Lesson108-应用Transfer Learning和Vgg 16识别厨房刀具.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 第108讲 应用Transfer Learning和Vgg 16识别厨房刀具\n",
    "\n",
    "\n",
    "## Python学习：https://www.ixigua.com/home/77346806707?utm_source=xiguastudio\n",
    "\n",
    "## Python源文件及数据下载链接: https://github.com/rayc2020/LessonPythonCode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#https://www.kaggle.com/c/dogs-vs-cats/data\n",
    "#http://kt.agh.edu.pl/~matiolanski/KnivesImagesDatabase/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "VGG16 is a convolutional neural network model proposed by K. Simonyan and A. Zisserman from the University of Oxford in the paper “Very Deep Convolutional Networks for Large-Scale Image Recognition”. The model achieves 92.7% top-5 test accuracy in ImageNet, which is a dataset of over 14 million images belonging to 1000 classes. It was one of the famous model submitted to ILSVRC-2014. It makes the improvement over AlexNet by replacing large kernel-sized filters (11 and 5 in the first and second convolutional layer, respectively) with multiple 3×3 kernel-sized filters one after another. VGG16 was trained for weeks and was using NVIDIA Titan Black GPU’s."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "VGG16 (also called OxfordNet) is a convolutional neural network architecture named after the Visual Geometry Group from Oxford, who developed it. It was used to win the ILSVR (ImageNet) competition in 2014"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Transfer learning generally refers to a process where a model trained on one problem is used in some way on a second related problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.3.0'"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "tf.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image(filename='D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson69-vgg16-1.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Input, Dense, Flatten\n",
    "from keras.models import Model\n",
    "from keras.applications.vgg16 import VGG16\n",
    "from keras.applications.vgg16 import preprocess_input\n",
    "from keras.preprocessing import image\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.models import Sequential\n",
    "import numpy as np\n",
    "import glob\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from os.path import isfile, isdir, getsize\n",
    "import shutil\n",
    "from os import mkdir, makedirs, remove, listdir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "datasetfolderpath = r\"D:\\python\\Project0-Python-MachineLearning\\Lesson108-knife\\KnivesImagesDatabase\"\n",
    "files = glob.glob(datasetfolderpath + \"/**/*.bmp\")\n",
    "\n",
    "labels = np.array([0]*9340 + [1]*3559)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10000.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10001.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10002.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10003.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10004.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10005.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10006.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10007.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10008.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10009.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10010.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10011.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10012.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10013.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10014.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10015.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10016.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10017.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10018.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10019.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10020.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10021.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10022.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10023.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10024.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10025.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10026.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10027.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10028.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10029.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10030.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10031.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10032.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10033.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10034.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10035.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10036.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10037.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10038.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10039.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10040.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10041.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10042.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10043.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10044.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10045.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10046.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10047.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10048.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10049.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10050.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10051.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10052.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10053.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10054.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10055.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10056.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10057.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10058.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10059.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10060.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10061.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10062.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10063.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10064.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10065.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10066.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10067.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10068.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10069.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10070.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10071.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10072.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10073.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10074.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10075.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10076.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10077.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10078.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10079.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10080.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10081.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10082.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10083.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10084.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10085.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10086.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10087.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10088.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10089.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10090.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10091.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10092.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10093.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10094.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10095.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10096.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10097.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10098.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10099.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10100.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10101.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10102.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10103.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10104.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10105.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10106.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10107.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10108.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10109.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10110.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10111.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10112.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10113.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10114.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10115.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10116.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10117.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10118.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10119.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10120.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10121.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10122.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10123.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10124.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10125.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10126.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10127.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10128.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10129.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10130.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10131.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10132.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10133.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10134.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10135.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10136.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10137.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10138.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10139.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10140.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10141.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10142.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10143.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10144.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10145.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10146.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10147.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10148.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10149.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10150.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10151.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10152.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10153.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10154.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10155.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10156.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10157.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10158.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10159.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10160.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10161.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10162.