niclas_svanström_knn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, accuracy_score\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {},
   "outputs": [],
   "source": [
    "test1 = pd.read_csv('./data/test_sample.csv')\n",
    "train1 = pd.read_csv('./data/train_sample.csv')\n",
    "test2 = pd.read_csv('./data/test_sample_2.csv')\n",
    "train2 = pd.read_csv('./data/train_sample_2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Det första testdatasetet har 1000st datapunkter och 785st attribut\n",
      "Det första träningsdatasetet har 5000st datapunkter och 785st attribut\n",
      "Det andra testdatasetet har 1000st datapunkter och 785st attribut\n",
      "Det andra träningsdatasetet har 5000st datapunkter och 782st attribut\n"
     ]
    }
   ],
   "source": [
    "print(f\"Det första testdatasetet har {test1.shape[0]}st datapunkter och {test1.shape[1]}st attribut\")\n",
    "print(f\"Det första träningsdatasetet har {train1.shape[0]}st datapunkter och {train1.shape[1]}st attribut\")\n",
    "print(f\"Det andra testdatasetet har {test2.shape[0]}st datapunkter och {test2.shape[1]}st attribut\")\n",
    "print(f\"Det andra träningsdatasetet har {train2.shape[0]}st datapunkter och {train2.shape[1]}st attribut\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Vi ser direkt att det andra träningssetet har 3st färre attribut än dom övriga"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label       0\n",
       "pixel0      0\n",
       "pixel1      0\n",
       "pixel2      0\n",
       "pixel3      0\n",
       "           ..\n",
       "pixel779    0\n",
       "pixel780    0\n",
       "pixel781    0\n",
       "pixel782    0\n",
       "pixel783    0\n",
       "Length: 785, dtype: int64"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test1.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label       0\n",
       "pixel0      0\n",
       "pixel1      0\n",
       "pixel2      0\n",
       "pixel3      0\n",
       "           ..\n",
       "pixel779    0\n",
       "pixel780    0\n",
       "pixel781    0\n",
       "pixel782    0\n",
       "pixel783    0\n",
       "Length: 785, dtype: int64"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train1.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label       0\n",
       "pixel0      0\n",
       "pixel1      0\n",
       "pixel2      0\n",
       "pixel3      0\n",
       "           ..\n",
       "pixel779    0\n",
       "pixel780    0\n",
       "pixel781    0\n",
       "pixel782    0\n",
       "pixel783    0\n",
       "Length: 785, dtype: int64"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test2.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label       0\n",
       "pixel0      0\n",
       "pixel1      0\n",
       "pixel2      0\n",
       "pixel3      0\n",
       "           ..\n",
       "pixel776    0\n",
       "pixel777    0\n",
       "pixel778    0\n",
       "pixel779    0\n",
       "pixel780    0\n",
       "Length: 782, dtype: int64"
      ]
     },
     "execution_count": 273,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train2.isnull().sum()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Efter att ha kontrollerat om det saknas värden så ser jag även att attributen är pixlar och att som vi skrev tidigare så saknar det andra träningssetet 3 pixlar.\n",
    "### Med tanke på att träningsdatan saknar 3 pixlar så om vi tränar modellen så när vi sedan ska testa den med testdatan så kommer det inte fungera eftersom att det är olika många attribut. En till sak som inte fungerar är att om till 28x28 när vi vill rita ut siffrorna.\n",
    "### Jag väljer därför att gå vidare med det första paret av träning och testdata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [],
   "source": [
    "def image_show(i, data, label):\n",
    "    x = data.values[i] # get the vectorized image\n",
    "    x = x.reshape((28,28)) # reshape it into 28x28 format\n",
    "    print(f'The image label of index {i} is {label[i]}.')\n",
    "    plt.imshow(x, cmap='gray') # show the image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [],
   "source": [
    "def one_of_each(df):\n",
    "    unique_labels = df['label'].unique()\n",
    "    num_labels = len(unique_labels)\n",
    "    fig, axes = plt.subplots(nrows=math.ceil(num_labels/10), ncols=10, figsize=(16, 7))\n",
    "    for i, label in enumerate(unique_labels):\n",
    "        index = df[df.label == label].index[0]\n",
    "        x = df.drop(columns='label').values[index]\n",
    "        x = x.reshape((28,28))\n",
    "        axes[i].imshow(x, cmap='gray')\n",
    "        axes[i].set_title(label)\n",
    "        axes[i].axis('off')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = train1.drop(columns='label')\n",
