07_regression.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>\n",
       "    \n",
       "    @import url('http://fonts.googleapis.com/css?family=Source+Code+Pro');\n",
       "    \n",
       "    @import url('http://fonts.googleapis.com/css?family=Kameron');\n",
       "    @import url('http://fonts.googleapis.com/css?family=Crimson+Text');\n",
       "    \n",
       "    @import url('http://fonts.googleapis.com/css?family=Lato');\n",
       "    @import url('http://fonts.googleapis.com/css?family=Source+Sans+Pro');\n",
       "    \n",
       "    @import url('http://fonts.googleapis.com/css?family=Lora'); \n",
       "\n",
       "    \n",
       "    body {\n",
       "        font-family: 'Lora', Consolas, sans-serif;\n",
       "       \n",
       "        -webkit-print-color-adjust: exact important !;\n",
       "        \n",
       "      \n",
       "       \n",
       "    }\n",
       "    \n",
       "    .alert-block {\n",
       "        width: 95%;\n",
       "        margin: auto;\n",
       "    }\n",
       "    \n",
       "    .rendered_html code\n",
       "    {\n",
       "        color: black;\n",
       "        background: #eaf0ff;\n",
       "        background: #f5f5f5; \n",
       "        padding: 1pt;\n",
       "        font-family:  'Source Code Pro', Consolas, monocco, monospace;\n",
       "    }\n",
       "    \n",
       "    p {\n",
       "      line-height: 140%;\n",
       "    }\n",
       "    \n",
       "    strong code {\n",
       "        background: red;\n",
       "    }\n",
       "    \n",
       "    .rendered_html strong code\n",
       "    {\n",
       "        background: #f5f5f5;\n",
       "    }\n",
       "    \n",
       "    .CodeMirror pre {\n",
       "    font-family: 'Source Code Pro', monocco, Consolas, monocco, monospace;\n",
       "    }\n",
       "    \n",
       "    .cm-s-ipython span.cm-keyword {\n",
       "        font-weight: normal;\n",
       "     }\n",
       "     \n",
       "     strong {\n",
       "         background: #f5f5f5;\n",
       "         margin-top: 4pt;\n",
       "         margin-bottom: 4pt;\n",
       "         padding: 2pt;\n",
       "         border: 0.5px solid #a0a0a0;\n",
       "         font-weight: bold;\n",
       "         color: darkred;\n",
       "     }\n",
       "     \n",
       "    \n",
       "    div #notebook {\n",
       "        # font-size: 10pt; \n",
       "        line-height: 145%;\n",
       "        }\n",
       "        \n",
       "    li {\n",
       "        line-height: 145%;\n",
       "    }\n",
       "\n",
       "    div.output_area pre {\n",
       "        background: #fff9d8 !important;\n",
       "        padding: 5pt;\n",
       "       \n",
       "       -webkit-print-color-adjust: exact; \n",
       "        \n",
       "    }\n",
       " \n",
       "    \n",
       " \n",
       "    h1, h2, h3, h4 {\n",
       "        font-family: Kameron, arial;\n",
       "\n",
       "\n",
       "    }\n",
       "    \n",
       "    div#maintoolbar {display: none !important;}\n",
       "</style>\n",
       "    <script>\n",
       "IPython.OutputArea.prototype._should_scroll = function(lines) {\n",
       "        return false;\n",
       "}\n",
       "    </script>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# IGNORE THIS CELL WHICH CUSTOMIZES LAYOUT AND STYLING OF THE NOTEBOOK !\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore', category=FutureWarning)\n",
    "warnings.filterwarnings('ignore', category=DeprecationWarning)\n",
    "warnings.filterwarnings = lambda *a, **kw: None\n",
    "from IPython.core.display import HTML; HTML(open(\"custom.html\", \"r\").read())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chapter 7: Regression\n",
    "\n",
    "Regression belongs like classification to the field of supervised learning. \n",
    "\n",
    "<div class=\"alert alert-block alert-warning\">\n",
    "<i class=\"fa fa-info-circle\"></i>&nbsp; \n",
    "<strong>Regression predicts numerical values</strong> \n",
    "in contrast to classification which predicts categories.\n",
    "</div>\n",
    "\n",
    "<img src=\"./images/30416v.jpg\" title=\"made at imgflip.com\" width=35%/>\n",
    "\n",
    "<div class=\"alert alert-block alert-warning\">\n",
    "<i class=\"fa fa-info-circle\"></i>&nbsp; \n",
    "    Other differences are:\n",
    "\n",
    "* Accuracy is measured differently\n",
    "\n",
    "\n",
    "* Other algorithms\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example: Salmon weight\n",
    "\n",
