{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"https://colab.research.google.com/github/visiont3lab/tecnologie_data_science/blob/master/book/docs/unsupervised/pca.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "b6UneNHRBg-J"
   },
   "source": [
    "# PCA: Regressione"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "PI1VdpMi_XUk"
   },
   "source": [
    "# Cosa andremo a fare oggi?\n",
    "  * Creazione di un dataset usando la libreria pandas. Come passare da numpy array to pandas dataframe\n",
    "  * Scikit-Learn datasets Spiegazione, nozioni su come scaricarli applicazione della regressione lineare a un dataset di scikit-learn (diabetes dataset)\n",
    "  * Cosa significa correlazione? Quando e perchè si utilizza\n",
    "  * Principal Component Analysis (PCA) dimensionality reduction\n",
    "    * Riduzione del numero di input a un numero fisso (es. 2)\n",
    "    * Come facciamo a riddure il numero degli input senza ridurre il contenuto di informazioni del nostro dataset iniziale? \n",
    "  * Esempio pratico: Regressione lineare applicata a un dataset con e senza PCA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sBCwXUEdGcjZ"
   },
   "source": [
    "# Creazione di un dataset usando la libreria pandas\n",
    "In questo paragrafo andiamo a vedere come creare una classe dataset usando sia numpy che pandas in modo che possiamo facilmente applicare le regressioni studiate a questo. Inoltre vedremo come utilizare i dataset di scikit-learn.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 480
    },
    "colab_type": "code",
    "id": "JTfex6C9G_49",
    "outputId": "8ef9d553-03c8-4179-a08b-0a53dca2da6b"
   },
   "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>età</th>\n",
       "      <th>ore mensili</th>\n",
       "      <th>esperienza</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>150</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>21</td>\n",
       "      <td>135</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22</td>\n",
       "      <td>160</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>158</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>26</td>\n",
       "      <td>155</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>33</td>\n",
       "      <td>120</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>29</td>\n",
       "      <td>130</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   età  ore mensili  esperienza\n",
       "0   19          150           6\n",
       "1   21          135           8\n",
       "2   22          160           5\n",
       "3   23          158           7\n",
       "4   26          155           7\n",
       "5   33          120          10\n",
       "6   29          130           9"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "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>solario</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1640</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   solario\n",
       "0     1100\n",
       "1     1150\n",
       "2     1155\n",
       "3     1170\n",
       "4     1200\n",
       "5     1750\n",
       "6     1640"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "Y = Salario al mese in euro\n",
    "X1 = Età del lavoratore\n",
    "X2 = Numero di ore mensili di lavoro\n",
    "X3 = Indice di esperienza da 1 a 10\n",
    "Equazione = Y = w0 + w1*X1 + w2*X2 + w3*X3\n",
    "1) Marco:    Y=1100   X1=19   X2=150   X3=6    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*19 + w2*150 + w3*6\n",
    "2) Daniele:  Y=1150   X1=21   X2=135   X3=8    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*21 + w2*135 + w3*8\n",
    "3) Davide:   Y=1155   X1=22   X2=160   X3=5    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*22 + w2*160 + w3*5 \n",
    "4) Marta:    Y=1170   X1=23   X2=158   X3=7    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*23 + w2*158 + w3*7\n",
    "6) Alessia:  Y=1200   X1=26   X2=155   X3=7    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*26 + w2*155 + w3*7\n",
    "9) Stella:   Y=1750   X1=33   X2=120   X3=10   Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*33 + w2*120 + w3*10\n",
    "10) Chiara   Y=1640   X1=29   X2=130   x3=9    Y= w0*1 + w1*X1 + w2*X2 + w3*X3 --> 1100 = w0*1 + w1*29 + w2*130 + w3*9\n",
    "'''\n",
    "\n",
    "import pandas as pd\n",
    "class Dataset():\n",
    "    def __init__(self):\n",
    "      self.X = np.array([[19,150,6],[21,135,8], [22,160,5], [23,158,7], [26,155,7], [33,120,10],[29,130,9]])\n",
    "      self.Y = np.array([[1100],[1150],[1155],[1170],[1200],[1750],[1640]])\n",
    "    def createPandasDataset(self):\n",
    "        df_X = pd.DataFrame(data=self.X, columns =[\"età\",\"ore mensili\", \"esperienza\"])\n",
    "        df_Y = pd.DataFrame(data=self.Y, columns =[\"salario\"])\n",
    "        return df_X, df_Y\n",
    "\n",
    "myDataset = Dataset()\n",
    "df_X, df_Y = myDataset.createPandasDataset()\n",
    "display(df_X)\n",
    "display(df_Y)\n",
    "\n",
    "      "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "9sT8LqWII21y"
   },
   "source": [
    "# Scikit-Learn datasets spiegazione e  nozioni su come scaricarli.\n",
    "\n",
    "Andremo a vedere quali dataset sono disponibili in scikit-learn, come scaricarli e capirne il contenuto.\n",
    "\n",
    "* [Sklearn dataset page](https://scikit-learn.org/stable/datasets/index.html#datasets)\n",
    "\n",
    "I dataset disponibili sono i seguenti: \n",
    "  * Regressione:\n",
    "      * [Boston houses price dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_boston.html#sklearn.datasets.load_boston)\n",
    "      * [Diabetes dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_diabetes.html#sklearn.datasets.load_diabetes)\n",
    "      * [Linnerrud Dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_linnerud.html#sklearn.datasets.load_linnerud)\n",
    "  * Classificazione:\n",
    "    * [Iris plant dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_iris.html#sklearn.datasets.load_iris)\n",
    "    * [Optical recognition of handwritten digits dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_digits.html#sklearn.datasets.load_digits)  \n",
    "    * [Wine Recognition dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_wine.html#sklearn.datasets.load_wine)\n",
    "    * [Breast cancer wisconsin (diagnostic) dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_breast_cancer.html#sklearn.datasets.load_breast_cancer)\n",
    "\n",
    "Implementazione di una classe capace di scaricare i dati da scikit-learn, visualizzarli e analizzarli."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "TcBjZAXdHnvP",
    "outputId": "4d4d341e-1e61-44fd-bf32-2b54d68ba602"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['data', 'target', 'DESCR', 'feature_names', 'data_filename', 'target_filename', 'target_names'])\n",
      "------------------------------------------\n",
      "data [[ 0.03807591  0.05068012  0.06169621 ... -0.00259226  0.01990842\n",
      "  -0.01764613]\n",
      " [-0.00188202 -0.04464164 -0.05147406 ... -0.03949338 -0.06832974\n",
      "  -0.09220405]\n",
      " [ 0.08529891  0.05068012  0.04445121 ... -0.00259226  0.00286377\n",
      "  -0.02593034]\n",
      " ...\n",
      " [ 0.04170844  0.05068012 -0.01590626 ... -0.01107952 -0.04687948\n",
      "   0.01549073]\n",
      " [-0.04547248 -0.04464164  0.03906215 ...  0.02655962  0.04452837\n",
      "  -0.02593034]\n",
      " [-0.04547248 -0.04464164 -0.0730303  ... -0.03949338 -0.00421986\n",
      "   0.00306441]]\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "target [151.  75. 141. 206. 135.  97. 138.  63. 110. 310. 101.  69. 179. 185.\n",
      " 118. 171. 166. 144.  97. 168.  68.  49.  68. 245. 184. 202. 137.  85.\n",
      " 131. 283. 129.  59. 341.  87.  65. 102. 265. 276. 252.  90. 100.  55.\n",
      "  61.  92. 259.  53. 190. 142.  75. 142. 155. 225.  59. 104. 182. 128.\n",
      "  52.  37. 170. 170.  61. 144.  52. 128.  71. 163. 150.  97. 160. 178.\n",
      "  48. 270. 202. 111.  85.  42. 170. 200. 252. 113. 143.  51.  52. 210.\n",
      "  65. 141.  55. 134.  42. 111.  98. 164.  48.  96.  90. 162. 150. 279.\n",
      "  92.  83. 128. 102. 302. 198.  95.  53. 134. 144. 232.  81. 104.  59.\n",
      " 246. 297. 258. 229. 275. 281. 179. 200. 200. 173. 180.  84. 121. 161.\n",
      "  99. 109. 115. 268. 274. 158. 107.  83. 103. 272.  85. 280. 336. 281.\n",
      " 118. 317. 235.  60. 174. 259. 178. 128.  96. 126. 288.  88. 292.  71.\n",
      " 197. 186.  25.  84.  96. 195.  53. 217. 172. 131. 214.  59.  70. 220.\n",
      " 268. 152.  47.  74. 295. 101. 151. 127. 237. 225.  81. 151. 107.  64.\n",
      " 138. 185. 265. 101. 137. 143. 141.  79. 292. 178.  91. 116.  86. 122.\n",
      "  72. 129. 142.  90. 158.  39. 196. 222. 277.  99. 196. 202. 155.  77.\n",
      " 191.  70.  73.  49.  65. 263. 248. 296. 214. 185.  78.  93. 252. 150.\n",
      "  77. 208.  77. 108. 160.  53. 220. 154. 259.  90. 246. 124.  67.  72.\n",
      " 257. 262. 275. 177.  71.  47. 187. 125.  78.  51. 258. 215. 303. 243.\n",
      "  91. 150. 310. 153. 346.  63.  89.  50.  39. 103. 308. 116. 145.  74.\n",
      "  45. 115. 264.  87. 202. 127. 182. 241.  66.  94. 283.  64. 102. 200.\n",
      " 265.  94. 230. 181. 156. 233.  60. 219.  80.  68. 332. 248.  84. 200.\n",
      "  55.  85.  89.  31. 129.  83. 275.  65. 198. 236. 253. 124.  44. 172.\n",
      " 114. 142. 109. 180. 144. 163. 147.  97. 220. 190. 109. 191. 122. 230.\n",
      " 242. 248. 249. 192. 131. 237.  78. 135. 244. 199. 270. 164.  72.  96.\n",
      " 306.  91. 214.  95. 216. 263. 178. 113. 200. 139. 139.  88. 148.  88.\n",
      " 243.  71.  77. 109. 272.  60.  54. 221.  90. 311. 281. 182. 321.  58.\n",
      " 262. 206. 233. 242. 123. 167.  63. 197.  71. 168. 140. 217. 121. 235.\n",
      " 245.  40.  52. 104. 132.  88.  69. 219.  72. 201. 110.  51. 277.  63.\n",
      " 118.  69. 273. 258.  43. 198. 242. 232. 175.  93. 168. 275. 293. 281.\n",
      "  72. 140. 189. 181. 209. 136. 261. 113. 131. 174. 257.  55.  84.  42.\n",
      " 146. 212. 233.  91. 111. 152. 120.  67. 310.  94. 183.  66. 173.  72.\n",
      "  49.  64.  48. 178. 104. 132. 220.  57.]\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "DESCR .. _diabetes_dataset:\n",
      "\n",
      "Diabetes dataset\n",
      "----------------\n",
      "\n",
      "Ten baseline variables, age, sex, body mass index, average blood\n",
      "pressure, and six blood serum measurements were obtained for each of n =\n",
      "442 diabetes patients, as well as the response of interest, a\n",
      "quantitative measure of disease progression one year after baseline.\n",
      "\n",
      "**Data Set Characteristics:**\n",
      "\n",
      "  :Number of Instances: 442\n",
      "\n",
      "  :Number of Attributes: First 10 columns are numeric predictive values\n",
      "\n",
      "  :Target: Column 11 is a quantitative measure of disease progression one year after baseline\n",
      "\n",
      "  :Attribute Information:\n",
      "      - Age\n",
      "      - Sex\n",
      "      - Body mass index\n",
      "      - Average blood pressure\n",
      "      - S1\n",
      "      - S2\n",
      "      - S3\n",
      "      - S4\n",
      "      - S5\n",
      "      - S6\n",
