Data train test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Librerias\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LinearRegression \n",
    "import statsmodels.formula.api as smf\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "from sklearn import linear_model,svm\n",
    "from sklearn.svm import SVR\n",
    "from time import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>Mes</th>\n",
       "      <th>Dia</th>\n",
       "      <th>Anio</th>\n",
       "      <th>Evaporacion</th>\n",
       "      <th>Calor</th>\n",
       "      <th>Precipitacion</th>\n",
       "      <th>TempMax</th>\n",
       "      <th>TempMin</th>\n",
       "      <th>TempProm</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1922</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>19.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1922</td>\n",
       "      <td>2.9</td>\n",
       "      <td>9.25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.5</td>\n",
       "      <td>14.0</td>\n",
       "      <td>19.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1922</td>\n",
       "      <td>3.7</td>\n",
       "      <td>8.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1922</td>\n",
       "      <td>3.1</td>\n",
       "      <td>8.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>18.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1922</td>\n",
       "      <td>4.3</td>\n",
       "      <td>9.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>19.50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  Mes  Dia  Anio  Evaporacion  Calor  Precipitacion  TempMax  \\\n",
       "0           0    2    1  1922          3.0   9.50            0.0     25.0   \n",
       "1           1    2    2  1922          2.9   9.25            0.0     24.5   \n",
       "2           2    2    3  1922          3.7   8.50            0.0     24.0   \n",
       "3           3    2    4  1922          3.1   8.00            0.0     25.0   \n",
       "4           4    2    5  1922          4.3   9.50            0.0     25.0   \n",
       "\n",
       "   TempMin  TempProm  \n",
       "0     14.0     19.50  \n",
       "1     14.0     19.25  \n",
       "2     13.0     18.50  \n",
       "3     11.0     18.00  \n",
       "4     14.0     19.50  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datamain = pd.read_csv('CICESEDATA.csv')\n",
    "datamain.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "datamain = datamain.fillna(0)\n",
    "datamain.describe()\n",
    "feature_cols = [\"Mes\", \"Dia\", \"Anio\", \"Evaporacion\", \"Calor\", \"Precipitacion\"]\n",
    "y_col = ['TempMax']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1922to2015 = datamain[(datamain['Anio']>=1922) & (datamain['Anio'] <= 2015)]\n",
    "df2016 = datamain[(datamain['Anio'] >= 2016)]\n",
    "dXtrain, dYtrain = df1922to2015[feature_cols] ,  df1922to2015[y_col]\n",
    "dXtest, dYtest = df2016[feature_cols] ,  df2016[y_col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "lm = smf.ols(formula=\"TempMax~Mes+Dia+Anio+Evaporacion+Calor+Precipitacion\", data=df1922to2015).fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td>Model:</td>               <td>OLS</td>         <td>Adj. R-squared:</td>      <td>0.764</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Dependent Variable:</td>      <td>TempMax</td>            <td>AIC:</td>         <td>177517.6861</td>\n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Date:</td>        <td>2019-03-28 13:45</td>        <td>BIC:</td>         <td>177576.7868</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>No. Observations:</td>        <td>34302</td>        <td>Log-Likelihood:</td>     <td>-88752.</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Df Model:</td>              <td>6</td>           <td>F-statistic:</td>      <td>1.849e+04</td> \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Df Residuals:</td>          <td>34295</td>      <td>Prob (F-statistic):</td>    <td>0.00</td>    \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>R-squared:</td>            <td>0.764</td>            <td>Scale:</td>          <td>10.351</td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>         <th>Coef.</th>  <th>Std.Err.</th>     <th>t</th>     <th>P>|t|</th> <th>[0.025</th>  <th>0.975]</th> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>     <td>-0.9463</td>  <td>1.9238</td>   <td>-0.4919</td> <td>0.6228</td> <td>-4.7170</td> <td>2.8243</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Mes</th>           <td>-0.0936</td>  <td>0.0051</td>  <td>-18.4578</td> <td>0.0000</td> <td>-0.1035</td> <td>-0.0836</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Dia</th>           <td>0.0003</td>   <td>0.0020</td>   <td>0.1414</td>  <td>0.8876</td> <td>-0.0036</td> <td>0.0041</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Anio</th>          <td>0.0066</td>   <td>0.0010</td>   <td>6.7268</td>  <td>0.0000</td> <td>0.0047</td>  <td>0.0085</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Evaporacion</th>   <td>0.0398</td>   <td>0.0086</td>   <td>4.6478</td>  <td>0.0000</td> <td>0.0230</td>  <td>0.0566</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Calor</th>         <td>1.6452</td>   <td>0.0052</td>  <td>317.4160</td> <td>0.0000</td> <td>1.6351</td>  <td>1.6554</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Precipitacion</th> <td>-0.0794</td>  <td>0.0030</td>  <td>-26.4843</td> <td>0.0000</td> <td>-0.0852</td> <td>-0.0735</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "     <td>Omnibus:</td>    <td>9598.858</td>  <td>Durbin-Watson:</td>     <td>0.247</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Prob(Omnibus):</td>   <td>0.000</td>  <td>Jarque-Bera (JB):</td> <td>30931.320</td>\n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Skew:</td>      <td>-1.425</td>      <td>Prob(JB):</td>       <td>0.000</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Kurtosis:</td>     <td>6.677</td>   <td>Condition No.:</td>    <td>218038</td>  \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary2.Summary'>\n",
       "\"\"\"\n",
       "                  Results: Ordinary least squares\n",
       "====================================================================\n",
       "Model:              OLS              Adj. R-squared:     0.764      \n",
       "Dependent Variable: TempMax          AIC:                177517.6861\n",
       "Date:               2019-03-28 13:45 BIC:                177576.7868\n",
       "No. Observations:   34302            Log-Likelihood:     -88752.    \n",
       "Df Model:           6                F-statistic:        1.849e+04  \n",
       "Df Residuals:       34295            Prob (F-statistic): 0.00       \n",
