diff --git a/.DS_Store b/.DS_Store
new file mode 100644
index 0000000..5008ddf
Binary files /dev/null and b/.DS_Store differ
diff --git a/.ipynb_checkpoints/fires-checkpoint.ipynb b/.ipynb_checkpoints/fires-checkpoint.ipynb
new file mode 100644
index 0000000..582e80b
--- /dev/null
+++ b/.ipynb_checkpoints/fires-checkpoint.ipynb
@@ -0,0 +1,3800 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Data is from here\n",
+    "\n",
+    "# http://archive.ics.uci.edu/ml/datasets/Parkinsons+Telemonitoring"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import io\n",
+    "import requests\n",
+    "\n",
+    "# Define URL\n",
+    "url = \"http://archive.ics.uci.edu/ml/machine-learning-databases/parkinsons/telemonitoring/parkinsons_updrs.data\"\n",
+    "content = requests.get(url).content\n",
+    "df = pd.read_csv(io.StringIO(content.decode('utf-8')), sep=',')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "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>subject#</th>\n",
+       "      <th>age</th>\n",
+       "      <th>sex</th>\n",
+       "      <th>test_time</th>\n",
+       "      <th>motor_UPDRS</th>\n",
+       "      <th>total_UPDRS</th>\n",
+       "      <th>Jitter(%)</th>\n",
+       "      <th>Jitter(Abs)</th>\n",
+       "      <th>Jitter:RAP</th>\n",
+       "      <th>Jitter:PPQ5</th>\n",
+       "      <th>...</th>\n",
+       "      <th>Shimmer(dB)</th>\n",
+       "      <th>Shimmer:APQ3</th>\n",
+       "      <th>Shimmer:APQ5</th>\n",
+       "      <th>Shimmer:APQ11</th>\n",
+       "      <th>Shimmer:DDA</th>\n",
+       "      <th>NHR</th>\n",
+       "      <th>HNR</th>\n",
+       "      <th>RPDE</th>\n",
+       "      <th>DFA</th>\n",
+       "      <th>PPE</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1</td>\n",
+       "      <td>72</td>\n",
+       "      <td>0</td>\n",
+       "      <td>5.6431</td>\n",
+       "      <td>28.199</td>\n",
+       "      <td>34.398</td>\n",
+       "      <td>0.00662</td>\n",
+       "      <td>0.000034</td>\n",
+       "      <td>0.00401</td>\n",
+       "      <td>0.00317</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.230</td>\n",
+       "      <td>0.01438</td>\n",
+       "      <td>0.01309</td>\n",
+       "      <td>0.01662</td>\n",
+       "      <td>0.04314</td>\n",
+       "      <td>0.014290</td>\n",
+       "      <td>21.640</td>\n",
+       "      <td>0.41888</td>\n",
+       "      <td>0.54842</td>\n",
+       "      <td>0.16006</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1</td>\n",
+       "      <td>72</td>\n",
+       "      <td>0</td>\n",
+       "      <td>12.6660</td>\n",
+       "      <td>28.447</td>\n",
+       "      <td>34.894</td>\n",
+       "      <td>0.00300</td>\n",
+       "      <td>0.000017</td>\n",
+       "      <td>0.00132</td>\n",
+       "      <td>0.00150</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.179</td>\n",
+       "      <td>0.00994</td>\n",
+       "      <td>0.01072</td>\n",
+       "      <td>0.01689</td>\n",
+       "      <td>0.02982</td>\n",
+       "      <td>0.011112</td>\n",
+       "      <td>27.183</td>\n",
+       "      <td>0.43493</td>\n",
+       "      <td>0.56477</td>\n",
+       "      <td>0.10810</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1</td>\n",
+       "      <td>72</td>\n",
+       "      <td>0</td>\n",
+       "      <td>19.6810</td>\n",
+       "      <td>28.695</td>\n",
+       "      <td>35.389</td>\n",
+       "      <td>0.00481</td>\n",
+       "      <td>0.000025</td>\n",
+       "      <td>0.00205</td>\n",
+       "      <td>0.00208</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.181</td>\n",
+       "      <td>0.00734</td>\n",
+       "      <td>0.00844</td>\n",
+       "      <td>0.01458</td>\n",
+       "      <td>0.02202</td>\n",
+       "      <td>0.020220</td>\n",
+       "      <td>23.047</td>\n",
+       "      <td>0.46222</td>\n",
+       "      <td>0.54405</td>\n",
+       "      <td>0.21014</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1</td>\n",
