stats.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyzing\n",
    "\n",
    "Start with pre-reqs.\n",
    "\n",
    "* Load modules\n",
    "* Set visuals for matplotlib\n",
    "* Load data to generic var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>prediction</th>\n",
       "      <th>exact_prediction</th>\n",
       "      <th>source</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>african_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.142970e-02</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>4.324376e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>9.800000e-09</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>2.099090e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>7.597720e-05</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.082998e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>5.988363e-04</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>9.574029e-04</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>9.517600e-06</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>2.559137e-04</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.184775e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>6.881700e-06</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.891300e-05</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.873720e-05</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>2.497690e-05</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>3.580000e-08</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>3.000000e-10</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_man.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.298883e-02</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_person.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>9.568221e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_woman.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>2.144596e-03</td>\n",
       "      <td>predict/stable-diffusion-turbo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>5.719743e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.995196e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>african_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999992e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.997694e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.981859e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.993564e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asian_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999459e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.996065e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999987e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>6.872771e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>black_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999964e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999981e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latinx_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.999200e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.991690e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.932262e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.817936e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>me_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>7.817080e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>8.502401e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_person_2.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>1.272374e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_person_3.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>4.615641e-04</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>na_person_4.jpg</th>\n",
       "      <td>real</td>\n",
       "      <td>2.455431e-04</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_person_1.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>8.955221e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_person_2.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.410242e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_person_3.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>white_person_4.jpg</th>\n",
       "      <td>synthetic</td>\n",
       "      <td>9.043637e-01</td>\n",
       "      <td>predict/fast-sdxl</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     prediction  exact_prediction  \\\n",