       "R-squared:          0.764            Scale:              10.351     \n",
       "---------------------------------------------------------------------\n",
       "                 Coef.   Std.Err.     t      P>|t|    [0.025   0.975]\n",
       "---------------------------------------------------------------------\n",
       "Intercept       -0.9463    1.9238   -0.4919  0.6228  -4.7170   2.8243\n",
       "Mes             -0.0936    0.0051  -18.4578  0.0000  -0.1035  -0.0836\n",
       "Dia              0.0003    0.0020    0.1414  0.8876  -0.0036   0.0041\n",
       "Anio             0.0066    0.0010    6.7268  0.0000   0.0047   0.0085\n",
       "Evaporacion      0.0398    0.0086    4.6478  0.0000   0.0230   0.0566\n",
       "Calor            1.6452    0.0052  317.4160  0.0000   1.6351   1.6554\n",
       "Precipitacion   -0.0794    0.0030  -26.4843  0.0000  -0.0852  -0.0735\n",
       "--------------------------------------------------------------------\n",
       "Omnibus:              9598.858      Durbin-Watson:         0.247    \n",
       "Prob(Omnibus):        0.000         Jarque-Bera (JB):      30931.320\n",
       "Skew:                 -1.425        Prob(JB):              0.000    \n",
       "Kurtosis:             6.677         Condition No.:         218038   \n",
       "====================================================================\n",
       "* The condition number is large (2e+05). This might indicate\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lm.summary2()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def regresion_validation(X_data, Y, Y_pred):\n",
    "    SSD = np.sum((Y - Y_pred)**2)\n",
    "    RSE = np.sqrt(SSD/(len(X_data)-1))\n",
    "    y_mean = np.mean(Y)\n",
    "    error = RSE/y_mean\n",
    "    print(\"SSD: \"+str(SSD)+\", RSE: \" +str(RSE) + \", Y_mean: \" +str(y_mean) +\", error: \" + str(error*100)+ \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regresión de grado 2\n",
      "R2:0.90754877554136\n",
      "[-1594.15788707]\n",
      "[[-2.07588348e-10 -4.51223713e+00 -2.02294747e-03  1.57373685e+00\n",
      "  -1.12616471e+01  2.07891243e+01 -5.80898836e-01  1.25632264e-01\n",
      "   6.81681518e-03  1.51394851e-03 -1.31231314e-02 -4.24003088e-02\n",
      "  -9.06002049e-03 -1.13605254e-04 -1.18568238e-05 -3.72275859e-04\n",
      "  -2.06927486e-03 -3.54379388e-04 -3.86182254e-04  5.70682790e-03\n",
      "  -8.58175608e-03  2.99780808e-04  8.86310099e-03  8.22172349e-03\n",
      "  -4.08299656e-03 -1.11826600e-01  1.17835383e-04  1.03313065e-03]]\n",
      "SSD: TempMax    138953.250941\n",
      "dtype: float64, RSE: TempMax    2.012709\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    7.482733\n",
      "dtype: float64%\n",
      "Regresión de grado 3\n",
      "R2:0.9330909044501062\n",
      "[-2326.95609659]\n",
      "[[-3.69834211e-03  1.00057728e+02  3.86620931e+01  6.00713407e-01\n",
      "  -2.88567101e+02  4.28833272e+02 -5.52248831e-01 -2.42239989e+00\n",
      "  -7.14783856e-03 -8.53102699e-02  3.50579876e-01 -5.27017086e-01\n",
      "  -1.00217285e-01  3.26079313e-02 -3.93085987e-02  2.48471457e-01\n",
      "  -1.56721780e-01 -3.70864003e-02  1.15784235e-03  2.82477494e-01\n",
      "  -4.21377945e-01  2.36331732e-03  1.60593037e+00 -9.24965806e-01\n",
      "   1.02137784e-01 -6.46957876e-01 -3.35099726e-01 -2.70117041e-03\n",
      "   1.20091310e-02  6.16782381e-04  1.16609501e-03  1.08134591e-03\n",
      "  -5.97687428e-03 -7.72217886e-04  5.11959949e-05  1.39314796e-06\n",
      "   4.73851094e-04 -3.74057350e-04  2.83164194e-04  1.74501783e-05\n",
      "  -1.72915179e-04  3.02011808e-04  4.50624379e-05 -1.84646338e-03\n",
      "  -3.07210204e-03  6.98714207e-05  6.98323601e-04  1.96161870e-03\n",
      "   9.46962653e-05 -1.71378905e-05 -1.67851945e-05  1.32591643e-04\n",
      "   5.95184628e-05  3.01353986e-05  9.99306057e-06 -1.26904111e-04\n",
      "   7.97041800e-05  1.68322171e-05  2.97265253e-04 -7.51969279e-04\n",
      "  -8.13401283e-05  1.05241650e-04  9.54162892e-06  3.44491287e-05\n",
      "  -4.37563671e-07 -6.92119139e-05  1.04886965e-04 -9.92049934e-07\n",
      "  -7.59664106e-04  4.92666911e-04 -4.55585618e-05  8.71941758e-05\n",
      "   1.35294953e-04  2.49028319e-06 -3.56538480e-03 -2.78119729e-03\n",
      "   2.07937460e-04  8.00821455e-04 -1.21030442e-03 -2.43138576e-05\n",
      "   1.43242073e-02  2.86607764e-03  7.55951769e-05 -3.47368389e-05]]\n",
      "SSD: TempMax    100563.690731\n",
      "dtype: float64, RSE: TempMax    1.71225\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    6.365705\n",
      "dtype: float64%\n",
      "Regresión de grado 4\n",
      "R2:0.9302360618098559\n",
      "[-15301.99767029]\n",
      "[[ 2.69898559e-04 -6.91363254e-01 -1.34428981e+00 -1.08033789e+00\n",
      "  -4.28359316e-01  1.24452975e+00  1.83130328e-01  4.96816246e+01\n",
      "   2.36835509e+00 -2.98112480e-01  5.58121525e+01 -1.68222188e+01\n",
      "  -7.06199359e+00  7.77918135e-01  6.11995512e-02  1.61194986e+01\n",
      "  -1.97610521e+01  1.12093293e+00  2.45429386e-02 -2.67430033e-01\n",
      "  -1.71290211e-02  2.26837044e-02  5.74626565e+01 -1.26220196e+02\n",
      "   2.71898285e+01  7.26417720e+01 -1.22371267e+01  2.18875473e+00\n",
      "   1.20123960e-01  9.33493358e-04 -5.21696934e-02 -5.57491942e-02\n",
      "   1.22117388e-01  1.11991212e-01 -1.16818209e-02 -1.98959276e-03\n",
      "  -3.99740318e-02 -1.23584708e-02  2.99259107e-02  3.09557956e-04\n",
      "  -5.46920857e-02  1.52205287e-02  5.80042892e-03  7.58032315e-02\n",
      "  -1.69425342e-01 -1.06882273e-01  9.96790246e-02  7.79143381e-02\n",
      "   8.98066513e-03 -2.98608849e-03 -6.79796536e-04 -3.99519503e-04\n",
      "  -5.70501309e-03  1.02309148e-02 -6.32717240e-05 -1.52676192e-02\n",
      "   2.00008630e-02 -1.26419112e-03  7.16040861e-02 -1.41487468e-01\n",
      "   2.18631288e-03  3.48889367e-02 -3.88439458e-04 -6.30035102e-04\n",
      "  -1.60887100e-05  2.65179109e-04  3.72140731e-05 -2.10823579e-05\n",
      "  -5.80782781e-02  1.24749007e-01 -2.61464437e-02 -7.66472373e-02\n",
      "   1.17505744e-02 -2.33891380e-03  1.21570675e-02 -8.26848663e-02\n",
      "  -4.37475294e-03  4.83428066e-01 -1.21345753e-01  1.66232654e-02\n",
      "   1.03258897e-01  7.78022146e-02  1.30526026e-02 -2.57215966e-04\n",
      "  -2.15845315e-03 -8.30938196e-04 -3.08939190e-05 -1.59217420e-04\n",
      "   2.83633163e-03 -7.79333717e-04  2.71853825e-04  2.86528251e-06\n",
      "   7.30426470e-06  2.57809717e-04  1.45706012e-04  1.35882023e-05\n",
      "   3.59625140e-05 -1.00802809e-04 -5.11517929e-05 -5.42210310e-03\n",
      "   4.01973410e-03  2.97425369e-04  4.37756382e-04 -8.80142580e-06\n",
      "   9.42026754e-05 -1.74544271e-06  4.16977804e-06  6.54484413e-05\n",
