3._Classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import fetch_openml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnist = fetch_openml('mnist_784', version=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'frame', 'categories', 'feature_names', 'target_names', 'DESCR', 'details', 'url'])"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mnist.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((70000, 784), (70000,))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X, y = mnist['data'].to_numpy(), mnist['target'].to_numpy()\n",
    "(X.shape, y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualizing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAGFCAYAAAASI+9IAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAIy0lEQVR4nO3cOWhWUR7G4ZsY16BGOxVrIY0LSgrBFbRSW7EQrSK4NAYRUlgK2mnsxEq0EVPYKApaiCApFBcwRUDEQpuQCFoo8k0zvM0MDP87Y/JNfJ7+5Vw04ZfTnJ5Op9NpAKBpmt75/gAAuocoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABB98/0B8J/8/v27vJmdnf0DX/K/MTY21mr348eP8mZycrK8uXHjRnkzMjJS3ty9e7e8aZqmWbZsWXlz8eLF8ubSpUvlzULgpgBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQHsRbYD59+lTe/Pz5s7x58eJFefP8+fPypmmaZmZmpry5d+9eq7MWmo0bN5Y3Z8+eLW/Gx8fLm5UrV5Y3TdM0mzdvLm92797d6qy/kZsCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQPR0Op3OfH8E/+rVq1etdvv27StvZmdnW53F3Fq0aFF5c+vWrfKmv7+/vGlj/fr1rXZr1qwpbzZt2tTqrL+RmwIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIA4ZXULjU9Pd1qNzQ0VN5MTU21OmuhafNv1+bFzqdPn5Y3TdM0S5YsKW+8gEuVmwIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBA9M33B/DvrV27ttXu6tWr5c2DBw/Km61bt5Y3586dK2/a2rJlS3nz5MmT8qa/v7+8effuXXnTNE1z7dq1VjuocFMAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQAiJ5Op9OZ749gfn379q28WblyZXkzPDxc3jRN09y8ebO8uX37dnlz7Nix8gYWGjcFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgOib7w9g/q1atWpOzlm9evWcnNM07R7RO3r0aHnT2+vvKhYWP9EAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoARE+n0+nM90fwd/j+/Xur3aFDh8qbZ8+elTcPHz4sbw4cOFDeQDdzUwAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAID+LR9aampsqbbdu2lTcDAwPlzd69e8ub7du3lzdN0zSnT58ub3p6elqdxd/LTQGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgPIjHgjQ+Pl7enDx5srz59u1bedPW5cuXy5vjx4+XN+vWrStvWDjcFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQDCg3jwT2/fvi1vzp8/X948efKkvGnr1KlT5c3o6Gh5s2HDhvKG7uSmAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABAexIP/wszMTHnz4MGDVmedOHGivGnz671///7y5vHjx+UN3clNAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYDwSir8n1i6dGl58+vXr/Jm8eLF5c2jR4/Kmz179pQ3/HluCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgDRN98fAN3izZs35c29e/fKm4mJifKmado9btfG4OBgebNr164/8CXMBzcFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgPAgHl1vcnKyvLl+/Xp5c//+/fLmy5cv5c1c6uur/4qvW7euvOnt9fflQuF/EoAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACA8iEcrbR6Cu3PnTquzxsbGypuPHz+2Oqub7dixo7wZHR0tbw4fPlzesHC4KQAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABCiAECIAgAhCgCEB/EWmK9fv5Y379+/L2/OnDlT3nz48KG86XZDQ0PlzYULF1qddeTIkfKmt9fffdT4iQEgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgvJI6B6anp8ub4eHhVme9fv26vJmammp1VjfbuXNneXP+/Pny5uDBg+XN8uXLyxuYK24KAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCAPFXP4j38uXL8ubKlSvlzcTERHnz+fPn8qbbrVixotXu3Llz5c3o6Gh509/fX97AQuOmAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAEKIAQIgCACEKAIQoABB/9YN44+Pjc7KZS4ODg+XNoUOHyptFixaVNyMjI+VN0zTNwMBAqx1Q56YAQIgCACEKAIQoABCiAECIAgAhCgCEKAAQogBAiAIAIQoAhCgAED2dTqcz3x8BQHdwUwAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAgRAGAEAUAQhQACFEAIEQBgBAFAEIUAAhRACBEAYAQBQBCFAAIUQAg/gEx1gSzbdeSSgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "some_digit = X[0]\n",
    "some_digit_image = some_digit.reshape(28,28)\n",
    "plt.imshow(some_digit_image, cmap=mpl.cm.binary, interpolation='nearest')\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.uint8(4)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np \n",
    "y = y.astype(np.uint8)\n",
    "y[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Binary classifier of 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_5 =(y_train==5)\n",
    "y_test_5 =(y_test==5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-1 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-1 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-1 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-1 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-1 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SGDClassifier(random_state=42)</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 fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;SGDClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.linear_model.SGDClassifier.html\">?<span>Documentation for SGDClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(random_state=42)</pre></div> </div></div></div></div>"
      ],
      "text/plain": [
       "SGDClassifier(random_state=42)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Classifier\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "sgd_clf = SGDClassifier(random_state=42)\n",
    "sgd_clf.fit(X_train, y_train_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.predict([X_train[0],X_train[0]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross validation using custom KFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.base import clone\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "skfolds = StratifiedKFold(n_splits=3, shuffle=True, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9669\n",
      "0.91625\n",
      "0.96785\n"
     ]
    }
   ],
   "source": [
    "for train_index, test_index in skfolds.split(X_train, y_train_5):\n",
    "    \n",
    "    clone_clf = clone(sgd_clf)\n",
    "    X_train_folds = X_train[train_index]\n",
    "    y_train_folds = y_train_5[train_index]\n",
    "    X_test_folds = X_train[test_index]\n",
    "    y_test_folds = y_train_5[test_index]\n",
    "    clone_clf.fit(X_train_folds, y_train_folds)\n",
    "    y_pred = clone_clf.predict(X_test_folds)\n",
    "    n_correct = sum(y_pred==y_test_folds)\n",
    "    print(n_correct/len(y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Same Cross Validation with Sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.95035, 0.96035, 0.9604 ])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "cross_val_score(sgd_clf, X_train, y_train_5, cv=3, scoring='accuracy')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Never5Classifier(BaseEstimator):\n",
    "    def fit(self, X, y=None):\n",
    "        pass\n",
    "    def predict(self, X):\n",
    "        return np.zeros((len(X), 1), dtype=bool)\n",
    "never_5_clf = Never5Classifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.91125, 0.90855, 0.90915])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(never_5_clf, X_train, y_train_5, cv=3, scoring='accuracy')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion matrix "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_5_pred = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, (60000,), array([ True, False]))"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(y_train_5_pred), y_train_5_pred.shape, y_train_5_pred[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[53892,   687],\n",
       "       [ 1891,  3530]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "real_cfm = confusion_matrix(y_train_5, y_train_5_pred)\n",
    "real_cfm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[54579,     0],\n",
       "       [    0,  5421]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perfect_pred_y5 = y_train_5\n",
    "perfect_cfm = confusion_matrix(y_train_5, perfect_pred_y5)\n",
    "perfect_cfm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "precision = ( 3530 /  3530 + 687 ) =  0.8370879772350012\n",
      "recall = ( 3530 /  3530 + 1891 ) =  0.6511713705958311\n"
     ]
    }
   ],
   "source": [
    "precision = real_cfm[1][1] / (real_cfm[1][1] + real_cfm[0][1])\n",
    "recall = real_cfm[1][1] / (real_cfm[1][1] + real_cfm[1][0])\n",
    "print(\"precision = (\" , real_cfm[1][1] , \"/ \", real_cfm[1][1], \"+\", real_cfm[0][1] , \") = \",precision)\n",
    "print(\"recall = (\" , real_cfm[1][1] , \"/ \", real_cfm[1][1], \"+\", real_cfm[1][0] , \") = \", recall)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "precision =  0.8370879772350012 recall = 0.6511713705958311 f1_score =  0.7325171197343847\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score, recall_score, f1_score\n",
    "\n",
    "precision_skl = precision_score(y_train_5, y_train_5_pred)\n",
    "recall_skl = recall_score(y_train_5, y_train_5_pred)\n",
    "f1_score_skl = f1_score(y_train_5, y_train_5_pred)\n",
    "print('precision = ',precision_skl,'recall =',recall_skl, 'f1_score = ', f1_score_skl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3, method='decision_function')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 60001, 60000)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "precisions, recalls, thresholds = precision_recall_curve(y_train_5, y_scores)\n",
