Non_Gen_LinCFA_experiments.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5703da24",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Preliminaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "42be8005",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "import pandas as pd\n",
    "sys.path.append(\"../PaperNon_Gen_LinCFA/\")\n",
    "from NonLinCFA.NonLinCFA import NonLinCFA\n",
    "from sklearn.linear_model import LinearRegression,LogisticRegression\n",
    "from sklearn.metrics import r2_score,accuracy_score\n",
    "import matplotlib.pyplot as plt\n",
    "from NonLinCFA.NonLinCFA_anyFunction import NonLinCFA_anyFunction\n",
    "from LinCFA.LinCFA import LinCFA\n",
    "from GenLinCFA.GenLinCFA import GenLinCFA\n",
    "from GenLinCFA.GenLinCFA_anyFunction import GenLinCFA_anyFunction\n",
    "import numpy as np\n",
    "from sklearn.feature_selection import SequentialFeatureSelector as SFS\n",
    "import random\n",
    "from DdimensionalExperiment import generate_dataset_n,generate_dataset_quadratic,run_GenLinCFA\n",
    "from DdimensionalExperiment_quadratic import generate_dataset_squared\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2e114354",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def compute_r2(x_train, y_train, x_val, y_val):\n",
    "    regr = LinearRegression().fit(x_train,y_train)\n",
    "    y_pred = regr.predict(x_val)\n",
    "    return r2_score(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b196ebf2",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def compute_accuracy(x_train, y_train, x_val, y_val):\n",
    "    regr = LogisticRegression().fit(x_train,y_train)\n",
    "    y_pred = regr.predict(x_val)\n",
    "    return accuracy_score(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8a75bc2f",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from mlxtend.feature_selection import SequentialFeatureSelector as SFS\n",
    "\n",
    "def compute_wrapper(x_trainVal, y_trainVal,n_features):\n",
    "    sfs = SFS(LinearRegression(),\n",
    "           k_features=n_features,\n",
    "           forward=True,\n",
    "           floating=False,\n",
    "           scoring = 'r2',\n",
    "           cv = 5)\n",
    "    \n",
    "    sfs.fit(x_trainVal, y_trainVal)\n",
    "    return pd.DataFrame(sfs.subsets_).transpose()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bb16884b",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def squared_aggregation(x):\n",
    "    x_sq = x**2\n",
    "    return x_sq.sum(axis=1)\n",
    "\n",
    "def mean_aggregation(x):\n",
    "    return x.mean(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9c79db8d",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "def linear_experiment_resampling():\n",
    "    score = []\n",
    "    list_of_length = []\n",
    "    wrapper_score = []\n",
    "    LinCFA_score = []\n",
    "    list_of_length_lin = []\n",
    "    GenLinCFA_score = []\n",
    "    list_of_length_gen = []\n",
    "    \n",
    "    for trials in range(10):\n",
    "        x,y,coeffs = generate_dataset_n(n_data=3000, noise=10, p1=0.3, p2=0.7, n_variables=100, coeffs=[0])\n",
    "        x = pd.DataFrame(x)\n",
    "        x['target'] = y\n",
    "        \n",
    "        for eps in [0.1,0.01,0.001,0.0001,0.00001,0.000001,0.0000001]:\n",
    "            output = NonLinCFA_anyFunction(x.iloc[:2000,:],'target', eps, 5 , neigh=0, customFunction=mean_aggregation).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "            \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "            \n",
    "            #print(aggregate_x.head())\n",
    "            score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(score[-1])\n",
    "        \n",
    "        for eps in [0.3,0.35,0.375,0.4,0.4125,0.425,0.45]:\n",
    "            output = GenLinCFA_anyFunction(x.iloc[:2000,:],'target', eps1=eps, n_val=5 , neigh=0, eps2=1, customFunction=mean_aggregation).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length_gen.append(len(output))\n",
    "            #list_of_length.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "        \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "        \n",
    "        #print(aggregate_x.head())\n",
    "            GenLinCFA_score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(GenLinCFA_score[-1])\n",
    "        \n",
    "        #for i in range(50):\n",
    "        res = compute_wrapper(x.iloc[:2000,:-1], x.iloc[:2000,-1], 50)\n",
    "        for i in range(50):\n",
    "            x_wrapper_red = x[list(res.loc[i+1,'feature_names'])]\n",
    "            wrapper_score.append(compute_r2(x_wrapper_red.iloc[:2000],x.loc[:,'target'].iloc[:2000],x_wrapper_red.iloc[2000:],x.loc[:,'target'].iloc[2000:]))\n",
    "            print(i+1,wrapper_score[-1])\n",
    "        \n",
    "        for eps in [0]:\n",
    "            output = LinCFA(x.iloc[:2000,:],'target', eps, neigh=0).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length_lin.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "            \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "            \n",
    "            #print(aggregate_x.head())\n",
    "            LinCFA_score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(LinCFA_score[-1])\n",
    "    return score,list_of_length,wrapper_score,LinCFA_score,list_of_length_lin,GenLinCFA_score,list_of_length_gen\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "76f3533c",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def linear_experiment_noResampling(n_reps):\n",
    "    score = []\n",
    "    list_of_length = []\n",
    "    wrapper_score = []\n",
    "    LinCFA_score = []\n",
    "    list_of_length_lin = []\n",
    "    GenLinCFA_score = []\n",
    "    list_of_length_gen = []\n",
    "    \n",
    "    x_all,y_all,coeffs = generate_dataset_n(n_data=3000*10, noise=10, p1=0.3, p2=0.7, n_variables=100, coeffs=[0])\n",
    "    \n",
    "    for trials in range(n_reps):\n",
    "        x = pd.DataFrame(x_all[3000*trials:3000*(trials+1)])\n",
    "        x['target'] = y_all[3000*trials:3000*(trials+1)]\n",
    "        \n",
    "        for eps in [0.1,0.01,0.001,0.0001,0.00001,0.000001,0.0000001]:\n",
    "            output = NonLinCFA_anyFunction(x.iloc[:2000,:],'target', eps, 5 , neigh=0, customFunction=mean_aggregation).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "            \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "            \n",
    "            #print(aggregate_x.head())\n",
    "            score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(score[-1])\n",
    "        \n",
    "        for eps in [0.3,0.35,0.375,0.4,0.4125,0.425,0.45]:\n",
    "            output = GenLinCFA_anyFunction(x.iloc[:2000,:],'target', eps1=eps, n_val=5 , neigh=0, eps2=1, customFunction=mean_aggregation).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length_gen.append(len(output))\n",
    "            #list_of_length.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "        \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "        \n",
    "        #print(aggregate_x.head())\n",
    "            GenLinCFA_score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(GenLinCFA_score[-1])\n",
    "        \n",
    "        #for i in range(50):\n",
    "        res = compute_wrapper(x.iloc[:2000,:-1], x.iloc[:2000,-1], 50)\n",
    "        for i in range(50):\n",
    "            x_wrapper_red = x[list(res.loc[i+1,'feature_names'])]\n",
    "            wrapper_score.append(compute_r2(x_wrapper_red.iloc[:2000],x.loc[:,'target'].iloc[:2000],x_wrapper_red.iloc[2000:],x.loc[:,'target'].iloc[2000:]))\n",
    "            print(i+1,wrapper_score[-1])\n",
    "        \n",
    "        for eps in [0]:\n",
    "            output = LinCFA(x.iloc[:2000,:],'target', eps, neigh=0).compute_clusters()\n",
    "            print(len(output))\n",
    "            list_of_length_lin.append(len(output))\n",
    "            aggregate_x = pd.DataFrame()\n",
    "            \n",
    "            for i in range(len(output)):\n",
    "                aggregate_x[i] = mean_aggregation(x[output[i]])\n",
    "            \n",
    "            #print(aggregate_x.head())\n",
    "            LinCFA_score.append(compute_r2(aggregate_x.iloc[:2000,:], x.loc[:,'target'].iloc[:2000], aggregate_x.iloc[2000:,:], x.loc[:,'target'].iloc[2000:]))\n",
    "            print(LinCFA_score[-1])\n",
    "    return score,list_of_length,wrapper_score,LinCFA_score,list_of_length_lin,GenLinCFA_score,list_of_length_gen\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "808422ea",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def compute_CI_fromList(lis, list_len, n_reps):\n",
    "    res = []\n",
    "    for i in range(list_len):\n",
    "        res.append([np.mean(lis[i::list_len])-1.96*np.std(lis[i::list_len])/np.sqrt(n_reps),np.mean(lis[i::list_len])+1.96*np.std(lis[i::list_len])/np.sqrt(n_reps)])\n",
    "    return res\n",
    "\n",
    "def compute_lower_fromList(lis, list_len, n_reps):\n",
    "    res = []\n",
    "    for i in range(list_len):\n",
    "        res.append(np.mean(lis[i::list_len])-1.96*np.std(lis[i::list_len])/np.sqrt(n_reps))\n",
    "    return res\n",
    "\n",
    "def compute_upper_fromList(lis, list_len, n_reps):\n",
    "    res = []\n",
    "    for i in range(list_len):\n",
    "        res.append(np.mean(lis[i::list_len])+1.96*np.std(lis[i::list_len])/np.sqrt(n_reps))\n",
    "    return res\n",
    "\n",
    "def compute_means_fromList(lis, list_len):\n",
    "    res = []\n",
    "    for i in range(list_len):\n",
    "        res.append(np.mean(lis[i::list_len]))\n",
    "    return res\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "102d853b",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Linear dependence, input=feature, mean as aggregation, first trial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7a2f17be",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "0.8817966330061661\n",
      "1\n",
      "0.8817966330061661\n",
      "5\n",
      "0.8820215451678672\n",
      "21\n",
      "0.8836656019498874\n",
      "25\n",
      "0.882466462744487\n",
      "23\n",
      "0.8825802084847787\n",
      "22\n",
      "0.8835012309635005\n",
      "40\n",
      "0.8807109875791379\n",
      "25\n",
      "0.8811377170596879\n",
      "15\n",
      "0.8824218989043883\n",
      "10\n",
      "0.8825791759538643\n",
      "6\n",
      "0.8819844243286494\n",
      "4\n",
      "0.8820360033544787\n",
      "1\n",
      "0.8817966330061661\n",
      "1 0.7538899708223442\n",
      "2 0.7807372947467764\n",
      "3 0.8018624376795207\n",
      "4 0.8121997255736462\n",
      "5 0.8223522064402358\n",
      "6 0.8301376030616282\n",
      "7 0.836971620504213\n",
      "8 0.8472091752798463\n",
      "9 0.8487582407976074\n",
      "10 0.8485366254834882\n",
      "11 0.8520882837685617\n",
      "12 0.85485028438627\n",
      "13 0.8569162541182482\n",
      "14 0.8590486947342697\n",
      "15 0.8610609071227546\n",
      "16 0.8625731877364119\n",
      "17 0.8621368680842687\n",
      "18 0.8634178501331029\n",
      "19 0.861534561980402\n",
      "20 0.8634617790187009\n",
      "21 0.8661574248553321\n",
      "22 0.8670858214956714\n",
      "23 0.8668547257571438\n",
      "24 0.8671502564259118\n",
      "25 0.8683785669960351\n",
      "26 0.8694981631041512\n",
      "27 0.8693846160024189\n",
      "28 0.8703976206949713\n",
      "29 0.8703485650157752\n",
      "30 0.8718388857601755\n",
      "31 0.8727298950072098\n",
      "32 0.8724391010744208\n",
      "33 0.8725643411605735\n",
      "34 0.8733190511739171\n",
      "35 0.8738477729966342\n",
      "36 0.8732359779655555\n",
      "37 0.8737819206255997\n",
      "38 0.8735903358742803\n",
      "39 0.8736312622260098\n",
      "40 0.8744395913367708\n",
      "41 0.8743138407009784\n",
      "42 0.8748185385250713\n",
      "43 0.8743870384327865\n",
      "44 0.8744340736756512\n",
      "45 0.8749512119424829\n",
      "46 0.8752977385925043\n",
      "47 0.8759654528999264\n",
      "48 0.8760621011901409\n",
      "49 0.876557929062442\n",
      "50 0.8763987174787007\n",
      "38\n",
      "0.8823358113396513\n",
      "1\n",
      "0.91106618849304\n",
      "1\n",
      "0.91106618849304\n",
      "5\n",
      "0.9101400075339009\n",
      "15\n",
      "0.91160082609768\n",
      "13\n",
      "0.9115319583526522\n",
      "23\n",
      "0.9114379744824214\n",
      "17\n",
      "0.9124817684719405\n",
      "53\n",
      "0.9092813922656334\n",
      "39\n",
      "0.9103958808794632\n",
      "27\n",
      "0.9110184904038476\n",
      "17\n",
      "0.9115009171589483\n",
      "14\n",
      "0.9106829797829942\n",
      "8\n",
      "0.9116945449100868\n",
      "3\n",
      "0.9115817937132668\n",
      "1 0.813198633949797\n",
      "2 0.854097649121891\n",
      "3 0.86741324172678\n",
      "4 0.8761411608734265\n",
      "5 0.8805287530320709\n",
      "6 0.8870426665260804\n",
      "7 0.8876569337190298\n",
      "8 0.8881053083320174\n",
      "9 0.8891972452213552\n",
      "10 0.8904818730749215\n",
      "11 0.8912806215748436\n",
      "12 0.89255839835785\n",
      "13 0.891558623312972\n",
      "14 0.8914180928416399\n",
      "15 0.8915981334414229\n",
      "16 0.8927320122222825\n",
      "17 0.8935023548956315\n",
      "18 0.8945722188973139\n",
      "19 0.8963083441949142\n",
      "20 0.8972692552648982\n",
      "21 0.8978030725721735\n",
      "22 0.8988556819625136\n",
      "23 0.8986581189749111\n",
      "24 0.8984781904833556\n",
      "25 0.8989749851910773\n",
      "26 0.9005810902141432\n",
      "27 0.9015062620833986\n",
      "28 0.9019222650803823\n",
      "29 0.9024768510423054\n",
      "30 0.9029567924768638\n",
      "31 0.9035925451361155\n",
      "32 0.9037319466579146\n",
      "33 0.9041207276536731\n",
      "34 0.9040577872132808\n",
      "35 0.9043269794920359\n",
      "36 0.9059617094231289\n",
      "37 0.9063857309848914\n",
      "38 0.906296989941653\n",
      "39 0.9067084334773216\n",
      "40 0.9075691514156058\n",
      "41 0.9078408526654093\n",
      "42 0.907540510426315\n",
      "43 0.9072926263049235\n",
      "44 0.9075035958730222\n",
      "45 0.9078283830885686\n",
      "46 0.9080863244618765\n",
      "47 0.9082795543506708\n",
      "48 0.9088216964638277\n",
      "49 0.9089588859424927\n",
      "50 0.9090780604142863\n",
      "38\n",
      "0.911477089706728\n",
      "1\n",
      "0.8938357827875488\n",
      "1\n",
      "0.8938357827875488\n",
      "2\n",
      "0.8933648935431088\n",
      "5\n",
      "0.8945052624538864\n",
      "6\n",
      "0.8949190228749424\n",
      "10\n",
      "0.8942782722152905\n",
      "10\n",
      "0.8942782722152905\n",
      "43\n",
      "0.8941882199795516\n",
      "25\n",
      "0.8957318764427504\n",
      "21\n",
      "0.8941607175833451\n",
      "11\n",
      "0.894032528030142\n",
      "6\n",
      "0.8938627017612424\n",
      "4\n",
      "0.893442090464509\n",
      "1\n",
      "0.8938357827875488\n",
      "1 0.7477291166317882\n",
      "2 0.8037540113557677\n",
      "3 0.8207265370915185\n",
      "4 0.8384392574416526\n",
      "5 0.8456115435596163\n",
      "6 0.8561059702289677\n",
      "7 0.8639692456284641\n",
      "8 0.8677249827365994\n",
      "9 0.8682415652743289\n",
      "10 0.8722150520769133\n",
      "11 0.8719918612812864\n",
      "12 0.8727996417500075\n",
      "13 0.8752734669072232\n",
      "14 0.8777758947144445\n",
      "15 0.8792863860415965\n",
      "16 0.8795724980892349\n",
      "17 0.8810813537508634\n",
      "18 0.8832015797252908\n",
      "19 0.8825307456512195\n",
      "20 0.8825925284689509\n",
      "21 0.8825471821884069\n",
      "22 0.883867306297145\n",
      "23 0.884937299732934\n",
      "24 0.8866793944191624\n",
      "25 0.8866421161871763\n",
      "26 0.8875253640741196\n",
      "27 0.8880591703704678\n",
      "28 0.8888950280444119\n",
      "29 0.8900608314172772\n",
      "30 0.8903339750878689\n",
      "31 0.8912525400604684\n",
      "32 0.891673383528609\n",
      "33 0.8926312416266551\n",
      "34 0.893778161521089\n",
      "35 0.8928574326441653\n",
      "36 0.8934794695261654\n",
      "37 0.8937060088529317\n",
      "38 0.8937043303180213\n",
      "39 0.8950001420098832\n",
      "40 0.895376851886726\n",
      "41 0.8956880563392468\n",
      "42 0.8957052803361024\n",
      "43 0.8952337796902394\n",
      "44 0.8953819206420749\n",
      "45 0.8955068902323094\n",
      "46 0.8954837474613424\n",
      "47 0.8953483199240362\n",
      "48 0.8953562632649206\n",
      "49 0.8962256619254497\n",
      "50 0.8966732024196364\n",
      "41\n",
      "0.8966476175581879\n",
      "1\n",
      "0.8923748660274042\n",
      "1\n",
      "0.8923748660274042\n",
      "1\n",
      "0.8923748660274042\n",
      "20\n",
      "0.8938888255909515\n",
      "18\n",
      "0.8934503265427532\n",
      "18\n",
      "0.8934503265427532\n",
      "18\n",
      "0.8934503265427532\n",
      "48\n",
      "0.8915082036782233\n",
      "31\n",
      "0.8923662582210266\n",
      "20\n",
      "0.890530448665404\n",
      "13\n",
      "0.8924477900187984\n",
      "7\n",
      "0.891975394420147\n",
      "4\n",
      "0.8921444595308147\n",
      "1\n",
      "0.8923748660274042\n",
      "1 0.7210765955805243\n",
      "2 0.7803847181818074\n",
      "3 0.821522661075646\n",
      "4 0.8325825117361629\n",
      "5 0.8402143938122486\n",
      "6 0.8473582682156438\n",
      "7 0.850095781695352\n",
      "8 0.853482833052041\n",
      "9 0.8567016890147588\n",
      "10 0.859807921561173\n",
      "11 0.8633712002008676\n",
      "12 0.8661781991521957\n",
      "13 0.8687361238172204\n",
      "14 0.8703120679789405\n",
      "15 0.8725655585499588\n",
      "16 0.8748724879889932\n",
      "17 0.8754554301969674\n",
      "18 0.8771787926404859\n",
      "19 0.8782710583035038\n",
      "20 0.8770367180181219\n",
      "21 0.8776774555437473\n",
      "22 0.879393402165475\n",
      "23 0.8804120146335086\n",
      "24 0.8814560529296013\n",
      "25 0.8818415041354333\n",
      "26 0.8818368726712688\n",
      "27 0.8827181472509313\n",
      "28 0.8827262974164433\n",
      "29 0.8830575992709702\n",
      "30 0.8833825500693503\n",
      "31 0.8844368995555789\n",
      "32 0.8842740857300093\n",
      "33 0.8842619676616424\n",
      "34 0.8851837940757131\n",
      "35 0.8852388120002783\n",
      "36 0.885572524341593\n",
      "37 0.8863215471593141\n",
      "38 0.8866925552445025\n",
      "39 0.8867249997533078\n",
      "40 0.8870235977619066\n",
      "41 0.887192509692501\n",
      "42 0.8875445483301121\n",
      "43 0.8877198036932292\n",
      "44 0.8879810217107125\n",
      "45 0.8889774760423483\n",
      "46 0.8890699087784499\n",
      "47 0.88956367792938\n",
      "48 0.889962445292012\n",
      "49 0.8898295148856503\n",
      "50 0.8901933703477753\n",
      "33\n",
      "0.8934249858245328\n",
      "1\n",
      "0.8988529951853931\n",
      "1\n",
      "0.8988529951853931\n",
      "12\n",
      "0.8997008243253966\n",
      "14\n",
      "0.9002220160517571\n",
      "13\n",
      "0.8990090833164567\n",
      "18\n",
      "0.9007805592893158\n",
      "17\n",
      "0.9006510522702806\n",
      "71\n",
      "0.8991686704782138\n",
      "52\n",
      "0.8994675765901823\n",
      "44\n",
      "0.8984309390417344\n",
      "22\n",
      "0.8980210249447347\n",
      "14\n",
      "0.8975331546909606\n",
      "13\n",
      "0.8985403511810863\n",
      "4\n",
      "0.8997152081149061\n",
      "1 0.8318142544094627\n",
      "2 0.8674090633026845\n",
      "3 0.878935110204864\n",
      "4 0.8822041298324192\n",
      "5 0.8840117699789536\n",
      "6 0.8843204569377823\n",
      "7 0.8826997242536319\n",
      "8 0.8816194310078896\n",
      "9 0.8840023828791114\n",
      "10 0.8868159679350566\n",
      "11 0.8875431937220708\n",
      "12 0.8885699602757614\n",
      "13 0.890000141464048\n",
      "14 0.891732103563853\n",
      "15 0.8923871605020688\n",
      "16 0.8935413299805814\n",
      "17 0.8939564076874961\n",
      "18 0.8934936057686075\n",
      "19 0.8932741863325997\n",
      "20 0.8940674787087837\n",
      "21 0.8942335809291578\n",
      "22 0.8942631765023282\n",
      "23 0.8944538824542241\n",
      "24 0.8940997644456179\n",
      "25 0.8940510331954669\n",
      "26 0.8942793075281292\n",
      "27 0.8946330001616154\n",
      "28 0.894865436891629\n",
      "29 0.8947362863193801\n",
      "30 0.8945341319580727\n",
      "31 0.894176960141455\n",
      "32 0.8944163726606109\n",
      "33 0.8951326512842828\n",
      "34 0.8955662581501974\n",
      "35 0.8960926751726515\n",
      "36 0.896228040367755\n",
      "37 0.8968632875649457\n",
      "38 0.8968281961160237\n",
      "39 0.8973046334550147\n",
      "40 0.8977374628523311\n",
      "41 0.8975463461182401\n",
      "42 0.8976186185962435\n",
      "43 0.8977760322557196\n",
      "44 0.8975986289605258\n",
      "45 0.8975647946459531\n",
      "46 0.8977087294724915\n",
      "47 0.8978563206782315\n",
      "48 0.8980500783945583\n",
      "49 0.8980567621039155\n",
      "50 0.8983743065825933\n",
      "44\n",
      "0.8985168932979759\n",
      "1\n",
      "0.9101938301012555\n",
      "1\n",
      "0.9101938301012555\n",
      "9\n",
      "0.9103741942675727\n",
      "9\n",
      "0.909714113131202\n",
      "20\n",
      "0.9099979692475726\n",
      "21\n",
      "0.908888902358745\n",
      "21\n",
      "0.908888902358745\n",
      "53\n",
      "0.9071364052241893\n",
      "33\n",
      "0.9084882471430875\n",
      "22\n",
      "0.9087660496187052\n",
      "17\n",
      "0.9089374665008947\n",
      "12\n",
      "0.9095542516192445\n",
      "6\n",
      "0.910756845966366\n",
      "1\n",
      "0.9101938301012555\n",
      "1 0.8018993823675564\n",
      "2 0.8264152798792486\n",
      "3 0.858073598984476\n",
      "4 0.8678169056437302\n",
      "5 0.8711112450175647\n",
      "6 0.8779990327606814\n",
      "7 0.8816695450330211\n",
      "8 0.8852003873906295\n",
      "9 0.8880014633699652\n",
      "10 0.889076681604355\n",
      "11 0.8898269291275189\n",
      "12 0.8916036221818263\n",
      "13 0.8927486769401739\n",
      "14 0.8929065493973224\n",
      "15 0.8954096673427046\n",
      "16 0.8957844978897384\n",
      "17 0.8962484534159569\n",
      "18 0.8967798004107588\n",
      "19 0.8968466899229816\n",
      "20 0.8977726205917325\n",
      "21 0.8987510724758159\n",
      "22 0.8989328092055361\n",
      "23 0.8995999497302316\n",
      "24 0.9013028150017732\n",
      "25 0.9012394650348011\n",
      "26 0.9015701813618076\n",
      "27 0.9021417421423329\n",
      "28 0.903063016763396\n",
      "29 0.9030474714175902\n",
      "30 0.9028495908306716\n",
      "31 0.902784277950796\n",
      "32 0.9030812931804132\n",
      "33 0.9029622303049862\n",
      "34 0.9030442726720361\n",
      "35 0.9026264302071471\n",
      "36 0.9026351681615964\n",
      "37 0.9023507032611972\n",
      "38 0.902623868183203\n",
      "39 0.9025408044129426\n",
      "40 0.9028520647059991\n",
      "41 0.9031356855181989\n",
      "42 0.9032993643953035\n",
      "43 0.9034033991764023\n",
      "44 0.9033302459088925\n",
      "45 0.9039074536182592\n",
      "46 0.904169612303657\n",
      "47 0.9045485652098533\n",
      "48 0.9050403599591866\n",
      "49 0.9045557042426171\n",
      "50 0.9051415620688396\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40\n",
      "0.9089741011878893\n",
      "1\n",
      "0.876230303021237\n",
      "1\n",
      "0.876230303021237\n",
      "11\n",
      "0.8770713485273053\n",
      "11\n",
      "0.8775482674469037\n",
      "20\n",
      "0.8778254582073721\n",
      "18\n",
      "0.8764752004275123\n",
      "18\n",
      "0.8764752004275123\n",
      "22\n",
      "0.877134955856377\n",
      "11\n",
      "0.8767120352684915\n",
      "6\n",
      "0.8781055298502877\n",
      "3\n",
      "0.8765147580978842\n",
      "2\n",
      "0.8761844473362717\n",
      "1\n",
      "0.876230303021237\n",
      "1\n",
      "0.876230303021237\n",
      "1 0.6292582595571139\n",
      "2 0.7335451247070128\n",
      "3 0.7764866956946761\n",
      "4 0.784934714763992\n",
      "5 0.8025035663256861\n",
      "6 0.8117842328060121\n",
      "7 0.8258976251085055\n",
      "8 0.8311161183027505\n",
      "9 0.8383827927195169\n",
      "10 0.8409541748829819\n",
      "11 0.8462093783257971\n",
      "12 0.8490644874716065\n",
      "13 0.8510061389040199\n",
      "14 0.8517118471563297\n",
      "15 0.8534829683854734\n",
      "16 0.852941216105561\n",
      "17 0.8535328871544329\n",
      "18 0.853582097217466\n",
      "19 0.854746934701174\n",
      "20 0.8561741097099385\n",
      "21 0.8567599988699144\n",
      "22 0.8577781894684234\n",
      "23 0.8589939886378106\n",
      "24 0.8587978379164836\n",
      "25 0.8596432151302809\n",
      "26 0.8612018151026146\n",
      "27 0.8623009214842121\n",
      "28 0.8643004665430194\n",
      "29 0.8648707620615598\n",
      "30 0.8649829836227607\n",
      "31 0.8659965494869908\n",
      "32 0.8668001875278719\n",
      "33 0.865647837529961\n",
      "34 0.8655024113268682\n",
      "35 0.864859131545203\n",
      "36 0.8654177745422638\n",
      "37 0.8664689782129784\n",
      "38 0.8673140596762023\n",
      "39 0.867656799154229\n",
      "40 0.8687669491528436\n",
      "41 0.8687896786055962\n",
      "42 0.8680022721737348\n",
      "43 0.8692976437487946\n",
      "44 0.8704153859681566\n",
      "45 0.871137180257465\n",
      "46 0.8714596123684298\n",
      "47 0.8720879932334566\n",
      "48 0.8725480472364465\n",
      "49 0.8722537500908445\n",
      "50 0.8718155212802041\n",
      "37\n",
      "0.8768545703007286\n",
      "1\n",
      "0.9022844334017679\n",
      "1\n",
      "0.9022844334017679\n",
      "5\n",
      "0.903978488497109\n",
      "11\n",
      "0.903391306234467\n",
      "16\n",
      "0.9042902999323111\n",
      "16\n",
      "0.9021261561776264\n",
      "16\n",
      "0.9021261561776264\n",
      "53\n",
      "0.9021128927174363\n",
      "35\n",
      "0.9033949553241896\n",
      "29\n",
      "0.903219651677476\n",
      "18\n",
      "0.904581809447137\n",
      "10\n",
      "0.903493032633341\n",
      "9\n",
      "0.9028572848372266\n",
      "2\n",
      "0.9025436581568786\n",
      "1 0.8144534104682635\n",
      "2 0.8486781335174317\n",
      "3 0.8582154153167647\n",
      "4 0.863426315389217\n",
      "5 0.8684773008981463\n",
      "6 0.8728829527897781\n",