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10163.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10164.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10165.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10166.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10167.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10168.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10169.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10170.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10171.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10172.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10173.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10174.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10175.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10176.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10177.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10178.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10179.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10180.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10181.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10182.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10183.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10184.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10185.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10186.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10187.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10188.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10189.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10190.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10191.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10192.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10193.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10194.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10195.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10196.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10197.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10198.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10199.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10200.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10201.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10202.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10203.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10204.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10205.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10206.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10207.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10208.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10209.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10210.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10211.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10212.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10213.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10214.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10215.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10216.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10217.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10218.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10219.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10220.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10221.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10222.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10223.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10224.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10225.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10226.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10227.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10228.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10229.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10230.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10231.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10232.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10233.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10234.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10235.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10236.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10237.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10238.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10239.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10240.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10241.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10242.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10243.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10244.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10245.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10246.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10247.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10248.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10249.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10250.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10251.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10252.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10253.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10254.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10255.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10256.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10257.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10258.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10259.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10260.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10261.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10262.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10263.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10264.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10265.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10266.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10267.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10268.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10269.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10270.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10271.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10272.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10273.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10274.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10275.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10276.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10277.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10278.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10279.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10280.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10281.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10282.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10283.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10284.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10285.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10286.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10287.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10288.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10289.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10290.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10291.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10292.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10293.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10294.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10295.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10296.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10297.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10298.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10299.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10300.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10301.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10302.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10303.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10304.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10305.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10306.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10307.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10308.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10309.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10310.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10311.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10312.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10313.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10314.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10315.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10316.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10317.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10318.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10319.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10320.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10321.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10322.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10323.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10324.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10325.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10326.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10327.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10328.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10329.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10330.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10331.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10332.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10333.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10334.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10335.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10336.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10337.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10338.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10339.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10340.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10341.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10342.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10343.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10344.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10345.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10346.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10347.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10348.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10349.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10350.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10351.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10352.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10353.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10354.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10355.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10356.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10357.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10358.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10359.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10360.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10361.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10362.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10363.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10364.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10365.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10366.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10367.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10368.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10369.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10370.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10371.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10372.