
      "label: U+6CE3\n",
      "label: U+6CE5\n",
      "label: U+6CE8\n",
      "label: U+6CEA\n",
      "label: U+6CF0\n",
      "label: U+6CF3\n",
      "label: U+6D0B\n",
      "label: U+6D12\n",
      "label: U+6D17\n",
      "label: U+6D1B\n",
      "label: U+6D1E\n",
      "label: U+6D25\n",
      "label: U+6D29\n",
      "label: U+6D2A\n",
      "label: U+6D32\n",
      "label: U+6D3B\n",
      "label: U+6D3E\n",
      "label: U+6D41\n",
      "label: U+6D44\n",
      "label: U+6D45\n",
      "label: U+6D5C\n",
      "label: U+6D66\n",
      "label: U+6D6A\n",
      "label: U+6D6E\n",
      "label: U+6D74\n",
      "label: U+6D77\n",
      "label: U+6D78\n",
      "label: U+6D85\n",
      "label: U+6D88\n",
      "label: U+6D95\n",
      "label: U+6D99\n",
      "label: U+6DAF\n",
      "label: U+6DB2\n",
      "label: U+6DB8\n",
      "label: U+6DBC\n",
      "label: U+6DBF\n",
      "label: U+6DC0\n",
      "label: U+6DCB\n",
      "label: U+6DE1\n",
      "label: U+6DE8\n",
      "label: U+6DEB\n",
      "label: U+6DF1\n",
      "label: U+6DF3\n",
      "label: U+6DF5\n",
      "label: U+6DF7\n",
      "label: U+6DF9\n",
      "label: U+6DFA\n",
      "label: U+6DFB\n",
      "label: U+6E05\n",
      "label: U+6E07\n",
      "label: U+6E08\n",
      "label: U+6E09\n",
      "label: U+6E0B\n",
      "label: U+6E13\n",
      "label: U+6E1B\n",
      "label: U+6E20\n",
      "label: U+6E21\n",
      "label: U+6E23\n",
      "label: U+6E24\n",
      "label: U+6E29\n",
      "label: U+6E2C\n",
      "label: U+6E38\n",
      "label: U+6E3E\n",
      "label: U+6E4A\n",
      "label: U+6E56\n",
      "label: U+6E5B\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+6E67\n",
      "label: U+6E6F\n",
      "label: U+6E7E\n",
      "label: U+6E7F\n",
      "label: U+6E80\n",
      "label: U+6E8F\n",
      "label: U+6E90\n",
      "label: U+6E9C\n",
      "label: U+6E9D\n",
      "label: U+6EA2\n",
      "label: U+6EBA\n",
      "label: U+6EC4\n",
      "label: U+6EC5\n",
      "label: U+6ED1\n",
      "label: U+6ED3\n",
      "label: U+6EDD\n",
      "label: U+6EF4\n",
      "label: U+6EF8\n",
      "label: U+6EFF\n",
      "label: U+6F01\n",
      "label: U+6F02\n",
      "label: U+6F06\n",
      "label: U+6F09\n",
      "label: U+6F0F\n",
      "label: U+6F14\n",
      "label: U+6F15\n",
      "label: U+6F22\n",
      "label: U+6F23\n",
      "label: U+6F2B\n",
      "label: U+6F2C\n",
      "label: U+6F32\n",
      "label: U+6F38\n",
      "label: U+6F3B\n",
      "label: U+6F3F\n",
      "label: U+6F45\n",
      "label: U+6F54\n",
      "label: U+6F5B\n",
      "label: U+6F5C\n",
      "label: U+6F5F\n",
      "label: U+6F64\n",
      "label: U+6F66\n",
      "label: U+6F6E\n",
      "label: U+6F6F\n",
      "label: U+6F70\n",
      "label: U+6F78\n",
      "label: U+6F84\n",
      "label: U+6FA1\n",
      "label: U+6FA4\n",
      "label: U+6FB3\n",
      "label: U+6FC1\n",
      "label: U+6FC2\n",
      "label: U+6FC3\n",
      "label: U+6FD5\n",
      "label: U+6FDF\n",
      "label: U+6FE1\n",
      "label: U+6FEB\n",
      "label: U+6FEF\n",
      "label: U+6FF1\n",
      "label: U+6FF5\n",
      "label: U+7009\n",
      "label: U+7027\n",
      "label: U+702C\n",
      "label: U+7030\n",
      "label: U+7058\n",
      "label: U+706B\n",
      "label: U+706F\n",
      "label: U+7070\n",
      "label: U+7078\n",
      "label: U+707D\n",
      "label: U+7089\n",
      "label: U+708A\n",
      "label: U+708E\n",
      "label: U+7092\n",
      "label: U+7099\n",
      "label: U+70AC\n",
      "label: U+70AD\n",
      "label: U+70AE\n",
      "label: U+70B3\n",
      "label: U+70B9\n",
      "label: U+70BA\n",
      "label: U+70C8\n",
      "label: U+70CF\n",
      "label: U+70D9\n",
      "label: U+70DF\n",
      "label: U+70F9\n",
      "label: U+7109\n",
      "label: U+7114\n",
      "label: U+7119\n",
      "label: U+711A\n",
      "label: U+7121\n",
      "label: U+7126\n",
      "label: U+7136\n",
      "label: U+713C\n",
      "label: U+7149\n",
      "label: U+714E\n",
      "label: U+7156\n",
      "label: U+7159\n",
      "label: U+7164\n",
      "label: U+7165\n",
      "label: U+7167\n",
      "label: U+7169\n",
      "label: U+716C\n",
      "label: U+716E\n",
      "label: U+717B\n",
      "label: U+7187\n",
      "label: U+718A\n",
      "label: U+719F\n",
      "label: U+71A8\n",
      "label: U+71AC\n",
      "label: U+71B1\n",
      "label: U+71BE\n",
      "label: U+71C3\n",
      "label: U+71C8\n",
      "label: U+71D2\n",
      "label: U+71D5\n",
      "label: U+71D7\n",
      "label: U+71DF\n",
      "label: U+71E5\n",
      "label: U+71E7\n",
      "label: U+71ED\n",
      "label: U+71F5\n",
      "label: U+71FF\n",
      "label: U+7208\n",
      "label: U+7210\n",
      "label: U+721B\n",
      "label: U+722A\n",
      "label: U+722C\n",
      "label: U+722D\n",
      "label: U+7230\n",
      "label: U+7232\n",
      "label: U+7235\n",
      "label: U+7236\n",
      "label: U+7239\n",
      "label: U+723A\n",
      "label: U+723C\n",
      "label: U+723D\n",
      "label: U+723E\n",
      "label: U+7240\n",
      "label: U+7246\n",
      "label: U+7247\n",
      "label: U+7248\n",
      "label: U+724C\n",
      "label: U+7252\n",
      "label: U+7259\n",
      "label: U+725B\n",
      "label: U+725D\n",
      "label: U+7261\n",
      "label: U+7262\n",
      "label: U+7267\n",
      "label: U+7269\n",
      "label: U+7279\n",
      "label: U+727D\n",
      "label: U+7280\n",
      "label: U+72A0\n",
      "label: U+72A2\n",
      "label: U+72AC\n",
      "label: U+72AF\n",
      "label: U+72B6\n",
      "label: U+72C2\n",
      "label: U+72C4\n",
      "label: U+72CE\n",
      "label: U+72D0\n",
      "label: U+72D7\n",
      "label: U+72DB\n",
      "label: U+72E9\n",
      "label: U+72EC\n",
      "label: U+72ED\n",
      "label: U+72F8\n",
      "label: U+72F9\n",
      "label: U+72FC\n",
      "label: U+731B\n",
      "label: U+731C\n",
      "label: U+731F\n",
      "label: U+7325\n",
      "label: U+7328\n",
      "label: U+7329\n",
      "label: U+732A\n",
      "label: U+732B\n",
      "label: U+732E\n",
      "label: U+7334\n",
      "label: U+7336\n",
      "label: U+733F\n",
      "label: U+7344\n",
      "label: U+7345\n",
      "label: U+734F\n",
      "label: U+7363\n",
      "label: U+7368\n",
      "label: U+736C\n",
      "label: U+7372\n",
      "label: U+7375\n",
      "label: U+7378\n",
      "label: U+737A\n",
      "label: U+737B\n",
      "label: U+7384\n",
      "label: U+7387\n",
      "label: U+7389\n",
      "label: U+738B\n",
      "label: U+7396\n",
      "label: U+73B2\n",
      "label: U+73CA\n",
      "label: U+73CD\n",
      "label: U+73E0\n",
      "label: U+73ED\n",
      "label: U+73FE\n",
      "label: U+7403\n",
      "label: U+7406\n",
      "label: U+7409\n",
      "label: U+7422\n",
      "label: U+7433\n",
      "label: U+7434\n",
      "label: U+7459\n",
      "label: U+745A\n",
      "label: U+745E\n",
      "label: U+7460\n",
      "label: U+746A\n",
      "label: U+7483\n",
      "label: U+74A7\n",
      "label: U+74B0\n",
      "label: U+74BD\n",
      "label: U+74CA\n",
      "label: U+74CF\n",
      "label: U+74DC\n",
      "label: U+74E0\n",
      "label: U+74E2\n",
      "label: U+74E6\n",
      "label: U+74EE\n",
      "label: U+74F6\n",
      "label: U+7500\n",
      "label: U+7511\n",
      "label: U+7515\n",
      "label: U+7518\n",
      "label: U+751A\n",
      "label: U+751C\n",
      "label: U+751E\n",
      "label: U+751F\n",
      "label: U+7523\n",
      "label: U+7528\n",
      "label: U+752B\n",
      "label: U+7530\n",
      "label: U+7531\n",
      "label: U+7532\n",
      "label: U+7533\n",
      "label: U+7537\n",
      "label: U+753A\n",
      "label: U+753B\n",
      "label: U+7544\n",
      "label: U+754C\n",
      "label: U+7551\n",
      "label: U+7552\n",
      "label: U+7554\n",
      "label: U+7559\n",
      "label: U+755A\n",
      "label: U+755C\n",
      "label: U+755D\n",
      "label: U+7560\n",
      "label: U+7562\n",
      "label: U+7565\n",
      "label: U+7566\n",
      "label: U+7567\n",
      "label: U+756A\n",
      "label: U+756B\n",
      "label: U+7570\n",
      "label: U+7573\n",
      "label: U+7576\n",
      "label: U+7578\n",
      "label: U+757F\n",
      "label: U+758A\n",
      "label: U+758B\n",
      "label: U+758E\n",
      "label: U+7591\n",
      "label: U+75A3\n",
      "label: U+75AB\n",
      "label: U+75B1\n",
      "label: U+75B2\n",
      "label: U+75B3\n",
      "label: U+75B5\n",
      "label: U+75B9\n",
      "label: U+75BE\n",
      "label: U+75C5\n",
      "label: U+75D8\n",
      "label: U+75DB\n",
      "label: U+75DE\n",
      "label: U+75E2\n",
      "label: U+75E9\n",
      "label: U+75F0\n",
      "label: U+75F4\n",
      "label: U+75FF\n",
      "label: U+761F\n",
      "label: U+7621\n",
      "label: U+7624\n",
      "label: U+7627\n",
      "label: U+7630\n",
      "label: U+7642\n",
      "label: U+7652\n",
      "label: U+7656\n",
      "label: U+7669\n",
      "label: U+7678\n",
      "label: U+767A\n",
      "label: U+767B\n",
      "label: U+767C\n",
      "label: U+767D\n",
      "label: U+767E\n",
      "label: U+7682\n",
      "label: U+7683\n",
      "label: U+7684\n",
      "label: U+7686\n",
      "label: U+7687\n",
      "label: U+7688\n",
      "label: U+76AE\n",
      "label: U+76B7\n",
      "label: U+76B9\n",
      "label: U+76BA\n",
      "label: U+76BF\n",
      "label: U+76C3\n",
      "label: U+76C6\n",
      "label: U+76CA\n",
      "label: U+76D2\n",
      "label: U+76D6\n",
      "label: U+76D7\n",
      "label: U+76DB\n",
      "label: U+76DC\n",
      "label: U+76DE\n",
      "label: U+76DF\n",
      "label: U+76E1\n",
      "label: U+76E4\n",
      "label: U+76E5\n",
      "label: U+76E7\n",
      "label: U+76EE\n",
      "label: U+76F2\n",
      "label: U+76F4\n",
      "label: U+76F8\n",
      "label: U+7701\n",
      "label: U+7709\n",
      "label: U+770B\n",
      "label: U+771B\n",
      "label: U+771E\n",
      "label: U+771F\n",
      "label: U+7720\n",
      "label: U+7724\n",
      "label: U+7737\n",
      "label: U+7738\n",
      "label: U+773A\n",
      "label: U+773C\n",
      "label: U+7740\n",
      "label: U+775A\n",
      "label: U+775C\n",
      "label: U+7761\n",
      "label: U+7762\n",
      "label: U+7763\n",
      "label: U+7766\n",
      "label: U+7768\n",
      "label: U+777E\n",
      "label: U+77AC\n",
      "label: U+77BB\n",
      "label: U+77BC\n",
      "label: U+77BF\n",
      "label: U+77DA\n",
      "label: U+77DB\n",
      "label: U+77E2\n",
      "label: U+77E3\n",
      "label: U+77E5\n",
      "label: U+77E9\n",
      "label: U+77ED\n",
      "label: U+77F3\n",
      "label: U+7802\n",
      "label: U+7814\n",
      "label: U+7815\n",
      "label: U+7825\n",
      "label: U+7827\n",
      "label: U+7832\n",
      "label: U+7834\n",
      "label: U+783A\n",
      "label: U+786B\n",
      "label: U+786F\n",
      "label: U+7881\n",
      "label: U+788E\n",
      "label: U+7891\n",
      "label: U+7893\n",
      "label: U+7897\n",
      "label: U+78A7\n",
      "label: U+78C1\n",
      "label: U+78D0\n",
      "label: U+78E4\n",
      "label: U+78E8\n",
      "label: U+78EF\n",
      "label: U+78FD\n",
      "label: U+7912\n",
      "label: U+7919\n",
      "label: U+792B\n",
      "label: U+793A\n",
      "label: U+793C\n",
      "label: U+793E\n",
      "label: U+7947\n",
      "label: U+7948\n",
      "label: U+7950\n",
      "label: U+7953\n",
      "label: U+7955\n",
      "label: U+7956\n",
      "label: U+7957\n",
      "label: U+795A\n",
      "label: U+795D\n",
      "label: U+795E\n",
      "label: U+795F\n",
      "label: U+7960\n",
      "label: U+7962\n",
      "label: U+7965\n",
      "label: U+796D\n",
      "label: U+7977\n",
      "label: U+797F\n",
      "label: U+7981\n",
      "label: U+7984\n",
      "label: U+7985\n",
      "label: U+798A\n",
      "label: U+798D\n",
      "label: U+798F\n",
      "label: U+79A6\n",
      "label: U+79AA\n",
      "label: U+79AE\n",
      "label: U+79B0\n",
      "label: U+79B1\n",
      "label: U+79BD\n",
      "label: U+79BF\n",
      "label: U+79C0\n",
      "label: U+79C1\n",
      "label: U+79CB\n",
      "label: U+79D1\n",
      "label: U+79D8\n",
      "label: U+79E1\n",
      "label: U+79E3\n",
      "label: U+79E4\n",
      "label: U+79E6\n",
      "label: U+79E7\n",
      "label: U+79E9\n",
      "label: U+79EC\n",
      "label: U+79F0\n",
      "label: U+79FB\n",
      "label: U+7A00\n",
      "label: U+7A0B\n",
      "label: U+7A0D\n",
      "label: U+7A17\n",
      "label: U+7A1A\n",
      "label: U+7A1C\n",
      "label: U+7A20\n",
      "label: U+7A2E\n",
      "label: U+7A31\n",
      "label: U+7A32\n",
      "label: U+7A37\n",
      "label: U+7A3B\n",
      "label: U+7A3C\n",
      "label: U+7A3D\n",
      "label: U+7A3E\n",
      "label: U+7A3F\n",
      "label: U+7A40\n",
      "label: U+7A42\n",
      "label: U+7A4D\n",
      "label: U+7A4F\n",
      "label: U+7A57\n",
      "label: U+7A62\n",
      "label: U+7A6A\n",
      "label: U+7A74\n",
      "label: U+7A76\n",
      "label: U+7A7A\n",
      "label: U+7A7F\n",
      "label: U+7A81\n",
      "label: U+7A88\n",
      "label: U+7A93\n",
      "label: U+7A95\n",
      "label: U+7A97\n",
      "label: U+7A9F\n",
      "label: U+7AA9\n",
      "label: U+7AAE\n",
      "label: U+7AB6\n",
      "label: U+7ABB\n",
      "label: U+7AC3\n",
      "label: U+7AC4\n",
      "label: U+7AC5\n",
      "label: U+7AC8\n",
      "label: U+7ACA\n",
      "label: U+7ACB\n",
      "label: U+7AD2\n",
      "label: U+7ADC\n",
      "label: U+7ADF\n",
      "label: U+7AE0\n",
      "label: U+7AE5\n",
      "label: U+7AEA\n",
      "label: U+7AEF\n",
      "label: U+7AF6\n",
      "label: U+7AF9\n",
      "label: U+7AFA\n",
      "label: U+7AFF\n",
      "label: U+7B06\n",
      "label: U+7B0B\n",
      "label: U+7B11\n",
      "label: U+7B18\n",
      "label: U+7B19\n",
      "label: U+7B1B\n",
      "label: U+7B20\n",
      "label: U+7B25\n",
      "label: U+7B26\n",
      "label: U+7B2C\n",
      "label: U+7B39\n",
      "label: U+7B46\n",
      "label: U+7B48\n",
      "label: U+7B49\n",
      "label: U+7B4B\n",
      "label: U+7B4D\n",
      "label: U+7B4F\n",
      "label: U+7B50\n",
      "label: U+7B51\n",
      "label: U+7B52\n",
      "label: U+7B54\n",
      "label: U+7B56\n",
      "label: U+7B65\n",
      "label: U+7B67\n",
      "label: U+7B6C\n",
      "label: U+7B6D\n",
      "label: U+7B72\n",
      "label: U+7B75\n",
      "label: U+7B87\n",
      "label: U+7B8F\n",
      "label: U+7B92\n",
      "label: U+7B94\n",
      "label: U+7B95\n",
      "label: U+7B97\n",
      "label: U+7BA1\n",
      "label: U+7BAA\n",
      "label: U+7BAD\n",
      "label: U+7BB1\n",
      "label: U+7BB8\n",
      "label: U+7BC0\n",
      "label: U+7BC4\n",
      "label: U+7BC6\n",
      "label: U+7BC7\n",
      "label: U+7BC9\n",
      "label: U+7BD7\n",
      "label: U+7BE0\n",
      "label: U+7BE4\n",
      "label: U+7BE6\n",
      "label: U+7BE9\n",
      "label: U+7BED\n",
      "label: U+7BF7\n",
      "label: U+7BFC\n",
      "label: U+7C00\n",
      "label: U+7C0D\n",
      "label: U+7C11\n",
      "label: U+7C12\n",
      "label: U+7C17\n",
      "label: U+7C1F\n",
      "label: U+7C21\n",
      "label: U+7C3E\n",
      "label: U+7C3F\n",
      "label: U+7C43\n",
      "label: U+7C46\n",
      "label: U+7C4D\n",
      "label: U+7C4F\n",
      "label: U+7C54\n",
      "label: U+7C5F\n",
      "label: U+7C60\n",
      "label: U+7C69\n",
      "label: U+7C6C\n",
      "label: U+7C73\n",
      "label: U+7C7E\n",
      "label: U+7C82\n",
      "label: U+7C83\n",
      "label: U+7C89\n",
      "label: U+7C8B\n",
      "label: U+7C92\n",
      "label: U+7C95\n",
      "label: U+7C97\n",
      "label: U+7C98\n",
      "label: U+7C9F\n",
      "label: U+7CA5\n",
      "label: U+7CA7\n",
      "label: U+7CAE\n",
      "label: U+7CB2\n",
      "label: U+7CB3\n",
      "label: U+7CB9\n",
      "label: U+7CBD\n",
      "label: U+7CBE\n",
      "label: U+7CC0\n",
      "label: U+7CC4\n",
      "label: U+7CCA\n",
      "label: U+7CD5\n",
      "label: U+7CD6\n",
      "label: U+7CDC\n",
      "label: U+7CDD\n",
      "label: U+7CDE\n",
      "label: U+7CDF\n",
      "label: U+7CE0\n",
      "label: U+7CE7\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+7CEF\n",
      "label: U+7CF6\n",
      "label: U+7CF8\n",
      "label: U+7CFA\n",
      "label: U+7CFB\n",
      "label: U+7D00\n",
      "label: U+7D02\n",
      "label: U+7D04\n",
      "label: U+7D05\n",
      "label: U+7D0B\n",
      "label: U+7D0D\n",
      "label: U+7D10\n",
      "label: U+7D17\n",
      "label: U+7D19\n",
      "label: U+7D1B\n",
      "label: U+7D20\n",
      "label: U+7D21\n",
      "label: U+7D22\n",
      "label: U+7D2B\n",
      "label: U+7D2C\n",
      "label: U+7D2F\n",
      "label: U+7D30\n",
      "label: U+7D35\n",
      "label: U+7D39\n",
      "label: U+7D3A\n",
      "label: U+7D42\n",
      "label: U+7D43\n",
      "label: U+7D44\n",
      "label: U+7D45\n",
      "label: U+7D46\n",
      "label: U+7D4C\n",
      "label: U+7D50\n",
      "label: U+7D5E\n",
      "label: U+7D66\n",
      "label: U+7D68\n",
      "label: U+7D6E\n",
      "label: U+7D71\n",
      "label: U+7D72\n",
      "label: U+7D75\n",
      "label: U+7D76\n",
      "label: U+7D79\n",
      "label: U+7D93\n",
      "label: U+7D99\n",
      "label: U+7D9A\n",
      "label: U+7DA2\n",
      "label: U+7DAD\n",
      "label: U+7DB1\n",
      "label: U+7DB2\n",
      "label: U+7DB8\n",
      "label: U+7DBA\n",
      "label: U+7DBE\n",
      "label: U+7DBF\n",
      "label: U+7DC7\n",
      "label: U+7DCB\n",
      "label: U+7DCF\n",
      "label: U+7DD1\n",
      "label: U+7DD2\n",
      "label: U+7DDA\n",
      "label: U+7DE1\n",
      "label: U+7DE8\n",
      "label: U+7DE9\n",
      "label: U+7DEC\n",
      "label: U+7DF4\n",
      "label: U+7DFB\n",
      "label: U+7E01\n",
      "label: U+7E04\n",
      "label: U+7E0A\n",
      "label: U+7E11\n",
      "label: U+7E1B\n",
      "label: U+7E1E\n",
      "label: U+7E21\n",
      "label: U+7E23\n",
      "label: U+7E26\n",
      "label: U+7E2B\n",
      "label: U+7E2E\n",
      "label: U+7E35\n",
      "label: U+7E37\n",
      "label: U+7E3A\n",
      "label: U+7E3B\n",
      "label: U+7E3D\n",
      "label: U+7E3E\n",
      "label: U+7E41\n",
      "label: U+7E45\n",
      "label: U+7E46\n",
      "label: U+7E4A\n",
      "label: U+7E4B\n",
      "label: U+7E4D\n",
      "label: U+7E54\n",
      "label: U+7E55\n",
      "label: U+7E5E\n",
      "label: U+7E61\n",
      "label: U+7E69\n",
      "label: U+7E6A\n",
      "label: U+7E6D\n",
      "label: U+7E70\n",
      "label: U+7E7B\n",
      "label: U+7E7C\n",
      "label: U+7E8C\n",
      "label: U+7E8F\n",
      "label: U+7E92\n",
      "label: U+7E94\n",
      "label: U+7F3A\n",
      "label: U+7F4C\n",
      "label: U+7F54\n",
      "label: U+7F5F\n",
      "label: U+7F60\n",
      "label: U+7F6A\n",
      "label: U+7F6E\n",
      "label: U+7F70\n",
      "label: U+7F75\n",
      "label: U+7F77\n",
      "label: U+7F78\n",
      "label: U+7F79\n",
      "label: U+7F85\n",
      "label: U+7F88\n",
      "label: U+7F8A\n",
      "label: U+7F8E\n",
      "label: U+7F94\n",
      "label: U+7F9A\n",
      "label: U+7F9E\n",
      "label: U+7FA4\n",
      "label: U+7FA8\n",
      "label: U+7FA9\n",
      "label: U+7FAE\n",
      "label: U+7FB9\n",
      "label: U+7FBD\n",
      "label: U+7FC1\n",
      "label: U+7FC5\n",
      "label: U+7FCC\n",
      "label: U+7FD2\n",
      "label: U+7FD4\n",
      "label: U+7FE0\n",
      "label: U+7FE1\n",
      "label: U+7FEB\n",
      "label: U+7FF0\n",
      "label: U+7FFC\n",
      "label: U+8001\n",
      "label: U+8003\n",
      "label: U+8004\n",
      "label: U+8005\n",
      "label: U+8006\n",
      "label: U+800C\n",
      "label: U+8010\n",
      "label: U+8012\n",
      "label: U+8015\n",
      "label: U+8018\n",
      "label: U+8033\n",
      "label: U+803B\n",
      "label: U+803D\n",
      "label: U+804A\n",
      "label: U+8056\n",
      "label: U+8058\n",
      "label: U+805A\n",
      "label: U+805E\n",
      "label: U+805F\n",
      "label: U+8061\n",
      "label: U+8062\n",
      "label: U+8072\n",
      "label: U+8073\n",
      "label: U+8074\n",
      "label: U+8077\n",
      "label: U+8086\n",
      "label: U+8089\n",
      "label: U+808C\n",
      "label: U+8096\n",
      "label: U+8098\n",
      "label: U+809A\n",
      "label: U+809D\n",
      "label: U+80A1\n",
      "label: U+80A5\n",
      "label: U+80A9\n",
      "label: U+80AF\n",
      "label: U+80B1\n",
      "label: U+80B2\n",
      "label: U+80B4\n",
      "label: U+80C3\n",
      "label: U+80C6\n",
      "label: U+80CC\n",
      "label: U+80CE\n",
      "label: U+80DD\n",
      "label: U+80DE\n",
      "label: U+80E1\n",
      "label: U+80E4\n",
      "label: U+80EF\n",
      "label: U+80F4\n",
      "label: U+80F8\n",
      "label: U+80FC\n",
      "label: U+80FD\n",
      "label: U+8102\n",
      "label: U+8107\n",
      "label: U+810A\n",
      "label: U+811A\n",
      "label: U+811B\n",
      "label: U+8123\n",
      "label: U+8129\n",
      "label: U+812F\n",
      "label: U+8131\n",
      "label: U+8133\n",
      "label: U+813E\n",
      "label: U+814B\n",
      "label: U+814E\n",
      "label: U+8150\n",
      "label: U+8155\n",
      "label: U+8165\n",
      "label: U+8166\n",
      "label: U+8168\n",
      "label: U+816B\n",
      "label: U+816E\n",
      "label: U+8170\n",
      "label: U+8178\n",
      "label: U+8179\n",
      "label: U+817F\n",
      "label: U+8182\n",
      "label: U+819A\n",
      "label: U+819D\n",
      "label: U+81A0\n",
      "label: U+81B3\n",
      "label: U+81BD\n",
      "label: U+81BE\n",
      "label: U+81C2\n",
      "label: U+81C6\n",
      "label: U+81C8\n",
      "label: U+81C9\n",
      "label: U+81CD\n",
      "label: U+81D1\n",
      "label: U+81D8\n",
      "label: U+81E3\n",
      "label: U+81E5\n",
      "label: U+81E8\n",
      "label: U+81EA\n",
      "label: U+81ED\n",
      "label: U+81F3\n",
      "label: U+81F4\n",
      "label: U+81FA\n",
      "label: U+81FC\n",
      "label: U+8201\n",
      "label: U+8202\n",
      "label: U+8205\n",
      "label: U+8207\n",
      "label: U+8208\n",
      "label: U+820A\n",
      "label: U+820C\n",
      "label: U+820D\n",
      "label: U+820E\n",
      "label: U+8216\n",
      "label: U+8218\n",
      "label: U+821B\n",
      "label: U+821C\n",
      "label: U+821E\n",
      "label: U+821F\n",
      "label: U+822A\n",
      "label: U+822C\n",
      "label: U+8236\n",
      "label: U+8239\n",
      "label: U+8258\n",
      "label: U+826A\n",
      "label: U+826F\n",
      "label: U+8271\n",
      "label: U+8272\n",
      "label: U+8276\n",
      "label: U+8278\n",
      "label: U+827E\n",
      "label: U+828B\n",
      "label: U+828D\n",
      "label: U+8292\n",
      "label: U+8299\n",
      "label: U+829D\n",
      "label: U+82A5\n",
      "label: U+82A6\n",
      "label: U+82AD\n",
      "label: U+82B1\n",
      "label: U+82B3\n",
      "label: U+82B8\n",
      "label: U+82B9\n",
      "label: U+82BD\n",
      "label: U+82D1\n",
      "label: U+82D3\n",
      "label: U+82D4\n",
      "label: U+82D7\n",
      "label: U+82DE\n",
      "label: U+82DF\n",
      "label: U+82E1\n",
      "label: U+82E3\n",
      "label: U+82E5\n",
      "label: U+82E6\n",
      "label: U+82E7\n",
      "label: U+82EA\n",
      "label: U+82EB\n",
      "label: U+82F1\n",
      "label: U+82F5\n",
      "label: U+8302\n",
      "label: U+8303\n",
      "label: U+8304\n",
      "label: U+8305\n",
      "label: U+830E\n",
      "label: U+8317\n",
      "label: U+8318\n",
      "label: U+831C\n",
      "label: U+8328\n",
      "label: U+8331\n",
      "label: U+8334\n",
      "label: U+8336\n",
      "label: U+8338\n",
      "label: U+8339\n",
      "label: U+833C\n",
      "label: U+8347\n",
      "label: U+8349\n",
      "label: U+834A\n",
      "label: U+8352\n",
      "label: U+8377\n",
      "label: U+837B\n",
      "label: U+837C\n",
      "label: U+8385\n",
      "label: U+838A\n",
      "label: U+838E\n",
      "label: U+8395\n",
      "label: U+8396\n",
      "label: U+839A\n",
      "label: U+839E\n",
      "label: U+839F\n",
      "label: U+83A7\n",
      "label: U+83A8\n",
      "label: U+83AB\n",
      "label: U+83B1\n",
      "label: U+83B5\n",
      "label: U+83BE\n",
      "label: U+83C1\n",
      "label: U+83C5\n",
      "label: U+83CA\n",
      "label: U+83CC\n",
      "label: U+83D3\n",
      "label: U+83D4\n",
      "label: U+83D6\n",
      "label: U+83D8\n",
      "label: U+83DC\n",
      "label: U+83DD\n",
      "label: U+83DF\n",
      "label: U+83E9\n",
      "label: U+83EA\n",
      "label: U+83EB\n",
      "label: U+83EF\n",
      "label: U+83F0\n",
      "label: U+83F1\n",
      "label: U+83F4\n",
      "label: U+83FD\n",
      "label: U+8404\n",
      "label: U+840A\n",
      "label: U+840C\n",
      "label: U+840D\n",
      "label: U+8413\n",
      "label: U+8420\n",
      "label: U+8429\n",
      "label: U+842C\n",
      "label: U+8431\n",
      "label: U+8438\n",
      "label: U+843D\n",
      "label: U+8449\n",
      "label: U+844E\n",
      "label: U+8457\n",
      "label: U+845B\n",
      "label: U+845C\n",
      "label: U+8461\n",
      "label: U+8463\n",
      "label: U+8466\n",
      "label: U+846C\n",
      "label: U+846D\n",
      "label: U+8471\n",
      "label: U+8475\n",
      "label: U+8476\n",
      "label: U+847A\n",
      "label: U+8494\n",
      "label: U+8499\n",
      "label: U+849C\n",
      "label: U+849F\n",
      "label: U+84A1\n",
      "label: U+84B2\n",
      "label: U+84B8\n",
      "label: U+84BB\n",
      "label: U+84BC\n",
      "label: U+84BF\n",
      "label: U+84C9\n",
      "label: U+84CB\n",
      "label: U+84CD\n",
      "label: U+84D9\n",
      "label: U+84EC\n",
      "label: U+84EE\n",
      "label: U+84F4\n",
      "label: U+84FC\n",
      "label: U+8500\n",
      "label: U+8513\n",
      "label: U+8514\n",
      "label: U+8515\n",
      "label: U+851E\n",
      "label: U+8521\n",
      "label: U+8526\n",
      "label: U+852D\n",
      "label: U+8535\n",
      "label: U+8540\n",
      "label: U+8543\n",
      "label: U+8549\n",
      "label: U+854E\n",
      "label: U+8557\n",
      "label: U+8562\n",
      "label: U+8563\n",
      "label: U+8568\n",
      "label: U+8569\n",
      "label: U+856A\n",
      "label: U+8577\n",
      "label: U+857A\n",
      "label: U+8581\n",
      "label: U+8584\n",
      "label: U+858A\n",
      "label: U+858F\n",
      "label: U+8591\n",
      "label: U+8594\n",
      "label: U+8597\n",
      "label: U+859C\n",
      "label: U+85A6\n",
      "label: U+85A8\n",
      "label: U+85A9\n",
      "label: U+85AA\n",
      "label: U+85AB\n",
      "label: U+85AC\n",
      "label: U+85AE\n",
      "label: U+85AF\n",
      "label: U+85B7\n",
      "label: U+85B9\n",
      "label: U+85BA\n",
      "label: U+85C1\n",
      "label: U+85CD\n",
      "label: U+85CF\n",
      "label: U+85D0\n",
      "label: U+85DC\n",
      "label: U+85DD\n",
      "label: U+85E4\n",
      "label: U+85E5\n",
      "label: U+85EA\n",
      "label: U+85F7\n",
      "label: U+85FA\n",
      "label: U+85FB\n",
      "label: U+85FF\n",
      "label: U+8602\n",
      "label: U+8607\n",
      "label: U+860B\n",
      "label: U+8613\n",
      "label: U+8618\n",
      "label: U+8621\n",
      "label: U+8629\n",
      "label: U+862D\n",
      "label: U+8630\n",
      "label: U+863C\n",
      "label: U+863F\n",
      "label: U+864E\n",
      "label: U+8655\n",
      "label: U+865A\n",
      "label: U+865E\n",
      "label: U+865F\n",
      "label: U+866B\n",
      "label: U+8671\n",
      "label: U+8679\n",
      "label: U+867B\n",
      "label: U+868A\n",
      "label: U+8693\n",
      "label: U+8695\n",
      "label: U+86A2\n",
      "label: U+86A3\n",
      "label: U+86A4\n",
      "label: U+86AB\n",
      "label: U+86AF\n",
      "label: U+86B6\n",
      "label: U+86B8\n",
      "label: U+86BA\n",
      "label: U+86C1\n",
      "label: U+86C4\n",
      "label: U+86C7\n",
      "label: U+86C9\n",
      "label: U+86CD\n",
      "label: U+86CE\n",
      "label: U+86D9\n",
      "label: U+86DB\n",
      "label: U+86DE\n",
      "label: U+86DF\n",
      "label: U+86E4\n",
      "label: U+86EE\n",
      "label: U+86EF\n",
      "label: U+86F8\n",
      "label: U+86F9\n",
      "label: U+86FE\n",
      "label: U+8700\n",
      "label: U+8702\n",
      "label: U+8703\n",
      "label: U+8706\n",
      "label: U+8708\n",
      "label: U+870A\n",
      "label: U+870D\n",
      "label: U+8710\n",
      "label: U+8711\n",
      "label: U+8718\n",
      "label: U+871C\n",
      "label: U+8725\n",
      "label: U+8729\n",
      "label: U+8734\n",
      "label: U+873B\n",
      "label: U+873E\n",
      "label: U+8749\n",
      "label: U+874B\n",
      "label: U+874E\n",
      "label: U+8753\n",
      "label: U+8755\n",
      "label: U+8757\n",
      "label: U+8759\n",
      "label: U+8760\n",
      "label: U+8764\n",
      "label: U+8766\n",
      "label: U+876E\n",
      "label: U+8774\n",
      "label: U+8776\n",
      "label: U+8778\n",
      "label: U+877F\n",
      "label: U+8782\n",
      "label: U+8787\n",
      "label: U+8789\n",
      "label: U+878D\n",
      "label: U+8796\n",
      "label: U+87A2\n",
      "label: U+87AB\n",
      "label: U+87BA\n",
      "label: U+87BB\n",
      "label: U+87C0\n",
      "label: U+87C4\n",
      "label: U+87C6\n",
      "label: U+87C7\n",
      "label: U+87CB\n",
      "label: U+87D2\n",
      "label: U+87DB\n",
      "label: U+87DF\n",
      "label: U+87E0\n",
      "label: U+87F2\n",
      "label: U+87F6\n",
      "label: U+87F7\n",
      "label: U+87F9\n",
      "label: U+87FE\n",
      "label: U+8803\n",
      "label: U+8805\n",
      "label: U+881C\n",
      "label: U+8821\n",
      "label: U+8822\n",
      "label: U+882E\n",
      "label: U+8831\n",
      "label: U+8836\n",
      "label: U+883B\n",
      "label: U+8840\n",
      "label: U+8846\n",
      "label: U+884C\n",
      "label: U+8852\n",
      "label: U+8853\n",
      "label: U+8857\n",
      "label: U+885B\n",
      "label: U+885D\n",
      "label: U+885E\n",
      "label: U+8862\n",
      "label: U+8863\n",
      "label: U+8868\n",
      "label: U+8870\n",
      "label: U+8872\n",
      "label: U+887D\n",
      "label: U+887E\n",
      "label: U+887F\n",
      "label: U+8881\n",
      "label: U+8882\n",
      "label: U+8888\n",
      "label: U+888B\n",
      "label: U+888D\n",
      "label: U+8896\n",
      "label: U+889E\n",
      "label: U+88AB\n",
      "label: U+88B1\n",
      "label: U+88B4\n",
      "label: U+88B7\n",
      "label: U+88C1\n",
      "label: U+88C2\n",
      "label: U+88C5\n",
      "label: U+88CF\n",
      "label: U+88D4\n",
      "label: U+88D8\n",
      "label: U+88D9\n",
      "label: U+88DC\n",
      "label: U+88DD\n",
      "label: U+88DF\n",
      "label: U+88E1\n",
      "label: U+88F3\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+88F8\n",
      "label: U+88F9\n",
      "label: U+88FD\n",
      "label: U+88FE\n",