      "  -2.78249199e-05  2.88870908e-04  4.08467724e-07  1.89855780e-05\n",
      "   6.92657658e-06 -2.11373598e-05 -5.16484030e-04  5.69110888e-04\n",
      "  -1.14329497e-04 -4.04736044e-04  2.30578860e-05  5.09817974e-05\n",
      "  -8.03052984e-08  1.33838441e-05 -3.25824118e-06 -1.11117994e-06\n",
      "  -1.08595168e-05  4.62194515e-05  5.11062720e-05 -5.20067590e-05\n",
      "  -4.16672634e-05 -5.44365602e-06 -2.04433528e-04  3.07261645e-03\n",
      "   4.93582683e-04 -4.27919362e-04 -5.77610587e-04  2.03117648e-04\n",
      "   2.92548194e-04  1.98392901e-04 -2.92437277e-05 -1.07505234e-05\n",
      "   1.84315951e-06  1.40989000e-06 -2.01826352e-05  1.50900315e-05\n",
      "   1.24114760e-05  1.47996271e-07 -1.01398091e-06  2.70772729e-06\n",
      "  -6.09921577e-06  7.68201807e-05  2.14043408e-04  5.27712687e-05\n",
      "   8.29255719e-06 -7.13393007e-05 -6.36088892e-06  1.65000293e-08\n",
      "   3.61855446e-06 -5.06413494e-06  3.66623060e-07 -3.65821096e-05\n",
      "   6.75767109e-05  1.79584450e-06 -1.65362339e-05  8.06455534e-07\n",
      "   1.61423595e-07  4.30416393e-04 -2.82573185e-04 -4.78080096e-04\n",
      "  -7.16621108e-06 -2.68365850e-04  3.80542928e-05 -7.79063691e-06\n",
      "   5.88895929e-05  2.08701115e-05 -2.55496510e-06  3.00049699e-09\n",
      "  -6.58420269e-08 -1.37986333e-08  4.79147300e-09  1.47022377e-05\n",
      "  -3.07271878e-05  6.30387341e-06  1.99683098e-05 -2.84349994e-06\n",
      "   6.27707485e-07 -1.15316922e-05  2.88153643e-05 -3.63511053e-06\n",
      "  -2.40748484e-04  5.78046431e-05 -1.20144013e-05 -2.75084980e-05\n",
      "  -3.37307892e-05 -5.15501631e-06 -1.00213677e-08  7.63487914e-04\n",
      "  -1.93271407e-03  3.59158557e-04  1.97218748e-03  2.75672147e-04\n",
      "   1.69065860e-04 -6.17218847e-04  3.97497101e-04  2.80594364e-04\n",
      "   5.38008739e-06 -1.10758712e-03 -2.79024455e-04 -2.15747015e-04\n",
      "   1.00544193e-05  1.93019608e-06]]\n",
      "SSD: TempMax    104854.49021\n",
      "dtype: float64, RSE: TempMax    1.748397\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    6.500091\n",
      "dtype: float64%\n",
      "Regresión de grado 5\n",
      "R2:0.9413129199088164\n",
      "[-7094.07679988]\n",
      "[[ 1.89006722e-02 -2.67164120e-03  9.40566330e-04  2.11077368e-04\n",
      "   7.02786011e-05  6.29622793e-05  1.87198458e-05  3.63675419e-06\n",
      "  -5.39320520e-07 -1.21870926e-04  2.55517372e-05 -2.13359074e-05\n",
      "  -2.25455749e-05 -3.75840585e-06  2.92376834e-04  1.23299362e-05\n",
      "  -1.09058519e-05 -3.81128636e-06  2.50310669e-03  8.71938079e-06\n",
      "   2.73124875e-04 -7.58361504e-05  7.27272662e-05 -6.63808160e-05\n",
      "   1.76128683e-05  5.01298132e-05  1.14827357e-05  4.04709655e-07\n",
      "   2.76620495e-04  8.45702351e-04  3.03851001e-02  8.86370560e-05\n",
      "  -4.71266120e-05 -1.64168187e-04 -2.79939413e-03  2.82534615e-04\n",
      "  -4.67501186e-04 -6.49254870e-04 -5.75814869e-05 -1.30551731e-04\n",
      "   1.63926951e-02 -1.59340759e-02 -1.22734808e-02  6.56737392e-05\n",
      "  -1.41566787e-05  5.81889200e-05 -5.23423823e-05 -2.72717333e-05\n",
      "  -7.47587548e-05  3.93211414e-04  1.95187351e-04  1.54832819e-04\n",
      "  -1.04892926e-04 -2.06326765e-03  3.04507764e-05  4.14863995e-03\n",
      "  -8.85438536e-03 -1.98349871e-03 -4.49407630e-04 -5.08712609e-04\n",
      "  -1.56481274e-04 -7.96016108e-05  2.86009099e-04  4.45940557e-03\n",
      "   5.09570883e-06 -2.29972760e-04  1.56083972e-05  2.92216327e-06\n",
      "   4.83023545e-02 -4.38483263e-02  1.14083995e-02  3.16689906e-02\n",
      "   7.87115256e-03 -2.27229990e-04  2.25906057e-04  1.47405926e-04\n",
      "   1.08854639e-04 -2.48070141e-04  8.63027213e-05  6.64537302e-05\n",
      "   7.33227366e-04  2.80359429e-04  2.97304946e-04  3.20874352e-03\n",
      "   8.38350193e-02  1.36139135e-02 -1.18846509e-03  2.30519080e-02\n",
      "  -2.34767226e-02 -9.55781674e-03  2.45190435e-03 -2.56910571e-04\n",
      "  -5.90413676e-03  1.87397605e-02  3.78877627e-03 -2.74844843e-05\n",
      "  -1.46178010e-04  1.08810571e-03  5.94012622e-05 -1.21325945e-02\n",
      "  -2.19855454e-02  1.94289916e-02 -8.30234324e-02  1.70956369e-02\n",
      "  -3.20350057e-03  4.32466717e-04 -7.96138750e-06 -6.76398963e-03\n",
      "  -2.23833108e-03  3.38550753e-03  5.76505047e-07  8.63909492e-05\n",
      "  -3.31676946e-05 -8.62996057e-05  5.55292320e-03  4.22522424e-03\n",
      "   5.46110716e-04 -2.54941350e-03  2.68948474e-03  6.49467036e-05\n",
      "   1.38257508e-07 -1.51141349e-05  1.01758212e-05  1.19558897e-05\n",
      "  -3.53944794e-04 -2.96331368e-05 -4.06003113e-04  2.99708242e-04\n",
      "   1.01188172e-04  8.47979992e-05  1.48490000e-02  1.11173517e-02\n",
      "   2.65741510e-02  1.77445703e-02  5.70886756e-03  1.72009209e-03\n",
      "   2.37826038e-02 -1.67343285e-02 -7.63389830e-03 -5.67556831e-04\n",
      "  -2.10362239e-04  2.26827358e-06  3.75993857e-04  2.84555516e-04\n",
      "   7.42298282e-04 -1.42604407e-07 -1.02481660e-05 -4.52954321e-05\n",
      "  -3.46188471e-05 -1.37838552e-03  4.18584214e-03 -9.44043254e-04\n",
      "   6.86164892e-03  9.36826839e-05 -1.56242117e-04 -3.22669817e-08\n",
      "  -3.01796265e-06  9.57553618e-06  2.65313420e-06 -1.31736173e-04\n",
      "  -3.52354181e-04 -1.22337843e-05 -9.93878298e-05 -1.06830146e-04\n",
      "   5.96618085e-06  3.92693640e-03  2.77522136e-02  9.35168924e-03\n",
      "   1.14176094e-02 -4.11395957e-03 -8.20644415e-04  1.22669607e-03\n",
      "   1.22328207e-02 -1.23693905e-03  3.48219222e-05 -4.77281243e-09\n",
      "   2.31600097e-07  8.18853049e-09 -3.27227498e-10 -5.06300197e-05\n",
      "   4.29166010e-05 -1.10527880e-05 -3.96947326e-05 -9.78806518e-06\n",
      "  -2.61821603e-07  5.47575874e-04  3.59752754e-04 -3.00776518e-05\n",
      "   3.22136980e-04  2.21755801e-04  2.08474635e-05  1.20733733e-03\n",
      "   3.74979328e-04  7.85397530e-05  1.38789778e-06 -3.19214683e-02\n",
      "  -3.44846617e-02 -2.18181297e-02 -3.10646769e-02  8.52843960e-03\n",
      "  -6.61721875e-04 -1.55076976e-02 -2.09655148e-02 -1.57763999e-03\n",
      "  -3.68434526e-04 -7.16267614e-02 -2.86161974e-02 -8.32534291e-04\n",
      "  -1.02160716e-03  5.29394701e-05 -4.64045551e-04 -7.57329169e-05\n",
      "  -3.42000613e-05 -1.58561348e-04 -2.97257236e-04  1.96390125e-05\n",
      "  -5.59177646e-06 -5.70689899e-06  1.34946302e-05 -7.26274699e-05\n",
      "  -3.81530767e-06  5.46971772e-07 -1.03488707e-05  1.69213025e-05\n",
      "   4.98359219e-06  3.43645462e-04 -2.16661509e-04 -3.70833907e-05\n",
      "   1.31530291e-04 -3.07495180e-04  3.46551990e-05 -3.61691374e-05\n",
      "  -3.42674119e-07 -2.62235412e-05 -5.28300134e-06 -7.54032357e-06\n",
