{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "41147902",
   "metadata": {},
   "source": [
    "# Integrating with the Scikit: Pipeline and Gridsearch\n",
    "\n",
    "ITEA implementations inherits scikits' base classes. This means that we can integrate the ITEA with methods like Pipeline and Gridsearch. In this notebook, we'll show some examples on how to take advantage of that to tune an predictor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d3847fde",
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "\n",
    "import pandas  as pd\n",
    "import numpy   as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from scipy                   import stats\n",
    "from sklearn.model_selection import train_test_split\n",
    "from IPython.display         import display\n",
    "\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import mutual_info_regression\n",
    "\n",
    "from sklearn import datasets\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Importing the halving gridsearch algorithm\n",
    "from sklearn.experimental import enable_halving_search_cv\n",
    "from sklearn.model_selection import HalvingGridSearchCV\n",
    "\n",
    "from itea.regression import ITEA_regressor\n",
    "from itea.inspection import *\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore', module=r'itea')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7eeb718c",
   "metadata": {},
   "source": [
    "## Loading the data\n",
    "\n",
    "First, let's load the data, and split it into a training and testing partition. The training partition will be used for the training and validation process, and only after obtaining a final method will we perform the training with this data and the test with the test partition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "abce135b",
   "metadata": {},
   "outputs": [],
   "source": [
    "boston_data = datasets.load_boston() \n",
    "X, y        = boston_data['data'], boston_data['target']\n",
    "labels      = boston_data['feature_names']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f280150",
   "metadata": {},
   "source": [
    "## Inspectioning the data\n",
    "\n",
    "Let's look at some descriptive statistics for the variables.\n",
    "\n",
    "Suppose that, to reduce the complexity of the final model, we are interested in obtaining a subset of these variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b72c9693",
   "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>CRIM</th>\n",
       "      <th>ZN</th>\n",
       "      <th>INDUS</th>\n",
       "      <th>CHAS</th>\n",
       "      <th>NOX</th>\n",
       "      <th>RM</th>\n",
       "      <th>AGE</th>\n",
       "      <th>DIS</th>\n",
       "      <th>RAD</th>\n",
       "      <th>TAX</th>\n",
       "      <th>PTRATIO</th>\n",
       "      <th>B</th>\n",
       "      <th>LSTAT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "      <td>506.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.613524</td>\n",
       "      <td>11.363636</td>\n",
       "      <td>11.136779</td>\n",
       "      <td>0.069170</td>\n",
       "      <td>0.554695</td>\n",
       "      <td>6.284634</td>\n",
       "      <td>68.574901</td>\n",
       "      <td>3.795043</td>\n",
       "      <td>9.549407</td>\n",
       "      <td>408.237154</td>\n",
       "      <td>18.455534</td>\n",
       "      <td>356.674032</td>\n",
       "      <td>12.653063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8.601545</td>\n",
       "      <td>23.322453</td>\n",
       "      <td>6.860353</td>\n",
       "      <td>0.253994</td>\n",
       "      <td>0.115878</td>\n",
       "      <td>0.702617</td>\n",
       "      <td>28.148861</td>\n",
       "      <td>2.105710</td>\n",
       "      <td>8.707259</td>\n",
       "      <td>168.537116</td>\n",
       "      <td>2.164946</td>\n",
       "      <td>91.294864</td>\n",
       "      <td>7.141062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.006320</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.460000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.385000</td>\n",
       "      <td>3.561000</td>\n",
       "      <td>2.900000</td>\n",
       "      <td>1.129600</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>187.000000</td>\n",
       "      <td>12.600000</td>\n",
       "      <td>0.320000</td>\n",
       "      <td>1.730000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.082045</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.190000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.449000</td>\n",
       "      <td>5.885500</td>\n",
       "      <td>45.025000</td>\n",
       "      <td>2.100175</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>279.000000</td>\n",
       "      <td>17.400000</td>\n",
       "      <td>375.377500</td>\n",
       "      <td>6.950000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.256510</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.690000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.538000</td>\n",
       "      <td>6.208500</td>\n",
       "      <td>77.500000</td>\n",
       "      <td>3.207450</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>330.000000</td>\n",
       "      <td>19.050000</td>\n",
       "      <td>391.440000</td>\n",
       "      <td>11.360000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3.677083</td>\n",
       "      <td>12.500000</td>\n",
       "      <td>18.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.624000</td>\n",
       "      <td>6.623500</td>\n",
       "      <td>94.075000</td>\n",
       "      <td>5.188425</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>666.000000</td>\n",
       "      <td>20.200000</td>\n",
       "      <td>396.225000</td>\n",
       "      <td>16.955000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>88.976200</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>27.740000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.871000</td>\n",