      "\n",
      "Note: Each of these 10 feature variables have been mean centered and scaled by the standard deviation times `n_samples` (i.e. the sum of squares of each column totals 1).\n",
      "\n",
      "Source URL:\n",
      "https://www4.stat.ncsu.edu/~boos/var.select/diabetes.html\n",
      "\n",
      "For more information see:\n",
      "Bradley Efron, Trevor Hastie, Iain Johnstone and Robert Tibshirani (2004) \"Least Angle Regression,\" Annals of Statistics (with discussion), 407-499.\n",
      "(https://web.stanford.edu/~hastie/Papers/LARS/LeastAngle_2002.pdf)\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "feature_names ['age', 'sex', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "data_filename /usr/local/lib/python3.6/dist-packages/sklearn/datasets/data/diabetes_data.csv.gz\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "target_filename /usr/local/lib/python3.6/dist-packages/sklearn/datasets/data/diabetes_target.csv.gz\n",
      "------------------------------------------\n",
      "------------------------------------------\n",
      "target_names ['Desease-Progression']\n",
      "------------------------------------------\n",
      "Dataset Parameters:  dict_keys(['data', 'target', 'DESCR', 'feature_names', 'data_filename', 'target_filename', 'target_names'])\n",
      "Feature Names:  ['age', 'sex', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n",
      "Output Names:  ['Desease-Progression']\n",
      "Input X Shape:  (442, 10)\n",
      "Output Y Shape:  (442,)\n"
     ]
    },
    {
     "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>bmi</th>\n",
       "      <th>bp</th>\n",
       "      <th>s1</th>\n",
       "      <th>s2</th>\n",
       "      <th>s3</th>\n",
       "      <th>s4</th>\n",
       "      <th>s5</th>\n",
       "      <th>s6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.038076</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.061696</td>\n",
       "      <td>0.021872</td>\n",
       "      <td>-0.044223</td>\n",
       "      <td>-0.034821</td>\n",
       "      <td>-0.043401</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.019908</td>\n",
       "      <td>-0.017646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.001882</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.051474</td>\n",
       "      <td>-0.026328</td>\n",
       "      <td>-0.008449</td>\n",
       "      <td>-0.019163</td>\n",
       "      <td>0.074412</td>\n",
       "      <td>-0.039493</td>\n",
       "      <td>-0.068330</td>\n",
       "      <td>-0.092204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.085299</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.044451</td>\n",
       "      <td>-0.005671</td>\n",
       "      <td>-0.045599</td>\n",
       "      <td>-0.034194</td>\n",
       "      <td>-0.032356</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.002864</td>\n",
       "      <td>-0.025930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.089063</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.011595</td>\n",
       "      <td>-0.036656</td>\n",
       "      <td>0.012191</td>\n",
       "      <td>0.024991</td>\n",
       "      <td>-0.036038</td>\n",
       "      <td>0.034309</td>\n",
       "      <td>0.022692</td>\n",
       "      <td>-0.009362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.005383</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.036385</td>\n",
       "      <td>0.021872</td>\n",
       "      <td>0.003935</td>\n",
       "      <td>0.015596</td>\n",
       "      <td>0.008142</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>-0.031991</td>\n",
       "      <td>-0.046641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>0.041708</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.019662</td>\n",
       "      <td>0.059744</td>\n",
       "      <td>-0.005697</td>\n",
       "      <td>-0.002566</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.031193</td>\n",
       "      <td>0.007207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>-0.005515</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>-0.015906</td>\n",
       "      <td>-0.067642</td>\n",
       "      <td>0.049341</td>\n",
       "      <td>0.079165</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>0.034309</td>\n",
       "      <td>-0.018118</td>\n",
       "      <td>0.044485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>0.041708</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>-0.015906</td>\n",
       "      <td>0.017282</td>\n",
       "      <td>-0.037344</td>\n",
       "      <td>-0.013840</td>\n",
       "      <td>-0.024993</td>\n",
       "      <td>-0.011080</td>\n",
       "      <td>-0.046879</td>\n",
       "      <td>0.015491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>-0.045472</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>0.039062</td>\n",
       "      <td>0.001215</td>\n",
       "      <td>0.016318</td>\n",
       "      <td>0.015283</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>0.026560</td>\n",
       "      <td>0.044528</td>\n",
       "      <td>-0.025930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>-0.045472</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.073030</td>\n",
       "      <td>-0.081414</td>\n",
       "      <td>0.083740</td>\n",
       "      <td>0.027809</td>\n",
       "      <td>0.173816</td>\n",
       "      <td>-0.039493</td>\n",
       "      <td>-0.004220</td>\n",
       "      <td>0.003064</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          age       sex       bmi  ...        s4        s5        s6\n",
       "0    0.038076  0.050680  0.061696  ... -0.002592  0.019908 -0.017646\n",
       "1   -0.001882 -0.044642 -0.051474  ... -0.039493 -0.068330 -0.092204\n",
       "2    0.085299  0.050680  0.044451  ... -0.002592  0.002864 -0.025930\n",
       "3   -0.089063 -0.044642 -0.011595  ...  0.034309  0.022692 -0.009362\n",
       "4    0.005383 -0.044642 -0.036385  ... -0.002592 -0.031991 -0.046641\n",
       "..        ...       ...       ...  ...       ...       ...       ...\n",
       "437  0.041708  0.050680  0.019662  ... -0.002592  0.031193  0.007207\n",
       "438 -0.005515  0.050680 -0.015906  ...  0.034309 -0.018118  0.044485\n",
       "439  0.041708  0.050680 -0.015906  ... -0.011080 -0.046879  0.015491\n",
       "440 -0.045472 -0.044642  0.039062  ...  0.026560  0.044528 -0.025930\n",
       "441 -0.045472 -0.044642 -0.073030  ... -0.039493 -0.004220  0.003064\n",
       "\n",
       "[442 rows x 10 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "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>Desease-Progression</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>151.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>75.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>206.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>135.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>178.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>104.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>132.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>220.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Desease-Progression\n",
       "0                  151.0\n",
       "1                   75.0\n",
       "2                  141.0\n",
       "3                  206.0\n",
       "4                  135.0\n",
       "..                   ...\n",
       "437                178.0\n",
       "438                104.0\n",
       "439                132.0\n",
       "440                220.0\n",
       "441                 57.0\n",
       "\n",
       "[442 rows x 1 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Importare i datasets\n",
    "from sklearn import datasets\n",
    "import pandas as pd\n",
    "\n",
    "class ScikitLearnDatasets:\n",
    "  def __init__(self, dataset_name):\n",
    "    # Load all scikit-learn dataset\n",
    "    if (\"iris\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_iris() # Classificazione iris dataset\n",
    "    elif (\"digits\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_digits() # Classificazione Load digits dataset\n",
    "    elif (\"wine\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_wine() # Classificazione Load wine dataset\n",
    "    elif (\"breast_cancer\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_breast_cancer() # Classificazione Load breast_cancer dataset\n",
    "    elif (\"boston\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_boston() # Regressione Load boston dataset\n",
    "      self.dataset_scelto.update([ ('target_names', ['Boston-House-Price'])] )\n",
    "    elif (\"diabetes\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_diabetes() # Regressione Load diabetes dataset\n",
    "      self.dataset_scelto.update([ ('target_names', ['Desease-Progression'])] )\n",
    "    elif (\"linnerud\"==dataset_name):\n",
    "      self.dataset_scelto = datasets.load_linnerud() # Regressione Load linnerud dataset\n",
    "    else:\n",
    "      self.dataset_scelto = diabetes # Regressione default choice\n",
    "    \n",
    "    # Print dataset information\n",
    "    self.printDatasetInformation()\n",
    "\n",
    "  def printDatasetInformation(self):\n",
    "    #print(dataset_scelto)\n",
    "    parametri = self.dataset_scelto.keys()\n",
    "    valore = self.dataset_scelto.values()\n",
    "    print(parametri)\n",
    "    # Print useful information\n",
    "    for name in parametri:\n",
    "      print(\"------------------------------------------\")\n",
    "      print(name , self.dataset_scelto[name])\n",
    "      print(\"------------------------------------------\")\n",
    "\n",
    "  def getXY(self):\n",
    "    # Get Input (X) Data\n",
    "    X = self.dataset_scelto['data'] # or  data = iris.get('data')\n",
    "    X_names = self.dataset_scelto['feature_names']\n",
    "    \n",
    "    # Get Output (Y) Target\n",
    "    parametri = self.dataset_scelto.keys()\n",
    "    Y = self.dataset_scelto['target']\n",
    "    Y_names = self.dataset_scelto['target_names']\n",
    "    \n",
    "    print(\"Dataset Parameters: \", parametri)\n",
    "    print(\"Feature Names: \", X_names)\n",
    "    print(\"Output Names: \", Y_names)\n",
    "    print(\"Input X Shape: \" , X.shape)\n",
    "    print(\"Output Y Shape: \" , Y.shape)\n",
    "    \n",
    "    return X,Y,X_names,Y_names\n",
    "          \n",
    "  def createPandasDataFrame(self,X,Y,X_names,Y_names,dataset_name):\n",
    "    df_X = pd.DataFrame(data=X, columns =X_names)\n",
    "    df_Y = pd.DataFrame(data=Y, columns =Y_names)\n",
    "    return df_X, df_Y\n",
    "\n",
    "  def writeDataFrameToCsv(self,df_X,df_Y):\n",
    "    # Create csv file\n",
    "    df_X.to_csv(dataset_name + '_X.csv', sep = ',', index = False)\n",
    "    df_Y.to_csv(dataset_name + '_Y.csv', sep = ',', index = False)\n",
    "    \n",
    "# Choose the dataset\n",
    "# Regressione: \"boston\", \"diabetes\",\n",
    "# Classificazione: \"iris\", \"digits\", \"wine\", \"breast_cancer \"\n",
    "# Regressione: \"diabetes\", \"boston\", \"linnerud\"\n",
    "dataset_name = \"diabetes\"\n",
    "myScikitLearnDatasets=ScikitLearnDatasets(dataset_name)\n",
    "X,Y,X_names,Y_names = myScikitLearnDatasets.getXY()\n",