      "7 0.8765808222689376\n",
      "8 0.8797109205995532\n",
      "9 0.8816508542899706\n",
      "10 0.8841819564955788\n",
      "11 0.8856968263343663\n",
      "12 0.8871649418118559\n",
      "13 0.8881056302443396\n",
      "14 0.8902048413204972\n",
      "15 0.8924972640030503\n",
      "16 0.8929619243630216\n",
      "17 0.8946079231535258\n",
      "18 0.8947142788839245\n",
      "19 0.8950316160264228\n",
      "20 0.8940092185948575\n",
      "21 0.8940615781731933\n",
      "22 0.8938960585197969\n",
      "23 0.8952380274974935\n",
      "24 0.8958441219391472\n",
      "25 0.8967427618757734\n",
      "26 0.8964940372371836\n",
      "27 0.8959958307791763\n",
      "28 0.8970580748065083\n",
      "29 0.8977239290426628\n",
      "30 0.8983346117632764\n",
      "31 0.8997868736798134\n",
      "32 0.8991070764026834\n",
      "33 0.8989975080375222\n",
      "34 0.8992542457096514\n",
      "35 0.8995446795193268\n",
      "36 0.9003483214463334\n",
      "37 0.9007832895810676\n",
      "38 0.9008342688750794\n",
      "39 0.9011648824706451\n",
      "40 0.901090037889485\n",
      "41 0.9012684275728549\n",
      "42 0.9014452530276876\n",
      "43 0.901926436548137\n",
      "44 0.9022995609498234\n",
      "45 0.9025536696500208\n",
      "46 0.9020557141821256\n",
      "47 0.9020789125820385\n",
      "48 0.9022616014315267\n",
      "49 0.9027382655942661\n",
      "50 0.9023433631702884\n",
      "36\n",
      "0.9042174576760456\n",
      "1\n",
      "0.8624938600737004\n",
      "1\n",
      "0.8624938600737004\n",
      "5\n",
      "0.8614756764001706\n",
      "15\n",
      "0.8628196819376828\n",
      "18\n",
      "0.8620346410173556\n",
      "17\n",
      "0.86343410337631\n",
      "17\n",
      "0.86343410337631\n",
      "41\n",
      "0.8597072716186143\n",
      "24\n",
      "0.8605944191412751\n",
      "17\n",
      "0.8616211469097311\n",
      "7\n",
      "0.8636781406607208\n",
      "5\n",
      "0.8632336579559321\n",
      "3\n",
      "0.8630405770063336\n",
      "1\n",
      "0.8624938600737004\n",
      "1 0.739465464050965\n",
      "2 0.7671166514519133\n",
      "3 0.7877657630989855\n",
      "4 0.7999750675746582\n",
      "5 0.807294442035795\n",
      "6 0.8081197285199966\n",
      "7 0.8135129205651257\n",
      "8 0.8208246504633734\n",
      "9 0.8280837301681814\n",
      "10 0.8318437422722621\n",
      "11 0.834178790446927\n",
      "12 0.8364293198489832\n",
      "13 0.8379721758523831\n",
      "14 0.8424676362706676\n",
      "15 0.8442436381038907\n",
      "16 0.8442790860245234\n",
      "17 0.8437616942098716\n",
      "18 0.8452650719119543\n",
      "19 0.8471438130265769\n",
      "20 0.8483665741189188\n",
      "21 0.8506133014098667\n",
      "22 0.8504123694731163\n",
      "23 0.8504113548296712\n",
      "24 0.8502246038774282\n",
      "25 0.8500575214328274\n",
      "26 0.8509660663978891\n",
      "27 0.8515522903489687\n",
      "28 0.8524141657063198\n",
      "29 0.8550244128405943\n",
      "30 0.8555246576091498\n",
      "31 0.8562026561208734\n",
      "32 0.8560004891376531\n",
      "33 0.856344561573463\n",
      "34 0.8569475641242797\n",
      "35 0.8570790399272676\n",
      "36 0.8566442493307747\n",
      "37 0.8565473334560666\n",
      "38 0.8567117537980392\n",
      "39 0.8566828373699221\n",
      "40 0.8572815405585522\n",
      "41 0.8576823544730043\n",
      "42 0.8577124513023063\n",
      "43 0.857169169856475\n",
      "44 0.8573011305181308\n",
      "45 0.8573468897002923\n",
      "46 0.8579496437875559\n",
      "47 0.8580487149978735\n",
      "48 0.858310445641992\n",
      "49 0.8585930017967488\n",
      "50 0.8588425134194261\n",
      "38\n",
      "0.8614033419163053\n",
      "1\n",
      "0.8558983527706091\n",
      "1\n",
      "0.8558983527706092\n",
      "8\n",
      "0.8569634556867887\n",
      "13\n",
      "0.8587421041902126\n",
      "19\n",
      "0.8568298818242935\n",
      "15\n",
      "0.85586672396107\n",
      "20\n",
      "0.859715282439359\n",
      "31\n",
      "0.8557312279567921\n",
      "18\n",
      "0.8552416594583914\n",
      "11\n",
      "0.8567304689078702\n",
      "4\n",
      "0.8567888344698533\n",
      "2\n",
      "0.8562754111449262\n",
      "2\n",
      "0.8559366419130128\n",
      "1\n",
      "0.8558983527706091\n",
      "1 0.6728590619910473\n",
      "2 0.7237487950462075\n",
      "3 0.7560300607812684\n",
      "4 0.7790263825133634\n",
      "5 0.7908304787574958\n",
      "6 0.7918926922905438\n",
      "7 0.7968761908024561\n",
      "8 0.8107364080813629\n",
      "9 0.8159555205252819\n",
      "10 0.818837217263821\n",
      "11 0.8216450332471104\n",
      "12 0.8230149561185668\n",
      "13 0.8254132156079272\n",
      "14 0.8291471348765351\n",
      "15 0.8316866530211623\n",
      "16 0.8325975363559349\n",
      "17 0.8350133904909798\n",
      "18 0.8375787120939346\n",
      "19 0.840147372233325\n",
      "20 0.8418745189468592\n",
      "21 0.8440060435922725\n",
      "22 0.8447353566593602\n",
      "23 0.84556645168674\n",
      "24 0.8458594212140191\n",
      "25 0.8454845240475759\n",
      "26 0.8453060102990204\n",
      "27 0.8466613953044988\n",
      "28 0.8472900336220284\n",
      "29 0.8472776843542137\n",
      "30 0.8473578058503352\n",
      "31 0.848839383457076\n",
      "32 0.848989302836468\n",
      "33 0.8495576764079367\n",
      "34 0.8498944554633296\n",
      "35 0.8490307759746659\n",
      "36 0.8504723532437465\n",
      "37 0.8509162435068042\n",
      "38 0.8518516558898459\n",
      "39 0.8518637919905397\n",
      "40 0.8513868739426861\n",
      "41 0.851982247038223\n",
      "42 0.8526291382909091\n",
      "43 0.8533304539598495\n",
      "44 0.8534933027041678\n",
      "45 0.8537616644419432\n",
      "46 0.8540217954063439\n",
      "47 0.8543547166438178\n",
      "48 0.8546586127179358\n",
      "49 0.8547403283662669\n",
      "50 0.8552130705604647\n",
      "38\n",
      "0.8577400411311383\n"
     ]
    }
   ],
   "source": [
    "lin_score,lin_list_of_length,lin_wrapper_score,lin_LinCFA_score,lin_list_of_length_lin,lin_GenLinCFA_score,lin_list_of_length_gen = linear_experiment_resampling()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "dbb35529",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x13a4c8df0>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = compute_means_fromList(lin_list_of_length, 7)\n",
    "rm = compute_means_fromList(lin_score, 7)\n",
    "us = compute_upper_fromList(lin_score, 7, 10)\n",
    "ls = compute_lower_fromList(lin_score, 7, 10)\n",
    "cm_sorted = [x for x,_ in sorted(zip(cm,ls))]\n",
    "rm_sorted = [x for _,x in sorted(zip(cm,rm))]\n",
    "ls_sorted = [x for _,x in sorted(zip(cm,ls))]\n",
    "us_sorted = [x for _,x in sorted(zip(cm,us))]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(cm_sorted,rm_sorted,'go')\n",
    "ax.fill_between(cm_sorted, (ls_sorted), (us_sorted), color='g', alpha=.1)\n",
    "ax.plot(compute_means_fromList(lin_wrapper_score,50),'bo')\n",
    "ax.fill_between(range(50), compute_lower_fromList(lin_wrapper_score,50,10), compute_upper_fromList(lin_wrapper_score,50,10), color='b', alpha=.1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "d9d6cced",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x15fea1e80>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = compute_means_fromList(lin_list_of_length_gen, 7)\n",
    "rm = compute_means_fromList(lin_GenLinCFA_score, 7)\n",
    "us = compute_upper_fromList(lin_GenLinCFA_score, 7, 10)\n",
    "ls = compute_lower_fromList(lin_GenLinCFA_score, 7, 10)\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(cm,rm,'ro')\n",
    "ax.fill_between(cm, (ls), (us), color='r', alpha=.1)\n",
    "ax.plot(compute_means_fromList(lin_wrapper_score,50),'bo')\n",
    "ax.fill_between(range(50), compute_lower_fromList(lin_wrapper_score,50,10), compute_upper_fromList(lin_wrapper_score,50,10), color='b', alpha=.1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "c31c91ef",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinCFA result: 0.8891591909939184 with [38, 38, 41, 33, 44, 40, 37, 36, 38, 38] reduced features\n"
     ]
    }
   ],
   "source": [
    "print(f'LinCFA result: {np.mean(lin_LinCFA_score)} with {lin_list_of_length_lin} reduced features')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e363d114",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Repeat using the same distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "9cca2fc0",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "0.8880956759252863\n",
      "1\n",
      "0.8880956759252863\n",
      "2\n",
      "0.8883636842956287\n",
      "11\n",
      "0.8921795104881671\n",
      "13\n",
      "0.8929325614443642\n",
      "14\n",
      "0.8926536236100118\n",
      "15\n",
      "0.8932155940780715\n",
      "57\n",
      "0.8945298150352698\n",
      "41\n",
      "0.892796287438592\n",
      "28\n",
      "0.8932199045259764\n",
      "19\n",
      "0.891463226302124\n",
      "14\n",
      "0.8913310458676756\n",
      "11\n",
      "0.8906953720331078\n",
      "2\n",
      "0.888420954434259\n",
      "1 0.8183609180275146\n",
      "2 0.8406006672475028\n",
      "3 0.8587618988617504\n",
      "4 0.8651077572239831\n",
      "5 0.8680982083869794\n",
      "6 0.8710996548941699\n",
      "7 0.8732285845884031\n",
      "8 0.8757022596306532\n",
      "9 0.8778071503922912\n",
      "10 0.8787127785908995\n",
      "11 0.8783013112919728\n",
      "12 0.8812084705343686\n",
      "13 0.883078859254669\n",
      "14 0.8844506202768785\n",
      "15 0.884610966781385\n",
      "16 0.8853251015698956\n",
      "17 0.8854361968615337\n",
      "18 0.8863670818589968\n",
      "19 0.885732405909858\n",
      "20 0.8862272553701469\n",
      "21 0.8877269648899154\n",
      "22 0.8890303329748063\n",
      "23 0.889667895478342\n",
      "24 0.8896055728722891\n",
      "25 0.8892971702804069\n",
      "26 0.8891251164029019\n",
      "27 0.8884810000980452\n",
      "28 0.8887253273410548\n",
      "29 0.8889184021212417\n",
      "30 0.8895168343543748\n",
      "31 0.8894942273757707\n",
      "32 0.8903461606330211\n",
      "33 0.8904879759521066\n",
      "34 0.8907526677599336\n",
      "35 0.8909320437124864\n",
      "36 0.8920623418279802\n",
      "37 0.8926142759385313\n",
      "38 0.8924922322644333\n",
      "39 0.8927075845017192\n",
      "40 0.8934467315737608\n",
      "41 0.8939166720285093\n",
      "42 0.8938036722694432\n",
      "43 0.8942696772588005\n",
      "44 0.894532895168243\n",
      "45 0.8950276802564534\n",
      "46 0.895211737598066\n",
      "47 0.8951181502879555\n",
      "48 0.8948995873255838\n",
      "49 0.8946834899146395\n",
      "50 0.8946802772985314\n",
      "38\n",
      "0.8925716349990058\n",
      "1\n",
      "0.88622116203857\n",
      "1\n",
      "0.88622116203857\n",
      "5\n",
      "0.8872753789882489\n",
      "16\n",
      "0.8879537664278376\n",
      "19\n",
      "0.887816169801194\n",
      "19\n",
      "0.887816169801194\n",
      "19\n",
      "0.887816169801194\n",
      "57\n",
      "0.8892843262658044\n",
      "41\n",
      "0.8894887141714973\n",
      "28\n",
      "0.8886484271394351\n",
      "18\n",
      "0.8874190900009057\n",
      "13\n",
      "0.8870781113969616\n",
      "9\n",
      "0.8872677758229601\n",
      "2\n",
      "0.8864609746358899\n",
      "1 0.8168484751109538\n",
      "2 0.8460492285444438\n",
      "3 0.8566094261423306\n",
      "4 0.86292630363043\n",
      "5 0.86416987046554\n",
      "6 0.86639928067001\n",
      "7 0.8693898148822863\n",
      "8 0.8714542601597182\n",
      "9 0.8719414454392709\n",
      "10 0.8726392413040052\n",
      "11 0.8721772731139916\n",
      "12 0.872218315888955\n",
      "13 0.8733454915753154\n",
      "14 0.8736544556630939\n",
      "15 0.8739325882713073\n",
      "16 0.8743267885811157\n",
      "17 0.8763950574601136\n",
      "18 0.8780830109487437\n",
      "19 0.8788545999586639\n",
      "20 0.8787016334213414\n",
      "21 0.8794511515226577\n",
      "22 0.8811287666320632\n",
      "23 0.8813428484589216\n",
      "24 0.8814288683760092\n",
      "25 0.8831340721184829\n",
      "26 0.8837508578183851\n",
      "27 0.8839037001593298\n",
      "28 0.8840230532980267\n",
      "29 0.8833095394058522\n",
      "30 0.8836214370551936\n",
      "31 0.8833229427526826\n",
      "32 0.8834024047340232\n",
      "33 0.8839380838730097\n",
      "34 0.8842306695789802\n",
      "35 0.8841935157318487\n",
      "36 0.8838388961390665\n",
      "37 0.8839853754908854\n",
      "38 0.8840343618030466\n",
      "39 0.8853480184480884\n",
      "40 0.8857018360108515\n",
      "41 0.8859067690189366\n",
      "42 0.8860682470056214\n",
      "43 0.8861389891139313\n",
      "44 0.8865346848373956\n",
      "45 0.8867740611673793\n",
      "46 0.886958654126587\n",
      "47 0.8868368761432197\n",
      "48 0.8867224488746419\n",
      "49 0.88653988341868\n",
      "50 0.8866151440185762\n",
      "38\n",
      "0.8872551836597906\n",
      "1\n",
      "0.8916599378413258\n",
      "1\n",
      "0.8916599378413258\n",
      "5\n",
      "0.8913912601773814\n",
      "11\n",
      "0.8913305225207072\n",
      "14\n",
      "0.8921380112463684\n",
      "14\n",
      "0.8921094742588993\n",
      "13\n",
      "0.8914632540417595\n",
      "57\n",
      "0.8909979017981448\n",
      "41\n",
      "0.8922154734693031\n",
      "29\n",
      "0.8916276128843407\n",
      "20\n",
      "0.8919616431883663\n",
      "14\n",
      "0.8933841850049982\n",
      "11\n",
      "0.8919977000656991\n",
      "2\n",
      "0.89145653697699\n",
      "1 0.820883565666384\n",
      "2 0.8479791645366691\n",
      "3 0.8619315969322943\n",
      "4 0.8659486903570591\n",
      "5 0.8691292710633878\n",
      "6 0.8697252608501727\n",
      "7 0.8697014441946023\n",
      "8 0.8702635315214173\n",
      "9 0.8719069068759373\n",
      "10 0.8731921911087556\n",
      "11 0.874820271888661\n",
      "12 0.8751965983174606\n",
      "13 0.8751517005188538\n",
      "14 0.8766786356939863\n",
      "15 0.8778666340104162\n",
      "16 0.8780234226503325\n",
      "17 0.8785819277079586\n",
      "18 0.8797020833511205\n",
      "19 0.8806056907793275\n",
      "20 0.8811062372342001\n",
      "21 0.8814946628534119\n",
      "22 0.8833271022687523\n",
      "23 0.8837494698905499\n",
      "24 0.884373815000428\n",
      "25 0.8838738787667595\n",
      "26 0.8845662609253356\n",
      "27 0.8849795224109583\n",
      "28 0.8854819379503884\n",
      "29 0.8863675344991233\n",
      "30 0.8873653358499441\n",
      "31 0.888396792491586\n",
      "32 0.8889103953505104\n",
      "33 0.888977286402727\n",
      "34 0.8893941069475644\n",
      "35 0.8893659523417853\n",
      "36 0.8896483936950762\n",
      "37 0.88993766880973\n",
      "38 0.8897097191351779\n",
      "39 0.8896959049753035\n",
      "40 0.8898409060616252\n",
      "41 0.8904513126046485\n",
      "42 0.8906654421823796\n",
      "43 0.8904392584766063\n",
      "44 0.8907070993821758\n",
      "45 0.8909102772605668\n",
      "46 0.8909618544594557\n",
      "47 0.8911079802520562\n",
      "48 0.8912480294328415\n",
      "49 0.8911869750187342\n",
      "50 0.8907939395312107\n",
      "38\n",
      "0.8913701253725557\n",
      "1\n",
      "0.89047374468368\n",
      "1\n",
      "0.89047374468368\n",
      "3\n",
      "0.8904261175305698\n",
      "8\n",
      "0.8934840359464031\n",
      "11\n",
      "0.8933340068966239\n",
      "11\n",
      "0.8933340068966239\n",
      "11\n",
      "0.8933340068966239\n",
      "58\n",
      "0.8926862862919273\n",
      "40\n",
      "0.8938756208514997\n",
      "30\n",
      "0.8923828786201753\n",
      "19\n",
      "0.8916833085045797\n",
      "14\n",
      "0.8911417587703363\n",
      "10\n",
      "0.8912530726673409\n",
      "2\n",
      "0.8903741540886593\n",
      "1 0.808196456620507\n",
      "2 0.8370789976057975\n",
      "3 0.8518485832302917\n",
      "4 0.8584428868016609\n",
      "5 0.8603036254529738\n",
      "6 0.864091219776806\n",
      "7 0.8675831061170401\n",
      "8 0.8689087311321777\n",
      "9 0.8706161613194554\n",
      "10 0.87268478264655\n",
      "11 0.8742674278472323\n",
      "12 0.8757001094186156\n",
      "13 0.8758050599536894\n",
      "14 0.8764063104115736\n",
      "15 0.875665311859159\n",
      "16 0.8771944029129857\n",
      "17 0.8774493426869243\n",
      "18 0.8778595891540769\n",
      "19 0.879073601930542\n",
      "20 0.8797970321211716\n",
      "21 0.8801402782832486\n",
      "22 0.8819980127814148\n",
      "23 0.883916712042135\n",
      "24 0.8847350525942205\n",
      "25 0.8851624211326428\n",
      "26 0.8852469761725844\n",
      "27 0.8866582925952279\n",
      "28 0.8868945746073914\n",
      "29 0.8871685774443039\n",
      "30 0.8875733437764444\n",
      "31 0.8874321509437626\n",
      "32 0.8873562530465897\n",
      "33 0.8878708142379104\n",
      "34 0.8883340566171547\n",
      "35 0.8884076865077442\n",
      "36 0.8883952043393446\n",
      "37 0.8887815341706512\n",
      "38 0.8892542032934984\n",
      "39 0.8898318596528011\n",
      "40 0.8901490398158004\n",
      "41 0.8904898068910383\n",
      "42 0.8903298853916183\n",
      "43 0.8908215487604909\n",
      "44 0.8912948163464962\n",
      "45 0.8913898902708797\n",
      "46 0.8913704466327862\n",
      "47 0.8919157933895778\n",
      "48 0.8921517349215762\n",
      "49 0.8926693403805871\n",
      "50 0.8927440216887008\n",
      "44\n",
      "0.892948409202056\n",
      "1\n",
      "0.8879149355324669\n",
      "1\n",
      "0.8879149355324669\n",
      "2\n",
      "0.8874061860817088\n",
      "11\n",
      "0.8890492706630035\n",
      "13\n",
      "0.889121598182219\n",
      "21\n",
      "0.8882599805003861\n",
      "21\n",
      "0.8882599805003861\n",
      "60\n",
      "0.8875436484053355\n",
      "40\n",
      "0.8887207872760767\n",
      "29\n",
      "0.8871607101452387\n",
      "17\n",
      "0.8877108419245748\n",
      "13\n",
      "0.8890458121474116\n",
      "11\n",
      "0.887533091313659\n",
      "2\n",
      "0.8878833480573853\n",
      "1 0.8048222513546393\n",
      "2 0.8312695120519201\n",
      "3 0.8497967339590216\n",
      "4 0.8548847708575118\n",
      "5 0.8613645563333141\n",
      "6 0.8686578346203098\n",
      "7 0.8717811526643384\n",
      "8 0.8726089257093089\n",
      "9 0.8747739895029869\n",
      "10 0.8743936385126335\n",
      "11 0.8762224881888417\n",
      "12 0.8777232297388189\n",
      "13 0.8788414058623876\n",
      "14 0.8799054598279638\n",
      "15 0.8799671400174149\n",
      "16 0.8809234751454228\n",
      "17 0.8823518861114763\n",
      "18 0.8825337932939632\n",
      "19 0.8840709028199853\n",
      "20 0.884871186087864\n",
      "21 0.8852511651258969\n",
      "22 0.8854252312261944\n",
      "23 0.8861436040945996\n",
      "24 0.8864074149613534\n",
      "25 0.8867962843695069\n",
      "26 0.8865647490204643\n",
      "27 0.886559424575891\n",
      "28 0.8869349522307467\n",
      "29 0.8867076826219016\n",
      "30 0.8871461931032074\n",
      "31 0.8874412814647369\n",
      "32 0.887801995192678\n",
      "33 0.887968184848858\n",
      "34 0.8881111765626829\n",
      "35 0.8877536839496386\n",
      "36 0.8882329836309021\n",
      "37 0.8880012213077563\n",
      "38 0.8883484598020295\n",
      "39 0.8887539729422638\n",
      "40 0.8894464322832736\n",
      "41 0.889648501998618\n",
      "42 0.8897042093687053\n",
      "43 0.8896452850549631\n",
      "44 0.8894875954068723\n",
      "45 0.8894893832841017\n",
      "46 0.8893078638320021\n",
      "47 0.8893099194136593\n",
      "48 0.8891109418979319\n",
      "49 0.8889756855112186\n",
      "50 0.8893019312199006\n",
      "40\n",
      "0.8889548689602971\n",
      "1\n",
      "0.8987513053865577\n",
      "1\n",
      "0.8987513053865577\n",
      "4\n",
      "0.8996344914532878\n",
      "11\n",
      "0.9005199478692586\n",
      "20\n",
      "0.9011414300021652\n",
      "20\n",
      "0.9011414300021652\n",
      "20\n",
      "0.9011414300021652\n",
      "56\n",
      "0.9014733945617504\n",
      "38\n",
      "0.8989414330900339\n",
      "25\n",
      "0.8991616017459587\n",
      "17\n",
      "0.8995038838535223\n",
      "12\n",
      "0.8987660361293425\n",
      "9\n",
      "0.8997543369246068\n",
      "2\n",
      "0.8986181961040118\n",
      "1 0.8108451434515123\n",
      "2 0.8325312911464703\n",
      "3 0.8600030731659973\n",
      "4 0.8668148865409976\n",
      "5 0.8709197988176379\n",
      "6 0.8739253339237002\n",
      "7 0.8752629321492484\n",
      "8 0.8762278895310974\n",
      "9 0.8799250375830463\n",
      "10 0.8810198194040708\n",
      "11 0.8824147706926369\n",
      "12 0.8838516682196219\n",
      "13 0.8847538083262725\n",
      "14 0.8861283657250674\n",
      "15 0.8880703327657269\n",
      "16 0.8893475005806125\n",
      "17 0.8896320998873626\n",
      "18 0.8891248141501559\n",
      "19 0.8887920775997282\n",
      "20 0.8884839496569544\n",
      "21 0.889394145341525\n",
      "22 0.8897437747173667\n",
      "23 0.8905778364802861\n",
      "24 0.891070724954901\n",
      "25 0.8909326554934075\n",
      "26 0.8915527623737178\n",
      "27 0.8922925661576823\n",
      "28 0.8924144346724534\n",
      "29 0.8928820462430661\n",
      "30 0.8930782642377527\n",
      "31 0.8938092706632078\n",
      "32 0.8943079448742199\n",
      "33 0.8950138676768951\n",
      "34 0.8957144268452145\n",
      "35 0.8958410691379965\n",
      "36 0.8962892554002746\n",
      "37 0.8962529538531117\n",
      "38 0.8965731596647298\n",
      "39 0.8965902391040892\n",
      "40 0.897046477438628\n",
      "41 0.8970424649685601\n",
      "42 0.8963933403599488\n",
      "43 0.8963119233870431\n",
      "44 0.8961501066875603\n",
      "45 0.8960274035451073\n",
      "46 0.8963295990613714\n",
      "47 0.8968193954324578\n",
      "48 0.8974662746756799\n",
      "49 0.8973834248367065\n",
      "50 0.8973605517068062\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "39\n",
      "0.9017470877428849\n",
      "1\n",
      "0.8826529317422573\n",
      "1\n",
      "0.8826529317422573\n",
      "3\n",
      "0.8825081561003176\n",
      "13\n",
      "0.8838196794077815\n",
      "15\n",
      "0.8834992261915244\n",
      "15\n",
      "0.8834992261915244\n",
      "15\n",
      "0.8834992261915244\n",
      "56\n",
      "0.885539871511144\n",
      "42\n",
      "0.8849307877306168\n",
      "28\n",
      "0.8847713697834391\n",
      "18\n",
      "0.8847247472766904\n",
      "13\n",
      "0.884662692913912\n",
      "10\n",
      "0.8842683227293496\n",
      "2\n",
      "0.882788368161381\n",
      "1 0.7946660939535615\n",
      "2 0.8254118700459598\n",
      "3 0.8442886962006421\n",
      "4 0.8474250106827179\n",
      "5 0.8516559572571335\n",
      "6 0.8557109611094559\n",
      "7 0.8615730235864182\n",
      "8 0.8620553429475759\n",
      "9 0.8631658019490236\n",
      "10 0.8655571301599612\n",
      "11 0.8673859118046101\n",
      "12 0.8686712168291028\n",
      "13 0.8707313232130925\n",
      "14 0.8712862408209432\n",
      "15 0.871599400280154\n",
      "16 0.8735046773888206\n",
      "17 0.8750457760726187\n",
      "18 0.8752359543183474\n",
      "19 0.8738213380519582\n",
      "20 0.875215240700723\n",
      "21 0.8765175627333275\n",
      "22 0.8767979808580965\n",
      "23 0.8763920636304696\n",
      "24 0.8766240395668373\n",
      "25 0.8762049627553947\n",
      "26 0.8763924050319308\n",
      "27 0.8769070631433783\n",
      "28 0.8773012341602064\n",
      "29 0.8776109716868699\n",
      "30 0.8774443013987856\n",
      "31 0.8781517787356776\n",
      "32 0.8777899201385986\n",
      "33 0.8786559765109884\n",
      "34 0.879304084058617\n",
      "35 0.8796112759941945\n",
      "36 0.8805321805795308\n",
      "37 0.8800273899745593\n",
      "38 0.8807522650698787\n",
      "39 0.8809406216940314\n",
      "40 0.8808382637425852\n",
      "41 0.8806930417674689\n",
      "42 0.8807784348324128\n",
      "43 0.8811983184833353\n",
      "44 0.881256161850402\n",
      "45 0.88188176823167\n",
      "46 0.8823565422155002\n",
      "47 0.8824109673407076\n",
      "48 0.8822296930083091\n",
      "49 0.8827570629920949\n",
      "50 0.8828707645369506\n",
      "39\n",
      "0.8846454852494771\n",
      "1\n",
      "0.8860529986946801\n",
      "1\n",
      "0.8860529986946801\n",
      "4\n",
      "0.8880172777872855\n",
      "13\n",
      "0.8865686292087527\n",
      "16\n",
      "0.8870637575668583\n",
      "14\n",
      "0.8877496446096089\n",
      "14\n",
      "0.8877496446096089\n",
      "59\n",
      "0.888185515381193\n",
      "41\n",
      "0.8888746521765305\n",
      "30\n",
      "0.8863954903146953\n",
      "21\n",
      "0.8868249735184386\n",
      "15\n",
      "0.8877424626874366\n",
      "12\n",
      "0.8872238760054749\n",
      "2\n",
      "0.8861832846002274\n",
      "1 0.8057561578301482\n",
      "2 0.8356126206686065\n",
      "3 0.855159683215001\n",
      "4 0.8604072816946028\n",
      "5 0.863428970896907\n",
      "6 0.8682518998879104\n",
      "7 0.871678029600482\n",
      "8 0.8759682124021663\n",
      "9 0.8774010505602174\n",
      "10 0.8780685972346406\n",
      "11 0.8775524247253339\n",
      "12 0.8793144142167226\n",
      "13 0.8788696162709184\n",
      "14 0.8802143042598156\n",
      "15 0.8793109402753063\n",
      "16 0.8792574584367039\n",
      "17 0.880895477084414\n",
      "18 0.8820073582918119\n",
      "19 0.8823259927148799\n",
      "20 0.8827156393433033\n",
      "21 0.8849752499037817\n",
      "22 0.8841810971569773\n",
      "23 0.8842838888133988\n",
      "24 0.8849687431213396\n",
      "25 0.8849786787964463\n",
      "26 0.884722219875512\n",
      "27 0.8849491406951691\n",
      "28 0.8857361928636496\n",
      "29 0.8851179485190408\n",
      "30 0.8862637137894729\n",
      "31 0.8862198402843096\n",
      "32 0.8863672224274622\n",
      "33 0.8864700004720515\n",
      "34 0.8866151969019287\n",
      "35 0.8868442588074893\n",
      "36 0.887045469647762\n",
      "37 0.8867843557215367\n",
      "38 0.886647382632239\n",
      "39 0.8868045768226672\n",
      "40 0.8869651029152958\n",
      "41 0.8869287241443834\n",
      "42 0.8869254739006869\n",
      "43 0.8870201458438907\n",
      "44 0.886989680464706\n",
      "45 0.8870956139489985\n",
      "46 0.8873148821336897\n",
      "47 0.8873332720320483\n",
      "48 0.8870306440400801\n",
