reproducible-ml/clean_modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# *Clean modeling*\n",
    "\n",
    "**Author:** [Kata Ferenc](https://github.com/ferenckata) ([k.t.ferenc@ncmm.uio.no](mailto:k.t.ferenc@ncmm.uio.no))\n",
    "\n",
    "**Achievement:** Using three different machine learning models predicting diabetes from health data\n",
    "\n",
    "## Introduction\n",
    "\n",
    "This notebook is a clean version of the previously seen machine learning notebook. The logic, such as model training and plotting have been moved to functions and built into a package (src/workshop). This lets us focus on the model performance itself.\n",
    "\n",
    "We trained machine learning models to predict diabetes from health indicators. The dataset used was downloaded from Kaggle [Diabetes Health Indicators](https://www.kaggle.com/datasets/alexteboul/diabetes-health-indicators-dataset). This dataset contains 20 columns, each a health indicator from the participants of the survey. The model used here is *support vector machine* from the popular scikit learn package. Two different subset of the data is used for prediction, and different hyperparameters are tested."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reproducibility and code formatting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To watermark the environment\n",
    "%load_ext watermark\n",
    "\n",
    "# For automatic code formatting in jupyter lab.\n",
    "%load_ext lab_black\n",
    "\n",
    "# For automatic code formatting in jupyter notebook\n",
    "%load_ext nb_black\n",
    "\n",
    "# For better logging\n",
    "%load_ext rich"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports\n",
    "# -------\n",
    "\n",
    "# System\n",
    "import sys\n",
    "\n",
    "# Logging\n",
    "import logging\n",
    "\n",
    "# Rich logging in jupyter\n",
    "from rich.logging import RichHandler\n",
    "\n",
    "FORMAT = \"%(message)s\"\n",
    "logging.basicConfig(\n",
    "    level=\"INFO\", format=FORMAT, datefmt=\"[%X]\", handlers=[RichHandler()]\n",
    ")\n",
    "log = logging.getLogger(\"rich\")\n",
    "\n",
    "# Other packages\n",
    "# Data processing\n",
    "import pandas as pd\n",
    "\n",
    "# Models\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "# Custom functions from module\n",
    "import workshop.model_utils as utils\n",
    "\n",
    "# Random seed to fix the random generators\n",
    "RANDOM_SEED = 42"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Selecting a few features that we think might be useful"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "datapath = \"../data/train/diabetes_binary_train.csv.zip\"\n",
    "data_in = pd.read_csv(datapath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "features_of_interest = [\"Diabetes_binary\", \"Age\", \"BMI\", \"Sex\", \"HighBP\", \"HighChol\"]\n",
    "X_train, X_valid, y_train, y_valid = utils.prepare_data(\n",
    "    data_in, features_of_interest, \"Diabetes_binary\", 0.2, seed=RANDOM_SEED\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:07:35] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.69</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">105</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:07:35]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.69\u001b[0m                                  \u001b]8;id=46363;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=399051;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\u001b\\\u001b[2m105\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "svm_model = SVC(kernel=\"linear\", random_state=RANDOM_SEED)\n",
    "fitted_svm_model = utils.train_eval_model(svm_model, X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:07:40] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.69</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">124</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:07:40]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.69\u001b[0m                                  \u001b]8;id=311147;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=98528;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\u001b\\\u001b[2m124\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.test_eval_model(fitted_svm_model, X_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Maybe we can make the model better by providing all features to learn from."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "features_of_interest_all = data_in.columns\n",
    "X_train, X_valid, y_train, y_valid = utils.prepare_data(\n",
    "    data_in, features_of_interest, \"Diabetes_binary\", 0.2, seed=RANDOM_SEED\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:09:38] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.70</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">105</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:09:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.70\u001b[0m                                  \u001b]8;id=711466;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=900190;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\u001b\\\u001b[2m105\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhkAAAGdCAYAAAC/02HYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABMtElEQVR4nO3de3zO9f/H8ce1I5aNbbYZ5tTIKTSnoRBG2Ub1Da2GEiK0HEISHVgOUVmhUiRFJ6cOc6jQYg5jyvl83GZkxphttuv3h1+X73VtDtPn+m7yvHf73G5dn8/r877e13VjXnu9Dx+T2Ww2IyIiImIwh6LugIiIiPw7KckQERERu1CSISIiInahJENERETsQkmGiIiI2IWSDBEREbELJRkiIiJiF0oyRERExC6UZIiIiIhdOBV1B/5WsuHAou6CSLGTtimmqLsgUiyVsPO/Xkb+m5S59c79e1xskgwREZFiw6RCvxH0LYqIiIhdqJIhIiJiy2Qq6h78KyjJEBERsaXhEkMoyRAREbGlSoYhlKqJiIiIXaiSISIiYkvDJYZQkiEiImJLwyWGUKomIiIidqFKhoiIiC0NlxhCSYaIiIgtDZcYQqmaiIiI2IUqGSIiIrY0XGIIJRkiIiK2NFxiCKVqIiIiYheqZIiIiNjScIkhlGSIiIjY0nCJIZRkiIiI2FIlwxD6FkVERMQuVMkQERGxpUqGIZRkiIiI2HLQnAwjKFUTERERu1AlQ0RExJaGSwyhJENERMSWlrAaQqmaiIiI2IUqGSIiIrY0XGIIJRkiIiK2NFxiCKVqIiIiYheqZIiIiNjScIkhlGSIiIjY0nCJIZRkiIiI2FIlwxD6FkVERIqJtWvXEhYWhr+/PyaTicWLF+eL2bVrF+Hh4Xh4eFC6dGmaNWvG0aNHLdezsrIYNGgQ3t7euLm5ER4ezvHjx63aSEtLIzIyEg8PDzw8PIiMjOTs2bNWMUePHiUsLAw3Nze8vb0ZPHgw2dnZhfo8SjJERERsmUzGHYVw4cIF6tevT0xMTIHXDxw4QMuWLbnnnntYvXo127ZtY8yYMZQoUcISExUVxaJFi1iwYAFxcXFkZGQQGhpKbm6uJSYiIoLExERiY2OJjY0lMTGRyMhIy/Xc3Fw6derEhQsXiIuLY8GCBXz77bcMHTq0cF+j2Ww2F+oOOynZcGBRd0Gk2EnbVPAPGpE7XQk7D/aXfPhdw9rK/PGFW7rPZDKxaNEiunTpYjnXvXt3nJ2dmTdvXoH3pKenU65cOebNm0e3bt0ASEpKolKlSvz444906NCBXbt2Ubt2beLj42natCkA8fHxBAcHs3v3bmrWrMlPP/1EaGgox44dw9/fH4AFCxbQq1cvUlNTcXd3v6nPoEqGiIiIHWVlZXHu3DmrIysrq9Dt5OXl8cMPP1CjRg06dOiAj48PTZs2tRpSSUhIICcnh5CQEMs5f39/6taty7p16wBYv349Hh4elgQDoFmzZnh4eFjF1K1b15JgAHTo0IGsrCwSEhJuus9KMkRERGwZOFwSHR1tmfvw9xEdHV3oLqWmppKRkcFbb71Fx44dWbFiBY888giPPvooa9asASAlJQUXFxfKli1rda+vry8pKSmWGB8fn3zt+/j4WMX4+vpaXS9btiwuLi6WmJuh1SUiIiK2DFxdMmrUKIYMGWJ1ztXVtdDt5OXlAdC5c2defPFFABo0aMC6deuYOXMmrVq1uua9ZrMZ03/NDzEVMFfkVmJuRJUMERERO3J1dcXd3d3quJUkw9vbGycnJ2rXrm11vlatWpbVJX5+fmRnZ5OWlmYVk5qaaqlM+Pn5cfLkyXztnzp1yirGtmKRlpZGTk5OvgrH9SjJEBERsWVyMO4wiIuLC40bN2bPnj1W5/fu3UvlypUBCAoKwtnZmZUrV1quJycns337dpo3bw5AcHAw6enpbNy40RKzYcMG0tPTrWK2b99OcnKyJWbFihW4uroSFBR0033WcImIiIitItrxMyMjg/3791teHzp0iMTERDw9PQkICGD48OF069aNBx54gDZt2hAbG8uyZctYvXo1AB4eHvTu3ZuhQ4fi5eWFp6cnw4YNo169erRr1w64Uvno2LEjffr0YdasWQD07duX0NBQatasCUBISAi1a9cmMjKSyZMnc+bMGYYNG0afPn1uemUJqJIhIiJSbGzevJmGDRvSsGFDAIYMGULDhg159dVXAXjkkUeYOXMmkyZNol69enz88cd8++23tGzZ0tLGtGnT6NKlC127dqVFixaUKlWKZcuW4ejoaImZP38+9erVIyQkhJCQEO69916rZbGOjo788MMPlChRghYtWtC1a1e6dOnClClTCvV5tE+GSDGmfTJECmb3fTI6zzKsrcwl/Qxr63aj4RIRERFbekCaIZRkiIiI2NID0gyhb1FERETsQpUMERERWxouMYSSDBERERuF2dVSrk3DJSIiImIXqmSIiIjYUCXDGEoyREREbCnHMISGS0RERMQuVMkQERGxoeESYyjJEBERsaEkwxgaLhERERG7UCVDRETEhioZxlCSISIiYkNJhjGUZIiIiNhSjmEIzckQERERu1AlQ0RExIaGS4yhJENERMSGkgxjaLhERERE7EKVDBERERuqZBhDSYaIiIgNJRnG0HCJiIiI2IUqGSIiIrZUyDCEkgwREREbGi4xhoZLRERExC5UyRAREbGhSoYxlGSIiIjYUJJhDCUZIiIitpRjGEJzMkRERMQuVMkQERGxoeESYyjJEBERsaEkwxgaLhERERG7