+       "      <td>72</td>\n",
+       "      <td>0</td>\n",
+       "      <td>25.6470</td>\n",
+       "      <td>28.905</td>\n",
+       "      <td>35.810</td>\n",
+       "      <td>0.00528</td>\n",
+       "      <td>0.000027</td>\n",
+       "      <td>0.00191</td>\n",
+       "      <td>0.00264</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.327</td>\n",
+       "      <td>0.01106</td>\n",
+       "      <td>0.01265</td>\n",
+       "      <td>0.01963</td>\n",
+       "      <td>0.03317</td>\n",
+       "      <td>0.027837</td>\n",
+       "      <td>24.445</td>\n",
+       "      <td>0.48730</td>\n",
+       "      <td>0.57794</td>\n",
+       "      <td>0.33277</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1</td>\n",
+       "      <td>72</td>\n",
+       "      <td>0</td>\n",
+       "      <td>33.6420</td>\n",
+       "      <td>29.187</td>\n",
+       "      <td>36.375</td>\n",
+       "      <td>0.00335</td>\n",
+       "      <td>0.000020</td>\n",
+       "      <td>0.00093</td>\n",
+       "      <td>0.00130</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.176</td>\n",
+       "      <td>0.00679</td>\n",
+       "      <td>0.00929</td>\n",
+       "      <td>0.01819</td>\n",
+       "      <td>0.02036</td>\n",
+       "      <td>0.011625</td>\n",
+       "      <td>26.126</td>\n",
+       "      <td>0.47188</td>\n",
+       "      <td>0.56122</td>\n",
+       "      <td>0.19361</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 22 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   subject#  age  sex  test_time  motor_UPDRS  total_UPDRS  Jitter(%)  \\\n",
+       "0         1   72    0     5.6431       28.199       34.398    0.00662   \n",
+       "1         1   72    0    12.6660       28.447       34.894    0.00300   \n",
+       "2         1   72    0    19.6810       28.695       35.389    0.00481   \n",
+       "3         1   72    0    25.6470       28.905       35.810    0.00528   \n",
+       "4         1   72    0    33.6420       29.187       36.375    0.00335   \n",
+       "\n",
+       "   Jitter(Abs)  Jitter:RAP  Jitter:PPQ5   ...     Shimmer(dB)  Shimmer:APQ3  \\\n",
+       "0     0.000034     0.00401      0.00317   ...           0.230       0.01438   \n",
+       "1     0.000017     0.00132      0.00150   ...           0.179       0.00994   \n",
+       "2     0.000025     0.00205      0.00208   ...           0.181       0.00734   \n",
+       "3     0.000027     0.00191      0.00264   ...           0.327       0.01106   \n",
+       "4     0.000020     0.00093      0.00130   ...           0.176       0.00679   \n",
+       "\n",
+       "   Shimmer:APQ5  Shimmer:APQ11  Shimmer:DDA       NHR     HNR     RPDE  \\\n",
+       "0       0.01309        0.01662      0.04314  0.014290  21.640  0.41888   \n",
+       "1       0.01072        0.01689      0.02982  0.011112  27.183  0.43493   \n",
+       "2       0.00844        0.01458      0.02202  0.020220  23.047  0.46222   \n",
+       "3       0.01265        0.01963      0.03317  0.027837  24.445  0.48730   \n",
+       "4       0.00929        0.01819      0.02036  0.011625  26.126  0.47188   \n",
+       "\n",
+       "       DFA      PPE  \n",
+       "0  0.54842  0.16006  \n",
+       "1  0.56477  0.10810  \n",
+       "2  0.54405  0.21014  \n",
+       "3  0.57794  0.33277  \n",
+       "4  0.56122  0.19361  \n",
+       "\n",
+       "[5 rows x 22 columns]"
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "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>subject#</th>\n",
+       "      <th>age</th>\n",
+       "      <th>sex</th>\n",
+       "      <th>test_time</th>\n",
+       "      <th>motor_UPDRS</th>\n",
+       "      <th>total_UPDRS</th>\n",
+       "      <th>Jitter(%)</th>\n",
+       "      <th>Jitter(Abs)</th>\n",
+       "      <th>Jitter:RAP</th>\n",
+       "      <th>Jitter:PPQ5</th>\n",
+       "      <th>...</th>\n",
+       "      <th>Shimmer(dB)</th>\n",
+       "      <th>Shimmer:APQ3</th>\n",