    "train_labels = train1['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x700 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "one_of_each(train1)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ovanför så ser vi klasserna som finns i datasetet, det är siffror från 0-9 och jag har även ritat ut en av vardera."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 386,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(n_neighbors=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 387,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-39 {color: black;background-color: white;}#sk-container-id-39 pre{padding: 0;}#sk-container-id-39 div.sk-toggleable {background-color: white;}#sk-container-id-39 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-39 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-39 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-39 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-39 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-39 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-39 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-39 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-39 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-39 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-39 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-39 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-39 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-39 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-39 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-39 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-39 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-39 div.sk-item {position: relative;z-index: 1;}#sk-container-id-39 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-39 div.sk-item::before, #sk-container-id-39 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-39 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-39 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-39 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-39 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-39 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-39 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-39 div.sk-label-container {text-align: center;}#sk-container-id-39 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-39 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-39\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>KNeighborsClassifier()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-39\" type=\"checkbox\" checked><label for=\"sk-estimator-id-39\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">KNeighborsClassifier</label><div class=\"sk-toggleable__content\"><pre>KNeighborsClassifier()</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 387,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(train_data, train_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 388,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_data = test1.drop(columns='label')\n",
    "test_labels = test1['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 389,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.945"
      ]
     },
     "execution_count": 389,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.score(test_data,test_labels)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Vi testar vad som är bästa värdet på K, då det är bäst att ha en ojämn siffra så steppar vi i loopen."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [],
   "source": [
    "k_value = []\n",
    "scores = []\n",
    "for x in range(1,100,2):\n",
    "    knn = KNeighborsClassifier(n_neighbors=x)\n",
    "    knn.fit(train_data, train_labels)\n",
    "    score = knn.score(test_data,test_labels)\n",
    "    k_value.append(x)\n",
    "    scores.append(score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'K_value')"
      ]
     },
     "execution_count": 283,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAGxCAYAAACKvAkXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABJ1ElEQVR4nO3de1xUdf4/8NcAwqDcVAQER1EkSVNUVMLsopKkLmvmlqmbRrfVBcuoDAt105Su/Lxm5a7lekkqyb5p4hLeoggU1LygYpiQcpFUBlAGmDm/P4yJgeEy1zOX1/PxmMdjPfM5w/uc3ZWXn897PkciCIIAIiIiIjviIHYBRERERObGAERERER2hwGIiIiI7A4DEBEREdkdBiAiIiKyOwxAREREZHcYgIiIiMjuMAARERGR3XESuwBLpFKpcOXKFbi7u0MikYhdDhEREXWAIAioqqqCv78/HBzanuNhANLiypUrkMlkYpdBREREeiguLkavXr3aHMMApIW7uzuA2zfQw8ND5GqIiIioI+RyOWQymfr3eFsYgLRoXPby8PBgACIiIrIyHWlfYRM0ERER2R0GICIiIrI7DEBERERkdxiAiIiIyO4wABEREZHdYQAiIiIiu8MARERERHaHAYiIiIjsDgMQERER2R0GICIiIrI7DEBERERkdxiAiIiIyO7wYagW6LfrN3Hp95vtjuvdrTNk3TqboSIiIiLbwgBkYX6vViAy+RBq61XtjpV2csD3C8ehh7uLGSojIiKyHQxAFubEbzdQW6+CaydH9G5jdqf4+k3crFMi5+I1TB7S04wVEhERWT8GIAuTX1IFAJgwyBerHx/W6rjXvzqJbdlFyCu6zgBERESkIzZBW5j8EjkAIMTPo81xw3t3BQAcK7pu8pqIiIhsDQOQhTlbensG6M6e7m2OG97ndgA6dVkORYPS5HURERHZEgYgC1Jbr0Th1WoAwJ09254BCuzeGd26OKNOqcLpK3JzlEdERGQzGIAsyPmyKqgEoFsXZ/i0880uiUSCYTIvAEDeJS6DERER6YIByIKc/aMBOsTPHRKJpN3xjctgx4pumLIsIiIim8MAZEHO/NEA3d7yV6Nhvb0AAHlshCYiItIJA5AFOVva+A2wthugG4X28oKDBCiprEVJ5S1TlkZERGRTGIAshCAI6j2AOjoD1MXFSf11+bxLN0xVGhERkc1hALIQpfJaVN6qh6ODBMG+bh0+r3EZjPsBERERdRwDkIVo3AAxqEcXuDg5dvi8xg0R2QdERETUcQxAFkLX5a9G3BCRiIhIdwxAFqKjj8BojhsiEhER6Y4ByEJ09BEYzXFDRCIiIt0xAFkAXR6BoQ03RCQiItINA5AFKCir7vAjMLThhohERES6YQCyAH/2/3TsERjNcUNEIiIi3VhEAFq/fj0CAwMhlUoRHh6OnJycVsfW19dj2bJlCAoKglQqRWhoKNLS0lod/9Zbb0EikWDBggUmqNw48kt1ewRGc003ROQyGBERUftED0ApKSmIj4/H0qVLkZeXh9DQUERFRaG8vFzr+MTERHz00UdYu3Ytzpw5g7lz52Lq1Kk4duxYi7FHjhzBRx99hCFDhpj6MgzSdAZIX+plMDZCExERtUv0AJScnIxnn30WMTExGDhwID788EN07twZmzZt0jp+y5YteO211zBp0iT069cP8+bNw6RJk/D+++9rjKuursasWbOwceNGdO3a1RyXohdBEJp8A0y/GSCAGyISERHpQtQAVFdXh9zcXERGRqqPOTg4IDIyEllZWVrPUSgUkEqlGsdcXV2RmZmpcSw2NhaTJ0/W+GxLVCqvxY2btx+B0d+n44/AaI4bIhIREXWck5g/vKKiAkqlEr6+vhrHfX19cfbsWa3nREVFITk5Gffddx+CgoKQkZGB1NRUKJV//tLfsWMH8vLycOTIkQ7VoVAooFAo1H+Wy823oeDZP3aADurRBdJOHX8ERnONGyJeq6nD6Sty9YwQERERtST6EpiuVq9ejeDgYISEhMDZ2RlxcXGIiYmBg8PtSykuLsYLL7yAbdu2tZgpak1SUhI8PT3VL5lMZspL0HCmxLAG6EbcEJGIiKjjRA1A3t7ecHR0RFlZmcbxsrIy+Pn5aT2nR48e2LVrF2pqanDp0iWcPXsWbm5u6NevHwAgNzcX5eXlGD58OJycnODk5IRDhw5hzZo1cHJy0pgparRo0SJUVlaqX8XFxca/2FY09v/o+ggMbbghIhERUceIGoCcnZ0RFhaGjIwM9TGVSoWMjAxERES0ea5UKkVAQAAaGhqwc+dOTJkyBQAwfvx4nDx5EsePH1e/RowYgVmzZuH48eNwdGy5zOTi4gIPDw+Nl7nkq2eA9P8GWKPGb4IdYyM0ERFRm0TtAQKA+Ph4zJkzByNGjMCoUaOwatUq1NTUICYmBgAwe/ZsBAQEICkpCQCQnZ2Ny5cvY+jQobh8+TL+9a9/QaVSYeHChQAAd3d33HXXXRo/o0uXLujevXuL42Iz9BEYzTVuiHilshallbXw8+zYEiAREZG9ET0ATZ8+HVevXsWSJUtQWlqKoUOHIi0tTd0YXVRUpO7vAYDa2lokJiaisLAQbm5umDRpErZs2QIvLy+RrkB/hj4Co7kuLk4Y4OeB/BI58oquY9LgnkaokoiIyPaIHoAAIC4uDnFxcVrfO3jwoMaf77//fpw5c0anz2/+GZaicQdofR+Boc3w3l63A9AlBiAiIqLWWN23wGxJvpG+AdYUN0QkIiJqHwOQiBr3ADLkERjNcUNEIiKi9jEAiUQQBIMfgqpN44aIdUoVTl8x34aORERE1oQBSCTGegRGc9wQkYiIqH0MQCIx1iMwtFFviFh8w6ifS0REZCsYgETS+AgMY+wA3Zx6Q0TOABEREWnFACSSxkdgGLP/p1HzDRGJiIhIEwOQSIz5CIzmGjdEBPh1eCIiIm0YgERg7EdgaDP8j2UwNkITERG1xAAkAmM/AkMbbohIRETUOgYgEZjiERjNcUNEIiKi1jEAicAUj8BorumGiLm/chaIiIioKQYgEZjiERjNSSQS3BvsDQB4fscxFJRVmexnERERWRsGIDMz1SMwtHnjr4MwyN8DFdV1mLHxJ4YgIiKiPzAAmVmZXGGSR2Bo49XZGdueCWcIIiIiaoYByMwa+39M8QgMbRiCiIiIWmIAMrM/vwFm2uWvphiCiIiINDEAmVl+iekegdGWliEomyGIiIjsFgOQmZ1tfAiqCR6B0R7NEKRgCCIiIrvFAGRGtfVKFFbUAAAGmnkGqBFDEBEREQOQWV0or4ZSJZj0ERgdoS0EXfwjmBEREdkDBiAzOlNi+kdgdFRjCBrY83YIem/fOVHrISIiMicGIDMa0acrlvxlIB4f1VvsUgDcDkHvPRoKAPjfmVJUVCtEroiIiMg8GIDMqF8PNzw1pi/+GuovdilqA/09MFTmhXqlgC9zfxO7HCIiIrNgACLMDL89I/VZThFUKkHkaoiIiEyPAYjwlyE94e7ihEu/30RW4e9il0NERGRyDECEzs5OeHhYAABge06RyNUQERGZHgMQAQBm/NGY/b/TbIYmIiLbxwBEANgMTURE9oUBiNRm/jELtIPN0EREZOMYgEjtL6G3m6F/ZTM0ERHZOAYgUmMzNBER2QsGINLAZmgiIrIHDECkgc3QRERkDxiAqAU2QxMRka2ziAC0fv16BAYGQiqVIjw8HDk5Oa2Ora+vx7JlyxAUFASpVIrQ0FCkpaVpjNmwYQOGDBkCDw8PeHh4ICIiAnv37jX1ZdgMNkMTEZGtEz0ApaSkID4+HkuXLkVeXh5CQ0MRFRWF8vJyreMTExPx0UcfYe3atThz5gzmzp2LqVOn4tixY+oxvXr1wltvvYXc3FwcPXoU48aNw5QpU3D69GlzXZZVYzM0ERHZOokgCKKucYSHh2PkyJFYt24dAEClUkEmk2H+/PlISEhoMd7f3x+vv/46YmNj1cemTZsGV1dXbN26tdWf061bN7z77rt4+umn261JLpfD09MTlZWV8PDw0OOqrN+ZK3JMWvM9OjlKkLVoPLzdXMQuiYiIqE26/P4WdQaorq4Oubm5iIyMVB9zcHBAZGQksrKytJ6jUCgglUo1jrm6uiIzM1PreKVSiR07dqCmpgYRERHGK97GDfT3QCiboYmIyEaJGoAqKiqgVCrh6+urcdzX1xelpaVaz4mKikJycjIKCgqgUqmQnp6O1NRUlJSUaIw7efIk3Nzc4OLigrlz5+Krr77CwIEDtX6mQqGAXC7XeBEwi83QRERko0TvAdLV6tWrERwcjJCQEDg7OyMuLg4xMTFwcNC8lAEDBuD48ePIzs7GvHnzMGfOHJw5c0brZyYlJcHT01P9kslk5rgUi/eX0J5wYzM0ERHZIFEDkLe3NxwdHVFWVqZxvKysDH5+flrP6dGjB3bt2oWamhpcunQJZ8+ehZubG/r166cxztnZGf3790dYWBiSkpIQGhqK1atXa/3MRYsWobKyUv0qLi42zgVaudvN0P4A2AxNRES2RdQA5OzsjLCwMGRkZKiPqVQqZGRktNuvI5VKERAQgIaGBuzcuRNTpkxpc7xKpYJCoX1nYxcXF/VX5htfdBt3hiYiIlsk+hJYfHw8Nm7ciM2bNyM/Px/z5s1DTU0NYmJiAACzZ8/GokWL1OOzs7ORmpqKwsJCfP/993jooYegUqmwcOFC9ZhFixbh8OHD+PXXX3Hy5EksWrQIBw8exKxZs8x+fdZukL8nm6GJiMjmOIldwPTp03H16lUsWbIEpaWlGDp0KNLS0tSN0UVFRRr9PbW1tUhMTERhYSHc3NwwadIkbNmyBV5eXuox5eXlmD17NkpKSuDp6YkhQ4Zg3759ePDBB819eTZh5igZThTfwM7c3zD3/iCxyyEiIjKY6PsAWSLuA6Sp8lY9RryZjnqlgPQX70Owr7vYJREREbVgNfsAkXXwdO2Ee4N7AAD2nCxpZzQREZHlYwCiDpk8uCcAYM/PDEBERGT9GICoQyIH+qKTowQF5dUoKKsSuxwiIiKDMABRh3AZjIiIbAkDEHUYl8GIiMhWMABRh3EZjIiIbAUDEHUYl8GIiMhWMACRTiZxGYyIiGwAAxDp5EEugxERkQ1gACKdcBmMiIhsAQMQ6YzLYEREZO0YgEhnXAYjIiJrxwBEOuMyGBERWTsGINJL4zLYtwxARERkhRiASC+Ny2Dny7gMRkRE1ocBiPSi7zJYjaIBV6sUpiqrhfKqWtQ1qMz284iIyDowAJHedF0Gu/R7DR5MPoTwld8hPuU4Cq9Wm7I8HP31GiKS9uOlL06Y9OcQEZH1YQAivemyDHbp9xo8/vFPuFJZC5UApB67jMjkQyYNQpt+uAilSsDun6/g8o1bJvkZRERknRiASG8dXQZrDD8llbXo7+OG/z41CpF3+pg0CJVX1eJ/p8sAAIIApOQUGe2ziYjI+jEAkUHaWwZrHn62PxuO++7ogX/PGYlv4saYLAh9mfsbGlQCujg7AgBSjhajQcleICIiuo0BiAzS1jKYtvDj4y5Vvz+4l2erQejlL07gVp1Sr5pUKgE7cooBAK9PHohuXZxRJlfgwLmr+l8oERHZFAYgMkhry2DthZ+mtAWhL3N/w4aDF/Sq6YdfKlB07SbcpU6YOiwAj4b1AgBsz76k1+cREZHtYQAigzVfBtMl/DTVGIRWTR8KAPjkh19x42adzvV89ke/zyPDAuDq7IjpI2UAgIPnr7IZmoiIADAAkRE0XQb77kyZXuGnqb+G+iPEzx1VigZsyryo07lXqxTq5ucZ4b0BAP16uCGiX3c2QxMRkRoDEBms6TLYs1uOGhR+AMDBQYIFkcEAdJ8F+iK3GA0qAcN7eyHEz0N9fOYfYYjN0EREBDAAkZE0LoMJAgwKP40mDPTTeRaoafPzjFG9NT9vkC+boYmISI0BiIxiwiBf+HtKMbCnh8HhB9BvFujHX35XNz//ZYi/xnsuTo5shiYiIjUGIDIKD2knfP/qOHwzf4zB4aeRrrNA23NuB5vG5ufm2AxNRESNGIDIaBwdJHB0kBjt83SZBdLW/NycRjP0kWKj1UlERNaHAYgsWkdngRp3fh7WrPm5OXUz9JEiNkMTEdkxBiCyaB2ZBVKpBPXePzNHaZ/9acRmaCIiAhiAyAq0NwvUVvNzc2yGJiIigAGIrICDgwQvjG99Fqix+XlqK83PzbEZmoiIGIDIKkQN0j4L1LT5eWYrzc/NsRmaiIgYgMgqtDYL1NHm5+bYDE1EZN8sIgCtX78egYGBkEqlCA8PR05OTqtj6+vrsWzZMgQFBUEqlSI0NBRpaWkaY5KSkjBy5Ei4u7vDx8cHDz/8MM6dO2fqyyATaz4L1LT5ufnOz+1hMzQRkX0TPQClpKQgPj4eS5cuRV5eHkJDQxEVFYXy8nKt4xMTE/HRRx9h7dq1OHPmDObOnYupU6fi2LFj6jGHDh1CbGwsfvrpJ6Snp6O+vh4TJkxATU2NuS6LTKD5LNC3p0rUzc/R7TQ/N8dmaCIi+yYRBEEQs4Dw8HCMHDkS69atAwCoVCrIZDLMnz8fCQkJLcb7+/vj9ddfR2xsrPrYtGnT4Orqiq1bt2r9GVevXoWPjw8OHTqE++67r92a5HI5PD09UVlZCQ+Pji+rkOmpVAImrfkeZ0urIO3kgNp6FWZH9MGyKXfp/FmFV6sx7v1DkEiAzFfHIcDL1QQVExGRuejy+1vUGaC6ujrk5uYiMjJSfczBwQGRkZHIysrSeo5CoYBUqvmoBVdXV2RmZrb6cyorKwEA3bp1M0LVJKams0C19bd7d3Rd/mrEZmgiIvslagCqqKiAUqmEr6+vxnFfX1+UlpZqPScqKgrJyckoKCiASqVCeno6UlNTUVJSonW8SqXCggULcM899+Cuu7TPEigUCsjlco0XWa7GXiAAGNbbC3f21H+WrrEZekdOERQNSqPUR0RElk/0HiBdrV69GsHBwQgJCYGzszPi4uIQExMDBwftlxIbG4tTp05hx44drX5mUlISPD091S+ZTGaq8skIHBwkWP7wXRjk74GFUSEGfdaEQb7w85CivEqBz4/+ZqQKiYjI0okagLy9veHo6IiysjKN42VlZfDz89N6To8ePbBr1y7U1NTg0qVLOHv2LNzc3NCvX78WY+Pi4rB7924cOHAAvXr1arWORYsWobKyUv0qLuZyiKUbGdgNe56/FxFB3Q36HBcnR/xzbBAA4IMDFzgLRERkJ0QNQM7OzggLC0NGRob6mEqlQkZGBiIiIto8VyqVIiAgAA0NDdi5cyemTJmifk8QBMTFxeGrr77C/v370bdv3zY/y8XFBR4eHhovsh+PjZDBz0OKkspazgIREdkJ0ZfA4uPjsXHjRmzevBn5+fmYN28eampqEBMTAwCYPXs2Fi1apB6fnZ2N1NRUFBYW4vvvv8dDDz0ElUqFhQsXqsfExsZi69at2L59O9zd3VFaWorS0lLcusXHHlBL0k7mmwW6VlOH5P+dwxU+goOISFSiB6Dp06fjvffew5IlSzB06FAcP34caWlp6sbooqIijQbn2tpaJCYmYuDAgZg6dSoCAgKQmZkJLy8v9ZgNGzagsrISDzzwAHr27Kl+paSkmPvyyEqYaxborb35WLP/Av6xJRf13IGaiEg0ou8DZIm4D5B9+m/Wr1jy9Wn09JTi4CsPwMWp/Qer6kJeW4/wFRm4VX97humVqAGIHdvfqD+DiMieWc0+QESW5LERMvh6uJhsFujrY5dxq16JLn88sX7Vd+dxrrTK6D+HiIjaxwBE9AdpJ0f884HbMzLG7gUSBAHbsm8/t+ylCQMQeacP6pUCXv7iBJfCiIhEwABE1MT0kaaZBTpefANnS6vg4uSAR4YHYOXUwfCQOuHk5Up8fLjQaD+HiIg6hgGIqAlTzQI1PrV+8uCe8OrsDB8PKf7110EAuBRGRCQGBiCiZow9CySvrcc3J25/k3FG+J/PLZs6LIBLYUREImEAImrG2LNAjc3PwT5uGNGnq/q4RCLBCi6FERGJggGISIums0BfGDAL1LT5ecao3pBIJBrv+3IpjIhIFAxARFoYaxaoefOzNlOHBWB8CJfCiIjMiQGIqBWNs0BXDJgFat78rI1EIsHKR7gURkRkTgxARK0wdBaoteZnbbgURkRkXgxARG0wZBaotebn1nApjIjIfBiAiNrQfBZIXlvfofPaa37WpvlS2NqMAv0LJyKiNjEAEbVj+kgZenpKcaWyFnM25XQoBJ34rbLd5mdtmi6Frdl/Af/+nv1ARESmwABE1A5pJ0dsnD0CXp074VjRjQ6FoO3ZlwC03fzcmqnDAvD8+GAAwJt78hmCiIhMgAGIqAPuCvDE1qfDOxSCdGl+1kYikeDFyGCGICIiE2IAIuqgjoYgXZuftWEIIiIyLQYgIh20F4L0aX5uDUMQEZHpMAAR6aitEKRv83Nr1CFo3O1vojEEEREZBwMQkR5aC0GGND+3RiKR4MUH72AIIiIyIgYgIj01D0FP/CfHoObntjAEEREZFwMQkQGahqATxTcMbn5uC0MQEZHxMAARGahpCAKAmeGGNT+3hSGIiMg4nMQugMgW3BXgidR5o/HDhQo8Psq4y1/NNYYgAcDa/Rfw5p58AMAz9/Yz6c8lIrIlDEBERtKvhxv69XAzy8+SSCSIf/AOAAxBRET64BIYkZVqDEHzuRxGRKQzBiAiK8YQRESkHwYgIivHEEREpDsGICIbwBBERKQbBiAiG8EQRETUcQYFoLq6Opw7dw4NDQ3GqoeIDKAtBH18+BcIgmDQ51YrGnDgbDkUDUpjlElEJDq9AtDNmzfx9NNPo3Pnzhg0aBCKim4//Xr+/Pl46623jFogEemmeQha+e1ZTF6TiX2nS3UOQtWKBqw/cAFj3t6PmE+P4IXPjhscpoiILIFeAWjRokU4ceIEDh48CKlUqj4eGRmJlJQUoxVHRPppDEGvT7oTXZwdcaZEjn9sye1wEGoafN7ddw43bt5+2n3a6VJ883OJOS6BiMikJIIe/5zr06cPUlJScPfdd8Pd3R0nTpxAv379cOHCBQwfPhxyudwUtZqNXC6Hp6cnKisr4eHhIXY5RAa5XlOHf2cW4tMffkVN3e0lrIE9PfBCZDAmDPTVeGxHtaIBm3/8FRu/L1SHnn7eXfD8+GAUXq3Gmv0X0LVzJ/zvxfvRw91FlOshImqNLr+/9doJ+urVq/Dx8WlxvKamxmTPQCIi/XTt4oxXokLwzJh+6iDUOCPUGIRGB3XHf7MuaQ0+0aH+cHSQoK5BhfT8cuSXyLF41yls+Ptw/v+diKyWXjNA9913Hx599FHMnz8f7u7u+Pnnn9G3b1/Mnz8fBQUFSEtLM0WtZsMZILJl2maEOjlKUK+8/VdB8+DT1OkrlZiy7gc0qASsmTEMfw31N3v9RESt0eX3t149QCtXrsRrr72GefPmoaGhAatXr8aECRPwySefYMWKFTp91vr16xEYGAipVIrw8HDk5OS0Ora+vh7Lli1DUFAQpFIpQkNDW4Stw4cPIzo6Gv7+/pBIJNi1a5c+l0hksxpnhDJfHYfYsUHo4uyIeqWAft5dsGr6UKTH34+HhwW0CD8AMMjfE3F/NFcv/foUrlYpzF0+EZFR6BWAxowZgxMnTqChoQGDBw/G//73P/j4+CArKwthYWEd/pyUlBTEx8dj6dKlyMvLQ2hoKKKiolBeXq51fGJiIj766COsXbsWZ86cwdy5czF16lQcO3ZMPaampgahoaFYv369PpdGZDcag9APCePwTdyYNoNPU/98oD/u7OmB6zfrsXjXKX4rjIisks5LYPX19fjHP/6BxYsXo2/fvgb98PDwcIwcORLr1q0DAKhUKshkMsyfPx8JCQktxvv7++P1119HbGys+ti0adPg6uqKrVu3thgvkUjw1Vdf4eGHH9apLi6BEbWNS2FEZIlMugTWqVMn7Ny5U+/iGtXV1SE3NxeRkZF/FuPggMjISGRlZWk9R6FQaHztHgBcXV2RmZlpcD1E1HGD/D0RO5ZLYURkvfRaAnv44YcN7q2pqKiAUqmEr6+vxnFfX1+UlpZqPScqKgrJyckoKCiASqVCeno6UlNTUVJi2L4kCoUCcrlc40VEbYsdy6UwIrJeen0NPjg4GMuWLcMPP/yAsLAwdOnSReP9559/3ijFNbd69Wo8++yzCAkJgUQiQVBQEGJiYrBp0yaDPjcpKQlvvPGGkaoksg/OTg5479EhmLLuB/UGiVwKIyJrodfX4Nvq/ZFIJCgsbP8BjHV1dejcuTO+/PJLjR6dOXPm4MaNG/j6669bPbe2tha///47/P39kZCQgN27d+P06dNaa+lID5BCoYBC8ecUvlwuh0wmYw8QUQf8v/TzWJ1RwA0SiUh0Jt8I8eLFi3oV1pSzszPCwsKQkZGhDigqlQoZGRmIi4tr81ypVIqAgADU19dj586deOyxxwyqxcXFBS4u/EubSB+xY/vjf2fKuEEiEVkVvQJQU40TSPr8hRcfH485c+ZgxIgRGDVqFFatWoWamhrExMQAAGbPno2AgAAkJSUBALKzs3H58mUMHToUly9fxr/+9S+oVCosXLhQ/ZnV1dW4cOGC+s8XL17E8ePH0a1bN/Tu3duQSyUiLZovhb30+Ql4du7U5jmjg7zx4EDfNscQEZmS3gHov//9L959910UFBQAAO644w688soreOKJJzr8GdOnT8fVq1exZMkSlJaWYujQoUhLS1M3RhcVFcHB4c8+7draWiQmJqKwsBBubm6YNGkStmzZAi8vL/WYo0ePYuzYseo/x8fHA7i9tPbpp5/qe7lE1IbGb4WtzihA6rHL7Y7fknUJPyaMg4+HtN2xRESmoFcPUHJyMhYvXoy4uDjcc889AIDMzEysX78eb775Jl588UWjF2pO3AeISHcNShX+m3UJFdVtfyU+7XQpCq/W4JWoAeqv0hMRGYMuv7/1boJ+4403MHv2bI3jmzdvxr/+9S+j9AiJiQGIyHS+OFqMV778Gb26uuLwK2Ph0M7O00REHWXyZ4GVlJRg9OjRLY6PHj3a4D15iMi2/WWIP9ylTvjt+i1kXqgQuxwislN6BaD+/fvj888/b3E8JSUFwcHBBhdFRLbL1dkRjwwLAABszy4SuRoisld6NUG/8cYbmD59Og4fPqzuAfrhhx+QkZGhNRgRETU1I7w3Nmddwnf5ZSiX17IZmojMTq8ZoGnTpiE7Oxve3t7YtWsXdu3aBW9vb+Tk5GDq1KnGrpGIbEyInweG9/ZCg0rAF7m/iV0OEdkhvZqgbR2boIlMr7EZWtbNFYdeZjM0ERnO5E3Q3377Lfbt29fi+L59+7B37159PpKI7ExjM3TxNTZDE5H56RWAEhISoFQqWxwXBAEJCQkGF0VEto/N0EQkJr0CUEFBAQYOHNjieEhIiMZjKIiI2jIj/PbjaRqboYmIzEWvAOTp6an1ie8XLlxAly5dDC6KiOwDm6GJSCx6BaApU6ZgwYIF+OWXX9THLly4gJdeegl//etfjVYcEdm+GaNuzwLtOFIElYrfySAi89ArAL3zzjvo0qULQkJC0LdvX/Tt2xchISHo3r073nvvPWPXSEQ2jM3QRCQGvTZC9PT0xI8//oj09HScOHECrq6uCA0Nxb333mvs+ojIxjU2Q2/OuoTPcopw3x09xC6JiOyATjNAWVlZ2L17NwBAIpFgwoQJ8PHxwXvvvYdp06bhueeeg0LR9pOgiYiaa2yGTj9ThvIqNkMTkenpFICWLVuG06dPq/988uRJPPvss3jwwQeRkJCAb775BklJSUYvkohsm0Yz9FE2QxOR6ekUgI4fP47x48er/7xjxw6MGjUKGzduRHx8PNasWcNngRGRXtgMTUTmpFMAun79Onx9fdV/PnToECZOnKj+88iRI1FcXGy86ojIbrAZmojMSacA5Ovri4sXLwIA6urqkJeXh7vvvlv9flVVFTp16mTcConILjTdGfqzHO4MTUSmpVMAmjRpEhISEvD9999j0aJF6Ny5s8Y3v37++WcEBQUZvUgisg9shiYic9EpAC1fvhxOTk64//77sXHjRmzcuBHOzs7q9zdt2oQJEyYYvUgisg9shiYic9FpHyBvb28cPnwYlZWVcHNzg6Ojo8b7X3zxBdzc3IxaIBHZlxmjeiOv6AZ2HCnC3+/uA09XLqsTkfHp/Syw5uEHALp166YxI0REpKu/DPGHxx/N0GPe3o9V351H5a16scsiIhujVwAiIjIVV2dHfDx7BIJ93FBV24BV3xUwCBGR0UkEQeCGG83I5XJ4enqisrISHh4eYpdDZJdUKgHfnirB6u8KUFBeDQBwlzrh6TF9EXNPXy6NEVELuvz+ZgDSggGIyHIwCBFRRzEAGYgBiMjyaAtCHlInbPh7GO7p7y1ydURkCXT5/c0eICKyCg4OEvxliD/2LbgP62YOQ7CPG+S1DXjq0yP4gTtHE5GOGICIyKo0BqHdz4/B+BAfKBpUDEFEpDMGICKySi5Ojvjg78MZgohILwxARGS1GIKISF8MQERk1RiCiEgfDEBEZPUYgohIVwxARGQTGIKISBcMQERkMxiCiKijGICIyKZoC0Hbs4tQr1SJXRoRWRCLCEDr169HYGAgpFIpwsPDkZOT0+rY+vp6LFu2DEFBQZBKpQgNDUVaWppBn0lEtqV5CHrtq5MY+95BpBxhECKi20QPQCkpKYiPj8fSpUuRl5eH0NBQREVFoby8XOv4xMREfPTRR1i7di3OnDmDuXPnYurUqTh27Jjen0lEtsfFyREb/h