cas-pml/SL/aufgaben/template/4_WS/Loesungen/WS 04 Loesung.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "## import libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns; sns.set()\n",
    "\n",
    "## load data\n",
    "datapath = '../../3_data'\n",
    "from os import chdir; chdir(datapath)\n",
    "data = pd.read_csv('bank_data_prep.csv')\n",
    "\n",
    "## features - target - split\n",
    "X = data.drop('y', axis=1)\n",
    "y = data['y']\n",
    "\n",
    "## test - train - split\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test, = train_test_split(X,\n",
    "                                                     y,\n",
    "                                                     train_size=2 / 3,\n",
    "                                                     random_state=1234)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "## standardiz features (lead: train data)\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler().fit(X_train)\n",
    "X_train_scaled = scaler.transform(X_train)\n",
    "X_test_scaled = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\joblib\\externals\\loky\\backend\\context.py:110: UserWarning: Could not find the number of physical cores for the following reason:\n",
      "found 0 physical cores < 1\n",
      "Returning the number of logical cores instead. You can silence this warning by setting LOKY_MAX_CPU_COUNT to the number of cores you want to use.\n",
      "  warnings.warn(\n",
      "  File \"C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\joblib\\externals\\loky\\backend\\context.py\", line 217, in _count_physical_cores\n",
      "    raise ValueError(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 0.7003346516580469 0.7550958320657134\n",
      "2 0.7027684818983876 0.7319744447824764\n",
      "3 0.7377547916032857 0.7715241861880134\n",
      "4 0.7423182233039245 0.7590508062062671\n",
      "5 0.7493154852449042 0.7912990568907818\n",
      "6 0.7505324003650745 0.7770003042287801\n",
      "7 0.7520535442652875 0.7982963188317614\n",
      "8 0.7517493154852449 0.7855187100699726\n",
      "9 0.7554000608457561 0.7952540310313355\n",
      "10 0.7532704593854579 0.7885609978703986\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Tune über n_neighbors\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "model = KNeighborsClassifier()\n",
    "\n",
    "params = range(1, 11)\n",
    "scores = []  ## scores ohne Standardisieren\n",
    "scores_sc = []  ## scores mit Standardisieren\n",
    "\n",
    "for param in params:\n",
    "    model.set_params(n_neighbors=param)\n",
    "    model.fit(X_train, y_train)\n",
    "    score = model.score(X_test, y_test)\n",
    "    scores.append(score)\n",
    "\n",
    "    model.fit(X_train_scaled, y_train)\n",
    "    score_sc = model.score(X_test_scaled, y_test)\n",
    "    scores_sc.append(score_sc)\n",
    "\n",
    "    print(param, score, score_sc)  ## zum Vergolgen der Fortschritte\n",
    "\n",
    "fig = sns.lineplot(x=params, y=scores, label='scaled: no')\n",
    "fig = sns.lineplot(x=params, y=scores_sc, label='scaled: yes')\n",
    "plt.xlabel(\"n_neighbors\")\n",
    "plt.ylabel(\"accuracy\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 0.7614846364466078 0.7992090051718893\n",
      "2 0.7520535442652875 0.7982963188317614\n",
      "3 0.7523577730453301 0.7940371159111652\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "C:\\Users\\werne\\anaconda3\\Lib\\site-packages\\seaborn\\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Tune über p\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "model = KNeighborsClassifier()\n",
    "\n",
    "params = range(1, 4) ## dasselbe wie [1, 2, 3]\n",
    "scores = []  ## scores ohne Standardisieren\n",
    "scores_sc = []  ## scores mit Standardisieren\n",
    "\n",
    "for param in params:\n",
    "    model.set_params(n_neighbors=7, p=param)\n",
    "    model.fit(X_train, y_train)\n",
    "    score = model.score(X_test, y_test)\n",
    "    scores.append(score)\n",
    "\n",
    "    model.fit(X_train_scaled, y_train)\n",
    "    score_sc = model.score(X_test_scaled, y_test)\n",
    "    scores_sc.append(score_sc)\n",
    "\n",
    "    print(param, score, score_sc)  ## zum Verfolgen der Fortschritte\n",
    "\n",
    "fig = sns.lineplot(x=params, y=scores, label='scaled: no')\n",
    "fig = sns.lineplot(x=params, y=scores_sc, label='scaled: yes')\n",
    "\n",
    "plt.xlabel(\"p\")\n",
    "plt.ylabel(\"accuracy\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fazit**:\n",
    "* standardisieren bringt bei diesem Learner tatsächlich eine markante Verbesserung der Performance\n",
    "* p=1 ist leicht besser als p=2\n",
    "* ab p = 3 nimmt der Zeitbedarf markant zu"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  },
  "toc": {
   "base_numbering": "",
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "WS 07 Klassifikation - KNeighborsClassifier - solution",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "205.2px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}