      "label: U+8901\n",
      "label: U+890C\n",
      "label: U+8910\n",
      "label: U+8912\n",
      "label: U+8913\n",
      "label: U+8925\n",
      "label: U+8941\n",
      "label: U+8944\n",
      "label: U+894C\n",
      "label: U+8956\n",
      "label: U+895F\n",
      "label: U+8972\n",
      "label: U+897F\n",
      "label: U+8981\n",
      "label: U+8986\n",
      "label: U+898B\n",
      "label: U+898F\n",
      "label: U+8996\n",
      "label: U+8997\n",
      "label: U+899A\n",
      "label: U+89A7\n",
      "label: U+89AA\n",
      "label: U+89B3\n",
      "label: U+89BA\n",
      "label: U+89C0\n",
      "label: U+89D2\n",
      "label: U+89DC\n",
      "label: U+89E3\n",
      "label: U+89E6\n",
      "label: U+89E7\n",
      "label: U+89F8\n",
      "label: U+8A00\n",
      "label: U+8A02\n",
      "label: U+8A08\n",
      "label: U+8A0E\n",
      "label: U+8A13\n",
      "label: U+8A17\n",
      "label: U+8A18\n",
      "label: U+8A1B\n",
      "label: U+8A1D\n",
      "label: U+8A23\n",
      "label: U+8A25\n",
      "label: U+8A2A\n",
      "label: U+8A2D\n",
      "label: U+8A31\n",
      "label: U+8A33\n",
      "label: U+8A34\n",
      "label: U+8A3B\n",
      "label: U+8A3C\n",
      "label: U+8A50\n",
      "label: U+8A54\n",
      "label: U+8A55\n",
      "label: U+8A5B\n",
      "label: U+8A5E\n",
      "label: U+8A60\n",
      "label: U+8A63\n",
      "label: U+8A66\n",
      "label: U+8A69\n",
      "label: U+8A6B\n",
      "label: U+8A6E\n",
      "label: U+8A70\n",
      "label: U+8A71\n",
      "label: U+8A73\n",
      "label: U+8A82\n",
      "label: U+8A85\n",
      "label: U+8A87\n",
      "label: U+8A89\n",
      "label: U+8A8C\n",
      "label: U+8A8D\n",
      "label: U+8A91\n",
      "label: U+8A93\n",
      "label: U+8A95\n",
      "label: U+8A98\n",
      "label: U+8A9E\n",
      "label: U+8AA0\n",
      "label: U+8AA4\n",
      "label: U+8AA6\n",
      "label: U+8AA8\n",
      "label: U+8AAC\n",
      "label: U+8AAD\n",
      "label: U+8AB0\n",
      "label: U+8AB2\n",
      "label: U+8AB9\n",
      "label: U+8ABF\n",
      "label: U+8AC7\n",
      "label: U+8ACB\n",
      "label: U+8ACC\n",
      "label: U+8ACF\n",
      "label: U+8AD6\n",
      "label: U+8ADA\n",
      "label: U+8ADB\n",
      "label: U+8AE7\n",
      "label: U+8AEB\n",
      "label: U+8AED\n",
      "label: U+8AF1\n",
      "label: U+8AF7\n",
      "label: U+8AF8\n",
      "label: U+8AFA\n",
      "label: U+8B00\n",
      "label: U+8B02\n",
      "label: U+8B0E\n",
      "label: U+8B17\n",
      "label: U+8B19\n",
      "label: U+8B1B\n",
      "label: U+8B1D\n",
      "label: U+8B20\n",
      "label: U+8B21\n",
      "label: U+8B2B\n",
      "label: U+8B49\n",
      "label: U+8B4F\n",
      "label: U+8B58\n",
      "label: U+8B5A\n",
      "label: U+8B5C\n",
      "label: U+8B6C\n",
      "label: U+8B70\n",
      "label: U+8B72\n",
      "label: U+8B77\n",
      "label: U+8B7D\n",
      "label: U+8B80\n",
      "label: U+8B83\n",
      "label: U+8B8A\n",
      "label: U+8B90\n",
      "label: U+8B92\n",
      "label: U+8B93\n",
      "label: U+8B9A\n",
      "label: U+8C37\n",
      "label: U+8C3F\n",
      "label: U+8C46\n",
      "label: U+8C48\n",
      "label: U+8C49\n",
      "label: U+8C4A\n",
      "label: U+8C4C\n",
      "label: U+8C50\n",
      "label: U+8C55\n",
      "label: U+8C5A\n",
      "label: U+8C61\n",
      "label: U+8C7A\n",
      "label: U+8C8C\n",
      "label: U+8C9D\n",
      "label: U+8C9E\n",
      "label: U+8CA0\n",
      "label: U+8CA1\n",
      "label: U+8CA2\n",
      "label: U+8CA7\n",
      "label: U+8CA8\n",
      "label: U+8CA9\n",
      "label: U+8CAA\n",
      "label: U+8CAB\n",
      "label: U+8CAC\n",
      "label: U+8CAF\n",
      "label: U+8CB0\n",
      "label: U+8CB3\n",
      "label: U+8CB4\n",
      "label: U+8CB6\n",
      "label: U+8CB7\n",
      "label: U+8CB8\n",
      "label: U+8CBB\n",
      "label: U+8CBC\n",
      "label: U+8CBF\n",
      "label: U+8CC0\n",
      "label: U+8CC3\n",
      "label: U+8CC4\n",
      "label: U+8CC7\n",
      "label: U+8CC8\n",
      "label: U+8CCA\n",
      "label: U+8CCE\n",
      "label: U+8CD1\n",
      "label: U+8CD3\n",
      "label: U+8CDB\n",
      "label: U+8CDC\n",
      "label: U+8CDE\n",
      "label: U+8CE2\n",
      "label: U+8CE3\n",
      "label: U+8CE4\n",
      "label: U+8CE6\n",
      "label: U+8CEA\n",
      "label: U+8CED\n",
      "label: U+8CFA\n",
      "label: U+8CFD\n",
      "label: U+8D04\n",
      "label: U+8D05\n",
      "label: U+8D08\n",
      "label: U+8D0F\n",
      "label: U+8D14\n",
      "label: U+8D16\n",
      "label: U+8D64\n",
      "label: U+8D66\n",
      "label: U+8D67\n",
      "label: U+8D6B\n",
      "label: U+8D70\n",
      "label: U+8D74\n",
      "label: U+8D77\n",
      "label: U+8D85\n",
      "label: U+8D8A\n",
      "label: U+8D99\n",
      "label: U+8DA3\n",
      "label: U+8DB3\n",
      "label: U+8DCB\n",
      "label: U+8DCC\n",
      "label: U+8DCF\n",
      "label: U+8DD6\n",
      "label: U+8DDF\n",
      "label: U+8DE1\n",
      "label: U+8DE8\n",
      "label: U+8DEA\n",
      "label: U+8DEF\n",
      "label: U+8DF3\n",
      "label: U+8DFC\n",
      "label: U+8E06\n",
      "label: U+8E0A\n",
      "label: U+8E0F\n",
      "label: U+8E1D\n",
      "label: U+8E1E\n",
      "label: U+8E2A\n",
      "label: U+8E30\n",
      "label: U+8E44\n",
      "label: U+8E47\n",
      "label: U+8E48\n",
      "label: U+8E4A\n",
      "label: U+8E64\n",
      "label: U+8E74\n",
      "label: U+8E81\n",
      "label: U+8E87\n",
      "label: U+8E8A\n",
      "label: U+8E8D\n",
      "label: U+8EA1\n",
      "label: U+8EAB\n",
      "label: U+8EAC\n",
      "label: U+8EAF\n",
      "label: U+8EB0\n",
      "label: U+8EBE\n",
      "label: U+8EC6\n",
      "label: U+8ECA\n",
      "label: U+8ECD\n",
      "label: U+8ED2\n",
      "label: U+8EE2\n",
      "label: U+8EF8\n",
      "label: U+8EFD\n",
      "label: U+8F03\n",
      "label: U+8F09\n",
      "label: U+8F14\n",
      "label: U+8F15\n",
      "label: U+8F1B\n",
      "label: U+8F1D\n",
      "label: U+8F29\n",
      "label: U+8F2A\n",
      "label: U+8F2F\n",
      "label: U+8F3F\n",
      "label: U+8F49\n",
      "label: U+8F4D\n",
      "label: U+8F5F\n",
      "label: U+8F9B\n",
      "label: U+8F9C\n",
      "label: U+8F9E\n",
      "label: U+8F9F\n",
      "label: U+8FA8\n",
      "label: U+8FB0\n",
      "label: U+8FB1\n",
      "label: U+8FB2\n",
      "label: U+8FBA\n",
      "label: U+8FBB\n",
      "label: U+8FBC\n",
      "label: U+8FC2\n",
      "label: U+8FC4\n",
      "label: U+8FCE\n",
      "label: U+8FD1\n",
      "label: U+8FD4\n",
      "label: U+8FDA\n",
      "label: U+8FE6\n",
      "label: U+8FE9\n",
      "label: U+8FEB\n",
      "label: U+8FEF\n",
      "label: U+8FF0\n",
      "label: U+8FF7\n",
      "label: U+8FF9\n",
      "label: U+8FFA\n",
      "label: U+8FFD\n",
      "label: U+9000\n",
      "label: U+9001\n",
      "label: U+9003\n",
      "label: U+9005\n",
      "label: U+9006\n",
      "label: U+900D\n",
      "label: U+900F\n",
      "label: U+9010\n",
      "label: U+9013\n",
      "label: U+9014\n",
      "label: U+9017\n",
      "label: U+9019\n",
      "label: U+901A\n",
      "label: U+901D\n",
      "label: U+901E\n",
      "label: U+901F\n",
      "label: U+9020\n",
      "label: U+9022\n",
      "label: U+9023\n",
      "label: U+9032\n",
      "label: U+9038\n",
      "label: U+903C\n",
      "label: U+9041\n",
      "label: U+9042\n",
      "label: U+9045\n",
      "label: U+9047\n",
      "label: U+904A\n",
      "label: U+904B\n",
      "label: U+904D\n",
      "label: U+904E\n",
      "label: U+9053\n",
      "label: U+9054\n",
      "label: U+9055\n",
      "label: U+9059\n",
      "label: U+9060\n",
      "label: U+9063\n",
      "label: U+9065\n",
      "label: U+9069\n",
      "label: U+906D\n",
      "label: U+9072\n",
      "label: U+9077\n",
      "label: U+9078\n",
      "label: U+907A\n",
      "label: U+907D\n",
      "label: U+907F\n",
      "label: U+9082\n",
      "label: U+9084\n",
      "label: U+9089\n",
      "label: U+908A\n",
      "label: U+9091\n",
      "label: U+90A3\n",
      "label: U+90A6\n",
      "label: U+90AA\n",
      "label: U+90AF\n",
      "label: U+90C1\n",
      "label: U+90CA\n",
      "label: U+90CE\n",
      "label: U+90DB\n",
      "label: U+90E1\n",
      "label: U+90E8\n",
      "label: U+90ED\n",
      "label: U+90F7\n",
      "label: U+90FD\n",
      "label: U+9119\n",
      "label: U+9132\n",
      "label: U+9149\n",
      "label: U+914C\n",
      "label: U+914D\n",
      "label: U+9152\n",
      "label: U+9154\n",
      "label: U+9156\n",
      "label: U+9162\n",
      "label: U+9169\n",
      "label: U+916C\n",
      "label: U+9174\n",
      "label: U+9177\n",
      "label: U+9178\n",
      "label: U+9189\n",
      "label: U+918D\n",
      "label: U+9190\n",
      "label: U+9192\n",
      "label: U+919C\n",
      "label: U+91A2\n",
      "label: U+91A4\n",
      "label: U+91AA\n",
      "label: U+91AB\n",
      "label: U+91AC\n",
      "label: U+91AF\n",
      "label: U+91B4\n",
      "label: U+91BF\n",
      "label: U+91C4\n",
      "label: U+91C7\n",
      "label: U+91C8\n",
      "label: U+91CB\n",
      "label: U+91CC\n",
      "label: U+91CD\n",
      "label: U+91CE\n",
      "label: U+91CF\n",
      "label: U+91D1\n",
      "label: U+91D8\n",
      "label: U+91DC\n",
      "label: U+91DD\n",
      "label: U+91E1\n",
      "label: U+91E3\n",
      "label: U+91F5\n",
      "label: U+91FF\n",
      "label: U+920D\n",
      "label: U+920E\n",
      "label: U+9214\n",
      "label: U+9234\n",
      "label: U+9237\n",
      "label: U+9244\n",
      "label: U+925B\n",
      "label: U+9262\n",
      "label: U+9264\n",
      "label: U+9266\n",
      "label: U+927E\n",
      "label: U+9280\n",
      "label: U+9283\n",
      "label: U+9285\n",
      "label: U+9295\n",
      "label: U+9298\n",
      "label: U+929A\n",
      "label: U+92AD\n",
      "label: U+92E4\n",
      "label: U+92F8\n",
      "label: U+9304\n",
      "label: U+9306\n",
      "label: U+9310\n",
      "label: U+9320\n",
      "label: U+9322\n",
      "label: U+9326\n",
      "label: U+932B\n",
      "label: U+932C\n",
      "label: U+9332\n",
      "label: U+934B\n",
      "label: U+9354\n",
      "label: U+935B\n",
      "label: U+936C\n",
      "label: U+936E\n",
      "label: U+937C\n",
      "label: U+937E\n",
      "label: U+938C\n",
      "label: U+9394\n",
      "label: U+9396\n",
      "label: U+939A\n",
      "label: U+93AD\n",
      "label: U+93AE\n",
      "label: U+93B0\n",
      "label: U+93CA\n",
      "label: U+93E1\n",
      "label: U+93E4\n",
      "label: U+93E5\n",
      "label: U+9403\n",
      "label: U+9410\n",
      "label: U+9418\n",
      "label: U+942B\n",
      "label: U+9435\n",
      "label: U+9444\n",
      "label: U+9451\n",
      "label: U+9453\n",
      "label: U+9470\n",
      "label: U+9475\n",
      "label: U+947F\n",
      "label: U+9577\n",
      "label: U+9580\n",
      "label: U+9589\n",
      "label: U+958B\n",
      "label: U+958F\n",
      "label: U+9591\n",
      "label: U+9593\n",
      "label: U+9594\n",
      "label: U+9599\n",
      "label: U+95A2\n",
      "label: U+95A3\n",
      "label: U+95A4\n",
      "label: U+95A8\n",
      "label: U+95BB\n",
      "label: U+95BC\n",
      "label: U+95C7\n",
      "label: U+95CC\n",
      "label: U+95CD\n",
      "label: U+95D5\n",
      "label: U+95D6\n",
      "label: U+95DC\n",
      "label: U+95E2\n",
      "label: U+9632\n",
      "label: U+963F\n",
      "label: U+9640\n",
      "label: U+9644\n",
      "label: U+964B\n",
      "label: U+964C\n",
      "label: U+964D\n",
      "label: U+9650\n",
      "label: U+9662\n",
      "label: U+9663\n",
      "label: U+9664\n",
      "label: U+9665\n",
      "label: U+966A\n",
      "label: U+9670\n",
      "label: U+9673\n",
      "label: U+9675\n",
      "label: U+9676\n",
      "label: U+9677\n",
      "label: U+9678\n",
      "label: U+967A\n",
      "label: U+967D\n",
      "label: U+9682\n",
      "label: U+9685\n",
      "label: U+9686\n",
      "label: U+9688\n",
      "label: U+968B\n",
      "label: U+968E\n",
      "label: U+968F\n",
      "label: U+9694\n",
      "label: U+9698\n",
      "label: U+9699\n",
      "label: U+969B\n",
      "label: U+969C\n",
      "label: U+96A0\n",
      "label: U+96A3\n",
      "label: U+96A8\n",
      "label: U+96B1\n",
      "label: U+96BB\n",
      "label: U+96C0\n",
      "label: U+96C1\n",
      "label: U+96C4\n",
      "label: U+96C5\n",
      "label: U+96C6\n",
      "label: U+96C7\n",
      "label: U+96C9\n",
      "label: U+96CC\n",
      "label: U+96D1\n",
      "label: U+96D5\n",
      "label: U+96D6\n",
      "label: U+96D9\n",
      "label: U+96DB\n",
      "label: U+96DC\n",
      "label: U+96DE\n",
      "label: U+96E2\n",
      "label: U+96E3\n",
      "label: U+96E8\n",
      "label: U+96EA\n",
      "label: U+96EB\n",
      "label: U+96F2\n",
      "label: U+96F6\n",
      "label: U+96F7\n",
      "label: U+96F9\n",
      "label: U+96FB\n",
      "label: U+9700\n",
      "label: U+9707\n",
      "label: U+970A\n",
      "label: U+970E\n",
      "label: U+9716\n",
      "label: U+971C\n",
      "label: U+971E\n",
      "label: U+9724\n",
      "label: U+9727\n",
      "label: U+9730\n",
      "label: U+9732\n",
      "label: U+9739\n",
      "label: U+973D\n",
      "label: U+9742\n",
      "label: U+9748\n",
      "label: U+974D\n",
      "label: U+9752\n",
      "label: U+9756\n",
      "label: U+9759\n",
      "label: U+975C\n",
      "label: U+975E\n",
      "label: U+9761\n",
      "label: U+9762\n",
      "label: U+9768\n",
      "label: U+9769\n",
      "label: U+976A\n",
      "label: U+976D\n",
      "label: U+977C\n",
      "label: U+9781\n",
      "label: U+9785\n",
      "label: U+9786\n",
      "label: U+978B\n",
      "label: U+978D\n",
      "label: U+9798\n",
      "label: U+97A0\n",
      "label: U+97C3\n",
      "label: U+97ED\n",
      "label: U+97EE\n",
      "label: U+97F3\n",
      "label: U+97FB\n",
      "label: U+97FF\n",
      "label: U+9802\n",
      "label: U+9803\n",
      "label: U+9805\n",
      "label: U+9806\n",
      "label: U+9808\n",
      "label: U+980C\n",
      "label: U+9810\n",
      "label: U+9811\n",
      "label: U+9813\n",
      "label: U+9817\n",
      "label: U+9818\n",
      "label: U+9821\n",
      "label: U+982C\n",
      "label: U+982D\n",
      "label: U+9830\n",
      "label: U+9834\n",
      "label: U+9837\n",
      "label: U+983B\n",
      "label: U+983C\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label: U+983D\n",
      "label: U+984B\n",
      "label: U+984C\n",
      "label: U+984D\n",
      "label: U+984F\n",
      "label: U+9854\n",
      "label: U+9855\n",
      "label: U+9858\n",
      "label: U+9859\n",
      "label: U+985B\n",
      "label: U+985E\n",
      "label: U+9867\n",
      "label: U+9870\n",
      "label: U+9871\n",
      "label: U+98A8\n",
      "label: U+98AD\n",
      "label: U+98AF\n",
      "label: U+98DB\n",
      "label: U+98DC\n",
      "label: U+98DF\n",
      "label: U+98E2\n",
      "label: U+98E9\n",
      "label: U+98EE\n",
      "label: U+98EF\n",
      "label: U+98F2\n",
      "label: U+98F4\n",
      "label: U+98FC\n",
      "label: U+98FD\n",
      "label: U+98FE\n",
      "label: U+9903\n",
      "label: U+9905\n",
      "label: U+9909\n",
      "label: U+990A\n",
      "label: U+990C\n",
      "label: U+9910\n",
      "label: U+9913\n",
      "label: U+9918\n",
      "label: U+991D\n",
      "label: U+9921\n",
      "label: U+9928\n",
      "label: U+992C\n",
      "label: U+9932\n",
      "label: U+993B\n",
      "label: U+9940\n",
      "label: U+9942\n",
      "label: U+9945\n",
      "label: U+9949\n",
      "label: U+9951\n",
      "label: U+9952\n",
      "label: U+9957\n",
      "label: U+9958\n",
      "label: U+9996\n",
      "label: U+9999\n",
      "label: U+99A8\n",
      "label: U+99AC\n",
      "label: U+99B3\n",
      "label: U+99B4\n",
      "label: U+99C4\n",
      "label: U+99C5\n",
      "label: U+99C6\n",
      "label: U+99C8\n",
      "label: U+99D0\n",
      "label: U+99D2\n",
      "label: U+99D5\n",
      "label: U+99DD\n",
      "label: U+99DF\n",
      "label: U+99ED\n",
      "label: U+99FF\n",
      "label: U+9A12\n",
      "label: U+9A13\n",
      "label: U+9A19\n",
      "label: U+9A28\n",
      "label: U+9A2B\n",
      "label: U+9A37\n",
      "label: U+9A45\n",
      "label: U+9A55\n",
      "label: U+9A57\n",
      "label: U+9A5A\n",
      "label: U+9A62\n",
      "label: U+9A6B\n",
      "label: U+9AA8\n",
      "label: U+9AB8\n",
      "label: U+9AC4\n",
      "label: U+9AD4\n",
      "label: U+9AD8\n",
      "label: U+9AD9\n",
      "label: U+9AE9\n",
      "label: U+9AEA\n",
      "label: U+9AED\n",
      "label: U+9AEE\n",
      "label: U+9AEF\n",
      "label: U+9AF7\n",
      "label: U+9AFB\n",
      "label: U+9B0C\n",
      "label: U+9B12\n",
      "label: U+9B18\n",
      "label: U+9B1A\n",
      "label: U+9B1B\n",
      "label: U+9B1F\n",
      "label: U+9B22\n",
      "label: U+9B29\n",
      "label: U+9B2A\n",
      "label: U+9B2E\n",
      "label: U+9B31\n",
      "label: U+9B3B\n",
      "label: U+9B3C\n",
      "label: U+9B41\n",
      "label: U+9B42\n",
      "label: U+9B44\n",
      "label: U+9B45\n",
      "label: U+9B4F\n",
      "label: U+9B51\n",
      "label: U+9B54\n",
      "label: U+9B58\n",
      "label: U+9B5A\n",
      "label: U+9B5F\n",
      "label: U+9B66\n",
      "label: U+9B6F\n",
      "label: U+9B73\n",
      "label: U+9B85\n",
      "label: U+9B87\n",
      "label: U+9B8E\n",
      "label: U+9B91\n",
      "label: U+9B92\n",
      "label: U+9B93\n",
      "label: U+9B97\n",
      "label: U+9B9F\n",
      "label: U+9BA0\n",
      "label: U+9BA5\n",
      "label: U+9BA6\n",
      "label: U+9BA7\n",
      "label: U+9BA8\n",
      "label: U+9BAA\n",
      "label: U+9BAB\n",
      "label: U+9BAC\n",
      "label: U+9BAD\n",
      "label: U+9BAE\n",
      "label: U+9BB2\n",
      "label: U+9BC9\n",
      "label: U+9BCA\n",
      "label: U+9BCE\n",
      "label: U+9BD1\n",
      "label: U+9BD2\n",
      "label: U+9BD4\n",
      "label: U+9BD6\n",
      "label: U+9BDB\n",
      "label: U+9BE8\n",
      "label: U+9BEF\n",
      "label: U+9BF0\n",
      "label: U+9BF2\n",
      "label: U+9BF5\n",
      "label: U+9BF7\n",
      "label: U+9C06\n",
      "label: U+9C08\n",
      "label: U+9C0A\n",
      "label: U+9C0C\n",
      "label: U+9C10\n",
      "label: U+9C12\n",
      "label: U+9C14\n",
      "label: U+9C15\n",
      "label: U+9C23\n",
      "label: U+9C24\n",
      "label: U+9C25\n",
      "label: U+9C29\n",
      "label: U+9C2D\n",
      "label: U+9C2F\n",
      "label: U+9C36\n",
      "label: U+9C39\n",
      "label: U+9C3A\n",
      "label: U+9C3B\n",
      "label: U+9C3E\n",
      "label: U+9C41\n",
      "label: U+9C45\n",
      "label: U+9C46\n",
      "label: U+9C47\n",
      "label: U+9C48\n",
      "label: U+9C4A\n",
      "label: U+9C4F\n",
      "label: U+9C52\n",
      "label: U+9C57\n",
      "label: U+9C5A\n",
      "label: U+9C60\n",
      "label: U+9C63\n",
      "label: U+9C67\n",
      "label: U+9C6A\n",
      "label: U+9C6E\n",
      "label: U+9C72\n",
      "label: U+9C76\n",
      "label: U+9C78\n",
      "label: U+9C7A\n",
      "label: U+9CE5\n",
      "label: U+9CE9\n",
      "label: U+9CEB\n",
      "label: U+9CF0\n",
      "label: U+9CF3\n",
      "label: U+9CF4\n",
      "label: U+9CF6\n",
      "label: U+9D08\n",
      "label: U+9D12\n",
      "label: U+9D1E\n",
      "label: U+9D28\n",
      "label: U+9D2B\n",
      "label: U+9D2C\n",
      "label: U+9D3D\n",
      "label: U+9D50\n",
      "label: U+9D51\n",
      "label: U+9D59\n",
      "label: U+9D5C\n",
      "label: U+9D5D\n",
      "label: U+9D5E\n",
      "label: U+9D72\n",
      "label: U+9D7A\n",
      "label: U+9D86\n",
      "label: U+9D89\n",
      "label: U+9D8F\n",
      "label: U+9DAB\n",
      "label: U+9DAF\n",
      "label: U+9DB4\n",
      "label: U+9DBA\n",
      "label: U+9DC2\n",
      "label: U+9DC3\n",
      "label: U+9DC4\n",
      "label: U+9DC9\n",
      "label: U+9DD7\n",
      "label: U+9DE6\n",
      "label: U+9DEF\n",
      "label: U+9DF2\n",
      "label: U+9DF9\n",
      "label: U+9DFA\n",
      "label: U+9E0A\n",
      "label: U+9E1E\n",
      "label: U+9E75\n",
      "label: U+9E7D\n",
      "label: U+9E7F\n",
      "label: U+9E81\n",
      "label: U+9E88\n",
      "label: U+9E8B\n",
      "label: U+9E8C\n",
      "label: U+9E93\n",
      "label: U+9E95\n",
      "label: U+9E97\n",
      "label: U+9E9D\n",
      "label: U+9E9F\n",
      "label: U+9EA5\n",
      "label: U+9EA6\n",
      "label: U+9EA9\n",
      "label: U+9EAA\n",
      "label: U+9EB4\n",
      "label: U+9EB5\n",
      "label: U+9EB8\n",
      "label: U+9EB9\n",
      "label: U+9EBA\n",
      "label: U+9EBB\n",
      "label: U+9EBF\n",
      "label: U+9EC4\n",
      "label: U+9ECD\n",
      "label: U+9ECE\n",
      "label: U+9ED0\n",
      "label: U+9ED2\n",
      "label: U+9ED8\n",
      "label: U+9EDB\n",
      "label: U+9EDE\n",
      "label: U+9EE0\n",
      "label: U+9EE8\n",
      "label: U+9EFD\n",
      "label: U+9F07\n",
      "label: U+9F08\n",
      "label: U+9F0E\n",
      "label: U+9F13\n",
      "label: U+9F20\n",
      "label: U+9F2B\n",
      "label: U+9F2C\n",
      "label: U+9F34\n",
      "label: U+9F3A\n",
      "label: U+9F3B\n",
      "label: U+9F3E\n",
      "label: U+9F4A\n",
      "label: U+9F4B\n",
      "label: U+9F52\n",
      "label: U+9F5F\n",
      "label: U+9F61\n",
      "label: U+9F62\n",
      "label: U+9F6C\n",
      "label: U+9F8D\n",
      "label: U+9F9C\n",
      "label: U+9F9D\n",
      "label: U+FA11\n",
      "label: U+FA1D\n",
      "label: U+FA38\n",
      "label: U+FA45\n",
      "label: U+FA4A\n",
      "label: U+FA55\n",
      "label: U+FA5C\n",
      "Images and labels saved\n"
     ]
    }
   ],
   "source": [
    "# Run this cell to get all images and labels of the Kanji dataset\n",
    "data_dir = \"..\\\\archive\\\\kkanji\\\\kkanji2\"\n",
    "\n",
    "array_of_images = []\n",
    "dict_of_image_data = {}\n",
    "for dirpath, dirnames, filenames in os.walk(data_dir):\n",
    "    label = os.path.basename(dirpath)\n",
    "    print(\"label:\", label)\n",
    "    for file in filenames:\n",
    "        filepath = os.path.join(dirpath, file)\n",
    "        img = np.array(Image.open(filepath))\n",
    "        array_of_images.append(img)  \n",
    "        if label in dict_of_image_data:\n",
    "            dict_of_image_data[label].append(filepath)\n",
    "        else:\n",
    "            dict_of_image_data[label] = [filepath]\n",
    "\n",
    "np.savez(\"all_images.npz\", array_of_images)\n",
    "\n",
    "array_of_label_numbers = []\n",
    "array_of_labels = []\n",
    "number = -1\n",
    "for dirpath, dirnames, filenames in os.walk(data_dir):\n",
    "    label = os.path.basename(dirpath)\n",
    "    if label != \"kkanji2\":\n",
    "        array_of_labels.append(label)\n",
    "    for file in filenames:\n",
    "        array_of_label_numbers.append(number)\n",
    "    number = number + 1\n",
    "\n",
    "np.savez(\"all_labels.npz\", array_of_labels)\n",
    "np.savez(\"all_label_numbers.npz\", array_of_label_numbers)\n",
    "\n",
    "print(\"Images and labels saved\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classes: 3832\n",
      "Number of samples: 140424 where each sample is of size: (64, 64)\n"
     ]
    }
   ],
   "source": [
    "kanji_images_path = \"all_images.npz\"\n",
    "kanji_labels_path = \"all_labels.npz\"\n",
    "kanji_label_numbers_path = \"all_label_numbers.npz\"\n",
    "\n",
    "# Load data from npz files\n",
    "images = np.load(kanji_images_path)['arr_0']\n",
    "unicode_labels = np.load(kanji_labels_path)['arr_0']\n",
    "labels = np.load(kanji_label_numbers_path)['arr_0']\n",
    "\n",
    "num_classes = len(np.unique(labels))\n",
    "\n",
    "print(\"Classes:\", num_classes)\n",
    "print(f\"Number of samples: {len(images)} where each sample is of size: {images.shape[1:]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 64 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n = 8\n",
    "fig, axs = plt.subplots(nrows=n, ncols=n, figsize=(8, 8))\n",
    "fig.suptitle('Kuzushiji-Kanji Images with Labels', fontsize=20)\n",
    "\n",
    "for i in range(n**2):\n",
    "    ax = axs[i // n, i % n]\n",
    "    ax.imshow(images[i], cmap='gray')\n",
    "    ax.set_title(labels[i])\n",
    "    ax.axis('off')\n",
    "\n",
    "plt.subplots_adjust(hspace=0.5)\n",
    "plt.savefig('images_kanji.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the data distribution between all classes\n",
    "label_numbers = pd.Series(labels)\n",
    "label_values = label_numbers.value_counts(sort=False).sort_index()\n",
    "fig, ax = plt.subplots(figsize=(18,6))\n",
    "fig.suptitle('Class Distribution of the Kuzushiji-Kanji Dataset', fontsize=25)\n",
    "ax.bar(label_values.index, label_values.values)\n",
    "plt.yticks(fontsize=14)\n",
    "plt.margins(x=0.001)\n",
    "plt.tick_params(\n",
    "    axis='x',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,      # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False) # labels along the bottom edge are off\n",
    "ax.set_xticks(label_values.index)\n",
    "ax.set_xlabel('Classes', fontsize=18, labelpad=14)\n",
    "ax.set_ylabel('Number of Images', fontsize=18, labelpad=14)\n",
    "fig.tight_layout()\n",
    "plt.subplots_adjust(top=0.85)\n",
    "plt.savefig('class_distribution_kanji.png', bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preset of Arguments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adapt these variables if you saved datasets to another location or want change something\n",
    "balancing = True\n",
    "batch_size = 128\n",
    "csv_file = \"\"\n",
    "data_dir = \"..\\\\archive\\\\kkanji\\\\kkanji2\"\n",
    "data_dir_cmnist = \"\"\n",
    "epochs = 50\n",
    "label = \"kanji\"\n",
    "learning_rate = 1e-4\n",
    "load_from_npz = False\n",
    "npz_paths = []\n",
    "save_to_npz = True\n",
    "resize_shape = (64,64)\n",
    "upsampling_size = 25\n",
    "use_grayscale = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load data from directory...\n",
      "140424 files found\n",
      "Balance data...\n",
      "Load images and labels from data...\n",
      "Number of folders for one array: 958\n",
      "Load from folder 0...\n",
      "Load from folder 1...\n",
      "Load from folder 2...\n",
      "Load from folder 3...\n",
      "Load from folder 4...\n",
      "Load from folder 5...\n",
      "Load from folder 6...\n",
      "Load from folder 7...\n",
      "Load from folder 8...\n",
      "Load from folder 9...\n",
      "Load from folder 10...\n",
      "Load from folder 11...\n",
      "Load from folder 12...\n",
      "Load from folder 13...\n",
      "Load from folder 14...\n",
      "Load from folder 15...\n",
      "Load from folder 16...\n",
      "Load from folder 17...\n",
      "Load from folder 18...\n",
      "Load from folder 19...\n",
      "Load from folder 20...\n",
      "Load from folder 21...\n",
      "Load from folder 22...\n",
      "Load from folder 23...\n",
      "Load from folder 24...\n",
      "Load from folder 25...\n",
      "Load from folder 26...\n",
      "Load from folder 27...\n",
      "Load from folder 28...\n",
      "Load from folder 29...\n",
      "Load from folder 30...\n",
      "Load from folder 31...\n",
      "Load from folder 32...\n",
      "Load from folder 33...\n",
      "Load from folder 34...\n",
      "Load from folder 35...\n",
      "Load from folder 36...\n",
      "Load from folder 37...\n",
      "Load from folder 38...\n",
      "Load from folder 39...\n",
      "Load from folder 40...\n",
      "Load from folder 41...\n",
      "Load from folder 42...\n",
      "Load from folder 43...\n",
      "Load from folder 44...\n",
      "Load from folder 45...\n",
      "Load from folder 46...\n",
      "Load from folder 47...\n",
      "Load from folder 48...\n",
      "Load from folder 49...\n",
      "Load from folder 50...\n",
      "Load from folder 51...\n",
      "Load from folder 52...\n",
      "Load from folder 53...\n",
      "Load from folder 54...\n",
      "Load from folder 55...\n",
      "Load from folder 56...\n",
      "Load from folder 57...\n",
      "Load from folder 58...\n",
      "Load from folder 59...\n",
      "Load from folder 60...\n",
      "Load from folder 61...\n",
      "Load from folder 62...\n",
      "Load from folder 63...\n",
      "Load from folder 64...\n",
      "Load from folder 65...\n",
      "Load from folder 66...\n",
      "Load from folder 67...\n",
      "Load from folder 68...\n",
      "Load from folder 69...\n",
      "Load from folder 70...\n",
      "Load from folder 71...\n",
      "Load from folder 72...\n",
      "Load from folder 73...\n",
      "Load from folder 74...\n",
      "Load from folder 75...\n",
      "Load from folder 76...\n",
      "Load from folder 77...\n",
      "Load from folder 78...\n",
      "Load from folder 79...\n",
      "Load from folder 80...\n",
      "Load from folder 81...\n",
      "Load from folder 82...\n",
      "Load from folder 83...\n",
      "Load from folder 84...\n",
      "Load from folder 85...\n",
      "Load from folder 86...\n",
      "Load from folder 87...\n",
      "Load from folder 88...\n",
      "Load from folder 89...\n",
      "Load from folder 90...\n",
      "Load from folder 91...\n",
      "Load from folder 92...\n",
      "Load from folder 93...\n",
      "Load from folder 94...\n",
      "Load from folder 95...\n",
      "Load from folder 96...\n",
      "Load from folder 97...\n",
      "Load from folder 98...\n",
      "Load from folder 99...\n",
      "Load from folder 100...\n",
      "Load from folder 101...\n",
      "Load from folder 102...\n",
      "Load from folder 103...\n",
      "Load from folder 104...\n",
      "Load from folder 105...\n",
      "Load from folder 106...\n",
      "Load from folder 107...\n",
      "Load from folder 108...\n",
      "Load from folder 109...\n",
      "Load from folder 110...\n",
      "Load from folder 111...\n",
      "Load from folder 112...\n",
      "Load from folder 113...\n",
      "Load from folder 114...\n",
      "Load from folder 115...\n",
      "Load from folder 116...\n",
      "Load from folder 117...\n",
      "Load from folder 118...\n",
      "Load from folder 119...\n",
      "Load from folder 120...\n",
      "Load from folder 121...\n",
      "Load from folder 122...\n",
      "Load from folder 123...\n",
      "Load from folder 124...\n",
      "Load from folder 125...\n",
      "Load from folder 126...\n",
      "Load from folder 127...\n",
      "Load from folder 128...\n",
      "Load from folder 129...\n",
      "Load from folder 130...\n",
      "Load from folder 131...\n",
      "Load from folder 132...\n",
      "Load from folder 133...\n",
      "Load from folder 134...\n",
      "Load from folder 135...\n",
      "Load from folder 136...\n",
      "Load from folder 137...\n",