+       "      <th>Shimmer:APQ5</th>\n",
+       "      <th>Shimmer:APQ11</th>\n",
+       "      <th>Shimmer:DDA</th>\n",
+       "      <th>NHR</th>\n",
+       "      <th>HNR</th>\n",
+       "      <th>RPDE</th>\n",
+       "      <th>DFA</th>\n",
+       "      <th>PPE</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>count</th>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>...</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "      <td>5875.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>mean</th>\n",
+       "      <td>21.494128</td>\n",
+       "      <td>64.804936</td>\n",
+       "      <td>0.317787</td>\n",
+       "      <td>92.863722</td>\n",
+       "      <td>21.296229</td>\n",
+       "      <td>29.018942</td>\n",
+       "      <td>0.006154</td>\n",
+       "      <td>0.000044</td>\n",
+       "      <td>0.002987</td>\n",
+       "      <td>0.003277</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.310960</td>\n",
+       "      <td>0.017156</td>\n",
+       "      <td>0.020144</td>\n",
+       "      <td>0.027481</td>\n",
+       "      <td>0.051467</td>\n",
+       "      <td>0.032120</td>\n",
+       "      <td>21.679495</td>\n",
+       "      <td>0.541473</td>\n",
+       "      <td>0.653240</td>\n",
+       "      <td>0.219589</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>std</th>\n",
+       "      <td>12.372279</td>\n",
+       "      <td>8.821524</td>\n",
+       "      <td>0.465656</td>\n",
+       "      <td>53.445602</td>\n",
+       "      <td>8.129282</td>\n",
+       "      <td>10.700283</td>\n",
+       "      <td>0.005624</td>\n",
+       "      <td>0.000036</td>\n",
+       "      <td>0.003124</td>\n",
+       "      <td>0.003732</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.230254</td>\n",
+       "      <td>0.013237</td>\n",
+       "      <td>0.016664</td>\n",
+       "      <td>0.019986</td>\n",
+       "      <td>0.039711</td>\n",
+       "      <td>0.059692</td>\n",
+       "      <td>4.291096</td>\n",
+       "      <td>0.100986</td>\n",
+       "      <td>0.070902</td>\n",
+       "      <td>0.091498</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>min</th>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>36.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>-4.262500</td>\n",
+       "      <td>5.037700</td>\n",
+       "      <td>7.000000</td>\n",
+       "      <td>0.000830</td>\n",
+       "      <td>0.000002</td>\n",
+       "      <td>0.000330</td>\n",
+       "      <td>0.000430</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.026000</td>\n",
+       "      <td>0.001610</td>\n",
+       "      <td>0.001940</td>\n",
+       "      <td>0.002490</td>\n",
+       "      <td>0.004840</td>\n",
+       "      <td>0.000286</td>\n",
+       "      <td>1.659000</td>\n",
+       "      <td>0.151020</td>\n",
+       "      <td>0.514040</td>\n",
+       "      <td>0.021983</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25%</th>\n",
+       "      <td>10.000000</td>\n",
+       "      <td>58.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>46.847500</td>\n",
+       "      <td>15.000000</td>\n",
+       "      <td>21.371000</td>\n",
+       "      <td>0.003580</td>\n",
+       "      <td>0.000022</td>\n",
+       "      <td>0.001580</td>\n",
+       "      <td>0.001820</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.175000</td>\n",
+       "      <td>0.009280</td>\n",
+       "      <td>0.010790</td>\n",
+       "      <td>0.015665</td>\n",
+       "      <td>0.027830</td>\n",
+       "      <td>0.010955</td>\n",
+       "      <td>19.406000</td>\n",
+       "      <td>0.469785</td>\n",
+       "      <td>0.596180</td>\n",
+       "      <td>0.156340</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>50%</th>\n",
+       "      <td>22.000000</td>\n",
+       "      <td>65.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>91.523000</td>\n",
+       "      <td>20.871000</td>\n",
+       "      <td>27.576000</td>\n",
