23_miscplots.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nplace to code up additional summary plots needed for paper\\n\\nShaina Lu\\nJune 2020\\nZador & Gillis Labs\\n'"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "place to code up additional summary plots needed for paper\n",
    "\n",
    "Shaina Lu\n",
    "June 2020\n",
    "Zador & Gillis Labs\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy as sp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def set_style():\n",
    "    #function for matplotlib formatting\n",
    "    plt.style.use(['seaborn-white','seaborn-notebook'])\n",
    "    plt.rcParams['figure.figsize'] = [6,4]\n",
    "    plt.rcParams['axes.spines.top'] = False       #remove top line\n",
    "    plt.rcParams['axes.spines.right'] = False     #remove right line\n",
    "    plt.rcParams['axes.linewidth'] = 2.0          #set weight of axes\n",
    "    plt.rcParams['axes.titlesize'] = 20           #set font size of title\n",
    "    plt.rcParams['axes.labelsize'] = 18           #set font size of x,y labels\n",
    "    plt.rcParams['axes.labelpad'] = 14            #space between labels and axes\n",
    "    plt.rcParams['xtick.labelsize'] = 14          #set x label size\n",
    "    plt.rcParams['ytick.labelsize'] = 14          #set y label size\n",
    "    plt.rcParams['legend.fontsize'] = 16          #set legend font size\n",
    "    \n",
    "set_style()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# within all by all summary figure (1c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "withinmeans = pd.read_csv(\"allbyall_means.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>alpha</th>\n",
       "      <th>variable</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.10</td>\n",
       "      <td>ABAtrain</td>\n",
       "      <td>0.999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.05</td>\n",
       "      <td>ABAtrain</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.01</td>\n",
       "      <td>ABAtrain</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.00</td>\n",
       "      <td>ABAtrain</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.10</td>\n",
       "      <td>ABAtest</td>\n",
       "      <td>0.996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.05</td>\n",
       "      <td>ABAtest</td>\n",
       "      <td>0.998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.01</td>\n",
       "      <td>ABAtest</td>\n",
       "      <td>0.998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.00</td>\n",
       "      <td>ABAtest</td>\n",
       "      <td>0.999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.10</td>\n",
       "      <td>STtrain</td>\n",
       "      <td>0.993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.05</td>\n",
       "      <td>STtrain</td>\n",
       "      <td>0.999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.01</td>\n",
       "      <td>STtrain</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.00</td>\n",
       "      <td>STtrain</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.10</td>\n",
       "      <td>STtest</td>\n",
       "      <td>0.883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.05</td>\n",
       "      <td>STtest</td>\n",
       "      <td>0.885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.01</td>\n",
       "      <td>STtest</td>\n",
       "      <td>0.878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.00</td>\n",
       "      <td>STtest</td>\n",
       "      <td>0.949</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    alpha   variable  value\n",
       "0    0.10   ABAtrain  0.999\n",
       "1    0.05   ABAtrain  1.000\n",
       "2    0.01   ABAtrain  1.000\n",
       "3    0.00   ABAtrain  1.000\n",
       "4    0.10    ABAtest  0.996\n",
       "5    0.05    ABAtest  0.998\n",
       "6    0.01    ABAtest  0.998\n",
       "7    0.00    ABAtest  0.999\n",
       "8    0.10    STtrain  0.993\n",
       "9    0.05    STtrain  0.999\n",
       "10   0.01    STtrain  1.000\n",
       "11   0.00    STtrain  1.000\n",
       "12   0.10     STtest  0.883\n",
       "13   0.05     STtest  0.885\n",
       "14   0.01     STtest  0.878\n",
       "15   0.00     STtest  0.949"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "withinmeans.head()\n",
    "withinmeans.melt(id_vars='alpha')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkEAAAEcCAYAAADA/GAZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzdeXhN1/7H8XckImYZEI1UCQkiIcYSiphaamwV7a1bXEpp+1Oa6kX1lnJpdYoxtHRAW1OKpKVmldas1CymiCERIRIyn98fuVKRwQnnZDqf1/Pkafbaa6/zPVua881aa69lZTAYDIiIiIhYmBIFHYCIiIhIQVASJCIiIhZJSZCIiIhYJCVBIiIiYpGUBN3Dw8MDDw+Pgg5DRERE8oGSIBEREbFISoJERETEIikJymer9l/EY8LPzN5y2rQNXz8DnzaAr56BlETTtl1IGQwG3vz+AI0+2MC+89dN2/jhFTDFGbbNMG27hVjkrQT8Pt5Kty92cCsh2bSNn90OmyZDQqxp2y3EbtxOYvovx9l3PqagQxGRHCgJykeRsQlM+ukIiSlpfPLrSU5cuWWahg0GCB4LN8PhQij8Pss07RZywYcv89PBS9y4ncw7Kw+TnJpmmobjoiD4LUi5A1unweVDpmm3kJsWcpwz1+I5cimWzzaeMl3DNy/C0n6w42P4ZZzp2i3k/FccYu7WMIZ8vYfr8UkFHY6IZENJUD6a9vNxbiWmAJCaZmDiT3+R3YLdd5JSWbjjDJ0/3Yb7+J9p9MEGxvz4J8cu5/BX9PF1ELbp7+NtH8GNcHO8hUIjPjGFKeuOZRyfjoxj0c6zOdYPi4ojYNMpPlh7lAXbz3A1NiHnxje+Dwk30783pEHIWEgzUYJVSO06E83qAxEZx4tDz+WYpKelGdh+Morpvxznw+CjrD5wkYTk1JwbX/9vSL6d/v3BJXDhD1OGXihtORHJhqNXAbhxO5mP1h837QukJEGqiXvrRCyQVVHYNmPPnj18+eWXHDlyhMjISKZNm0afPn1yvebEiRNMnjyZQ4cOUbFiRfr168fIkSOxsrLK8Zq7T4adOHHCpPFD+odMv8Csv/w/79+Ino1cMo5v3k7mpS//4K+IrAlPSWsrPu/vQ1evan8XJt2G2c3Te4HuVa8H9PvWZPEXNtN+Psb8bWcylZW1tWbTmHY4V7TLKLuTlMo7Kw+x5s9LmeralLDiX21q4d/FgxIl7vmZCN8NX3bK+oK95kKjF036HgqL5NQ0nv3iN05czZz0NK/pwA/Dnsz0/0xYVBwjvtvHyatxmeo6lLVlxnPedKxfNXPjYZvh296Zy6p6wbCtYG1jwndReCQkp9Lls+2cj76dUWZlBatf86WRa6Us9c9ExfHDnnCOX7mFrU0JnnKvTG8fF8qVyuH+xF6ChR2hhDX8azOUq2yutyJS7BWJnqDbt2/j7u7O+PHjsbOze2D9uLg4Bg8ejKOjIytWrGDChAl8+eWXLFq0KB+izSo5NY33fjqS7bkpwccyzb8YH3Q42wQovR0D//f9QcKv//3LlR0zsyZAAMfWwOlNWcuLgdORt/hyR9Zen/ikVD4M+bt36O6cofsTIICUNAPztoXx0YZ7Et60VAgek/2LbpgId248cuyF0Te/n8+SAAHsPns9072LupXIgMA/siRAANfjkxj+3T7+OBP9d2FKEoT4Z33Bq4dh71cmib0wWrD9TKYECNJHrN/76S9S0wz3lBn4bONJ/GZuY/72M2w7GcWvR68yMegv2s7Ywv4LOcwlWj8eYiPgxgXY9L4Z34lI8VckkqC2bdvy1ltv8fTTT1OixINDXrNmDXfu3GH69Om4u7vTpUsXhg4dyqJFi7IdfjK3nD5kIP2D5fP/zb+4dOMOIYcv59pWUmoa3/1xPv0gOgxCv8i58s/+xW6StMFgYNKaI6SkZf/vuPbPS4SGXQNg/4WYjCGJnCzccYaoW/+7R3u/gis5zP+5fQ22TH3ouAuryNgEPvv1ZI7n703Sv9p5lshbOf88paQZ+Hj9PUnlH7MhOoe5RZunQFzkQ8VcmIVfv82sHB56OHTxJt/vuZBxvHT3hRznXkXHJ/HKV7u5fPNO5hNntsKRVX8fH/guvfdSRB5KkUiC8urgwYM0bdo0U69R69atiYyM5OLFi/kez4LtZ3I9//Xv50hITuWPM9Hk8Nmeyc7/fcizaz6k5jLhMvo0nPjZ+ECLgKOXY9l5OjrXOnfv98r9EbnWg/TetbV3eztCA3KvvGchJMUbFWdR8f2e8Ix5atmJupVI0MH0+7Nq/4P/39l7PoYL0bchPjp9blpOEm+mTzovZj5af4LElJznj320/gS3k1JISU1j9ubcnxCNTUjh69DzfxekJEHI21krBo9J78UUkTwrlknQtWvXcHR0zFTm5OSUcS6/eVWvmOt596rlKWVTwuinm5Lu/pJ9zCf3iiVswNnLqDaLCleHMlQqUzLXOl7V0+ddXL2Zy+Tne1y99b96D7qfletCyTJGtVlUPOhnE8DLpSIGg4Grscb1Kl69lQAJNyD5AQljbNZhyqLuygN+5mLvJBOXmMKfF29wyYifz+DD99yjXXPhWja9dlcOFevhRRFzKpZJEJBlAvTdYbDcJkaby4Ru9bC1yflWf9DTEysrK+pVq2BUexn1vPuBS9OcKz75Gji65SXUQq+CXUne7pLz1iYulUozvG0tAOzL2hrVpkOZ/9Xr9AHYlM65YteP0me4FiPt3CvjV7dKjuefb1KdRq6VsLKywsHI+2lfxjb95867Xy61rKDtO3mMtvD7v451cj0/sOUTVClvx43bxj3ZdfNuvZsRsHV6zhU3T05f2kFE8qRYJkFOTk5Zenyio9OHUO7vIcoPNRzLMrxt9snI802q06SGA5D+F7eXy4P/Mn+pRY30b0qUgG4zgWw+mMtXg7bZTEotBvo3ezzH+zTx2XqUsU1/qqZ7w8ce2JaVFX8/bWdfA9rkMDHa6wV4wveh4i3MrKysmNS9frZJenk7G8Y9Uzfj+Fnvalnq3K9etQq4VS6bftBpMpTKIbFv8gq4NH6YkAu1VrWdcrxPTuVsGd3JHYBqFXNJtu/xWKX/1ftzWe49awk34fDyPMUqIsU0CWrUqBF79+4lMfHv7vvQ0FCqVKlC9erVCySm19q5Ud0+8y+++z9krKysmNbHK+dHY4GBLWvQvKbD3wWPNYKmg7NW7PIhlCr/yHEXRtYlrJjcq0GWTpmn3CvTxdM547hNbSea1rDPta0Xmrji6nDPEFer18GhVuZKtuWh8+RHDbvQyilJH9vZA6dypTKO/9W6FuVz+dkEeLNDnb97W8tXhXbvZq1U2gE6vPdIMRdmE7rVp4ytdZbycc/Uo2Lp9KHcetXKU9+Int/nm/zv99UTrR9Q0wpqtMprqCIWr0gkQfHx8Rw7doxjx46RlpbGpUuXOHbsGJcupY+Xz5w5k3/+858Z9bt3707p0qUZN24cJ0+eZMOGDQQGBjJo0KACGQ4DsCtpzaTunpnK7v+QAWjgUpEVI1rylHvmtT+qVbRj4rP1+U+PzG0A4DcBytzTw1XzKfDMfR2loq6RayX6NXXNOC5pbcX73etn+vctUcKKBQOb0uyJ7BOhHg0f44Ne993PknbwzH1bZbR/F8o7U5zdn6TXr1aBl1o8nqnO445lWDy4GU7lsg6L2VqX4MPeDXi6wX33qfkwqHLfPe74PpRxoLhyrmiXZVisaQ17+vj8vR6YlZUV/+5aD+sSOf8+cqtclv7N//dv8PiT6b2ROWnyz/Q/iEQkT4rEYom7du1i4MCBWcp79+7Nf//7X8aNG8fu3bvZvHlzxrkTJ07wwQcfZCyW2L9//wJdLBHS5yUN+Xovm49HUr9aBdaM8sXGOuc8NOLGHc5Hx1PW1gbPxyrkWpf938KaUemToYfvhCp1c65bTFyPT8Jv5lZu3E5mZHs33u6S/Xs2GAyEhkWz7tAlYuKTca5ox/NNqtMgt6HH719KX4m7Sn14dTtY5z4Zuzj49ehVhn6zF4CVI1pmDNPe705SKmv+jCA0LJrk1DQ8H6tI36bVqVI+hzW8zofComfSv3dpAkM2pg/lFmPJqWk88/kOTkfGUcIK1r3ehvqPZe35+fXoVd5ddZhrcZknnbeo6cAXA3yoWuGee3rrCgQ0haT7ltsobQ+v7y/WiaWIuRSJJCi/mDsJAohLTCHoQARdPJ2pXL7Ugy8wlsGQvkBiOWd4vIXp2i3kTl69xaGLN+nt45LrX9V5lhgHh36Aut2KfS/QvTYfv4qdjTWtajuZtuGt0+FECPSeB1XqmbbtQurQxRuMW3mYPo1d+FebWjnWS0pJ49ejVzlx9RalbErwVJ3KOT+19/scWH/fEGP3z9PnWIlInikJukd+JEEiIg8tNQXmt4HIo+nHjzWGf20q9j1rIuai/3NERIoKaxvo+jFgBVbW0O1jJUAij6B47mAoIlJcPeELQ35N30C1GC4zIJKflASJiBQ1rs0KOgKRYkH9qCIiImKRlASJiIiIRVISJCIiIhZJSZCIiIhYJCVBlioxDhJi0xdZFBERsUB6OsySpKXBwSWwez5cOZxe5uQOzYamb8JqrR8HERGxHFox+h7FesXotDQIGgGHvs/+vPvT0O87i9gjS0REBDQcZjkOfJtzAgRw8hf4fVb+xSMiIlLAlARZit2BRtRZCGmp5o9FRESkEFASZAkSYuHqXw+uF3sRblwwfzwiIiKFgJIgS2BIM09dERGRIkxJkCWwqwgObg+uV8YJKj1u/nhEREQKASVBlsDKCpr968H1mg7S02EiImIxlARZiuZDoXbHnM9Xbw6t38q/eERERAqYkiBLYV0S+i8DvwlQvtrf5WWcoM0YGBgEtmUKLj4REZF8psUS71GsF0u8V2oK3DifPgm6Ug2wsS3oiERERPKd9kmwRNY24GjERGkREZFiTMNhIiIiYpGUBImIiIhFUhIkIiIiFklJkIiIiFgkJUEiIiJikZQEiYiIiEVSEiQiIiIWSUmQiIiIWCQlQSIiImKRlASJiIiIRSoySdCSJUvw8/PDy8uLPn36sHfv3gfWf+aZZ/D29qZLly4EBQXlU6QiIiJSFBSJJCgkJISpU6cyfPhwgoKC8PHxYejQoVy6dCnb+kuXLuXjjz9m5MiRBAcH88Ybb/Cf//yHzZs353PkIiIiUlgViSRo0aJF9O7dmxdeeAE3NzcmTpxI5cqVWbZsWbb116xZQ9++fXn22WdxdXWlW7du9OvXjwULFuRz5CIiIlJYFfokKCkpiSNHjuDr65up3NfXlwMHDuR4TalSpTKVlSpVisOHD5OcnGy2WEVERKToyDUJSkhIYPTo0QQGBubaSGBgIGPGjCEpKcmkwQHExMSQmpqKk5NTpnJHR0eioqKyvaZ169asXLmSQ4cOYTAYOHz4MCtWrCA5OZmYmBiTxygiJpKWChf+gOMhELEfDIaCjkhEijGb3E6uWLGCX3/9lf/7v//LtZEuXbrwxRdf0KxZM/r372/SAO+ysrIyqgzgtddeIyoqigEDBmAwGHB0dKRXr14sXLgQa2trs8QnIo9o/7ewbQbcvPB3WeW60PF98HimoKIq2q4egTPbIDUJnBtArfZQQr8DRe7KtSdow4YNtGvXjho1auTaSI0aNfDz8+OXX34xaXAA9vb2WFtbZ+n1iY6OztI7dJednR3Tpk3j4MGDbN68ma1bt+Li4kLZsmWxt7c3eYwi8oh++wzWjMqcAAFEHYdlA+DwioKJq6i6dQW+6QlzW8H6d2HjJPjuOQhoDOd+K+joRAqNXJOgEydO0LRpU6Maaty4McePHzdJUPeytbXF09OT0NDQTOWhoaH4+Pjkem3JkiVxdnbG2tqakJAQ2rdvT4kShX4alBQlBgNc2AUHvoMjq+H29YKOqOi5EQ6bPsilggHWvQVJ8fkWUpGWcBMWPwtntmY9F3MuPRm6mPsSIyKWItfhsPj4eCpUqGBUQxUqVCAuLs4kQd1v0KBB+Pv74+3tTePGjVm2bBmRkZEZQ2/+/v4AzJgxA4CzZ8/y559/0qhRI2JjY1m0aBGnTp3iv//9r1niEwt1dgeEjE3vrbjLxg4a/xM6TwabUjlfK3878B0YUnOvk3gzPcn0+Uf+xFSU7VkI0adyPp+SAL9OgkHB+ReTSCGVaxJUvnz5HCcf3+/atWuUL1/eJEHdr2vXrsTExDB37lwiIyNxd3cnMDAQFxcXAC5fvpypflpaGosXL+bs2bPY2NjQokULli1bRvXq1c0Sn1igszvg296Qdt/ThikJsHs+3LgA/ZeCeh4fLPKokfWOmTeO4uLAkgfXOf8bXD8LDjXNH49IIZZrElSvXj22bNnCq6+++sCGNm/eTL169UwW2P1eeuklXnrppWzPffvtt5mO3dzctEK0mI/BkN4DdH8CdK+TP6d/1e2Wf3EVVTZ2RtZTz5pRblx4cB2Am+FKgsTi5fpnateuXfnzzz9ZsiT3vyyWLl3Kn3/+Sbdu+oUvFuDinsxDYDnZ97X5YykOanc0bT1LZ2fcFAZKGVlPpBjLNQnq3bs33t7eTJkyhTFjxrBr1y5iY2NJS0vj1q1b7Nq1i7FjxzJ58mQaNmxIr1698itukYJzLZf5FvfKbV6G/K1+T6jwgKHqxxrD4y3zJ56irl6PB9exfwKcvc0eikhhl+twmLW1NXPnzuWNN94gODiYkJCQLHUMBgPNmjXj888/1xo8Yhlsy5q2nqUraQcvfg/f9ILb17Ket68JL3wDOawLJvdpORL+/B5S7uRcp/Vbmq8mAlgZDMYtybp582Y2bNjAyZMniY+Pp2zZsri7u9O5c2f8/PzMHWe+8PDwANKXBhDJ0Z0YmFkv9w8ZgLbvQPt/509MxUFcZPqTTX+thNvRUP4xaNgfmvwT7CoWdHRFy+lN8OM/IelW1nNt34F27yqpFCEPSZAlUBIkRvv5Hdg1L+fztuVh1B6oUC3/YhK5V3w0HPgmfb2g1GSo2gCaDoIq5nuARaSoyXMSdOfOHW7dukX58uUpXbq0ueIqEEqCxGgpiel/aZ/8Oes52/IwYCnUfCr/4xIREaPlOiforuvXr7Nw4UI2bNhARERERrmLiwudO3dmyJAhODo6mi1IkULHplT6OkAnf05/Cux6GJQsA+5PQ9PB6gESESkCHtgTdPDgQUaOHEl0dDQ2NjbUrFmT8uXLExcXx9mzZ0lOTsbJyYnZs2fTsGHD/IrbLNQTJCIiYjlyTYKio6Pp1q0bqampjB49mj59+mBn9/fCZgkJCaxevZpPP/0Ua2tr1q1bV6R7hJQEiYiIWI5cn5H88ssviYuLY/Hixbz44ouZEiBI3619wIABLFq0iLi4OL766iuzBisiIiJiKrkmQdu2baNnz554enrm2oinpyc9evRgy5YtJg1ORERExFxyTYIiIiJo1KiRUQ01bNiQS5cumSQoEREREXPLNQkqUaIEycm5bBJ5j5SUFEpoBVIREREpInLNWp544gl27dplVEO7d++mRo0aJglKRERExNxyTYI6dOjAr7/+yvbt23NtZMeOHfz666906tTJpMGJiIiImEuuj8jHxcXRo0cPoqKiGDRoEH379sXV1TXjfHh4OMuXL2fx4sU4OTmxZs0aypUrly+Bm4MekRcREbEcD1ws8ezZswwfPpzz589jZWVF2bJlMxZLjIuLw2AwUKNGDebMmYObm1t+xW0WSoJEREQsh1F7hyUkJLB8+XLWr1/PqVOnMu0i36lTJ/r27Vss9hFTEiQiImI5tIv8PZQEiYiIWA6TPtO+d+9eUzYnIiIiYjYmSYL279/PoEGDePnll03RnIiIiIjZ2Tyowrlz5/juu+84f/48FStWpGfPnrRp0waAkydPMn36dEJDQ7GysqJr165mD1hERETEFHJNgk6dOkX//v2Jj4/PKAsODmb69OmkpaUxYcIEAHr27Mmrr75KzZo1zRutiIiIiInkmgTNmTOHpKQkxo8fT8uWLTl//jwffvghH3/8MTdu3KBVq1aMHz9eK0WLiIhIkZNrErR3716ee+65jLk+tWvXxmAwMGrUKNq1a8e8efPyJUgRERERU8t1YnRMTAyenp6Zyho0aABAr169zBeViIiIiJnlmgSlpKRgZ2eXqaxUqVIAVKxY0XxRiYiIiJjZAx+Rt7KyylO5iIhIcZGYkkrfeaG0mLqRM1FxBR1OFnPmzMHDw4NRo0Zle37Xrl14eHhk+vL19WXo0KEcOnTokdoNCAggLS3NJO/jXgEBARmLF5tbritG161bl1KlSmFtbZ2p/Pbt29jZ2VGiROYcysrKin379pkn0nygFaNFRORes7ec5qP16Z8JrWs78e2Q5oWqE6BLly6cO3eOkiVLsmPHDuzt7TOd37VrFwMHDmTChAl4eXlhMBi4cuUKCxYs4Ny5c/z000+ZNkY3tt2AgABmzZrFkSNHsLF54Go7eXLlyhWuXLlCo0aNTNpudnKNvFmzZmYPQEREpDCKuHGHgM2nMo5/O32NkMNX6OZdrQCj+tv+/fs5d+4cbdu2Zdu2bQQHB/OPf/wj27pubm6Zkor69evTuXNnduzYwYsvvvjQ7RojOTkZGxsbo5NHZ2dnnJ2dH/r18iLXJOjbb7/NlyBEREQKmynrjpKQnHm4Z0rwUdp5VKZsKdP2fjyMoKAgrK2tmTx5Mv369SMoKMjoZKVcuXJA+tzfvLZ7txcIyPTw1IkTJ7h48SIdOnTgvffeIyIigjVr1nDt2jV27dpFamoqn332Gbt27eLKlStUqlSJpk2b4u/vT9WqVbO0f++ojIeHB8OHD8fR0ZHFixdnPLg1adIk6tSpk7cbdw+T7h1mTkuWLMHPzw8vLy/69OnzwH3K1q5dS8+ePWnYsCG+vr6MHTuWqKiofIpWRESKsm0no/j5rytZyi/fTGDWltMFEFFmiYmJ/Pzzz7Rq1YqqVavSo0cPDh8+TFhYWLb109LSSElJITk5mfDwcD788ENKly5N+/bt89xu3759ef755wFYunQpP/zwAz/88EOmdubNm8e5c+eYPHkys2bNolSpUty4cQNbW1veeustFi5ciL+/P+fOnWPAgAEkJiY+8D2vXbuWbdu2MX78eKZNm8alS5d47bXXsk3kjFUkkqCQkBCmTp3K8OHDCQoKwsfHh6FDh3Lp0qVs6+/btw9/f3969+7NunXrmD17NmFhYYwdOzafIxcRkaImMSWV99ccyfH8wh1nCCvgSdKbNm0iNjY2Y7mau/9dvXp1tvWHDBmCp6cnDRo0oGPHjmzfvp3PP/88y3wgY9q9d7iqYcOGNGrUKMv8HScnJ2bPnk379u3p2LEjdnZ21KpViwkTJtClSxeaNWvG008/zaxZs4iIiGD79u0PfM82NjbMmzePDh068PTTT+Pv78+FCxdyneD9wDZzO9mhQ4ccz1lZWWFnZ0f16tXp2LEjvXv3zjKB2lQWLVpE7969eeGFFwCYOHEiO3bsYNmyZYwZMyZL/YMHD+Ls7Mwrr7wCgKurK//4xz+YMmWKWeITEZHiY8vxSM5ei8/xfHKqga9Dz/FBzwb5GFVmq1evply5cnTs2BGAWrVq0bBhQ9asWcNbb72V5cGl9957D29vbwCuX79OUFAQb775JoGBgTRv3vyh281Jhw4dsp0DtHTpUr7//nvCw8O5fft2RvnZs2cf2GarVq0oWbJkxrG7uzsAly9fNiqm7OT6bgwGQ45faWlpREdHs3XrViZOnMjgwYNJTk5+6EBykpSUxJEjR/D19c1U7uvry4EDB7K9pnHjxkRFRbF582YMBgPXr18nJCSEp556yuTxiYhI8VKvWgVKWuc+ibdh9Ur5FE1WUVFR7Ny5k3bt2pGUlERsbCyxsbF07tyZq1ev8vvvv2e5pmbNmnh5eeHl5UXbtm355JNPcHV15aOPPnqkdnNSpUqVLGXffvst//nPf2jVqhUBAQEsX76cH3/8EcCo4bD71ye0tbU1+tqc5NoTtHnz5gc2cP36dRYvXkxgYCBff/01//rXvx46mOzExMSQmpqKk5NTpnJHR0dCQ0OzvcbHx4eZM2cyduxYEhMTSUlJwdfXl+nTp5s0NhERKX5qOJblX21qMXdr9vNrGrlWorePSz5H9bc1a9aQmprKunXrWLduXZbzq1evztJxcD8rKytq1arFli1bTNruve3fLzg4mJYtWzJu3LiMsvDwcKPaM5dHnt7u4ODAW2+9xbFjxwgJCTF5EnRXdjc0p8ftTp8+zZQpU3jttddo3bo1UVFRzJgxg/fee48ZM2aYJT4RESk+XverTdCBCC7fTMhUbmUFU3o1oESJglsrKCgoCBcXF6ZNm5bl3IIFC9i4cSNxcXEZT4BlJy0tjdOnT+Pg4PBQ7d7thUlISMj1de6VXd1Vq1YZda25mOwZv5YtWxIQEGCq5jLY29tjbW2d5cmu6OjoLL1Dd82fPx9vb++MhKxu3bqULl2al156idGjR1OtWuFY40FERAqnMrY2THy2Pq8t2Z+p/KUWj9PApeC2jTpy5AgnT57k9ddfp0WLFlnOJyYmsmPHDtavX89zzz2XUR4WFkaZMmWAv+cEnT59mrfffvuh2nVzcwPS5+w+9dRTlChRAi8vr1xjb9OmDQsWLGDevHl4e3vzxx9/sH79+oe+F6ZgsiSoVKlSpKammqq5DLa2tnh6ehIaGsozzzyTUR4aGkrnzp2zvSYhISHLJO27x7kskC0iIpLhmQbOtKnjxI5T1wBwKGvL2M75s51DTlavXo2VlRW9e/fO9nzr1q2pVq0aQUFBmZKgex8MqlixIjVr1mTmzJk8++yzD9Vu+/btefHFF1m6dCmzZ8/GYDA8cLeFkSNHEhsby+LFi0lMTKR58+YsXLgwYxJ2Qch124y8mDBhArt372bDhg2maC6TkJAQ/P39mTRpEo0bN2bZsmWsXLmSdevW4eLigr+/P0DGUNeqVauYOHEi48ePp02bNkRGRjJ16lQMBkOuXW/aNkNERO4VFhXHM5/tICk1jRnPefNCs6xbTEjRZZKeoJ07dxIUFET//v1N0VwWXbt2JSYmhrlz5xIZGYm7uzuBgYG4uKRPTC1DggwAACAASURBVLv/8bg+ffoQHx/PkiVLmD59OuXLl6dFixYZ3X4iIiLGcKtcjh+Ht+TKzQS6eFZ98AVSpOTaE/Tuu+/menFCQgJhYWGcOnUKJycnVq1aReXKlU0eZH5RT5CIiIjlyLUnKKeVJ+9VpkwZunbtytixY4t0AiQiIiKWJdckaNOmTblebGdnh729vdErSIqIiIgUFrkmQXfn3IiIiIgUNybpwrl27RoLFiyga9eupmhORERExOwe+umwtLQ0tmzZwooVK9ixYwcpKSmULVvWlLGJiIiImE2ek6AzZ86wcuVKfvrpJ6Kjo6lQoQLdu3enS5cutGrVyhwxioiIiJicUUnQ7du3CQkJYeXKlRw8eBBra2saN25MdHQ0kydPznHlZhEREZHCKtckaP/+/axYsYJffvmF27dvU79+fd599126d+9ObGwsXbp0ya84RUREREwq14nRL774Itu3b6dfv3789NNPrFq1ioEDB2Jvb5/jDu4iIiKSP+bMmYOHhwejRo3K9vyuXbvw8PDI9OXr68vQoUM5dOjQI7UbEBBAWlqaSd7H/WJjYwkICODIkSNmaf+uBz4dlpCQwK1bt4iLizNrICIiIoXNxZjbLN8bzrLdF/gz/Eah24T7p59+AmDr1q3ExMTkWG/ChAn88MMPfP/990yYMIHo6GheeeUVwsPDH6rd3bt3M2vWLLMmQbNmzeLo0aNmaf+uXIfDgoODWbFiBWvWrGHlypVUr16dXr160atXL7MGJSIiUpBi4pN4d9Vh1h+9wr15TwOXCvy3jzcNXCoWXHD/s3//fs6dO0fbtm3Ztm0bwcHB/OMf/8i2rpubG40aNco4rl+/Pp07d2bHjh28+OKLD91uUZdrT5CbmxvvvPMO27Zt4/PPP6dmzZrMmTOHTp068eabb2JlZVXosmIREZFHEZ+YwosLd/HLkcwJEMBfEbH0D/yDE1duFUxw9wgKCsLa2prJkydTrVo1goKCjL62XLlyAKSkpOS53YCAAGbNmgWAp6dnxjDbXXfu3OGjjz7Cz8+PBg0a4Ofnx9y5czP1GsXHxzN58mTatWtHgwYNaNWqFa+88gphYWFcvHiRDh06AOk9WHfbX7VqlfE3x0hGPR1mY2ND586d6dy5M5GRkaxatYrVq1djMBjw9/dnzZo1dO7cGT8/P8qXL2/yIEVERPLLd3+c59jl2BzPxyWmMDXkGF8Pbp6PUWWWmJjIzz//TKtWrahatSo9evRg/vz5hIWF4ebmlqV+WloaKSkpGAwGrly5wqeffkrp0qVp3759ntvt27cvV65cYcWKFSxduhRra+uM61NSUhgyZAhhYWGMGDECDw8PDh48yJw5c7h58ybjxo0DYNq0aWzevJnRo0fzxBNPcOPGDfbv38+tW7dwdXVl1qxZjBo1ildffRU/Pz8AHn/8cZPfxzyvE1SlShWGDx/O8OHD2b17NytWrGDDhg1s2rSJkiVLcvjwYZMHKSIikl9+3Jv9PJl7bT8VxeWbd6hWsXQ+RJTVpk2biI2NzZie0qtXL+bPn8/q1asZO3ZslvpDhgzJdFy+fHk+//xzXF1d89yus7Mzzs7OADRs2BAbm79TiXXr1rFv3z6+++47mjVrBkDLli0BmD17NkOHDsXR0ZGDBw/SvXt3+vbtm3Ftp06dMr6vV68eAK6urpmG8UztkbbNaN68OTNmzOC3335j0qRJmbrDREREiqJz0bcfWMdggAtG1DOX1atXU65cOTp27AhArVq1aNiwIWvWrMl2svJ7773HihUrWLFiBYGBgbRp04Y333yT3bt3P1K799uxYwcuLi74+PiQkpKS8eXr60tycjIHDx4EwMvLi9WrVzNv3jwOHz5Mamrqo96Sh2KSvcPKlSvHgAEDWLFihSmaExERKTBlbK0fXAkoW+qhd556JFFRUezcuZN27dqRlJREbGwssbGxdO7cmatXr/L7779nuaZmzZp4eXnh5eVF27Zt+eSTT3B1deWjjz56pHbvd/36dSIiIvD09Mz0dbfH58aNG0D6XJ9+/fqxcuVKnn/+eVq1asXUqVO5c+eOie6ScQrmX1BERKSQ6lS/Kqv2R+Rax6VSaepVq5BPEWW2Zs0aUlNTWbduHevWrctyfvXq1fj6+ubahpWVFbVq1WLLli0mbbdSpUpUr16dzz77LNvzLi4uAJQtW5YxY8YwZswYIiIiWL9+PTNnzqRkyZK8/fbbub6GKSkJEhERuceQ1jVZ++clklNzfvr51ba1sC5RMIsGBwUF4eLiwrRp07KcW7BgARs3biQuLi7jCbDspKWlcfr0aRwcHB6qXVtbWyB9LcF7X6dNmzZs2LCBMmXKZDtBOzsuLi4MHjyYtWvXcurUKYBM7ZuTkiAREZF7eD5Wkc/6+TD6h4MkpWadBzPYtyYvP1mjACKDI0eOcPLkSV5//XVatGiR5XxiYiI7duxg/fr1PPfccxnlYWFhlClTBkgfsgoKCuL06dMZvS55bfdugrNo0SKeeuopSpQogZeXF927d2fVqlW88sorDB48mLp165KUlER4eDibN29m9uzZlC5dmn79+uHn54e7uztlypRhz549HD9+PGNCtpOTE5UqVSI4OBgPDw9Kly5N9erVsbe3N+n9tDJooZ8Mdyd2nzhxooAjERGRghZ+/Tbf7TpP6OloklPTqF+tAi89WYMmNUz7QZwXU6ZM4bvvvmPTpk0ZQ0v3SktLw8/PD1dXV7799lt27drFwIEDM9WpWLEiNWvW5OWXX+bZZ599qHZTU1OZMmUKv/zyCzExMRgMhozPzsTERAIDAwkODubixYuUKVMGV1dX2rVrx4gRI7CxseGjjz5i586dhIeHk5qaiqurK