      "Load from folder 138...\n",
      "Load from folder 139...\n",
      "Load from folder 140...\n",
      "Load from folder 141...\n",
      "Load from folder 142...\n",
      "Load from folder 143...\n",
      "Load from folder 144...\n",
      "Load from folder 145...\n",
      "Load from folder 146...\n",
      "Load from folder 147...\n",
      "Load from folder 148...\n",
      "Load from folder 149...\n",
      "Load from folder 150...\n",
      "Load from folder 151...\n",
      "Load from folder 152...\n",
      "Load from folder 153...\n",
      "Load from folder 154...\n",
      "Load from folder 155...\n",
      "Load from folder 156...\n",
      "Load from folder 157...\n",
      "Load from folder 158...\n",
      "Load from folder 159...\n",
      "Load from folder 160...\n",
      "Load from folder 161...\n",
      "Load from folder 162...\n",
      "Load from folder 163...\n",
      "Load from folder 164...\n",
      "Load from folder 165...\n",
      "Load from folder 166...\n",
      "Load from folder 167...\n",
      "Load from folder 168...\n",
      "Load from folder 169...\n",
      "Load from folder 170...\n",
      "Load from folder 171...\n",
      "Load from folder 172...\n",
      "Load from folder 173...\n",
      "Load from folder 174...\n",
      "Load from folder 175...\n",
      "Load from folder 176...\n",
      "Load from folder 177...\n",
      "Load from folder 178...\n",
      "Load from folder 179...\n",
      "Load from folder 180...\n",
      "Load from folder 181...\n",
      "Load from folder 182...\n",
      "Load from folder 183...\n",
      "Load from folder 184...\n",
      "Load from folder 185...\n",
      "Load from folder 186...\n",
      "Load from folder 187...\n",
      "Load from folder 188...\n",
      "Load from folder 189...\n",
      "Load from folder 190...\n",
      "Load from folder 191...\n",
      "Load from folder 192...\n",
      "Load from folder 193...\n",
      "Load from folder 194...\n",
      "Load from folder 195...\n",
      "Load from folder 196...\n",
      "Load from folder 197...\n",
      "Load from folder 198...\n",
      "Load from folder 199...\n",
      "Load from folder 200...\n",
      "Load from folder 201...\n",
      "Load from folder 202...\n",
      "Load from folder 203...\n",
      "Load from folder 204...\n",
      "Load from folder 205...\n",
      "Load from folder 206...\n",
      "Load from folder 207...\n",
      "Load from folder 208...\n",
      "Load from folder 209...\n",
      "Load from folder 210...\n",
      "Load from folder 211...\n",
      "Load from folder 212...\n",
      "Load from folder 213...\n",
      "Load from folder 214...\n",
      "Load from folder 215...\n",
      "Load from folder 216...\n",
      "Load from folder 217...\n",
      "Load from folder 218...\n",
      "Load from folder 219...\n",
      "Load from folder 220...\n",
      "Load from folder 221...\n",
      "Load from folder 222...\n",
      "Load from folder 223...\n",
      "Load from folder 224...\n",
      "Load from folder 225...\n",
      "Load from folder 226...\n",
      "Load from folder 227...\n",
      "Load from folder 228...\n",
      "Load from folder 229...\n",
      "Load from folder 230...\n",
      "Load from folder 231...\n",
      "Load from folder 232...\n",
      "Load from folder 233...\n",
      "Load from folder 234...\n",
      "Load from folder 235...\n",
      "Load from folder 236...\n",
      "Load from folder 237...\n",
      "Load from folder 238...\n",
      "Load from folder 239...\n",
      "Load from folder 240...\n",
      "Load from folder 241...\n",
      "Load from folder 242...\n",
      "Load from folder 243...\n",
      "Load from folder 244...\n",
      "Load from folder 245...\n",
      "Load from folder 246...\n",
      "Load from folder 247...\n",
      "Load from folder 248...\n",
      "Load from folder 249...\n",
      "Load from folder 250...\n",
      "Load from folder 251...\n",
      "Load from folder 252...\n",
      "Load from folder 253...\n",
      "Load from folder 254...\n",
      "Load from folder 255...\n",
      "Load from folder 256...\n",
      "Load from folder 257...\n",
      "Load from folder 258...\n",
      "Load from folder 259...\n",
      "Load from folder 260...\n",
      "Load from folder 261...\n",
      "Load from folder 262...\n",
      "Load from folder 263...\n",
      "Load from folder 264...\n",
      "Load from folder 265...\n",
      "Load from folder 266...\n",
      "Load from folder 267...\n",
      "Load from folder 268...\n",
      "Load from folder 269...\n",
      "Load from folder 270...\n",
      "Load from folder 271...\n",
      "Load from folder 272...\n",
      "Load from folder 273...\n",
      "Load from folder 274...\n",
      "Load from folder 275...\n",
      "Load from folder 276...\n",
      "Load from folder 277...\n",
      "Load from folder 278...\n",
      "Load from folder 279...\n",
      "Load from folder 280...\n",
      "Load from folder 281...\n",
      "Load from folder 282...\n",
      "Load from folder 283...\n",
      "Load from folder 284...\n",
      "Load from folder 285...\n",
      "Load from folder 286...\n",
      "Load from folder 287...\n",
      "Load from folder 288...\n",
      "Load from folder 289...\n",
      "Load from folder 290...\n",
      "Load from folder 291...\n",
      "Load from folder 292...\n",
      "Load from folder 293...\n",
      "Load from folder 294...\n",
      "Load from folder 295...\n",
      "Load from folder 296...\n",
      "Load from folder 297...\n",
      "Load from folder 298...\n",
      "Load from folder 299...\n",
      "Load from folder 300...\n",
      "Load from folder 301...\n",
      "Load from folder 302...\n",
      "Load from folder 303...\n",
      "Load from folder 304...\n",
      "Load from folder 305...\n",
      "Load from folder 306...\n",
      "Load from folder 307...\n",
      "Load from folder 308...\n",
      "Load from folder 309...\n",
      "Load from folder 310...\n",
      "Load from folder 311...\n",
      "Load from folder 312...\n",
      "Load from folder 313...\n",
      "Load from folder 314...\n",
      "Load from folder 315...\n",
      "Load from folder 316...\n",
      "Load from folder 317...\n",
      "Load from folder 318...\n",
      "Load from folder 319...\n",
      "Load from folder 320...\n",
      "Load from folder 321...\n",
      "Load from folder 322...\n",
      "Load from folder 323...\n",
      "Load from folder 324...\n",
      "Load from folder 325...\n",
      "Load from folder 326...\n",
      "Load from folder 327...\n",
      "Load from folder 328...\n",
      "Load from folder 329...\n",
      "Load from folder 330...\n",
      "Load from folder 331...\n",
      "Load from folder 332...\n",
      "Load from folder 333...\n",
      "Load from folder 334...\n",
      "Load from folder 335...\n",
      "Load from folder 336...\n",
      "Load from folder 337...\n",
      "Load from folder 338...\n",
      "Load from folder 339...\n",
      "Load from folder 340...\n",
      "Load from folder 341...\n",
      "Load from folder 342...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 343...\n",
      "Load from folder 344...\n",
      "Load from folder 345...\n",
      "Load from folder 346...\n",
      "Load from folder 347...\n",
      "Load from folder 348...\n",
      "Load from folder 349...\n",
      "Load from folder 350...\n",
      "Load from folder 351...\n",
      "Load from folder 352...\n",
      "Load from folder 353...\n",
      "Load from folder 354...\n",
      "Load from folder 355...\n",
      "Load from folder 356...\n",
      "Load from folder 357...\n",
      "Load from folder 358...\n",
      "Load from folder 359...\n",
      "Load from folder 360...\n",
      "Load from folder 361...\n",
      "Load from folder 362...\n",
      "Load from folder 363...\n",
      "Load from folder 364...\n",
      "Load from folder 365...\n",
      "Load from folder 366...\n",
      "Load from folder 367...\n",
      "Load from folder 368...\n",
      "Load from folder 369...\n",
      "Load from folder 370...\n",
      "Load from folder 371...\n",
      "Load from folder 372...\n",
      "Load from folder 373...\n",
      "Load from folder 374...\n",
      "Load from folder 375...\n",
      "Load from folder 376...\n",
      "Load from folder 377...\n",
      "Load from folder 378...\n",
      "Load from folder 379...\n",
      "Load from folder 380...\n",
      "Load from folder 381...\n",
      "Load from folder 382...\n",
      "Load from folder 383...\n",
      "Load from folder 384...\n",
      "Load from folder 385...\n",
      "Load from folder 386...\n",
      "Load from folder 387...\n",
      "Load from folder 388...\n",
      "Load from folder 389...\n",
      "Load from folder 390...\n",
      "Load from folder 391...\n",
      "Load from folder 392...\n",
      "Load from folder 393...\n",
      "Load from folder 394...\n",
      "Load from folder 395...\n",
      "Load from folder 396...\n",
      "Load from folder 397...\n",
      "Load from folder 398...\n",
      "Load from folder 399...\n",
      "Load from folder 400...\n",
      "Load from folder 401...\n",
      "Load from folder 402...\n",
      "Load from folder 403...\n",
      "Load from folder 404...\n",
      "Load from folder 405...\n",
      "Load from folder 406...\n",
      "Load from folder 407...\n",
      "Load from folder 408...\n",
      "Load from folder 409...\n",
      "Load from folder 410...\n",
      "Load from folder 411...\n",
      "Load from folder 412...\n",
      "Load from folder 413...\n",
      "Load from folder 414...\n",
      "Load from folder 415...\n",
      "Load from folder 416...\n",
      "Load from folder 417...\n",
      "Load from folder 418...\n",
      "Load from folder 419...\n",
      "Load from folder 420...\n",
      "Load from folder 421...\n",
      "Load from folder 422...\n",
      "Load from folder 423...\n",
      "Load from folder 424...\n",
      "Load from folder 425...\n",
      "Load from folder 426...\n",
      "Load from folder 427...\n",
      "Load from folder 428...\n",
      "Load from folder 429...\n",
      "Load from folder 430...\n",
      "Load from folder 431...\n",
      "Load from folder 432...\n",
      "Load from folder 433...\n",
      "Load from folder 434...\n",
      "Load from folder 435...\n",
      "Load from folder 436...\n",
      "Load from folder 437...\n",
      "Load from folder 438...\n",
      "Load from folder 439...\n",
      "Load from folder 440...\n",
      "Load from folder 441...\n",
      "Load from folder 442...\n",
      "Load from folder 443...\n",
      "Load from folder 444...\n",
      "Load from folder 445...\n",
      "Load from folder 446...\n",
      "Load from folder 447...\n",
      "Load from folder 448...\n",
      "Load from folder 449...\n",
      "Load from folder 450...\n",
      "Load from folder 451...\n",
      "Load from folder 452...\n",
      "Load from folder 453...\n",
      "Load from folder 454...\n",
      "Load from folder 455...\n",
      "Load from folder 456...\n",
      "Load from folder 457...\n",
      "Load from folder 458...\n",
      "Load from folder 459...\n",
      "Load from folder 460...\n",
      "Load from folder 461...\n",
      "Load from folder 462...\n",
      "Load from folder 463...\n",
      "Load from folder 464...\n",
      "Load from folder 465...\n",
      "Load from folder 466...\n",
      "Load from folder 467...\n",
      "Load from folder 468...\n",
      "Load from folder 469...\n",
      "Load from folder 470...\n",
      "Load from folder 471...\n",
      "Load from folder 472...\n",
      "Load from folder 473...\n",
      "Load from folder 474...\n",
      "Load from folder 475...\n",
      "Load from folder 476...\n",
      "Load from folder 477...\n",
      "Load from folder 478...\n",
      "Load from folder 479...\n",
      "Load from folder 480...\n",
      "Load from folder 481...\n",
      "Load from folder 482...\n",
      "Load from folder 483...\n",
      "Load from folder 484...\n",
      "Load from folder 485...\n",
      "Load from folder 486...\n",
      "Load from folder 487...\n",
      "Load from folder 488...\n",
      "Load from folder 489...\n",
      "Load from folder 490...\n",
      "Load from folder 491...\n",
      "Load from folder 492...\n",
      "Load from folder 493...\n",
      "Load from folder 494...\n",
      "Load from folder 495...\n",
      "Load from folder 496...\n",
      "Load from folder 497...\n",
      "Load from folder 498...\n",
      "Load from folder 499...\n",
      "Load from folder 500...\n",
      "Load from folder 501...\n",
      "Load from folder 502...\n",
      "Load from folder 503...\n",
      "Load from folder 504...\n",
      "Load from folder 505...\n",
      "Load from folder 506...\n",
      "Load from folder 507...\n",
      "Load from folder 508...\n",
      "Load from folder 509...\n",
      "Load from folder 510...\n",
      "Load from folder 511...\n",
      "Load from folder 512...\n",
      "Load from folder 513...\n",
      "Load from folder 514...\n",
      "Load from folder 515...\n",
      "Load from folder 516...\n",
      "Load from folder 517...\n",
      "Load from folder 518...\n",
      "Load from folder 519...\n",
      "Load from folder 520...\n",
      "Load from folder 521...\n",
      "Load from folder 522...\n",
      "Load from folder 523...\n",
      "Load from folder 524...\n",
      "Load from folder 525...\n",
      "Load from folder 526...\n",
      "Load from folder 527...\n",
      "Load from folder 528...\n",
      "Load from folder 529...\n",
      "Load from folder 530...\n",
      "Load from folder 531...\n",
      "Load from folder 532...\n",
      "Load from folder 533...\n",
      "Load from folder 534...\n",
      "Load from folder 535...\n",
      "Load from folder 536...\n",
      "Load from folder 537...\n",
      "Load from folder 538...\n",
      "Load from folder 539...\n",
      "Load from folder 540...\n",
      "Load from folder 541...\n",
      "Load from folder 542...\n",
      "Load from folder 543...\n",
      "Load from folder 544...\n",
      "Load from folder 545...\n",
      "Load from folder 546...\n",
      "Load from folder 547...\n",
      "Load from folder 548...\n",
      "Load from folder 549...\n",
      "Load from folder 550...\n",
      "Load from folder 551...\n",
      "Load from folder 552...\n",
      "Load from folder 553...\n",
      "Load from folder 554...\n",
      "Load from folder 555...\n",
      "Load from folder 556...\n",
      "Load from folder 557...\n",
      "Load from folder 558...\n",
      "Load from folder 559...\n",
      "Load from folder 560...\n",
      "Load from folder 561...\n",
      "Load from folder 562...\n",
      "Load from folder 563...\n",
      "Load from folder 564...\n",
      "Load from folder 565...\n",
      "Load from folder 566...\n",
      "Load from folder 567...\n",
      "Load from folder 568...\n",
      "Load from folder 569...\n",
      "Load from folder 570...\n",
      "Load from folder 571...\n",
      "Load from folder 572...\n",
      "Load from folder 573...\n",
      "Load from folder 574...\n",
      "Load from folder 575...\n",
      "Load from folder 576...\n",
      "Load from folder 577...\n",
      "Load from folder 578...\n",
      "Load from folder 579...\n",
      "Load from folder 580...\n",
      "Load from folder 581...\n",
      "Load from folder 582...\n",
      "Load from folder 583...\n",
      "Load from folder 584...\n",
      "Load from folder 585...\n",
      "Load from folder 586...\n",
      "Load from folder 587...\n",
      "Load from folder 588...\n",
      "Load from folder 589...\n",
      "Load from folder 590...\n",
      "Load from folder 591...\n",
      "Load from folder 592...\n",
      "Load from folder 593...\n",
      "Load from folder 594...\n",
      "Load from folder 595...\n",
      "Load from folder 596...\n",
      "Load from folder 597...\n",
      "Load from folder 598...\n",
      "Load from folder 599...\n",
      "Load from folder 600...\n",
      "Load from folder 601...\n",
      "Load from folder 602...\n",
      "Load from folder 603...\n",
      "Load from folder 604...\n",
      "Load from folder 605...\n",
      "Load from folder 606...\n",
      "Load from folder 607...\n",
      "Load from folder 608...\n",
      "Load from folder 609...\n",
      "Load from folder 610...\n",
      "Load from folder 611...\n",
      "Load from folder 612...\n",
      "Load from folder 613...\n",
      "Load from folder 614...\n",
      "Load from folder 615...\n",
      "Load from folder 616...\n",
      "Load from folder 617...\n",
      "Load from folder 618...\n",
      "Load from folder 619...\n",
      "Load from folder 620...\n",
      "Load from folder 621...\n",
      "Load from folder 622...\n",
      "Load from folder 623...\n",
      "Load from folder 624...\n",
      "Load from folder 625...\n",
      "Load from folder 626...\n",
      "Load from folder 627...\n",
      "Load from folder 628...\n",
      "Load from folder 629...\n",
      "Load from folder 630...\n",
      "Load from folder 631...\n",
      "Load from folder 632...\n",
      "Load from folder 633...\n",
      "Load from folder 634...\n",
      "Load from folder 635...\n",
      "Load from folder 636...\n",
      "Load from folder 637...\n",
      "Load from folder 638...\n",
      "Load from folder 639...\n",
      "Load from folder 640...\n",
      "Load from folder 641...\n",
      "Load from folder 642...\n",
      "Load from folder 643...\n",
      "Load from folder 644...\n",
      "Load from folder 645...\n",
      "Load from folder 646...\n",
      "Load from folder 647...\n",
      "Load from folder 648...\n",
      "Load from folder 649...\n",
      "Load from folder 650...\n",
      "Load from folder 651...\n",
      "Load from folder 652...\n",
      "Load from folder 653...\n",
      "Load from folder 654...\n",
      "Load from folder 655...\n",
      "Load from folder 656...\n",
      "Load from folder 657...\n",
      "Load from folder 658...\n",
      "Load from folder 659...\n",
      "Load from folder 660...\n",
      "Load from folder 661...\n",
      "Load from folder 662...\n",
      "Load from folder 663...\n",
      "Load from folder 664...\n",
      "Load from folder 665...\n",
      "Load from folder 666...\n",
      "Load from folder 667...\n",
      "Load from folder 668...\n",
      "Load from folder 669...\n",
      "Load from folder 670...\n",
      "Load from folder 671...\n",
      "Load from folder 672...\n",
      "Load from folder 673...\n",
      "Load from folder 674...\n",
      "Load from folder 675...\n",
      "Load from folder 676...\n",
      "Load from folder 677...\n",
      "Load from folder 678...\n",
      "Load from folder 679...\n",
      "Load from folder 680...\n",
      "Load from folder 681...\n",
      "Load from folder 682...\n",
      "Load from folder 683...\n",
      "Load from folder 684...\n",
      "Load from folder 685...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 686...\n",
      "Load from folder 687...\n",
      "Load from folder 688...\n",
      "Load from folder 689...\n",
      "Load from folder 690...\n",
      "Load from folder 691...\n",
      "Load from folder 692...\n",
      "Load from folder 693...\n",
      "Load from folder 694...\n",
      "Load from folder 695...\n",
      "Load from folder 696...\n",
      "Load from folder 697...\n",
      "Load from folder 698...\n",
      "Load from folder 699...\n",
      "Load from folder 700...\n",
      "Load from folder 701...\n",
      "Load from folder 702...\n",
      "Load from folder 703...\n",
      "Load from folder 704...\n",
      "Load from folder 705...\n",
      "Load from folder 706...\n",
      "Load from folder 707...\n",
      "Load from folder 708...\n",
      "Load from folder 709...\n",
      "Load from folder 710...\n",
      "Load from folder 711...\n",
      "Load from folder 712...\n",
      "Load from folder 713...\n",
      "Load from folder 714...\n",
      "Load from folder 715...\n",
      "Load from folder 716...\n",
      "Load from folder 717...\n",
      "Load from folder 718...\n",
      "Load from folder 719...\n",
      "Load from folder 720...\n",
      "Load from folder 721...\n",
      "Load from folder 722...\n",
      "Load from folder 723...\n",
      "Load from folder 724...\n",
      "Load from folder 725...\n",
      "Load from folder 726...\n",
      "Load from folder 727...\n",
      "Load from folder 728...\n",
      "Load from folder 729...\n",
      "Load from folder 730...\n",
      "Load from folder 731...\n",
      "Load from folder 732...\n",
      "Load from folder 733...\n",
      "Load from folder 734...\n",
      "Load from folder 735...\n",
      "Load from folder 736...\n",
      "Load from folder 737...\n",
      "Load from folder 738...\n",
      "Load from folder 739...\n",
      "Load from folder 740...\n",
      "Load from folder 741...\n",
      "Load from folder 742...\n",
      "Load from folder 743...\n",
      "Load from folder 744...\n",
      "Load from folder 745...\n",
      "Load from folder 746...\n",
      "Load from folder 747...\n",
      "Load from folder 748...\n",
      "Load from folder 749...\n",
      "Load from folder 750...\n",
      "Load from folder 751...\n",
      "Load from folder 752...\n",
      "Load from folder 753...\n",
      "Load from folder 754...\n",
      "Load from folder 755...\n",
      "Load from folder 756...\n",
      "Load from folder 757...\n",
      "Load from folder 758...\n",
      "Load from folder 759...\n",
      "Load from folder 760...\n",
      "Load from folder 761...\n",
      "Load from folder 762...\n",
      "Load from folder 763...\n",
      "Load from folder 764...\n",
      "Load from folder 765...\n",
      "Load from folder 766...\n",
      "Load from folder 767...\n",
      "Load from folder 768...\n",
      "Load from folder 769...\n",
      "Load from folder 770...\n",
      "Load from folder 771...\n",
      "Load from folder 772...\n",
      "Load from folder 773...\n",
      "Load from folder 774...\n",
      "Load from folder 775...\n",
      "Load from folder 776...\n",
      "Load from folder 777...\n",
      "Load from folder 778...\n",
      "Load from folder 779...\n",
      "Load from folder 780...\n",
      "Load from folder 781...\n",
      "Load from folder 782...\n",
      "Load from folder 783...\n",
      "Load from folder 784...\n",
      "Load from folder 785...\n",
      "Load from folder 786...\n",
      "Load from folder 787...\n",
      "Load from folder 788...\n",
      "Load from folder 789...\n",
      "Load from folder 790...\n",
      "Load from folder 791...\n",
      "Load from folder 792...\n",
      "Load from folder 793...\n",
      "Load from folder 794...\n",
      "Load from folder 795...\n",
      "Load from folder 796...\n",
      "Load from folder 797...\n",
      "Load from folder 798...\n",
      "Load from folder 799...\n",
      "Load from folder 800...\n",
      "Load from folder 801...\n",
      "Load from folder 802...\n",
      "Load from folder 803...\n",
      "Load from folder 804...\n",
      "Load from folder 805...\n",
      "Load from folder 806...\n",
      "Load from folder 807...\n",
      "Load from folder 808...\n",
      "Load from folder 809...\n",
      "Load from folder 810...\n",
      "Load from folder 811...\n",
      "Load from folder 812...\n",
      "Load from folder 813...\n",
      "Load from folder 814...\n",
      "Load from folder 815...\n",
      "Load from folder 816...\n",
      "Load from folder 817...\n",
      "Load from folder 818...\n",
      "Load from folder 819...\n",
      "Load from folder 820...\n",
      "Load from folder 821...\n",
      "Load from folder 822...\n",
      "Load from folder 823...\n",
      "Load from folder 824...\n",
      "Load from folder 825...\n",
      "Load from folder 826...\n",
      "Load from folder 827...\n",
      "Load from folder 828...\n",
      "Load from folder 829...\n",
      "Load from folder 830...\n",
      "Load from folder 831...\n",
      "Load from folder 832...\n",
      "Load from folder 833...\n",
      "Load from folder 834...\n",
      "Load from folder 835...\n",
      "Load from folder 836...\n",
      "Load from folder 837...\n",
      "Load from folder 838...\n",
      "Load from folder 839...\n",
      "Load from folder 840...\n",
      "Load from folder 841...\n",
      "Load from folder 842...\n",
      "Load from folder 843...\n",
      "Load from folder 844...\n",
      "Load from folder 845...\n",
      "Load from folder 846...\n",
      "Load from folder 847...\n",
      "Load from folder 848...\n",
      "Load from folder 849...\n",
      "Load from folder 850...\n",
      "Load from folder 851...\n",
      "Load from folder 852...\n",
      "Load from folder 853...\n",
      "Load from folder 854...\n",
      "Load from folder 855...\n",
      "Load from folder 856...\n",
      "Load from folder 857...\n",
      "Load from folder 858...\n",
      "Load from folder 859...\n",
      "Load from folder 860...\n",
      "Load from folder 861...\n",
      "Load from folder 862...\n",
      "Load from folder 863...\n",
      "Load from folder 864...\n",
      "Load from folder 865...\n",
      "Load from folder 866...\n",
      "Load from folder 867...\n",
      "Load from folder 868...\n",
      "Load from folder 869...\n",
      "Load from folder 870...\n",
      "Load from folder 871...\n",
      "Load from folder 872...\n",
      "Load from folder 873...\n",
      "Load from folder 874...\n",
      "Load from folder 875...\n",
      "Load from folder 876...\n",
      "Load from folder 877...\n",
      "Load from folder 878...\n",
      "Load from folder 879...\n",
      "Load from folder 880...\n",
      "Load from folder 881...\n",
      "Load from folder 882...\n",
      "Load from folder 883...\n",
      "Load from folder 884...\n",
      "Load from folder 885...\n",
      "Load from folder 886...\n",
      "Load from folder 887...\n",
      "Load from folder 888...\n",
      "Load from folder 889...\n",
      "Load from folder 890...\n",
      "Load from folder 891...\n",
      "Load from folder 892...\n",
      "Load from folder 893...\n",
      "Load from folder 894...\n",
      "Load from folder 895...\n",
      "Load from folder 896...\n",
      "Load from folder 897...\n",
      "Load from folder 898...\n",
      "Load from folder 899...\n",
      "Load from folder 900...\n",
      "Load from folder 901...\n",
      "Load from folder 902...\n",
      "Load from folder 903...\n",
      "Load from folder 904...\n",
      "Load from folder 905...\n",
      "Load from folder 906...\n",
      "Load from folder 907...\n",
      "Load from folder 908...\n",
      "Load from folder 909...\n",
      "Load from folder 910...\n",
      "Load from folder 911...\n",
      "Load from folder 912...\n",
      "Load from folder 913...\n",
      "Load from folder 914...\n",
      "Load from folder 915...\n",
      "Load from folder 916...\n",
      "Load from folder 917...\n",
      "Load from folder 918...\n",
      "Load from folder 919...\n",
      "Load from folder 920...\n",
      "Load from folder 921...\n",
      "Load from folder 922...\n",
      "Load from folder 923...\n",
      "Load from folder 924...\n",
      "Load from folder 925...\n",
      "Load from folder 926...\n",
      "Load from folder 927...\n",
      "Load from folder 928...\n",
      "Load from folder 929...\n",
      "Load from folder 930...\n",
      "Load from folder 931...\n",
      "Load from folder 932...\n",
      "Load from folder 933...\n",
      "Load from folder 934...\n",
      "Load from folder 935...\n",
      "Load from folder 936...\n",
      "Load from folder 937...\n",
      "Load from folder 938...\n",
      "Load from folder 939...\n",
      "Load from folder 940...\n",
      "Load from folder 941...\n",
      "Load from folder 942...\n",
      "Load from folder 943...\n",
      "Load from folder 944...\n",
      "Load from folder 945...\n",
      "Load from folder 946...\n",
      "Load from folder 947...\n",
      "Load from folder 948...\n",
      "Load from folder 949...\n",
      "Load from folder 950...\n",
      "Load from folder 951...\n",
      "Load from folder 952...\n",
      "Load from folder 953...\n",
      "Load from folder 954...\n",
      "Load from folder 955...\n",
      "Load from folder 956...\n",
      "Load from folder 957...\n",
      "Load from folder 958...\n",
      "Load from folder 959...\n",
      "Load from folder 960...\n",
      "Load from folder 961...\n",
      "Load from folder 962...\n",
      "Load from folder 963...\n",
      "Load from folder 964...\n",
      "Load from folder 965...\n",
      "Load from folder 966...\n",
      "Load from folder 967...\n",
      "Load from folder 968...\n",
      "Load from folder 969...\n",
      "Load from folder 970...\n",
      "Load from folder 971...\n",
      "Load from folder 972...\n",
      "Load from folder 973...\n",
      "Load from folder 974...\n",
      "Load from folder 975...\n",
      "Load from folder 976...\n",
      "Load from folder 977...\n",
      "Load from folder 978...\n",
      "Load from folder 979...\n",
      "Load from folder 980...\n",
      "Load from folder 981...\n",
      "Load from folder 982...\n",
      "Load from folder 983...\n",
      "Load from folder 984...\n",
      "Load from folder 985...\n",
      "Load from folder 986...\n",
      "Load from folder 987...\n",
      "Load from folder 988...\n",
      "Load from folder 989...\n",
      "Load from folder 990...\n",
      "Load from folder 991...\n",
      "Load from folder 992...\n",
      "Load from folder 993...\n",
      "Load from folder 994...\n",
      "Load from folder 995...\n",
      "Load from folder 996...\n",
      "Load from folder 997...\n",
      "Load from folder 998...\n",
      "Load from folder 999...\n",
      "Load from folder 1000...\n",
      "Load from folder 1001...\n",
      "Load from folder 1002...\n",
      "Load from folder 1003...\n",
      "Load from folder 1004...\n",
      "Load from folder 1005...\n",
      "Load from folder 1006...\n",
      "Load from folder 1007...\n",
      "Load from folder 1008...\n",
      "Load from folder 1009...\n",
      "Load from folder 1010...\n",
      "Load from folder 1011...\n",
      "Load from folder 1012...\n",
      "Load from folder 1013...\n",
      "Load from folder 1014...\n",
      "Load from folder 1015...\n",
      "Load from folder 1016...\n",
      "Load from folder 1017...\n",
      "Load from folder 1018...\n",
      "Load from folder 1019...\n",
      "Load from folder 1020...\n",
      "Load from folder 1021...\n",
      "Load from folder 1022...\n",
      "Load from folder 1023...\n",
      "Load from folder 1024...\n",
      "Load from folder 1025...\n",
      "Load from folder 1026...\n",
      "Load from folder 1027...\n",
      "Load from folder 1028...\n",
      "Load from folder 1029...\n",
      "Load from folder 1030...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 1031...\n",