      "   1.18463633e-07  2.97974343e-06 -8.59026471e-06 -1.75299573e-06\n",
      "  -5.72620578e-05  9.17325177e-05 -1.48855489e-05 -8.39235437e-06\n",
      "  -4.08480596e-05  7.88516274e-06  6.32399603e-09  6.81075358e-08\n",
      "  -5.95730591e-07 -3.37623119e-08  2.32225663e-06  1.64427101e-05\n",
      "  -1.10783181e-05  4.51570570e-05 -5.83490403e-06  1.52196842e-06\n",
      "   3.71318717e-04 -6.58487633e-04  5.15916235e-05 -1.22564558e-04\n",
      "   2.86998249e-04  3.25535331e-05 -4.15594477e-04  9.62950052e-05\n",
      "  -2.70202746e-05 -7.86051523e-06  1.35129808e-07  1.33108834e-08\n",
      "   1.27370480e-05 -6.17529672e-06 -4.63501688e-06 -8.49082447e-10\n",
      "   3.40119286e-06  1.22778085e-06 -1.50030156e-06 -5.09032782e-05\n",
      "   9.77913522e-06  1.56897475e-05  1.48736990e-05 -1.05730450e-05\n",
      "  -4.52951595e-07 -3.17352576e-10 -4.39831530e-08  1.29629761e-08\n",
      "   4.21878210e-08 -1.82175100e-06 -2.65143029e-06 -6.77980771e-07\n",
      "   1.46429648e-06 -1.05109807e-06 -1.33535705e-07  1.13707971e-05\n",
      "   4.83971019e-06 -3.15458616e-06 -1.06513649e-05  5.52137681e-05\n",
      "   3.77351685e-06 -3.56860026e-05 -1.56469480e-05  1.53767739e-05\n",
      "  -1.46582296e-06 -3.69569273e-11  3.48329085e-09 -9.01812952e-10\n",
      "  -2.89723133e-09  1.94182745e-07 -1.87419721e-08  2.16770218e-07\n",
      "  -1.75200266e-07 -5.58622246e-08 -4.32678931e-08 -1.14662915e-05\n",
      "  -1.07729755e-06 -1.48547591e-05 -7.46073889e-06 -3.99092922e-06\n",
      "  -8.97283784e-07 -9.81183084e-06  7.26282752e-06  3.87847558e-06\n",
      "   3.55472117e-07  1.08087906e-04  1.67430159e-05  1.30695183e-04\n",
      "  -1.59469257e-05 -2.91992456e-05  1.43655702e-05 -1.98152749e-05\n",
      "  -1.11237812e-04 -2.76289233e-05 -5.54382783e-06  5.76785138e-05\n",
      "   5.61607359e-05 -1.71149695e-06  3.06396023e-06 -5.13781330e-08\n",
      "   8.15649578e-08  1.12682896e-07 -1.60078018e-06 -1.30043399e-06\n",
      "  -2.08931935e-07 -1.77512449e-09 -2.74341956e-07  8.02723905e-09\n",
      "  -3.91865218e-07  7.88486486e-06  9.25992333e-06  1.84745296e-06\n",
      "  -1.51855380e-05  4.55271747e-06  4.21750508e-07  3.03842084e-11\n",
      "   7.96620569e-09  2.15729363e-08  1.82784537e-08  8.03967037e-07\n",
      "  -2.83252953e-06  2.65499900e-07 -3.55133247e-06 -1.47937332e-07\n",
      "   8.04552768e-08 -3.11326935e-05  8.39762180e-06  3.74477573e-05\n",
      "   5.39346304e-05 -6.35627163e-06 -1.30875282e-06  2.70361936e-05\n",
      "   4.70084608e-06 -8.59927267e-07 -5.99459468e-08  8.44588917e-12\n",
      "   4.43538532e-10 -2.57271653e-09 -8.36330207e-10  5.94185831e-08\n",
      "   1.90551999e-07  1.14001764e-08  4.95470200e-08  5.45461921e-08\n",
      "  -4.24053799e-09 -3.93392655e-07 -1.44070452e-05 -5.57340619e-06\n",
      "  -7.16722965e-06  1.39685331e-06  4.61716333e-07 -1.51069661e-07\n",
      "  -6.24294019e-06  6.59608064e-07 -1.04929227e-08 -5.66982454e-05\n",
      "  -6.07363752e-05  1.94102312e-05  5.98305497e-05  1.02000171e-05\n",
      "   9.38777067e-07  2.58330154e-05  6.77670985e-05 -5.15730557e-06\n",
      "  -3.23998249e-07 -4.62980818e-05  1.08266643e-06 -6.53206874e-06\n",
      "   4.36530154e-08 -5.59046345e-10  1.02179550e-12 -5.83589169e-11\n",
      "  -7.75123301e-12 -5.48427162e-13  1.32995784e-08 -1.04740309e-08\n",
      "   2.67068778e-09  1.17866920e-08  2.90569021e-09  1.91347691e-10\n",
      "  -2.90778011e-07 -1.84052356e-07  1.84733501e-08 -1.66474340e-07\n",
      "  -1.23678257e-07 -1.28590630e-08 -5.63914881e-07 -1.81346172e-07\n",
      "  -3.69632845e-08 -1.79823577e-09  1.76202737e-05  1.78359576e-05\n",
      "   1.11703232e-05  1.54049217e-05 -3.60596849e-06  2.58413554e-07\n",
      "   8.94351195e-06  1.30576884e-05  1.28761918e-06  2.46979287e-07\n",
      "   3.04034349e-05  1.38599115e-05 -2.30631866e-08  4.56248002e-07\n",
      "  -1.97408616e-08 -6.04487322e-05 -1.78457949e-05 -6.92749440e-06\n",
      "   3.96167420e-05 -8.96185297e-05 -5.49733467e-06 -6.41517165e-05\n",
      "  -3.43100299e-05  1.95949901e-05 -1.69011431e-06 -5.25044893e-05\n",
      "  -1.40073490e-04 -2.61318213e-05 -4.00920145e-06 -1.03189890e-07\n",
      "   2.55147439e-04  1.65142444e-05  2.02014747e-05  6.13914005e-06\n",
      "  -1.89028323e-07 -5.95923050e-08]]\n",
      "SSD: TempMax    88206.085042\n",
      "dtype: float64, RSE: TempMax    1.603599\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    5.96177\n",
      "dtype: float64%\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regresión de grado 6\n",
      "R2:0.9433188163972992\n",
      "[-3591.93884192]\n",
      "[[ 1.41026675e-06  5.42631267e-08 -1.19195536e-08  8.91722515e-10\n",
      "  -4.40938927e-10 -3.82412508e-10 -4.25515111e-11  9.79972336e-11\n",
      "   1.95339845e-11 -8.92617448e-12 -1.17357735e-12  2.75747015e-12\n",
      "  -2.06790193e-11  1.37882238e-11 -2.28579214e-11  6.78565670e-12\n",
      "  -4.35363820e-12 -1.53267496e-11 -9.25407773e-10 -2.34151963e-10\n",
      "   4.00288265e-10  9.92337058e-12  3.72767176e-11 -3.12099153e-11\n",
      "   1.13383392e-11  2.24255827e-11  9.25156466e-12 -1.33871653e-12\n",
      "   1.82578846e-10  1.85770657e-10  1.43368882e-08  9.29648199e-11\n",
      "   3.03789393e-11  1.52845800e-11 -4.25706338e-12  3.67684740e-09\n",
      "   1.78409549e-11 -1.42402231e-10 -5.79478815e-11 -1.89830981e-08\n",
      "   9.61357273e-09 -7.75356120e-09 -3.49575256e-09  1.00681534e-10\n",
      "  -6.60803701e-11  4.46613564e-11  2.46801995e-11  4.07106277e-11\n",
      "  -8.71705214e-11  6.15538068e-10  3.00154292e-10 -1.49678033e-10\n",
      "  -1.89853079e-10  1.22504329e-10 -1.00381886e-09  4.69307232e-10\n",
      "  -6.92578765e-09 -1.19442268e-09  2.33817214e-10 -3.46574326e-10\n",
      "  -1.63761408e-11  1.22275116e-10  1.37842761e-10  2.57978563e-10\n",
      "  -6.81740509e-07 -1.21038876e-07  1.97853873e-07  1.23705729e-08\n",
      "   3.26653199e-08 -3.22170949e-08  1.00538736e-08  2.23493692e-08\n",
      "   9.12847934e-09 -4.58545436e-10  2.10740155e-10  8.59187227e-11\n",
      "   7.42101247e-11 -2.88149459e-10  4.19456321e-11  2.42198796e-10\n",
      "   4.29994493e-10  2.52704915e-10  1.96098118e-10 -2.27127759e-10\n",
      "   3.72863729e-08  6.88609829e-09  1.20784869e-07 -1.25006451e-09\n",
      "  -6.86948351e-09  5.15121156e-10  2.45367495e-09  1.22780637e-07\n",
      "  -3.53136129e-09  1.07054662e-08 -1.45371689e-09  9.56143568e-06\n",
      "   6.19413498e-08  1.97506148e-08  1.00138855e-08 -3.57149623e-09\n",
      "  -8.12463653e-09 -3.18669978e-09 -3.39181019e-08 -2.52184710e-09\n",
      "   1.67807677e-09  1.66905470e-09 -2.57539738e-09 -4.53009436e-09\n",