      "49 0.8867854390926486\n",
      "50 0.8869560691401517\n",
      "36\n",
      "0.8899498784146602\n",
      "1\n",
      "0.8859388467308895\n",
      "1\n",
      "0.8859388467308895\n",
      "3\n",
      "0.8869181033466343\n",
      "13\n",
      "0.8897065137861949\n",
      "18\n",
      "0.8907436428251682\n",
      "20\n",
      "0.8904411256695808\n",
      "20\n",
      "0.8904411256695808\n",
      "57\n",
      "0.8925203043815422\n",
      "38\n",
      "0.8906177759072379\n",
      "30\n",
      "0.8905934141950239\n",
      "22\n",
      "0.8887837354860061\n",
      "14\n",
      "0.8898334078854541\n",
      "10\n",
      "0.8898944039388578\n",
      "2\n",
      "0.8866314391043546\n",
      "1 0.818104910046602\n",
      "2 0.8369451612685495\n",
      "3 0.8588061911227158\n",
      "4 0.8655484722586697\n",
      "5 0.8679665710351152\n",
      "6 0.8707958908137743\n",
      "7 0.8735506751792237\n",
      "8 0.8759085094780086\n",
      "9 0.8773361264461846\n",
      "10 0.8769868321612881\n",
      "11 0.8775949972119447\n",
      "12 0.8787995009686348\n",
      "13 0.8793007382465489\n",
      "14 0.8813531291446088\n",
      "15 0.8827383521583302\n",
      "16 0.8829142678479717\n",
      "17 0.8837974280525336\n",
      "18 0.8852654184253814\n",
      "19 0.8859957380598422\n",
      "20 0.8866448167132824\n",
      "21 0.8867187315591021\n",
      "22 0.8861704222434222\n",
      "23 0.8866983399718518\n",
      "24 0.8871152750111222\n",
      "25 0.8871894460534123\n",
      "26 0.8874766155914732\n",
      "27 0.8874742978214455\n",
      "28 0.8881613204854344\n",
      "29 0.887708379766793\n",
      "30 0.8878374732034514\n",
      "31 0.8881360936633873\n",
      "32 0.8882175356144575\n",
      "33 0.8883021316963752\n",
      "34 0.8882335594048156\n",
      "35 0.8879560448164243\n",
      "36 0.887945634722679\n",
      "37 0.8876315159646319\n",
      "38 0.8879796041886915\n",
      "39 0.8880023158427846\n",
      "40 0.8883236546165024\n",
      "41 0.8885950464463359\n",
      "42 0.8893163293691263\n",
      "43 0.8893066070813157\n",
      "44 0.889352435187687\n",
      "45 0.8895241773248419\n",
      "46 0.8895347062221429\n",
      "47 0.8891149218811706\n",
      "48 0.8893097135269773\n",
      "49 0.8889503583082905\n",
      "50 0.8887017006505306\n",
      "41\n",
      "0.8905688512563519\n",
      "1\n",
      "0.8918266503573463\n",
      "1\n",
      "0.8918266503573463\n",
      "2\n",
      "0.8919994737814386\n",
      "12\n",
      "0.8922662830392423\n",
      "13\n",
      "0.8933378300916447\n",
      "14\n",
      "0.8912338081265949\n",
      "14\n",
      "0.8912338081265949\n",
      "61\n",
      "0.8928367507426123\n",
      "42\n",
      "0.8935502524859995\n",
      "31\n",
      "0.8928425798562314\n",
      "20\n",
      "0.8920310450120124\n",
      "15\n",
      "0.8918516391683631\n",
      "12\n",
      "0.8927079383830537\n",
      "3\n",
      "0.8920373922855815\n",
      "1 0.8171131700003265\n",
      "2 0.8390552650755362\n",
      "3 0.856389677449956\n",
      "4 0.8616555817892572\n",
      "5 0.8653137751899665\n",
      "6 0.8676133305433652\n",
      "7 0.8729130742425715\n",
      "8 0.874218056433273\n",
      "9 0.8770549981842057\n",
      "10 0.8799582690900267\n",
      "11 0.8813935044633998\n",
      "12 0.8826847580062624\n",
      "13 0.8832110750550913\n",
      "14 0.882563820332706\n",
      "15 0.8842128342025237\n",
      "16 0.8845559111867821\n",
      "17 0.8853061726533308\n",
      "18 0.886071256155359\n",
      "19 0.8865875139890199\n",
      "20 0.8867841536398239\n",
      "21 0.8879780552332637\n",
      "22 0.8886967789095815\n",
      "23 0.889892324694473\n",
      "24 0.8896874091638437\n",
      "25 0.8905081237019967\n",
      "26 0.8904794023582818\n",
      "27 0.8908453076643611\n",
      "28 0.8905211140666849\n",
      "29 0.8915422740146942\n",
      "30 0.8913412652407071\n",
      "31 0.8919981635629323\n",
      "32 0.8923115607033872\n",
      "33 0.8914460169160612\n",
      "34 0.8914407249913088\n",
      "35 0.8914526162984764\n",
      "36 0.8909142241187981\n",
      "37 0.8911681660730955\n",
      "38 0.8915720923705742\n",
      "39 0.891838620988104\n",
      "40 0.8917506183836464\n",
      "41 0.8913145102824442\n",
      "42 0.8912091033703265\n",
      "43 0.8914555252923946\n",
      "44 0.8919336052770908\n",
      "45 0.8918733429335036\n",
      "46 0.8921151686211808\n",
      "47 0.8919161623151066\n",
      "48 0.8924956854068363\n",
      "49 0.8926509347177983\n",
      "50 0.8929115472697068\n",
      "39\n",
      "0.8928617664582797\n"
     ]
    }
   ],
   "source": [
    "lin2_score,lin2_list_of_length,lin2_wrapper_score,lin2_LinCFA_score,lin2_list_of_length_lin,lin2_GenLinCFA_score,lin2_list_of_length_gen = linear_experiment_noResampling(10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "bd73ad2e",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x15fe25610>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = compute_means_fromList(lin2_list_of_length, 7)\n",
    "rm = compute_means_fromList(lin2_score, 7)\n",
    "us = compute_upper_fromList(lin2_score, 7, 10)\n",
    "ls = compute_lower_fromList(lin2_score, 7, 10)\n",
    "cm_sorted = [x for x,_ in sorted(zip(cm,ls))]\n",
    "rm_sorted = [x for _,x in sorted(zip(cm,rm))]\n",
    "ls_sorted = [x for _,x in sorted(zip(cm,ls))]\n",
    "us_sorted = [x for _,x in sorted(zip(cm,us))]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(cm_sorted,rm_sorted,'go')\n",
    "ax.fill_between(cm_sorted, (ls_sorted), (us_sorted), color='g', alpha=.1)\n",
    "ax.plot(compute_means_fromList(lin2_wrapper_score,50),'bo')\n",
    "ax.fill_between(range(50), compute_lower_fromList(lin2_wrapper_score,50,10), compute_upper_fromList(lin2_wrapper_score,50,10), color='b', alpha=.1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "18d500bc",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x15fee36d0>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = compute_means_fromList(lin2_list_of_length_gen, 7)\n",
    "rm = compute_means_fromList(lin2_GenLinCFA_score, 7)\n",
    "us = compute_upper_fromList(lin2_GenLinCFA_score, 7, 10)\n",
    "ls = compute_lower_fromList(lin2_GenLinCFA_score, 7, 10)\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(cm,rm,'ro')\n",
    "ax.fill_between(cm, (ls), (us), color='r', alpha=.1)\n",
    "ax.plot(compute_means_fromList(lin2_wrapper_score,50),'bo')\n",
    "ax.fill_between(range(50), compute_lower_fromList(lin2_wrapper_score,50,10), compute_upper_fromList(lin2_wrapper_score,50,10), color='b', alpha=.1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "d5f67704",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinCFA result: 0.8912873291315359 with [38, 38, 38, 44, 40, 39, 39, 36, 41, 39] reduced features\n"
     ]
    }
   ],
   "source": [
    "print(f'LinCFA result: {np.mean(lin2_LinCFA_score)} with {lin2_list_of_length_lin} reduced features')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7656990a",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Structured experiment, linear: run script, save result, plot here from script"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "097288b3",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "761bc6fe",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import math\n",
    "def plot_linear_result(res,title,wrapper_stop,savename,res2):\n",
    "    # in 0 ho NonLinCFA, in 1 wrapper\n",
    "\n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(7):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[0][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "    print(f'NonLinCFA:\\n number of red features:{count_means[1:-1]} +- {count_plusminus[1:-1]},\\n R2 score: {R2_means[1:-1]} +- {R2_means_plusminus[1:-1]}\\n')\n",
    "\n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"R2 score\")\n",
    "    ax.plot([round(x) for x in count_means[1:-1]],R2_means[1:-1],'go',label='NonLinCFA score')\n",
    "    ax.fill_between([round(x) for x in count_means[1:-1]], ([i-j for i,j in zip(R2_means[1:-1], R2_means_plusminus[1:-1])]), ([i+j for i,j in zip(R2_means[1:-1], R2_means_plusminus[1:-1])]), color='g', alpha=.1)\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[1][i::50]))\n",
    "        wrapper_std.append(np.std(results[1][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    \n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "    ax.plot(range(1,wrapper_stop),wrapper_means[:(wrapper_stop-1)],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,wrapper_stop), ([i-j for i,j in zip(wrapper_means[:(wrapper_stop-1)], wrapper_plusminus[:(wrapper_stop-1)])]), ([i+j for i,j in zip(wrapper_means[:(wrapper_stop-1)], wrapper_plusminus[:(wrapper_stop-1)])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(res2[0][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "\n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "    \n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "    \n",
    "    print(f'LinCFA:\\n mean {np.mean(results[2])}, confidence interval [{np.mean(results[2])-1.96*np.std(results[2])/np.sqrt(10)},{np.mean(results[2])+1.96*np.std(results[2])/np.sqrt(10)}],')\n",
    "    print(f'with number of reduced features: {np.mean(results[3])}, confidence interval [{np.mean(results[3])-1.96*np.std(results[3])/np.sqrt(10)},{np.mean(results[3])+1.96*np.std(results[3])/np.sqrt(10)}]')\n",
    "    \n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    \n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "    plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "534dcc80",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Low noise, 100 variables, linear dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9ea86f2a",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('results/linear_lowNoise_100var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "\n",
    "with open('results/linear_lowNoise_100var_onlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5e8d29a0",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.76, 0.77, 0.78, 0.79, 0.8]\n"
     ]
    }
   ],
   "source": [
    "print(f'NonLinCFA epsilon: {[results[0][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[0][0][i][0] for i in range(5)]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "640a5c0d",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 8.0, 11.4, 14.4, 14.7] +- [0.0, 0.8765386471799175, 1.4456245709035245, 1.276259534734217, 1.2098138699816594],\n",
      " R2 score: [0.8655215785894974, 0.8663963367073707, 0.8660527311262882, 0.8658764264928565, 0.8663889970491097] +- [0.005065200704867381, 0.004751893342010827, 0.004874233714607369, 0.005024884367078448, 0.00433416556615672]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.7618800883682314, 0.7925098529490984, 0.8078723723579652, 0.8205446073009851, 0.8280159426628593, 0.8344789502588549, 0.839275234057725, 0.8423316960004839, 0.8450262859322881, 0.846936063320413, 0.8485963098734693, 0.8493849164649946, 0.8505980355515804, 0.8516217530522064, 0.8522112865194309, 0.8533402455408263, 0.8544591147489904, 0.8550426722766467, 0.8559530916633713, 0.8567911573105098, 0.8572701588739269, 0.857522040149471, 0.8580024068871177, 0.8584481586103827, 0.8589109744821428, 0.859463111885964, 0.8600552311120204, 0.8602174018174683, 0.8604303642636342, 0.860654511035789, 0.8607254379668718, 0.8608853082271907, 0.8608753937071084, 0.8611032409875087, 0.8617216000845854, 0.8618645750026486, 0.8620183662920737, 0.8620973655206061, 0.8623693869742215, 0.8626313132191827, 0.8626672591172817, 0.8626819968627528, 0.8627444166038918, 0.8630251910399258, 0.863259207095662, 0.8634395898505172, 0.8636274507520181, 0.8637901317277825, 0.8638944053510411, 0.8639233475938635] +- [0.008220607642768468, 0.006196003395355399, 0.005842316549442443, 0.005825670234183862, 0.005682510413553197, 0.005693409116350873, 0.005207184691005083, 0.004431649347206837, 0.004240248799838386, 0.0039699427435608645, 0.003960584085073433, 0.0035233574613290214, 0.0036763563286088675, 0.00412930513024872, 0.0041657533380626026, 0.004304399937200329, 0.004033454695833844, 0.004101344018934775, 0.004300915251557756, 0.004418047705719388, 0.004504534168597786, 0.004527345335419924, 0.004511213374617063, 0.0044742288454108735, 0.004450987540186716, 0.004363418068444588, 0.004272913859785621, 0.00430530404503947, 0.0043735628791483185, 0.004443296155328581, 0.004646493735973628, 0.004671137262176592, 0.004798278103131063, 0.004884821671896453, 0.004831273733494956, 0.00485601067664049, 0.004908912018345857, 0.004858519633348742, 0.004840375298527235, 0.004882318756041078, 0.004640672975014993, 0.0046953741585380425, 0.004718905676727892, 0.004718980558599017, 0.004722625661683489, 0.00476735046867315, 0.004806289536717754, 0.004768525739948911, 0.004751608892994505, 0.004735763287432148]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[21.6, 16.6, 13.6, 5.4, 2.0] +- [1.471958695072657, 0.6320810074666061, 1.3060131699182822, 1.14956965861143, 0.39199999999999996],\n",
      " R2 score: [0.8656042073177955, 0.8663234949478145, 0.8660285582535272, 0.8659283775007541, 0.8656828514751457] +- [0.004680524446198159, 0.004648113464700666, 0.004639869122686313, 0.004856140820748642, 0.00489056620458126]\n",
      "\n",
      "LinCFA:\n",
      " mean 0.8659457480503938, confidence interval [0.8610119225119705,0.8708795735888172],\n",
      "with number of reduced features: 39.0, confidence interval [37.481790528286695,40.518209471713305]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_linear_result(results,'Low variance, 100 features, linear model',23,'LowVar_100Feat_lin.png',results2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "73957198",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.004933825538423364"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "0.8708795735888172-0.8659457480503938"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "811a3b22",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### High noise, 1000 variables, linear dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5300ae75",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/Linear_HighNoise_1000var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    \n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/Linear_HighNoise_1000var_OnlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "cf0743c7",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.76, 0.77, 0.78, 0.79, 0.8]\n"
     ]
    }
   ],
   "source": [
    "print(f'NonLinCFA epsilon: {[results[0][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[0][0][i][0] for i in range(5)]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d40ef070",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 3.1, 18.0, 21.9, 22.3] +- [0.0, 0.6470969015533917, 1.9203999583420115, 2.1009517843111007, 1.901301028243555],\n",
      " R2 score: [0.7331785859981281, 0.7318399096256514, 0.7273622438417259, 0.7264645636402756, 0.7267467306766292] +- [0.006919557090134497, 0.007097774274391802, 0.007900826451524923, 0.0072330984949896685, 0.00756874782221514]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.6459180200226504, 0.6645209752516574, 0.6786353707203743, 0.6847866260321733, 0.6875357649150693, 0.6901587001803189, 0.6917246227991605, 0.6933028523696785, 0.6937751671759953, 0.6933643294721039, 0.6946437425962897, 0.6950026384006219, 0.6956886629049504, 0.6967577013154108, 0.6978174877271465, 0.6970549203477496, 0.6975132397612498, 0.6974245666079647, 0.6977761794010575, 0.6977246056912791, 0.6985350772862586, 0.6981120241123071, 0.6980480380510614, 0.6983439293965918, 0.69823144204184, 0.6986214652365897, 0.698558348147529, 0.6987993427553463, 0.6994468222397401, 0.6991977427113227, 0.6989442581872305, 0.6987710605621522, 0.6989801003904375, 0.6990477545792395, 0.6983126879628078, 0.6982639091048592, 0.6980110439225408, 0.698154095807801, 0.6978567696804092, 0.6978784405507614, 0.6971444724390055, 0.6965669109901469, 0.6962696671626102, 0.696196426759169, 0.6962439551088322, 0.6961101938071931, 0.6957493072543623, 0.6953159905984541, 0.6950394800456842, 0.6942702532229175] +- [0.011045863965845198, 0.008532253866982824, 0.0084028735816561, 0.007154601871693961, 0.007312228584368754, 0.007839421182655733, 0.007080718075079096, 0.0067901291938330735, 0.006884627460973554, 0.006755078629134668, 0.006770281987210967, 0.006492806974306642, 0.0065732019862272395, 0.006335291164240179, 0.005925426588414607, 0.0059920561905469705, 0.006146081721615774, 0.006226488500132371, 0.006647318890842079, 0.007215443915558918, 0.006824108727759053, 0.006840128408821022, 0.007077262292964883, 0.007489615921653974, 0.007228136038252376, 0.00694268519259202, 0.007027052632614871, 0.006908457999501887, 0.007000405708432023, 0.007082938697196098, 0.0069766808398437526, 0.007070613394795596, 0.00698192865740588, 0.0071168976103670005, 0.006774783771749133, 0.006936402275581737, 0.006831149412751283, 0.006536347264111841, 0.006890139697148878, 0.007022715632744235, 0.007019582959759711, 0.006713720786595209, 0.006828884973847238, 0.007163097250484455, 0.007340869958083846, 0.007087842480619774, 0.006977547966052804, 0.007055925254413887, 0.007166960526053795, 0.007260909071308444]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[7.3, 3.4, 1.0, 1.0, 1.0] +- [1.0753239511886636, 0.6320810074666061, 0.0, 0.0, 0.0],\n",
      " R2 score: [0.7325572570168253, 0.732515254337499, 0.7331785859981281, 0.7331785859981281, 0.7331785859981281] +- [0.006910954363303619, 0.007224711100725818, 0.006919557090134497, 0.006919557090134497, 0.006919557090134497]\n",
      "\n",
      "LinCFA:\n",
      " mean 0.7069992174682609, confidence interval [0.6996361393195503,0.7143622956169715],\n",
      "with number of reduced features: 193.2, confidence interval [190.83496942937305,195.56503057062693]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAHHCAYAAABXx+fLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACKDUlEQVR4nO3dd3xTZcMG4DtJ23QvSieFsveQVRFZghZRFEGWKENffZXKEERBX6ZKQcSvCshQBFRURAEVkSFLRPYSEMrelDJbOtMkz/fH4YSstmmbNE1zX5pfk3NOTp4kDbn7TIUQQoCIiIjIjSidXQAiIiKissYARERERG6HAYiIiIjcDgMQERERuR0GICIiInI7DEBERETkdhiAiIiIyO0wABEREZHbYQAiIiIit8MARGUiLi4OgwcPLvF9n3zySfsWyAHOnTsHhUKBxYsXO7soZMXatWvRrFkzeHt7Q6FQ4M6dO84ukkuw9ns9adIkKBQK5xXKxQ0ePBhxcXElum/Hjh3RsWNHu5bHXTEAUbEtXrwYCoUCe/futbq/Y8eOaNSoURmXihxt/fr1eOmll9CoUSOoVKpC/wHX6/X48MMPUb16dXh7e6NJkyb47rvvrB577NgxdO3aFf7+/ggNDcULL7yA69evl+qc5m7evIk+ffrAx8cHc+bMwddffw0/Pz+b7lsc//77LyZNmoRz587Z/dxEZF8ezi4AuYeUlBQolRU7b1erVg05OTnw9PR0dlEc4ttvv8WyZcvQvHlzREdHF3rsu+++i2nTpuHll19Gq1at8PPPP+O5556DQqFAv379DMddunQJ7du3R1BQEKZOnYrMzEx89NFHOHz4MHbv3g0vL69in9OaPXv24O7du3jvvffQpUuX0r0Qhfj3338xefJkdOzYscR/4buC//3vfxg7dqyzi0FUKhX7G4nKDbVaXWGDgVarhUajgUKhgLe3N1QqlbOL5BBTp05FRkYGtm/fjqZNmxZ43OXLlzFz5kwkJiZiwYIFePnll/Hrr7+iXbt2GDNmDHQ6nck5s7KysGnTJgwfPhzvvPMOfvjhBxw6dMikyaU457QmLS0NABAcHFyq18BZsrKynF0EEx4eHvD29nZ2MQokfyaJCsMARGXCWh+gf/75Bx06dICPjw+qVKmC999/H4sWLYJCobDahPDXX3+hdevW8Pb2Ro0aNfDVV18V+pj5+fkIDQ3FkCFDLPZlZGTA29sbb775JgBAo9FgwoQJaNGiBYKCguDn54d27dph8+bNJveT+0N89NFHSE5ORs2aNaFWq/Hvv/9a7Svxzz//YPDgwahRowa8vb0RGRmJF198ETdv3jQ5r9yn4tSpUxg8eDCCg4MRFBSEIUOGIDs726L833zzDVq3bg1fX1+EhISgffv2WL9+vckxv//+O9q1awc/Pz8EBATgiSeewNGjRwt9zQoTHR1tU4j9+eefkZ+fj6FDhxq2KRQKvPbaa7h06RJ27Nhh2P7TTz/hySefRNWqVQ3bunTpgjp16uCHH34o0TnNdezYEYMGDQIAtGrVCgqFwuR3cdeuXejatSuCgoLg6+uLDh06YPv27SbnOH/+PIYOHYq6devCx8cHlSpVQu/evU1+TxcvXozevXsDADp16gSFQgGFQoEtW7YYyjtp0iSL8pl/NuQm5q1bt2Lo0KEIDw9HlSpVDPtteV9TU1MxZMgQVKlSBWq1GlFRUXj66aft1jRnrQ+QQqHA66+/jlWrVqFRo0ZQq9Vo2LAh1q5da3H/y5cv48UXX0RERIThuC+//NLkGHt8Jgsil3X58uVo0KABfHx80KZNGxw+fBgAMH/+fNSqVQve3t7o2LGj1ddt+fLlaNGiBXx8fBAWFobnn38ely9ftjhOfj28vb3RqFEjrFy50mqZ9Ho9kpOT0bBhQ3h7eyMiIgL//e9/cfv27QKfB5UOm8CoxNLT03Hjxg2L7fn5+UXe9/Lly4YviXHjxsHPzw9ffPEF1Gq11eNPnTqFZ599Fi+99BIGDRqEL7/8EoMHD0aLFi3QsGFDq/fx9PTEM888gxUrVmD+/PkmzSmrVq1CXl6eoekkIyMDX3zxBfr374+XX34Zd+/excKFC5GQkIDdu3ejWbNmJudetGgRcnNz8corr0CtViM0NBR6vd6iDBs2bMCZM2cwZMgQREZG4ujRo1iwYAGOHj2KnTt3WnyJ9OnTB9WrV0dSUhL279+PL774AuHh4Zg+fbrhmMmTJ2PSpEl46KGHMGXKFHh5eWHXrl3YtGkTHnvsMQDA119/jUGDBiEhIQHTp09HdnY25s6di4cffhgHDhxwaPPMgQMH4Ofnh/r165tsb926tWH/ww8/jMuXLyMtLQ0tW7a0OEfr1q2xZs2aYp/TmnfffRd169bFggULMGXKFFSvXh01a9YEAGzatAmPP/44WrRogYkTJ0KpVGLRokV45JFHsG3bNsP59+zZg7///hv9+vVDlSpVcO7cOcydOxcdO3bEv//+C19fX7Rv3x7Dhw/Hp59+infeecdQVvMy22ro0KGoXLkyJkyYYKgBsvV97dWrF44ePYphw4YhLi4OaWlp2LBhAy5cuODQ9/6vv/7CihUrMHToUAQEBODTTz9Fr169cOHCBVSqVAkAcO3aNTz44IOGEFK5cmX8/vvveOmll5CRkYGRI0cCsM9nsjDbtm3DL7/8gsTERABAUlISnnzySbz11lv47LPPMHToUNy+fRsffvghXnzxRWzatMlw38WLF2PIkCFo1aoVkpKScO3aNXzyySfYvn07Dhw4YKhpXL9+PXr16oUGDRogKSkJN2/eNARTc//9738N5x0+fDjOnj2L2bNn48CBA9i+fXuFrUF3KkFUTIsWLRIACr00bNjQ5D7VqlUTgwYNMtweNmyYUCgU4sCBA4ZtN2/eFKGhoQKAOHv2rMl9AYg///zTsC0tLU2o1WoxevToQsu6bt06AUD8+uuvJtu7desmatSoYbit1WpFXl6eyTG3b98WERER4sUXXzRsO3v2rAAgAgMDRVpamsnx8r5FixYZtmVnZ1uU6bvvvrN4PhMnThQATB5LCCGeeeYZUalSJcPtkydPCqVSKZ555hmh0+lMjtXr9UIIIe7evSuCg4PFyy+/bLI/NTVVBAUFWWwviSeeeEJUq1atwH3Gr60sKytLABBjx44VQgixZ88eAUB89dVXFseOGTNGABC5ubnFOmdB5N/ZPXv2GLbp9XpRu3ZtkZCQYHjthJDes+rVq4tHH33UZJu5HTt2WJR/+fLlAoDYvHmzxfEAxMSJEy22m3825LI+/PDDQqvVGrbb+r7evn1bABAzZswo+AUpBmu/1/LvqzEAwsvLS5w6dcqw7dChQwKAmDVrlmHbSy+9JKKiosSNGzdM7t+vXz8RFBRkeK3t8ZksCAChVqtN/p2ZP3++ACAiIyNFRkaGYfu4ceNM/k3SaDQiPDxcNGrUSOTk5BiOW716tQAgJkyYYNjWrFkzERUVJe7cuWPYtn79egHA5POzbds2AUAsXbrUpJxr16612N6hQwfRoUMHm54nFY5NYFRic+bMwYYNGywuTZo0KfK+a9euRZs2bUz+igsNDcWAAQOsHt+gQQO0a9fOcLty5cqoW7cuzpw5U+jjPPLIIwgLC8OyZcsM227fvo0NGzagb9++hm0qlcpQQ6TX63Hr1i1otVq0bNkS+/fvtzhvr169ULly5SKfp4+Pj+F6bm4ubty4gQcffBAArJ731VdfNbndrl073Lx5ExkZGQCkmiu9Xo8JEyZYdCqXa5M2bNiAO3fuoH///rhx44bholKpEB8fb9GEYG85OTlWa/LkPiM5OTkmP2091pbjiuPgwYM4efIknnvuOdy8edPwOmVlZaFz5874888/DbV6xu9jfn4+bt68iVq1aiE4ONjq+2gPL7/8skl/MlvfVx8fH3h5eWHLli1l3nzSpUsXQ+0aADRp0gSBgYGGz6kQAj/99BO6d+8OIYTJ80hISEB6errh9XTUZ1LWuXNnk9qw+Ph4w3kCAgIstsvPYe/evUhLS8PQoUNN+kE98cQTqFevHn777TcAwNWrV3Hw4EEMGjQIQUFBhuMeffRRNGjQwKQsy5cvR1BQEB599FGT16RFixbw9/d3+GfWXbEJjEqsdevWVpsvQkJCrDaNGTt//jzatGljsb1WrVpWjzfuI2L8OEX9A+/h4YFevXrh22+/RV5eHtRqNVasWIH8/HyTAAQAS5YswcyZM3H8+HGTZrzq1atbnNfaNmtu3bqFyZMn4/vvvzd0xJWlp6dbHG/+PENCQgBIoS0wMBCnT5+GUqm0+AfU2MmTJwFI4c+awMBAm8peUj4+PsjLy7PYnpuba9hv/NPWY205rjjk10nuH2RNeno6QkJCkJOTg6SkJCxatAiXL1+GEMLkGEcw/x2z9X1Vq9WYPn06Ro8ejYiICDz44IN48sknMXDgQERGRjqkrLKiPqfXr1/HnTt3sGDBAixYsMDqOYw/J474TBZUVjmkxMbGWt0uP4fz588DAOrWrWtxznr16uGvv/4yOa527doWx9WtW9ckxJ08eRLp6ekIDw+3WlbzfzvIPhiAyCUUNLLK+IuoIP369cP8+fPx+++/o0ePHvjhhx9Qr149k5FM33zzDQYPHowePXpgzJgxCA8Ph0qlQlJSEk6fPm1xTlu/cPv06YO///4bY8aMQbNmzeDv7w+9Xo+uXbta7TNUmucpk8/79ddfW/3C8/Bw7Mc+KioKmzdvhhDCpI/T1atXAcAwhD4qKspku7GrV68iNDTUUOtj6zmLQ36dZsyYYdGfRObv7w8AGDZsGBYtWoSRI0eiTZs2CAoKMgy/t/Y+FkdBI9jMf8eK876OHDkS3bt3x6pVq7Bu3TqMHz8eSUlJ2LRpEx544IFSlbcwRf3+ys/h+eefLzB4yjXIjvpMFlVWe3wGi0uv1yM8PBxLly61ur84NVtkOwYgcopq1arh1KlTFtutbSut9u3bIyoqCsuWLcPDDz+MTZs24d133zU55scff0SNGjWwYsUKky/YiRMnlvhxb9++jY0bN2Ly5MmYMGGCYbv8l3xJ1KxZE3q9Hv/++2+BX9pyE0R4eLhD57wpSLNmzfDFF1/g2LFjJjVVu3btMuwHgJiYGFSuXNnqhJrmnVxtPWdxyK9TYGBgka/Tjz/+iEGDBmHmzJmGbbm5uRazSRc2O3JISIjF8RqNxmoALKy8tr6vNWvWxOjRozF69GicPHkSzZo1w8yZM/HNN9/Y9HiOULlyZQQEBECn09n0mtv7M2kP1apVAyDNbWZeG5eSkmLYL/+09nlPSUkxuV2zZk388ccfaNu2bYlqM6lk2AeInCIhIQE7duzAwYMHDdtu3bpV4F9ApaFUKvHss8/i119/xddffw2tVmvR/CX/1Wf8V96uXbsKHV5dFGvnBIDk5OQSn7NHjx5QKpWYMmWKRc2D/DgJCQkIDAzE1KlTrY7IszbLsj09/fTT8PT0xGeffWZStnnz5iEmJgYPPfSQYXuvXr2wevVqXLx40bBt48aNOHHihGFIeXHPaasWLVqgZs2a+Oijj5CZmWmx3/h1UqlUFu/jrFmzLGpv5NmlrS2zUbNmTfz5558m2xYsWFDkHEYyW9/X7OxsQ9Og8WMHBARYbUYsSyqVCr169cJPP/2EI0eOWOw3f80B+34m7aFly5YIDw/HvHnzTF7P33//HceOHcMTTzwBQKq1bNasGZYsWWLSTLphwwaLIfp9+vSBTqfDe++9Z/F4Wq2Wy7Y4CGuAyCneeustfPPNN3j00UcxbNgwwzD4qlWr4tatW3ZfZ6hv376YNWsWJk6ciMaNG1sMTX7yySexYsUKPPPMM3jiiSdw9uxZzJs3Dw0aNLD65WiLwMBAtG/fHh9++CHy8/MRExOD9evX4+zZsyV+HrVq1cK7776L9957D+3atUPPnj2hVquxZ88eREdHIykpCYGBgZg7dy5eeOEFNG/eHP369UPlypVx4cIF/Pbbb2jbti1mz54NQJpDpXr16hg0aFCRa5j9888/+OWXXwBINXXp6el4//33AQBNmzZF9+7dAQBVqlTByJEjMWPGDOTn56NVq1ZYtWoVtm3bhqVLl5o0MbzzzjtYvnw5OnXqhBEjRiAzMxMzZsxA48aNTeZvKs45baVUKvHFF1/g8ccfR8OGDTFkyBDExMTg8uXL2Lx5MwIDA/Hrr78CkH4/vv76awQFBaFBgwbYsWMH/vjjD8PQblmzZs2gUqkwffp0pKenQ61W45FHHkF4eDj+85//4NVXX0WvXr3w6KOP4tChQ1i3bh3CwsJsKq+t7+uJEyfQuXNn9OnTBw0aNICHhwdWrlyJa9eumcyYLQ+5XrRoUYnX6SuJadOmYfPmzYiPj8fLL7+MBg0a4NatW9i/fz/++OMP3Lp1C4BjPpP24OnpienTp2PIkCHo0KED+vfvbxgGHxcXhzfeeMNwbFJSEp544gk8/PDDePHFF3Hr1i3MmjULDRs2NHkOHTp0wH//+18kJSXh4MGDeOyxx+Dp6YmTJ09i+fLl+OSTT/Dss8864+lWbGU/8IxcnbUhxcY6dOhQ5DB4IYQ4cOCAaNeunVCr1aJKlSoiKSlJfPrppwKASE1NNbnvE088YfVxbB0OqtfrRWxsrAAg3n//fav7p06dKqpVqybUarV44IEHxOrVq8WgQYNMhqvKQ26tDTG2Nlz40qVL4plnnhHBwcEiKChI9O7dW1y5csViSLQ8rPj69esm55Rfa+PhukII8eWXX4oHHnhAqNVqERISIjp06CA2bNhgcszmzZtFQkKCCAoKEt7e3qJmzZpi8ODBYu/evYZjDh8+bNMwcuOyWLuYv7c6nc7wenp5eYmGDRuKb775xup5jxw5Ih577DHh6+srgoODxYABA0ze/5Kcs6CyW/udPXDggOjZs6eoVKmSUKvVolq1aqJPnz5i48aNhmNu374thgwZIsLCwoS/v79ISEgQx48ft/p7/fnnn4saNWoIlUplMiRep9OJt99+W4SFhQlfX1+RkJAgTp06VeAw+II+X0W9rzdu3BCJiYmiXr16ws/PTwQFBYn4+Hjxww8/mJxn1qxZAoBYu3Ztoa9dcYbBJyYmWtzf2mt07do1kZiYKGJjY4Wnp6eIjIwUnTt3FgsWLDAcY4/PZEGslbWg82zevFkAEMuXLzfZvmzZMsNnMDQ0VAwYMEBcunTJ4rF++uknUb9+faFWq0WDBg3EihUrLJ6DbMGCBaJFixbCx8dHBAQEiMaNG4u33npLXLlyxXAMh8Hbj0IIB/bsIiqmkSNHYv78+cjMzKywS0qUJ5999hneeustnD59GhEREc4uDpWhPn364Ny5c9i9e7ezi0LkFGwCI6fJyckx6fB38+ZNfP3113j44YcZfsrI5s2bMXz4cIYfNyOEwJYtW5zaIZrI2VgDRE7TrFkzdOzYEfXr18e1a9ewcOFCXLlyBRs3bkT79u2dXTwiIqrAWANETtOtWzf8+OOPWLBgARQKBZo3b46FCxcy/BARkcOxBoiIiIjcDucBIiIiIrfDAERERERuh32ArNDr9bhy5QoCAgLsPiEfEREROYYQAnfv3kV0dDSUysLreBiArLhy5YrFisBERETkGi5evIgqVaoUegwDkBUBAQEApBcwMDDQyaUhIiIiW2RkZCA2NtbwPV4YBiAr5GavwMBABiAiIiIXY0v3FXaCJiIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjtMAARERGR22EAIiIiIrfDAERERERup1wEoDlz5iAuLg7e3t6Ij4/H7t27Czy2Y8eOUCgUFpcnnnjC6vGvvvoqFAoFkpOTHVR6IiIicjVOD0DLli3DqFGjMHHiROzfvx9NmzZFQkIC0tLSrB6/YsUKXL161XA5cuQIVCoVevfubXHsypUrsXPnTkRHRzv6aRAREZELcXoA+vjjj/Hyyy9jyJAhaNCgAebNmwdfX198+eWXVo8PDQ1FZGSk4bJhwwb4+vpaBKDLly9j2LBhWLp0KTw9PcviqRAREZGLcOpq8BqNBvv27cO4ceMM25RKJbp06YIdO3bYdI6FCxeiX79+8PPzM2zT6/V44YUXMGbMGDRs2LDIc+Tl5SEvL89wOyMjoxjPoniEXo/83CwolSooFUoolSpAXrVWobh/kW8TERGR3Tk1AN24cQM6nQ4REREm2yMiInD8+PEi7797924cOXIECxcuNNk+ffp0eHh4YPjw4TaVIykpCZMnT7a94KWQdScNN04dBoSAAoBCqYSH0gueSg+oVB5SMFKqoFKqoFQqoVB5QAmltE/lAYVSKQUj+af5dfMAVdh1W48zVthtBjYiInIRTg1ApbVw4UI0btwYrVu3Nmzbt28fPvnkE+zfvx8KG7+Qx40bh1GjRhluZ2RkIDY21u7lBaQaIE1OJnxCw6HX66ETeuQLLbKFBiJfD6EXEEIPhZCOV0IBhQJQQQUFpJ8eKpUUmKS9UCqU967DULOkUiihVNxr4RTCejixRwAqbJ9Saftt88cyv12cc1krS2FBzZ63bT3W09OyzEREVGacGoDCwsKgUqlw7do1k+3Xrl1DZGRkoffNysrC999/jylTpphs37ZtG9LS0lC1alXDNp1Oh9GjRyM5ORnnzp2zOJdarYZarS75EykmhUIJT7WPTcfqhR56vV76KfTIh0Cu0EEILfRCDwAQEFDc+0+pUEKhUEClUEGhUMBD6QGVUgVPpSdUinvNboaLAsp791FCcS9CQQpMQty/LjO+XtRtIQC9vuT3tfXYgm4rFJbXS3K7oPMWdKyt4ahyZaBSJeuPQ0REDufUAOTl5YUWLVpg48aN6NGjBwCp/87GjRvx+uuvF3rf5cuXIy8