6GxMl3wtvNBb9dv4VXdzIIEdFtoj8LLDw8HCNHjsS6desAACqVCjKZDPPnz0dCQkKL8f7+/nj99dcRGxurPjZt2jS4urpi69aten1mc3wWGJFtuVWnxLbsS/jwUCEqqhUAgF5dXTF/XH88MrwXOjmK/m9BIjICq3kWWF1dHXJzcxEZGak+5uDggMjISGRlZWk9R6FQQCqVahxzdXVFZmam3p9JRLbN1dkRz9zbD98vHMsZISICIHIAqqiogFKphK+vr8ZxX19flJaWaj0nKioKycnJKCgogEqlQnp6OlJTU1FSUqL3ZyoUCsjlco0XEdmetoLQ858da/8DiMhmWN287+rVqxEcHIyQkBA4OzsjLi4OMTExcHDQ/1KSkpLg6empfslkMiNWTESWpnkQcnSQYO+pUuQVXRe7NCIyE1EDkLe3NxwdHVFWVqZxvKysDH5+flrP6dGjB3bt2oWamhpcunQJZ8+ehZubG/r166f3Zy5atAiVlZXqV3FxsRGujogsXWMQemRYAABg9XcFIldEROYiagBydnZGWFgYMjIy1MdUKhUyMjIQERHR5rlSqRQBAQFoaGjAzp07MWXKFL0/08XFBR4eHhovIrIfceP6w9FBgkPnr3IWiMhOiL4EFh8fj40bN2Lz5s3Iz8/HvHnzUFNTg5iYGADA7NmzsWjRIvX47OxspKamorCwEN9//z0eeughqFQqLFy4sMOfSUTUVJ/uXTgLRGRnnMQuYPr06bh69SqWLFmC0tJSDB06FGlpaeom5qKiIo3+ntraWiQmJqKwsBBubm6YNGkStmzZAi8vrw5/JhFRc3Hj+iP12GX1LNDw3l3FLomITEj0fYAsEfcBIrJPr3xxAl/k/ob77+iBzU+NErscItKR1ewDRERkSdgLRGQ/GICIiP7AXiAi+8EARETUBGeBiOwDAxARUROcBSKyDwxARETNcBaIyPYxABERNcNZICLbxwBERKSFqWaBahQN+Knwd3AHEiJxMQAREWlhilmg36sVmLbhRzz+8U/Yll1klM8kIv0wABERtcKYs0C/Vysw69/ZOFtaBQBYt/8CauuVxiiTiPTAAERE1ApjzQI1DT8+7i7wcXdBqbwWnx8tNlapRKQjBiAiojYYOgvUPPx89tzdmD+uPwDggwO/cBaISCQMQEREbWg6C7RiTz5KKm91+Fxt4SeohxseGymDn4eUs0BEImIAIiJqR9y4/nB2dEDupeu4/52DWPL1qXaDUGvhBwBcnBwROzYIAGeBiMTCAERE1I4+3bvgs+fuxqi+3VCnVOG/WZfaDEJthZ9GnAUiEhcDEBFRB4T16YqU5+7G9mfDWwShpV+fQmllLYCOhR+As0BEYpMI3I2rBblcDk9PT1RWVsLDw0PscojIwgiCgKzC37EqvQA5v14DADg7OuDxUTLkXLzWbvhppGhQ4v53DqJUXotlUwZhdkSgma6AyDbp8vubM0BERDqSSCQYHeSNlH/8MSMU+OeMUEfDD9ByFkjRwFkgInNhACIi0lPzIDQ6qDtC/Nw7FH4aafQCHWEvEJG5cAlMCy6BEZE5bcn6FYu/Pg0/DykOLXwALk6OYpdEZJW4BEZEZEU4C0RkfgxAREQia9oLtJ69QERmwQBERGQBOAtEZF4MQEREFoCzQETmxQBERGQhOAtEZD4MQEREFoKzQETm4yR2AURE9KfHRsqw/sAvKJXXYuSb38HJsfV/p0oAhPfrhvnjgnFnT27ZQaQL7gOkBfcBIiIxfX60GAu//Fmncybe5YfnxzMIkX3T5fc3A5AWDEBEJLbfrt/Erbq2l8DktQ349MdfsfvnK2j8m5xBiOwZA5CBGICIyJqcL6vCmowC7DlZwiBEdo0ByEAMQERkjVoLQi9NGID+Ph17NhmRNWMAMhADEBFZs+ZByNO1Ew6+/AC6dnEWuzQik+KzwIiI7Ngdvu5YN3M49i24D/193FB5qx7/ybwodllEFoUBiIjIRt3h645XogYAAD798Vdcr6kTuSIiy8EARERkwyYM9MXAnh6oVjRwFoioCQYgIiIbJpFI8EJkMADOAhE1JXoAWr9+PQIDAyGVShEeHo6cnJw2x69atQoDBgyAq6srZDIZXnzxRdTW1qrfr6qqwoIFC9CnTx+4urpi9OjROHLkiKkvg4jIYnEWiKglUQNQSkoK4uPjsXTpUuTl5SE0NBRRUVEoLy/XOn779u1ISEjA0qVLkZ+fj//85z9ISUnBa6+9ph7zzDPPID09HVu2bMHJkycxYcIEREZG4vLly+a6LCIii8JZIKKWRP0afHh4OEaOHIl169YBAFQqFWQyGebPn4+EhIQW4+Pi4pCfn4+MjAz1sZdeegnZ2dnIzMzErVu34O7ujq+//hqTJ09WjwkLC8PEiRPx5ptvdqgufg2eiGyNIAiYvCYTZ0rkiBvbHy//0RxNZEus4mvwdXV1yM3NRWRk5J/FODggMjISWVlZWs8ZPXo0cnNz1ctkhYWF+PbbbzFp0iQAQENDA5RKJaRSqcZ5rq6uyMzMNNGVEBFZPs4CEWkSLQBVVFRAqVTC19dX47ivry9KS0u1njNz5kwsW7YMY8aMQadOnRAUFIQHHnhAvQTm7u6OiIgILF++HFeuXIFSqcTWrVuRlZWFkpKSVmtRKBSQy+UaLyIiW8NeIKI/id4ErYuDBw9i5cqV+OCDD5CXl4fU1FTs2bMHy5cvV4/ZsmULBEFAQEAAXFxcsGbNGsyYMQMODq1falJSEjw9PdUvmUxmjsshIjIrzgIR/Um0AOTt7Q1HR0eUlZVpHC8rK4Ofn5/WcxYvXownnngCzzzzDAYPHoypU6di5cqVSEpKgkqlAgAEBQXh0KFDqK6uRnFxMXJyclBfX49+/fq1WsuiRYtQWVmpfhUXFxvvQomILAhngYhuEy0AOTs7IywsTKOhWaVSISMjAxEREVrPuXnzZouZHEdHRwC3G/ya6tKlC3r27Inr169j3759mDJlSqu1uLi4wMPDQ+NFRGSLOAtEdJuoS2Dx8fHYuHEjNm/ejPz8fMybNw81NTWIiYkBAMyePRuLFi1Sj4+OjsaGDRuwY8cOXLx4Eenp6Vi8eDGio6PVQWjfvn1IS0tTvz927FiEhISoP5OIyN5xFogIcBLzh0+fPh1Xr17FkiVLUFpaiqFDhyItLU3dGF1UVKQx45OYmAiJRILExERcvnwZPXr0QHR0NFasWKEeU1lZiUWLFuG3335Dt27dMG3aNKxYsQKdOnUy+/UREVmixlmgf2zJxac//oqnx/Tlk+LJ7oi6D5Cl4j5ARGTruC8Q2SKr2AeIiIjEw14gsncMQEREdoq9QGTPGICIiOxU01mgT364iOJrN0WuiMh8GICIiOzYhIG+GNLLEzV1Sjz+8U8MQWQ3GICIiOyYRCLBv2ePQD/vLrh84xZDENkNBiAiIjvn4yHFjufuZggiu8IAREREDEFkdxiAiIgIAEMQ2RcGICIiUmMIInvBAERERBoYgsgeMAAREVELDEFk6xiAiIhIK20h6McLFeAjJMkWMAAREVGrmoegmf/OxvSPf8KPvzAIkXVjACIiojb5eEjxxdwIPDk6EM5ODsi5eA0zNzIIkXWTCPxfbgtyuRyenp6orKyEh4eH2OUQEVmM0spafHjoF2zPKUJdgwoAMKpvNyyIDEZEv+6QSCQiV0j2TJff3wxAWjAAERG1jUGILBEDkIEYgIiIOqa0shYbDl7AZznFqFMyCJG4GIAMxABERKQbBiGyBAxABmIAIiLSD4MQiYkByEAMQEREhmEQIjEwABmIAYiIyDgYhMicGIAMxABERGRcDEJkDgxABmIAIiIyDQYhMiUGIAMxABERmVZJ5S18ePAXBiEyKgYgAzEAERGZB4MQGRMDkIEYgIiIzItBiIyBAchADEBEROJgECJDMAAZiAGIiEhcbQWh0UHeIldHlooByEAMQERElqGk8hY2HPwFO5oEofC+3bAg8g5EBHUXuTqyNAxABmIAIiKyLK0FoZcmDMCovt1Ero4sBQOQgRiAiIgsU/MgJJEAW54Kx5hgLouRbr+/HcxUExERkcF6erpi2ZS7cGjhA4ga5AtBAF7d+TOqauvFLo2sDAMQERFZnZ6erkh+bChk3Vxx+cYtJO09K3ZJZGUYgIiIyCp1cXHCO9NCAQDbs4uQWVAhckVkTRiAiIjIakUEdcfsiD4AuBRGuhE9AK1fvx6BgYGQSqUIDw9HTk5Om+NXrVqFAQMGwNXVFTKZDC+++CJqa2vV7yuVSixevBh9+/aFq6srgoKCsHz5crDXm4jINr36UAiXwkhnogaglJQUxMfHY+nSpcjLy0NoaCiioqJQXl6udfz27duRkJCApUuXIj8/H//5z3+QkpKC1157TT3m7bffxoYNG7Bu3Trk5+fj7bffxjvvvIO1a9ea67KIiMiMuBRG+hA1ACUnJ+PZZ59FTEwMBg4ciA8//BCdO3fGpk2btI7/8ccfcc8992DmzJkIDAzEhAkTMGPGDI1Zox9//BFTpkzB5MmTERgYiL/97W+YMGFCuzNLRERkvbgURroSLQDV1dUhNzcXkZGRfxbj4IDIyEhkZWVpPWf06NHIzc1Vh5nCwkJ8++23mDRpksaYjIwMnD9/HgBw4sQJZGZmYuLEiSa8GiIiEps9LoWxvUN/TmL94IqKCiiVSvj6+moc9/X1xdmz2v+HO3PmTFRUVGDMmDEQBAENDQ2YO3euxhJYQkIC5HI5QkJC4OjoCKVSiRUrVmDWrFmt1qJQKKBQKNR/lsvlBl4dERGZW+NS2IyNP2F7dhEm3dXTZjdIPF9WhdUZBdh7sgSJkwfiqTF9xS7J6ojeBK2LgwcPYuXKlfjggw+Ql5eH1NRU7NmzB8uXL1eP+fzzz7Ft2zZs374deXl52Lx5M9577z1s3ry51c9NSkqCp6en+iWTycxxOUREZGS2vhR2vqwKsdvzELXqMPb8XAKVAHxw8ALqGlRil2Z1RHsURl1dHTp37owvv/wSDz/8sPr4nDlzcOPGDXz99dctzrn33ntx9913491331Uf27p1K5577jlUV1fDwcEBMpkMCQkJiI2NVY958803sXXr1lZnlrTNAMlkMj4Kg4jICtUoGvDQ6sMovnYLM8N7Y+XUwWKXZLDGGZ9vT5ag8bf2xLv8kHvpOsqrFFg/czgmD+kpbpEWwCoeheHs7IywsDBkZGSoj6lUKmRkZCAiIkLrOTdv3oSDg2bJjo6OAP5cB21tjErVejp2cXGBh4eHxouIiKxTFxcnvD1tCADr/1ZY8xkfQbgdfPa+cC82/D0M00feXrH4LKdI5Eqtj2g9QAAQHx+POXPmYMSIERg1ahRWrVqFmpoaxMTEAABmz56NgIAAJCUlAQCio6ORnJyMYcOGITw8HBcuXMDixYsRHR2tDkLR0dFYsWIFevfujUGDBuHYsWNITk7GU089Jdp1EhGReY0O8sYTd/fBlp8u4dWdP+Or2NHwcZeKXVaHtTbj8/z4YNzZ889/pE8fKcO6AxeQeaECv1bUINC7i0gVWx9RA9D06dNx9epVLFmyBKWlpRg6dCjS0tLUjdFFRUUaszmJiYmQSCRITEzE5cuX0aNHD3XgabR27VosXrwY//znP1FeXg5/f3/84x//wJIlS8x+fUREJJ6EiSE4eL4cxddu4b53DmB2RCCeu68fvN1cxC6tVefLqrAmowB72gk+jXp17Yz77+iBg+euYseRYiRMDDFzxdZLtB4gS6bLGiIREVmu/BI5ElJP4kTxDQCAtJODRQYhXYNPU/tOl+IfW3Lh7eaMHxPGw9nJqr7fZFS6/P5mANKCAYiIyHYIgoCD569i1XcF6iDk2skRT0T0ET0IGRJ8GjUoVRj91n42Q4MByGAMQEREtseSgpAxgk9T7//vHNbuv4Ax/b2x9ZlwI1drPRiADMQARERku9RBKP08TvxWCQBwcXIwawC6UnnLKMGn0W/Xb+Ledw5AEIBDrzyAPt3tsxlal9/fojZBExERmZtEIsHYAT544I/m4VXf3Q5Cl2/cMmsdxgg+jZo2Q3+Ww2bojmAAIiIiuySRSDA2xAcPDOiBgvJq3KpTmu1nd3dzRq+unY36mTNG9cbBc1fxZW4x4h+8w66boTuCAYiIiOyaRCLBHb7uYpdhsHEhPvBxd0F5lQLf5Zdh0mD7bYbuCMZDIiIiG9DJ0UG9M/T2bO4M3R4GICIiIhvx2AgZJBIg80IFLv1eI3Y5Fo0BiIiIyEbIunXGfcE9AACf5RSLXI1lYwAiIiKyITPDewMAvswtRl1D6w8Ct3cMQERERDaksRm6oroO3+WXiV2OxWIAIiIisiG6NkNfrVIg+X/nsOHgL1Cp7GdvZH4NnoiIyMY8NkKGdQcuqJuhte0MfbVKgY8P/4ItP11Cbf3tpbKiazex4uG74OAgMXfJZscZICIiIhvTtBl6xxHNZuirVQqs2HMG976zHxu/v4jaehVC/NzhIAE+yynC67tO2cVMEAMQERGRDWpshv7i6O1maG3BZ6jMC5/GjMTeF+7F+4+FQmJHIYhLYERERDao6c7Qc7fm4sdfKtRLXaEyL7wYGYz77+gBieT2ctfUYb0AAPGfn8BnObd7h2x5OYwBiIiIyAZ1cnRQ9wLtP1sO4HbwWRAZjAeaBJ+m7CkEMQARERHZqCci+uDbkyXw6twJ88e3Hnyaah6CJBLgzSm2F4IYgIiIiGyUr4cU+19+QOfzmoagxq/S21oIYgAiIiKiFmw9BDEAERERkVbNQ1DlzXrcFeDZ5jlenTthylB/dHa27Ihh2dURERGRqJqGoD0nS7DnZEm753yVdxmfxIxEFxfLjRmWWxkRERFZhKnDeqFbFxd8+3MJVELr+wMJAPadKkXOr9cQ88kRiw5BEkFo40rslFwuh6enJyorK+Hh4SF2OURERFbjePENPPHvbFQpGjAqsJtZQ5Auv7+5EzQREREZzVCZF7Y8Ew53Fyf1TFCNokHsslpgACIiIiKjsoYQxABERERERmfpIYgBiIiIiEzCkkMQAxARERGZjKWGIAYgIiIiMilLDEEMQERERGRyzUPQ05uPQKUSbyceBiAiIiIyi8YQ5OnaCX8Lk4n6XDHL3J6RiIiIbNJQmRcOvzIWnp07iVoHZ4CIiIjIrMQOPwADEBEREdkhBiAiIiKyOxYRgNavX4/AwEBIpVKEh4cjJyenzfGrVq3CgAED4OrqCplMhhdffBG1tbXq9wMDAyGRSFq8YmNjTX0pREREZAVEb4JOSUlBfHw8PvzwQ4SHh2PVqlWIiorCuXPn4OPj02L89u3bkZCQgE2bNmH06NE4f/48nnzySUgkEiQnJwMAjhw5AqVSqT7n1KlTePDBB/Hoo4+a7bqIiIjIckkEQRDvS/gAwsPDMXLkSKxbtw4AoFKpIJPJMH/+fCQkJLQYHxcXh/z8fGRkZKiPvfTSS8jOzkZmZqbWn7FgwQLs3r0bBQUFkEja/8qdXC6Hp6cnKisr4eHhoeeVERERkTnp8vtb1CWwuro65ObmIjIyUn3MwcEBkZGRyMrK0nrO6NGjkZubq14mKywsxLfffotJkya1+jO2bt2Kp556qtXwo1AoIJfLNV5ERERku0RdAquoqIBSqYSvr6/GcV9fX5w9e1brOTNnzkRFRQXGjBkDQRDQ0NCAuXPn4rXXXtM6fteuXbhx4waefPLJVutISkrCG2+8ofd1EBERkXWxiCZoXRw8eBArV67EBx98gLy8PKSmpmLPnj1Yvny51vH/+c9/MHHiRPj7+7f6mYsWLUJlZaX6VVxcbKryiYiIyAKIOgPk7e0NR0dHlJWVaRwvKyuDn5+f1nMWL16MJ554As888wwAYPDgwaipqcFzzz2H119/HQ4Of2a6S5cu4bvvvkNqamqbdbi4uMDFxcXAqyEiIiJrIeoMkLOzM8LCwjQamlUqFTIyMhAREaH1nJs3b2qEHABwdHQEADTv5/7kk0/g4+ODyZMnG7lyIiIismaifw0+Pj4ec+bMwYgRIzBq1CisWrUKNTU1iImJAQDMnj0bAQEBSEpKAgBER0cjOTkZw4YNQ3h4OC5cuIDFixcjOjpaHYSA20Hqk08+wZw5c+DkJPplEhERkQURPRlMnz4dV69exZIlS1BaWoqhQ4ciLS1N3RhdVFSkMeOTmJgIiUSCxMREXL58GT169EB0dDRWrFih8bnfffcdioqK8NRTT5n1eoiIiMjyib4PkCXiPkBERETWR5ff36LPAFmixkzI/YCIiIisR+Pv7Y7M7TAAaVFVVQUAkMlkIldCREREuqqqqoKnp2ebY7gEpoVKpcKVK1fg7u7eoUdntEYul0Mmk6G4uJhLaWbA+21evN/mxfttXrzf5mWs+y0IAqqqquDv79/iG+PNcQZICwcHB/Tq1cton+fh4cH/A5kR77d58X6bF++3efF+m5cx7nd7Mz+NrG4naCIiIiJDMQARERGR3WEAMiEXFxcsXbqUj9kwE95v8+L9Ni/eb/Pi/TYvMe43m6CJiIjI7nAGiIiIiOwOAxARERHZHQYgIiIisjsMQCa0fv16BAYGQiqVIjw8HDk5OWKXZPWSkpIwcuRIuLu7w8fHBw8//DDOnTunMaa2thaxsbHo3r073NzcMG3aNJSVlYlUsW156623IJFIsGDBAvUx3m/junz5Mv7+97+je/fucHV1xeDBg3H06FH1+4IgYMmSJejZsydcXV0RGRmJgoICESu2XkqlEosXL0bfvn3h6uqKoKAgLF++XOMxCrzfhjl8+DCio6Ph7+8PiUSCXbt2abzfkft77do1zJo1Cx4eHvDy8sLTTz+N6upqg2tjADKRlJQUxMfHY+nSpcjLy0NoaCiioqJQXl4udmlW7dChQ4iNjcVPP/2E9PR01NfXY8KECaipqVGPefHFF/HNN9/giy++wKFDh3DlyhU88sgjIlZtG44cOYKPPvoIQ4YM0TjO+208169fxz333INOnTph7969OHPmDN5//3107dpVPeadd97BmjVr8OGHHyI7OxtdunRBVFQUamtrRazcOr399tvYsGED1q1bh/z8fLz99tt45513sHbtWvUY3m/D1NTUIDQ0FOvXr9f6fkfu76xZs3D69Gmkp6dj9+7dOHz4MJ577jnDixPIJEaNGiXExsaq/6xUKgV/f38hKSlJxKpsT3l5uQBAOHTokCAIgnDjxg2hU6dOwhdffKEek5+fLwAQsrKyxCrT6lVVVQnBwcFCenq6cP/99wsvvPCCIAi838b26quvCmPGjGn1fZVKJfj5+Qnvvvuu+tiNGzcEFxcX4bPPPjNHiTZl8uTJwlNPPaVx7JFHHhFmzZolCALvt7EBEL766iv1nztyf8+cOSMAEI4cOaIes3fvXkEikQiXL182qB7OAJlAXV0dcnNzERkZqT7m4OCAyMhIZGVliViZ7amsrAQAdOvWDQCQm5uL+vp6jXsfEhKC3r17894bIDY2FpMnT9a4rwDvt7H93//9H0aMGIFHH30UPj4+GDZsGDZu3Kh+/+LFiygtLdW4356enggPD+f91sPo0aORkZGB8+fPAwBOnDiBzMxMTJw4EQDvt6l15P5mZWXBy8sLI0aMUI+JjIyEg4MDsrOzDfr5fBaYCVRUVECpVMLX11fjuK+vL86ePStSVbZHpVJhwYIFuOeee3DXXXcBAEpLS+Hs7AwvLy+Nsb6+vigtLRWhSuu3Y8cO5OXl4ciRIy3e4/02rsLCQmzYsAHx8fF47bXXcOTIETz//PNwdnbGnDlz1PdU298tvN+6S0hIgFwuR0hICBwdHaFUKrFixQrMmjULAHi/Tawj97e0tBQ+Pj4a7zs5OaFbt24G/3fAAERWKzY2FqdOnUJmZqbYpdis4uJivPDCC0hPT4dUKhW7HJunUqkwYsQIrFy5EgAwbNgwnDp1Ch9++CHmzJkjcnW25/PPP8e2bduwfft2DBo0CMePH8eCBQvg7+/P+20HuARmAt7e3nB0dGzxTZiysjL4+fmJVJVtiYuLw+7du3HgwAH06tVLfdzPzw91dXW4ceOGxnjee/3k5uaivLwcw4cPh5OTE5ycnHDo0CGsWbMGTk5O8PX15f02op49e2LgwIEax+68804UFRUBgPqe8u8W43jllVeQkJCAxx9/HIMHD8YTTzyBF198EUlJSQB4v02tI/fXz8+vxZeHGhoacO3aNYP/O2AAMgFnZ2eEhYUhIyNDfUylUiEjIwMREREiVmb9BEFAXFwcvvrqK+zfvx99+/bVeD8sLAydOnXSuPfnzp1DUVER770exo8fj5MnT+L48ePq14gRIzBr1iz1f+b9Np577rmnxbYO58+fR58+fQAAffv2hZ+fn8b9lsvlyM7O5v3Ww82bN+HgoPlr0NHRESqVCgDvt6l15P5GRETgxo0byM3NVY/Zv38/VCoVwsPDDSvAoBZqatWOHTsEFxcX4dNPPxXOnDkjPPfcc4KXl5dQWloqdmlWbd68eYKnp6dw8OBBoaSkRP26efOmeszcuXOF3r17C/v37xeOHj0qRERECBERESJWbVuafgtMEHi/jSknJ0dwcnISVqxYIRQUFAjbtm0TOnfuLGzdulU95q233hK8vLyEr7/+Wvj555+FKVOmCH379hVu3bolYuXWac6cOUJAQICwe/du4eLFi0Jqaqrg7e0tLFy4UD2G99swVVVVwrFjx4Rjx44JAITk5GTh2LFjwqVLlwRB6Nj9feihh4Rhw4YJ2dnZQmZmphAcHCzMmDHD4NoYgExo7dq1Qu/evQVnZ2dh1KhRwk8//SR2SVYPgNbXJ598oh5z69Yt4Z///KfQtWtXoXPnzsLUqVOFkpIS8Yq2Mc0DEO+3cX3zzTfCXXfdJbi4uAghISHCxx9/rPG+SqUSFi9eLPj6+gouLi7C+PHjhXPnzolUrXWTy+XCCy+8IPTu3VuQSqVCv379hNdff11QKBTqMbzfhjlw4IDWv7PnzJkjCELH7u/vv/8uzJgxQ3BzcxM8PDyEmJgYoaqqyuDa+DR4IiIisjvsASIiIiK7wwBEREREdocBiIiIiOwOAxARERHZHQYgIiIisjsMQERERGR3GICIiIjI7jAAERERkd1hACIiaiIwMBCrVq0SuwwiMjEGICKyKE8++SQefvhhjWNffvklpFIp3n//fXGKIiKb4yR2AUREbfn3v/+N2NhYfPjhh4iJiRG7HCKyEZwBIiKL9c4772D+/PnYsWNHu+Hn448/hr+/P1QqlcbxKVOm4KmnngIA/PLLL5gyZQp8fX3h5uaGkSNH4rvvvmv1M3/99VdIJBIcP35cfezGjRuQSCQ4ePCg+tipU6cwceJEuLm5wdfXF0888QQqKip0v2AiMhsGICKySK+++iqWL1+O3bt3Y+rUqe2Of/TRR/H777/jwIED6mPXrl1DWloaZs2aBQCorq7GpEmTkJGRgWPHjuGhhx5CdHQ0ioqK9K7zxo0bGDduHIYNG4ajR48iLS0NZWVleOyxx/T+TCIyPS6BEZHF2bt3L77++mtkZGRg3LhxHTqna9eumDhxIrZv347x48cDuN075O3tjbFjxwIAQkNDERoaqj5n+fLl+Oqrr/B///d/iIuL06vWdevWYdiwYVi5cqX62KZNmyCTyXD+/Hnccccden0uEZkWZ4CIyOIMGTIEgYGBWLp0Kaqrqzt83qxZs7Bz504oFAoAwLZt2/D444/DweH2X3XV1dV4+eWXceedd8LLywtubm7Iz883aAboxIkTOHDgANzc3NSvkJAQALeX3IjIMnEGiIgsTkBAAL788kuMHTsWDz30EPbu3Qt3d/d2z4uOjoYgCNizZw9GjhyJ77//Hv/v//0/9fsvv/wy0tPT8d5776F///5wdXXF3/72N9TV1Wn9vMbgJAiC+lh9fb3GmOrqakRHR+Ptt99ucX7Pnj07dL1EZH4MQERkkfr06YNDhw6pQ1BaWlq7IUgqleKRRx7Btm3bcOHCBQwYMADDhw9Xv//DDz/gySefVPcUVVdX49dff23183r06AEAKCkpwbBhwwBAoyEaAIYPH46dO3ciMDAQTk78K5XIWnAJjIgslkwmw8GDB1FeXo6oqCjI5fJ2z5k1axb27NmDTZs2qZufGwUHByM1NRXHjx/HiRMnMHPmzBbfGmvK1dUVd999N9566y3k5+fj0KFDSExM1BgTGxuLa9euYcaMGThy5Ah++eUX7Nu3DzExMVAqlfpdOBGZHAMQEVm0Xr164eDBg6ioqOhQCBo3bhy6deuGc+fOYebMmRrvJScno2vXrhg9ejSio6MRFRWlMUOkzaZNm9DQ0ICwsDAsWLAAb775psb7/v7++OGHH6BUKjFhwgQMHjwYCxYsgJeXl3oJjYgsj0RourhNREREZAf4zxMiIiKyOwxARGQVioqKNL5q3vxlyFfZicj+cAmMiKxCQ0NDm9/Y4rewiEgXDEBERERkd7gERkRERHaHAYiIiIjsDgMQERER2R0GICIiIrI7DEBERERkdxiAiIiIyO4wABEREZHdYQAiIiIiu/P/Aem4z52zRk1tAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(k_value,scores)\n",
    "plt.ylabel('Score')\n",
    "plt.xlabel('K_value')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Som man ser i grafen så får vi bäst score med antingen 3 eller 5 som K-värde, en vanlig sak annars är att man tar roten ur antalet datapunkter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 390,
   "metadata": {},
   "outputs": [],
   "source": [
    "label_pred = knn.predict(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 391,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "lab = test_labels.unique()\n",
    "lab = sorted(lab)\n",
    "cm = confusion_matrix(test_labels, label_pred)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=lab)\n",
    "disp.plot()\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Som vi ser så har nästan varje siffra ett antal fel, men 1or verkar den ha lätt att klassificera, jag kommer att rita ut och fortsätta min formulering nedan om vad den gör för fel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 392,
   "metadata": {},
   "outputs": [],
   "source": [
    "incorrect_indexes = [i for i in range(len(label_pred)) if label_pred[i] != test_labels[i]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 393,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_images(indexes, data, labels, model):\n",
    "    num_images = len(indexes)\n",
    "    fig, axs = plt.subplots(math.ceil(num_images/8), 8, figsize=(16,math.ceil(num_images/8)))\n",
    "    axs = axs.flatten()\n",
    "\n",
    "    for i, index in enumerate(indexes):\n",
    "        predicted_label = model.predict([data.values[index]])[0]\n",
    "        img = data.iloc[index].values.reshape(28,28)\n",
    "        label = labels[index]\n",
    "        axs[i].imshow(img, cmap='gray')\n",
    "        axs[i].set_title(f\"Label: {label}\\n Predicted Label: {predicted_label}\")\n",
    "        axs[i].axis('off')\n",
    "    \n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 394,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x700 with 56 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_images(incorrect_indexes, test_data, test_labels, knn)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Jag blev nyfiken på att se siffrorna som modellen gjorde fel på så jag gjorde en funktion för att rita ut den sanna siffran och sedan skriva vad modellen predictar att den ska vara, och en observation är att det känns som att modellen nöjer sig när den hittar en likhet, om man tar vissa 8:or exempelvis så kan dom se väldigt självklara ut men den gissar på 5or och 3or och det antagligen för att man kan ju faktiskt klämma in en 5a eller 3a i en 8a i många fall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 384,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_neighbors(model, new_number, x_train, y_train):\n",
    "    label_pred = model.predict([new_number])\n",
    "    print(f'The prediction of the new number is: {label_pred}.')\n",
    "    nearest = model.kneighbors([new_number], n_neighbors = model.n_neighbors, return_distance = False)\n",
    "    fig, axs = plt.subplots(math.ceil(len(nearest[0])/5), 5, figsize=(16, math.ceil(len(nearest[0])/5)))\n",
    "    axs = axs.flatten()\n",
    "    for i, index in enumerate(nearest[0]):\n",
    "        img = x_train.values[index].reshape(28,28)\n",
    "        label = y_train[index]\n",
    "        axs[i].imshow(img, cmap='gray')\n",
    "        axs[i].set_title(f\"Label: {label}\")\n",