       "african_man.jpg            real      1.142970e-02   \n",
       "african_person.jpg         real      4.324376e-03   \n",
       "african_woman.jpg          real      9.800000e-09   \n",
       "asian_man.jpg              real      2.099090e-03   \n",
       "asian_person.jpg           real      7.597720e-05   \n",
       "asian_woman.jpg            real      1.082998e-03   \n",
       "black_man.jpg              real      5.988363e-04   \n",
       "black_person.jpg           real      9.574029e-04   \n",
       "black_woman.jpg            real      9.517600e-06   \n",
       "latinx_man.jpg             real      2.559137e-04   \n",
       "latinx_person.jpg          real      1.184775e-03   \n",
       "latinx_woman.jpg           real      6.881700e-06   \n",
       "me_man.jpg                 real      1.891300e-05   \n",
       "me_person.jpg              real      1.873720e-05   \n",
       "me_woman.jpg               real      2.497690e-05   \n",
       "na_man.jpg                 real      3.580000e-08   \n",
       "na_person.jpg              real      3.000000e-10   \n",
       "na_woman.jpg               real      0.000000e+00   \n",
       "white_man.jpg              real      1.298883e-02   \n",
       "white_person.jpg           real      9.568221e-03   \n",
       "white_woman.jpg            real      2.144596e-03   \n",
       "african_person_1.jpg  synthetic      1.000000e+00   \n",
       "african_person_2.jpg  synthetic      5.719743e-01   \n",
       "african_person_3.jpg  synthetic      9.995196e-01   \n",
       "african_person_4.jpg  synthetic      9.999992e-01   \n",
       "asian_person_1.jpg    synthetic      9.997694e-01   \n",
       "asian_person_2.jpg    synthetic      9.981859e-01   \n",
       "asian_person_3.jpg    synthetic      9.993564e-01   \n",
       "asian_person_4.jpg    synthetic      9.999459e-01   \n",
       "black_person_1.jpg    synthetic      9.996065e-01   \n",
       "black_person_2.jpg    synthetic      9.999987e-01   \n",
       "black_person_3.jpg    synthetic      6.872771e-01   \n",
       "black_person_4.jpg    synthetic      1.000000e+00   \n",
       "latinx_person_1.jpg   synthetic      1.000000e+00   \n",
       "latinx_person_2.jpg   synthetic      9.999964e-01   \n",
       "latinx_person_3.jpg   synthetic      9.999981e-01   \n",
       "latinx_person_4.jpg   synthetic      9.999200e-01   \n",
       "me_person_1.jpg       synthetic      9.991690e-01   \n",
       "me_person_2.jpg       synthetic      9.932262e-01   \n",
       "me_person_3.jpg       synthetic      9.817936e-01   \n",
       "me_person_4.jpg       synthetic      7.817080e-01   \n",
       "na_person_1.jpg       synthetic      8.502401e-01   \n",
       "na_person_2.jpg            real      1.272374e-01   \n",
       "na_person_3.jpg            real      4.615641e-04   \n",
       "na_person_4.jpg            real      2.455431e-04   \n",
       "white_person_1.jpg    synthetic      8.955221e-01   \n",
       "white_person_2.jpg    synthetic      9.410242e-01   \n",
       "white_person_3.jpg    synthetic      1.000000e+00   \n",
       "white_person_4.jpg    synthetic      9.043637e-01   \n",
       "\n",
       "                                              source  \n",
       "african_man.jpg       predict/stable-diffusion-turbo  \n",
       "african_person.jpg    predict/stable-diffusion-turbo  \n",
       "african_woman.jpg     predict/stable-diffusion-turbo  \n",
       "asian_man.jpg         predict/stable-diffusion-turbo  \n",
       "asian_person.jpg      predict/stable-diffusion-turbo  \n",
       "asian_woman.jpg       predict/stable-diffusion-turbo  \n",
       "black_man.jpg         predict/stable-diffusion-turbo  \n",
       "black_person.jpg      predict/stable-diffusion-turbo  \n",
       "black_woman.jpg       predict/stable-diffusion-turbo  \n",
       "latinx_man.jpg        predict/stable-diffusion-turbo  \n",
       "latinx_person.jpg     predict/stable-diffusion-turbo  \n",
       "latinx_woman.jpg      predict/stable-diffusion-turbo  \n",
       "me_man.jpg            predict/stable-diffusion-turbo  \n",
       "me_person.jpg         predict/stable-diffusion-turbo  \n",
       "me_woman.jpg          predict/stable-diffusion-turbo  \n",
       "na_man.jpg            predict/stable-diffusion-turbo  \n",