vD88//7zJ9hdeeAFdunQx2ZaQkIAXXngBQ4YMsWv5y4JSoYRSZVtNgRDCEJT0Qg8hBPK0edDj/m3gfmCSw5ISynu1RlLTm4fSAx4qDylE4d5UA1BAoVQYbgO4v72Qn27F1qCWkWG5j4iIypTTm8BGjRqFQYMGoWXLlmjdujWSk5ORlZVlCCsDBw5ETEwMkpKSTO63cOFC9OjRA5XM/oquVKmSxTZPT09ERkaibt26jn0yTibX/Kigsul4ORTp9DrohR5aoYVGq4GAtM2cgPSlLdc2yUFH+l9hsl0ujxJKKO819RjXPsn7bA1R5rfl5r1yFbhsbS50djmJiMj5Aahv3764fv06JkyYgNTUVDRr1gxr1641dIy+cOGC4QtUlpKSgr/++gvr1693RpErDKVCCSgAldK2wCQzrkkyXBfCcNvwUwhooYXQ3tsGy2NtUVjYAmA1cCmggFKpNNwuKGyZn5+1W0RE7kEhbP0WciMZGRkICgpCeno6AgMD7XruuzeuIPXfPQiIdEwn64rIPDBZC1sALANWMQOXcV+q4tRuydvNa7jMz2X4mZEBVK4MReXKhYYs+XEYuIiIbFOc72+n1wARFcU4jDiaHJbk68bbilu7ZdxkaEx59y50XrnQeWTa3PwHwKQ2y9rPQs9RxOOwdouI3A0DEJER49oXRwUuRZ4Cep9gCJ8Q6+HKStgSENAJHaBHgcfZ9Ng2Bq7CQpZJDVcpf8qPR0RU1hiAiJzIGbVb5n24iqzdKqLJ0fBczJ6ErWFL/q8ktVvyYxo6xbN2i4hsxABE5CbKMmwBBTQdWglbekijEaVJQK2Hs6IU1nfK1tot4xGL9qjdYtgiKt8YgIjIIZxVuyXXTBkmCi1B7RZQeA1XcWq3zCcnLU7tVnH7chGR7RiAiMjllefaLa1eW2g4K4otQQiwrXarJE2GDFtUUTEAEREVkzNrtwrsy1WM2i35fhbPC7YFHzlsWUz/YNaXqyShioGLygoDEBFROVZea7e00Er9thwwKtF4H2B77VZpp39g2HIvDEBERGRQXmq3TGq5ilm7Bdyf2FSu6Spu7ZbxDPIqhQpKpRIeSg+Tfl2GWq9CblP5xQBEREROUdZhCyhk+ocimhKNF5SWzyMHLOPmQOMAJIcpD6UHVEppsWkllFApVUUGKIYox2MAKgsrVwJHj0IHgcup/+LGjQvI8w9DrF80FAoFtLVrQPPoI84uJRFRhWXcLGbvwCXEvYBkFJTkn/kiHxq9xmQ7cD9AGcp3rxaqoBAlBygPpYfNIcr4Npv2LDEAOdrKlUDPngAAFYB6Vg4RAG7PnsEQRETkguRmstIwr2mSA5VcI6XRaiy2G+5rFKYKaoYzbsaTa6WUSqXNAUq+bS93cu8gV5uLSP9Iu52zuBiAHO3oUQDAhurA5jjTXQoAHc8Bj54FPE6eKRcBSHnlKpS37xS4Xx8SDH10VNkVqIIwfl0VmVnQX0sDgoIM+/WVQqGLjXFS6YjI2YznjSop46Y8vdCbhiihhV5rGa4A67VR1prhlIp7NVEK6VKaEJWdn12KV8s+GIAcTAcBFaTwk9TB+jGPnrVtPhBHU165irBuz0KZk1vgMXofb9xY8yNDUDHY9Lr6+uD6nq0MQURUYsZ9qlQoeY2UeS2UfFsndNBqtRbbgYJDlHkznEKhgIfSA0IIeKo8S/2cS4MByMHO3z6HGoXsl2PPxawrCCuLAhVCefsOlDm5uDPjPWiqV8OR2ym4mXcbldQhaBRSF15nzyN4zHgob99hACoG49dVW7O6VAMUEmyoAfJIOYmQl4dBefMWAxAROZ1SoSx1iDIPUMahKk+bh7uau6jsW9mOpS4+BiAHu6u5a9Nxmdpspwcg2Rb/Gxh6ehau5qQZtkXdCsdngf3Rw3nFcnnamtWhbVgfivQM6CuHQVQKdXaRiIgcoqh+UTqhg6fSuTVAHF/nYAFeATYd5+/h6+CS2O69A5+YhB8ASM1Jw3sHPnFSiYiIiOyLAcjBqoXEASh41KW8PdYvuiyKUyi9Xh6eaUlYOY6IiMhVsQnMwVT3Ik7Hc9Jt4yChMNpeHuZoOHI7BYWNQxNGxzVAw7IoEhERkUMwADlaQykoPHpWulgjAGhrF9ZVumzczLtt1+OIiIjKKwYgR3vmGWDFCsNM0CfvzQQdXA5ngq6kDgEA1L9hfb+8XT6OisfjtJSAzecB8kg56cxiERG5JQagsvDMM8Azz0AFIObGFaj+3YOAyFjkOLtcZhrUiEe2J7B0RcHHZHtKx5Ht9CHB0Pt4I3jM+IKP8fWBnqPCiIjKDAMQ3VclBr8sfgcztk4FYNlfCQDGdHgHHatwrpri0EdH4caaH01ngjaaBwjgTNBERGWNAYhMdGzZC1kRIRi/b4bJUPho3whMaf4mOsY6v6nOFemjowyTR3IeICIi52MAIgtPxD6CrjEdsOv6AVzLuYEInzDEV34AKmXpFvsjIiIqLxiAyCqVUoWHIlo6uxhEREQOwYkQiYiIyO0wABEREZHbYQAiIiIit8M+QERERFbo9DrsurwLaVlpCPcLR3xMPAeDVCAMQERERGbWnFyDCZsn4GrmVcO2KP8oTOk0Bd1qd3Niyche2ARGRERkZM3JNXjl11dMwg8ApGam4pVfX8Gak2ucVLKKQaPVYMnBJXhn0ztI3pkMjVbjlHKwBoiIiOgenV6HCZsnQJjMhS8REFBAgYlbJiKhZgKbw0rg/T/fx/x986EXesO2N9e/iVFtRuHDRz8s07KwBoiIiOieXZd3WdT8GBMQuHL3CnZd3lWGpaoY3v/zfczdO9ck/ACATugw4+8ZeGvDW2VaHgYgIiKie9Ky0oo+qBjHkUSj1WD+vvmFHvPxjo/LtDmMAYiIiOiecL9wux5HkiWHlljU/JjTCR0+2/tZGZWIAYiIiMggPiYeUf5RUEBhdb8CCkQHRCM+Jr6MS+bazqeft+m407dOO7gk97ETdBlTKJVQqtW4m3rR0KFOoVBApVRBpVBBofKAysMDSqUHoFICSiWgUJj+JCIih1ApVZjSaQpe+fUVKKAw6Qwth6LJHSezA3QxVQuqZtNxNUNrOrgk9ymEEJZd3d1cRkYGgoKCkJ6ejsDAQLueW+j1yM1Kh06rgU6nhU6XD22+BvnaPORrciA0+dBpNRBaLYReB+j1UAgBJRRQCQWUQgpRKqUKSiihkEORUgFAui6UZoFJvk5lx/hjJV+/91NxNxP68MoQlUKdUDAisoW1eYCiA6IxueNkzgNUAhqtBjVn1Sy0GUylUCH7nWx4eXiV+HGK8/3NGqAyplAq4RMQYnWfEAJ6oYdO6KDT66DTa6HT5kOny4dGkyuFpPw86HVa6PU66HT5UOgFoNUCOh1UegEPHaDUKaCCAko5MEEAQhgqdA1fzXJwUtwLTeY1TcbBydoXus3XTZ6lzfdTGN+voONRxGMW9jjGz83KdcvIaPQK3jumwL8fTAKnwnQ7/3IkKve61e6GhJoJnAnaTrw8vPDfFv/F3L1zCzxmVJtRpQo/xcUAVI4oFAqoFCqooALkz5ja9BghxP2AZPRTq9MiX58vXfLzoNHroNPmQ6/TAnq9oSZJJdckQQGVXkClE1DpAei0UpjK10ohQC8AoTcNDda+1I23mexWFOu6UBhtk8MFACiMmvyMm/+UCovjTa4bHSuM71fMcpX0uijqtfD0BBGVbyqlCg/FPuTsYlQY/2v/PwCwmAdIpVA5ZR4gNoFZ4cgmsLJQWEjS6DXQaDWG7XKNEwCjkKQ0CUpKhVJq+1YqDH2WTEKHzBHXiYioQpGHxF/LuoYmEU0wtOVQu9X8sAnMzSkUCngoPOChLPjt1Qu9RUCSmt10JiFJK/TQCz2E0ENA3E/thiYmmHTmVhgFJvk6BKRtxvsVRmGKiIjchpeHFwY1G4RgdTAiAyKdVg4GIDelVCihVCnhiYKbYvT3wo8UgKTwIyAKvK7X66HVa6GH3hCm9JDuq9PrDAFKCGEapnA/RMmshSXlveYwheJ+yGKYIiKikmAAogIpFUpD6CgN89BT2HXjMKUTUs2UcbAS4t4+6EyPLyBMCQirYclamDK/TkREFRcDEDmcUqG0NqSqRMxDjy3BSq/XQ497IcooWBUUpoQQ0ENvmAPEWpgyD0vWtjFMERGVXwxA5FLkkXL2YFOzXhFhSr4uhHS/fH2+Za3WvTAFmNZOFRScLPpTMUwREdkdAxC5rbIOU+bBSg5Qxv2ljMOUXEtlHsQMj2nWb6qwjubWmgAZpojInTEAEdmBvcNUUX2kzK/LAco4TMnTHBQWpsxDlIxhiogqOgYgonLGvKN2adja+dw8ZMmdz407ohv6TQk98kW+ZRAz6y9l3NRX0Ki9gvpOERE5GgMQUQXmiDBV3I7o5p3P5aY/+TzWarUKez7F7YhORGQNAxAR2UQOGVBAWq6lFKyFKVuClXnnc/MwZa25ELDsLyU/H2thCoDVzucMU0QVCwMQEZW5sgxTBV0373xuHKaKmrjT8NgQBdY8mYcpTtxJVL6UiwA0Z84czJgxA6mpqWjatClmzZqF1q1bWz22Y8eO2Lp1q8X2bt264bfffkN+fj7+97//Yc2aNThz5gyCgoLQpUsXTJs2DdHR0Y5+KkRUxuwZpoCiJ+505izoDFNE9uP0ALRs2TKMGjUK8+bNQ3x8PJKTk5GQkICUlBSEh4dbHL9ixQpoNBrD7Zs3b6Jp06bo3bs3ACA7Oxv79+/H+PHj0bRpU9y+fRsjRozAU089hb1795bZ8yIi12TPiTvtPQu6HKbMg5i1pj7jiTu5pAyRJaevBh8fH49WrVph9uzZAAC9Xo/Y2FgMGzYMY8eOLfL+ycnJmDBhAq5evQo/Pz+rx+zZswetW7fG+fPnUbVq1SLP6eqrwRNRxWPvWdALajosqBN6URN3chZ0Ko6MvAyHLIbqMqvBazQa7Nu3D+PGjTNsUyqV6NKlC3bs2GHTORYuXIh+/foVGH4AID09HQqFAsHBwVb35+XlIS8vz3A7IyPDtidARFRGOAs6wxTZl1MD0I0bN6DT6RAREWGyPSIiAsePHy/y/rt378aRI0ewcOHCAo/Jzc3F22+/jf79+xeYBpOSkjB58uTiFZ6IyEU5cuLOks6CbsvEnYbHhPVZ0IsKU5y4k4w5vQ9QaSxcuBCNGzcusMN0fn4++vTpAyEE5s6dW+B5xo0bh1GjRhluZ2RkIDY21u7lJSKqaMpq4s6SzoJuy8SdxgrqaF7Qc7fYZmW6BWceQwVzagAKCwuDSqXCtWvXTLZfu3YNkZGFtwtmZWXh+++/x5QpU6zul8PP+fPnsWnTpkLbAtVqNdRqdfGfABER2U15nQVdD73hnOaPUdhtABZ9qmw5xuq5S3CegkJeUczvZ8t5bH0sOaTlanPh51lw15Wy4NQA5OXlhRYtWmDjxo3o0aMHAKkT9MaNG/H6668Xet/ly5cjLy8Pzz//vMU+OfycPHkSmzdvRqVKlRxRfCIiKqfsGaYA+4WbooJNRX4svf7+9A95ujyn11g5vQls1KhRGDRoEFq2bInWrVsjOTkZWVlZGDJkCABg4MCBiImJQVJSksn9Fi5ciB49eliEm/z8fDz77LPYv38/Vq9eDZ1Oh9TUVABAaGgovLy8yuaJERFRhWFLE5S9pk+gsuH0ANS3b19cv34dEyZMQGpqKpo1a4a1a9caOkZfuHABSqVpgk9JScFff/2F9evXW5zv8uXL+OWXXwAAzZo1M9m3efNmdOzY0SHPg4iIiFyH0+cBKo84DxAREZHrKc73N1f3IyIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjtMAARERGR22EAIiIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjtMAARERGR22EAIiIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjtMAARERGR22EAIiIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjtMAARERGR22EAIiIiIrfDAERERERuhwGIiIiI3A4DEBEREbkdBiAiIiJyOwxARERE5HYYgIiIiMjteDi7AERERFQ+CCFd9HrpIl9XKACl8v5P+borYwAiIiKqIOTAYv7T2jadTvqp1UoXeRtQdABSKKSLh4d028MDUKmki3lIsvZTvjgTAxAREVE5ZBxeCrrodPcDjFZbcICRQ4w5a0FFoQA8PU23WTunEEBenuU28+cgn9P8sYKDgUqVHPoSFooBiIiIqIzodPcvtoQZ8xBkHjDMa1SMA4x5sCmt0tTYmIekrCwgP7/0ZSoNBiAiIiI7kEOM+SU/H9BopIvxMbaEGZXqfphx5X43cnhSqaSfubnOK4uMAYiIiKwybjpx5S9fexDCerjRau+HG/PaHZlSKX3xyz+9vMpHHxh3xwBERORm5C9oaz+1WqnGIj///nY5AHl4SF/e8he43OnV/LojGDcTGf807sQrd+QtCTncWQt5er30ehjX3hjfz/i5q9WOfR3IfhiAiIgqAPORPcY1EXIzjHyRazOMR/3I5JoJeTSPp+f9c+fmAtnZ95tv5LBgHoLkoGQ8Msj8GIXC9lCj0Vg+P/PmI/n8RdVSmd/PFnIA9PQEvL1ZG1ZRMAAREbkA8yYYuZOstf4lcogwZhxAFAqppsJe/UqMw5ZWK40Munv3flAxHgVkHFRsDTXyhQGE7IkBiIioHDAOOHK4kQNOXt79JinjJhghLOdgcUb/EvmxC2Ne66TXM9SQczEAERGVgYJGCGk09wOO+Qgh41oTuWnJVcOCPGkeUXnBX0ciolKyxwghebizt3fRtSlEVHoMQEREhSgs3Mj9b4yXEeAIISLXwABERC7LfHSTVnt/n3Eti7zAo63Xje9vPBOv+eR1xuGGfVmIXEu5+Dtkzpw5iIuLg7e3N+Lj47F79+4Cj+3YsSMUCoXF5YknnjAcI4TAhAkTEBUVBR8fH3Tp0gUnT54si6dCRA6i1QI5OUBGBnDjBnDxInDunHS5cAG4elXafvOmdLl9G7hzR7qkp0ujku7eBTIzpaHc2dnS+fLypItck2M8n4wcdjw8AB8fICgICAm5fwkMBPz8pH1eXvdHWBFR+ef0GqBly5Zh1KhRmDdvHuLj45GcnIyEhASkpKQgPDzc4vgVK1ZAo9EYbt+8eRNNmzZF7969Dds+/PBDfPrpp1iyZAmqV6+O8ePHIyEhAf/++y+8vb3L5HkRuSrj0TpyLYjxUGRHN98IYToZn0YjhRX5tlwmT08pmPj5sc8MERWfQoiSTAtlP/Hx8WjVqhVmz54NANDr9YiNjcWwYcMwduzYIu+fnJyMCRMm4OrVq/Dz84MQAtHR0Rg9ejTefPNNAEB6ejoiIiKwePFi9OvXr8hzZmRkICgoCOnp6QgMDCzdEyRyMvO5VswXXyxoAUbjzrrmAcjDQ7p4elrO1WLtUlCtiBCmTVh5eVKtjPGEffJkfPLjsf8MkevLyJBWg4+MtPd5bf/+dmoNkEajwb59+zBu3DjDNqVSiS5dumDHjh02nWPhwoXo168f/Pz8AABnz55FamoqunTpYjgmKCgI8fHx2LFjh9UAlJeXh7y8PMPtjIyMkj4lojJnXFsiBxvjPjHmK0qbz/wLmIYYueOuPJ8MYDlbr/FjFbago9wkJAcm+aJQSLMK5+TcLz9wvy+Npyfg68vmJCJyHKcGoBs3bkCn0yEiIsJke0REBI4fP17k/Xfv3o0jR45g4cKFhm2pqamGc5ifU95nLikpCZMnTy5u8YnKlHnTUF6eadOQcRAxDyClXVFarvWxhXktkxD3a3aMm9XkMqnVUjMWww5R+aLTAbt2AWlpQHg4EB9fsZqbnd4HqDQWLlyIxo0bo3Xr1qU6z7hx4zBq1CjD7YyMDMTGxpa2eEQlZlyLk59vWVsizwAsNw35+JSfpiGuck3k+tasASZMkAYXyKKigClTgG7dSndunQ7YvRvIygLq1wfatXNOsHJqAAoLC4NKpcK1a9dMtl+7dg2RRTQMZmVl4fvvv8eUKVNMtsv3u3btGqKiokzO2axZM6vnUqvVUKvVJXgGRKVnPpQ7J0cKPHKfHMB0gUnWllBZqOh//TuLo15Xe553zRrglVcsm7dTU6XtCxaUPARZC1ZVqgCffAL07Fmyc5aUUwOQl5cXWrRogY0bN6JHjx4ApE7QGzduxOuvv17ofZcvX468vDw8//zzJturV6+OyMhIbNy40RB4MjIysGvXLrz22muOeBpENjGeRE8e3ZSTI/2U13kC7veT8fHh0gHkHI78698RXCFUAI57Xe15Xp1OOpe14VHyoISJE4GEhOK/FgUFq8uXgWefBX78sWxDkNNHgS1btgyDBg3C/Pnz0bp1ayQnJ+OHH37A8ePHERERgYEDByImJgZJSUkm92vXrh1iYmLw/fffW5xz+vTpmDZtmskw+H/++cfmYfAcBUbFVdhCluYrdRuv82Q+morI2Qr6kpJrHUvz1z/gnqFCPp8jXld7n/fvvwGjWWUKtHw58NBDtp9Xp5Pea+PX05hCIdUEnT1but8HlxkFBgB9+/bF9evXMWHCBKSmpqJZs2ZYu3atoRPzhQsXoDTrUJCSkoK//voL69evt3rOt956C1lZWXjllVdw584dPPzww1i7di3nAKISMx8ybryQpVyDY74UQkErdXOyPLI3e4UKR/71D5RdqChtU429z+uo19UR501Ls+9xsl27Cg4/gFTeixeBbduAjh2Ld+6ScnoNUHnEGiASQhphdfeu1B/HfCFL45W6PTxMF7RkTQ6VJXuGCkf99S+X0541FbbUKERFATt3Fj9U2Pu8jnpdHXFeR5V11SogMbHo4779Fujf3/bzmivO9zfHahAZ0eul0HPpEnD+PHDrllTjo1BIw7UDAqQlEEJDpZ9BQZZLIRCVFTlUmH9ZyzUVa9YU73yO+uu/qJoKQKqpsDZHVUFsqVG4ckU6rjgccV5Hva6OOG98vBTwCqqlViiA6GjpuOKwsrCDVUZjlxyOAYgI0j+8GRnSmlIXLkjDMwMCpJlKfX2lRS45CzGVJ44IFbZ+Sdl6nMzdQ4WjXldHnFelkmoPAcsQJN+ePLn4f+zZEqxiY6Uh8WWF/5yTW9NqpcUyL1yQ2p81GqlWJzCQtTlkfzqd1MSwapX0szjhxJwjQoWj/vp391DhqNfVUeft1k1qkjSfjSYqquT9qmwJVsnJZfvvLgMQuaX8fGnF8PPnpeYurVaq7fH3Zy0POcaaNdIXUe/eUl+I3r2l28VtppI5IlQ46q9/dw8VjnpdHXVeQAo5u3ZJfX3mzJF+7txZupF1BQWrKlXKfgg8UIoAdOrUKaxbtw45OTkAAPalJleQlwfcuAGcO3f/r+eQEE4uSJbsWVtj7746gONqQBzx1z9DhWNeV0eeF5Ce40MPAT16SD/tUTsjB6slS4DPPgM2b5aGvpd1+AFKMArs5s2b6Nu3LzZt2gSFQoGTJ0+iRo0aePHFFxESEoKZM2c6qqxlhqPAKp6cHKmPT3q61Mzl4yNdyPU5YhI8e08s58jRSqmp1vsBlfS8xue393w9r7wiXTcurz3mwTF/r6KjpZBi73mA7HFeV5m00dHKw2rwxQ5AAwcORFpaGr744gvUr18fhw4dQo0aNbBu3TqMGjUKR48eLVXhywMGoIpBCCn4pKdLF51O6tDMVU8qDkdMgucqE8sBjgsVjsJQQbLyEICKPRHi+vXrsW7dOlSpUsVke+3atXH+/Pnino7I7oSQRnGlp0sfMiGkJi5PT2eXjOzJEZPgudLEcsD95g9rIbC0ocIRunWTXjt7hwq5qcbeHHVeKh+KHYCysrLg6+trsf3WrVtcUJScSq8HMjOlUV2ZmdKXlZ8f19MqifL+l6+jZtYtzsgqW78YHdVXR+aoUOEoDBVUXhT7q6Fdu3b46quv8N577wEAFAoF9Ho9PvzwQ3Tq1MnuBSQqilZrGnw8PaU5fMrrF4C9ucraSvYsqyOCCuDYieWK6qtT3NFKxhgqiIqv2AHoww8/ROfOnbF3715oNBq89dZbOHr0KG7duoXt27c7ooxEVuXm3g8+ubnSTMzBwe41jN1V1layd1kd1azkyInlXnlFCjvW+uqUdKgyEZVcsb8qGjVqhBMnTuDhhx/G008/jaysLPTs2RMHDhxAzZo1HVFGIgO5f8/Vq9IcPqmp0pdISIj7zeFj76HVjphZ2FFldVSzkitNLEdEpVOsUWD5+fno2rUr5s2bh9q1azuyXE7FUWDlj1YrBR/j/j2+vlKtjztypQUbHVFWRw4Bd+TIqvLet4qorJSHUWDF+nvZ09MT//zzT6kKR1QceXn3Z2y+eFG6HRgofXDcNfwArrW2kiPK6ugZcF1pYjkiKpliNxg8//zzWLhwoSPKQgTAspnLeMZmf3/X/tKw1+zCrrS2kqOClSODiiOWASCi8qXYnaC1Wi2+/PJL/PHHH2jRogX8/PxM9n/88cd2Kxy5F53OtJkLkJq5/P2dWiy7sWcnYEeurWTv0UqOHAbuyCHgHFlFVLEVeybowoa6KxQKbNq0qdSFcjb2ASpbeXmmo7k8PKTg48o1PebsPbuwo/rAOKL/i6OXbCAi11Me+gAVOwC5AwYgx5OXqbh79/76XN7e0vpcFW1RUketBeVKayu52pINRORY5SEAlWrQ8KVLl3Dp0qXSnILcjBBS6Ll0Serfc/OmNHFhaKhU61PRwg/gmE7AgGNXl7Z3/xcOA3dNer1UQ5uXJ43E5J/LVJEUuw+QXq/H+++/j5kzZyLzXkeNgIAAjB49Gu+++y6U7jQRCxWLVgvcuAHcuiXN11Oel6mw53BlR68F5SprK7nakg3uRgjpM5qfL9XIykuKyKMt8/JMO+0rFNJ7Z36piH/EuBshpN8Dne7+TyGkf7eVSuk9Nr5ufttVYkCxv37effddLFy4ENOmTUPbtm0BAH/99RcmTZqE3NxcfPDBB3YvJLm+7GzpSy8zUxrGXl6DD2D/2ZUdvRaUK3XWdaWyVnQ6nRR0tFrpAki1sZ6eQFgYoFZL4cfLS/ry0+lML/n5UijSaO6fS6eTao0A6UtQDkVKpfSZtzUgCWF50evvX5ePMd5mvE9+DPnL2dr1on5auy4r7LarBUD5/ZfDrxx0PDykS1CQ9LsASK+3fKz8eyD/bsjvh/yeAPdDNGAZlOTfOWcqdh+g6OhozJs3D0899ZTJ9p9//hlDhw7F5cuX7VpAZ2AfIPvR66XOzdevSx+GgIDy/Q+EvTsrA+wETM4n/0Uv1+4A0peQl5fU906eVFQOQMX9jMpfhsZfjFqtFJDy8y2/MK31AzMuq/xFKX9pGn95yveRv0jlgCX/BEyDUWHX5cczfmzjL2/j68bHmJe3oNvGAcD8tnk4MH4Oxhd7/Zug11vW6si1eJ6e0vvv4yP99PCQtnl4FP27YBx6CrtuHJ7lEBUUBFSqZJ/nJyvO93ex/w6/desW6tWrZ7G9Xr16uHXrVnFPRxWYRiM1ed2+LX2wvL2dXaLCOWqFca4FRWVN/mteDh9yU5anpzSfllyz4+lpvykDVKqCJyc1rz3SaqUvxYKCTmEByFqNTGkUFnqKG4CKG47kkGBeG2ccUoyDmvzcCwpKSuX9pkzjCyDdTw64AQFSrY4cdDw9S95sJT++Kyp2AGratClmz56NTz/91GT77Nmz0bRpU7sVjFzb3btSk1dOjpTyXeHL3VErjAP3OwFba1orzegqIvnLUv6yE+L+X/VBQdIfH/Jf+J6ezimjHJDKI3sHqtIyD0bGP+VaHDngymFJo7m/H7jffOXvf//9N67VIUmJVoN/4okn8Mcff6BNmzYAgB07duDixYtYU9wVDanC0emkGp/r16V/8EJDnV0i2zmyszLATsBUesZffPJf9nLY8fGRLnLfndL8VU/OY9y5vKjAaty8JP8E7gcdvv+FK3YA6tChA1JSUvDZZ5/h+PHjAICePXti6NChiI6OtnsByXXk5krBJz1dGuEld5xzFY7urAywEzDZzrwZC7j/V7zxX/byhUHa/RiHJSo+ToRoBTtBF48Q0qRW169LVbGBga75lwc7K5MzGPfZkPt9APebMeSaHeOw44qfL6Ky4NBO0IsWLYK/vz969+5tsn358uXIzs7GoEGDintKcmFarTSZ4c2bUrV7cLCzS1Ry7KxMjmLej0MeTiz32ZCbLIKCpMECchMWmzGIHKfYH62kpCSEhYVZbA8PD8fUqVPtUihyDdnZ0ozO169LTV6+vs4ph71WWAc4YzHZRp77RJ4LJydHWsj37l1p2ofbt6UJP2/fli5375pOJOjlJfWPi4kB4uKkS/Xq0u9ZSIj0efLyYvghcqRi1wBduHAB1atXt9herVo1XLhwwS6FovJNr5f6+aSlSddDQpw3isLekxYC7KzszqzV1Mi35WHR8u+6+VBk42HFcs2N+dwu8vXyNOqIyF0VOwCFh4fjn3/+QVxcnMn2Q4cOoZK9ZzSickee2+fWrft9E5yloEkLU1Ol7aWpsWFn5YpDrq2Rw4z5beMwYh5Y5FDj5WU54Z75dYYaItdS7ADUv39/DB8+HAEBAWjfvj0AYOvWrRgxYgT69etn9wJS+ZGZCVy7Jo32cvbcPo6atJBcj/HcKMa1NTLzWXY9PO73s5GXZyhoVmGGGqKKq9gB6L333sO5c+fQuXNneNybUUmv12PgwIHsA1RB6fVSjc/169KXQnCw878YHDlpIZVfxpP+Ga9bJI+O8ve/PyTcPMwYBx0iomIHIC8vLyxbtgzvv/8+Dh48CB8fHzRu3BjVqlVzRPnIyfLypH4w5W1uH0dPWkjOZT403HweHB8fqROx8fpVnOGWiIqjxP9k1K5dG7Vr14ZOp8Phw4cRGBiIkJAQe5aNnCw7W6pFycuTan3K01/OZTFpobsxXrjQ+Cdgfa0m80tpHtd4kUbjoeEqlTSvFOfBISJ7K3YAGjlyJBo3boyXXnoJOp0OHTp0wN9//w1fX1+sXr0aHTt2dEAxqawJIc3to9FIo7zKm/h4abRXUZMWxseXfdmcRZ5Xxnila2uhBrjfT8p4VJPc58X4p8x8JW3z1bWLYr6gpVzDY96EFRhoWqtTkpXJiYhsUewA9OOPP+L5558HAPz66684c+YMjh8/jq+//hrvvvsutm/fbvdCUtnLzpbmLgkIcHZJrOOkhVL4yMuTLnr9/ZoR47ChVt/v/2LcF8Y86JivtG0egsyDT0GBqKCL8dByeS4c4wn/2IRFRGWt2P/k3LhxA5H3Zolbs2YN+vTpgzp16uDFF1/EJ598YvcCUtkTQprMDbBvgNDp7Du3jrutsC6EVCOn0UhNRfIw7UqV7i+CaR5u7IW1MERU0RQ7AEVERODff/9FVFQU1q5di7lz5wIAsrOzoarIf267kZwcaW0vf3/7ndMRExYCFX/SQq32fugRQqox8fWV3htv7/s1PEREVDzFDkBDhgxBnz59EBUVBYVCgS5dugAAdu3ahXr16tm9gFT20tOlL1t7NUk4csJCoGJNWiiE1KSl0Ujhx8NDCjnh4fcDj6ens0tJROT6iv0VN2nSJDRq1AgXL15E7969ob43LlqlUmHs2LF2LyCVrZyc+0Pe7YETFhYtP18KPHl5Um2Ol5fUGViedkCtZhMUEZG9lehv/GeffdZiG1eBrxjS06XQYq9aBk5YaMla52W1WprXRq7lYYdgIiLH4j+zZJCba9/aH4ATFsqEkF7f3Fzptre3aedlLy/W8hARlSUGIDJIT5eaY+w59N2dJyy0FnrCw6VOzD4+7LxMRORMDEAEQGqOsXftD+B+ExbKnZhzc6XmLW9vICxMel19fNy3nxMRUXnDv0EJgDTsXaOx/1pf8oSFgGUTT0WasFAOkHfuSH2oQkKAatWAuDggIkIatu7qz5GIqCJhACJoNNIXt6+vY84vT1h4b/5Mg6io0g+BdyaNRgqOt25JQ9aDgoDYWCn0REVJTYnszExEVD7Z/M9zfn4+3n33XaxYsQKhoaF49dVX8eKLLxr2X7t2DdHR0dDJ89yTy8jIkGowQkMd9xgVZcLC/HypeSs/X+q4HBAg1e74+Ei3iYjINdgcgD744AN89dVXePPNN3Hnzh2MGjUKu3btwvz58w3HCFtWRaRyJT9fqv3x8XH8Y7nqhIVarRR6NBppyLqvrxR85BFcRETkemwOQEuXLsUXX3yBJ598EgAwePBgPP744xgyZAi+/PJLAICC43hdTkaG9OXuyNofV2Qcejw8pLBTufL90MNfdSIi12ZzH6DLly+jUaNGhtu1atXCli1b8Pfff+OFF15g05cL0mrLrvbHFWg0wN27Up+e7Gwp6MTESH16qlaVOjZ7ezP8EBFVBDbXAEVGRuL06dOIi4szbIuJicHmzZvRqVMnDB482AHFI0e6e1da+iIkxNklcQ55yHpenjRyS15oVG7e4uSEREQVl801QI888gi+/fZbi+3R0dHYtGkTzp49a9eCkWNptVJNh7vVaOj1Uui7fVuq/dLrgeBgqYYnLk4axRUczGYuIqKKzuYANH78ePTp08fqvpiYGGzdutXQF6g45syZg7i4OHh7eyM+Ph67d+8u9Pg7d+4gMTERUVFRUKvVqFOnDtasWWPYr9PpMH78eFSvXh0+Pj6oWbMm3nvvPXbQNpOZKQUBd2j+0umkJq3bt6W5egCpP488T09UlLT4KEdxERG5D5ubwKpVq4Zq1aoVuD86OrrAgFSQZcuWYdSoUZg3bx7i4+ORnJyMhIQEpKSkINzK2ggajQaPPvoowsPD8eOPPyImJgbnz59HcHCw4Zjp06dj7ty5WLJkCRo2bIi9e/diyJAhCAoKwvDhw4tVvopKp5PCQEWu5cjPl5q2NBpp9Jm3tzQhoa+v9Lxdbfg9ERHZl0LYoWokLy8Ps2fPxowZM5Cammrz/eLj49GqVSvMnj0bAKDX6xEbG4thw4Zh7NixFsfPmzcPM2bMwPHjx+FZwHLlTz75JCIiIrBw4ULDtl69esHHxwfffPONTeXKyMhAUFAQ0tPTERgYaPPzcRXp6cDFi1Lfn4oUgOT+PFqtNFzdx0fqz+Pt7X5NfURE7qg43982N4Hl5eVh3LhxaNmyJR566CGsWrUKALBo0SJUr14dycnJeOONN2wupEajwb59+9ClS5f7hVEq0aVLF+zYscPqfX755Re0adMGiYmJiIiIQKNGjTB16lSTEWgPPfQQNm7ciBMnTgAADh06hL/++guPP/54oc8tIyPD5FJR6fVS7U9RHXx1OuDvv4FVq6Sf5XWQn1Z7f+SWVisFHnk25thYKeT5+DD8EBGRKZubwCZMmID58+ejS5cu+Pvvv9G7d28MGTIEO3fuxMcff4zevXtDVYx2hRs3bkCn0yEiIsJke0REBI4fP271PmfOnMGmTZswYMAArFmzBqdOncLQoUORn5+PiRMnAgDGjh2LjIwM1KtXDyqVCjqdDh988AEGDBhQYFmSkpIwefJkm8vuyjIzpYtRq6GFNWuACROAq1fvb4uKktb0Kg/LVsirrOfkSE1Zfn7SMhucjZmIiGxlcwBavnw5vvrqKzz11FM4cuQImjRpAq1Wi0OHDpXZBIh6vR7h4eFYsGABVCoVWrRogcuXL2PGjBmGAPTDDz9g6dKl+Pbbb9GwYUMcPHgQI0eORHR0NAYNGmT1vOPGjcOoUaMMtzMyMhAbG1smz6ksybU/np6AsoC6vzVrgFdesVy5PTVV2u7Mtbs0Gin06HRS2ImMlMIPm7eIiKi4bA5Aly5dQosWLQAAjRo1glqtxhtvvFHi8BMWFgaVSoVr166ZbL927RoizVfNvCcqKgqenp4mNU3169dHamoqNBoNvLy8MGbMGIwdOxb9+vUDADRu3Bjnz59HUlJSgQFIrVZD7QZrGmRlSbU/QUHW9+t0Us2PtV5hQkghY+JEaU2vsupELA9bz8uTgltgoHTx9WVHZiIiKjmb+wDpdDp4GbUveHh4wN/fv8QP7OXlhRYtWmDjxo2GbXq9Hhs3bkSbNm2s3qdt27Y4deoU9Hq9YduJEycQFRVlKFt2djaUZtUbKpXK5D7uSAhp3huVquDan127TJu9rJ3jyhXpOEfLzZVqqzIypKUo5BmZY2Kkfj4MP0REVBo21wAJITB48GBDTUlubi5effVV+Pn5mRy3YsUKmx981KhRGDRoEFq2bInWrVsjOTkZWVlZGDJkCABg4MCBiImJQVJSEgDgtddew+zZszFixAgMGzYMJ0+exNSpU02Gt3fv3h0ffPABqlatioYNG+LAgQP4+OOPTVaud0dZWVKYKKxTfFqabeey9bjiMl5/S60GwsLur7ReUGgjIiIqCZsDkHnz0fPPP1/qB+/bty+uX7+OCRMmIDU1Fc2aNcPatWsNHaMvXLhgUpsTGxuLdevW4Y033kCTJk0QExODESNG4O233zYcM2vWLIwfPx5Dhw5FWloaoqOj8d///hcTJkwodXldlVz7o1QWXnNiZeqlUh1nC7lDc26uVD4/v/vz9RQw0wEREVGp2WUeoIqmos0DlJUFnD8v1aZ4FBJ5dTogPl7q8Gztt0KhkEaD7dxZ+iao/Pz7HZq9vaV+SezQTEREpeGQeYDINQlxf/mHwsIPIIWaKVOk6+YhRL49eXLpwo9OJ83Zk5sr9eWpWlVakiIsjPP1EBFR2WEAquBycqQAZNZVq0DduklD3c0H4kVFlX4IvF4vlSUkxLRDc1HBjIiIyN741VPBpadLtUDF6U/TrZs01H3XLqnDc3i41DRWmpofuR9SUJDUx4ehh4iInIlfQxWYXPvj61v8+6pUwEMP2a8s6elSHySGHyIiKg/YBFaBZWRIfW6cvTxEero0rD0qyvllISIiAhiAKqzcXKnJqSS1P/Z0967U/BYdLYUgIiKi8oABqILKyJAmFnRm6MjOlkZ1yQuVEhERlRcMQBVQXp5U++PM0JGTIwWwyEip7w8REVF5wgBUAd29Ky0n4e3tnMfPy5MukZGFL71BRETkLAxAFYxGIy0i6qzan/x8qekrIqLgVeeJiIicjQGogrl7V6p9cUYA0mqlxw8PB0JDOaszERGVXwxAFUh+vvNqf3Q6qeN1WBhQqRLDDxERlW8MQBVIZqY0/L2s+/4YL3FRubK0qjsREVF5xq+qCkKrlRYZLevV1M2XuCjtKvFERERlgQGogsjMlIael3XzF5e4ICIiV8QAVAHodFLtj1pdtrU/8hIXkZFc4oKIiFwLA1AFkJ1d9rU/mZnSEhdRUc6bb4iIiKikGIBcnBBSTYxKVXadj7OzpZ+Rkc5fa4yIiKgkGIBcXG6uVBtTVkEkJ0cabs8lLoiIyJUxALm4zExpBFhZdECWl7iIiuISF0RE5NoYgFxYfr7U/FUWfX/y84GsLC5xQUREFQMDkAvLzi6biQ+5xAUREVU0DEAuSq+Xlr3w8nJsIDFe4iIsjOGHiIgqBgYgF5WTIzVJObL5i0tcEBFRRcWvNBeVkSHVxjhy6Ym7d6XOzlzigoiIKhoGIBeUlyeFE0cOfdfppBqg0FAucUFERBUPA5ALyswENBrHLj+RlSXV/vj5Oe4xiIiInIUByMXodNLq644c+aXXS48THMxOz0REVDExALmYrCzHr/uVnS3V/LD2h4iIKioGIBcir/vl4eG4mhkhpEkPQ0M56ouIiCoufsW5kLJY9ys7Wzo/a3+IiKgiYwByIXfvSv1zHDUqSwhphFlICIe9ExFRxcYA5CLy86W5fxzZ9yc3Vzo/V3knIqKKjgHIRWRlOX7dr5wczvtDRETugQHIBej10tB3R877k5MjhSvW/hARkTtgAHIB2dn3Oyc7Sk6ONO+Pp6fjHoOIiKi8YAByAXfvSj8dNSw9N1eqXQoIcMz5iYiIyhv29ijn8vKkzs8F1f7odMCuXUBaGhAeDsTHF38EV3a2dF+1uvTlJSIicgUMQOVcZqY0Asxa7cyaNcCECcDVq/e3RUUBU6YA3brZdn6NRmr2Cgy0T3mJiIhcAZvAyjF53S9rQ9/XrAFeecU0/ABAaqq0fc0a2x4jOxsICnLs6DIiIqLyhgGoHCto6LtOJ9X8CGF5H3nbxInScYXJz5f6FQUF2ae8REREroIBqJyS1/1SqSzX/dq1y7Lmx/y+V65IxxUmK0tq+mLtDxERuRsGoHIqJ6fgdb/S0mw7R2HHabXSz6Agxy2sSkREVF4xAJVTmZkFr/sVHm7bOQo7Ljtb6ljtyLmFiIiIyisGoHIoP19q/ipo3a/4eGm0V0E1NwoFEB0tHWeNTieFq5AQ1v4QEZF7YgAqh7KypPl/Cuqbo1JJQ90BywAj3548ueD5gLKzAT8/1v4QEZH7YgAqZ+R1v4qalLBbN2DBAiAy0nR7VJS0vaB5gPR6qYYpJMRxM0sTERGVd5wIsZyR1/2yZWLCbt2AhITizQSdnS0teMpFT4mIyJ0xAJUzGRnST1trZ1Qq4KGHbDtWCGnm54gI1v4QEZF749dgOZKbKy186qi+OTk5Usdq1v4QEZG7YwAqR+R1v7y87H9uIaSAFRpa/MVSiYiIKhoGoHJCqy186Htp5eZK57a2qCoREZG7cXoAmjNnDuLi4uDt7Y34+Hjs3r270OPv3LmDxMREREVFQa1Wo06dOlhjtvLn5cuX8fzzz6NSpUrw8fFB48aNsXfvXkc+jVLLzpaaqBy1LEVODhAcbH1iRSIiInfj1K/DZcuWYdSoUZg3bx7i4+ORnJyMhIQEpKSkINzKNMYajQaPPvoowsPD8eOPPyImJgbnz59HcHCw4Zjbt2+jbdu26NSpE37//XdUrlwZJ0+eREhISBk+s+KR1/3y9HTMxIS5udKwetb+EBERSRRCWFtTvGzEx8ejVatWmD17NgBAr9cjNjYWw4YNw9ixYy2OnzdvHmbMmIHjx4/D09PT6jnHjh2L7du3Y9u2bSUuV0ZGBoKCgpCeno5AW8ajl1J2NnD+vNQ52RH9c27flobIV65s/3MTERGVF8X5/nZaE5hGo8G+ffvQpUuX+4VRKtGlSxfs2LHD6n1++eUXtGnTBomJiYiIiECjRo0wdepU6HQ6k2NatmyJ3r17Izw8HA888AA+//xzhz+f0rh7V6oFckT4ycuTapbKIMcRERG5DKcFoBs3bkCn0yEiIsJke0REBFJTU63e58yZM/jxxx+h0+mwZs0ajB8/HjNnzsT7779vcszcuXNRu3ZtrFu3Dq+99hqGDx+OJUuWFFiWvLw8ZGRkmFzKikYjzf3jqL4/2dnSiu9FzSxNRETkTlyqS6xer0d4eDgWLFgAlUqFFi1a4PLly5gxYwYmTpxoOKZly5aYOnUqAOCBBx7AkSNHMG/ePAwaNMjqeZOSkjB58uQyex7G5HW/QkPtf+78fKlWibU/REREppxWAxQWFgaVSoVr166ZbL927RoizRe4uicqKgp16tSByqitqH79+khNTYVGozEc06BBA5P71a9fHxcuXCiwLOPGjUN6errhcvHixZI+rWKxdd2vksrKkmp/HDW0noiIyFU5LQB5eXmhRYsW2Lhxo2GbXq/Hxo0b0aZNG6v3adu2LU6dOgW9Xm/YduLECURFRcHr3uyBbdu2RUpKisn9Tpw4gWrVqhVYFrVajcDAQJNLWZDX/XJEQNFqpRFlQUH2PzcREZGrc+o8QKNGjcLnn3+OJUuW4NixY3jttdeQlZWFIUOGAAAGDhyIcePGGY5/7bXXcOvWLYwYMQInTpzAb7/9hqlTpyIxMdFwzBtvvIGdO3di6tSpOHXqFL799lssWLDA5JjyIj1dWpPLEetyZWVJTV+s/SEiIrLk1D5Affv2xfXr1zFhwgSkpqaiWbNmWLt2raFj9IULF6A0SgexsbFYt24d3njjDTRp0gQxMTEYMWIE3n77bcMxrVq1wsqVKzFu3DhMmTIF1atXR3JyMgYMGFDmz68wubnS0heOWPdLp5Oa14KCHDOvEBERkatz6jxA5VVZzAN04wZw7RrgiPkZMzIAPz+gShUGICIich8uMQ+QO5PX/XLE0He9XqoBCg5m+CEiIioIA5ATZGU5bt2v7Gyp9sfPz/7nJiIiqigYgMqYI9f9EkKaWDEkxDEdq4mIiCoKfk2WsZwcx3V+lmt//P3tf24iIqKKhAGojOl0jln3SwhpRumQEMesKUZERFSRMABVELm50pw/rP0hIiIqGgNQBZGTI4388nCp1d2IiIicgwGoAtBoAC8vICDA2SUhIiJyDQxAFUBentT8dW85NCIiIioCA1AFkJ/P2h8iIqLiYABycfn50pxCXPSUiIjIdgxALi4vT5pRms1fREREtmMAcnEaDRAYyHW/iIiIioMByIVptdKkh2z+IiIiKh4GIBcmj/5Sq51dEiIiItfCAOTCNBpp9Bebv4iIiIqH8wa7KJ1OCj5s/iKisqDT6ZCfn+/sYpCb8/T0hMpOC14yALkoefSXt7ezS0JEFZkQAqmpqbhz546zi0IEAAgODkZkZCQUpWz+YAByUXl5QEQEoGQjJhE5kBx+wsPD4evrW+ovHaKSEkIgOzsbaWlpAICoqKhSnY8ByAXp9dJPNn8RkSPpdDpD+KlUqZKzi0MEn3tffGlpaQgPDy9VcxjrD1wQm7+IqCzIfX58fX2dXBKi++Tfx9L2SWMAckG5udLoLzv1AyMiKhSbvag8sdfvIwOQixFC+sk/yIiIiEqOAcjF5OVJEx+y+YuIXIlOr8OWc1vw3eHvsOXcFuj0OmcXqcwsXrwYwcHBzi4GmWEAcjG5uYCfH+DB7utE5CJWHFuBuE/i0GlJJzy34jl0WtIJcZ/EYcWxFQ57zMGDB0OhUGDatGkm21etWuWQJj2FQoFVq1ZZ3de3b1+cOHGi2OdMTU3FsGHDUKNGDajVasTGxqJ79+7YuHGj4Zi4uDgoFAqTS5UqVUzOk5CQAJVKhT179hS7DBUZA5ALEUK6+Ps7uyRERLZZcWwFnv3hWVzKuGSy/XLGZTz7w7MODUHe3t6YPn06bt++7bDHsIWPjw/Cw8OLdZ9z586hRYsW2LRpE2bMmIHDhw9j7dq16NSpExITE02OnTJlCq5evWq4HDhwwLDvwoUL+Pvvv/H666/jyy+/tMvzsSdnTq7JAORCNBrAy4vNX0TkGnR6HUasHQEBYbFP3jZy7UiHNYd16dIFkZGRSEpKKvS4n376CQ0bNoRarUZcXBxmzpxpsj8uLg5Tp07Fiy++iICAAFStWhULFiywuRzmTWCTJk1Cs2bN8PXXXyMuLg5BQUHo168f7t69azhm6NChUCgU2L17N3r16oU6deqgYcOGGDVqFHbu3Gly/oCAAERGRhoulStXNuxbtGgRnnzySbz22mv47rvvkJOTU2hZz58/j+7duyMkJAR+fn5o2LAh1qxZY9h/9OhRPPnkkwgMDERAQADatWuH06dPAwD0ej2mTJmCKlWqQK1Wo1mzZli7dq3hvufOnYNCocCyZcvQoUMHeHt7Y+nSpQCAL774AvXr14e3tzfq1auHzz77zObXt6QYgFxIXp7U+dnT09klISIq2rYL2yxqfowJCFzMuIhtF7Y55PFVKhWmTp2KWbNm4dIl6+XYt28f+vTpg379+uHw4cOYNGkSxo8fj8WLF5scN3PmTLRs2RIHDhzA0KFD8dprryElJaXEZTt9+jRWrVqF1atXY/Xq1di6dauhue7WrVtYu3YtEhMT4efnZ3FfW/sTCSGwaNEiPP/886hXrx5q1aqFH3/8sdD7JCYmIi8vD3/++ScOHz6M6dOnw/9es8Ply5fRvn17qNVqbNq0Cfv27cOLL74IrVYLAPjkk08wc+ZMfPTRR/jnn3+QkJCAp556CidPnjR5jLFjx2LEiBE4duwYEhISsHTpUkyYMAEffPABjh07hqlTp2L8+PFYsmSJTc+zpBiAXIhWKw1/JyJyBVfvXrXrcSXxzDPPoFmzZpg4caLV/R9//DE6d+6M8ePHo06dOhg8eDBef/11zJgxw+S4bt26YejQoahVqxbefvtthIWFYfPmzSUul16vx+LFi9GoUSO0a9cOL7zwgqFvz6lTpyCEQL169Ww619tvvw1/f3/D5dNPPwUA/PHHH8jOzkZCQgIA4Pnnn8fChQsLPdeFCxfQtm1bNG7cGDVq1MCTTz6J9u3bAwDmzJmDoKAgfP/992jZsiXq1KmDIUOGoG7dugCAjz76CG+//Tb69euHunXrYvr06WjWrBmSk5NNHmPkyJHo2bMnqlevjqioKEycOBEzZ840bOvZsyfeeOMNzJ8/3+bXsyQYgFwEm7+IyNVEBdi2VIGtx5XU9OnTsWTJEhw7dsxi37Fjx9C2bVuTbW3btsXJkyeh091vmmvSpInhukKhQGRkpGFJhpKIi4tDgNFftFFRUYbzCWHZZFiYMWPG4ODBg4bLwIEDAQBffvkl+vbtC497o2b69++P7du3G5qsrBk+fDjef/99tG3bFhMnTsQ///xj2Hfw4EG0a9cOnlaaITIyMnDlyhWrr6X5696yZUvD9aysLJw+fRovvfSSSYh7//33Cy2nPTAAuYi8PGnpCy8vZ5eEiMg27aq2Q5XAKlDA+qgrBRSIDYxFu6rtHFqO9u3bIyEhAePGjSvxOcy/9BUKBfTyukR2Pl/t2rWhUChw/Phxm84VFhaGWrVqGS7BwcG4desWVq5cic8++wweHh7w8PBATEwMtFptoZ2h//Of/+DMmTN44YUXcPjwYbRs2RKzZs0CcH8ZitIybtbLzMwEAHz++ecmIe7IkSMWfZ3sjQHIReTnA4GBzi4FEZHtVEoVPun6CQBYhCD5dnLXZKiUjp/Wftq0afj111+xY8cOk+3169fH9u3bTbZt374dderUKdU6U6URGhqKhIQEzJkzB1lZWRb779y5U+Q5li5diipVquDQoUMmwWLmzJlYvHixSe2WudjYWLz66qtYsWIFRo8ejc8//xyAVAu2bds2qyO3AgMDER0dbfW1bNCgQYGPFRERgejoaJw5c8YkxNWqVQvVq1cv8nmWBmeTcQH5+VLHZzZ/EZGr6Vm/J37s8yNGrB1h0iG6SmAVJHdNRs/6PcukHI0bN8aAAQMM/WNko0ePRqtWrfDee++hb9++2LFjB2bPnl2iUUhnz57FwYMHTbbVrl27ROWdM2cO2rZti9atW2PKlClo0qQJtFotNmzYgLlz51ptzjO2cOFCPPvss2jUqJHJ9tjYWIwbNw5r167FE088YXG/kSNH4vHHH0edOnVw+/ZtbN68GfXr1wcAvP7665g1axb69euHcePGISgoCDt37kTr1q1Rt25djBkzBhMnTkTNmjXRrFkzLFq0CAcPHjSM9CrI5MmTMXz4cAQFBaFr167Iy8vD3r17cfv2bYwaNaqYr5ztGIBcAJu/iMiV9azfE0/XfRrbLmzD1btXERUQhXZV25VJzY+xKVOmYNmyZSbbmjdvjh9++AETJkzAe++9h6ioKEyZMgWDBw8u9vmtfVlv21ayEW41atTA/v378cEHH2D06NG4evUqKleujBYtWmDu3LmF3nffvn04dOiQoebGWFBQEDp37oyFCxdaDUA6nQ6JiYm4dOkSAgMD0bVrV/zf//0fAKBSpUrYtGkTxowZgw4dOkClUqFZs2aGfj/Dhw9Heno6Ro8ejbS0NDRo0AC//PJLkSHwP//5D3x9fTFjxgyMGTMGfn5+aNy4MUaOHGnjq1UyClHc3lZuICMjA0FBQUhPT0egHduddDpg3Trg6FGgenUgPt62BU1v3QJiYoCQELsVhYioSLm5uTh79iyqV68Ob1ZBUzlR2O9lcb6/WQNURlasAEaMAIynooiKAqZMAbp1K/h+Wq207IWd+p4RERER2Am6TKxYATz7rGn4AYDUVOCVVwCjSTYt5OVJfX/UaseWkYiIyJ0wADmYTifV/FhraJS3TZwoHWeNRiNNfuiAtfuIiIjcFgOQg23bZlnzY0wI4MoVYNcuy306nRR8fH0dVz4iIiJ3xADkYFdtnOHd2oSibP4iIiJyDAYgB4uycYb38HDLbXl50uSHSr5LREREdsWvVgdr1w6oUqXgPjwKBRAdLQ2JNybPsM7RX0RERPbHAORgKhXwiTQTvEUIkm9Pnmw5H5Dc/MWpN4iIiOyPAagM9OwJ/PijNJmhsagoYMEC6/MA5eVJo7+ctBQNERFRhcYAVEZ69gTOnQN++w348ENg+XJg507r4UcI6cLRX0RUUeh0wJYtwHffST8LWYuz1ObNm4eAgABotVrDtszMTHh6eqJjx44mx27ZsgUKhQKnT592XIGoXGIAKkMqldQnqFs34KGHCq7dycuTRn6x+YuIKoIVK4C4OKBTJ+C556SfcXHSdkfo1KkTMjMzsXfvXsO2bdu2ITIyErt27UJubq5h++bNm1G1alXUrFnT4jwajcYxBbQjayuzlzVXeJ2sYQAqh3JzpeYvDy5UQkQurqCZ8C9flrY7IgTVrVsXUVFR2LJli2Hbli1b8PTTT6N69erYuXOnyfZOnToBAAYPHowePXrggw8+QHR0NOrWrQsA+Prrr9GyZUsEBAQgMjISzz33HNKM5i6Ra5F+++03NGnSBN7e3njwwQdx5MgRwzGLFy9GcHAwVq1ahdq1a8Pb2xsJCQm4ePGiSdl//vlnNG/eHN7e3qhRowYmT55sUpOlUCgwd+5cPPXUU/Dz88MHH3xg9TX47LPPDI8TERGBZ5991rBPr9fjww8/RK1ataBWq1G1alWT8xw+fBiPPPIIfHx8UKlSJbzyyivIzMw07C/odbp48SL69OmD4OBghIaG4umnn8a5c+eKfL+chQGonGHzFxFVFLbMhD9ypGOawzp16oTNmzcbbm/evBkdO3ZEhw4dDNtzcnKwa9cuQwACgI0bNyIlJQUbNmzA6tWrAUi1LO+99x4OHTqEVatW4dy5c1ZXix8zZgxmzpyJPXv2oHLlyujevbtJDU12djY++OADfPXVV9i+fTvu3LmDfv36GfZv27YNAwcOxIgRI/Dvv/9i/vz5WLx4sUXImTRpEp555hkcPnwYL774okU59u7di+HDh2PKlClISUnB2rVr0b59e8P+cePGYdq0aRg/fjz+/fdffPvtt4iIiAAAZGVlISEhASEhIdizZw+WL1+OP/74A6+//rrJY5i/Tvn5+UhISEBAQAC2bduG7du3w9/fH127di2/NUSCLKSnpwsAIj093e7nzsgQ4sgRIS5ftn45c0aIEyeE0Gjs/tBERMWSk5Mj/v33X5GTk1Oi+2/eLP9JV/hl82a7FlsIIcTnn38u/Pz8RH5+vsjIyBAeHh4iLS1NfPvtt6J9+/ZCCCE2btwoAIjz588LIYQYNGiQiIiIEHl5eYWee8+ePQKAuHv37r3nuVkAEN9//73hmJs3bwofHx+xbNkyIYQQixYtEgDEzp07DcccO3ZMABC7du0SQgjRuXNnMXXqVJPH+vrrr0VUVJThNgAxcuTIQsv3008/icDAQJGRkWGxLyMjQ6jVavH5559bve+CBQtESEiIyMzMNGz77bffhFKpFKmpqUII66/T119/LerWrSv0er1hW15envDx8RHr1q0rtLzFVdjvZXG+v1kDVM7k5gJ+foCnp7NLQkRUOrbOhG/rccXRsWNHZGVlYc+ePdi2bRvq1KmDypUro0OHDoZ+QFu2bEGNGjVQtWpVw/0aN24MLy8vk3Pt27cP3bt3R9WqVREQEIAOHToAAC5cuGByXJs2bQzXQ0NDUbduXRw7dsywzcPDA61atTLcrlevHoKDgw3HHDp0CFOmTIG/v7/h8vLLL+Pq1avIzs423K9ly5aFPvdHH30U1apVQ40aNfDCCy9g6dKlhvsfO3YMeXl56Ny5s9X7Hjt2DE2bNoWfn59hW9u2baHX65GSklLg63To0CGcOnUKAQEBhrKHhoYiNze33HYwZy+TckanA/z9nV0KIqLSs3UmfFuPK45atWqhSpUq2Lx5M27fvm0ILdHR0YiNjcXff/+NzZs345FHHjG5n/EXP3C/SSghIQFLly5F5cqVceHCBSQkJNi9aSczMxOTJ09Gz549LfZ5G42KMS+juYCAAOzfvx9btmzB+vXrMWHCBEyaNAl79uyBj51m1zUvQ2ZmJlq0aIGlS5daHFu5cmW7PKa9sQaoHNFoAC8vjv4ioorBlpnwY2Ol4xyhU6dO2LJlC7Zs2WIy/L19+/b4/fffsXv3bpP+P9YcP34cN2/exLRp09CuXTvUq1fPpAO0MePO1bdv38aJEydQv359wzatVmsyMi0lJQV37twxHNO8eXOkpKSgVq1aFhdlMddE8vDwQJcuXfDhhx/in3/+wblz57Bp0ybUrl0bPj4+2Lhxo9X71a9fH4cOHUJWVpZh2/bt26FUKg2dna1p3rw5Tp48ifDwcIuyBwUFFavsZYUBqBzJy5OWvjCrfSUickm2zISfnOy4CV87deqEv/76CwcPHjTUAAFAhw4dMH/+fGg0miIDUNWqVeHl5YVZs2bhzJkz+OWXX/Dee+9ZPXbKlCnYuHEjjhw5gsGDByMsLAw9evQw7Pf09MSwYcOwa9cu7Nu3D4MHD8aDDz6I1q1bAwAmTJiAr776CpMnT8bRo0dx7NgxfP/99/jf//5XrOe9evVqfPrppzh48CDOnz+Pr776Cnq9HnXr1oW3tzfefvttvPXWW/jqq69w+vRp7Ny5EwsXLgQADBgwAN7e3hg0aBCOHDmCzZs3Y9iwYXjhhRcMHaWtGTBgAMLCwvD0009j27ZtOHv2LLZs2YLhw4fjkvkQwHKiXASgOXPmIC4uDt7e3oiPj8fu3bsLPf7OnTtITExEVFQU1Go16tSpgzVr1lg9dtq0aVAoFBg5cqQDSm5f+fnS4qdERBVFQTPhV6kibbfS2mM3nTp1Qk5ODmrVqmXy5d2hQwfcvXvXMFy+MJUrV8bixYuxfPlyNGjQANOmTcNHH31k9dhp06ZhxIgRaNGiBVJTU/Hrr7+a9JPx9fXF22+/jeeeew5t27aFv78/li1bZtifkJCA1atXY/369WjVqhUefPBB/N///R+qVatWrOcdHByMFStW4JFHHkH9+vUxb948fPfdd2jYsCEAYPz48Rg9ejQmTJiA+vXro2/fvoZaLV9fX6xbtw63bt1Cq1at8Oyzz6Jz586YPXt2oY/p6+uLP//8E1WrVkXPnj1Rv359vPTSS8jNzUVgOf1iUwhhbYBi2Vm2bBkGDhyIefPmIT4+HsnJyVi+fDlSUlIQbmWJdI1Gg7Zt2yI8PBzvvPMOYmJicP78eQQHB6Np06Ymx+7Zswd9+vRBYGAgOnXqhOTkZJvKlJGRgaCgIKSnp9v9jbt7F7hwAQgJMd2eny91gI6LkyZBJCJyttzcXJw9exbVq1c36YNSEjodsG2b1OE5Kkpq9qooS/3Icwndvn0bwcHBVo9ZvHgxRo4ciTt37pRp2Sqiwn4vi/P97fRO0B9//DFefvllDBkyBIA0hflvv/2GL7/8EmPHjrU4/ssvv8StW7fw999/w/PeUKm4uDiL4zIzMzFgwAB8/vnneP/99x36HOyBzV9EVJGpVIDZKhRETuXUJjCNRoN9+/ahS5cuhm1KpRJdunTBjh07rN7nl19+QZs2bZCYmIiIiAg0atQIU6dOhc5sJq3ExEQ88cQTJucuzzQaafbngjoLEhERkf04tQboxo0b0Ol0Fh2rIiIicPz4cav3OXPmDDZt2oQBAwZgzZo1OHXqFIYOHYr8/HxMnDgRAPD9999j//792LNnj03lyMvLQ15enuF2RkZGCZ9RyWi10rIXdhqdSEREZahjx44oqjfJ4MGDrc4eTc7j9Caw4tLr9QgPD8eCBQugUqnQokULXL58GTNmzMDEiRNx8eJFjBgxAhs2bLC5zTopKQmTJ092cMkLlpcnDX1n3x8iIqKy4dQmsLCwMKhUKly7ds1k+7Vr1xAZGWn1PlFRUahTpw5URr3n6tevj9TUVEOTWlpaGpo3bw4PDw94eHhg69at+PTTT+Hh4WHRVAZI66Kkp6cbLuaL0zmaRiON/mLzFxERUdlwagDy8vJCixYtTCZk0uv12Lhxo8mU4sbatm2LU6dOQa/XG7adOHECUVFR8PLyQufOnXH48GEcPHjQcGnZsiUGDBiAgwcPmgQnmVqtRmBgoMmlrOh0gFLJ5i8iIqKy5PQmsFGjRmHQoEFo2bIlWrdujeTkZGRlZRlGhQ0cOBAxMTFISkoCALz22muYPXs2RowYgWHDhuHkyZOYOnUqhg8fDkCaArxRo0Ymj+Hn54dKlSpZbC8P8vKkpi82fxEREZUdpwegvn374vr165gwYQJSU1PRrFkzrF271tAx+sKFCyZTgMfGxmLdunV444030KRJE8TExGDEiBF4++23nfUUSiUvD4iIkGqBiIiIqGw4fSLE8qisJkLU64H0dGnywyLWtiMiKnP2nAiRyF7sNREi6x2cSB79xf4/REQVm0KhwKpVq5xdDDLCAOREeXnS5Ids/iKiCuvCBWD//oIvFy447KFTU1MxYsQI1KpVC97e3oiIiEDbtm0xd+5cZGdn2/WxBg8ebLLwqbmrV6/i8ccfL9Y5hRBYsGAB4uPj4e/vj+DgYLRs2RLJycmG8k+aNAkKhcLi8scffxjO891330GlUiExMbFEz62icnofIHclhHTx9XV2SYiIHOTCBaB+faCwsOHrCxw7BlStateHPnPmDNq2bYvg4GBMnToVjRs3hlqtxuHDh7FgwQLExMTgqaeesutjFqagqV0K88ILL2DFihX43//+h9mzZ6Ny5co4dOgQkpOTERcXZwhcDRs2NAk8ABAaGmq4vnDhQrz11luYP38+Zs6cWa6aMzUajcmCsWWJdQ9Okpsrjfxi8xcRVVg3bkjh55tvgH37LC/ffCPtv3HD7g89dOhQeHh4YO/evejTpw/q16+PGjVq4Omnn8Zvv/2G7t27G469c+cO/vOf/6By5coIDAzEI488gkOHDhn2T5o0Cc2aNcPXX3+NuLg4BAUFoV+/frh7967N5TFuAjt37hwUCgVWrFiBTp06wdfXF02bNjVZAuqHH37A0qVL8d133+Gdd95Bq1atEBcXh6effhqbNm1Cp06dDMd6eHggMjLS5CKHirNnz+Lvv//G2LFjUadOHaxYsaLQcgohMGnSJFStWhVqtRrR0dGGUdaAtHLC22+/jdjYWKjVatSqVQsLFy407N+6dStat24NtVqNqKgojB07Flqt1rC/Y8eOeP311zFy5EiEhYUhISEBAHDkyBE8/vjj8Pf3R0REBF544QXccMDvhTEGICeRm78qymrIREQFql8faN7c8lK/vkMe7ubNm1i/fj0SExPhV8AIE4XRzLO9e/dGWloafv/9d+zbtw/NmzdH586dcevWLcMxp0+fxqpVq7B69WqsXr0aW7duxbRp00pVznfffRdvvvkmDh48iDp16qB///6GsLB06VLUrVsXTz/9tNWyBwUF2fQYixYtwhNPPIGgoCA8//zzJmHFmp9++gn/93//h/nz5+PkyZNYtWoVGjdubNg/cOBAfPfdd/j0009x7NgxzJ8/H/7+/gCAy5cvo1u3bmjVqhUOHTqEuXPnYuHChRYLki9ZsgReXl7Yvn075s2bhzt37uCRRx7BAw88gL1792Lt2rW4du0a+vTpY9NzLDFBFtLT0wUAkZ6ebvdzZ2QIcfiwdMnIsPvpiYjsJicnR/z7778iJyenZCfYt09q7d+3r2T7S2jnzp0CgFixYoXJ9kqVKgk/Pz/h5+cn3nrrLSGEENu2bROBgYEiNzfX5NiaNWuK+fPnCyGEmDhxovD19RUZRv9ojxkzRsTHxxtuDxo0SDz99NMFlgmAWLlypRBCiLNnzwoA4osvvjDsP3r0qAAgjh07JoQQon79+uKpp54q8rlOnDhRKJVKw/Py8/MTrVq1EkIIodPpRGxsrFi1apUQQojr168LLy8vcebMmQLPN3PmTFGnTh2h0Wgs9qWkpAgAYsOGDVbv+84774i6desKvV5v2DZnzhzh7+8vdDqdEEKIDh06iAceeMDkfu+995547LHHTLZdvHhRABApKSkWj1PY72Vxvr9ZA+QEQkjNX+WoGZaIqMLbvXs3Dh48iIYNGxoWwD506BAyMzNRqVIl+Pv7Gy5nz57F6dOnDfeNi4tDQECA4XZUVBTS0tJKVZ4mTZqYnA+A4ZyiGDPU1K1b12T1g59++gkAsGHDBmRlZaFbt24ApOWnHn30UXz55ZcFnqt3797IyclBjRo18PLLL2PlypWGWil5NYUOHTpYve+xY8fQpk0bk9q1tm3bIjMzE5cuXTJsa9Gihcn9Dh06hM2bN5u8/vXq1QMAk/fA3tgJ2kn8/QFPT2eXgoio4qlVqxYUCgVSUlJMtteoUQMA4GPU+TIzMxNRUVHYsmWLxXmCg4MN1z3N/sFWKBQmSzKVhPE55dAgn7NOnTo4fvy4Tefx8vJCrVq1LLYvXLgQt27dMnm+er0e//zzDyZPnmwyybAsNjYWKSkp+OOPP7BhwwYMHToUM2bMwNatW03OUxrmzZKZmZno3r07pk+fbnGsHAwdgTVATuDhwYkPiYgcpVKlSnj00Ucxe/ZsZGVlFXps8+bNkZqaCg8PD9SqVcvkEhYWVkYltvTcc8/hxIkT+Pnnny32CSGQnp5e6P1v3ryJn3/+Gd9//71J7dCBAwdw+/ZtrF+/vsD7+vj4oHv37vj000+xZcsW7NixA4cPH0bjxo2h1+uxdetWq/erX78+duzYYVJ7tX37dgQEBKBKlSoFPl7z5s1x9OhRxMXFWbwHBfXhsgcGICdg8xcRuZVjx6zPAXTsmMMe8rPPPoNWq0XLli2xbNkyHDt2DCkpKfjmm29w/Phxw8LYXbp0QZs2bdCjRw+sX78e586dw99//413330Xe/fuLdZjpqenm4SNgwcP4uLFiyUqf58+fdC3b1/0798fU6dOxd69e3H+/HmsXr0aXbp0webNmwu9/9dff41KlSqhT58+aNSokeHStGlTdOvWrcDO0IsXL8bChQtx5MgRnDlzBt988w18fHxQrVo1xMXFYdCgQXjxxRexatUqnD17Flu2bMEPP/wAQBp5d/HiRQwbNgzHjx/Hzz//jIkTJ2LUqFFWa5tkiYmJuHXrFvr37489e/bg9OnTWLduHYYMGQKdTlei188WbAIrY0qlVPvjpGkPiIjKTliYNM/P888XfIyvr3ScndWsWRMHDhzA1KlTMW7cOFy6dAlqtRoNGjTAm2++iaFDhwKQmp7WrFmDd999F0OGDMH169cRGRmJ9u3bG9aktNWWLVvwwAMPmGx76aWX8MUXXxS7/AqFAt9++y0WLFiAL7/8Eh988AE8PDxQu3ZtDBw40DB8vCBffvklnnnmGZP+OLJevXoZhpmb13IFBwdj2rRpGDVqFHQ6HRo3boxff/0VlSpVAgDMnTsX77zzDoYOHYqbN2+iatWqeOeddwAAMTExWLNmDcaMGYOmTZsiNDQUL730Ev73v/8VWtbo6Ghs374db7/9Nh577DHk5eWhWrVq6Nq1a6HBqbS4FpgVjlwLDJA6QVv5nSQiKlfsshbYhQuFz/MTFmb3SRCpYrPXWmCsAXIChh8ichtVqzLgULnEPkBERETkdhiAiIiIyO0wABEREZHbYQAiIiIit8MAREREheJgYSpP7PX7yABERERWyUs1ZGdnO7kkRPfJv4/my5MUF4fBExGRVSqVCsHBwYYFOn19fa1OrEdUFoQQyM7ORlpaGoKDgw2zeZcUAxARERUoMjISAEq98jmRvQQHBxt+L0uDAYiIiAqkUCgQFRWF8PBw5OfnO7s45OY8PT1LXfMjYwAiIqIiqVQqu33xEJUH7ARNREREbocBiIiIiNwOAxARERG5HfYBskKeZCkjI8PJJSEiIiJbyd/btkyWyABkxd27dwEAsbGxTi4JERERFdfdu3cRFBRU6DEKwTnOLej1eqSkpKBBgwa4ePEiAgMDnV0kKkJGRgZiY2P5frkAvleug++Va+H7JdX83L17F9HR0VAqC+/lwxogK5RKJWJiYgAAgYGBbvuL5Ir4frkOvleug++Va3H396uomh8ZO0ETERGR22EAIiIiIrfDAFQAtVqNiRMnQq1WO7soZAO+X66D75Xr4HvlWvh+FQ87QRMREZHbYQ0QERERuR0GICIiInI7DEBERETkdhiAiIiIyO0wABVgzpw5iIuLg7e3N+Lj47F7925nF4nMTJo0CQqFwuRSr149ZxeL7vnzzz/RvXt3REdHQ6FQYNWqVSb7hRCYMGECoqKi4OPjgy5duuDkyZPOKaybK+q9Gjx4sMVnrWvXrs4prJtLSkpCq1atEBAQgPDwcPTo0QMpKSkmx+Tm5iIxMRGVKlWCv78/evXqhWvXrjmpxOUXA5AVy5Ytw6hRozBx4kTs378fTZs2RUJCAtLS0pxdNDLTsGFDXL161XD566+/nF0kuicrKwtNmzbFnDlzrO7/8MMP8emnn2LevHnYtWsX/Pz8kJCQgNzc3DIuKRX1XgFA165dTT5r3333XRmWkGRbt25FYmIidu7ciQ0bNiA/Px+PPfYYsrKyDMe88cYb+PXXX7F8+XJs3boVV65cQc+ePZ1Y6nJKkIXWrVuLxMREw22dTieio6NFUlKSE0tF5iZOnCiaNm3q7GKQDQCIlStXGm7r9XoRGRkpZsyYYdh2584doVarxXfffeeEEpLM/L0SQohBgwaJp59+2inlocKlpaUJAGLr1q1CCOlz5OnpKZYvX2445tixYwKA2LFjh7OKWS6xBsiMRqPBvn370KVLF8M2pVKJLl26YMeOHU4sGVlz8uRJREdHo0aNGhgwYAAuXLjg7CKRDc6ePYvU1FSTz1lQUBDi4+P5OSuntmzZgvDwcNStWxevvfYabt686ewiEYD09HQAQGhoKABg3759yM/PN/ls1atXD1WrVuVnywwDkJkbN25Ap9MhIiLCZHtERARSU1OdVCqyJj4+HosXL8batWsxd+5cnD17Fu3atcPdu3edXTQqgvxZ4ufMNXTt2hVfffUVNm7ciOnTp2Pr1q14/PHHodPpnF00t6bX6zFy5Ei0bdsWjRo1AiB9try8vBAcHGxyLD9blrgaPLmsxx9/3HC9SZMmiI+PR7Vq1fDDDz/gpZdecmLJiCqWfv36Ga43btwYTZo0Qc2aNbFlyxZ07tzZiSVzb4mJiThy5Aj7PpYQa4DMhIWFQaVSWfSYv3btGiIjI51UKrJFcHAw6tSpg1OnTjm7KFQE+bPEz5lrqlGjBsLCwvhZc6LXX38dq1evxubNm1GlShXD9sjISGg0Gty5c8fkeH62LDEAmfHy8kKLFi2wceNGwza9Xo+NGzeiTZs2TiwZFSUzMxOnT59GVFSUs4tCRahevToiIyNNPmcZGRnYtWsXP2cu4NKlS7h58yY/a04ghMDrr7+OlStXYtOmTahevbrJ/hYtWsDT09Pks5WSkoILFy7ws2WGTWBWjBo1CoMGDULLli3RunVrJCcnIysrC0OGDHF20cjIm2++ie7du6NatWq4cuUKJk6cCJVKhf79+zu7aAQpkBrXEJw9exYHDx5EaGgoqlatipEjR+L9999H7dq1Ub16dYwfPx7R0dHo0aOH8wrtpgp7r0JDQzF58mT06tULkZGROH36NN566y3UqlULCQkJTiy1e0pMTMS3336Ln3/+GQEBAYZ+PUFBQfDx8UFQUBBeeukljBo1CqGhoQgMDMSwYcPQpk0bPPjgg04ufTnj7GFo5dWsWbNE1apVhZeXl2jdurXYuXOns4tEZvr27SuioqKEl5eXiImJEX379hWnTp1ydrHons2bNwsAFpdBgwYJIaSh8OPHjxcRERFCrVaLzp07i5SUFOcW2k0V9l5lZ2eLxx57TFSuXFl4enqKatWqiZdfflmkpqY6u9huydr7BEAsWrTIcExOTo4YOnSoCAkJEb6+vuKZZ54RV69edV6hyymFEEKUfewiIiIich72ASIiIiK3wwBEREREbocBiIiIiNwOAxARERG5HQYgIiIicjsMQEREROR2GICIiIjI7TAAEbmZc+fOQaFQ4ODBg84uisHx48fx4IMPwtvbG82aNXP443Xs2BEjR450+OMURaFQYNWqVQXuF0LglVdeQWhoaLl7z4hcHQMQURkbPHgwFAoFpk2bZrJ91apVUCgUTiqVc02cOBF+fn5ISUkxWcPI3a1duxaLFy/G6tWrcfXqVTRq1Mgu5x08eDCXHCG3xwBE5ATe3t6YPn06bt++7eyi2I1GoynxfU+fPo2HH34Y1apVQ6VKlRz+eK5CXtz3oYceQmRkJDw8ytfyjTqdDnq93tnFICoRBiAiJ+jSpQsiIyORlJRU4DGTJk2yaA5KTk5GXFyc4bb8l/zUqVMRERGB4OBgTJkyBVqtFmPGjEFoaCiqVKmCRYsWWZz/+PHjeOihh+Dt7Y1GjRph69atJvuPHDmCxx9/HP7+/oiIiMALL7yAGzduGPZ37NgRr7/+OkaOHImwsLACF8bU6/WYMmUKqlSpArVajWbNmmHt2rWG/QqFAvv27cOUKVOgUCgwadIkq+cp6PGKKmdWVhYGDhwIf39/REVFYebMmRbnttYUFRwcjMWLFxtuX7p0Cf3790doaCj8/PzQsmVL7Nq1y7D/559/RvPmzeHt7Y0aNWpg8uTJ0Gq1hv0nT55E+/bt4e3tjQYNGmDDhg1Wn6ds8ODBGDZsGC5cuACFQmF43/V6PZKSklC9enX4+PigadOm+PHHHw330+l0eOmllwz769ati08++cSwf9KkSViyZAl+/vlnKBQKKBQKbNmyBVu2bIFCocCdO3cMxx48eBAKhQLnzp0DACxevBjBwcH45Zdf0KBBA6jValy4cAF5eXl48803ERMTAz8/P8THx2PLli2G85w/fx7du3dHSEgI/Pz80LBhQ6xZs6bQ50/kaAxARE6gUqkwdepUzJo1C5cuXSrVuTZt2oQrV67gzz//xMcff4yJEyfiySefREhICHbt2oVXX30V//3vfy0eZ8yYMRg9ejQOHDiANm3aoHv37rh58yYA4M6dO3jkkUfwwAMPYO/evVi7di2uXbuGPn36mJxjyZIl8PLywvbt2zFv3jyr5fvkk08wc+ZMfPTRR/jnn3+QkJCAp556CidPngQAXL16FQ0bNsTo0aNx9epVvPnmmwU+V/PHs6WcY8aMwdatW/Hzzz9j/fr12LJlC/bv31+s1zgzMxMdOnTA5cuX8csvv+DQoUN46623DLUf27Ztw8CBAzFixAj8+++/mD9/PhYvXowPPvgAgBRaevbsCS8vL+zatQvz5s3D22+/XehjfvLJJ4bgePXqVezZswcAkJSUhK+++grz5s3D0aNH8cYbb+D55583BFi9Xo8qVapg+fLl+PfffzFhwgS88847+OGHHwAAb775Jvr06YOuXbvi6tWruHr1Kh566CGbX4vs7GxMnz4dX3zxBY4ePYrw8HC8/vrr2LFjB77//nv8888/6N27N7p27Wp4jxMTE5GXl4c///wThw8fxvTp0+Hv71+s94DI7py8GCuR2xk0aJB4+umnhRBCPPjgg+LFF18UQgixcuVKYfyRnDhxomjatKnJff/v//5PVKtWzeRc1apVEzqdzrCtbt26ol27dobbWq1W+Pn5ie+++04IIcTZs2cFADFt2jTDMfn5+aJKlSpi+vTpQggh3nvvPfHYY4+ZPPbFixcFAMOK7R06dBAPPPBAkc83OjpafPDBBybbWrVqJYYOHWq43bRpUzFx4sRCz2Pt8Yoq5927d4WXl5f44YcfDPtv3rwpfHx8xIgRIwzbAIiVK1eanCcoKMiwwvb8+fNFQECAuHnzptWyde7cWUydOtVk29dffy2ioqKEEEKsW7dOeHh4iMuXLxv2//7771Yf15j5+52bmyt8fX3F33//bXLcSy+9JPr371/geRITE0WvXr0Mt41/B2XyivC3b982bDtw4IAAIM6ePSuEEGLRokUCgDh48KDhmPPnzwuVSmXy3ISQXpNx48YJIYRo3LixmDRpUoHlI3KG8tWgTORmpk+fjkceeaTQWo+iNGzYEErl/crciIgIk86yKpUKlSpVQlpamsn92rRpY7ju4eGBli1b4tixYwCAQ4cOYfPmzVb/Sj99+jTq1KkDAGjRokWhZcvIyMCVK1fQtm1bk+1t27bFoUOHbHyG95k/XlHlzMnJgUajQXx8vGF7aGgo6tatW6zHPXjwIB544AGEhoZa3X/o0CFs377dUOMDSE1Rubm5yM7OxrFjxxAbG4vo6GjDfuPX31anTp1CdnY2Hn30UZPtGo0GDzzwgOH2nDlz8OWXX+LChQuG18Beo+u8vLzQpEkTw+3Dhw9Dp9MZfidkeXl5hv5cw4cPx2uvvYb169ejS5cu6NWrl8k5iJyBAYjIidq3b4+EhASMGzcOgwcPNtmnVCohhDDZlp+fb3EOT09Pk9sKhcLqtuJ0Vs3MzET37t0xffp0i31RUVGG635+fjaf0x7MH6+ocp46dcqm8yoUikJfax8fn0Lvn5mZicmTJ6Nnz54W+7y9vW0qgy0yMzMBAL/99htiYmJM9qnVagDA999/jzfffBMzZ85EmzZtEBAQgBkzZpj0V7JGDtHGr4O13zcfHx+T0YqZmZlQqVTYt28fVCqVybFyMP3Pf/6DhIQE/Pbbb1i/fj2SkpIwc+ZMDBs2zNanTmR3DEBETjZt2jQ0a9bMolaicuXKSE1NhRDC8IVjz3lgdu7cifbt2wMAtFot9u3bh9dffx0A0Lx5c/z000+Ii4sr1cijwMBAREdHY/v27ejQoYNh+/bt29G6devSPQEbylmzZk14enpi165dqFq1KgDg9u3bOHHihEl5KleujKtXrxpunzx5EtnZ2YbbTZo0wRdffIFbt25ZrQVq3rw5UlJSUKtWLavlrF+/Pi5evIirV68aAuTOnTuL/XyNOx4bl9/Y9u3b8dBDD2Ho0KGGbadPnzY5xsvLCzqdzmRb5cqVAUh9skJCQgDY9vv2wAMPQKfTIS0tDe3atSvwuNjYWLz66qt49dVXMW7cOHz++ecMQORU7ARN5GSNGzfGgAED8Omnn5ps79ixI65fv44PP/wQp0+fxpw5c/D777/b7XHnzJmDlStX4vjx40hMTMTt27fx4osvApA6rd66dQv9+/fHnj17cPr0aaxbtw5Dhgyx+OIsypgxYzB9+nQsW7YMKSkpGDt2LA4ePIgRI0aU+jkUVU5/f3+89NJLGDNmDDZt2oQjR45g8ODBJk2GAPDII49g9uzZOHDgAPbu3YtXX33VpBatf//+iIyMRI8ePbB9+3acOXMGP/30E3bs2AEAmDBhAr766itMnjwZR48exbFjx/D999/jf//7HwBp1F+dOnUwaNAgHDp0CNu2bcO7775b7OcbEBCAN998E2+88QaWLFmC06dPY//+/Zg1axaWLFkCAKhduzb27t2LdevW4cSJExg/fryhA7UsLi4O//zzD1JSUnDjxg3k5+ejVq1aiI2NxaRJk3Dy5En89ttvVkfMmatTpw4GDBiAgQMHYsWKFTh79ix2796NpKQk/PbbbwCAkSNHYt26dTh79iz279+PzZs3o379+sV+/kT2xABEVA5MmTLFoomqfv36+OyzzzBnzhw0bdoUu3fvLlVfIXPTpk3DtGnT0LRpU/z111/45ZdfEBYWBgCGWhudTofHHnsMjRs3xsiRIxEcHGwRHooyfPhwjBo1CqNHj0bjxo2xdu1a/PLLL6hdu3apn4Mt5ZwxYwbatWuH7t27o0uXLnj44Yct+hLNnDkTsbGxaNeuHZ577jm8+eab8PX1Nez38vLC+vXrER4ejm7duqFx48aYNm2aocknISEBq1evxvr169GqVSs8+OCD+L//+z9Uq1YNgNS8tHLlSuTk5KB169b4z3/+Y9JfqDjee+89jB8/HklJSahfvz66du2K3377DdWrVwcA/Pe//0XPnj3Rt29fxMfH4+bNmya1QQDw8ssvo27dumjZsiUqV66M7du3w9PTE9999x2OHz+OJk2aYPr06Xj//fdtKtOiRYswcOBAjB49GnXr1kWPHj2wZ88eQ62bTqdDYmKiobx16tTBZ599VqLnT2QvCmHe8E1ERERUwbEGiIiIiNwOAxARERG5HQYgIiIicjsMQEREROR2GICIiIjI7TAAERERkdthACIiIiK3wwBEREREbocBiIiIiNwOAxARERG5HQYgIiIicjsMQEREROR2/h9rNS4VVJYLCwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_linear_result(results, 'High variance, 1000 features, linear model',23, 'HighVar_1000Feat_lin.png',results2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "458302d2",
   "metadata": {},
   "source": [
    "# Quadratic relationship, plot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9874fcde",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "dc7bae3b",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# in 0 NonLinCFA, in 1 wrapper\n",
    "def plot_quadratic(results,title,savename,results2,wrapper_stop):\n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(7):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[0][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "    \n",
    "    print(f'NonLinCFA:\\n number of red features:{count_means[1:-1]} +- {count_plusminus[1:-1]},\\n R2 score: {R2_means[1:-1]} +- {R2_means_plusminus[1:-1]}\\n')\n",
    "    \n",
    "    count_means_sorted = [x for x in sorted(count_means[1:-1])]\n",
    "    R2_means = [x for _,x in sorted(zip(count_means[1:-1],R2_means[1:-1]))]\n",
    "    R2_means_plusminus = [x for _,x in sorted(zip(count_means[1:-1],R2_means_plusminus[1:-1]))]\n",
    "    \n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"R2 score\")\n",
    "    ax.plot([round(x) for x in count_means_sorted],R2_means,'go',label='NonLinCFA score')\n",
    "    ax.fill_between([round(x) for x in count_means_sorted], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='g', alpha=.1)\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[1][i::50]))\n",
    "        wrapper_std.append(np.std(results[1][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "\n",
    "    ax.plot(range(1,wrapper_stop+1),wrapper_means[:wrapper_stop],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,wrapper_stop+1), ([i-j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), ([i+j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results2[0][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    \n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "\n",
    "    \n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "    \n",
    "    print(f'SQUARED FEATURES:\\nLinCFA mean {np.mean(results[2])}, confidence interval [{np.mean(results[2])-1.96*np.std(results[2])/np.sqrt(10)},{np.mean(results[2])+1.96*np.std(results[2])/np.sqrt(10)}],')\n",
    "    print(f'with number of reduced features: {np.mean(results[3])}, confidence interval [{np.mean(results[3])-1.96*np.std(results[3])/np.sqrt(10)},{np.mean(results[3])+1.96*np.std(results[3])/np.sqrt(10)}]')\n",
    "\n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    \n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "    \n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1a44fb6",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Low noise, 100 variables, quadratic dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "03995f70",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_lowNoise_100var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    \n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_lowNoise_100var_onlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "941a348c",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.68, 0.7, 0.72, 0.74, 0.76]\n"
     ]
    }
   ],
   "source": [
    "print(f'NonLinCFA epsilon: {[results[0][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[0][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d170b505",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 7.1, 10.3, 12.9, 12.4] +- [0.0, 0.852092952675939, 1.6410245580124632, 1.5789865103920298, 1.3914619649850297],\n",
      " R2 score: [0.7435796504110674, 0.7469435190462297, 0.7483113277255279, 0.7478827217848412, 0.7475181023920491] +- [0.009613153691609828, 0.009719453041330275, 0.009980777648615158, 0.009921865757770227, 0.009696986760829673]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.4588728244122411, 0.5223470777812327, 0.5646268658280296, 0.5911582610061302, 0.6139461193772527, 0.633126609370802, 0.651206783516486, 0.6609334203086007, 0.6694324178672105, 0.674824077892848, 0.6810727971064592, 0.6865674037826511, 0.6908673425311546, 0.6960644836032694, 0.6994004249102292, 0.7019649158242594, 0.7049070110434679, 0.7077232182124751, 0.7117070206655665, 0.7146729578041104, 0.7165480358673113, 0.7180030302107419, 0.719319041540922, 0.7216983702263332, 0.7232596652903627, 0.7247329222494996, 0.7263887028401307, 0.7285888979030128, 0.7300831310567156, 0.7312239192768402, 0.7331974819762085, 0.7338023716555, 0.7346754059137537, 0.7359211334400829, 0.736694587666199, 0.7388225313774626, 0.7394431825738886, 0.7397825898095913, 0.7405708662661434, 0.7411979065970671, 0.7416902201764433, 0.7422288761946885, 0.7427923292040309, 0.743221203684307, 0.7439941267759522, 0.7446451092048063, 0.7448990863144258, 0.7454168203220553, 0.7461528954287907, 0.7461098727796005] +- [0.010763949748541193, 0.011219275582565224, 0.010852341985137674, 0.010036390005056064, 0.008926595223703022, 0.009076937860279851, 0.009458436422263603, 0.008756860722419185, 0.009401403828624703, 0.010636717712816142, 0.01130818601686541, 0.012585043394847093, 0.013139021809547844, 0.012450415229619606, 0.01287164115581747, 0.012705748472468845, 0.012779200641954347, 0.0127604625203422, 0.012871435880905129, 0.01252679450779362, 0.012133057244713726, 0.011498475443196272, 0.010894877554701017, 0.011368259922264493, 0.011111423891708933, 0.01094486232550529, 0.011310752062920551, 0.011274320735814667, 0.01117548243296749, 0.011402342689337316, 0.011319220306618999, 0.011399880994360045, 0.011353796346407688, 0.011747714536043835, 0.011543893448332284, 0.011512532334856782, 0.011784316337655289, 0.012216554104048032, 0.011844826072527508, 0.01207080852023962, 0.012116875478518111, 0.012316075084858052, 0.011805210254389697, 0.011699222834067232, 0.011768682856588, 0.011612004399250097, 0.011647655336281171, 0.011449162023787422, 0.011344411295157909, 0.011464997180436627]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[17.6, 14.5, 7.2, 4.9, 3.0] +- [0.0, 0.0, 0.852092952675939, 1.6410245580124632, 1.5789865103920298, 1.3914619649850297, 1.3914619649850297],\n",
      " R2 score: [0.7462293348566434, 0.7459997559948034, 0.7453698408958116, 0.7455734110573703, 0.7455088278937119] +- [0.010461654277760428, 0.010125744631908948, 0.00934479290343028, 0.009496845765980522, 0.009656862241569916]\n",
      "\n",
      "SQUARED FEATURES:\n",
      "LinCFA mean 0.7502973318651838, confidence interval [0.7422736485575209,0.7583210151728467],\n",
      "with number of reduced features: 34.3, confidence interval [33.29867208168353,35.30132791831647]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_quadratic(results,\"Low variance, 100 features, squared model:\\nsquared input, mean aggregation\",'LowVar_100Feat_quadr.png',results2,18)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "318db81b",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### High noise, 1000 variables, quadratic dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "090589ec",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_highNoise_1000var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    \n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_highNoise_1000var_onlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "496bbad0",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.68, 0.7, 0.72, 0.74, 0.76]\n"
     ]
    }
   ],
   "source": [
    "print(f'NonLinCFA epsilon: {[results[0][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[0][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7e4065d3",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 5.9, 11.6, 13.2, 13.7] +- [0.0, 0.7565701553722562, 0.690187220977033, 0.6072837886853228, 0.6228977444171715],\n",
      " R2 score: [0.4043600294140658, 0.3873536998825163, 0.3735210038242839, 0.36277098329044266, 0.36496033201194733] +- [0.013070969124468677, 0.01841887174306367, 0.017922417169732162, 0.020044624454245933, 0.019976669055864083]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.24682387520675286, 0.2663863841926727, 0.28242638483621507, 0.2908549685697016, 0.29951079060876334, 0.3043620111862685, 0.30657915144761166, 0.30732995095040727, 0.3062713287400328, 0.305364729531867, 0.3060518187972073, 0.30698081837724756, 0.3074702817241948, 0.3062614512088471, 0.3070664881062214, 0.3062544994565551, 0.3060496585868935, 0.30713468799494825, 0.3050265647413321, 0.30536591168334504, 0.305025483413231, 0.30446346787810435, 0.30419846986843363, 0.30513943790726933, 0.30545914797713025, 0.30554195124994404, 0.30612798137442143, 0.3057718599119472, 0.3058740575979249, 0.30467001645238284, 0.3053580895353435, 0.3048656505169346, 0.30491241943007635, 0.3057651412348791, 0.30467867348358935, 0.30369658536350785, 0.30134039262615303, 0.3008327396235875, 0.30021517163773803, 0.2999949207256128, 0.2991095664800785, 0.29876866242434447, 0.29765022987177825, 0.2973414311415395, 0.29785823499408354, 0.2973324920196715, 0.29620876115372485, 0.29492605798416255, 0.2942396145689213, 0.29365728497606114] +- [0.011494199045477264, 0.01203594319738569, 0.012151437989402896, 0.01401196528215874, 0.014303034942622637, 0.017336882019862634, 0.017613721124728546, 0.017410895169906956, 0.016952085541063876, 0.01842417666936419, 0.01890827979343883, 0.019136020277592324, 0.01944225661691717, 0.020128504480057364, 0.02009250772414834, 0.01944260794876157, 0.01918184596749585, 0.018862751739827554, 0.020011351868491966, 0.020333984469568116, 0.02065052309044267, 0.019681245748292248, 0.019929986430083237, 0.01934549177450598, 0.019572558517793295, 0.0199927122954906, 0.019120837029353573, 0.018513819512021513, 0.01969357508638251, 0.019244666703201324, 0.019342234397212364, 0.018806731641310005, 0.018338180840184626, 0.01841866001115172, 0.017581985376004552, 0.01815989088488773, 0.01818766993485748, 0.017999266738036705, 0.01869315353666232, 0.01940465823127852, 0.018974940206740627, 0.018438658267239293, 0.018048271419010352, 0.01823847383678278, 0.017813759774417924, 0.01865618209614833, 0.018728998105558566, 0.019348241851337385, 0.020196247119276683, 0.02083376739242741]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[5.5, 3.5, 2.0, 1.2, 1.1] +- [0.0, 0.0, 0.7565701553722562, 0.690187220977033, 0.6072837886853228, 0.6228977444171715, 0.690187220977033],\n",
      " R2 score: [0.3772021135137055, 0.37609719592618324, 0.39869736646614956, 0.40379146269789457, 0.4043484508078589] +- [0.015648958214555683, 0.014814067069408934, 0.01274007157178324, 0.012846193354291775, 0.01306699540082782]\n",
      "\n",
      "SQUARED FEATURES:\n",
      "LinCFA mean 0.36186557832585, confidence interval [0.3432706835864724,0.38046047306522757],\n",
      "with number of reduced features: 120.5, confidence interval [118.78570014291549,122.21429985708451]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_quadratic(results,\"High variance, 1000 features, squared model:\\nsquared input, mean aggregation\",'HIghVar_1000Feat_quadr.png',results2,14)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b462f65b",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Repeat considering linear input and squared sums to aggregate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82848e85",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9424815e",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def plot_squaredAggr(results,title,savename,results2,wrapper_stop):\n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(7):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[5][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std[1:-1]]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std[1:-1]]\n",
    "    \n",
    "    print(f'NonLinCFA:\\n number of red features:{count_means[1:-1]} +- {count_plusminus},\\n R2 score: {R2_means[1:-1]} +- {R2_means_plusminus}\\n')\n",
    "    \n",
    "    count_means_sorted = [x for x in sorted(count_means[1:-1])]\n",
    "    R2_means = [x for _,x in sorted(zip(count_means[1:-1],R2_means[1:-1]))]\n",
    "    R2_means_plusminus = [x for _,x in sorted(zip(count_means[1:-1],R2_means_plusminus))]\n",
    "    \n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"R2 score\")\n",
    "    ax.plot([round(x) for x in count_means_sorted],R2_means,'go',label='NonLinCFA score')\n",
    "    ax.fill_between([round(x) for x in count_means_sorted], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='g', alpha=.1)\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[1][i::50]))\n",
    "        wrapper_std.append(np.std(results[1][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "\n",
    "    ax.plot(range(1,wrapper_stop+1),wrapper_means[:wrapper_stop],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,wrapper_stop+1), ([i-j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), ([i+j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results2[1][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    \n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "\n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "\n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "    \n",
    "    \n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5b66fb4",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Low noise, 100 variables, quadratic dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b416e731",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_lowNoise_100var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    \n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_lowNoise_100var_onlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9ca628f0",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.68, 0.7, 0.72, 0.74, 0.76]\n"
     ]
    }
   ],
   "source": [
    "print(f'NonLinCFA epsilon: {[results[5][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[1][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8be3fed7",
   "metadata": {
    "hidden": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 2.8, 3.8, 4.1, 4.1] +- [0.0, 0.4638206549950098, 0.3718838528358014, 0.43386449497510166, 0.43386449497510166],\n",