    "df_X,df_Y = myScikitLearnDatasets.createPandasDataFrame(X,Y,X_names,Y_names,dataset_name)\n",
    "myScikitLearnDatasets.writeDataFrameToCsv(df_X,df_Y)\n",
    "\n",
    "display(df_X)\n",
    "display(df_Y)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "QfQrzbm1UoK3"
   },
   "source": [
    "# Cosa significa correlazione?\n",
    "Andiamo a vedere come si interpreta la matrice di correlazione.\n",
    "Rispondiamo alla domanda:\n",
    "<center><h3> Come deve essere la matrice di correlazione?</h3> <center>\n",
    "La correlazione esprime quanto due feature (esempio età e sesso) sono simili tra loro. Al fine di avere un dataset utile alla nostra regressione lineare è necessario che non vi sia troppa correlazione tra i dati. Se ciò accadesse significherebbe che stiamo usando diverse volte informazioni molto simili per risolvere un problema."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 258
    },
    "colab_type": "code",
    "id": "potZTF2_lNSK",
    "outputId": "c8b773fd-3aa9-495c-ea45-cd115f7912a0"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARAAAADxCAYAAAADZeiGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n79IAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "# Correlation Matrix\n",
    "plt.matshow(df_X.corr())\n",
    "plt.xticks(range(len(df_X.columns)), df_X.columns)\n",
    "plt.yticks(range(len(df_X.columns)), df_X.columns)\n",
    "plt.colorbar()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 216
    },
    "colab_type": "code",
    "id": "AaSc8cXzlost",
    "outputId": "a796df1b-876c-4b9b-e273-e12df9e77dc9"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col0 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col1 {\n",
       "            background-color:  #c0d4f5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col2 {\n",
       "            background-color:  #c1d4f4;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col3 {\n",
       "            background-color:  #cbd8ee;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col4 {\n",
       "            background-color:  #86a9fc;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col5 {\n",
       "            background-color:  #aec9fc;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col6 {\n",
       "            background-color:  #bad0f8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col7 {\n",
       "            background-color:  #e7d7ce;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col8 {\n",
       "            background-color:  #d7dce3;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col9 {\n",
       "            background-color:  #cfdaea;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col0 {\n",
       "            background-color:  #86a9fc;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col1 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col2 {\n",
       "            background-color:  #aac7fd;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col3 {\n",
       "            background-color:  #b2ccfb;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col4 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col5 {\n",
       "            background-color:  #98b9ff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col6 {\n",
       "            background-color:  #7da0f9;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col7 {\n",
       "            background-color:  #f3c7b1;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col8 {\n",
       "            background-color:  #bed2f6;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col9 {\n",
       "            background-color:  #b9d0f9;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col0 {\n",
       "            background-color:  #89acfd;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col1 {\n",
       "            background-color:  #abc8fd;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col2 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col3 {\n",
       "            background-color:  #d9dce1;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col4 {\n",
       "            background-color:  #82a6fb;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col5 {\n",
       "            background-color:  #bad0f8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col6 {\n",
       "            background-color:  #80a3fa;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col7 {\n",
       "            background-color:  #f7b99e;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col8 {\n",
       "            background-color:  #f2cab5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col9 {\n",
       "            background-color:  #e2dad5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col0 {\n",
       "            background-color:  #bad0f8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col1 {\n",
       "            background-color:  #cfdaea;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col2 {\n",
       "            background-color:  #ead4c8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col3 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col4 {\n",
       "            background-color:  #80a3fa;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col5 {\n",
       "            background-color:  #a5c3fe;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col6 {\n",
       "            background-color:  #a6c4fe;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col7 {\n",
       "            background-color:  #edd1c2;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col8 {\n",
       "            background-color:  #ecd3c5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col9 {\n",
       "            background-color:  #e2dad5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col0 {\n",
       "            background-color:  #a2c1ff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col1 {\n",
       "            background-color:  #9ebeff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col2 {\n",
       "            background-color:  #cfdaea;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col3 {\n",
       "            background-color:  #b2ccfb;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col4 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col5 {\n",
       "            background-color:  #d24b40;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col6 {\n",
       "            background-color:  #d1dae9;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col7 {\n",
       "            background-color:  #f59f80;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col8 {\n",
       "            background-color:  #f7bca1;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col9 {\n",
       "            background-color:  #d5dbe5;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col0 {\n",
       "            background-color:  #96b7ff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col1 {\n",
       "            background-color:  #b9d0f9;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col2 {\n",
       "            background-color:  #d2dbe8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col3 {\n",
       "            background-color:  #a2c1ff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col4 {\n",
       "            background-color:  #d85646;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col5 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col6 {\n",
       "            background-color:  #a2c1ff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col7 {\n",
       "            background-color:  #ed8366;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col8 {\n",
       "            background-color:  #e0dbd8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col9 {\n",
       "            background-color:  #cdd9ec;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col0 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col1 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col2 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col3 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col4 {\n",
       "            background-color:  #3f53c6;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col5 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col6 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col7 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col8 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col9 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col0 {\n",
       "            background-color:  #90b2fe;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col1 {\n",
       "            background-color:  #e1dad6;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col2 {\n",
       "            background-color:  #edd1c2;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col3 {\n",
       "            background-color:  #b6cefa;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col4 {\n",
       "            background-color:  #e3d9d3;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col5 {\n",
       "            background-color:  #f7a688;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col6 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col7 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col8 {\n",
       "            background-color:  #f6a283;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col9 {\n",
       "            background-color:  #e7d7ce;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col0 {\n",
       "            background-color:  #a6c4fe;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col1 {\n",
       "            background-color:  #bbd1f8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col2 {\n",
       "            background-color:  #f1ccb8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col3 {\n",
       "            background-color:  #d9dce1;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col4 {\n",
       "            background-color:  #dcdddd;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col5 {\n",
       "            background-color:  #cad8ef;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col6 {\n",
       "            background-color:  #7a9df8;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col7 {\n",
       "            background-color:  #f18d6f;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col8 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col9 {\n",
       "            background-color:  #efcfbf;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col0 {\n",
       "            background-color:  #afcafc;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col1 {\n",
       "            background-color:  #c9d7f0;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col2 {\n",
       "            background-color:  #ead5c9;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col3 {\n",
       "            background-color:  #d8dce2;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col4 {\n",
       "            background-color:  #9fbfff;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col5 {\n",
       "            background-color:  #c3d5f4;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col6 {\n",
       "            background-color:  #93b5fe;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col7 {\n",
       "            background-color:  #f7b89c;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col8 {\n",
       "            background-color:  #f4c6af;\n",
       "            color:  #000000;\n",
       "        }    #T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col9 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }</style><table id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >age</th>        <th class=\"col_heading level0 col1\" >sex</th>        <th class=\"col_heading level0 col2\" >bmi</th>        <th class=\"col_heading level0 col3\" >bp</th>        <th class=\"col_heading level0 col4\" >s1</th>        <th class=\"col_heading level0 col5\" >s2</th>        <th class=\"col_heading level0 col6\" >s3</th>        <th class=\"col_heading level0 col7\" >s4</th>        <th class=\"col_heading level0 col8\" >s5</th>        <th class=\"col_heading level0 col9\" >s6</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row0\" class=\"row_heading level0 row0\" >age</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col0\" class=\"data row0 col0\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col1\" class=\"data row0 col1\" >0.17</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col2\" class=\"data row0 col2\" >0.19</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col3\" class=\"data row0 col3\" >0.34</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col4\" class=\"data row0 col4\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col5\" class=\"data row0 col5\" >0.22</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col6\" class=\"data row0 col6\" >-0.075</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col7\" class=\"data row0 col7\" >0.2</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col8\" class=\"data row0 col8\" >0.27</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row0_col9\" class=\"data row0 col9\" >0.3</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row1\" class=\"row_heading level0 row1\" >sex</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col0\" class=\"data row1 col0\" >0.17</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col1\" class=\"data row1 col1\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col2\" class=\"data row1 col2\" >0.088</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col3\" class=\"data row1 col3\" >0.24</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col4\" class=\"data row1 col4\" >0.035</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col5\" class=\"data row1 col5\" >0.14</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col6\" class=\"data row1 col6\" >-0.38</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col7\" class=\"data row1 col7\" >0.33</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col8\" class=\"data row1 col8\" >0.15</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row1_col9\" class=\"data row1 col9\" >0.21</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row2\" class=\"row_heading level0 row2\" >bmi</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col0\" class=\"data row2 col0\" >0.19</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col1\" class=\"data row2 col1\" >0.088</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col2\" class=\"data row2 col2\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col3\" class=\"data row2 col3\" >0.4</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col4\" class=\"data row2 col4\" >0.25</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col5\" class=\"data row2 col5\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col6\" class=\"data row2 col6\" >-0.37</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col7\" class=\"data row2 col7\" >0.41</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col8\" class=\"data row2 col8\" >0.45</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row2_col9\" class=\"data row2 col9\" >0.39</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row3\" class=\"row_heading level0 row3\" >bp</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col0\" class=\"data row3 col0\" >0.34</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col1\" class=\"data row3 col1\" >0.24</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col2\" class=\"data row3 col2\" >0.4</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col3\" class=\"data row3 col3\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col4\" class=\"data row3 col4\" >0.24</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col5\" class=\"data row3 col5\" >0.19</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col6\" class=\"data row3 col6\" >-0.18</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col7\" class=\"data row3 col7\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col8\" class=\"data row3 col8\" >0.39</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row3_col9\" class=\"data row3 col9\" >0.39</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row4\" class=\"row_heading level0 row4\" >s1</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col0\" class=\"data row4 col0\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col1\" class=\"data row4 col1\" >0.035</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col2\" class=\"data row4 col2\" >0.25</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col3\" class=\"data row4 col3\" >0.24</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col4\" class=\"data row4 col4\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col5\" class=\"data row4 col5\" >0.9</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col6\" class=\"data row4 col6\" >0.052</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col7\" class=\"data row4 col7\" >0.54</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col8\" class=\"data row4 col8\" >0.52</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row4_col9\" class=\"data row4 col9\" >0.33</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row5\" class=\"row_heading level0 row5\" >s2</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col0\" class=\"data row5 col0\" >0.22</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col1\" class=\"data row5 col1\" >0.14</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col2\" class=\"data row5 col2\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col3\" class=\"data row5 col3\" >0.19</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col4\" class=\"data row5 col4\" >0.9</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col5\" class=\"data row5 col5\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col6\" class=\"data row5 col6\" >-0.2</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col7\" class=\"data row5 col7\" >0.66</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col8\" class=\"data row5 col8\" >0.32</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row5_col9\" class=\"data row5 col9\" >0.29</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row6\" class=\"row_heading level0 row6\" >s3</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col0\" class=\"data row6 col0\" >-0.075</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col1\" class=\"data row6 col1\" >-0.38</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col2\" class=\"data row6 col2\" >-0.37</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col3\" class=\"data row6 col3\" >-0.18</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col4\" class=\"data row6 col4\" >0.052</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col5\" class=\"data row6 col5\" >-0.2</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col6\" class=\"data row6 col6\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col7\" class=\"data row6 col7\" >-0.74</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col8\" class=\"data row6 col8\" >-0.4</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row6_col9\" class=\"data row6 col9\" >-0.27</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row7\" class=\"row_heading level0 row7\" >s4</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col0\" class=\"data row7 col0\" >0.2</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col1\" class=\"data row7 col1\" >0.33</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col2\" class=\"data row7 col2\" >0.41</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col3\" class=\"data row7 col3\" >0.26</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col4\" class=\"data row7 col4\" >0.54</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col5\" class=\"data row7 col5\" >0.66</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col6\" class=\"data row7 col6\" >-0.74</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col7\" class=\"data row7 col7\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col8\" class=\"data row7 col8\" >0.62</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row7_col9\" class=\"data row7 col9\" >0.42</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row8\" class=\"row_heading level0 row8\" >s5</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col0\" class=\"data row8 col0\" >0.27</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col1\" class=\"data row8 col1\" >0.15</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col2\" class=\"data row8 col2\" >0.45</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col3\" class=\"data row8 col3\" >0.39</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col4\" class=\"data row8 col4\" >0.52</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col5\" class=\"data row8 col5\" >0.32</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col6\" class=\"data row8 col6\" >-0.4</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col7\" class=\"data row8 col7\" >0.62</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col8\" class=\"data row8 col8\" >1</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row8_col9\" class=\"data row8 col9\" >0.46</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002level0_row9\" class=\"row_heading level0 row9\" >s6</th>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col0\" class=\"data row9 col0\" >0.3</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col1\" class=\"data row9 col1\" >0.21</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col2\" class=\"data row9 col2\" >0.39</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col3\" class=\"data row9 col3\" >0.39</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col4\" class=\"data row9 col4\" >0.33</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col5\" class=\"data row9 col5\" >0.29</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col6\" class=\"data row9 col6\" >-0.27</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col7\" class=\"data row9 col7\" >0.42</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col8\" class=\"data row9 col8\" >0.46</td>\n",
       "                        <td id=\"T_c9d8c402_1e82_11ea_bd4e_0242ac1c0002row9_col9\" class=\"data row9 col9\" >1</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f3437020dd8>"
      ]
     },
     "execution_count": 5,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_X.corr().style.background_gradient(cmap='coolwarm').set_precision(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 280