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10373.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10374.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10375.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10376.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10377.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10378.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10379.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10380.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10381.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10382.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10383.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10384.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10385.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10386.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10387.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10388.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10389.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10390.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10391.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10392.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10393.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10394.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10395.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10396.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10397.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10398.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10399.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10400.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10401.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10402.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10403.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10404.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10405.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10406.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10407.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10408.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10409.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10410.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10411.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10412.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10413.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10414.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10415.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10416.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10417.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10418.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10419.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10420.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10421.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10422.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10423.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10424.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10425.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10426.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10427.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10428.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10429.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10430.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10431.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10432.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10433.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10434.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10435.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10436.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10437.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10438.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10439.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10440.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10441.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10442.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10443.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10444.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10445.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10446.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10447.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10448.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10449.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10450.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10451.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10452.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10453.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10454.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10455.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10456.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10457.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10458.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10459.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10460.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10461.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10462.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10463.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10464.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10465.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10466.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10467.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10468.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10469.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10470.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10471.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10472.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10473.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10474.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10475.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10476.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10477.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10478.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10479.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10480.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10481.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10482.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10483.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10484.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10485.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10486.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10487.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10488.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10489.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10490.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10491.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10492.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10493.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10494.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10495.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10496.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10497.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10498.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10499.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10500.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10501.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10502.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10503.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10504.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10505.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10506.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10507.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10508.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10509.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10510.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10511.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10512.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10513.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10514.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10515.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10516.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10517.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10518.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10519.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10520.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10521.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10522.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10523.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10524.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10525.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10526.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10527.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10528.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10529.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10530.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10531.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10532.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10533.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10534.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10535.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10536.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10537.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10538.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10539.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10540.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10541.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10542.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10543.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10544.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10545.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10546.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10547.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10548.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10549.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10550.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10551.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10552.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10553.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10554.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10555.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10556.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10557.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10558.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10559.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10560.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10561.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10562.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10563.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10564.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10565.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10566.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10567.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10568.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10569.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10570.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10571.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10572.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10573.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10574.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10575.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10576.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10577.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10578.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10579.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10580.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10581.