      "  -4.61381104e-09  9.77855218e-10  2.45170547e-06  1.27263170e-08\n",
      "  -9.66977158e-08 -3.86752103e-08  2.69757774e-09 -1.13089714e-10\n",
      "   6.07728928e-10 -2.53512908e-09  3.23763446e-09  5.47752203e-10\n",
      "  -5.53571853e-08  6.48999275e-06 -5.43120067e-06 -2.30857804e-06\n",
      "   6.69810146e-08 -4.30741002e-08  2.96374852e-08  1.20562833e-08\n",
      "   2.63495735e-08 -5.56716455e-08  5.49741331e-09  4.21975162e-09\n",
      "   6.16787387e-10  3.28452507e-09 -5.83066193e-09  3.47515287e-09\n",
      "   1.61184544e-08 -5.28003217e-09 -5.68857759e-09 -2.79675264e-09\n",
      "   1.30158471e-08  3.98808766e-07 -2.20629895e-09 -4.31426852e-10\n",
      "   2.34495755e-09  3.00614582e-08 -9.37742303e-08 -1.37203366e-07\n",
      "   7.69783344e-08 -3.86935446e-10 -1.85308461e-09 -8.19775386e-10\n",
      "   4.19455333e-09  2.63861168e-09  5.99667020e-09  2.32415337e-08\n",
      "   8.51212166e-07 -5.65697476e-06 -1.01965172e-06  1.56214462e-07\n",
      "  -2.26404456e-07 -1.06313711e-08  6.85606952e-08  8.80167203e-08\n",
      "   1.70148768e-07  7.41816575e-10  1.38351211e-08  1.31069368e-09\n",
      "   1.21715063e-09  1.07559830e-09  7.63653734e-10  4.19711698e-09\n",
      "   4.97507117e-09  3.75662797e-09  4.18429147e-09  4.42079889e-09\n",
      "  -1.18020614e-07  1.91648433e-08 -2.51819977e-08  2.18314501e-05\n",
      "  -2.09142204e-05  6.68841321e-06  1.32699037e-05  5.82740254e-06\n",
      "   4.89937947e-08  1.40391437e-07  5.79983757e-08  4.92127512e-08\n",
      "  -1.88516102e-07  2.81518282e-08  1.60546298e-07  2.70633165e-07\n",
      "   1.64477817e-07  1.29816897e-07 -1.36161982e-07 -1.69607521e-08\n",
      "  -1.26424684e-08  1.45692632e-10 -6.86588747e-09 -3.95605207e-10\n",
      "  -6.96648951e-10 -1.22839265e-08 -1.58712165e-08  2.40081925e-10\n",
      "   2.81514956e-09 -3.30477434e-08 -1.27704303e-08  5.95831385e-10\n",
      "   1.13064247e-09 -7.91092437e-09  5.90665173e-07  3.34652816e-07\n",
      "   3.56573792e-05  1.63226721e-08  6.59810113e-08  5.20282399e-09\n",
      "   1.09826153e-07  5.30876426e-06 -6.10245797e-08  1.98406983e-08\n",
      "  -2.10214633e-08 -5.38714502e-07 -1.70814033e-06 -7.17694895e-06\n",
      "   3.38896861e-07 -4.55502044e-08 -9.14770542e-08 -2.63442413e-08\n",
      "  -3.81650567e-07 -2.11585743e-08  3.90316850e-09  8.68076048e-08\n",
      "   9.71614694e-07 -9.60016733e-08  1.39719076e-07 -2.25132998e-08\n",
      "  -1.26217555e-07 -3.81647228e-06  1.05242491e-05 -1.36289615e-06\n",
      "  -3.38466518e-08 -5.88576717e-08 -2.90663967e-08 -1.84024366e-07\n",
      "  -1.95299003e-09  2.92079125e-09 -7.73941559e-09  9.47881522e-08\n",
      "   4.06019379e-07  3.53160038e-08 -3.86142454e-06 -8.05774306e-06\n",
      "  -3.30717919e-06 -3.35005723e-05 -2.67978141e-06  1.93976508e-06\n",
      "   1.94106729e-08 -1.02044868e-08 -8.28308317e-09 -5.18537205e-08\n",
      "  -4.93867776e-08  1.91499314e-08 -1.93340712e-07 -4.33267195e-08\n",
      "  -4.19931834e-08 -1.71388628e-09 -2.79943349e-07  4.21304383e-07\n",
      "  -4.99753942e-08 -9.47334171e-08  4.69276898e-08 -2.65463660e-09\n",
      "  -4.11916083e-06 -2.91454899e-06  1.26637248e-06  2.90915615e-08\n",
      "   1.30648537e-08  6.72414397e-09 -3.69843666e-08  4.78162859e-08\n",
      "   2.15342511e-08 -2.08154508e-09  8.95802676e-08 -1.48284078e-09\n",
      "  -1.77361792e-08  2.31559134e-06  5.93525742e-07  6.22047065e-07\n",
      "  -2.07791041e-06  3.05810164e-06  6.84361184e-07  3.80646448e-08\n",
      "   7.55455166e-08  1.51438870e-08  4.72317177e-08 -3.31159711e-08\n",
      "   4.26927534e-08  1.32596302e-07 -7.93435787e-08 -3.83279250e-08\n",
      "  -5.80879405e-08  5.62373524e-11 -6.77686478e-09  3.49799328e-09\n",
      "   2.06807235e-09 -6.62366953e-08 -5.95113041e-08 -1.00037436e-07\n",
      "   4.87836837e-08  8.46209204e-08  8.76117738e-09  5.08457328e-06\n",
      "   4.03801428e-06  4.22074478e-07  3.78814969e-06 -5.94438366e-06\n",
      "   2.81313665e-06  1.50484116e-05 -5.86400003e-06 -5.83558545e-06\n",
      "  -9.11342129e-08 -3.74928303e-09  1.66672363e-08  1.07665524e-08\n",
      "   3.34271196e-08 -2.23652423e-08  9.68902105e-10  1.63681251e-08\n",
      "  -1.07044754e-07 -3.97840293e-08  2.41848556e-07  9.05758192e-08\n",
      "  -1.16551450e-07 -1.29856810e-07 -2.48673108e-07 -2.83778185e-07\n",
      "   2.59248495e-07 -3.32486359e-07 -1.63741701e-08  6.21582003e-08\n",
      "  -2.64244927e-08  7.60031285e-09  1.86916980e-07 -5.42981356e-08\n",
      "  -1.04363632e-09 -1.64699345e-07  1.31530219e-08  1.24596783e-07\n",
      "   3.47299238e-07  7.26214360e-08  2.65731838e-07 -5.65666257e-11\n",
      "   7.58637380e-09 -2.29587696e-08 -5.23255912e-09 -1.60701406e-06\n",
      "   1.35804632e-06 -3.50116249e-07  3.89918969e-06  9.23114448e-07\n",
      "  -1.49428711e-07 -4.63466835e-08  2.19605782e-07  1.57328893e-08\n",
      "   7.21366373e-08  3.45756016e-08 -3.25325651e-07  2.35176814e-07\n",
      "   1.05576872e-07  3.08108406e-07  9.17867364e-07 -2.43471080e-11\n",
      "  -5.08042216e-10  6.19313538e-09  1.21744139e-09 -6.53637284e-08\n",
      "  -1.33043393e-07 -2.21215726e-08 -7.63356860e-08 -5.80297888e-08\n",
      "  -3.93773102e-09 -2.91033908e-07  1.35176423e-05  1.16423810e-06\n",
      "   3.28117975e-06  1.20058339e-06  2.28160069e-07  2.18307575e-06\n",
      "   3.04126154e-06  6.81065798e-07 -9.84423949e-08 -1.36238970e-07\n",
      "  -1.62998171e-08 -4.55862129e-08  9.59382897e-08  5.99109928e-09\n",
      "   7.88109370e-08 -7.50005524e-08 -1.10603330e-07 -2.40770193e-08\n",
      "   8.59933544e-08 -1.24945197e-07 -5.57300221e-08  3.26258501e-08\n",
      "  -3.39384795e-09  2.00765390e-06 -3.23794898e-12  1.18979095e-10\n",
      "  -1.07413667e-11  2.62622521e-11 -2.32936190e-08  2.12572117e-08\n",
      "  -7.17548318e-09 -1.66579108e-08 -6.97746296e-09  3.41643523e-12\n",
      "   3.33267530e-07  7.36586784e-08  2.48093888e-08  1.21852312e-07\n",
      "   2.59581654e-07  5.19294877e-09  5.70523426e-07  2.25865758e-07\n",
      "   2.68911428e-08 -6.78044753e-09 -1.75072901e-05 -1.31087148e-05\n",
      "  -1.46107316e-07 -6.91496491e-06 -6.59211273e-07 -1.33164067e-06\n",
      "  -1.07958896e-05 -1.58933077e-05 -6.58789975e-07 -3.11404051e-07\n",
      "  -3.56740608e-05 -1.45030801e-05 -1.68273230e-06  1.08191585e-07\n",
      "   1.93815566e-07 -2.90175162e-07 -2.46001866e-07 -3.18129561e-08\n",
      "  -1.86639042e-07 -1.49286801e-08  1.47349775e-08 -1.26617122e-07\n",
      "  -5.13201848e-08 -1.32425693e-08 -2.03249673e-08 -2.07115736e-07\n",
      "  -2.32871267e-07 -1.47470627e-07  8.68882504e-08  1.61958624e-07\n",
      "  -7.01046186e-07 -2.21104836e-07 -1.36243000e-07 -1.08518878e-08\n",
      "  -9.16803646e-07 -1.50276772e-06  6.53083435e-06  5.77542524e-06\n",
      "  -4.05834338e-07 -1.20005957e-06 -7.56109137e-06 -3.24956640e-07\n",
      "   9.82425325e-07 -4.82669807e-08  2.86372579e-06 -2.21359564e-05\n",
      "   1.80485069e-06 -1.28873827e-08 -4.54579276e-08  1.86962530e-07\n",
      "   1.45286946e-08  2.28120959e-06 -5.48992828e-06  2.56137393e-06\n",
      "  -1.19738395e-05  1.96567816e-06  6.05787236e-06  1.65552297e-06\n",
      "  -6.05666696e-08  6.72181187e-07  1.04381520e-06 -6.06828206e-06\n",
      "  -2.34999793e-09 -1.36444054e-07  2.45308395e-07  8.45756178e-08\n",
      "   4.95716156e-06  1.34425699e-05 -3.50020464e-06  2.60253984e-06\n",
      "  -4.10203130e-06  9.76694531e-08  2.45746889e-10  2.51464529e-09\n",
      "   3.16575702e-09 -5.50878880e-10 -7.86702187e-08 -1.85640016e-07\n",
      "   5.65020162e-08  2.54799446e-08 -2.66486757e-07 -7.77712822e-08\n",
      "   9.27062311e-07  1.37682069e-06  3.14802305e-07  2.44476153e-06\n",
      "   1.24777342e-06  6.71262365e-06  1.56613871e-05  3.96736144e-06\n",
      "   1.14776933e-05 -8.63546435e-07 -5.94929553e-07  5.72087793e-09\n",
      "   8.08819910e-07 -8.81638639e-07 -2.83351226e-06 -2.68637015e-10\n",
      "  -2.06439392e-08  2.24631868e-08  1.14864768e-07  4.39892332e-06\n",
      "  -5.10289669e-06  3.92390583e-07 -8.81698221e-07  3.88836799e-07\n",
      "   1.37649332e-06  6.32667388e-11  1.96557579e-09 -6.59805286e-09\n",
      "  -6.56461026e-10  5.02293889e-08  2.07534937e-07 -2.65447459e-08\n",
      "   4.10113043e-08  9.74827993e-08 -3.58820547e-08 -1.16026762e-05\n",
      "  -1.00687021e-05  3.68910382e-06 -1.65332900e-05 -5.46510862e-07\n",
      "   2.60651193e-06 -6.52986543e-06  6.72327264e-07 -8.11197094e-07\n",
      "   4.99112351e-07  1.54023747e-12 -5.17286091e-11 -2.20158112e-10\n",
      "  -1.52589206e-11  6.03374081e-10  5.39244525e-09 -2.22454070e-10\n",
      "   2.19296845e-08  3.59007301e-09 -7.47782817e-11  3.60097288e-07\n",
      "  -3.38297418e-07  7.78551234e-08  1.13431432e-07  1.35326766e-07\n",
      "   5.92332928e-09 -8.87195330e-07 -5.64994293e-08 -1.41263561e-07\n",
      "  -1.64846124e-08 -1.79537956e-06  2.14760536e-07  4.63865812e-07\n",
      "   1.36350655e-06 -4.76613601e-07 -5.51365404e-06  5.19964163e-06\n",
      "  -4.30749246e-06  1.53547328e-07 -8.78731206e-07  3.83049315e-05\n",
      "   4.82284659e-06 -1.11444276e-06 -1.83321531e-07  4.24205029e-07\n",
      "   2.66640257e-07 -6.62976975e-09  1.82227681e-07  3.62141558e-08\n",
      "  -9.81411843e-07 -1.65385399e-10 -5.53238860e-09  2.34916785e-10\n",
      "   5.74116864e-08  1.24739316e-06 -1.18486766e-06 -3.09737029e-07\n",
      "   7.46135351e-07 -2.03032577e-06  5.78257376e-07  1.05516743e-12\n",
      "   2.08292767e-09  1.42612623e-09 -1.67639884e-09 -4.86975512e-08\n",
      "   7.89688081e-08 -3.06500123e-10 -1.04274739e-08  4.26210892e-08\n",
      "  -1.87573732e-08  7.47948842e-07 -3.81364747e-06  5.67189459e-06\n",
      "  -1.71796922e-06 -2.56789371e-06  6.75443485e-07 -4.08420619e-07\n",
      "   1.54206640e-06 -1.45993305e-06 -1.64047865e-08  4.23398969e-13\n",
      "  -4.00502580e-11  2.80127344e-12  1.67747072e-11 -1.62911529e-09\n",
      "  -2.45823303e-09  3.51921207e-10  1.07952401e-09 -2.19089755e-09\n",
      "  -1.08840547e-10  1.27725513e-07  7.05884565e-08 -7.44550905e-08\n",
      "   1.07737869e-07 -1.24700988e-07 -1.34770438e-08 -3.92416315e-09\n",
      "  -9.57098502e-08  6.35865457e-08 -8.66942334e-09 -7.80739080e-06\n",
      "   1.18328690e-06 -8.32845842e-06  1.56688974e-06  2.83051961e-06\n",
      "  -1.74856133e-06  3.80107647e-06  1.95628630e-05 -3.28466461e-06\n",
      "   4.96134700e-07  3.13923181e-06  4.00936355e-06 -1.35323671e-06\n",
      "  -6.10354528e-07  1.54863044e-07 -1.44022583e-14  1.74071116e-12\n",
      "  -2.99515082e-13 -4.62871153e-13  4.65572620e-11  2.67465394e-11\n",
      "   5.92046324e-11 -5.07874785e-11 -4.94209377e-11 -8.55688272e-12\n",
      "  -5.61974985e-09  3.33829299e-10 -2.47327290e-09 -1.66171003e-09\n",
      "   4.08101457e-09 -1.49467222e-09 -4.28535628e-09  3.96268009e-09\n",
      "   2.93048906e-09  2.02571674e-10  6.38294883e-08 -3.89148392e-08\n",
      "   2.72560451e-07 -3.73536202e-09 -5.03054643e-08  7.59025112e-08\n",
      "  -2.30708975e-07 -1.77659389e-07 -1.04181448e-07  1.99250157e-09\n",
      "   3.55664284e-08 -1.17070037e-07  9.32532235e-08  1.23571017e-08\n",
      "  -3.16056677e-09  3.70245042e-07  5.08325509e-06 -3.36496221e-06\n",
      "   3.26784746e-07 -1.63536631e-06 -3.36892674e-06 -3.50459010e-06\n",
      "   1.45189242e-06  9.15089761e-06 -1.30733695e-06  1.07867389e-05\n",
      "   3.31340495e-06  2.47265197e-06  6.58094522e-07  1.04402311e-07\n",
      "  -1.45717763e-05  4.17367357e-06 -8.06129015e-06  1.16729361e-06\n",
      "  -3.48696860e-07  8.49786069e-09  1.23089675e-07 -7.21558193e-09\n",
      "   2.07690755e-08  7.70897947e-08 -8.23785150e-08  3.25821122e-10\n",
      "  -5.93767208e-10 -5.52254794e-09  8.76343545e-09 -4.40270135e-07\n",
      "  -9.24635016e-08 -8.89441100e-08  2.38473581e-07 -3.11425365e-07\n",
      "   1.94111137e-08 -5.65490717e-12 -9.07622276e-11  1.63796988e-10\n",
      "  -2.13760208e-10  1.67151119e-08 -8.40677234e-09  1.17206660e-08\n",
      "  -7.96414942e-09 -1.79717838e-09 -5.09727512e-09 -8.65703094e-07\n",
      "   8.42770070e-07  1.08368414e-07  1.23951433e-06 -1.15442080e-06\n",
      "  -4.48040853e-08 -3.99446128e-07  1.80074187e-06  2.88237516e-08\n",
      "   4.75196762e-08  2.99985605e-14 -3.79646786e-12  1.05405252e-11\n",
      "   4.48919174e-12  8.43669639e-10 -6.25653449e-10  3.29282842e-12\n",
      "  -1.98048936e-09 -3.26158154e-10  1.92416696e-10  3.58887954e-09\n",
      "  -7.65155066e-08 -1.60937208e-08  9.85511146e-09  2.50662170e-08\n",
      "  -5.43970424e-09  2.01268074e-08 -3.22276442e-08  6.86511329e-09\n",
      "  -2.12308033e-09 -7.06993614e-07  3.69050687e-06  1.15277272e-06\n",
      "   3.87474782e-06 -2.87861496e-06  1.71579555e-07 -2.33377785e-06\n",
      "   1.64875693e-06  2.56560493e-07  6.93832168e-08 -5.48847215e-07\n",
      "  -1.22888180e-06 -4.86807603e-07  8.17307547e-08 -5.14256263e-10\n",
      "   6.35103598e-15  3.90006873e-14 -1.68307891e-12 -3.62023706e-13\n",
      "   2.74702506e-11  6.97957154e-11  1.24071942e-11  3.88301421e-11\n",
      "   2.77704087e-11  1.38067064e-12  2.92113796e-10 -4.21193610e-09\n",
      "  -4.87936610e-10 -3.03495921e-09 -1.20752946e-09 -1.14346592e-10\n",
      "  -1.17468491e-09 -6.94792103e-10 -6.93751698e-10  7.59503201e-11\n",
      "   2.26549736e-08 -1.51243286e-07 -2.83996100e-09 -2.26313277e-07\n",
      "  -2.72905618e-08  1.76100484e-08  1.94782045e-07  8.77819001e-08\n",