UCVDRETEhioZxlCSISIiYkNJhjE0XCIid4wli76jZbNGRd0NkTuGKhnFVObWmOten7c0nr5jP/+f9OXD154iMrwZY95bwpRPV1rOh7W+l6+m9aVkw4H/k36IAIx5eSRLlyzKd37ZjysIqFy5CHp01ZJF3/HqK6Msr729y3FfUBAvDBlGxYqVirBnUmgqZBhCSUYxVaXd1R9U/wkJYkz/TtR/5HXLucysHKt4JycHLl/Os1t/Mi9lM6RXez7+Jo6z5zPt9j4iN6NFy/t5/c1oq3NlPT2LqDfW7rrrLpZ8H4sZM4cOHuSN18bywsABfPXtYhwdHYu6e3KTNFxiDA2XFFMn/zpvOdIzMjFjtrx2dXHm5G9TeKx9Q5Z/9AJp8dN44uEmjO73MPELRlq1MzCiNbt/eM3qXGR4M7Z++wpp8dNI/O4V+j5+/w3788uGPZw8fY7hz4RcN65Z/aqsnB3FmfVT2ffTG7z90n8oVcLFct3P253v3nuOM+unsuv7cXTr2IjdP7zGwIjWN//lyB3PxcUF73LlrA5HR0c+m/Mpj3UJo2mjBoS0bcX418dx8cKFa7azZ/dueveKJLhxQ5o3uY/ujz/Kju1/Wq4nbt3C0z2epMl99xLSthVvTXiTixcvXrdvJpMJ73LlKFfOhyZNm/HcgOfZv28vx44eAeCrBV/QqWM7gurXJbxTB5YtXWx1/4z3p9OhbWsaNahLu9YteWvCm7f8PYkUNSUZt7E3X+jMB1+upsGjb7Jq/a6buufpR5rz2sAwxr2/jAaPvsnYmGW8OiCUJ8OaXve+vLw8xsYspX/3VlTwKVNgTJ27/Vn6/vMs+TmRxt2iiRz5CcENqjNtZFdLzMdv9KB8OQ869HmXJ4Z9zDOPtaBc2dI3/ZlFrsfBwcSIUaP5dvEy3hj/Fhs3xjPt7cnXjB81Yhi+fn58sfAbvvz6O555tg9OTs4A7Nu7h/59e9O2XXu+XrSUSVOmsXVLAtHj3yhUn1xdSwCQc/kyP69aycToCfTo+TTfLlnGfx7vzthXXmbjhngAVi6P5fPP5jBm3Gss+3EF0977gMDAGrf4bcg/YTKZDDvuZEoybmMx81ez5JdtHEn6i+RT6Td1z6g+HRk59TvLfUt+2cb0+b/w7GMtbnjv0l//4I+9J3il/8MFXn+xZ1sW/rSZmC9Wc+DoKeK3HWLYpK95MrQJri5O1KjiS9tm9/D8G1+yafsREncfp//r8ylV0qXA9kSuZe2a1TRr1NByDHtxMABP9ehFk6bNqFixEk2bBfP8oBdYsfyna7aTkpxEs2bNqVqtOpUrVyGkw0PUvOceAOZ8OpuHOoXxVI9eVK5chQYN72PEqNF8v3QxWVlZN9XPkykpzP10Nr5+flSpXIXPPp1N5y6P0O2JJ6lSpSo9ej1N23bt+WzOJwAkJyfj5e1N02bNKe/vT7177+Wxx7ve4F3EHpRkGENzMm5jW3YeLVS8d9m7qFTekxmvPsn7YyIs550cHUjPuLl5FqPfXUzsrMG8O++XfNca1gqgeiVvuj/c2HLOZAJHRweqVPAisLIPOTm5bN11zHL94LHTnEm/djlbpCCNmzRl9JhxltclS5UEYOOGeGZ/NIsDB/ZzISOD3NxcsrKyuHjxIqVKlcrXTmTPp3lt7Ct8v2wJTZs1J6RDRyoFBACwc8cOjh09wo/fL7PEmzGTl5fHiePHqVa9eoF9O3/+PM0aNcSMmUuZmdSqXYep70zH2cWFgwcP8tjj3aziGzS8j/mffwZASIeOzJ83l04d29Gixf20fKAVrVq3wclJP6r/5+7s3MAw+pN7G7uQaf3bVJ45D9uk2dnp6kQzh/+/+PwbX7Bx+2GruNxc80295+9bDrBy/S5eHxjGvKUbrK45mEzM/vZ33v9ydb77jiWnUaOyb4Ft3umZvhReyZIl860kSUo6wcD+fXm8a3eeH/QC7h4ebN2SwLgxo7l8+XKB7fR/fhAPdQrltzVriItby4z332PilGm0bdceszmP/3TtTsSTkfnuK1++/DX75ubmxoKvF+Hg4ICnl1e+5Mb2z7vZbLac8ytfniU/xBK/7nfi49cz4Y3XmPvpbGbPmYezs/NNfTcixYmSjH+R02kZ+Hq5W527t2ZFy/+nnjnPiZNpVKnozYKfNt/y+4x5bykbFoxk35FUq/OJu49Rq1p5Dh47XeB9ew6fxNnZkQb3VLRUM6pV8qase/7fMEUKa+f27eTm5jL0pZE4OFwZCV4Re+2hkr9VqVKVKlWqEtmzFyOGDWHJom9p2649tWrV5sD+fYVeFuvg4HDNe6pVq8bWLQmEde5iObctcStVq12tipQoUYLWD7al9YNt6f5EBJ1DH2L/vr3Uql2nUP2Qf0a//BhDczL+RdZu3ke5sncxtFc7qlb0pl/XBwhpUdsq5s1ZPzL86RCef6I1dwf4UOdufyLDmzH4qQdv+n127E9iwU+b6N+9ldX5t+espOm9VZk2siv31qhA9YBydGpVj6kjHgdg7+GT/By/m5hXnqBRncrUr1mR9195gouZ2ZhvrpAick0VKwVw+fJlvpw/j+PHjrFs6WK+/mrBNeMvXbrEhDdfZ9PGDSQlnWDrlgR2bP/T8g/+07378Me2RCa88Rq7d+3iyJHDrP7l50JP/PxvPZ95liWLF/HVwi85cuQwn835lJ9XraRnr2eAK/tsfPft1+zbt5fjx47x/dIllChRgvL+/rf8nnJrNCfDGKpk/IvsOXSSF6K/4qVnQhjZ5yEW/5zIO5/9TO//mtQ5Z9F6MjNziOrZlvFRnbmQmc2O/UnEzP+1UO/1+gff81j7+6zObd+XRMiz7zBuYBirPnkRk8nEweOn+Gb5FkvMs2M+Y8bYJ1k5O4qTf53j1elLqVW9PJeyc2zfQqRQ7qlVi2EvjeLT2R/x3jtTuS+oEYOjhvDKqBEFxjs6OJB+9iyvjBrBX3+dpkzZsrRtF8KAgVcmkdaoeQ+z58xj+nvv8HSPCMxmqFSpEh0eKnji8814sG07Rox6mbmfzmbihPFUqFiB196cQOMmV1Z3lXZ355OPP+TtSW+Rm5tHYI0avPf+TMqUKXvL7ylSlExmc/H4HVK7Rt6ZKviUYf/yN3mo33us3ri3qLtT7KRtuv7OryJ3qhJ2/hW5ygvfG9bW4XdDDWvrdqNKhvxPtWpcg7tKubJ9XxLly7kz/oUuHD5xmrgt+4u6ayIiFnf6MIdRCp1kHD9+nBkzZrBu3TpSUlIwmUz4+vrSvHlznnvuOSpV0v78cm3OTo68NjCMqhW8OX/xEhu2HeLp0XPsuiW6iIgUjUIlGXFxcTz00ENUqlSJkJAQQkJCMJvNpKamsnjxYqZPn85PP/1EixbX39gpKysr32Y25rxcTA7a1//fbtX6XTR6/OZ2JxURKTIqZBiiUEnGiy++yLPPPsu0adOueT0qKopNmzZdt53o6Ghee836eRqOvo1xLt+kMN0RERGxCw2XGKNQEz9LlixJYmIiNWvWLPD67t27adiwIZmZ1989sqBKhs/9I1TJELGhiZ8iBbP3xM9qQ340rK2DU299RdLtrlD7ZJQvX55169Zd8/r69euvuxPe31xdXXF3d7c6lGDcWJ/HW7Jx4ShO/jaZk79NZvXcoVb7YGRujSnweLFH22u26eTkwKi+HdmxdCxp8dPYsHAk7ZvXsooZ9kwIcZ8PJzVuCkd+juarqX0IrOxjFRMV2ZbDqyZweNUEBj3Zxupa47qV+X3+Szg46DcDsY+EzZsYNOA52rVuSf06Nfnl51WWazk5OUx7e7Ll6aztWrdk9KiXSE09ecN2V61YziNhD9OoQV0eCXuYn1etvOn3/dvcT2fT5oHmtHmgOfPmzrG69scf2+j++KPk5ube2gcXu9E+GcYoVJIxbNgwnnvuOQYOHMiSJUuIj49nw4YNLFmyhIEDB9K/f39eeukle/X1jnfi5FnGTF9Ciycn0+LJyazeuJevp/WlVjU/AKq0G2V19B37OXl5eSz6OfGabY4bEMazj7VkyKSvafjYm3z8TRwL3+5D/f/aKfT+++5m5sK1tOoxhdD+MTg6OvL9jIGWR7jXudufMf070XPUHHq9PIfXBoZRu/qVZNPJyYH3Rndn8PgF5OUVi9XS8i+UmXmRmjVrMnL0q/muXbp0id27dtL3uf4s/Po7pr4bw5HDh3lhYP/rtrktcSsvDXuR0PDOfP3dEkLDO/PS0Cj++GPbTb0vXHmS6wcx7/HWpLd5a9LbTH93Kvv2XVmqnZOTw5uvjeWVV8fh6Khfsoobk8m4ozDWrl1LWFgY/v7+mEwmFi9efM3Yfv36YTKZeOedd6zOZ2VlMWjQILy9vXFzcyM8PJzjx49bxaSlpREZGYmHhwceHh5ERkZy9uxZq5ijR48SFhaGm5sb3t7eDB48mOzs7EJ9nkIVnAYMGICXlxfTpk1j1qxZluzb0dGRoKAgPvvsM7p21RMD7eXHtdutXo97fxl9Hm9Jk3ursutgCif/Om91Pax1PdZs2sfhE39ds82I0CZM/Hg5y+N2AvDR13G0D67FC5EP8swrVx7a1HngB1b39Bv3Ocd+eYuGtSvx+5YD3FPVl+37TrBm05Ufntv3JXFPVT92HkjmxR7t+H3LfhIK+TA3kcJoeX8rWt7fqsBrpUuXZtbHn1qdG/nyKzzZ/XGSk5KuuZvm5/Pm0iy4Ob379AOgd7XqbN60kfmfzeXeKVNv+L4ABw8eILBGTZo2CwYgsEZNDh08QGBgDeZ+OpugRo2oW+/eQn9esb+iqkBcuHCB+vXr8/TTT/PYY49dM27x4sVs2LAB/wL+/EZFRbFs2TIWLFiAl5cXQ4cOJTQ0lISEBEtCGxERwfHjx4mNjQWgb9++REZGsmzZlQcC5ubm0qlTJ8qVK0dcXBx//fUXPXv2xGw2M3369Jv+PIUe1erWrRvdunUjJyeH06evPKPC29tbD+/5H3NwMPFY+/twK+nChj8O5bvu41maji3r0ufVeddtx8XZKd9um5lZOTRvWPATJgHc7yoBQFr6RQC270/i7so+VPIri8kEd1f2YceBJKpV8iYyvBnNIyYW9uOJ2FVGRgYmk4nS7u7XjPkjMZGnevSyOte8xf3Mnzf3pt8nMLAmRw4fJjkpCTNmjhw5zN131+DokSMsWbyIBV9/e6sfQf6lHnroIR566KHrxpw4cYKBAweyfPlyOnXqZHUtPT2d2bNnM2/ePNq1awfA559/TqVKlVi1ahUdOnRg165dxMbGEh8fT9OmV3ab/eijjwgODmbPnj3UrFmTFStWsHPnTo4dO2ZJZN5++2169erF+PHjcb/O353/dstTZ5ydnW9q/oUYq87d/qyeO5QSLk5kZGbRbehH7D6Yki/uqbCmnL94icW/JF63vVXrdzH4qQeJ27Kfg8dO06ZJTUJb3Yuj47Wz+IlDH+P3LfvZeSAZuLKd+diYZXw/48qura9OX8qeQyf5YeZARr+zmPbNazG638PkXM5l2ORv+H3LgVv/AkT+oaysLN6dNoWHOoVy1113XTPu9OnTeHl5WZ3z8vLi9OlTN/1e1apXZ1DUi/Tr8zQAg6OGUK16dfr27sWLQ4ezLi6OGR/E4OTkxIhRowlq1PjWPpQYzshCRkGLHVxdXXF1dS10W3l5eURGRjJ8+HDq1Mn/0LyEhARycnIICQmxnPP396du3bqsW7eODh06sH79ejw8PCwJBkCzZs3w8PBg3bp11KxZk/Xr11O3bl2rSkmHDh3IysoiISGBNm2s595di3b8vM3sPXySpt2jKVO6FF3aNuCj1yMJefbdfIlGj87NWPjTZrKyC37E9d+GTf6GD8Y8wbbvxmA2mzl4/DSfLY2nR3izAuOnjexKvUB/2j5tvYz542/i+PibOMvrp8KaknEhiw1/HGLb4jG0fGoyFXzKMO+tZ7in01iyc67fLxF7yMnJYcSwF8nLMzN6zLgbxud7LDvmQpfRu3Z7gq7dnrC8XrLoO0q5uVG/fgM6h3Zk/sJvOJmSwohhL/Ljil9wcXEpVPtiH0YOlxS0bcPYsWMZN25coduaOHEiTk5ODB48uMDrKSkpuLi4ULas9fNufH19SUlJscT4+Pjku9fHx8cqxtfX1+p62bJlcXFxscTcDCUZt5mcy7mWR6lv2XmUoDoBPP9EawaNv/q0yRYNq1Ozqh+RIz+9VjMWp9My6DrkI1xdnPDycCPpVDpvDu7M4aT88zimjnic0Fb1aNf7HU6knr1mm15l3Hi570O07/0OjetVYf+RVA4cPcWBo6dwcnIgsLIPO/YnFf7Di/wDOTk5DB8axYnjx/no07nXrWLAlWHgv4eE/3bmrzN4eXnfch/S0s4wa+b7fDp3Pn/+sY2AylWo/P/H5cuXOXL4EIE1Ct4iQG5fo0aNYsiQIVbnbqWKkZCQwLvvvsuWLVsKnQSZzdYJckH330rMjehR77c5EyZcXaxzxZ5dgknYeZQ/95646Xaysi+TdCodJycHurRtwPer/7C6Pm3E43R+sD4d+73HkQISkP82edhjTJ//KydSz+LoYMLJ6erMeSdHRxy1lFX+x/5OMI4eOcKs2XNu6qmm9zZoQPz6363OrV8XR/0GDW+5H5PemsBTPXrh6+dHbl4ely9frehdzs0lN1fb6xcXRq4uKWjbhltJMn777TdSU1MJCAjAyckJJycnjhw5wtChQ6lSpQoAfn5+ZGdnk5aWZnVvamqqpTLh5+fHyZP5l3CfOnXKKsa2YpGWlkZOTk6+Csf1KMm4jbw2MIwWDasTUN6TOnf7M+75MB5oFMiCHzdbYkq7leDR9g2Zs6jg/Uw+fiOS1weFW143rluZzg/Wp0oFL1o0rM7SmOdxcDAxdc7V9f7vjOpK906N6fnyHDIuXMLXqzS+XqUp4Zp/su+DTe/h7gAfZi5cC8Dm7UeoWcWXkBa1eebRFuTm5rH3SKpRX4kIABcvXGD3rl3s3nVly/oTx4+ze9cukpOSuHz5MsNeHMzOHduJnjiFvNxcTp86xelTp8j5r+V4o0e9xLvT3ra8fvKpHqxf9zuffPwhhw4e4JOPP2RD/Hqe7NHzpt7X1vp1v3P0yBG6P/EkAHXr3cvhQweJ+20N33y1EEcHB6pUrWqX70cKz8HBZNhhlMjISP744w8SExMth7+/P8OHD2f58uUABAUF4ezszMqVV/d0SU5OZvv27TRv3hyA4OBg0tPT2bhxoyVmw4YNpKenW8Vs376d5ORkS8yKFStwdXUlKCjopvus4ZLbiI9XaWa/2QM/b3fSMy6xfd8Jwp//gF827LbEPN4hCBMmvordXGAblfw8rfarcHV1ZuzzoVSt4E3GxSyW/76D3mM+Iz3j6q6t/bo+AMDKj6Os2urz6jw+X7bB8rqEqzPTRj5O5IhP+Hsj2aRT6QyZ9DWzxj1Fds5l+rw6j0tZ1qtZRP6pHTu28+zTPSyvp0yKBiC88yM89/xAVv/6CwBdH+tsdd/Hn35G4yZXJr+lJCfjYLr6e1eDhvcxcfJUYqa/w/vT36NSQCUmTpnGvffWv6n3fWPCW5bzly5dInr860ya8g4ODlfew9fXl5Evj+HV0S/j4uLCGxMmUqJECUO+D7l9ZWRksH//1adSHzp0iMTERDw9PQkICMg3GdnZ2Rk/Pz/LTtweHh707t2boUOH4uXlhaenJ8OGDaNevXqW1Sa1atWiY8eO9OnTh1mzZgFXlrCGhoZa2gkJCaF27dpERkYyefJkzpw5w7Bhw+jTp89NryyBQm4rbk8lGw4s6i6IFDvaVlykYPbeVrzO6BWGtbVjfMiNg/7f6tWrC1y50bNnT+bMmZPvfJUqVYiKiiIqKspy7tKlSwwfPpwvvviCzMxM2rZtywcffGD1lPQzZ84wePBgli5dCkB4eDgxMTGUKVPGEnP06FEGDBjAL7/8QsmSJYmIiGDKlCmFGupRkiFSjCnJECmYvZOMuq+svHHQTdr+ZnvD2rrdaE6GiIiI2IXmZIiIiNi4w59rZhglGSIiIjbu9KenGkVJhoiIiA0lGcbQnAwRERGxC1UyREREbKiQYQwlGSIiIjY0XGIMDZeIiIiIXaiSISIiYkOFDGMoyRAREbGh4RJjaLhERERE7EKVDBERERsqZBhDSYaIiIgNDZcYQ8MlIiIiYheqZIiIiNhQIcMYSjJERERsaLjEGEoyREREbCjHMIbmZIiIiIhdqJIhIiJiQ8MlxlCSISIiYkM5hjE0XCIiIiJ2oUqGiIiIDQ2XGENJhoiIiA3lGMbQcImIiIjYhSoZIiIiNjRcYgwlGSIiIjaUZBhDwyUiIiJiF6pkiIiI2FAhwxhKMkRERGxouMQYSjJERERsKMcwhuZkiIiIiF2okiEiImJDwyXGUJIhIiJiQzmGMTRcIiIiInahSoaIiIgNB5UyDKEkQ0RExIZyDGNouERERETsQpUMERERG1pdYgwlGSIiIjYclGMYQkmGiIiIDVUyjKE5GSIiImIXqmSIiIjYUCHDGEoyREREbJhQlmEEDZeIiIiIXaiSISIiYkOrS4yhJENERMSGVpcYQ8MlIiIiYheqZIiIiNhQIcMYSjJERERs6CmsxtBwiYiIiNiFKhkiIiI2VMgwhpIMERERG1pdYgwNl4iIiNgwmYw7CmPt2rWEhYXh7++PyWRi8eLFlms5OTmMGDGCevXq4ebmhr+/Pz169CApKcmqjaysLAYNGoS3tzdubm6Eh4dz/Phxq5i0tDQiIyPx8PDAw8ODyMhIzp49axVz9OhRwsLCcHNzw9vbm8GDB5OdnV2oz6MkQ0REpJi4cOEC9evXJyYmJt+1ixcvsmXLFsaMGcOWLVv47rvv2Lt3L+Hh4VZxUVFRLFq0iAULFhAXF0dGRgahoaHk5uZaYiIiIkhMTCQ2NpbY2FgSExOJjIy0XM/NzaVTp05cuHCBuLg4FixYwLfffsvQoUML9XlMZrPZXMjvwC5KNhxY1F0QKXbSNuX/QSMiUMLOg/3d5m41rK2FPRve0n0mk4lFixbRpUuXa8Zs2rSJJk2acOTIEQICAkhPT6dcuXLMmzePbt26AZCUlESlSpX48ccf6dChA7t27aJ27drEx8fTtGlTAOLj4wkODmb37t3UrFmTn376idDQUI4dO4a/vz8ACxYsoFevXqSmpuLu7n5Tn0GVDBERERsmA4+srCzOnTtndWRlZRnSz/T0dEwmE2XKlAEgISGBnJwcQkJCLDH+/v7UrVuXdevWAbB+/Xo8PDwsCQZAs2bN8PDwsIqpW7euJcEA6NChA1lZWSQkJNx0/5RkiIiI2FF0dLRl7sPfR3R09D9u99KlS4wcOZKIiAhLZSElJQUXFxfKli1rFevr60tKSoolxsfHJ197Pj4+VjG+vr5W18uWLYuLi4sl5mZodYmIiIgNI1eXjBo1iiFDhlidc3V1/Udt5uTk0L17d/Ly8vjggw9uGG82m60+U0Gf71ZibkSVDBERERsOJuMOV1dX3N3drY5/kmTk5OTQtWtXDh06xMqVK63mR/j5+ZGdnU1aWprVPampqZbKhJ+fHydPnszX7qlTp6xibCsWaWlp5OTk5KtwXI+SDBERkdvE3wnGvn37WLVqFV5eXlbXg4KCcHZ2ZuXKlZZzycnJbN++nebNmwMQHBxMeno6GzdutMRs2LCB9PR0q5jt27eTnJxsiVmxYgWurq4EBQXddH81XCIiImKjqDbjysjIYP/+/ZbXhw4dIjExEU9PT/z9/fnPf/7Dli1b+P7778nNzbVUGzw9PXFxccHDw4PevXszdOhQvLy88PT0ZNiwYdSrV4927doBUKtWLTp27EifPn2YNWsWAH379iU0NJSaNWsCEBISQu3atYmMjGTy5MmcOXOGYcOG0adPn5teWQJKMkRERPIpqg0/N2/eTJs2bSyv/57L0bNnT8aNG8fSpUsBaNCggdV9v/76K61btwZg2rRpODk50bVrVzIzM2nbti1z5szB0dHREj9//nwGDx5sWYUSHh5utTeHo6MjP/zwAwMGDKBFixaULFmSiIgIpkyZUqjPo30yRIox7ZMhUjB775MROX+bYW3Ne7K+YW3dblTJEBERsaFnlxhDSYaIiIgNB+UYhlCSISIiYkOVDGNoCauIiIjYhSoZIiIiNlTHMIaSDBERERsOGi4xhIZLRERExC5UyRAREbGhQoYxlGSIiIjY0OoSY2i4REREROxClQwREREbKmQYQ0mGiIiIDa0uMYaGS0RERMQuVMkQERGxoUKGMZRkiIiI2NDqEmMUmyTj90UTiroLIsVOtYHfFXUXRIqlpJmP2rV9zSUwhr5HERERsYtiU8kQEREpLjRcYgwlGSIiIjYclGMYQsMlIiIiYheqZIiIiNhQJcMYSjJERERsaE6GMTRcIiIiInahSoaIiIgNDZcYQ0mGiIiIDY2WGEPDJSIiImIXqmSIiIjY0KPejaEkQ0RExIbK/MZQkiEiImJDhQxjKFkTERERu1AlQ0RExIbmZBhDSYaIiIgN5RjG0HCJiIiI2IUqGSIiIja046cxlGSIiIjY0JwMY2i4REREROxClQwREREbKmQYQ0mGiIiIDc3JMIaGS0RERMQuVMkQERGxYUKlDCMoyRAREbGh4RJjKMkQERGxoSTDGJqTISIiInahSoaIiIgNk9awGkJJhoiIiA0NlxhDwyUiIiJiF6pkiIiI2NBoiTGUZIiIiNjQA9KMoeESERERsQtVMkRERGxo4qcxlGSIiIjY0GiJMTRcIiIiInahJENERMSGAybDjsJYu3YtYWFh+Pv7YzKZWLx4sdV1s9nMuHHj8Pf3p2TJkrRu3ZodO3ZYxWRlZTFo0CC8vb1xc3MjPDyc48ePW8WkpaURGRmJh4cHHh4eREZGcvbsWauYo0ePEhYWhpubG97e3gwePJjs7OxCfR4lGSIiIjZMJuOOwrhw4QL169cnJiamwOuTJk1i6tSpxMTEsGnTJvz8/Gjfvj3nz5+3xERFRbFo0SIWLFhAXFwcGRkZhIaGkpuba4mJiIggMTGR2NhYYmNjSUxMJDIy0nI9NzeXTp06ceHCBeLi4liwYAHffvstQ4cOLdz3aDabzYX7Cuxjy+FzRd0FkWIn9K1VRd0FkWIpaeajdm1/5vrDhrX1XHCVW7rPZDKxaNEiunTpAlypYvj7+xMVFcWIESOAK1ULX19fJk6cSL9+/UhPT6dcuXLMmzePbt26AZCUlESlSpX48ccf6dChA7t27aJ27drEx8fTtGlTAOLj4wkODmb37t3UrFmTn376idDQUI4dO4a/vz8ACxYsoFevXqSmpuLu7n5Tn0GVDBERETvKysri3LlzVkdWVlah2zl06BApKSmEhIRYzrm6utKqVSvWrVsHQEJCAjk5OVYx/v7+1K1b1xKzfv16PDw8LAkGQLNmzfDw8LCKqVu3riXBAOjQoQNZWVkkJCTcdJ+VZIiIiNhwMJkMO6Kjoy1zH/4+oqOjC92nlJQUAHx9fa3O+/r6Wq6lpKTg4uJC2bJlrxvj4+OTr30fHx+rGNv3KVu2LC4uLpaYm6ElrCIiIjaMXMI6atQohgwZYnXO1dX1ltuzfUKs2Wy+4VNjbWMKir+VmBtRJUNERMSOXF1dcXd3tzpuJcnw8/MDyFdJSE1NtVQd/Pz8yM7OJi0t7boxJ0+ezNf+qVOnrGJs3yctLY2cnJx8FY7rUZIhIiJiw8jhEqNUrVoVPz8/Vq5caTmXnZ3NmjVraN68OQBBQUE4OztbxSQnJ7N9+3ZLTHBwMOnp6WzcuNESs2HDBtLT061itm/fTnJysiVmxYoVuLq6EhQUdNN91nCJiIiIjaLa8TMjI4P9+/dbXh86dIjExEQ8PT0JCAggKiqKCRMmEBgYSGBgIBMmTKBUqVJEREQA4OHhQe/evRk6dCheXl54enoybNgw6tWrR7t27QCoVasWHTt2pE+fPsyaNQuAvn37EhoaSs2aNQEICQmhdu3aREZGMnnyZM6cOcOwYcPo06fPTa8sASUZIiIixcbmzZtp06aN5fXfczl69uzJnDlzeOmll8jMzGTAgAGkpaXRtGlTVqxYQenSpS33TJs2DScnJ7p27UpmZiZt27Zlzpw5ODo6WmLmz5/P4MGDLatQwsPDrfbmcHR05IcffmDAgAG0aNGCkiVLEhERwZQpUwr1ebRPhkgxpn0yRApm730y5mw6alhbvRoHGNbW7UaVDBERERuFWUEh16aJnyIiImIXqmSIiIjYUB3DGEoyREREbBi59PROpiRDRETEhlIMY2hOhoiIiNiFKhkiIiI2NFpiDCUZIiIiNrSE1RgaLhERERG7UCVDRETEhn4DN4aSDBERERsaLjGGkjURERGxC1UyREREbKiOYQwlGSIiIjY0XGIMDZeIiIiIXaiSISIiYkO/gRtDSYaIiIgNDZcYQ0mGiIiIDaUYxlBFSEREROxClQwREREbGi0xhpIMERERGw4aMDGEhktERETELlTJEBERsaHhEmMoyRAREbFh0nCJITRcIiIiInahSoaIiIgNDZcYQ0mGiIiIDa0uMYaGS0RERMQuVMkQERGxoeESYyjJEBERsaEkwxgaLvkXWbNiGb0fbVPU3RARue2ZDPzvTqZKRjEzY8o41q78Id/5aZ98h1+FSkXQo6vWrFjGzLdf596gZoyaMN1y/kLGeZ597EHGTJpJ7fpBRdhDuZMkzXz0utcXrj/Ci3MT/id9mdYziG7BlQHIyc0j6UwmPyWeYPKyXWRm5/5P+iBSHCnJKIbqNwrmuaGvWp1z9yhbRL2x5ujoyI7ETexI3EydBo2KujtyB6v/0tVkPLxRRYaH1eb+sSss5y7Z/OPu5GDicp7Zbv35ZXsKL36WgLOjA03v9mJK5H2UdHFi1JeJdntPsR+HO7sAYRglGcWQs7MLZTy9853/4dv5rFmxjNTkE7iVdue+Zvfz5LODKVGyVIHtHDmwl89mTuXgvl2YTCb8/CvR+4VRVK9RG4C9O7bx5Sfvc2DvTkq7e9C4RRu6P/M8JUqUvGbfXEuUpNkD7fjykxjefG/ONePOnE5l3qxp/LllAyaTAzXr1Kdn/6GU8/MHIDf3MvNmvcNvq37AwcGRNh07czbtLzIvZDB03JRCfFtypzp1Lsvy/+czczCbzZZzFb1KsW1SJ/p9tIGerapxX1VPRn2RSEWvUnSsX57243+x3Pvsg9Xp0/Zumo5ebjnXLbgyA0ICqeTtxvG/LjL71wPMXXPwuv3Jvpxnef9Fm47TvGY5OjbwZ9SXibg4OTDm0bp0blSRu0o688eRNMZ+/SfbjqQB4FHKmfHd69Oqli+lXJ1IPpvJ9J/2sHD9EcO+LymcO32Ywyiak3EbMZlM9Ow/jEkfLqD/8HHsTNzM/I/fu2Z8zMQxeJbz4c335jI+5jPCu/XEyfFKXnn00H6iRw+mcYvWTJz5BS+8PIE9OxKZEzPphv14LLIvxw7vZ8NvPxd4PevSJd54qT8lSpbi1SkfMm7qR5QoWZK3Rg/mck4OAEsXfsbvv8Ty3NBXGTf1YzIvXmDzutWF/1JErmP0I3WZ/csBWo1byeqdJ2/qnoiWVRjRuTZvLdlJq3EriV68g+FhtXi8WUCh3vtSdi7Ojlf+oXrl0bo8fF8FXpibQIcJv3Do1AW+GNyCMqWcAXgpvDY1/Nx5cvrvtHptJaO+2MqZjKzrNS9yW1CSUQxt2RBHr84PWI533hwJwMOPRlCnQSN8/CpQt0FjHu/5HPFrV12znb9OnaRuwyZUCKhC+QoBNHugHZWr1wDg+6/n0aJNBx5+NILyFQKoUac+PfsPY+3PP5Kdff0fbp5e5ejYpTsL53xAbu7lfNfXrVmBg8lE3xdfIaDq3VQIqMpzQ8dy+lQKO/+4Mka+fOlCOnfrSeMWbagQUIWnnx+O212lb/UrEynQR7/s56fEJI79dZGT6Zdu6p4XH76H17/503LfT4lJfPTzfiLvr3rT79ugSlkeaVKJuN2nKOniSI8HqvHmt3/y646T7Es+z/B5W7iUk8sTLaoAUKFsKbYfO8sfR89y/K+L/Lb7FCv/TLmVjywGMZmMO+5kGi4phmrXD6L3oJGW167/P3yxI3Ezixd8yomjh8i8eIHc3FxysrO4dCmzwCGOhx+N4KNpbxK36kfq3teEZve3w9e/IgAH9+3iZPJx4n6JvXqD2Yw5L49TKUlUCLj+D9Twrj35+YdFrF6+lGYPtLe6dmjfLlKSjvN0l1ZW53OyszmZdJyL92SQnnaG6vfUsVxzcHSkauA9mO04Zi53nj/+fzjiZnne5UIFz1K83eM+Jj91n+W8o6OJ85k51723XT0/9r0TjqOjCWdHB5ZvS2L0wm1UKeeGi5MDGw/8ZYm9nGcm8XAagX5XEuu5aw/ycb+m1Asow5pdqcQmJrH54JlC9V2MpeESYyjJKIZKlCiZbyXJqZPJTBwTRbtOj9K153PcVdqd3Tu28eHUN8i9nL+aAPCfyL60aNOBrRt/J3HTOr6Z9yGDR42ncYs2mM1m2j78KB07d8t3n7eP3w376HZXaTp378W3n39Mw6b3W10z55mpGngPA0e8ke8+9zJXJ7Da/iU2K78Qg13Msp78mZdnxvbfDmfHqwVdh///tXPY51vZesj6H/ncGyTA6/aeYuQXieTk5nHy7CXLJFNf9xJXAmxuN/3XqV93nKTxy7G0q+fH/ff4sDDqfuauOcDr326/8YcUKcY0XHKbOLh3F3m5l3mqbxSBtepRvmJl0v46dcP7yleszMOPRvBydAyNW7Rh9YplAFS9uybHjxzAr0KlfIeTs/NN9alD566YTCZiFy2wOl/l7pqknDiGe5my+dou5XYXpdzuwqOsJ/v37LDck5eby+EDewrxjYgU3l8ZWfj8/Y/+/6tTycPy/6fPZ5GUlkllbzcOn7pgdRz76+J1276YlcvhUxc4cSbTahXLoVMZZOXk0uRuL8s5JwcT91Yuy76U85ZzZzKy+Wr9UQZ9upmxX//Bky1vfnhGjOdgMu64kynJuE34+lcgNzeX5UsWcjL5OL+t+pGff/jumvHZWZf4NGYSO7clcOpkMnt2bOPgnp1UqHTlB1dY157s2/Unn8RM5PCBPSSfOMrm9Wv49P3JN90nFxdX/tOjH7FLFlqdb/ngQ5T2KMPb44ax+8+tpKacYOcfCcydMYW/Tl2ZfNchvBtLFs5h87o1JB07zNwZb3Ph/DkNYIpdrdt7Gq+7XHk+pAaVvd3o1aoabepYV+6mfr+LQR1r0PvB6lTzuYt7/N3pFlyZvm3vvqX3zMzO5bO1h3jlsXq0ru1LYPnSTI68j5Iujnz5+2EAhofVokP98lQp50aN8qVpX8/PKgGR/z1txmUMDZfcJqpUr0lkvxdZ+tVnLPj0fe6p25DuTz/PB5PHFhjv4ODI+fPpfDB5LOlnz1DavQyNW7ThPz36AlC5WiCvTp7FwjkzeG1oX8xmM77lKxLcqn2B7V1Lq3ad+OGbzzlx9JDlnGuJEoydMosvZscw9Y2XuHTxImW9y1G3QWNKlnIDILxbD86m/cUHk8fi4OBI24e7cG9QMA6OynvFfvannGfUl4kMfqgmUQ/fww9bTzBz5T6eur+KJeaL3w+TmX2Z/u1r8MojdbmYncvuE+l89MuBW37fCYu242CC6U83wq2EE38cSSPivd9Jv3hlnkf25TxGdalDJa9SZGbnsnH/X/T/eOM//bgiRc5kNhePkfAth88VdRekCOXl5TGsz+M0e6AdXXv2L+ruFBuhb1179ZDInexGO77+U3H7Cjdp+HpaBhaPzRSLgioZUiROnUzmj4R4at97Hzk5OaxY+hWpKUm0aNOxqLsmInKHD3IYR0mGFAkHk4m1K79n/kfvAlCxcjVGv/X+DZfOioj8LzhofpghDE8yjh07xtixY/nkk0+uGZOVlUVWlvWGT9lZWbi4uhrdHSmmvHz8eG3a7KLuhoiI2JHhs+zOnDnD3LlzrxsTHR2Nh4eH1fHpjKlGd0VEROSWmAw87mSFrmQsXbr0utcPHrz+Q4QARo0axZAhQ6zO7UzWPv0iIlJM3OnZgUEKnWR06dIFk8nE9RalmG4wluXq6oqrzdCIyxmtLrkZu/7cwvdfz+Pgvt2cPXOaIWMn07h5a8v1Jzo0LvC+iGcHE/Z45DXbvZBxnoVzPmDT779y4fx5yvn581TfKBo2aQHAoB7hnD6ZnO++9mH/4ZmBI4Arz0NZ9s3nAHTu1pOHH42wxO3fvZ1Ppk/kzffm4ODoWOjPLXI9PR6oSo8HqlHJ68oTifckn2PaD7v5dcdJnBxMjOhcmwfr+lHZ241zmTn8tjuVCYt23PTzTDo3qsiMZ5sQm5jEMzPjLeeb3u3FgJAa1Asog1+ZkjwzYz2x26z/njzXPpD+7QMBiFm+l49+3m+51rBKWaKfaMDDb/2KdtSXf6NCJxnly5fn/fffp0uXLgVeT0xMJCgo6J/2S64h61ImAdVq0CokjGlvjMh3fcaXP1m9Tty0jg+nvUmTlm2u2eblnBwmjHoe9zKeRL0yEU9vH/46dZKS//UI+fHvzSUv7+oWzccOH2DCqIE0u78dcOWprl/Pm8Xw16eB2cykV4dQ774mVKpyN5cvX+bj96Lp88LLSjDELpLTMpmweDuHUy8A8HhwAJ/2DyZk/M8kp2VSL6AM7/y4m53H0/Eo5cxrj9dnzoBgHor+9YZtV/AsyZjH6hG/73S+a6VcndhxPJ0F644w+7lm+a7f4+/O8LBa9Hh/PSZg7vPNWbsrlT1J53ByMDExoiHD529VglEM3embaBml0ElGUFAQW7ZsuWaScaMqh/wzDRq3oEHjFte8XsbT2+p1wvq11K4fhG/5ite859flS8k4f47Xpn2Ck9OVPxLlfMtbxfz3M0cAliyci2/5itS698pDpE4cPURA1UDqNrhSSQmoejcnjh6mUpW7+f7redSq15DqNesgYg+2TyyduGQnPR6oRlBVT75MPkL3d3+3uv7Kwm38NKoNFcqW5ERa5jXbdTDB+8805u1lO2kS6I1HSest93/dcZJfd1z7EfKB5Uuz8/g5ft9z5REAu06kE+hXmj1J5+gfUoP4/afZVsiHuMn/hhaXGKPQScbw4cO5cOHCNa/ffffd/PrrjX87EPs7m/YXWzfG0X/YuOvGbYlfS2CtenwaM5HN69fi7lGGFm06Et61R4GVh8s5OcT98hOdHn3SMjQWUPVuko8f5XRqCmazmZQTR6lUpTopJ46xZuX3TIj5zB4fUSQfBxOEBVWklIsjmw8V/CRT95JO5OWZSb/Bk1WHdKrFXxnZfLnuCE0Cva8bW5BdJ85RzfcuKpQtCSao5nMXu5POUaWcG12DA+g4QT8r5d+t0EnG/ffff93rbm5utGrV6rox8r+xduUPlCjpRuPrDJUApCaf4FTiZlo82JERb75DyoljfBozidzcyzz2VJ988ZvWreZiRgYPhIRazlUIqEq3pwcwYdTzAHR7+nkqBFRl/IgBRDw7iD8S4vlm3oc4OjnRs/9QatW7L1+7Iv/EPf7uLHupNa7ODlzIukzvWfHsS87//A9XJwdefqQuizYdI+NSwU8wBmhc3ZPuLaoQ8ubPt9yn/SnneWvxDha80BKA6MU72J9ynoUvtGT8d9tpXceHoZ1qcTnXzJivtrFh/183aFH+V1TIMIYeFPEvtmb5Ulo82BEXl+vvP5JnNuNepix9XniZaoG1aN46hC5PPM2qH74tMH718qU0aByMp1c5q/PtQx9j6uxvmTr7W9qHPsaaFcsoUcqNwFr1+HDamwwZO5nIvi/y3oTR5GRnG/Y5RQAOnDxP+/E/EzpxNZ+tPcS7PRsRWL60VYyTg4kZzzbBwWRi1JeJ12zLzdWJ6U83ZvjnWzhz4Z/9WZ332yHuH7eS+8etZN5vh+gaHEBG1mU2HzzDlKfuo/eseMZ98wcznm2Ci5N+JBcbRbSG9fLly7zyyitUrVqVkiVLUq1aNV5//XXy8vIsMWazmXHjxuHv70/JkiVp3bo1O3bssGonKyuLQYMG4e3tjZubG+Hh4Rw/ftwqJi0tjcjISMtWEpGRkZw9e7ZwHb4B/Yn+l9r951aSjh/hwY6dbxhbxtOL8hUCrIZGKgRU4eyZv7icY11OPnUymT+3bqRNxy7XbfNc+lm+m/8xvQYMY//u7ZSvEED5CgHUadCI3NzLJJ84ekufS+RacnLNHD51gT+OniV68Q52Hk/n2TZXn5zq5GBiVt+mVPIuRfd3465bxahSzo0AbzfmDgjm6PtdOPp+Fx5vGkDIveU5+n4XKnu73VIfPd1cePHhWryyYBv3VfXkYGoGh1IvsG7vaZwdHajmc9cttSv/HhMnTmTmzJnExMSwa9cuJk2axOTJk5k+fbolZtKkSUydOpWYmBg2bdqEn58f7du35/z5q5W7qKgoFi1axIIFC4iLiyMjI4PQ0FByc69O4I+IiCAxMZHY2FhiY2NJTEwkMvLaqxBvhbYV/5f6dfkSqgbWonL1GjeMrVm7Pr+vXk5eXh4ODlfyzuTjRynj6Y2Ts/VEtzUrluFRpiwNm1578inAZzPf5qFHI/Aq58vBvTu5nHv1B3pebq7VShURuzCBi/OVP89/JxhVy7nxn2m/kXaD6sT+lPO0ed364XQjwmvjVsKJV7/6g6S0i7fUpde63stHP+8j+Wwm9auUxdnh6u95jg4mHB1UpC8uimp1yfr16+ncuTOdOnUCoEqVKnz55Zds3rwZuFLFeOeddxg9ejSPPnrlIXFz587F19eXL774gn79+pGens7s2bOZN28e7dpdWQH4+eefU6lSJVatWkWHDh3YtWsXsbGxxMfH07RpUwA++ugjgoOD2bNnDzVr1jTk86iScZu5lHmRwwf2cPjAHgBOpSRx+MAeTqdenV1/8UIGG9b+TJtrVDE+mDSWLz+JsbxuH/oYGefSmTvjbZKPH2HLhjgWL5hDSNjjVvfl5eWxZsUyHmjXCUfHa+enfyRsIOXEMcv91WvWIenYERI3/c7PP36HycEB/4qVb/k7ELE1snMdmtztRUWvUtzj786IzrVpXqMcizYew9HBxEf9mlI/oAwDP9mMo4OJcu6ulHN3xdnx6j8k7/YKYlSXKyugsi7nsSfpnNWRnpnDhUuX2ZN0jpzcKyvoSrk6UqeiB3UqegBQyduNOhU9rkz0tPFALR+q+tzFp2uubFiYePgM1f1K06aOL0+2rEKe2cyBk/nnkEjRMJmMO7Kysjh37pzVYftojb+1bNmSn3/+mb179wKwbds24uLiePjhhwE4dOgQKSkphISEWO5xdXWlVatWrFu3DoCEhARycnKsYvz9/albt64lZv369Xh4eFgSDIBmzZrh4eFhiTGCKhm3mYN7d/HGS89ZXs+bNQ2AB9p3sqwiWb9mBWbMtGjTocA2Tp9KwfRfvzF5+fgxasJ05s2axojnIijrXY6HunQnvGsPq/u2b93I6dQUWncIv2b/srMuMeeDSQx+eYKlKuLp7UOvAcOY+fbrODu70H/YOFxcS9zS5xcpSDl3V6Y/3Qgf9xKcz8xh14lzPDn9d9buSqWiVyk61PcHYNWYtlb3PTZ1Lev3Xtn/ooJnqULvV1G/clm+HfKA5fVrj98LwML1R3hxboLlfAlnB8Z3q89zH2/k7xX+KWcvMWbhNqb1CCL7ch4vzEngUk4eUjwYWceIjo7mtddeszo3duxYxo0bly92xIgRpKenc8899+Do6Ehubi7jx4/niSeeACAl5covlL6+vlb3+fr6cuTIEUuMi4sLZcuWzRfz9/0pKSn4+Pjke38fHx9LjBFM5mKyqcWWw9rxU8RW6FurbhwkcgdKmvmoXds38t+kOuVd81UuCtr5GmDBggUMHz6cyZMnU6dOHRITE4mKimLq1Kn07NmTdevW0aJFC5KSkihf/up+Rn369OHYsWPExsbyxRdf8PTTT+d7z/bt21O9enVmzpzJhAkTmDt3Lnv27LGKCQwMpHfv3owcOdKQz65KhoiIiC0DSxnXSigKMnz4cEaOHEn37t0BqFevHkeOHCE6OpqePXvi5+cHXKlE/HeSkZqaaqlu+Pn5kZ2dTVpamlU1IzU1lebNm1tiTp7Mv5HcqVOn8lVJ/gnNyRAREbFhMvC/wrh48aJlqPlvjo6OliWsVatWxc/Pj5UrV1quZ2dns2bNGksCERQUhLOzs1VMcnIy27dvt8QEBweTnp7Oxo0bLTEbNmwgPT3dEmMEVTJERESKibCwMMaPH09AQAB16tRh69atTJ06lWeeeQa48uiOqKgoJkyYQGBgIIGBgUyYMIFSpUoREXHloZQeHh707t2boUOH4uXlhaenJ8OGDaNevXqW1Sa1atWiY8eO9OnTh1mzZgHQt29fQkNDDVtZAkoyRERE8imqZ5dMnz6dMWPGMGDAAFJTU/H396dfv368+uqrlpiXXnqJzMxMBgwYQFpaGk2bNmXFihWULn1187lp06bh5ORE165dyczMpG3btsyZMwfH/9oPaf78+QwePNiyCiU8PJyYmKsrD42giZ8ixZgmfooUzN4TP7cdNW45cf2A0jcO+pfSnAwRERGxCw2XiIiI2NLmq4ZQkiEiImKjqLYV/7fRcImIiIjYhSoZIiIiNopqdcm/jZIMERERG8oxjKEkQ0RExJayDENoToaIiIjYhSoZIiIiNrS6xBhKMkRERGxo4qcxNFwiIiIidqFKhoiIiA0VMoyhJENERMSWsgxDaLhERERE7EKVDBERERtaXWIMJRkiIiI2tLrEGBouEREREbtQJUNERMSGChnGUJIhIiJiS1mGIZRkiIiI2NDET2NoToaIiIjYhSoZIiIiNrS6xBhKMkRERGwoxzCGhktERETELlTJEBERsaVShiGUZIiIiNjQ6hJjaLhERERE7EKVDBERERtaXWIMJRkiIiI2lGMYQ8MlIiIiYheqZIiIiNhSKcMQSjJERERsaHWJMZRkiIiI2NDET2NoToaIiIjYhSoZIiIiNlTIMIaSDBERERsaLjGGhktERETELlTJEBERyUelDCMoyRAREbGh4RJjaLhERERE7EKVDBERERsqZBhDSYaIiIgNDZcYQ8MlIiIiYheqZIiIiNjQs0uMoSRDRETElnIMQyjJEBERsaEcwxiakyEiIiJ2oUqGiIiIDa0uMYaSDBERERua+GkMDZeIiIiIXaiSISIiYkuFDEMoyRAREbGhHMMYGi4REREpRk6cOMFTTz2Fl5cXpUqVokGDBiQkJFium81mxo0bh7+/PyVLlqR169bs2LHDqo2srCwGDRqEt7c3bm5uhIeHc/z4cauYtLQ0IiMj8fDwwMPDg8jISM6ePWvoZ1GSISIiYsNkMu4ojLS0NFq0aIGzszM//fQTO3fu5O2336ZMmTKWmEmTJjF16lRiYmLYtGkTfn5+tG/fnvPnz1tioqKiWLRoEQsWLCAuLo6MjAxCQ0PJzc21xERERJCYmEhsbCyxsbEkJiYSGRn5T786Kyaz2Ww2tMVbtOXwuaLugkixE/rWqqLugkixlDTzUbu2f+ZC7o2DbpKnm+NNx44cOZLff/+d3377rcDrZrMZf39/oqKiGDFiBHClauHr68vEiRPp168f6enplCtXjnnz5tGtWzcAkpKSqFSpEj/++CMdOnRg165d1K5dm/j4eJo2bQpAfHw8wcHB7N69m5o1a/7DT32FKhkiIiJ2lJWVxblz56yOrKysAmOXLl1Ko0aNePzxx/Hx8aFhw4Z89NFHluuHDh0iJSWFkJAQyzlXV1datWrFunXrAEhISCAnJ8cqxt/fn7p161pi1q9fj4eHhyXBAGjWrBkeHh6WGCMoyRAREbFh5HBJdHS0Zd7D30d0dHSB73vw4EFmzJhBYGAgy5cv57nnnmPw4MF89tlnAKSkpADg6+trdZ+vr6/lWkpKCi4uLpQtW/a6MT4+Pvne38fHxxJjBK0uERERsaNRo0YxZMgQq3Ourq4Fxubl5dGoUSMmTJgAQMOGDdmxYwczZsygR48eljiTzWQPs9mc75wt25iC4m+mncJQJUNERMSGkZUMV1dX3N3drY5rJRnly5endu3aVudq1arF0aNHAfDz8wPIV21ITU21VDf8/PzIzs4mLS3tujEnT57M9/6nTp3KVyX5J5RkiIiIFBMtWrRgz549Vuf27t1L5cqVAahatSp+fn6sXLnScj07O5s1a9bQvHlzAIKCgnB2draKSU5OZvv27ZaY4OBg0tPT2bhxoyVmw4YNpKenW2KMoOESERERG0X17JIXX3yR5s2bM2HCBLp27crGjRv58MMP+fDDD6/0y2QiKiqKCRMmEBgYSGBgIBMmTKBUqVJEREQA4OHhQe/evRk6dCheXl54enoybNgw6tWrR7t27YAr1ZGOHTvSp08fZs2aBUDfvn0JDQ01bGUJKMkQERHJp6iewtq4cWMWLVrEqFGjeP3116latSrvvPMOTz75pCXmpZdeIjMzkwEDBpCWlkbTpk1ZsWIFpUuXtsRMmzYNJycnunbtSmZmJm3btmXOnDk4Ol5dTjt//nwGDx5sWYUSHh5OTEyMoZ9H+2SIFGPaJ0OkYPbeJ+PcpTzD2nIvcefOTFAlQ0RExIaeXWIMJRkiIiK2lGUY4s6t4YiIiIhdqZIhIiJio6hWl/zbKMkQERGxUVSrS/5tNFwiIiIidqFKhoiIiA0VMoyhJENERMSWsgxDKMkQERGxoYmfxtCcDBEREbELVTJERERsaHWJMYrNs0ukeMjKyiI6OppRo0bh6upa1N0RKRb090Lk1ijJECvnzp3Dw8OD9PR03N3di7o7IsWC/l6I3BrNyRARERG7UJIhIiIidqEkQ0REROxCSYZYcXV1ZezYsZrcJvJf9PdC5NZo4qeIiIjYhSoZIiIiYhdKMkRERMQulGSIiIiIXSjJEBEREbtQkiEWH3zwAVWrVqVEiRIEBQXx22+/FXWXRIrU2rVrCQsLw9/fH5PJxOLFi4u6SyK3FSUZAsDChQuJiopi9OjRbN26lfvvv5+HHnqIo0ePFnXXRIrMhQsXqF+/PjExMUXdFZHbkpawCgBNmzblvvvuY8aMGZZztWrVokuXLkRHRxdhz0SKB5PJxKJFi+jSpUtRd0XktqFKhpCdnU1CQgIhISFW50NCQli3bl0R9UpERG53SjKE06dPk5ubi6+vr9V5X19fUlJSiqhXIiJyu1OSIRYmk8nqtdlszndORETkZinJELy9vXF0dMxXtUhNTc1X3RAREblZSjIEFxcXgoKCWLlypdX5lStX0rx58yLqlYiI3O6ciroDUjwMGTKEyMhIGjVqRHBwMB9++CFHjx7lueeeK+quiRSZjIwM9u/fb3l96NAhEhMT8fT0JCAgoAh7JnJ70BJWsfjggw+YNGkSycnJ1K1bl2nTpvHAAw8UdbdEiszq1atp06ZNvvM9e/Zkzpw5//sOidxmlGSIiIiIXWhOhoiIiNiFkgwRERGxCyUZIiIiYhdKMkRERMQulGSIiIiIXSjJEBEREbtQkiEiIiJ2oSRDRERE7EJJhoiIiNiFkgwRERGxCyUZIiIiYhdKMkRERMQu/g8m4OvM6brKCQAAAABJRU5ErkJggg=="
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "svm_model_lin = SVC(kernel=\"poly\", degree=5, random_state=RANDOM_SEED)\n",