    "len(thresholds), len(precisions), len(y_train_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_precision_recall_vs_threshold(precisions, recalls, thresholds):\n",
    "    plt.plot(thresholds, precisions[:-1], \"b--\", label='Precision')\n",
    "    plt.plot(thresholds, recalls[:-1], 'g-', label='Recall')\n",
    "    plt.legend(loc='center right', fontsize=16)\n",
    "    plt.grid(True)\n",
    "    plt.axis([-50000,50000,0,1])\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "plot_precision_recall_vs_threshold(precisions, recalls, thresholds)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "recall_90_precision = recalls[np.argmax(precisions >= 0.90)]\n",
    "threshold_90_precision = thresholds[np.argmax(precisions>=0.90)]\n",
    "recall_90_precision, threshold_90_precision\n",
    "plt.figure(figsize=(10,6))\n",
    "plot_precision_recall_vs_threshold(precisions, recalls, thresholds)\n",
    "plt.plot([threshold_90_precision, threshold_90_precision], [0,.9],\"r:\")\n",
    "plt.plot([-50000, threshold_90_precision], [0.9, 0.9], \"r:\")  \n",
    "plt.plot([-50000, threshold_90_precision], [recall_90_precision, recall_90_precision], \"r:\")\n",
    "plt.plot([threshold_90_precision], [0.9], \"ro\")\n",
    "plt.plot([threshold_90_precision], [recall_90_precision], \"ro\")                             # Not shown\n",
    "                                              # Not shown\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(np.float64(0.9000345901072293), np.float64(0.4799852425751706))"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_pred_90 =(y_scores >= threshold_90_precision)\n",
    "\n",
    "precision_scr = precision_score(y_train_5, y_train_pred_90)\n",
    "recall_scr = recall_score(y_train_5, y_train_pred_90)\n",
    "precision_scr, recall_scr\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Precision vs recall tradeoff graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2IAAAIbCAYAAACJ9S0vAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABrl0lEQVR4nO3dd3hUxf7H8c8m2VQ6odegKCgKGiyAqKigWBBFxQtKEbgiNkRRFL0CFqyIXsUKckVUBBUsWFABKcoPERAEBAWMlFBCT90k+/tjDJtNT7acLe/X88yTM7Oze75LDtl8M3NmbE6n0ykAAAAAgN9EWB0AAAAAAIQbEjEAAAAA8DMSMQAAAADwMxIxAAAAAPAzEjEAAAAA8DMSMQAAAADwMxIxAAAAAPAzEjEAAAAA8DMSMQAAAADwMxIxAAAAAPCzgE/EfvjhB1111VVq3LixbDab5s6dW+5zFi9erOTkZMXGxqpVq1Z67bXXfB8oAAAAAFRQwCdi6enpat++vV5++eUK9d+2bZsuv/xyde3aVatXr9ZDDz2ku+66Sx999JGPIwUAAACAirE5nU6n1UFUlM1m0yeffKLevXuX2ueBBx7Qp59+qo0bNx5vGz58uNauXasff/zRD1ECAAAAQNmirA7A23788Uf16NHDre3SSy/V1KlT5XA4ZLfbiz0nOztb2dnZx+v5+fk6cOCA6tatK5vN5vOYAQAAAAQmp9Opo0ePqnHjxoqI8N6EwpBLxFJTU9WgQQO3tgYNGig3N1f79+9Xo0aNij1n4sSJGj9+vL9CBAAAABBk/v77bzVt2tRrrxdyiZikYqNYBbMvSxvdevDBBzVq1Kjj9cOHD6t58+aS/ta99ybowQfzfRYrwpvD4dDChQvVrVu3EkdrAW/hWoO/cK3BX7jW4C8HDhzQSSedpOrVq3v1dUMuEWvYsKFSU1Pd2vbu3auoqCjVrVu3xOfExMQoJiamhEdqKD6+hkp5GuAxh8Oh+Ph41a1blw8R+BTXGvyFaw3+wrUGf/P2LUsBv2piZXXq1EkLFixwa/vmm2/UsWNH/pMCAAAACAgBn4gdO3ZMa9as0Zo1aySZ5enXrFmjlJQUSWZa4YABA473Hz58uP766y+NGjVKGzdu1LRp0zR16lTdd999VoQPAAAAAMUE/NTEn3/+Wd26dTteL7iXa+DAgZo+fbp27959PCmTpKSkJM2fP1/33HOPXnnlFTVu3FgvvfSS+vTp4/fYAQAAAKAkAZ+IXXjhhSprq7Pp06cXa7vgggv0yy+/+DAqAAAAAKi6gJ+aCAAAAAChhkQMAAAAAPyMRAwAAAAA/IxEDAAAAAD8LOAX60DlpaZKX38t7dsn/fvfUo0aVkcEAAAAoDASsRCQny+tWCF98YX05ZdS4QUjt26Vpkyp2OscOSLt2CG1aSNFMFYKAAAA+AyJWJByOk3C9cEH0qxZ0t9/l9zv1Velf/1L6tq1+GOHDklLlkiLF5vyyy8mqRs2THrjjeL9s7Ol9HSpTh2vvhUAAAAg7JCIBQGnU7LZzPH+/dLbb0tvvilt2VJy/1NOkTZscNXPP1/q2FF66SXzOl98Ic2fL61ebV67qDfflNaulTZvlgYNkux2adky6eefpZwck4jde680erR5TJLy8szXyEivve1S5eZK27ZJGzdKmzaZBLFHD5OMbtlipmZedpl09Kj055+u8ssvZtTvv/+VOnc2I3/x8b6PFwAAACiKRCxAHTggTZpkRqZOOEF64QUzxfDDD03iUVhUlElE+vSRevaUGjVyJW4Ffv7ZJB8V9X//Z75OnlxybGPHmnLTTWY644oVUmamifmee9z7p6a6Xq9nT5M8rVkj1a1rYipI+s4+W6pZU1q3Tvr1V2nvXunSS83I3caNrsRryxaTEBb2n/+41196qfT3duedruOHHzbn7tRJSkgwCd7Wrebr//2fVL261K2btGiRdOWV0jvvmGTT6ZSOHZOqVSv+bw0AAACUh0QswBw7Jj33nEm8jhwxbfv2mUShqG7dzLTDa681SU1h8+dLl19e9rnat5cuvFC64AIzdbFbN2n9+srF++677vVRo0wylZZmEpkVK0qfNlkRZSVU3vD44+brhx+W/PjRo9Knn5rj994z5fTTTaJ29KhpnzTJ/BuefLIZYTt4UIqNNcdZWWbU0B8jhQAAAAgeJGIBwumUPvrIjCbt2FF6vzp1pMGDpVtvlVq3Lr1fz56SwyH973/S0KGmrVo1qXt36YorzOONG7s/Z9Eik1g1a2ZGqVaskNq1k7p0MSNXO3dKc+ZIEyeW/V6GDKnQW64Su1066SQzrbBNGxNv9erSmWeaf49ly8wIXHKyGUk84QTpxBOlpCTps8/Mv++BA57F8Ouv7vVRoyr2vIgI6amnzJROAAAAhDcSsQCwe7dJXr780tUWFWXuhSrQrJn00EPSwIFSXFzFXjcqyrzugAHmHqmkJCkmpvT+detKd99tjq+9tvjjDRqYhOfxx80iIDt2SGedJZ17rlngY/784s9JSDAjb8uXm/pJJ0nNm0vffmum9HXqZEYBf/3VtJ9+uimLFpmRwM6dpbZtXSUpybyvAgUjWhUxYIAp+fkmQT1wwIzc/fyz1KGDee1WrczXOnWkn36SatUy7+HMM12vY7ebJLey8vOl++8379NmM6Nq69dHqnnzBuWOXgIAACC0kIhZbN48M2K1f7+rrWdP6cUXTUJQMOJz9dXuCUhl2O1m9MhbIiKk2293b3vlFenGG839a2efLZ1zjvnatq3rniqHQ4qO9l4cVRURYUYVy3Plla7jY8fMPWoNGpiRxMhIk8y99JJru4CoKLNnW8GIW9FkusCNN7pFI+lcrViRr2++YdsAAACAcEEiZpH8fLPYxVNPudoaNTIJTe/ergUgBg2yIrrKa9nSjCCVxmYLjCSsqhISzMqThQ0caIpkEk2n0yRShZPOvDyTxF16qZnqWZrvvotQZKRJ4mrXdr0mC4EAAACEJhIxC6Snm9UG5851tfXubZaNT0y0Kip4wmZzJU2Fk87ISLN4yU8/SU8/LU2fbhb2SEqS/vhDeust99cp2KOtVi2zWqRU8hYDAAAACG4kYn525IhZzXDZMlOPjDQrJN5xB6Mfoe6BB0wpbOhQh849116sb0ESJrmui48+MveX/fGHuW4K7ptjRUYAAIDgQyLmR4cPm42GC6bw1aghzZ5t9gBDeDrzTGnWrM80evSVSkkxGVdEhJm6WlSfPqW/TmqquX9NMvellXY/YW6utGePVK+e2YstIYE/AAAAAFiBRMzHfvrJ7PXVpImrLpkVCr/91qzWh/AWE5OvP/7Ild1u17FjZlpjRobrXrGKaNhQatHCJFo7d5q2M880r/XTT+b+w4QEafv2khcQuflms/gISRkAAIB/kIj5UOGNmLdvd7UnJkrffWeWaQcKq1bNfI2ONveG5eaaewdHjJDOP98k9O+/X/Jz//rLvV6wmqNktkgoy4wZpkRGmgVGChs3zozetm5tlvvv0MFsSQAAAICqIxHzEaez5CXSY2Olzz8nCUPFREVJt91mSoH33jNf8/Ol+vVNclQRCQlSZmbJ0x4LFE3CJJOIlaRxY+nTT6XTTgvuFTEBAACsQCLmI+++K33xRcnt55zj/3gQeiIizP5z27ebZOyEE8xqizt3ujbITkw09aZNzX1hRacerlhhNuSuil27XEv6Dx1qRsq2bnWVX381j7VpY0by6tc3I37jx5uksEaNKr5xAACAEEAi5gNpadKoUcXbb7217AUXgKpo2dKUAk2auO5JlFyLeJTknHNc+54dOmTuXdy/3yyj/9ln0scfS1u2mFG0unWlr74q+XWKLsNf2KZNphR4803X8WuvSatWScnJJnlbskTq1s1Ms9y2TVq+XPrPf8z7W7fOjCSnpJjHN20yj8fHm5VI580zfe+7z4w8AwAABDISMR8YO9b8MitJ551npnAlJUlPPGFtXEBp7HYzYiaZkStJuuYaUwrLy5N++026+mr3+x6ravhw87Vwcvbjj+59Jkwo+zUyMqQ5c8zxI4+Yct555p66ZcvMyN2AASaZ69WLkTgAABAYSMS8bPNm1+hA9erSrFkmEQNCQWSkGZXats3UFy+Wbr/dJGatWrlK3bpmpGvPHtO3TRuzaXl8vEmcfG3pUlMKPPKI++PnnCOtX282V5fMfmwbN5qpnU8+6X5PHgAAgC+QiHnZI4+4FjwYPZokDKHtggtMQlPaY4U5neZrVpZ00UVS587S2rUmSTvlFJMInXSSudetZUtp0SLpuefMqNyKFdKpp5qEqWVLs1S/w+Ealduxo+TpwKVZscK9vnGj+XrokFmhcsQI6f77pb//NtMgN282o9z9+5t74S64wNwfx3L/AACgqkjEvGjjRunDD81xvXrSPfdYGw8QiGJjzb1d5WnbtvyRqbPOch3fdZc0c6b0888moWvYUPrgA7OR+rffVj7OZ54p3jZzpilFPfGEGR1s1szEHRdnVpNMTjajgAAAAEWRiHnRCy+4jh94wLUnFADfi4w094INGOBqu+469z47dpg91Zo3NyNzeXlmqmRkpEmwqnof59ixFes3bJj5A02bNq7RtILFUtgCAACA8BJhdQChYt8+6Z13zHH16uYXLgCBpWlTM4rWoIGZ3tiqldSunRnFevxxkxRNnmxWc/zmG+n3302itnixdO+9np//zTfNNMyICJOItWsn1awpxcSYeknlhBOk77/3/NwAACCwMCLmJW+/LWVnm+Nhw1iZDQhWd99dvO3880157jlX28GD5j6ypk3N//enn5YuucQs+3/kSMXO9dtv5ffZulW6+GJzfPbZ0v/9nzmOjDRTMIcMMcnaFVeY1S8BAEBwIBHzAqdTmj7dVR8xwrJQAPhJ7druy+6XdE+o02lG1D74QJoyRfrlF/fHY2Jcf8CpiIIkTDLTKjduNPumFUhONitV7tjharv1VrNKZGKiTfPnJ+mii0pP2PLzzWgdAADwPRIxL1i50rXqWteu5q/TAGCzSQkJZtRqyBDTlptrVmFMTDSL+hS+V6zgODXV/Ezp39/c01ZRq1YVb3v9dVPMj/vT9cYbpj0pybUNQcOGJmEsGMlbscIsOJKfbxZXiY01CdqBA9LJJ7NhNgAA3kAi5gX/+5/reNAgy8IAEASiosx9YkUVXgq/YUNTdu2S/vxTSkszI2CNG5tVID/80CRzmzebLQCqoiAJk0ziV9g551TsNbp2la66yiRnDRqYRG79eqlHD3PfHaNrAACUjkTMQ/n50iefmOOYmOKrtAGAJ044ofgo++mnu9f37DHJWrNmZtphwVYaTz0ltW5tFiYpawn/yk6RLLBkiSlladNG2rTJVe/Xz+wFV7eu1KiR9K9/Vf68AACEAhIxD61c6Zo61KMHi3QA8L8GDUwpcMYZpkyc6GpzOByaP3++unW7XA6HXXXqmGmSR45IdeqYEbkBA6QZM8zKkitXup6bkOC6362yCidhkvTee+71fv2KP2f5cqlTp8qfCwCAYMLEEQ/Nnes67t3bqigAoGLi4sxolM1mRs8KjiWzBYfTaRYFcTpd5dgxKT3dVf/qKzPKVqeO63X793eve6Jz5+LL+L/3nhnlKzqNskx5edKiRdL775uveXneCRAAAC9gRMxDn31mvkZEmHslACDUXXqptH176Y87HNJff5kkr3p1M1Vy3jxp6lSzsmNkpEmqKqN/f/f6O+9I330nde9uthbIzjYLkERG/tPh44/NXgSFl5Bs2lR68UXp2msrd3IAAHyARMwDqamufYDOOsusgAYA4c5ul0480VU/7TRTHn7Y1TZrlmuAKivL3Gt2991mAZKKGDDAfC28WFKBa/SxPtJ1kpyyFX5g505zI++cOSRjAADLMTWxDHH6Zy5OgZwcMz/nn7vav//eNMcrXZd1TTcrdxRwOEzfrCz3F01PN6UifTMyTHvh6TS5uaYtM7PqfTMzTXturqstL6/yfYveMJKVZdodjqr1zc93/fsUlp1t2nJyqtbX6XT1Len7WZm+RVc0KOn7WZm+Docis7Iq9r33xnVS8P309Dop+v309Dop7fvp6XVS+PtZmb6V+d5X9Tqx4GdEZFZWQP2MiIyUIpWnBKXrsvMz9PvvrumPeelZSktJ13nnONSqlXm6TfmKV7ri5f79jFa24pUuu3IUoTy9qLvlLJqESa7v1ciRTFMEAFiORKwMqWqs+Iz9roZnn5WqVZPuuEOSKxHbq/oa91w1KSXF1feVV0zfgs2DCrRsadoLNh6TzG7Q1apJN97o3veUU0x74V1gZ80ybb16ufc96yzTXngJs88/N22XXOLe9/zzTfvXX7vavv/etBW9Q75nT9NesDSkJP30k2lr3969b58+pn3mTFfbunWmrXVr974332zaCzY1ksw63dWqSU2auPe99VbT/uKLrrbdu01brVrufUeNMu1PPulqO3zYtFWr5v7L4tixpm3sWFdbbq6r7+HDrvYnnzRto0a5n69WLdNeeLOnF180bbfe6t63SRPT/uefx5si3npLV954oyKL7nvQurXpu26dq23mTNPWp4973/btTftPP7naPvnEtPXs6d63UyfTXnDxSuY6qFbNXBeFXXKJaf/8c1fbkiWm7ayz3Pv26mXaZ81ytf3yi2krulb7jTea9sK7oG/caNpatnTvO2SIaX/lFVdbSoppq1/fve8dd5j2Z591te3f7/p+FvbAA6Zt/HhXW0aGq2/hBGL8eNP2wAPur1HQd3/pPyOOq1/ftFv4MyKqUyddeeONsi1d6moM4J8REQNvVp3m1bTk5jf0558mh3Js/FPpqqZ90U3UoYOr7+u6Vemqprv1orpqiZppR+kfbk6n9PffyvmunOUeAQDwMRIxDxT+XRYA4FsF93/Fx0mrV7tGzwYNNO1PPSX1aFexHbAHXrr7+IbbR4/6KGAAAMpgczoLz6uBJB05ckQ1a9ZUnHbpvocbasJj/0xwyckxU2qiorT7QIwaNzbNl56Xrq++klmOrGAHU4fD9I+MlGJjXS9eMEWqIn0zMsxvGbGxrt9AcnPNtKeICPMaVembmWmmPcXEmN1lJTNNJyurcn1tNik+3tW3YNpTdLS5SaSyffPzXdOeEhJcfbOzzXux203/yvYtvO52fLxribiC72dl+kZFmX+LAiV9PyvR15GRoa8/+0yXXn657NWru/qW9P30xnVS8P309Dop+v309Dop7fvp6XVS+Pvp6XVS2vezqteJn39GOA4f1tdffaVLr75a9oLXCMWfEcuXS926qTwXaqEW60K3ttGjzb5mp59eaNEPVFrBVgmXX3657AXfO8AHuNbgL2lpaUpMTNThw4dVw4t7VbFYRxkyleD6BUsyvxD880vbihWu5g5dEqSEIk+2212/PBSWULRjGX0L/1JSICrK9ctOVfsW/iWqQGRkybFVpm/hXxCr0jciouS+MTHuv9RWtm/Bn72LKvT9rFJfyfO+drvyYmOL/3uU9P30xnVS0vczEK6T0r6fnl4npX0/Pb1OJK987/39MyKvcMJWVt9g/hnRtatZHXHnTvf79/6RL5t2qKmWqGuxx5591n1m6403mlmhRS8VAAC8gamJVVQ4ETvnHOviAAAUEhnpup/UVmS5DptNETap+UeTtWlz+UNeH3xg8sGCvcxOPtnc6lmrlqkPHGg2xAYAoCpIxKro//7PdUwiBgAB5NprzRL1RRf+adr0+NL1rVubAbPcXPf1esqyebO0a5drHZ933pFq1nQlamvXSgcPSps2mQG53Fzp779NAQCgKBKxKsjPl1auNMdNmuj4vWIAgABx7bXS77+b3ae7djUrQG7bVmz/sMhI6fHHXQt/5OZKDz5YtVN26CDVqSO1bWtyPrtdat7clIJkbfLkEmdMAgDCEIlYFWzf7lplKznZ0lAAAKVxOk0CtmSJ1KVLhVbgiIw0u1UUJGb/rHavzz83q+6vXCnddlvVQ7rnHnOrW0FiZrOZza6bNSu+/RwAILSxWEcVFN7a6bTTrIsDAFAGu1165hnXcRU1bWpKgY4dpSlTzOjZ9u1mW8SCbfxatjRtknTGGWaZ/fKsX2++FqxPEhVltoG7+Wbp6quL3+oGAAgNJGJVQCIGAEEgOtqsSe8jUVHSiSdKP/5Yft/p06XBgyv2urm50scfmyKZEbRZs8wMjKSkKocLAAgwJGJVUDgRa9fOujgAAMFh0CBTJGn/fiktTVq0yKzAeOONZT83P1+6/nr315Kk9u2lxETpooukhg1d284BAIIDiVg5HntMmjdPeuopqWdP01aQiNnt0kknWRcbAKAM+fnS7t3muFGjgMlUEhNNOflkU+/b13zds0e66Sazpsiff5b+/OnTy379a6+VPvrIK6ECAHwoMD6VAtyvv0qXXy7NnStlZJgljCWzMhYbuQNAgMrMdN3glZlpdTTlatBAWrBA+uMPs0jI5s1m6mNlffyxua+MPc4AILCRiFXCNddId90l5eWZesFfMwEAASoqypQg1Lq1tGWLScpSUqR+/aTTT5duuaVizy/Y4+yss8zgIAAgsJCIVdLUqa7jqvylEgDgJwkJksNhSkKC1dF4pFkzs3z+2rXmc6jw8voFZdGikp/7889mWf6C5fIfeUT64gspJ8evbwEAUASJmAdIxAAAgeKCC6StW8vv9/jj0pVXSjExUu3axfc0W7TIrNwIAPAtEjEPkIgBAAJJUpJrhGzjxvL7HzrkXl+/XurWzdz/bLNJQ4dKzz8vrVrlk3ABIKwF58T5AEEiBgABLDtbGjXKHE+aZIaAwkibNiYhKzBxovTyy9KuXaYeEVH+vWOFp+NL0uzZ5t6zCy9ksSoA8BQjYlUUH29WQwYABKjcXGnKFFOYa6cHH5R27jTJWX6+uXWu4HjrVunWW81+ZGW5/nqpRw+zV3bhKY0zZ7JKIwBUFiNiVXTCCebDBwAQoOx26dFHXcc4riCBKjhOSpJee80USVqyRHrhBSk2Vnr//fJf76abXMdxcVGqV6+bevc2/+br10unnurlNwAAIYBErIpatLA6AgBAmaKjpXHjrI4iKHXtaopkBhSfekpatkxaurT852Zm2pSSUuN4vV078/Xmm6W33zbHNlvA7K8NAJbhx2AVNWtmdQQAAPherVomEVuyxLUQSH6+lJUlvfSSWa2xImbMcG3rVng5/VGj3O9lA4BwQSJWRSRiABDgnE6zLOChQ/ym72U2m1n75M47zXL3eXnS/v3Sjh3SwYMOzZ07T/36VWwX6RdeMKNjNpt0//3S8uVsQA0gPJCIVVHTplZHAAAoU0aG2Sirdm1zDJ+JiJDq1pWaNHHtnT19ep6cTik9XTrjDNPWsmXZr/Pss1KXLu4jZuPHm+SMXBpAqCERqyJGxAAAKF98vPTLLyaR2rbNfM3Jkdatq9jzx40zyVlEhFm1kdEyAKGCRKyKSMQAIMDFx5vf+HNyzDECht1uFvEouOdszRrpuuvK3xZmzhwzWjZjhvTVV9Jff/klXADwCRKxKmJqIgAEOJvN/MZvt7PfSIBr395sFr1rl3tyduedJfcfMEDq2dNMdSy8n9l77zFiBiB4kIhVUUyM1REAABC62rc3qzI6nWYRkIro39/9/rJTTzXJHQAEIhIxAEBoysmRRo82JSfH6mjggSZNTEL26admufwbb6zY8zZsMM+12aSGDc2CIgMHSk8/LR075tuYAaA8JGJV8L//WR0BAKBcDof03HOmOBxWRwMvuOoqs1z+++/r+IqMR46YkbPy7NkjHTggvfOONGaMVL26lJnp85ABoFQkYlUwYIDVEQAAymW3S/fdZ4rdbnU08IH4eJNQ3Xmn694yp1M6eFAaOrRizy+Yxvivf1V8JUcA8AYSMQBAaIqONhtTPfusOUbYqFVLevNNk5Tt3Cn9+qs0darUq1fpz/ngA+n0001Sdv31ZpNqAPAlEjEAABCyGjeWTjtNuuUWad48k5yVtwXNnDlSVJSZxggAvkIiBgAITU6nuTfM4TDHwD9SUlzTGNevlx54oOR+AweaEbKKrtoIAJVBIlZJy5ZZHQEAoEIyMsyUxOhocwyU4NRTpaeeMklZamrJfZo1MwnZvfdKv/0mff+9tHkze5YB8AyJWCWsWyd17mx1FAAAwBcaNDAJWWl/dJ00SWrXTrr4Yunkk117lvXv7984AYQGErFKqFbN6ggAABUWH2+Wzzt40BwDFdS5sxntuvTSivV/7z2TkI0aJb3wAlMZAVQMiVglkIgBQBCx2czyebVqmWOgEmw26auvzAjZli1m9KtuXSk5ufTnvPCCScYKpjKOHeu/eAEEHxKxSiARAwAg/Jx4opSbK+3fL/38s2uhj61by37ek0+ahOyyy6RDh/wSKoAgQiJWCTExVkcAAKiwnBxp3DhTcnKsjgYhKClJOnpUGj/erLB47rkl9/v6a6l2bdfm0TabNHmydOut0p49fg0ZQACJsjqAYMLMFgAIIg6H+Q1ZkkaPZlNn+ES1atJ//uPetmSJdP75ZT/vnnvM1zfeMF+fe07697+l6tW9HyOAwMSIGAAgNEVFSSNGmBLF3x3hP127mqmM77xT8dk0990n1aghXXmlNH++eT6A0EYiBgAITTEx0iuvmMLccvhZZKR0881SVpbrnrLUVOmSS8zoV69eJT/viy+kK66Q7HYpLo69yIFQFhSJ2JQpU5SUlKTY2FglJydryZIlZfafOXOm2rdvr/j4eDVq1EiDBw9WWlqan6IFAAAorkEDacECszH0vHlmifz//a/0/llZUkSENG2a/2IE4D8Bn4jNmjVLI0eO1NixY7V69Wp17dpVPXv2VEpKSon9ly5dqgEDBmjIkCH67bffNHv2bK1cuVJDhw6tcgytWklz51b56QAAAMXYbNKAAWbUa+FCs1l0SYYMMX3HjJEOH/ZvjAB8J+ATsUmTJmnIkCEaOnSo2rZtq8mTJ6tZs2Z69dVXS+z/008/qWXLlrrrrruUlJSk8847T7feeqt+/vnnKsfw55/S1VdX+ekAACukp5v5XXa7OQYC2IUXSuvWmaSstEU+n37abIv39NP+jAyArwT03cs5OTlatWqVxowZ49beo0cPLV++vMTndO7cWWPHjtX8+fPVs2dP7d27V3PmzNEVV1xR6nmys7OVnZ19vH7kyBG3xx0OhwfvAihdwbXFNQZfC8trzeGQ/Z8VDxwOh1lFET4XlteaD+TkSC+8EKEHHogs9tiYMaZ8912uunYN35vIuNbgL766xmxOZ+DeBrpr1y41adJEy5YtU+fOnY+3P/nkk/rf//6n33//vcTnzZkzR4MHD1ZWVpZyc3PVq1cvzZkzR3a7vcT+48aN0/iCJY7dHJZUQ3PnzvPCuwEA+FV+vmIPHpQkZdWubW62AYLQp5+20rRpp5XZ55135qtGDRISwBcyMjLUr18/HT58WDVq1PDa6wZFIrZ8+XJ16tTpePsTTzyhGTNmaNOmTcWes2HDBl1yySW65557dOmll2r37t0aPXq0zjrrLE2dOrXE85Q0ItasWTMVJGI5Ofxgg284HA4tWLBA3bt3L/UPBYA3cK3BX7jWfGfGDJuGDCl7MlOjRk5t3pwbFguFcq3BX9LS0tSoUSOvJ2IBPTUxMTFRkZGRSk1NdWvfu3evGjRoUOJzJk6cqC5dumj06NGSpNNPP10JCQnq2rWrHn/8cTVq1KjYc2JiYhRTxk8s/nPD1+x2O9cZ/IJrDf7CteZ9t9xiypw50vXXl9xn926bqle367PPzJ5k4YBrDb7mq+sroOdpREdHKzk5WQsWLHBrX7BggdtUxcIyMjIUUWT6SWSkmV8dwIN/AABvy8mRnn3WlNJWPwCC0HXXmUU98vOlc88tuc9VV0lJSdLpp3P5A4EqoEfEJGnUqFG6+eab1bFjR3Xq1ElvvPGGUlJSNHz4cEnSgw8+qJ07d+qdd96RJF111VUaNmyYXn311eNTE0eOHKmzzz5bjRs3tvKtAAD8yeGQ7r/fHI8YIUVHWxsP4GU2m/Tjj+Y4M1OKj3d/fPt287XwpJ/sbP4rAIEi4BOxvn37Ki0tTRMmTNDu3bvVrl07zZ8/Xy1atJAk7d69221PsUGDBuno0aN6+eWXde+996pWrVq66KKL9DRrvQJAeImKkgYOdB0DISwuzoySnXWWVNaOPTExUkqK1KyZ/2IDULKg+GQaMWKERowYUeJj06dPL9Z255136s477/RxVACAgBYTI5XwGQGEspUrpb17pR07pFtvLTkpa95cOv98afFi/8cHwCWg7xEDAABA5dSvL515pknKStsg+ocfzNTGrVv9Hx8Ag0QMAAAghNntJiH7Z0FpNyecYBKyWbNMHwD+QyIGAAhN6elSrVqmpKdbHQ1guWeeKX0E7MYbzZ7nRXYMAuBDJGIAgNB1+LApACSZJe2dTqlevZIfL2G7VQA+QiJWhiuvzNfKlVZHAQCokrg4afNmU+LirI4GCCh795ol73v3Lv7Ym2/6PRwgLJGIlWH69Dx17Gh1FACAKomIkFq3NiWCjzugqNhY6ZNPit8b9u9/S8eOWRMTEE74ZAIAAAhzRZeyr17dLOKxbZs18QDhgEQMABCaHA7plVdMcTisjgYIaOefX3J7q1ZmGiMA7wuKDZ0BAKi0nBzpjjvM8aBBZg1vAKVyOs0oWFENGpivDocUxW+OgNfw3wkAEJoiI6XrrnMdAyiX02lKSbdV2u3SX39JzZv7Py4gFJGIAQBCU2ysNHu21VEAQcdmk7Zvl3r0MIuOFtaihfmak8MgM+Ap7hEDAACAmxYtpN9/L76iYoHoaJOwvfSSf+MCQgmJGAAAAErldErXX1/yY3ffbRIy9l0FKo9EDAAQmjIypCZNTMnIsDoaIKh9+GHpo2OSdPbZZk2cgwf9FhIQ9EjEAAChyemUdu0ypazfIAFUmNMpHTkiTZ5c/LH//U+qU8eMkC1Zwn87oDwkYgCA0BQbK61ebUpsrNXRACGjenUzJdHplM48s+Q+559vVl4cOtS/sQHBhEQMABCaIiOlDh1MYfl6wCdWrSq+smJhU6dK2dn+iwcIJiRiAAAAqLLWrV37j738cvHHY2PNdMXcXP/HBgQyEjEAQGhyOKTp001xOKyOBggLt99e+r1hdrv08cf+jQcIZCRiAIDQlJMjDR5sSk6O1dEAYSU3V2rUqHh7nz5mdCw11f8xAYGGRAwAEJoiI6XLLzeFe8QAv4qMNAuWlvY3kEaNTELGCBnCGYkYACA0xcZKX3xhCqsmApaw281UxaefLvnxghEyIByRiAEAAMCn7r/fJGTR0SU/TjKGcEQiBgAAAL/IzjYJ2auvFn/MZjNl927/xwVYgUQMABCaMjLMutqtW5tjAAFj+HBpyZKSH2vcWLr+ev/GA1iBRAwAEJqcTumPP0wpbT1tAJY577zSR7/mzDGjY5dc4t+YAH8iEQMAhKbYWGnpUlNYrAMISA0bujaDbtiw+OPffWcSsunT/R4a4HMkYgCA0BQZKXXpYgrL1wMBb/duac2akh8bPFj6+mu/hgP4HIkYAAAAAkL79mZ0bP364o9ddpn099/+jwnwFRIxAEBoys2VZs82JTfX6mgAVMKpp5qEbNQo9/bmzaXXX7cmJsDbSMQAAKEpO1u64QZTsrOtjgZAFTz/fPG24cOlgQP9HwvgbSRiAIDQFBEhXXCBKRF83AHByumUxoxxb3vnHSk62q4DB1iIB8EryuoAAADwibg4adEiq6MA4AUTJ0qnnSb17+/efsstl2rv3jw98IA1cQGe4E+EAAAACHj9+kk//li8fcyYSD37rP/jATxFIgYAAICgcO65UkZG8fb775euv97/8QCeIBEDAISmzEypQwdTMjOtjgaAl8TFmfvGJkzIc2ufM0c66SSLggKqgEQMABCa8vOltWtNyc+3OhoAXjZmTL6uvXaLW9uWLZLNZlFAQCWRiAEAQlNsrPTNN6bEsrIaEIoGDNigTZscxdptNumzzywICKgEVk0EAISmyEipe3erowDgY61aSbt3S40aubf36mW+7tkj1a/v/7iA8jAiBgAAgKDWsKH0558lP9aggX9jASqKRAwAEJpyc6UvvjAlN9fqaAD4WKtWZhGPBQuKP7Zpk//jAcpDIgYACE3Z2dKVV5qSnW11NAD85JJLTEJWWNu20rhxJS99D1iFRAwAEJoiIqSOHU2J4OMOCDd9+7rXx4+XEhKkF1+0Jh6gKD6ZAAChKS5OWrnSlLg4q6MB4GcffFBy+8iRxUfMACuQiAEAACAkOZ3SV18Vb2eQHIGAyxAAAAAh69JLSx4Bs9nMsveAVUjEAAChKTNT6tLFlMxMq6MBYLGSVk5s3NgkZH/84f94ADZ0BgCEpvx8afly1zGAsHbyydLRo1L16sUfa91aysqSYmL8HxfCF4kYACA0xcRIn3ziOgYQ9qpVM9MUbbbij8XGSqtWSWee6f+4EJ6YmggACE1RUVLv3qZE8XdHAC5Op5SeXrw9OVm66iozOgb4GokYAAAAwk58fMmLeHz+uUnGAF8jEQMAhKa8PGnRIlPy8qyOBkCAcjqlUaPc2779VvrtN2viQfggEQMAhKasLKlbN1OYZwSgDM8/L6Wlube1ayc99hibP8N3SMQAAKHJZpNOOcWUku7MB4BC6tQxf7cp7D//MZs/M6gOX+DuZQBAaIqPZ24RgEr57juTeBUVFcXy9vA+RsQAAAAAmcFzp1PasKH4Y7GxJa+0CFQViRgAAABQSNu20p49xdurVZO2b/d7OAhRJGIAgNCUmSl1725KZqbV0QAIMvXrS/n5xduTkvwfC0ITiRgAIDTl55s1qL/9tuTfpgCgHAVTFYsu4jF+vDXxILSQiAEAQlNMjPTuu6Zwhz0AD3z/vXt93DhLwkCIIREDAISmqCipf39TolgkGIBnfvjBvX7//dbEgdBBIgYAAACUo2tX9/qzz7JXPDxDIgYACE15edLKlaawGysAL/jlF/d6XJzUtKk1sSD4kYgBAEJTVpZ09tmm8GdrAF5wxhnF23buNIt6/Pe//o8HwY1EDAAQmmw2qUULU2w2q6MBECJyc6Xo6OLtd91VfHVFoCwkYgCA0BQfb3Ze3b7dHAOAF0RGStnZ0uHDxR9btEh66SW/h4QgRSIGAAAAVFKNGmaPsfXr3dvvvtuaeBB8SMQAAACAKjr1VOnhh93bmA2NiiARAwCEpqwsqXdvU1isA4APPfZY8TaSMZSHHS4BAKEpL0+aN891DAA+lJtbfO94m838+Ilg6AMlIBEDAISm6GjpjTdcxwDgQ5GR0pYtUuvWxds3bpTatLEmLgQuEjEAQGiy26Vhw6yOAkAYOfFE6bvvpIsvdm9v29Ys7AEUxkApAAAA4CUXXWSmKZbUDhTGiBgAIDTl55v5QJL5czQ3aQDwk8hIMwJWeMGOhQtNnZExFOBTCQAQmjIzpXbtTMnMtDoaAGFo8eLibWPG+D8OBCYSMQBA6EpMNAUALHD++dI337i3Pf00yRgMEjEAQGhKSJD27TMlIcHqaACEqe7dpd273dueftqaWBBYSMQAAAAAH2rYUBoxwr2NDZ9BIgYAAAD42CuvFG+z2aScHP/HgsBAIgYACE1ZWVL//qZkZVkdDQBo//7ibTEx/o8DgYFEDAAQmvLypPfeMyUvz+poAEB165qdNYrib0XhiX3EAAChKTpaeuEF1zEABACbzSRjhbc2rFWLZCwcBcWI2JQpU5SUlKTY2FglJydryZIlZfbPzs7W2LFj1aJFC8XExOiEE07QtGnT/BQtACAg2O3SyJGm2O1WRwMAx9lsZjXFAtnZ0ty5loUDiwT8iNisWbM0cuRITZkyRV26dNHrr7+unj17asOGDWrevHmJz7nhhhu0Z88eTZ06VSeeeKL27t2r3NxcP0cOAAAAlOzLL6WoQr+JX3ON5HRaFw/8L+BHxCZNmqQhQ4Zo6NChatu2rSZPnqxmzZrp1VdfLbH/V199pcWLF2v+/Pm65JJL1LJlS5199tnq3LmznyMHAFgqP1/avt2Ukm7KAAALRUZK//2ve9sFF1gTC6wR0CNiOTk5WrVqlcYU2X68R48eWr58eYnP+fTTT9WxY0c988wzmjFjhhISEtSrVy899thjiouLK/E52dnZys7OPl4/cuSIJMnhcMjhcHjp3QDFFVxfXGfwtbC81tLTZU9KkiQ5Dh5kU2c/CctrDZYIhWvt1lulO+90TZ3+4QczbTE728E+YwHEV9dYQCdi+/fvV15enho0aODW3qBBA6Wmppb4nK1bt2rp0qWKjY3VJ598ov3792vEiBE6cOBAqfeJTZw4UePHjy/WvnDhQsXHx3v+RoByLFiwwOoQECbC6VqLzMrSZf+sC/31118rLzbW4ojCSzhda7BWsF9rjz9eVw8/fJ5bW0yMXTNmzFf16sGbZIaSjIwMn7yuzekM3Nmou3btUpMmTbR8+XJ16tTpePsTTzyhGTNmaNOmTcWe06NHDy1ZskSpqamqWbOmJOnjjz/Wddddp/T09BJHxUoaEWvWrJl2796tunXr+uCdAYbD4dCCBQvUvXt32VlMAD7EtQZ/4VqDv4TStbZhg9ShQ/H3kJXlcFtdEdZIS0tTo0aNdPjwYdWoUcNrr+u1ETGn06lff/1VW7du1bFjx1RWfjdgwIAKvWZiYqIiIyOLjX7t3bu32ChZgUaNGqlJkybHkzBJatu2rZxOp3bs2KHWrVsXe05MTIxiSthNz263B/1/bAQHrjX4C9ca/IVrDf4SCtda+/Zm+fqiA/exsXbl5YlkzGK+ur68koi99957euCBB7Rr164K9a9oIhYdHa3k5GQtWLBA11xzzfH2BQsW6Oqrry7xOV26dNHs2bN17NgxVatWTZK0efNmRUREqGnTphU6LwAAAOBPMTHF9xeTzKIegTt/DZ7wOBGbPXu2brrpJklSw4YN1b59e9WvX18RXkrdR40apZtvvlkdO3ZUp06d9MYbbyglJUXDhw+XJD344IPauXOn3nnnHUlSv3799Nhjj2nw4MEaP3689u/fr9GjR+uWW24pdbEOAEAIys6W7rjDHL/8svktBwACmM1mkq6iC3UcOSJ5cUYcAoTHidjTTz8tm82mBx54QBMmTFBUlHfX/+jbt6/S0tI0YcIE7d69W+3atdP8+fPVokULSdLu3buVkpJyvH+1atW0YMEC3XnnnerYsaPq1q2rG264QY8//rhX4wIABLjcXOmtt8zx5MkkYgCCRtFkrGZN8yMtMtK6mOB9HmdNGzZsUL169fTkk096I54SjRgxQiNGjCjxsenTpxdra9OmTdCvoAMA8JDdLhX8ES7I7x8BEH769JE++shVj4oyUxdZ1j50eJyIJSQkqHnz5t6IBQAA74mOlsaOtToKAKiS2bOL3y/WoYO0dq0l4cAHPL6R68ILL9TmzZuVk5PjjXgAAACAsGezSQcPurf9+quUmWlNPPA+jxOxxx9/XPn5+br//vu9EQ8AAN7hdEr79pnCkmMAglCtWtKePe5t8fGWhAIf8Hhq4p49ezRu3Dg9+OCDWrJkiQYPHqwTTjhBCQkJpT7n/PPP9/S0AACULSNDql/fHB87JpXxuQQAgap+falOHenAAVfbffdJzz1nXUzwDo8TsQsvvFA2m01Op1OrV6/WmjVryuxvs9mUm5vr6WkBAACAsLBjh/tI2PPPS9dfL51zjnUxwXMeJ2Lnn3++bCzfAgAINAkJTEkEEBLi4qR166TTTnO1depkVlFE8PI4EVu0aJEXwgAAAABQmnbtpAsvlAp+9XY6pVmzpL59rYwKnvB4sQ4AAAAAvvf55+71G2907VuP4EMiBgAITdnZ0siRpmRnWx0NAHgsIUH68EP3tmHDpEOHLAkHHvJaIlawemLnzp2VmJiomJgYJSYmqnPnzpowYYL27t3rrVMBAFC+3FzpxRdNYZEoACHi+uulu+92b6td25pY4BmP7xGTpC+//FL9+/fX4cOH5Sx0Y/SBAwf0008/acWKFXrxxRc1c+ZMXXbZZd44JQAAZbPbpYcech0DQIiYPFn65RdpyRJXW26uFOWV3+zhLx6PiG3atEl9+vTRoUOHdMopp+j111/X0qVLtWXLFi1dulSvv/66TjnlFB08eFDXXnutNm3a5I24AQAoW3S09MQTpkRHWx0NAHjV11+714cMsSYOVJ3HidjEiROVlZWl22+/XevWrdOwYcPUuXNnnXDCCercubOGDRumdevW6Y477lBWVpaeeuopb8QNAAAAhK24OKl6dVf9nXekgwetiweV53Ei9v3336t27dqaNGlSmf2ef/551apVS999952npwQAoHxOp5Sebgr7iQEIQRs2uNevvtqaOFA1Hidie/fu1Yknnih7OfPv7Xa7WrdurX379nl6SgAAypeRIVWrZkpGhtXRAIDXNW3qXi98zxgCn8eJWO3atZWSklJuP6fTqZSUFNWqVcvTUwIAAAAQ0xGDmceJWOfOnbV3795ypya+8MIL2rNnj7p06eLpKQEAKF98vHTsmCnx8VZHAwA+UXSMo+g+YwhcHidi9913nyRp9OjR6tOnjxYuXKg9e/bI6XRqz549Wrhwoa699lqNHj1aERERx/sDAOBTNpvZ/TQhwRwDQBjo21fKz7c6ClSEx7sNdO7cWS+//LLuvvtuzZ07V3Pnzi3Wx+l0KioqSi+99JI6derk6SkBAAAA/OPdd6WbbnLVIyNZoygYeDwiJkm33XabVq5cqX/9619KTEyU0+k8XhITE3XTTTdp5cqVGj58uDdOBwBA+XJypLFjTcnJsToaAPCZ/v2Lt5UwNoIA47X9t9u3b693331XknT48GEdO3ZM1apVU82aNb11CgAAKs7hkJ580hw/9BCbOgMIaQ6HVHgR82uukXJzzegYApPXErHCatasSQIGALBWVJR0992uYwAIYVFR0tSp0pAh7m35+dwmG6i8MjURAICAExMjTZ5sSkyM1dEAgM8NHly8rXlz/8eBiqnUnwgnTJggSUpMTNSIESPc2irKZrPpkUceqdRzAAAAAJTNZjO3xBaeib1jh5SZKcXFWRcXSlapRGzcuHGy2Ww6+eSTjydiBW3OcpZmKehDIgYAAAD4ht0u7d0r1a/vaouPZxXFQFSpROzRRx+VZEbEirYBABBQ0tOlatXM8bFjZj8xAAgD9eoVb7vvPum55/wfC0pXpUSsvDYAAAAA1nE63RfpeP55UxgZCxws1gEACE3x8WZ+zt695hgAwsyMGcXbzjjD/3GgZD5PxPbs2aPVq1crIyPD16cCAMDFZjPzc+rVY+1mAGHpppukAwfc29asYY/7QOFxIrZixQqNGjVKX3zxhVv7kSNHdPXVV6tx48bq2LGjGjZsqLffftvT0wEAAACooNq1pdRU9zZ29AgMHidib731ll588UVVr17drX306NH67LPPZLPZVKtWLR07dkzDhg3TunXrPD0lAADly8mRnnjCFP78CyCMNWhgVlMsjIkC1vM4EVu2bJkSEhJ0/vnnH287duyYZsyYoerVq2v9+vVKS0vT5MmTlZ+fr+eff97TUwIAUD6HQ3r4YVMcDqujAQBL7d9fvO2mm/wfB1w8TsT27NmjZs2aubUtXrxYWVlZ6tu3r9q0aSNJuuOOO5SYmKgVK1Z4ekoAAMoXFSUNHWpKVKUWCQaAkFOjhrR7t3vbzJnWxALD40Ts6NGjii+yGtXSpUtls9nUvXt314kiItSyZUv9/fffnp4SAIDyxcRIb75pCjdEAIAaNpR+/tm9bcIEa2KBFxKxunXr6q+//pKz0KYE3377rSTpggsucOvrcDgUHR3t6SkBAAAAVEFysnv90UfZW8wqHidi5557rtLS0vTmm29KMknYqlWr1L59e9WvX/94P6fTqT/++EONGjXy9JQAAAAAquinn9zrRdbcg594nIjde++9stlsuu2225SYmKjLLrtMNptN9957r1u/H374Qenp6TrrrLM8PSUAAOVLT5cSEkxJT7c6GgAIGOec415PT2cVRSt4nIidd955+uijj3TKKacoPT1drVq10ssvv6z+/fu79XvttdckST169PD0lAAAVExGhikAADe//FK8bdQo/8cRzryyjNTVV1+tq6++usw+b7zxhl577bVi+40BAOATcXHStm2uYwDAcWecIR05YlZTLPDCC9KkSdbFFG48HhGrqOrVq6tmzZqKiPDbKQEA4SwiQmrZ0hQ+ewCgmOrVpXnz3NvuuceaWMIRn0wAAABAmOrVy70+ebIlYYSlSk1NvOWWWyRJjRo10hNPPOHWVlE2m01Tp06t1HMAAKg0h0N65RVzfPvtkt1ubTwAEKC++EK64gpX3eHgR6Y/VCoRmz59uiSpTZs2xxOxgraKIhEDAPhFTo5rjs2wYfxWAQClKLqW3jPPSGPHWhNLOKlUIvb2229LkmrWrFmsDQCAgBIZKfXr5zoGAJQoqkhG8PDDJGL+UKlEbODAgRVqAwDAcrGx0syZVkcBAEFh2jSp8B1HeXn8DcvXWKwDAAAACHODBrnXn3zSkjDCiseJmMPhUEpKitLS0srsl5aWppSUFOXm5np6SgAAAABeZLO5j4D95z/WxRIuPE7E3nzzTSUlJWnGjBll9psxY4aSkpI0bdo0T08JAED50tOlevVMSU+3OhoACHjffONe37zZmjjChceJ2Jw5cxQREaFBRcczixg0aJAiIiI0e/ZsT08JAEDF7N9vCgCgXBdd5F4/+WRr4ggXHidiv//+u5o1a6ZatWqV2a9WrVpq1qyZfv/9d09PCQBA+eLipPXrTYmLszoaAAgKd93lXs/KsiaOcOBxIpaWlqZ69epVqG+9evW0b98+T08JAED5IiKkU081JYK1qQCgIl580b3O37F8x+NPprp162rbtm0V6rtt2za3PcgAAAAABJbWrd3rdetaE0eo8zgRO+ecc5SWlqb333+/zH4ffPCB9u/fr3POOcfTUwIAUD6HQ3rzTVMcDqujAYCgUfROogMHrIkj1HmciN12221yOp3697//rffee6/EPu+//76GDRsmm82m2267zdNTAgBQvpwc6d//NiUnx+poACBo2GxSSop720svWRNLKIvy9AW6d++uESNGaMqUKbr55pv1wAMP6Oyzz1atWrV06NAhrVy5Ujt37pTT6dSIESN02WWXeSNuAADKFhkpXX216xgAUGHNmrnX775buu02yW63Jp5Q5HEiJkkvv/yyWrdurccff1w7d+7UJ5984vZ4YmKiHnnkEd15553eOB0AAOWLjZXmzrU6CgAIWr/+Kp1+uqseHS05ndbFE2q8kohJ0t13361bb71Vy5Yt08aNG3XkyBFVr15dp556qrp06aKYmBhvnQoAAACAj512WvG2Vauk5GT/xxKKvJaISVJsbKwuvvhiXXzxxd58WQAAAAAWyMtzn93dsaOUn2/uI4NnvL6xSn5+vvbt26eUonf4AQDgTxkZUsuWpmRkWB0NAASliAhpyBD3tnfesSaWUOO1RGz+/Pnq3r27qlevroYNG6pVq1Zujz/xxBPq168fGzoDAPzD6ZT++ssUbmoAgCp76y33+qBBUna2JaGEFK8kYvfff7+uuuoqfffdd8rLy5PdbpezyIdeo0aNNGvWrGILeQAA4BOxsdL//Z8psbFWRwMAQa3o2kdTp1oSRkjxOBH76KOP9Nxzz6lx48b6/PPPlZ6errPOOqtYv2uuuUaS9Omnn3p6SgAAyhcZKZ11liksXw8AHinYDaTAtGnWxBFKPE7EXnnlFdlsNs2ePVuXX365Ikv5sKtdu7aSkpK0ZcsWT08JAAAAwM/++1/X8apV1sURKjxOxFavXq1mzZrp3HPPLbdvvXr1tHPnTk9PCQBA+XJzpZkzTcnNtToaAAh611/vXn/1VWviCBUeJ2LZ2dmqVatWhfpmZGSUOmIGAIBXZWdLN91kCneVA4DHGjRwr48YYU0cocLjRKxZs2b6448/5HA4yux3+PBhbdq0SSeccIKnpwQAoHwREdIll5gS4fXdWgAgLM2b517v0MGSMEKCx59Ml156qTIzM/XCCy+U2W/ChAnKzc3VlVde6ekpAQAoX1yctGCBKXFxVkcDACGhVy/3+tq11sQRCjxOxB544AFVr15dDz30kEaPHq1NmzYdfyw/P1+//vqrbrnlFr3wwgtKTEzU3Xff7ekpAQAAAFik6EQ4lrKvGo8TsSZNmmjevHmqWbOmJk2apFNPPVXLly+XJNntdp1xxhmaPn266tSpo08++UR169b1OGgAAAAA1oiKcq8zzlI1Xpk0f8EFF2j9+vUaOXKkWrRoIafTebw0atRId9xxh9auXavOnTt743QAAJQvI0M69VRTMjKsjgYAQsrnn7uO09OtiyOYRZXfpWIaNWqk559/Xs8//7zS09N1+PBhVatWTTVq1PDWKQAAqDinU9qwwXUMAPCaSy5xr6elSUx8qxyPE7GIiAjVqVNHO3fuVExMjCQpISFBCQkJHgcHAECVxcZKCxe6jgEAXvPPr/3HNW0qZWZaE0uw8nhqYrVq1XTCCSccT8IAAAgIkZHShReawh6WAOB13bq5jrOypPx862IJRh4nYm3atNGePXu8EQsAAACAIDFnjnv9p5+siSNYeZyIDRs2TCkpKfriiy+8EQ8AAN6RmyvNnWtKbq7V0QBAyKlTx73epQv7ilWGVxKx4cOH61//+pdefPFFHThwwBtxAQDgmexs6ZprTMnOtjoaAAhJb7/tXu/QwZIwgpLHi3W0atVKkpSZmalRo0Zp1KhRSkxMLHWxDpvNpj///NPT0wIAULaICKlg25QIr+zWAgAoYtAg6a67pKNHXW1Ll0rnnWdZSEHD40Rs+/btxdr27dunffv2ldjfZrN5ekoAAMoXFyctW2Z1FAAQ8o4ckQr/iv/88yRiFeFxIrZt2zZvxAEAAAAgSA0fLr32mjmeO9fSUIJGlROxjIwMLViwQFu2bJEknXjiierevTv7hwEAAABhZsAAVyImSZ9/Ll15pXXxBIMqJWJffPGFBg8erLS0NLf22rVr66233lLv3r29ERsAAFWXmSmdf745/uEHM1URAOAT557rXr/qKsnptCaWYFHpu5c3bNig6667Tvv371d0dLROPfVUnXLKKYqOjtaBAwd044036tdff/VqkFOmTFFSUpJiY2OVnJysJUuWVOh5y5YtU1RUlDqwfAsAhJ/8fOnnn01hl1EA8CmbTZoxw71t3jxrYgkWlU7Enn/+eWVnZ6t79+7avn27fv31V61bt07btm3TxRdfrJycHE2aNMlrAc6aNUsjR47U2LFjtXr1anXt2lU9e/ZUSkpKmc87fPiwBgwYoIsvvthrsQAAgkhMjJkb8/nn5hgA4FP9+rnX77/fmjiCRaUTscWLFysmJkbvvvuuGjRocLy9YcOGmjlzpqKjo7V48WKvBThp0iQNGTJEQ4cOVdu2bTV58mQ1a9ZMr776apnPu/XWW9WvXz916tTJa7EAAIJIVJR0xRWmRHm8NhUAoBwREdKtt7rqmzdbF0swqPQn065du9S6dWvVq1ev2GP169dX69at9ccff3gluJycHK1atUpjxoxxa+/Ro4eWL19e6vPefvtt/fnnn3r33Xf1+OOPl3ue7OxsZRfa7PPIkSOSJIfDIYfDUcXogfIVXF9cZ/A1rjX4C9ca/IVrLTA9/bT0+uv24/VVqxw6/XQLA/ICX11jlU7EsrKyVKtWrVIfr1WrlnJycjyJ6bj9+/crLy/PbeRNkho0aKDU1NQSn7NlyxaNGTNGS5YsUVQF/wI6ceJEjR8/vlj7woULFR8fX/nAgUpasGCB1SEgTITVtZaXp3rr1kmS9p12mhQZaXFA4SWsrjVYimstEF19/OiKK3L0+uvfWhiL5zIyMnzyukExV6PoJtBOp7PEjaHz8vLUr18/jR8/XieddFKFX//BBx/UqFGjjtePHDmiZs2aqVu3bqpbt27VAwfK4XA4tGDBAnXv3l12u738JwBVFJbXWnq67H36SJIcBw9KbK/iF2F5rcESXGuB67rr8jVnjrkDqnr1eF1++eUWR+SZoivFe0uVErG9e/fqnXfeKfUxSZoxY4acpaxZOWDAgAqdJzExUZGRkcVGv/bu3VtslEySjh49qp9//lmrV6/WHXfcIUnKz8+X0+lUVFSUvvnmG1100UXFnhcTE6OYEm7kttvt/MeGX3CtwV/C6lqLiZHat5ck2WNipHB53wEirK41WIprLfD85z/SnDnm+I8/bEH//fFV/FVKxLZs2aLBgweX2WfQoEEltttstgonYtHR0UpOTtaCBQt0zTXXHG9fsGCBrr766mL9a9SooXX/TEMpMGXKFH3//feaM2eOkpKSKnReAEAIiIuT1qyxOgoACDtt2rjX8/KYHV6SSidizZs3L3FaoK+MGjVKN998szp27KhOnTrpjTfeUEpKioYPHy7JTCvcuXOn3nnnHUVERKhdu3Zuz69fv75iY