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10582.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10583.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10584.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10585.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10586.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10587.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10588.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10589.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10590.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10591.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10592.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10593.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10594.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10595.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10596.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10597.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10598.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10599.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10600.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10601.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10602.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10603.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10604.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10605.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10606.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10607.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10608.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10609.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10610.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10611.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10612.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10613.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10614.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10615.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10616.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10617.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10618.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10619.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10620.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10621.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10622.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10623.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10624.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10625.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10626.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10627.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10628.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10629.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10630.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10631.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10632.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10633.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10634.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10635.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10636.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10637.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10638.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10639.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10640.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10641.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10642.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10643.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10644.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10645.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10646.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10647.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10648.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10649.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10650.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10651.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10652.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10653.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10654.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10655.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10656.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10657.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10658.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10659.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10660.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10661.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10662.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10663.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10664.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10665.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10666.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10667.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10668.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10669.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10670.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10671.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10672.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10673.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10674.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10675.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10676.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10677.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10678.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10679.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10680.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10681.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10682.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10683.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10684.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10685.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10686.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10687.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10688.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10689.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10690.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10691.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10692.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10693.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10694.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10695.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10696.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10697.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10698.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10699.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10700.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10701.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10702.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10703.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10704.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10705.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10706.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10707.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10708.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10709.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10710.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10711.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10712.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10713.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10714.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10715.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10716.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10717.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10718.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10719.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10720.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10721.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10722.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10723.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10724.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10725.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10726.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10727.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10728.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10729.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10730.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10731.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10732.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10733.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10734.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10735.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10736.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10737.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10738.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10739.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10740.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10741.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10742.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10743.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10744.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10745.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10746.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10747.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10748.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10749.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10750.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10751.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10752.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10753.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10754.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10755.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10756.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10757.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10758.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10759.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10760.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10761.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10762.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10763.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10764.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10765.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10766.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10767.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10768.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10769.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10770.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10771.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10772.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10773.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10774.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10775.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10776.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10777.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10778.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10779.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10780.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10781.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10782.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10783.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10784.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10785.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10786.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10787.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10788.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10789.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10790.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10791.