       "na_person.jpg         predict/stable-diffusion-turbo  \n",
       "na_woman.jpg          predict/stable-diffusion-turbo  \n",
       "white_man.jpg         predict/stable-diffusion-turbo  \n",
       "white_person.jpg      predict/stable-diffusion-turbo  \n",
       "white_woman.jpg       predict/stable-diffusion-turbo  \n",
       "african_person_1.jpg               predict/fast-sdxl  \n",
       "african_person_2.jpg               predict/fast-sdxl  \n",
       "african_person_3.jpg               predict/fast-sdxl  \n",
       "african_person_4.jpg               predict/fast-sdxl  \n",
       "asian_person_1.jpg                 predict/fast-sdxl  \n",
       "asian_person_2.jpg                 predict/fast-sdxl  \n",
       "asian_person_3.jpg                 predict/fast-sdxl  \n",
       "asian_person_4.jpg                 predict/fast-sdxl  \n",
       "black_person_1.jpg                 predict/fast-sdxl  \n",
       "black_person_2.jpg                 predict/fast-sdxl  \n",
       "black_person_3.jpg                 predict/fast-sdxl  \n",
       "black_person_4.jpg                 predict/fast-sdxl  \n",
       "latinx_person_1.jpg                predict/fast-sdxl  \n",
       "latinx_person_2.jpg                predict/fast-sdxl  \n",
       "latinx_person_3.jpg                predict/fast-sdxl  \n",
       "latinx_person_4.jpg                predict/fast-sdxl  \n",
       "me_person_1.jpg                    predict/fast-sdxl  \n",
       "me_person_2.jpg                    predict/fast-sdxl  \n",
       "me_person_3.jpg                    predict/fast-sdxl  \n",
       "me_person_4.jpg                    predict/fast-sdxl  \n",
       "na_person_1.jpg                    predict/fast-sdxl  \n",
       "na_person_2.jpg                    predict/fast-sdxl  \n",
       "na_person_3.jpg                    predict/fast-sdxl  \n",
       "na_person_4.jpg                    predict/fast-sdxl  \n",
       "white_person_1.jpg                 predict/fast-sdxl  \n",
       "white_person_2.jpg                 predict/fast-sdxl  \n",
       "white_person_3.jpg                 predict/fast-sdxl  \n",
       "white_person_4.jpg                 predict/fast-sdxl  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.rcParams.update({\"text.usetex\": True, \"font.family\": \"monospace\", \"font.size\": \"14\"})\n",
    "\n",
    "data = pd.read_json(open(\"reports/predictions.json\", \"r\"), orient=\"index\")\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "selected = data.sort_values(by=[\"exact_prediction\"])\n",
    "selected.plot.bar(y=\"exact_prediction\", figsize=(16, 4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "zoom_selected = selected[\n",
    "    (selected.exact_prediction > 0.1) & (selected.exact_prediction < 0.99)\n",
    "]\n",
    "zoom_selected_ax = zoom_selected.plot.bar(\n",
    "    y=\"exact_prediction\", figsize=(8, 4), color=\"gray\", width=0.4, legend=False\n",
    ")\n",
    "zoom_selected_ax.set_title(\"Zoomed for prediction \\(> 0.1\\) \\& prediction \\(< 0.99\\)\")\n",
    "zoom_selected_ax.set_xlabel(\"image file\")\n",
    "zoom_selected_ax.set_ylabel(\"prediction value\")\n",
    "zoom_selected_ax.set_yticks([0.1, 0.5, 0.99])\n",
    "zoom_selected_ax.set_yticklabels([\"real\", \"0,5\", \"synthetic\"])\n",
    "zoom_selected_bar = plt.setp(\n",
    "    zoom_selected_ax.get_xticklabels(), rotation=45, ha=\"right\", rotation_mode=\"anchor\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoIAAAGFCAYAAABpHZIwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA+UElEQVR4nO3dX3Bb9333+Q9ISnTkiDgA1aqt61Q8oJ8kdp80Bqh95tluZ2sTcDK1ctEEkDK+24kJ2LcZBxB9k+lNZbDudQLI3ktPLTD2RZydTQAmz0yfZ3czIuA2W/lJHwuQEkdt1UjEART/0T+evdACIQiQBMgDgOB5v2Ywic754Xe+/JE0vvz99di2bQsAAACuMzbsAAAAADAcJIIAAAAuRSIIAADgUiSCAAAALkUiCAAA4FIkggAAAC5FIggAAOBSJIKQJNm2rXq9LraVBADAPUgEIUm6deuWvF6vbt26NexQAADAgJAIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEtNDDsAjLa7d+/q/v37ww4DkCSNj4/r0KFDww4DAEYGiSB2pV6v68aNG7p9+/awQwFaTE5O6tixY5qamhp2KACw77k+ESwUCgqHwx3vpVIpLS0tyTAMVavVgT13s0gkItM0lclkHI1ht+r1uq5du6ZPf/rTOnbsmA4dOiSPxzPssOBytm3r7t27qtVqunbtmiSRDALADjy2bdvDDmJYLMuSz+dTuVyWaZodyxQKBUUiEVWrVRmGMbDnbtRIRuPxuCPP76Rer8vr9apWq+344VmpVHTo0CH94R/+IQkg9h3btvWrX/1Kd+/e7er3C+gH217XR7WrunfnliYOH9UR7wl5PGM9l9mvdRG783UNi+t7BHeyHz5IksnksENounv3rm7fvq1jx46RBGJf8ng88nq9unbtmu7evcucQThupw/1+o1Lun75Hd27U29emzg8peOzpzR17Imuy+zXuojd+bqGaV+ko9lsVqFQSB6PRx6PR5FIRIFAoHl/eXm5ea+RfFiW1XKtUChoaWlJgUBAoVCo2ZPn8XgUCoVUKpXa6vP5fJKkQCDQrKdxbbPt6tsokUgoEAjI5/MpFovJsqxdPXfz1xeJRLZtw+XlZUUiEfl8Pvl8PoVCIVUqlW3fsxuNhSF8uGI/a/x8spAJvbDtdX1oVVT793/Uh1ZFtr3eVqZ+45Iu//Rv9Mufva5/+fkF/fJnr+vyT/9G9RuXmvevvfdGy4e+JN27U9e1995Q/calrsrs17qI3fm6hm3oiWA2m1UqldLi4qLK5bKKxaKkB8OPjSQqGo02r5fLZUmSYRgt18LhsOLxuNLptEqlkiKRiGKxWHP4dWFhofnMaDSqarXafH8+n1e1WlW1WtWVK1c6xhmLxbasryEUCml1dVWZTEYrKyuSpJmZmV09tzEvsVqtKplMam1tbcs2XFpaUiwWUyQS0crKiorFoiqVipaXl7d8z17RG4j9jJ9P9GqnBK9RZrsP9dqv/19dv/zOts/5t8vv7FjmevkHWl+/t+/qInbn67pe/kHHPzgGaehDw8ViUadPn1Y0Gm1ey2QyCgQCLXPygsGgwuGwCoVCc67c6uqqwuFwc/jWMAwFg0FJUjqdbpZLJBJtPWqGYTTfZ5rmjvP/dqpvaWlJlUqlZVFJLpdTIBDQ0tJSc3i3l+c27gUCARUKhY5lLMtSKpVSPp9vWXxy5cqVbeu+fft2y4rfer2+ZVkAOMgaCd5mjQRPjz+no9Of7yJB+L7W7364bZn7d3b+b+292zVV/+WnbQnnsOsidufrune7po9qV/WwMbxpaEPvEUwkEspms4pEIlpaWlKhUJBpmuq0hiUWiymXyzX/nc/ntxwy3ZhY+v3+Pce5U335fL7jYo5wOKyLFy/u+flbuXDhgkzTbFuBvFNie+7cOXm93ubr0Ucf7VuMALBf2fZ6V702H1pXdvxQ3ykJ7MWdT7YeBRpmXYN+nhtiv3fnlmPP3I2hJ4LBYFDlclnBYLCZ2Hk8HqVSqbayp0+fVqFQaA4ZLy8vtyRoGzmR/PVS3+rqqpaWllrm9Xk8HmWzWUfj2KzblcebLS4uqlarNV8ffPBBH6IDgP1hq/l/DxZ9dNFrYzk/53o7hx9y7jPMyboG/Tw3xD5x+Khjz9yNoQ8NN4Y80+l0y7VYLKaTJ0+2JHqGYSgcDjd7wRovp+OZm5vreauYubk5BYPBlq9jEM8NBAK7mgs4OTmpycnJnt8HAKNmu1Wb9vo9R581fuhh3d+mZ3D88JQ80rbJ58SkV74/+E9a+9V/3bbcoOsidufrmpj06oj3xJb3B2HoPYKZTEaJRKLlWmPhRz6fbyvfGB7O5/Nb9gb2auPK3kQioQsXLvRcRywWUzabbalLetBr2al306nnhsNhVSqVtjmE2z13FKyv27r+wZp+8fN/1fUP1rS+vr+2u9xqzmY/pFKpbVe070eJRGJPMcdisZae9aWlJYcjhFvstMDjzsc3u6rniDGjicPb77E6MenV781+Zdsyvzd7SsdnT21b5njgWY2NTexYbtB1EbvzdR0PPDv0/QSHnghKD1YIx2IxFQqF5mrXbDarWCzWVrYxPNzYLmWzRnK1ceuUxorbzdupGIYhwzCUyWRUqVSUzWZVqVRa5tt1W188HpdpmgqFQlpeXm7WF4vFWrbC6fa5jWd3em1kmqbS6XRzjmWj/RYWFtqeOyo+eP+6vv/a3+snuaL+7//jn/STXFHff+3v9cH714cdmqQH35dIJNKX7Xk6SafTyufzHb//+1Umk2nGvBu5XK65aj4cDjd3CwB60c38v+q/XuwqwXvYMLv6UJ/6nf+oRx5/rq3OiUmvHnn8OU0de0JTx57YsYykrsoNui5id76uYRv6ySKNXquN2500kputevwikYgKhULbgpLGkXAN6XRa5XK5ZZ5eOp1u2aC5VCopFoupUqm0PXc39S0tLTUTvGAwqEQi0XERyXbPlX67b2EnuVyurW2Wl5d17tw5lUqlbZ+7lW5PFvnkk0905coVzczM6KGHHuq6/m598P51/bfv/2zL+3/6lS/o0ceOO/7cXvR6MowTKpWKAoGAoyfc9FupVFIoFOq48KsXsVhMfr+/pyMW+/1zitHwoVXRL3/2+o7ljv3RvG78YmXL+xs/sDsOM096HySBGz7UR/mEi/34PDfEPixDTwR3I5VKqVKptKwgxt7sh0Rwfd3W91/7e338m9tbljlydFKnvvFnGhsb/D5xy8vLHXuppd/u+yh1PqO6sTp+47VGctdQrVZ17ty55hQD0zRVLBZlGEazbCaTUS6X0+rqqkzT1Pnz55tbJm3UmGrg9/sVjUZ3PXdVerDXZyaTaW6i3piOsLmXrlQqaWFhofnHyJkzZ5RKpZqJYCMxbCiXyyoUCs2pIVud6U0iiN2q/fs/6l9+vvOUmz/43Gl5xia6SvCk/f2hDvRqJH9yt+stw+j69bXqtkmgJH1067Z+fa09WRiEbjcE7zSU22mo1DTNlvrm5+dVKBS0srKicrncsecvkUgoFotpZWVF4XC44yk3gUBAlUpFKysrymQye/p96WbDd+nB72QoFFI4HFaxWFQikWiboxoMBlWtVhWNRpv7f54+fbo5TWKrzdyB3ep2NebE4aOaOvaEZv/Tt/SZL3xDf/C50/rMF76h2f/ppY5Ddx7PmB42THl/90/0sGGSBGKkDX3VcLcaQ6iWZalUKun06dPDDgkO++TD7ZPAXsv1Q7cbgncaMu60BdHG+jZvSN5IujbKZDLNIf9gMCjLsrSwsNAs25i2sHGh1crKimZmZlq2W9qcyG2OvfF1dbvheyKRUDKZbPY8NmLbnAwahqHz588348nn85qbm+tpGgPQrSPeE5o4PNX1qs1Ggge4yUgkgpZltQyLJZPJkZknhe499HB329l0W27ULC4u7lhm8x9AjSHnhk7zRw3D0OnTp1tW2sdisS3Py45Go81pF4lEonludSQSaZ7ws3FGiWVZqlQqbfFvXvy0MZ5cLtccZqcnEP3i8Yzp+OypjqeGNOyHVZvAMI3ET79hGIrH40qlUs2FFTh4fucRnz716e2TvCNHJ/U7j4zONiq92Cpx2s7G3sTG/3ZaLd6YZ9hQLBZl23bH18a5t91s+L66utp8RrfC4bD8fn9zaBjol1FYtQkM00j0CEoPhqN6mSyO0TM25lHwqc9uu2r4yT//7FAWiuxktxuCb9TNe9fW1lrKNZK7jcPVnbZaaUyt6FU3G7436rUsqyW2xjZLnWSzWZmmqUKh0FxcAvTL1LEndHT68yzwADrgtwD7yqOPHdeffuULbT2DR45O7outYzbqdUPwRs/ZXmw+RebcuXMtPYmRSKTtWEPLsrbcd3Mn3Wz43jjhZ/Nzt/rDrVKpKJVKNf+422olNuAkFngAW7AB27ZrtZotya7VatuW+/jjj+333nvP/vjjj/saz/376/a//fKmffW//4v9b7+8ad+/v97X5/XKMAw7Ho/b5XLZzmQytiS7XC63lJFkJ5NJu1wu27lczjYMw5Zk5/N5u1qt2rZt29Vq1S4Wi7Yku1gsNq9vVi6XbUm2JDuXy9nlctmOx+Mdn2uaph0MBu1isWjn83nbNE07HA7v6uuMRqO2JDsajdr5fL7la8nn881yuVzOlmSn02m7XC7byWSyGe/mrzcYDNrJZLLlGdFotO1rr1ardrVatcPhsB2Px5v/7sagfk4BYNSRCMK27f2XCO53xWLRNk3TlmSbpmnncrm2MrlcrlkmGAza6XTaNk3TNgzDzmQyLcndxlcwGGyrq5FA5fN5OxgM2oZh2OFwuC0JbIjH47ZhGLZpmi1JV6+SyaSdTCabCeF2X+/G2KLRaLONGl9vI1lsvGzbbmuDYrFo27bd8rzNr26+Hn5OAaA7I7mhNJy3HzaUBpzCz6m7sMEzsHsjs1gEAIDNOh75dnhKx2dPsSIY6AJ/MgEARlL9xiVde++Ntg2j792p69p7b6h+49KQIgNGB4kgAGDk2Pa6rl9+Z9sy18s/kG2vDygiYDSRCAIARs6DOYFbHx0nSfdu1/RR7epgAgJGFIkgAGDk3Ltzy9FygFuRCAIARs7E4aOOlgPcikQQADByjnhPtJ0fvNnEpFdHvCcGExAwokgEAQAjx+MZ0/HZU9uWOR54lv0EgR3wGwIAGElTx57QI48/19YzODHp1SOPP8c+gkAX2FAaADCypo49oaPTn+dkEWCXSAQBACPN4xnTw4Y57DCAkUQiCADYlzhDGOg/EkEAwL7DGcLAYPCnFQBgX+EMYWBw6BHEvsSQEHCwbfU73u0ZwkenP89/EwAH8FuEfad+45Iu//Rv9Mufva5/+fkF/fJnr+vyT/9m3/UCLC8vKxAIyOPxKBQKybKsvj+zUCj0/RnbSSQS8ng88vl8u3p/LBaTx+NpvpaWlhyOEKNgu99xzhAGBotEEPvKqAwJFQoFxWIxpdNpFYtFzc3N9f2ZlmUpEomoUqn0/VlbyWQyyufzu056c7mcqtWqqtWqwuGwyuWyswFi39vpd/zWzf/eVT2cIQw4g6Fh7BujNCRUKpUUDAYVjUYlPUiQ3MLv9+/p/YZhtPwv3KOb3/Hav/9jV3VxhjDgDHoEsW/s9yEhy7KaQ5qpVEqlUqn570gk0lY2EonI5/M1h46z2WzHerPZrEKhUEtdgUCgeX95ebllOLYxHL2XIdpunttQKpWa5UKhUNvw9MZ28Hg8qlQqymazjsSIg6Wb3/H1ux9q/NDD25bhDGHAOSSC2De6HeoZ1pCQYRjNYc1kMqlgMNj8dz6fbym7urqqQqGglZUVlctlLS4uKpVKtc2Jy2azSqVSWlxcVLlcVrFYlCRVKpXm8Gs0GlW1Wm3ey+fzzedeuXJlV19LN8+VHgyBh0IhhcNhFYtFJRIJpVKplroa7RCNRhUOh2Wapk6fPi3DMJTJZHYdIw6ebn93p373T7a9zxnCgHMYGsa+0e1QzzCHhBrDmdPT0/L7/VsOb4bDYdm23fy3aZpaW1tTKpVSMplsXi8Wizp9+nRziFl6MMwcCARa6jYMQ6ZpNuva6rmbE7mNNr6v2+cmEgklk0ml02lJD5I+y7LakkHDMHT+/HnNzMxoeXlZ+Xxec3NzisfjHWOBO3X7u3t0+vM64j3Rvo/gpFfHA8+yjyDgIBJB7BtHvCc0cXhq26GjURsSWlpa0sWLF1UqlbS2ttaWpCUSCYVCIVUqFUUiEQWDwbYkshexWEylUqnjvWg0qlwu1/VzLctSpVLR4uJiSz3hcLhj/YZhKJfLKRaLSRI9gWjTy++4xzPGGcLAAPAbhX3D4xnT8dlT25YZlSGhUqkkn8+nN998U5FIRLlcrtmrtlEwGFS5XFYwGFQ+n1ckEmnOQdyNYrEo27Y7vhpJYLfPXV1dldTboo5wOCy/398cGgY26vV3vHGGsPd3/0QPG+ZI/O4Do4YeQewrU8eekB5/buSHhBYWFhQOh1uSr7W1tbZyjYUXG5PExtY0J0+ebBm67aRQKGhubq7npKub5zaGoi3Laqm/09fRkM1mZZqmCoVCc2U1sNFB+R0HDgoSQew7U8eeGPkhocaq4c3XNstkMiqVSi376YXDYcXjceXz+Y6J4MZ6Gos3ep2L181zTdOUaZrKZrMt8xq32iqnUqkolUqpWCyqUqkoFouxTyA6Ogi/48BBQSKIfakxJLTfNJKwmzdvtsz529wjFw6HlUql5Pf7FQwGVSqVtLCwIOnBdjDBYLDZ49ZImhKJhEzTVKlUUjabbelNbDyjsRI3lUqpUCioUqlsOWdvJ908N51ON+f8RaNRZTIZLS8vS2rtjbQsS7FYTPF4vJlABoNBxWIxnT9/vqV9Gm1mWZb8fv+WbYiDbb/+jgOuYwO2bddqNVuSXavVti338ccf2++995798ccfDyiy/cUwDFtS26tYLLaVjcfjzfLBYNDOZDJ2PB63TdO0o9Gobdu2nUwm7WQyaUej0WZdpmnauVyu4/OLxaJtmuaO5XbSy3Pz+bwdDAZtwzDsaDTajMEwDDuTydi5XK6lLWzbtsvlcsf22fi8za9kMrmrr6UTt/+cAkC3PLa9y+WJOFDq9bq8Xq9qtZqmpqa2LPfJJ5/oypUrmpmZ0UMPPTTACIHu8XMKAN1hQgYAAIBLkQgCAAC4FItFAACOse11VgMDI4REEADgiPqNS+37Ax6e0vHZU+wPCOxT/JkGANiz+o1LuvbeG23Hx927U9e1995Q/calIUUGYDskggCAPbHtdV2//M62Za6XfyDbXh9QRAC6RSKIXWHXIexn/HwO1oM5gfVty9y7XdNHtauDCQhA10gE0ZPx8XFJ0t27d4ccCbC1xs9n4+cV/XXvzi1HywEYHBJB9OTQoUOanJxUrVaj1wX7km3bqtVqmpyc1KFDh4YdjitMHD7qaDkAg8OqYfTs2LFjunbtmn71q1/J6/Xq0KFD8ng8ww4LLmfbtu7evatarabf/OY3euSRR4Ydkmsc8Z7QxOGpbYeHJya9OuI9MbigAHSFRBA9axxBd+PGDV27dm3I0QCtJicn9cgjj2x7VCKc5fGM6fjsKV17740tyxwPPMt+gsA+xFnDkNT9WcOb3b17V/fv3+9jZED3xsfHGQ4eoo77CE56dTzwLPsIAvsUPYLYk0OHDvHBC0CSNHXsCR2d/jwniwAjhEQQAOAYj2dMDxvmsMMA0CX+TAMAAHApEkEAAACXIhEEAABwKRJBAAAAlyIRBAAAcCkSQQAAAJciEQQAAHApEkEAAACXIhEEAABwKRJBAAAAlyIRBAAAcCnOGgYAdMW21/VR7aru3bmlicNHdcR7Qh4P/QnAKCMRBADsqH7jkq5ffkf37tSb1yYOT+n47ClNHXtiiJEB2Av+lAMAbKt+45KuvfdGSxIoSffu1HXtvTdUv3FpSJEB2CsSQQDAlmx7Xdcvv7NtmevlH8i21wcUEQAnkQgCALb0YE5gfdsy927X9FHt6mACAuAoEkEAwJbu3bnlaDkA+wuJIABgSxOHjzpaDsD+QiIIANjSEe8JTRye2rbMxKRXR7wnBhMQAEeRCAIAtuTxjOn47KltyxwPPMt+gsCI4jcXALCtqWNP6JHHn2vrGZyY9OqRx59jH0FghLGhNABgR1PHntDR6c9zsghwwJAIAgC64vGM6WHDHHYYABzEn3IAAAAuRSIIAADgUiSCAAAALkUiCAAA4FIkggAAAC7lSCJYq9X02muv6erVq05UBwAAgAHoORGcnp5WvV5vuba6uqpXXnlFgUBAb7/9tmPBAQAAoH96TgSr1Wrbtfn5eV2+fFnf+c53lEwmHQkMAAAA/eXoHMFIJKJKpeJklQAAAOiTrk8W2Tj/r1gsamZmpuW+ZVnKZDJt1wEAALA/dZUIrqysKBKJyOPxSHowFNyJz+fTysqKc9EBAACgbzy2bdvdFKzVarJtW9PT01pdXZVptp836fV6HQ8Qg1Gv1+X1elWr1TQ1NTXscAAAwAB0PTTcSPJs21YgECBZAAAAGHE9LxbJ5XIkgQAAAAdA10PDONgYGgYAwH0cP2JucXHR6SoBAADQB472CNZqNfn9ft2/f9+pKjEg9AgCAOA+jvcIAgAAYDR0tWr41VdfVblc7ncsAAAAGKCuEsFkMqlwOCzDMLYtZ1mWAyEBAAbNttf1Ue2q7t25pYnDR3XEe0IeD4NGwEHX1RzBsbExWZa149wxy7I0PT3NHMERxBxBwL3qNy7p+uV3dO9OvXlt4vCUjs+e0tSxJ4YYGYB+c/TPPY/HI3ajAYDRUb9xSdfee6MlCZSke3fquvbeG6rfuDSkyAAMQleJYDQa7aqXyOv1KhqN7jkoAED/2fa6rl9+Z9sy18s/kG2vDygiAIPGhtKQxNAw4EYfWhX98mev71juM1/4hh422s+XBzD6HB0artVqeu2115ysEgDQJ/fu3HK0HIDR42giuLq6qlQq5WSVAIA+mTh81NFyAEZPV9vHbDQ9Pb3tNjGmyfABAIyCI94Tmjg81bZQZKOJSa+OeE8MLigAA9VzIlitVpXP5+X3+1uuX7x4UWfPnlUikXAsOABA/3g8Yzo+e0rX3ntjyzLHA8+ynyBwgPWcCMbjcc3Pz7ddf/LJJ+X3+5VOp/XSSy85EhwAoL+mjj0hPf5c+z6Ck14dDzzLPoLAAefoquErV65odnaWDaVHEKuGAXfjZBHAnXruEZQeJA2dZLNZ5ggCwAjyeMbYIgZwIccWizQ6FpeXl/ccFAAAAPpv14tFOvX8zczMOBIUAAAA+q/nRDCZTHZcLAIAAIDRwhFzkMRiEQAA3GhXS8I2LxZ566239OKLL+onP/mJI0EBAACg/3pOBGdnZ+Xz+TQ+Pi5JOnv2rBYWFuT1ehWNRvX22287HiQAAACc1/PQ8DPPPCNJymQympmZ0fj4uPL5vJ5++mktLy9rcXFR77//fl+CRf8wNAwAgPv0vFhkZWVFxWJRMzMzWllZkSQ9/fTTkqRQKKRKpeJshAAAAOiLnoeGbduWYRiSpFKppHA43LxHEggAADA6ek4Eg8Gg3n33XUlSPp9XMBhs3stkMopGo85FBwAAgL7pORGMx+N6/vnn9eKLL6pQKCiRSGhlZUWPPfaYvve973HEHAAAwIjoeY5gPB6X9GBYuFgs6sSJEyqXy3ryySf15JNPOh4gAAAA+oMNpSGJVcMAALjRrjaUBgAAwOjbdSL46quv6syZM/r6178u6cHG0q+//rpjgQEAAKC/ek4Ea7Wa/H6/MpmMfD6fcrmcJGlubk4LCwskgwAAACOi5zmCZ86ckW3bunDhgiRpbGxM6+vrkqRsNqvz58/r4sWLzkeKvmKOIAAA7tNzIuj3+/XjH/9YX/ziFyVJ4+Pjun//viTpypUrmp2dbf4bo4NEEAAA9+l5aNjv96tarXa8VyqVNDMzs+egAAAA0H89J4Lz8/NKJpNt12u1ms6ePasXXnjBkcAAAADQXz0ngplMRuvr65qentbi4qIkaXFxUTMzMzIMQy+99JLjQQIAAMB5u9o+plgsKpVKqVgsyuv1qlgsKp1Os0gEAABghHCyCCSxWAQAADfquUewXq+rXq/3IxYAAAAMUM+JYDAY1Pz8fD9iAQAAwABN9PqGSqWiSqXSj1gAAAAwQD33CBqGoVqttuX9kydP7ikgAAAADEbPieDCwoKef/75jvdqtZpKpdKegwIAOMO21/WhVVHt3/9RH1oV2fb6sEMCsI/0PDT88ssva2VlRceOHdPi4qLm5+dlGIYkbXniCABg8Oo3Lun65Xd0785vF/hNHJ7S8dlTmjr2xBAjA7Bf9Lx9zPj4uDq9xePxyLZteTwezhoeQWwfAxws9RuXdO29N7a8/8jjz5EMAuh9aNi2bZVKJVWr1ZbX2tqaVldX+xEjAKAHtr2u65ff2bbM9fIPGCYG0PvQsCSZptmx1ygQCHTsLQQADM5Htastw8Gd3Ltd00e1q3rYMAcUFYD9qOcewXK5vOXQodfrVblc3nNQAIDdu3fnlqPlABxcPSeCMzMze7oPAOivicNHHS0H4ODqOREEAOxvR7wnNHF4+0VfE5NeHfGeGExAAPYtEkEAOGA8njEdnz21bZnjgWfl8fARALgd/xUAgANo6tgTeuTx59p6BicmvWwdA6BpV6uGAQD739SxJ3R0+vP//yriW5o4fFRHvCfoCQTQRCIIAAeYxzPGFjEAtsSfhQAAAC5FIggAAOBSJIIAAAAu5Xgi+OqrrzpdJQAAAPrAYzt4OPCVK1c0Ozur+/fvO1UlBqRer8vr9apWq215hCAAADhYeu4RfPfdd3Xy5EmNj4+3vQKBQD9iBAAAQB/03CN48uRJzczMKJFIyO/3t9y7efOmnnnmGa2vrzsaJPqPHkEAANyn530Ei8WicrmcTpw40fF+PB7fa0wAAAAYgJ6HhsPhsFZWVra8Pzs7u6eAAAAAMBg9Dw2vrKzomWee0dLSkr72ta+1DA+Xy2XNzc2xWGQEMTQMAID79JwIzs3NqVQqPXizx9Nyz7ZteTweEsERRCIIAID79Dw0bFmWcrmcyuWyLl++3PJaXV2Vg7vRHHipVEqBQECBQEDZbHbX9ZRKJXk8no4vy7KcCxgAABwoPS8WSSQSikQiHXuNZmZmlEwmHQnsoEskEioUCspkMrIsSwsLC7Isa1ftt7a2JkmqVqtt9wzD2GuoAADggHJ0Q2l0p1QqKRQKqVqtNhO1SqWiQCDQcq1bhUJBsVisYyLYLYaGAQBwn10fMffWW2/pxRdf1Je+9CW9+OKLevvtt52M60A7d+6c4vF4S8JnmqbC4bDOnTs3vMAAAICr7CoRfOaZZ/T888/Ltm09+eSTunnzpr72ta/py1/+stPxHUilUkmRSKTteiQSaS7E6ZVlWYrFYvL5fPL5fIpEIqpUKnsNFQAAHGA9zxE8e/as1tbWdOXKFXm93uZ1y7I0Nzenl19+WX/913/taJAHTaVSkWmabddN09xz8pbL5eT3+5XJZLYdar59+7Zu377d/HetVpP0YIgYAACMlqNHj7bt5tIVu0ezs7P2yspKx3v5fN6enZ3ttUrXkWTn8/m265lMxjZNc1d1dqovHA7b8Xi8Y/lvf/vbtiRevHjx4sWL1wF41Wq1XeUPPfcI3rx5U9PT0x3vTU9PN1ewYmtb9fyVy+WOPYXdCIfDbddisZjS6XTH8ouLi/rmN7/Z/HetVtNnPvMZffDBB44vFqnX63r00Uf7Uje2RrsPB+0+HLT74NHmw7FVux89enRX9fWcCMZiMSWTSf3whz9suxePx3X69OldBeImwWBQ+Xy+7VzmQqHQMaHbSSgU0uLioqLRaMt1y7K2XIE8OTmpycnJtutTU1N9+4XuZ93YGu0+HLT7cNDug0ebD4dT7d5zItiYe/bYY48pGo0qEAioXC4rm83K7/frO9/5zp6DOugWFxcVCoVa5gqWSiWVSqVtz3HeSqVS6djDmMlk2pJDAACAhl2tGi6Xy/rWt76lYrGoV155RcViUWfPntX777/vdHwHUjAYVDweVyQSUaFQ0PLysubn55VOp7fswWusBu50Usji4qJSqZSWlpZUqVRUKBQUCoUkacuhYQAAgJ4Xi8A5yWTSNk3TNk3TzmQy25YNBoO2JDuXy3W8n8vlmmVM09xykchWarXaniabbueTTz6xv/3tb9uffPKJ43Vja7T7cNDuw0G7Dx5tPhxOtzsni4wIy7I0MzOjlZUVBYNBx+vnZBEAANxn1yeLbOWtt95yukrXKxQKmpmZ0enTp/uSBAIAAHdytEewVqvJ7/fr/v37TlUJPVgMYllWX5NAegQBAHCfnlcN74SRZuftdm/BYatUKkqn0woEApIkwzDatszB3mWzWZXL5Y4Lg/ge9EehUFA+n5dlWapUKorFYm3tSts7r1AoNI/hLJfLCgQCSiaTLWVo9/6LRCLK5/Mt12h35zW2hmtsK5fNZiWp5WfekXbvZiLhCy+8YJ8/f962bdt+5pln7Lm5uY6v2dlZe2xszJHJixgspxeLlMtl2zAMu1qtNq8lk0k7nU47Ur/blctlOx6P2/F43DYMw04mkx3L8D1wXj6fb2nDarXatkCLtndesVhsWyxHuw9eOp22N6cOtHt/aNPJIZsXgTrV7l0NDT/zzDN65pln9NJLL2lsbEzJZLKZfW5UrVa1uLjI0PAIcnpoOJFIyDCMll4qy7Lk8/noNXZYKBRSOBxu6xHke9AfsVhMuVyu5Vo2m1UikWieDkTbOy+RSKhQKKhcLrdcy2azzTal3furUqkolUppeXm5pT1p9/5IJBLNreDC4XDb6KBT7d7VYpEf/ehHeumllyRJHo9HL7/8shYWFtpeiUSi6wfjYLtw4ULbHwuNPRILhcIQInIfvgf9sby8rFQq1XJtbm5O0m/blbZ3XiQS2fHkJdq9v5aXl3XmzJm267R7fwQCAcXjccXj8Y5TxJxq955XDe+UZZL9w7IsWZbV8QfXMIzmHB/0D9+D/mmcqLQV2r4/otGoMplMy7ULFy40e0No9/5aXl7ueFIV7d5flmW1zI3deN2pdu95sUi1Wt1y6NDr9bZNIIX7dDrursHv9+vmzZsDjMad+B70z+ZhYUlaXV2V9GD4hrbvr+XlZV28eFGlUkm5XK7ZS0i7949lWVpbW5Npmm0JBu3eP/l8XqZpNv+7EolElE6nFQwGHW33nnsEvV7vtvfn5+d7rRIu0+mYPAwW3wNnpdNppdPprlb40/Z7E41GlU6nlUgklEqltv1A3Ih2371sNrvrFcC0++5lMhlFo1EZhqFgMKhEIqFYLNbVe3tpd8c3lAa2Oi9ZktbW1gYXiIvxPRicWCymcDjc3NKBth+MaDSqcDisUCgky7Jo9z4pFArbzs2k3ftn8x+WjZ7AQqHgaLs7kgjWajW99tprunr1qhPVYcT5/X5Jnf8i2ek/2HAG34PByGaz8vv9LXPXaPvBiUQisiyr+X2QaHenlUqlbQ8zoN37I5VKtQ3DN9q6Uqk42u49J4LT09Oq1+st11ZXV/XKK68oEAjo7bff7rVKHDCGYcgwjC3/KolEIgOOyH34HvTf8vKyLMtqSQIb/wGm7Z3n8/m0tLTUcq3xYVgul2n3PmhsVp9KpZqvxs97KpXS0tIS7d4nS0tLzbnHDY02Nk3T0Xbf1WKRzebn53X58mVls1klk0n95V/+Za/V4oA5ffp0y35f0m8nFe+0BQScwfegf0qlktbW1lp2+G+s7otGo7S9wxq9HpuHyhpt2thrjXZ3Vqd5gdlsVoVCoWXvOtrdeel0uq39G0PCjTZ1qt0dnSMYiUS6nriLg62x6ehGmUymbfsH7F1jG4HN+B70R6VS0blz5+T3+7W8vNx8pVKpZqJC2zurcWzW5iHKTCajYDDY/MCk3fuP/9YMRjAYbGlTy7KUTqd1/vz55jWn2r2rk0UkNef/maaplZUVzczMtNxvDJHk83ldvny5pyAwfE6fLCI96DV58803dfLkyeYfCJvPBcXuWJalc+fONedHGYah06dPt529yvfAeT6fb8sVeRv/