      "   2.25866036e-08 -3.07267672e-09  7.38006546e-10 -7.66955507e-08\n",
      "   2.98790040e-08  2.51126061e-09 -1.18267054e-09  2.64117771e-06\n",
      "  -5.76149149e-07  7.60996646e-07  2.85522507e-06  4.29705827e-06\n",
      "  -2.62829556e-07  8.00938976e-06  7.99605747e-06 -2.01338027e-06\n",
      "   3.14450194e-08 -9.64742120e-06 -1.08957613e-05 -1.63996428e-06\n",
      "   3.50719088e-07 -3.40403432e-08 -5.15421607e-07  4.45884119e-06\n",
      "  -1.05225099e-06 -9.51058290e-08 -1.46201973e-08 -3.27797515e-09\n",
      "   6.55258276e-16 -2.99467119e-14  8.23665010e-16 -6.80275700e-15\n",
      "   6.23639217e-12 -5.47137730e-12  1.90330925e-12  4.83302555e-12\n",
      "   2.01200607e-12 -1.85907549e-14 -1.87992056e-10 -3.41379916e-11\n",
      "  -7.38536509e-12 -2.20133254e-11 -1.32060786e-10 -2.19400631e-12\n",
      "  -2.20267198e-10 -1.06871862e-10 -5.74214573e-12  3.52602206e-12\n",
      "   1.41131153e-08  5.89359273e-09 -8.80208907e-10  7.74339585e-10\n",
      "   1.67278671e-09  4.84652622e-10 -3.76445279e-09  6.65425703e-09\n",
      "   4.37996001e-10  1.27051101e-10  6.16726044e-09  5.93265698e-09\n",
      "  -7.63173293e-10 -2.91485881e-10 -9.25452704e-11 -6.30512367e-07\n",
      "   8.92297928e-08  9.00185800e-08  1.64611027e-07 -1.46027053e-07\n",
      "   1.59636617e-08  2.72513161e-07 -7.45254546e-08 -1.51839660e-08\n",
      "  -4.52861521e-09  1.12724898e-06  3.07037616e-07  3.00666007e-08\n",
      "   7.62001769e-09  6.10148104e-10  9.53167095e-07  1.58556327e-07\n",
      "   1.38122174e-07  1.88180527e-08  4.05286132e-09  5.56543147e-10\n",
      "   2.65916388e-05 -4.63603713e-06 -5.12935812e-06 -4.28959064e-06\n",
      "   5.29517967e-06 -1.33102266e-06 -7.67701993e-07 -1.43449475e-06\n",
      "   1.36453154e-06  5.28023525e-07 -1.47782067e-05  4.88241185e-07\n",
      "  -6.24893322e-07 -6.80613942e-08  5.11266950e-08 -5.15228780e-05\n",
      "  -1.59564758e-05 -2.65002238e-06 -5.58174280e-07 -9.79762425e-08\n",
      "  -4.34900638e-09 -2.78056466e-05 -7.76911789e-06 -3.19470726e-06\n",
      "  -8.95115852e-07 -1.28615209e-07 -8.29568438e-09  2.05084633e-09]]\n",
      "SSD: TempMax    85191.243003\n",
      "dtype: float64, RSE: TempMax    1.575956\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    5.858999\n",
      "dtype: float64%\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regresión de grado 7\n",
      "R2:0.9443813585566408\n",
      "[-2270.31223268]\n",
      "[[-3.84180582e-11 -1.33156432e-11  3.89227681e-13 ...  2.74276465e-09\n",
      "  -4.43159772e-10 -5.24680022e-11]]\n",
      "SSD: TempMax    83594.252934\n",
      "dtype: float64, RSE: TempMax    1.561115\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    5.803823\n",
      "dtype: float64%\n",
      "Regresión de grado 8\n",
      "R2:0.9439248894478868\n",
      "[-1135.41747239]\n",
      "[[-3.50152183e-13  1.75025270e-14 -2.08489632e-15 ...  1.81754388e-13\n",
      "   8.58811487e-14  3.32494225e-12]]\n",
      "SSD: TempMax    84280.321366\n",
      "dtype: float64, RSE: TempMax    1.567508\n",
      "dtype: float64, Y_mean: TempMax    26.898044\n",
      "dtype: float64, error: TempMax    5.82759\n",
      "dtype: float64%\n"
     ]
    },
    {
     "ename": "MemoryError",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mMemoryError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-16-0bc3c3cd8e50>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      3\u001b[0m     \u001b[0mX_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpoly\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_transform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdXtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m     \u001b[0mlm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlinear_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLinearRegression\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0mlm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdYtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Regresión de grado \"\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0md\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"R2:\"\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscore\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdYtrain\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/base.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight)\u001b[0m\n\u001b[1;32m    463\u001b[0m         X, y, X_offset, y_offset, X_scale = self._preprocess_data(\n\u001b[1;32m    464\u001b[0m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfit_intercept\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_intercept\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnormalize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnormalize\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 465\u001b[0;31m             copy=self.copy_X, sample_weight=sample_weight)\n\u001b[0m\u001b[1;32m    466\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    467\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0msample_weight\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/base.py\u001b[0m in \u001b[0;36m_preprocess_data\u001b[0;34m(X, y, fit_intercept, normalize, copy, sample_weight, return_mean, check_input)\u001b[0m\n\u001b[1;32m    116\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mcheck_input\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    117\u001b[0m         X = check_array(X, copy=copy, accept_sparse=['csr', 'csc'],\n\u001b[0;32m--> 118\u001b[0;31m                         dtype=FLOAT_DTYPES)\n\u001b[0m\u001b[1;32m    119\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    120\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0msp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0missparse\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/sklearn/utils/validation.py\u001b[0m in \u001b[0;36mcheck_array\u001b[0;34m(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, warn_on_dtype, estimator)\u001b[0m\n\u001b[1;32m    596\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    597\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mcopy\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmay_share_memory\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0marray_orig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 598\u001b[0;31m         \u001b[0marray\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0morder\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0morder\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    599\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    600\u001b[0m     if (warn_on_dtype and dtypes_orig is not None and\n",
      "\u001b[0;31mMemoryError\u001b[0m: "
     ]
    }
   ],
   "source": [