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10792.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10793.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10794.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10795.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10796.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10797.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10798.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10799.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10800.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10801.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10802.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10803.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10804.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10805.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10806.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10807.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10808.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10809.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10810.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10811.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10812.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10813.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10814.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10815.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10816.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10817.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10818.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10819.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10820.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10821.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10822.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10823.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10824.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10825.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10826.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10827.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10828.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10829.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10830.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10831.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10832.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10833.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10834.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10835.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10836.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10837.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10838.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10839.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10840.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10841.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10842.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10843.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10844.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10845.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10846.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10847.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10848.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10849.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10850.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10851.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10852.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10853.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10854.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10855.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10856.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10857.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10858.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10859.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10860.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10861.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10862.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10863.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10864.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10865.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10866.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10867.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10868.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10869.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10870.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10871.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10872.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10873.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10874.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10875.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10876.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10877.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10878.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10879.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10880.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10881.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10882.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10883.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10884.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10885.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10886.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10887.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10888.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10889.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10890.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10891.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10892.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10893.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10894.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10895.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10896.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10897.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10898.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10899.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10900.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10901.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10902.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10903.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10904.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10905.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10906.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10907.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10908.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10909.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10910.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10911.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10912.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10913.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10914.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10915.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10916.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10917.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10918.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10919.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10920.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10921.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10922.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10923.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10924.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10925.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10926.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10927.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10928.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10929.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10930.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10931.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10932.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10933.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10934.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10935.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10936.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10937.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10938.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10939.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10940.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10941.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10942.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10943.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10944.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10945.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10946.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10947.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10948.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10949.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10950.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10951.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10952.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10953.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10954.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10955.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10956.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10957.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10958.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10959.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10960.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10961.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10962.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10963.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10964.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10965.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10966.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10967.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10968.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10969.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10970.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10971.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10972.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10973.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10974.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10975.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10976.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10977.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10978.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10979.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10980.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10981.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10982.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10983.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10984.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10985.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10986.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10987.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10988.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10989.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10990.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10991.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10992.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10993.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10994.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10995.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10996.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10997.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10998.bmp',\n",
       " 'D:\\\\python\\\\Project0-Python-MachineLearning\\\\Lesson108-knife\\\\KnivesImagesDatabase\\\\NEGATIVES_ALL\\\\10999.bmp',\n",