       "      <td>8.780000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>12.126500</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>711.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>396.900000</td>\n",
       "      <td>37.970000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             CRIM          ZN       INDUS        CHAS         NOX          RM  \\\n",
       "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
       "mean     3.613524   11.363636   11.136779    0.069170    0.554695    6.284634   \n",
       "std      8.601545   23.322453    6.860353    0.253994    0.115878    0.702617   \n",
       "min      0.006320    0.000000    0.460000    0.000000    0.385000    3.561000   \n",
       "25%      0.082045    0.000000    5.190000    0.000000    0.449000    5.885500   \n",
       "50%      0.256510    0.000000    9.690000    0.000000    0.538000    6.208500   \n",
       "75%      3.677083   12.500000   18.100000    0.000000    0.624000    6.623500   \n",
       "max     88.976200  100.000000   27.740000    1.000000    0.871000    8.780000   \n",
       "\n",
       "              AGE         DIS         RAD         TAX     PTRATIO           B  \\\n",
       "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
       "mean    68.574901    3.795043    9.549407  408.237154   18.455534  356.674032   \n",
       "std     28.148861    2.105710    8.707259  168.537116    2.164946   91.294864   \n",
       "min      2.900000    1.129600    1.000000  187.000000   12.600000    0.320000   \n",
       "25%     45.025000    2.100175    4.000000  279.000000   17.400000  375.377500   \n",
       "50%     77.500000    3.207450    5.000000  330.000000   19.050000  391.440000   \n",
       "75%     94.075000    5.188425   24.000000  666.000000   20.200000  396.225000   \n",
       "max    100.000000   12.126500   24.000000  711.000000   22.000000  396.900000   \n",
       "\n",
       "            LSTAT  \n",
       "count  506.000000  \n",
       "mean    12.653063  \n",
       "std      7.141062  \n",
       "min      1.730000  \n",
       "25%      6.950000  \n",
       "50%     11.360000  \n",
       "75%     16.955000  \n",
       "max     37.970000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(data=X, columns=labels).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8b0a129",
   "metadata": {},
   "source": [
    "One way to reduce the amount of attributes is to use attribute engineering methods (such as PCA), but we can also select a subset of attributes.\n",
    "\n",
    "Let's use a scikit method that finds a subset of k attributes based on a passed metric. Let's get the 4 best variables based on the mutual information of continuous variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e99b8858",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['INDUS' 'NOX' 'RM' 'LSTAT']\n"
     ]
    }
   ],
   "source": [
    "feature_selector = SelectKBest(mutual_info_regression, k=4)\n",
    "\n",
    "X_new      = feature_selector.fit_transform(X, y)\n",
    "mask       = feature_selector.get_support()\n",
    "labels_new = labels[mask]\n",
    "\n",
    "print(labels_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac3bf5c4",
   "metadata": {},
   "source": [
    "Without going into further details or making use of pre-processing rigor, let's just look at the correlation between the selected variables and the dependent variable (target variable)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e5b50eda",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x900 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def corrfunc(x,y, ax=None, **kws):\n",
    "    \n",
    "    r, _ = stats.pearsonr(x, y)\n",
    "    ax = ax or plt.gca()\n",
    "\n",
    "    ax.annotate(\n",
    "        r'$\\rho' + f'= {r:.2f}$',\n",
    "        xy=(.1, .9),\n",
    "        xycoords=ax.transAxes,\n",
    "        bbox=dict(\n",
    "            facecolor='white',\n",
    "            edgecolor='black',\n",
    "            boxstyle='round, pad=0.35'\n",
    "        )\n",
    "    )\n",
    "\n",
    "g = sns.pairplot(\n",
    "    pd.DataFrame(\n",
    "        data=np.hstack( (X_new, y.reshape(-1, 1)) ),\n",
    "        columns=labels_new.tolist()+['target variable']\n",
    "    ),\n",
    "    markers=\".\", \n",
    "    corner=True,\n",
    ")\n",
    "\n",
    "g.map_lower(corrfunc)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "342231ff",
   "metadata": {},
   "source": [
    "## Creating a pipeline\n",
    "\n",
    "scikit provides a Pipeline class, which serves to nest a sequence of transformations and a final estimator. With this, we can automate the date transformation steps for an estimator.\n",
    "\n",
    "Let's do the variable selection step and fit an ITEA regressor into a pipeline. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0c7ccabb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gen | smallest fitness | mean fitness | highest fitness | remaining time\n",
      "----------------------------------------------------------------------------\n",
      "  0 |         4.768589 |     5.768432 |        9.387225 | 0min2sec    \n",
      " 10 |         4.301443 |     4.560781 |        4.914515 | 0min2sec    \n",
      " 20 |         4.218327 |     4.367407 |        4.679878 | 0min2sec    \n",
      " 30 |         4.058739 |     4.215670 |        6.612622 | 0min2sec    \n",
      " 40 |         3.962236 |     4.207883 |        6.405873 | 0min2sec    \n",
      " 50 |         3.962236 |     4.176876 |        7.135065 | 0min1sec    \n",
      " 60 |         3.962236 |     4.186288 |        4.764542 | 0min1sec    \n",
      " 70 |         3.916771 |     4.141421 |        4.779798 | 0min1sec    \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('selectKbest',\n",
       "                 SelectKBest(k=4,\n",
       "                             score_func=<function mutual_info_regression at 0x7f89b6d52dd0>)),\n",
       "                ('itea',\n",
       "                 ITEA_regressor(gens=75,\n",