      "Load from folder 1032...\n",
      "Load from folder 1033...\n",
      "Load from folder 1034...\n",
      "Load from folder 1035...\n",
      "Load from folder 1036...\n",
      "Load from folder 1037...\n",
      "Load from folder 1038...\n",
      "Load from folder 1039...\n",
      "Load from folder 1040...\n",
      "Load from folder 1041...\n",
      "Load from folder 1042...\n",
      "Load from folder 1043...\n",
      "Load from folder 1044...\n",
      "Load from folder 1045...\n",
      "Load from folder 1046...\n",
      "Load from folder 1047...\n",
      "Load from folder 1048...\n",
      "Load from folder 1049...\n",
      "Load from folder 1050...\n",
      "Load from folder 1051...\n",
      "Load from folder 1052...\n",
      "Load from folder 1053...\n",
      "Load from folder 1054...\n",
      "Load from folder 1055...\n",
      "Load from folder 1056...\n",
      "Load from folder 1057...\n",
      "Load from folder 1058...\n",
      "Load from folder 1059...\n",
      "Load from folder 1060...\n",
      "Load from folder 1061...\n",
      "Load from folder 1062...\n",
      "Load from folder 1063...\n",
      "Load from folder 1064...\n",
      "Load from folder 1065...\n",
      "Load from folder 1066...\n",
      "Load from folder 1067...\n",
      "Load from folder 1068...\n",
      "Load from folder 1069...\n",
      "Load from folder 1070...\n",
      "Load from folder 1071...\n",
      "Load from folder 1072...\n",
      "Load from folder 1073...\n",
      "Load from folder 1074...\n",
      "Load from folder 1075...\n",
      "Load from folder 1076...\n",
      "Load from folder 1077...\n",
      "Load from folder 1078...\n",
      "Load from folder 1079...\n",
      "Load from folder 1080...\n",
      "Load from folder 1081...\n",
      "Load from folder 1082...\n",
      "Load from folder 1083...\n",
      "Load from folder 1084...\n",
      "Load from folder 1085...\n",
      "Load from folder 1086...\n",
      "Load from folder 1087...\n",
      "Load from folder 1088...\n",
      "Load from folder 1089...\n",
      "Load from folder 1090...\n",
      "Load from folder 1091...\n",
      "Load from folder 1092...\n",
      "Load from folder 1093...\n",
      "Load from folder 1094...\n",
      "Load from folder 1095...\n",
      "Load from folder 1096...\n",
      "Load from folder 1097...\n",
      "Load from folder 1098...\n",
      "Load from folder 1099...\n",
      "Load from folder 1100...\n",
      "Load from folder 1101...\n",
      "Load from folder 1102...\n",
      "Load from folder 1103...\n",
      "Load from folder 1104...\n",
      "Load from folder 1105...\n",
      "Load from folder 1106...\n",
      "Load from folder 1107...\n",
      "Load from folder 1108...\n",
      "Load from folder 1109...\n",
      "Load from folder 1110...\n",
      "Load from folder 1111...\n",
      "Load from folder 1112...\n",
      "Load from folder 1113...\n",
      "Load from folder 1114...\n",
      "Load from folder 1115...\n",
      "Load from folder 1116...\n",
      "Load from folder 1117...\n",
      "Load from folder 1118...\n",
      "Load from folder 1119...\n",
      "Load from folder 1120...\n",
      "Load from folder 1121...\n",
      "Load from folder 1122...\n",
      "Load from folder 1123...\n",
      "Load from folder 1124...\n",
      "Load from folder 1125...\n",
      "Load from folder 1126...\n",
      "Load from folder 1127...\n",
      "Load from folder 1128...\n",
      "Load from folder 1129...\n",
      "Load from folder 1130...\n",
      "Load from folder 1131...\n",
      "Load from folder 1132...\n",
      "Load from folder 1133...\n",
      "Load from folder 1134...\n",
      "Load from folder 1135...\n",
      "Load from folder 1136...\n",
      "Load from folder 1137...\n",
      "Load from folder 1138...\n",
      "Load from folder 1139...\n",
      "Load from folder 1140...\n",
      "Load from folder 1141...\n",
      "Load from folder 1142...\n",
      "Load from folder 1143...\n",
      "Load from folder 1144...\n",
      "Load from folder 1145...\n",
      "Load from folder 1146...\n",
      "Load from folder 1147...\n",
      "Load from folder 1148...\n",
      "Load from folder 1149...\n",
      "Load from folder 1150...\n",
      "Load from folder 1151...\n",
      "Load from folder 1152...\n",
      "Load from folder 1153...\n",
      "Load from folder 1154...\n",
      "Load from folder 1155...\n",
      "Load from folder 1156...\n",
      "Load from folder 1157...\n",
      "Load from folder 1158...\n",
      "Load from folder 1159...\n",
      "Load from folder 1160...\n",
      "Load from folder 1161...\n",
      "Load from folder 1162...\n",
      "Load from folder 1163...\n",
      "Load from folder 1164...\n",
      "Load from folder 1165...\n",
      "Load from folder 1166...\n",
      "Load from folder 1167...\n",
      "Load from folder 1168...\n",
      "Load from folder 1169...\n",
      "Load from folder 1170...\n",
      "Load from folder 1171...\n",
      "Load from folder 1172...\n",
      "Load from folder 1173...\n",
      "Load from folder 1174...\n",
      "Load from folder 1175...\n",
      "Load from folder 1176...\n",
      "Load from folder 1177...\n",
      "Load from folder 1178...\n",
      "Load from folder 1179...\n",
      "Load from folder 1180...\n",
      "Load from folder 1181...\n",
      "Load from folder 1182...\n",
      "Load from folder 1183...\n",
      "Load from folder 1184...\n",
      "Load from folder 1185...\n",
      "Load from folder 1186...\n",
      "Load from folder 1187...\n",
      "Load from folder 1188...\n",
      "Load from folder 1189...\n",
      "Load from folder 1190...\n",
      "Load from folder 1191...\n",
      "Load from folder 1192...\n",
      "Load from folder 1193...\n",
      "Load from folder 1194...\n",
      "Load from folder 1195...\n",
      "Load from folder 1196...\n",
      "Load from folder 1197...\n",
      "Load from folder 1198...\n",
      "Load from folder 1199...\n",
      "Load from folder 1200...\n",
      "Load from folder 1201...\n",
      "Load from folder 1202...\n",
      "Load from folder 1203...\n",
      "Load from folder 1204...\n",
      "Load from folder 1205...\n",
      "Load from folder 1206...\n",
      "Load from folder 1207...\n",
      "Load from folder 1208...\n",
      "Load from folder 1209...\n",
      "Load from folder 1210...\n",
      "Load from folder 1211...\n",
      "Load from folder 1212...\n",
      "Load from folder 1213...\n",
      "Load from folder 1214...\n",
      "Load from folder 1215...\n",
      "Load from folder 1216...\n",
      "Load from folder 1217...\n",
      "Load from folder 1218...\n",
      "Load from folder 1219...\n",
      "Load from folder 1220...\n",
      "Load from folder 1221...\n",
      "Load from folder 1222...\n",
      "Load from folder 1223...\n",
      "Load from folder 1224...\n",
      "Load from folder 1225...\n",
      "Load from folder 1226...\n",
      "Load from folder 1227...\n",
      "Load from folder 1228...\n",
      "Load from folder 1229...\n",
      "Load from folder 1230...\n",
      "Load from folder 1231...\n",
      "Load from folder 1232...\n",
      "Load from folder 1233...\n",
      "Load from folder 1234...\n",
      "Load from folder 1235...\n",
      "Load from folder 1236...\n",
      "Load from folder 1237...\n",
      "Load from folder 1238...\n",
      "Load from folder 1239...\n",
      "Load from folder 1240...\n",
      "Load from folder 1241...\n",
      "Load from folder 1242...\n",
      "Load from folder 1243...\n",
      "Load from folder 1244...\n",
      "Load from folder 1245...\n",
      "Load from folder 1246...\n",
      "Load from folder 1247...\n",
      "Load from folder 1248...\n",
      "Load from folder 1249...\n",
      "Load from folder 1250...\n",
      "Load from folder 1251...\n",
      "Load from folder 1252...\n",
      "Load from folder 1253...\n",
      "Load from folder 1254...\n",
      "Load from folder 1255...\n",
      "Load from folder 1256...\n",
      "Load from folder 1257...\n",
      "Load from folder 1258...\n",
      "Load from folder 1259...\n",
      "Load from folder 1260...\n",
      "Load from folder 1261...\n",
      "Load from folder 1262...\n",
      "Load from folder 1263...\n",
      "Load from folder 1264...\n",
      "Load from folder 1265...\n",
      "Load from folder 1266...\n",
      "Load from folder 1267...\n",
      "Load from folder 1268...\n",
      "Load from folder 1269...\n",
      "Load from folder 1270...\n",
      "Load from folder 1271...\n",
      "Load from folder 1272...\n",
      "Load from folder 1273...\n",
      "Load from folder 1274...\n",
      "Load from folder 1275...\n",
      "Load from folder 1276...\n",
      "Load from folder 1277...\n",
      "Load from folder 1278...\n",
      "Load from folder 1279...\n",
      "Load from folder 1280...\n",
      "Load from folder 1281...\n",
      "Load from folder 1282...\n",
      "Load from folder 1283...\n",
      "Load from folder 1284...\n",
      "Load from folder 1285...\n",
      "Load from folder 1286...\n",
      "Load from folder 1287...\n",
      "Load from folder 1288...\n",
      "Load from folder 1289...\n",
      "Load from folder 1290...\n",
      "Load from folder 1291...\n",
      "Load from folder 1292...\n",
      "Load from folder 1293...\n",
      "Load from folder 1294...\n",
      "Load from folder 1295...\n",
      "Load from folder 1296...\n",
      "Load from folder 1297...\n",
      "Load from folder 1298...\n",
      "Load from folder 1299...\n",
      "Load from folder 1300...\n",
      "Load from folder 1301...\n",
      "Load from folder 1302...\n",
      "Load from folder 1303...\n",
      "Load from folder 1304...\n",
      "Load from folder 1305...\n",
      "Load from folder 1306...\n",
      "Load from folder 1307...\n",
      "Load from folder 1308...\n",
      "Load from folder 1309...\n",
      "Load from folder 1310...\n",
      "Load from folder 1311...\n",
      "Load from folder 1312...\n",
      "Load from folder 1313...\n",
      "Load from folder 1314...\n",
      "Load from folder 1315...\n",
      "Load from folder 1316...\n",
      "Load from folder 1317...\n",
      "Load from folder 1318...\n",
      "Load from folder 1319...\n",
      "Load from folder 1320...\n",
      "Load from folder 1321...\n",
      "Load from folder 1322...\n",
      "Load from folder 1323...\n",
      "Load from folder 1324...\n",
      "Load from folder 1325...\n",
      "Load from folder 1326...\n",
      "Load from folder 1327...\n",
      "Load from folder 1328...\n",
      "Load from folder 1329...\n",
      "Load from folder 1330...\n",
      "Load from folder 1331...\n",
      "Load from folder 1332...\n",
      "Load from folder 1333...\n",
      "Load from folder 1334...\n",
      "Load from folder 1335...\n",
      "Load from folder 1336...\n",
      "Load from folder 1337...\n",
      "Load from folder 1338...\n",
      "Load from folder 1339...\n",
      "Load from folder 1340...\n",
      "Load from folder 1341...\n",
      "Load from folder 1342...\n",
      "Load from folder 1343...\n",
      "Load from folder 1344...\n",
      "Load from folder 1345...\n",
      "Load from folder 1346...\n",
      "Load from folder 1347...\n",
      "Load from folder 1348...\n",
      "Load from folder 1349...\n",
      "Load from folder 1350...\n",
      "Load from folder 1351...\n",
      "Load from folder 1352...\n",
      "Load from folder 1353...\n",
      "Load from folder 1354...\n",
      "Load from folder 1355...\n",
      "Load from folder 1356...\n",
      "Load from folder 1357...\n",
      "Load from folder 1358...\n",
      "Load from folder 1359...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 1360...\n",
      "Load from folder 1361...\n",
      "Load from folder 1362...\n",
      "Load from folder 1363...\n",
      "Load from folder 1364...\n",
      "Load from folder 1365...\n",
      "Load from folder 1366...\n",
      "Load from folder 1367...\n",
      "Load from folder 1368...\n",
      "Load from folder 1369...\n",
      "Load from folder 1370...\n",
      "Load from folder 1371...\n",
      "Load from folder 1372...\n",
      "Load from folder 1373...\n",
      "Load from folder 1374...\n",
      "Load from folder 1375...\n",
      "Load from folder 1376...\n",
      "Load from folder 1377...\n",
      "Load from folder 1378...\n",
      "Load from folder 1379...\n",
      "Load from folder 1380...\n",
      "Load from folder 1381...\n",
      "Load from folder 1382...\n",
      "Load from folder 1383...\n",
      "Load from folder 1384...\n",
      "Load from folder 1385...\n",
      "Load from folder 1386...\n",
      "Load from folder 1387...\n",
      "Load from folder 1388...\n",
      "Load from folder 1389...\n",
      "Load from folder 1390...\n",
      "Load from folder 1391...\n",
      "Load from folder 1392...\n",
      "Load from folder 1393...\n",
      "Load from folder 1394...\n",
      "Load from folder 1395...\n",
      "Load from folder 1396...\n",
      "Load from folder 1397...\n",
      "Load from folder 1398...\n",
      "Load from folder 1399...\n",
      "Load from folder 1400...\n",
      "Load from folder 1401...\n",
      "Load from folder 1402...\n",
      "Load from folder 1403...\n",
      "Load from folder 1404...\n",
      "Load from folder 1405...\n",
      "Load from folder 1406...\n",
      "Load from folder 1407...\n",
      "Load from folder 1408...\n",
      "Load from folder 1409...\n",
      "Load from folder 1410...\n",
      "Load from folder 1411...\n",
      "Load from folder 1412...\n",
      "Load from folder 1413...\n",
      "Load from folder 1414...\n",
      "Load from folder 1415...\n",
      "Load from folder 1416...\n",
      "Load from folder 1417...\n",
      "Load from folder 1418...\n",
      "Load from folder 1419...\n",
      "Load from folder 1420...\n",
      "Load from folder 1421...\n",
      "Load from folder 1422...\n",
      "Load from folder 1423...\n",
      "Load from folder 1424...\n",
      "Load from folder 1425...\n",
      "Load from folder 1426...\n",
      "Load from folder 1427...\n",
      "Load from folder 1428...\n",
      "Load from folder 1429...\n",
      "Load from folder 1430...\n",
      "Load from folder 1431...\n",
      "Load from folder 1432...\n",
      "Load from folder 1433...\n",
      "Load from folder 1434...\n",
      "Load from folder 1435...\n",
      "Load from folder 1436...\n",
      "Load from folder 1437...\n",
      "Load from folder 1438...\n",
      "Load from folder 1439...\n",
      "Load from folder 1440...\n",
      "Load from folder 1441...\n",
      "Load from folder 1442...\n",
      "Load from folder 1443...\n",
      "Load from folder 1444...\n",
      "Load from folder 1445...\n",
      "Load from folder 1446...\n",
      "Load from folder 1447...\n",
      "Load from folder 1448...\n",
      "Load from folder 1449...\n",
      "Load from folder 1450...\n",
      "Load from folder 1451...\n",
      "Load from folder 1452...\n",
      "Load from folder 1453...\n",
      "Load from folder 1454...\n",
      "Load from folder 1455...\n",
      "Load from folder 1456...\n",
      "Load from folder 1457...\n",
      "Load from folder 1458...\n",
      "Load from folder 1459...\n",
      "Load from folder 1460...\n",
      "Load from folder 1461...\n",
      "Load from folder 1462...\n",
      "Load from folder 1463...\n",
      "Load from folder 1464...\n",
      "Load from folder 1465...\n",
      "Load from folder 1466...\n",
      "Load from folder 1467...\n",
      "Load from folder 1468...\n",
      "Load from folder 1469...\n",
      "Load from folder 1470...\n",
      "Load from folder 1471...\n",
      "Load from folder 1472...\n",
      "Load from folder 1473...\n",
      "Load from folder 1474...\n",
      "Load from folder 1475...\n",
      "Load from folder 1476...\n",
      "Load from folder 1477...\n",
      "Load from folder 1478...\n",
      "Load from folder 1479...\n",
      "Load from folder 1480...\n",
      "Load from folder 1481...\n",
      "Load from folder 1482...\n",
      "Load from folder 1483...\n",
      "Load from folder 1484...\n",
      "Load from folder 1485...\n",
      "Load from folder 1486...\n",
      "Load from folder 1487...\n",
      "Load from folder 1488...\n",
      "Load from folder 1489...\n",
      "Load from folder 1490...\n",
      "Load from folder 1491...\n",
      "Load from folder 1492...\n",
      "Load from folder 1493...\n",
      "Load from folder 1494...\n",
      "Load from folder 1495...\n",
      "Load from folder 1496...\n",
      "Load from folder 1497...\n",
      "Load from folder 1498...\n",
      "Load from folder 1499...\n",
      "Load from folder 1500...\n",
      "Load from folder 1501...\n",
      "Load from folder 1502...\n",
      "Load from folder 1503...\n",
      "Load from folder 1504...\n",
      "Load from folder 1505...\n",
      "Load from folder 1506...\n",
      "Load from folder 1507...\n",
      "Load from folder 1508...\n",
      "Load from folder 1509...\n",
      "Load from folder 1510...\n",
      "Load from folder 1511...\n",
      "Load from folder 1512...\n",
      "Load from folder 1513...\n",
      "Load from folder 1514...\n",
      "Load from folder 1515...\n",
      "Load from folder 1516...\n",
      "Load from folder 1517...\n",
      "Load from folder 1518...\n",
      "Load from folder 1519...\n",
      "Load from folder 1520...\n",
      "Load from folder 1521...\n",
      "Load from folder 1522...\n",
      "Load from folder 1523...\n",
      "Load from folder 1524...\n",
      "Load from folder 1525...\n",
      "Load from folder 1526...\n",
      "Load from folder 1527...\n",
      "Load from folder 1528...\n",
      "Load from folder 1529...\n",
      "Load from folder 1530...\n",
      "Load from folder 1531...\n",
      "Load from folder 1532...\n",
      "Load from folder 1533...\n",
      "Load from folder 1534...\n",
      "Load from folder 1535...\n",
      "Load from folder 1536...\n",
      "Load from folder 1537...\n",
      "Load from folder 1538...\n",
      "Load from folder 1539...\n",
      "Load from folder 1540...\n",
      "Load from folder 1541...\n",
      "Load from folder 1542...\n",
      "Load from folder 1543...\n",
      "Load from folder 1544...\n",
      "Load from folder 1545...\n",
      "Load from folder 1546...\n",
      "Load from folder 1547...\n",
      "Load from folder 1548...\n",
      "Load from folder 1549...\n",
      "Load from folder 1550...\n",
      "Load from folder 1551...\n",
      "Load from folder 1552...\n",
      "Load from folder 1553...\n",
      "Load from folder 1554...\n",
      "Load from folder 1555...\n",
      "Load from folder 1556...\n",
      "Load from folder 1557...\n",
      "Load from folder 1558...\n",
      "Load from folder 1559...\n",
      "Load from folder 1560...\n",
      "Load from folder 1561...\n",
      "Load from folder 1562...\n",
      "Load from folder 1563...\n",
      "Load from folder 1564...\n",
      "Load from folder 1565...\n",
      "Load from folder 1566...\n",
      "Load from folder 1567...\n",
      "Load from folder 1568...\n",
      "Load from folder 1569...\n",
      "Load from folder 1570...\n",
      "Load from folder 1571...\n",
      "Load from folder 1572...\n",
      "Load from folder 1573...\n",
      "Load from folder 1574...\n",
      "Load from folder 1575...\n",
      "Load from folder 1576...\n",
      "Load from folder 1577...\n",
      "Load from folder 1578...\n",
      "Load from folder 1579...\n",
      "Load from folder 1580...\n",
      "Load from folder 1581...\n",
      "Load from folder 1582...\n",
      "Load from folder 1583...\n",
      "Load from folder 1584...\n",
      "Load from folder 1585...\n",
      "Load from folder 1586...\n",
      "Load from folder 1587...\n",
      "Load from folder 1588...\n",
      "Load from folder 1589...\n",
      "Load from folder 1590...\n",
      "Load from folder 1591...\n",
      "Load from folder 1592...\n",
      "Load from folder 1593...\n",
      "Load from folder 1594...\n",
      "Load from folder 1595...\n",
      "Load from folder 1596...\n",
      "Load from folder 1597...\n",
      "Load from folder 1598...\n",
      "Load from folder 1599...\n",
      "Load from folder 1600...\n",
      "Load from folder 1601...\n",
      "Load from folder 1602...\n",
      "Load from folder 1603...\n",
      "Load from folder 1604...\n",
      "Load from folder 1605...\n",
      "Load from folder 1606...\n",
      "Load from folder 1607...\n",
      "Load from folder 1608...\n",
      "Load from folder 1609...\n",
      "Load from folder 1610...\n",
      "Load from folder 1611...\n",
      "Load from folder 1612...\n",
      "Load from folder 1613...\n",
      "Load from folder 1614...\n",
      "Load from folder 1615...\n",
      "Load from folder 1616...\n",
      "Load from folder 1617...\n",
      "Load from folder 1618...\n",
      "Load from folder 1619...\n",
      "Load from folder 1620...\n",
      "Load from folder 1621...\n",
      "Load from folder 1622...\n",
      "Load from folder 1623...\n",
      "Load from folder 1624...\n",
      "Load from folder 1625...\n",
      "Load from folder 1626...\n",
      "Load from folder 1627...\n",
      "Load from folder 1628...\n",
      "Load from folder 1629...\n",
      "Load from folder 1630...\n",
      "Load from folder 1631...\n",
      "Load from folder 1632...\n",
      "Load from folder 1633...\n",
      "Load from folder 1634...\n",
      "Load from folder 1635...\n",
      "Load from folder 1636...\n",
      "Load from folder 1637...\n",
      "Load from folder 1638...\n",
      "Load from folder 1639...\n",
      "Load from folder 1640...\n",
      "Load from folder 1641...\n",
      "Load from folder 1642...\n",
      "Load from folder 1643...\n",
      "Load from folder 1644...\n",
      "Load from folder 1645...\n",
      "Load from folder 1646...\n",
      "Load from folder 1647...\n",
      "Load from folder 1648...\n",
      "Load from folder 1649...\n",
      "Load from folder 1650...\n",
      "Load from folder 1651...\n",
      "Load from folder 1652...\n",
      "Load from folder 1653...\n",
      "Load from folder 1654...\n",
      "Load from folder 1655...\n",
      "Load from folder 1656...\n",
      "Load from folder 1657...\n",
      "Load from folder 1658...\n",
      "Load from folder 1659...\n",
      "Load from folder 1660...\n",
      "Load from folder 1661...\n",
      "Load from folder 1662...\n",
      "Load from folder 1663...\n",
      "Load from folder 1664...\n",
      "Load from folder 1665...\n",
      "Load from folder 1666...\n",
      "Load from folder 1667...\n",
      "Load from folder 1668...\n",
      "Load from folder 1669...\n",
      "Load from folder 1670...\n",
      "Load from folder 1671...\n",
      "Load from folder 1672...\n",
      "Load from folder 1673...\n",
      "Load from folder 1674...\n",
      "Load from folder 1675...\n",
      "Load from folder 1676...\n",
      "Load from folder 1677...\n",
      "Load from folder 1678...\n",
      "Load from folder 1679...\n",
      "Load from folder 1680...\n",
      "Load from folder 1681...\n",
      "Load from folder 1682...\n",
      "Load from folder 1683...\n",
      "Load from folder 1684...\n",
      "Load from folder 1685...\n",
      "Load from folder 1686...\n",
      "Load from folder 1687...\n",
      "Load from folder 1688...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 1689...\n",
      "Load from folder 1690...\n",
      "Load from folder 1691...\n",
      "Load from folder 1692...\n",
      "Load from folder 1693...\n",
      "Load from folder 1694...\n",
      "Load from folder 1695...\n",
      "Load from folder 1696...\n",
      "Load from folder 1697...\n",
      "Load from folder 1698...\n",
      "Load from folder 1699...\n",
      "Load from folder 1700...\n",
      "Load from folder 1701...\n",
      "Load from folder 1702...\n",
      "Load from folder 1703...\n",
      "Load from folder 1704...\n",
      "Load from folder 1705...\n",
      "Load from folder 1706...\n",
      "Load from folder 1707...\n",
      "Load from folder 1708...\n",
      "Load from folder 1709...\n",
      "Load from folder 1710...\n",
      "Load from folder 1711...\n",
      "Load from folder 1712...\n",
      "Load from folder 1713...\n",
      "Load from folder 1714...\n",
      "Load from folder 1715...\n",
      "Load from folder 1716...\n",
      "Load from folder 1717...\n",
      "Load from folder 1718...\n",
      "Load from folder 1719...\n",
      "Load from folder 1720...\n",
      "Load from folder 1721...\n",
      "Load from folder 1722...\n",
      "Load from folder 1723...\n",
      "Load from folder 1724...\n",
      "Load from folder 1725...\n",
      "Load from folder 1726...\n",
      "Load from folder 1727...\n",
      "Load from folder 1728...\n",
      "Load from folder 1729...\n",
      "Load from folder 1730...\n",
      "Load from folder 1731...\n",
      "Load from folder 1732...\n",
      "Load from folder 1733...\n",
      "Load from folder 1734...\n",
      "Load from folder 1735...\n",
      "Load from folder 1736...\n",
      "Load from folder 1737...\n",
      "Load from folder 1738...\n",
      "Load from folder 1739...\n",
      "Load from folder 1740...\n",
      "Load from folder 1741...\n",
      "Load from folder 1742...\n",
      "Load from folder 1743...\n",
      "Load from folder 1744...\n",
      "Load from folder 1745...\n",
      "Load from folder 1746...\n",
      "Load from folder 1747...\n",
      "Load from folder 1748...\n",
      "Load from folder 1749...\n",
      "Load from folder 1750...\n",
      "Load from folder 1751...\n",
      "Load from folder 1752...\n",
      "Load from folder 1753...\n",
      "Load from folder 1754...\n",
      "Load from folder 1755...\n",
      "Load from folder 1756...\n",
      "Load from folder 1757...\n",
      "Load from folder 1758...\n",
      "Load from folder 1759...\n",
      "Load from folder 1760...\n",
      "Load from folder 1761...\n",
      "Load from folder 1762...\n",
      "Load from folder 1763...\n",
      "Load from folder 1764...\n",
      "Load from folder 1765...\n",
      "Load from folder 1766...\n",
      "Load from folder 1767...\n",
      "Load from folder 1768...\n",
      "Load from folder 1769...\n",
      "Load from folder 1770...\n",
      "Load from folder 1771...\n",
      "Load from folder 1772...\n",
      "Load from folder 1773...\n",
      "Load from folder 1774...\n",
      "Load from folder 1775...\n",
      "Load from folder 1776...\n",
      "Load from folder 1777...\n",
      "Load from folder 1778...\n",
      "Load from folder 1779...\n",
      "Load from folder 1780...\n",
      "Load from folder 1781...\n",
      "Load from folder 1782...\n",
      "Load from folder 1783...\n",
      "Load from folder 1784...\n",
      "Load from folder 1785...\n",
      "Load from folder 1786...\n",
      "Load from folder 1787...\n",
      "Load from folder 1788...\n",
      "Load from folder 1789...\n",
      "Load from folder 1790...\n",
      "Load from folder 1791...\n",
      "Load from folder 1792...\n",
      "Load from folder 1793...\n",
      "Load from folder 1794...\n",
      "Load from folder 1795...\n",
      "Load from folder 1796...\n",
      "Load from folder 1797...\n",
      "Load from folder 1798...\n",
      "Load from folder 1799...\n",
      "Load from folder 1800...\n",
      "Load from folder 1801...\n",
      "Load from folder 1802...\n",
      "Load from folder 1803...\n",
      "Load from folder 1804...\n",
      "Load from folder 1805...\n",
      "Load from folder 1806...\n",
      "Load from folder 1807...\n",
      "Load from folder 1808...\n",
      "Load from folder 1809...\n",
      "Load from folder 1810...\n",
      "Load from folder 1811...\n",
      "Load from folder 1812...\n",
      "Load from folder 1813...\n",
      "Load from folder 1814...\n",
      "Load from folder 1815...\n",
      "Load from folder 1816...\n",
      "Load from folder 1817...\n",
      "Load from folder 1818...\n",
      "Load from folder 1819...\n",
      "Load from folder 1820...\n",
      "Load from folder 1821...\n",
      "Load from folder 1822...\n",
      "Load from folder 1823...\n",
      "Load from folder 1824...\n",
      "Load from folder 1825...\n",
      "Load from folder 1826...\n",
      "Load from folder 1827...\n",
      "Load from folder 1828...\n",
      "Load from folder 1829...\n",
      "Load from folder 1830...\n",
      "Load from folder 1831...\n",
      "Load from folder 1832...\n",
      "Load from folder 1833...\n",
      "Load from folder 1834...\n",
      "Load from folder 1835...\n",
      "Load from folder 1836...\n",
      "Load from folder 1837...\n",
      "Load from folder 1838...\n",
      "Load from folder 1839...\n",
      "Load from folder 1840...\n",
      "Load from folder 1841...\n",
      "Load from folder 1842...\n",
      "Load from folder 1843...\n",
      "Load from folder 1844...\n",
      "Load from folder 1845...\n",
      "Load from folder 1846...\n",
      "Load from folder 1847...\n",
      "Load from folder 1848...\n",
      "Load from folder 1849...\n",
      "Load from folder 1850...\n",
      "Load from folder 1851...\n",
      "Load from folder 1852...\n",
      "Load from folder 1853...\n",
      "Load from folder 1854...\n",
      "Load from folder 1855...\n",
      "Load from folder 1856...\n",
      "Load from folder 1857...\n",
      "Load from folder 1858...\n",
      "Load from folder 1859...\n",
      "Load from folder 1860...\n",
      "Load from folder 1861...\n",
      "Load from folder 1862...\n",
      "Load from folder 1863...\n",
      "Load from folder 1864...\n",
      "Load from folder 1865...\n",
      "Load from folder 1866...\n",
      "Load from folder 1867...\n",
      "Load from folder 1868...\n",
      "Load from folder 1869...\n",
      "Load from folder 1870...\n",
      "Load from folder 1871...\n",
      "Load from folder 1872...\n",
      "Load from folder 1873...\n",
      "Load from folder 1874...\n",
      "Load from folder 1875...\n",
      "Load from folder 1876...\n",
      "Load from folder 1877...\n",
      "Load from folder 1878...\n",
      "Load from folder 1879...\n",
      "Load from folder 1880...\n",
      "Load from folder 1881...\n",
      "Load from folder 1882...\n",
      "Load from folder 1883...\n",
      "Load from folder 1884...\n",
      "Load from folder 1885...\n",
      "Load from folder 1886...\n",
      "Load from folder 1887...\n",
      "Load from folder 1888...\n",
      "Load from folder 1889...\n",
      "Load from folder 1890...\n",
      "Load from folder 1891...\n",
      "Load from folder 1892...\n",
      "Load from folder 1893...\n",
      "Load from folder 1894...\n",
      "Load from folder 1895...\n",
      "Load from folder 1896...\n",
      "Load from folder 1897...\n",
      "Load from folder 1898...\n",
      "Load from folder 1899...\n",
      "Load from folder 1900...\n",
      "Load from folder 1901...\n",