c0rbOy+bzapYLMowDFUqFZmm2XaiDu3eH5VKRZlMRsvLy6pUKopGo4pEIs0knHZ3XuNMc+lB+ycSibaePifavatEcGVlRZFIRB6PZ9sNo30+n1ZWVvTFL36xpyAwfP1IBAEAwP7WdY9grVaTbduanp7W6upqx/2ydtpjEPsXiSAAAO7T9WKRRpJn27YCgQDJAgAAwIjrebFILpcjCQQAADgAuh4axsHG0DAAAO7Tc4/g1atX9eqrr7ZdP3v2rF5//XVHggIAAED/9byhdDQa1ZUrVxSPx1t6jubm5prLyL/xjW84FyEAAAD6oucewUqlomKx2DZ8GI1Glc1m9d3vftex4AAAANA/uzpZZKvNVEOhUNshyQAAANifek4E5+bmtLKy0vHe8vKynnzyyT0HBQDDFolElEgkhh0G9OCEBQD90fMcwXg8rjNnzigYDOqpp55qXl9ZWel47h0AjKJIJCLDMIYdhutZlqVIJKJyudzxIAMAe7OrxSLnzp3T/Py8AoGATNNUpVJRpVJROp3WV7/61X7ECQADxTmpANxgV3MEk8mkqtWqksmknnzySSWTSV2+fFnf+ta3nI4PAAbGsix5PJ7mKxKJtJUJhUIKBAJKJBLy+XyKxWIqFAoKBAIKBALNg98b9UUiEfl8Pnk8HoVCIWWz2Y7PLpVKCoVCzeemUqlmHJvfk0gkFAgEms/fat52t5aXl5tx+nw+hUKhlq9j43N9Pp8CgYBSqVTLvUa8Pp+vpfzma0tLSwoEAgqFQioUCs1z7DfPMV9eXm55byAQaLbHxvoA7JEN2LZdq9VsSXatVht2KMBQVatVu1qt2slk0g4Gg233y+WyLck2TdPO5/O2YRi2JDuXy9nhcNhOJpPNsvl83pZkF4tFu1wu27lczjYMw06n0x3rTKfTdrlctqPRaPN9mUzGzufzzbLBYNAOBoN2Pp+3i8WiHY1GbcMwdv31ptPp5rMbcXaK0TRNOxwO28Vi0c7n83YwGLTD4XBLmcbXW61W2641VKtVO5fL2ZJsSXYmk2l+zZvbu1qt2sVi0ZZk5/P55vdmY/0A9oZEELZtkwgCm2UymY6JoG3bzcTPtm07HA7b0Wi0+Z7NyVGnejcnbp2Szo3PaEin0x2TPtM02xK3blSr1WaStfn65phN02wrYxhGS4yNhHbj+xuJ3EYbE9+GzQnj5hjL5XKvXx6ALuxqaHg7b731ltNVAsC+EwwGJUmGYejkyZOSJL/f37Hs0tKSYrFYc0h181BupVLR3NxcW/2bh2fz+Xxz4/6NwuGwLl682PPXcOHCBZmmqXA43HJ98yKZXC6naDTaVub06dPK5/M9P7dhY51btR2A/nI0EazVaorFYk5WCQD70sbEZavVxaVSST6fT2+++aYikYhyuZzS6XRbOdM0tbq62vbezatkV1dXtbS01DKPsdMcwm51uxK3UqkoEAi0XTcMo+Ncwm6R/AHD53iPoG3bTlcJACNpYWFB4XBYxWJR8XhcwWCwY+J15swZlUolLS0tqVKpKBaLyTTNtl64ubk5JZNJ2Q+m9bS8crlcz/FtXtyyFdM0VS6X265XKpWhbOlSKBT2vEAGwANdJYIvvviiXnvtNUnSl770JZ08ebLja25uTh6Pp68BA8CoaKwa3nxts0ZClclkFAgEZFlWxyHXWCymbDbbVsfy8nLbKt5uhMNhVSqVtg2bN9cXiUTaeh0ty2quNt7O5p7O3dr4NScSCV24cMGRegG36yoRLJfLzV/CfD6v+fl5xePxttfCwkI/YwWAvrMsq+Nrs7W1tZbyG681hMPh5kb7lUpFy8vLzf9ONq413tfodbNtW/l8vmNPWzwel2maCoVCzfdns9nm/MNemaapdDqtSCTS7I1sxLixvmQyKb/f39zipVAoKBQKKRwOt/RaNmI+d+5cs65GQrmxF6/xvxt7Ixttt7mH0jAMGYahTCbT/HorlUrbvEYAu9Tz6pKxsS1XllqWZY+Nje1+6QqGhlXDwG9XrnZ6NVbHmqbZsvXJxvuN/x+Px5t1xuPx5hYzwWDQzmQydjwet03TbK42bqys3fgyTdOOx+Mdt0pJp9PNOBp17kUul7ODweCO9TW+FtM0W7bJ2VzXxtgasRqGYWcyGTuZTLZ8nel02o7H423XNioWi806TdNsW00NYPc8tt3bpL6xsTFZlqWpqam2e7VaTT6fT+vr6z0loxi+er0ur9erWq3W8XsLoD8sy9LMzIwWFxebK4IrlYrW1taUyWRUKpU6zs8DACf0fMRctVrdMlHwer172koAANymUCjINM2WI+0aW9OEw2H5fD4VCgWGQgH0Rc+rhr1e77b35+fndx0MALiNYRg7roBlmxUA/eL4PoKN1cUAgJ2Fw2EZhtE8e7dSqTQXWgQCAc3NzTV7CAHAaT3PEdzOysqKTp8+rZs3bzpVJQaEOYLAcC0tLenNN99UqVSSYRgyTVOJRKLjSSIA4JSuEsHp6emuN+80TVPvv//+XuPCgJEIAgDgPl0NDVerVf3oRz/S6uqqXnnlFdm2rVdeeUWrq6taXV3VD3/4Q331q1/d9e72AAAAGLyuegRfeOEFffe735UknTx5UolEQs8//3xbuVQqpV/84hf6u7/7O+cjRV/RIwgAgPvsah/BUqmkL37xi233rly5orm5OeYIjiASQQAA3KfnVcPBYHDLMx6/973vaWZmZs9BAQAAoP963lA6m81qbm5OknT27FlNTU2pXq8rk8no7NmzbCgNAAAwInbVI/ijH/1IFy5ckM/n0/j4uHw+n7LZrC5cuKCnn366H3ECAADAYXvaR9CyLBWLRZmmyZDwiGOOIAAA7uPohtIYXSSCAAC4z66PmHv11Vd15swZff3rX5f0YL7g66+/7lhgAAAA6K+eE8FarSa/369MJiOfz9fcQHpubk4LCwskgwAAACOi56HhM2fOyLbt5hYyY2NjWl9fl/RgRfH58+d18eJF5yNFXzE0DACA+/ScCPr9fv34xz9ubig9Pj6u+/fvS3qwofTs7Gzz3xgdJIIAALhPz0PDfr9f1Wq1471SqcTqYQAAgBHRcyI4Pz+vZDLZdr1Wq+ns2bN64YUXHAkMAAAA/dVzIpjJZLS+vq7p6WktLi5KkhYXFzUzMyPDMPTSSy85HiQAAACct6vtY4rFolKplIrForxer4rFotLpNItEAAAARggbSkMSi0UAAHCjnnsE6/W66vV6P2IBAADAAPWcCAaDQc3Pz/cjFgAAAAzQRK9vqFQqqlQq/YgFAAAAA9Rzj6BhGKrValveP3ny5J4CAgAAwGD0nAguLCzo+eef73ivVqupVCrtOSgAAAD0X8+J4MsvvyzbtnXs2DH97d/+rf7hH/5BV69e1dWrVxkyBgAAGCE9bx8zPj6uTm/xeDyybVsej4ezhkcQ28cAAOA+PS8WsW17yzOFy+UycwQBAABGRM+JoCSZptmx1ygQCHTsLQQAAMD+0/McwXK5vOXQodfrVblc3nNQAAAA6L+eE8HNQ8JXr17d9j4AAAD2p54TQUn68Y9/rLm5OY2Pj8s0TY2Pj+vLX/6yfvGLXzgdHwAAAPqk50Twe9/7nsLhsMLhsFZXV1WtVnXx4kUdPXpUpmmSDAIAAIyInrePmZ6e1iuvvKKFhYW2e4lEQlevXtUPf/hDxwLEYLB9DAAA7tNzIjg2NibLsjomC++++67C4bBu3rzpWIAYDBJBAADcp+ehYdM0VSwWO95bXV3tuFjk1Vdf7T0yAAAA9FXPiWAqlVI0GtXbb7+ter2uer2uq1ev6tVXX9ULL7ygl19+uXm9Xq/r3XffVSqV6kfsAAAA2IOeh4b9fr8sy3rwZo+neX1jNZuvc+zc/sfQMAAA7tPzySKWZalQKHS9X2C1WtXc3FzPgQEAAKC/ek4Ek8mknn766a7Lz8zMKJlM9voYAAAA9FnPQ8M4mBgaBgDAfXZ1sggAAABGH4kgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuNTEsAMAAOyOba/ro9pV3btzSxOHj+qI94Q8Hv6+B9A9EkEAGEH1G5d0/fI7unen3rw2cXhKx2dPaerYE0OMDMAo4U9HABgx9RuXdO29N1qSQEm6d6eua++9ofqNS0OKDMCoIREEgBFi2+u6fvmdbctcL/9Atr0+oIgAjDISQQAYIQ/mBNa3LXPvdk0f1a4OJiAAI41EEABGyL07txwtB8DdSAQBYIRMHD7qaDkA7kYiCAAj5Ij3hCYOT21bZmLSqyPeE4MJCMBIIxEEgBHi8Yzp+OypbcscDzzLfoIAusJ/KQBgxEwde0KPPP5cW8/gxKRXjzz+HPsIAugaG0oDwAiaOvaEjk5/npNFAOwJiSAAjCiPZ0wPG+awwwAwwvjTEQAAwKVIBAEAAFyKRBAAAMClSAQBAABcikQQAADApUgEAQAAXIpEEAAAwKVIBAEAAFyKRBAAAMClSAQBAABcikQQAADApUgEAQAAXIpEEAAAwKVIBAEAAFyKRBAAAMClSAQBAABcikQQAADApUgEAQAAXIpEEAAAwKVIBAEAAFyKRBAAAMClSAQBAABcikQQAADApUgEAQAAXIpEEAAAwKVIBAEAAFyKRBAAAMClJoYdAAC4yfq6rV9fq+qTD2/roYcn9TuP+DQ25um5DAA4gUQQAByyUwL3wfvXVfrJP+vj39xuXvvUpycVfOqzevSx412XAQCneGzbtocdBIavXq/L6/WqVqtpampq2OEAI2enBO6D96/rv33/Z1u+/0+/8gVJ2rEMySAAJ5EIQhKJIA6ebodXnRiq3SnJ+59P/Ue9+1/+R0uSuNmnPn1YkmfbMkeOTurUN/6MYWIAjmFo+ACIxWJaXl5u/judTiuZTA4xot9yaj7UID/UD0Lstr2uj2pXde/OLU0cPqoj3hPyeMZ6rqubevZjO3Q7vOrEUO36uq3ST/657evYqLjyc93++O62ZT7+zZ1t70vSR7du69fXqjr+qH/HsgDQDXoEDwjLsiQ9SApN01Qmk+np/f3oEXRqPtQgP9QPQuz1G5d0/fI7unen3rw2cXhKx2dPaerYE13X1U09+7EduhmCdXKo9vBDh/STXHHLMk77z3/xx/qjz/3+wJ4H4GAjETxgYrGY/H7/0BNBpz5kuynj9PyrUY69fuOSrr33xpblHnn8OdWqx3asy+u7sWM9U8ee2Hft3s0Q7JGjk/qL/+1/0Q/+9//qyFDtF/70Mf0//+c/bVnGaU/FQvQIAnAM+wg6pFKpyOPxNF+WZSmVSsnn88nj8SgQCDR77RoSiYQCgYB8Pp9isVjbfcuyFIlEmnWEQiFls9nBfVG71M1QWeknP3ekzLv/5Z917976vqtrGLHfv39f1y+/s225f7v8gy7q+vmO9Vwv/0D379/fd+1QXPn5tomb9GB49fI/frBjuY9/c6eruj75eOchXUma/NShbe9/6tOH9alPT25b5sjRB8PgAOAU5gg6xDRNVatVVSoVhUIhzc/PS5JWVlZkGIZisVhL+VAoJEnKZDLy+/06d+6cZmZmVK1Wm2VWV1dVKBRULBZlGIZKpZIWFhZkWdae5wDevn1bt2//9kOuXq9vU7o3v75W7epDdifdzpnq9kN9kHUNI/Z/Lf9TyzBuJ7Ub2/dwSZJH13es597tmn71P3ZOugbdDjvNw2v4sPZRV+W68dCRQ/rUpyd37Dn84v/6Wf1f72zdmxl86nOStu/xfPLPP8tCEQCOIhF0kGEYMk1T0oMewo1JXbH42zlES0tLbfdzuZwCgYCWlpaaSV44HNbGkXvTNLW2tqZUKrXnRPDcuXP6q7/6qz3VsZVPPtz+A91pTn6oO1nXoJ93+yNrxzJ37x7escyhQ931cH1U/01X5box6HZ/2HvEsbo+9emHFHzqszsmcI8+dlyer3yhbX7jkaOTzfvSgyHwncoAgFNIBPtkcXFxy3v5fF7xeLztejgc1sWLF9uuLy0t6eLFiyqVSlpbW2sbQt5tfN/85jeb/67X63r00Uf3XK8kPfTw9sNbTnPyQ93Jugb9vMkjhnboyOsqyesmWZSkI1OfluRMT7Kj7fCpQ9v2DB45OqnZP3lU/1z8hSNzBBsrlrtJ4B597LgeCfzutiueuykDAE5hjmCfhMPhLe+trq5qaWmpZU6hx+Npm/9XKpXk8/n05ptvKhKJKJfLKZ1OOxLf5OSkpqamWl5O+Z1HfDvOdepmPlS3c6Zm/+TRfVfXMGL//cAfa+Lw9t9H7zF7x7psHd+xnolJr/7wP3xuX7ZDaP7z25Z58s8/q4mJMQWf+uy25YJPfW7HMhuHah997Li+8vyf6alYSP/5L/5YT8VCOvWNP2vrxRsb8+j4o3790ed+X8cf9XdM8LopAwBOIBHsE8Mwtrw3NzenZDIp27bbXrlcrlluYWFB4XBYxWJR8XhcwWCwOfS8n42NeRz5kO32g9jJD3Wn6hpG7OPj4zo+e2rbcr83+2wXdX1ux3qOB57V+Pj4vmyHz/yH4/rTr3yhLWk8cnSy5WSORx/buVw3ZTYigQMwatg+xmGWZcnn86lcLm+ZtGWzWaVSKV25cqUlYVxeXtbFixebvX6BQECpVKplGHl5eVmxWExbfdv2y/YxUuf93jYPlTlVZr/WNYzYO+7/N+nV8cCz2+4juLmuburZz+0wjI28AWDUkAg6yLKs5qrhYrEo0zS37BkMhUKyLEvpdFrBYFCFQkGJREKZTKaZ+CUSCV24cEHnz59XMBhsWTWcy+Vaegg3byjdSCa365ncqF9HzI3y6RyjHLvbTxYBAHSHRNAhlUpFgUCg7XowGGxZMbzR0tKSMpmMKpWKgsGgEolE2yKSRjJoWVazTLFYVKFQUDAYVC6XaztibqNkMtnVvELOGgYAwH1IBCGJRBAAADdisQgAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBRHzEGSZNu2bt26paNHj8rj8Qw7HAAAMAAkggAAAC7F0DAAAIBLkQgCAAC4FIkgAACAS5EIAgAAuBSJIAAAgEuRCAIAALgUiSAAAIBL/X+hyjcK1bpBGAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Subdivide our measures based on image source\n",
    "\n",
    "selected.source.mask(\n",
    "    selected.source == \"predict/fast-sdxl\", \"#d6c17a\", inplace=True\n",
    ")\n",
    "selected.source.mask(\n",
    "    selected.source == \"predict/stable-diffusion-turbo\", \"#a78dba\", inplace=True\n",
    ")\n",
    "\n",
    "selected_fig = plt.figure()\n",
    "selected_ax = selected_fig.add_subplot(1, 1, 1)\n",
    "\n",
    "# Format\n",
    "selected_ax.spines[\"bottom\"].set_position(\"center\")\n",
    "selected_ax.spines[\"right\"].set_color(\"none\")\n",
    "selected_ax.spines[\"top\"].set_color(\"none\")\n",
    "selected_ax.set_ylabel(\"prediction result\")\n",
    "selected_ax.set_yticks([0, 0.5, 1])\n",
    "selected_ax.set_yticklabels([\"real\", \"0,5\", \"synthetic\"])\n",
    "selected_ax.set_xlabel(\"image count\")\n",
    "\n",
    "# Build\n",
    "count = 0\n",
    "for _, i in selected.iterrows():\n",
    "    plt.plot(\n",
    "        count,\n",
    "        i.exact_prediction,\n",
    "        label='fast-sdxl' if i.source == \"#d6c17a\" else 'turbo-sdxl',\n",
    "        color=i.source,\n",
    "        linestyle=\"\",\n",
    "        marker=\"o\",\n",
    "    )\n",
    "    count += 1\n",
    "\n",
    "# Legend\n",
    "handles, labels = selected_ax.get_legend_handles_labels()\n",
    "unique = [(handle, label) for i, (handle, label) in enumerate(zip(handles, labels)) if label not in labels[:i]]\n",
    "selected_ax.legend(*zip(*unique))\n",
    "\n",
    "# Display\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}