    "The dataset `data/salmon.csv` holds measurements of `circumference`, `length` and `weight` for  `atlantic` and `sockeye` salmons.\n",
    "\n",
    "Our goal is to predict `weight` based on the other three features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>circumference</th>\n",
       "      <th>length</th>\n",
       "      <th>kind</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19.0</td>\n",
       "      <td>59.5</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>5.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>4.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.0</td>\n",
       "      <td>75.5</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.5</td>\n",
       "      <td>89.0</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>15.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23.5</td>\n",
       "      <td>63.0</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>5.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   circumference  length      kind  weight\n",
       "0           19.0    59.5   sockeye     5.1\n",
       "1           18.0    53.0   sockeye     4.1\n",
       "2           28.0    75.5  atlantic     9.1\n",
       "3           33.5    89.0  atlantic    15.6\n",
       "4           23.5    63.0  atlantic     5.2"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv(\"data/salmon.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>circumference</th>\n",
       "      <th>length</th>\n",
       "      <th>kind</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>24.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>6.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>18.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>18.0</td>\n",
       "      <td>59.5</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>4.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>20.0</td>\n",
       "      <td>64.5</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>4.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>23.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>7.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    circumference  length      kind  weight\n",
       "95           24.0    76.0  atlantic     6.7\n",
       "96           18.5    67.0   sockeye     5.0\n",
       "97           18.0    59.5   sockeye     4.7\n",
       "98           20.0    64.5  atlantic     4.1\n",
       "99           23.0    75.0   sockeye     7.2"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us inspect the features and their distributions:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 619.85x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 526,
       "width": 612
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set(style=\"ticks\")\n",
    "\n",
    "sns.pairplot(df, hue=\"kind\", diag_kind=\"hist\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In contrast to our previous examples, our data set contains a non-numerical text column `kind`.\n",
    "\n",
    "<code>sklearn.preprocessing.OneHotEncoder</code> is a preprocessor which encodes text values to according flags:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "features = df.iloc[:, :-1]\n",
    "values = df.iloc[:, -1]\n",
    "\n",
    "# needs 2d data structure, features.iloc[2] has dimension 1\n",
    "encoder = OneHotEncoder(sparse=False)\n",
    "one_hot = encoder.fit_transform(features.iloc[:, 2: 3]) \n",
    "\n",
    "one_hot[:5, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So the one-hot encoder computes two columns with exclusive flags 0 and 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>circumference</th>\n",
       "      <th>length</th>\n",
       "      <th>kind</th>\n",
       "      <th>is_atlantic</th>\n",
       "      <th>is_sockeye</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19.0</td>\n",
       "      <td>59.5</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>sockeye</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.0</td>\n",
       "      <td>75.5</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33.5</td>\n",
       "      <td>89.0</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23.5</td>\n",
       "      <td>63.0</td>\n",
       "      <td>atlantic</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   circumference  length      kind  is_atlantic  is_sockeye\n",
       "0           19.0    59.5   sockeye          0.0         1.0\n",
       "1           18.0    53.0   sockeye          0.0         1.0\n",
       "2           28.0    75.5  atlantic          1.0         0.0\n",
       "3           33.5    89.0  atlantic          1.0         0.0\n",