    },
    "colab_type": "code",
    "id": "uMy_YwZ2KMph",
    "outputId": "c15d7c9b-2236-44b8-963b-f2eb8e162897"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3407ea9320>"
      ]
     },
     "execution_count": 6,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWYAAAD4CAYAAADfPUyRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "corr = df_X.corr()\n",
    "sns.heatmap(corr,  xticklabels=corr.columns.values, yticklabels=corr.columns.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "JAf08kA3IP8o"
   },
   "source": [
    "# Dimensionality Reduction (Principal Component Analysis PCA)\n",
    "Nel caso vi sia una situazione in cui la correlazione tra le features è molto alta possiamo sia manualmente rimuovere le feature che consideriamo superflue oppure utilizzare la PCA. Quest'ultima si occupa di creare nuove features (se prima ne avevamo 10 adesso ne avremo un numero minore) che non hanno un significato fisico ma che sono sufficienti a rappresentare il nostro dataset. In poche parole semplichiamo gli input (features) al minimo numero necessario. Questò fara si che tra gli input vi sia pochissima correlazione in quanto ogni input features avrà un valore diverso dalle altre.\n",
    "\n",
    "IMPORTANTE: Il calcolo della principal componet analysis (PCA) è fortemente influenzato dalla scala. Quindi è necesseria avere per tutti gli input (features ) una scala comune.\n",
    "\n",
    "Standardizzare i dati significarli ricondurli a una scala il cui mean=0 e la variance=1.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 777
    },
    "colab_type": "code",
    "id": "PPpLpDqrMCrH",
    "outputId": "ded399e1-daac-43b3-9874-060e6546e679"
   },
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>bmi</th>\n",
       "      <th>bp</th>\n",
       "      <th>s1</th>\n",
       "      <th>s2</th>\n",
       "      <th>s3</th>\n",
       "      <th>s4</th>\n",
       "      <th>s5</th>\n",
       "      <th>s6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.038076</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.061696</td>\n",
       "      <td>0.021872</td>\n",
       "      <td>-0.044223</td>\n",
       "      <td>-0.034821</td>\n",
       "      <td>-0.043401</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.019908</td>\n",
       "      <td>-0.017646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.001882</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.051474</td>\n",
       "      <td>-0.026328</td>\n",
       "      <td>-0.008449</td>\n",
       "      <td>-0.019163</td>\n",
       "      <td>0.074412</td>\n",
       "      <td>-0.039493</td>\n",
       "      <td>-0.068330</td>\n",
       "      <td>-0.092204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.085299</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.044451</td>\n",
       "      <td>-0.005671</td>\n",
       "      <td>-0.045599</td>\n",
       "      <td>-0.034194</td>\n",
       "      <td>-0.032356</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.002864</td>\n",
       "      <td>-0.025930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.089063</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.011595</td>\n",
       "      <td>-0.036656</td>\n",
       "      <td>0.012191</td>\n",
       "      <td>0.024991</td>\n",
       "      <td>-0.036038</td>\n",
       "      <td>0.034309</td>\n",
       "      <td>0.022692</td>\n",
       "      <td>-0.009362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.005383</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.036385</td>\n",
       "      <td>0.021872</td>\n",
       "      <td>0.003935</td>\n",
       "      <td>0.015596</td>\n",
       "      <td>0.008142</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>-0.031991</td>\n",
       "      <td>-0.046641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>0.041708</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>0.019662</td>\n",
       "      <td>0.059744</td>\n",
       "      <td>-0.005697</td>\n",
       "      <td>-0.002566</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>-0.002592</td>\n",
       "      <td>0.031193</td>\n",
       "      <td>0.007207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>-0.005515</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>-0.015906</td>\n",
       "      <td>-0.067642</td>\n",
       "      <td>0.049341</td>\n",
       "      <td>0.079165</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>0.034309</td>\n",
       "      <td>-0.018118</td>\n",
       "      <td>0.044485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>0.041708</td>\n",
       "      <td>0.050680</td>\n",
       "      <td>-0.015906</td>\n",
       "      <td>0.017282</td>\n",
       "      <td>-0.037344</td>\n",
       "      <td>-0.013840</td>\n",
       "      <td>-0.024993</td>\n",
       "      <td>-0.011080</td>\n",
       "      <td>-0.046879</td>\n",
       "      <td>0.015491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>-0.045472</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>0.039062</td>\n",
       "      <td>0.001215</td>\n",
       "      <td>0.016318</td>\n",
       "      <td>0.015283</td>\n",
       "      <td>-0.028674</td>\n",
       "      <td>0.026560</td>\n",
       "      <td>0.044528</td>\n",
       "      <td>-0.025930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>-0.045472</td>\n",
       "      <td>-0.044642</td>\n",
       "      <td>-0.073030</td>\n",
       "      <td>-0.081414</td>\n",
       "      <td>0.083740</td>\n",
       "      <td>0.027809</td>\n",
       "      <td>0.173816</td>\n",
       "      <td>-0.039493</td>\n",
       "      <td>-0.004220</td>\n",
       "      <td>0.003064</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          age       sex       bmi  ...        s4        s5        s6\n",
       "0    0.038076  0.050680  0.061696  ... -0.002592  0.019908 -0.017646\n",
       "1   -0.001882 -0.044642 -0.051474  ... -0.039493 -0.068330 -0.092204\n",
       "2    0.085299  0.050680  0.044451  ... -0.002592  0.002864 -0.025930\n",
       "3   -0.089063 -0.044642 -0.011595  ...  0.034309  0.022692 -0.009362\n",
       "4    0.005383 -0.044642 -0.036385  ... -0.002592 -0.031991 -0.046641\n",
       "..        ...       ...       ...  ...       ...       ...       ...\n",
       "437  0.041708  0.050680  0.019662  ... -0.002592  0.031193  0.007207\n",
       "438 -0.005515  0.050680 -0.015906  ...  0.034309 -0.018118  0.044485\n",
       "439  0.041708  0.050680 -0.015906  ... -0.011080 -0.046879  0.015491\n",
       "440 -0.045472 -0.044642  0.039062  ...  0.026560  0.044528 -0.025930\n",
       "441 -0.045472 -0.044642 -0.073030  ... -0.039493 -0.004220  0.003064\n",
       "\n",
       "[442 rows x 10 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>bmi</th>\n",
       "      <th>bp</th>\n",
       "      <th>s1</th>\n",
       "      <th>s2</th>\n",
       "      <th>s3</th>\n",
       "      <th>s4</th>\n",
       "      <th>s5</th>\n",
       "      <th>s6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.800500</td>\n",
       "      <td>1.065488</td>\n",
       "      <td>1.297088</td>\n",
       "      <td>0.459840</td>\n",
       "      <td>-0.929746</td>\n",
       "      <td>-0.732065</td>\n",
       "      <td>-0.912451</td>\n",
       "      <td>-0.054499</td>\n",
       "      <td>0.418551</td>\n",
       "      <td>-0.370989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.039567</td>\n",
       "      <td>-0.938537</td>\n",
       "      <td>-1.082180</td>\n",
       "      <td>-0.553511</td>\n",
       "      <td>-0.177624</td>\n",
       "      <td>-0.402886</td>\n",
       "      <td>1.564414</td>\n",
       "      <td>-0.830301</td>\n",
       "      <td>-1.436551</td>\n",
       "      <td>-1.938479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.793307</td>\n",
       "      <td>1.065488</td>\n",
       "      <td>0.934533</td>\n",
       "      <td>-0.119218</td>\n",
       "      <td>-0.958674</td>\n",
       "      <td>-0.718897</td>\n",
       "      <td>-0.680245</td>\n",
       "      <td>-0.054499</td>\n",
       "      <td>0.060207</td>\n",
       "      <td>-0.545154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.872441</td>\n",
       "      <td>-0.938537</td>\n",
       "      <td>-0.243771</td>\n",
       "      <td>-0.770658</td>\n",
       "      <td>0.256292</td>\n",
       "      <td>0.525397</td>\n",
       "      <td>-0.757647</td>\n",
       "      <td>0.721302</td>\n",
       "      <td>0.477072</td>\n",
       "      <td>-0.196823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.113172</td>\n",
       "      <td>-0.938537</td>\n",
       "      <td>-0.764944</td>\n",
       "      <td>0.459840</td>\n",
       "      <td>0.082726</td>\n",
       "      <td>0.327890</td>\n",
       "      <td>0.171178</td>\n",
       "      <td>-0.054499</td>\n",
       "      <td>-0.672582</td>\n",
       "      <td>-0.980568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>0.876870</td>\n",
       "      <td>1.065488</td>\n",
       "      <td>0.413360</td>\n",
       "      <td>1.256044</td>\n",
       "      <td>-0.119769</td>\n",
       "      <td>-0.053957</td>\n",
       "      <td>-0.602843</td>\n",
       "      <td>-0.054499</td>\n",
       "      <td>0.655795</td>\n",
       "      <td>0.151508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>-0.115937</td>\n",
       "      <td>1.065488</td>\n",
       "      <td>-0.334410</td>\n",
       "      <td>-1.422098</td>\n",
       "      <td>1.037341</td>\n",
       "      <td>1.664355</td>\n",
       "      <td>-0.602843</td>\n",
       "      <td>0.721302</td>\n",
       "      <td>-0.380915</td>\n",
       "      <td>0.935254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>0.876870</td>\n",
       "      <td>1.065488</td>\n",
       "      <td>-0.334410</td>\n",
       "      <td>0.363330</td>\n",
       "      <td>-0.785107</td>\n",
       "      <td>-0.290965</td>\n",
       "      <td>-0.525441</td>\n",
       "      <td>-0.232934</td>\n",
       "      <td>-0.985585</td>\n",
       "      <td>0.325674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>-0.956004</td>\n",
       "      <td>-0.938537</td>\n",
       "      <td>0.821235</td>\n",
       "      <td>0.025547</td>\n",
       "      <td>0.343075</td>\n",
       "      <td>0.321306</td>\n",
       "      <td>-0.602843</td>\n",
       "      <td>0.558384</td>\n",
       "      <td>0.936155</td>\n",
       "      <td>-0.545154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>-0.956004</td>\n",
       "      <td>-0.938537</td>\n",
       "      <td>-1.535374</td>\n",
       "      <td>-1.711626</td>\n",
       "      <td>1.760535</td>\n",
       "      <td>0.584649</td>\n",
       "      <td>3.654268</td>\n",
       "      <td>-0.830301</td>\n",
       "      <td>-0.088717</td>\n",
       "      <td>0.064426</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          age       sex       bmi  ...        s4        s5        s6\n",
       "0    0.800500  1.065488  1.297088  ... -0.054499  0.418551 -0.370989\n",
       "1   -0.039567 -0.938537 -1.082180  ... -0.830301 -1.436551 -1.938479\n",
       "2    1.793307  1.065488  0.934533  ... -0.054499  0.060207 -0.545154\n",