    "fitted_svm_model_lin = utils.train_eval_model(svm_model_lin, X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:09:44] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.70</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">124</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:09:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.70\u001b[0m                                  \u001b]8;id=599634;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=163483;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\u001b\\\u001b[2m124\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.test_eval_model(fitted_svm_model_lin, X_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Furthermore, we can train an even more flexible model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:11:42] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.70</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">105</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:11:42]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.70\u001b[0m                                  \u001b]8;id=973406;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=235666;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\u001b\\\u001b[2m105\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "svm_model_poly = SVC(kernel=\"poly\", degree=5, random_state=RANDOM_SEED)\n",
    "fitted_svm_model_poly = utils.train_eval_model(svm_model_poly, X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:11:48] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.70</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">124</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:11:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.70\u001b[0m                                  \u001b]8;id=447450;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=708295;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\u001b\\\u001b[2m124\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.test_eval_model(fitted_svm_model_poly, X_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "... and an even more flexible model..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:13:18] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.69</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">105</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:13:18]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.69\u001b[0m                                  \u001b]8;id=353429;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=892090;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#105\u001b\\\u001b[2m105\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "svm_model_rbf = SVC(kernel=\"rbf\", gamma=\"scale\", random_state=RANDOM_SEED)\n",
    "fitted_svm_model_rbf = utils.train_eval_model(svm_model_rbf, X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A more flexible model might cause reduced validation accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"color: #7fbfbf; text-decoration-color: #7fbfbf\">[18:13:30] </span><span style=\"color: #000080; text-decoration-color: #000080\">INFO    </span> The accuracy on the validation data is <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">0.69</span>                                  <a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">model_utils.py</span></a><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">:</span><a href=\"file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\" target=\"_blank\"><span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\">124</span></a>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[2;36m[18:13:30]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO    \u001b[0m The accuracy on the validation data is \u001b[1;36m0.69\u001b[0m                                  \u001b]8;id=217238;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py\u001b\\\u001b[2mmodel_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=197264;file:///home/katalitf/Documents/conferences/NBD_22_workshop/responsible-datascience-ws/src/workshop/model_utils.py#124\u001b\\\u001b[2m124\u001b[0m\u001b]8;;\u001b\\\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.test_eval_model(fitted_svm_model_rbf, X_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Incuding more features did not make the model much better. Feature engineering would let us know which features are most important."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAHFCAYAAAAOmtghAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABOJklEQVR4nO3dd1QU1/8+8GepSxMFlCbNCkbF2MVuIqhYYzcRQaxosCVRNMYSI5ZYkthiFNBIlNg1VixgQz9qsESMsYMKYkOs1Pv7wx/7dd0VWVxYGJ/XOZzD3r0z857Zu+zDlB2ZEEKAiIiISCL0dF0AERERkTYx3BAREZGkMNwQERGRpDDcEBERkaQw3BAREZGkMNwQERGRpDDcEBERkaQw3BAREZGkMNwQERGRpDDcvOHnn3+GTCZDzZo1dV2KZN24cQO+vr6wsrKCTCbD6NGj39p35syZ2LJlS5HVIZPJEBERUSTzf5dWrVqhVatWOlm2Jvz9/eHq6lqoaUvLOua5c+cOpk6dijNnzhTJ/CMiIiCTyXDjxg2Np9X1eC0uU6dOhUwmKxHLLej4fZ/XJiEhAVOnTlU7Jt7nvVcUimo9iwLDzRvCwsIAABcuXMCJEyd0XI00jRkzBidOnEBYWBji4uIwZsyYt/YtynBjb2+PuLg4+Pr6Fsn8qfS5c+cOpk2bVmThxtfXF3FxcbC3t9d4Wo7X4rdkyRIsWbKkSJeRkJCAadOmqf3Qnzx5MjZv3lykyy8u+a1nUTAolqWUEqdOncLZs2fh6+uLHTt2YOXKlWjUqJGuy1Lr+fPnMDU11XUZhfLPP/+gYcOG6Nq1q1bn++LFC8jl8gL/12dsbIzGjRtrtQb6sGj6PixfvjzKly9fqGVxvBa/GjVq6HT5lStX1unySzPuuXnNypUrAQCzZs2Cl5cX1q1bh+fPn6v0u337NoYMGQInJycYGRnBwcEBPXr0wN27dxV90tLSMG7cOFSqVAnGxsaoUKECOnTogH///RcAEBMTA5lMhpiYGKV5q9vt5+/vD3Nzc5w/fx7e3t6wsLDAJ598AgCIjo5Gly5dULFiRcjlclSpUgVDhw7F/fv3Ver+999/0bdvX9ja2sLY2BjOzs7w8/NDRkYGbty4AQMDA4SGhqpMd+jQIchkMqxfvz7f7ZeYmIgvvvgCFSpUgLGxMTw8PDBv3jzk5uYqrfOVK1ewa9cuyGSyfHfRy2QyPHv2DKtWrVL0zdtFnLd7f+/evRg4cCDKly8PU1NTZGRk4MqVKwgICEDVqlVhamoKR0dHdOrUCefPn3/nts7bNX3hwgX07dsXlpaWsLW1xcCBA/H48WOl6YUQWLJkCerUqQMTExOUK1cOPXr0wLVr11T6zZkzBy4uLpDL5ahbty527dqV77Z8czuMHDkS4eHhqF69OkxMTFC/fn0cP34cQgjMnTsXbm5uMDc3R5s2bXDlyhWVeYSFhcHT0xNyuRxWVlbo1q0bLl68qNIvIiIC1atXV7x+q1evVltTZmYmZsyYAXd3dxgbG6N8+fIICAjAvXv33rk+S5cuhaenJ8zNzWFhYQF3d3dMnDjxndM9fPgQQUFBcHR0hJGRESpVqoRJkyYhIyND7fb6/fff4eHhAVNTU3h6euKvv/7Kd/4xMTFo0KABACAgIEAx5qZOnQpAO+9DdYelWrVqhZo1a+LkyZNo3rw5TE1NUalSJcyaNUvx3gHef7ympaUhMDAQVlZWMDc3h6+vL65du6a0jm/z8uVLjBs3DnXq1IGlpSWsrKzQpEkTbN26VaWvJtt/x44dqFOnDoyNjeHm5oYff/wx3zryjB49GmZmZkhPT1d5rnfv3rC1tUVWVhYAICoqCt7e3rC3t4eJiQk8PDwwYcIEPHv27J3LUXdY6s6dO+jVqxcsLCxgaWmJ3r17IyUlRWXaU6dOoU+fPnB1dYWJiQlcXV3Rt29f3Lx5U9EnIiICPXv2BAC0bt1aMebyXmN1h6VevnyJkJAQuLm5wcjICI6OjhgxYgTS0tKU+rm6uqJjx47YvXs36tatCxMTE7i7uyuOTrxLca6nJp9jBSZICCHE8+fPhaWlpWjQoIEQQogVK1YIACIiIkKp361bt4S9vb2wsbER8+fPF/v27RNRUVFi4MCB4uLFi0IIIdLT08VHH30kzMzMxPTp08WePXvExo0bxahRo8SBAweEEEIcPHhQABAHDx5Umv/169cFABEeHq5oGzBggDA0NBSurq4iNDRU7N+/X+zZs0cIIcTSpUtFaGio2LZtm4iNjRWrVq0Snp6eonr16iIzM1MxjzNnzghzc3Ph6uoqli1bJvbv3y/WrFkjevXqJdLT04UQQnTr1k04OzuL7OxspZp69uwpHBwcRFZW1lu3X2pqqnB0dBTly5cXy5YtE7t37xYjR44UAMTw4cOFEEI8fvxYxMXFCTs7O9G0aVMRFxcn4uLixMuXL9XOMy4uTpiYmIgOHToo+l64cEEIIUR4eLgAIBwdHcWQIUPErl27xIYNG0R2draIjY0V48aNExs2bBCxsbFi8+bNomvXrsLExET8+++/+W7rKVOmCACievXq4rvvvhPR0dFi/vz5wtjYWAQEBCjVN3jwYGFoaCjGjRsndu/eLf744w/h7u4ubG1tRUpKiso8AwMDxa5du8Ty5cuFo6OjsLOzEy1btnzrNs0DQLi4uAgvLy+xadMmsXnzZlGtWjVhZWUlxowZI7p06SL++usvERkZKWxtbUXt2rVFbm6uYvqZM2cKAKJv375ix44dYvXq1aJSpUrC0tJS/Pfff4p+edu0S5cuYvv27WLNmjWiSpUqwsnJSbi4uCj65eTkiHbt2gkzMzMxbdo0ER0dLVasWCEcHR1FjRo1xPPnzxV9W7ZsqbSOa9euFQDEl19+Kfbu3Sv27dsnli1bJoKDg/PdBi9evBC1a9cWZmZm4scffxR79+4VkydPFgYGBqJDhw4q28vV1VU0bNhQ/Pnnn2Lnzp2iVatWwsDAQFy9evWty3j8+LFiG3z77beKMZeUlCSE0M77MG/+169fV9pG1tbWomrVqmLZsmUiOjpaBAUFCQBi1apVin7vM15zcnJEs2bNhFwuF7NmzRJ79+4V06ZNE1WrVhUAxJQpU/Ld/mlpacLf31/8/vvv4sCBA2L37t3iq6++Enp6eko1arL99+3bJ/T19UWzZs3Epk2bxPr160WDBg2Es7OzeNdH09mzZwUA8dtvvym1P3r0SBgbG4uxY8cq2r7//nuxYMECsWPHDhETEyOWLVsm3NzcROvWrZWmzduWr3tz/D5//lx4eHgIS0tL8csvv4g9e/aI4OBgRc2vvzbr168X3333ndi8ebOIjY0V69atEy1bthTly5cX9+7dE0K8+ruZ9/5cvHixYsylpqYKIV6Nudffe7m5ucLHx0cYGBiIyZMni71794off/xRmJmZiY8//ljpb6mLi4uoWLGiqFGjhli9erXYs2eP6NmzpwAgYmNj892+xb2eBX3/aILh5v9bvXq1ACCWLVsmhBDiyZMnwtzcXDRv3lyp38CBA4WhoaFISEh467ymT58uAIjo6Oi39tE03AAQYWFh+a5Dbm6uyMrKEjdv3hQAxNatWxXPtWnTRpQtW1YxmPKrafPmzYq227dvCwMDAzFt2rR8lz1hwgQBQJw4cUKpffjw4UImk4lLly4p2lxcXISvr2++88tjZmYmBgwYoNKe9yHh5+f3znlkZ2eLzMxMUbVqVTFmzBhFe34fFnPmzFGaR1BQkJDL5YrQEBcXJwCIefPmKfVLSkoSJiYm4ptvvhFCvPpjK5fLRbdu3ZT6HT16VAAocLixs7MTT58+VbRt2bJFABB16tRRCjILFy4UAMS5c+cUy88LiK9LTEwUxsbGol+/fkKIVx9+Dg4Oom7dukrzu3HjhjA0NFT6A5sXUDZu3Kg0z5MnTwoAYsmSJYq2Nz8cRo4cKcqWLfvOdX7TsmXLBADx559/KrXPnj1bABB79+5VtAEQtra2itAuhBApKSlCT09PhIaG5rucvHV4fUzk0cb78G3hRt17p0aNGsLHx0fx+H3G644dOwQAsXTpUqV+oaGhBQo3b8rOzhZZWVkiMDBQfPzxx0rPFXT7N2rUSDg4OIgXL14o2tLT04WVldU7w40QQtStW1d4eXkptS1ZskQAEOfPn1c7Td5rExsbKwCIs2fPKp4rSLhZunSpymsqxKt/dN42bvJkZ2eLp0+fCjMzM/HTTz8p2tevX6/2s0AI1XCze/duta93VFSUACCWL1+uaHNxcRFyuVzcvHlT0fbixQthZWUlhg4d+tY6dbGer8vv/aMJHpb6/1auXAkTExP06dMHAGBubo6ePXvi8OHDuHz5sqLfrl270Lp1a3h4eLx1Xrt27UK1atXw6aefarXG7t27q7SlpqZi2LBhcHJygoGBAQwNDeHi4gIAisMOz58/R2xsLHr16pXv8f5WrVrB09MTixcvVrQtW7YMMpkMQ4YMybe2AwcOoEaNGmjYsKFSu7+/P4QQOHDgQIHXUxPqtkl2djZmzpyJGjVqwMjICAYGBjAyMsLly5fVHopRp3PnzkqPa9eujZcvXyI1NRUA8Ndff0Emk+GLL75Adna24sfOzg6enp6Kw41xcXF4+fIlPv/8c6X5eXl5KV6ngmjdujXMzMwUj/PGX/v27ZXOMcprz9slHBcXhxcvXsDf319pfk5OTmjTpg32798PALh06RLu3LmDfv36Kc3PxcUFXl5eStP+9ddfKFu2LDp16qS07nXq1IGdnZ3KodbXNWzYEGlpaejbty+2bt1a4N3OBw4cgJmZGXr06KHUnrdeeeuRp3Xr1rCwsFA8trW1RYUKFZR2lRdWYd+H+bGzs1N579SuXbvA9b5rvMbGxgIAevXqpdSvb9++BZo/AKxfvx5NmzaFubm5Yh1Xrlypdv3etf2fPXuGkydP4rPPPoNcLlf0s7CwQKdOnQpUT0BAAI4dO4ZLly4p2sLDw9GgQQOlq12vXbuGfv36wc7ODvr6+jA0NETLli0BFOy1ed3BgwdhYWGhsr379eun0vfp06cYP348qlSpAgMDAxgYGMDc3BzPnj3TeLl58v6Ovvl+7tmzJ8zMzFTeB3Xq1IGzs7PisVwuR7Vq1d45rop7Pd/3/aMOww2AK1eu4NChQ/D19YUQAmlpaUhLS1P8IX39GOW9e/dQsWLFfOdXkD6aMjU1RZkyZZTacnNz4e3tjU2bNuGbb77B/v378b///Q/Hjx8H8OoEWwB49OgRcnJyClRTcHAw9u/fj0uXLiErKwu//fYbevToATs7u3yne/DggdorQBwcHBTPFwV1yxw7diwmT56Mrl27Yvv27Thx4gROnjwJT09PxTZ5F2tra6XHxsbGAP5vm969exdCCNja2sLQ0FDp5/jx44oP7bz1Vrf93rVNX2dlZaX02MjIKN/2ly9fKi3/ba9N3vOa1Hn37l2kpaXByMhIZd1TUlLyDSz9+/dHWFgYbt68ie7du6NChQpo1KgRoqOj377y/78+Ozs7lZPFK1SoAAMDA5Xx9ebrB7x6DQv6+r/N+7wP8/O+9b5rvD548AAGBgYq48XW1rZA89+0aRN69eoFR0dHrFmzBnFxcTh58iQGDhyoGGuarM+jR4+Qm5v7Xu+Lzz//HMbGxorzNhISEnDy5EkEBAQo+jx9+hTNmzfHiRMnMGPGDMTExODkyZPYtGkTgIK9Nq978OCB2m2mruZ+/fph0aJFGDRoEPbs2YP//e9/OHnyJMqXL1/ocZj3Or75T6pMJoOdnZ3W3gfFuZ7aeP+ow6ul8Cq8CCGwYcMGbNiwQeX5VatWYcaMGdDX10f58uVx69atfOdXkD55/628eTLk2z4Y1F0B9M8//+Ds2bOIiIjAgAEDFO1vnlBqZWUFfX39d9YEvBqo48ePx+LFi9G4cWOkpKRgxIgR75zO2toaycnJKu137twBANjY2LxzHoWhbrusWbMGfn5+mDlzplL7/fv3UbZsWa0s18bGBjKZDIcPH1Z8kLwury3vj4u6E/FSUlKK/Dss8pb/ttcm73V5V52vs7GxgbW1NXbv3q12ma//x65OQEAAAgIC8OzZMxw6dAhTpkxBx44d8d9//711b5a1tTVOnDgBIYTSa56amors7OwiG19vep/3oS5ZW1sjOzsbDx8+VAo46l5vddasWQM3NzdERUUpbYM3/34VVLly5SCTyQo03vKbR5cuXbB69WrMmDED4eHhkMvlSnujDhw4gDt37iAmJkaxtwaAysm3BWVtbY3//e9/76z58ePH+OuvvzBlyhRMmDBB0Z6RkYGHDx8Watl5y8/Ozsa9e/eUAo4QAikpKYoT4t9Xca5nUb1/Pvg9Nzk5OVi1ahUqV66MgwcPqvyMGzcOycnJiqtb2rdvj4MHDyrtCn1T+/bt8d9//+V7KCbvQ+3cuXNK7du2bStw7Xl/ZN78cP3111+VHpuYmKBly5ZYv379Ow8DyOVyDBkyBKtWrcL8+fNRp04dNG3a9J21fPLJJ0hISMDff/+t1L569WrIZDK0bt26IKukojD/bctkMpVtsmPHDty+fbtQNajTsWNHCCFw+/Zt1K9fX+WnVq1aAIDGjRtDLpcjMjJSafpjx45p5RDJuzRp0gQmJiZYs2aNUvutW7dw4MABxdU+1atXh729PdauXQshhKLfzZs3cezYMaVpO3bsiAcPHiAnJ0ftulevXr1AtZmZmaF9+/aYNGkSMjMzceHChbf2/eSTT/D06VOV7zzKu5orbz3e15t7PAqioO9DXcr7YI+KilJqX7duXYGml8lkMDIyUgo2KSkpaq+WKggzMzM0bNgQmzZtUtrz8+TJE2zfvr3A8wkICMCdO3ewc+dOrFmzBt26dVP6B0bbr03r1q3x5MkTlb/Tf/zxh9JjmUwGIYTKclesWIGcnBylNk3GXN44f/P9vHHjRjx79kxr74PiXM+iev988Htudu3ahTt37mD27Nlqv4myZs2aWLRoEVauXImOHTti+vTp2LVrF1q0aIGJEyeiVq1aSEtLw+7duzF27Fi4u7tj9OjRiIqKQpcuXTBhwgQ0bNgQL168QGxsLDp27IjWrVvDzs4On376KUJDQ1GuXDm4uLhg//79it2lBeHu7o7KlStjwoQJEELAysoK27dvV7uLf/78+WjWrBkaNWqECRMmoEqVKrh79y62bduGX3/9Vem/7aCgIMyZMwenT5/GihUrClTLmDFjsHr1avj6+mL69OlwcXHBjh07sGTJEgwfPhzVqlUr8Hq9rlatWoiJicH27dthb28PCwuLd354duzYEREREXB3d0ft2rVx+vRpzJ07V6uHCps2bYohQ4YgICAAp06dQosWLWBmZobk5GQcOXIEtWrVwvDhw1GuXDl89dVXmDFjBgYNGoSePXsiKSkJU6dO1eiwVGGVLVsWkydPxsSJE+Hn54e+ffviwYMHmDZtGuRyOaZMmQIA0NPTw/fff49BgwahW7duGDx4MNLS0tTW2adPH0RGRqJDhw4YNWoUGjZsCENDQ9y6dQsHDx5Ely5d0K1bN7X1DB48GCYmJmjatCns7e2RkpKC0NBQWFpa5vtfp5+fHxYvXowBAwbgxo0bqFWrFo4cOYKZM2eiQ4cOWju/rXLlyjAxMUFkZCQ8PDxgbm4OBwcHxeFVdTR5H+pKu3bt0LRpU4wbNw7p6emoV68e4uLiFOFQTy///3M7duyITZs2ISgoCD169EBSUhK+//572NvbK52TqInvv/8e7dq1Q9u2bTFu3Djk5ORg9uzZMDMzK/B//d7e3qhYsSKCgoKQkpKidEgKeHVuW7ly5TBs2DBMmTIFhoaGiIyMxNmzZwtVs5+fHxYsWAA/Pz/88MMPqFq1Knbu3Ik9e/Yo9StTpgxatGiBuXPnwsbGBq6uroiNjcXKlStV9h7nnR+0fPlyWFhYQC6Xw83NTe0hpbZt28LHxwfjx49Heno6mjZtinPnzmHKlCn4+OOP0b9//0Ktly7Xs8jeP4U6DVlCunbtKoyMjPK9iqhPnz7CwMBAcXlvUlKSGDhwoLCzsxOGhobCwcFB9OrVS9y9e1cxzaNHj8SoUaOEs7OzMDQ0FBUqVBC+vr5KlyInJyeLHj16CCsrK2FpaSm++OILcerUKbVXS5mZmamtLSEhQbRt21ZYWFiIcuXKiZ49e4rExES1V0AkJCSInj17Cmtra2FkZCScnZ2Fv7+/2kuxW7VqJaysrJQu632Xmzdvin79+glra2thaGgoqlevLubOnStycnKU+mlytdSZM2dE06ZNhampqdLVRXlXnZw8eVJlmkePHonAwEBRoUIFYWpqKpo1ayYOHz6scuVDflef5F3CmEfdVS5CCBEWFiYaNWokzMzMhImJiahcubLw8/MTp06dUvTJzc0VoaGhwsnJSRgZGYnatWuL7du3q9TzNgDEiBEjlNryap87d65Se94Vb+vXr1dqX7Fihahdu7YwMjISlpaWokuXLorL6t/sV7VqVWFkZCSqVasmwsLCVK7YEEKIrKws8eOPPwpPT08hl8uFubm5cHd3F0OHDhWXL19W9HtzHVetWiVat24tbG1thZGRkeK9k3d1V34ePHgghg0bJuzt7YWBgYFwcXERISEhKuNX3fYS4tW4U3fl3ZvWrl0r3N3dhaGhodL7SBvvw7ddLfXRRx+pzPPN7f6+4/Xhw4ciICBAlC1bVpiamoq2bduK48ePCwBKV7W8zaxZs4Srq6swNjYWHh4e4rffflN7hZEm23/btm2Kcens7CxmzZqldp75mThxogAgnJycVP7WCCHEsWPHRJMmTYSpqakoX768GDRokPj777/fui1fp+49euvWLdG9e3dhbm4uLCwsRPfu3cWxY8dU5pfXr1y5csLCwkK0a9dO/PPPP2q3w8KFC4Wbm5vQ19dXmo+6996LFy/E+PHjhYuLizA0NBT29vZi+PDh4tGjR0r93vZ3tqB/d4pzPTX5HCsomRCv7YMmwqvzGFxcXPDll19izpw5ui6HiIrIH3/8gc8//xxHjx5VuSqOqDT74A9L0f+5desWrl27hrlz50JPTw+jRo3SdUlEpCVr167F7du3UatWLejp6eH48eOYO3cuWrRowWBDksNwQworVqzA9OnT4erqisjISDg6Ouq6JCLSEgsLC6xbtw4zZszAs2fPYG9vD39/f8yYMUPXpRFpHQ9LERERkaTo9FLwQ4cOoVOnTnBwcIBMJlO5zFOd2NhY1KtXD3K5HJUqVcKyZcuKvlAiIiIqNXQabp49ewZPT08sWrSoQP2vX7+ODh06oHnz5oiPj8fEiRMRHByMjRs3FnGlREREVFqUmMNSMpkMmzdvRteuXd/aZ/z48di2bZvSvSaGDRuGs2fPIi4urhiqJCIiopKuVJ1QHBcXB29vb6U2Hx8frFy5EllZWTA0NFSZJiMjQ+krwnNzc/Hw4UNYW1ur/Sp1IiIiKnmEEHjy5AkcHBze+cWTpSrcpKSkqNzMy9bWFtnZ2bh//77amwOGhoZi2rRpxVUiERERFaGkpKR3fuN8qQo3gOqN6/KOqr1tL0xISAjGjh2rePz48WM4OzsjKSlJ5e6+REREVDKlp6fDycnpnTfnBUpZuLGzs1O5K2lqaioMDAzU3ocDeHUzLnV3bS5TpgzDDRERUSlTkFNKStVdwZs0aaJyM629e/eifv36as+3ISIiog+PTsPN06dPcebMGZw5cwbAq0u9z5w5g8TERACvDin5+fkp+g8bNgw3b97E2LFjcfHiRYSFhWHlypX46quvdFE+ERERlUA6PSx16tQptG7dWvE479yYAQMGICIiAsnJyYqgAwBubm7YuXMnxowZg8WLF8PBwQE///wzunfvXuy1ExERUclUYr7nprikp6fD0tISjx8/5jk3REREpYQmn9+l6pwbIiIiondhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSdF5uFmyZAnc3Nwgl8tRr149HD58ON/+kZGR8PT0hKmpKezt7REQEIAHDx4UU7VERERU0uk03ERFRWH06NGYNGkS4uPj0bx5c7Rv3x6JiYlq+x85cgR+fn4IDAzEhQsXsH79epw8eRKDBg0q5sqJiIiopNJpuJk/fz4CAwMxaNAgeHh4YOHChXBycsLSpUvV9j9+/DhcXV0RHBwMNzc3NGvWDEOHDsWpU6eKuXIiIiIqqXQWbjIzM3H69Gl4e3srtXt7e+PYsWNqp/Hy8sKtW7ewc+dOCCFw9+5dbNiwAb6+vm9dTkZGBtLT05V+iIiISLp0Fm7u37+PnJwc2NraKrXb2toiJSVF7TReXl6IjIxE7969YWRkBDs7O5QtWxa//PLLW5cTGhoKS0tLxY+Tk5NW14OIiIhKFp2fUCyTyZQeCyFU2vIkJCQgODgY3333HU6fPo3du3fj+vXrGDZs2FvnHxISgsePHyt+kpKStFo/ERERlSwGulqwjY0N9PX1VfbSpKamquzNyRMaGoqmTZvi66+/BgDUrl0bZmZmaN68OWbMmAF7e3uVaYyNjWFsbKz9FSAiIqISSWd7boyMjFCvXj1ER0crtUdHR8PLy0vtNM+fP4eennLJ+vr6AF7t8SEiIiLS6WGpsWPHYsWKFQgLC8PFixcxZswYJCYmKg4zhYSEwM/PT9G/U6dO2LRpE5YuXYpr167h6NGjCA4ORsOGDeHg4KCr1SAiIqISRGeHpQCgd+/eePDgAaZPn47k5GTUrFkTO3fuhIuLCwAgOTlZ6Ttv/P398eTJEyxatAjjxo1D2bJl0aZNG8yePVtXq0BEREQljEx8YMdz0tPTYWlpicePH6NMmTK6LoeIiIgKQJPPb51fLUVERESkTQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkGui5Aalwn7NB1CTp1Y5avrkv44HEMcgwSfegYbki7plrqugLdm/pY1xV82DgGOQbpg8fDUkRERCQpDDdEREQkKTwsRUREWsXzvnjel64x3BAREWkTz/vS+XlfPCxFREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREksJwQ0RERJLCcENERESSwnBDREREkqLzcLNkyRK4ublBLpejXr16OHz4cL79MzIyMGnSJLi4uMDY2BiVK1dGWFhYMVVLREREJZ2BLhceFRWF0aNHY8mSJWjatCl+/fVXtG/fHgkJCXB2dlY7Ta9evXD37l2sXLkSVapUQWpqKrKzs4u5ciIiIiqpdBpu5s+fj8DAQAwaNAgAsHDhQuzZswdLly5FaGioSv/du3cjNjYW165dg5WVFQDA1dW1OEsmIiKiEk5nh6UyMzNx+vRpeHt7K7V7e3vj2LFjaqfZtm0b6tevjzlz5sDR0RHVqlXDV199hRcvXrx1ORkZGUhPT1f6ISIiIunS2Z6b+/fvIycnB7a2tkrttra2SElJUTvNtWvXcOTIEcjlcmzevBn3799HUFAQHj58+NbzbkJDQzFt2jSt109EREQlk85PKJbJZEqPhRAqbXlyc3Mhk8kQGRmJhg0bokOHDpg/fz4iIiLeuvcmJCQEjx8/VvwkJSVpfR2IiIio5NDZnhsbGxvo6+ur7KVJTU1V2ZuTx97eHo6OjrC0tFS0eXh4QAiBW7duoWrVqirTGBsbw9jYWLvFExERUYmlsz03RkZGqFevHqKjo5Xao6Oj4eXlpXaapk2b4s6dO3j69Kmi7b///oOenh4qVqxYpPUSERFR6aDTw1Jjx47FihUrEBYWhosXL2LMmDFITEzEsGHDALw6pOTn56fo369fP1hbWyMgIAAJCQk4dOgQvv76awwcOBAmJia6Wg0iIiIqQXR6KXjv3r3x4MEDTJ8+HcnJyahZsyZ27twJFxcXAEBycjISExMV/c3NzREdHY0vv/wS9evXh7W1NXr16oUZM2boahWIiIiohNFpuAGAoKAgBAUFqX0uIiJCpc3d3V3lUBYRERFRHp1fLUVERESkTRqHG1dXV0yfPl3pcBERERFRSaFxuBk3bhy2bt2KSpUqoW3btli3bh0yMjKKojYiIiIijWkcbr788kucPn0ap0+fRo0aNRAcHAx7e3uMHDkSf//9d1HUSERERFRghT7nxtPTEz/99BNu376NKVOmYMWKFWjQoAE8PT0RFhYGIYQ26yQiIiIqkEJfLZWVlYXNmzcjPDwc0dHRaNy4MQIDA3Hnzh1MmjQJ+/btwx9//KHNWomIiIjeSeNw8/fffyM8PBxr166Fvr4++vfvjwULFsDd3V3Rx9vbGy1atNBqoUREREQFoXG4adCgAdq2bYulS5eia9euMDQ0VOlTo0YN9OnTRysFEhEREWlC43Bz7do1xTcIv42ZmRnCw8MLXRQRERFRYWl8QnFqaipOnDih0n7ixAmcOnVKK0URERERFZbG4WbEiBFISkpSab99+zZGjBihlaKIiIiICkvjcJOQkIC6deuqtH/88cdISEjQSlFEREREhaVxuDE2Nsbdu3dV2pOTk2FgoPP7cBIREdEHTuNw07ZtW4SEhODx48eKtrS0NEycOBFt27bVanFEREREmtJ4V8u8efPQokULuLi44OOPPwYAnDlzBra2tvj999+1XiARERGRJjQON46Ojjh37hwiIyNx9uxZmJiYICAgAH379lX7nTdERERExalQJ8mYmZlhyJAh2q6FiIiI6L0V+gzghIQEJCYmIjMzU6m9c+fO710UERERUWEV6huKu3XrhvPnz0Mmkynu/i2TyQAAOTk52q2QiIiISAMaXy01atQouLm54e7duzA1NcWFCxdw6NAh1K9fHzExMUVQIhEREVHBabznJi4uDgcOHED58uWhp6cHPT09NGvWDKGhoQgODkZ8fHxR1ElERERUIBrvucnJyYG5uTkAwMbGBnfu3AEAuLi44NKlS9qtjoiIiEhDGu+5qVmzJs6dO4dKlSqhUaNGmDNnDoyMjLB8+XJUqlSpKGokIiIiKjCNw823336LZ8+eAQBmzJiBjh07onnz5rC2tkZUVJTWCyQiIiLShMbhxsfHR/F7pUqVkJCQgIcPH6JcuXKKK6aIiIiIdEWjc26ys7NhYGCAf/75R6ndysqKwYaIiIhKBI3CjYGBAVxcXPhdNkRERFRiaXy11LfffouQkBA8fPiwKOohIiIiei8an3Pz888/48qVK3BwcICLiwvMzMyUnv/777+1VhwRERGRpjQON127di2CMoiIiIi0Q+NwM2XKlKKog4iIiEgrND7nhoiIiKgk03jPjZ6eXr6XffNKKiIiItIljcPN5s2blR5nZWUhPj4eq1atwrRp07RWGBEREVFhaBxuunTpotLWo0cPfPTRR4iKikJgYKBWCiMiIiIqDK2dc9OoUSPs27dPW7MjIiIiKhSthJsXL17gl19+QcWKFbUxOyIiIqJC0/iw1Js3yBRC4MmTJzA1NcWaNWu0WhwRERGRpjQONwsWLFAKN3p6eihfvjwaNWqEcuXKabU4IiIiIk1pHG78/f2LoAwiIiIi7dD4nJvw8HCsX79epX39+vVYtWqVVooiIiIiKiyNw82sWbNgY2Oj0l6hQgXMnDlTK0URERERFZbG4ebmzZtwc3NTaXdxcUFiYqJWiiIiIiIqLI3DTYUKFXDu3DmV9rNnz8La2lorRREREREVlsbhpk+fPggODsbBgweRk5ODnJwcHDhwAKNGjUKfPn2KokYiIiKiAtP4aqkZM2bg5s2b+OSTT2Bg8Gry3Nxc+Pn58ZwbIiIi0jmNw42RkRGioqIwY8YMnDlzBiYmJqhVqxZcXFyKoj4iIiIijWgcbvJUrVoVVatW1WYtRERERO9N43NuevTogVmzZqm0z507Fz179tRKUURERESFpXG4iY2Nha+vr0p7u3btcOjQIa0URURERFRYGoebp0+fwsjISKXd0NAQ6enpWimKiIiIqLA0Djc1a9ZEVFSUSvu6detQo0YNrRRFREREVFgan1A8efJkdO/eHVevXkWbNm0AAPv378cff/yBDRs2aL1AIiIiIk1oHG46d+6MLVu2YObMmdiwYQNMTEzg6emJAwcOoEyZMkVRIxEREVGBFepScF9fX8VJxWlpaYiMjMTo0aNx9uxZ5OTkaLVAIiIiIk1ofM5NngMHDuCLL76Ag4MDFi1ahA4dOuDUqVParI2IiIhIYxrtubl16xYiIiIQFhaGZ8+eoVevXsjKysLGjRt5MjERERGVCAXec9OhQwfUqFEDCQkJ+OWXX3Dnzh388ssvRVkbERERkcYKvOdm7969CA4OxvDhw3nbBSIiIiqxCrzn5vDhw3jy5Anq16+PRo0aYdGiRbh3715R1kZERESksQKHmyZNmuC3335DcnIyhg4dinXr1sHR0RG5ubmIjo7GkydPirJOIiIiogLR+GopU1NTDBw4EEeOHMH58+cxbtw4zJo1CxUqVEDnzp2LokYiIiKiAiv0peAAUL16dcyZMwe3bt3C2rVrtVUTERERUaG9V7jJo6+vj65du2Lbtm0aT7tkyRK4ublBLpejXr16OHz4cIGmO3r0KAwMDFCnTh2Nl0lERETSpZVwU1hRUVEYPXo0Jk2ahPj4eDRv3hzt27dHYmJivtM9fvwYfn5++OSTT4qpUiIiIiotdBpu5s+fj8DAQAwaNAgeHh5YuHAhnJycsHTp0nynGzp0KPr164cmTZoUU6VERERUWugs3GRmZuL06dPw9vZWavf29saxY8feOl14eDiuXr2KKVOmFGg5GRkZSE9PV/ohIiIi6dJZuLl//z5ycnJga2ur1G5ra4uUlBS101y+fBkTJkxAZGQkDAwK9v2DoaGhsLS0VPw4OTm9d+1ERERUcun0sBQAyGQypcdCCJU2AMjJyUG/fv0wbdo0VKtWrcDzDwkJwePHjxU/SUlJ710zERERlVwa3ThTm2xsbKCvr6+ylyY1NVVlbw4APHnyBKdOnUJ8fDxGjhwJAMjNzYUQAgYGBti7dy/atGmjMp2xsTGMjY2LZiWIiIioxNHZnhsjIyPUq1cP0dHRSu3R0dHw8vJS6V+mTBmcP38eZ86cUfwMGzYM1atXx5kzZ9CoUaPiKp2IiIhKMJ3tuQGAsWPHon///qhfvz6aNGmC5cuXIzExEcOGDQPw6pDS7du3sXr1aujp6aFmzZpK01eoUAFyuVylnYiIiD5cOg03vXv3xoMHDzB9+nQkJyejZs2a2LlzJ1xcXAAAycnJ7/zOGyIiIqLX6TTcAEBQUBCCgoLUPhcREZHvtFOnTsXUqVO1XxQRERGVWjq/WoqIiIhImxhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSGG6IiIhIUhhuiIiISFIYboiIiEhSdB5ulixZAjc3N8jlctSrVw+HDx9+a99Nmzahbdu2KF++PMqUKYMmTZpgz549xVgtERERlXQ6DTdRUVEYPXo0Jk2ahPj4eDRv3hzt27dHYmKi2v6HDh1C27ZtsXPnTpw+fRqtW7dGp06dEB8fX8yVExERUUml03Azf/58BAYGYtCgQfDw8MDChQvh5OSEpUuXqu2/cOFCfPPNN2jQoAGqVq2KmTNnomrVqti+fXsxV05EREQllc7CTWZmJk6fPg1vb2+ldm9vbxw7dqxA88jNzcWTJ09gZWX11j4ZGRlIT09X+iEiIiLp0lm4uX//PnJycmBra6vUbmtri5SUlALNY968eXj27Bl69er11j6hoaGwtLRU/Dg5Ob1X3URERFSy6fyEYplMpvRYCKHSps7atWsxdepUREVFoUKFCm/tFxISgsePHyt+kpKS3rtmIiIiKrkMdLVgGxsb6Ovrq+ylSU1NVdmb86aoqCgEBgZi/fr1+PTTT/Pta2xsDGNj4/eul4iIiEoHne25MTIyQr169RAdHa3UHh0dDS8vr7dOt3btWvj7++OPP/6Ar69vUZdJREREpYzO9twAwNixY9G/f3/Ur18fTZo0wfLly5GYmIhhw4YBeHVI6fbt21i9ejWAV8HGz88PP/30Exo3bqzY62NiYgJLS0udrQcRERGVHDoNN71798aDBw8wffp0JCcno2bNmti5cydcXFwAAMnJyUrfefPrr78iOzsbI0aMwIgRIxTtAwYMQERERHGXT0RERCWQTsMNAAQFBSEoKEjtc28GlpiYmKIviIiIiEo1nV8tRURERKRNDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpOr8rOBERUWHJ9WUoZ6IHPZmuK/k/L42ddF2C7r18WajJDA0Noa+v/96LZ7ghIqJSqaqVIYbXLwsLuT6AkpNursvm6boE3bt+vVCTyWQyVKxYEebm5u+1eIYbIiIqdeT6MgyvXxYutuVgYFoGkJWccOPGEz6ACm4aTyKEwL1793Dr1i1UrVr1vfbgMNwQEVGpU85EDxZyfRiYloHM0FjX5SiRl6RjZLoilxdqsvLly+PGjRvIysp6r3DDfElERKXOq/wgK1F7bOj9ybT0ejLcEBERkaQw3BAREZVirXoMxujv5uq6jBKF59wQEZFkdF50tFiXt21k0wL3lTnWzff5AT07IWLhNI1r2PTbjzA05Mf567g1iIiIikFy/F7F71Hb9uK7H5fh0qFNijYTufKJ0VlZWTA0NHznfK3KWWqvSIngYSkiIqJiYFfBRvFjaWEOmez/2l5mZKKsR0v8uW0vWvUYDHmlxlizaScePExD36AQVKzXDqaVvVDrk15Yu2W30nzfPCzl2sgXM39eiYFjp8KiWjM4N+iA5Ws2FvPa6hbDDRERUQkxfubPCB7YBxdjNsKnpRdeZmSiXm0P/LXqJ/xz4E8M+fwz9A+ejBN/n893PvN+XYP6tWsgfs8fCBrQE8NDQvHvlcJ9sV5pxMNSREREJcToQf3wWYdPlNq+Guan+P3LgX2w++AxrP9rHxrVrfXW+XRo0xRB/r0AAONH+GPBb5GIOXYa7lU0/3K90ojhhoiIqISo71lD6XFOTg5mLQpH1Pa9uJ18DxmZmcjIzIKZqUm+86ldo6rid5lMBrvy1kh98LBIai6JGG6IiIhKCDMT5dAy79ffseC3P7Bw2jjUcq8KM1M5Rk/5EZlZWfnOx9BA+eNdJpMhNzdX6/WWVAw3REREJdThE/Ho4tMSX3T3BQDk5ubi8vUkeFT9MA4vFRZPKCYiIiqhqrg6IfrQCRw7eRYXL1/D0PE/IOXeA12XVeJxzw0REVEJNXn0YFxPugOfz0fA1ESOIZ9/hq4+rfD4yVNdl1aiMdwQEZFkaPKNwbrk37sz/Ht3Vjx2dXKAuP23Sj+rcpbYEjY/33nFbPhN6fGNEztU+pyJXlfISksnHpYiIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIiSWG4ISIiIklhuCEiIiJJYbghIiIqJVr1GIzR381VPHZt5IuFv0XmO43MsS627D743svW1nyKA2+/QEREklF7hUuxLu/coJsF7ttpwCi8eJmBfVHLVJ6LO3UWXl0CcHp3JOrW8ijwPE/uXAMzU3mB+xfE1HnLsGV3jMotG5Lj96KcZRmtLquocM8NERFRMQjs2xUHjp7EzVt3VJ4Li9qGOh9V1yjYAEB563IwNTHRVon5sqtgA2Njo2JZ1vtiuCEiIioGHT9tjgo2Voj4c7tS+/MXLxC1bS+6tmuFvkEhqFivHUwre6HWJ72wdsvufOf55mGpy9cS0eKzQMgrNUaNVt0Rfei4yjTjf/gJ1Zp1hWllL1Rq0gmT5yxBVlYWACAiahumzV+Oswn/QeZYFzLHuoiI2gZA9bDU+YuX0abnEJhUbgLrj1pjyDff4+mz54rn/f390bVrV/z444+wt7eHtbU1RowYoVhWUeJhKSIiomJgYGAAvx6+iPhzO74bMwQymQwAsH77PmRmZWFQ325Yu3U3xgf5o4yFGXbsP4L+wZNRydkRjerWeuf8c3Nz8dngr2BTriyOb1uF9KdPMXrKPJV+FmZmiFgwDQ525XH+4mUM/mYGLMxN8U2QP3p39sY/l65id8wx7Fu3FABgaWGuMo/nL16g3Rcj0bhuLZzc8TtS7z/EoK+/x8hJsxGxcJqi38GDB2Fvb4+DBw/iypUr6N27N+rUqYPBgwcXdjMWCPfcEBERFZOBfbrgRtIdxBw7pWgLi9qKz9q3gaN9BXw1zA91alZHJZeK+HJgH/i0bIL1f+0r0Lz3HT6Bi5ev4/efv0edmtXRonE9zJwwQqXft6MHwauBJ1ydHNDJuyXGDf0Cf26PBgCYmMhhbmYCA3192FWwgV0FG5iYqJ7TE7lpF168zMDqn75HTfcqaNOsIRbNGI/fN+7A3XsPFP3KlSuHRYsWwd3dHR07doSvry/279+v6WbTGPfcEBERFRP3Km7wqu+JsHVb0bppA1y9kYTDJ+Kx948lyMnJwaxF4Yjavhe3k+8hIzMTGZlZMDMt2Dk1Fy9fh7OjHSo62CramtSrrdJvw1/7sHDFH7hyIwlPnz1Hdk4OypibabQeFy9fh6dHNaXamjbwRG5uLi5dvQFbz1dtH330EfT19RV97O3tcf78eY2WVRjcc0NERFSMAvt2wcadB5D+5CnCo7bBpaI9PmneEPN+/R0LfvsD3wwfgAN//ooze9fCp2VjZBbwHBUhhEpb3qGvPMdPn0OfoBC0b+2Fv1b9hPg9azHpy8ACL+P1Zb0xa7XLNDQ0VHkuNzdXo2UVBsMNERFRMerVyRv6+nr4Y/NurFr/FwJ6d4ZMJsPhE/Ho4tMSX3T3hedH1VDJpSIuX08q8HxrVKuExNspuJNyT9EWd/qcUp+jJ8/CpaI9Jo0ahPqeNVC1kjNu3k5W6mNkaIicdwSQGtUq4UzCf3j2/IXSvPX09FCtUvFejq8Oww0REVExMjczRe/O3pg4exHu3L0H/16dAABVXJ0QfegEjp08i4uXr2Ho+B+Q8tr5K+/yafNGqF7ZBX6jJuPshf9w+MTfmDR7sVKfKm5OSLydgnVb9+DqjST8vHItNu9S/mI+VycHXE+8jTP/XML9h4+QkZGpsqzPP2sPubERBoz6Dv/8ewUHj57El5PnoH93X9iWty7EVtEuhhsiIqJiFtinCx6lpePT5g3h7GgPAJg8ejDq1nKHz+cj0KrHENiVt0ZXn1YFnqeenh42r5iHjMxMNOzYH