       "4           23.5    63.0  atlantic          1.0         0.0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features[\"is_atlantic\"] = one_hot[:, 0]\n",
    "features[\"is_sockeye\"] = one_hot[:, 1]\n",
    "\n",
    "features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we remove the categorical column now:\n",
    "del features[\"kind\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we prepare the data for training and testing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "(features_train, features_test, \n",
    " values_train, \n",
    " values_test) = train_test_split(features, values, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Without further explanation we pick a regression algorithm, more about regrssion algorithms will be discussed later:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.kernel_ridge import KernelRidge\n",
    "kr = KernelRidge(alpha=.001, kernel=\"rbf\", gamma=.05)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-warning\">\n",
    "    <i class=\"fa fa-info-circle\"></i>&nbsp; Regression methods in <code>scikit-learn</code> also have <code>fit</code> and <code>predict</code> methods. Thus cross validation, pipelines and hyperparameter-optimization will be available.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "kr.fit(features_train, values_train)\n",
    "predicted = kr.predict(features_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us plot how good given and predicted values match on the training data set (sic !)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 269,
       "width": 724
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "def plot_fit_quality(values_test, predicted):\n",
    "\n",
    "   \n",
    "    \n",
    "    plt.figure(figsize=(12, 4))\n",
    "    plt.subplot(1, 2, 1)\n",
    "\n",
    "    x = np.arange(len(predicted))\n",
    "    plt.plot(x, predicted - values_test, 'o') \n",
    "\n",
    "    plt.plot([0, len(predicted)], [0, 0], \"k:\")\n",
    "    \n",
    "    max_diff = np.max(np.abs(predicted - values_test))\n",
    "    plt.ylim([-max_diff, max_diff])\n",
    "    \n",
    "    plt.ylabel(\"error\")\n",
    "    plt.xlabel(\"sample id\")\n",
    "\n",
    "    plt.subplot(1, 2, 2)\n",
    "\n",
    "    plt.plot(x, (predicted - values_test) / values_test, 'o') \n",
    "    plt.plot([0, len(predicted)], [0, 0], \"k:\")\n",
    "    plt.ylim([-.5, .5])\n",
    "      \n",
    "    plt.ylabel(\"relative error\")\n",
    "    plt.xlabel(\"sample id\")\n",
    "\n",
    "    \n",
    "plot_fit_quality(values_test, predicted)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For assessing the quality of the predictions of a regression method, we can use multiple methods which we will discuss later in this script.\n",
    "\n",
    "For our current example we compute the average absolute difference between given values $y_i$ and predicted values  $\\hat{y}_i$:\n",
    "\n",
    "$$\n",
    "\\frac{1}{n} \\left(\\, |y_1 - \\hat{y}_1| \\, + \\, |y_2 - \\hat{y}_2| \\, + \\, \\ldots \\,+ \\,|y_n - \\hat{y}_n| \\,\\right)\n",
    "$$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7122581321318665\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "error = np.sum(np.abs(predicted - values_test)) / len(values_test)\n",
    "print(error)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metrics / error measures"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When we used classification metrics (like accuracy, precision, recall, F1) high values indicated good classification performance. \n",
    "\n",
    "Most regression metrics turn this upside down. E.g. smaller values indicate a better regression model.\n",
    "\n",
    "The hyperparameter optimization functions from `scikit-learn` select configurations which yield a large score. To make regression functions work in this framework, we have to flip the sign of the error value to achieva a usable score.\n",
    "\n",
    "E.g.\n",
    "\n",
    "- an average absolute error of 0.1 is scored as -0.1\n",
    "- an average absolute error of 0.2 is scored as -0.2\n",
    "\n",
    "In this situation the first case would be prefered: higher score indicates lower error.\n",
    "   \n",
    "\n",
    "`scikit-learn` offers the following metrics for measuring regression quality:\n",
    "\n",
    "### 1. Mean absolute error\n",
    "\n",
    "This is the metric we used before. Taking absolute values before adding up the deviatons assures that deviations with different signs can not cancel out.\n",
    "\n",
    "<div class=\"alert alert-block alert-warning\">\n",
    "    <i class=\"fa fa-info-circle\"></i>&nbsp; <strong>mean absolute error</strong> is defined as \n",