       "                                labels=array(['INDUS', 'NOX', 'RM', 'LSTAT'], dtype='<U7'),\n",
       "                                popsize=75,\n",
       "                                tfuncs={'exp': <ufunc 'exp'>,\n",
       "                                        'id': <function <lambda> at 0x7f89ab6f3dd0>,\n",
       "                                        'log': <ufunc 'log'>,\n",
       "                                        'sqrt.abs': <function <lambda> at 0x7f89ab6f3950>},\n",
       "                                tfuncs_dx={'exp': <ufunc 'exp'>,\n",
       "                                           'id': <function <lambda> at 0x7f89aafe2a70>,\n",
       "                                           'log': <function <lambda> at 0x7f89aafe2950>,\n",
       "                                           'sqrt.abs': <function <lambda> at 0x7f89aafe29e0>},\n",
       "                                verbose=10))])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfuncs = {\n",
    "    'log'      : np.log,\n",
    "    'sqrt.abs' : lambda x: np.sqrt(np.abs(x)), \n",
    "    'id'       : lambda x: x,\n",
    "    'exp'      : np.exp\n",
    "}\n",
    "\n",
    "tfuncs_dx = {\n",
    "    'log'      : lambda x: 1/x,\n",
    "    'sqrt.abs' : lambda x: x/( 2*(np.abs(x)**(3/2)) ),\n",
    "    'id'       : lambda x: np.ones_like(x),\n",
    "    'exp'      : np.exp,\n",
    "}\n",
    "\n",
    "# Creating our ITEA regressor instance\n",
    "itea = ITEA_regressor(\n",
    "    gens         = 75,\n",
    "    popsize      = 75,\n",
    "    expolim      = (-2, 2),\n",
    "    tfuncs       = tfuncs,\n",
    "    tfuncs_dx    = tfuncs_dx,\n",
    "    verbose      = 10,\n",
    "    labels       = labels_new\n",
    ")\n",
    "\n",
    "pipeline = Pipeline([\n",
    "    ('selectKbest', feature_selector),\n",
    "    ('itea', itea)\n",
    "])\n",
    "\n",
    "pipeline.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7752a47f",
   "metadata": {},
   "source": [
    "We can access the Pipeline ITEA with the index operator. Let's save the ITEA in one variable, and let's save the final expression (``ITExpr_regressor``) in another. Finally, let's look at the final expression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6e134e5c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.086*id(INDUS * NOX^-2 * RM^2) + -18.892*log(NOX^-2 * RM * LSTAT^2) + -18.139*log(NOX^-1 * LSTAT^-2) + -0.002*sqrt.abs(INDUS^2 * NOX^2 * RM^2 * LSTAT) + 0.705*id(INDUS * NOX^-2 * RM^2 * LSTAT^-2) + -21.588\n"
     ]
    }
   ],
   "source": [
    "print(pipeline['itea'].bestsol_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0bbcf428",
   "metadata": {},
   "source": [
    "## Finetuning with gridsearch\n",
    "\n",
    "ITEA has several hyperparameters, and although the method can be used with default values (which deliver fast execution with satisfactory results), it may be necessary to further investigate a suitable configuration for the domain of the problem in which the regressor is being applied.\n",
    "\n",
    "Imagine we want to limit the final expression to something that isn't too complex. We can achieve this by several ways.\n",
    "\n",
    "We can have several expression sizes, exponents limits, and different transformation functions. Let's choose some values for each configuration to perform the gridsearch.\n",
    "\n",
    "Here, we'll look to find a subset of functions and exponents that, combined, deliver good performance in the dataset we're using.\n",
    "\n",
    "Gridsearch can receive either an estimator or a pipeline to make the adjustment.\n",
    "\n",
    "A detail that is worth mentioning is that, in the case of a Pipeline, the variables will have a name with a prefix to be used in gridsearch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cf66793b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import permutations\n",
    "\n",
    "two_tfuncs = permutations(['log', 'sqrt.abs', 'exp'], r=2)\n",
    "\n",
    "parameters = {\n",
    "    'itea__gens'      : [100],\n",
    "    'itea__popsize'   : [100],\n",
    "    'itea__tfuncs_dx' : [tfuncs_dx],\n",
    "    'itea__expolim'   : [(-2, 2), (-1, 1), (0, 1), (0, 2)],\n",
    "    'itea__max_terms' : [10],\n",
    "    'itea__tfuncs'    : [\n",
    "        {t1:tfuncs[t1], t2:tfuncs[t2], 'id':tfuncs['id']}\n",
    "        for (t1, t2) in set(two_tfuncs)\n",
    "    ],\n",
    "    'itea__verbose': [False]\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9bb6cb4b",
   "metadata": {},
   "source": [
    "The scikit provides GridSearchCV, a method that does an exhaustive search for the best configuration by cross-validating past data.\n",
    "\n",
    "Since ITEA is an evolutionary algorithm, exhaustive testing can be computationally expensive. Let's use HalvingGridSearchCV (which is in experimental stage at the time of creation of this notebook), imported at the beginning of the notebook.\n",
    "\n",
    "This method makes the gridsearch with several interactions, but allocating few resources for the first runs, in order to get a possible direction of where it should apply more effort to obtain the best configuration.\n",
    "\n",
    "To use the standard gridsearch, just change ``HalvingGridSearchCV`` to ``GridSearchCV``."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a7ed8417",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_iterations: 5\n",
      "n_required_iterations: 5\n",
      "n_possible_iterations: 5\n",
      "min_resources_: 21\n",
      "max_resources_: 339\n",
      "aggressive_elimination: False\n",
      "factor: 2\n",
      "----------\n",
      "iter: 0\n",
      "n_candidates: 24\n",
      "n_resources: 21\n",
      "Fitting 3 folds for each of 24 candidates, totalling 72 fits\n",
      "----------\n",
      "iter: 1\n",
      "n_candidates: 12\n",
      "n_resources: 42\n",
      "Fitting 3 folds for each of 12 candidates, totalling 36 fits\n",
      "----------\n",
      "iter: 2\n",
      "n_candidates: 6\n",
      "n_resources: 84\n",