3379s0U68aNG/nkk084f/48KSkpTJs2jT59+pjydioJupeSIBEREcthkqfDRERERIqaPM0Jun37NuvWrePcuXPcuJF1IzkrKyumTp1q0gBFREREzMHo4bBDhw4xbNiwjGf8s23Myopjx46ZLLj8puEwERERy2F0T9C0adNISUnhs88+48knn6RSpUrmjEtERETErIxOgo4cOcKrr77K008/bc54RERERPKF0ROjy5Urp94fERERKTaMToI6derEb7/9Zs5YRERERPKN0UnQ22+/zfXr15k8eTIXLlzI8mSYiIiISFFi9NNhdevWxcoq931SrKysOHr0qEkCKwh6OkxERMRyGD0xulevXg9MgkRERESKiiKzbcaSJUv48ssviYqKok6dOvz73/+madOm2dYdN24cq1evzlJeunRpDh48mONrqCdIREQySUmEb3pCzDn45zpwql3QEQGwe/du5s+fz4kTJ7h58yb29vbUqVOHnj170qNHD15++WV27979wHY2bdpEREQEu3fvZuTIkZQoYdqNJAICApg1a1ah/VwtEklQSEgIb7/9NpMmTaJJkyYsXbqUVatWERwczGOPPZal/q1bt0hISMhUNmDAAJo1a8a0adNyfB0lQSIiksmOmbDpg/Tva7WDl4OggEdFNm7cyKhRo/Dz86NXr15UrFiRS5cusXPnTgwGAzNnzuT06dPExcVlXDNnzhwOHz7M3LlzM7VVv3595s+fz6xZszhy5Ag2NnnaSOKBrly5wpUrV2jUqJFJ2zWVh3q38fHx3Lp1i7S0tCznsktKHtWiRYvo3bs3L7zwAgATJ05kx44dLFu2jDFjxmSpX758ecqXL59xvG/fPsLDw5kxY4bJYxMRkWLqRjhs++jv4zNb4WgQePYusJAg/TOxfv36zJ49O9M0ld69e2d8LteunbnHysHBgZIlSz5yMpKcnIyNjY3R02OcnZ1xdnZ+pNc0pzwlQcHBwcydO5ewsLAc65h624ykpCSOHDnC4MGDM5X7+vpy4MABo9pYvnw5derUoXHjxiaNTUREirH1/4aUO/eVjYfanaBUuYKJCbh58ybOzs7ZJiJ5Hc66O1wF4OnpmVF+4sQJLl68SIcOHXjvvfeIiIhgzZo1XLt2jV27dpGamspnn33Grl27uHLlCpUqVaJp06b4+/tTtWrVLO3fO8Li4eHB8OHDcXR0ZPHixcTExODp6cmkSZOoU6dOXm/HIzH6bm3cuJExY8aQkpJCv379MBgMdOvWjaeffhobGxvq16/PyJEjTR5gTEwMqampODk5ZSp3dHQkKirqgdffunWLX375hb59+5o8NhERKaZOb4Rja7KWx0bAjo/zP557eHl5sXPnTj799FOOHz/+SEvW9O3bl+effx6ApUuX8sMPP/DDDz9kqjNv3jzOnTvH5MmTmTVrFqVKleLGjRvY2try1ltvsXDhQvz9/Tl37hwDBgwgMTHxga+7du1atm3bxvjx45k2bRqXLl3itddeIyUl5aHfy8Mwuifoyy+/xM3NjVWrVhEfH8/333/Pc889R8uWLTl58iQDBgygbt26Zgs0u4zXmO64NWvWkJqaSs+ePc0RloiIFDcpiRDin/P50FnQ6CVwyt9ei7vGjh3LhQsXmDdvHvPmzaNcuXK0aNGCZ599lq5du+aprXuHqxo2bJjtnCAnJ6csQ2+1atViwoQJGcepqak0btyYdu3asX37djp16pTr69rY2DBv3jxKliyZUfbmm29y6NChfB21MToJOnHiBCNGjKBUqVLcuZPePXh37NHd3Z0XXniBwMBAOnbsaNIA7e3tsba2ztLrEx0dnaV3KDs//vgjnTt31pYfIiJinJPr4XrO0z5IS4Zd86FbwfQIOTo6smTJEg4dOsSOHTs4dOgQv//+O5s2bSI0NJQpU6aY9PU6dOiQbafD0qVL+f777wkPD+f27dsZ5WfPnn1gm61atcqUALm7uwNw+fJlE0RsPKOHw9LS0jISCTs7OyB9qOmuWrVqcerUKROHB7a2tnh6ehIaGpqpPDQ0FB8fn1yvPXToEMePH8+YUC0iIvJAzg2gRMnc67g0yZ9YcuHt7c3IkSOZP38+27Zto2XLlixfvpyTJ0+a9HWqVKmSpezbb7/lP//5D61atSIgIIDly5fz448/Ahg1HFaxYsVMx7a2tkZfa0pGJ0FVq1bl0qVLQHoS5OjoyF9//ZVx/syZM5QuXdr0EQKDBg1i9erVLF++nLCwMKZMmUJkZCT9+/cHwN/fH3//rF2XP/zwA0888QTNmzc3S1wiIlIMOdSCVqNyPu/SFLz75V88RqhQoQIvv/wyAKdPnzZp29n1AgUHB9OyZUvGjRtH69at8fb2xsHBwaSvmx+MHg5r3Lgxv//+O2+++SYAfn5+fPPNN9jZ2WEwGFi6dCnt27c3S5Bdu3YlJiaGuXPnEhkZibu7O4GBgbi4uADZd5/FxcUREhLCa6+9ppWuRUQkb556Gw79mD4ROhMr6DYTTLyoYF5cvnyZatWqZSk/c+YMkH3PTW7u9sIkJCRQrpxxT71lV3fVqlV5et3CwOgkaMCAAWzcuJGEhATs7OwYPXo0hw4dyni0rk6dOrzzzjtmC/Sll17ipZdeyvbct99+m6WsXLlyRj9CLyIikoltWegyFZb/M3N508HwWMEu/Dds2DAcHBzo2rUrNWvWJCEhgb179/LVV1/h4+OT54nFbm5uQPr6Q0899RQlSpTAy8sr12vatGnDggULmDdvHt7e3vzxxx+sX7/+od9TQTE6CfL29sbb2zvj2MHBgZ9++onjx49jbW2Nm5ubyZfbFhERKTD1e0Kt9nBmS/pxGUfwm5D7NflgxIgRrF+/ngULFhAVFYXBYKB69eoMHjyYYcOG5fmzuH379rz44ossXbqU2bNnYzAYHrhzwsiRI4mNjWXx4sUkJibSvHlzFi5caPKHo8ytSGybkV+0bYaIiGRy7RTMbQWpSdBjFjR+uaAjEhPK87YZe/bs4bfffiM6OppBgwbh5uZGfHw8R48excPDgwoVKpgjThERkfznVAcG/QK3LkHdZws6GjExo5Og1NRUxowZw/r16zEYDFhZWdGtWzfc3NywsbFh5MiRDB48mOHDh5szXhERkfxVvQlQ8I/Ei+kZPXC4YMECNmzYwLhx4wgJCcm0THepUqXo2LEj27ZtM0uQIiIiIqZmdBIUFBREz549+ec//4m9vX2W825uboSHh5s0OBERERFzMToJioiIyHWF5goVKnDz5k2TBCUiIiJibkYnQWXLluXGjRs5nj9//nyRXC1SRERELJPRSVCTJk1Yu3Yt2T1Rf/PmTVauXEmLFi1MGpyIiIiIuRidBA0fPpxz584xcOBAtm7dCqSvp/P999/Tu3dv7ty5w7Bhw8wVp4iIiIhJ5WmxxG3btjF+/HiuXbuWfrGVFQaDAUdHR6ZPn07r1q3NFmh+0GKJIiIiliPPK0YnJSWxc+dOwsLCMBgMPPHEE7Ru3dpsO8jnJyVBIiIilkPbZtxDSZCIiBQFu3fvZv78+Zw4cYKbN29ib29PnTp16NmzJz169ODll19m9+7dD2xn06ZNREREsHv3bkaOHGmWPUBjY2P5+uuv8fPzw9PT0+TtP4o8b5shIiJiMW5cgLM7IC0ZnL3gscZgZVWgIW3cuJFRo0bh5+fHe++9R8WKFbl06RI7d+5k27Zt9OjRg0mTJhEXF5dxzZw5czh8+DBz587N1FaVKlVYvXo1s2bNYsSIEWZLgmbNmoWzs3PRToLWrl3LkiVLOH/+fLaPy1tZWXH06FGTBSciIlIgbl+HtW/AsXXAPQMm1RpCj4D0/xaQRYsWUb9+fWbPno3VPQlZ7969SUtLA6B27dqZrnFwcKBkyZI0atQoX2Mt7IxOgubMmUNAQACOjo74+PhQsWJFc8YlIiJSMBLj4OsecPVw1nOX/4RF3WDIBqhaP/9jI31ZGmdn50wJ0F157ckJCAhg1qxZAJl6ae5OC7lz5w6zZs3i559/JjIykipVqtC3b19effXVjNeKj4/nk08+YdOmTVy7do0KFSrg7u7OxIkTKVWqFB06dABgwoQJTJgwAYBp06bRp0+fvL95EzM6CVq6dCnNmzdn4cKFlCxZ0pwxiYiIFJy9X2afAN2VdAt+nQj/WJl/Md3Dy8uLoKAgPv30U5555hk8PDyyTYiM0bdvX65cucKKFStYunQp1tbWGedSUlIYMmQIYWFhjBgxAg8PDw4ePMicOXO4efMm48aNA9ITms2bNzN69GieeOIJbty4wf79+7l16xaurq7MmjWLUaNG8eqrr+Ln5wfA448//ug3wgSMToLi4+N55plnlACJiEjxduC7B9c5vQluRkBFF/PHc5+xY8dy4cIF5s2bx7x58yhXrhwtWrTg2WefpWvXrnlqy9nZGWdnZwAaNmyIjc3facG6devYt28f3333Hc2aNQOgZcuWAMyePZuhQ4fi6OjIwYMH6d69O3379s24tlOnThnf16tXDwBXV9dCNxxndL9ZvXr1uHz5sjljERERKXjRYUZUMkDMOXNHki1HR0eWLFnC8uXLeeONN2jatCm///47o0ePzhhuMoUdO3bg4uKCj48PKSkpGV++vr4kJydz8OBBIL1navXq1cybN4/Dhw+TmppqshjMzeieoP/7v//j9ddfp3PnzoVudreIiIjJ2JaDRCM2BLcta/5YcuHt7Y23tzeQ/gTWG2+8wfLlyxk4cCDu7u6P3P7169eJiIjI8TP/7gNSEyZMwMnJiZUrV/Lpp59SqVIlevbsyejRowv9GoJGJ0HNmzfnww8/pF+/fjRq1AgXF5csE7CsrKyYOnWqyYMUERHJN3W7wp/Lcq9T8fH0R+YLiQoVKvDyyy/z+++/c/r0aZMkQZUqVaJ69ep89tln2Z53cUkfCixbtixjxoxhzJgxREREsH79embOnEnJkiV5++23HzkOczI6Cfrzzz959913SUlJYe/evezduzdLHSVBIiJS5D35Ghxekb42UE5834AS1jmfN6PLly9TrVq1LOVnzpwB0tf+yQtbW1sAEhISKFeuXEZ5mzZt2LBhA2XKlMHNzc2otlxcXBg8eDBr167l1KlTWdovbIxOgj788ENsbGyYM2cOTZs2pUKFCuaMS0REpGBU84bnFsCqYZCalPX8k69Bs3/lf1z/M2zYMBwcHOjatSs1a9YkISGBvXv38tVXX+Hj40Pjxo3z1N7dBGfRokU89dRTlChRAi8vL7p3786qVat45ZVXGDx4MHXr1iUpKYnw8HA2b97M7NmzKV26NP369cPPzw93d3fKlCnDnj17OH78OL169QLAycmJSpUqERwcjIeHB6VLl6Z69erY29ub/N7kldHbZjRs2JBRo0YxdOhQc8dUYLRthoiIZIg5n/64/JltkPq/FaObDQHX5gUaVkhICOvXr+fIkSNERUVhMBioXr06HTt2ZNiwYZl6c+4aN24coaGhbN++Pcu51NRUpkyZwi+//EJMTAwGgyHjczAxMZHAwECCg4O5ePEiZcqUwdXVlXbt2jFixAhsbGz46KOP2LlzJ+Hh4aSmpuLq6krfvn0ZOHBgxmts3LiRTz75hPPnz5OSklJo1gkyOgny8/Nj4MCBvPLKK2YOqeAoCRIREbEcRj8i36dPH9asWUNKSoo54xERERHJF0bPCWrSpAlbt27lhRde4MUXX6R69eqZVpa86+6CSiIiIiKFmdHDYXXr1s184X1LdBsMBqysrDh27JjpostnGg4TERGxHEb3BE2bNs2ccYiIiIjkK6N7giyBeoJEREQsh9ETowvakiVL8PPzw8vLiz59+mS7WOO9kpKS+Pzzz/Hz86NBgwa0a9eOb775Jp+iFRERkcLO6OGwghQSEsLUqVOZNGkSTZo0YenSpQwdOpTg4GAee+yxbK8ZM2YMly9fZvLkydSoUYPo6OhCuVqliIiIFIwiMRzWt29fPDw8mDJlSkZZ586d6dKlC2PGjMlS/7fffuPNN9/k119/xcHBwejX0XCYiIiI5Sj0w2FJSUkcOXIEX1/fTOW+vr4cOHAg22s2btyIl5cXixcv5qmnnqJz585