      "Load from folder 1902...\n",
      "Load from folder 1903...\n",
      "Load from folder 1904...\n",
      "Load from folder 1905...\n",
      "Load from folder 1906...\n",
      "Load from folder 1907...\n",
      "Load from folder 1908...\n",
      "Load from folder 1909...\n",
      "Load from folder 1910...\n",
      "Load from folder 1911...\n",
      "Load from folder 1912...\n",
      "Load from folder 1913...\n",
      "Load from folder 1914...\n",
      "Load from folder 1915...\n",
      "Load from folder 1916...\n",
      "Load from folder 1917...\n",
      "Load from folder 1918...\n",
      "Load from folder 1919...\n",
      "Load from folder 1920...\n",
      "Load from folder 1921...\n",
      "Load from folder 1922...\n",
      "Load from folder 1923...\n",
      "Load from folder 1924...\n",
      "Load from folder 1925...\n",
      "Load from folder 1926...\n",
      "Load from folder 1927...\n",
      "Load from folder 1928...\n",
      "Load from folder 1929...\n",
      "Load from folder 1930...\n",
      "Load from folder 1931...\n",
      "Load from folder 1932...\n",
      "Load from folder 1933...\n",
      "Load from folder 1934...\n",
      "Load from folder 1935...\n",
      "Load from folder 1936...\n",
      "Load from folder 1937...\n",
      "Load from folder 1938...\n",
      "Load from folder 1939...\n",
      "Load from folder 1940...\n",
      "Load from folder 1941...\n",
      "Load from folder 1942...\n",
      "Load from folder 1943...\n",
      "Load from folder 1944...\n",
      "Load from folder 1945...\n",
      "Load from folder 1946...\n",
      "Load from folder 1947...\n",
      "Load from folder 1948...\n",
      "Load from folder 1949...\n",
      "Load from folder 1950...\n",
      "Load from folder 1951...\n",
      "Load from folder 1952...\n",
      "Load from folder 1953...\n",
      "Load from folder 1954...\n",
      "Load from folder 1955...\n",
      "Load from folder 1956...\n",
      "Load from folder 1957...\n",
      "Load from folder 1958...\n",
      "Load from folder 1959...\n",
      "Load from folder 1960...\n",
      "Load from folder 1961...\n",
      "Load from folder 1962...\n",
      "Load from folder 1963...\n",
      "Load from folder 1964...\n",
      "Load from folder 1965...\n",
      "Load from folder 1966...\n",
      "Load from folder 1967...\n",
      "Load from folder 1968...\n",
      "Load from folder 1969...\n",
      "Load from folder 1970...\n",
      "Load from folder 1971...\n",
      "Load from folder 1972...\n",
      "Load from folder 1973...\n",
      "Load from folder 1974...\n",
      "Load from folder 1975...\n",
      "Load from folder 1976...\n",
      "Load from folder 1977...\n",
      "Load from folder 1978...\n",
      "Load from folder 1979...\n",
      "Load from folder 1980...\n",
      "Load from folder 1981...\n",
      "Load from folder 1982...\n",
      "Load from folder 1983...\n",
      "Load from folder 1984...\n",
      "Load from folder 1985...\n",
      "Load from folder 1986...\n",
      "Load from folder 1987...\n",
      "Load from folder 1988...\n",
      "Load from folder 1989...\n",
      "Load from folder 1990...\n",
      "Load from folder 1991...\n",
      "Load from folder 1992...\n",
      "Load from folder 1993...\n",
      "Load from folder 1994...\n",
      "Load from folder 1995...\n",
      "Load from folder 1996...\n",
      "Load from folder 1997...\n",
      "Load from folder 1998...\n",
      "Load from folder 1999...\n",
      "Load from folder 2000...\n",
      "Load from folder 2001...\n",
      "Load from folder 2002...\n",
      "Load from folder 2003...\n",
      "Load from folder 2004...\n",
      "Load from folder 2005...\n",
      "Load from folder 2006...\n",
      "Load from folder 2007...\n",
      "Load from folder 2008...\n",
      "Load from folder 2009...\n",
      "Load from folder 2010...\n",
      "Load from folder 2011...\n",
      "Load from folder 2012...\n",
      "Load from folder 2013...\n",
      "Load from folder 2014...\n",
      "Load from folder 2015...\n",
      "Load from folder 2016...\n",
      "Load from folder 2017...\n",
      "Load from folder 2018...\n",
      "Load from folder 2019...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 2020...\n",
      "Load from folder 2021...\n",
      "Load from folder 2022...\n",
      "Load from folder 2023...\n",
      "Load from folder 2024...\n",
      "Load from folder 2025...\n",
      "Load from folder 2026...\n",
      "Load from folder 2027...\n",
      "Load from folder 2028...\n",
      "Load from folder 2029...\n",
      "Load from folder 2030...\n",
      "Load from folder 2031...\n",
      "Load from folder 2032...\n",
      "Load from folder 2033...\n",
      "Load from folder 2034...\n",
      "Load from folder 2035...\n",
      "Load from folder 2036...\n",
      "Load from folder 2037...\n",
      "Load from folder 2038...\n",
      "Load from folder 2039...\n",
      "Load from folder 2040...\n",
      "Load from folder 2041...\n",
      "Load from folder 2042...\n",
      "Load from folder 2043...\n",
      "Load from folder 2044...\n",
      "Load from folder 2045...\n",
      "Load from folder 2046...\n",
      "Load from folder 2047...\n",
      "Load from folder 2048...\n",
      "Load from folder 2049...\n",
      "Load from folder 2050...\n",
      "Load from folder 2051...\n",
      "Load from folder 2052...\n",
      "Load from folder 2053...\n",
      "Load from folder 2054...\n",
      "Load from folder 2055...\n",
      "Load from folder 2056...\n",
      "Load from folder 2057...\n",
      "Load from folder 2058...\n",
      "Load from folder 2059...\n",
      "Load from folder 2060...\n",
      "Load from folder 2061...\n",
      "Load from folder 2062...\n",
      "Load from folder 2063...\n",
      "Load from folder 2064...\n",
      "Load from folder 2065...\n",
      "Load from folder 2066...\n",
      "Load from folder 2067...\n",
      "Load from folder 2068...\n",
      "Load from folder 2069...\n",
      "Load from folder 2070...\n",
      "Load from folder 2071...\n",
      "Load from folder 2072...\n",
      "Load from folder 2073...\n",
      "Load from folder 2074...\n",
      "Load from folder 2075...\n",
      "Load from folder 2076...\n",
      "Load from folder 2077...\n",
      "Load from folder 2078...\n",
      "Load from folder 2079...\n",
      "Load from folder 2080...\n",
      "Load from folder 2081...\n",
      "Load from folder 2082...\n",
      "Load from folder 2083...\n",
      "Load from folder 2084...\n",
      "Load from folder 2085...\n",
      "Load from folder 2086...\n",
      "Load from folder 2087...\n",
      "Load from folder 2088...\n",
      "Load from folder 2089...\n",
      "Load from folder 2090...\n",
      "Load from folder 2091...\n",
      "Load from folder 2092...\n",
      "Load from folder 2093...\n",
      "Load from folder 2094...\n",
      "Load from folder 2095...\n",
      "Load from folder 2096...\n",
      "Load from folder 2097...\n",
      "Load from folder 2098...\n",
      "Load from folder 2099...\n",
      "Load from folder 2100...\n",
      "Load from folder 2101...\n",
      "Load from folder 2102...\n",
      "Load from folder 2103...\n",
      "Load from folder 2104...\n",
      "Load from folder 2105...\n",
      "Load from folder 2106...\n",
      "Load from folder 2107...\n",
      "Load from folder 2108...\n",
      "Load from folder 2109...\n",
      "Load from folder 2110...\n",
      "Load from folder 2111...\n",
      "Load from folder 2112...\n",
      "Load from folder 2113...\n",
      "Load from folder 2114...\n",
      "Load from folder 2115...\n",
      "Load from folder 2116...\n",
      "Load from folder 2117...\n",
      "Load from folder 2118...\n",
      "Load from folder 2119...\n",
      "Load from folder 2120...\n",
      "Load from folder 2121...\n",
      "Load from folder 2122...\n",
      "Load from folder 2123...\n",
      "Load from folder 2124...\n",
      "Load from folder 2125...\n",
      "Load from folder 2126...\n",
      "Load from folder 2127...\n",
      "Load from folder 2128...\n",
      "Load from folder 2129...\n",
      "Load from folder 2130...\n",
      "Load from folder 2131...\n",
      "Load from folder 2132...\n",
      "Load from folder 2133...\n",
      "Load from folder 2134...\n",
      "Load from folder 2135...\n",
      "Load from folder 2136...\n",
      "Load from folder 2137...\n",
      "Load from folder 2138...\n",
      "Load from folder 2139...\n",
      "Load from folder 2140...\n",
      "Load from folder 2141...\n",
      "Load from folder 2142...\n",
      "Load from folder 2143...\n",
      "Load from folder 2144...\n",
      "Load from folder 2145...\n",
      "Load from folder 2146...\n",
      "Load from folder 2147...\n",
      "Load from folder 2148...\n",
      "Load from folder 2149...\n",
      "Load from folder 2150...\n",
      "Load from folder 2151...\n",
      "Load from folder 2152...\n",
      "Load from folder 2153...\n",
      "Load from folder 2154...\n",
      "Load from folder 2155...\n",
      "Load from folder 2156...\n",
      "Load from folder 2157...\n",
      "Load from folder 2158...\n",
      "Load from folder 2159...\n",
      "Load from folder 2160...\n",
      "Load from folder 2161...\n",
      "Load from folder 2162...\n",
      "Load from folder 2163...\n",
      "Load from folder 2164...\n",
      "Load from folder 2165...\n",
      "Load from folder 2166...\n",
      "Load from folder 2167...\n",
      "Load from folder 2168...\n",
      "Load from folder 2169...\n",
      "Load from folder 2170...\n",
      "Load from folder 2171...\n",
      "Load from folder 2172...\n",
      "Load from folder 2173...\n",
      "Load from folder 2174...\n",
      "Load from folder 2175...\n",
      "Load from folder 2176...\n",
      "Load from folder 2177...\n",
      "Load from folder 2178...\n",
      "Load from folder 2179...\n",
      "Load from folder 2180...\n",
      "Load from folder 2181...\n",
      "Load from folder 2182...\n",
      "Load from folder 2183...\n",
      "Load from folder 2184...\n",
      "Load from folder 2185...\n",
      "Load from folder 2186...\n",
      "Load from folder 2187...\n",
      "Load from folder 2188...\n",
      "Load from folder 2189...\n",
      "Load from folder 2190...\n",
      "Load from folder 2191...\n",
      "Load from folder 2192...\n",
      "Load from folder 2193...\n",
      "Load from folder 2194...\n",
      "Load from folder 2195...\n",
      "Load from folder 2196...\n",
      "Load from folder 2197...\n",
      "Load from folder 2198...\n",
      "Load from folder 2199...\n",
      "Load from folder 2200...\n",
      "Load from folder 2201...\n",
      "Load from folder 2202...\n",
      "Load from folder 2203...\n",
      "Load from folder 2204...\n",
      "Load from folder 2205...\n",
      "Load from folder 2206...\n",
      "Load from folder 2207...\n",
      "Load from folder 2208...\n",
      "Load from folder 2209...\n",
      "Load from folder 2210...\n",
      "Load from folder 2211...\n",
      "Load from folder 2212...\n",
      "Load from folder 2213...\n",
      "Load from folder 2214...\n",
      "Load from folder 2215...\n",
      "Load from folder 2216...\n",
      "Load from folder 2217...\n",
      "Load from folder 2218...\n",
      "Load from folder 2219...\n",
      "Load from folder 2220...\n",
      "Load from folder 2221...\n",
      "Load from folder 2222...\n",
      "Load from folder 2223...\n",
      "Load from folder 2224...\n",
      "Load from folder 2225...\n",
      "Load from folder 2226...\n",
      "Load from folder 2227...\n",
      "Load from folder 2228...\n",
      "Load from folder 2229...\n",
      "Load from folder 2230...\n",
      "Load from folder 2231...\n",
      "Load from folder 2232...\n",
      "Load from folder 2233...\n",
      "Load from folder 2234...\n",
      "Load from folder 2235...\n",
      "Load from folder 2236...\n",
      "Load from folder 2237...\n",
      "Load from folder 2238...\n",
      "Load from folder 2239...\n",
      "Load from folder 2240...\n",
      "Load from folder 2241...\n",
      "Load from folder 2242...\n",
      "Load from folder 2243...\n",
      "Load from folder 2244...\n",
      "Load from folder 2245...\n",
      "Load from folder 2246...\n",
      "Load from folder 2247...\n",
      "Load from folder 2248...\n",
      "Load from folder 2249...\n",
      "Load from folder 2250...\n",
      "Load from folder 2251...\n",
      "Load from folder 2252...\n",
      "Load from folder 2253...\n",
      "Load from folder 2254...\n",
      "Load from folder 2255...\n",
      "Load from folder 2256...\n",
      "Load from folder 2257...\n",
      "Load from folder 2258...\n",
      "Load from folder 2259...\n",
      "Load from folder 2260...\n",
      "Load from folder 2261...\n",
      "Load from folder 2262...\n",
      "Load from folder 2263...\n",
      "Load from folder 2264...\n",
      "Load from folder 2265...\n",
      "Load from folder 2266...\n",
      "Load from folder 2267...\n",
      "Load from folder 2268...\n",
      "Load from folder 2269...\n",
      "Load from folder 2270...\n",
      "Load from folder 2271...\n",
      "Load from folder 2272...\n",
      "Load from folder 2273...\n",
      "Load from folder 2274...\n",
      "Load from folder 2275...\n",
      "Load from folder 2276...\n",
      "Load from folder 2277...\n",
      "Load from folder 2278...\n",
      "Load from folder 2279...\n",
      "Load from folder 2280...\n",
      "Load from folder 2281...\n",
      "Load from folder 2282...\n",
      "Load from folder 2283...\n",
      "Load from folder 2284...\n",
      "Load from folder 2285...\n",
      "Load from folder 2286...\n",
      "Load from folder 2287...\n",
      "Load from folder 2288...\n",
      "Load from folder 2289...\n",
      "Load from folder 2290...\n",
      "Load from folder 2291...\n",
      "Load from folder 2292...\n",
      "Load from folder 2293...\n",
      "Load from folder 2294...\n",
      "Load from folder 2295...\n",
      "Load from folder 2296...\n",
      "Load from folder 2297...\n",
      "Load from folder 2298...\n",
      "Load from folder 2299...\n",
      "Load from folder 2300...\n",
      "Load from folder 2301...\n",
      "Load from folder 2302...\n",
      "Load from folder 2303...\n",
      "Load from folder 2304...\n",
      "Load from folder 2305...\n",
      "Load from folder 2306...\n",
      "Load from folder 2307...\n",
      "Load from folder 2308...\n",
      "Load from folder 2309...\n",
      "Load from folder 2310...\n",
      "Load from folder 2311...\n",
      "Load from folder 2312...\n",
      "Load from folder 2313...\n",
      "Load from folder 2314...\n",
      "Load from folder 2315...\n",
      "Load from folder 2316...\n",
      "Load from folder 2317...\n",
      "Load from folder 2318...\n",
      "Load from folder 2319...\n",
      "Load from folder 2320...\n",
      "Load from folder 2321...\n",
      "Load from folder 2322...\n",
      "Load from folder 2323...\n",
      "Load from folder 2324...\n",
      "Load from folder 2325...\n",
      "Load from folder 2326...\n",
      "Load from folder 2327...\n",
      "Load from folder 2328...\n",
      "Load from folder 2329...\n",
      "Load from folder 2330...\n",
      "Load from folder 2331...\n",
      "Load from folder 2332...\n",
      "Load from folder 2333...\n",
      "Load from folder 2334...\n",
      "Load from folder 2335...\n",
      "Load from folder 2336...\n",
      "Load from folder 2337...\n",
      "Load from folder 2338...\n",
      "Load from folder 2339...\n",
      "Load from folder 2340...\n",
      "Load from folder 2341...\n",
      "Load from folder 2342...\n",
      "Load from folder 2343...\n",
      "Load from folder 2344...\n",
      "Load from folder 2345...\n",
      "Load from folder 2346...\n",
      "Load from folder 2347...\n",
      "Load from folder 2348...\n",
      "Load from folder 2349...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 2350...\n",
      "Load from folder 2351...\n",
      "Load from folder 2352...\n",
      "Load from folder 2353...\n",
      "Load from folder 2354...\n",
      "Load from folder 2355...\n",
      "Load from folder 2356...\n",
      "Load from folder 2357...\n",
      "Load from folder 2358...\n",
      "Load from folder 2359...\n",
      "Load from folder 2360...\n",
      "Load from folder 2361...\n",
      "Load from folder 2362...\n",
      "Load from folder 2363...\n",
      "Load from folder 2364...\n",
      "Load from folder 2365...\n",
      "Load from folder 2366...\n",
      "Load from folder 2367...\n",
      "Load from folder 2368...\n",
      "Load from folder 2369...\n",
      "Load from folder 2370...\n",
      "Load from folder 2371...\n",
      "Load from folder 2372...\n",
      "Load from folder 2373...\n",
      "Load from folder 2374...\n",
      "Load from folder 2375...\n",
      "Load from folder 2376...\n",
      "Load from folder 2377...\n",
      "Load from folder 2378...\n",
      "Load from folder 2379...\n",
      "Load from folder 2380...\n",
      "Load from folder 2381...\n",
      "Load from folder 2382...\n",
      "Load from folder 2383...\n",
      "Load from folder 2384...\n",
      "Load from folder 2385...\n",
      "Load from folder 2386...\n",
      "Load from folder 2387...\n",
      "Load from folder 2388...\n",
      "Load from folder 2389...\n",
      "Load from folder 2390...\n",
      "Load from folder 2391...\n",
      "Load from folder 2392...\n",
      "Load from folder 2393...\n",
      "Load from folder 2394...\n",
      "Load from folder 2395...\n",
      "Load from folder 2396...\n",
      "Load from folder 2397...\n",
      "Load from folder 2398...\n",
      "Load from folder 2399...\n",
      "Load from folder 2400...\n",
      "Load from folder 2401...\n",
      "Load from folder 2402...\n",
      "Load from folder 2403...\n",
      "Load from folder 2404...\n",
      "Load from folder 2405...\n",
      "Load from folder 2406...\n",
      "Load from folder 2407...\n",
      "Load from folder 2408...\n",
      "Load from folder 2409...\n",
      "Load from folder 2410...\n",
      "Load from folder 2411...\n",
      "Load from folder 2412...\n",
      "Load from folder 2413...\n",
      "Load from folder 2414...\n",
      "Load from folder 2415...\n",
      "Load from folder 2416...\n",
      "Load from folder 2417...\n",
      "Load from folder 2418...\n",
      "Load from folder 2419...\n",
      "Load from folder 2420...\n",
      "Load from folder 2421...\n",
      "Load from folder 2422...\n",
      "Load from folder 2423...\n",
      "Load from folder 2424...\n",
      "Load from folder 2425...\n",
      "Load from folder 2426...\n",
      "Load from folder 2427...\n",
      "Load from folder 2428...\n",
      "Load from folder 2429...\n",
      "Load from folder 2430...\n",
      "Load from folder 2431...\n",
      "Load from folder 2432...\n",
      "Load from folder 2433...\n",
      "Load from folder 2434...\n",
      "Load from folder 2435...\n",
      "Load from folder 2436...\n",
      "Load from folder 2437...\n",
      "Load from folder 2438...\n",
      "Load from folder 2439...\n",
      "Load from folder 2440...\n",
      "Load from folder 2441...\n",
      "Load from folder 2442...\n",
      "Load from folder 2443...\n",
      "Load from folder 2444...\n",
      "Load from folder 2445...\n",
      "Load from folder 2446...\n",
      "Load from folder 2447...\n",
      "Load from folder 2448...\n",
      "Load from folder 2449...\n",
      "Load from folder 2450...\n",
      "Load from folder 2451...\n",
      "Load from folder 2452...\n",
      "Load from folder 2453...\n",
      "Load from folder 2454...\n",
      "Load from folder 2455...\n",
      "Load from folder 2456...\n",
      "Load from folder 2457...\n",
      "Load from folder 2458...\n",
      "Load from folder 2459...\n",
      "Load from folder 2460...\n",
      "Load from folder 2461...\n",
      "Load from folder 2462...\n",
      "Load from folder 2463...\n",
      "Load from folder 2464...\n",
      "Load from folder 2465...\n",
      "Load from folder 2466...\n",
      "Load from folder 2467...\n",
      "Load from folder 2468...\n",
      "Load from folder 2469...\n",
      "Load from folder 2470...\n",
      "Load from folder 2471...\n",
      "Load from folder 2472...\n",
      "Load from folder 2473...\n",
      "Load from folder 2474...\n",
      "Load from folder 2475...\n",
      "Load from folder 2476...\n",
      "Load from folder 2477...\n",
      "Load from folder 2478...\n",
      "Load from folder 2479...\n",
      "Load from folder 2480...\n",
      "Load from folder 2481...\n",
      "Load from folder 2482...\n",
      "Load from folder 2483...\n",
      "Load from folder 2484...\n",
      "Load from folder 2485...\n",
      "Load from folder 2486...\n",
      "Load from folder 2487...\n",
      "Load from folder 2488...\n",
      "Load from folder 2489...\n",
      "Load from folder 2490...\n",
      "Load from folder 2491...\n",
      "Load from folder 2492...\n",
      "Load from folder 2493...\n",
      "Load from folder 2494...\n",
      "Load from folder 2495...\n",
      "Load from folder 2496...\n",
      "Load from folder 2497...\n",
      "Load from folder 2498...\n",
      "Load from folder 2499...\n",
      "Load from folder 2500...\n",
      "Load from folder 2501...\n",
      "Load from folder 2502...\n",
      "Load from folder 2503...\n",
      "Load from folder 2504...\n",
      "Load from folder 2505...\n",
      "Load from folder 2506...\n",
      "Load from folder 2507...\n",
      "Load from folder 2508...\n",
      "Load from folder 2509...\n",
      "Load from folder 2510...\n",
      "Load from folder 2511...\n",
      "Load from folder 2512...\n",
      "Load from folder 2513...\n",
      "Load from folder 2514...\n",
      "Load from folder 2515...\n",
      "Load from folder 2516...\n",
      "Load from folder 2517...\n",
      "Load from folder 2518...\n",
      "Load from folder 2519...\n",
      "Load from folder 2520...\n",
      "Load from folder 2521...\n",
      "Load from folder 2522...\n",
      "Load from folder 2523...\n",
      "Load from folder 2524...\n",
      "Load from folder 2525...\n",
      "Load from folder 2526...\n",
      "Load from folder 2527...\n",
      "Load from folder 2528...\n",
      "Load from folder 2529...\n",
      "Load from folder 2530...\n",
      "Load from folder 2531...\n",
      "Load from folder 2532...\n",
      "Load from folder 2533...\n",
      "Load from folder 2534...\n",
      "Load from folder 2535...\n",
      "Load from folder 2536...\n",
      "Load from folder 2537...\n",
      "Load from folder 2538...\n",
      "Load from folder 2539...\n",
      "Load from folder 2540...\n",
      "Load from folder 2541...\n",
      "Load from folder 2542...\n",
      "Load from folder 2543...\n",
      "Load from folder 2544...\n",
      "Load from folder 2545...\n",
      "Load from folder 2546...\n",
      "Load from folder 2547...\n",
      "Load from folder 2548...\n",
      "Load from folder 2549...\n",
      "Load from folder 2550...\n",
      "Load from folder 2551...\n",
      "Load from folder 2552...\n",
      "Load from folder 2553...\n",
      "Load from folder 2554...\n",
      "Load from folder 2555...\n",
      "Load from folder 2556...\n",
      "Load from folder 2557...\n",
      "Load from folder 2558...\n",
      "Load from folder 2559...\n",
      "Load from folder 2560...\n",
      "Load from folder 2561...\n",
      "Load from folder 2562...\n",
      "Load from folder 2563...\n",
      "Load from folder 2564...\n",
      "Load from folder 2565...\n",
      "Load from folder 2566...\n",
      "Load from folder 2567...\n",
      "Load from folder 2568...\n",
      "Load from folder 2569...\n",
      "Load from folder 2570...\n",
      "Load from folder 2571...\n",
      "Load from folder 2572...\n",
      "Load from folder 2573...\n",
      "Load from folder 2574...\n",
      "Load from folder 2575...\n",
      "Load from folder 2576...\n",
      "Load from folder 2577...\n",
      "Load from folder 2578...\n",
      "Load from folder 2579...\n",
      "Load from folder 2580...\n",
      "Load from folder 2581...\n",
      "Load from folder 2582...\n",
      "Load from folder 2583...\n",
      "Load from folder 2584...\n",
      "Load from folder 2585...\n",
      "Load from folder 2586...\n",
      "Load from folder 2587...\n",
      "Load from folder 2588...\n",
      "Load from folder 2589...\n",
      "Load from folder 2590...\n",
      "Load from folder 2591...\n",
      "Load from folder 2592...\n",
      "Load from folder 2593...\n",
      "Load from folder 2594...\n",
      "Load from folder 2595...\n",
      "Load from folder 2596...\n",
      "Load from folder 2597...\n",
      "Load from folder 2598...\n",
      "Load from folder 2599...\n",
      "Load from folder 2600...\n",
      "Load from folder 2601...\n",
      "Load from folder 2602...\n",
      "Load from folder 2603...\n",
      "Load from folder 2604...\n",
      "Load from folder 2605...\n",
      "Load from folder 2606...\n",
      "Load from folder 2607...\n",
      "Load from folder 2608...\n",
      "Load from folder 2609...\n",
      "Load from folder 2610...\n",
      "Load from folder 2611...\n",
      "Load from folder 2612...\n",
      "Load from folder 2613...\n",
      "Load from folder 2614...\n",
      "Load from folder 2615...\n",
      "Load from folder 2616...\n",
      "Load from folder 2617...\n",
      "Load from folder 2618...\n",
      "Load from folder 2619...\n",
      "Load from folder 2620...\n",
      "Load from folder 2621...\n",
      "Load from folder 2622...\n",
      "Load from folder 2623...\n",
      "Load from folder 2624...\n",
      "Load from folder 2625...\n",
      "Load from folder 2626...\n",
      "Load from folder 2627...\n",
      "Load from folder 2628...\n",
      "Load from folder 2629...\n",
      "Load from folder 2630...\n",
      "Load from folder 2631...\n",
      "Load from folder 2632...\n",
      "Load from folder 2633...\n",
      "Load from folder 2634...\n",
      "Load from folder 2635...\n",
      "Load from folder 2636...\n",
      "Load from folder 2637...\n",
      "Load from folder 2638...\n",
      "Load from folder 2639...\n",
      "Load from folder 2640...\n",
      "Load from folder 2641...\n",
      "Load from folder 2642...\n",
      "Load from folder 2643...\n",
      "Load from folder 2644...\n",
      "Load from folder 2645...\n",
      "Load from folder 2646...\n",
      "Load from folder 2647...\n",
      "Load from folder 2648...\n",
      "Load from folder 2649...\n",
      "Load from folder 2650...\n",
      "Load from folder 2651...\n",
      "Load from folder 2652...\n",
      "Load from folder 2653...\n",
      "Load from folder 2654...\n",
      "Load from folder 2655...\n",
      "Load from folder 2656...\n",
      "Load from folder 2657...\n",
      "Load from folder 2658...\n",
      "Load from folder 2659...\n",
      "Load from folder 2660...\n",
      "Load from folder 2661...\n",
      "Load from folder 2662...\n",
      "Load from folder 2663...\n",
      "Load from folder 2664...\n",
      "Load from folder 2665...\n",
      "Load from folder 2666...\n",
      "Load from folder 2667...\n",
      "Load from folder 2668...\n",
      "Load from folder 2669...\n",
      "Load from folder 2670...\n",
      "Load from folder 2671...\n",
      "Load from folder 2672...\n",
      "Load from folder 2673...\n",
      "Load from folder 2674...\n",
      "Load from folder 2675...\n",
      "Load from folder 2676...\n",
      "Load from folder 2677...\n",
      "Load from folder 2678...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 2679...\n",
      "Load from folder 2680...\n",
      "Load from folder 2681...\n",
      "Load from folder 2682...\n",
      "Load from folder 2683...\n",
      "Load from folder 2684...\n",
      "Load from folder 2685...\n",
      "Load from folder 2686...\n",
      "Load from folder 2687...\n",
      "Load from folder 2688...\n",
      "Load from folder 2689...\n",
      "Load from folder 2690...\n",
      "Load from folder 2691...\n",
      "Load from folder 2692...\n",
      "Load from folder 2693...\n",
      "Load from folder 2694...\n",
      "Load from folder 2695...\n",
      "Load from folder 2696...\n",
      "Load from folder 2697...\n",
      "Load from folder 2698...\n",
      "Load from folder 2699...\n",
      "Load from folder 2700...\n",
      "Load from folder 2701...\n",
      "Load from folder 2702...\n",
      "Load from folder 2703...\n",
      "Load from folder 2704...\n",
      "Load from folder 2705...\n",
      "Load from folder 2706...\n",
      "Load from folder 2707...\n",
      "Load from folder 2708...\n",
      "Load from folder 2709...\n",
      "Load from folder 2710...\n",
      "Load from folder 2711...\n",
      "Load from folder 2712...\n",
      "Load from folder 2713...\n",
      "Load from folder 2714...\n",
      "Load from folder 2715...\n",
      "Load from folder 2716...\n",
      "Load from folder 2717...\n",
      "Load from folder 2718...\n",
      "Load from folder 2719...\n",
      "Load from folder 2720...\n",
      "Load from folder 2721...\n",
      "Load from folder 2722...\n",
      "Load from folder 2723...\n",
      "Load from folder 2724...\n",
      "Load from folder 2725...\n",
      "Load from folder 2726...\n",
      "Load from folder 2727...\n",
      "Load from folder 2728...\n",
      "Load from folder 2729...\n",
      "Load from folder 2730...\n",
      "Load from folder 2731...\n",
      "Load from folder 2732...\n",
      "Load from folder 2733...\n",
      "Load from folder 2734...\n",
      "Load from folder 2735...\n",
      "Load from folder 2736...\n",
      "Load from folder 2737...\n",
      "Load from folder 2738...\n",
      "Load from folder 2739...\n",
      "Load from folder 2740...\n",
      "Load from folder 2741...\n",
      "Load from folder 2742...\n",
      "Load from folder 2743...\n",
      "Load from folder 2744...\n",
      "Load from folder 2745...\n",
      "Load from folder 2746...\n",
      "Load from folder 2747...\n",
      "Load from folder 2748...\n",
      "Load from folder 2749...\n",
      "Load from folder 2750...\n",
      "Load from folder 2751...\n",
      "Load from folder 2752...\n",
      "Load from folder 2753...\n",
      "Load from folder 2754...\n",
      "Load from folder 2755...\n",
      "Load from folder 2756...\n",
      "Load from folder 2757...\n",
      "Load from folder 2758...\n",
      "Load from folder 2759...\n",
      "Load from folder 2760...\n",
      "Load from folder 2761...\n",
      "Load from folder 2762...\n",
      "Load from folder 2763...\n",
      "Load from folder 2764...\n",
      "Load from folder 2765...\n",
      "Load from folder 2766...\n",
      "Load from folder 2767...\n",
      "Load from folder 2768...\n",
      "Load from folder 2769...\n",
      "Load from folder 2770...\n",
