{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Feature Engineering\n",
    "# Klassifikation\n",
    "# Regression\n",
    "## Einleitung\n",
    "## Klassische Methoden \n",
    "## ML Methoden\n",
    "## Vergleich über alle Modelle"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "## for scikit-learn 1.4.2, to silence warnings regarding physical cores\n",
    "import os\n",
    "os.environ['LOKY_MAX_CPU_COUNT'] = '4' ## depending on the hardware used"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('./')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-08T10:06:24.890328Z",
     "start_time": "2020-04-08T10:06:23.220148Z"
    }
   },
   "outputs": [],
   "source": [
    "## preparation\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns; sns.set()\n",
    "%matplotlib inline\n",
    "\n",
    "datapath = '../3_data'\n",
    "from os import chdir; chdir(datapath)\n",
    "\n",
    "from bfh_cas_pml import prep_data, prep_demo_data\n",
    "X_train, X_test, y_train, y_test = prep_data('melb_data_prep.csv', 'Price', seed = 1234)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-08T10:06:55.122481Z",
     "start_time": "2020-04-08T10:06:55.110322Z"
    }
   },
   "outputs": [],
   "source": [
    "## import classes & functions\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import AdaBoostRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.ensemble import HistGradientBoostingRegressor\n",
    "from catboost import CatBoostRegressor\n",
    "from lightgbm.sklearn import LGBMRegressor\n",
    "\n",
    "from sklearn.metrics import r2_score, mean_squared_error\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-08T10:06:55.137655Z",
     "start_time": "2020-04-08T10:06:55.126100Z"
    }
   },
   "outputs": [],
   "source": [
    "## define models\n",
    "models = [\n",
    "    LinearRegression(),\n",
    "    Ridge(),\n",
    "    Lasso(),\n",
    "    KNeighborsRegressor(),\n",
    "    DecisionTreeRegressor(max_depth=8),\n",
    "    RandomForestRegressor(random_state = 1234),\n",
    "    AdaBoostRegressor(DecisionTreeRegressor(max_depth=15), learning_rate=0.07),\n",
    "    GradientBoostingRegressor(),\n",
    "    HistGradientBoostingRegressor(),\n",
    "    CatBoostRegressor(logging_level='Silent'),\n",
    "    LGBMRegressor()\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-08T10:06:55.153478Z",
     "start_time": "2020-04-08T10:06:55.141155Z"
    }
   },
   "outputs": [],
   "source": [
    "## for collection of results\n",
    "names = []\n",
    "scores = []\n",
    "used_times = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-08T10:07:07.654300Z",
     "start_time": "2020-04-08T10:06:55.155608Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regressor                        r2         time\n",
      "=================================================\n",
      "LinearRegression                 0.5601     0.004\n",
      "Ridge                            0.5601     0.002\n",
      "Lasso                            0.5601     0.012\n",
      "KNeighborsRegressor              0.4497     0.051\n",
      "DecisionTreeRegressor            0.6713     0.033\n",
      "RandomForestRegressor            0.7778     3.800\n",
      "AdaBoostRegressor                0.7652     2.297\n",
      "GradientBoostingRegressor        0.7250     1.137\n",
      "HistGradientBoostingRegressor    0.7831     0.404\n",
      "CatBoostRegressor                0.8003     0.945\n",
      "[LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.000331 seconds.\n",
      "You can set `force_col_wise=true` to remove the overhead.\n",
      "[LightGBM] [Info] Total Bins 1630\n",
      "[LightGBM] [Info] Number of data points in the train set: 12262, number of used features: 23\n",
      "[LightGBM] [Info] Start training from score 1055902.695237\n",
      "LGBMRegressor                    0.7882     0.411\n"
     ]
    }
   ],
   "source": [
    "## print output header\n",
    "print('Regressor                        r2         time')\n",
    "print('=================================================')\n",
    "\n",
    "## iteration over all defined models\n",
    "for model in models:\n",
    "    start_time = time.time()        ## start timer\n",
    "    model.fit(X_train, y_train)     ## train\n",
    "    y_pred = model.predict(X_test)  ## predict\n",
    "    name = model.__class__.__name__ ## pick model name (f. Output)\n",
    "    score = r2_score(y_test, y_pred)             ## calculate r2\n",
    "    used_time = time.time() - start_time    ## stop timer\n",
    "    names.append(name)\n",
    "    scores.append(score)\n",
    "    used_times.append(used_time)\n",
    "    print('%-30s   %0.4f    %6.3f' %(name, score, used_time)) ## console output  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## vizualisation, ordered by decreasing accuracy\n",
    "\n",
    "results = pd.DataFrame({\n",
    "    'models' : names,\n",
    "    'scores' : scores,\n",
    "    'used_times' : used_times}\n",
    ").sort_values(by='scores', ascending=False)\n",
    "#print(results)\n",
    "\n",
    "order = results.sort_values('scores', ascending=False).models\n",
    "\n",
    "sns.set(font_scale=1.5) \n",
    "f, axes = plt.subplots(nrows=1, ncols=2, figsize=(10, 4), sharex=False)\n",
    "\n",
    "ax = sns.barplot(data=results, x=scores, y=names, color='lightblue', order=order, ax=axes[0])\n",
    "ax.set_xlabel('r2_score')\n",
    "#ax.set(xlim=(0.5, 1))\n",
    "\n",
    "ax = sns.barplot(x=used_times, y=names, color='darksalmon', order=order, ax=axes[1])\n",
    "ax.set_xlabel('time')\n",
    "ax.set_yticklabels('')\n",
    "ax.set_ylabel('');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best_regressor : CatBoostRegressor\n",
      "best_score     : 0.8003\n",
      "best_time      : 0.9447\n"
     ]
    }
   ],
   "source": [
    "## show results of best regressor\n",
    "print('best_regressor :', names[scores.index(max(scores))])\n",
    "print('best_score     : %0.4f' % (max(scores)))\n",
    "print('best_time      : %6.4f' % (used_times[scores.index(max(scores))]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            names    scores\n",
      "9               CatBoostRegressor  0.800349\n",
      "10                  LGBMRegressor  0.788166\n",
      "8   HistGradientBoostingRegressor  0.783064\n"
     ]
    }
   ],
   "source": [
    "## top 3!!\n",
    "print(pd.DataFrame({\n",
    "    'names'  : names,\n",
    "    'scores' : scores\n",
    "}).sort_values(by='scores', ascending=False).head(3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fazit:**  vgl. Präsentation"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "teaching",
   "language": "python",
   "name": "teaching"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.0"
  },
  "toc": {
   "base_numbering": "3.4",
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "3.4 Regression - Modellvergleiche",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "toc-autonumbering": 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()) "
    }
   },
   "position": {
    "height": "321.85px",
    "left": "787px",
    "right": "20px",
    "top": "115px",
    "width": "350px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}