MmTKF+Pj4/AhZREREioBCPxwWExNDamoqTk5OmcodHR0JDQ3N9prw8HD27duHra0tAQEBxMbGMmXKFCIjI/niiy/yI2wREREp5Ap9EnTX/esS5VQGf69ZNHPmTMqXLw/AxIkTGTJkCNeuXcuSUImIiIjlKfTDYfb29lhbWxMVFZWpPDo6OsdkpnLlylStWjUjAYK/d8m9dOmS+YIVERGRIqPQJ0G2trZ4enpmGfoKDQ3Fx8cn22saN25MZGRkpjlA586dA8DFxcVssYqIiEjRUeiTIIBBgwaxevVqli9fTlhYWMb8nv79+wPg7++Pv79/Rv1nn32WSpUq8e6773Lq1Cn27dvHhx9+SJcuXXB0dCyotyEiIiKFSJGYE9S1a1diYmKYO3cukZGRuLu7ExgYmNGrc/ny5Uz1y5Yty6JFi5gyZQrPP/88FSpUoGPHjtk+Ti8iIiKWqUisE5RftE6QiIiI5SgSw2EiIiIipqYkSERERCySkiARERGxSEqCRERExCIpCRIRERGLpCRIRERELJKSIBEREbFISoJERETEIikJEhEREYukJEhEREQskpIgERERsUhKgkRERMQiKQkSERERi6QkSERERCySkiARERGxSEqCRERExCIpCRIRERGLpCRIRERELJKSIBEREbFISoJERETEIikJEhEREYukJEhEREQskpIgERERsUhKgkRERMQiKQkSERERi6QkSERERCySkiARERGxSEqCRERExCIpCRIRERGLpCRIRERELJKSIBEREbFIRSYJWrJkCX5+fnh5edGnTx/27t2bY91du3bh4eGR5SssLCwfIxYREZHCzKagAzBGSEgIU6dOZdKkSTRp0oSlS5cydOhQgoODeeyxx3K8Ljg4mIoVK2YcOzg45Ee4IiIiUgQUiZ6gRYsW0bt3b1544QXc3NyYOHEilStXZtmyZble5+DgQOXKlTO+rK2t8yliERERKewKfU9QUlISR44cYfDgwZnKfX19OXDgQK7XPv/88yQlJeHm5saIESN48sknjXpNDw+Ph45XRMRSnThxoqBDEMmTQt8TFBMTQ2pqKk5OTpnKHR0diYqKyvaaypUr8/777/PFF18QEBBAzZo1eeWVV9izZ09+hCwiIiJFQKHvCbrLysrKqDKAWrVqUatWrYxjHx8fIiIi+PLLL2nWrFmOr6G/YkRERCxHoe8Jsre3x9raOkuvT3R0dJbeodw0bNiQ8+fPmzo8ERERKaIKfRJka2uLp6cnoaGhmcpDQ0Px8fExup1jx45RuXJlU4cnIiIiRVSRGA4bNGgQ/v7+eHt707hxY5YtW0ZkZCT9+/cHwN/fH4AZM2YAsHjxYqpXr07t2rVJTk5mzZo1bNy4kYCAgAJ7DyIiIlK4FIkkqGvXrsTExDB37lwiIyNxd3cnMDAQFxcXAC5fvpypfnJyMtOnT+fq1avY2dlRu3ZtAgMDadu2bUGELyIiIoWQlcFgMBR0ECIiIiL5rdDPCRIRERExByVBIiIiYpGUBBVSedkwFmD37t306dMHLy8vOnTo8MAtRYozU9+7PXv2MHz4cNq0aYOHhwerVq0yZ/iFjqnvZ0BAQJbNjX19fc35Fgq1vNzfyMhIxowZw9NPP029evUYN25cPkYqUvwoCSqE7m4YO3z4cIKCgvDx8WHo0KFcunQp2/rh4eEMGzYMHx8fgoKCePXVV5kyZQrr16/P58gLnjnu3e3bt3F3d2f8+PHY2dnl11spFMz1s1izZk1+++23jK+1a9fmx9spdPJ6f5OSkrC3t2fYsGE0bNgwn6MVKYYMUug8//zzhvHjx2cq69Spk+Hjjz/Otv6MGTMMnTp1ylT273//2/DCCy+YLcbCytz3rlGjRoaVK1eaJtgiwBz384svvjB069bN9MEWQXm9v/caNmyY4Z133jFXaCIWQT1BhczdDWPvHx7IbcPYgwcPZqnfunVr/vrrL5KTZ4N7GgAAB0ZJREFUk80Wa2Gje2da5ryf4eHhtGnTBj8/P0aPHk14eLjp30Ah9zD3V0RMS0lQIfMwG8Zeu3YNR0fHTGVOTk6kpKQQExNjtlgLG9070zLX/fT29mbatGksWLCAKVOmcO3aNfr3729x9/th7q+ImFaRWCzREuVlw9jszhn+t/xTbtcUV7p3pmXq+3n/oqUNGzakY8eOBAUFMWjQoEcNt8jJ6/0VEdNRT1Ah8zAbxjo5OXHt2rUs9W1sbKhUqZLZYi1sdO9MK7/uZ9myZalduzbnzp0zSdxFhak2hxaRh6ckqJB5mA1jGzVqlG39Bg0aULJkSbPFWtjo3plWft3PxMREzp49a3EbHJtqc2gReXjW77///vsFHYRkVq5cOQICAqhcuTJ2dnbMmTOHvXv3MnXqVCpUqIC/vz+//vornTp1AuDxxx9nwYIFREdH4+LiwqZNm5g3bx7jxo2jdu3aBfxu8pc57l18fDxhYWFcu3aN5cuX4+7uTvny5UlOTqZ8+fIF+XbNzhz3c/r06dja2pKWlsa5c+f44IMPOH/+PB988AEVKlQoyLeb7/J6fwGOHTvGtWvX2LhxIwaDATc3N27evImDg0MBvhORoklzggqhvG4Y6+rqSmBgINOmTWPZsmVUqVKF8ePH06VLl4IIv0CZ49799ddfDBw4MOM4ICCAgIAAevfuzX//+9/8eWMFxBz388qVK7z11lvcuHEDe3t7GjVqxI8//pjRpiXJ6/0F6NWrV6bjLVu24OLiwubNm/MlZpHiRBuoioiIiEXSnCARERGxSEqCRERExCIpCRIRERGLpCRIRERELJKSIBEREbFISoJERETEIikJEhEREYukJEhEREQskpIgERERsUhKgkRERMQiKQkSERERi6QkSERERCySkiARERGxSEqCRERExCIpCRIRERGLpCRIRERELJKSIJECcPHiRTw8PAgICHjoNsaNG4eHh4cJoxIRsSxKgkRERMQiKQkSERERi6QkSERERCySTUEHIFKcxMXFsWDBAkJDQ7lw4QLx8fFUq1aNLv/fzt2ERLXGcRz/Hl9GBUFMDURchC5EQXNVSCClaWlg4EJlRAVfoNHAaGPSwo2ooOAiEFGZEgQXYyCIkO+IuHAhqMhsjBRFcCNOGS0cPa2a21zrqnPncm3m99k9z3nO/zznwMCPZ55zCgtpbGwkKirqt+fu7e2Rl5dHU1MTt27dor+/n+3tbeLi4igtLeXZs2eEhZ3/yX758oXu7m6mpqY4Pj4mIyODV69ekZWV5RlzdnZGf38/S0tLbG9v43K5iI+PJzc3l+bmZmJjY/+T5yEicp0pBIn40cHBAQ6Hg4KCAp48eUJYWBgrKysMDg7idDoZGhq6sMb8/Dzv3r3DarUSHx/P3Nwcb968YX9/n46OjnPja2truXHjBo2NjRwdHWG322loaGB2dpbo6GgATk5OGBoaoqCggLy8PKKiotjY2GBsbIzV1VXGxsawWCx+fx4iIteZQpCIHyUnJ7OwsEB4eLinz2q10tvbS19fH+vr62RmZv5jDafTicPhICMjA4DKykqampp4//49ZWVl3L5922t8eno6bW1tnnZKSgrNzc1MTExQXl4OgMViYWlpicjISM+4iooKsrOzef36NTMzMxQVFf3b2xcR+aNoT5CIH1ksFk8AcrvduFwuDg8PycnJAWBtbe3CGjk5OZ4ABGAYBnV1dQBMT0+fG19TU+PVvnv3LgA7OzteNX4EoNPTUz5//szh4aFn7Pr6+mVvUUQkYGglSMTPRkZGGB0dZWtri7OzM69jLpfrwvNTUlLO9aWmpgKwu7t77lhycrJX+8f+nqOjI6/+yclJ7HY7TqeTk5OTK89LRCTQKASJ+JHdbqezs5N79+5RVVXFzZs3CQ8P5+DggJaWFkzTvLCGYRhXumZoaOgv+3++1tTUFC9evCAzM5PW1lYSExOJiIjg9PSUurq6S81LRCTQKASJ+NH4+DhJSUkMDAwQEvLXv82Li4uXrrG1tfXbvr+v+lxlXhEREQwPD3u9ofbx40ef6omIBALtCRLxo5CQEAzD8FpZcbvdDAwMXLrG8vIym5ubnrZpmgwODgKQn5/v07xCQ0MxDMPr7znTNOnr6/OpnohIINBKkIgfPXr0iJ6eHurr63n48CHHx8dMTEz88vs+v5OWlkZ1dTVWq5WEhARmZ2dZXl6mpKSE7Oxsn+ZVWFjIhw8fqK6u5unTp7jdbmZmZvj27ZtP9UREAoFCkIgf1dbWYpomDoeD9vZ2EhISePz4MaWlpZd+Bf3BgweejyV++vSJuLg4bDYbNpvN53kVFxfz9etX3r59S1dXFzExMdy/f5+XL19y584dn+uKiPzJDFM7IkWuhZ+/GP38+fP/ezoiIgFPe4JEREQkKCkEiYiISFBSCBIREZGgpD1BIiIiEpS0EiQiIiJBSSFIREREgpJCkIiIiAQlhSAREREJSgpBIiIiEpS+Az1J4E7gUzaAAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pointplot(x='alpha', y='value', hue=\"variable\", palette=['C0','C0','C1','C1'], markers=[\"d\",\"o\",\"d\",\"o\"], data=withinmeans.melt(id_vars='alpha'), dodge=0.4, join=False)\n",
    "plt.ylabel(\"mean AUROC\")\n",
    "plt.xlabel(\"alpha\")\n",
    "plt.ylim([0.5,1.02])\n",
    "plt.legend(loc=[1.1,0.2]) #, title=\"\", title_fontsize=\"16\")\n",
    "plt.savefig(\"051921_1.pdf\", bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkEAAAEiCAYAAAD+jAbEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3de1zO9/8/8MelMxUdkFXbiBqto9PIMWSaQ7GcGlOGiJlFH77CPhPNNoeto0I2K0yUqC2UQ+SjYS0zQpOlWAcpRTpdvz/8XNNKrnK9q8v1uN9u3W6u9+F5Pa/rfevWw/vweonEYrEYRERERAqmTUs3QERERNQSGIKIiIhIITEEERERkUJiCCIiIiKFxBBERERECokhiIiIiBQSQxAREREpJLkIQb/88gs8PDwwePBgmJmZ4cCBAy/cJyMjAx988AEsLS0xePBgBAQEgEMiERER0VNyEYIePnwIU1NTrFy5Eurq6i/cvrS0FO7u7tDT00NUVBR8fHywfft2hIeHN0O3REREJA+UW7oBaQwdOhRDhw4FAKxYseKF28fGxuLRo0fYsGED1NXVYWpqiszMTISHh8PNzQ0ikUjolomIiKiVk4szQY2VlpaGPn361DprNGjQIOTl5eH27dvP3c/MzAxmZmbN0SIRERG1MLk4E9RYBQUF6Ny5c61l+vr6knXGxsYye6/ey76XWa1nXfhqpiB17fztBKkLAGcWnRGsthCEOnaAcMePiIhk55UMQQDqXPJ6elO0vFwK++tzC2EK62gLU5dqEer4TRPo+MlbgAXk7z8gRNT6vJIhSF9fHwUFBbWWFRYWAgD09PRaoiUikhNCBdjXV18SpC4RNd0rGYKsra3x9ddf4/Hjx1BTUwMApKSkoFOnTjAyMmrh7ohIEfFSNFHrIxc3RpeVleHKlSu4cuUKampqkJubiytXriA3NxcAsHHjRnz44YeS7ceNGwcNDQ0sX74c165dw5EjRxAaGsonw4iIiEhCLkLQ77//DicnJzg5OaG8vBz+/v5wcnLCt99+CwDIz89Hdna2ZHstLS3s2LEDeXl5mDRpEj7//HO4u7vDzc2tpT4CERERtTJycTmsf//+yMjIeO76L774os4yMzMzRERECNkWERERyTG5OBNEREREJGsMQURERKSQGIKIiIhIITEEERERkUJiCCIiIiKFxBBERERECokhiIiIiBQSQxAREREpJIYgIiIiUkgMQURERKSQGIKIiIhIITEEERERkUJiCCIiIiKFxBBERERECokhiIiIiBQSQxAREREpJOWWboBeHSeHDBWk7tBTJwWpS0REio1ngoiIiEghMQQRERGRQmIIIiIiIoXEEEREREQKiSGIiIiIFBJDEBERESkkhiAiIiJSSAxBREREpJAYgoiIiEghMQQRERGRQmIIIiIiIoXEEEREREQKiSGIiIiIFBJDEBERESkkhiAiIiJSSAxBREREpJDkJgRFRETA3t4eFhYWmDhxIs6fP//C7ceMGQNLS0uMHj0aMTExzdQpERERyQO5CEHx8fFYv349PDw8EBMTAxsbG8yZMwe5ubn1bh8ZGYmvv/4anp6eiIuLw8cff4z//ve/SEpKaubOiYiIqLWSixAUHh4OZ2dnTJ48GSYmJli1ahU6duyI3bt317t9bGwsXFxcMHbsWBgbG+O9997DlClTEBYW1sydExERUWvV6kNQRUUFLl++DDs7u1rL7ezs8Ouvvz53HzU1tVrL1NTUcOnSJVRWVgrWKxEREcmPVh+CioqKUF1dDX19/VrL9fT0kJ+fX+8+gwYNwv79+5Geng6xWIxLly4hKioKlZWVKCoqao62iYiIqJVTbukGpCUSiaRaBgALFixAfn4+pk2bBrFYDD09PTg5OWHbtm1QUlISulUiIiKSA63+TJCOjg6UlJTqnPUpLCysc3boKXV1dfj5+SEtLQ1JSUk4ceIEDA0N0a5dO+jo6DRH20RERNTKtfoQpKqqCnNzc6SkpNRanpKSAhsbmwb3VVFRgYGBAZSUlBAfH4/hw4ejTZtW/5GJiIioGcjF5TA3Nzd4e3vD0tIStra22L17N/Ly8jB16lQAgLe3NwDgyy+/BADcvHkTv/32G6ytrVFSUoLw8HBcv34dX3zxRYt9BiIiImpd5CIEOTo6oqioCMHBwcjLy4OpqSlCQ0NhaGgIALhz506t7WtqarBz507cvHkTysrK6N+/P3bv3g0jI6OWaJ+IiIhaoQZDUHl5OVasWIGePXti7ty5z90uNDQUGRkZ8PPzg6qqqsybBABXV1e4urrWu27Xrl21XpuYmHCEaKJGODlkqCB1h546KUhdIiJZaDAERUVF4ejRo/jkk08aLDJ69Gh8++236Nu3r+QSFRERNQ+GWKKmafAu4SNHjmDYsGF44403GizyxhtvwN7eHj///LNMmyMiIiISSoMhKCMjA3369JGqkK2tLa5evSqTpoiIiIiE1mAIKisrg7a2tlSFtLW1UVpaKpOmiIiIiITWYAjS0tJ67tQU/1ZQUAAtLS2ZNEVEREQktAZDUM+ePXH8+HGpCiUlJaFnz54yaYqIiIhIaA2GIEdHR/z222+IiIhosEhkZCR+++03vPfeezJtjoiIiEgoDT4i7+zsjH379sHX1xcXL17E5MmT0bNnT2hqaqKsrAx//PEH9u3bh7i4OFhZWcHJyam5+iYiIiJ6KQ2GICUlJQQHB+Pjjz9GXFwc4uPj62wjFovRt29ffPPNN5yhnYiIiOTGC6fN0NXVxQ8//ICkpCQcOXIE165dQ1lZGdq1awdTU1M4ODjA3t6+OXolIiIikhmp5w6zt7dn2CEiIqJXRoM3Rtfn0aNHyMvLw6NHj4Toh4iIiKhZSHUm6N69e9i2bRuOHDmCnJwcyXJDQ0M4ODhg9uzZ0NPTE6xJIiIiIll7YQhKS0uDp6cnCgsLoaysjO7du0NLSwulpaW4efMmduzYgdjYWAQGBsLKyqo5eiYiIiJ6aQ2GoMLCQnh4eKC6uhqrV6/GxIkToa6uLllfXl6O6OhobN68GR4eHjh8+DDPCBEREZFcaPCeoO3bt6O0tBQ7d+7E9OnTawUgAFBXV8e0adMQHh6O0tJS7NixQ9BmiYiIiGSlwRB08uRJTJgwAebm5g0WMTc3x/jx46WeYoOIiIiopTUYgnJycmBtbS1VISsrK+Tm5sqkKSIiIiKhNRiC2rRpg8rKSqkKVVVVoU2bRj9xT0RERNQiGkwtb775Js6dOydVodTUVLzxxhsyaYqIiIhIaA2GoBEjRuDo0aM4depUg0WSk5Nx9OhRjBo1SqbNEREREQmlwRD04YcfwsDAAJ6enti0aROys7Nrrc/OzsamTZvg6emJzp07Y+bMmYI2S0RERCQrDY4TpKmpie3bt8PDwwOhoaEICwtDu3btJIMllpaWQiwW44033kBQUBA0NTWbq28iIiKil/LCEaO7du2KgwcPYt++fUhISMD169eRn5+Pdu3aoU+fPhg1ahRcXFygoaHRHP0SERERyYRUc4epq6tjxowZmDFjhtD9EBERETULmT7Tfv78eVmWIyIiIhKMTELQxYsX4ebmxjNFREREJDdeeDksKysLP/zwA27duoX27dtjwoQJGDx4MADg2rVr2LBhA1JSUiASieDo6Ch4w0RERESy0GAIun79OqZOnYqysjLJsri4OGzYsAE1NTXw8fEBAEyYMAHz5s1D165dhe2WiIiISEYaDEFBQUGoqKjAypUrMWDAANy6dQvr1q3D119/jfv372PgwIFYuXIlR4omIiIiudNgCDp//jwmTZokudene/fuEIvFWLhwIYYNG4aQkJBmaZKIiIhI1hq8MbqoqAjm5ua1lr399tsAACcnJ+G6IiIiIhJYgyGoqqoK6urqtZapqakBANq3by9cV/WIiIiAvb09LCwsMHHixBc+jn/o0CFMmDABVlZWsLOzw9KlS5Gfn99M3RIREVFr98JH5EUiUaOWCyE+Ph7r16+Hh4cHYmJiYGNjgzlz5iA3N7fe7S9cuABvb284Ozvj8OHDCAwMRGZmJpYuXdpsPRMREQktKCgIZmZmWLhwYb3rz507BzMzs1o/dnZ2mDNnDtLT01+qrr+/P2pqamTyOZ7l7+8PMzMzmdetzwsfkV+5ciVWr15dZ7mHhwfatKmdoUQiES5cuCC77v6/8PBwODs7Y/LkyQCAVatWITk5Gbt374aXl1ed7dPS0mBgYIBZs2YBAIyNjfHBBx/A19dX5r0REVHr13vZ9y36/he+EmaC8YMHDwIATpw4gaKiIujo6NS7nY+PDywsLCAWi3H37l2EhYVh1qxZOHjwIIyNjRtdNzU1FQEBAZg/f36dLPCyXFxcJEPxCK3BENS3b99maaIhFRUVuHz5Mtzd3Wstt7Ozw6+//lrvPra2tti8eTOSkpIwfPhwFBUVIT4+HkOGDGmOlomIiAR38eJFZGVlYejQoTh58iTi4uLwwQcf1LutiYkJrK2tJa979eoFBwcHJCcnY/r06U2uK43KykooKytLfQXJwMAABgYGTX6/xmgwBO3atatZmmhIUVERqquroa+vX2u5np4eUlJS6t3HxsYGGzduxNKlS/H48WNUVVXBzs4OGzZsaI6WiYiIBBcTEwMlJSWsXbsWU6ZMQUxMjNRhRVNTE8CTe38bW9ff3x8BAQEAUOvhqYyMDNy+fRsjRozA6tWrkZOTg9jYWBQUFODcuXOorq7Gli1bcO7cOdy9excdOnRAnz594O3tjc6dO9epn5GRIVlmZmYGDw8P6OnpYefOnZIHt9asWYMePXo07ot7hmzPYQmovgT5vFR548YN+Pr6YsGCBdi/fz+2bduG/Pz8ei/rERERyZvHjx/jp59+wsCBA9G5c2eMHz8ely5dQmZmZr3b19TUoKqqCpWVlcjOzsa6deugoaGB4cOHN7qui4sL3n//fQBAZGQk9u7di71799aqExISgqysLKxduxYBAQFQU1PD/fv3oaqqik8//RTbtm2Dt7c3srKyMG3aNDx+/PiFn/nQoUM4efIkVq5cCT8/P+Tm5mLBggX1BjlpSTWLfEvS0dGBkpJSnSe7CgsL65wdemrr1q2wtLTERx99BAB46623oKGhAVdXVyxZsgRdunQRvG8iIiKhJCYmoqSkRDJcjZOTE7Zu3Yro6Oh6HwKaPXt2rddaWlr45ptv6twPJE3dZy9XWVlZQVm5bpTQ19dHYGBgrZMV3bp1k8w0AQDV1dWwtbXFsGHDcOrUKYwaNarBz6ysrIyQkBCoqKhIli1evBjp6emwtbVtcN/n1mxo5YgRI567TiQSQV1dHUZGRhg5ciScnZ2hpKTUpCYaoqqqCnNzc6SkpGDMmDGS5SkpKXBwcKh3n/Ly8jq9PH0tFotl3iMREVFzio6OhqamJkaOHAngScCwsrJCbGwsPv300zo3K69evRqWlpYAgHv37iEmJgaLFy9GaGgo+vXr1+S6zzNixIh6r9ZERkZiz549yM7OxsOHDyXLb968+cKaAwcOrBWATE1NAQB37tyRqqf6NPhpxGLxc39qampQWFiIEydOYNWqVXB3d0dlZWWTG2mIm5sboqOjsW/fPmRmZsLX1xd5eXmYOnUqAMDb2xve3t6S7YcPH47ExERERkYiOzsbFy5cgK+vL8zNzfHaa68J0iMREVFzyM/Px5kzZzBs2DBUVFSgpKQEJSUlcHBwwN9//42zZ8/W2adr166wsLCAhYUFhg4dik2bNsHY2BhfffXVS9V9nk6dOtVZtmvXLvz3v//FwIED4e/vj3379uHHH38EAKkuh/17fEJVVVWp932eBs8EJSUlvbDAvXv3sHPnToSGhuK7776TXIKSJUdHRxQVFSE4OBh5eXkwNTVFaGgoDA0NAdRNgRMnTkRZWRkiIiKwYcMGaGlpoX///li2bJnMeyMiImpOsbGxqK6uxuHDh3H48OE666Ojo2FnZ9dgDZFIhG7duuH48eMyrfts/X+Li4vDgAEDsHz5csmy7OxsqeoJ5aXvCdLV1cWnn36KK1euID4+XpAQBACurq5wdXWtd119T7HNmDFDMucZERHRqyImJgaGhobw8/Orsy4sLAzHjh1DaWmp5Amw+tTU1ODGjRvQ1dVtUt2nZ2HKy8sbfJ9n1bftgQMHpNpXKDK7MXrAgAHw9/eXVTkiIiL6l8uXL+PatWtYtGgR+vfvX2f948ePkZycjISEBEyaNEmyPDMzE23btgXwzz1BN27ckFwhaWxdExMTAE8GMx4yZAjatGkDCwuLBnsfPHgwwsLCEBISAktLS/zvf/9DQkJCk78LWZBZCFJTU0N1dbWsyhEREdG/REdHQyQSwdnZud71gwYNQpcuXRATE1MrBD07Y0L79u3RtWtXbNy4EWPHjm1S3eHDh2P69OmIjIxEYGAgxGJxrXF96uPp6YmSkhLs3LkTjx8/Rr9+/bBt2zbJTdgtQSSW0eNSPj4+SE1NxZEjR2RRrkU8navkRQfyWUINxR6t9dWLN2qCaTragtQFgPX7hBlxYeipk4LUFXIYfXk7fvJ27AD+7j1LHo8fUWsgk8ESz5w5g5iYGE5LQURERHKjwf8+rFixosGdy8vLkZmZievXr0NfXx/z5s2TaXNEREREQmkwBEVHR7+wQNu2beHo6IilS5eiY8eOMmuMiIiISEgNhqDExMQGd1ZXV4eOjo7UI0gSERERtRYNhqCngxESERERvWpkcgqnoKAAYWFhcHR0lEU5IiIiIsE1+bnKmpoaHD9+HFFRUUhOTkZVVRXatWsny96IiIiIBNPoEPTnn39i//79OHjwIAoLC6GtrY1x48Zh9OjRGDhwoBA9EhEREcmcVCHo4cOHiI+Px/79+5GWlgYlJSXY2tqisLAQa9euhYODg9B9EhEREclUgyHo4sWLiIqKws8//4yHDx+iV69eWLFiBcaNG4eSkhKMHj26ufokIiIikqkGb4yePn06Tp06hSlTpuDgwYM4cOAAZs6cCR0dHYhEoubqkYiIiOoRFBQEMzMzLFy4sN71586dg5mZWa0fOzs7zJkzB+np6S9V19/fHzU1NTL5HP9WUlICf39/XL58WZD6T73wclh5eTkePHiA0tJSQRshIiISyl+fNzzDudBeX31JkLoHDx4EAJw4cQJFRUXQ0dGpdzsfHx9YWFhALBbj7t27CAsLw6xZs3Dw4EEYGxs3um5qaioCAgIwf/58QcYKLCkpQUBAAAwMDGBubi7z+k812HlcXBxcXFxw/PhxfPDBBxg1ahQCAwORk5MjWENERET0YhcvXkRWVhaGDh2KyspKxMXFPXdbExMTWFtbw8bGBmPGjMHmzZtRVlaG5OTkl6or7xoMQSYmJvjPf/6DkydP4ptvvkHXrl0RFBSEUaNGYfHixRCJRJDRJPRERETUCDExMVBSUsLatWvRpUsXxMTESL2vpqYmAKCqqqrRdf39/REQEAAAMDc3l1xme+rRo0f46quvYG9vj7fffhv29vYIDg6udemsrKwMa9euxbBhw/D2229j4MCBmDVrFjIzM3H79m2MGDECwJMzWE/rHzhwQPovR0pSPR2mrKwMBwcHODg4IC8vDwcOHEB0dDTEYjG8vb0RGxsLBwcH2NvbQ0tLS+ZNEhER0T8eP36Mn376CQMHDkTnzp0xfvx4bN26FZmZmTAxMamzfU1NDaqqqiSXwzZv3gwNDQ0MHz680XVdXFxw9+5dREVFITIyEkpKSpL9q6qqMHv2bGRmZmL+/PkwMzNDWloagoKCUFxcjOXLlwMA/Pz8kJSUhCVLluDNN9/E/fv3cfHiRTx48ADGxsYICAjAwoULMW/ePNjb2wMAXn/9dZl/j40eJ6hTp07w8PCAh4cHUlNTERUVhSNHjiAxMREqKiq4dEmY655ERET0RGJiIkpKSuDk5AQAcHJywtatWxEdHY2lS5fW2X727Nm1XmtpaeGbb76pcz+QNHUNDAxgYGAAALCysoKy8j9R4vDhw7hw4QJ++OEH9O3bFwAwYMAAAEBgYCDmzJkDPT09pKWlYdy4cXBxcZHsO2rUKMm/e/bsCQAwNjaGtbV1E74h6bzU3Uz9+vXDl19+idOnT2PNmjW1TocRERGRMKKjo6GpqYmRI0cCALp16wYrKyvExsbW+8TW6tWrERUVhaioKISGhmLw4MFYvHgxUlNTX6ruvyUnJ8PQ0BA2NjaoqqqS/NjZ2aGyshJpaWkAAAsLC0RHRyMkJASXLl1CdXX1y34lTSKTW7o1NTUxbdo0REVFyaIcERERPUd+fj7OnDmDYcOGoaKiAiUlJSgpKYGDgwP+/vtvnD17ts4+Xbt2hYWFBSwsLDB06FBs2rQJxsbG+Oqrr16q7r/du3cPOTk5MDc3r/Xz9IzP/fv3ATy512fKlCnYv38/3n//fQwcOBDr16/Ho0ePZPQtSafJc4cRERFR84uNjUV1dTUOHz6Mw4cP11kfHR0NOzu7BmuIRCJ069YNx48fl2ndDh06wMjICFu2bKl3vaGhIQCgXbt28PLygpeXF3JycpCQkICNGzdCRUUFy5Yta/A9ZIkhiIiISI7ExMTA0NAQfn5+ddaFhYXh2LFjKC0tlTwBVp+amhrcuHEDurq6TaqrqqoK4MlYgs++z+DBg3HkyBG0bdu23hu062NoaAh3d3ccOnQI169fB4Ba9YXEEERERCQnLl++jGvXrmHRokXo379/nfWPHz9GcnIyEhISMGnSJMnyzMxMtG3bFsCTS1YxMTG4ceOG5KxLY+s+DTjh4eEYMmQI2rRpAwsLC4wbNw4HDhzArFmz4O7ujrfeegsVFRXIzs5GUlISAgMDoaGhgSlTpsDe3h6mpqZo27YtfvnlF1y9elVyQ7a+vj46dOiAuLg4mJmZQUNDA0ZGRs8dDLKpGIKIiIjkRHR0NEQiEZydnetdP2jQIMnYPs+GIF9fX8m/27dvj65du2Ljxo0YO3Zsk+oOHz4c06dPR2RkJAIDAyEWi5GRkQEVFRVs374doaGh2Lt3L27fvo22bdvC2NgYw4YNg4qKCgCgT58++OmnnxAaGorq6moYGxtjxYoVmDlzJgCgTZs2WLduHTZt2gQ3NzdUVVXBz88PEydOlMn3+JRIzNEOJZ4+3ZaRkSH1Pr2XfS9IL9FaX714oyaYpqMtSF0AWL9PmEw99NRJQeoKdewA+Tt+8nbsAP7uPUsejx9RayD7CT+IiIiI5ECj/vvw8OFDHD58GFlZWbh//36dKTNEIhHWr18v0waJiIiIhCB1CEpPT8fcuXMlz/jXhyGIiIiI5IXUIcjPzw9VVVXYsmUL3nnnHXTo0EHIvoiIiIgEJXUIunz5MubNm4d3331XyH6IiIiImoXUN0Zramry7A8RERG9MqQOQaNGjcLp06eF7IWIiIio2UgdgpYtW4Z79+5h7dq1+Ouvv+o8GSa0iIgI2Nvbw8LCAhMnTsT58+efu+3y5cthZmZW58fa2roZOyYiIqLWTOp7gvr06QORSIT09HRERkbWu41IJMIff/whs+aeio+Px/r167FmzRr07t0bkZGRmDNnDuLi4vDaa6/V2X7lypXw8vKqtWzatGno27evzHsjIiIi+SR1CHJycoJIJBKyl+cKDw+Hs7MzJk+eDABYtWoVkpOTsXv37jphBwC0tLSgpaUleX3hwgVkZ2fjyy+/bLaeiYiIqHWTOgR98cUXQvbxXBUVFbh8+TLc3d1rLbezs8Ovv/4qVY19+/ahR48esLW1FaJFIiKiZpWamoqtW7ciIyMDxcXF0NHRQY8ePTBhwgSMHz8eM2bMQGpq6gvrJCYmIicnB6mpqfD09ESbNrKdSMLf3x8BAQGNmo6qObX6CVSLiopQXV0NfX39Wsv19PSQkpLywv0fPHiAn3/+GUuWLBGqRSIiauXs/O1a9P3PLDojs1rHjh3DwoULYW9vj9WrV6N9+/bIzc3FmTNncPLkSYwfPx5r1qxBaWmpZJ+goCBcunQJwcHBtWp16tQJ0dHRCAgIwPz582UeglxcXDB48GCZ1pSlJoWgsrIyPHjwADU1NXXW1XePjizUdylOmstzsbGxqK6uxoQJE4Roi4iIqFmFh4ejV69eCAwMrPV30NnZWfJ3uXv37rX20dXVhYqKyks/IFRZWQllZWWpb48xMDCAgYHBS72nkBoVguLi4hAcHIzMzMznbnPlypWXbupZOjo6UFJSQn5+fq3lhYWFdc4O1efHH3+Eg4MDxzgiIqJXQnFxMQwMDOoNIo09k/P0chUAmJubS5ZnZGTg9u3bGDFiBFavXo2cnBzExsaioKAA586dQ3V1NbZs2YJz587h7t276NChA/r06QNvb2907ty5Tv1nL4eZmZnBw8MDenp62LlzJ4qKimBubo41a9agR48ejf06XorU39axY8fg5eWFqqoqTJkyBWKxGO+99x7effddKCsro1evXvD09JR5g6qqqjA3N69z6SslJQU2NjYN7pueno6rV69KbqgmIiKSdxYWFjhz5gw2b96Mq1evvtSQNS4uLnj//fcBAJGRkdi7dy/27t1ba5uQkBBkZWVh7dq1CAgIgJqaGu7fvw9VVVV8+umn2LZtG7y9vZGVlYVp06bh8ePHL3zfQ4cO4eTJk1i5ciX8/PyQm5uLBQsWoKqqqsmfpSmkPhO0fft2mJiY4MCBAygrK8OePXswadIkDBgwANeuXcO0adPw1ltvCdKkm5sbvL29YWlpCVtbW+zevRt5eXmYOnUqAMDb2xsA6jz9tXfvXrz55pvo16+fIH0RERE1t6VLl+Kvv/5CSEgIQkJCoKmpif79+2Ps2LFwdHRsVK1nL1dZWVlBWbluLNDX169z6a1bt27w8fGRvK6uroatrS2GDRuGU6dOYdSoUQ2+r7KyMkJCQqCioiJZtnjxYqSnpzfrQ0xSh6CMjAzMnz8fampqePToEQBIrj2amppi8uTJCA0NxciRI2XepKOjI4qKihAcHIy8vDyYmpoiNDQUhoaGAIA7d+7U2ae0tBTx8fFYsGBBiz3aT0REJGt6enqIiIhAeno6kpOTkZ6ejrNnzyIxMREpKSnw9fWV6fuNGDGi3r+jkZGR2LNnD7FdSkkAABdaSURBVLKzs/Hw4UPJ8ps3b76w5sCBA2sFIFNTUwD1/z0XktQhqKamRnJfjbq6OoAnT1491a1bN+zZs0fG7f3D1dUVrq6u9a7btWtXnWWamppSP0JPREQkbywtLWFpaQkAKCkpwccff4x9+/Zh5syZklAhC506daqzbNeuXfD19YWbmxsGDRoEbW1tiMViTJ48WarLYe3bt6/1WlVVFQCk2leWpL4nqHPnzsjNzQXwJATp6enh999/l6z/888/oaGhIfsOiYiIqEHa2tqYMWMGAODGjRsyrV3fWaC4uDgMGDAAy5cvx6BBg2BpaQldXV2Zvm9zkPpMkK2tLc6ePYvFixcDAOzt7fH9999DXV0dYrEYkZGRGD58uGCNEhER0ZNLRl26dKmz/M8//wRQ/5mbhjw9C1NeXg5NTU2p9qlv2wMHDjTqfVsDqUPQtGnTcOzYMZSXl0NdXR1LlixBenq65NG6Hj164D//+Y9gjRIREREwd+5c6OrqwtHREV27dkV5eTnOnz+PHTt2wMbGptE3FpuYmAB4Mv7QkCFD0KZNG1hYWDS4z+DBgxEWFoaQkBBYWlrif//7HxISEpr8mVqK1CHo2WuPwJOBlw4ePIirV69CSUkJJiYmMh9pkoiIiGqbP38+EhISEBYWhvz8fIjFYhgZGcHd3R1z585t9N/i4cOHY/r06YiMjERgYCDEYvELp7nw9PRESUkJdu7cicePH6Nfv37Ytm2bIA9HCUkkfpkBBl4xZmZmANCoOU56L/tekF6itb4SpO40HW1B6gLA+n3CzMIy9NRJQeoKdewA+Tt+8nbsAP7uPUsejx9Ra9Do35xffvkFp0+fRmFhIdzc3GBiYoKysjL88ccfMDMzg7a2cL/oRERERLIidQiqrq6Gl5cXEhISIBaLIRKJ8N5778HExATKysrw9PSEu7s7PDw8hOyXiIiISCakvnAYFhaGI0eOYPny5YiPj681TLeamhpGjhyJkyd56pSIiIjkg9QhKCYmBhMmTMCHH34IHR2dOutNTEyQnZ0t0+aIiIiIhCJ1CMrJyWlwwlJtbW0UFxfLpCkiIiIioUkdgtq1a4f79+8/d/2tW7fkcrRIIiIiUkxSh6DevXvj0KFDqO+J+uLiYuzfvx/9+/eXaXNEREREQpE6BHl4eCArKwszZ87EiRMnADwZT2fPnj1wdnbGo0ePMHfuXKH6JCIiIpIpqR+Rt7CwQEBAAFauXIkVK1YAADZs2ACxWAw9PT0EBASge/fugjVKREREJEuNGixx6NChSEpKwpkzZ5CZmQmxWIw333wTgwYN4gzyREREJFcaPWK0qqoqhg8fzhnjiYiIWkhqaiq2bt2KjIwMFBcXQ0dHBz169MCECRMwfvx4zJgxA6mpqS+sk5iYiJycHKSmpsLT01OQOUBLSkrw3Xffwd7eHubm5jKv/zKEmXCGiIioFTk5ZGiLvr8s52E7duwYFi5cCHt7e6xevRrt27dHbm4uzpw5g5MnT2L8+PFYs2YNSktLJfsEBQXh0qVLCA4OrlWrU6dOiI6ORkBAAObPny9YCAoICICBgYF8h6BDhw4hIiICt27dqvdxeZFIhD/++ENmzREREVFt4eHh6NWrFwIDAyESiSTLnZ2dUVNTAwB17tHV1dWFiooKrK2tm7XX1k7qEBQUFAR/f3/o6enBxsYG7du3F7IvIiIiqkdxcTEMDAxqBaCnGnsmx9/fHwEBAQBQ6yxNRkYGAODRo0cICAjATz/9hLy8PHTq1AkuLi6YN2+e5L3KysqwadMmJCYmoqCgANra2jA1NcWqVaugpqaGESNGAAB8fHzg4+MDAPDz88PEiRMb/+FlTOoQFBkZiX79+mHbtm1QUVERsiciIiJ6DgsLC8TExGDz5s0YM2YMzMzM6g1E0nBxccHdu3cRFRWFyMhIKCkpSdZVVVVh9uzZyMzMxPz582FmZoa0tDQEBQWhuLgYy5cvB/Ak0CQlJWHJkiV48803cf/+fVy8eBEPHjyAsbExAgICsHDhQsybNw/29vYAgNdff/3lvwgZkDoElZWVYcyYMQxARERELWjp0qX466+/EBISgpCQEGhqaqJ///4YO3YsHB0dG1XLwMAABgYGAAArKysoK/8TCw4fPowLFy7ghx9+QN++fQEAAwYMAAAEBgZizpw50NPTQ1paGsaNGwcXFxfJvqNGjZL8u2fPngAAY2PjVnc5TurzZj179sSdO3eE7IWIiIheQE9PDxEREdi3bx8+/vhj9OnTB2fPnsWSJUskl5tkITk5GYaGhrCxsUFVVZXkx87ODpWVlUhLSwPw5MxUdHQ0QkJCcOnSJVRXV8usB6FJfSbok08+waJFi+Dg4NDq7u4mIiJSNJaWlrC0tATw5Amsjz/+GPv27cPMmTNhamr60vXv3buHnJyc5/7Nf/qAlI+PD/T19bF//35s3rwZHTp0wIQJE7BkyZJWP4ag1CGoX79+WLduHaZMmQJra2sYGhrWuQFLJBJh/fr1Mm+SiIiInk9bWxszZszA2bNncePGDZmEoA4dOsDIyAhbtmypd72hoSGAJxOse3l5wcvLCzk5OUhISMDGjRuhoqKCZcuWvXQfQpI6BP32229YsWIFqqqqcP78eZw/f77ONgxBREREwrpz5w66dOlSZ/mff/4J4MnYP42hqqoKACgvL4empqZk+eDBg3HkyBG0bdsWJiYmUtUyNDSEu7s7Dh06hOvXr9ep39pIHYLWrVsHZWVlBAUFoU+fPtDW1hayLyIiIqrH3LlzoaurC0dHR3Tt2hXl5eU4f/48duzYARsbG9ja2jaq3tOAEx4ejiFDhqBNmzawsLDAuHHjcODAAcyaNQvu7u546623UFFRgezsbCQlJSEwMBAaGhqYMmUK7O3tYWpqirZt2+KXX37B1atX4eTkBADQ19dHhw4dEBcXBzMzM2hoaMDIyAg6Ojoy/24aS+oQlJGRIRmhkoiIiFrG/PnzkZCQgLCwMOTn50MsFsPIyAju7u6YO3duo8cKGj58OKZPn47IyEgEBgZCLBYjIyMDKioq2L59O0JDQ7F3717cvn0bbdu2hbGxMYYNGyZ5WrxPnz746aefEBoaiurqahgbG2PFihWYOXMmgCdjF61btw6bNm2Cm5sbqqqq5G+cID09PT4eT0REckmW01a0NEdHx0Y/Cv/FF188d52SkhLWrFmDNWvW1FmnpqaGRYsWYdGiRc/df9myZS+892fkyJEYOXKk9A03E6nj4sSJExEbG4uqqioh+yEiIiJqFlKfCerduzdOnDiByZMnY/r06TAyMqo1suRTTwdUIiIiImrNpA5Bbm5ukn/7+PjUGaJbLBZDJBLhypUrsuuOiIiISCBShyA/Pz8h+yAiIiJqVlKHIGdnZyH7eKGIiAhs374d+fn56NGjB/7v//4Pffr0ee72FRUVCA4OxsGDB5GXlwd9fX24u7tL7lYnIiIixSZ1CGpJ8fHxWL9+PdasWYPevXsjMjISc+bMQVxcHF577bV69/Hy8sKdO3ewdu1avPHGGygsLGyVAzURERFRy5CLEBQeHg5nZ2dMnjwZALBq1SokJydj9+7d8PLyqrP96dOnkZKSgqNHj0JXVxcAYGRk1Kw9ExERUevWuBGVWkBFRQUuX74MOzu7Wsvt7Ozw66+/1rvPsWPHYGFhgZ07d2LIkCFwcHCAr68vysrKmqNlIiIikgOt/kxQUVERqquroa+vX2u5np4eUlJS6t0nOzsbFy5cgKqqKvz9/VFSUgJfX1/k5eXh22+/bY62iYiIqJVr9SHoqX8/kv+8ZcA/j+tv3LgRWlpaAJ5cQps9ezYKCgrqBCoiIiJSPK3+cpiOjg6UlJSQn59fa3lhYeFzw0zHjh3RuXNnSQAC/pkgLjc3V7hmiYiISG60+hCkqqoKc3PzOpe+UlJSYGNjU+8+tra2yMvLq3UPUFZWFgDA0NBQsF6JiIhIfrT6EAQ8Ga06Ojoa+/btQ2ZmpuT+nqlTpwIAvL294e3tLdl+7Nix6NChA1asWIHr16/jwoULWLduHUaPHg09Pb2W+hhERETUisjFPUGOjo4oKipCcHAw8vLyYGpqitDQUMlZnTt37tTavl27dggPD4evry/ef/99aGtrY+TIkfU+Tk9ERESKSS5CEAC4urrC1dW13nW7du2qs6xbt27YsWOH0G0RERGRnJKLy2FEREREssYQRERERAqJIYiIiIgUEkMQERERKSSGICIiIlJIDEFERESkkBiCiIiISCExBBEREZFCYggiIiIihcQQRERERAqJIYiIiIgUEkMQERERKSSGICIiIlJIDEFERESkkBiCiIiISCExBBEREZFCYggiIiIihcQQRERERAqJIYiIiIgUEkMQERERKSSGICIiIlJIDEFERESkkBiCiIiISCExBBEREZFCYggiIiIihcQQRERERAqJIYiIiIgUEkMQERERKSSGICIiIlJIDEFERESkkBiCiIiISCExBBEREZFCYggiIiIihSQ3ISgiIgL29vawsLDAxIkTcf78+edue+7cOZiZmdX5yczMbMaOiYiIqDVTbukGpBEfH4/169djzZo16N27NyIjIzFnzhzExcXhtddee+5+cXFxaN++veS1rq5uc7RLREREckAuzgSFh4fD2dkZkydPhomJCVatWoWOHTti9+7dDe6nq6uLjh07Sn6UlJSaqWMiIiJq7Vr9maCKigpcvnwZ7u7utZbb2dnh119/bXDf999/HxUVFTAxMcH8+fPxzjvvSPWeZmZmTe5XVkYJVrlAsMpzhSrcCo5HY8nb8eOx+4e8HTug9Ry/jIwMgRohEkarPxNUVFSE6upq6Ovr11qup6eH/Pz8evfp2LEjPvvsM3z77bfw9/dH165dMWvWLPzyyy/N0TIRERHJgVZ/JugpkUgk1TIA6NatG7p16yZ5bWNjg5ycHGzfvh19+/Z97nvwfzFERESKo9WfCdLR0YGSklKdsz6FhYV1zg41xMrKCrdu3ZJ1e0RERCSnWn0IUlVVhbm5OVJSUmotT0lJgY2NjdR1rly5go4dO8q6PSIiIpJTcnE5zM3NDd7e3rC0tIStrS12796NvLw8TJ06FQDg7e0NAPjyyy8BADt37oSRkRG6d++OyspKxMbG4tixY/D392+xz0BERESti1yEIEdHRxQVFSE4OBh5eXkwNTVFaGgoDA0NAQB37typtX1lZSU2bNiAv//+G+rq6ujevTtCQ0MxdOjQlmifiIiIWiGRWCwWt3QTRERERM2t1d8TRERERCQEhqBXTGPmWAOA1NRUTJw4ERYWFhgxYsQLR+Em2ZH1sfL3968zX56dnZ2QH4Ge0ZjjmZeXBy8vL7z77rvo2bMnli9f3oydEtFTDEGvkKdzrHl4eCAmJgY2NjaYM2cOcnNz690+Ozsbc+fOhY2NDWJiYjBv3jz4+voiISGhmTtXPEIdq65du+L06dOSn0OHDjXHx1F4jT2eFRUV0NHRwdy5c2FlZdXM3RLRUwxBr5DGzrG2Z88edOrUCatWrYKJiQkmT54MJycn7Nixo5k7VzxCHStlZeVa8+Vx0uDm0djjaWRkBB8fH0ycOLHWJM9E1LwYgl4RT+dY+/flj4bmWEtLS6uz/aBBg/D777+jsrJSsF4VnZDHKjs7G4MHD4a9vT2WLFmC7Oxs2X8AqqUpx5OIWgeGoFdEU+ZYKygogJ6eXq1l+vr6qKqqQlFRkWC9KjqhjpWlpSX8/PwQFhYGX19fFBQUYOrUqTyWAmvK8SSi1kEuxgki6TVmjrX61j0dMaGhfUg2ZH2s/j0OlpWVFUaOHImYmBi4ubm9bLv0Ao09nkTU8ngm6BXRlDnW9PX1UVBQUGd7ZWVldOjQQbBeFV1zHat27dqhe/fuyMrKkknfVD9ZzW9IRM2PIegV0ZQ51qytrevd/u2334aKiopgvSq65jpWjx8/xs2bNzlnnsBkNb8hETU/pc8+++yzlm6CZENTUxP+/v7o2LEj1NXVERQUhPPnz2P9+vXQ1taGt7c3jh49ilGjRgEAXn/9dYSFhaGwsBCGhoZITExESEgIli9fju7du7fwp3m1CXGsNmzYAFVVVdTU1CArKwuff/45bt26hc8//xza2tot+XFfeY09nsCTSZ0LCgpw7NgxiMVimJiYoLi4mE/0ETUj3hP0CmnsHGvGxsYIDQ2Fn58fdu/ejU6dOmHlypUYPXp0S7SvUIQ4Vnfv3sWnn36K+/fvQ0dHB9bW1vjxxx8lNUk4jT2eAODk5FTr9fHjx2FoaIikpKRm6ZmIOHcYERERKSjeE0REREQKiSGIiIiIFBJDEBERESkkhiAiIiJSSAxBREREpJAYgoiIiEghMQQRERGRQmIIIiIiIoXEEEREREQKiSGIiIiIFBJDEBERESkkhiAiIiJSSAxBREREpJAYgoiIiEghMQQRERGRQmIIIiIiIoXEEETUAm7fvg0zMzP4+/s3ucby5cthZmYmw66IiBQLQxAREREpJIYgIiIiUkgMQURERKSQlFu6AaJXSWlpKcLCwpCSkoK//voLZWVl6NKlC0aPHg1PT09oaGg8d9/bt29jxIgRWLhwIbp27YqtW7ciKysLenp6mDRpEubPnw9l5bq/sg8ePMDXX3+NI0eOoLS0FObm5lixYgWsrKwk29TU1GDr1q04ffo0srKyUFxcDH19fQwdOhSffPIJdHR0BPk+iIhaM4YgIhn6+++/ERUVBQcHB4wdOxbKyspITU3Ftm3bcOXKFWzfvv2FNY4fP47vvvsOrq6u0NfXR1JSEgICApCbmws/P78628+ePRu6urrw9PTE/fv3ER4ejrlz5yIxMRGampoAgMrKSmzfvh0ODg4YMWIENDQ0cOnSJezfvx8XL17E/v37oaqqKvPvg4ioNWMIIpIhY2NjnDhxAioqKpJlrq6u2LJlC4KDg5Geng5LS8sGa1y5cgVRUVEwNzcHAHzwwQdYuHAhDhw4gClTpsDa2rrW9r169cJnn30meW1iYoJPPvkEhw8fxtSpUwEAqqqqOH36NNTV1SXbTZs2DTY2NvDx8cGxY8fg6Oj4sh+fiEiu8J4gIhlSVVWVBKCqqioUFxfj3r17GDhwIADgt99+e2GNgQMHSgIQAIhEInz00UcAgKNHj9bZftasWbVev/POOwCAW7du1arxNABVV1ejpKQE9+7dk2ybnp4u7UckInpl8EwQkYxFRERgz549uHHjBmpqamqtKy4ufuH+JiYmdZZ1794dAJCdnV1nnbGxca3XT+/vuX//fq3l8fHxCA8Px5UrV1BZWdnovoiIXjUMQUQyFB4eji+++AKDBg3CzJkz0alTJ6ioqODvv//G8uXLIRaLX1hDJBI16j2VlJTqXf7sex05cgRLliyBpaUl/u///g9dunSBmpoaqqur8dFHH0nVFxHRq4YhiEiGDh48CENDQ4SFhaFNm3+uNp86dUrqGjdu3Hjusn+f9WlMX2pqavj+++9rPaGWmZnZpHpERK8C3hNEJENt2rSBSCSqdWalqqoKYWFhUtdISUnB5cuXJa/FYjG2bdsGABg5cmST+lJSUoJIJKp1eU4sFiM4OLhJ9YiIXgU8E0QkQ++++y42btyIOXPmYNSoUSgtLcXhw4frHd/ned566y18+OGHcHV1RceOHZGYmIiUlBRMmDABNjY2Tepr9OjRSEhIwIcffggnJydUVVXh2LFjePToUZPqERG9ChiCiGRo9uzZEIvFiIqKwrp169CxY0eMGTMGkyZNkvoRdHt7e8lgiTdv3oSenh4WLFiABQsWNLmv9957D2VlZdi5cyc2bNiA9u3bY/jw4fDy8kL//v2bXJeISJ6JxLwjkqhVeHbE6EWLFrV0O0RErzzeE0REREQKiSGIiIiIFBJDEBERESkk3hNEREREColngoiIiEghMQQRERGRQmIIIiIiIoXEEEREREQKiSGIiIiIFBJDEBERESmk/weCqD/Kwi28lAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(x='alpha', y='value', hue=\"variable\", data=withinmeans.melt(id_vars='alpha'))#, dodge=0.3, join=False)\n",
    "plt.ylabel(\"mean AUROC\")\n",
    "plt.xlabel(\"alpha\")\n",
    "plt.legend(loc=[1.1,0.2]) #, title=\"\", title_fontsize=\"16\")\n",
    "plt.ylim([0.5, 1.0])\n",
    "plt.savefig(\"062620_2.pdf\", bbox_inches='tight')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}