      "Fitting 3 folds for each of 6 candidates, totalling 18 fits\n",
      "----------\n",
      "iter: 3\n",
      "n_candidates: 3\n",
      "n_resources: 168\n",
      "Fitting 3 folds for each of 3 candidates, totalling 9 fits\n",
      "----------\n",
      "iter: 4\n",
      "n_candidates: 2\n",
      "n_resources: 336\n",
      "Fitting 3 folds for each of 2 candidates, totalling 6 fits\n",
      "----------\n",
      "222.4 seconds\n"
     ]
    }
   ],
   "source": [
    "gridsearch = HalvingGridSearchCV(\n",
    "    estimator=pipeline,\n",
    "    param_grid=parameters,\n",
    "    verbose=2, \n",
    "    n_jobs=-1,\n",
    "    refit=True, # If true, then 'gridsearch' will have a best_estimator_\n",
    "    cv=3,\n",
    "    factor=2,\n",
    "    scoring='neg_root_mean_squared_error'\n",
    ")\n",
    "\n",
    "t_start = time.time()\n",
    "\n",
    "gridsearch.fit(X_train, y_train)\n",
    "\n",
    "t_end = time.time() - t_start\n",
    "\n",
    "print('----------')\n",
    "print(f'{round(t_end, 2)} seconds')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "231aab72",
   "metadata": {},
   "source": [
    "Now that we have the best result, let's preview the grid of different settings for exponent limits and subsets of transform functions, and let's also create a final model with the best found setting.\n",
    "\n",
    "The heatmap is based on [this example from the scikits' documentation](https://scikit-learn.org/stable/auto_examples/model_selection/plot_successive_halving_heatmap.html#sphx-glr-auto-examples-model-selection-plot-successive-halving-heatmap-py)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f34a1277",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.97417141e+02 -2.46071938e+01 -8.44492769e+01 -1.03811080e+01]\n",
      " [-1.15262941e+01 -6.56115272e+15 -5.41812874e+02 -1.82725318e+01]\n",
      " [-3.63525617e+01 -2.17213496e+01 -9.62046613e+01 -3.41900820e+01]\n",
      " [-1.26314994e+01 -7.06547429e+00 -4.19566596e+00 -3.05124383e+01]\n",
      " [-2.26318495e+01 -7.44435538e+01 -5.82059580e+02 -5.72124969e+00]\n",
      " [-2.61751169e+01 -1.14023381e+01 -4.14897391e+00 -5.76394867e+00]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1, figsize=(8, 5))\n",
    "\n",
    "results_aux = gridsearch.cv_results_.copy()\n",
    "    \n",
    "# One of the parameters is a dictionary, the other a tuple. When we create a data frame\n",
    "# with this, it ends up becoming a mess when the pandas parser handles these values.\n",
    "# Let's turn it into strings.\n",
    "results_aux['param_itea__tfuncs']  = [\n",
    "    str(list(k.keys())[:-1]) # Left out the 'id' tfunc\n",
    "    for k in results_aux['param_itea__tfuncs']]\n",
    "\n",
    "results_aux['param_itea__expolim'] = [\n",
    "    str(k)\n",
    "    for k in results_aux['param_itea__expolim']]\n",
    "\n",
    "results = pd.DataFrame.from_dict(results_aux)\n",
    "\n",
    "scores_matrix = results.sort_values('iter').pivot_table(\n",
    "    index   = 'param_itea__tfuncs',\n",
    "    columns = 'param_itea__expolim',\n",
    "    values  = 'mean_test_score',\n",
    "    aggfunc = 'last'\n",
    ")\n",
    "\n",
    "print(scores_matrix.values)\n",
    "\n",
    "im = ax.imshow(scores_matrix*-1, aspect='auto', cmap='viridis_r')\n",
    "\n",
    "expolims_gs = set(results_aux['param_itea__expolim'])\n",
    "ax.set_xlabel('expolim', fontsize=15)\n",
    "ax.set_xticks(np.arange(len(expolims_gs)))\n",
    "ax.set_xticklabels(expolims_gs)\n",
    "\n",
    "tfuncs_gs = set(results_aux['param_itea__tfuncs'])\n",
    "ax.set_ylabel('tfuncs', fontsize=15)\n",
    "ax.set_yticks(np.arange(len(tfuncs_gs)))\n",
    "ax.set_yticklabels(tfuncs_gs)\n",
    "\n",
    "iterations = results.pivot_table(\n",
    "    index='param_itea__tfuncs',\n",
    "    columns='param_itea__expolim',\n",
    "    values='iter',\n",
    "    aggfunc='max'\n",
    ").values\n",
    "\n",
    "for i in range(len(tfuncs_gs)):\n",
    "    for j in range(len(expolims_gs)):\n",
    "        ax.text(j, i, iterations[i, j],\n",
    "            ha=\"center\", va=\"center\", color=\"k\", fontsize=15)\n",
    "\n",
    "fig.subplots_adjust(right=0.8)\n",
    "cbar_ax = fig.add_axes([0.85, 0.15, 0.05, 0.7])\n",
    "\n",
    "fig.colorbar(im, cax=cbar_ax)\n",
    "cbar_ax.set_ylabel(\n",
    "    'mean test score (RMSE)', rotation=-90,\n",
    "    va=\"bottom\", fontsize=15)\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be5e5a68",
   "metadata": {},
   "source": [
    "## Creating and fitting a model with the best configuration\n",
    "\n",
    "Finally, let's create an instance of ITEA with the best configuration, and then use the test data to see how the final model performs as a predictor.\n",
    "\n",
    "Additionally, let's look at some interpretability graphs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8392c853",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7876885603830583"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# best_pipeline is be a pipeline!\n",
    "best_pipeline = gridsearch.best_estimator_\n",
    "\n",
    "# predict(), score() and other estimators methods will\n",
    "# perform the transformations and then call the method on the final\n",
    "# estimator.\n",
    "\n",
    "best_pipeline.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "484a7e27",
   "metadata": {},
   "source": [
    "ITEA is an estimator, and the interpretability classes only work with instances of ``ITEA`` or ``ITExpr``.\n",
    "\n",
    "To be able to use the entire pipeline created, let's create a method that receives a pipeline and iterates over all transformations successively until finishing the treatment, and returning this new data.\n",
    "\n",
    "So we'll use this method to handle the data before calling the ``fit`` of the explainers. Thus, we use the pipeline with the classes from ``itea.inspection``."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "6d1e46d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def just_transformers(pipeline, X):\n",