2OLtQMAAADwvqIDSG+/LQ0dak0sgazSidj27dt9EEbp+vbtq7S0NE2YMEG7d+9Wu3btNH/+fLVo0UKStHv37nL3FAMAAABgjVGjSMRKUul9xKwwYsQIbd++XdnZ2Vq1apXOP//8449Nnz5dixYtKvW548aN0xqmpgBA+MnMlC680JTMTKujAYCwMnu26/joUem336yLJVAFxaqJAABUWn6+tHix6xgA4De9ernX27WTSlnHL2wFxYgYAACVFhMjffihKSWsjAsA8J3oaOm559zbvv/emlgCFYkYACA0RUVJ119vShQTQADA3wpt0ytJuvhiRsUKIxEDAAAA4HU2mzR5snvb/PmWhBKQSMQAAKEpL09atsyUvDyrowGAsHT33e71K6+0Jo5ARCIGAAhNWVnSeeeZkpVldTQAELbeece9zkK2BokYACA02WzSiSeaYrNZHQ0AhK2bb3avF13EI1yRiAEAQlN8vLRliynx8VZHAwBhLaJQ1vGf/1gXRyAhEQMAAADgU59/7l7PzrYmjkBCIgYAAADAp3r2dK/HxloTRyAhEQMAhKasLOmKK0xhsQ4ACDi7dlkdgbXY4RIAEJry8lwb1rB8PQBYLjtbiolx1Zs0Ce8NnhkRAwCEpuho6e23TYmOtjoaAAh70dHS7be7t40da00sgYBEDAAQmux2adAgU+x2q6MBAEh6+WX3+pNPWhNHICARAwAAAOA3Cxe6148etSYOq5GIAQBCU16etGaNKdwjBgAB44IL3Os33WRNHFZjsQ4AQGjKypLOOMMcHzsmJSRYGw8AQJJks0l16kgHDpj6p59aG49VSMQAAKHJZpMaN3YdAwACxm+/SY0aWR2FtZiaCAAITfHx0s6dpsTHWx0NAKCQhg3d61u2WBOHlUjEAAAAAFiq6H1j4YBEDAAAAIDf3XKL63j37vDb3JlEDAAQmrKypOuvNyUry+poAABF/Pe/7vWIMMtMwuztAgDCRl6eNGeOKSxfDwABJ9xv32XVRABAaIqOll5+2XUMAAg4GRnuCdn06dKgQVZF418kYgCA0GS3S7ffbnUUAIAyxMW517dutSYOKzA1EQAAAIBlZs1yHT/2mHVx+BsjYgCA0JSfL/35pzk+4YTwuwscAIJEgwbudZstPFZQ5FMJABCaMjOlk04yJTPT6mgAAKUoaQ+xuXP9HobfkYgBAEJXzZqmAAAC2tGj7vVrrgn9BW9JxAAAoSkhQTp0yJSEBKujAQCUoVq14qNgUSF+ExWJGAAAAADLXX211RH4F4kYAAAAgICQne1ev+sua+LwBxIxAEBoys42u4IOGlT8kx0AEJCio93r//2vNXH4A4kYACA05eZK//ufKbm5VkcDAKigNWvc63//bUkYPhfit8ABAMKW3S4984zrGAAQFNq3d683bx6a+4oxIgYACE3R0dLo0aYUnesCAAhoDz9sdQS+RyIGAAAAIKA89ph7fdMma+LwJRIxAEBoys+Xdu40JT/f6mgAAJUUGek6HjjQujh8hUQMABCaMjOlpk1Nycy0OhoAQCU9+6zr+P/+L/TuEyMRAwCErqgoUwAAQWfwYPd6kybWxOErfDoBAEJTQoLkcFgdBQCgimrVcq/v3m1JGD7DiBgAAACAgHT0aNn1YEYiBgAAACAgVavmXn/wQWvi8AUSMQBAaMrOlm6/3ZTsbKujAQBUUbNmruNXXrEuDm8jEQMAhKbcXGnKFFNyc62OBgBQRVOnutfXr7cmDm9jsQ4AQGiy26VHH3UdAwCCUvfu7vVly6R27ayJxZsYEQMAhKboaGncOFOio62OBgDggbvvdh2//rp1cXgTiRgAAACAgFZ40Y46dayLw5tIxAAAocnplA4dMsXptDoaAIAHrrvOdfzdd9bF4U3cIwYACE0ZGVLt2ub42DGzwTMAICg1bGh1BN7HiBgAAACAgFa/vnv9wAFr4vAmRsQAAKEpPl7KyTHHUXzcAUAwiygyfFS3bvDPOmdEDAAQmmw2s2y93W6OAQBBregy9nl51sThLSRiAAAAAALe11+714N9sgOJGAAgNOXkSKNHm1IwRREAELRCbXIDiRgAIDQ5HNJzz5nicFgdDQDAC7Kz3evHjlkThzcE+YAeAAClsNul++5zHQMAgl50tHv9ttukGTOsicVTJGIAgNAUHS09+6zVUQAAvKxOHdfy9e++G7yJGFMTAQAAAASNhQvd63/9ZU0cniIRAwCEJqfT3BvmcAT/ZjMAgONOP929/uCD1sThKRIxAEBoysgw0xOjo80xACBknH226/j994NzTzESMQAAAABB5eOP3evNm1sThydIxAAAoSk+Xjp40JT4eKujAQB4UZMm7vVdu6SlS62JpapYNREAEJpsNqlWLaujAAD4SF6eFBnpqnftGly3BDMiBgAAACDoRERI997r3vbDD9bEUhUkYgCA0JSTI40bZ0pOjtXRAAB84Kmn3OuzZ1sTR1WQiAEAQpPDIY0fb4rDYXU0AAAfiIqSHnnEVX/5ZetiqSwSMQBAaIqKkkaMMCWKW6IBIFT16eNez8+3Jo7K4pMJABCaYmKkV16xOgoAgI+1b+9eX7NGOvNMS0KpFEbEAAAAAAS16tVdx8nJ1sVRGSRiAAAAAILapZe613//3Zo4KoNEDAAQmtLTJbvdlPR0q6MBAPjQBx+41595xpo4KoNEDAAQunJzTQEAhLTISOmxx1z1adMCPxkjEQMAhKa4OGnHDlPi4qyOBgDgY7ff7l5/4AFr4qgoEjEAQGiKiJCaNDElgo87AAh1tWtLM2e6t2VnWxNLRfDJBAAAACAk9OvnXl+71po4KoJEDAAQmnJypGefNSUnx+poAAB+cuqpruNly6yLozwkYgCA0ORwSPffb4rDYXU0AAA/iYpyHY8aZV0c5SERAwCEpqgoaeBAUwp/KgMAQtqdd7rX8/KsiaM8JGIAgNAUEyNNn25KTIzV0QAA/OSWW9zrX35pTRzlIREDAAAAEDJsNvf6+PHWxFEeEjEAAAAAIWXYMKsjKF9QJGJTpkxRUlKSYmNjlZycrCVLlpTa9+OPP1b37t1Vr1491ahRQ506ddLXX3/tx2gBAAEhPV2qVcuU9HSrowEA+NHw4a7j1auti6MsAZ+IzZo1SyNHjtTYsWO1evVqde3aVT179lRKSkqJ/X/44Qd1795d8+fP16pVq9StWzddddVVWh2o3wEAgO8cPmwKACCsNG7sOg7UxToCfhmpSZMmaciQIRo6dKgkafLkyfr666/16quvauLEicX6T5482a3+5JNPat68efrss890xhlnlHiO7OxsZRfadvvIkSOSJIfDIQdLHsOHCq4vrjP4Wlhea1FR0m+/uY7D6b1bKCyvNViCaw1lqVtXkuzH6zk5jmL3jlWUr66xgE7EcnJytGrVKo0ZM8atvUePHlq+fHmFXiM/P19Hjx5VnTp1Su0zceJEjS/hLr6FCxcqPj6+ckEDVbBgwQKrQ0CYCNtrbcsWqyMIO2F7rcHvuNZQuquPH11zzd/697/XVelVMjIyvBWQm4BOxPbv36+8vDw1aNDArb1BgwZKTU2t0Gs8//zzSk9P1w033FBqnwcffFCjCu32duTIETVr1kzdunVTXZNOAz7hcDi0YMECde/eXXa7vfwnAFXEtQZ/4VqDv3CtoTLmz2+luXObVem5aWlpXo7GCOhErICtyDii0+ks1laS999/X+PGjdO8efNUv379UvvFxMQopoQ9Zux2O/+x4Rdca/CXsLrWHA7pjTfM8b//LYXL+w4QYXWtwVJcayjNzz9LHTu66ps22XXaaZV/HV9dXwG9WEdiYqIiIyOLjX7t3bu32ChZUbNmzdKQIUP04Ycf6pJLLvFlmACAQJSTI91xhyk5OVZHAwDws+Rk9/rTT1sTR2kCOhGLjo5WcnJysbm/CxYsUOfOnUt93vvvv69Bgwbpvffe0xVXXOHrMAEAgSgyUrruOlMiI62OBgBggcLr+C1caFkYJQroREySRo0apbfeekvTpk3Txo0bdc899yglJUXD/9kc4MEHH9SAAQOO93///fc1YMAAPf/88zr33HOVmpqq1NRUHWb5YgAIL7Gx0uzZpsTGWh0NAMACV17pOt61S6rgen9+EfD3iPXt21dpaWmaMGGCdu/erXbt2mn+/Plq0aKFJGn37t1ue4q9/vrrys3N1e23367bb7/9ePvAgQM1ffp0f4cPAAAAwCItW7rXu3SRnE5LQikm4BMxSRoxYoRGjBhR4mNFk6tFixb5PiAAAAAAAS8yUho3zpRAE/BTEwEAqJKMDKlJE1N8tAcMACDwjR1rdQQlC4oRMQAAKs3pNDcEFBwDAMJSVJGMJz9figiA4agACAEAAB+IjZVWrzaFxToAAP/44QerIzBIxAAAoSkyUurQwRSWrweAsBYX5zr+8Ufr4iiMRAwAAABASLvnHtfxQw9ZF0dh3CMGAAhNDoc0c6Y57t9fstutjQcAYJnTTnOvO52SzWZNLAUYEQMAhKacHGnwYFNycqyOBgBgoRtucK9/9JE1cRRGIgYACE2RkdLll5vCPWIAENaKrpJIIgYAgK/ExkpffGEKqyYCQNgbNcp1/MEH1sVRgEQMAAAAQMgbMMC9npVlTRwFSMQAAAAAhLz27d3rycnWxFGARAwAEJoyMqTWrU3JyLA6GgBAAKhZ03W8YYN1cUgsXw8ACFVOp/THH65jAEDY27PH/bbh/PziC3n4CyNiAIDQFBsrLV1qCot1AAAkxcS4163c3JkRMQBAaIqMlLp0sToKAEAA27zZunMzIgYAAAAgbCxZ4jq2cptJRsQAAKEpN1f65BNzfM01UhQfeQAAqVEj1/GcOdbFwacSACA0ZWdLN9xgjo8dIxEDAEiSGjZ0r3/3nXTxxf6Pg08lAEBoioiQLrjAdQwAgKSEBPf6JZdYs7gun0wAgNAUFyctWmRKXJzV0QAAAsisWe71w4f9HwOJGAAAAICwUjBzvcDKlf6PgUQMAAAAQNjp0cN13L27/89PIgYACE2ZmVKHDqZkZlodDQAgwHTu7F5fu9a/5ycRAwCEpvx886m6dq05BgCgkEcfda+PHevf85OIAQBCU2ys9M03psTGWh0NACAAXXml63jxYv+em0QMABCaIiPNpP/u3c0xAABFjBjhOj52zL/nJhEDAAAAEJYKL9jhbyRiAIDQlJsrffGFKbm5VkcDAAhARSdM7Nnjv3NH+e9UAAD4UXa2a/L/sWNSFB95AICy9e4t/fijf87FiBgAIDRFREgdO5oSwccdAKBk3bq5jn/6yX/n5c+DAIDQFBcnrVxpdRQAgAD35ZfWLK7LnwgBAAAAhK2YGPf60aP+OS+JGAAAAAD849RT/XMeEjEAQGjKzJS6dDElM9PqaAAAAaxLF9fx33/755wkYgCA0JSfLy1fbkp+vtXRAAAC2IcfutfnzPH9OVmsAwAQmmJipE8+cR0DAFCKxo3d69dfLzmdvj0niRgAIDRFRZkNYQAAqIA1a6QOHVz1rVulVq18dz6mJgIAAAAIe+3bu9dPOEE6fNh35yMRAwCEprw8adEiU/LyrI4GABAEXnjBvV6rlu/ORSIGAAhNWVlSt26mZGVZHQ0AIAiMHFm8rVUr39zNxT1iAIDQZLNJp5ziOgYAoALy86WIQsNVR4745jOERAwAEJri46XffrM6CgBAkLHZpH37pHr1fHsepiYCAAAAQCGJidLrr/v2HCRiAAAAAFDEv//t29cnEQMAhKbMTKl7d1MyM62OBgAQhJxO6YEHfLPyLokYACA05edL335rSn6+1dEAAILU6NG++QxhsQ4AQGiKiZHefdd1DABAACERAwCEpqgoqX9/q6MAAKBETE0EAAAAAD9jRAwAEJry8qRffjHHZ54pRUZaGw8AAIWQiAEAQlNWlnT22eb42DEpIcHaeAAAKIREDAAQmmw2qUUL1zEAAAGERAwAEJri46Xt262OAgCAErFYBwAAAAD4GYkYAAAAAPgZiRgAIDRlZUm9e5uSlWV1NAAAuOEeMQBAaMrLk+bNcx0DABBASMQAAKEpOlp64w3XMQAAAYREDAAQmux2adgwq6MAAKBE3CMGAAAAAH7GiBgAIDTl50sbN5rjtm2lCP72CAAIHCRiAIDQlJkptWtnjo8dkxISrI0HAIBCSMQAAKErMdHqCAAAKBGJGAAgNCUkSPv2WR0FAAAlYsI8AAAAAPgZiRgAAAAA+BmJGAAgNGVlSf37m5KVZXU0AAC4IREDAISmvDzpvfdMycuzOhoAANywWAcAIDRFR0svvOA6BgAggJCIAQBCk90ujRxpdRQAAJSIqYkAAAAA4GeMiAEAQlN+vpSSYo6bN5ci+NsjACBwkIgBAEJTZqaUlGSOjx0zGzwDABAgSMQAAKErPt7qCAAAKBGJGAAgNCUkSOnpVkcBAECJmDAPAAAAAH5GIgYAAAAAfkYiBgAITdnZ0rBhpmRnWx0NAABuSMQAAKEpN1d66y1TcnOtjgYAADcs1gEACE12u/T4465jAAACCIkYACA0RUdLY8daHQUAACUKiqmJU6ZMUVJSkmJjY5WcnKwlS5aU2X/x4sVKTk5WbGysWrVqpddee81PkQIAAABA+QI+EZs1a5ZGjhypsWPHavXq1eratat69uyplJSUEvtv27ZNl19+ubp27arVq1froYce0l133aWPPvrIz5EDACzldEr79pnidFodDQAAbgI+EZs0aZKGDBmioUOHqm3btpo8ebKaNWumV199tcT+r732mpo3b67Jkyerbdu2Gjp0qG655RY999xzfo4cAGCpjAypfn1TMjKsjgYAADcBfY9YTk6OVq1apTFjxri19+jRQ8uXLy/xOT/++KN69Ojh1nbppZdq6tSpcjgcspdww3Z2drayCy1tfPjwYUnSgQMHPH0LQJkcDocyMjKUlpZW4rUJeEtYXmvp6Sp4p460NCkry9JwwkVYXmuwBNca/KUgJ3B6eXZFQCdi+/fvV15enho0aODW3qBBA6Wmppb4nNTU1BL75+bmav/+/WrUqFGx50ycOFHjx48v1n7SSSd5ED0AIGC0aGF1BACAIJeWlqaaNWt67fUCOhErYLPZ3OpOp7NYW3n9S2ov8OCDD2rUqFHH64cOHVKLFi2UkpLi1X9soKgjR46oWbNm+vvvv1WjRg2rw0EI41qDv3CtwV+41uAvhw8fVvPmzVWnTh2vvm5AJ2KJiYmKjIwsNvq1d+/eYqNeBRo2bFhi/6ioKNWtW7fE58TExCgmJqZYe82aNfmPDb+oUaMG1xr8gmsN/sK1Bn/hWoO/RER4d3mNgF6sIzo6WsnJyVqwYIFb+4IFC9S5c+cSn9OpU6di/b/55ht17NiR+cMAAAAAAkJAJ2KSNGrUKL311luaNm2aNm7cqHvuuUcpKSkaPny4JDOtcMCAAcf7Dx8+XH/99ZdGjRqljRs3atq0aZo6daruu+8+q94CAAAAALgJ6KmJktS3b1+lpaVpwoQJ2r17t9q1a6f58+erxT83Xu/evdttT7GkpCTNnz9f99xzj1555RU1btxYL730kvr06VPhc8bExOjRRx8tcboi4E1ca/AXrjX4C9ca/IVrDf7iq2vN5vT2OowAAAAAgDIF/NREAAAAAAg1JGIAAAAA4GckYgAAAADgZyRiAAAAAOBnYZuITZkyRUlJSYqNjVVycrKWLFlSZv/FixcrOTlZsbGxatWqlV577TU/RYpgV5lr7eOPP1b37t1Vr1491ahRQ506ddLXX3/tx2gRzCr7c63AsmXLFBUVpQ4dOvg2QISMyl5r2dnZGjt2rFq0aKGYmBidcMIJmjZtmp+iRTCr7LU2c+ZMtW/fXvHx8WrUqJEGDx6stLQ0P0WLYPXDDz/oqquuUuPGjWWz2TR37txyn+ON3CAsE7FZs2Zp5MiRGjt2rFavXq2uXbuqZ8+ebsvgF7Zt2zZdfvnl6tq1q1avXq2HHnpId911lz766CM/R45gU9lr7YcfflD37t01f/58rVq1St26ddNVV12l1atX+zlyBJvKXmsFDh8+rAEDBujiiy/2U6QIdlW51m644QZ99913mjp1qn7//Xe9//77atOmjR+jRjCq7LW2dOlSDRgwQEOGDNFvv/2m2bNna+XKlRo6dKifI0ewSU9PV/v27fXyyy9XqL/XcgNnGDr77LOdw4cPd2tr06aNc8yYMSX2v//++51t2rRxa7v11lud5557rs9iRGio7LVWklNOOcU5fvx4b4eGEFPVa61v377Ohx9+2Pnoo48627dv78MIESoqe619+eWXzpo1azrT0tL8ER5CSGWvtWeffdbZqlUrt7aXXnrJ2bRpU5/FiNAjyfnJJ5+U2cdbuUHYjYjl5ORo1apV6tGjh1t7jx49tHz58hKf8+OPPxbrf+mll+rnn3+Ww+HwWawIblW51orKz8/X0aNHVadOHV+EiBBR1Wvt7bff1p9//qlHH33U1yEiRFTlWvv000/VsWNHPfPMM2rSpIlOOukk3XfffcrMzPRHyAhSVbnWOnfurB07dmj+/PlyOp3as2eP5syZoyuuuMIfISOMeCs3iPJ2YIFu//79ysvLU4MGDdzaGzRooNTU1BKfk5qaWmL/3Nxc7d+/X40aNfJZvAheVbnWinr++eeVnp6uG264wRchIkRU5VrbsmWLxowZoyVLligqKuw+ClBFVbnWtm7dqqVLlyo2NlaffPKJ9u/frxEjRujAgQPcJ4ZSVeVa69y5s2bOnKm+ffsqKytLubm56tWrl/773//6I2SEEW/lBmE3IlbAZrO51Z1OZ7G28vqX1A4UVdlrrcD777+vcePGadasWapfv76vwkMIqei1lpeXp379+mn8+PE66aST/BUeQkhlfq7l5+fLZrNp5syZOvvss3X55Zdr0qRJmj59OqNiKFdlrrUNGzborrvu0n/+8x+tWrVKX331lbZt26bhw4f7I1SEGW/kBmH3Z9DExERFRkYW+2vK3r17i2W2BRo2bFhi/6ioKNWtW9dnsSK4VeVaKzBr1iwNGTJEs2fP1iWXXOLLMBECKnutHT16VD///LNWr16tO+64Q5L5ZdnpdCoqKkrffPONLrroIr/EjuBSlZ9rjRo1UpMmTVSzZs3jbW3btpXT6dSOHTvUunVrn8aM4FSVa23ixInq0qWLRo8eLUk6/fTTlZCQoK5du+rxxx9nBhO8xlu5QdiNiEVHRys5OVkLFixwa1+wYIE6d+5c4nM6depUrP8333yjjh07ym63+yxWBLeqXGuSGQkbNGiQ3nvvPea1o0Iqe63VqFFD69at05o1a46X4cOH6+STT9aaNWt0zjnn+Ct0BJmq/Fzr0qWLdu3apWPHjh1v27x5syIiItS0aVOfxovgVZVrLSMjQxER7r/aRkZGSnKNVgDe4LXcoFJLe4SIDz74wGm3251Tp051btiwwTly5EhnQkKCc/v27U6n0+kcM2aM8+abbz7ef+vWrc74+HjnPffc49ywYYNz6tSpTrvd7pwzZ45VbwFBorLX2nvvveeMiopyvvLKK87du3cfL4cOHbLqLSBIVPZaK4pVE1FRlb3Wjh496mzatKnzuuuuc/7222/OxYsXO1u3bu0cOnSoVW8BQaKy19rbb7/tjIqKck6ZMsX5559/OpcuXers2LGj8+yzz7bqLSBIHD161Ll69Wrn6tWrnZKckyZNcq5evdr5119/OZ1O3+UGYZmIOZ1O5yuvvOJs0aKFMzo62nnmmWc6Fy9efPyxgQMHOi+44AK3/osWLXKeccYZzujoaGfLli2dr776qp8jRrCqzLV2wQUXOCUVKwMHDvR/4Ag6lf25VhiJGCqjstfaxo0bnZdccokzLi7O2bRpU+eoUaOcGRkZfo4awaiy19pLL73kPOWUU5xxcXHORo0aOfv37+/csWOHn6NGsFm4cGGZv3/5KjewOZ2M1QIAAACAP4XdPWIAAAAAYDUSMQAAAADwMxIxAAAAAPAzEjEAAAAA8DMSMQAAAADwMxIxAAAAAPAzEjEAAAAA8DMSMQAAAADwMxIxAAD8bPv27bLZbGrZsmWxx1q2bCmbzabt27f7PS4AgP+QiAEAAl5BclK4xMbGKikpSTfddJNWrlxpdYgAAFQKiRgAIGi0bt1aXbp0UZcuXdS6dWulpqZq5syZ6tSpk2bMmGF1eAAAVBiJGAAgaDz00ENaunSpli5dqnXr1mnXrl267rrrlJeXp9tvv10HDx60OkQAACqERAwAELRq166tqVOnKiEhQUePHtU333xjdUgAAFQIiRgAIKjVqFFDJ510kiSVuMDF119/rV69eqlBgwaKiYlR06ZNNXjwYP3555+lvmZGRoaee+45nXvuuapVq5bi4+PVunVr3XzzzVq8eLFb361bt+rpp5/WhRdeqGbNmikmJkb16tXTZZddpi+++MKr7xUAEDpIxAAAQS8jI0OSFB8f79Y+cuRIXXbZZfrss88kSaeeeqqOHj2q6dOn68wzz9Ty5cuLvVZKSoo6duyo0aNHa8WKFapfv77atm2rAwcO6N1339Wjjz7q1v/JJ5/UmDFjtGrVKsXHx+v000+X3W7X119/rSuvvFJPP/20j941ACCYkYgBAILali1bjo9udejQ4Xj766+/rhdffFFJSUlauHCh9uzZo19++UUHDhzQ448/riNHjqhv377Kyso6/py8vDxde+212rhxozp27KgNGzZo8+bNWrVqldLS0rR69Wr17dvX7fx9+vTRTz/9pCNHjuj333/XypUrtWvXLv3www9q1KiRxo4dW+boGwAgPJGIAQCC0pEjR/Ttt9+qd+/eys3NVZcuXdS1a1dJUk5OjsaNG6fIyEh99NFHuvDCC48/LzIyUmPHjlWfPn20Y8cOzZ49+/hjH3/8sVatWqX69evrq6++Utu2bd3O2aFDB912221ubT179tQ555wjm83m1t61a1c99thjysvL06xZs7z87gEAwS7K6gAAAKiowYMHa/DgwW5tERER6tu3r1599dXjbT/++KNSU1N11lln6YwzzijxtXr16qWPPvpIixcv1s033yxJmjdvniTplltuUd26dSsc1759+/Tee+9pxYoV2rt37/FRtsOHD0uS1q5dW/E3CQAICyRiAICg0bp1a9WvX19Op1OpqanaunWr7Ha7zjrrLNWuXft4v3Xr1kkyi3ecd955Jb7WoUOHJEk7d+483rZx40ZJ0rnnnlvhmL755hvdcMMNx5Oukhw4cKDCrwcACA8kYgCAoPHQQw9p0KBBx+vLli1T7969dd9996lBgwa66aabJLlGovbt26d9+/aV+ZqZmZnHj48cOSJJqlWrVoXiOXTokG688UYdPnxYAwYM0IgRI3TyySerRo0aioiI0Lfffqvu3bvL4XBU4l0CAMIB94gBAIJWly5d9Oabb0qS7r777uOJVLVq1SRJ/fv3l9PpLLMsWrTo+OtVr15dkmu0rDxffvmlDh48qE6dOmn69Ok655xzVKtWLUVEmI/Xv//+20vvFAAQakjEAABBrXfv3jr33HN14MABTZo0SZJ0yimnSJLWr19fqdc69dRTJUk//fRThfoX7FvWqVOnYot1SNwbBgAoHYkYACDojRkzRpL00ksv6dixY+ratasSExO1du1atxGv8vTu3VuSNG3atArd1xUXFydJ2rNnT7HH0tLSNHXq1AqfGwAQXkjEAABBr1evXmrbtq0OHjyoV199VbGxsZowYYIk6frrr9cnn3wip9Pp9pz169frgQce0LJly4639e7dWx07dtTevXt1+eWX6/fff3d7ztq1a91WZyxYLv/DDz/Ut99+e7x99+7d6tOnj3Jzc73+XgEAoYFEDAAQ9Gw2m+677z5J0qRJk5SVlaXbbrtNY8aM0f79+3XttdcqMTFRZ599tpKTk1W3bl2ddtppeuaZZ3T06NHjr1Ow79jJJ5+sFStWqE2bNjr55JPVsWNHJSYmqkOHDm57giUnJ+u6666Tw+FQ9+7d1bp1a51xxhlq3ry5fvnlFz311FN+/7cAAAQHEjEAQEi46aab1LhxY6WmpmratGmSpIkTJ2rZsmXq16+fEhIStHbtWm3fvl1NmzbVLbfcoi+++EIXX3yx2+s0b95cq1at0sSJE3XmmWdq165d2rhxo+rUqaOBAwfqsccec+s/c+ZMPfLII2rZsqX++usvpaam6rrrrtPKlSvVvn17v71/AEBwsTmLztUAAAAAAPgUI2IAAAAA4GckYgAAAADgZyRiAAAAAOBnJGIAAAAA4GckYgAAAADgZyRiAAAAAOBnJGIAAAAA4GckYgAAAADgZyRiAAAAAOBnJGIAAAAA4GckYgAAAADgZyRiAAAAAOBnJGIAAAAA4Gf/DxrXVrnw2qH6AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(recalls, precisions,  \"b-\", linewidth=2)\n",
    "plt.xlabel(\"Recall\", fontsize=16)\n",
    "plt.ylabel(\"Precision\", fontsize=16)\n",
    "plt.plot([recall_90_precision, recall_90_precision], [0., 0.9], \"r:\")\n",
    "plt.plot([0.0, recall_90_precision], [0.9, 0.9], \"r:\")\n",
    "plt.plot([recall_90_precision], [0.9], \"ro\")\n",
    "plt.axis([0, 1, 0, 1])\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ROC Curve( TPR vs FPR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(np.float64(0.006271905552481092),\n",
       " np.float64(9.161032631598233e-05),\n",
       " np.float64(27687.178595082736))"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr,tpr, thresholds = roc_curve(y_train_5,y_scores)\n",
    "tpr[10], fpr[10], thresholds[10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_roc_curve(fpr,tpr, labels=None):\n",
    "    plt.plot(fpr,tpr, linewidth=2)\n",
    "    plt.plot([0,1],[0,1], \"k--\")\n",
    "    plt.xlabel(labels[0])\n",
    "    plt.ylabel(labels[1])\n",
    "    plt.axis([0,1, 0, 1])\n",
    "    plt.grid(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_roc_curve(fpr, tpr, labels=['False Positive Rate(Fallout)', 'True Positive Rate(Recall)'])\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.9604938554008616)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Precision vs recall is preferred when the positive class is rare (imbalance class), AuC is used otherwise"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ROC Curve of two classifers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "forest_clf = RandomForestClassifier(random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probas_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3, method='predict_proba')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores_forest = y_probas_forest[:, 1]\n",
    "fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.plot(fpr, tpr, \"b:\",label=\"SGD\")\n",
    "plot_roc_curve(fpr_forest, tpr_forest, labels=[\"False Positive Rate(Fallout)\", \"True Positive Rate(recall)\"])\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.9983436731328145)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc_score = roc_auc_score(y_train_5, y_scores_forest)\n",
    "auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(fpr, tpr, \":b\",label='SGD')\n",
    "plot_roc_curve(fpr_forest, tpr_forest, 'Random Forest')\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.9983436731328145)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argmax(y_scores_forest>=0.90)\n",
    "roc_auc_score(y_train_5, y_scores_forest)\n",
    "#precision = precision_score(y_train_5, np.argmax[y_pred_forest])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multiclassification\n",
    "\n",
    "## OvA( One versus All)\n",
    "10 classifier for 10 digits, select the highest score per prediction\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-2 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-2 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-2 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-2 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-2 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-2 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>OneVsRestClassifier(estimator=SGDClassifier(random_state=42))</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 sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;OneVsRestClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.multiclass.OneVsRestClassifier.html\">?<span>Documentation for OneVsRestClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>OneVsRestClassifier(estimator=SGDClassifier(random_state=42))</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">estimator: SGDClassifier</label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(random_state=42)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;SGDClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.linear_model.SGDClassifier.html\">?<span>Documentation for SGDClassifier</span></a></label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(random_state=42)</pre></div> </div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "OneVsRestClassifier(estimator=SGDClassifier(random_state=42))"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "ova = OneVsRestClassifier(SGDClassifier(random_state=42))\n",
    "ova.fit(X_train[:300], y_train[:300])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ova.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8))"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ova.estimators_), ova.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.int64(5)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# gets the index of the class with highest score\n",
    "np.argmax(ova.decision_function([some_digit]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OvO(One Versus One)\n",