       "3   -1.872441 -0.938537 -0.243771  ...  0.721302  0.477072 -0.196823\n",
       "4    0.113172 -0.938537 -0.764944  ... -0.054499 -0.672582 -0.980568\n",
       "..        ...       ...       ...  ...       ...       ...       ...\n",
       "437  0.876870  1.065488  0.413360  ... -0.054499  0.655795  0.151508\n",
       "438 -0.115937  1.065488 -0.334410  ...  0.721302 -0.380915  0.935254\n",
       "439  0.876870  1.065488 -0.334410  ... -0.232934 -0.985585  0.325674\n",
       "440 -0.956004 -0.938537  0.821235  ...  0.558384  0.936155 -0.545154\n",
       "441 -0.956004 -0.938537 -1.535374  ... -0.830301 -0.088717  0.064426\n",
       "\n",
       "[442 rows x 10 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "Index(['age', 'sex', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6'], dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "features = X_names\n",
    "# Separating out the features\n",
    "x = df_X\n",
    "# Separating out the target\n",
    "y = df_Y\n",
    "# Standardizing the features\n",
    "x = StandardScaler().fit_transform(x)\n",
    "df_X_Standard = pd.DataFrame(data = x , columns = X_names)\n",
    "display(df_X)\n",
    "display(df_X_Standard)\n",
    "df_X.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "2cUPVclyMtIK"
   },
   "source": [
    "## Riduciamo il numero delle features (input) a 2 usando la PCA\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 405
    },
    "colab_type": "code",
    "id": "2_E00Vt4M0vl",
    "outputId": "4e1e3dde-07ef-45b9-a996-54a074f6e673"
   },
   "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>principal component 1</th>\n",
       "      <th>principal component 2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.587208</td>\n",
       "      <td>-1.946828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.831612</td>\n",
       "      <td>1.372085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.272148</td>\n",
       "      <td>-1.634898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.049310</td>\n",
       "      <td>0.382253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.756451</td>\n",
       "      <td>0.811968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>1.239531</td>\n",
       "      <td>-1.035955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>1.264676</td>\n",
       "      <td>0.761301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>-0.205246</td>\n",
       "      <td>-1.205446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>0.692866</td>\n",
       "      <td>0.210117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>-1.903934</td>\n",
       "      <td>3.975771</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     principal component 1  principal component 2\n",
       "0                 0.587208              -1.946828\n",
       "1                -2.831612               1.372085\n",
       "2                 0.272148              -1.634898\n",
       "3                 0.049310               0.382253\n",
       "4                -0.756451               0.811968\n",
       "..                     ...                    ...\n",
       "437               1.239531              -1.035955\n",
       "438               1.264676               0.761301\n",
       "439              -0.205246              -1.205446\n",
       "440               0.692866               0.210117\n",
       "441              -1.903934               3.975771\n",
       "\n",
       "[442 rows x 2 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "array([0.40242142, 0.14923182])"
      ]
     },
     "execution_count": 8,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=2)\n",
    "principalComponents = pca.fit_transform(x)\n",
    "principalDf = pd.DataFrame(data = principalComponents , columns = ['principal component 1', 'principal component 2'])\n",
    "display(principalDf)\n",
    "\n",
    "# Varianza associata ad ogni componente\n",
    "pca.explained_variance_ratio_\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "G6yZS2AlVKF5"
   },
   "source": [
    "Dobbiamo capire il levello di informazione di ogni singola componente trovata. Quando riduciamo la dimensionalità perdiamo delle informazioni in quanto il numero di input è stato ridotto. Vogliamo chiederci adesso le componenti 1 e 2 quanta informazione contengono? La compomente 1 contiene il 40% della varianza mentre la componete 2 il 14 %. Insieme essi contengono il 54% della varianza. Cioè significa che rispetto alle informazioni iniziali abbiamo perso il 46%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5J4Ryq1kQrfH"
   },
   "source": [
    "## Come facciamo a riddure il numero degli input senza ridurre il contenuto di informazioni del nostro dataset iniziale?\n",
    "Mantenere esattamente il 100 % delle informazioni è impossibile. Pertanto ridurremo di poco il contenuto delle informazioni (es. 90%). Questo ci permetterà di ridurre il numero di input ed avere allo stesso tempo un predizione ottima.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 405
    },
    "colab_type": "code",
    "id": "SBjO2Lt-R0ou",
    "outputId": "5bdc0b15-fb72-474b-d3dc-738138a79534"
   },
   "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>component_0</th>\n",
       "      <th>component_1</th>\n",
       "      <th>component_2</th>\n",
       "      <th>component_3</th>\n",
       "      <th>component_4</th>\n",
       "      <th>component_5</th>\n",
       "      <th>component_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.587208</td>\n",
       "      <td>-1.946828</td>\n",
       "      <td>0.589233</td>\n",
       "      <td>-0.082812</td>\n",
       "      <td>-0.256630</td>\n",
       "      <td>-1.011214</td>\n",
       "      <td>-0.179807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.831612</td>\n",
       "      <td>1.372085</td>\n",
       "      <td>0.027915</td>\n",
       "      <td>-0.469999</td>\n",
       "      <td>-0.143229</td>\n",
       "      <td>-1.013015</td>\n",
       "      <td>0.224414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.272148</td>\n",
       "      <td>-1.634898</td>\n",
       "      <td>0.739270</td>\n",
       "      <td>-0.791475</td>\n",
       "      <td>-1.163821</td>\n",
       "      <td>-1.112806</td>\n",
       "      <td>-0.462406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.049310</td>\n",
       "      <td>0.382253</td>\n",
       "      <td>-2.013037</td>\n",
       "      <td>1.373239</td>\n",
       "      <td>0.255528</td>\n",
       "      <td>0.445315</td>\n",
       "      <td>0.482147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.756451</td>\n",
       "      <td>0.811968</td>\n",
       "      <td>-0.057259</td>\n",
       "      <td>-0.137520</td>\n",
       "      <td>-0.133344</td>\n",
       "      <td>-0.814591</td>\n",
       "      <td>0.436451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>1.239531</td>\n",
       "      <td>-1.035955</td>\n",
       "      <td>0.928691</td>\n",
       "      <td>-0.656266</td>\n",
       "      <td>0.204315</td>\n",
       "      <td>-0.479371</td>\n",
       "      <td>0.394431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>1.264676</td>\n",
       "      <td>0.761301</td>\n",
       "      <td>-1.750210</td>\n",
       "      <td>-1.133482</td>\n",
       "      <td>-0.094014</td>\n",
       "      <td>0.973430</td>\n",
       "      <td>-1.173570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>439</th>\n",
       "      <td>-0.205246</td>\n",
       "      <td>-1.205446</td>\n",
       "      <td>0.496078</td>\n",
       "      <td>-1.353349</td>\n",
       "      <td>-0.141686</td>\n",
       "      <td>-0.045289</td>\n",
       "      <td>-0.635451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>0.692866</td>\n",
       "      <td>0.210117</td>\n",
       "      <td>-0.868730</td>\n",
       "      <td>1.616790</td>\n",
       "      <td>0.119649</td>\n",
       "      <td>-0.556900</td>\n",
       "      <td>0.545703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>-1.903934</td>\n",
       "      <td>3.975771</td>\n",
       "      <td>-0.048381</td>\n",
       "      <td>-0.220612</td>\n",
       "      <td>0.599824</td>\n",
       "      <td>1.647108</td>\n",
       "      <td>0.245265</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>442 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     component_0  component_1  ...  component_5  component_6\n",
       "0       0.587208    -1.946828  ...    -1.011214    -0.179807\n",
       "1      -2.831612     1.372085  ...    -1.013015     0.224414\n",
       "2       0.272148    -1.634898  ...    -1.112806    -0.462406\n",
       "3       0.049310     0.382253  ...     0.445315     0.482147\n",
       "4      -0.756451     0.811968  ...    -0.814591     0.436451\n",
       "..           ...          ...  ...          ...          ...\n",
       "437     1.239531    -1.035955  ...    -0.479371     0.394431\n",
       "438     1.264676     0.761301  ...     0.973430    -1.173570\n",
       "439    -0.205246    -1.205446  ...    -0.045289    -0.635451\n",
       "440     0.692866     0.210117  ...    -0.556900     0.545703\n",
       "441    -1.903934     3.975771  ...     1.647108     0.245265\n",
       "\n",
       "[442 rows x 7 columns]"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "94.79436357350414\n"
     ]
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "# If 0 < n_components < 1 and svd_solver == 'full', select the number of \n",
    "# components such that the amount of variance that needs to be explained \n",
    "# is greater than the percentage specified by n_components.\n",
    "pca = PCA(n_components=0.90)\n",
    "x = df_X.values\n",
    "x = StandardScaler().fit_transform(x)\n",
    "principalComponents = pca.fit_transform(x)\n",
    "\n",
    "row_number = principalComponents.shape[1]\n",
    "X_names_new = []\n",
    "for i in range(0,row_number): \n",
    "  name = \"component_\" + str(i)\n",
    "  X_names_new.append(name)\n",
    "\n",
    "\n",
    "principalDf = pd.DataFrame(data = principalComponents ,columns=X_names_new)\n",
    "display(principalDf)\n",
    "\n",
    "# Varianza associata ad ogni componente\n",
    "variance_arr = pca.explained_variance_ratio_\n",
    "tot_variance = 0\n",
    "for variance in variance_arr:\n",
    "  temp_variance = variance*100\n",
    "  tot_variance += temp_variance\n",
    "print(tot_variance)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "9xdIC8Xr82uL"
   },
   "source": [
    "# Esempio Pratico\n",
    "1. Scarichiamo il boston dataset\n",
    "2. Dividiamo i dati in training e test\n",
    "3. Applichiamo la PCA\n",
    "4. Compariamo la predizione di un Regressore Lineare con e senza pca "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iN3zGXhP-AM1"
   },
   "outputs": [],
   "source": [
    "  from sklearn.preprocessing import StandardScaler\n",
    "  from sklearn.decomposition import PCA\n",
    "  import pandas as pd\n",
    "  import matplotlib.pyplot as plt\n",
    "  import numpy as np\n",
    "  from sklearn import datasets\n",
    "  import os\n",
    "  from sklearn import linear_model\n",
    "  from sklearn.metrics import mean_squared_error\n",
    "  from sklearn.model_selection import train_test_split\n",
    "  plt.rcParams['figure.figsize'] = [15, 10]\n",