       " ...]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 1, 1, 1])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12899"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12899"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "len_data = len(files)\n",
    "train_examples = 8000\n",
    "test_examples = len_data - train_examples\n",
    "\n",
    "permutation = np.random.permutation(len_data)\n",
    "train_set = [files[i] for i in permutation[:train_examples]]\n",
    "test_set = [files[i] for i in permutation[-test_examples:]]\n",
    "train_labels = labels[permutation[:train_examples]]\n",
    "test_labels = labels[permutation[-test_examples:]]\n",
    "\n",
    "train_folder = datasetfolderpath + '/train'\n",
    "test_folder = datasetfolderpath + '/test'\n",
    "\n",
    "if isdir(train_folder):\n",
    "    shutil.rmtree(train_folder)\n",
    "if isdir(test_folder):\n",
    "    shutil.rmtree(test_folder)\n",
    "    \n",
    "makedirs(train_folder + '/positive/')\n",
    "makedirs(train_folder + '/negative/')\n",
    "makedirs(test_folder + '/positive/')\n",
    "makedirs(test_folder + '/negative/')\n",
    "\n",
    "for f,i in zip(train_set, train_labels):\n",
    "    if i == 1: \n",
    "        shutil.copy2(f, train_folder+'/positive/')\n",
    "    else:\n",
    "        shutil.copy2(f, train_folder+'/negative/')\n",
    "\n",
    "for f,i in zip(test_set, test_labels):\n",
    "    if i == 1: \n",
    "        shutil.copy2(f, test_folder+'/positive/')\n",
    "    else:\n",
    "        shutil.copy2(f, test_folder+'/negative/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# re-size all the images to this\n",
    "IMAGE_SIZE = [100, 100]\n",
    "\n",
    "#Keras generator alway looks for subfolders (representing the classes). Images insight the subfolders are associated with a class\n",
    "\n",
    "train_path = 'D:/python/Project0-Python-MachineLearning/Lesson108-knife/KnivesImagesDatabase/train'\n",
    "valid_path = 'D:/python/Project0-Python-MachineLearning/Lesson108-knife/KnivesImagesDatabase/test'\n",
    "\n",
    "# add preprocessing layer to the front of VGG\n",
    "vgg = VGG16(input_shape=IMAGE_SIZE + [3], weights='imagenet', include_top=False)\n",
    "\n",
    "# don't train existing weights\n",
    "for layer in vgg.layers:\n",
    "  layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_3\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 100, 100, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 100, 100, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 100, 100, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 50, 50, 64)        0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 50, 50, 128)       73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 50, 50, 128)       147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 25, 25, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 25, 25, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 25, 25, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 25, 25, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 12, 12, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 12, 12, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 12, 12, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 12, 12, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 6, 6, 512)         0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 6, 6, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 6, 6, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 6, 6, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 3, 3, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 4608)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 4609      \n",
      "=================================================================\n",
      "Total params: 14,719,297\n",
      "Trainable params: 4,609\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# useful for getting number of classes\n",
    "folders = glob.glob('D:/python/Project0-Python-MachineLearning/Lesson108-knife/KnivesImagesDatabase/train')\n",
    "  \n",
    "# our layers - you can add more if you want\n",
    "x = Flatten()(vgg.output)\n",
    "prediction = Dense(len(folders), activation='sigmoid')(x)\n",
    "\n",
    "# create a model object\n",
    "model = Model(inputs=vgg.input, outputs=prediction)\n",
    "\n",
    "# view the structure of the model\n",
    "model.summary()\n",
    "\n",
    "# tell the model what cost and optimization method to use\n",
    "model.compile(\n",
    "  loss='categorical_crossentropy',\n",
    "  optimizer='adam',\n",
    "  metrics=['accuracy']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 8000 images belonging to 2 classes.\n",
      "Found 4899 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "train_datagen = ImageDataGenerator(rescale = 1./255,\n",
    "                                   shear_range = 0.2,\n",
    "                                   zoom_range = 0.2,\n",
    "                                   horizontal_flip = True)\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale = 1./255)\n",
    "\n",
    "#Keras generator alway looks for subfolders (representing the classes). Images insight the subfolders are associated with a class\n",
    "\n",
    "training_set = train_datagen.flow_from_directory('D:/python/Project0-Python-MachineLearning/Lesson108-knife/KnivesImagesDatabase/train',\n",
    "                                                 target_size = (100, 100),\n",
    "                                                 batch_size = 32,\n",
    "                                                 class_mode = 'binary')\n",
    "\n",
    "test_set = test_datagen.flow_from_directory('D:/python/Project0-Python-MachineLearning/Lesson108-knife/KnivesImagesDatabase/test',\n",
    "                                            target_size = (100, 100),\n",
    "                                            batch_size = 32,\n",
    "                                            class_mode = 'binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "250"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(training_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "154"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "250/250 [==============================] - 3944s 16s/step - loss: 3.2976e-08 - accuracy: 0.5900 - val_loss: 3.2753e-08 - val_accuracy: 0.5805\n"
     ]
    }
   ],
   "source": [
    "# fit the model\n",
    "r = model.fit_generator(\n",
    "  training_set,\n",
    "  validation_data=test_set,\n",
    "  epochs=1,\n",
    "  steps_per_epoch=len(training_set),\n",
    "  validation_steps=len(test_set)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'negative': 0, 'positive': 1}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_set.class_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'negative': 0, 'positive': 1}"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set.class_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.9999678]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<PIL.Image.Image image mode=RGB size=100x100 at 0x1B582F75048>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img = image.load_img('D:/python/Project0-Python-MachineLearning/Lesson108-knife/IMG_7920.jpg',target_size=(100,100))\n",
    "x = image.img_to_array(img)\n",
    "x = np.expand_dims(x,axis=0)\n",
    "img_data = preprocess_input(x)\n",
    "classes = model.predict(img_data)\n",
    "\n",
    "print(classes)\n",
    "img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.9945714]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGQAAABkCAIAAAD/gAIDAAABVmlDQ1BJQ0MgUHJvZmlsZQAAeJxjYGBSSSwoyGFhYGDIzSspCnJ3UoiIjFJgf8jADoS8DGIMConJxQWOAQE+QCUMMBoVfLvGwAiiL+uCzDolNbVJtV7A12Km8NWLr0SbMNWjAK6U1OJkIP0HiFOTC4pKGBgYU4Bs5fKSAhC7A8gWKQI6CsieA2KnQ9gbQOwkCPsIWE1IkDOQfQPIVkjOSASawfgDyNZJQhJPR2JD7QUBbpfM4oKcxEqFAGMCriUDlKRWlIBo5/yCyqLM9IwSBUdgKKUqeOYl6+koGBkYmjMwgMIcovpzIDgsGcXOIMSa7zMw2O7/////boSY134Gho1AnVw7EWIaFgwMgtwMDCd2FiQWJYKFmIGYKS2NgeHTcgYG3kgGBuELQD3RxWnGRmB5Rh4nBgbWe///f1ZjYGCfzMDwd8L//78X/f//dzFQ8x0GhgN5ABUhZe5sUv9jAAAITGVYSWZNTQAqAAAACAANAQ8AAgAAAAYAAACqARAAAgAAAAkAAACwARIAAwAAAAEABgAAARoABQAAAAEAAAC6ARsABQAAAAEAAADCASgAAwAAAAEAAgAAATEAAgAAAAcAAADKATIAAgAAABQAAADSATwAAgAAAAkAAADmAUIABAAAAAEAAAIAAUMABAAAAAEAAAIAAhMAAwAAAAEAAQAAh2kABAAAAAEAAADwAAAAAEFwcGxlAGlQaG9uZSBYAAAAAABIAAAAAQAAAEgAAAABMTQuNC4yAAAyMDIxOjA0OjIwIDE0OjQwOjQ3AGlQaG9uZSBYAAAAI4KaAAUAAAABAAACmoKdAAUAAAABAAACoogiAAMAAAABAAIAAIgnAAMAAAABACAAAJAAAAcAAAAEMDIzMpADAAIAAAAUAAACqpAEAAIAAAAUAAACvpAQAAIAAAAHAAAC0pARAAIAAAAHAAAC2pASAAIAAAAHAAAC4pEBAAcAAAAEAQIDAJIBAAoAAAABAAAC6pICAAUAAAABAAAC8pIDAAoAAAABAAAC+pIEAAoAAAABAAADApIHAAMAAAABAAUAAJIJAAMAAAABABAAAJIKAAUAAAABAAADCpIUAAMAAAAEAAADEpJ8AAcAAATmAAADGpKRAAIAAAAENTExAJKSAAIAAAAENTExAKAAAAcAAAAEMDEwMKABAAMAAAAB//8AAKACAAQAAAABAAAPwKADAAQAAAABAAAL0KIXAAMAAAABAAIAAKMBAAcAAAABAQAAAKQCAAMAAAABAAAAAKQDAAMAAAABAAAAAKQFAAMAAAABABwAAKQGAAMAAAABAAAAAKQyAAUAAAAEAAAIAKQzAAIAAAAGAAAIIKQ0AAIAAAAkAAAIJgAAAAAAAAABAAAAHgAAAAkAAAAFMjAyMTowNDoyMCAxNDo0MDo0NwAyMDIxOjA0OjIwIDE0OjQwOjQ3ACswODowMAAAKzA4OjAwAAArMDg6MDAAAAABAbYAADSDAADWJwAAfkUAAN2fAAA5oAAAAAAAAAABAAAABAAAAAEH3wXnCKkFMkFwcGxlIGlPUwAAAU1NACEAAQAJAAAAAQAAAAwAAgAHAAACLgAAAaAAAwAHAAAAaAAAA84ABAAJAAAAAQAAAAEABQAJAAAAAQAAALUABgAJAAAAAQAAALQABwAJAAAAAQAAAAEACAAKAAAAAwAABDYADAAKAAAAAgAABE4ADQAJAAAAAQAAAAYADgAJAAAAAQAAAAAADwAJAAAAAQAAAAIAEAAJAAAAAQAAAAEAFAAJAAAAAQAAAAUAFgACAAAAHQAABF4AFwAJAAAAAQAAIAAAGQAJAAAAAQAAAAAAGgACAAAABgAABHwAHwAJAAAAAQAAAAAAIAACAAAAJQAABIIAIQAKAAAAAQAABKgAIwAJAAAAAgAABLAAJQAJAAAAAQAAAAAAJgAJAAAAAQAAAAAAJwAKAAAAAQAABLgAKwACAAAAJQAABMAALQAJAAAAAQAAG1oALgAJAAAAAQAAAAIALwAJAAAAAQAAAD4ANgAJAAAAAQAABHQANwAJAAAAAQAAAAQAOwAJAAAAAQAAAAAAPAAJAAAAAQAAAAAAAAAAYnBsaXN0MDBPEQIAmwCSALMA1QDvAAMBvAA4AEgANwArACsAOABQAFEAVwCiAK0A0gD7ACEBMgHTADIASAAzAC4ALQA0AFAAUQBXAKcA9QAUATABigGZAfEALwBEADAALQAsADoAUABTAFgAsQAGASgBRQGwAcAB+QAvAEUAMQArACoAMwBPAFEAUQC7ABMBOgEtASoBkAEFAS8ARwA0ADAAKQAwAEoATQBPAMQAFwFJAW4B5ADTAOEAMABJADUALAAvAC8ASgBOAFEAygAeAVMBnQExAekAjwA3AE0ANwAqACUAMgBMAFAAUwDOACsBXQGIAQwCqwC3AHQATwA4ADAALQA8AFMAUABVAMkAJAFlAY0BGQKCAZgAswBoADUAMgA1AEUAYgBVAFsAxAAvAW4BhAG1AcEB/ACcAIcAXQAwADYASQBvAGQAZgDIADMBWwFrAREBsABkAacAkgBvAFEAOABLAH8AdwB5ANUALQFbAWUBBwHsACwBsgCYAH0AZgBHAFsAjwCOAIoA5QBkAYIBxwHtAdEBjgHfAKIAjQBuAFYAbQDrAO0A3QDmAAsBMQFgAYQBqgH6AWIBuwCXAHkAZABrAAkBIQEdAeUA1AD8ACYBTQGGAa0CIQI4AasAiQBzAHAAFwE/ATwB5wDNAO8AFwE3AVwBTgL1AkgCCgGmAIMAcwC1AFwBWgEACAAAAAAAAAIBAAAAAAAAAAEAAAAAAAAAAAAAAAAAAAIMYnBsaXN0MDDUAQIDBAUGBwhVZmxhZ3NVdmFsdWVZdGltZXNjYWxlVWVwb2NoEAETAAEJccRirYcSO5rKABAACBEXHSctLzg9AAAAAAAAAQEAAAAAAAAACQAAAAAAAAAAAAAAAAAAAD////qJAAFys///rD0AAGly///lAAAALWsAAADdAAABAAAAADUAAACAQVFEemFVb2ZpY25IempHeEJUdFdWMEZSdHgwRgAAcTgyNXMAMDMzNDU2RDEtNkE5NC00NEY1LTgwQzYtRjIwNjZDNjQ2MDY1AAAAAAAAAAAAAQAAABsgAAA3AAAAAAAAAAExNjAxQjNBNS1GMjVFLTREMjctQjU2NS1GNjlDOTk4OUE1MkQAAAAAAAQAAAABAAAABgAAAAEAAAAJAAAABQAAAAwAAAAFQXBwbGUAaVBob25lIFggYmFjayBkdWFsIGNhbWVyYSA0bW0gZi8xLjgAAAAV8trnAABK+ElEQVR4nGX9ebhtaVUfCo/mbWazmt2fvqm+oIqi7wXpFBC4JJHELl7RLyY+V29ibHK9N49fbGLM55ebXHM1N9GQGFHUiEqjAgFBEJCuLKCqKIqqoqrOOXXq9Ltda80533eM8f3xrnPgeb71xz777L323mu9czS/8Ru/MSa+6vmnuyEhUDs5VFXNi170TCIiRiI68/g1RWLmGCsiQkQi9N4Tg/cegZEMAOqqFREARMcAxoyMqKrMzMwAUI8iADBz1y2ci0iEyIQ229v97Xf8P977YRicx5xVVUXkzpuOgcrRrfH9X7tw4tTGV792Po5GO1d2Y8Cjx7aGxfCF+x87cXjtQx/87Td+x//86FNXT22tMONksvLFhx6969ZT3dD33ZBSIsas9pJn3VLXNTPXkBAt54zIKfVETlVVFdHMzDEjoqgyMwLEGC0n8g7RiNEhxRhJxAgQEZmZvQtNxRSr2MY43p8doGNmh4hZybu6qmrnHKEjdM4576J3VTJjX7mq9t7HWHlfGbpYt9579p5jACBmj8ixmsZQR1+1dXPfvZ8Zj8fMnE0BgMjlnE2haRozI++8973oQ49c6LIebO/WAQX9zs7erJuJWlVVb3nz23POjhAAEHF/fx9AF33HzETknFNVMmjqWvIwO9hrmibnHEIAAO8jAJhZ+VlENIAs4pxDRCISSeLIzMrJoOOu68g5x8wA5Hxd+fr97/4TQxi1kxBdTkjomTwCe8/ZcjY1IvIemNA5IG/smD05ZmZ2LsboQ6jq2jli70OoQqjYVUwVYqyrxseqrmti+JZXvjqplBcKTN6zqprZ0Kchp8W8G5LEGJ++eGlUBe85K+zvzS5dOo/ggV2fhsWwYPbMCADlnZuhqvZ9PwyDqoIhMB0c7CGyKsxm+8zX/8ow5JyZWTU754iImc0s50xEamYIKAqgqlqeUI9ayjkjYlVVxJBEYj0WMYGMaFUViBwRKZJzLoTAzOXtMTtRMATvfbkaHHwIFSCbITp2wZPzAp6Qg6+dj1XVEFH0AZl8CN5H4phUUkqqurM3a+qxMuc0vPjF9xgCIpLjQ5sbi374xz/+dw01Vn517ehisWBG55waG8K8GxySiCEiEeSczcz7SERElLNWTT0M3dVruyGE4ncppWJ6xayKfZWzK0dWoodzrvzaxbwbhmEYBlIwVZ3PDyQbAHzbG18fYgVGhIFCJKKkYqgpCQAhBSRSw8l0nX0kcuQCkCMXmKIRAhP7yMyE3ii6EEM1YS4HaugIA4VQMce+k+gDACGiqu7t98+9fXO1Dsj0V5+6FxGzKZi8+MXPQrRjR0+i0WzeSZLdLm1N6hIEAWBlMikWWt48GCJiSr2qfu/bvzdGDwCrq6uXd2YiVqyv+KmIiEixtRu2CQBy/Xnlgqmq994UVJVMAZDIVSK6v7//+U//jWh2gZGdc87MmmZkhugdu4BoAOR93Ds4AABkBmLnnHOOGZnZgMixc04pOq68q5wj55xzBJ59jB4bx5GQN7bWFYSIVMGQ+u5ABVQVgRgdAOQ+DUP35te/pq7bT3/qU1mFkS5cuzZu6kt7i2IjALCYd6GKYtr3PZkL0Xvvfe3mGd71W+9ySCmlg9nCe04qdP0BAESkqiKWUioHV06KmYu3DcMAhOW/xQaXP5lVYozT6fTQyVuGLETOheCcU7CsUsWGABQByCF7MVNVACLnTZGcB2L2gZnZBWYmrgljjJVjCsRKloGDb7xrow9mAp52Z/MMKCLf8uJbAMBxqGMlIjEEIiruLxn/xc/9m6Hr73rmcwlQLR8/vLW2MlZVU4hNfcsddzimlJL3nr1zzs27fp5MBxsFAgDvOPjKeTIzh1SOQ8CKEXnv4fpjaZgAJQncMNUCA2KMKkbMXA41Jw2+6ruZGYIxIgJAjF6BANXHaCYlbTnnyHlmT+i998wMyGZC7J1zTBGwmCGGQIJghkwRwTswwwxMgFRXoyqMDOEtr/7WnLOibB8ciGHXdaI6n88VYDQZHz5565DTL/7L/5BVDGhtbaXv5oiMpE9f3nvw/q80VV0CUOqHOGqOHz1kw1xNnGfnOWvKOUu24PxiMSs5F9XMTFVLgCsPESkGdcOUiuUWu8s5i2ZSVUQcuh7MBtF2st73PRIQqIgAkCMwM2RyLiAyImYhpkDsiAiREK0cmQEZBna14ypEYiYFBHYhTEIIIRg6U0ZzNVE4dWwtOETggcw5B+aa2JhZ3da/8ss/47xnxJTS1x55aDKZvPWtryoA4/GvP9r1SVVVcOy5aeOVvd0QQhytXNmdBZT9/VlJRIQ8SC75UUSQzMUq5+y9L6dQ3IqZzVAVzJb/TSkVH7wOLHiJJ5BJRFQ1xkgEJewxs3MOmHJWESHnkBmAFAzJmS3tcxkDr4dGVVDxhAERGQnAyKHzFdMIyYiViBQUsfHkg4sXnt4GFSL6q0/fZ2ZVVT19+RIQIod//OM/W9KQSr++vrpYLP7wjz586NhG24SmHaeUnHMh+nZUmYKZXbi4u33p/GRc100YsoQQPvaX/w3QYozouBhOylAcSFWLZ5U4VaLQdfDli9+VtOi9BzUASGkJcch7T0RiICKLg/1hGFJKWRWMqnoM5ESMyAEAggNAH2oAAmQiZwDsGQiJCMnj9YTtIhqZggcLIQTHoAoJBaBSYEZCGa5uXxVGAFirCYwc+5QSqPXzmQ+jlAQAnI//9Md+oM+pF3nq3BUievqps7fcfodqYkYz67oOEUctN7Uvb9sRIuKb3vxj3aKXPlc+5JxV9dTRTU9cQFO5wCXAl0+KxeSsBYUWtJVzNoTrECSb2RKkppRcHef9UPlKNRQTzdIV5K2CpsTsRQxsabrOOXSeiMFIgQkrNCIiREupk0xMFQLl3AEyMhC0gK3DILljBw88/EU1EpH3fOBjxez3DvoCF80sBFdsPKUUQ9VUdVVVRnjk9J1ffegRzzjr+r4bHBIitqG6dnXPzPpuWMw7VR26hQ9OwPq+r6qq6zrVVH5h3/fFMQsmGIahnF2MsXgoERVjLJ8AgHO0dNtBsgACkpmFUAG7w4fWiBgRTREAmBz74FwAgBBrEclizF7NEI0ZgDwiE0YE74icJ2YmroicY2IywAxEkpnIO7boGUBf9vLXIEci1+FaeU2GXlUxhAKvkXlnZ+9X/t1/QdPJdIxo3cGwONhdmbSLlLfWN2KMRggACXSyMv07b31DjLGtR6paVdUb3/LmLsHm+kZJ3NPpFIBKPCo2WDJb8bhiTSWul2MqTlpqphtfoaUDa5ZsyFDSgSqoOeccIRM5771ky2IiZkjl4Jh9OXtUlEwKzIzOEYAoRLXgIBArsCAzWEQIZoLSE0E2N56uEYOY5pzFdECt6gBIi4PhF3/xp0tg5hAVuJ2Mf+nn/umi74BlNG6z6XQ8PTg4MMJq1NQljGR9z/v/hwIq27e+9jUXr84+8ucfRE1Xt68URD6fd8sqD1FEysfy+pcudh1/FaMLITjnTFQ1l6cxs0NFBStHZsoGsLc/P3v26rOfe2vXDS4MqiAKROQQQS1LKrgWERWVAAkZyNIg6nOeD8FhFnAuSu5EF71l7xFVmTQPGagXZ8lyVgMBM4sxLvrOAdz36NljmxuhCT/zM/+qahqHrq7rEKv5fP9nf/7fq8KQkpmde/KxeQp0bOOJR85urI17yQCA3Gnm3b2ZX1/56098KmASdVUVDvYXTTNaLBbT6ZiHWc5Wwnxxq2EYloEcoNhR+S4RlW854hvQynvviIgdzxYZEZ1jAHChWdscg/m77nnW4SPTKoQQHBEoUlVVWYwZmb3DhN4FTtlqxqb2AfycMaNjsCmzG3FKHBRIMhG2CC7S4MF1CCnrE2dn5QKWS3fk8Ebc3p93fR3YEAF1kDzbPejTHmA99Dt333PX4499HRGPH7/pyw8/UVfxyKGVggFmi26ysfHI4187tLE6DAOoHDq89YxnPOMrX334zJlzl65eCexmM2rcsio2s5RSOZQbqB0A+r6P0ZeEQOQIABFz1hCoIDGXc5ac5vO5mfV9KnSPGhDj7s58SHrHrUcUiAAIPRidPbd/+tSkqkdnnrx44hQr4vlLilnGTcYgSIpqxw67SAGhd46yOvKcMzsmxyDiHMi1AwHQfL0uSykhjcvVBoBQ1YQEAD64l7zopbsdvP9PPzSfDaH2k5W2y7lp6gsXLjgXhmEQsZOnjj35xPlxE+u6Ho/HoEk1/9VffdyFuh2PVJU87O/vj1ZaF0IBmd77AkRv8BBVVaWUzCyEUJIe4hK+IyIB0HVKj4BC13UhQqw8ESAAsDIjEQlmRkFEdmBmps67mPqsAmBMhM5cCAGAyMAAclIkx4zgAIAcqUNHzIAqxsbRgBwuc3bhCZi5ZCURGZIwo4jE6Hf29z7x2QfOPPnk0UMbLvLu7uLlr3nFF7784HRUhRBQxcyQ6NzZp1FlbW0lpf7ipStdGn7+n//g5vohNO1mfR18245DqFSh64ZyeUpCLJZVYllKvXOUUrpO7yyL6hLaU0qFq6DiDnXb+CpqFiJixhA5pZ4Z3vlffwecOUdg5DwUosP7yMxwvcgCgJQWAECIZgjmSshERMNl5cTkyTkDdghgnJUcCCyDIS4WiyXBZLngkmEYxuOxq2K36AuvkvLwe+/674fW1gG995yt8AG5quOHPvK7OSsAmdlisfjZX3yHsv7Cz/4MM3ddt1gs5n3niJmxaUaFlrnBN5S6pxxNVVU38kBxVVUdhiENYooEAFVVDcMwdD1YQHJIpXTKPgYxQ+3BkBjIIQCoDSU0IiJAoTsQFH1wimoAWr7IKOAM2ZEaMTsAzCmzIbNStuDMO0MzS6kvMKcEjpTSeNwiGSKura8PB3MR2d7eVgEXwg//g79dRUa02UHnvU8pMSBIfv3rf6BcEkRsYvWiF71k3NQ/90u/gg5irFV1a30tycDM+/v7xaBu8FYA4IgRgGnJ7iGighSckcX+4MMPPrU/f/TSNplJYCIGAJjPFsyO0AGyKeWsiPjCl74MkYesxMqoRGggCAzmicgAAcAQCD0iM7qmrQHAlIwYkR2jATAnBQMyVEsUsyCgEqKIlBhRWLcCf+bzeSm5zp87v765tr+/v7a2xswi+ta3vNUMvfdJcs55nvPqyjgbIFrbtoWMTKLNyuSBh57oBgmIALA3OxhNV5wL34zgbxBYnp3aN0gYRJScq2a02+VHLu1e2Nt/7UtujcykmZxzs24h2ZBs69CmZERkcoxQmTICP/u5zwdzCF7EFBDJmCnGCBaJyAwBldmbOgAihmHRIZkSAlQOPSozQSAzRTFQ1IQM6D3VJqGwkQCwWCyKXyC5PIjjUCi5/b1d770Y1G3DCG/92z9You94sjJfJDbthn7oU/E+z2Qg80X6k99/96iNZPrUpUsiEmPc3t4uQSelvjjdEkAYZE3FmkT19C037/a6kHR1e2d1Mm49y3WkioiUswYOBW4czGYu+GHIMTQA5JxHCLt7MyBidoTBsMCtPF8cKAKYA9BzZx5T1T4JoTezI0dWDNWRN3ViaFA5UgRhZ4EVCUQJABeDGJZCREtZX14QgBY37/t+Y3O9beubb72lbkY7O3vjtkEgZMg5H+zvN7Vvqnrcjgzh77/9+3ulJIoG3WJ/NG6r4EMTqzji4BFxZ2fHORKRuq4L4CQiR+y9J3JHTp/menrQpfsf+Cpq33UDAfZ9b2Y3Whjee3KOiMAMF91QxWg6AGjOcyJTVQP90Ps/AEYpKyKZOpFkMCAaAqsCkXv4sS8DZOcCAAWHi8WccFAb1BChSuqJHIJjMqbBzAiTSBajgkjLdeu6rnzivXfBA4BzIaV0MOv297a3r16rK3d1d1Zsuaoqk1zHsCxQRP/gd3+fTRhpED114nhVhde//vWBnRoQcupnpY1UslmB7yZ64tabdhddl+XRh7+2u33JQAiNkWKMIqJgJZsVTF8ABAxDHoahrkJd16VuTIMSFx+Gv/u9b4vBOecQPKBHZEfgvBCRqTN1z7jtDuNMRmZ1TsjeOZcNMhMhOMTWlDNGRiMERHM4fOiD70OWZF3OmWBZ5ZfyNQ8yn8+H1CHr9va2ZpFBquhNMaX00KNP9V06ODggov35rCSEQnuEAM7RMOSmqa5e3f7kJz9R1/UguU99U4/WV1aXaNO7F7/i5RDiQuRrDz/KiGnoSq3DSKWQzjknyYzUp74wNje4nSWhIxmHDs2iAiI5NQREInrHO96hKgDABOQqhIDACCA6IAS05tOf/hRhj2gpO7F2Ns+APWAymJe4oxZNvUFIEhgNGd/wHa9n6Bm1eB8BSloWaLHyiBZdZKSmaZxz874T2yuX9+YTW3XlqhCzmikxs2MENUSUrH2fvPfXru2srKzsHSwuXr6SkhDifN6dOXN2ZfPQLOVZ13/0f3y0m+0bSBo6AynVWzmOQnsUgm8YhsLKlGCHiJRSapqmNGabNoiISWXqwDwiimE3G8gxOyQD6RUsJPWes0EnIqrwfd/9ow6yi3MAYGiYGjMCUM89gCBkEQVAFWfgRF3K8ZGHnvTOOcYblG5J/KradymEkHOWQXKStm37Ia2ungAAQ73nmVvMnFNfng+iLlaT6biElevxBYzQsqyurXjvb77jDmO/stG+74OfGI9qS7IkOUQLBVQY9xvgqzRS+753zg1qJTLcCPA55+w9o5pqJkbvS/sMcwZmd9ddd+UshE5wIBYVUwsZCJ0SIWj96//+3zhPCjPCHow0R4AIFM3EKCPPHAlhFhHEgg/o9ltuS1IR1OXtee9vMCdt2/Z9XxJ5XdfOOe/93t4eAFShfuiRq7/26z+ths6557/weY8+eelGuQcAOQ/z+Xw+n6sYB//o44+bySMPPdQ2jGqXrswKmiOiUtnc4E7Lx4JISyusmJuhjptW1ZYeEGNMKaUk7NvJygRUUl4ggHNUVVXgajHr1DCJAmiWzjGZNGkgAOEwN4Qf+4mfW6TKyLxPCHOj2jTmDFld6QsACkEiVtSuQP6eVBXEANSICiutaAAART9RVSFBXswOdvZ2vI933/2Mq9d2VydRVf/RP/xXhpoG+cxff2FlWu8e7C/mncNlpjtx/LBxvP8rj80W3ahqVkfBBW8izL4dVXt7e4WP/0Zv+DouJaKSmpm5+DU5bpvx/mw+5FwoQ4oxFixuII4Cu2hmhgagkoYuda/81lcTeDAnyuwAeU6AmiNT5dxArv/Yn3+QpWWl1O+xU8pdd++7n3zvb1JeqKqoz0CqyKgA4IjIus/99afVerNULiMA5KwAamYxegCad8PKaBTq6skzF5nhC5//0uEjm1eu7pLzAGSqomltbXTo8OraeNqnocv9dHPzsTMXhm7RNrGOjozrur60vSCi1dXVGKOSqQAQlh54OZelyRB9c++rBHtJWRWIqKpjeSb1ffF/M1REcy4w1WhYGjlE9PjXHiyG6mhs6swE4pzZS2pNGXF2ZfvrJ5/6L/V9/905l/PByd0/Xts5d0Qvf/H//sUv/tr/vpDeFHvwqstCksi99pWvcKQG8s01bXnRK9NJFXwVwqLXnPOp44dm+4t2FCUPYPzt3/7acxcuiVJV+e//rrc4YtHSjIHLT18gsHmfZwfzQ1sbSFY8tF90LoT9/f3gY93EG4DuepNViksWE7vx3fKSvvrQgyXhLl+5iiyPGSFED6DkcNkoNjazjpgcEmczK+iBnTqfTB1AC2C/9v1HRgfzlW7R/dn/dfV9v7ZF/XRrPB63G5Oac37o1//FwSOfhZxm6LKxKargvMuEVVHvlBzEjGZooiEEQyCCbnagg7ztba8rgACMBOz97/vgqG7rWJng77zrzwqIvemmm4LnUVsdP7JxdXef2XvPZEoGAIpMe3t7ZobI81knIqPR6MZxFLRZUt4Nnu/Gec0HREQoBZYKxRiJyAV/cDA3QDNDyArKzAZgirfdcrspmonBAMaamzQo+x5pAK0JRnE8PXR4xXmYxJrS/PLla2FzLThPCCvjUNnwwB//t4/90k+4oVfApCKA73nXH4Ix0jfAccq9aka0NrpCPxSs/Du/+/4Qwi/+i58sZ/qbv/H/Xl8bGWpSGbRPkonc5SsX3vsnv5pFmXnos0gqKcLMRAEAJKUYPQMiIjHcYP6K6xUavpSoJckoGBAC4ZFDR0WEHWXTZTb03qsOi8Us+BFyJAzMDJAdo2d81+/9vhFnJQAl6hx50DYlpThnEtUGbr+jXlsdj1vvdNL6v/zIF4Fcs7Jy+PR6jRnyEMymcfjoL/+kk14U0Ia7nnOPIgA4VdUsIuI4lHC7tj42w9LjclWcTqci8gv/8ledp3bc3HnnMxERDUoHHw36Yei7/Oa/9aNHpuMkCkveDpf9QVATW/LFtrSaEiiX4DOlkl4KoVZ+FnTZoXj04cfEtJT3qko5Z9XMRE0VVtbGJmJIkhUIe8mIOJpWqIbokxE5BFoQEdqaGYK7xk7lYA7HDjfjuqqgRt3aGv3+f/jjuDZZq3V1VE/aqgoWUVZr/bN/+eOx9GPSoHlhJjcS9trKaqEAD/YXW2vTQ4cOZYNuvj+fLa7tbAPZ69/whkPrq695zXebIbGqSNWOnr5wzTtX17UJPPbUozeUVqi2JNHYA7Kq9n3yhSbNJW5SAVCF7Sj+uCRFicrxee85htIfG4aBiKhuGiJCYOecWma2WCQeUDmiJPamN/xPSFI5NKnMOOuAMGeOeQgiHnDvoUemcNMpFxfTmpMKI/jAj3/x/mZzOl5rJ6tVCIEIIA815ff9/I+y8sa0ASAAQNOu60z06tWrJR63VXvpyrVr164BqDkYshw5dHh20L/3ve+/fOlaSVg5q3fxwvmLSFJoQiKKcbMAxtKtWRIJIgfDQXG3LN/g3QsYLhUffJNKq1CAS9YIidmnlCRrCQtUroaBqOpiNi8uYGZIVlWV6hL4GKTg1dCbOqXUDzseKtQVMHcpZ7j49frExnjStrWvvRsF/tDHviR9nhxZXV9bqRuvCEwQyRj79//CP/nCxz+rGBC0/CFAdZ7QwMwUIXpvZjnrzTedesFzbt3e3z1yeKuuwvbefmmd3X77raGK9933gcNbm6DStq2qtuPRnc+8Y311ckOTVo6sqerScGGE6+oovnFMRRxaCj7VfMPoCDCr7Ozs3ThfVKPrzQxama6tbxxW1SK3NZOiKvjcZz4P5AB50Blol3NDGINTcgfIJFK9//0fh/Xnw8paXYeqrV2kqg5H15v/9BsfCRuHDt185PDRydrKCNGq2o88h5CPD19mEVNkZhAtobcIgy5eupRVzaSqwsWnLr/iZS+tiA0GRGTn9vYXovDgAw+vTSfve897S7Gyvz9DhtnO3hfvvV/ysETCSwVGvnDh2rwbiKioVVVVDZGXOo+c8w2inciNx+MbYgjPbnV1bGaAhoiKsPRPZt7evcQ4MOMwDM5VhAEADPNkuhIAhh4QvMEAbsEUZ73LmtSuIeLrv/3Nf/F5042N6ujaympTj8f9kGMdguN7P/LFxc588/jhja2mrZ1HIDSHWlf65Pv/DxMEHUpc15RTSmgQQlVHjwD7+zPn3K/9p98l53/sR39QRNTyqK2b2ld12DvY/b9+9Xd+8p/9v1KSRdd5dm1bT8b15e35N4cexKKQseI9sa6WNSD5UpB+c3GqqoXPKJk353zTiQ0A4Ot0G5XSMed8a9UYMAAMQ48oiOg4aqar184riHPkkNRahMFk4XyD1hoE5YMnHvkSQvjU52+lO08dWp9uHZqG4ILzG6vNJ/76wQe+/PV688iJmw6fOrU5GflR5MBkZi3axT//OREzQkn5umhJLA1DVjNrYlip49q08cH98i//OgCA0a//2s8WlHgw63PO/+7f/Nckw6GNzRusU3mfxcWMsPCI3nsXQ+VD1/U3nsPMuhSueCAkxzeKeUQEQkRMgyGiiZXn0zIRAG+7ihyLYKzaYRgUk6oQUVOvJgPRRAwOia3JJuwODFS1YavOnH86wzz7Bg49j++4aWVST9dGProQaGstfuTTj/zhf/zv4a7n3XznyaNHV5rWR1LTodcsQ//25x1GykpMBJZTCTfjtm3bNoSwtrby2JkLpfp1jryjn/yJf/3nf/6fiwShdJLR4MqVS2RgsBROAgAxdrkv1cKsW/jAJ0+erBvPzEUlRdeVksMwDDkBLJW7JcmQYxFBpq6fE5EPbqm3AVyWGuLanDORmibDCOYUASi/4ltfXrkI5rMNhnMFh1KZovAiuAG0ffzMJUM1nz75cYY7X1Y9767TpzenK61zbtRWW+vhK0/u3v+uP8Pp2sbRyeq03tyatsE1gaqKEPHv330ysLlYxRiLJu9gcaCqg+hdd9/pVPb3DkRp6NLWytQ594bX/0MAQOQzZy4t+kEM6roujfZShzMzMHkXSqkcqybGeP78+el0usx9TCklFSB05L6RCsSUmYecCj+Th9S2FSGUVhaqERqpiaqmPBuPpgAEhJ6FmZkCAK1ujhXEex+oEXVAs+By6gNoFEvmDsZx4t1ERBLPAaZw4q61m4+sr7e+8g5pZVSd3mx/+72fuf+zXx6dOHXi1Na4JQd4bH06rqs+S5b+jbdtSN8NQy6BFqTwcDKb7f/uH/yimE6ncTRqnrp8zczUBgKMTWxHlSOOPjSjRsDEABGDi4XwKSc19Lmp6wLZU+oEtJDD7B17p2CmGGMsqLVoaL8h8PaucjwajZacsneEDCoWOBA5H8i54DiqgujCZHDk//A//2HSJDCYDkwjRM5k09UquAppgmi33HlcJDFOUOOnPrsNQHD8yPETm8e21jiw974eV7efXP3td33+0gNnp1tro2lVBdrd3V+Z1pUPdeNX2vDSWza992aS82CSi+TOkv4v/+jfbl+9ttL4lIQBX/v613QpX7225yCPx9F7NpOnz1/JSUQksHOBxZSQCyTyFe3u7hZ0Nh1NyyfMrgQ4M+vTkHLu+57QmeINgFaO+/ixQ92iuDOJCC2GZGbZ8nx/bxiGnIdBBhdD4JGQGvW72zseRohoqGozMGYezxZ7ih2jmIzuefYLBAfjAXSM1n7ucwhHb6GjR44emR7amiCZoTZNdfOx+p1/8rkwGo2nk77v6zp2vVA+GOYD5vndG9WUe0Qcte0NQu7atStvfOOrQwhff/Lpoh957x/9KShKkUwZaFriTx8rQ1rkvNLG0ihd9sPNmrqt67odjYh5ZbpKnlWViG/AdFUg5iWbQL6gsKUMIqcsqa5rQGxiIAQqrSG1xDHGWA9Dn/OgkD04wviKt36bUmKqCCtEJk4eJPcOkZP20euHP/hxwmiWnE+aPcgI4Gg+cWR8/NB0VE8mo+hDiL4f8u729u/+x/fHuk2DXXjy0mJv+/iR9a1xtTJuxm39t59zTAB39/YAQDWb8srKyic/+YXnPOc5bTMpqb1t60kTt7YmO7szMPziox8rYagfdOj64Onilf0QQtM0RYYrIrP5ICZ93xPa3t5eCe2lLfSN8ggZCF3wRSdwg8AZ1dW9Dzz1xFP70mGnzgXHBmpmhi7lObA2sSUk0YTeCO3Wk3ciZuMFaXTUOpwnXfgQGRvHlWB61be9BIwRKjEhXgCF82f9kRMvgpsPjl28cvapa1VwInLy5Fbbzh56cmfzU4+87FXPHIaun9Fjj5+pK4eiydLIhx9+wbG/OeAcPBE5ArR87tz5S1evFCHzskwBBUUiQsfPvfN1qxuHZuee2lpvZ7tLmng8HvfdImW59bZbHn3yaeepbduOeiKnXZfScnipVDbF+rz3ReonKtEH0SVds+iG59193DlmJrBUgAOHUG2tb7H6S0+fSTZPOuSEptwPwyc/8xekkbQxgGQHYuRoUtfRbObQgYy7ea+2QFSTBq1SVgXM2ezI3eHYxi03HQLQtq339+beWeXh3R++/9N/+RVXjcer1Ynj602EGNibjGu30sYXr1KFpjkjGhq+/YfevOiHpy9fSRlOnzxkZm0rDKiqgOac29+79tM/9Q9OnjhWwnAhW7wLGxvrZ86cYeaqqj5z31e//sSTCEv+YEk5KFLRxAKIqHO+9DGHnJCo6LzFtHQ2EdG7QABQ1XE229/bvQKYIhNkBsDYkqrGMFodrSvPiCUJEVdGPbGk3jsMmXpyC9XgXQuYjTsFlxMrDhcvGMIxPXFqY7WeOjr31BUijIGqmm89Pr3c6cXzF2Lb1HVc35pOmjiuY406cjSO/IrDvnG+X3SIcOrY0dlBP52MquifOPM0AOztYtIkWWPVzhbzfjF85rP33f+Vx77wxf9e9KHMnCXNZrP54qCtqxe96EWvfPE9dz3zjhBj04yqqiqJEtFEJIkgOzPr+76co3Ou6EAdcwyuxChmVhNKKe3t7tdNzDJk1cvbV9mBJ448coEVFp/56y+wNiY9uh7NBWjEzMV5VnDYAvEjT35VRE0DIiosEIigNqCnzh7w0edW3/ndN92ydWxrrWmag9mwPp00EZ546sqX73ty/9oeM7d1045CU0fvY/DYRGsDvWBdK+2bpvnXv/JfEW3SVA6BgEUzESFyU4ednWt1XTfjcOTYcQf5xc//ntJD29nZDiGkYWiqtqrCY489Mh6PAcB5LvhzNJooLDXaJRSKGXs/SB4kFxhRwlZVVapSqkXniEzBe68Ck9WtKrgjh08SuUHSvN+HZCbNHXfcBL7//Gfu97jGfp6hd9yaNMSq1oG4g6tzZEM0VQSt1WUzA2OBCOBg2FltG5+GaePrKjLJ6rg+vTn57Nd3v/w3X/V1M1qdjtZWRuPoPDAhGDnA1bq6e6t68gufY+vEwEwM1EzaZtKlLCLBc9vUTHjtWveRD390ZWWFGOfzuXM8mUwKpZdzrqoKAIahK4Hcex6ydENP5Mh5REZk72Np9jj2pZNYTAmJhpRKzURoUNRfiJjVkELOGqIzMwbnsRm0E5w/fP9jCP7o8aMCc7QRIosdIKtpIAyK3U2nbyarEAFIVRU1GgKgmtnZswHCM8LLnvP8553QIU1r1wTvHU1H/u7j0+5Zf+/BLz6UsiZQ11RV5b1nUFXNjtGb3Xpo9Npn3nJopS15vW6q2Xxv3NTR8ZWruwz4l3/xzrWV+uKl3a8/fm7eCTl/9ND63t5e1/dqQOxCcOfPny84gICGYQDQ0otExFBFQ+jT8I1CB9nFgI5v1NhmVujJJUdYuhU7V86K9A984bOqamQKixBbMveq73gJAHzus/eieYFtBCbdMOjID2bI0J7fvaKYRBEAnVeFDECIioiqbufKAcRq9e0/cNcdR6ajajJqIpN3tDZyD73vN77y5N6Fs5faZhzrql0btxWPmcd1BZqHoSPVrdC/4Tm3EgB5BwBNrEyFHWUVEbly5UpKaTEfDh9aH7p06cL2fJFKlBmN2hgDGpHjUs85x4h46623liZrgQ6FoimMTfFKZi5fZOaVlZVATGjMHJ1fjmR478fTY8GPDp8+raqBgxhL7gj5/i89AWRmPUUMsIZOlS5JqlhqowXS8IqXfZuZMQuCFwUiVlU1Z2iAunOwiofuzN3+sRfeedvpzZqBwJrAdeTDY3/fk1e/9JWnr1y7Nu8GGZL3frpSjRkJcFpX08ZPmzBOu5sg8919UFmdTsrgVU6GiG972z8ZjyY5D5PReGUlHD+5eXV7Z7ZYHDt6NA1D13Xs0BGT47qOCuacu3p1uxuG8XSqAApQ1/VkMhGxm2++2cdQUJv33sUqqe3vz5KZGoJmA1kKQ0xkZVLN+w5FCUx0+PpDjzRhTIxtK4zuVa97E5up29HkAh8GWiRcELQi1Wc+9REwUnUGiZmHJMZOVVUQAADzE49NXHVa77xz6+at4yc3D29OakeS8uokPufk2vu+8vSDXz7TdX3Vjvq+39/ddx4mtSeF6LBB2hz5b7lt9W3PPuXZdvcPxIwAg2PVbGbdbH7k8Obu3o4ppX4AgKaqL168mHN25Pb29sBcUTMQkY81EU0mk2EYQqi8j0POzDydTs+dO39DvFzQPzMaoXNcxqDMjNpRUyrJxVyInAo6F8BwfXNjkN7Map4CyalTh0CjxxE5zbCL1jiMxJkpXbm8D6gIisBi6h1pNkQu/QhVFXMXHjdq7uFnPWNrc7K22jaVG49qzzwK8OITk9///JPnnrjQ9/2w6NcObfhATaC1JrIiE0CWSe2nnN5y96ZJLgWzrziECoiHYYiBRGRULym6/d09JE5ZDKGqqv3ZwY3WKQCEqkLEkjedc6PR5MakYQiVj6GKtWS93p2GIqQwZCIiSUNpDW1vn9dhtljsDipm9md//r6HH3qICJ586lEP9bt+5z3oe+XsoSYYqe0YAlg0pLXViOCBEFAQtMuCaFl12YMTULBFdgBzOPya9ll3HDq6OWqrOkRmrOt4ZHX0yttW//gv7n/owcfjpO3nHQJ775EyZtU+eUKvOh03J6bNyw/He55793RzPTDUde0IQ/SXLm+D6P5+b2YE5mM1n88AIOfEjMvxuBiISEwvXr6Uc+77vmpqsZzzUF5n6cyLyJCTD0u5CoiqLns/5Mk551KSGGNd197XR47dBlnUue/7gR954N570eB7vvvtfUr3POdmgposC10F34W8znyA3IGGO299DmD+vf/ym9/3Qz8CDCyohj/+Q99roJoF0SIqmTVVeOcHf9ue8ez6wsX17T3BvWt7e2waAhyKk9vf+Ny7nnHiqa+dIUMiXBk3qgowKIaKaa/LkWFSR0EZLj5671MHIVRDt0A0Vcml8UWWcgaArMsuTs65rpuVleicSymFWIUwmNmJE6fOnTsnIt6329tXC13vlsKjYko6pBS8J4bW4zDkUMWhW1BRzg19JscAevbMI8UTmRfnn76IyE+efcIHqvAQYg+EaKzDRGAmYpIqAFo/sY6Ib/o7301EoACgBiIiwzCoZhMZhgGJ+6Q//OYfQzhGt966cezIdHU0nUbPRihsOt+dX7mwf/zmk6qSBA5mnQKEygFYSuJAVDU4Cqxbo+r1t60ezHogRsgAxEj7s+4nf+p/3d3b+emf/BFfutmw7OXMZoumaQBAcjpy5Mh4PD579uxisZBs29vbIVTsQ9Hi5+vjhyEE71yM3ogXvYSqSqknIqrr2jnyVXTczOcdmrEDg+Hd//WdTz75uKF+5cFHReJ0yxHWnsNTZ74W4kEVVomYcEDEnJwKXDh3tmw9IHI56Q0RQwkXqkAYnG/+t+/9ZTtxDx1Z3zi0cXhjpfLeA1VVMIT9/f2Dg/l0fc0Qdvbms0WPjtlZzhmMomcCa2OYjvxKpO96/hGUvDaZoikivuQlL/qpf/YLqu4//IffKkLQvu9TyuPxaBiG3d1dIiLm+Xy+sjJJOW9tbaXcr66uFqrPBU/k2rYVsBgjIjnnclYA2NzcPDiYIzIyUd/3jhjRTOZgdMddd1WxBeOXvvbb7nnms0yqS+euoM+T+jC6NOTF9tNPQd8i9n/yzv/n0Ye/jmjoJYTq9mfeDYAILDo45wAN1NBsGLJnR4COY59soULQwqHjzaidbqysjcKoiqBGwKY4258ZkffMnnLWNGQfQ9fL7nyOaI4ITb2j6SiuVficdby8Myuw6OKlC7fdfOzUqSM56zCkEEIIPsZ44cJFESu4KXg/m82OHDk2auuu69q27bqu8DmIWJBnHeqUxDnnQwDEqq6vXbvGbikiI+/9bDYri0C6fv7U2Z0kg5ldurhz5fwZ5O7V3/F6SfEd7/z1shljvLohYW5Ux+mh03fekhU+8Ie/O6T5mSceL5CttNfSkEtBj4jj6eSZL3tbu3nah5otfu2BGZ64E1bbtOhO33Vi5JARiUBExHSxWAxZc1J2rusFjOqWwPGQIRlUdUDEwBTYHZ6OnrXZKhiTm+/vdd1gWY4fO7S7u0uIZZw5xpgtT8cTVUWi6XR63333lR0wADCZTGKMZlbXtYIBAHvfLMPlcsmGGjTt1HsfQqByqKraD4nJ7+9dQADv/fOefxdNG4Rw38c/63n+D3/4xwBANN32jDu9rWlejB2DegD4trd9j/dubb0GAEB1nqFsCTI7fcsLuh6f+bw3bx05+tyXfcdo9chLDp178Df/OQPy6dNN247a6eHDKxPnCGAY0tBrn4UcU/RGaMRZhlETQbRIa8wsFC7FOyY9vRYw4/Gjq3lIr37t3YRy6crl5UgJAICBFY2vMFI37xaLhXNuvuhTktXVaVG13RgBXtvYzDmbIjAl1RCC975tWxFRIDMjzRkA6rp2LhwcHMR2AsgA0Od01zOfqwg/+lPfRzS6fP4JA2lGk//xkQ/tLS4Q1t/yhr/lPBLon7zr95DiZz/7BQCwfqn7KqKUu5//8je8+jvvftErT998+2Ixf9nq05AWDuyP/vH/DqsrzcbkYGd7ujoKjiP71GczBKasxt65KhiqITLiOMbFfDmXNZvN6iqg5eiYQF53++qZc08z8yc+8mUwGrcjuD5B75hT6lWWZV09ivP5ASKr6vr66v7+zHs/GjWTyaT0ImezWayrtbW1wuFUbWvXZ8NCCGiwnBYmougds5vPdgnNzKaTFUJm1N/7w/cRpk9/8tOM1M0Wr3jVd6xMJkgyH66Vtvv3vv2Huvns8NYWAHzoox/QJYuC7MKFJx5Zu+2uc0888ZE/fc9f/NG/GePBehvHFY1YwPFo84gpr2yur65GS4MjnM8WnrjQvvP5vHQNqsq31XL8SkSqECzlKkZmriLXwf7eS29VwEIzhODW1tbKmxqGYWNjKwQPACUwtW0bgmva6tq1ayVsGVAZBiv8RJmE8i7WdT2bzQBgyKmUljFGWqYr4Mr5lNIznnk3oEf2BPLJj3+cKdxx+62I+Pq3vFVEiC0f7PZJVNNau7G3uwvkfu+33sGe9g9OAQBnJSUViTGqDH/zhQ++552/+Bfv/XfnH/l4225uHhlPxoEZqxhgd++phx66eO7yo185c+TEkY31yrKB2tBrod3qtl0uaEKoauz7hQuh70RVmZHATGXkq3HdrAZ6xlqVwbz3MiTvXYy+beu1tZWU0rHjR0aj0Q2mocD3ugqjUVNc20dX2IS2bctljpU3s+h8EQUWuJ9lWM6D+br+1re8NSVRrpkNAO793L39MBMdxmuHUfGpJx8j9pLh839zr2pWxY987INtW6PZaIUYq2MnFwDwhr/1d3Qpubs+h42Uh0E1Hw55bVJzIBHLeQBQG1SQru7MdxaLZjxa22jVxAgLQ5KSuComM1Nyzq1Mmq7rxJaNaEm9A0spmQ4OdOKSCe7P5i6GjfXVtm1nB90X7v3qZz775dVpUzRodazatnWOPJOqNk2zsbGy6GYba6uIOB63w9A558RUstV1XYZ+qiqWXTfOx9LbwL/1nW/buXZJRAKKmUWPQ8rteI2ZP/exD4Oz9773T9UyO33Vt75GBJ2jV736DXXdqqbbb/27Ynlv54qZfeW+vwaAM09cKr8WAKwMnJrMpZ+s1Wtr06oKoW7g7OVmZQXIDVm+9uDZfpDYRACY7e2Tkar2aZjPO8nWDYmI1qZjz0i0nINoqjp6FxxblrapVir/slOj06dP7lzeJoMvfukrX3vsTPBuc2N1f2d3Y2PDe980lfN87OhRAHCOtrd3+6FbnU7Onz9/9Ojh1dXVAvSJaDxpi00NIkePHJGspbpeyt0unb/SjqciMp8vYhWGIUvKL3rus8ysHq2nrv+hf/D2KnhT95/f8VtN60z5E3/xga7vvXOf+OQ7TfGrX/0qAGydODGu/dpmS8jMTM55750nRFTnVo9vrWyuVq07dstR0xzqQCyrWytx1MwOEhiNp6O2afZ29sxsUGHnBSCLUYhAWDqviFiUlXX0ZOCI+753jKt1WKP+JS977oXLV9707a98w+te+FM/8UOve82LHn/8qXe/56O/+mt/8Jo3vPrS5WtffehreRBTPXbs0MmTJ4dhCCFcvXr1woULRZsbgpvN513XpZRijLGqbigkHICGGEZrdXcwMzNNB33fs69e/PIXfe2RnaoKF84+5OO3u8ViGDIg33Z67elzT588efzYiWNMoAZ/97u+E0m+5/v/5zzI6rjp+7SxWf/8v38HAyJ2bOYYdTC13ib36uMX1tdWgAy3Nu3ieWR6+qmLZByCH7K54IcuFeDTpWHos3OOEfb2F+wwNr7f74WEAABgsegBOUvyhAqcctoK/pHZPmTb3d0dTcaPPvp11Tzv+m99+bNf98oXfvzDn37Z8++pmto5uu+++3/jHe9enbTgq0Ob67ffcdN00l67tgMAjtGAqhhPnDh+4cLFvd3dEL2qJhWHyCmllWZUT9bPXdxe2bqJyaVuQS6ePL0xrsOpf/AjQ1pMVldUQTlvnjh+5PDm/v7sBS991pCTI77llkPBG6NDFNVC/chjX3zgrufdZcYAkgWQgIzw8M1rx57Gpy4AGTx9IXB967PuxBjOPnExqc529h1YrOOQZH97J3ovDPP5PIQKJU2qtgYaelE1YGTiism5YIR786EbOh9in/NRz2dSWl1dVdW9vT3vfc4ZjL33qR+atTEi5Kz33HP3r/7qXSCKyC6Eknmv7Wx/9KOffeD+r7aTljTfcvPNq+uT226/ucuZckIyV3YAzvouXbvsuNI0F5moCWgiYAFzbGZoJsiAim984xtBcgi8MTk+zHaz6tknLt5223EzMVNGzJYB6NZbTpXGnyECGBiY2dceXb/tjpvjzt4w359d3c7dwqFbX1092O8A6OKFS66K5qiNHhFGk7HNZ2XsvaqCIRn0hJYkI7oMkJL0SRy7uvICppJZsYH0j97+1vu+8tjBwRyQ+y4xs4qYyY1OF5T9dCmxc0RObTlXePTwse/+e/8TftdblvyfSV3XKfUbK+1v/bcPYDUpBSbPdvfGa4eTdECI0BPy9vZerLwPlJMUzCYZCT1H3bm6k1Keza99/MN/gYi33HxMNS/VcqpkhGaPPPiQmQEu5YqAmg0027B5KyK2oxVYDKPRZH9nd8j96sa0Hvnp+gSjr6rKRYcG1y5drut6bWOFnVNDZg7IoyowgSFnBTHz3uchyZAcgpkxKuT+4U9+/GBnxzPefNOpoV80TVVV8dZbb40xVlUgorqOzi1nBfT6rpAQgpkg2g2RLREtFgtmv7Ky9uP/5Hv+lx/8NpIhIRp69+UvfpqZ510i5y9fvryysjI7WOScsQxCIBODSP7oB78wDPPg/JWLB89+/vNEJN0YSjOE60Lgu1/ygtIyMbMSmAF5/73/5yP/+t/GjTFX7uoTZ/j2m7buuH3URshy5MgxEusX3d7eXs6KTHWspMuIXC7n/v4+MiFaW1VqubAaQ9epKoIykmcktOBZu+7wSgUUHnroYWQaut45d/bs2RtzQvZNGxGDo6YKBY4UdrR0tksqj01dWhh1VVVVtZzRQ7U7b7lbMwYUE90/2Hnooa8QK+I3dkGYABG96rXP2zh6tOuGjc3paG0FEZ2/rkIwwaJwVVVYanrMTMEkG6itrsaACkeP2Hi8tj5+7Pffy6HaOHnT+pGtSxfOx8p77/N1SV8/5G7oD2YdIqqIIGaFWPkYQsBl+elj9J6rqqqcMzMP5AA8W/fUU1evXu66g/u+/PCz777Ne74+F+2WI0uIIimEICKDZEAtbCgAFB7Ce4+ONeUQwnx+AABV0d7lnAnx2v7VEByT77ru9OmTdz3reSIiGURKw3JIKaU+mSiAVk0QWc503VBplqK0rAYsQ8h2faav6DLX1+ssxsdvkrpp7rxzZW164W/u42k7nYxEhJibysXKA1kzGvnAach5SMvC2ACZEFh1QFJJfUppPp8Pw4CmBtLGYCCMEJmrgDcF+eKXn5jtdVeuXBuGjGZE5BBuDDSVzoP3vvIBEYmXFeUwDP2wQOBivF3XNc1IzXLODgC8j76OMkDXz+umQZLVtdUnHn9wbfUQaFIkQHMIRgBoaVAwBkjsyCwjoqeIzixj13UIcPXKTtcNYDSfz1NKfZ+RHCM5htPT1gd++Fd+4/bvfF2etnE0Xt/agtmwvb27XBEyafMuFEa8VAJGKDnHuskGKYmLgRhGVS06N/QFJYHqMGREjN4t+sEhB0TW4Ud++K1psP/8O+8nUETU5ViAMmNKSqXPXBaaqFb1KOW+5ijZioQmD0U2QyGEIipxRISE/bCQr5+Tes3z2s23HGNmg+nffPYBNfKhGY1Gg6r3MZuGEBwHtxwIAnZLndw3TTkaLzcO1K5ufAXOE4N5h6OTR9qLV67s7ghSClV10ylY34K8WN1ZJWbv3NPnnqr6tLd30MRKTH0VJEl0frGYxRiz5M4gupiSIADhcvFxlxIRAQErBMfJaOTCft9tf+X+m1/+Kkc8HY/Iueh93/exqe36EBoiIoEZIKKBdIvBBybionUo2dN7D6BZhBDJeUI1G/TQ857rfQQczMTMmECRXYjocJBMiGbinCNT0CwgAsJopQRV1VJ/mQki3dDsy/VNqVj06OuH14+ur0ymbmulIWLvrnzus7C2NXruM4+eODTkngNPx20IrojXqzoEz845MJOcSkcKnDczT2SiAMtFV8EtN6cxs2fMOXviKDo/mJno/v6+mfTDokhomLnsLMXr28XbyRgAiCHGyG7ZlEbEtq0R0bnlHohlI/tL937h85/+HJIOw2ClmhABKxIBKL8RzFCX45SoZYBFZUgMZa5YRMRMNaeCS1WHwhfq9cdn7l+Lp0+u33oC2OmVq2goSBc//DHYPbDxpPHx8OGt8Xi8ur4Wgmvb5YC0FQ26QRUqM8sqVVMv9d6kIiYiQFSFWKKkmXk2xwQq1770eRQNVRSRvu+dCw6d915xOYrTNI33XPaVlJmmEAKgIpmAdd1AVNyFrKw2RkSzBGbd7CAnyFlNs6oCWM6DmYHmYehVNQ2DYpk9TYiF3RHVzGBkgNf1IDdgl0pffLPk4z11urXRvOKFUI+GtADvvY9iGbreRk1cmyBTNQ7T1YmRxRjNRFLu+x7AHHOZOZoddGUFLzKZoSP11xWh3vvIjAaz2UxzMhDJXTNqnXNGuL6+rqjAkFIq29dv9HJi9GbWtm2IsfSlEdETl85r6dEuV0Jly8y+G+Z1XRPlcorl7ZlmkRsaSwGA1PVmYtfZ2OJlAABoaAYCIglAsWz0WR5UVhMzywlp8+Y0mxue4Hu+FTTnIYHZ9pNnmH37LS8braw2zcg7DJXPptGVdb6KiFlEU07DUFfVzs5ebGpC857p+hKQstU/51xHP52OY4zeOzL6rpffMgzDqG4ODg5QTVXrEIuJkMH29vb29n7gUGbFVaQMYE4mEyLy3h8c7DlHiDhqW1f2j+U8VHGUkrSNL3AtS0Y0RECDnDMtCVV2jkXVOzZTUSNcvlz7prEFEQEqS6+BDKCMp0siR3/2R198wwsZV8TDUbjnFattxCee2r62u/OlB1Z2dv14FPp+89AWktu5cK1PXdPUoMJIkJMoSk6F4VwufTcEgOB4kfIwzESMHRIqmhEDZXFMkOblhS1RKDsi0pwQsWobCj7GGkCzaM5DVdUlNBFRbOrcD6Xr03ddnEzIFSUOYlVVzhHkJXoiXK63JQZmc0SguQTyMnSgmgmQEAwUwQBKq17KXFxR/ZfHUmoAimihOsWHj8CTn7aH/gTguL/l9e7Vb9p8yQvRsxzMZ3u7vWRm9N6zx9XVlXbsmmZJ+IJlBiAMKSVHMGrikiM3DUXcWP5xjgw8ABFoSjKkv/7AB1I/ROcFrDQHovNlMUEIoYQaIqiqymy5ro3BCkAtDtSORmrmyg/I4EzVWBS4yFsNIKUe0JsBGyKWzUgDaJndQ1BTLVIsNESHDmC5pwRQAUEVmElNRJSRFDBnJAfgb9b2Gi3m+sif0m2vAzwOxwdPxrO5v6w84qYesa9VNToUSLlPkI2QEH2eLRjz0A9ceTRo6jAk7bquaBu992SGRCG4pFI5RqY0iBt6ImLnNGUXKySQnNuqDsGJqWeWbASQup6D77p527bR+fl8niWLyGKxsEL+1XVNROxQNd95+0133H58kKH4LghU0QOYSiqJhgGJoViMWRktKhsIRC0bCJIZCICiKYEVNAhqNwK/GX7kg4/RxrPBO2q8yZOKCLjWHDsOK2v+plPNqZtg3PiNldWN9VDFdjyaTMaxqZiZGFZXpyoSQmAkIiID76iqKjPzYXkXhL5fLPERmkNSVczpL9/9R5cvX2ZCH1wBADfqQe99rHwIoSwoGY1GBfQU7WSMcTweV1UVY6S+73MutykICEoIJQqiCjPmPIAaIiiIaQZQy8ksAeSSCtHEQMpqloKzGKzcVkI15yGZDgZiZvk6hhBlgNV8/OTiyiV75EsE+wAThSlUDfugoxHcdBqbEXvXTMfMHBu/tbFWVX46ncTK1bUPvuQrJCIGILQQHCI6guCobSrH6Aiq4J2jGJxzDN28F+iH/dI6cuzJuxtrOHNS55yAlXZGCEHAyjmGEArIYEYKnq9P0oGpyypmVkafVBKqAWaE5XZFFZGcEUBFTDOYACiDmQmIEhioqCVVNclgUgxqGd0R0FQ1m+gnP/Qgx1fUd7+UThwz2DZoCCYwOQobq6AZDuZweN0dP15V9dqhTWYmT5OVKXvHzE0bY/TOExEwAZhF50rk5uujAHp9WQsZIGgeOmL76kf+rB+0dHHs+jqe0rWOlU+5JwNPzIDR+cJALMVZZlVV+dKw8OyqqlLNSJlJTUVE1MRMswyq2XuPICJZVYhQVZYztSpgaioIRgigRmBoYJLVcom+oGKihFBMT0VEZJYGBFA+Cs3N6cKjCE8ArBk4AE/rqzCdiPl4620wXW1vOb25tRoqT568Z2QyAwPx3rMrEjVDRCZSzQW1MHPZn+QIEITImhgCaIX0sT94d2mI8f/frSsQsYxqs0MRAc2AGqIrggEkUs1UiJucE5FTgdxn1WySTZQYAAzNyiYkBCUUtaH4o17/KJLUBoMEoESklsESmoJls6yqooOCmKQsQ5JBRKSXD//phxGOCK6EQ4fnF74C8AgAAq4MQwIVIgBJdmQL6rh60+nRdGJm6BAdrmyutG1DhE3TGIIPVc65gEkiKutSPTFY0pxAc2QqBRxp8pJ/9//8d/1iXjcVIoQQHC23yJRZjHJwrujWnDMRSamqg2hi78hMsgw5D8PsClhvJpIWJgKaLfeERoSSFqhilnIeNCeArNKbZpUEOZlmzAo5qfSaFqRZRDT3mgbISWWhKad51/d96ofUd33qUu7me/O/+vM/ZTim+Mzm8MsAaoQKwELYykkYxdhh3Yp5WNscr21ONyaxDswokmIdmNlM6zqGSN77IfXlrQ4yAAOAOiQw8+y856aKhTYgMBi6P/pPv9n3PSOqLimmguOXzJ1zen1qpXRYTdQRaxYHAHUdB+EqjrpZ5+IAQCgGAFkGFFNcVFUjkNCo/JasSoRZe++9ImLG0m+JFBQ1m5LBkjVESSld275ce3ewt7uzu22yUBlMc0q9I3jld7wJYBWAFWYEOwAOwFw9ynvXpLsUp1MAAOfi1toWCrsrzrn57gwdhlFz7fx+U4eDWecDSwYpN50K/rqJQYwkpmqUUlEUGGARB6Tf+9V//70/8U8ZoGlaW8Ly5fZNAIilzQGAiCklZgLRGCP+0x96NSL22Qeu6rpuRpPLly9WjvcX3e7u7jDkbui9910aDBhsYGBlMwFIaTxp29GICSK7LBJC6PueGMCIUFM2YjMxJGcgAEaAqlq0FIzmPUfmX/6t9yIowhWAGUAGEIWOwMPiCgAMl8/7WOHubtrb944vP35m58rl3GcE3r52zXG1s7u3P+uBMGd17AfJqsqFnhcFtGw0H1KflIiymnPBzBQpB/8D/9vPMKKPsZQ4N+7zUeL6UmCTl6UiIuLP/MjrASCLmZkaFpVx36fow7zr4PqtjsoNBkRTIStKp5NoWf8RoqgSkYiWMeRi20lyKZjg+lJ6FTFLhcwgMO99YPhX/+VDChcIwOAs2EyGnigQG/Rd2tmX8+errfV08TKlBKBXn356tru7d3XHBd91w/bOouvVRBWs3HQDANCWflRu5NFnmS0SlztxGCgCkUtZoar//j/76aqpTbTYl3zTciozE8mShL3LQ3LOETPx9VtmBc/egal6RymlMmdXEEYBnI4YEQgAzULwhAgmjpEIHCMTOAYmQLDSuudCB+MSE6nlQjrnYSjaqGEYuj79zA++weCogQgg4gjSMJw/a/v7QujXp8wOUvabGxAcN03btqPRaHVrfX19bTQaNVUYVR5QTbNz5AhBhRkdATN6Js/kCcd1VegVBCMDM3OOMPX/8Zd+JSuYLQm466Rz4RGAsbz95cR1WeTCjrmbL5g5qyCyiPR9GcJORMvxjGKriEAMIQRQi9GP27ZPyTM2VQOEu7u7qmCmTV27wBub6yp2/OiJy5cvLxaLK1euMHO/mPV9IqIYPYKJpOmkYbhkcDNBBXDJxejW1hY7uzjUvHnYnz5u5y4ImNs6BN2iPXJEJBegmFLKk2ZndxEHHtCYfJfFBwZDLeu4EY0M2HWd1p7EIOeM7BUgiYDCiNNXP/7pO1/+AkdkauTZrivmy4JLACDicjcnfPKB3796eWcyXtvt9tePHDq6eei+e7+0v7NL3r3/Pe/XIe/uLdpRtdi36Zj3d6/0uV9b3Xztm171ile9dOXoaZgvoE6ABLABEMEuG7a7n/q9lZf9890v/8bffOzTr/7xn1e7xR76v/mZ/6t276aqlbzHXANuGHQP/+n/uPPNP/yl3/m3W6vtkTf9awBQuIpwBuWqznaoXVGOBAy7T8PlywAA49oeewJNdUhD3/Wz+ZXL2zvbB7PZYr7oACjnjA4BqBRkntgIRaFPIopJ1MzQYFBRYEMyoGTw2r//PaduvbNM+cDylmpyoyd2nXRxuP/x/6OtWmA4e+5inmXsUtv4cpKK2TLubR+EyiMBAISqgqwAtvl9Pw6wDRBVLhOjwRrASQBEuAqwk84/2n/9wfZbfiE/+fv7EtZuPgLwCoOvIowN7gNIBongDoArAAbwgkf/+BcCy7EXvIiPfReAA5gJnGO4AKAAFUAE2Ifdy3D1Mozq9ORZ7jszQ+9o1p0/d/7a1YOdnb2Ucj8MJVYaoRmaGRVKxKwXVcGUtcTvrJYV1GiQTOw7oO//mZ9q6jEAIBMYmRnx0jGJKA8JEWm00oLLw9B579pJ3N3dr7w1rWPSynEMtLk5HjURDTyT5qFpYmQCUIBoAMQRwAA8ACJ4AASY+CPfNvqWN88//0v+1Heu3vw2gBcIPIhwAWAbAVXNrs4gPZH7bbMEkM6ee1rV8/YFsC8bAEBNsKVwBKA1EAMEGMP0iB06Agn8yhhCwHFL0QNm79g78o6IyDtXxeiYGZBv3JjDDBEjlbuXWPCMiJ4dGaRhERyDSUX22/+fX8l5qfEE1HJrULq+Xb8sXyEwAbAsQx5SjIyG3nuTjGjd/iIPS/oYQQnNMRqqUNl6TVm/DoCgMwOHwAoEYAJRcA/she0LX54feYfCZ8w+gXAWwMvOvbbYpcvb+cyF+QMPH3zhQXvs6wCLdto8/NjjT37tcb32EEBhomuEIwaVAiOUHWyC7WpyAqGGqoIY1dCwbJKGMq1FRKDKTITomMiUCG5sr2PS6L1zbKoGiZ3V0ZdArpK84W/98v+3T70pFsvi6w8zK7t7/n96U+CwcIacoAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<PIL.Image.Image image mode=RGB size=100x100 at 0x1B580288C88>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img1 = image.load_img('D:/python/Project0-Python-MachineLearning/Lesson108-knife/IMG_7921.jpg',target_size=(100,100))\n",
    "x1 = image.img_to_array(img1)\n",
    "x1 = np.expand_dims(x1,axis=0)\n",
    "img_data1 = preprocess_input(x1)\n",
    "classes = model.predict(img_data1)\n",
    "\n",
    "print(classes)\n",
    "img1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}