    "\n",
    "\n",
    "$$\n",
    "\\frac{1}{n} \\left(\\, |y_1 - \\hat{y}_1| \\, + \\, |y_2 - \\hat{y}_2| \\, + \\, \\ldots \\,+ \\,|y_n - \\hat{y}_n| \\,\\right)\n",
    "$$\n",
    "\n",
    "\n",
    "</div>\n",
    "\n",
    "\n",
    "The name of the corresponding score in `scikit-learn` is `neg_mean_absolute_error`.\n",
    "\n",
    "\n",
    "### 2. Mean squared error\n",
    "\n",
    "Here we replace the absolute difference by its squared difference. Squaring also insures positive differeces.\n",
    "\n",
    "\n",
    "<div class=\"alert alert-block alert-warning\">\n",
    "    <i class=\"fa fa-info-circle\"></i>&nbsp; <strong>mean squared error</strong> is defined as \n",
    "\n",
    "\n",
    "\n",
    "$$\n",
    "\\frac{1}{n} \\left(\\, (y_1 - \\hat{y}_1)^2 \\, + \\, (y_2 - \\hat{y}_2)^2 \\, \\, \\ldots \\,+ \\,(y_n - \\hat{y}_n)^2 \\,\\right)\n",
    "$$\n",
    "\n",
    "\n",
    "</div>\n",
    "\n",
    "\n",
    "\n",
    "This measure is more sensitive to outliers: A few larger differences contribute more significantly to a larger mean squared error. The name of the corresponding score in `scikit-learn` is `neg_mean_squared_error`.\n",
    "\n",
    "\n",
    "### 3. Median absolute error\n",
    "\n",
    "Here we replace mean calculation by median. \n",
    "\n",
    "<div class=\"alert alert-block alert-warning\">\n",
    "    <i class=\"fa fa-info-circle\"></i>&nbsp; <strong>median absolute error</strong> is defined as \n",
    "\n",
    "\n",
    "\n",
    "$$\n",
    "\\text{median}\\left(\\,|y_1 - \\hat{y}_1|, \\,|y_2 - \\hat{y}_2|, \\,\\ldots, \\,|y_n - \\hat{y}_n| \\, \\right)\n",
    "$$\n",
    "\n",
    "\n",
    "</div>\n",
    "\n",
    "\n",
    "This measure is less sensitive to outliers than the metrics we discussed before: A few larger differences will not contribute significantly to a larger error value. The name of the corresponding score in `scikit-learn` is `neg_median_absolute_error`.\n",
    "\n",
    "### 4. Mean squared log error\n",
    "\n",
    "The formula for this metric can be found [here](https://scikit-learn.org/stable/modules/model_evaluation.html#mean-squared-log-error). \n",
    "\n",
    "This metric is recommended when your target values are distributed over a huge range of values, like popoluation numbers. \n",
    "The previous error metrics would put a larger weight on large target values. One could consider relative deviations to compensate such effects but relative deviations come with other problems like division by zero.\n",
    "\n",
    "\n",
    "The name is `neg_mean_squared_log_error`\n",
    "\n",
    "\n",
    "### 5. Explained variance and $r^2$-score\n",
    "\n",
    "Two other scores to mention are *explained variance* and $r^2$-score. For both larger values indicate better regression results.\n",
    "\n",
    "The formula for [r2 can be found here](https://scikit-learn.org/stable/modules/model_evaluation.html#r2-score), the score takes values in the range $0 .. 1$. The name within `scikit-learn` is `r2`.\n",
    "\n",
    "The formula for [explained variance](https://scikit-learn.org/stable/modules/model_evaluation.html#explained-variance-score), the score takes values up to $1$. The name within `scikit-learn` is `explained_variance`.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some algorithms from sklearn\n",
    "\n",
    "- `sklearn.linear_model.LinearRegression` is a linear regression method, which only works well for target values which can be described as a linear combination of feature values.\n",
    "\n",
    "\n",
    "- `sklearn.kernel_ridge.KernelRidge` is [documented here](https://scikit-learn.org/stable/modules/kernel_ridge.html#kernel-ridge). It combines the kernel trick from SVMs with classical least squares regression.\n",
    "\n",
    "\n",
    "- `sklearn.svm.SVR` is an extension of support vector classification concept to regression, [you find examples here](https://scikit-learn.org/stable/modules/svm.html#svm-regression)\n",
    "\n",
    "\n",
    "- `sklearn.neighbors.KNeighborsRegressor` extends the idea of nearest neighbour classification to regression: Search for similar data points in the learning data set and compute the predicted value from the values from the neighbourhood, e.g. by averaging or by linear interpolation. [Documentation is available here](https://scikit-learn.org/stable/modules/neighbors.html#regression)\n",
    "\n",
    "\n",
    "- `sklearn.tree.DecisionTreeRegressor` expands the concept of decision trees to regression [is documented here](https://scikit-learn.org/stable/modules/tree.html#regression).\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A full pipeline\n",