    "\n",
    "  # FUNCTION: Standard data with 0 mean and unit variance (Gaussian)\n",
    "  def standardScaler(df):\n",
    "      # Input pandas dataframe Output pandas dataframe scaled\n",
    "      names = list(df.keys())\n",
    "      data = df.values\n",
    "      data_scaled = StandardScaler().fit_transform(data)\n",
    "      df_scaled = pd.DataFrame(data = data_scaled , columns=names)\n",
    "      #print(df)\n",
    "      #print(df_scaled)\n",
    "      return df_scaled\n",
    "\n",
    "  # FUNCTION: Create pandas dataframe from numpy array\n",
    "  def createPandasDataFrame(X,Y,X_names,Y_names):\n",
    "    df_X = pd.DataFrame(data=X, columns =X_names)\n",
    "    df_Y = pd.DataFrame(data=Y, columns =Y_names)\n",
    "    return df_X, df_Y\n",
    "\n",
    "  # FUNCTION: Write pandas dataframe to csv file\n",
    "  def writeDataFrameToCsv(df_X,df_Y,directory_path,dataset_name):\n",
    "    if not os.path.exists(directory_path):\n",
    "      os.makedirs(directory_path)\n",
    "    \n",
    "    # Create csv file\n",
    "    path_write = os.path.join(directory_path, dataset_name)\n",
    "    df_X.to_csv(path_write + '_X.csv', sep = ',', index = False)\n",
    "    df_Y.to_csv(path_write + '_Y.csv', sep = ',', index = False)\n",
    "\n",
    "  # FUNCTION: Read from csv file a dataset\n",
    "  def readDataFrameFromCsv(directory_path, dataset_name):\n",
    "    path_read = os.path.join(directory_path, dataset_name)\n",
    "    if not os.path.exists(directory_path):\n",
    "      print(\"Directory path does not exist\")\n",
    "    df_X = pd.read_csv(path_read + '_X.csv') \n",
    "    df_Y = pd.read_csv(path_read + '_Y.csv') \n",
    "    return df_X, df_Y  \n",
    "\n",
    "  # CLASS: PCA class to do input dimensionality reduction\n",
    "  class dimensionalityReduction():\n",
    "\n",
    "      def __init__(self,n_components): \n",
    "          # define pca\n",
    "          self.pca = PCA(n_components)\n",
    "          \n",
    "      def create_names(self,col_number):\n",
    "          names = []\n",
    "          for i in range(0,col_number): \n",
    "              name = \"component_\" + str(i)\n",
    "              names.append(name)\n",
    "          return names\n",
    "\n",
    "      def fit(self,df):\n",
    "          df_scaled = standardScaler(df)\n",
    "          model = self.pca.fit(df_scaled.values)\n",
    "          information_array = model.explained_variance_ratio_ *100.00\n",
    "          total_information = np.sum(information_array)\n",
    "          return model,information_array, total_information\n",
    "\n",
    "      def transform(self, model,df):  \n",
    "          # Input dataframe\n",
    "          principalComponents = model.transform(df.values)\n",
    "          names = self.create_names(col_number=principalComponents.shape[1])\n",
    "          df_scaled = pd.DataFrame(data = principalComponents , columns = names)\n",
    "          return df_scaled\n",
    "\n",
    "  # CLASS: Easily import different datasets\n",
    "  class ScikitLearnDatasets():\n",
    "    \n",
    "    def __init__(self, dataset_name):\n",
    "      # Load all scikit-learn dataset\n",
    "      if (\"iris\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_iris() # Classificazione iris dataset\n",
    "      elif (\"digits\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_digits() # Classificazione Load digits dataset\n",
    "      elif (\"wine\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_wine() # Classificazione Load wine dataset\n",
    "      elif (\"breast_cancer\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_breast_cancer() # Classificazione Load breast_cancer dataset\n",
    "      elif (\"boston\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_boston() # Regressione Load boston dataset\n",
    "        self.dataset_scelto.update([ ('target_names', ['Boston-House-Price'])] )\n",
    "      elif (\"diabetes\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_diabetes() # Regressione Load diabetes dataset\n",
    "        self.dataset_scelto.update([ ('target_names', ['Desease-Progression'])] )\n",
    "      elif (\"linnerud\"==dataset_name):\n",
    "        self.dataset_scelto = datasets.load_linnerud() # Regressione Load linnerud dataset\n",
    "      else:\n",
    "        self.dataset_scelto = \"diabetes\" # Regressione default choice\n",
    "      \n",
    "      # Print dataset information\n",
    "      #self.printDatasetInformation()\n",
    "\n",
    "    def printDatasetInformation(self):\n",
    "      #print(dataset_scelto)\n",
    "      parameters = self.dataset_scelto.keys()\n",
    "      data = self.dataset_scelto.values()\n",
    "      #print(parameters)\n",
    "      # Print useful information\n",
    "      for name in parameters:\n",
    "        print(\"------------------------------------------\")\n",
    "        print(name , self.dataset_scelto[name])\n",
    "        print(\"------------------------------------------\")\n",
    "\n",
    "    def getXY(self):\n",
    "      # Get Input (X) Data\n",
    "      X = self.dataset_scelto['data'] # or  data = iris.get('data')\n",
    "      X_names = self.dataset_scelto['feature_names']\n",
    "      \n",
    "      # Get Output (Y) Target\n",
    "      parameters = self.dataset_scelto.keys()\n",
    "      Y = self.dataset_scelto['target']\n",
    "      Y_names = self.dataset_scelto['target_names']\n",
    "      \n",
    "      return X,Y,X_names,Y_names\n",
    "            \n",
    "  # CLASS: Linar Regression          \n",
    "  class LinearRegression(): \n",
    "\n",
    "      def __init__(self):\n",
    "        # Inizializzazione\n",
    "        # https://scikit-learn.org/stable/modules/linear_model.html#ridge-regression-and-classification\n",
    "        self.model = linear_model.LinearRegression(fit_intercept=True, normalize=False)\n",
    "        \n",
    "      def train(self,X,Y):\n",
    "        # Stimare w0, w1 .. wN\n",
    "        trained_model = self.model.fit(X,Y)\n",
    "        #print(\"w1,w2 .. wN : \",self.model.coef_)\n",
    "        #print(\"w0 : \", self.model.intercept_) \n",
    "        return trained_model\n",
    "      \n",
    "      def predict(self,X_test,trained_model):\n",
    "        Y_pred = trained_model.predict(X_test)\n",
    "        return Y_pred\n",
    "      \n",
    "      def evaluate(self,X_test, Y_test, trained_model):\n",
    "        # R2 score\n",
    "        Y_pred = trained_model.predict(X_test)\n",
    "        R2_score = trained_model.score(X_test, Y_test)\n",
    "        RMSE_score = (np.sqrt(mean_squared_error(Y_test, Y_pred)))\n",
    "        return Y_pred,R2_score, RMSE_score\n",
    "\n",
    "      def plot(self,Y_test,Y_pred):\n",
    "          length = Y_pred.shape[0] # 20\n",
    "          index_bar = np.linspace(0,length,length)\n",
    "          plt.plot(index_bar, Y_test, label='Test')\n",
    "          plt.plot(index_bar, Y_pred, label='Prediction')\n",
    "          plt.legend()\n",
    "          plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "IfeiGK_c8190",
    "outputId": "9d962321-9a32-412a-c462-4aa7e088665f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#---------- DATASET INFORMATION ------------#\n",
      "X Input or feature_names:  ['age', 'sex', 'bmi', 'bp', 's1', 's2', 's3', 's4', 's5', 's6']\n",
      "Y Output or target_names:  ['Desease-Progression']\n",
      "Input X Shape:  (442, 10)\n",
      "Output Y Shape:  (442,)\n",
      "Dataframe df_X Input Describe: \n",
      "                 age           sex  ...            s5            s6\n",
      "count  4.420000e+02  4.420000e+02  ...  4.420000e+02  4.420000e+02\n",
      "mean  -3.634285e-16  1.308343e-16  ... -3.830854e-16 -3.412882e-16\n",
      "std    4.761905e-02  4.761905e-02  ...  4.761905e-02  4.761905e-02\n",
      "min   -1.072256e-01 -4.464164e-02  ... -1.260974e-01 -1.377672e-01\n",
      "25%   -3.729927e-02 -4.464164e-02  ... -3.324879e-02 -3.317903e-02\n",
      "50%    5.383060e-03 -4.464164e-02  ... -1.947634e-03 -1.077698e-03\n",
      "75%    3.807591e-02  5.068012e-02  ...  3.243323e-02  2.791705e-02\n",
      "max    1.107267e-01  5.068012e-02  ...  1.335990e-01  1.356118e-01\n",
      "\n",
      "[8 rows x 10 columns]\n",
      "Dataframe df_Y Output Describe: \n",
      "        Desease-Progression\n",
      "count           442.000000\n",
      "mean            152.133484\n",
      "std              77.093005\n",
      "min              25.000000\n",
      "25%              87.000000\n",
      "50%             140.500000\n",
      "75%             211.500000\n",
      "max             346.000000\n",
      "#-------------------------------------------#\n",
      "#-------------- PCA ANALYSIS ---------------#\n",
      "Information for each new component:  [39.68814054 14.77973436 12.51654914 10.10869693  6.58293305  5.93511774\n",
      "  5.2036642   4.33649442] %\n",
      "Total Information of the reduced dataset:  99.15133037867369  %\n",
      "Number of inputs with PCA:  8\n",
      "Number of inputs without PCA:  10\n",
      "#-------------------------------------------#\n",
      "#----- LINEAR REGRESSION PCA RESULTS -------#\n",
      "w1,w2 .. wN :  [[ 453.41998089 -244.95319406  372.8080337   524.14406459  -30.20353053\n",
      "  -252.44550156  125.2975691    34.12174383]]\n",
      "w0 :  [151.30071414]\n",
      "Score Linear Regression PCA:  R2 Score:  0.4557153915064335  RMSE Score:  53.7001159233466\n",
      "#-------------------------------------------#\n",
      "#------- LINEAR REGRESSION RESULTS ---------#\n",
      "w1,w2 .. wN :  [[ 453.41998089 -244.95319406  372.8080337   524.14406459  -30.20353053\n",
      "  -252.44550156  125.2975691    34.12174383]]\n",
      "w0 :  [151.30071414]\n",
      "Score Linear regression without PCA:  R2 Score:  0.45260660216173787  RMSE Score:  53.8532569849144\n",
      "#-------------------------------------------#\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAI/CAYAAADkwzGCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "if __name__ == \"__main__\":\n",
    "\n",
    "    # ----------- SKLEARN DATASET LOADING ------------ #\n",
    "    # Load sklearn dataset\n",
    "    # Classificazione: \"iris\", \"digits\", \"wine\", \"breast_cancer \"\n",
    "    # Regressione: \"diabetes\", \"boston\", \"linnerud\"\n",
    "    # 1. Select dataset \n",
    "    dataset_name = \"diabetes\"\n",
    "    # 2. Create class object ScikitLearnDatasets \n",
    "    myScikitLearnDatasets=ScikitLearnDatasets(dataset_name)\n",
    "    # 3. Print dataset information\n",
    "    #myScikitLearnDatasets.printDatasetInformation()\n",
    "    # 4. Get dataset data as numpy array X=input, Y=output and X_names=input_names, Y_names=output_names\n",
    "    X,Y,X_names,Y_names = myScikitLearnDatasets.getXY()\n",
    "    # 5. Convert numpy array data to Pandas Dataframe\n",
    "    df_X,df_Y = createPandasDataFrame(X,Y,X_names,Y_names)\n",
    "    print(\"#---------- DATASET INFORMATION ------------#\")\n",
    "    print(\"X Input or feature_names: \", X_names)\n",
    "    print(\"Y Output or target_names: \", Y_names)\n",
    "    print(\"Input X Shape: \" , X.shape)\n",