4O++h4/jFc+obiLTyuMGdwPIyfNRh3vvjh26iwmjx6k1Kd7h0/QrpUXWvcagvK1PlF7ObqpiQn2RC7Gw7THaODbHz2GfINPmjXAoh/Ga74xioBMqDtIJ2Hp6emwtLTE48ePUaaM9r9p0XXCDq3PszS5Ie+n6xJ0b+pjnS6eY5Bj8EMYg44W+pjaugIqOFSEzKBkfbFcbb3rui5B9xw+LtRkL1++xPXr1+Hm5ga5XPkqLU0+v7nnhoiIiCSF4YaIiIgkheGGiIiIJIXhhoiIiCSF4YaIiIgkheGGiIhKndwP6jrfD4e2LuBmuCEiolLnWZZATm4uRE62rkshLcrMfPWFga/fj6oweONMIiIqdZ5k5OL83ZewNHsE07L6eOuNjnTgpR53K+HlS40nyc3Nxb1792BqagoDg/eLJww3RERU6ggAa88/gYulIcq9eAmg5IQbI9m9d3eSumeF+yJDPT09ODs7q9zwU1MMN0REVCo9fJmLifvvw8ZUH/ol6CSL/cZf6boE3Rt5qlCTGRkZQU/v/V9MnYebJUuWYO7cuUhOTsZHH32EhQsXonnz5m/tHxsbi7Fjx+LChQtwcHDAN998g2HDhhVjxUREVFJkCyDlWY6uy1Aizyr4nbwl641bJxQ3nWbdqKgojB49GpMmTUJ8fDyaN2+O9u3bIzExUW3/69evo0OHDmjevDni4+MxceJEBAcHY+PGjcVcOREREZVUOg038+fPR2BgIAYNGgQPDw8sXLgQTk5OWLp0qdr+y5Ytg7OzMxYuXAgPDw8MGjQIAwcOxI8//ljMlRMREVFJpbNwk5mZidOnT8Pb21up3dvbG8eOHVM7TVxcnEp/Hx8fnDp1CllZWUVWKxEREZUeOjvn5v79+8jJyYGtra1Su62tLVJSUtROk5KSorZ/dnY27t+/D3t7e5VpMjIykJGRoXj8+PFjAK9unV4UcjOeF8l8S4t0GS+BRBGNrYLiGOQY5BjULY5BFMkYzPvcLsgX/en8hOI3L/cSQuR7CZi6/ura84SGhmLatGkq7U5OTpqWSgVgqesCSoJZ3Aq6xK0PjkEd49ZHkY7BJ0+ewNIy//nrLNzY2NhAX19fZS9Namqqyt6ZPHZ2dmr7GxgYwNraWu00ISEhGDt2rOJxbm4uHj58CGtr6/e+jp6Upaenw8nJCUlJSShTpoyuy6EPEMcg6RrHYNERQuDJkydwcHB4Z1+dhRsjIyPUq1cP0dHR6Natm6I9OjoaXbp0UTtNkyZNsH37dqW2vXv3on79+jA0NFQ7jbGxMYyNjZXaypYt+37FU77KlCnDNzXpFMcg6RrHYNF41x6bPDq9Wmrs2LFYsWIFwsLCcPHiRYwZMwaJiYmK760JCQmBn5+fov+wYcNw8+ZNjB07FhcvXkRYWBhWrlyJr77iFyYRERHRKzo956Z379548OABpk+fjuTkZNSsWRM7d+6Ei4sLACA5OVnpO2/c3Nywc+dOjBkzBosXL4aDgwN+/vlndO/eXVerQERERCWMTGjr/uL0wcvIyEBoaChCQkJUDgUSFQeOQdI1jsGSgeGGiIiIJKUE3WqMiIiI6P0x3BAREZGkMNwQERGRpDDcULFzdXXFwoULFY9lMhm2bNnyzulu3LgBmUyGM2fOAABiYmIgk8mQlpZWJHVS6VbYcaYtrVq1wujRo4tteVQ6aHtcRkREFOi725YvXw4nJyfo6ekpLV+qGG5KudTUVAwdOhTOzs4wNjaGnZ0dfHx8EBsbCxsbG8yYMUPtdKGhobCxsUFmZiYiIiIgk8ng4eGh0u/PP/+ETCaDq6trEa+J5ry8vJCcnFzgL3WiwvuQxxmVXByXBZOeno6RI0di/PjxuH37NoYMGaLrkoocw00p1717d5w9exarVq3Cf//9h23btqFVq1Z4+vQpvvjiC0RERKi9yVh4eDj69+8PIyMjAICZmRlSU1MRFxen1C8sLAzOzs7Fsi6aMjIygp2dHW+jUQw+5HFGJdeHPi6zsrIK1C8xMRFZWVnw9fWFvb09TE1Ni7gy3WO4KcXS0tJw5MgRzJ49G61bt4aLiwsaNmyIkJAQ+Pr6IjAwEFevXsWhQ4eUpjt8+DAuX76MwMBARZuBgQH69euHsLAwRdutW7cQExODfv36Fbimq1evokuXLrC1tYW5uTkaNGiAffv2vf/KqvHmYam83bN79uyBh4cHzM3N0a5dOyQnJxfJ8j8UH9o4yzv8uW7dOnh5eUEul+Ojjz5CTEyMUr/Y2Fg0bNgQxsbGsLe3x4QJE5Cdna12ntOnT0etWrVU2uvVq4fvvvtOK3V/aD7Ucfnnn3+iVatWkMvlWLNmjeL5LVu2oFq1apDL5Wjbti2SkpIAvPq7mDf2KlWqBJlMhhs3bmilppKM4aYUMzc3h7m5ObZs2YKMjAyV52vVqoUGDRogPDxcqT0sLAwNGzZEzZo1ldoDAwMRFRWF58+fA3j1pmjXrt1bb2SqztOnT9GhQwfs27cP8fHx8PHxQadOnZS+abooPX/+HD/++CN+//13HDp0CImJibw9x3v6UMfZ119/jXHjxiE+Ph5eXl7o3LkzHjx4AAC4ffs2OnTogAYNGuDs2bNYunQpVq5c+dbDIAMHDkRCQgJOnjypaDt37hzi4+Ph7++vtZo/JB/quBw/fjyCg4Nx8eJF+Pj4AHj1d++HH37AqlWrcPToUaSnp6NPnz4AXt0JIC9g/e9//0NycjKcnJy0Vk+JJahU27BhgyhXrpyQy+XCy8tLhISEiLNnzyqeX7p0qTAzMxNPnjwRQgjx5MkTYWZmJn799VdFn/DwcGFpaSmEEKJOnTpi1apVIjc3V1SuXFls3bpVLFiwQLi4uBS6xho1aohffvlF8djFxUUsWLBA8RiA2Lx58zvnc/36dQFAxMfHCyGEOHjwoAAgHj16pFgPAOLKlSuKaRYvXixsbW0LXTu98iGOs1mzZinasrKyRMWKFcXs2bOFEEJMnDhRVK9eXeTm5ir6LF68WJibm4ucnBwhhBAtW7YUo0aNUjzfvn17MXz4cMXj0aNHi1atWmm6mvSaD3FcLly4UKk97+/e8ePHFW0XL14UAMSJEyeEEELEx8cLAOL69euFXo/ShntuSrnu3bvjzp072LZtG3x8fBATE4O6desiIiICANC3b1/k5uYiKioKABAVFQUhhCLVv2ngwIEIDw9HbGys4r8QTTx79gzffPMNatSogbJly8Lc3Bz//vtvse25MTU1ReXKlRWP7e3tkZqaWizLlrIPcZw1adJE8buBgQHq16+PixcvAgAuXryIJk2aKJ3v1bRpUzx9+hS3bt1SO7/Bgwdj7dq1ePnyJbKyshAZGYmBAwdqrd4P0Yc4LuvXr6/Sljc+87i7u6Ns2bKK8fohYriRgLxjrN999x2OHTsGf39/TJkyBcCr28P36NFDsWs2PDwcPXr0QJkyZdTO6/PPP8fx48cxdepU+Pn5wcBAs3urfv3119i4cSN++OEHHD58GGfOnEGtWrWQmZn5fitZQIaGhkqPZTKZ2hMKSXMcZ1CEGSGEyonseePsbSe4d+rUCcbGxti8eTO2b9+OjIwM3vRXCz60cWlmZqa2Xd24+5AvtmC4kaAaNWrg2bNniseBgYE4evQo/vrrLxw9elTpRLo3WVlZoXPnzoiNjS3Uf5WHDx+Gv78/unXrhlq1asHOzu6DOHntQyT1cXb8+HHF79nZ2Th9+jTc3d0BvFr3Y8eOKQXnY8eOwcLCAo6OjmrnZ2BggAEDBiA8PBzh4eHo06fPB3HVSnGT+rhUJzs7G6dOnVI8vnTpEtLS0hTj9UPEcFOKPXjwAG3atMGaNWtw7tw5XL9+HevXr8ecOXPQpUsXRb+WLVuiSpUq8PPzQ5UqVdCiRYt85xsREYH79+8X6o1RpUoVbNq0CWfOnMHZs2fRr18/5ObmajwfKjk+1HG2ePFibN68Gf/++y9GjBiBR48eKT7wgoKCkJSUhC+//BL//vsvtm7diilTpmDs2LHQ03v7n9VBgwbhwIED2LVrFw9JvacPdVyqY2hoiC+//BInTpzA33//jYCAADRu3BgNGzYs8mWXVJrtc6MSxdzcHI0aNcKCBQtw9epVZGVlwcnJCYMHD8bEiROV+g4cOBATJ07E119//c75mpiYwMTEpFA1LViwAAMHDoSXlxdsbGwwfvx4pKenF2peVDJ8qONs1qxZmD17NuLj41G5cmVs3boVNjY2AABHR0fs3LkTX3/9NTw9PWFlZYXAwEB8++23+c6zatWq8PLywoMHD9CoUSOt1vuh+VDHpTqmpqYYP348+vXrh1u3bqFZs2ZKl7V/iGSCJyQQESncuHEDbm5uiI+PR506dbQ6byEE3N3dMXToUIwdO1ar8yai/8M9N0RExSA1NRW///47bt++jYCAAF2XQyRpPOeGNPLRRx8pvjzrzZ/IyMj3mvfMmTPfOu/27dtraQ2oNJDiOLO1tcWsWbOwfPlylCtXrsiWQ0VHiuNSqnhYijRy8+bNt97PxNbWFhYWFoWe98OHD/Hw4UO1z5mYmLz1KhSSHo4zKok4LksPhhsiIiKSFB6WIiIiIklhuCEiIiJJYbghIiIiSWG4ISLJiYmJgUwmQ1paWoGncXV1xcKFC4usJiIqPgw3RFTs/P39IZPJMGzYMJXngoKCIJPJ4O/vX/yFEZEkMNwQkU44OTlh3bp1ePHihaLt5cuXWLt2LZydnXVYGRGVdgw3RKQTdevWhbOzMzZt2qRo27RpE5ycnPDxxx8r2jIyMhAcHIwKFSpALpejWbNmOHnypNK8du7ciWrVqsHExAStW7dWeyfmY8eOoUWLFjAxMYGTkxOCg4OV7h79pqlTp8LZ2RnGxsZwcHBAcHDw+680ERULhhsi0pmAgACEh4crHoeFhancLfubb77Bxo0bsWrVKvz999+oUqUKfHx8FF94lpSUhM8++wwdOnTAmTNnMGjQIEyYMEFpHufPn4ePjw8+++wznDt3DlFRUThy5AhGjhyptq4NGzZgwYIF+PXXX3H58mVs2bIFtWrV0vLaE1GREURExWzAgAGiS5cu4t69e8LY2Fhcv35d3LhxQ8jlcnHv3j3RpUsXMWDAAPH06VNhaGgoIiMjFdNmZmYKBwcHMWfOHCGEECEhIcLDw0Pk5uYq+owfP14AEI8ePRJCCNG/f38xZMgQpRoOHz4s9PT0xIsXL4QQQri4uIgFCxYIIYSYN2+eqFatmsjMzCzCrUBERYV7bohIZ2xsbODr64tVq1YhPDwcvr6+sLGxUTx/9epVZGVloWnTpoo2Q0NDNGzYEBcvXgQAXLx4EY0bN4ZMJlP0adKkidJyTp8+jYiICKX79fj4+CA3NxfXr19Xqatnz5548eIFKlWqhMGDB2Pz5s3Izs7W9uoTURHhXcGJSKcGDhyoODy0ePFipefE/787zOvBJa89r00U4A4yubm5GDp0qNrzZtSdvOzk5IRLly4hOjoa+/btQ1BQEObOnYvY2FgYGhoWbMWISGe454aIdKpdu3bIzMxEZmYmfHx8lJ6rUqUKjIyMcOTIEUVbVlYWTp06BQ8PDwBAjRo1cPz4caXp3nxct25dXLhwAVWqVFH5MTIyUluXiYkJOnfujJ9//hkxMTGIi4vD+fPntbHKRFTEuOeGiHRKX19fcYhJX19f6TkzMzMMHz4cX3/9NaysrODs7Iw5c+bg+fPnCAwMBAAMGzYM8+bNw9ixYzF06FDFIajXjR8/Ho0bN8aIESMwePBgmJmZ4eLFi4iOjsYvv/yiUlNERARycnLQqFEjmJqa4vfff4eJiQlcXFyKZiMQkVZxzw0R6VyZMmVQpkwZtc/NmjUL3bt3R//+/VG3bl1cuXIFe/bsQbly5QC8Oqy0ceNGbN++HZ6enli2bBlmzpypNI/atWsjNjYWly9fRvPmzfHxxx9j8uTJsLe3V7vMsmXL4rfffkPTpk1Ru3Zt7N+/H9u3b4e1tbV2V5yIioRMFOSANREREVEpwT03REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKQw3REREJCkMN0RERCQpDDdEREQkKf8Pc/ubeXsmy2gAAAAASUVORK5CYII="
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_list_names = [\"SVM_all_lin\", \"SVM_all_poly\", \"SVM_all_rbf\"]\n",
    "model_list = [fitted_svm_model_lin, fitted_svm_model_poly, fitted_svm_model_rbf]\n",
    "utils.compare_models(model_list_names, model_list, X_train, y_train, X_valid, y_valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Watermark\n",
    "\n",
    "This should be the last section of your notebook, since it watermarks all your environment.\n",
    "\n",
    "When commiting this notebook, remember to restart the kernel, rerun the notebook and run this cell last, to watermark the environment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Python implementation: CPython\n",
      "Python version       : 3.8.13\n",
      "IPython version      : 7.31.1\n",
      "\n",
      "Compiler    : GCC 10.3.0\n",
      "OS          : Linux\n",
      "Release     : 5.19.9-200.fc36.x86_64\n",
      "Machine     : x86_64\n",
      "Processor   : x86_64\n",
      "CPU cores   : 12\n",
      "Architecture: 64bit\n",
      "\n",
      "Git hash: ea49738d78d4c40bd08db659ddf9953d814507c8\n",
      "\n",
      "Git branch: splits_analysis\n",
      "\n",
      "pandas  : 1.4.3\n",
      "sys     : 3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 06:04:10) \n",
      "[GCC 10.3.0]\n",
      "workshop: 0.0.1\n",
      "logging : 0.5.1.2\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%watermark -gb -iv -m -v"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.13 ('ml_workshop')",
   "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": "3.8.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "8e4ad55d24ca2819cba0036a3b0296ce849220695bf3b2f3682ab4bcc2c4fdd7"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}