      "Load from folder 2771...\n",
      "Load from folder 2772...\n",
      "Load from folder 2773...\n",
      "Load from folder 2774...\n",
      "Load from folder 2775...\n",
      "Load from folder 2776...\n",
      "Load from folder 2777...\n",
      "Load from folder 2778...\n",
      "Load from folder 2779...\n",
      "Load from folder 2780...\n",
      "Load from folder 2781...\n",
      "Load from folder 2782...\n",
      "Load from folder 2783...\n",
      "Load from folder 2784...\n",
      "Load from folder 2785...\n",
      "Load from folder 2786...\n",
      "Load from folder 2787...\n",
      "Load from folder 2788...\n",
      "Load from folder 2789...\n",
      "Load from folder 2790...\n",
      "Load from folder 2791...\n",
      "Load from folder 2792...\n",
      "Load from folder 2793...\n",
      "Load from folder 2794...\n",
      "Load from folder 2795...\n",
      "Load from folder 2796...\n",
      "Load from folder 2797...\n",
      "Load from folder 2798...\n",
      "Load from folder 2799...\n",
      "Load from folder 2800...\n",
      "Load from folder 2801...\n",
      "Load from folder 2802...\n",
      "Load from folder 2803...\n",
      "Load from folder 2804...\n",
      "Load from folder 2805...\n",
      "Load from folder 2806...\n",
      "Load from folder 2807...\n",
      "Load from folder 2808...\n",
      "Load from folder 2809...\n",
      "Load from folder 2810...\n",
      "Load from folder 2811...\n",
      "Load from folder 2812...\n",
      "Load from folder 2813...\n",
      "Load from folder 2814...\n",
      "Load from folder 2815...\n",
      "Load from folder 2816...\n",
      "Load from folder 2817...\n",
      "Load from folder 2818...\n",
      "Load from folder 2819...\n",
      "Load from folder 2820...\n",
      "Load from folder 2821...\n",
      "Load from folder 2822...\n",
      "Load from folder 2823...\n",
      "Load from folder 2824...\n",
      "Load from folder 2825...\n",
      "Load from folder 2826...\n",
      "Load from folder 2827...\n",
      "Load from folder 2828...\n",
      "Load from folder 2829...\n",
      "Load from folder 2830...\n",
      "Load from folder 2831...\n",
      "Load from folder 2832...\n",
      "Load from folder 2833...\n",
      "Load from folder 2834...\n",
      "Load from folder 2835...\n",
      "Load from folder 2836...\n",
      "Load from folder 2837...\n",
      "Load from folder 2838...\n",
      "Load from folder 2839...\n",
      "Load from folder 2840...\n",
      "Load from folder 2841...\n",
      "Load from folder 2842...\n",
      "Load from folder 2843...\n",
      "Load from folder 2844...\n",
      "Load from folder 2845...\n",
      "Load from folder 2846...\n",
      "Load from folder 2847...\n",
      "Load from folder 2848...\n",
      "Load from folder 2849...\n",
      "Load from folder 2850...\n",
      "Load from folder 2851...\n",
      "Load from folder 2852...\n",
      "Load from folder 2853...\n",
      "Load from folder 2854...\n",
      "Load from folder 2855...\n",
      "Load from folder 2856...\n",
      "Load from folder 2857...\n",
      "Load from folder 2858...\n",
      "Load from folder 2859...\n",
      "Load from folder 2860...\n",
      "Load from folder 2861...\n",
      "Load from folder 2862...\n",
      "Load from folder 2863...\n",
      "Load from folder 2864...\n",
      "Load from folder 2865...\n",
      "Load from folder 2866...\n",
      "Load from folder 2867...\n",
      "Load from folder 2868...\n",
      "Load from folder 2869...\n",
      "Load from folder 2870...\n",
      "Load from folder 2871...\n",
      "Load from folder 2872...\n",
      "Load from folder 2873...\n",
      "Load from folder 2874...\n",
      "Load from folder 2875...\n",
      "Load from folder 2876...\n",
      "Load from folder 2877...\n",
      "Load from folder 2878...\n",
      "Load from folder 2879...\n",
      "Load from folder 2880...\n",
      "Load from folder 2881...\n",
      "Load from folder 2882...\n",
      "Load from folder 2883...\n",
      "Load from folder 2884...\n",
      "Load from folder 2885...\n",
      "Load from folder 2886...\n",
      "Load from folder 2887...\n",
      "Load from folder 2888...\n",
      "Load from folder 2889...\n",
      "Load from folder 2890...\n",
      "Load from folder 2891...\n",
      "Load from folder 2892...\n",
      "Load from folder 2893...\n",
      "Load from folder 2894...\n",
      "Load from folder 2895...\n",
      "Load from folder 2896...\n",
      "Load from folder 2897...\n",
      "Load from folder 2898...\n",
      "Load from folder 2899...\n",
      "Load from folder 2900...\n",
      "Load from folder 2901...\n",
      "Load from folder 2902...\n",
      "Load from folder 2903...\n",
      "Load from folder 2904...\n",
      "Load from folder 2905...\n",
      "Load from folder 2906...\n",
      "Load from folder 2907...\n",
      "Load from folder 2908...\n",
      "Load from folder 2909...\n",
      "Load from folder 2910...\n",
      "Load from folder 2911...\n",
      "Load from folder 2912...\n",
      "Load from folder 2913...\n",
      "Load from folder 2914...\n",
      "Load from folder 2915...\n",
      "Load from folder 2916...\n",
      "Load from folder 2917...\n",
      "Load from folder 2918...\n",
      "Load from folder 2919...\n",
      "Load from folder 2920...\n",
      "Load from folder 2921...\n",
      "Load from folder 2922...\n",
      "Load from folder 2923...\n",
      "Load from folder 2924...\n",
      "Load from folder 2925...\n",
      "Load from folder 2926...\n",
      "Load from folder 2927...\n",
      "Load from folder 2928...\n",
      "Load from folder 2929...\n",
      "Load from folder 2930...\n",
      "Load from folder 2931...\n",
      "Load from folder 2932...\n",
      "Load from folder 2933...\n",
      "Load from folder 2934...\n",
      "Load from folder 2935...\n",
      "Load from folder 2936...\n",
      "Load from folder 2937...\n",
      "Load from folder 2938...\n",
      "Load from folder 2939...\n",
      "Load from folder 2940...\n",
      "Load from folder 2941...\n",
      "Load from folder 2942...\n",
      "Load from folder 2943...\n",
      "Load from folder 2944...\n",
      "Load from folder 2945...\n",
      "Load from folder 2946...\n",
      "Load from folder 2947...\n",
      "Load from folder 2948...\n",
      "Load from folder 2949...\n",
      "Load from folder 2950...\n",
      "Load from folder 2951...\n",
      "Load from folder 2952...\n",
      "Load from folder 2953...\n",
      "Load from folder 2954...\n",
      "Load from folder 2955...\n",
      "Load from folder 2956...\n",
      "Load from folder 2957...\n",
      "Load from folder 2958...\n",
      "Load from folder 2959...\n",
      "Load from folder 2960...\n",
      "Load from folder 2961...\n",
      "Load from folder 2962...\n",
      "Load from folder 2963...\n",
      "Load from folder 2964...\n",
      "Load from folder 2965...\n",
      "Load from folder 2966...\n",
      "Load from folder 2967...\n",
      "Load from folder 2968...\n",
      "Load from folder 2969...\n",
      "Load from folder 2970...\n",
      "Load from folder 2971...\n",
      "Load from folder 2972...\n",
      "Load from folder 2973...\n",
      "Load from folder 2974...\n",
      "Load from folder 2975...\n",
      "Load from folder 2976...\n",
      "Load from folder 2977...\n",
      "Load from folder 2978...\n",
      "Load from folder 2979...\n",
      "Load from folder 2980...\n",
      "Load from folder 2981...\n",
      "Load from folder 2982...\n",
      "Load from folder 2983...\n",
      "Load from folder 2984...\n",
      "Load from folder 2985...\n",
      "Load from folder 2986...\n",
      "Load from folder 2987...\n",
      "Load from folder 2988...\n",
      "Load from folder 2989...\n",
      "Load from folder 2990...\n",
      "Load from folder 2991...\n",
      "Load from folder 2992...\n",
      "Load from folder 2993...\n",
      "Load from folder 2994...\n",
      "Load from folder 2995...\n",
      "Load from folder 2996...\n",
      "Load from folder 2997...\n",
      "Load from folder 2998...\n",
      "Load from folder 2999...\n",
      "Load from folder 3000...\n",
      "Load from folder 3001...\n",
      "Load from folder 3002...\n",
      "Load from folder 3003...\n",
      "Load from folder 3004...\n",
      "Load from folder 3005...\n",
      "Load from folder 3006...\n",
      "Load from folder 3007...\n",
      "Load from folder 3008...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 3009...\n",
      "Load from folder 3010...\n",
      "Load from folder 3011...\n",
      "Load from folder 3012...\n",
      "Load from folder 3013...\n",
      "Load from folder 3014...\n",
      "Load from folder 3015...\n",
      "Load from folder 3016...\n",
      "Load from folder 3017...\n",
      "Load from folder 3018...\n",
      "Load from folder 3019...\n",
      "Load from folder 3020...\n",
      "Load from folder 3021...\n",
      "Load from folder 3022...\n",
      "Load from folder 3023...\n",
      "Load from folder 3024...\n",
      "Load from folder 3025...\n",
      "Load from folder 3026...\n",
      "Load from folder 3027...\n",
      "Load from folder 3028...\n",
      "Load from folder 3029...\n",
      "Load from folder 3030...\n",
      "Load from folder 3031...\n",
      "Load from folder 3032...\n",
      "Load from folder 3033...\n",
      "Load from folder 3034...\n",
      "Load from folder 3035...\n",
      "Load from folder 3036...\n",
      "Load from folder 3037...\n",
      "Load from folder 3038...\n",
      "Load from folder 3039...\n",
      "Load from folder 3040...\n",
      "Load from folder 3041...\n",
      "Load from folder 3042...\n",
      "Load from folder 3043...\n",
      "Load from folder 3044...\n",
      "Load from folder 3045...\n",
      "Load from folder 3046...\n",
      "Load from folder 3047...\n",
      "Load from folder 3048...\n",
      "Load from folder 3049...\n",
      "Load from folder 3050...\n",
      "Load from folder 3051...\n",
      "Load from folder 3052...\n",
      "Load from folder 3053...\n",
      "Load from folder 3054...\n",
      "Load from folder 3055...\n",
      "Load from folder 3056...\n",
      "Load from folder 3057...\n",
      "Load from folder 3058...\n",
      "Load from folder 3059...\n",
      "Load from folder 3060...\n",
      "Load from folder 3061...\n",
      "Load from folder 3062...\n",
      "Load from folder 3063...\n",
      "Load from folder 3064...\n",
      "Load from folder 3065...\n",
      "Load from folder 3066...\n",
      "Load from folder 3067...\n",
      "Load from folder 3068...\n",
      "Load from folder 3069...\n",
      "Load from folder 3070...\n",
      "Load from folder 3071...\n",
      "Load from folder 3072...\n",
      "Load from folder 3073...\n",
      "Load from folder 3074...\n",
      "Load from folder 3075...\n",
      "Load from folder 3076...\n",
      "Load from folder 3077...\n",
      "Load from folder 3078...\n",
      "Load from folder 3079...\n",
      "Load from folder 3080...\n",
      "Load from folder 3081...\n",
      "Load from folder 3082...\n",
      "Load from folder 3083...\n",
      "Load from folder 3084...\n",
      "Load from folder 3085...\n",
      "Load from folder 3086...\n",
      "Load from folder 3087...\n",
      "Load from folder 3088...\n",
      "Load from folder 3089...\n",
      "Load from folder 3090...\n",
      "Load from folder 3091...\n",
      "Load from folder 3092...\n",
      "Load from folder 3093...\n",
      "Load from folder 3094...\n",
      "Load from folder 3095...\n",
      "Load from folder 3096...\n",
      "Load from folder 3097...\n",
      "Load from folder 3098...\n",
      "Load from folder 3099...\n",
      "Load from folder 3100...\n",
      "Load from folder 3101...\n",
      "Load from folder 3102...\n",
      "Load from folder 3103...\n",
      "Load from folder 3104...\n",
      "Load from folder 3105...\n",
      "Load from folder 3106...\n",
      "Load from folder 3107...\n",
      "Load from folder 3108...\n",
      "Load from folder 3109...\n",
      "Load from folder 3110...\n",
      "Load from folder 3111...\n",
      "Load from folder 3112...\n",
      "Load from folder 3113...\n",
      "Load from folder 3114...\n",
      "Load from folder 3115...\n",
      "Load from folder 3116...\n",
      "Load from folder 3117...\n",
      "Load from folder 3118...\n",
      "Load from folder 3119...\n",
      "Load from folder 3120...\n",
      "Load from folder 3121...\n",
      "Load from folder 3122...\n",
      "Load from folder 3123...\n",
      "Load from folder 3124...\n",
      "Load from folder 3125...\n",
      "Load from folder 3126...\n",
      "Load from folder 3127...\n",
      "Load from folder 3128...\n",
      "Load from folder 3129...\n",
      "Load from folder 3130...\n",
      "Load from folder 3131...\n",
      "Load from folder 3132...\n",
      "Load from folder 3133...\n",
      "Load from folder 3134...\n",
      "Load from folder 3135...\n",
      "Load from folder 3136...\n",
      "Load from folder 3137...\n",
      "Load from folder 3138...\n",
      "Load from folder 3139...\n",
      "Load from folder 3140...\n",
      "Load from folder 3141...\n",
      "Load from folder 3142...\n",
      "Load from folder 3143...\n",
      "Load from folder 3144...\n",
      "Load from folder 3145...\n",
      "Load from folder 3146...\n",
      "Load from folder 3147...\n",
      "Load from folder 3148...\n",
      "Load from folder 3149...\n",
      "Load from folder 3150...\n",
      "Load from folder 3151...\n",
      "Load from folder 3152...\n",
      "Load from folder 3153...\n",
      "Load from folder 3154...\n",
      "Load from folder 3155...\n",
      "Load from folder 3156...\n",
      "Load from folder 3157...\n",
      "Load from folder 3158...\n",
      "Load from folder 3159...\n",
      "Load from folder 3160...\n",
      "Load from folder 3161...\n",
      "Load from folder 3162...\n",
      "Load from folder 3163...\n",
      "Load from folder 3164...\n",
      "Load from folder 3165...\n",
      "Load from folder 3166...\n",
      "Load from folder 3167...\n",
      "Load from folder 3168...\n",
      "Load from folder 3169...\n",
      "Load from folder 3170...\n",
      "Load from folder 3171...\n",
      "Load from folder 3172...\n",
      "Load from folder 3173...\n",
      "Load from folder 3174...\n",
      "Load from folder 3175...\n",
      "Load from folder 3176...\n",
      "Load from folder 3177...\n",
      "Load from folder 3178...\n",
      "Load from folder 3179...\n",
      "Load from folder 3180...\n",
      "Load from folder 3181...\n",
      "Load from folder 3182...\n",
      "Load from folder 3183...\n",
      "Load from folder 3184...\n",
      "Load from folder 3185...\n",
      "Load from folder 3186...\n",
      "Load from folder 3187...\n",
      "Load from folder 3188...\n",
      "Load from folder 3189...\n",
      "Load from folder 3190...\n",
      "Load from folder 3191...\n",
      "Load from folder 3192...\n",
      "Load from folder 3193...\n",
      "Load from folder 3194...\n",
      "Load from folder 3195...\n",
      "Load from folder 3196...\n",
      "Load from folder 3197...\n",
      "Load from folder 3198...\n",
      "Load from folder 3199...\n",
      "Load from folder 3200...\n",
      "Load from folder 3201...\n",
      "Load from folder 3202...\n",
      "Load from folder 3203...\n",
      "Load from folder 3204...\n",
      "Load from folder 3205...\n",
      "Load from folder 3206...\n",
      "Load from folder 3207...\n",
      "Load from folder 3208...\n",
      "Load from folder 3209...\n",
      "Load from folder 3210...\n",
      "Load from folder 3211...\n",
      "Load from folder 3212...\n",
      "Load from folder 3213...\n",
      "Load from folder 3214...\n",
      "Load from folder 3215...\n",
      "Load from folder 3216...\n",
      "Load from folder 3217...\n",
      "Load from folder 3218...\n",
      "Load from folder 3219...\n",
      "Load from folder 3220...\n",
      "Load from folder 3221...\n",
      "Load from folder 3222...\n",
      "Load from folder 3223...\n",
      "Load from folder 3224...\n",
      "Load from folder 3225...\n",
      "Load from folder 3226...\n",
      "Load from folder 3227...\n",
      "Load from folder 3228...\n",
      "Load from folder 3229...\n",
      "Load from folder 3230...\n",
      "Load from folder 3231...\n",
      "Load from folder 3232...\n",
      "Load from folder 3233...\n",
      "Load from folder 3234...\n",
      "Load from folder 3235...\n",
      "Load from folder 3236...\n",
      "Load from folder 3237...\n",
      "Load from folder 3238...\n",
      "Load from folder 3239...\n",
      "Load from folder 3240...\n",
      "Load from folder 3241...\n",
      "Load from folder 3242...\n",
      "Load from folder 3243...\n",
      "Load from folder 3244...\n",
      "Load from folder 3245...\n",
      "Load from folder 3246...\n",
      "Load from folder 3247...\n",
      "Load from folder 3248...\n",
      "Load from folder 3249...\n",
      "Load from folder 3250...\n",
      "Load from folder 3251...\n",
      "Load from folder 3252...\n",
      "Load from folder 3253...\n",
      "Load from folder 3254...\n",
      "Load from folder 3255...\n",
      "Load from folder 3256...\n",
      "Load from folder 3257...\n",
      "Load from folder 3258...\n",
      "Load from folder 3259...\n",
      "Load from folder 3260...\n",
      "Load from folder 3261...\n",
      "Load from folder 3262...\n",
      "Load from folder 3263...\n",
      "Load from folder 3264...\n",
      "Load from folder 3265...\n",
      "Load from folder 3266...\n",
      "Load from folder 3267...\n",
      "Load from folder 3268...\n",
      "Load from folder 3269...\n",
      "Load from folder 3270...\n",
      "Load from folder 3271...\n",
      "Load from folder 3272...\n",
      "Load from folder 3273...\n",
      "Load from folder 3274...\n",
      "Load from folder 3275...\n",
      "Load from folder 3276...\n",
      "Load from folder 3277...\n",
      "Load from folder 3278...\n",
      "Load from folder 3279...\n",
      "Load from folder 3280...\n",
      "Load from folder 3281...\n",
      "Load from folder 3282...\n",
      "Load from folder 3283...\n",
      "Load from folder 3284...\n",
      "Load from folder 3285...\n",
      "Load from folder 3286...\n",
      "Load from folder 3287...\n",
      "Load from folder 3288...\n",
      "Load from folder 3289...\n",
      "Load from folder 3290...\n",
      "Load from folder 3291...\n",
      "Load from folder 3292...\n",
      "Load from folder 3293...\n",
      "Load from folder 3294...\n",
      "Load from folder 3295...\n",
      "Load from folder 3296...\n",
      "Load from folder 3297...\n",
      "Load from folder 3298...\n",
      "Load from folder 3299...\n",
      "Load from folder 3300...\n",
      "Load from folder 3301...\n",
      "Load from folder 3302...\n",
      "Load from folder 3303...\n",
      "Load from folder 3304...\n",
      "Load from folder 3305...\n",
      "Load from folder 3306...\n",
      "Load from folder 3307...\n",
      "Load from folder 3308...\n",
      "Load from folder 3309...\n",
      "Load from folder 3310...\n",
      "Load from folder 3311...\n",
      "Load from folder 3312...\n",
      "Load from folder 3313...\n",
      "Load from folder 3314...\n",
      "Load from folder 3315...\n",
      "Load from folder 3316...\n",
      "Load from folder 3317...\n",
      "Load from folder 3318...\n",
      "Load from folder 3319...\n",
      "Load from folder 3320...\n",
      "Load from folder 3321...\n",
      "Load from folder 3322...\n",
      "Load from folder 3323...\n",
      "Load from folder 3324...\n",
      "Load from folder 3325...\n",
      "Load from folder 3326...\n",
      "Load from folder 3327...\n",
      "Load from folder 3328...\n",
      "Load from folder 3329...\n",
      "Load from folder 3330...\n",
      "Load from folder 3331...\n",
      "Load from folder 3332...\n",
      "Load from folder 3333...\n",
      "Load from folder 3334...\n",
      "Load from folder 3335...\n",
      "Load from folder 3336...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 3337...\n",
      "Load from folder 3338...\n",
      "Load from folder 3339...\n",
      "Load from folder 3340...\n",
      "Load from folder 3341...\n",
      "Load from folder 3342...\n",
      "Load from folder 3343...\n",
      "Load from folder 3344...\n",
      "Load from folder 3345...\n",
      "Load from folder 3346...\n",
      "Load from folder 3347...\n",
      "Load from folder 3348...\n",
      "Load from folder 3349...\n",
      "Load from folder 3350...\n",
      "Load from folder 3351...\n",
      "Load from folder 3352...\n",
      "Load from folder 3353...\n",
      "Load from folder 3354...\n",
      "Load from folder 3355...\n",
      "Load from folder 3356...\n",
      "Load from folder 3357...\n",
      "Load from folder 3358...\n",
      "Load from folder 3359...\n",
      "Load from folder 3360...\n",
      "Load from folder 3361...\n",
      "Load from folder 3362...\n",
      "Load from folder 3363...\n",
      "Load from folder 3364...\n",
      "Load from folder 3365...\n",
      "Load from folder 3366...\n",
      "Load from folder 3367...\n",
      "Load from folder 3368...\n",
      "Load from folder 3369...\n",
      "Load from folder 3370...\n",
      "Load from folder 3371...\n",
      "Load from folder 3372...\n",
      "Load from folder 3373...\n",
      "Load from folder 3374...\n",
      "Load from folder 3375...\n",
      "Load from folder 3376...\n",
      "Load from folder 3377...\n",
      "Load from folder 3378...\n",
      "Load from folder 3379...\n",
      "Load from folder 3380...\n",
      "Load from folder 3381...\n",
      "Load from folder 3382...\n",
      "Load from folder 3383...\n",
      "Load from folder 3384...\n",
      "Load from folder 3385...\n",
      "Load from folder 3386...\n",
      "Load from folder 3387...\n",
      "Load from folder 3388...\n",
      "Load from folder 3389...\n",
      "Load from folder 3390...\n",
      "Load from folder 3391...\n",
      "Load from folder 3392...\n",
      "Load from folder 3393...\n",
      "Load from folder 3394...\n",
      "Load from folder 3395...\n",
      "Load from folder 3396...\n",
      "Load from folder 3397...\n",
      "Load from folder 3398...\n",
      "Load from folder 3399...\n",
      "Load from folder 3400...\n",
      "Load from folder 3401...\n",
      "Load from folder 3402...\n",
      "Load from folder 3403...\n",
      "Load from folder 3404...\n",
      "Load from folder 3405...\n",
      "Load from folder 3406...\n",
      "Load from folder 3407...\n",
      "Load from folder 3408...\n",
      "Load from folder 3409...\n",
      "Load from folder 3410...\n",
      "Load from folder 3411...\n",
      "Load from folder 3412...\n",
      "Load from folder 3413...\n",
      "Load from folder 3414...\n",
      "Load from folder 3415...\n",
      "Load from folder 3416...\n",
      "Load from folder 3417...\n",
      "Load from folder 3418...\n",
      "Load from folder 3419...\n",
      "Load from folder 3420...\n",
      "Load from folder 3421...\n",
      "Load from folder 3422...\n",
      "Load from folder 3423...\n",
      "Load from folder 3424...\n",
      "Load from folder 3425...\n",
      "Load from folder 3426...\n",
      "Load from folder 3427...\n",
      "Load from folder 3428...\n",
      "Load from folder 3429...\n",
      "Load from folder 3430...\n",
      "Load from folder 3431...\n",
      "Load from folder 3432...\n",
      "Load from folder 3433...\n",
      "Load from folder 3434...\n",
      "Load from folder 3435...\n",
      "Load from folder 3436...\n",
      "Load from folder 3437...\n",
      "Load from folder 3438...\n",
      "Load from folder 3439...\n",
      "Load from folder 3440...\n",
      "Load from folder 3441...\n",
      "Load from folder 3442...\n",
      "Load from folder 3443...\n",
      "Load from folder 3444...\n",
      "Load from folder 3445...\n",
      "Load from folder 3446...\n",
      "Load from folder 3447...\n",
      "Load from folder 3448...\n",
      "Load from folder 3449...\n",
      "Load from folder 3450...\n",
      "Load from folder 3451...\n",
      "Load from folder 3452...\n",
      "Load from folder 3453...\n",
      "Load from folder 3454...\n",
      "Load from folder 3455...\n",
      "Load from folder 3456...\n",
      "Load from folder 3457...\n",
      "Load from folder 3458...\n",
      "Load from folder 3459...\n",
      "Load from folder 3460...\n",
      "Load from folder 3461...\n",
      "Load from folder 3462...\n",
      "Load from folder 3463...\n",
      "Load from folder 3464...\n",
      "Load from folder 3465...\n",
      "Load from folder 3466...\n",
      "Load from folder 3467...\n",
      "Load from folder 3468...\n",
      "Load from folder 3469...\n",
      "Load from folder 3470...\n",
      "Load from folder 3471...\n",
      "Load from folder 3472...\n",
      "Load from folder 3473...\n",
      "Load from folder 3474...\n",
      "Load from folder 3475...\n",
      "Load from folder 3476...\n",
      "Load from folder 3477...\n",
      "Load from folder 3478...\n",
      "Load from folder 3479...\n",
      "Load from folder 3480...\n",
      "Load from folder 3481...\n",
      "Load from folder 3482...\n",
      "Load from folder 3483...\n",
      "Load from folder 3484...\n",
      "Load from folder 3485...\n",
      "Load from folder 3486...\n",
      "Load from folder 3487...\n",
      "Load from folder 3488...\n",
      "Load from folder 3489...\n",
      "Load from folder 3490...\n",
      "Load from folder 3491...\n",
      "Load from folder 3492...\n",
      "Load from folder 3493...\n",
      "Load from folder 3494...\n",
      "Load from folder 3495...\n",
      "Load from folder 3496...\n",
      "Load from folder 3497...\n",
      "Load from folder 3498...\n",
      "Load from folder 3499...\n",
      "Load from folder 3500...\n",
      "Load from folder 3501...\n",
      "Load from folder 3502...\n",
      "Load from folder 3503...\n",
      "Load from folder 3504...\n",
      "Load from folder 3505...\n",
      "Load from folder 3506...\n",
      "Load from folder 3507...\n",
      "Load from folder 3508...\n",
      "Load from folder 3509...\n",
      "Load from folder 3510...\n",
      "Load from folder 3511...\n",
      "Load from folder 3512...\n",
      "Load from folder 3513...\n",
      "Load from folder 3514...\n",
      "Load from folder 3515...\n",
      "Load from folder 3516...\n",
      "Load from folder 3517...\n",
      "Load from folder 3518...\n",
      "Load from folder 3519...\n",
      "Load from folder 3520...\n",
      "Load from folder 3521...\n",
      "Load from folder 3522...\n",
      "Load from folder 3523...\n",
      "Load from folder 3524...\n",
      "Load from folder 3525...\n",
      "Load from folder 3526...\n",
      "Load from folder 3527...\n",
      "Load from folder 3528...\n",
      "Load from folder 3529...\n",
      "Load from folder 3530...\n",
      "Load from folder 3531...\n",
      "Load from folder 3532...\n",
      "Load from folder 3533...\n",
      "Load from folder 3534...\n",
      "Load from folder 3535...\n",
      "Load from folder 3536...\n",
      "Load from folder 3537...\n",
      "Load from folder 3538...\n",
      "Load from folder 3539...\n",
      "Load from folder 3540...\n",
      "Load from folder 3541...\n",
      "Load from folder 3542...\n",
      "Load from folder 3543...\n",
      "Load from folder 3544...\n",
      "Load from folder 3545...\n",
      "Load from folder 3546...\n",
      "Load from folder 3547...\n",
      "Load from folder 3548...\n",
      "Load from folder 3549...\n",
      "Load from folder 3550...\n",
      "Load from folder 3551...\n",
      "Load from folder 3552...\n",
      "Load from folder 3553...\n",
      "Load from folder 3554...\n",
      "Load from folder 3555...\n",
      "Load from folder 3556...\n",
      "Load from folder 3557...\n",
      "Load from folder 3558...\n",
      "Load from folder 3559...\n",
      "Load from folder 3560...\n",
      "Load from folder 3561...\n",
      "Load from folder 3562...\n",
      "Load from folder 3563...\n",
      "Load from folder 3564...\n",
      "Load from folder 3565...\n",
      "Load from folder 3566...\n",
      "Load from folder 3567...\n",
      "Load from folder 3568...\n",
      "Load from folder 3569...\n",
      "Load from folder 3570...\n",
      "Load from folder 3571...\n",
      "Load from folder 3572...\n",
      "Load from folder 3573...\n",
      "Load from folder 3574...\n",
      "Load from folder 3575...\n",
      "Load from folder 3576...\n",
      "Load from folder 3577...\n",
      "Load from folder 3578...\n",
      "Load from folder 3579...\n",
      "Load from folder 3580...\n",
      "Load from folder 3581...\n",
      "Load from folder 3582...\n",
      "Load from folder 3583...\n",
      "Load from folder 3584...\n",
      "Load from folder 3585...\n",
      "Load from folder 3586...\n",
      "Load from folder 3587...\n",
      "Load from folder 3588...\n",
      "Load from folder 3589...\n",
      "Load from folder 3590...\n",
      "Load from folder 3591...\n",
      "Load from folder 3592...\n",
      "Load from folder 3593...\n",
      "Load from folder 3594...\n",
      "Load from folder 3595...\n",
      "Load from folder 3596...\n",
      "Load from folder 3597...\n",
      "Load from folder 3598...\n",
      "Load from folder 3599...\n",
      "Load from folder 3600...\n",
      "Load from folder 3601...\n",
      "Load from folder 3602...\n",
      "Load from folder 3603...\n",
      "Load from folder 3604...\n",
      "Load from folder 3605...\n",
      "Load from folder 3606...\n",
      "Load from folder 3607...\n",
      "Load from folder 3608...\n",
      "Load from folder 3609...\n",
      "Load from folder 3610...\n",
      "Load from folder 3611...\n",
      "Load from folder 3612...\n",
      "Load from folder 3613...\n",
      "Load from folder 3614...\n",
      "Load from folder 3615...\n",
      "Load from folder 3616...\n",
      "Load from folder 3617...\n",
      "Load from folder 3618...\n",
      "Load from folder 3619...\n",
      "Load from folder 3620...\n",
      "Load from folder 3621...\n",
      "Load from folder 3622...\n",
      "Load from folder 3623...\n",
      "Load from folder 3624...\n",
      "Load from folder 3625...\n",
      "Load from folder 3626...\n",
      "Load from folder 3627...\n",
      "Load from folder 3628...\n",
      "Load from folder 3629...\n",
      "Load from folder 3630...\n",
      "Load from folder 3631...\n",
      "Load from folder 3632...\n",
      "Load from folder 3633...\n",
      "Load from folder 3634...\n",
      "Load from folder 3635...\n",
      "Load from folder 3636...\n",
      "Load from folder 3637...\n",
      "Load from folder 3638...\n",
      "Load from folder 3639...\n",
      "Load from folder 3640...\n",