    "N/(N-1)/2 estimators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-3 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-3 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-3 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-3 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-3 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-3 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-3 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-3 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-3 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>OneVsOneClassifier(estimator=SGDClassifier(random_state=42))</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 sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;OneVsOneClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.multiclass.OneVsOneClassifier.html\">?<span>Documentation for OneVsOneClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>OneVsOneClassifier(estimator=SGDClassifier(random_state=42))</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">estimator: SGDClassifier</label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(random_state=42)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;SGDClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.linear_model.SGDClassifier.html\">?<span>Documentation for SGDClassifier</span></a></label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(random_state=42)</pre></div> </div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "OneVsOneClassifier(estimator=SGDClassifier(random_state=42))"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.multiclass import OneVsOneClassifier\n",
    "ovo_clf = OneVsOneClassifier(SGDClassifier(random_state=42))\n",
    "ovo_clf.fit(X_train[:300], y_train[:300])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ovo_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45, array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8))"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ovo_clf.estimators_), ovo_clf.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf.fit(X_train[:300], y_train[:300])\n",
    "forest_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.03, 0.03, 0.  , 0.17, 0.01, 0.66, 0.01, 0.03, 0.02, 0.04]])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf.predict_proba([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluating classifier using cross_validation score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.8 , 0.77, 0.79])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(sgd_clf, X_train[:300], y_train[:300], cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.79, 0.84, 0.79])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(forest_clf, X_train[:300], y_train[:300], cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Same scores with scaled inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "X_train_scaled_300 = scaler.fit_transform(X_train[:300].astype(np.float64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.78, 0.78, 0.73])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(sgd_clf, X_train_scaled_300, y_train[:300], cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.79, 0.84, 0.79])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(forest_clf, X_train_scaled_300, y_train[:300], cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Error Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_scaled = scaler.fit_transform(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred = cross_val_predict(sgd_clf, X_train_scaled, y_train, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[5577,    0,   22,    5,    8,   43,   36,    6,  225,    1],\n",
       "       [   0, 6400,   37,   24,    4,   44,    4,    7,  212,   10],\n",
       "       [  27,   27, 5220,   92,   73,   27,   67,   36,  378,   11],\n",
       "       [  22,   17,  117, 5227,    2,  203,   27,   40,  403,   73],\n",
       "       [  12,   14,   41,    9, 5182,   12,   34,   27,  347,  164],\n",
       "       [  27,   15,   30,  168,   53, 4444,   75,   14,  535,   60],\n",
       "       [  30,   15,   42,    3,   44,   97, 5552,    3,  131,    1],\n",
       "       [  21,   10,   51,   30,   49,   12,    3, 5684,  195,  210],\n",
       "       [  17,   63,   48,   86,    3,  126,   25,   10, 5429,   44],\n",
       "       [  25,   18,   30,   64,  118,   36,    1,  179,  371, 5107]])"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "conf_mx = confusion_matrix(y_train, y_train_pred)\n",
    "conf_mx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib_scalebar.scalebar import ScaleBar\n",
    "plt.matshow(conf_mx,cmap=plt.cm.gray)\n",
    "plt.gca().add_artist(ScaleBar(1.0))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing predicted(columns) images against available images per classes(rows)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rows_sum = conf_mx.sum(axis=1,keepdims=True)\n",
    "norm_conf_mx = conf_mx / rows_sum\n",
    "np.fill_diagonal(norm_conf_mx,0)\n",
    "plt.matshow(norm_conf_mx, cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting different images classes(3,5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "cl_a, cl_b = 3, 5\n",
    "X_aa = X_train[(y_train== cl_a) & (y_train_pred == cl_a)]\n",
    "X_ab = X_train[(y_train== cl_a) & (y_train_pred == cl_b)]\n",
    "X_ba = X_train[(y_train== cl_b) & (y_train_pred == cl_a)]\n",
    "X_bb = X_train[(y_train== cl_b) & (y_train_pred == cl_b)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def plot_digits(instances, images_per_row=10, **options):\n",
    "    size = 28\n",
    "    images_per_row = min(len(instances), images_per_row)\n",
    "    images = [instance.reshape(size,size) for instance in instances]\n",
    "    n_rows = (len(instances) - 1) // images_per_row + 1\n",
    "    row_images = []\n",
    "    n_empty = n_rows * images_per_row - len(instances)\n",
    "    images.append(np.zeros((size, size * n_empty)))\n",
    "    for row in range(n_rows):\n",
    "        rimages = images[row * images_per_row : (row + 1) * images_per_row]\n",
    "        row_images.append(np.concatenate(rimages, axis=1))\n",
    "    image = np.concatenate(row_images, axis=0)\n",
    "    plt.imshow(image, cmap = plt.cm.binary, **options)\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "plt.subplot(221);plt.title(\"True Positive 3\");plot_digits(X_aa[:25], images_per_row=5)\n",
    "plt.subplot(222);plt.title(\"False Negative 3\");plot_digits(X_ab[:25], images_per_row=5)\n",
    "plt.subplot(223);plt.title(\"FPos3/False Negative 5\");plot_digits(X_ba[:25], images_per_row=5)\n",
    "plt.subplot(224);plt.title(\"True Neg 3/True Positive 5\");plot_digits(X_bb[:25], images_per_row=5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multilabel  Classification\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_large = (y_train >= 7)\n",
    "y_train_odd = (y_train % 2 == 1)\n",
    "y_multilabel = np.c_[y_train_large, y_train_odd]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-4 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-4 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-4 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-4 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-4 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-4 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-4 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-4 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-4 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-4 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-4 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-4 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-4 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-4 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-4 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-4\" 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 fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" checked><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;KNeighborsClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.neighbors.KNeighborsClassifier.html\">?<span>Documentation for KNeighborsClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>KNeighborsClassifier()</pre></div> </div></div></div></div>"
      ],
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf = KNeighborsClassifier()\n",
    "knn_clf.fit(X_train, y_multilabel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False,  True]])"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluating each prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_knn_pred = cross_val_predict(knn_clf, X_train, y_multilabel, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.9764102655606048)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_multilabel, y_train_knn_pred, average='macro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.9778357403921755)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Weighted labels(each label is assigned multiplied by the ratio of its instance in the training set)\n",
    "f1_score(y_multilabel, y_train_knn_pred, average='weighted')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multi output classification\n",