    "for d in range(2,12):\n",
    "    poly = PolynomialFeatures(degree=d)\n",
    "    X_data = poly.fit_transform(dXtrain)\n",
    "    lm = linear_model.LinearRegression()\n",
    "    lm.fit(X_data, dYtrain)\n",
    "    print(\"Regresión de grado \" +str(d))\n",
    "    print(\"R2:\" +str(lm.score(X_data, dYtrain)))\n",
    "    print(lm.intercept_)\n",
    "    print(lm.coef_)\n",
    "    regresion_validation(X_data, dYtrain, lm.predict(X_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R2:0.8187752963613997\n",
      "SSD: TempMax    597.512802\n",
      "dtype: float64, RSE: TempMax    1.399664\n",
      "dtype: float64, Y_mean: TempMax    29.609477\n",
      "dtype: float64, error: TempMax    4.72708\n",
      "dtype: float64%\n"
     ]
    }
   ],
   "source": [
    "X2_data = poly.fit_transform(dXtest)\n",
    "print(\"R2:\" +str(lm.score(X2_data, dYtest)))\n",
    "regresion_validation(X2_data, dYtest, lm.predict(X2_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>TempMax2016</th>\n",
       "      <th>Prediccion2016</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>29.5</td>\n",
       "      <td>30.148935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>29.0</td>\n",
       "      <td>29.807606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.0</td>\n",
       "      <td>29.604457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28.5</td>\n",
       "      <td>30.020084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21.5</td>\n",
       "      <td>22.369699</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   TempMax2016  Prediccion2016\n",
       "0         29.5       30.148935\n",
       "1         29.0       29.807606\n",
       "2         28.0       29.604457\n",
       "3         28.5       30.020084\n",
       "4         21.5       22.369699"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res1 = dYtest['TempMax'].tolist()\n",
    "res2 = pd.DataFrame(lm.predict(X2_data),columns = ['Prediccion2016'])\n",
    "res3 = pd.DataFrame(res1,columns=['TempMax2016'])\n",
    "Predic2016 = pd.concat([res3, res2], axis=1, sort = False)\n",
    "Predic2016.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((34302, 6), (34302,))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = dXtrain\n",
    "Y = dYtrain['TempMax']\n",
    "X.shape, Y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "C=1e3\n",
    "X = dXtrain\n",
    "Y = dYtrain\n",
    "svr_lin = SVR(kernel=\"linear\", C=C)\n",
    "svr_rbf = SVR(kernel=\"rbf\", C=C, gamma=0.1)\n",
    "svr_pol = SVR(kernel=\"poly\", C=C, degree=3)\n",
    "\n",
    "y_lin = svr_lin.fit(X,Y).predict(X)\n",
    "y_rbf = svr_rbf.fit(X,Y).predict(X)\n",
    "y_pol = svr_pol.fit(X,Y).predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = [\n",
    "    {\n",
    "        'kernel': ['rbf'],\n",
    "        'gamma' : [1e-4,1e-3,1e-2, 0.1, 0.2, 0.5],\n",
    "        'C': [1,10,100,1000]\n",
    "    },\n",
    "    {\n",
    "        'kernel':[\"linear\"],\n",
    "        'C':[1,10,100,1000]\n",
    "    }\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = GridSearchCV(svm.SVR(kernel='rbf'), param_grid=parameters, cv=5)\n",
    "clf.fit(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.6125739097595215"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "start_time = time()\n",
    "svr_lin = SVR(kernel=\"linear\", C=1.0)\n",
    "y_lin = svr_lin.fit(X,Y).predict(X)\n",
    "elapsed_time = (time()-start_time)/60\n",
    "elapsed_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.214698576927185"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "start_time = time()\n",
    "svr_rbf = SVR(kernel=\"rbf\", C=1.0, gamma=0.1)\n",
    "y_rbf = svr_rbf.fit(X,Y).predict(X)\n",
    "y_rbf\n",
    "elapsed_time = (time()-start_time)/60\n",
    "elapsed_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([25.09974343, 24.83926826, 24.68132294, ..., 25.98717083,\n",
       "       27.90014242, 28.22688837])"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_rbf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/eea/anaconda3/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "5.694342557589213"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "start_time = time()\n",
    "svr_pol = SVR(kernel=\"poly\", C=1.0, degree=2)\n",
    "y_pol = svr_pol.fit(X,Y).predict(X)\n",
    "elapsed_time = (time()-start_time)/60\n",
    "elapsed_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>TempMax</th>\n",
       "      <th>PredicLinear</th>\n",
       "      <th>PredicRBF</th>\n",
       "      <th>PredicPoly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>25.0</td>\n",
       "      <td>38.192581</td>\n",
       "      <td>25.099743</td>\n",
       "      <td>71588.181565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24.5</td>\n",
       "      <td>37.168093</td>\n",
       "      <td>24.839268</td>\n",
       "      <td>62745.336289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>24.0</td>\n",
       "      <td>35.249774</td>\n",
       "      <td>24.681323</td>\n",
       "      <td>56918.656174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25.0</td>\n",
       "      <td>32.780651</td>\n",
       "      <td>24.708672</td>\n",
       "      <td>54211.211533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25.0</td>\n",
       "      <td>39.557983</td>\n",
       "      <td>25.099769</td>\n",
       "      <td>26703.130539</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   TempMax  PredicLinear  PredicRBF    PredicPoly\n",
       "0     25.0     38.192581  25.099743  71588.181565\n",
       "1     24.5     37.168093  24.839268  62745.336289\n",
       "2     24.0     35.249774  24.681323  56918.656174\n",
       "3     25.0     32.780651  24.708672  54211.211533\n",
       "4     25.0     39.557983  25.099769  26703.130539"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res1 = pd.DataFrame(y_lin,columns = ['PredicLinear'])\n",
    "res2 = pd.DataFrame(y_rbf,columns = ['PredicRBF'])\n",
    "res3 = pd.DataFrame(y_pol,columns = ['PredicPoly'])\n",
    "\n",
    "result = pd.concat([Y, res1, res2, res3], axis=1, sort=False)\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lw = 2\n",
    "plt.figure(figsize=(16,9))\n",
    "#plt.scatter(X,Y,color=\"darkorange\", label =\"data\")\n",
    "plt.plot(X,y_lin, color=\"navy\", lw = lw, label = \"SVM Lineal\")\n",
    "plt.plot(X,y_rbf, color=\"c\", lw=lw, label=\"SVM Radial\")\n",
    "#plt.plot(X,y_pol, color=\"cornflowerblue\", label=\"SVM Polinómico\")\n",
    "plt.xlabel(\"x\")\n",
    "plt.ylabel(\"y\")\n",
    "plt.title(\"Support Vector Regression\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f4c69c314e0>"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}