      "Load from folder 3641...\n",
      "Load from folder 3642...\n",
      "Load from folder 3643...\n",
      "Load from folder 3644...\n",
      "Load from folder 3645...\n",
      "Load from folder 3646...\n",
      "Load from folder 3647...\n",
      "Load from folder 3648...\n",
      "Load from folder 3649...\n",
      "Load from folder 3650...\n",
      "Load from folder 3651...\n",
      "Load from folder 3652...\n",
      "Load from folder 3653...\n",
      "Load from folder 3654...\n",
      "Load from folder 3655...\n",
      "Load from folder 3656...\n",
      "Load from folder 3657...\n",
      "Load from folder 3658...\n",
      "Load from folder 3659...\n",
      "Load from folder 3660...\n",
      "Load from folder 3661...\n",
      "Load from folder 3662...\n",
      "Load from folder 3663...\n",
      "Load from folder 3664...\n",
      "Load from folder 3665...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load from folder 3666...\n",
      "Load from folder 3667...\n",
      "Load from folder 3668...\n",
      "Load from folder 3669...\n",
      "Load from folder 3670...\n",
      "Load from folder 3671...\n",
      "Load from folder 3672...\n",
      "Load from folder 3673...\n",
      "Load from folder 3674...\n",
      "Load from folder 3675...\n",
      "Load from folder 3676...\n",
      "Load from folder 3677...\n",
      "Load from folder 3678...\n",
      "Load from folder 3679...\n",
      "Load from folder 3680...\n",
      "Load from folder 3681...\n",
      "Load from folder 3682...\n",
      "Load from folder 3683...\n",
      "Load from folder 3684...\n",
      "Load from folder 3685...\n",
      "Load from folder 3686...\n",
      "Load from folder 3687...\n",
      "Load from folder 3688...\n",
      "Load from folder 3689...\n",
      "Load from folder 3690...\n",
      "Load from folder 3691...\n",
      "Load from folder 3692...\n",
      "Load from folder 3693...\n",
      "Load from folder 3694...\n",
      "Load from folder 3695...\n",
      "Load from folder 3696...\n",
      "Load from folder 3697...\n",
      "Load from folder 3698...\n",
      "Load from folder 3699...\n",
      "Load from folder 3700...\n",
      "Load from folder 3701...\n",
      "Load from folder 3702...\n",
      "Load from folder 3703...\n",
      "Load from folder 3704...\n",
      "Load from folder 3705...\n",
      "Load from folder 3706...\n",
      "Load from folder 3707...\n",
      "Load from folder 3708...\n",
      "Load from folder 3709...\n",
      "Load from folder 3710...\n",
      "Load from folder 3711...\n",
      "Load from folder 3712...\n",
      "Load from folder 3713...\n",
      "Load from folder 3714...\n",
      "Load from folder 3715...\n",
      "Load from folder 3716...\n",
      "Load from folder 3717...\n",
      "Load from folder 3718...\n",
      "Load from folder 3719...\n",
      "Load from folder 3720...\n",
      "Load from folder 3721...\n",
      "Load from folder 3722...\n",
      "Load from folder 3723...\n",
      "Load from folder 3724...\n",
      "Load from folder 3725...\n",
      "Load from folder 3726...\n",
      "Load from folder 3727...\n",
      "Load from folder 3728...\n",
      "Load from folder 3729...\n",
      "Load from folder 3730...\n",
      "Load from folder 3731...\n",
      "Load from folder 3732...\n",
      "Load from folder 3733...\n",
      "Load from folder 3734...\n",
      "Load from folder 3735...\n",
      "Load from folder 3736...\n",
      "Load from folder 3737...\n",
      "Load from folder 3738...\n",
      "Load from folder 3739...\n",
      "Load from folder 3740...\n",
      "Load from folder 3741...\n",
      "Load from folder 3742...\n",
      "Load from folder 3743...\n",
      "Load from folder 3744...\n",
      "Load from folder 3745...\n",
      "Load from folder 3746...\n",
      "Load from folder 3747...\n",
      "Load from folder 3748...\n",
      "Load from folder 3749...\n",
      "Load from folder 3750...\n",
      "Load from folder 3751...\n",
      "Load from folder 3752...\n",
      "Load from folder 3753...\n",
      "Load from folder 3754...\n",
      "Load from folder 3755...\n",
      "Load from folder 3756...\n",
      "Load from folder 3757...\n",
      "Load from folder 3758...\n",
      "Load from folder 3759...\n",
      "Load from folder 3760...\n",
      "Load from folder 3761...\n",
      "Load from folder 3762...\n",
      "Load from folder 3763...\n",
      "Load from folder 3764...\n",
      "Load from folder 3765...\n",
      "Load from folder 3766...\n",
      "Load from folder 3767...\n",
      "Load from folder 3768...\n",
      "Load from folder 3769...\n",
      "Load from folder 3770...\n",
      "Load from folder 3771...\n",
      "Load from folder 3772...\n",
      "Load from folder 3773...\n",
      "Load from folder 3774...\n",
      "Load from folder 3775...\n",
      "Load from folder 3776...\n",
      "Load from folder 3777...\n",
      "Load from folder 3778...\n",
      "Load from folder 3779...\n",
      "Load from folder 3780...\n",
      "Load from folder 3781...\n",
      "Load from folder 3782...\n",
      "Load from folder 3783...\n",
      "Load from folder 3784...\n",
      "Load from folder 3785...\n",
      "Load from folder 3786...\n",
      "Load from folder 3787...\n",
      "Load from folder 3788...\n",
      "Load from folder 3789...\n",
      "Load from folder 3790...\n",
      "Load from folder 3791...\n",
      "Load from folder 3792...\n",
      "Load from folder 3793...\n",
      "Load from folder 3794...\n",
      "Load from folder 3795...\n",
      "Load from folder 3796...\n",
      "Load from folder 3797...\n",
      "Load from folder 3798...\n",
      "Load from folder 3799...\n",
      "Load from folder 3800...\n",
      "Load from folder 3801...\n",
      "Load from folder 3802...\n",
      "Load from folder 3803...\n",
      "Load from folder 3804...\n",
      "Load from folder 3805...\n",
      "Load from folder 3806...\n",
      "Load from folder 3807...\n",
      "Load from folder 3808...\n",
      "Load from folder 3809...\n",
      "Load from folder 3810...\n",
      "Load from folder 3811...\n",
      "Load from folder 3812...\n",
      "Load from folder 3813...\n",
      "Load from folder 3814...\n",
      "Load from folder 3815...\n",
      "Load from folder 3816...\n",
      "Load from folder 3817...\n",
      "Load from folder 3818...\n",
      "Load from folder 3819...\n",
      "Load from folder 3820...\n",
      "Load from folder 3821...\n",
      "Load from folder 3822...\n",
      "Load from folder 3823...\n",
      "Load from folder 3824...\n",
      "Load from folder 3825...\n",
      "Load from folder 3826...\n",
      "Load from folder 3827...\n",
      "Load from folder 3828...\n",
      "Load from folder 3829...\n",
      "Load from folder 3830...\n",
      "Load from folder 3831...\n",
      "\n",
      "Number of samples: 95800 where each sample is of size: (64, 64)\n",
      "\n",
      "Elapsed time for data preparation: 00:05:54\n",
      "\n",
      "Number of classes: 3832\n",
      "Data splitted in train and test sets\n",
      "Save train and test images and labels to npz files...\n",
      "Train set splitted in train and validation sets\n",
      "\n",
      "Train images dimensions: (77598, 64, 64, 1)\n",
      "\n",
      "Validation images dimensions: (8622, 64, 64, 1)\n",
      "\n",
      "Test images dimensions: (9580, 64, 64, 1)\n"
     ]
    }
   ],
   "source": [
    "# Get the dataloader that load the data\n",
    "data_loader_kanji = data_loading.DataLoader(\n",
    "    data_dir=data_dir,\n",
    "    balancing=balancing, upsampling_size=upsampling_size,\n",
    "    load_from_npz=load_from_npz, npz_paths=npz_paths,\n",
    "    save_to_npz=save_to_npz,\n",
    "    csv_file=csv_file,\n",
    "    use_grayscale=use_grayscale,\n",
    "    label=label,\n",
    "    resize_shape=resize_shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the properties of the dataloader\n",
    "input_shape = data_loader_kanji.input_shape\n",
    "num_classes = data_loader_kanji.num_classes\n",
    "train_images = data_loader_kanji.train_images\n",
    "train_labels = data_loader_kanji.train_labels\n",
    "validation_images = data_loader_kanji.validation_images\n",
    "validation_labels = data_loader_kanji.validation_labels\n",
    "test_images = data_loader_kanji.test_images\n",
    "test_labels = data_loader_kanji.test_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classes: 3832\n",
      "Number of train images: 77598 where each sample is of size: (64, 64, 1)\n",
      "Number of validation images: 8622 where each sample is of size: (64, 64, 1)\n",
      "Number of test images: 9580 where each sample is of size: (64, 64, 1)\n"
     ]
    }
   ],
   "source": [
    "print(\"Classes:\", num_classes)\n",
    "print(f\"Number of train images: {len(train_images)} where each sample is of size: {train_images.shape[1:]}\")\n",
    "print(f\"Number of validation images: {len(validation_images)} where each sample is of size: {validation_images.shape[1:]}\")\n",
    "print(f\"Number of test images: {len(test_images)} where each sample is of size: {test_images.shape[1:]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train a Base CNN Model on the Kanji Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "QdR71x49fcjb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 64, 64, 32)        832       \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 32, 32, 32)        25632     \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 16, 16, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 16, 16, 32)       128       \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 16, 16, 64)        18496     \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 16, 16, 64)        36928     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 8, 8, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 8, 8, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 8, 8, 96)          55392     \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 8, 8, 96)          83040     \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 4, 4, 96)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " batch_normalization_2 (Batc  (None, 4, 4, 96)         384       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 4, 4, 96)          0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 1536)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               196736    \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 3832)              494328    \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 912,152\n",
      "Trainable params: 911,768\n",
      "Non-trainable params: 384\n",
      "_________________________________________________________________\n",
      "Epoch 1/50\n",
      "607/607 [==============================] - 34s 39ms/step - loss: 9.0145 - accuracy: 0.0015 - val_loss: 8.0025 - val_accuracy: 0.0080\n",
      "Epoch 2/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 6.6747 - accuracy: 0.1181 - val_loss: 5.3088 - val_accuracy: 0.3285\n",
      "Epoch 3/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 4.3824 - accuracy: 0.4465 - val_loss: 3.6936 - val_accuracy: 0.5373\n",
      "Epoch 4/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 3.1991 - accuracy: 0.5981 - val_loss: 3.0360 - val_accuracy: 0.6124\n",
      "Epoch 5/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 2.6054 - accuracy: 0.6679 - val_loss: 2.5250 - val_accuracy: 0.6699\n",
      "Epoch 6/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 2.2186 - accuracy: 0.7158 - val_loss: 2.2205 - val_accuracy: 0.7088\n",
      "Epoch 7/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.9356 - accuracy: 0.7497 - val_loss: 1.9838 - val_accuracy: 0.7392\n",
      "Epoch 8/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 1.7068 - accuracy: 0.7803 - val_loss: 1.7872 - val_accuracy: 0.7713\n",
      "Epoch 9/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.5199 - accuracy: 0.8056 - val_loss: 1.6353 - val_accuracy: 0.7883\n",
      "Epoch 10/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.3583 - accuracy: 0.8269 - val_loss: 1.5463 - val_accuracy: 0.7994\n",
      "Epoch 11/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.2253 - accuracy: 0.8438 - val_loss: 1.4012 - val_accuracy: 0.8200\n",
      "Epoch 12/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.1079 - accuracy: 0.8623 - val_loss: 1.3006 - val_accuracy: 0.8307\n",
      "Epoch 13/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 1.0099 - accuracy: 0.8760 - val_loss: 1.2226 - val_accuracy: 0.8394\n",
      "Epoch 14/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.9261 - accuracy: 0.8879 - val_loss: 1.1515 - val_accuracy: 0.8512\n",
      "Epoch 15/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.8564 - accuracy: 0.8965 - val_loss: 1.0963 - val_accuracy: 0.8556\n",
      "Epoch 16/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.7932 - accuracy: 0.9070 - val_loss: 1.0297 - val_accuracy: 0.8700\n",
      "Epoch 17/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.7381 - accuracy: 0.9140 - val_loss: 1.0175 - val_accuracy: 0.8686\n",
      "Epoch 18/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.6868 - accuracy: 0.9213 - val_loss: 0.9617 - val_accuracy: 0.8758\n",
      "Epoch 19/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 0.6444 - accuracy: 0.9276 - val_loss: 0.9408 - val_accuracy: 0.8783\n",
      "Epoch 20/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 0.6065 - accuracy: 0.9331 - val_loss: 0.9275 - val_accuracy: 0.8833\n",
      "Epoch 21/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.5746 - accuracy: 0.9377 - val_loss: 0.8875 - val_accuracy: 0.8877\n",
      "Epoch 22/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.5465 - accuracy: 0.9414 - val_loss: 0.8682 - val_accuracy: 0.8896\n",
      "Epoch 23/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.5152 - accuracy: 0.9463 - val_loss: 0.8534 - val_accuracy: 0.8897\n",
      "Epoch 24/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.4902 - accuracy: 0.9500 - val_loss: 0.8250 - val_accuracy: 0.8934\n",
      "Epoch 25/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.4673 - accuracy: 0.9532 - val_loss: 0.8150 - val_accuracy: 0.8946\n",
      "Epoch 26/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.4488 - accuracy: 0.9558 - val_loss: 0.8101 - val_accuracy: 0.8923\n",
      "Epoch 27/50\n",
      "607/607 [==============================] - 25s 41ms/step - loss: 0.4290 - accuracy: 0.9585 - val_loss: 0.7935 - val_accuracy: 0.8977\n",
      "Epoch 28/50\n",
      "607/607 [==============================] - 26s 43ms/step - loss: 0.4149 - accuracy: 0.9596 - val_loss: 0.7847 - val_accuracy: 0.8998\n",
      "Epoch 29/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.4005 - accuracy: 0.9610 - val_loss: 0.7692 - val_accuracy: 0.9022\n",
      "Epoch 30/50\n",
      "607/607 [==============================] - 23s 37ms/step - loss: 0.3826 - accuracy: 0.9656 - val_loss: 0.7474 - val_accuracy: 0.9036\n",
      "Epoch 31/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.3701 - accuracy: 0.9664 - val_loss: 0.7517 - val_accuracy: 0.8992\n",
      "Epoch 32/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.3623 - accuracy: 0.9668 - val_loss: 0.7385 - val_accuracy: 0.9047\n",
      "Epoch 33/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.3530 - accuracy: 0.9671 - val_loss: 0.7432 - val_accuracy: 0.9037\n",
      "Epoch 34/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.3416 - accuracy: 0.9693 - val_loss: 0.7211 - val_accuracy: 0.9040\n",
      "Epoch 35/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "607/607 [==============================] - 24s 39ms/step - loss: 0.3294 - accuracy: 0.9711 - val_loss: 0.7342 - val_accuracy: 0.8994\n",
      "Epoch 36/50\n",
      "607/607 [==============================] - 24s 40ms/step - loss: 0.3252 - accuracy: 0.9708 - val_loss: 0.7256 - val_accuracy: 0.8992\n",
      "Epoch 37/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.3202 - accuracy: 0.9708 - val_loss: 0.7127 - val_accuracy: 0.9065\n",
      "Epoch 38/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.3088 - accuracy: 0.9730 - val_loss: 0.7159 - val_accuracy: 0.9033\n",
      "Epoch 39/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2994 - accuracy: 0.9748 - val_loss: 0.6853 - val_accuracy: 0.9096\n",
      "Epoch 40/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2946 - accuracy: 0.9747 - val_loss: 0.7286 - val_accuracy: 0.9016\n",
      "Epoch 41/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2943 - accuracy: 0.9731 - val_loss: 0.7031 - val_accuracy: 0.9059\n",
      "Epoch 42/50\n",
      "607/607 [==============================] - 23s 39ms/step - loss: 0.2853 - accuracy: 0.9753 - val_loss: 0.6808 - val_accuracy: 0.9112\n",
      "Epoch 43/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2810 - accuracy: 0.9751 - val_loss: 0.6812 - val_accuracy: 0.9106\n",
      "Epoch 44/50\n",
      "607/607 [==============================] - 26s 43ms/step - loss: 0.2733 - accuracy: 0.9769 - val_loss: 0.6873 - val_accuracy: 0.9037\n",
      "Epoch 45/50\n",
      "607/607 [==============================] - 25s 41ms/step - loss: 0.2712 - accuracy: 0.9758 - val_loss: 0.6764 - val_accuracy: 0.9103\n",
      "Epoch 46/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.2682 - accuracy: 0.9764 - val_loss: 0.6637 - val_accuracy: 0.9095\n",
      "Epoch 47/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2616 - accuracy: 0.9774 - val_loss: 0.6775 - val_accuracy: 0.9079\n",
      "Epoch 48/50\n",
      "607/607 [==============================] - 24s 39ms/step - loss: 0.2574 - accuracy: 0.9776 - val_loss: 0.6577 - val_accuracy: 0.9148\n",
      "Epoch 49/50\n",
      "607/607 [==============================] - 23s 38ms/step - loss: 0.2528 - accuracy: 0.9785 - val_loss: 0.6611 - val_accuracy: 0.9125\n",
      "Epoch 50/50\n",
      "607/607 [==============================] - 24s 40ms/step - loss: 0.2530 - accuracy: 0.9776 - val_loss: 0.6656 - val_accuracy: 0.9064\n",
      "\n",
      "Elapsed time for training: 00:19:41\n"
     ]
    }
   ],
   "source": [
    "model, history = train.train_model(data_loader_kanji, batch_size, epochs, learning_rate, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot and Evaluate the Trained Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plot loss and accuracy...\n",
      "Evaluate model on test data...\n",
      "75/75 [==============================] - 1s 17ms/step - loss: 0.6538 - accuracy: 0.9108\n",
      "The test loss is: 0.6538242697715759\n",
      "The test accuracy is: 0.9107515811920166\n",
      "\n",
      "\n",
      "\n",
      "75/75 [==============================] - 1s 10ms/step\n",
      "The Precision is: 0.9088483157113332\n",
      "The Recall is: 0.9042994747218143\n",
      "The F1-score is: 0.8925897047143961\n",
      "Plot some output predictions...\n",
      "2/2 [==============================] - 0s 3ms/step\n"
     ]
    }
   ],
   "source": [
    "test.plot_accuracy_and_loss(history, label)\n",
    "test.evaluate_model(model, test_images, test_labels, batch_size)\n",
    "test.plot_predictions(model, test_images, test_labels, resize_shape, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the transfer learning was successful, you should see accuracy_kanji.png, loss_kanji.png and predictions_kanji.png as well as a saved model file named something like kanji_model_Aug30-2022-10-52-42.h5 and the saved train and test images and labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load a Local Pretrained Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adapt the following line with the correct path of the local pretrained model\n",
    "model_path = \"kanji_model_Aug-31-2022-21-47-17.h5\"\n",
    "\n",
    "# Load the model\n",
    "model = keras.models.load_model(model_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transfer Learning on Chinese-MNIST Dataset using a Local Pretrained Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To use a local pretrained model we have to update some variables to get the right dataloader\n",
    "balancing = False       # We don't want to balance it because the dataset is balanced\n",
    "csv_file = \"..\\\\chinese-MNIST\\\\chinese_mnist.csv\"\n",
    "data_dir_cmnist = \"..\\\\chinese-MNIST\\\\data\\\\data\"\n",
    "label = \"cmnist\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load data from directory with csv file...\n",
      "15000 files found\n",
      "Load images and labels from data...\n",
      "Number of folders for one array: 3\n",
      "Load from folder 0...\n",
      "Load from folder 1...\n",
      "Load from folder 2...\n",
      "Load from folder 3...\n",
      "Load from folder 4...\n",
      "Load from folder 5...\n",
      "Load from folder 6...\n",
      "Load from folder 7...\n",
      "Load from folder 8...\n",
      "Load from folder 9...\n",
      "Load from folder 10...\n",
      "Load from folder 11...\n",
      "Load from folder 12...\n",
      "Load from folder 13...\n",
      "Load from folder 14...\n",
      "\n",
      "Number of samples: 15000 where each sample is of size: (64, 64)\n",
      "\n",
      "Elapsed time for data preparation: 00:02:02\n",
      "\n",
      "Number of classes: 15\n",
      "Data splitted in train and test sets\n",
      "Save train and test images and labels to npz files...\n",
      "Train set splitted in train and validation sets\n",
      "\n",
      "Train images dimensions: (12150, 64, 64, 1)\n",
      "\n",
      "Validation images dimensions: (1350, 64, 64, 1)\n",
      "\n",
      "Test images dimensions: (1500, 64, 64, 1)\n"
     ]
    }
   ],
   "source": [
    "# Create a new dataloader of the Chinese-MNIST dataset\n",
    "data_loader_cmnist = data_loading.DataLoader(\n",
    "    data_dir=data_dir_cmnist,\n",
    "    balancing=balancing, upsampling_size=upsampling_size,\n",
    "    load_from_npz=load_from_npz, npz_paths=npz_paths,\n",
    "    save_to_npz=save_to_npz,\n",
    "    csv_file=csv_file,\n",
    "    use_grayscale=use_grayscale,\n",
    "    label=label,\n",
    "    resize_shape=resize_shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 64, 64, 1)]       0         \n",
      "                                                                 \n",
      " sequential (Sequential)     (None, 3832)              912152    \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 1024)              3924992   \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 512)               524800    \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 256)               131328    \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 128)               32896     \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_6 (Dense)             (None, 15)                1935      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 5,528,103\n",
      "Trainable params: 4,615,951\n",
      "Non-trainable params: 912,152\n",
      "_________________________________________________________________\n",
      "Epoch 1/50\n",
      "95/95 [==============================] - 4s 34ms/step - loss: 2.8105 - accuracy: 0.1733 - val_loss: 2.6530 - val_accuracy: 0.1911\n",
      "Epoch 2/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 2.4403 - accuracy: 0.2277 - val_loss: 2.2244 - val_accuracy: 0.2993\n",
      "Epoch 3/50\n",
      "95/95 [==============================] - 3s 27ms/step - loss: 2.1973 - accuracy: 0.2855 - val_loss: 2.0124 - val_accuracy: 0.3874\n",
      "Epoch 4/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 2.0125 - accuracy: 0.3498 - val_loss: 1.8342 - val_accuracy: 0.4659\n",
      "Epoch 5/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 1.8602 - accuracy: 0.4092 - val_loss: 1.6827 - val_accuracy: 0.5089\n",
      "Epoch 6/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 1.7209 - accuracy: 0.4660 - val_loss: 1.5592 - val_accuracy: 0.5437\n",
      "Epoch 7/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 1.6079 - accuracy: 0.5122 - val_loss: 1.4942 - val_accuracy: 0.5600\n",
      "Epoch 8/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.5313 - accuracy: 0.5341 - val_loss: 1.4001 - val_accuracy: 0.5993\n",
      "Epoch 9/50\n",
      "95/95 [==============================] - 2s 22ms/step - loss: 1.4562 - accuracy: 0.5666 - val_loss: 1.3318 - val_accuracy: 0.6200\n",
      "Epoch 10/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 1.3847 - accuracy: 0.5893 - val_loss: 1.2823 - val_accuracy: 0.6252\n",
      "Epoch 11/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.3204 - accuracy: 0.6111 - val_loss: 1.2374 - val_accuracy: 0.6326\n",
      "Epoch 12/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.2708 - accuracy: 0.6269 - val_loss: 1.1890 - val_accuracy: 0.6570\n",
      "Epoch 13/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 1.2312 - accuracy: 0.6419 - val_loss: 1.1390 - val_accuracy: 0.6815\n",
      "Epoch 14/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 1.1886 - accuracy: 0.6543 - val_loss: 1.0953 - val_accuracy: 0.7030\n",
      "Epoch 15/50\n",
      "95/95 [==============================] - 2s 21ms/step - loss: 1.1597 - accuracy: 0.6665 - val_loss: 1.0720 - val_accuracy: 0.7163\n",
      "Epoch 16/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 1.1188 - accuracy: 0.6775 - val_loss: 1.0416 - val_accuracy: 0.7141\n",
      "Epoch 17/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.0985 - accuracy: 0.6889 - val_loss: 1.0305 - val_accuracy: 0.7156\n",
      "Epoch 18/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 1.0632 - accuracy: 0.6978 - val_loss: 1.0135 - val_accuracy: 0.7304\n",
      "Epoch 19/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.0460 - accuracy: 0.6994 - val_loss: 0.9816 - val_accuracy: 0.7319\n",
      "Epoch 20/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 1.0214 - accuracy: 0.7101 - val_loss: 0.9596 - val_accuracy: 0.7459\n",
      "Epoch 21/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 1.0000 - accuracy: 0.7198 - val_loss: 0.9483 - val_accuracy: 0.7356\n",
      "Epoch 22/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.9814 - accuracy: 0.7286 - val_loss: 0.9237 - val_accuracy: 0.7452\n",
      "Epoch 23/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.9648 - accuracy: 0.7317 - val_loss: 0.9118 - val_accuracy: 0.7444\n",
      "Epoch 24/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.9421 - accuracy: 0.7367 - val_loss: 0.9002 - val_accuracy: 0.7622\n",
      "Epoch 25/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 0.9282 - accuracy: 0.7472 - val_loss: 0.8943 - val_accuracy: 0.7563\n",
      "Epoch 26/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.9137 - accuracy: 0.7503 - val_loss: 0.8754 - val_accuracy: 0.7607\n",
      "Epoch 27/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.9030 - accuracy: 0.7527 - val_loss: 0.8578 - val_accuracy: 0.7733\n",
      "Epoch 28/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.8852 - accuracy: 0.7579 - val_loss: 0.8660 - val_accuracy: 0.7711\n",
      "Epoch 29/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 0.8758 - accuracy: 0.7621 - val_loss: 0.8410 - val_accuracy: 0.7696\n",
      "Epoch 30/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.8588 - accuracy: 0.7644 - val_loss: 0.8292 - val_accuracy: 0.7778\n",
      "Epoch 31/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 0.8484 - accuracy: 0.7679 - val_loss: 0.8270 - val_accuracy: 0.7793\n",
      "Epoch 32/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.8324 - accuracy: 0.7735 - val_loss: 0.8256 - val_accuracy: 0.7726\n",
      "Epoch 33/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.8298 - accuracy: 0.7731 - val_loss: 0.8076 - val_accuracy: 0.7881\n",
      "Epoch 34/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.8144 - accuracy: 0.7822 - val_loss: 0.7903 - val_accuracy: 0.7867\n",
      "Epoch 35/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.8051 - accuracy: 0.7848 - val_loss: 0.7830 - val_accuracy: 0.7859\n",
      "Epoch 36/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7996 - accuracy: 0.7853 - val_loss: 0.7768 - val_accuracy: 0.7963\n",
      "Epoch 37/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7894 - accuracy: 0.7887 - val_loss: 0.7724 - val_accuracy: 0.7941\n",
      "Epoch 38/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7834 - accuracy: 0.7908 - val_loss: 0.7630 - val_accuracy: 0.7948\n",
      "Epoch 39/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.7759 - accuracy: 0.7937 - val_loss: 0.7760 - val_accuracy: 0.7919\n",
      "Epoch 40/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.7647 - accuracy: 0.7944 - val_loss: 0.7516 - val_accuracy: 0.7911\n",
      "Epoch 41/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 0.7714 - accuracy: 0.7932 - val_loss: 0.7483 - val_accuracy: 0.7948\n",
      "Epoch 42/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7501 - accuracy: 0.8007 - val_loss: 0.7376 - val_accuracy: 0.8030\n",
      "Epoch 43/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7495 - accuracy: 0.8012 - val_loss: 0.7444 - val_accuracy: 0.7933\n",
      "Epoch 44/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7367 - accuracy: 0.8044 - val_loss: 0.7234 - val_accuracy: 0.8044\n",
      "Epoch 45/50\n",
      "95/95 [==============================] - 2s 19ms/step - loss: 0.7314 - accuracy: 0.8061 - val_loss: 0.7172 - val_accuracy: 0.8096\n",
      "Epoch 46/50\n",
      "95/95 [==============================] - 2s 20ms/step - loss: 0.7271 - accuracy: 0.8074 - val_loss: 0.7605 - val_accuracy: 0.7978\n",
      "Epoch 47/50\n",
      "95/95 [==============================] - 2s 18ms/step - loss: 0.7194 - accuracy: 0.8113 - val_loss: 0.6990 - val_accuracy: 0.8200\n",
      "Epoch 48/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95/95 [==============================] - 2s 18ms/step - loss: 0.7131 - accuracy: 0.8121 - val_loss: 0.7053 - val_accuracy: 0.8111\n",
      "Epoch 49/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7114 - accuracy: 0.8114 - val_loss: 0.6981 - val_accuracy: 0.8193\n",
      "Epoch 50/50\n",
      "95/95 [==============================] - 2s 17ms/step - loss: 0.7028 - accuracy: 0.8186 - val_loss: 0.6993 - val_accuracy: 0.8156\n",
      "\n",
      "Elapsed time for training: 00:01:31\n",
      "Plot loss and accuracy...\n",
      "Evaluate model on test data...\n",
      "12/12 [==============================] - 1s 43ms/step - loss: 0.7143 - accuracy: 0.8153\n",
      "The test loss is: 0.7143165469169617\n",
      "The test accuracy is: 0.8153333067893982\n",
      "\n",
      "\n",
      "\n",
      "12/12 [==============================] - 0s 11ms/step\n",
      "The Precision is: 0.8197436796529228\n",
      "The Recall is: 0.8159249219725215\n",
      "The F1-score is: 0.8143977073199054\n",
      "Plot some output predictions...\n",
      "2/2 [==============================] - 0s 7ms/step\n"
     ]
    }
   ],
   "source": [
    "# Use the loaded model and the dataloader for transfer learning\n",