    "    \n",
    "    Xt = X.copy()\n",
    "    for name, transformer in pipeline.steps[:-1]:\n",
    "        Xt = transformer.transform(Xt)\n",
    "        \n",
    "    return Xt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1018f628",
   "metadata": {},
   "source": [
    "Let's create the explainer instance. We'll pass an ``ITExpr`` to the explainer and use the transformations to fit the data. Note how these values are used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b7dd04e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "explainer = ITExpr_explainer(\n",
    "    itexpr    = best_pipeline['itea'].bestsol_,\n",
    "    tfuncs    = tfuncs,\n",
    "    tfuncs_dx = tfuncs_dx\n",
    ").fit(just_transformers(best_pipeline, X_train), y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "728b37ae",
   "metadata": {},
   "source": [
    "Now we can create the interpretability plots we saw on the other notebooks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c210c635",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "explainer.plot_feature_importances(\n",
    "    X=just_transformers(best_pipeline, X_train),\n",
    "    importance_method='pe',\n",
    "    grouping_threshold=0.0,\n",
    "    barh_kw={'color':'green'}\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1aff74ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "explainer.plot_normalized_partial_effects(\n",
    "    grouping_threshold=0.1, show=False, num_points=10)\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "69c346b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAADQCAYAAAAasZepAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABA3UlEQVR4nO3dd3zcZ5Xv8c+ZPqPeLMuy5Ra3xHGcxHESEiANSKPDEsou9YbdZbnAstxd4MI22F3Iwi5cWJYsPfSQEEpCIEAKEFJsx3Gc2LETd7mq1+nP/eM3kmXFkhVrpJmRvu/XSy9pZn766Xj86DdHz5znPOacQ0REREREPL5CByAiIiIiUkyUIIuIiIiIjKAEWURERERkBCXIIiIiIiIjKEEWERERERkhUOgAplJ9fb1btGhRocOQGWTjxo1tzrmGyZ5nrLGZyTrS2cJ0ljHAZ4bZ8c9SOqZ6bIqcLo1NKWZjjc8ZnSAvWrSIDRs2FDoMmUHMbG8+zjPW2OweSNHen8jHj5i0gM9HNOQnGvITC/rx+ZQxF7OpHpsip0tjU4rZWONzRifIInL60tksvfEsvfEUZkY44KMsFCAa8hMKqDpLRERmrpJLkM3MD2wAWp1z1xc6HpHZwDlHPJUhnspAPwT9PsrCAcrCfsIBf6HDExERyauSS5CB9wHbgMpCByIyW6UyWboGknQNeMlyLOSnLBwgElSyLCIipa+k3ic1s/nAdcBXCh2LiHhSmSzdgykOdg2yv2OAzv4kyXS20GGJiIictpJKkIH/BP4PMOarr5ndaGYbzGzDsWPHpi0wkVOZDWMzlcnSOZDkQOcArV2DdA+kyBSoK4dM3GwYm1KaNDalUEqmxMLMrgeOOuc2mtllYx3nnLsZuBlg3bp1emWWonGqsbn9cA9v+PJD+H1G0G+E/F4XiVjIT1koQEUkSGU0QHU0SE1ZiNqyEPXlYeZUhCkLF9+vciKVIZHK0DGQJBbyUx4OEAv5MfWPKzq6bkqx0tiUQim+V9WxXQK8wsyuBSJApZl92zn3lgLHJZIXFZEgV6+eS38iTSqTJZ7KEk9lGEimOdQd5+kjvfTE0yctXygL+5lbGaGpKkpzdYTmmhgLaqMsrI1RHQsV4F9znHOO/kSa/kQav88oD3vJvjphiIhIsSqZBNk592HgwwC5GeS/UXIsM0lzdZSPXLNq3D7IXjeJLB39Sdr7E7T1JTnam+BoT5zDPXH2dwzw8O52UpnjEy3V0SCL6stY0lDGGQ3lnDGnnIV1MYL+6U9QM1lH92CK7sEUkaCfikiA8nBAs8oiIlJUSiZBFhEwM6IhP82hKM010ZMek8k6jvbG2dcxwL6OQfa09bO7rZ+7thwinpt9DvqNJQ3lrJxbwZlNlZzZVMm86si0JqpDbeM6+pOaVRYRkaJSkgmyc+4+4L4ChyFSlPw+o6kqSlNVlAsXH78/k3Uc7Bpk59E+nj7cy44jvfzyycP8ZPNBAGpiQVY3V3F2cxVr5lextKEc/zTsnjdyVjka8lMZCRZlTbWIiMweehUSmSX8PmNBbYwFtTGuWDkH8JLTPe39PHWwh60He9ja2s3vdrYBUB4OsGZ+Fee1VHPewhoW1samfIZ5MJlhMJkh6PdREfFmlacjSRcRERlJCbLILOb3GUsbylnaUM7Lz5kHwLHeBFsOdPHY/i4e29fFg8+2A1BfHmLdwlrWL65l3cIayiNTd/lIZbw6686BFOXhAJXRgHbsExGRaaMEWURO0FAR5spVjVy5qhGAQ92DbNzbxYa9Hfz+mTbufvIwPoPVzVVctKSOFyyto6U2NiWxOOfojafojXvlF1XRILGQLlsiIjK19EojIuNqqopy/Zoo169pIpN1bDvUw8O7O3hoVzs3P7CLmx/YxfyaKJcsrePSZfWsaqrENwWlGCPLL6piQSrU/UJERKaIEmQRmTC/z1jdXMXq5ireeelijvbEefDZdh58tp3bNrXygw0HqCsPcekZ9Vy2vIHVzVV5ryFOZbK09Sbo7E9SEQlSFVWdsoiI5JcSZBE5bXMqI7zq3GZedW4zffE0D+1u54Edbfxiq9cdo64sxIuWN3D5igbOnJffmeVM1tE1kKR70KtTroqqTZyIiOSHEmQRyYvySICrVjVy1apGBpMZHtrVzn07jvHzLQf58WOtNFaGuWLlHK5a1cji+rK8/dyRdcpluUQ5EtSCPhEROX1KkEUk76IhP5evnMPlK+fQn0jzh2fb+e22I/zg0f1875H9nNFQzlVneslybVn+tsIe2tI6HPRTHVU/ZREROT169RApIpXRABW59mlDm0U758g6yDqHc5BxjkzWkc060llH1nmf05ksmawb++QFUhYO8NIzG3npmY10DiS5d/tR7tl2lP++31vgt35xLVefNZeLl9blbfvrRCrDkZQW9ImIyOlRgixSRMyM5+ZxE0/snHOkMo50Nksq7UhmsqRyH8WQPNfEQrzmvPm85rz57G3v51dPHeFXTx3hoV1PURkJcNWZjVx3dlPeSjCGFvR19aeoigapiATwaUGfiIicghJkkRnEzAgFjBA+GFW5kMk6kuksyXSWRCYz/HWhLKwr43+9cAnvuGQxG/d28outh/np5oPcvqmVM5squG7NPC5b0UA0D/XE6WyW9v4EnQNJKqPqfCEiIuNTgiwyS/h9RjTkJxryA0HAm3FOpLO5jwyJlDfbPN1xrV/s7dDXPZDiV9uOcOeWQ9z0y6f5r3uf4aozG3nFOfPyMqucdcc7X1REvAV9+SrrEBGRmUMJssgsZmZEgv5c1wcvac5kHfFUxvvIzTI7Nz3lGVWxIK8/fz6vO6+Zra09/GzLQe564hA/2XyQs5ureOXaebxwWf2kk1rnHD2DKXqGWsTFgtrKWkREhpVMgmxmEeABIIwX94+cc39f2KhEZh6/zygLB4Y7QGSz3izzYCrDYCpDIpWZ8hjMjLPnV3H2/Crec9kZ/OLJw/zs8YN84s5t1JaFuP7sJq4/p4n68vCkf1ZfIk1fIk0sFKA6phZxIiJSQgkykACucM71mVkQ+L2Z/cI591ChAxOZyXwnlGZ4CfNQsjyYzEx5SUZVLMgNFyzgT9bN59E9Hfxk80FueWgv33lkHy9e3sBrzm3mzHmVk/45A8k0A0m1iBMRkRJKkJ33Hm9f7mYw91H4Zfkis4xv1AxzKpNlIOkly4OpzJSVY/jMuHBxHRcurqO1a5A7Hmvl7q2H+e32o6xqquC1583nRcvqCUyy/GJki7jqWJBytYgTEZl1SiZBBjAzP7AROAP4onPu4ZMccyNwI0BLS8v0Bigyjpk6NoN+H1VRH1XRIM55s8sDyQwDiQzp7NTMLjdXR3nP5Wfw9ksW8csnj/Djx1r5xJ3bmFMR5tXnNnPdmibKJzkDnMpkOdaboHMWtIibqWNTSp/GphSKTdfim3wys2rgx8B7nXNbxzpu3bp1bsOGDdMWl8x8ZrbRObdusueZLWMznivD6Eukp7QUI+scD+1q50cbW9m8v4to0M91a+bymvPmM7cykpef4fcZlZEglUXaIk5jU4qVxqYUs7HGZ0nNIA9xznWZ2X3A1cCYCbKIFNZQh4yashDJdJaBpLcgLt/9l31mvGBpPS9YWs/OI73cuvEAP37M66l8+Yo53HDBApbOKZ/Uz8hkHZ0DSbrUIk5EZMYrmQTZzBqAVC45jgJXAZ8qcFgiMkGhgI9QIER1LEQqk6U/MTXJ8rLGCj5y7Sreeelibt/Uys+3HOI3249ywaIabrhgAWsXVE+qpnh0i7jKqDpfiIjMNCWTIANNwDdzdcg+4IfOuZ8XOCYROQ3eAjgvWU6mjyfL+SzDaKyM8BeXLeVPL1rITx8/yG2bDvDBW7ewqqmCN61v4eKldfgmufhuqEVcNOSnKhokFiqlS6qIiIylZK7mzrktwLmFjkNE8mtoZrmmLEQ85dUr9yfSZLL5WR9RHgnwpgtbeO15zdz95BF+uGE/H/vJkyyqi/GmC1u4fMWcSdcUD+a6eIQC3mJFdb4QESltKqATkaIRCfqpLw/TUhujsTKS117E4aCfV66dx7fesZ4PX7MSgH+5aztv/foj3LnlUF5mr5Npr/PF/o5BugaSZPOU5IuIyPQqmRlkEZk9zI73Ws5kHX3xND3xVF6SWL/PeMmZjVy5ag4PPtPOtx/ey2fu2cEtD+3ljesXcM3qJkKByc0dpLNZOvqTdA14C/oqtaBPRKSkKEEWkaLm9xlVsSBVsSDxVIaeeIr+xOQ3JPGZcemyei45o45H9nTw7Yf28bnfPMO3H97HGy9o4fo1k0+Us87RPZiiWwv6RERKihJkESkZQ23jMmX5m1W23A596xfV8ti+Lr710F6+cO8zfO+Rfbxx/QKuXzNv0okyHF/QFw56C/rKQn7VKYuIFCklyCJSckbOKg8mh2aV05M6p5lx3sIazltYw+b9XXzzwT184d5n+d6j+/M2owzeVtZHc1tZV0Zm9g59IiKlSgmyiJS0aMhPNOQnlcnSM5iiN54mO8nyi7ULqln7hrVs3t/F1/+why/c+wzff3Qfb75wIdeePTcv9cSpTJb2/gSdA0nKIwEqI8G8JOAiIjJ5uhqLyIwQ9Puoy3XAqCsP5yWJXbugmv98wzn8++vWMKciwud+s5O3fu1RfrH1cN7a0GVzG48c6BzgcHecwWQmL+cVEZHTpxlkEZlRfD6jKhqkKhpkIJmmayBFPHX6SedQ6cW5LdU8sqeDb/xhLzf98mm+98g+3vaCRVy2omHSG44MGUimGUimCfp9VMWClIdUfiEiUghKkEVkxoqFAsRCAeKpDN2Dk6tTHrmY78Fn2/n6H/bwiTu38d1H9vHOSxZz0ZLavC26S2WytPUm6LCk2sSJiBSAEmQRmfGGul8k01m6BpOTahNnZlxyRj0XL63j3u1H+fqDe/joHVs5a14l77p0MecsqM5b3CPbxMVCAWrLQqpTFhGZBkqQRWTWCAV8zKmIkIpl6c4t6DvdRNlnxpWrGnnx8gbufvII3/rjHj7ww8dZv6iGd166mGWNFXmNfSCZpizsV4IsIjINdKUVkVkn6PdRXx5mQU2UymhwUqURAb+P69c0ccs71vPuFy1h2+Fe3v3tTXzizm0c7BrMY9QiIjJdNIMsIrNWIJcoV0eDdE1yRjkc9POGCxZw3dlNfP/Rfdy2qZUHdhzj+jVN/OnFC6mJhfIcvYiITJWSmUE2swVmdq+ZbTOzJ83sfYWOSURmhsCIGeWKSHBS5yqPBHjXC5dwyzvXc/Xqufz08YO85SuPcMsf9zI4iW4aIiIyfUomQQbSwAedc6uAi4D3mNmZBY5JRGaQgN9HQ0WY+TUxysOTe4OtvjzMX79kOV976wWcv7CGrz+4hz/96iP8fMuhvPVQFhGRqVEyCbJz7pBzblPu615gG9Bc2KhEZCYKBXzMqYzQXBMlGvJP6lwtdTH+6ZVn8fkb1tJUFeGz9+zgXd/cwB+fbT/tcg4REZlaJZMgj2Rmi4BzgYdP8tiNZrbBzDYcO3Zs2mMTGYvGZukJB/w0VUVpqopOunvE6uYqPn/DWv7pFWeRcY6P3rGVD976ODuO9OYp2tOnsSnFSmNTCuWUV3wzC0/kvhGPXWJmZbmv32JmnzWzhZML84TzlwO3Ae93zvWMftw5d7Nzbp1zbl1DQ0O+fqzIpGlslq5oyM/8mhj1FWECvtNPlM2MS5fV87W3ruO9V5zB7rYB/vzbm/jXX2znaE88jxE/PxqbUqw0NqVQJnKl/+ME7xvyJWDAzM4B/g+wF/jWacT2HGYWxEuOv+Ocuz0f5xQRmajKSJD5NVGqY6FJt4Z79bnN3PLO9dxwwQLue/oof/b1R/nq73czkDz93f5ERCQ/xlyFYmZz8Wp8o2Z2LjD0alAJxMY5Z9o558zslcDnnHNfNbO3TjZQ816Nvgpsc859drLnExE5HT6fUVsWoiISoLM/Sd8ktq8uDwe48UVLeMXaeXzld7v5zsP7uOuJQ7zjksVcvXoufl9+tq4WEZHnZ7xl2i8D3gbMBz7D8QS5B/jION/Xa2YfBt4CvMjM/MDk+iZ5LgH+FHjCzDbn7vuIc+6uPJxbROR5Cfq9hXwVyQzt/QmS6expn2tuZYT/e90qXnteM/9137N85p4d3LG5lb+4bCnntdTkMWoREZmIMRNk59w3gW+a2Wudc7c9j3O+AXgT8E7n3GEzawFummScOOd+z/EkXUSkKERDfpqDUXoG03QOJMlOojPFqqZKPn/DWu7fcYwvP7CLv7l1C5csrePdL17C/Jrx3rgTEZF8mkgN8vlmVj10w8xqzOwTYx3snDvsnPusc+53udv7nHN5qUEWESlGZkZVzKtPnmz/ZDPjshVz+Obb1/OuSxezaV8X7/jGBr5037P0xlN5ilhERMYzkQT5Gudc19AN51wncO1YB5vZa8xsp5l1m1mPmfWa2XO6TYiIzDSBXNlFU1WUoH9ybeFCAR9vurCFW965npec2ciPNh7g+s//vijawomIzHQTmerwm1nYOZcAMLMoMGabN+DTwMudc9vyEaCISKnx2sJF6RpI0TWYmtSGILVlIT70shW8cu08fvnkEZbUl+UxUhEROZmJJMjfBn5jZl8HHPAO4JvjHH9EybGIzHZmRk1ZiLJwgLa+BPFUZlLnW95YwSVn1BOY5My0iIic2ikTZOfcp81sC3AV3iK5f3bO/XKcb9lgZj8A7gASI86jvsUiMuuEAj7mVUfpiafo6JvcIj4REZkeE11Nsg2vv/GvzSxmZhXOubEK4SqBAeClI+5zgBJkEZm1KiNBYkE/bX1JbQYiIlLkTpkgm9n/Am4EaoGleJuH/Ddw5cmOd869PZ8BiojMFAG/j7lVEfoSadr7EmSymk0WESlGE5lBfg+wHngYwDm308zmjHWwmUWAdwJnAZGh+51z75hcqCIiM0N5OEA06KetL0H/JHbiExGRqTGR1R4J51xy6IaZBfBKJsZyCzAXbye++/F24lNfIhGREfw+o7EywpzKiLaUFhEpMhNJkO83s48AUTN7CXAr8LNxjj/DOfcxoD+3G991wNmTD1VEZOYpDwdoro4SC01ugxEREcmfiSTIfwccA54A3g3cBfzfcY4f2uqpy8xWA1XAoknEKCIyow3VJtdXhPGZZpNFRAptzCkLM/uNc+5K4F+dc38L/M8Ez3mzmdUAHwN+CpQDH590pCIiM1xlJEg06Odob4LEJPsmi4jI6RvvPb0mM3sx8Aoz+z5eD+RhzrlNJ/sm59xXcl/eDyzJS5QiIrNE0O9jXlWE7sEUnQOT24VPREROz3gJ8sfxyivmA58d9ZgDrjjZN5lZI/AvwDzn3DVmdiZwsXPuq5MJ1My+BlwPHHXOrZ7MuUREipmZUR0LEQn6OdabIJXJFjokEZFZZbwa5EPOuWuAm5xzl4/6OGlynPMN4JfAvNztHcD78xDrN4Cr83AeEZGSEAn6aa6OUh7RAj4Rkek0XoL8+dznVz3Pc9Y7534IZAGcc2lg0sV0zrkHgI7JnkdEpJT4fMacCq8dnBbwiYhMj/GmJVJm9nWg2cw+P/pB59z/HuP7+s2sjlyvZDO7COiedKQiIrNYeVizyCIi02W8K+71wFV4tcYbn8c5/xqve8VSM/sD0AC87rQjfJ7M7Ea8rbFpaWmZrh8rckoam1KsNDalWGlsSqGMmSA759qA75vZNufc4xM9oXNuU677xQq8zhdPO+dSp/i2vHHO3QzcDLBu3Tot/5aiobEpxUpjU4qVxqYUykTesxs0s98Ajc651Wa2BniFc+4TJzvYzPzAtXibgwSAl5oZzrnRnTBERERERIrORHbS+x/gw+R2yHPObQFuGOf4nwFvA+qAihEfk2Jm3wP+CKwwswNm9s7JnlNEREREZLSJzCDHnHOP2Imrp9PjHD/fObdmcmE9l3Pujfk+p4iIiIjIaBOZQW4zs6Uc70rxOuDQOMf/wsxemo/gRERERESm20RmkN+DVyC/0sxagd3Am8c5/iHgx2bmwyvLMMA55yonG6yIiIiIyFQ7ZYLsnNsFXGVmZYDPOdd7im/5DHAx8IRzTitORURERKSkTLjzvHOuf4KH7gS2KjkWERERkVI0FVszHQLuM7NfAImhO9XmTURERERKwVQkyLtzH6Hch4iIiIhIyRgzQTaz14z3jc6528e4/x8nG5SIiIiISKGMN4P88nEec8BJE2QRERERkVI2ZoLsnHv7dAYiIiIiIlIMJlSDbGbXAWcBkaH7nHP/NFVBiYhIcclmHemsI+g3Ru2sKiIy45wyQTaz/wZiwOXAV4DXAY+Mc/xy4EtAo3NutZmtAV7hnPtEfkIWmbxs1pF1jqyDrHM4Bxnn3VcWCuD3KQEQGSmZyXKwaxCAoN9HKOAj5PcRDPgI+o2Q36fEWURmjInMIL/AObfGzLY45/7RzD7D+PXH/wN8CPgygHNui5l9F1CCLHnjnCOTPZ7gZkfczmSyDCQz9MRT9Ccy9CZS9MXT9Ccy9CfT9CczDCTSDCYzDKZyH8kMA8kM8XSGmliIr73tgkL/E0WKViqTJZXJMro5/lDifPyzEmcRKU0TSZAHc58HzGwe0A4sHuf4mHPukVEXxPRpxiezQCY7lNw60pks/ck03QNpugZTdA+m6BlM0RP3PvriaXrjafoSafoTXrLbn0gzkPvcn/QS3+wEt6mJBH1Eg36iIb/3OegnHPBN7T9YZIYaSpxHC/pPTJqDfm/22ad3akSkSE0kQf65mVUDNwGb8DpYfGWc49vMbGnuOMzsdXibh0yamV0NfA7wA19xzv1bPs4r+ZXJeoluXyJNe3+S9r4knf1J2vuTdA8m6RpInZD89sZTw0lvbzxN+hTZbdBvlIcDlIUDlIUClIX91JaFiOWS3JH3l4UDlIcDxMJ+73PIT3k4SFnYTywUIODz6ikN8JmBQUV4KtqDi8xeQ4nzQPLE+wO+EUnzUMmG36cSJxHJi3QmS8Y5wgH/8/7eU2YCzrl/zn15m5n9HIg457rH+Zb3ADcDK82sFW/TkLc878hGMTM/8EXgJcAB4FEz+6lz7qnJnltOLZt1DCQzHOtNcKQ3zrGeBMf6E7T3JejoT9HRn6CjP0nngJf0dg2mSKafO5M0JBbyUxkJUhEJUBEJ0FARoSoaoCJ3X1U06D0eDVAZCeZuB6iMhAiHfPjM8FkuqYUTbpuht3RF8uyym+4j6xzRoJ9Y2E8sePwP0PLc7/HQ72pVNEhVLEhNNERFNDD8ezpaOpslnXzudSLg8xEM2PGZ59zssxJnERnJOUcq40hns97nTJaugRT7Owdo7RrkUFecY30JLlvewDVnNz2vc4+3UcgVzrnfnmzDEDMbb6OQXcBVZlYG+Jxzvc8rorGtB57JnR8z+z7wSkAJ8iRks47BVIbD3XEOdg9yuDvOkZ44R3sTHOmJ096XpK3PS3574ievlIkEfVRHQ1THgjSUh1k2p5yashA1sSA1sRA1sRC1ZUFqy8JU5+4LBbwk1+/zElsltCLFyznHFavm0N6XGK7XH0imaetLeu/8JNJj/kHs9xk1sSC1ZSHqysLUl4eorwjTUB6mocL7aKwIEw4en+HxEmcYJPOcc40s11DiLDI7DL0LlUo74ukMrZ0D7O0YYH/HIIe6BznYFedwd5xD3YPPyVXCAR8N5eH8JcjAi4HfcvINQ8bcKMTMwsBrgUVAYCjxyUNbuGZg/4jbB4ALJ3nOGS2bdaSyWboHUuzrGGB/xwAHugY52DXIoRGJcEdfktFFDQGfUVsWor48REtdjPNaaqivCOVe1CLUl4eYUxmmvjxMRSSI30z1hCIzlJnx8evPHO5icTKJVIbeRJruAa98qmswReeAV1419C7Tkd44Tx3qoXsw9Zzvr4kFmVMZoakywtyqCPOqI8yrijKvOkpDRRi/z3LrFTLEU89NnEcuDgypxlmk5GSzjmQuEY6nsuzvGGBXWx/72gc40DlIa5f3cbg7fkIpZsBnzK2KMLcywvLGBuZWRWiqijKvOsL8mhhNVRHqysPPO57xNgr5+9yX/+Sc2z3yMTMbb5HeT4BuYCOQeN4Rje1kV7rnFKua2Y3AjQAtLS15/PHFJ5N1w39VHe6Os7utnz3tA+zr6Kc1N5gOdcfpHfXXVCjgY05FmMaKCBctLqepKkJjZZimqihNuRel+vIQAa0+z6vZNDaltORjbIaDfsJBP/UTeCFKprO09SVyJVsJjvbEOdwT50hPgp1H+/j9M20nvAAG/UZTVZTm6ijza7yPltoYC2pj1MSCZLJ4XWlGzTiPbkcXziXRUjp03Zx5hhLhZCZLR5/3O7/rWD972vvZ3zHI/s4BDnYNksocvwZEgj7mV8dYUl/GpWfUM78myoKaGAvqosyrihEJ+obfYQr4jEAefs8nshrpNuC8Uff9CDh/jOPnO+eunlRUJ3cAWDDy5wAHRx/knLsZrwaadevWTbCXQfFKZbKkM95giqfS7G0fYOeRPna19bO3vZ99Hd5fVgPJ4y8Mfp8xt9KbgVnVVMn8mhgLaqLMr42xsDbKnIoIoYCS3+k208amzBzTPTZDAR/zqr3Z4ZPJZB3H+hIc6hqktSvOwdzM0YHOQTbu6zyhnKM8HKClNsaiuhgL62Isqi9jUV0Z9eWhk7aj85k32zz84fcSZ10Pi5Oum6VrqD44mclytCfO04d7c8lwH3s7BtjbPkBH//GVuwGfDf8RfNHiWlrqylhUF2NxfRmNlRGv57rPRyDXCWeqjVeDvBJv97yqUXXIlYzYUe8kHjSzs51zT+QpxiGPAstys9etwA3Am/L8MwrCuaG3FRzJtHdBb+tNsP1IL88c6eWZY/3sPtbP3o7+E/6imlMRZkFtjJedVcXCuhhLGspYUl9GS22M6IgODSIi+eCtGfDKqfy5hbEYGLnFsZy4nsA573rlAOe821kHDq+1o8v1Mc+M6lwz9Ef+3MoI546aNMw6x9HeBPtzZWN7OwbY1z7Ag8+2c9fWw8PHlYX9LK4rY3FDGUvqy1naUMbi+jLKwgHiqRPLNMzM69kc8BEOeK0elTSLTNzQrHBPPMXTh3rZdriXnUd62dXWz+62/hMS4VjIT0ttjAsW1bCwzstbljaUs7A+RjToH15nUOi1BePNIK8ArgeqObEOuRf4X6MPNrMn8K6DAeDtZrYLr8TCAOecWzOZQJ1zaTP7K+CXeG3evuace3Iy5yyEdO5thWTa+0ikvQG143Av2w71sP1wL08f6eVIz/HqlLqyEEsayjh/YTNLGspZNqeC5XPLqYmFht9CFBGZSpGgn4V1ZVNy7qHEOZ3Nks16nzO5ra2HPqcz3n0+O548X7Co9oTzdA+k2NPuvSDvbvcmFn67/Sg/SxzvNNpUFWFpQzlnzCnjjDne9bS+PETSGcl0lr4RbfuHE+aglzBr0xMR7/c1kc5ytCfBltYunjrYw44jvTxztI99HQPD+xCEAj4W1XmJ8JL6cpY1lrO8sYL5NVHCAT9Bf35KIabKeDXIP8m1dftb59y/TOBc1+cvrDFjugu4a6p/Tr5ks94gSqQz3udUlnQ2S0d/ki0HutlyoIutB3vYdaxveEA1VUVYObeSV5zjDaQz51UytzJCJLeBRTEPJhGR02Fm+A38vqFOFifvWTq08DidccOlZ+mst7I9nc1SFQtyTqyacxZUD3+Py8047zrWz662Pp452s+zx/r4wzNtw4tYqqNBljWWs2yOd91dPreCxorw8ERGb/x4nKGAj0jA59Vcq6ZZZoFE2mvxunlfF0+0drPtUA87jvRxuCc+fExDeZilc8p44bJ6VsytZGVTBUsbyodnhAs9G3w6xq1Bds5lzOwlwCkTZOfcXgAzu8U596cjHzOzW4A/Pek3ziBDLdPiqQzxdJZE7i28wWSGx/Z3snFvF5v2dbK3fQCASMDHqnmVvOnCFs5squTMpkrmVkWJBH3DCbFmK0REPD6fEfb5OdlePiNXwI8sV0tlsjRWRmisjHDx0rrh4weSaXYd62fn0T52Hulj59FefrDhwHC5R1U0yPLGclbMrWDl3ApWzq2ktixEIpXxru25Thx+nxEO+IkEj5dnqHuGlKps1tEbT/NEq5evbDnQzfbDvRzoPN7BZl51hJVzK3jVufM4a14lZzdXM6cyTMg/sybxJrJI70Ez+wLwAzi+1sE5t2mM488aeSO3wcdYC/pKXjyV8fqCDl00cw51D/KHZ9p5aFc7Ww50k846wgEfa+ZX8dIzG1m7oJplc8opiwSJ5bY6VkIsInJ6fD4j4vMTCZ44+zxyxfzQjHAynSUWCrC6uYrVzVXDxybTWXa19fH04T52HPHK3b778L7hd/jmVIS9ZLmpklVNFSxvrCAa9DOQTJ+wS2Ao4E1yRDTLLEUuk3Uc6h7k4d0dbNzbyRMHunj6SN/wQti6shArmyq4bk0TZzdXsXZBNfXl4VmRr0wkQX5B7vPIPsYOuGLkQWb2YeAjQNTMeobuBpLkVqDOBM454ilvG+XBZIZ09vhq6tauQe7dfpQHdrTxzLE+ABbWxnjtec1csLiW1fOqiAT93pbIIW+r41J820FEpFSMlTinMsfXgSRzpXChgI+VcytZObdy+Lh4KsPOI31sP9LL9tw6kQd2tnnnNljSUO69AzivkrOaKplXHRlOwntys8wBn8+bYQ4en2kWKYRM1rGvY4AHn2njkT0dbN7fNfyudsBnLGss51Vr57G2pZrzW2ppqY0SCfpnfDJ8MhPZavryiZzIOfevwL+a2b865z486ciKTCKdoS+epi+RPmHFdV88zW+2H+VXTx1m2yFv08Cz5lXyFy9ewgvOqKe5OorPjFjI25Y1FpqdA01EpJgMrZQvG9G2OZ0ZmTB7SXMk6Ofs+VWcPf/4THPXQJLth3t58mAP2w718OttR/jp417X0epo0EuW51Wyel4VK+ZWQAD6Et7ECnit5iK5ZFnldDKVnPNmiH+3s40/PtvOxr2d7M+VS8RCflbPq+Lq1XO5YFEt57VUUxUNaeIuZyIzyJjZdXilE8Pt3cbaGW8mJcfOOfoSaXri6RPKJwCePtzLHZtbuffpYyTTWRbXl3HjCxdz+co5NFZ6T1M05Kc8HKAsFFBNmohIkQvkaihHJs2pXNKcSOUWW6ezVMdCXLSkjouWeDXNmaxjb3s/Tx7s4alDPTx5sIcHn233zukzljdWsLq5krObq1g9r4qqWPCEsgwzI5wry4gGVccsk9OfSPPQs+3cv/MYD+1qZ+eRPhwQDfo5Z0EVrzq3mYuX1LFmfhVl4YD+OBvDKRNkM/tvIAZcDnwFeB3wyBTHVVDZrKMnnqJnMH1CCUXWOR58pp0fbNjPkwd7iAb9vOzMRq5b08SyOeXeSmyfUREJUhEJqO5MRKTEDc00l+dWBg71rU+ks8RTGRKpLJBlSUM5SxrKefk58wBvlvnJg16yvLW1mx8/1soPNxwAvNK71c1VrJnvfTRWRoZ7M3flfm446CcS8BEN+YkE/EqYZUzOeWUT9zx1hAd2HGPD3k4Gkhl8BmfOq+RdL1zMpWc0cP6iasrDwUKHWzImVIPsnFtjZlucc/9oZp8Bbp/qwArBOUfPYJquweQJZRRZ53hgxzG+9ce97GkfoKkqwnsuX8rVZ82lLHfRDAV8VEWDlOuvMRGRGcub7fUTDvipjHjJRibrSKQzxFO5tp4pb5b5kjPqueSMesBbALj9cA9bW3t4orWb+3Yc5c4nvP7McyrCuWS5mnPmVzG/JjrcLaM7V8ccCviI5hZ0K2GWTCbLltZu7t56mPt3HGP7Ya/Es6E8zFWrGnnhsnouW9FAfXlYOclpmkiCPNTbY8DM5gHtwOKxDh6rzdvo+4pNfyJNR3+SVCZ7wv2b9nXy3/fv4pmjfSysjfHRa1dy2Yo5wzU64aCfmliQWGhC1SoiIjLD+H1GLBQgFvJuD80yx1NeaUY8lYUArJlfzZr51YCXVO9u6/d64rd2sXFvJ7/edhSA2rIQ5+QS5rULqmipjQ0v/FPCPHulM1k27O3kricO8dvtR4dbr61qquDPX7yEq1Z5HbJmUqu1QppIVvdzM6sGbgI24XWw+J9xji+pNm+ZrKOtL0F/In3C/Ud74nzxvmf53c42GivD/N01K7ly5fHEOBTwUVsWUmIsIiInGDnLTNSbZU6NKMuIpzIk01nOmFPOGXPKec15zTjn2N85yJYDXTy+v5vNB7q49+ljANTEgpwz39sA5dwF1SyojSphniWyWcdj+zv56eaD3PPUEQ52x/H7jHMXVPOWixZyzeq5tNTGNEs8BSbSxeKfc1/elttZL+Kc6x59XCm2eRtIpjnWm3hOOcXPHj/IzQ/sJusc77hkEX+ybsHwds4Bn4+asiAVEdXxiIjIxIyuZc5k3XCyPLSxVEttjJbaGNevmYdzjoNdcR4/0MXm/d7HfTu8hLmuLDScLJ/bUk1TVeQ5CbNqmEvb7rY+btvYyp1PHGJ3Wz8+g3ULa3nXCxdz7dlNzK2KFjrEGW8ii/QiwF8Cl+LNHv/ezL7knIuPPK7U2rx19ifpHNnZHWjvS/Cpu59mw95Ozl9Ywwdfspy5VcONO6iMBqmNhXShERGRSfH7jLJwYHgdSzbriOfqmAdzM8zNNVGaa6Jce3YTzjlauwbZvL97OGH+7XavJKOxMsy5C2o4t8VLmOvLw8+pYQ7nOmREgj4lzEWqL57i51sOcfumVh7d04HDaxv7oZet4NXnNjOvWknxdJpIfcC3gF7g/+VuvxG4BXj9yQ52zn3YzGqAZZzYFu6ByYWaPx39SbpGJceP7+/in37+FAPJDO+/ahkvX9M0/JZF0O+joSL8nEbzIiIi+eAbVceczToSaS9ZHswlu/NrYsyviXH9mqbhzgWP7eti074u/vBsG3c/eRjwumR4Gz3UsHZBNeWRwPEtsnOUMBcH5xxPHerh2w/t5a4nDtM9mGJORZi3vWARr183n1VNlSqfKJCJJMgrnHPnjLh9r5k9PtbBZvYu4H3AfGAzcBHwR0btvFdI6VEL8X6+5SD/+eudNFdH+ffXn8Pi+rLhxyqjQerKQhqgIiIybXw+I5rbdRWOzzAPJjPDM8wL68pYWFfGq85tJusczx7tY+O+Lh7b18kvtx7mJ5sP4jNY3ljB+QtrOK+lmrPmVREK+E5ImM3seA2z+jBPi0Qqwy+2HuY7D+/l0T2d+H3GpWfU84YLFnDlyjmENSFXcBNJkB8zs4uccw8BmNmFwB/GOf59wAXAQ865y81sJfCPkw81/5xzfP3BPXz7oX2sX1zLx65bNfx2l8+M+orwcL2YiIhIoRyfYT6xhnkw5SXNqUyWZY0VLGus4IYLFpBMZ9l2uIdNezvZuLeL7z2yj+88vI9wwMea+VWc11LDukU1LMlNCA0lzF2cmDBrhjm/OvoTfPfhfXzvkf20dg1SVx7ixhcu4c0XtbCwruzUJ5BpM5Hs70Lgz8xsX+52C7DNzJ4AnHNuzajj4865uJlhZmHn3HYzWzGZIM3s9cA/AKuA9c65DZM5H3jJ8Rfve5bbN7Vy7dlz+cBVy4c7VAT9PuZUhr0VyCIiIkVmdA1zKuOVY8RzM8yhgM/rfDG/mrdf4rUy3bzfK8fYtLeTLz+wiy8/4HXIOH9hDesW1nD+whrqysM4505akqFFf6evtXOAr/5+Nz/aeICeeJpVTRW85/LVvOrcZnXDKlIT+V+5+nme80CuLdwdwD1m1gkcfJ7nGG0r8Brgy5M8D+Alxzc/sIvbN7Xy2vOa+cvLlg6XUISDfuZWRrQXuYiIlIyhLhlDm5cMdcgYSHpbZJeFAydsXHKsN8HGvZ1s3NvJhj3HezAvri9j3UJvdnlNc9XwW/1jbVwSyX3oNfO5nPO2IP/yA7v48WOtxFNZLjmjjndduoQXLW/Qc1bkJtLmbe/zOaFz7tW5L//BzO4FqoC7TyO2kefcBuStDvj7j+7nBxsO8Mq1805IjiO55Fh/GYuISCkbSlyrY1798uCocoyGijBXr57L1avnknWOXcf62bCngw17O7ljcyu3bjxA0G+saa5i3aJaLlhUw+L6suHXy9Ft5apjIWrLQoX8JxeVAx0DfOn+Z7lt4wGSmSxXrJzDn794Kee11CjHKBFTOq/vnLt/Ks9/MmZ2I3AjQEtLy3Mev3vrYf7z1zt50bJ63nvFGUqOZdqcamyKFIrG5szmG1WOkUxnGUxmGEiliaey+GB405Ib1rcQT2XYcqCbDXs7eHTP8XKMurIQ6xbVcMGiWs5fWENV9Ph+ACP3E8inUhubbb0Jbv7dLr7z8F4GEhmuXDWH916+jDULqrTYv8QUTeGLmf0amHuShz7qnPvJRM/jnLuZ3MYk69ate85v7NyqCC9cVs+Hr1mJb1RZhZJjmUqnGpsihaKxObuEAj5CAR9VBIdnlweS3uxyOpslEvSzfnEt6xfXAl45xoY9XrL84LPt/PLJIxiwYm4F6xfVcsHiGi5YVDclsZbK2BxIpPn+hv186d5nOdaX4AVL63j/lctYt6hWuUWJKpoE2Tl31XT8nLULqvn3159DX25r6aDfp+RYRERmpdGzy4lcK7n+5PFFeg0VYa45u4lrzm4ik3XsONLLI7s7eHRPB99+eC/femgvlZEAn37dOVy9+mTzXDNXOpPlD8+08am7t/PUoV6WN5bzz68+i6tWNhLw+wodnkxC0STIheAzo1EL8kRERAAIB/yEA17tcibr6E+mvXKMZAbnHH6fsaqpklVNlbz1BYvoGUyxaV8nj+3rOmEPgZnOOceBzgE+e89O7nislepYkI9cu4q3XNhCTO1hZ4SS+F80s1fj7eTXANxpZpudcy+b7HnrK8KEAvoLT0REZDS/z6iMBKmMBHHOK8XoT2QYSKaHa44ro0EuWzGHl5/TTENFuMART494KsMdj7Vy06+epqMvyavObeZvXrac5upYoUOTPCqJBNk592Pgx/k8Z0UkqE1AREREJsBs5EYlYeKpDP2JNAO5rhizgXOOPW39fPKubfx621GW1Jdx0+vW8KJlDSqnmIFmZYYY8PuoHrH6VkRERCZuqI1cHV7dcnaG58jxVIa7tx7mk3duo70/wVsvXsj7rlxGbfnsmDWfjWZlglwTC6rdioiISB7M9F1n2/rifPaenXzv4X3Mr4nylbeu06zxLDArE2QlxyIiIjKeTNbx1KFuPnL7Vp5o7ebas+fysevOpKk6WujQZBrMygRZREREZCzxVIbfPHWEj/30SQaSaf7vdat484ULiYZm9my5HKcEWURERCSnN57ilj/u5bP37KCxMsLnbljLxUvqVFIxyyhBFhEREQHa+hJ89ldP891H9nN+SzWfeu0als4pV2nmLKQEWURERGY15xyHugb5+E+f5NfbjnL9mib+/uVn0lARKXRoUiBKkEVERGTWymYdezv6+bvbnuDh3R2845JFvO+qZVRFQ4UOTQpICbKIiIjMStmsY3dbPx+89XEe39/FX79kOe+4dLE2EhMlyCIiIjL7ZLOOXW39fPCHm3mitZsPX7uKt1zUktstUGY7jQIRERGZVZzzyir+9keP80RrNx+5dhVvulDJsRynniUiIiIyazjnaO0a5GN3PMnGfV186GUreON6JcdyIiXIIiIiMmsc7Y3zqbu38/tn2viry5fylosWUqaaYxmlJBJkM7vJzLab2RYz+7GZVRc6JhERESktnf1Jvvr7Pfzs8UPccMEC3vXCJVREgoUOS4pQSSTIwD3AaufcGmAH8OECxyMiIiIlpC+R5s4nDvI/D+zisuUNfOAly6mOqZWbnFxJJMjOuV8559K5mw8B8wsZj4iIiJSORDrDxj0d/Mtd21neWMHfv+JM5lSECx2WFLGSSJBHeQfwi7EeNLMbzWyDmW04duzYNIYlMj6NTSlWGptSrPIxNrNZx562fj7+kycJ+X188tWraakt0/bRMq6iSZDN7NdmtvUkH68cccxHgTTwnbHO45y72Tm3zjm3rqGhYTpCF5kQjU0pVhqbUqzyMTbb+hL8+692sL9zgI+9/ExWN1fh9yk5lvEVzbJN59xV4z1uZm8FrgeudM656YlKRERESlVPPMXtj7Vyz1NHeOvFC7lqVSORoL/QYUkJKJoEeTxmdjXwt8CLnXMDhY5HREREilsqk+Wp1h4+/5udnN1cxY0vWkpVVB0rZGKKpsTiFL4AVAD3mNlmM/vvQgckIiIixetIT5x/u3s7AB+9biVzqyIFjkhKSUnMIDvnzih0DCIiIlIaeuIpbt90gM37u/jgS5ZzdnO16o7leSmVGWQRERGRU0pnsjx9qJcv37+Lc1uq+ZN187VTnjxvSpBFRERkxugYSPJf9z1DMpPlQy9dQX2FSivk+VOCLCIiIjNCPJXhgR3HuPfpY7xpfQur56ulm5weJcgiIiIyIxztTfDFe5+lqSrCW1+wiMqIulbI6VGCLCIiIiWvL5Hmjsda2d3Wz40vWsK86mihQ5ISpgRZRERESppzjoOdg3zjwT2cNa+Sa8+eqw1BZFKUIIuIiEhJ60ukuXXjfjr6k7z7xUupKwsXOiQpcUqQRUREpKS1dg7yg0f3s35RDZcsrSPgV3ojk6MRJCIiIiXt1o376Ymneceli6mOhQodjswASpBFRESkZA0mM9y2sZX1i2pYv7hObd0kL5Qgi4iISMn64Yb9dA2meNOFC6mKqq2b5IcSZBERESlZt27cz+p5lVxyhmaPJX+0ObmIiIiUrB/ceDGb9nWq9ljyqiRmkM3sn81si5ltNrNfmdm8QsckIiIihVcWDrCisYKgOldIHpXKaLrJObfGObcW+Dnw8QLHIyIiIkWiUrXHkmclkSA753pG3CwDXKFiERERkeKiXfMk30oiQQYws0+a2X7gzYwzg2xmN5rZBjPbcOzYsekLUOQUNDalWGlsSrHS2JRCKZoE2cx+bWZbT/LxSgDn3EedcwuA7wB/NdZ5nHM3O+fWOefWNTQ0TFf4IqeksSnFSmNTipXGphRK0XSxcM5dNcFDvwvcCfz9FIYjIiIiIrOUOVf85bxmtsw5tzP39XuBFzvnXjeB7zsG7J3q+EaoB9qm8ec9H4rt9IyObaFzbtLTGAUYm1OtmP8PC2k6n5eZODZLaVwp1rHNxLE5EaU0JvKhVP+9Jx2fpZIg3wasALJ4vxx/7pxrLWxUz2VmG5xz6wodx8kottNTzLEVEz1PJ6fnZXJK6flTrDLabHueZ9q/t2hKLMbjnHttoWMQERERkdmhaBbpiYiIiIgUAyXI+XVzoQMYh2I7PcUcWzHR83Ryel4mp5SeP8Uqo82253lG/XtLogZZRERERGS6aAZZRERERGQEJcgiIiIiIiMoQc4TM9tjZk+Y2WYz21DgWL5mZkfNbOuI+2rN7B4z25n7XFNEsf2DmbXmnrvNZnZtgWJbYGb3mtk2M3vSzN6Xu78onrtiZWbVZvYjM9uee+4uLnRMhWRmK0aM5c1m1mNm7y90XMXuZNeG3P3vNbOnc7+Tny5UfCONcR1ba2YPDb0GmNn6QsaYi0nXtGlSzK+7+TZbxpUS5Py63Dm3tgj6AH4DuHrUfX8H/MY5twz4Te52IXyD58YG8B+5526tc+6uaY5pSBr4oHNuFXAR8B4zO5Piee6K1eeAu51zK4FzgG0FjqegnHNPD41l4HxgAPhxYaMqCd9g1LXBzC4HXgmscc6dBfx7AeI6mW/w3OvYp4F/zP2/fzx3u9B0TZs+36B4X3fzbVaMKyXIM5Bz7gGgY9TdrwS+mfv6m8CrpjOmIWPEVhScc4ecc5tyX/fiJXrNFMlzV4zMrBJ4EfBVAOdc0jnXVdCgisuVwLPOuVLa/asgxrg2/AXwb865RO6Yo9Me2EmMEasDKnNfVwEHpzWok9A1bfoU8+tuvs2WcaUEOX8c8Csz22hmNxY6mJNodM4dAm9wA3MKHM9of2VmW3JvUxX8bRkzWwScCzxM8T93hbQEOAZ83cweM7OvmFlZoYMqIjcA3yt0ECVsOfBCM3vYzO43swsKHdA43g/cZGb78Wa6P1zYcE6ka1pBzPjneSaPKyXI+XOJc+484Bq8txteVOiASsiXgKXAWuAQ8JlCBmNm5cBtwPudcz2FjKUEBIDzgC85584F+inxt9XyxcxCwCuAWwsdSwkLADV4b+N+CPihmVlhQxrTXwAfcM4tAD5A7l2VYqBrmkyFmT6ulCDniXPuYO7zUbx6w4Iv0BjliJk1AeQ+F8VblQDOuSPOuYxzLgv8DwV87swsiPcL/x3n3O25u4v2uSsCB4ADzrmHc7d/hJcwi/fH8ibn3JFCB1LCDgC3O88jQBaoL3BMY3krMHTNuJUieQ3QNa2gZuzzPBvGlRLkPDCzMjOrGPoaeCmwdfzvmnY/xbuAk/v8kwLGcoKhX6icV1Og5y43M/VVYJtz7rMjHira567QnHOHgf1mtiJ315XAUwUMqZi8EZVXTNYdwBUAZrYcCAFthQxoHAeBF+e+vgLYWcBYAF3TisCMfJ5ny7jSTnp5YGZLOL5KPQB81zn3yQLG8z3gMryZliPA3+O90PwQaAH2Aa93zk37YrkxYrsMr7zCAXuAdw/VMU1zbJcCvwOewJupAvgIXm1VwZ+7YmVma4Gv4CUvu4C3O+c6CxpUgZlZDNgPLHHOdRc6nlIwxrXhFuBreNeHJPA3zrnfFijEYWPE+jReR5cAEAf+0jm3sVAxgq5p06mYX3fzbbaMKyXIIiIiIiIjqMRCRERERGQEJcgiIiIiIiMoQRYRERERGUEJsoiIiIjICEqQRURERERGUII8Q5hZX+7zIjNzZvbeEY99wczelvv6G2a228weN7MdZvYtM2sefZ4Rt99mZl/Ifb3CzO4zs81mts3Mbp6Wf5zMSLlx+pkRt//GzP5hxO0bzWx77uORXGshzOyvzeyrI457s5ndOa3By6xhZpncNW+rmf3MzKpz9w9da/95xLH1ZpYaumaKTNbo1+Tcfc95LTazl+VubzazPjN7Ovf1t3Lf8+rceF2Zu/1w7vF9ZnZsxPcumuZ/YtFSgjwzHQXel9vq9mQ+5Jw7B1gBPAbcO86xI30e+A/n3Frn3Crg/+UnXJmlEsBrzOw5O6OZ2fXAu4FLnXMrgT8Hvmtmc/HG4flmdkkuWfkE8N7R5xDJk8HcNW810AG8Z8Rju4DrR9x+PfDkdAYns9JzXoudc7/M3V4LbADenLv9Z7nveSPwe+AGAOfchbljPw78YOh7nXN7pvsfU6yUIM9Mx4DfcHxHm5PKbd/6H8BhvG1xT6UJb+vXoe9/YjJByqyXBm4GPnCSx/4W7w+5NgDn3Cbgm8B7nHNp4C+BLwKfBr7mnNs1PSHLLPdHoHnE7UFgm5mty91+A95GCSJT6Xm9FptZOXAJ8E5yCbKcmhLkmevfgA+amX8Cx24CVk7guP8AfmtmvzCzDwy91SgyCV8E3mxmVaPuPwsYvQvZhtz9OOceBLYBV+ElySJTKnctvRJvO92Rvg/cYGbzgQzeltMiU+n5vha/CrjbObcD6DCz86Y6wJlACfIM5ZzbDTwCvGkCh9upTpc759eBVcCteFtqPmRm4UmEKbOcc64H+BbwvydwuJEbi7kZkXVAEGiYsgBFIGpmm4F2oBa4Z9TjdwMvwXsL+wfTG5rMRqfxWvxGvD/kyH1+45QGOEMoQZ7Z/gXvrepT/T+fizcbBzA4qh65FmgbuuGcO+ic+5pz7pV4b5GvzmO8Mjv9J95bf2Uj7nsKOH/Ucefl7gf4R+DbwCfxZlNEpspgrlZzIRDixBpknHNJvHc7PgjcNu3Ryaw00ddiM6sDrgC+YmZ7gA8BbzCzU02MzXpKkGcw59x2vITi+pM9bp7/jVfPdHfu7vuBt+QejwJ/Atybu321mQVzX88F6oDWqfw3yMznnOvAq9t854i7Pw18Kndxx8zWAm8D/svMzgauAz6FV8O80MxeMp0xy+zjnOvGe6fjb4augyN8Bvhb51z79Ecms83zfC1+HfAt59xC59wi59wCYDdw6fREW7oChQ5Aptwn8TpVjHSTmX0MiAEPAZfnZkEA3gd8OZc4G94v1gO5x14KfM7M4rnbH3LOHZ7a8GWW+AzwV0M3nHM/zbUffNDMHNCL94fbYby3FT/gnIsDmNlfAt8ys7UjxrFI3jnnHjOzx/EWOv1uxP1Pou4VMjViZnZgxO3PAvOZ+GvxG/HWJI10G1755e+ee7gMMedcoWMQERERESkaKrEQERERERlBCbKIiIiIyAhKkEVERERERlCCLCIiIiIyghJkEREREZERlCCLiIiIiIygBFlEREREZIT/D8Y0c/Ql/WHYAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x216 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 4, figsize=(10, 3))\n",
    "\n",
    "explainer.plot_partial_effects_at_means(\n",
    "    X=just_transformers(best_pipeline, X_test),\n",
    "    features=range(4),\n",
    "    ax=axs,\n",
    "    num_points=100,\n",
    "    share_y=True,\n",
    "    show_err=True,\n",
    "    show=False\n",
    ")\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}