      " R2 score: [0.7435796504110674, 0.7455923801513561, 0.7514748171087419, 0.7529332012932627, 0.7539048076870722] +- [0.009613153691609823, 0.009332800682799089, 0.009579663806688241, 0.009468732130713182, 0.009728952173565836]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.4588728244122411, 0.5223470777812327, 0.5646268658280296, 0.5911582610061302, 0.6139461193772527, 0.633126609370802, 0.651206783516486, 0.6609334203086007, 0.6694324178672105, 0.674824077892848, 0.6810727971064592, 0.6865674037826511, 0.6908673425311546, 0.6960644836032694, 0.6994004249102292, 0.7019649158242594, 0.7049070110434679, 0.7077232182124751, 0.7117070206655665, 0.7146729578041104, 0.7165480358673113, 0.7180030302107419, 0.719319041540922, 0.7216983702263332, 0.7232596652903627, 0.7247329222494996, 0.7263887028401307, 0.7285888979030128, 0.7300831310567156, 0.7312239192768402, 0.7331974819762085, 0.7338023716555, 0.7346754059137537, 0.7359211334400829, 0.736694587666199, 0.7388225313774626, 0.7394431825738886, 0.7397825898095913, 0.7405708662661434, 0.7411979065970671, 0.7416902201764433, 0.7422288761946885, 0.7427923292040309, 0.743221203684307, 0.7439941267759522, 0.7446451092048063, 0.7448990863144258, 0.7454168203220553, 0.7461528954287907, 0.7461098727796005] +- [0.010763949748541193, 0.011219275582565224, 0.010852341985137674, 0.010036390005056064, 0.008926595223703022, 0.009076937860279851, 0.009458436422263603, 0.008756860722419185, 0.009401403828624703, 0.010636717712816142, 0.01130818601686541, 0.012585043394847093, 0.013139021809547844, 0.012450415229619606, 0.01287164115581747, 0.012705748472468845, 0.012779200641954347, 0.0127604625203422, 0.012871435880905129, 0.01252679450779362, 0.012133057244713726, 0.011498475443196272, 0.010894877554701017, 0.011368259922264493, 0.011111423891708933, 0.01094486232550529, 0.011310752062920551, 0.011274320735814667, 0.01117548243296749, 0.011402342689337316, 0.011319220306618999, 0.011399880994360045, 0.011353796346407688, 0.011747714536043835, 0.011543893448332284, 0.011512532334856782, 0.011784316337655289, 0.012216554104048032, 0.011844826072527508, 0.01207080852023962, 0.012116875478518111, 0.012316075084858052, 0.011805210254389697, 0.011699222834067232, 0.011768682856588, 0.011612004399250097, 0.011647655336281171, 0.011449162023787422, 0.011344411295157909, 0.011464997180436627]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[15.4, 10.9, 6.3, 3.3, 2.1] +- [0.0, 0.4638206549950098, 0.3718838528358014, 0.43386449497510166, 0.43386449497510166],\n",
      " R2 score: [0.7457892802208258, 0.7450984086923766, 0.7460187713691055, 0.7442656756326018, 0.7435965077654504] +- [0.009051215225571549, 0.00917167163472808, 0.009434669218915151, 0.009602893757632921, 0.009672700781425233]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_squaredAggr(results,\"Low variance, 100 features, squared model:\\nlinear input, squared sum aggregation\",'LowVar_100Feat_squaredAggr.png',results2,15)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c3ada72",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### High noise, 1000 variables, quadratic dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "8500588d",
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA epsilon: [0.1, 0.01, 0.001, 0.0001, 1e-05, 1e-06, 1e-07]\n",
      "GenLinCFA epsilon: [0.68, 0.7, 0.72, 0.74, 0.76]\n"
     ]
    }
   ],
   "source": [
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_highNoise_1000var.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    \n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/squared_highNoise_1000var_onlyGen.pkl', 'rb') as f:\n",
    "    results2 = pickle.load(f)\n",
    "\n",
    "print(f'NonLinCFA epsilon: {[results[5][0][i][0] for i in range(7)]}')\n",
    "print(f'GenLinCFA epsilon: {[results2[1][0][i][0] for i in range(5)]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "060dfbac",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NonLinCFA:\n",
      " number of red features:[1.0, 6.3, 15.5, 17.7, 18.0] +- [0.0, 0.7864461837913641, 1.5495160534825059, 1.9213999063183071, 1.8386429778507842],\n",
      " R2 score: [0.4043600294140659, 0.40066945402735515, 0.3843090728533701, 0.3785464497064927, 0.3772973315928946] +- [0.013070969124468663, 0.013480973372648168, 0.014034228333703535, 0.015643215713622927, 0.01662247144619444]\n",
      "\n",
      "Wrapper:\n",
      " Mean values [0.24682387520675286, 0.2663863841926727, 0.28242638483621507, 0.2908549685697016, 0.29951079060876334, 0.3043620111862685, 0.30657915144761166, 0.30732995095040727, 0.3062713287400328, 0.305364729531867, 0.3060518187972073, 0.30698081837724756, 0.3074702817241948, 0.3062614512088471, 0.3070664881062214, 0.3062544994565551, 0.3060496585868935, 0.30713468799494825, 0.3050265647413321, 0.30536591168334504, 0.305025483413231, 0.30446346787810435, 0.30419846986843363, 0.30513943790726933, 0.30545914797713025, 0.30554195124994404, 0.30612798137442143, 0.3057718599119472, 0.3058740575979249, 0.30467001645238284, 0.3053580895353435, 0.3048656505169346, 0.30491241943007635, 0.3057651412348791, 0.30467867348358935, 0.30369658536350785, 0.30134039262615303, 0.3008327396235875, 0.30021517163773803, 0.2999949207256128, 0.2991095664800785, 0.29876866242434447, 0.29765022987177825, 0.2973414311415395, 0.29785823499408354, 0.2973324920196715, 0.29620876115372485, 0.29492605798416255, 0.2942396145689213, 0.29365728497606114] +- [0.011494199045477264, 0.01203594319738569, 0.012151437989402896, 0.01401196528215874, 0.014303034942622637, 0.017336882019862634, 0.017613721124728546, 0.017410895169906956, 0.016952085541063876, 0.01842417666936419, 0.01890827979343883, 0.019136020277592324, 0.01944225661691717, 0.020128504480057364, 0.02009250772414834, 0.01944260794876157, 0.01918184596749585, 0.018862751739827554, 0.020011351868491966, 0.020333984469568116, 0.02065052309044267, 0.019681245748292248, 0.019929986430083237, 0.01934549177450598, 0.019572558517793295, 0.0199927122954906, 0.019120837029353573, 0.018513819512021513, 0.01969357508638251, 0.019244666703201324, 0.019342234397212364, 0.018806731641310005, 0.018338180840184626, 0.01841866001115172, 0.017581985376004552, 0.01815989088488773, 0.01818766993485748, 0.017999266738036705, 0.01869315353666232, 0.01940465823127852, 0.018974940206740627, 0.018438658267239293, 0.018048271419010352, 0.01823847383678278, 0.017813759774417924, 0.01865618209614833, 0.018728998105558566, 0.019348241851337385, 0.020196247119276683, 0.02083376739242741]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[12.7, 7.2, 4.1, 2.4, 1.7] +- [0.0, 0.7864461837913641, 1.5495160534825059, 1.9213999063183071, 1.8386429778507842],\n",
      " R2 score: [0.39956801488902294, 0.4020114076760901, 0.4037567069771029, 0.40435312644306415, 0.40424714837847375] +- [0.012287046790761263, 0.01262300861170105, 0.012997949803957253, 0.01306475332140161, 0.01306490849065005]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_squaredAggr(results,\"High variance, 1000 features, squared model:\\nlinear input, squared sum aggregation\",'HighVar_1000Feat_squaredAggr.png',results2,18)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "170b7bda",
   "metadata": {},
   "source": [
    "# classification"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b880e0a9",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "b3aff554",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# in 0 ho wrapper, in 1 GenLinCFA\n",
    "def plot_class(result,title,wrapper_stop,savename):\n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"Accuracy score\")\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[0][i::50]))\n",
    "        wrapper_std.append(np.std(results[0][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    \n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "\n",
    "    ax.plot(range(1,(wrapper_stop+1)),wrapper_means[:wrapper_stop],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,(wrapper_stop+1)), ([i-j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), ([i+j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[1][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "\n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "\n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "    \n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2cb93c16",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Low noise, 100 variables, linear dependence, class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "31d53e93",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/class_retry.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "a1a6a6f0",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GenLinCFA epsilon: [0.71, 0.72, 0.73, 0.75, 0.77]\n"
     ]
    }
   ],
   "source": [
    "print(f'GenLinCFA epsilon: {[results[1][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "48bf7dc8",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.866, 0.8708, 0.8769, 0.8774, 0.8769, 0.8775000000000001, 0.8786000000000002, 0.8794000000000001, 0.8815000000000002, 0.8814, 0.8833, 0.8831, 0.883, 0.884, 0.8840999999999999, 0.8842000000000001, 0.8841000000000001, 0.8850999999999999, 0.8851000000000001, 0.8851000000000001, 0.8836, 0.8844999999999998, 0.8835000000000001, 0.8834, 0.8833, 0.8834, 0.8840000000000001, 0.8840999999999999, 0.8847000000000002, 0.8833, 0.8829, 0.8834, 0.8831999999999999, 0.8834, 0.8836999999999999, 0.884, 0.8849, 0.8847000000000002, 0.8857000000000002, 0.8863, 0.8854, 0.8852, 0.8853, 0.8867, 0.8862, 0.8871, 0.8865000000000001, 0.8868, 0.8867, 0.8872] +- [0.007083169911840325, 0.005859056033184873, 0.007429666910434148, 0.006224041645104896, 0.006746785782874693, 0.005762866647771753, 0.006607264365832509, 0.005978481947785747, 0.005769528923577735, 0.0057089521280178955, 0.006429576782339572, 0.006176021891152918, 0.006646683383462767, 0.006278122012194416, 0.007074758398701688, 0.007011213418517517, 0.0073255245818985605, 0.0070257183547307155, 0.006926596740102609, 0.00738298465391877, 0.0075819326296136455, 0.006897696427068973, 0.006969722519584268, 0.006796425413406674, 0.006990636137004992, 0.007255707160573673, 0.007570777503004568, 0.00811189043318511, 0.008721669656665524, 0.008014698596952982, 0.008459619755048104, 0.0076173175330952365, 0.007141504155288301, 0.0072662886262520624, 0.007217751561255076, 0.007354571911403143, 0.007683849972507277, 0.007565955432065411, 0.007565955432065411, 0.0077515460135382085, 0.007510662713769011, 0.007211095450761977, 0.007281340645787703, 0.00717504575595167, 0.00666284695907088, 0.007351698470421654, 0.006665441020667729, 0.006928537623481602, 0.006635403348704588, 0.006440024347780067]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[25.2, 19.4, 15.6, 4.3, 1.0] +- [1.5874790077352205, 1.6906065183832695, 1.3914619649850297, 1.2098138699816594, 0.0],\n",
      " R2 score: [0.8927999999999999, 0.8946999999999999, 0.8956000000000002, 0.8958, 0.8975] +- [0.006404133165386247, 0.006618011906909814, 0.006489938551326975, 0.006939617972194153, 0.005369469992466673]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class(results,\"Low variance, 100 features, linear model\",25,'LowVar_100Feat_lin_class.png')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fde1f697",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### High noise, 100 variables, linear dependence, class\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "f67779fa",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/class_retry_highNoise.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ab90b5b9",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GenLinCFA epsilon: [0.71, 0.72, 0.73, 0.75, 0.77]\n"
     ]
    }
   ],
   "source": [
    "print(f'GenLinCFA epsilon: {[results[1][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "b6c4d33f",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.8271, 0.829, 0.8310000000000001, 0.8310000000000001, 0.8318, 0.8339000000000001, 0.8335999999999999, 0.8343, 0.8344999999999999, 0.8343, 0.835, 0.8343, 0.8347, 0.8345, 0.834, 0.8335999999999999, 0.8337, 0.834, 0.834, 0.8337, 0.834, 0.8335999999999999, 0.8333999999999999, 0.8332, 0.8329000000000001, 0.8334999999999999, 0.8337999999999999, 0.8335999999999999, 0.8346, 0.8344999999999999, 0.8346, 0.8342, 0.8341, 0.8338999999999999, 0.8337999999999999, 0.8339000000000001, 0.8334999999999999, 0.8338000000000001, 0.8336, 0.8332, 0.8327, 0.8333, 0.8333999999999999, 0.8337999999999999, 0.8338999999999999, 0.8335000000000001, 0.8337999999999999, 0.8346, 0.8352, 0.8347] +- [0.007377779503346514, 0.008434834438209193, 0.00848931186846141, 0.008743445545092616, 0.008188954121253816, 0.007915342974249432, 0.008181444762387623, 0.008114731394199062, 0.008191064888035003, 0.0080290653005191, 0.008264602107784732, 0.007859819947047124, 0.007499657698855318, 0.007469888888062517, 0.007489149484420768, 0.007423200819053723, 0.007291884914067135, 0.00728107684343462, 0.007328408285569237, 0.007751546013538192, 0.007296888103842613, 0.007127504359872383, 0.007122112495601278, 0.007033096160298098, 0.007053004919890518, 0.007319228921136423, 0.007316866651784755, 0.007276854705159355, 0.007371267896366258, 0.007176116498496929, 0.0070570888048826375, 0.007264173566208329, 0.007166474056326432, 0.006887664219457844, 0.00664552733799206, 0.007351698470421637, 0.007516031399615083, 0.007700580341766447, 0.007687598740829273, 0.007839019938742331, 0.008019490357871868, 0.008052952849731572, 0.007707561378283008, 0.007779990334184218, 0.007429666910434131, 0.0077475802674125205, 0.008127739759613362, 0.008237598096532742, 0.008497453359683703, 0.008861497932065428]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[20.0, 11.1, 5.7, 1.0, 1.0] +- [3.5388721367124862, 2.0640577511300404, 0.8787272614412277, 0.0, 0.0],\n",
      " R2 score: [0.8462, 0.8452999999999999, 0.8428999999999999, 0.852, 0.852] +- [0.006668610170042936, 0.00749453358121772, 0.006408031086066926, 0.004840842901809564, 0.004840842901809564]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkgAAAHHCAYAAABEEKc/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACXL0lEQVR4nOzdd1xT1/sH8M9NIIMtGxTBVdyjqIi7Llr9Ugetrfp11erXOupq694/RTssHc7W0aHV2qodWltFUesu7qrUjQsQZG+S8/vj9IYkBCSQEALP+/WKkpuTe8/N4D6c85xzBMYYAyGEEEII0ZBYugKEEEIIIVUNBUiEEEIIIXooQCKEEEII0UMBEiGEEEKIHgqQCCGEEEL0UIBECCGEEKKHAiRCCCGEED0UIBFCCCGE6KEAiRBCCCFEDwVIpMoICAjAqFGjyv3c//znP6atkBncvXsXgiBgy5Ytlq4KMWD//v1o3bo1FAoFBEFAamqqpatkFQx9rhctWgRBECxXKSs3atQoBAQElOu53bt3R/fu3U1an5qIAiRiFlu2bIEgCPjrr78MPt69e3c0b968kmtFzO2PP/7AmDFj0Lx5c0il0lJ/wavVarz//vuoV68eFAoFWrZsie+++85g2WvXruHFF1+Eg4MDXF1dMXz4cDx58qRC+9SXnJyMwYMHQ6lUYvXq1fjmm29gb29fpuca4+rVq1i0aBHu3r1r8n0TQkzHxtIVIEQUGxsLiaR6x+z+/v7IycmBra2tpatiFtu2bcOOHTvw/PPPw9fXt9Syc+fOxYoVKzB27Fi0a9cOP/30E4YOHQpBEPD6669ryj148ABdu3aFs7Mzli9fjszMTHz44Ye4fPkyzpw5A5lMZvQ+DTl79iwyMjKwdOlS9OrVq2IvRCmuXr2KxYsXo3v37uVuIbAG8+bNw6xZsyxdDULKrXpfjYhVkcvl1TZwKCwsRH5+PgRBgEKhgFQqtXSVzGL58uVIT0/H8ePH0apVqxLLPXz4EB999BEmTpyIDRs2YOzYsfjll1/QpUsXvPvuu1CpVDr7zMrKwqFDh/D2229jzpw5+P7773Hx4kWdLh1j9mlIYmIiAMDFxaVCr4GlZGVlWboKOmxsbKBQKCxdjRKJ30lCSkIBEqkyDOUgXbp0Cd26dYNSqUSdOnXwf//3f9i8eTMEQTDYRfHnn3+iffv2UCgUqF+/Pr7++utSj1lQUABXV1eMHj262GPp6elQKBR45513AAD5+flYsGABgoKC4OzsDHt7e3Tp0gWHDx/WeZ6Yj/Hhhx8iMjISDRo0gFwux9WrVw3maly6dAmjRo1C/fr1oVAo4O3tjTfeeAPJyck6+xVzOm7evIlRo0bBxcUFzs7OGD16NLKzs4vV/9tvv0X79u1hZ2eHWrVqoWvXrvjjjz90yvz222/o0qUL7O3t4ejoiH79+uHvv/8u9TUrja+vb5mC3J9++gkFBQWYMGGCZpsgCHjrrbfw4MEDnDx5UrP9xx9/xH/+8x/UrVtXs61Xr1547rnn8P3335drn/q6d++OkSNHAgDatWsHQRB0PounT5/Giy++CGdnZ9jZ2aFbt244fvy4zj7u3buHCRMmIDAwEEqlEm5ubnj11Vd1PqdbtmzBq6++CgB44YUXIAgCBEFAdHS0pr6LFi0qVj/974bYhX3kyBFMmDABnp6eqFOnjubxsryv8fHxGD16NOrUqQO5XA4fHx/079/fZF1/hnKQBEHApEmTsGfPHjRv3hxyuRzNmjXD/v37iz3/4cOHeOONN+Dl5aUpt2nTJp0ypvhOlkSs686dO9G0aVMolUqEhITg8uXLAID169ejYcOGUCgU6N69u8HXbefOnQgKCoJSqYS7uzv++9//4uHDh8XKia+HQqFA8+bNsXv3boN1UqvViIyMRLNmzaBQKODl5YX//e9/SElJKfE8SPlRFxsxq7S0NCQlJRXbXlBQ8MznPnz4UHMRmT17Nuzt7fHll19CLpcbLH/z5k288sorGDNmDEaOHIlNmzZh1KhRCAoKQrNmzQw+x9bWFgMHDsSuXbuwfv16ne6aPXv2IC8vT9M1k56eji+//BJDhgzB2LFjkZGRgY0bNyI0NBRnzpxB69atdfa9efNm5ObmYty4cZDL5XB1dYVarS5WhwMHDuD27dsYPXo0vL298ffff2PDhg34+++/cerUqWIXmcGDB6NevXqIiIjAuXPn8OWXX8LT0xMrV67UlFm8eDEWLVqEjh07YsmSJZDJZDh9+jQOHTqEPn36AAC++eYbjBw5EqGhoVi5ciWys7Oxdu1adO7cGefPnzdr98/58+dhb2+PJk2a6Gxv37695vHOnTvj4cOHSExMRNu2bYvto3379ti3b5/R+zRk7ty5CAwMxIYNG7BkyRLUq1cPDRo0AAAcOnQIL730EoKCgrBw4UJIJBJs3rwZPXr0wLFjxzT7P3v2LE6cOIHXX38dderUwd27d7F27Vp0794dV69ehZ2dHbp27Yq3334bn376KebMmaOpq36dy2rChAnw8PDAggULNC1IZX1fw8PD8ffff2Py5MkICAhAYmIiDhw4gLi4OLO+93/++Sd27dqFCRMmwNHREZ9++inCw8MRFxcHNzc3AEBCQgI6dOigCVI8PDzw22+/YcyYMUhPT8fUqVMBmOY7WZpjx47h559/xsSJEwEAERER+M9//oP33nsPa9aswYQJE5CSkoL3338fb7zxBg4dOqR57pYtWzB69Gi0a9cOERERSEhIwCeffILjx4/j/PnzmpbKP/74A+Hh4WjatCkiIiKQnJysCVz1/e9//9Ps9+2338adO3fw+eef4/z58zh+/Hi1bYG3GEaIGWzevJkBKPXWrFkznef4+/uzkSNHau5PnjyZCYLAzp8/r9mWnJzMXF1dGQB2584dnecCYEePHtVsS0xMZHK5nM2YMaPUuv7+++8MAPvll190tvft25fVr19fc7+wsJDl5eXplElJSWFeXl7sjTfe0Gy7c+cOA8CcnJxYYmKiTnnxsc2bN2u2ZWdnF6vTd999V+x8Fi5cyADoHIsxxgYOHMjc3Nw092/cuMEkEgkbOHAgU6lUOmXVajVjjLGMjAzm4uLCxo4dq/N4fHw8c3Z2Lra9PPr168f8/f1LfEz7tRVlZWUxAGzWrFmMMcbOnj3LALCvv/66WNl3332XAWC5ublG7bMk4mf27Nmzmm1qtZo1atSIhYaGal47xvh7Vq9ePda7d2+dbfpOnjxZrP47d+5kANjhw4eLlQfAFi5cWGy7/ndDrGvnzp1ZYWGhZntZ39eUlBQGgH3wwQclvyBGMPS5Fj+v2gAwmUzGbt68qdl28eJFBoB99tlnmm1jxoxhPj4+LCkpSef5r7/+OnN2dta81qb4TpYEAJPL5Tq/Z9avX88AMG9vb5aenq7ZPnv2bJ3fSfn5+czT05M1b96c5eTkaMr9+uuvDABbsGCBZlvr1q2Zj48PS01N1Wz7448/GACd78+xY8cYALZ161adeu7fv7/Y9m7durFu3bqV6TxJyaiLjZjV6tWrceDAgWK3li1bPvO5+/fvR0hIiM5fga6urhg2bJjB8k2bNkWXLl009z08PBAYGIjbt2+XepwePXrA3d0dO3bs0GxLSUnBgQMH8Nprr2m2SaVSTQuTWq3G06dPUVhYiLZt2+LcuXPF9hseHg4PD49nnqdSqdT8nJubi6SkJHTo0AEADO53/PjxOve7dOmC5ORkpKenA+AtX2q1GgsWLCiW9C62Rh04cACpqakYMmQIkpKSNDepVIrg4OBiXRSmlpOTY7AlUMxZycnJ0fm/rGXLUs4YFy5cwI0bNzB06FAkJydrXqesrCz07NkTR48e1bQKar+PBQUFSE5ORsOGDeHi4mLwfTSFsWPH6uSzlfV9VSqVkMlkiI6OrvTumV69emla5wCgZcuWcHJy0nxPGWP48ccfERYWBsaYznmEhoYiLS1N83qa6zsp6tmzp05rWnBwsGY/jo6OxbaL5/DXX38hMTEREyZM0MnD6tevHxo3boy9e/cCAB4/fowLFy5g5MiRcHZ21pTr3bs3mjZtqlOXnTt3wtnZGb1799Z5TYKCguDg4GD272xNRF1sxKzat29vsHukVq1aBrvetN27dw8hISHFtjds2NBgee0cFe3jPOsCYGNjg/DwcGzbtg15eXmQy+XYtWsXCgoKdAIkAPjqq6/w0Ucf4fr16zrdhPXq1Su2X0PbDHn69CkWL16M7du3axKFRWlpacXK659nrVq1APCgzsnJCbdu3YJEIin2C1bbjRs3APDg0BAnJ6cy1b28lEol8vLyim3Pzc3VPK79f1nLlqWcMcTXScxPMiQtLQ21atVCTk4OIiIisHnzZjx8+BCMMZ0y5qD/GSvr+yqXy7Fy5UrMmDEDXl5e6NChA/7zn/9gxIgR8Pb2NktdRc/6nj558gSpqanYsGEDNmzYYHAf2t8Tc3wnS6qrGMT4+fkZ3C6ew7179wAAgYGBxfbZuHFj/PnnnzrlGjVqVKxcYGCgTpB348YNpKWlwdPT02Bd9X93kIqjAIlUGyWNDNO+UJXk9ddfx/r16/Hbb79hwIAB+P7779G4cWOdkVjffvstRo0ahQEDBuDdd9+Fp6cnpFIpIiIicOvWrWL7LOsFefDgwThx4gTeffddtG7dGg4ODlCr1XjxxRcN5ixV5DxF4n6/+eYbgxdEGxvz/mrw8fHB4cOHwRjTybF6/PgxAGimCPDx8dHZru3x48dwdXXVtBqVdZ/GEF+nDz74oFg+i8jBwQEAMHnyZGzevBlTp05FSEgInJ2dNdMLGHofjVHSCDz9z5gx7+vUqVMRFhaGPXv24Pfff8f8+fMRERGBQ4cOoU2bNhWqb2me9fkVz+G///1viYGp2AJtru/ks+pqiu+gsdRqNTw9PbF161aDjxvTMkbKhgIkUmX5+/vj5s2bxbYb2lZRXbt2hY+PD3bs2IHOnTvj0KFDmDt3rk6ZH374AfXr18euXbt0LsALFy4s93FTUlIQFRWFxYsXY8GCBZrtYktAeTRo0ABqtRpXr14t8aIudnF4enqadc6fkrRu3Rpffvklrl27ptPSdfr0ac3jAFC7dm14eHgYnHBUPwm3rPs0hvg6OTk5PfN1+uGHHzBy5Eh89NFHmm25ubnFZuMubXbpWrVqFSufn59vMEAsrb5lfV8bNGiAGTNmYMaMGbhx4wZat26Njz76CN9++22ZjmcOHh4ecHR0hEqlKtNrburvpCn4+/sD4HO76bfmxcbGah4X/zf0fY+NjdW536BBAxw8eBCdOnUqV2soMR7lIJEqKzQ0FCdPnsSFCxc0254+fVriX1AVIZFI8Morr+CXX37BN998g8LCwmLda+Jfjdp/JZ4+fbrU4ePPYmifABAZGVnufQ4YMAASiQRLliwp1nIhHic0NBROTk5Yvny5wRGFhmapNqX+/fvD1tYWa9as0anbunXrULt2bXTs2FGzPTw8HL/++ivu37+v2RYVFYV//vlHM2Te2H2WVVBQEBo0aIAPP/wQmZmZxR7Xfp2kUmmx9/Gzzz4r1vojzs5taBmTBg0a4OjRozrbNmzY8Mw5nERlfV+zs7M1XY/ax3Z0dDTYTVmZpFIpwsPD8eOPP+LKlSvFHtd/zQHTfidNoW3btvD09MS6det0Xs/ffvsN165dQ79+/QDwVs/WrVvjq6++0umGPXDgQLEpCAYPHgyVSoWlS5cWO15hYSEti2MG1IJEqqz33nsP3377LXr37o3JkydrhvnXrVsXT58+Nfk6T6+99ho+++wzLFy4EC1atCg29Po///kPdu3ahYEDB6Jfv364c+cO1q1bh6ZNmxq8eJaFk5MTunbtivfffx8FBQWoXbs2/vjjD9y5c6fc59GwYUPMnTsXS5cuRZcuXTBo0CDI5XKcPXsWvr6+iIiIgJOTE9auXYvhw4fj+eefx+uvvw4PDw/ExcVh79696NSpEz7//HMAfA6ZevXqYeTIkc9cQ+7SpUv4+eefAfCWvrS0NPzf//0fAKBVq1YICwsDANSpUwdTp07FBx98gIKCArRr1w579uzBsWPHsHXrVp0ujDlz5mDnzp144YUXMGXKFGRmZuKDDz5AixYtdOavMmafZSWRSPDll1/ipZdeQrNmzTB69GjUrl0bDx8+xOHDh+Hk5IRffvkFAP98fPPNN3B2dkbTpk1x8uRJHDx4UDN0XdS6dWtIpVKsXLkSaWlpkMvl6NGjBzw9PfHmm29i/PjxCA8PR+/evXHx4kX8/vvvcHd3L1N9y/q+/vPPP+jZsycGDx6Mpk2bwsbGBrt370ZCQoLOjOPikPLNmzeXe53E8lixYgUOHz6M4OBgjB07Fk2bNsXTp09x7tw5HDx4EE+fPgVgnu+kKdja2mLlypUYPXo0unXrhiFDhmiG+QcEBGDatGmashEREejXrx86d+6MN954A0+fPsVnn32GZs2a6ZxDt27d8L///Q8RERG4cOEC+vTpA1tbW9y4cQM7d+7EJ598gldeecUSp1t9Vf7AOVITGBoyra1bt27PHObPGGPnz59nXbp0YXK5nNWpU4dFRESwTz/9lAFg8fHxOs/t16+fweOUdbirWq1mfn5+DAD7v//7P4OPL1++nPn7+zO5XM7atGnDfv31VzZy5Eid4bjikGJDQ6gNDYd+8OABGzhwIHNxcWHOzs7s1VdfZY8ePSo25FscNv3kyROdfYqvtfZwZMYY27RpE2vTpg2Ty+WsVq1arFu3buzAgQM6ZQ4fPsxCQ0OZs7MzUygUrEGDBmzUqFHsr7/+0pS5fPlymYbJa9fF0E3/vVWpVJrXUyaTsWbNmrFvv/3W4H6vXLnC+vTpw+zs7JiLiwsbNmyYzvtfnn2WVHdDn9nz58+zQYMGMTc3NyaXy5m/vz8bPHgwi4qK0pRJSUlho0ePZu7u7szBwYGFhoay69evG/xcf/HFF6x+/fpMKpXqDPlXqVRs5syZzN3dndnZ2bHQ0FB28+bNEof5l/T9etb7mpSUxCZOnMgaN27M7O3tmbOzMwsODmbff/+9zn4+++wzBoDt37+/1NfOmGH+EydOLPZ8Q69RQkICmzhxIvPz82O2trbM29ub9ezZk23YsEFTxhTfyZIYqmtJ+zl8+DADwHbu3KmzfceOHZrvoKurKxs2bBh78OBBsWP9+OOPrEmTJkwul7OmTZuyXbt2FTsH0YYNG1hQUBBTKpXM0dGRtWjRgr333nvs0aNHmjI0zN80BMbMmFVGiBlMnToV69evR2ZmZrVdsqMqWbNmDd577z3cunULXl5elq4OqUSDBw/G3bt3cebMGUtXhZBKR11spErLycnRSUhMTk7GN998g86dO1NwVEkOHz6Mt99+m4KjGoYxhujoaIsmbBNiSdSCRKq01q1bo3v37mjSpAkSEhKwceNGPHr0CFFRUejataulq0cIIaSaohYkUqX17dsXP/zwAzZs2ABBEPD8889j48aNFBwRQggxK2pBIoQQQgjRQ/MgEUIIIYTooQCJEEIIIUQP5SCVk1qtxqNHj+Do6GjyCQsJIYQQYh6MMWRkZMDX1xcSScntRBQgldOjR4+KrehMCCGEEOtw//591KlTp8THKUAqJ0dHRwD8BXZycrJwbQghhBBSFunp6fDz89Ncx0tCAVI5id1qTk5OFCARQgghVuZZ6TGUpE0IIYQQoocCJEIIIYQQPRQgEUIIIYTooRwkQgghJqFSqVBQUGDpapAaztbW1iSLmVOARAghpEIYY4iPj0dqaqqlq0IIAMDFxQXe3t4VmqeQAiRCCCEVIgZHnp6esLOzo8lzicUwxpCdnY3ExEQAgI+PT7n3RQESIYSQclOpVJrgyM3NzdLVIQRKpRIAkJiYCE9Pz3J3t1GSNiGEkHITc47s7OwsXBNCioifx4rkxFGARAghpMKoW41UJab4PFKARAghhBCihwIkQgghhBA9FCARQgixOJUKiI4GvvuO/69Sme9Y69atg6OjIwoLCzXbMjMzYWtri+7du+uUjY6OhiAIuHXrlvkqRKokCpAIIYRY1K5dQEAA8MILwNCh/P+AAL7dHF544QVkZmbir7/+0mw7duwYvL29cfr0aeTm5mq2Hz58GHXr1kWDBg2K7Sc/P988FTShqjBxpzW8ToZQgFQVqVSAWm3pWhBCiNnt2gW88grw4IHu9ocP+XZzBEmBgYHw8fFBdHS0Zlt0dDT69++PevXq4dSpUzrbX3jhBQDAqFGjMGDAACxbtgy+vr4IDAwEAHzzzTdo27YtHB0d4e3tjaFDh2rm4RH3IQgC9u7di5YtW0KhUKBDhw64cuWKpsyWLVvg4uKCPXv2oFGjRlAoFAgNDcX9+/d16v7TTz/h+eefh0KhQP369bF48WKdljBBELB27Vq8/PLLsLe3x7Jlywy+BmvWrNEcx8vLC6+88ormMbVajffffx8NGzaEXC5H3bp1dfZz+fJl9OjRA0qlEm5ubhg3bhwyMzM1j5f0Ot2/fx+DBw+Gi4sLXF1d0b9/f9y9e/eZ75elUIBUFSUl8d8OWVmWrgkhhJiNSgVMmQIwVvwxcdvUqebpbnvhhRdw+PBhzf3Dhw+je/fu6Natm2Z7Tk4OTp8+rQmQACAqKgqxsbE4cOAAfv31VwC8lWbp0qW4ePEi9uzZg7t372LUqFHFjvnuu+/io48+wtmzZ+Hh4YGwsDCdFp7s7GwsW7YMX3/9NY4fP47U1FS8/vrrmsePHTuGESNGYMqUKbh69SrWr1+PLVu2FAuCFi1ahIEDB+Ly5ct44403itXjr7/+wttvv40lS5YgNjYW+/fvR9euXTWPz549GytWrMD8+fNx9epVbNu2DV5eXgCArKwshIaGolatWjh79ix27tyJgwcPYtKkSTrH0H+dCgoKEBoaCkdHRxw7dgzHjx+Hg4MDXnzxxarbwsRIuaSlpTEALC0tzfQ7v3+fsQsXGLt6lbHHjxnLzTX9MQghxARycnLY1atXWU5OjtHPPXyYMR4KlX47fNjk1WZffPEFs7e3ZwUFBSw9PZ3Z2NiwxMREtm3bNta1a1fGGGNRUVEMALt37x5jjLGRI0cyLy8vlpeXV+q+z549ywCwjIyMf8/zMAPAtm/frimTnJzMlEol27FjB2OMsc2bNzMA7NSpU5oy165dYwDY6dOnGWOM9ezZky1fvlznWN988w3z8fHR3AfApk6dWmr9fvzxR+bk5MTS09OLPZaens7kcjn74osvDD53w4YNrFatWiwzM1Ozbe/evUwikbD4+HjGmOHX6ZtvvmGBgYFMrVZrtuXl5TGlUsl+//33UutbHqV9Lst6/aYWpKpKoQDs7HhrUlwckJwMaDWjEkKItXv82LTljNG9e3dkZWXh7NmzOHbsGJ577jl4eHigW7dumjyk6Oho1K9fH3Xr1tU8r0WLFpDJZDr7iomJQVhYGOrWrQtHR0d069YNABAXF6dTLiQkRPOzq6srAgMDce3aNc02GxsbtGvXTnO/cePGcHFx0ZS5ePEilixZAgcHB81t7NixePz4MbKzszXPa9u2bann3rt3b/j7+6N+/foYPnw4tm7dqnn+tWvXkJeXh549exp87rVr19CqVSvY29trtnXq1AlqtRqxsbElvk4XL17EzZs34ejoqKm7q6srcnNzq2wCPC01UpXZ2gKurkBODv8NkZ4OuLkBDg6AhGJbQoh1K+syWRVYTqtEDRs2RJ06dXD48GGkpKRoghpfX1/4+fnhxIkTOHz4MHr06KHzPO3AACjqcgoNDcXWrVvh4eGBuLg4hIaGmrzrKDMzE4sXL8agQYOKPaZQKEqsoz5HR0ecO3cO0dHR+OOPP7BgwQIsWrQIZ8+e1SzTUVH6dcjMzERQUBC2bt1arKyHh4dJjmlqdJW1BkolUKsWUFAA3L/P85O0/loghBBr1KULUKcOUNKkx4IA+PnxcubwwgsvIDo6GtHR0TrD+7t27YrffvsNZ86c0ck/MuT69etITk7GihUr0KVLFzRu3FgnQVubdvJ3SkoK/vnnHzRp0kSzrbCwUGdkXWxsLFJTUzVlnn/+ecTGxqJhw4bFbhIj/2i2sbFBr1698P777+PSpUu4e/cuDh06hEaNGkGpVCIqKsrg85o0aYKLFy8iSytH9vjx45BIJJpkbEOef/553LhxA56ensXq7uzsbFTdKwsFSNZCEHjLkZMTkJEB3LsHJCQAVTW5jRBCnkEqBT75hP+sHySJ9yMjeTlzeOGFF/Dnn3/iwoULmhYkAOjWrRvWr1+P/Pz8ZwZIdevWhUwmw2effYbbt2/j559/xtKlSw2WXbJkCaKionDlyhWMGjUK7u7uGDBggOZxW1tbTJ48GadPn0ZMTAxGjRqFDh06oH379gCABQsW4Ouvv8bixYvx999/49q1a9i+fTvmzZtn1Hn/+uuv+PTTT3HhwgXcu3cPX3/9NdRqNQIDA6FQKDBz5ky89957+Prrr3Hr1i2cOnUKGzduBAAMGzYMCoUCI0eOxJUrV3D48GFMnjwZw4cP1yRyGzJs2DC4u7ujf//+OHbsGO7cuYPo6Gi8/fbbeKA/hLGKoADJ2kilgIsLb1VKSuKBUnKyeWdVI4QQMxk0CPjhB6B2bd3tderw7QZ6k0zmhRdeQE5ODho2bKhzce/WrRsyMjI00wGUxsPDA1u2bMHOnTvRtGlTrFixAh9++KHBsitWrMCUKVMQFBSE+Ph4/PLLLzp5OnZ2dpg5cyaGDh2KTp06wcHBATt27NA8Hhoail9//RV//PEH2rVrhw4dOuDjjz+Gv7+/Ueft4uKCXbt2oUePHmjSpAnWrVuH7777Ds2aNQMAzJ8/HzNmzMCCBQvQpEkTvPbaa5pWMTs7O/z+++94+vQp2rVrh1deeQU9e/bE559/Xuox7ezscPToUdStWxeDBg1CkyZNMGbMGOTm5sLJycmo+lcWgTFDAyzJs6Snp8PZ2RlpaWmmf3MfPOBD/B0dn102J4ff7O15fpKjY8nt1YQQYmK5ubm4c+cO6tWrp5MHYyyVCjh2jKdb+vjwbjVztRxVNnEupZSUFLi4uBgss2XLFkydOhWpqamVWrfqqrTPZVmv35Skbe2USj7iLSuL5ye5uPB8JTs7S9eMEELKTCoF9Fb5IMSiKECqDsT8JJUKSEvjOUq1avGb3nBUQgghhDwb5SBVJ2J+kkIBPHnC509KSaH8JEIIsaDu3buDMVZi9xrAl+eg7rWqhQKk6kgm4/MnCQKfEuD+fd6qROlmhBBCSJlYPEBavXo1AgICoFAoEBwcjDNnzpRaPjIyEoGBgVAqlfDz88O0adN0Vl5etGgRBEHQuTVu3FhnH927dy9WZvz48WY5P4sS50/Ky+NB0qNHPKGbEEIIIaWyaA7Sjh07MH36dKxbtw7BwcGIjIxEaGgoYmNj4enpWaz8tm3bMGvWLGzatAkdO3bEP//8g1GjRkEQBKxatUpTrlmzZjh48KDmvo1N8dMcO3YslixZorlvV12TmgWBj2wT85MyM4vyk2xtLV07QgghpEqyaIC0atUqjB07FqNHjwYArFu3Dnv37sWmTZswa9asYuVPnDiBTp06YejQoQCAgIAADBkyBKdPn9YpZ2NjA29v71KPbWdn98wy1YqYn5SXByQm8mVL3N158FRdxtISQgghJmKxLrb8/HzExMSgV69eRZWRSNCrVy+cPHnS4HM6duyImJgYTTfc7du3sW/fPvTt21en3I0bN+Dr64v69etj2LBhxRYMBICtW7fC3d0dzZs3x+zZs3UW+qvW5HKenwTw+Zbu3+etSpSfRAghhGhYrAUpKSkJKpWq2NTkXl5euH79usHnDB06FElJSejcuTMYYygsLMT48eMxZ84cTZng4GBs2bIFgYGBePz4MRYvXowuXbrgypUrcPx34sWhQ4fC398fvr6+uHTpEmbOnInY2Fjs2rWrxPrm5eUhLy9Pcz89Pb0ip295dnZF8yfdu1fU7WaihQoJIYQQa2ZV8yBFR0dj+fLlWLNmDYKDg3Hz5k1MmTIFS5cuxfz58wEAL730kqZ8y5YtERwcDH9/f3z//fcYM2YMAGDcuHGaMi1atICPjw969uyJW7duoUGDBgaPHRERgcWLF5vx7CxAIuFdbIWFQGoqH+nm6sq74ig/iRBCKo0gCNi9e7fO2mzEsizWxebu7g6pVIqEhASd7QkJCSXmBs2fPx/Dhw/Hm2++iRYtWmDgwIFYvnw5IiIioFarDT7HxcUFzz33HG7evFliXYKDgwGg1DKzZ89GWlqa5nb//v1nnaL1sLHhQZFMxhfAjYvjAVMJrykhhJhMXBxw7lzJNwMpEqYSHx+PKVOmoGHDhlAoFPDy8kKnTp2wdu1ak6ddjBo1qtTg5/Hjxzp/4JcFYwwbNmxAcHAwHBwc4OLigrZt2yIyMlJTf0MjuwVB0BnI9N1330EqlWLixInlOrfqymItSDKZDEFBQYiKitJ8aNRqNaKiojBp0iSDz8nOzoZEohvTSf9NMC5pSbnMzEzcunULw4cPL7EuFy5cAIBSFyWUy+WQy+UlPl4tyOU8SMrJ4flJjo58fTd7e1rfjRBienFxQJMmQGnBiJ0dcO0aULeuSQ99+/ZtdOrUCS4uLli+fDlatGgBuVyOy5cvY8OGDahduzZefvllkx6zNOUZNDR8+HDs2rUL8+bNw+effw4PDw9cvHgRkZGRCAgI0Fxb9Ud2A4CrmIsKYOPGjXjvvfewfv16fPTRRxVaU8/U8vPzdRb0rUwWnQdp+vTp+OKLL/DVV1/h2rVreOutt5CVlaUZ1TZixAjMnj1bUz4sLAxr167F9u3bcefOHRw4cADz589HWFiYJlB65513cOTIEdy9excnTpzAwIEDIZVKMWTIEADArVu3sHTpUsTExODu3bv4+eefMWLECHTt2hUtW7as/BehqhEE/gvJxYUHSnFxQHw8oDXXFCGEmERSEg+Ovv0WiIkpfvv2W/54UpLJDz1hwgTY2Njgr7/+wuDBg9GkSRPUr18f/fv3x969exEWFqYpm5qaijfffBMeHh5wcnJCjx49cPHiRc3jixYtQuvWrfHNN98gICAAzs7OeP3115GRkVHm+giCgD179gAA7t69C0EQsGvXLrzwwguws7NDq1atdAYwff/999i6dSu+++47zJkzB+3atUNAQAD69++PQ4cO4YUXXtCUFUd2a9/EoOPOnTs4ceIEZs2aheeee67UXFyAN0YsWrQIdevWhVwuh6+vL95++23N43l5eZg5cyb8/Pwgl8vRsGFDbNy4UfP4kSNH0L59e8jlcvj4+GDWrFkoLCzUPN69e3dMmjQJU6dOhbu7O0JDQwEAV65cwUsvvQQHBwd4eXlh+PDhSDLD50KbRXOQXnvtNTx58gQLFixAfHw8Wrdujf3792sSt+Pi4nRajObNmwdBEDBv3jw8fPgQHh4eCAsLw7JlyzRlHjx4gCFDhiA5ORkeHh7o3LkzTp06BQ8PDwC85ergwYOIjIxEVlYW/Pz8EB4ejnnz5lXuyVd1Egng5MTzk54+1c1PMjCvFCGElFuTJsDzz1fa4ZKTk/HHH39g+fLlsLe3N1hG0Go1f/XVV6FUKvHbb7/B2dkZ69evR8+ePfHPP/9oWmJu3bqFPXv24Ndff0VKSgoGDx6MFStW6FyfjDV37lx8+OGHaNSoEebOnYshQ4bg5s2bsLGxwdatWxEYGIj+/fsbrLuzs3OZjrF582b069cPzs7O+O9//4uNGzdqptIx5Mcff8THH3+M7du3o1mzZoiPj9cJFkeMGIGTJ0/i008/RatWrXDnzh1NIPPw4UP07dsXo0aNwtdff43r169j7NixUCgUWLRokWYfX331Fd566y0cP34cAA9Qe/TogTfffBMff/wxcnJyMHPmTAwePBiHDh0q03mWCyPlkpaWxgCwtLQ00+/8/n3Grl9n7OHDqnO7dYuxy5cZu3mTsdRUxlQq0583IcTq5OTksKtXr7KcnBzjnxwTwxjA/y/P4+V06tQpBoDt2rVLZ7ubmxuzt7dn9vb27L333mOMMXbs2DHm5OTEcnNzdco2aNCArV+/njHG2MKFC5mdnR1LT0/XPP7uu++y4OBgzf2RI0ey/v37l1gnAGz37t2MMcbu3LnDALAvv/xS8/jff//NALBr164xxhhr0qQJe/nll595rgsXLmQSiURzXvb29qxdu3aMMcZUKhXz8/Nje/bsYYwx9uTJEyaTydjt27dL3N9HH33EnnvuOZafn1/ssdjYWAaAHThwwOBz58yZwwIDA5lardZsW716NXNwcGCqf68p3bp1Y23atNF53tKlS1mfPn10tt2/f58BYLGxsQaPVdrnsqzXb4svNUKshELBpwFQq/ncSQ8e8CkCaP4kQkg1cebMGVy4cAHNmjXTTOty8eJFZGZmws3NDQ4ODprbnTt3cOvWLc1zAwICNFPJADynNTExsUL10U77EHNkxX0yI373BgYG4sKFC5rbjz/+CAA4cOAAsrKyNHMJuru7o3fv3ti0aVOJ+3r11VeRk5OD+vXrY+zYsdi9e7emi+zChQuQSqXo1q2bwedeu3YNISEhOq1znTp1QmZmJh48eKDZFhQUpPO8ixcv4vDhwzqvv7iEmPZ7YGrUV0LKThB4wrZSybvcxGVLXF15gjchhFiBhg0bQhAExMbG6myvX78+AECpNR9cZmYmfHx8EB0dXWw/Li4ump9t9aZGEQShxNHVZaW9TzGoEPf53HPPlThnoD6ZTIaGDRsW275x40Y8ffpU53zVajUuXbqExYsXFxsUBQB+fn6IjY3FwYMHceDAAUyYMAEffPABjhw5orOfitDv9szMzERYWBhWrlxZrGxpg6sqilqQiPEkEsDZmQdLT5/yiSaTk3m+EiGEVHFubm7o3bs3Pv/8c2RlZZVa9vnnn0d8fDxsbGzQsGFDnZu7u3sl1bi4oUOH4p9//sFPP/1U7DHGGNLS0kp9fnJyMn766Sds375dp3Xp/PnzSElJwR9//FHic5VKJcLCwvDpp58iOjoaJ0+exOXLl9GiRQuo1WocOXLE4POaNGmCkydP6rR+HT9+HI6OjqhTp06Jx3v++efx999/IyAgoNh7UFIOmSlQgETKz9aWtyDZ2ACPH/NAKS2N5k8ihBjn2jXDcyBdu2a2Q65ZswaFhYVo27YtduzYgWvXriE2Nhbffvstrl+/rhkZ3atXL4SEhGDAgAH4448/NCOk586di7/++suoY6alpekEIxcuXCj3nHqDBw/Ga6+9hiFDhmD58uX466+/cO/ePfz666/o1asXDh8+XOrzv/nmG7i5uWHw4MFo3ry55taqVSv07dtXZ+SZti1btmDjxo24cuUKbt++jW+//RZKpRL+/v4ICAjAyJEj8cYbb2DPnj24c+cOoqOj8f333wPgIwfv37+PyZMn4/r16/jpp5+wcOFCTJ8+3WBrlWjixIl4+vQphgwZgrNnz+LWrVv4/fffMXr0aKhUqnK9fmVBXWyk4hQK3sWWnc3zk5ydebebGSN7Qkg14O7OpxX5739LLmNnx8uZWIMGDXD+/HksX74cs2fPxoMHDyCXy9G0aVO88847mDBhAgDetbVv3z7MnTsXo0ePxpMnT+Dt7Y2uXbsWWyrrWaKjo9GmTRudbWPGjMGXX35pdP0FQcC2bduwYcMGbNq0CcuWLYONjQ0aNWqEESNGaIbHl2TTpk0YOHCgTj6QKDw8XDOMXr+VzMXFBStWrMD06dOhUqnQokUL/PLLL3BzcwMArF27FnPmzMGECROQnJyMunXrapYDq127Nvbt24d3330XrVq1gqurK8aMGfPMUeS+vr44fvw4Zs6ciT59+iAvLw/+/v548cUXSw2sKkpgxmR6EY309HQ4OzsjLS0NTk5Opt25mACtlfBnNVQqnpsEFK3vRvlJhFRbubm5uHPnDurVq1e+CQbj4kqf58jd3eSTRJLqr7TPZVmv39SCRExLKuUtSAUF/JeeOH+SszPNn0QIKa5uXQqASJVEOUjEPGxteWAkkfD8pPv3gfR0mhaAEEKIVaAAiZiXUsm72QoKiuZPMvEikIQQQoipUZ8HMT9BABwceH5SRgbPrxLzkyy0CCEhhBBSGgqQSOWRSvlabvn5wJMnuvlJ/w6pJYRYJxrvQ6oSU3weqYuNVD6ZjAdGggA8esRHsWRkUH4SIVZInO05m7rOSRUifh71Zzg3BrUgEctRKvkcSllZPEhyceGBk4mmqyeEmJ9UKoWLi4tmjTA7OzuDc+tYrerwh5s1n4Mg8FsZMcaQnZ2NxMREuLi4aCb8LA8KkIhlaecnpaUVdbu5uFB+Eqn6GOMDEPLySl9qpzwXKL6OvXnLa/9fgfLejAESCRIfPeIbBMG6L8r6qtO5WBuplI+GNpKLiwu8vb0rdGgKkEjVoJ2flJjIpwRwcwOcnCg/iVQthYU8IMrN5QF9fj4Pkhgr/S9d7YussS0s5WmRMeY5FayPAMAHgKetLQpMFUyY+5zLewxz7r8yztnc+zblOWdkAD4+fDZ1I9ja2lao5UhEARKpWsT8pJwc4OFD3qrk5sZbmapTsz2xHioVD4jy8vgs8bm5PCgC+OdVLuefTwLpvzdCTEIi4d+v8szQbgIUIFUFJ08Ct28X3X/6lP8yFj8U/v5AUJBl6mYpYn5SZqbu+m6Un0TMTa3mAVBeHp+zKzub/6xW8wlQZTL+Fy0F7IRUaxQgWdrJk0DHjs8u9/PPNS9IEgS+Hl1hIW9Jyswsmj+pAiMTCNGhnUeUk8M/Z/n5vOVIKuWfNSencuVBEEKsFwVIlia2HL31FtCpE/9ZuwXp7Fng66+Be/dqXoAksrHh+Ul5eTw/KSOjKD+JLlqkPMSAKDeXB0R5eXyb2KRvb0+5b4TUcBQgVRWdOgHDhvGfHzzgQ98dHfn9r7+2XL2qErmc37Kz+Wvk6MgDJXt76u4gpXtWHpFCUfR9I4QQUIBErJGdXVF+kvb8SRZK5CNVEOUREUIqiAIkYp0kEt7FVlgIpKQUdbs5O1N+Uk3EWFFApD38XqXiXbQyGXXJEkKMQgESsW42NjxpOy8PiI8vmhaALobV37PyiBwcKI+IEFJuFCCR6kEu560EOTmUn1RdUR4RIaQSUYBUVRw/XvSz/ig2UjaCUDw/SZwWgPKTrI9aXRQQZWXx4Dcvjz9GeUSEEDOjAMnS6tfn/69dy28l8fevnPpUB2J+UkEBDza113ezoY98laWdR6Q/H5GYR+TsTF2nhJBKQVcLSwsJAU6coJm0zcHWlrce5eYW5Se5u/NuGLrIVg2UR0QIqaIoQKoKQkL4TaQ/DxKpGIWiaP4k7WVLqHum8okLvernEUkkvIVIqaTPPSGkSqAAidQMgsATtpVK3uWWmVk0f5JcbunaVV9lySOiRHpCSBVEARKpWSQS3oJUUAAkJ/NAydWVb6P8pIorSx6RiwsFRISQKo+uCKRmsrXlgVFODvD4cVF+koMD5ScZSwyI8vJ461xeHu9Kk0p5QER5RIQQK0QBEqnZlEqeo5SVxacFEPOT7O0tXbOqS8wjys3lr5uYRyQIvLvSzo5a4wghVo9+ixEiCLyVQ6XiXUJit5uLC+UnASXnETHGW4goj4gQUg1RgESISCrlLUj5+cCTJ0B6elGgVJO6iBjjAVB+Ph/5l5VVFBDZ2PDuScojIoRUcxQgEaJPJtPNT0pP58uWODpW36BAe6FXcT4i7TwiWtuOEFLDWPw33urVqxEQEACFQoHg4GCcOXOm1PKRkZEIDAyEUqmEn58fpk2bhtzcXM3jixYtgiAIOrfGjRvr7CM3NxcTJ06Em5sbHBwcEB4ejoSEBLOcH7FiSiWfaLKggM+f9PAhb1GpDgoLectQcjJw7x5w9y7PwUpM5F2NdnZFo/uUSgqOCCE1jkVbkHbs2IHp06dj3bp1CA4ORmRkJEJDQxEbGwtPT89i5bdt24ZZs2Zh06ZN6NixI/755x+MGjUKgiBg1apVmnLNmjXDwYMHNfdt9BJGp02bhr1792Lnzp1wdnbGpEmTMGjQIBzXXg+NEEA3Pyk9nbeuiOu7yWSWrl3ZqVRFrURZWTzQKyjg3WbifEQODpauJSGEVBkWDZBWrVqFsWPHYvTo0QCAdevWYe/evdi0aRNmzZpVrPyJEyfQqVMnDB06FAAQEBCAIUOG4PTp0zrlbGxs4O3tbfCYaWlp2LhxI7Zt24YePXoAADZv3owmTZrg1KlT6NChgylPkVQXUinPuxHzk8T13Zydq2Z+kphHlJfHg6HsbP6zWk0LvRJCSBlYrN08Pz8fMTEx6NWrV1FlJBL06tULJ0+eNPicjh07IiYmRtMNd/v2bezbtw99+/bVKXfjxg34+vqifv36GDZsGOLi4jSPxcTEoKCgQOe4jRs3Rt26dUs8LgDk5eUhPT1d50ZqIDE/SRCAR49411tGBg9ILEmcoDEjg3eT3bnDu83u3wdSU3kZJyded0dHPjqPgiNCCCmRxVqQkpKSoFKp4OXlpbPdy8sL169fN/icoUOHIikpCZ07dwZjDIWFhRg/fjzmzJmjKRMcHIwtW7YgMDAQjx8/xuLFi9GlSxdcuXIFjo6OiI+Ph0wmg4uLS7HjxsfHl1jfiIgILF68uPwnTKoXcf6kzEzd9d2Uysqrg/Z8RBkZPEAqKOCP0XxEhBBSIVaVeRkdHY3ly5djzZo1OHfuHHbt2oW9e/di6dKlmjIvvfQSXn31VbRs2RKhoaHYt28fUlNT8f3331fo2LNnz0ZaWprmdv/+/YqeDrF2gsBbYxwdeSuNmOScn2+e46lUvKssJYUHZWIrUXw8D4zk8qL8KAqOCCGkQiz2G9Td3R1SqbTY6LGEhIQS84fmz5+P4cOH48033wQAtGjRAllZWRg3bhzmzp0LiYGRNi4uLnjuuedw8+ZNAIC3tzfy8/ORmpqq04pU2nEBQC6XQ06TBhJDpFIelOTl8QBJnBbAyali+UlqdVFiNeUREUJIpbJYC5JMJkNQUBCioqI029RqNaKiohASEmLwOdnZ2cWCIOm/FyBWQg5IZmYmbt26BR8fHwBAUFAQbG1tdY4bGxuLuLi4Eo9LSJnI5UX5SQ8fGp+fpJ9HdPcu5RERQoiFWLQNfvr06Rg5ciTatm2L9u3bIzIyEllZWZpRbSNGjEDt2rUREREBAAgLC8OqVavQpk0bBAcH4+bNm5g/fz7CwsI0gdI777yDsLAw+Pv749GjR1i4cCGkUimGDBkCAHB2dsaYMWMwffp0uLq6wsnJCZMnT0ZISAiNYCOmoVTy4CUr69n5SQUFxSdoLCjg8w7J5XwJj6o4So4QQqo5iwZIr732Gp48eYIFCxYgPj4erVu3xv79+zWJ23FxcTotRvPmzYMgCJg3bx4ePnwIDw8PhIWFYdmyZZoyDx48wJAhQ5CcnAwPDw907twZp06dgoeHh6bMxx9/DIlEgvDwcOTl5SE0NBRr1qypvBMn1Z9Ewlt5CguBtLSi9d2cnHguUV4e3yYu9ArwLjOFgj+PEEKIRQmspL4pUqr09HQ4OzsjLS0NTk5Opt35gwe89YEulNWHOEGjTMaDJu08IpmMusoIIURfSgoQEMBb0k2orNdvGuZCSGWQy/ktP58nVtPSHYQQUqVRgERIZbKm5UkIIaQGoz9jCSGEEEL0UIBECCGEEKKHAiRCCCGEED0UIBFCCCGE6KEAiRBCCCFEDwVIhBBCCCF6KEAihBBCCNFDARIhhBBCiB4KkAghhBBC9FCARAghhBCihwIkQgghhBA9FCARQgghhOihAIkQQgghRA8FSIQQQggheihAIoQQQgjRQwESIYQQQogeCpCqKsYsXQNCCCGkxqIAqSqyt+cBUlaWpWtCCCGE1EgUIFVFtWoBdeoAggCkplJrEiGEEFLJKECqqhwdgdq1ATs7ICUFUKksXSNCCCGkxrCxdAVIKZRKHiQ9eQI8fQo4OAAymaVrRQghhJjew4f8WidKT+e9KEolv+/uDtStW2nVoQCpqrO1Bby9eWCUkAAUFvJWJUIIIaS6ePgQ6NYNyMkpuYydHXDtWqUFSRQgWQOJBHBz48FSQgKPqh0deY4SIYQQYu2ePuXB0WefAY0a8W3p6YCvL29BunYN+O9/gaQkCpCIHkEAnJ2LgqSUFMDFhQdPhBBCSHXQqBHQogX/OSUFCAjgI7stgK6u1sbOjuclOTvzvtmCAkvXiBBCCKl2KECyRjIZb3b08AAyM0vvsyWEEEKI0ShAslZSKeDpyQOlggIgI8PSNSKEEEKqDQqQrJkgFE0qaWPDu9zUakvXihBCCLF6lKRdHTg4FCVvp6YCTk48YCKEmI/+nC36XF15viAhpOxu3Cj6WXsepGvXKr0qdBWtLuRy3t1mawskJ/Osf7nc0rUipHoqy5wtSiVw5AgFSYSUhasr/85MnlxyGTs7PllkJaEAqTqxsQG8vHQnlbTQ8EhCqjVDc7Zou3GD/6J/+pQCJELKonZt/geF/kza4jxIAM2kTSpIf1LJ1FQ+JQBNKkmI6WnP2UIIqZjatXX/oKjp8yCtXr0aAQEBUCgUCA4OxpkzZ0otHxkZicDAQCiVSvj5+WHatGnIzc01WHbFihUQBAFTp07V2d69e3cIgqBzGz9+vKlOqWpwcuLJ27TYLSGEEGI0i7Yg7dixA9OnT8e6desQHByMyMhIhIaGIjY2Fp6ensXKb9u2DbNmzcKmTZvQsWNH/PPPPxg1ahQEQcCqVat0yp49exbr169Hy5YtDR577NixWLJkiea+XXVc30xc7DYxkQdJtNgtIYQQUiYWbUFatWoVxo4di9GjR6Np06ZYt24d7OzssGnTJoPlT5w4gU6dOmHo0KEICAhAnz59MGTIkGKtTpmZmRg2bBi++OIL1KpVy+C+7Ozs4O3trbk5OTmZ/PyqBFtbwMeH5yZlZ9OkkoQQQkgZWCxAys/PR0xMDHr16lVUGYkEvXr1wsmTJw0+p2PHjoiJidEERLdv38a+ffvQt29fnXITJ05Ev379dPatb+vWrXB3d0fz5s0xe/ZsZGdnl1rfvLw8pKen69yshkTCk9t8fXlXW3o6wJila0UIIYRUWRbrYktKSoJKpYKXl5fOdi8vL1y/ft3gc4YOHYqkpCR07twZjDEUFhZi/PjxmDNnjqbM9u3bce7cOZw9e7bEYw8dOhT+/v7w9fXFpUuXMHPmTMTGxmLXrl0lPiciIgKLFy828iyrEEHgi9va2gLx8UXJ27TYLSHlpz1nS1m2E0KshlWNYouOjsby5cuxZs0aBAcH4+bNm5gyZQqWLl2K+fPn4/79+5gyZQoOHDgAhUJR4n7GjRun+blFixbw8fFBz549cevWLTRo0MDgc2bPno3p06dr7qenp8PPz890J1dZ7O0BP7+iEW6OjjxoIoSUXVnmbFEqeTlCiFWyWIDk7u4OqVSKhIQEne0JCQnw9vY2+Jz58+dj+PDhePPNNwHw4CYrKwvjxo3D3LlzERMTg8TERDz//POa56hUKhw9ehSff/458vLyIJVKi+03ODgYAHDz5s0SAyS5XA55dZl4UVzsViYDnjzhI91KCSgJIXoMzdmij2bSJsSqWSxAkslkCAoKQlRUFAYMGAAAUKvViIqKwqRJkww+Jzs7GxK9LiEx4GGMoWfPnrh8+bLO46NHj0bjxo0xc+ZMg8ERAFy4cAEA4OPjU4EzsjLiYre2tnyUW2EhH+VGCCkb/TlbCCHVikW72KZPn46RI0eibdu2aN++PSIjI5GVlYXRo0cDAEaMGIHatWsjIiICABAWFoZVq1ahTZs2mi62+fPnIywsDFKpFI6OjmjevLnOMezt7eHm5qbZfuvWLWzbtg19+/aFm5sbLl26hGnTpqFr164lTglQbQkC/ytXnHk7JYXnKdGkkoQQQmo4iwZIr732Gp48eYIFCxYgPj4erVu3xv79+zWJ23FxcTotRvPmzYMgCJg3bx4ePnwIDw8PhIWFYdmyZWU+pkwmw8GDBzXBmJ+fH8LDwzFv3jyTn5/VcHDgy5SIQZKzM29hIoQQQmoogTEa710e6enpcHZ2RlpaWvWZQ6mwkOck0WK3hBBCLM1MS42U9fptVaPYiJmJi92KeUkqFU/gJoQQQmoYmgSH6BInlaxTh08mmZZGk0oSQgipcShAIoaJi90qFLTYLSGEkBqnXAFSYWEhDh48iPXr1yMjIwMA8OjRI2RmZpq0csTCxMVua9XiLUn5+ZauESGEEFIpjM5BunfvHl588UXExcUhLy8PvXv3hqOjI1auXIm8vDysW7fOHPUkliKTAd7ePC/pyRPekqRUWrpWhBBCiFkZ3YI0ZcoUtG3bFikpKVBqXSgHDhyIqKgok1aOVBFSKeDhwVuTCguBf1sNCSGEkOrK6BakY8eO4cSJE5DJZDrbAwIC8PDhQ5NVjFQx+ovdivMl0WK3hBBCqiGjAyS1Wg2VgYTdBw8ewNHR0SSVIlWYvT1P3k5M5IvdOjnx6QGIYQ8f0npdhBBihYy+svXp0weRkZHYsGEDAEAQBGRmZmLhwoXo27evyStIqiC5nC92a2sLJCXRYrclefgQ6NYNyMkpuYxSyRc9pSDJqhUUAGo1za1KSHVidID04Ycf4sUXX0TTpk2Rm5uLoUOH4saNG3B3d8d3331njjqSqkgq1Z1Ukha7Le7pUx4cffYZ0KhR8cdv3AAmT+blKECyOioVf3uzsvhNrQYcHfnXQKmk3mdCrJ3RAZKfnx8uXryIHTt24OLFi8jMzMSYMWMwbNgwnaRtUgMIAuDmVrTYbWoqz0uixW51NWoEtGhh6VoQE1CrgdxcIDsbyMwE8vL4x12h4P+npfGbUsl7n+3s+N8QhBDrY1SAVFBQgMaNG+PXX3/FsGHDMGzYMHPVi1gTR8fiydu02C2pJhjjQVFODh/AmZvLgyGZjAdB2n8P2NryIConB3j8mHe5iYGSGEQRQqyDUQGSra0tcnNzzVUXYs0UCt5N9OQJ7zJycOBXEEKsVF4eD4YyMnjAo1bzj7SjY+ndZxJJ0dqaeXk8TU8q5dscHXnrEv39QEjVZ3Qv+cSJE7Fy5UoUFhaaoz7Emtna8kklvb15UkZ2tqVrRKyEWs3T2NRqy9ajoABITwcePQLu3+etQHl5vAXI2dn43CK5vOh5WVk8b//BA94bTRPTE1K1GZ2DdPbsWURFReGPP/5AixYtYC/+qfSvXbt2maxyxApJJDwvydaW5yWlpRXvhyA1BmM86FGp+I2xop9VKh6QFBTwn9Vq/vGxseEtNTIZb2mxsSnabo6WF/1k6/x8fiyFwnQzWEilvFFVreYBV3x8UWuUgwN1vxFSFRn99XdxcUF4eLg56kKqC0HgfzaLQVJKCp9ksqYO67lxw7jtVkIMfNRq3SBIrS4KfMRWIXE7Y/y5jPGPg0TCgwdBKMrfyc8v6tJijD8mlRYFS7a2vGXGUPBU1iCjtGRrOzvzvWYSCW9NUir5eaak8Ju9fVGuEnW/EVI1GB0gbd682Rz1INWRnR3PSxInlRSTuWsKV1d+JZw8ueQySiUvV4WpVPxinp/PAx79Vh8xABKDGaAoWBEDG5mM3y9PjCy2QhUW8lteHm+YFEmlRYGWTKYbPGkHVkDZk60rg9hKJrZgZWTwAM3JiQdMNKcSIZZV7gbkJ0+eIDY2FgAQGBgIDw8Pk1WKVCMyGeDjUzSppEJRcxa7rV2bTwJpZTNpi0FQXh6/cOfmFk2EaKjVR9xmrgBDO9AyRLvlKiuLBxqM8Zt28CSR8HMqa7J1ZRG73xjj9UtM5AGdoyO/KRRVo56E1DRGB0hZWVmYPHkyvv76a6j/zaiUSqUYMWIEPvvsM9iZs32aWCepFPD05FelxER+Baspy9LUrl3lAiB9YotQfn5RDk5+Pr9gi11aDg5V9yJdWvCk3fWnUlXtLiyxi0+h4O9HWhpveLWz461KMllRFyWg212pv83Q488qJwa6hm7GdF8SUl0YHSBNnz4dR44cwS+//IJOnToBAP7880+8/fbbmDFjBtauXWvySpJqQBCAWrWK8pLEddyq6lW3mmKsKCASW4jy8vh9oKibqrrk1YsXeGvr2bW1LT6nkiDodmMaIj5e1nJA0f/agZTYHSr+LwZJUimvl5j7VVIwRV9rUh0IjGn/LfFs7u7u+OGHH9C9e3ed7YcPH8bgwYPx5MkTU9avykpPT4ezszPS0tLg5ORk6epYl7w8HiSlp9Nit2bGGG8N0m8hEmfpEEeM2dhUj4CImI6YJK+dYC/mnInTMYjBlHYgJQZWNjZFLZBikKp9n5BnSkkBAgKKJhYzkbJev42+MmVnZ8PLy6vYdk9PT2TTvDekLLQXu336lPch1JCMVLHrSjsY0f9Lvjw/i8QRZPn5PA7NzuY/q1RFFy2ZjL/kFBCR0ohBTFm6JPWDJ3H2cXG7mBOmHTQplUVBk3ijzySpSowOkEJCQrBw4UJ8/fXXUPy7gntOTg4WL16MkJAQk1eQVFM2NnxCSTEvqbDQ5H8lVCUqFc8pSUnRzf8obyCkv128mKlURaO9xL/aTTmfDyGGPCuRXiR+PnNz+fQK4nQP2tM3yOXFW58IsQSjf21+8sknCA0NRZ06ddCqVSsAwMWLF6FQKPD777+bvIKkGhMXu7W1LZoKoBoudpuTwxvKMjL4X83awcqzEmfL+rPYZSYItJQFqbrEIEq7wVhMpBcDJ7G1U5yeQSbjQb52Fx0F/KQyGP0xa968OW7cuIGtW7fi+vXrAIAhQ4Zg2LBhUNaU4dvEtJycdCeVrCaL3arVPM0qOZn/0qecdEKKKymRXpxpXZzMU+yaFoMkhYIHT9rddPT9IqZUrjjczs4OY8eONXVdSE2mVBZNKpmSYvWL3ebl8VajtDT+i7wa9x4SYhaGuuy0JwxNTX12axPlNlUuMddM+6adgwbo3tcvoz3xrFoNSNIBVx9AbqHfn0YHSBEREfDy8sIbb7yhs33Tpk148uQJZs6cabLKkRrG1pZPKimTAU+e8G+JlbVKMsb/2k1K4snRjo7VojGMkCpBIin+d5N4YdVubdJe108q5UGTXM5/xYhTFVTX76Wh9Q9LmhOrLPNmaQc22v8bCm5UKt3HSrrpB6xiLpo4IhLg91UZgFMhYKkhPEYHSOvXr8e2bduKbW/WrBlef/11CpBIxUgkgLs7/y0oTgXg6GgVfwIWFPDGr9RU/ovZ2dnSNSKk+ispQVw7tykjg38vxZF04nxf+q1NtrZV+1eNfvCj/XNp6x/qz49V1vmyRPojbfUHkejftIMd/ZsxMiz8XhgdIMXHx8PHx6fYdg8PDzx+/NgklSI1nLjYrY0NX/ZcTN6uwgkGWVk81yg7m/cOUhIpIZZVWm6TuKZfVhYPIsTcJjGBXAycxNYm7V89ZRlhWh7inFOGFoAuS/CjP6GnTGbeJYBqAqN/jfv5+eH48eOoV6+ezvbjx4/D19fXZBUjBPb2gJ9f0czbVXCxW5WKV+3p06K4jn4hEVJ1GWptEoOTwkIeNKWnF63lV9rM4IZaUvR/1i5X0nZx/UP9uaREFPxYhtEB0tixYzF16lQUFBSgR48eAICoqCi89957mDFjhskrSGo4mYxPKimT8cQepZL/eVcF5OTwVqPMTB7LVbHYjRBSRtozf2sTk8IN5exo3y9pe2k/6x9fTDQXAzIKfizP6ADp3XffRXJyMiZMmID8/HwAgEKhwMyZMzF79myTV5AQzWK34nxJhYW8H8tC1OqiViPGaPg+IdWVoaRwUnMYvRabKDMzE9euXYNSqUSjRo0gryFLRYhoLTYLyczkXW55eRbpz8rL461G6em8MauGfewJIaTSZNxPgW9IAOw9LbMWW7n/7nVwcEC7du1Qt25d/Pbbb7h27Vp5d0VI2Tk48PmS7O35kDFxCmkzY4wHRY8e8RExjo4UHBFCSHVmdIA0ePBgfP755wD4Gmxt27bF4MGD0bJlS/z4449GV2D16tUICAiAQqFAcHAwzpw5U2r5yMhIBAYGQqlUws/PD9OmTUNubq7BsitWrIAgCJg6darO9tzcXEycOBFubm5wcHBAeHg4EhISjK47sRCFggdJbm48asnLM+vh8vN5z544SLOaTPRNCCGkFEYHSEePHkWXLl0AALt37wZjDKmpqfj000/xf//3f0bta8eOHZg+fToWLlyIc+fOoVWrVggNDUViYqLB8tu2bcOsWbOwcOFCXLt2DRs3bsSOHTswZ86cYmXPnj2L9evXo2XLlsUemzZtGn755Rfs3LkTR44cwaNHjzBo0CCj6k4szMYG8PLiC97m5PChJ2aQmclbjcTJva1s3kpCCCHlZHSAlJaWBldXVwDA/v37ER4eDjs7O/Tr1w83btwwal+rVq3C2LFjMXr0aDRt2hTr1q2DnZ0dNm3aZLD8iRMn0KlTJwwdOhQBAQHo06cPhgwZUqzVKTMzE8OGDcMXX3yBWrVqFav/xo0bsWrVKvTo0QNBQUHYvHkzTpw4gVOnThlVf2Jh4qSSderw+2lpJQ8TMVJhIR809+gRH/5LrUaEEFKzGB0g+fn54eTJk8jKysL+/fvRp08fAEBKSgoURgy/zs/PR0xMDHr16lVUGYkEvXr1wsmTJw0+p2PHjoiJidEERLdv38a+ffvQt29fnXITJ05Ev379dPYtiomJQUFBgc5jjRs3Rt26dUs8LgDk5eUhPT1d50aqCCcnHiQplUULNFVAdjbvThNnFbC3pyG3hBBS0xg9zH/q1KkYNmwYHBwc4O/vj+7duwPgXW8tWrQo836SkpKgUqng5eWls93LywvXr183+JyhQ4ciKSkJnTt3BmMMhYWFGD9+vE4X2/bt23Hu3DmcPXvW4D7i4+Mhk8ng4uJS7Ljx8fEl1jciIgKLFy8u49mRSmfEYrfipHDijLTa9wsL+dMBmvSREEJqMqMDpAkTJiA4OBhxcXHo3bs3JP9OAFO/fn2jc5CMFR0djeXLl2PNmjUIDg7GzZs3MWXKFCxduhTz58/H/fv3MWXKFBw4cMCo1qyymD17NqZPn665n56eDj8/P5Meg5QfY4BaYguVmw/UKluonyRBbaOCWq7Uma6/sFB3Kn/96fqBooUtCSGE1FzlWjEqKCgIQUFBOtv69etn1D7c3d0hlUqLjR5LSEiAt7e3wefMnz8fw4cPx5tvvgkAaNGiBbKysjBu3DjMnTsXMTExSExMxPPPP695jkqlwtGjR/H5558jLy8P3t7eyM/PR2pqqk4rUmnHBQC5XF7j5nqqCvTXJNJeh0il4iPMxGn6+UKOEqgKPYB8GaSPEiAgHWoHJ83MtFIp/19cX0lcw4gQQgjRZrElNWUyGYKCghAVFYUBAwYAANRqNaKiojBp0iSDz8nOzta0WImk/2bOMsbQs2dPXL58Wefx0aNHo3Hjxpg5cyakUimCgoJga2uLqKgohIeHAwBiY2MRFxeHkJAQE58lKYmhYEd79W0x8NF/XDsHWwxuxKn5xYUlpXYCJC4uEDxsISTGQ8hLAbOv2ovdEkIIqVosuub49OnTMXLkSLRt2xbt27dHZGQksrKyMHr0aADAiBEjULt2bURERAAAwsLCsGrVKrRp00bTxTZ//nyEhYVBKpXC0dERzZs31zmGvb093NzcNNudnZ0xZswYTJ8+Ha6urnBycsLkyZMREhKCDh06VO4LUE3l5xcPbAoLdQMf7ZYhMfARhOILRIqrcSsUxq9PxOzswXzqQJKUCElGKtT2TsUXWyKEEGI54gVArQYYvzAI/94XYJpRyeVl0avFa6+9hidPnmDBggWIj49H69atsX//fk3idlxcnE6L0bx58yAIAubNm4eHDx/Cw8MDYWFhWLZsmVHH/fjjjyGRSBAeHo68vDyEhoZizZo1Jj23mkil4uuTpaQUD3wA3ZYeMfAR75uNXA61ty/wVAbJ0ydQK+wAedVY7JYQQqoNMcjRD3T0gh8AEP79h6lZ0V/CAv+fCQKYjQ0gtUGhs5NFVwEv91psNR2txaZLnG06NbVoZfsq1aPFGISUp5AkJwISKZi95Ra7JYSQKk0v0IGaQWDqYoEOUNTyr8l5EIqSO5kg8FZ7qQ1gYwMmtdEKiMQykqJt2o+B/7EdEMCvKaZU1uu30S1IAQEBeOONNzBq1CjUrVu3QpUk1UNWFhAfzye0dnGpYoGRSBDAXN2glskgeZIAIS0FzMmFxvETQqovnaG6ut1XxVp1/v1VyCDotOhAIgGTCGBSGQ90pFLewiOR6gQ1mkBHDJTEmxUr1zxIW7ZswZIlS/DCCy9gzJgxGDhwII3wqoEY4y1G4sowtWpV/XiDOThCZWMLSdK/QZIjTZFNCKnizNB9BRsb/rOBFpxirTqCUPV/uZtBubvYzp07hy1btuC7776DSqXC0KFD8cYbb+gMsa/OanoXW2EhkJzMZ5tWKKxwjbLCQp68nfYUaqU9IKMAnxBiZoaCGnN0X/0b1JTWfWUNLN3FVuEcpIKCAqxZswYzZ85EQUEBWrRogbfffhujR4+GYEVvhLFqcoCUm8tbjdLT+SofVjswTK2GkPoUkqREwFYGprSzdI0IIdZCu0VHHHWl39Lzb0BUaveVOEFbDeu+KgtLB0jlvrQVFBRg9+7d2Lx5Mw4cOIAOHTpgzJgxePDgAebMmYODBw9i27Zt5d09qaIyMoCEBJ6UXWXzjcpKIgFzdYfaxhaSpEQIGWlgDk5W9RcWIaSCytN9pd+qQ91X1ZLRAdK5c+ewefNmfPfdd5BIJBgxYgQ+/vhjNG7cWFNm4MCBaNeunUkrSixLreZD+J884S1GekvZWTXm5Ay1rS0E7eRtq478CKmBTNF9JXZNSSTVvvuKPJvRAVK7du3Qu3dvrF27FgMGDICtgTkK6tWrh9dff90kFSSWV1DAA6OnT3lTZ3XMx2dKOzDv2pAkP4EkPQVqO0eLzr9BSI1mqu4riQBmI+ddVs/qvtIbfk6I0QHS7du34e/vX2oZe3t7bN68udyVIlVHdjbPN8rM5KvbV+sBXzIZ1J7egK0thOQnYHIFoLC27HNCqpCSuq/0hp4D1H1Fqh6jA6TExETEx8cjODhYZ/vp06chlUrRtm1bk1WOWA5jQFoaD45UKusYwm8SUinUbh4QbPl8ScjMAHNwtHStCLGssnZf/Tt1fkndV0zsvpLJqPuKVHlGB0gTJ07Ee++9VyxAevjwIVauXInTp0+brHLEMlSqoiH8cjngUNMmnRYEMGcXnpeUGF80XxI1u5PqgLqvCCkTowOkq1evGpzrqE2bNrh69apJKkUsJy+PtxqlpQGONTwNhxa7JVWaObuvJNJiQ8oNtuoQUo0Z/dteLpcjISEB9evX19n++PFj2NDFw6plZvIh/Lm51WAIv6nI5VB7+QA2NpCkJtNit8T0jOi+EgTwFk7qviLE7IyOaPr06YPZs2fjp59+grOzMwAgNTUVc+bMQe/evU1eQWJ+jBUN4RcEnm9EtNjYQO3pDWYr44vdqlRgdiaeuYxUD/prXDFmuPsKgAD+/zO7r2xswKRS6r4ipJIZHSB9+OGH6Nq1K/z9/dGmTRsAwIULF+Dl5YVvvvnG5BUk5lVYyAOj5GTAzo4vG0IM0F/sNj2V5yXRX941g0r1b5CjgiAGO2qVVhcW0wp0hGd3X+nNikzdV4RUPUYHSLVr18alS5ewdetWXLx4EUqlEqNHj8aQIUMMzolEqq6cHJ5vlJFh5UuGVCLm4AiV1IYWu60uyhr4SKU86JFKeTeWXAHY2oLZ2PLHSgt0KIgmxCqV65Job2+PcePGmboupBKlp/N8o4KCGjSE31SUSqjFSSVTn0Jt7wDYyixdK6JNK/CBmkFQq0pu8RFHYYmBj40NmK2sKPCRSnngI5UWbSOEVHvlbjO4evUq4uLikJ+fr7P95ZdfrnCliPkwxrvTEhN5Lmd1WjKkUtna8rwkmQySpESwQhWgpEklze5ZgY84Uksn8BHA5HYU+BBCjFKumbQHDhyIy5cvQxAEMDHh8N8mCJVKZdoaEpMRg6P4eD63kYwaPSpGIgGr5cYXu32SAGSk80klqTnOeGr1v8GPgcCHMQgSQTfwkUjAJBLdwEdSFOxoAh9xGyGEGMnoAGnKlCmoV68eoqKiUK9ePZw5cwbJycmYMWMGPvzwQ3PUkZiIODO2vT0FRyYjCLTYbWkMBj7/tgCVFvjIlDzHhwIfQoiFGB0gnTx5EocOHYK7uzskEgkkEgk6d+6MiIgIvP322zh//rw56kkqKCODtxzJ5dVzsVlL0yx2+yQBkvRUqO2r+Sybpgx8JBIwiZQCH0JIlWJ0gKRSqeDoyNemcnd3x6NHjxAYGAh/f3/ExsaavIKk4rKyeHAklVKajFnJZFB7+wIyGYSnSda52G1FAx9xVBcFPoQQK2d0gNS8eXNcvHgR9erVQ3BwMN5//33IZDJs2LCh2OzaxPJyc3lwpFLxofzEzKRSqN09+WK3SQlAlgrMvgosZleWwEfN/g1mKPAhhBCjA6R58+YhKysLALBkyRL85z//QZcuXeDm5oYdO3aYvIKk/PLzeXCUl0ej1SqVIIC51ILa1tb8k0qKgQ9TAyq1buCDf0d1GQx8FICtrPTAh+bwIYTUYEYHSKGhoZqfGzZsiOvXr+Pp06eoVauWZiQbsbzCQj7PUWYmLR1iKczegU8qmZwISVoK1A5GzMZZ4cDn39mbtQMfrZwfCnwIIaR0RgVIBQUFUCqVuHDhApo3b67Z7urqavKKkfJTq/lotbQ03nJE10ILUiig9vIFbGyLFruVSEsPfAzm+FDgQwghlcmoAMnW1hZ169aluY6qMMb42mpPnwLOzjTavEqwsYHaw4svdpuSDLACzczNFPgQQkjVZHQX29y5czFnzhx888031HJUBT19ygMkR0fKna1SJBIwVzeo7B14sEOBDyGEVGlGB0iff/45bt68CV9fX/j7+8Pe3l7n8XPnzpmscsQ4qak878jevnpPwWPVaBIqQgixCkYHSAMGDDBDNUhF0USQhBBCiOkYHSAtXLjQHPUgFZCdTRNBEkIIIaZEKbxWLjcXePyYjwjX6+0khBBCSDkZ3YIkkUhKne+IRrhVnoICmgiSEEIIMQejA6Tdu3fr3C8oKMD58+fx1VdfYfHixSarGCmdSsUTsrOyKDgihBBCTM3oAKl///7Ftr3yyito1qwZduzYgTFjxpikYqRk4kSQqak0ESQhhBBiDibLQerQoQOioqKMft7q1asREBAAhUKB4OBgnDlzptTykZGRCAwMhFKphJ+fH6ZNm4bc3FzN42vXrkXLli3h5OQEJycnhISE4LffftPZR/fu3SEIgs5t/PjxRtfdEhgDkpKA5GSaCJKQqkClAk6cAPbs4f9TlgEh1YPRLUiG5OTk4NNPP0Xt2rWNet6OHTswffp0rFu3DsHBwYiMjERoaChiY2Ph6elZrPy2bdswa9YsbNq0CR07dsQ///yDUaNGQRAErFq1CgBQp04drFixAo0aNQJjDF999RX69++P8+fPo1mzZpp9jR07FkuWLNHct7OzK+fZVy6aCJKQqmPfPmDBAj5QQuTjAyxZAvTta7l6EUIqzugASX9RWsYYMjIyYGdnh2+//daofa1atQpjx47F6NGjAQDr1q3D3r17sWnTJsyaNatY+RMnTqBTp04YOnQoACAgIABDhgzB6dOnNWXCwsJ0nrNs2TKsXbsWp06d0gmQ7Ozs4O3tbVR9LS0tjecdKZU0ESQhlrZvHzBu3L9r52mJj+fbN2yo+kGSSgWcPs277D09geBg+sOLEJHRAdLHH3+sEyBJJBJ4eHggODgYtYxYNj4/Px8xMTGYPXu2zr569eqFkydPGnxOx44d8e233+LMmTNo3749bt++jX379mH48OEGy6tUKuzcuRNZWVkICQnReWzr1q349ttv4e3tjbCwMMyfP79KtyJlZhZNBKlQWLo2hNRsKhVvOdIPjgC+TRCAhQuB0NCqG3BQ6xchpTM6QBo1apRJDpyUlASVSgUvLy+d7V5eXrh+/brB5wwdOhRJSUno3LkzGGMoLCzE+PHjMWfOHJ1yly9fRkhICHJzc+Hg4IDdu3ejadOmOvvx9/eHr68vLl26hJkzZyI2Nha7du0qsb55eXnIy8vT3E9PTy/PaZdLTg4PjgSBJoIk1Y81tmKcPq0bWOhjDHj0iJfr2LHy6lVWldX6ZY3vLSEiowOkzZs3w8HBAa+++qrO9p07dyI7OxsjR440WeX0RUdHY/ny5VizZg2Cg4Nx8+ZNTJkyBUuXLsX8+fM15QIDA3HhwgWkpaXhhx9+wMiRI3HkyBFNkDRu3DhN2RYtWsDHxwc9e/bErVu30KBBA4PHjoiIsMg0Bnl5/BdxQQFPyiakOrHWVozERNOWq0yV1fplre8tISKjx0BFRETA3d292HZPT08sX768zPtxd3eHVCpFQkKCzvaEhIQSc4Pmz5+P4cOH480330SLFi0wcOBALF++HBEREVCr1ZpyMpkMDRs2RFBQECIiItCqVSt88sknJdYlODgYAHDz5s0Sy8yePRtpaWma2/3798t8ruUlTgSZkwM4OZn9cIQYZK5RWmIrhn5LjNiKsW+faY5jjvobGENSoXIlMUfdjWn9Kq/Kem8JMSejA6S4uDjUq1ev2HZ/f3/ExcWVeT8ymQxBQUE6UwOo1WpERUUVyxcSZWdnQ6I3rl367584zNCfQ1r71e4e03fhwgUAgI+PT4ll5HK5ZuoA8WZO4kSQGRk01xGxnH37eLfIq68CEyfy/4ODK36Be1YrBsBbMSoaEJir/sHBvDWkpO+lIAC+vrxceZmr7uZu/aqs95YQczM6QPL09MSlS5eKbb948SLc3NyM2tf06dPxxRdf4KuvvsK1a9fw1ltvISsrSzOqbcSIETpJ3GFhYVi7di22b9+OO3fu4MCBA5g/fz7CwsI0gdLs2bNx9OhR3L17F5cvX8bs2bMRHR2NYcOGAQBu3bqFpUuXIiYmBnfv3sXPP/+MESNGoGvXrmjZsqWxL4dZ0ESQpCowZyuAtbdiSKW8qwgo/v0U7y9eXP4uKnPW3dytX5Xx3hJSGYzOQRoyZAjefvttODo6omvXrgCAI0eOYMqUKXj99deN2tdrr72GJ0+eYMGCBYiPj0fr1q2xf/9+TeJ2XFycTovRvHnzIAgC5s2bh4cPH8LDwwNhYWFYtmyZpkxiYiJGjBiBx48fw9nZGS1btsTvv/+O3r17A+AtVwcPHkRkZCSysrLg5+eH8PBwzJs3z9iXwmxSUvhEkE5ONBEksQxz56lYuhXDFHk2ffvyZGZDeTaLF5c/z8bcdRdbv+LjDR9DEPjj5W39qsz8LEoCL5m5X5ua8NoLrLS+KQPy8/MxfPhw7Ny5EzY2PL5Sq9UYMWIE1q1bB5lMZpaKVjXp6elwdnZGWlqaybvbHjzga6w5Opp0t8SCrO2XyYkTvEvnWXbuLN8oLWvfvzZTv7eVUXexhQrQDZLE1q+KjGKrrNeeksBLZu7XprJe+5QUICAAsLc33T6Bsl+/jW6fkMlk2LFjB2JjY7F161bs2rULt27dwqZNm2pMcESIMcyVS2JO5m4FMHcOT2W2Ykil/EI/YAD/v6KBb2XUXWz90h8P4+NT8SH+lZWfZe1J4NY6+KEyXnvxtdm3Dzh61HL5auVeaqRRo0Zo1KiRKetCSLVjrfPNmDtPRczhGTeOXzANtWJUJIenskaZmUNl1b1vX95NZ+qWTXO/t5U1TYE5W33N1QJj7temMl57Q69NnTrAJ58AgwaVb5/lZXQLUnh4OFauXFls+/vvv19sbiRCajJrHqlVGa0A1t6KYS6VWXdTt36JzPneVlaCv7lafa158IO591/Sa/PwIfDKK0ApczmbhdEB0tGjR9HXwKf7pZdewtGjR01SKUKqA2seqWXuUVqivn35+e/cCaxezf8/darirWqVVX9zsOa6azPXe2vuLkhzBjDm/qPJ3K+NOfdfltdm6tTK7W4zOkDKzMw0mGtka2tbqctvEFLVWXqkFlCxX7bmbAXQZo2tGOZmzXXXZo731pxdkOb+Tpn7jyZzd8+ac/9leW3u3weOHTN+3+VldA5SixYtsGPHDixYsEBn+/bt23XWOyOkpqtK882Ud7SQufJUKos119+a625O5pymwNzfqcoa/GCuKRzMuf+ynnNp74+pGR0gzZ8/H4MGDcKtW7fQo0cPAEBUVBS+++477Ny50+QVJMRaVZf5ZsRWAGtlzfW35rqbizmTwM39nbL2wQ/m3H9Zz7mUBS9MzugutrCwMOzZswc3b97EhAkTMGPGDDx48AAHDx7EgAEDzFBFQqyTuXNJrHmkFiEVYa4uSHN/p6x98IM591+W18bPD+jSpXz7Lw+jJ4oszZUrV9C8eXNT7a5Ko4kiSVkZGrbq61ux2ZYBngcRHPzsFqpTp6hbhlRPph6KXxnfKXNO0qnNGmfSftZr88MPphnqX9brd4UDpIyMDHz33Xf48ssvERMTA1UNWYGQAiRiDHP9sqqsX7aE1BSV8Z0y1x9N1YGh18bPD4iMNN08SGYPkI4ePYovv/wSu3btgq+vLwYNGoTw8HC0a9eu3JW2JhQgkaqCftkSYlqV8Z2ytuWHKpP42ty5AzRvDvTpY9rXxiwBUnx8PLZs2YKNGzciPT0dgwcPxrp163Dx4sUaN4KNAiRSldAvW0JMi75TlmfptdjKPIotLCwMR48eRb9+/RAZGYkXX3wRUqkU69atM0mFCSHlR6OdCDEt+k6RMgdIv/32G95++2289dZbtAYbIYQQQqq1Mg/z//PPP5GRkYGgoCAEBwfj888/R1JSkjnrRgghhBBiEWUOkDp06IAvvvgCjx8/xv/+9z9s374dvr6+UKvVOHDgADIyMsxZT0IIIYSQSmP0RJH29vZ444038Oeff+Ly5cuYMWMGVqxYAU9PT7z88svmqCMhhBBCSKUyOkDSFhgYiPfffx8PHjzAd999Z6o6EUIIIYRYlNFrsRkilUoxYMAAWmqEWC0a0ksIIUSbSQIkQqyZoUnhfHz4Omo00SIhhNRMFepiI8TaicsKaAdHAF+Ladw4/jghhJCahwIkUmOpVLzlyNBc8uK2hQt5OUIIITULBUikxjp9unjLkTbGgEePeDlCCCE1CwVIpMZKTDRtOUIIIdUHBUikxvL0NG05Qggh1QeNYiMmY+6h8qbef3AwH60WH284D0kQ+OPBweU/BiGEEOtEARIxCXMPlTfH/qVS/vxx43gwpB0kCQL/f/Fimg+JEEJqIupiIxVm7qHy5tx/377Ahg2At7fudh8fvp3mQSKEkJpJYMxQ5wJ5lvT0dDg7OyMtLQ1OTk4m3feDB0BWFuDoaNLdmoVKxbugShoNJnZTnTpVvpYYc+9f+zg0kzYhhFQdKSlAQABgb2/a/Zb1+k1dbKRCjBkq37Fj1du/SCqt2PMJIYRUL9TFRirE3EPlaSg+IYQQS6AAiVSIuYfK01B8QgghlkABEqkQcai8OOpLnyAAvr7lHypv7v0TQgghhlCARCpEHCoPFA9iTDFU3tz7J4QQQgyhAIlUmLmHytNQfEIIIZXN4gHS6tWrERAQAIVCgeDgYJw5c6bU8pGRkQgMDIRSqYSfnx+mTZuG3NxczeNr165Fy5Yt4eTkBCcnJ4SEhOC3337T2Udubi4mTpwINzc3ODg4IDw8HAkJCWY5v5qib18+kmznTmD1av7/qVOmC17MvX9CCCFEm0WH+e/YsQPTp0/HunXrEBwcjMjISISGhiI2NhaeBrJut23bhlmzZmHTpk3o2LEj/vnnH4waNQqCIGDVqlUAgDp16mDFihVo1KgRGGP46quv0L9/f5w/fx7NmjUDAEybNg179+7Fzp074ezsjEmTJmHQoEE4fvx4pZ5/dWPuofI0FJ8QQkhlsehEkcHBwWjXrh0+//xzAIBarYafnx8mT56MWbNmFSs/adIkXLt2DVFRUZptM2bMwOnTp/Hnn3+WeBxXV1d88MEHGDNmDNLS0uDh4YFt27bhlVdeAQBcv34dTZo0wcmTJ9GhQ4cy1Z0miiSEEELMx9ITRVqsiy0/Px8xMTHo1atXUWUkEvTq1QsnT540+JyOHTsiJiZG0w13+/Zt7Nu3D31L6GdRqVTYvn07srKyEBISAgCIiYlBQUGBznEbN26MunXrlnhcAMjLy0N6errOjRBCCCHVk8W62JKSkqBSqeDl5aWz3cvLC9evXzf4nKFDhyIpKQmdO3cGYwyFhYUYP3485syZo1Pu8uXLCAkJQW5uLhwcHLB79240bdoUABAfHw+ZTAYXF5dix42Pjy+xvhEREVi8eHE5zpQQQggh1sbiSdrGiI6OxvLly7FmzRqcO3cOu3btwt69e7F06VKdcoGBgbhw4QJOnz6Nt956CyNHjsTVq1crdOzZs2cjLS1Nc7t//36F9kcIIYSQqstiLUju7u6QSqXFRo8lJCTAW38897/mz5+P4cOH48033wQAtGjRAllZWRg3bhzmzp0LiYTHezKZDA0bNgQABAUF4ezZs/jkk0+wfv16eHt7Iz8/H6mpqTqtSKUdFwDkcjnkcnlFTpkQQgghVsJiLUgymQxBQUE6CddqtRpRUVGafCF92dnZmiBIJP13hsDScs3VajXy8vIA8IDJ1tZW57ixsbGIi4sr8biEEEIIqVksOsx/+vTpGDlyJNq2bYv27dsjMjISWVlZGD16NABgxIgRqF27NiIiIgAAYWFhWLVqFdq0aYPg4GDcvHkT8+fPR1hYmCZQmj17Nl566SXUrVsXGRkZ2LZtG6Kjo/H7778DAJydnTFmzBhMnz4drq6ucHJywuTJkxESElLmEWzWSqXicwklJvK1y4KDaQZqQgghxBCLBkivvfYanjx5ggULFiA+Ph6tW7fG/v37NYnbcXFxOi1G8+bNgyAImDdvHh4+fAgPDw+EhYVh2bJlmjKJiYkYMWIEHj9+DGdnZ7Rs2RK///47evfurSnz8ccfQyKRIDw8HHl5eQgNDcWaNWsq78QtYN8+YMEC4PHjom0+PnwZD5pskRBCCNFl0XmQrJk1zYO0bx8wbhyg/06La5nRch2EEEKqmho7DxKpHCoVbzkyFAaL2xYu5OUIIYQQwlGAVM2dPq3braaPMeDRI16OEEIIIRwFSNVcYqJpyxFCCCE1AQVI1ZyBNX8rVI4QQgipCShAquaCg/loNTEhW58gAL6+vBwhhBBCOAqQqjmplA/lB4oHSeL9xYtpPiRCCCFEGwVINUDfvnwov/5KKj4+NMSfEEIIMcSiE0WSytO3LxAaSjNpE0IIIWVBAVINIpUCHTtauhaEEEJI1UddbIQQQggheihAIoQQQgjRQwESIYQQQogeCpAIIYQQQvRQgEQIIYQQoodGsVUhKhVw7Bhw5Qrg5AS88AINwyeEEEIsgQKkKmLXLmDKFODBg6JtPj58FmyayJEQQgipXNTFVgXs2gW88opucAQA8fHAuHHAvn2WqRchhBBSU1GAZGEqFW85Yqz4Y+K2hQt5OUIIIYRUDupis7Bjx4q3HGljDHj0iC8RQrNgE0JMRaUC8vOB3NzijzEGSCT8JghFPwNFP2s/Rkh1RAGShT1+XLZyiYnmrQchpPrLzy+6SaWATAZ4eABKJX+cMUCtLroVFvKb+DNjPLAqKODbxPLicwWh6FjaQZRUym+2trplCKnKKECyMB+fspXz9DRvPQgh1Y9aDeTl8ZtazQMipZIHRQoFIJcb1wKkHRSJN+374s8qlW6AVVjIg7KsLL4fiYQHSzY2FDSRqosCJAvr0gWoUwd4+NBwHpIg8CAqOLjy60YIsT4FBTwgKijgvz/kcsDNDbCz4z/LZOXftxhMGTv9CGNFQZJYv+zsoqCJsaIWJgqaSFVBAZKFSaXAJ5/wUWyCoBskib8gFi+m+ZAIqWrEriaxhUTM2xG7k7R/NufFXq3mgUZeHq+TrS1vHXJ3L2olsvTvD0Hg9bK1LdomBk0FBfyWmwvk5PCfxaBJIuEBHQVNZSN2gWrf1Oqi103/f0Pb9F/jZz1X/KxXRxQgVQGDBgE//GB4HqTFi2keJEIsSb+rSKXiFyIbG35TKHi3lVTKL+5iK4mYBC2WF5kiiCos5AFRfj6/r1AALi6AvX1RK1FVDyb0gyZn5+JBk9jSVFDA/1eri3fPVdeLsyH6gY/4s0g730si4Z8LW9uiz5+YL1bS/4wVldX/X7uM9v9ijppMxm/VKZClAKmKGDQI6N+fZtImxFLEi7N+ICReyG1sAAcHHoBoX6ANfUe1c3H0b2KStHgMQ0GUfgAlkRR1URUW8uPK5YCrK78IKhTV43dFWVqatIOmnJyi183Gpui1En+2pgu1duuP9mdHzO0SRwyKnw0xIBFbCA3dKlIX/Z9L+l98TzIz+fuRmVnU8ie2/lkrK6569SOVAt27Aw0b8ibm6vALj5CqSOweE/8XL0BSKf+Fbm/PW4XEi7XYWlTWC672X/IlKak7RDuIEoMClYof39GRB2kKhXW0EpmCoaAJ0M1pElvuxAAyJ0e3e0n7/dC+Vcbrp520buimfZ7agbGdXVGLjKG6m7PlzFAXXElsbXlda9UqCpZycoCMDB7IFhbyz64YzFnTZ5YCJEJIhYjN8mKrifZ9/Zv+4+LztX9pat/Xf6y0x0vLqxAVFuoGQjIZ79qRy4tahGxsKqfbRhCKAq+SaAdRYp0JV9Jrpx1wiq10YpekmKclBlIi8b3Qb4EyRHvUntjypx0EiZ9t8bMnTncgBmRisKf9WdMOfMRt1hRIiMTzcnDgOXDia56dzf/oT03lr40YLOkHvVUNfd0IIaUqaUJB8SKgfRN/sWvfxIuOeBMvANoTDYpltfet/f+ztgGG8yi0/0IHytY9VpWUJYgiup7VcqcdOIk/67dAiYGU+DnSDsi1P8vizVCArd09qv2zNQY+5SEIRd2/zs789czN5a9tRkbR/+LrJ5NVve8jfe0IIcWUNKGg2EReUkBkKECqKRcEYh1KC6C0W+y0gyjt2cT1A5+aFPRUhFTKu67t7Q13x2VmFo3CFPOXLI0CJEIqSU4OvwFFf2lWlb+aDE0oqFCUf0JBQqyRdoudXG7p2lRfglAUBDk66nbHZWXxLrnsbEvXkgIkQiqFOOrG15f/csjO5s3NmZlF3UBiQmZlDZPVzssQm8NdXXnCpZgETAgh5iaR8EERSiWfrqKwkP9+zM+37O8hCpAIMbPMTN507+vL++IB3sQsjqASu7KysoovxyAGTabIQWGsaO6c6jxUnBBi3cQpNSyNAiRCzCgtjQci3t7Fv/BiLoRCwe+7u+sOWc7J4S1N4uzCYvO/2Edfli4vlaooKGKMB0Q1cag4IYQYiwIkQsyAMR4cyeV8RnQ7u7I9T+xis7PjTc2MFc31IrYuiaM/VKqiWYW1u+bEZScKCngAJpfzxY7FViIaEUUIIc9m8bTL1atXIyAgAAqFAsHBwThz5kyp5SMjIxEYGAilUgk/Pz9MmzYNuVrjjyMiItCuXTs4OjrC09MTAwYMQGxsrM4+unfvDkEQdG7jx483y/mRmketBlJSeJBTu3bZgyNDxMVGHRx4V5ifH1CvHhAQANSty1udbG15UJSWBjx9yn+2s+OLIAcE8Ju7O98HBUeEEFI2Fv11uWPHDkyfPh3r1q1DcHAwIiMjERoaitjYWHh6ehYrv23bNsyaNQubNm1Cx44d8c8//2DUqFEQBAGrVq0CABw5cgQTJ05Eu3btUFhYiDlz5qBPnz64evUq7O3tNfsaO3YslixZorlvV5GrGCH/Uql4oOLsDHh5mSfBUCotSmh0ctJdKV3sRqvqE7ARQkhVZ9EAadWqVRg7dixGjx4NAFi3bh327t2LTZs2YdasWcXKnzhxAp06dcLQoUMBAAEBARgyZAhOnz6tKbN//36d52zZsgWenp6IiYlB165dNdvt7Ozg7e1tjtMiNVRhIZCezhOwvbwqr7WmpKUYCCGElJ/Futjy8/MRExODXr16FVVGIkGvXr1w8uRJg8/p2LEjYmJiNN1wt2/fxr59+9C3lOXu09LSAACurq4627du3Qp3d3c0b94cs2fPRvYzJl3Iy8tDenq6zo0QUX4+D47c3XlCNnVlEUKIdbPYr/GkpCSoVCp4eXnpbPfy8sL169cNPmfo0KFISkpC586dwRhDYWEhxo8fjzlz5hgsr1arMXXqVHTq1AnNmzfX2Y+/vz98fX1x6dIlzJw5E7Gxsdi1a1eJ9Y2IiMDixYvLcaakuhPXGvLyAtzcaEJFQgipDqzq79zo6GgsX74ca9asQXBwMG7evIkpU6Zg6dKlmD9/frHyEydOxJUrV/Dnn3/qbB83bpzm5xYtWsDHxwc9e/bErVu30KBBA4PHnj17NqZPn665n56eDj8/PxOdGbFWOTk8QPL25knUNGSeEEKqB4sFSO7u7pBKpUhISNDZnpCQUGJu0Pz58zF8+HC8+eabAHhwk5WVhXHjxmHu3LmQaP3pPmnSJPz66684evQo6tSpU2pdgoODAQA3b94sMUCSy+WQ09zzREtWFk/K9vXlQ/IJIYRUHxbrDJDJZAgKCkJUVJRmm1qtRlRUFEJCQgw+Jzs7WycIAgDpv1P/sn+XXWaMYdKkSdi9ezcOHTqEevXqPbMuFy5cAAD4+PiU51RIDZSRwUeM1a5NwREhhFRHFu1imz59OkaOHIm2bduiffv2iIyMRFZWlmZU24gRI1C7dm1EREQAAMLCwrBq1Sq0adNG08U2f/58hIWFaQKliRMnYtu2bfjpp5/g6OiI+Ph4AICzszOUSiVu3bqFbdu2oW/fvnBzc8OlS5cwbdo0