    "\n",
    "Let us now try to find a good regressor using `scikit-learn`s hyper-parameter tuning:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross val score: -0.7568859642342642\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 269,
       "width": 733
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler, PolynomialFeatures\n",
    "from sklearn.kernel_ridge import KernelRidge\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "\n",
    "def eval_regression(p, features, values):\n",
    "    score = cross_val_score(p, features, values, scoring=\"neg_median_absolute_error\", cv=4).mean()\n",
    "    print(\"cross val score:\", score)\n",
    "  \n",
    "    predicted = p.fit(features_train, values_train).predict(features_test)\n",
    "    plot_fit_quality(values_test, predicted)\n",
    "\n",
    "    \n",
    "p = make_pipeline(PolynomialFeatures(2), PCA(2), LinearRegression())\n",
    "eval_regression(p, features, values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "p = make_pipeline(PolynomialFeatures(), PCA(), LinearRegression())\n",
    "\n",
    "param_grid = {'polynomialfeatures__degree': range(3, 6),\n",
    "              'pca__n_components': range(3, 11),\n",
    "             }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'pca__n_components': 10, 'polynomialfeatures__degree': 3}\n",
      "cross val score: -0.22752607270361858\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 269,
       "width": 733
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "search = GridSearchCV(p, param_grid, scoring=\"neg_median_absolute_error\", cv=4, n_jobs=4)\n",
    "\n",
    "search.fit(features, values)\n",
    "\n",
    "\n",
    "print(search.best_params_)\n",
    "eval_regression(search, features, values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exercise section\n",
    "\n",
    "- Play with the examples above and try different algorithms, metrics and pipelines.\n",
    "\n",
    "\n",
    "### Optional exercise: Timeseries prediction\n",
    "\n",
    "The file  `data/sales.csv` holds sales data of a swiss sportsshop selling skiing equipment. The time axis is in units of months, starting with January.\n",
    "\n",
    "- Load the data and plot it\n",
    "\n",
    "For time series prediction we want to construct a method which uses regression to predict $y_{n}$ based on previous  `W` values $y_{n - W}, y_{n - W + 1} \\ldots y_{n - 1}$. `W` is a parameter called *window size*.\n",
    "\n",
    "E.r. For window size $W$ = 3 we create a feature matrix `X`\n",
    "\n",
    "    y_0, y_1, y_2\n",
    "    y_1, y_2, y_3\n",
    "    y_2, y_3, y_4\n",
    "    ...\n",
    "    \n",
    "and a right hand side vector `y`\n",
    "\n",
    "    y_3\n",
    "    y_4\n",
    "    y_5\n",
    "    ...\n",
    "    \n",
    "- Write a function which takes window size `W` and the vector `y` to construct the feature matrix `X` and the right hand side `y`\n",
    "\n",
    "- Find optimal configurations (use the `r2` metric) for the regressors `Lasso`, `SVR` and `KernelRidge(kernel=\"rbf\")` and several window sizes. Plot `y` and predicted values as time series and scatter plots.\n",
    "\n",
    "Now we want to check long term predictions starting at month `N` given window size `W`.\n",
    "\n",
    "We start with a data window of `W` values $(y_N, \\ldots, y_{N+W-1})$ and update  this window by iteratively\n",
    "\n",
    "1. predict first value after given time window.\n",
    "2. update data window: shift by one time step, thus: discard the first entry and append the previously predicted value\n",
    "3. continue with step 1.\n",
    "\n",
    "\n",
    "We demonstrate this for `W=4` and `N=100` and name the predicted values `z`:\n",
    "\n",
    "<div style=\"margin-left: 1em;\">\n",
    "<code>\n",
    "predict <span style=\"font-weight: bold;\">z_104</span> from y_100, y_101, y_102, y_103\n",
    "predict <span style=\"font-weight: bold;\">z_105</span> from y_101, y_102, y_103, <span style=\"font-weight: bold;\">z_104</span>\n",
    "predict <span style=\"font-weight: bold;\">z_106</span> from y_102, y_103, <span style=\"font-weight: bold;\">z_104, z_105</span>\n",
    "predict <span style=\"font-weight: bold;\">z_107</span> from y_103, <span style=\"font-weight: bold;\">z_104, z_105, z_106</span>\n",
    "predict <span style=\"font-weight: bold;\">z_108</span> from <span style=\"font-weight: bold;\">z_104, z_105, z_106, z_107</span>\n",
    "...                                          \n",
    "</code> \n",
    "</div>\n",
    "\n",