    "    print(\"Output Y Shape: \" , Y.shape)\n",
    "    print(\"Dataframe df_X Input Describe: \\n\", df_X.describe())\n",
    "    print(\"Dataframe df_Y Output Describe: \\n\", df_Y.describe())\n",
    "    print(\"#-------------------------------------------#\")\n",
    "    # 6. Write Pandas dataframe df_X, df_Y to csv file\n",
    "    directory_path = os.path.join(os.getcwd(), \"ScikitLearnDatasets\")\n",
    "    writeDataFrameToCsv(df_X,df_Y,directory_path, dataset_name)\n",
    "    # ------------------------------------------------ #\n",
    "\n",
    "    # ----------- READ DATASET FROM CSV -------------- #\n",
    "    # Read previously saved dataset\n",
    "    # 1. Read csv dataset (examvle boston_X.csv and boston_Y.csv) and transform to pandas daframe\n",
    "    #dataset_name = \"boston\" # desired dataset name \n",
    "    #directory_path = os.path.join(os.getcwd(), \"ScikitLearnDatasets\") # dataset folder\n",
    "    df_X,df_Y = readDataFrameFromCsv(directory_path, dataset_name)\n",
    "    # ------------------------------------------------ #\n",
    "\n",
    "    # -------- Split data into train and test -------- #\n",
    "    # Split dataset into training and test set\n",
    "    X_train, X_test, Y_train, Y_test = train_test_split(df_X.values, df_Y.values, test_size=0.20, random_state=42)\n",
    "    # ------------------------------------------------ #\n",
    "\n",
    "    # -------------------- PCA ----------------------- #\n",
    "    # Principal component analysis (PCA) or dimensionality reduction\n",
    "    # Number of input is reduced while keeping overall dataset information\n",
    "    # 1. Covert numpy array to pandas dataframe\n",
    "    df_X_train = pd.DataFrame(data = X_train , columns=df_X.keys())\n",
    "    df_X_test = pd.DataFrame(data = X_test , columns=df_X.keys())\n",
    "    # 2. Initialize PCA (Principal Component Analysis)\n",
    "    n_components = 0.95 # 90% of the variance\n",
    "    mydimensionalityReduction = dimensionalityReduction(n_components)\n",
    "    # 3. Create PCA model (using input training data)\n",
    "    pcaModel,information_array, total_information = mydimensionalityReduction.fit(df_X_train)\n",
    "    print(\"#-------------- PCA ANALYSIS ---------------#\")\n",
    "    print(\"Information for each new component: \", information_array, \"%\")\n",
    "    print(\"Total Information of the reduced dataset: \", total_information, \" %\")\n",
    "    # 4. Apply created PCA model to both training and test dataset\n",
    "    df_X_train_scaled = mydimensionalityReduction.transform(pcaModel,df_X_train)\n",
    "    df_X_test_scaled = mydimensionalityReduction.transform(pcaModel,df_X_test)\n",
    "    X_train_scaled = df_X_train_scaled.values\n",
    "    X_test_scaled = df_X_test_scaled.values\n",
    "    #print(\"Dataset X_train: \", X_train)\n",
    "    #print(\"Dataset X_train Reduced: \", X_train_scaled)\n",
    "    print(\"Number of inputs with PCA: \",X_train_scaled.shape[1])\n",
    "    print(\"Number of inputs without PCA: \",X_train.shape[1])\n",
    "    print(\"#-------------------------------------------#\")\n",
    "    # --------------------------------------------------#\n",
    "    \n",
    "    # -------- LINEAR REGRESSION WITH PCA DATA -------- #\n",
    "    # we use reduce input data\n",
    "    myModelPCA = LinearRegression()\n",
    "    trained_modelPCA = myModelPCA.train(X_train_scaled, Y_train)\n",
    "    Y_predPCA,R2_scorePCA, RMSE_scorePCA = myModelPCA.evaluate(X_test_scaled,Y_test,trained_modelPCA)\n",
    "    print(\"#----- LINEAR REGRESSION PCA RESULTS -------#\")\n",
    "    print(\"w1,w2 .. wN : \",trained_modelPCA.coef_)\n",
    "    print(\"w0 : \", trained_modelPCA.intercept_) \n",
    "    print(\"Score Linear Regression PCA: \", \"R2 Score: \", R2_scorePCA, \" RMSE Score: \", RMSE_scorePCA)\n",
    "    print(\"#-------------------------------------------#\")\n",
    "    #myModelPCA.plot(Y_test,Y_pred)\n",
    "    # ------------------------------------------------ #\n",
    "\n",
    "    # -------------- LINEAR REGRESSOR ---------------- #\n",
    "    # We use initial data \n",
    "    myModel = LinearRegression()\n",
    "    trained_model = myModel.train(X_train, Y_train)\n",
    "    Y_pred,R2_score, RMSE_score = myModel.evaluate(X_test,Y_test,trained_model)\n",
    "    print(\"#------- LINEAR REGRESSION RESULTS ---------#\")\n",
    "    print(\"w1,w2 .. wN : \",trained_modelPCA.coef_)\n",
    "    print(\"w0 : \", trained_modelPCA.intercept_) \n",
    "    print(\"Score Linear regression without PCA: \", \"R2 Score: \", R2_score, \" RMSE Score: \", RMSE_score)\n",
    "    print(\"#-------------------------------------------#\")\n",
    "    #myModel.plot(Y_test,Y_pred)\n",
    "    # ------------------------------------------------ #\n",
    "\n",
    "    #----------------- COMPARISON -------------------- #    \n",
    "    length = Y_pred.shape[0] # 20\n",
    "    index_bar = np.linspace(0,length,length)\n",
    "    plt.plot(index_bar, Y_test, label='Test')\n",
    "    plt.plot(index_bar, Y_predPCA, label='PredictionPCA')\n",
    "    plt.plot(index_bar, Y_pred, label='Prediction')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "    # ------------------------------------------------ #\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "uEfz-0X1nX1p"
   },
   "source": [
    "# Referenze utili\n",
    "* [PCA SPiegazione ed esempio](https://towardsdatascience.com/pca-using-python-scikit-learn-e653f8989e60)\n",
    "* [Scikit-Learn Regressione](https://scikit-learn.org/stable/supervised_learning.html)\n",
    "* [L'importanza di Standardizzare i dati](https://scikit-learn.org/stable/auto_examples/preprocessing/plot_scaling_importance.html#sphx-glr-auto-examples-preprocessing-plot-scaling-importance-py)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "XUuyk3-o6onC"
   },
   "source": [
    "# Extra : Come funzionano i dizionari in python "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 528
    },
    "colab_type": "code",
    "id": "EEHD-7vS6qmr",
    "outputId": "a33b9336-dd2c-46e6-bac6-838ffc9859e6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------------------------\n",
      "-------- Approach 1 ---------\n",
      "-----------------------------\n",
      "dict_keys(['data', 'target', 'target_names', 'DESCR', 'feature_names', 'filename'])\n",
      "data\n",
      "target\n",
      "target_names\n",
      "DESCR\n",
      "feature_names\n",
      "filename\n",
      "All keys inside array  ['data', 'target', 'target_names', 'DESCR', 'feature_names', 'filename']\n",
      "----------------------------------\n",
      "<class 'numpy.ndarray'> (6,) ()\n",
      "<class 'numpy.ndarray'> (6,) (150, 4)\n",
      "----------------------------------\n",
      "data <class 'numpy.ndarray'> (150, 4)\n",
      "target <class 'numpy.ndarray'> (150,)\n",
      "target_names <class 'numpy.ndarray'> (3,)\n",
      "DESCR <class 'str'>\n",
      "feature_names <class 'list'>\n",
      "filename <class 'str'>\n",
      "-----------------------------\n",
      "-------- Approach 2 ---------\n",
      "-----------------------------\n",
      "<class 'list'>\n",
      "<class 'list'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'list'>\n",
      "<class 'list'>\n",
      "The lists are identical\n"
     ]
    }
   ],
   "source": [
    "#---------------------------------------------------------------------\n",
    "#---------------- Trasformare dizionari in Array --------- -----------\n",
    "#---------------------------------------------------------------------\n",
    "# Usiamo Le funzioni\n",
    "# Importare i datasets\n",
    "\n",
    "from sklearn import datasets\n",
    "import numpy as np\n",
    "\n",
    "iris = datasets.load_iris() # Load iris dataset\n",
    "digits = datasets.load_digits() # Load digits dataset\n",
    "boston = datasets.load_boston() # Load boston dataset\n",
    "diabetes = datasets.load_diabetes() # Load diabetes dataset\n",
    "linnerud = datasets.load_linnerud() # Load linnerud dataset\n",
    "wine = datasets.load_wine() # Load wine dataset\n",
    "breast_cancer = datasets.load_breast_cancer() # Load breast_cancer dataset\n",
    "\n",
    "dataset_scelto = iris\n",
    "\n",
    "def approccio_1(dataset_scelto):\n",
    "  # --------------------- Approccio 1\n",
    "  # Get dictionar keys, value\n",
    "  print(dataset_scelto.keys())\n",
    "  list_keys = []\n",
    "  list_values = []\n",
    "  for key in dataset_scelto:\n",
    "    list_keys.append(key)\n",
    "    print(key)\n",
    "    value = dataset_scelto[key]\n",
    "    list_values.append(value)\n",
    "  print(\"All keys inside array \", list_keys) \n",
    "  #print(\"All values inside array \", list_values) \n",
    "\n",
    "  # Convert list to numpy array\n",
    "  print(\"----------------------------------\")\n",
    "  array_keys = np.asarray(list_keys)\n",
    "  array_values = np.array(list_values)\n",
    "  print(type(array_keys), array_keys.shape, array_keys[0].shape)\n",
    "  print(type(array_values), array_values.shape,array_values[0].shape)\n",
    "\n",
    "  # Going deeper inside data shape\n",
    "  print(\"----------------------------------\")\n",
    "  for i in range(0,len(array_keys)):\n",
    "    if isinstance(array_values[i],np.ndarray):\n",
    "      print(array_keys[i], type(array_values[i]), array_values[i].shape )\n",
    "    else:\n",
    "      print(array_keys[i], type(array_values[i]))\n",
    "    \n",
    "  # Other Useful Solutions\n",
    "  '''\n",
    "  for value in diabetes.values():\n",
    "    print(value) \n",
    "\n",
    "  for key, value in diabetes.items():\n",
    "    print(key, value)\n",
    "  '''\n",
    "\n",
    "def approccio_2(dataset_scelto):\n",
    "  #--------------------- Approccio 2\n",
    "  # Convert a dictionary to an array of string\n",
    "  list_keys = list(dataset_scelto.keys())\n",
    "  list_values = list(dataset_scelto.values())\n",
    "  #print(list_keys)\n",
    "  print(type(list_keys))\n",
    "  #print(list_values)\n",
    "  print(type(list_values))\n",
    "  # Convert list as numpy narray\n",
    "  array_keys = np.asarray(list_keys)\n",
    "  array_values = np.array(list_values)\n",
    "  print(type(array_keys))\n",
    "  print(type(array_values))\n",
    "  # Covert back numpy ndarray to list\n",
    "  new_list_keys = array_keys.tolist()\n",
    "  new_list_values = array_values.tolist()\n",
    "  print(type(new_list_keys))\n",
    "  print(type(new_list_values))\n",
    "  # Check if the list are equal\n",
    "  if list_keys == new_list_keys and list_values==new_list_values: \n",
    "    print (\"The lists are identical\") \n",
    "  else : \n",
    "    print (\"The lists are not identical\")\n",
    "\n",
    "print(\"-----------------------------\")\n",
    "print(\"-------- Approach 1 ---------\")\n",
    "print(\"-----------------------------\")\n",
    "approccio_1(dataset_scelto)\n",
    "print(\"-----------------------------\")\n",
    "print(\"-------- Approach 2 ---------\")\n",
    "print(\"-----------------------------\")\n",
    "approccio_2(dataset_scelto)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "04_scikit-learn-Dataset_PCA.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}