    "transfer_learning.transfer_learning_with_local_model(model, data_loader_cmnist, batch_size, epochs, learning_rate, label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the transfer learning was successful, you should see accuracy_cmnist.png, loss_cmnist.png and predictions_cmnist.png as well as a saved model file named something like cmnist_model_Aug30-2022-11-03-33.h5 and the saved train and test images and labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transfer Learning on Chinese-MNIST Dataset using a Pretained Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To use a pretrained model like ResNet50 we have to update some parameter to get the right dataloader\n",
    "batch_size = 64\n",
    "label = \"pretrained_resnet50\"\n",
    "resize_shape = (224,224)    # ResNet50 accepts only images of size 224 x 224 so we need to resize the images\n",
    "use_grayscale = True        # ResNet50 was trained with RGB images so we have to transfer \n",
    "                            # the grayscale images of the Chinese-MNIST dataset to RGB images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load data from directory with csv file...\n",
      "15000 files found\n",
      "Load images and labels from data...\n",
      "Number of folders for one array: 3\n",
      "Load from folder 0...\n",
      "Load from folder 1...\n",
      "Load from folder 2...\n",
      "Load from folder 3...\n",
      "Load from folder 4...\n",
      "Load from folder 5...\n",
      "Load from folder 6...\n",
      "Load from folder 7...\n",
      "Load from folder 8...\n",
      "Load from folder 9...\n",
      "Load from folder 10...\n",
      "Load from folder 11...\n",
      "Load from folder 12...\n",
      "Load from folder 13...\n",
      "Load from folder 14...\n",
      "\n",
      "Number of samples: 15000 where each sample is of size: (224, 224)\n",
      "\n",
      "Elapsed time for data preparation: 00:08:11\n",
      "\n",
      "Number of classes: 15\n",
      "Data splitted in train and test sets\n",
      "Save train and test images and labels to npz files...\n",
      "Train set splitted in train and validation sets\n",
      "\n",
      "Train images dimensions: (12150, 224, 224, 3)\n",
      "\n",
      "Validation images dimensions: (1350, 224, 224, 3)\n",
      "\n",
      "Test images dimensions: (1500, 224, 224, 3)\n"
     ]
    }
   ],
   "source": [
    "# Create a new dataloader of the Chinese-MNIST dataset for the pretrained model\n",
    "data_loader_cmnist_pretrained = data_loading.DataLoader(\n",
    "    data_dir=data_dir_cmnist,\n",
    "    balancing=balancing, upsampling_size=upsampling_size,\n",
    "    load_from_npz=load_from_npz, npz_paths=npz_paths,\n",
    "    save_to_npz=save_to_npz,\n",
    "    csv_file=csv_file,\n",
    "    use_grayscale=use_grayscale,\n",
    "    label=label,\n",
    "    resize_shape=resize_shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_2 (InputLayer)           [(None, 224, 224, 3  0           []                               \n",
      "                                )]                                                                \n",
      "                                                                                                  \n",
      " conv1_pad (ZeroPadding2D)      (None, 230, 230, 3)  0           ['input_2[0][0]']                \n",
      "                                                                                                  \n",
      " conv1_conv (Conv2D)            (None, 112, 112, 64  9472        ['conv1_pad[0][0]']              \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv1_bn (BatchNormalization)  (None, 112, 112, 64  256         ['conv1_conv[0][0]']             \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv1_relu (Activation)        (None, 112, 112, 64  0           ['conv1_bn[0][0]']               \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " pool1_pad (ZeroPadding2D)      (None, 114, 114, 64  0           ['conv1_relu[0][0]']             \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " pool1_pool (MaxPooling2D)      (None, 56, 56, 64)   0           ['pool1_pad[0][0]']              \n",
      "                                                                                                  \n",
      " conv2_block1_1_conv (Conv2D)   (None, 56, 56, 64)   4160        ['pool1_pool[0][0]']             \n",
      "                                                                                                  \n",
      " conv2_block1_1_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_1_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block1_2_conv (Conv2D)   (None, 56, 56, 64)   36928       ['conv2_block1_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block1_2_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_2_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block1_0_conv (Conv2D)   (None, 56, 56, 256)  16640       ['pool1_pool[0][0]']             \n",
      "                                                                                                  \n",
      " conv2_block1_3_conv (Conv2D)   (None, 56, 56, 256)  16640       ['conv2_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block1_0_bn (BatchNormal  (None, 56, 56, 256)  1024       ['conv2_block1_0_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_3_bn (BatchNormal  (None, 56, 56, 256)  1024       ['conv2_block1_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_add (Add)         (None, 56, 56, 256)  0           ['conv2_block1_0_bn[0][0]',      \n",
      "                                                                  'conv2_block1_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv2_block1_out (Activation)  (None, 56, 56, 256)  0           ['conv2_block1_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv2_block2_1_conv (Conv2D)   (None, 56, 56, 64)   16448       ['conv2_block1_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv2_block2_1_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block2_1_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block2_2_conv (Conv2D)   (None, 56, 56, 64)   36928       ['conv2_block2_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block2_2_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block2_2_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block2_3_conv (Conv2D)   (None, 56, 56, 256)  16640       ['conv2_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block2_3_bn (BatchNormal  (None, 56, 56, 256)  1024       ['conv2_block2_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block2_add (Add)         (None, 56, 56, 256)  0           ['conv2_block1_out[0][0]',       \n",
      "                                                                  'conv2_block2_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv2_block2_out (Activation)  (None, 56, 56, 256)  0           ['conv2_block2_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv2_block3_1_conv (Conv2D)   (None, 56, 56, 64)   16448       ['conv2_block2_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv2_block3_1_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv2_block3_1_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block3_2_conv (Conv2D)   (None, 56, 56, 64)   36928       ['conv2_block3_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block3_2_bn (BatchNormal  (None, 56, 56, 64)  256         ['conv2_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block3_2_relu (Activatio  (None, 56, 56, 64)  0           ['conv2_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block3_3_conv (Conv2D)   (None, 56, 56, 256)  16640       ['conv2_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block3_3_bn (BatchNormal  (None, 56, 56, 256)  1024       ['conv2_block3_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block3_add (Add)         (None, 56, 56, 256)  0           ['conv2_block2_out[0][0]',       \n",
      "                                                                  'conv2_block3_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv2_block3_out (Activation)  (None, 56, 56, 256)  0           ['conv2_block3_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block1_1_conv (Conv2D)   (None, 28, 28, 128)  32896       ['conv2_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block1_1_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_1_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block1_2_conv (Conv2D)   (None, 28, 28, 128)  147584      ['conv3_block1_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block1_2_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_2_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block1_0_conv (Conv2D)   (None, 28, 28, 512)  131584      ['conv2_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block1_3_conv (Conv2D)   (None, 28, 28, 512)  66048       ['conv3_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block1_0_bn (BatchNormal  (None, 28, 28, 512)  2048       ['conv3_block1_0_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_3_bn (BatchNormal  (None, 28, 28, 512)  2048       ['conv3_block1_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_add (Add)         (None, 28, 28, 512)  0           ['conv3_block1_0_bn[0][0]',      \n",
      "                                                                  'conv3_block1_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv3_block1_out (Activation)  (None, 28, 28, 512)  0           ['conv3_block1_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block2_1_conv (Conv2D)   (None, 28, 28, 128)  65664       ['conv3_block1_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block2_1_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block2_1_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block2_2_conv (Conv2D)   (None, 28, 28, 128)  147584      ['conv3_block2_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block2_2_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block2_2_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block2_3_conv (Conv2D)   (None, 28, 28, 512)  66048       ['conv3_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block2_3_bn (BatchNormal  (None, 28, 28, 512)  2048       ['conv3_block2_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block2_add (Add)         (None, 28, 28, 512)  0           ['conv3_block1_out[0][0]',       \n",
      "                                                                  'conv3_block2_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv3_block2_out (Activation)  (None, 28, 28, 512)  0           ['conv3_block2_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block3_1_conv (Conv2D)   (None, 28, 28, 128)  65664       ['conv3_block2_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block3_1_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv3_block3_1_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block3_2_conv (Conv2D)   (None, 28, 28, 128)  147584      ['conv3_block3_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block3_2_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block3_2_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block3_3_conv (Conv2D)   (None, 28, 28, 512)  66048       ['conv3_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block3_3_bn (BatchNormal  (None, 28, 28, 512)  2048       ['conv3_block3_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block3_add (Add)         (None, 28, 28, 512)  0           ['conv3_block2_out[0][0]',       \n",
      "                                                                  'conv3_block3_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv3_block3_out (Activation)  (None, 28, 28, 512)  0           ['conv3_block3_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block4_1_conv (Conv2D)   (None, 28, 28, 128)  65664       ['conv3_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block4_1_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block4_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block4_1_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block4_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block4_2_conv (Conv2D)   (None, 28, 28, 128)  147584      ['conv3_block4_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block4_2_bn (BatchNormal  (None, 28, 28, 128)  512        ['conv3_block4_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block4_2_relu (Activatio  (None, 28, 28, 128)  0          ['conv3_block4_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block4_3_conv (Conv2D)   (None, 28, 28, 512)  66048       ['conv3_block4_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block4_3_bn (BatchNormal  (None, 28, 28, 512)  2048       ['conv3_block4_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block4_add (Add)         (None, 28, 28, 512)  0           ['conv3_block3_out[0][0]',       \n",
      "                                                                  'conv3_block4_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv3_block4_out (Activation)  (None, 28, 28, 512)  0           ['conv3_block4_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block1_1_conv (Conv2D)   (None, 14, 14, 256)  131328      ['conv3_block4_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block1_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block1_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block1_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block1_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block1_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block1_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block1_0_conv (Conv2D)   (None, 14, 14, 1024  525312      ['conv3_block4_out[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block1_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block1_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block1_0_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block1_0_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block1_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block1_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block1_add (Add)         (None, 14, 14, 1024  0           ['conv4_block1_0_bn[0][0]',      \n",
      "                                )                                 'conv4_block1_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block1_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block1_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block2_1_conv (Conv2D)   (None, 14, 14, 256)  262400      ['conv4_block1_out[0][0]']       \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv4_block2_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block2_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block2_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block2_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block2_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block2_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block2_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block2_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block2_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block2_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block2_add (Add)         (None, 14, 14, 1024  0           ['conv4_block1_out[0][0]',       \n",
      "                                )                                 'conv4_block2_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block2_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block2_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block3_1_conv (Conv2D)   (None, 14, 14, 256)  262400      ['conv4_block2_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block3_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block3_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block3_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block3_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block3_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block3_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block3_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block3_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block3_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block3_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block3_add (Add)         (None, 14, 14, 1024  0           ['conv4_block2_out[0][0]',       \n",
      "                                )                                 'conv4_block3_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block3_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block3_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block4_1_conv (Conv2D)   (None, 14, 14, 256)  262400      ['conv4_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block4_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block4_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block4_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block4_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block4_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block4_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block4_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block4_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block4_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block4_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block4_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block4_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block4_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block4_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block4_add (Add)         (None, 14, 14, 1024  0           ['conv4_block3_out[0][0]',       \n",
      "                                )                                 'conv4_block4_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block4_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block4_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " conv4_block5_1_conv (Conv2D)   (None, 14, 14, 256)  262400      ['conv4_block4_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block5_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block5_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block5_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block5_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block5_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block5_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block5_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block5_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block5_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block5_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block5_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block5_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block5_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block5_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block5_add (Add)         (None, 14, 14, 1024  0           ['conv4_block4_out[0][0]',       \n",
      "                                )                                 'conv4_block5_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block5_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block5_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block6_1_conv (Conv2D)   (None, 14, 14, 256)  262400      ['conv4_block5_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block6_1_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block6_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block6_1_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block6_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block6_2_conv (Conv2D)   (None, 14, 14, 256)  590080      ['conv4_block6_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block6_2_bn (BatchNormal  (None, 14, 14, 256)  1024       ['conv4_block6_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block6_2_relu (Activatio  (None, 14, 14, 256)  0          ['conv4_block6_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block6_3_conv (Conv2D)   (None, 14, 14, 1024  263168      ['conv4_block6_2_relu[0][0]']    \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block6_3_bn (BatchNormal  (None, 14, 14, 1024  4096       ['conv4_block6_3_conv[0][0]']    \n",
      " ization)                       )                                                                 \n",
      "                                                                                                  \n",
      " conv4_block6_add (Add)         (None, 14, 14, 1024  0           ['conv4_block5_out[0][0]',       \n",
      "                                )                                 'conv4_block6_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv4_block6_out (Activation)  (None, 14, 14, 1024  0           ['conv4_block6_add[0][0]']       \n",
      "                                )                                                                 \n",
      "                                                                                                  \n",
      " conv5_block1_1_conv (Conv2D)   (None, 7, 7, 512)    524800      ['conv4_block6_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block1_1_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_1_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block1_2_conv (Conv2D)   (None, 7, 7, 512)    2359808     ['conv5_block1_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block1_2_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_2_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block1_0_conv (Conv2D)   (None, 7, 7, 2048)   2099200     ['conv4_block6_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block1_3_conv (Conv2D)   (None, 7, 7, 2048)   1050624     ['conv5_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block1_0_bn (BatchNormal  (None, 7, 7, 2048)  8192        ['conv5_block1_0_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_3_bn (BatchNormal  (None, 7, 7, 2048)  8192        ['conv5_block1_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_add (Add)         (None, 7, 7, 2048)   0           ['conv5_block1_0_bn[0][0]',      \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                  'conv5_block1_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv5_block1_out (Activation)  (None, 7, 7, 2048)   0           ['conv5_block1_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block2_1_conv (Conv2D)   (None, 7, 7, 512)    1049088     ['conv5_block1_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block2_1_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block2_1_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block2_2_conv (Conv2D)   (None, 7, 7, 512)    2359808     ['conv5_block2_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block2_2_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block2_2_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block2_3_conv (Conv2D)   (None, 7, 7, 2048)   1050624     ['conv5_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block2_3_bn (BatchNormal  (None, 7, 7, 2048)  8192        ['conv5_block2_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block2_add (Add)         (None, 7, 7, 2048)   0           ['conv5_block1_out[0][0]',       \n",
      "                                                                  'conv5_block2_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv5_block2_out (Activation)  (None, 7, 7, 2048)   0           ['conv5_block2_add[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block3_1_conv (Conv2D)   (None, 7, 7, 512)    1049088     ['conv5_block2_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv5_block3_1_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block3_1_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block3_2_conv (Conv2D)   (None, 7, 7, 512)    2359808     ['conv5_block3_1_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block3_2_bn (BatchNormal  (None, 7, 7, 512)   2048        ['conv5_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block3_2_relu (Activatio  (None, 7, 7, 512)   0           ['conv5_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block3_3_conv (Conv2D)   (None, 7, 7, 2048)   1050624     ['conv5_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block3_3_bn (BatchNormal  (None, 7, 7, 2048)  8192        ['conv5_block3_3_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block3_add (Add)         (None, 7, 7, 2048)   0           ['conv5_block2_out[0][0]',       \n",
      "                                                                  'conv5_block3_3_bn[0][0]']      \n",
      "                                                                                                  \n",
      " conv5_block3_out (Activation)  (None, 7, 7, 2048)   0           ['conv5_block3_add[0][0]']       \n",
      "                                                                                                  \n",
      " global_average_pooling2d (Glob  (None, 2048)        0           ['conv5_block3_out[0][0]']       \n",
      " alAveragePooling2D)                                                                              \n",
      "                                                                                                  \n",
      " dense_7 (Dense)                (None, 1024)         2098176     ['global_average_pooling2d[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " dense_8 (Dense)                (None, 15)           15375       ['dense_7[0][0]']                \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 25,701,263\n",
      "Trainable params: 25,648,143\n",
      "Non-trainable params: 53,120\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "ename": "InternalError",
     "evalue": "Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /job:localhost/replica:0/task:0/device:GPU:0 in order to run _EagerConst: Dst tensor is not initialized.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mInternalError\u001b[0m                             Traceback (most recent call last)",
      "Input \u001b[1;32mIn [35]\u001b[0m, in \u001b[0;36m<cell line: 2>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Use the pretrained model ResNet50 and the dataloader for transfer learning\u001b[39;00m\n\u001b[1;32m----> 2\u001b[0m \u001b[43mtransfer_learning\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtransfer_learning_with_pretrained_model\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mresnet50\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata_loader_cmnist_pretrained\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mepochs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_rate\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mD:\\Users\\Alexander\\Desktop\\DL\\Neuer Ordner\\transfer_learning.py:114\u001b[0m, in \u001b[0;36mtransfer_learning_with_pretrained_model\u001b[1;34m(model_name, data_loader, batch_size, epochs, learning_rate)\u001b[0m\n\u001b[0;32m    108\u001b[0m model\u001b[38;5;241m.\u001b[39mcompile(\n\u001b[0;32m    109\u001b[0m     optimizer\u001b[38;5;241m=\u001b[39mAdam(learning_rate),\n\u001b[0;32m    110\u001b[0m     loss\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcategorical_crossentropy\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[0;32m    111\u001b[0m     metrics\u001b[38;5;241m=\u001b[39m[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124maccuracy\u001b[39m\u001b[38;5;124m\"\u001b[39m])\n\u001b[0;32m    113\u001b[0m \u001b[38;5;66;03m# Train the model on the new data\u001b[39;00m\n\u001b[1;32m--> 114\u001b[0m history \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    115\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtrain_images\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrain_labels\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    116\u001b[0m \u001b[43m    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mepochs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    117\u001b[0m \u001b[43m    \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mvalidation_images\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mvalidation_labels\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    118\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbatch_size\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    120\u001b[0m train\u001b[38;5;241m.\u001b[39msave_model(model, save_name\u001b[38;5;241m=\u001b[39mmodel_name)\n\u001b[0;32m    121\u001b[0m test\u001b[38;5;241m.\u001b[39mplot_accuracy_and_loss(history, model_name)\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python39\\site-packages\\keras\\utils\\traceback_utils.py:67\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     65\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:  \u001b[38;5;66;03m# pylint: disable=broad-except\u001b[39;00m\n\u001b[0;32m     66\u001b[0m   filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n\u001b[1;32m---> 67\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m e\u001b[38;5;241m.\u001b[39mwith_traceback(filtered_tb) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;28mNone\u001b[39m\n\u001b[0;32m     68\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[0;32m     69\u001b[0m   \u001b[38;5;28;01mdel\u001b[39;00m filtered_tb\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python39\\site-packages\\tensorflow\\python\\framework\\constant_op.py:102\u001b[0m, in \u001b[0;36mconvert_to_eager_tensor\u001b[1;34m(value, ctx, dtype)\u001b[0m\n\u001b[0;32m    100\u001b[0m     dtype \u001b[38;5;241m=\u001b[39m dtypes\u001b[38;5;241m.\u001b[39mas_dtype(dtype)\u001b[38;5;241m.\u001b[39mas_datatype_enum\n\u001b[0;32m    101\u001b[0m ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[1;32m--> 102\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mops\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mEagerTensor\u001b[49m\u001b[43m(\u001b[49m\u001b[43mvalue\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdtype\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[1;31mInternalError\u001b[0m: Failed copying input tensor from /job:localhost/replica:0/task:0/device:CPU:0 to /job:localhost/replica:0/task:0/device:GPU:0 in order to run _EagerConst: Dst tensor is not initialized."
     ]
    }
   ],
   "source": [
    "# Use the pretrained model ResNet50 and the dataloader for transfer learning\n",
    "transfer_learning.transfer_learning_with_pretrained_model('resnet50', data_loader_cmnist_pretrained, batch_size, epochs, learning_rate)\n",
    "\n",
    "### An error occurs here because it runs out of memory. \n",
    "### Our PCs have not enough RAM when we create this demo file. \n",
    "### But it should work on a PC with better hardware."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the transfer learning was successful, you should see accuracy_pretrained_resnet50.png, loss_pretrained_resnet50.png and predictions_pretrained_resnet50.png as well as a saved model file named something like pretrained_resnet50_model_Aug30-2022-12-10-23.h5 and the saved train and test images and labels."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyPJHBRT0cN3ryQqEqGw4dVA",
   "name": "Untitled0.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.9.0"
  },
  "vscode": {
   "interpreter": {
    "hash": "3107ee7e48c87bc0659fdfa1e5e48f81abb78cf8d4b3aa059a62b014212b10ba"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}