    "Predicting Pixel Values given an image + noise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "noise_train = np.random.randint(0,100, (len(X_train), 784))\n",
    "X_train_mod = X_train + noise_train\n",
    "noise_test = np.random.randint(0,100,(len(X_test),784))\n",
    "X_test_mod = X_test + noise_test\n",
    "y_train_mod = X_train\n",
    "y_test_mod = X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# left is a 4 with noise, right is a 4 without\n",
    "plot_digits([X_train_mod[2], X_train[2]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-5 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-5 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-5 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-5 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-5 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-5 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-5 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-5 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-5 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-5 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-5 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-5 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-5 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-5 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-5 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-5 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-5 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-5 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-5\" 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 fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" checked><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;KNeighborsClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.neighbors.KNeighborsClassifier.html\">?<span>Documentation for KNeighborsClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>KNeighborsClassifier()</pre></div> </div></div></div></div>"
      ],
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.fit(X_train_mod, y_train_mod)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "clean_digit = knn_clf.predict([X_test_mod[2]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_digits([X_test_mod[2],clean_digit])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <blockquote>GridSearch with KNN</blockquote>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = {\"weights\": (\"distance\", \"uniform\"),\"n_neighbors\":[2,3,5]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_grid = GridSearchCV(knn_clf, parameters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-6 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-6 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-6 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-6 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-6 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-6 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-6 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-6 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-6 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-6 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-6 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-6 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-6 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-6\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(estimator=KNeighborsClassifier(),\n",
       "             param_grid={&#x27;n_neighbors&#x27;: [2, 3, 5],\n",
       "                         &#x27;weights&#x27;: (&#x27;distance&#x27;, &#x27;uniform&#x27;)})</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 sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;GridSearchCV<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.model_selection.GridSearchCV.html\">?<span>Documentation for GridSearchCV</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>GridSearchCV(estimator=KNeighborsClassifier(),\n",
       "             param_grid={&#x27;n_neighbors&#x27;: [2, 3, 5],\n",
       "                         &#x27;weights&#x27;: (&#x27;distance&#x27;, &#x27;uniform&#x27;)})</pre></div> </div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">best_estimator_: KNeighborsClassifier</label><div class=\"sk-toggleable__content fitted\"><pre>KNeighborsClassifier(n_neighbors=3, weights=&#x27;distance&#x27;)</pre></div> </div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;KNeighborsClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.neighbors.KNeighborsClassifier.html\">?<span>Documentation for KNeighborsClassifier</span></a></label><div class=\"sk-toggleable__content fitted\"><pre>KNeighborsClassifier(n_neighbors=3, weights=&#x27;distance&#x27;)</pre></div> </div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(estimator=KNeighborsClassifier(),\n",
       "             param_grid={'n_neighbors': [2, 3, 5],\n",
       "                         'weights': ('distance', 'uniform')})"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'mean_fit_time': array([0.16558652, 0.15966797, 0.16320415, 0.15787368, 0.16236935,\n",
       "        0.15839634]),\n",
       " 'std_fit_time': array([0.0034334 , 0.00600156, 0.00475314, 0.00429515, 0.0043528 ,\n",
       "        0.00317037]),\n",
       " 'mean_score_time': array([26.13389506, 26.11585479, 26.02369814, 25.82551599, 25.22329211,\n",
       "        26.17589955]),\n",
       " 'std_score_time': array([0.76221632, 0.36503564, 0.5799688 , 0.27729886, 0.18110568,\n",
       "        0.5040632 ]),\n",
       " 'param_n_neighbors': masked_array(data=[2, 2, 3, 3, 5, 5],\n",
       "              mask=[False, False, False, False, False, False],\n",
       "        fill_value=999999),\n",
       " 'param_weights': masked_array(data=['distance', 'uniform', 'distance', 'uniform',\n",
       "                    'distance', 'uniform'],\n",
       "              mask=[False, False, False, False, False, False],\n",
       "        fill_value=np.str_('?'),\n",
       "             dtype=object),\n",
       " 'params': [{'n_neighbors': 2, 'weights': 'distance'},\n",
       "  {'n_neighbors': 2, 'weights': 'uniform'},\n",
       "  {'n_neighbors': 3, 'weights': 'distance'},\n",
       "  {'n_neighbors': 3, 'weights': 'uniform'},\n",
       "  {'n_neighbors': 5, 'weights': 'distance'},\n",
       "  {'n_neighbors': 5, 'weights': 'uniform'}],\n",
       " 'split0_test_score': array([0.96983333, 0.96416667, 0.97233333, 0.97175   , 0.97033333,\n",
       "        0.96975   ]),\n",
       " 'split1_test_score': array([0.97075   , 0.96283333, 0.97166667, 0.97066667, 0.97133333,\n",
       "        0.97016667]),\n",
       " 'split2_test_score': array([0.96866667, 0.9615    , 0.97008333, 0.96891667, 0.97033333,\n",
       "        0.96933333]),\n",
       " 'split3_test_score': array([0.96941667, 0.96316667, 0.97008333, 0.96858333, 0.96941667,\n",
       "        0.96816667]),\n",
       " 'split4_test_score': array([0.96833333, 0.96408333, 0.97141667, 0.97041667, 0.97066667,\n",
       "        0.969     ]),\n",
       " 'mean_test_score': array([0.9694    , 0.96315   , 0.97111667, 0.97006667, 0.97041667,\n",
       "        0.96928333]),\n",
       " 'std_test_score': array([0.00085862, 0.0009724 , 0.00089536, 0.00116952, 0.00061914,\n",
       "        0.00068232]),\n",
       " 'rank_test_score': array([4, 6, 1, 3, 2, 5], dtype=int32)}"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_grid.cv_results_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.ndimage import shift"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_img = shift(X_test[2].reshape(28,28), [10,4] ,cval=0)\n",
    "plot_digits([X_test[2], test_img])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "def shift_img(img):\n",
    "    d_p = random.randint(3,8)\n",
    "    r_p = random.randint(2,5)\n",
    "    img = shift(img.reshape(28,28), [d_p, r_p], cval=0)\n",
    "    img = img.flatten()\n",
    "    return img\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_shifted = []\n",
    "\n",
    "for img in list(X_train):\n",
    "    X_train_shifted.append(shift_img(img))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_shifted = np.array(X_train_shifted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_digits([X_train_shifted[2], X_train_shifted[6000]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_shifted_pred = cross_val_predict(clf_grid, X_train_shifted, y_train, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test_pred = clf_grid.predict(X_test)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.6"
  },
  "vscode": {
   "interpreter": {
    "hash": "76f3d5e912a864e6de7d4e86a584e56bbbd14237fc223dd3755ec0a4422cfb2f"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}