    "        axs[i].axis('off')\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Här kommer det att ritas ut lika många grannar som K-värdet i modellen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 395,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Programmering\\MachineLearning-SchoolProjects\\env\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The prediction of the new number is: [6].\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x100 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "new_number = test_data.iloc[97]\n",
    "plot_neighbors(knn, new_number, train_data, train_labels)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# VG"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Jag bygger en egen KNN från grunden, nu har jag använt mig av ett par enstaka numpy funktioner som jag anser är okej då dom enbart är där för ett par beräkningar."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.spatial.distance import cdist\n",
    "class KNN:\n",
    "    def __init__(self, k=3, metric='euclidean'):\n",
    "        self.K = k\n",
    "        self.metric = metric\n",
    "    def euc_dist(self, x1, x2):\n",
    "        return np.sqrt(np.sum(np.square(x1-x2)))\n",
    "    def manhattan_dist(self, x1, x2):\n",
    "        return np.sum(np.abs(x1 - x2))\n",
    "    def fit(self, x_train, y_train):\n",
    "        self.X_train = x_train\n",
    "        self.Y_train = y_train\n",
    "    def predict(self, X_test):\n",
    "        predictions = [] \n",
    "        for i in range(len(X_test)):\n",
    "            distances = []\n",
    "            if self.metric == 'manhattan':\n",
    "                for n in range(len(self.X_train)):\n",
    "                    distance = self.manhattan_dist(X_test[i],self.X_train[n])\n",
    "                    distances.append(distance)\n",
    "            elif self.metric == 'euclidean':\n",
    "                for n in range(len(self.X_train)):\n",
    "                    distance = self.euc_dist(X_test[i],self.X_train[n])\n",
    "                    distances.append(distance)\n",
    "            else:\n",
    "                raise TypeError(\"metric must be a euclidean or manhattan\")\n",
    "            k_indices = np.argsort(distances)[:self.K]\n",
    "            k_nearest_labels = [self.Y_train[i] for i in k_indices]\n",
    "            counts = np.bincount(k_nearest_labels)\n",
    "            best_prediction = np.argmax(counts)\n",
    "            predictions.append(best_prediction)\n",
    "        return predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNN(k=5)\n",
    "knn.fit(train_data.values,train_labels.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = knn.predict(test_data.values)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Efter att ha kört med den vanliga euklidiska formel som även den färdiga modellen kör om man inte ändrar så tar det min laptop dryga 45 sekunder att köra predicten, jag testar att lägga till manhattan formeln för beräkna avståndet då den har lite kortare beräkning, och nu får jag dryga 40 sekunder, så inte så stor skillnad i tid, men min score går ner med 0.015, vilket inte är mycket men tiden kan ju också variera en aning från gång till gång.\n",
    "### Eftersom att jag har tre loopar i min KNN så är det många beräkningar därav tiden, jag förstår inte riktigt hur den färdiga modellen är klar på nolltid. Jag funderar på om det går att göra alla träningsberäkningar enbart en gång så att varje rad i testdatan inte behöver gå igenom varje rad i träningsdatan, men jag har svårt att skriva det utan att importera funktioner som jag anser vara lite fusk.\n",
    "### Nu ska jag försöka beskriva tidskomplexiteten nedanför med mina egna ord:\n",
    "### Vi loopar igenom alla test datan O(n_test) och sedan så ska dom loopas igenom alla träningsdata och tar fram distansen O(n_träning) = O(n_test * n_träning) men eftersom att jag är osäker på om np.sum loopar igenom för att summera eller om dom gör det på något annat sätt så är jag osäker men jag skriver det som en O(log(k)) och då blir det = O(n_test * n_träning * log(k))\n",
    "### Nu ska jag försöka beskriva det med den faktan vi har fått:\n",
    "### O(n) är generellt för en loop vilket gör att mina första två loopar med testdata och träningsdata blir O(n^2) eftersom att det blir också när jag har en till loop i testdata loopen så blir det O(n) + O(n) = 2O(n) = O(n) men sedan eftersom att jag är osäker på np.sum om den loopar så är det potentiellt O(n^3) alternativt O(n^2 * log n)\n",
    "### Intressant att se då det ökade förståelsen för hur lång tid det faktiskt tar."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.945"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(test_labels, predictions)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Jag får precis samma score som den färdiga modellen i G-delen, även när jag ändrar K värdet i min egna och den färdiga så blir det samma score. Nu är ju den färdiga modellens algoritm inställd på 'auto' så den går igenom vilken den tror är bäst för datasetet men i min så har jag bara en att gå på, men med detta sagt så har jag testat att ställa in den färdiga på olika algoritmer men den visar ett och samma score för samtliga."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "env",
   "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": "3.11.0"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}