dO3aFS1btrTMC0GshvYEkN7efGVqQggh1Y9FA6TXXnsNT548wYIFCxAfH4/WrVtj//79msTtuLg4nRajefPmQRAEzJs3Dw8fPoSHhwfCwsKwbNkyTZm1a9cC4JNBatu8eTNGjRoFmUyGgwcPaoIxPz8/hIeHY968eeY/YWLVGANSU/kkjN7efB4iQggh1ZPAxL4pYpT09HQ4OzsjLS0NTk5OJt33gwc8v8XR0aS7JRWgVvPgyNGRB0eUjkYIIdaprNdvqxrFRogliLNju7jwofw0ISMhhFR/FCARUoqCAp6Q7erKF3ylCSAJIaRmoF/3hJQgP593dXp48BtNAEkIITUHBUiEGJCby2/i7Ng0ASQhhNQsFCARoic7m3et+fjwvCMKjgghpOahAImQfzHGu9QY47NjOztbukaEEEIshQIkUuMxxtdUy83lcxt5eQEODpauFSGEEEuiAInUaDk5/KZU8mVDHB1ppBohhBAKkEgNJQZGCgXPNXJ2psCIEEJIEbokkBolN5cnYYtrqTk708SPhBBCiqMAidQIeXk8MJLJeGDk5MR/JoQQQgyhAIlUa/n5QGYmD4Y8PHiLEa2jRggh5FkoQCLVUkEBD4xsbIoCI4XC0rUihBBiLShAItVKQQGfy0gi4eunubjwEWqEEEKIMShAItVCYSFvMRIEHhTVqkWBESGEkPKjAImYVGEhT4gWBN6KI5EAUin/3xxLdqhUPDBijHejiYERLQ9CCCGkIihAIiYjzi3k6MgDF8Z4krRaze8DfJt+8CQIRUGUeP9ZVCrelaZW8xFptWoBdnYUGBFCCDENCpBIhTEGpKfz4KZOHd6SIwhFgZH4v/bPhYU8X6iggN8vKNAtLwh8v/qtUBIJD7oKC4sCI3t7CowIIYSYFgVIpEIKC3lw5OAAeHryVhyRGNA8i1pdPJjSDqrEgEgMphQKnoBtb1+2/RNCCCHGogCJlJu4wKuHB+DmVv6lOsRA6lnPZ4wHTubKZyKEEEJEFCARozEGpKXxJTrq1OFdXZURsIi5SoQQQoi5UYBEjFJQAGRk8ERsT08aSk8IIaR6ogCJlFl2Ns8HqmiXGiGEEFLV0SWOPJNazROxbW2B2rUrr0uNEEIIsRQKkEipxMVenZ15yxGtZ0YIIaQmoACJlCgri+cceXryLjVKkCaEEFJTUIBEilGr+Sg1uZyPUnN0pC41QgghNQsFSESHdpeapycPkgghhJCahgIkopGZyWes9vLiM1VTlxohhJCaigIkApWKj1JTKAAfH96lRgghhNRkFCDVcHl5PBnbxYWPUqMuNUIIIYQCpBotI4MnZHt78y41WviVEEII4ShAqoFUKj5KTank+UYODpauESGEEFK1UIBUQzDGR6jl5vJWo1q1eJeaTGbpmhFCCCFVj8U7VVavXo2AgAAoFAoEBwfjzJkzpZaPjIxEYGAglEol/Pz8MG3aNOTm5moej4iIQLt27eDo6AhPT08MGDAAsbGxOvvIzc3FxIkT4ebmBgcHB4SHhyMhIcEs52dJjPGAKDWV3woL+fB9Pz+ejE3BESGEEGKYRQOkHTt2YPr06Vi4cCHOnTuHVq1aITQ0FImJiQbLb9u2DbNmzcLChQtx7do1bNy4ETt27MCcOXM0ZY4cOYKJEyfi1KlTOHDgAAoKCtCnTx9kZWVpykybNg2//PILdu7ciSNHjuDRo0cYNGiQ2c+3MqjVQE4OkJLCgyKxtahuXSAgAPD15WupUb4RIYQQUjKBMcYsdfDg4GC0a9cOn3/+OQBArVbDz88PkydPxqxZs4qVnzRpEq5du4aoqCjNthkzZuD06dP4888/DR7jyZMn8PT0xJEjR9C1a1ekpaXBw8MD27ZtwyuvvAIAuH79Opo0aYKTJ0+iQ4cOZap7eno6nJ2dkZaWBicnJ2NPvVQPHvCRZWUdbq9W89FoeXn8vlzO84rs7fnQfRvqSCWEEEIAlP36bbF2hPz8fMTExKBXr15FlZFI0KtXL5w8edLgczp27IiYmBhNN9zt27exb98+9O3bt8TjpKWlAQBcXV0BADExMSgoKNA5buPGjVG3bt0SjwsAeXl5SE9P17lZkkoFZGfzliKxKh4egL8/bykSk68pOCKEEEKMZ7HLZ1JSElQqFby8vHS2e3l54fr16wafM3ToUCQlJaFz585gjKGwsBDjx4/X6WLTplarMXXqVHTq1AnNmzcHAMTHx0Mmk8HFxaXYcePj40usb0REBBYvXmzEGZpeYSFvJcrP511kCgUPhOzs+M/UbUYIIYSYhlVdUqOjo7F8+XKsWbMG586dw65du7B3714sXbrUYPmJEyfiypUr2L59e4WPPXv2bKSlpWlu9+/fr/A+y6KwkC8B8vQpbzGyteXzFgUE8NYid3ceIFFwRAghhJiOxVqQ3N3dIZVKi40eS0hIgLe3t8HnzJ8/H8OHD8ebb74JAGjRogWysrIwbtw4zJ07FxKtKGHSpEn49ddfcfToUdSpU0ez3dvbG/n5+UhNTdVpRSrtuAAgl8shr8RppvPzeVBkY8PnK/Lw4P/L5YAgVFo1CCGEkBrJYu0OMpkMQUFBOgnXarUaUVFRCAkJMfic7OxsnSAIAKT/rqgq5pozxjBp0iTs3r0bhw4dQr169XTKBwUFwdbWVue4sbGxiIuLK/G4lc3Wli/9UacObymqW5ePRFMoKDgihBBCKoNFU3inT5+OkSNHom3btmjfvj0iIyORlZWF0aNHAwBGjBiB2rVrIyIiAgAQFhaGVatWoU2bNggODsbNmzcxf/58hIWFaQKliRMnYtu2bfjpp5/g6OioyStydnaGUqmEs7MzxowZg+nTp8PV1RVOTk6YPHkyQkJCyjyCzdw8PSkQIoQQQizJogHSa6+9hidPnmDBggWIj49H69atsX//fk3idlxcnE6L0bx58yAIAubNm4eHDx/Cw8MDYWFhWLZsmabM2rVrAQDdu3fXOdbmzZsxatQoAMDHH38MiUSC8PBw5OXlITQ0FGvWrDHvyRqBgiNCCCHEsiw6D5I1M+c8SIQQQggxjyo/DxIhhBBCSFVFARIhhBBCiB4KkAghhBBC9FCARAghhBCihwIkQgghhBA9FCARQgghhOihAIkQQgghRA8FSIQQQggheihAIoQQQgjRQwESIYQQQogeCpAIIYQQQvRQgEQIIYQQosfG0hWwVuIav+np6RauCSGEEELKSrxui9fxklCAVE4ZGRkAAD8/PwvXhBBCCCHGysjIgLOzc4mPC+xZIRQxSK1W49GjR3B0dIQgCJaujtmkp6fDz88P9+/fh5OTk6WrY1Y16VyBmnW+dK7VV006XzpX02CMISMjA76+vpBISs40ohakcpJIJKhTp46lq1FpnJycqv0XUlSTzhWoWedL51p91aTzpXOtuNJajkSUpE0IIYQQoocCJEIIIYQQPRQgkVLJ5XIsXLgQcrnc0lUxu5p0rkDNOl861+qrJp0vnWvloiRtQgghhBA91IJECCGEEKKHAiRCCCGEED0UIBFCCCGE6KEAiRBCCCFEDwVINVhERATatWsHR0dHeHp6YsCAAYiNjS31OVu2bIEgCDo3hUJRSTUuv0WLFhWrd+PGjUt9zs6dO9G4cWMoFAq0aNEC+/btq6TaVlxAQECx8xUEARMnTjRY3pre16NHjyIsLAy+vr4QBAF79uzReZwxhgULFsDHxwdKpRK9evXCjRs3nrnf1atXIyAgAAqFAsHBwThz5oyZzqDsSjvXgoICzJw5Ey1atIC9vT18fX0xYsQIPHr0qNR9lue7UFme9d6OGjWqWN1ffPHFZ+7X2t5bAAa/v4Ig4IMPPihxn1X1vS3LtSY3NxcTJ06Em5sbHBwcEB4ejoSEhFL3W97vellRgFSDHTlyBBMnTsSpU6dw4MABFBQUoE+fPsjKyir1eU5OTnj8+LHmdu/evUqqccU0a9ZMp95//vlniWVPnDiBIUOGYMyYMTh//jwGDBiAAQMG4MqVK5VY4/I7e/aszrkeOHAAAPDqq6+W+BxreV+zsrLQqlUrrF692uDj77//Pj799FOsW7cOp0+fhr29PUJDQ5Gbm1viPnfs2IHp06dj4cKFOHfuHFq1aoXQ0FAkJiaa6zTKpLRzzc7Oxrlz5zB//nycO3cOu3btQmxsLF5++eVn7teY70JletZ7CwAvvviiTt2/++67Uvdpje8tAJ1zfPz4MTZt2gRBEBAeHl7qfqvie1uWa820adPwyy+/YOfOnThy5AgePXqEQYMGlbrf8nzXjcII+VdiYiIDwI4cOVJimc2bNzNnZ+fKq5SJLFy4kLVq1arM5QcPHsz69eunsy04OJj973//M3HNKseUKVNYgwYNmFqtNvi4tb6vANju3bs199VqNfP29mYffPCBZltqaiqTy+Xsu+++K3E/7du3ZxMnTtTcV6lUzNfXl0VERJil3uWhf66GnDlzhgFg9+7dK7GMsd8FSzF0viNHjmT9+/c3aj/V5b3t378/69GjR6llrOW91b/WpKamMltbW7Zz505NmWvXrjEA7OTJkwb3Ud7vujGoBYlopKWlAQBcXV1LLZeZmQl/f3/4+fmhf//++PvvvyujehV248YN+Pr6on79+hg2bBji4uJKLHvy5En06tVLZ1toaChOnjxp7mqaXH5+Pr799lu88cYbpS6sbK3vq7Y7d+4gPj5e571zdnZGcHBwie9dfn4+YmJidJ4jkUjQq1cvq3u/09LSIAgCXFxcSi1nzHehqomOjoanpycCAwPx1ltvITk5ucSy1eW9TUhIwN69ezFmzJhnlrWG91b/WhMTE4OCggKd96lx48aoW7duie9Teb7rxqIAiQAA1Go1pk6dik6dOqF58+YllgsMDMSmTZvw008/4dtvv4VarUbHjh3x4MGDSqyt8YKDg7Flyxbs378fa9euxZ07d9ClSxdkZGQYLB8fHw8vLy+dbV5eXoiPj6+M6prUnj17kJqailGjRpVYxlrfV33i+2PMe5eUlASVSmX173dubi5mzpyJIUOGlLq4p7HfharkxRdfxNdff42oqCisXLkSR44cwUsvvQSVSmWwfHV5b7/66is4Ojo+s8vJGt5bQ9ea+Ph4yGSyYoF9ae9Teb7rxrIxyV6I1Zs4cSKuXLnyzP7qkJAQhISEaO537NgRTZo0wfr167F06VJzV7PcXnrpJc3PLVu2RHBwMPz9/fH999+X6a8ya7Zx40a89NJL8PX1LbGMtb6vhCsoKMDgwYPBGMPatWtLLWvN34XXX39d83OLFi3QsmVLNGjQANHR0ejZs6cFa2ZemzZtwrBhw545cMIa3tuyXmuqAmpBIpg0aRJ+/fVXHD58GHXq1DHquba2tmjTpg1u3rxpptqZh4uLC5577rkS6+3t7V1sBEVCQgK8vb0ro3omc+/ePRw8eBBvvvmmUc+z1vdVfH+Mee/c3d0hlUqt9v0Wg6N79+7hwIEDpbYeGfKs70JVVr9+fbi7u5dYd2t/bwHg2LFjiI2NNfo7DFS997aka423tzfy8/ORmpqqU76096k833VjUYBUgzHGMGnSJOzevRuHDh1CvXr1jN6HSqXC5cuX4ePjY4Yamk9mZiZu3bpVYr1DQkIQFRWls+3AgQM6rSzWYPPmzfD09ES/fv2Mep61vq/16tWDt7e3znuXnp6O06dPl/jeyWQyBAUF6TxHrVYjKiqqyr/fYnB048YNHDx4EG5ubkbv41nfharswYMHSE5OLrHu1vzeijZu3IigoCC0atXK6OdWlff2WdeaoKAg2Nra6rxPsbGxiIuLK/F9Ks93vTwVJzXUW2+9xZydnVl0dDR7/Pix5padna0pM3z4cDZr1izN/cWLF7Pff/+d3bp1i8XExLDXX3+dKRQK9vfff1viFMpsxowZLDo6mt25c4cdP36c9erVi7m7u7PExETGWPHzPH78OLOxsWEffvghu3btGlu4cCGztbVlly9fttQpGE2lUrG6deuymTNnFnvMmt/XjIwMdv78eXb+/HkGgK1atYqdP39eM3JrxYoVzMXFhf3000/s0qVLrH///qxevXosJydHs48ePXqwzz77THN/+/btTC6Xsy1btrCrV6+ycePGMRcXFxYfH1/p56ettHPNz89nL7/8MqtTpw67cOGCznc4Ly9Psw/9c33Wd8GSSjvfjIwM9s4777CTJ0+yO3fusIMHD7Lnn3+eNWrUiOXm5mr2UR3eW1FaWhqzs7Nja9euNbgPa3lvy3KtGT9+PKtbty47dOgQ++uvv1hISAgLCQnR2U9gYCDbtWuX5n5ZvusVQQFSDQbA4G3z5s2aMt26dWMjR47U3J86dSqrW7cuk8lkzMvLi/Xt25edO3eu8itvpNdee435+PgwmUzGateuzV577TV28+ZNzeP658kYY99//z177rnnmEwmY82aNWN79+6t5FpXzO+//84AsNjY2GKPWfP7evjwYYOfW/F81Go1mz9/PvPy8mJyuZz17Nmz2Gvg7+/PFi5cqLPts88+07wG7du3Z6dOnaqkMypZaed6586dEr/Dhw8f1uxD/1yf9V2wpNLONzs7m/Xp04d5eHgwW1tb5u/vz8aOHVss0KkO761o/fr1TKlUstTUVIP7sJb3tizXmpycHDZhwgRWq1YtZmdnxwYOHMgeP35cbD/azynLd70ihH8PSgghhBBC/kU5SIQQQggheihAIoQQQgjRQwESIYQQQogeCpAIIYQQQvRQgEQIIYQQoocCJEIIIYQQPRQgEUIIIYTooQCJEFLM3bt3IQgCLly4YOmqaFy/fh0dOnSAQqFA69atzX687t27Y+rUqWY/zrMIgoA9e/aU+DhjDOPGjYOrq2uVe88IsWYUIBFSBY0aNQqCIGDFihU62/fs2QNBECxUK8tauHAh7O3tERsbW2ydvJps//792LJlC3799Vc8fvwYzZs3N8l+R40ahQEDBphkX4RYIwqQCKmiFAoFVq5ciZSUFEtXxWTy8/PL/dxbt26hc+fO8Pf3L/OirBU5nrUQFyPt2LEjvL29YWNjY+kq6VCpVFCr1ZauBiFGowCJkCqqV69e8Pb2RkRERIllFi1aVKy7KTIyEgEBAZr7YkvA8uXL4eXlBRcXFyxZsgSFhYV499134erqijp16mDz5s3F9n/9+nV07NgRCoUCzZs3x5EjR3Qev3LlCl566SU4ODjAy8sLw4cPR1JSkubx7t27Y9KkSZg6dSrc3d0RGhpq8DzUajWWLFmCOnXqQC6Xo3Xr1ti/f7/mcUEQEBMTgyVLlkAQBCxatMjgfko63rPqmZWVhREjRsDBwQE+Pj746KOPiu3bUFeXi4sLtmzZorn/4MEDDBkyBK6urrC3t0fbtm1x+vRpzeM//fQTnn/+eSgUCtSvXx+LFy9GYWGh5vEbN26ga9euUCgUaNq0KQ4cOGDwPEWjRo3C5MmTERcXB0EQNO+7Wq1GREQE6tWrB6VSiVatWuGHH37QPE+lUmHMmDGaxwMDA/HJJ59oHl+0aBG++uor/PTTTxAEAYIgIDo6GtHR0RAEAampqZqyFy5cgCAIuHv3LgBgy5YtcHFxwc8//4ymTZtCLpcjLi4OeXl5eOedd1C7dm3Y29sjODgY0dHRmv3cu3cPYWFhqFWrFuzt7dGsWTPs27ev1PMnxJwoQCKkipJKpVi+fDk+++wzPHjwoEL7OnToEB49eoSjR49i1apVWLhwIf7zn/+gVq1aOH36NMaPH4///e9/xY7z7rvvYsaMGTh//jxCQkIQFhaG5ORkAEBqaip69OiBNm3a4K+//sL+/fuRkJCAwYMH6+zjq6++gkwmw/Hjx7Fu3TqD9fvkk0/w0Ucf4cMPP8SlS5cQGhqKl19+GTdu3AAAPH78GM2aNcOMGTPw+PFjvPPOOyWeq/7xylLPd999F0eOHMFPP/2EP/74A9HR0Th37pxRr3FmZia6deuGhw8f4ueff8bFixfx3nvvaVpPjh07hhEjRmDKlCm4evUq1q9fjy1btmDZsmUAeFAzaNAgyGQynD59GuvWrcPMmTNLPeYnn3yiCSwfP36Ms2fPAgAiIiLw9ddfY926dfj7778xbdo0/Pe//9UEuGq1GnXq1MHOnTtx9epVLFiwAHPmzMH3338PAHjnnXcwePBgvPjii3j8+DEeP36Mjh07lvm1yM7OxsqVK/Hll1/i77//hqenJyZNmoSTJ09i+/btuHTpEl599VW8+OKLmvd44sSJyMvLw9GjR3H58mWsXLkSDg4ORr0HhJiUyZa9JYSYzMiRI1n//v0ZY4x16NCBvfHGG4wxxnbv3s20v7YLFy5krVq10nnuxx9/zPz9/XX25e/vz1QqlWZbYGAg69Kli+Z+YWEhs7e3Z9999x1jjGlWil+xYoWmTEFBAatTpw5buXIlY4yxpUuXsj59+ugc+/79+wyAZkXtbt26sTZt2jzzfH19fdmyZct0trVr145NmDBBc79Vq1bFVmnXZ+h4z6pnRkYGk8lk7Pvvv9c8npyczJRKJZsyZYpmGwC2e/dunf04OztrVhdfv349c3R0ZMnJyQbr1rNnT7Z8+XKdbd988w3z8fFhjDH2+++/MxsbG/bw4UPN47/99pvB42rTf79zc3OZnZ0dO3HihE65MWPGsCFDhpS4n4kTJ7Lw8HDNfe3PoEhcgT4lJUWz7fz58wwAu3PnDmOMsc2bNzMA7MKFC5oy9+7dY1KpVOfcGOOvyezZsxljjLVo0YItWrSoxPoRUtmqVmc1IaSYlStXokePHqW2mjxLs2bNIJEUNRh7eXnpJPNKpVK4ubkhMTFR53khISGan21sbNC2bVtcu3YNAHDx4kUcPnzY4F/5t27dwnPPPQcACAoKKrVu6enpePToETp16qSzvVOnTrh48WIZz7CI/vGeVc+cnBzk5+cjODhYs93V1RWBgYFGHffChQto06YNXF1dDT5+8eJFHD9+XNNiBPCurtzcXGRnZ+PatWvw8/ODr6+v5nHt17+sbt68iezsbPTu3Vtne35+Ptq0aaO5v3r1amzatAlxcXGa18BUowNlMhlatmypuX/58mWoVCrNZ0KUl5enySd7++238dZbb+GPP/5Ar169EB4errMPQiobBUiEVHFdu3ZFaGgoZs+ejVGjRuk8JpFIwBjT2VZQUFBsH7a2tjr3BUEwuM2YZNrMzEyEhYVh5cqVxR7z8fHR/Gxvb1/mfZqC/vGeVc+bN2+Wab+CIJT6WiuVylKfn5mZicWLF2PQoEHFHlMoFGWqQ1lkZmYCAPbu3YvatWvrPCaXywEA27dvxzvvvIOPPvoIISEhcHR0xAcffKCTL2WIGGRrvw6GPm9KpVJntGVmZiakUiliYmIglUp1yoqB65tvvonQ0FDs3bsXf/zxByIiIvDRRx9h8uTJZT11QkyKAiRCrMCKFSvQunXrYq0aHh4eiI+PB2NMc0Ey5Tw4p06dQteuXQEAhYWFiImJwaRJkwAAzz//PH788UcEBARUaOSUk5MTfH19cfz4cXTr1k2z/fjx42jfvn3FTqAM9WzQoAFsbW1x+vRp1K1bFwCQkpKCf/75R6c+Hh4eePz4seb+jRs3kJ2drbnfsmVLfPnll3j69KnBVqTnn38esbGxaNiwocF6NmnSBPfv38fjx481AeapU6eMPl/txGjt+ms7fvw4OnbsiAkTJmi23bp1S6eMTCaDSqXS2ebh4QGA54TVqlULQNk+b23atIFKpUJiYiK6dOlSYjk/Pz+MHz8e48ePx+zZs/HFF19QgEQshpK0CbECLVq0wLBhw/Dpp5/qbO/evTuePHmC999/H7du3cLq1avx22+/mey4q1evxu7du3H9+nVMnDgRKSkpeOONNwDwpNqnT/+/fXt5hX+P4zj+PAsWoiyQjVCMZjFm47YwGwkpJRsN5RKilOsoxcitZibJwmWrscECJVNmFiyYlJnGJXJrzD/A2VpIv7P4daZ8nfJz/Dan83psv5v3e/V99X6/P39it9sJhUJEo1H8fj/t7e0ffqyfGRkZwePxsLm5yd3dHaOjo5yfn9Pf3//tHj6rMzk5mY6ODkZGRjg4OODq6oq2trZ3K0mAiooKlpaWODs7IxwO09PT824KZ7fbyczMpL6+nmAwyOPjI1tbW5ycnAAwMTHB2toaU1NTXF9fc3Nzw8bGBuPj48DPV4smk4nW1lYuLi44OjpibGzsy/2mpKTgcDgYHBzE6/USjUaJRCIsLi7i9XoByM/PJxwO4/f7ub+/x+l0xg+8/5aTk8Pl5SV3d3c8PT3x+vpKXl4eWVlZTE5O8vDwgM/n+8cXf0Ymk4nm5mZaWlrY3t4mFotxenqKy+XC5/MBMDAwgN/vJxaLEYlEODw8xGw2f7l/kd9FAUnkP2J6evrDCsxsNrOyssLy8jJWq5XT09Nv3SoZud1u3G43VquV4+Njdnd3SUtLA4hPfd7e3qiqqsJisTAwMEBqauqHcPGZvr4+hoaGGB4exmKxsL+/z+7uLvn5+d/u4VfqnJubw2azUVdXR2VlJeXl5R9umebn58nKysJms9HU1ITD4SApKSn+PTExkUAgQEZGBrW1tVgsFtxud3ylVF1dzd7eHoFAgOLiYsrKylhYWCA7Oxv4ub7a2dnh5eWFkpISOjs7390rfcXMzAxOpxOXy4XZbKampgafz0dubi4A3d3dNDQ00NjYSGlpKc/Pz++mSQBdXV0UFBRQVFREeno6wWCQhIQE1tfXub29pbCwEI/Hw+zs7C/VtLq6SktLC8PDwxQUFFBfX08oFIpP7d7e3ujt7Y3XazKZWFlZ+Vf9i/wOf/wwLtVFRERE/uc0QRIRERExUEASERERMVBAEhERETFQQBIRERExUEASERERMVBAEhERETFQQBIRERExUEASERERMVBAEhERETFQQBIRERExUEASERERMVBAEhERETH4C3voBRuGXMzeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class(results,\"High variance, 1000 features, linear model\",20,'HighVar_1000Feat_lin_class.png')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40d74a41",
   "metadata": {},
   "source": [
    "# Quadratic"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6eee377",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### plot function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d5d22070",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# in 0 ho wrapper, in 1 GenLinCFA\n",
    "def plot_class(result,title,wrapper_stop,savename):\n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"Accuracy score\")\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[2][i::50]))\n",
    "        wrapper_std.append(np.std(results[2][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    \n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "\n",
    "    ax.plot(range(1,(wrapper_stop+1)),wrapper_means[:wrapper_stop],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,(wrapper_stop+1)), ([i-j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), ([i+j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[0][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "\n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "\n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "    \n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "529a6f47",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# in 0 ho wrapper, in 1 GenLinCFA\n",
    "def plot_class_squaredAggr(result,title,wrapper_stop,savename):\n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel(\"Number of reduced features\")\n",
    "    ax.set_ylabel(\"Accuracy score\")\n",
    "    \n",
    "    wrapper_means = []\n",
    "    wrapper_std = []\n",
    "    \n",
    "    for i in range(50):\n",
    "        wrapper_means.append(np.mean(results[2][i::50]))\n",
    "        wrapper_std.append(np.std(results[2][i::50]))\n",
    "    wrapper_plusminus = [x*1.96/np.sqrt(10) for x in wrapper_std]\n",
    "    \n",
    "    print(f'Wrapper:\\n Mean values {wrapper_means} +- {wrapper_plusminus}\\n')\n",
    "\n",
    "    ax.plot(range(1,(wrapper_stop+1)),wrapper_means[:wrapper_stop],'bo',label='Wrapper score')\n",
    "    ax.fill_between(range(1,(wrapper_stop+1)), ([i-j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), ([i+j for i,j in zip(wrapper_means[:wrapper_stop], wrapper_plusminus[:wrapper_stop])]), color='b', alpha=.1)\n",
    "    \n",
    "    count_means = []\n",
    "    R2_means = []\n",
    "    count_std = []\n",
    "    R2_std = []\n",
    "    for j in range(5):\n",
    "        dummy_list = []\n",
    "        for i in range(10):\n",
    "            dummy_list.append(results[1][i][j])\n",
    "        count_means.append(np.mean([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_means.append(np.mean([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        count_std.append(np.std([dummy_list[i][1] for i in range(len(dummy_list))]))\n",
    "        R2_std.append(np.std([dummy_list[i][2] for i in range(len(dummy_list))]))\n",
    "        \n",
    "    R2_means_plusminus = [x*1.96/np.sqrt(10) for x in R2_std]\n",
    "    count_plusminus = [x*1.96/np.sqrt(10) for x in count_std]\n",
    "\n",
    "    print(f'GenLinCFA:\\n number of red features:{count_means} +- {count_plusminus},\\n R2 score: {R2_means} +- {R2_means_plusminus}\\n')\n",
    "\n",
    "    ax.plot([round(x) for x in count_means],R2_means,'rs',label='GenLinCFA score',fillstyle='none')\n",
    "    ax.fill_between([round(x) for x in count_means], ([i-j for i,j in zip(R2_means, R2_means_plusminus)]), ([i+j for i,j in zip(R2_means, R2_means_plusminus)]), color='r', alpha=.1)\n",
    "    \n",
    "    plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.savefig(savename, format='png', dpi=1200)\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9c828d2",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Low noise, 100 variables, quadratic dependence, class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "0e6dc442",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/class_quadratic_retry.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "14a19ff5",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GenLinCFA epsilon: [0.77, 0.79, 0.81, 0.85, 0.95]\n"
     ]
    }
   ],
   "source": [
    "print(f'GenLinCFA epsilon: {[results[0][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "a4717bef",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.7758, 0.7847999999999999, 0.796, 0.8006, 0.8051999999999999, 0.8094000000000001, 0.8112999999999999, 0.8099000000000001, 0.8125, 0.8139, 0.8145999999999999, 0.8136000000000001, 0.8143, 0.8143, 0.8158999999999998, 0.8158999999999998, 0.8154, 0.8175000000000001, 0.8195, 0.8202999999999999, 0.8206000000000001, 0.8208, 0.8206999999999999, 0.8201, 0.8220000000000001, 0.8213000000000001, 0.8227, 0.8244999999999999, 0.8240999999999999, 0.8244, 0.8245000000000001, 0.8251999999999999, 0.8257, 0.8248, 0.8246, 0.8241999999999999, 0.8259000000000001, 0.8253, 0.8245999999999999, 0.8251, 0.8253, 0.8253, 0.8238999999999999, 0.8249000000000001, 0.8251, 0.8251, 0.8255000000000001, 0.8255000000000001, 0.8253999999999999, 0.8234999999999999] +- [0.0073116144318474624, 0.00851551769418631, 0.008380002863961315, 0.008139075524898379, 0.007863484825444743, 0.007255707160573645, 0.009457076165496374, 0.008045316550639859, 0.007737657009715515, 0.005764199718954898, 0.0069473637014337815, 0.006624684022653436, 0.007185746001634049, 0.005734800571946661, 0.005456057037825008, 0.005203790772119859, 0.0050824126554226004, 0.0037624853488086633, 0.005593731491589468, 0.005259597094835283, 0.004826537309500442, 0.004925804056192223, 0.005266896011884019, 0.0051741772679335, 0.005080900707551746, 0.0049355430906841274, 0.0049433205034672715, 0.0051279383771648335, 0.005406545514466683, 0.005105038138936851, 0.005060065612222823, 0.0045275575755588165, 0.004240581280909496, 0.004372162119592538, 0.0058090114821714636, 0.006033490001649111, 0.005560668880629369, 0.006041443668528234, 0.006382501578534854, 0.005980088159885263, 0.006333257108313217, 0.005860694975853964, 0.005106919071220914, 0.004923073674037381, 0.0051741772679335125, 0.0056904170673158835, 0.005661991875656472, 0.00556619295389586, 0.005887834780290622, 0.006478385601367042]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[30.2, 22.9, 18.5, 13.0, 1.0] +- [1.9756177767979308, 2.0074457402380768, 0.8430278761701773, 1.176, 0.0],\n",
      " R2 score: [0.8412, 0.8403, 0.8436999999999999, 0.8431, 0.845] +- [0.0061219808558995075, 0.0060731541722567894, 0.0045466085822291795, 0.006088316548932066, 0.0057344656246245]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class(results,\"Low variance, 100 features, squared model:\\nsquared input, mean aggregation\",30,'LowVar_100Feat_quadr_class.png')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4a188f1",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Low noise, 100 variables, quadratic dependence (squared sums), class\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "4fffe59f",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GenLinCFA epsilon: [0.77, 0.79, 0.81, 0.85, 0.95]\n"
     ]
    }
   ],
   "source": [
    "print(f'GenLinCFA epsilon: {[results[1][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "bb5d6b8c",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.7758, 0.7847999999999999, 0.796, 0.8006, 0.8051999999999999, 0.8094000000000001, 0.8112999999999999, 0.8099000000000001, 0.8125, 0.8139, 0.8145999999999999, 0.8136000000000001, 0.8143, 0.8143, 0.8158999999999998, 0.8158999999999998, 0.8154, 0.8175000000000001, 0.8195, 0.8202999999999999, 0.8206000000000001, 0.8208, 0.8206999999999999, 0.8201, 0.8220000000000001, 0.8213000000000001, 0.8227, 0.8244999999999999, 0.8240999999999999, 0.8244, 0.8245000000000001, 0.8251999999999999, 0.8257, 0.8248, 0.8246, 0.8241999999999999, 0.8259000000000001, 0.8253, 0.8245999999999999, 0.8251, 0.8253, 0.8253, 0.8238999999999999, 0.8249000000000001, 0.8251, 0.8251, 0.8255000000000001, 0.8255000000000001, 0.8253999999999999, 0.8234999999999999] +- [0.0073116144318474624, 0.00851551769418631, 0.008380002863961315, 0.008139075524898379, 0.007863484825444743, 0.007255707160573645, 0.009457076165496374, 0.008045316550639859, 0.007737657009715515, 0.005764199718954898, 0.0069473637014337815, 0.006624684022653436, 0.007185746001634049, 0.005734800571946661, 0.005456057037825008, 0.005203790772119859, 0.0050824126554226004, 0.0037624853488086633, 0.005593731491589468, 0.005259597094835283, 0.004826537309500442, 0.004925804056192223, 0.005266896011884019, 0.0051741772679335, 0.005080900707551746, 0.0049355430906841274, 0.0049433205034672715, 0.0051279383771648335, 0.005406545514466683, 0.005105038138936851, 0.005060065612222823, 0.0045275575755588165, 0.004240581280909496, 0.004372162119592538, 0.0058090114821714636, 0.006033490001649111, 0.005560668880629369, 0.006041443668528234, 0.006382501578534854, 0.005980088159885263, 0.006333257108313217, 0.005860694975853964, 0.005106919071220914, 0.004923073674037381, 0.0051741772679335125, 0.0056904170673158835, 0.005661991875656472, 0.00556619295389586, 0.005887834780290622, 0.006478385601367042]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[8.8, 5.5, 3.2, 1.4, 1.0] +- [1.380374441954066, 1.574113083612483, 1.0296997620665937, 0.4111330684826994, 0.0],\n",
      " R2 score: [0.844, 0.8457999999999999, 0.8452, 0.8459, 0.8455999999999999] +- [0.006559414608027158, 0.00656994806676583, 0.0057465114286843685, 0.006069357659588044, 0.006023293982531491]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class_squaredAggr(results,\"Low variance, 100 features, squared model:\\nlinear input, squared sum aggregation\",9,'LowVar_100Feat_sqsum_class.png')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0867910",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### High noise, 1000 variables, quadratic dependence, class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "1d8fb7d7",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GenLinCFA epsilon: [0.77, 0.79, 0.81, 0.83, 0.85]\n",
      "GenLinCFA epsilon: [0.77, 0.79, 0.81, 0.83, 0.85]\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "with open('/Users/paolo/Documents/PaperNon_Gen_LinCFA/results/class_quadratic_retry_highVar.pkl', 'rb') as f:\n",
    "    results = pickle.load(f)\n",
    "print(f'GenLinCFA epsilon: {[results[0][0][i][0] for i in range(5)]}')\n",
    "print(f'GenLinCFA epsilon: {[results[1][0][i][0] for i in range(5)]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "107b199f",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.683, 0.6875, 0.6857, 0.6876, 0.6883000000000001, 0.6889000000000001, 0.6882, 0.6891999999999999, 0.6879, 0.6873, 0.6885000000000001, 0.6882999999999999, 0.6881999999999999, 0.6884, 0.6887000000000001, 0.6892999999999999, 0.6884, 0.689, 0.6889000000000001, 0.6881999999999999, 0.6887, 0.6886, 0.6895, 0.6892, 0.6887, 0.69, 0.689, 0.6885, 0.6885999999999999, 0.6883, 0.6890999999999999, 0.6896, 0.6902, 0.6889, 0.6890000000000001, 0.6893, 0.6900000000000002, 0.6895, 0.6896000000000001, 0.6893, 0.6889999999999998, 0.6889, 0.689, 0.6884, 0.688, 0.6881999999999999, 0.6881999999999999, 0.6889, 0.6894, 0.6889] +- [0.007401433374691672, 0.008120410334459685, 0.007721753272411628, 0.00871351722325719, 0.008852823368846779, 0.007668836574083428, 0.007482991487366513, 0.006928537623481574, 0.00676384612480205, 0.0068742651680015745, 0.007035553709552627, 0.007072586061689155, 0.006703085379136966, 0.006768104490919134, 0.00706715229777877, 0.006658521277280695, 0.006624684022653437, 0.006749916740227224, 0.006809131398350288, 0.006850473385102649, 0.006935464915923064, 0.006796425413406648, 0.007510918452492992, 0.006956205402372744, 0.00788422029118921, 0.007473744978255528, 0.007509639671781843, 0.007392344688933255, 0.007607224355834376, 0.007581172574213024, 0.007336005343509488, 0.007213226074371978, 0.0066223640491896675, 0.007225199125283659, 0.006606101422170247, 0.007023530849935791, 0.007499401576125906, 0.007826513399975737, 0.0071382758702644555, 0.007318179117786044, 0.007344117646116498, 0.007802917300599804, 0.007932553687180422, 0.007801686381802317, 0.00811922754946551, 0.00783411779334468, 0.007853708015962888, 0.008275982503606427, 0.008067965691548256, 0.008436883690083661]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[9.4, 5.9, 3.5, 1.6, 1.0] +- [1.14956965861143, 0.6470969015533916, 0.49970391233209294, 0.4111330684826994, 0.0],\n",
      " R2 score: [0.7051000000000001, 0.7093, 0.7091, 0.7152000000000001, 0.7186] +- [0.0056971640664456955, 0.006806874290010068, 0.007812757669350823, 0.008323887168865285, 0.008376334663801352]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class(results,\"High variance, 1000 features, squared model:\\nsquared input, mean aggregation\",9,'HighVar_1000Feat_quadr_class.png')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "53792e47",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wrapper:\n",
      " Mean values [0.683, 0.6875, 0.6857, 0.6876, 0.6883000000000001, 0.6889000000000001, 0.6882, 0.6891999999999999, 0.6879, 0.6873, 0.6885000000000001, 0.6882999999999999, 0.6881999999999999, 0.6884, 0.6887000000000001, 0.6892999999999999, 0.6884, 0.689, 0.6889000000000001, 0.6881999999999999, 0.6887, 0.6886, 0.6895, 0.6892, 0.6887, 0.69, 0.689, 0.6885, 0.6885999999999999, 0.6883, 0.6890999999999999, 0.6896, 0.6902, 0.6889, 0.6890000000000001, 0.6893, 0.6900000000000002, 0.6895, 0.6896000000000001, 0.6893, 0.6889999999999998, 0.6889, 0.689, 0.6884, 0.688, 0.6881999999999999, 0.6881999999999999, 0.6889, 0.6894, 0.6889] +- [0.007401433374691672, 0.008120410334459685, 0.007721753272411628, 0.00871351722325719, 0.008852823368846779, 0.007668836574083428, 0.007482991487366513, 0.006928537623481574, 0.00676384612480205, 0.0068742651680015745, 0.007035553709552627, 0.007072586061689155, 0.006703085379136966, 0.006768104490919134, 0.00706715229777877, 0.006658521277280695, 0.006624684022653437, 0.006749916740227224, 0.006809131398350288, 0.006850473385102649, 0.006935464915923064, 0.006796425413406648, 0.007510918452492992, 0.006956205402372744, 0.00788422029118921, 0.007473744978255528, 0.007509639671781843, 0.007392344688933255, 0.007607224355834376, 0.007581172574213024, 0.007336005343509488, 0.007213226074371978, 0.0066223640491896675, 0.007225199125283659, 0.006606101422170247, 0.007023530849935791, 0.007499401576125906, 0.007826513399975737, 0.0071382758702644555, 0.007318179117786044, 0.007344117646116498, 0.007802917300599804, 0.007932553687180422, 0.007801686381802317, 0.00811922754946551, 0.00783411779334468, 0.007853708015962888, 0.008275982503606427, 0.008067965691548256, 0.008436883690083661]\n",
      "\n",
      "GenLinCFA:\n",
      " number of red features:[6.1, 3.7, 2.6, 1.6, 1.3] +- [2.2252690623832434, 1.3307808234266076, 0.6320810074666061, 0.3036418943426615, 0.2840309842253129],\n",
      " R2 score: [0.6968, 0.7114999999999999, 0.7178, 0.7189, 0.7184] +- [0.021832323779204073, 0.008476858380319911, 0.0077898596649747225, 0.008140255425967916, 0.008063202738366443]\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_class_squaredAggr(results,\"High variance, 1000 features, squared model:\\nlinear input, squared sum aggregation\",6,'HighVar_1000Feat_sqsum_class.png')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b8c507e0",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5d0084bb",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4ee68e2a",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0ed9090e",
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9 (tensorflow)",
   "language": "python",
   "name": "tensorflow"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}