+       "      <td>0.004900</td>\n",
+       "      <td>0.000035</td>\n",
+       "      <td>0.002250</td>\n",
+       "      <td>0.002490</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.253000</td>\n",
+       "      <td>0.013700</td>\n",
+       "      <td>0.015940</td>\n",
+       "      <td>0.022710</td>\n",
+       "      <td>0.041110</td>\n",
+       "      <td>0.018448</td>\n",
+       "      <td>21.920000</td>\n",
+       "      <td>0.542250</td>\n",
+       "      <td>0.643600</td>\n",
+       "      <td>0.205500</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>75%</th>\n",
+       "      <td>33.000000</td>\n",
+       "      <td>72.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>138.445000</td>\n",
+       "      <td>27.596500</td>\n",
+       "      <td>36.399000</td>\n",
+       "      <td>0.006800</td>\n",
+       "      <td>0.000053</td>\n",
+       "      <td>0.003290</td>\n",
+       "      <td>0.003460</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.365000</td>\n",
+       "      <td>0.020575</td>\n",
+       "      <td>0.023755</td>\n",
+       "      <td>0.032715</td>\n",
+       "      <td>0.061735</td>\n",
+       "      <td>0.031463</td>\n",
+       "      <td>24.444000</td>\n",
+       "      <td>0.614045</td>\n",
+       "      <td>0.711335</td>\n",
+       "      <td>0.264490</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>max</th>\n",
+       "      <td>42.000000</td>\n",
+       "      <td>85.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>215.490000</td>\n",
+       "      <td>39.511000</td>\n",
+       "      <td>54.992000</td>\n",
+       "      <td>0.099990</td>\n",
+       "      <td>0.000446</td>\n",
+       "      <td>0.057540</td>\n",
+       "      <td>0.069560</td>\n",
+       "      <td>...</td>\n",
+       "      <td>2.107000</td>\n",
+       "      <td>0.162670</td>\n",
+       "      <td>0.167020</td>\n",
+       "      <td>0.275460</td>\n",
+       "      <td>0.488020</td>\n",
+       "      <td>0.748260</td>\n",
+       "      <td>37.875000</td>\n",
+       "      <td>0.966080</td>\n",
+       "      <td>0.865600</td>\n",
+       "      <td>0.731730</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>8 rows × 22 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          subject#          age          sex    test_time  motor_UPDRS  \\\n",
+       "count  5875.000000  5875.000000  5875.000000  5875.000000  5875.000000   \n",
+       "mean     21.494128    64.804936     0.317787    92.863722    21.296229   \n",
+       "std      12.372279     8.821524     0.465656    53.445602     8.129282   \n",
+       "min       1.000000    36.000000     0.000000    -4.262500     5.037700   \n",
+       "25%      10.000000    58.000000     0.000000    46.847500    15.000000   \n",
+       "50%      22.000000    65.000000     0.000000    91.523000    20.871000   \n",
+       "75%      33.000000    72.000000     1.000000   138.445000    27.596500   \n",
+       "max      42.000000    85.000000     1.000000   215.490000    39.511000   \n",
+       "\n",
+       "       total_UPDRS    Jitter(%)  Jitter(Abs)   Jitter:RAP  Jitter:PPQ5  \\\n",
+       "count  5875.000000  5875.000000  5875.000000  5875.000000  5875.000000   \n",
+       "mean     29.018942     0.006154     0.000044     0.002987     0.003277   \n",
+       "std      10.700283     0.005624     0.000036     0.003124     0.003732   \n",
+       "min       7.000000     0.000830     0.000002     0.000330     0.000430   \n",
+       "25%      21.371000     0.003580     0.000022     0.001580     0.001820   \n",
+       "50%      27.576000     0.004900     0.000035     0.002250     0.002490   \n",
+       "75%      36.399000     0.006800     0.000053     0.003290     0.003460   \n",
+       "max      54.992000     0.099990     0.000446     0.057540     0.069560   \n",
+       "\n",
+       "          ...       Shimmer(dB)  Shimmer:APQ3  Shimmer:APQ5  Shimmer:APQ11  \\\n",
+       "count     ...       5875.000000   5875.000000   5875.000000    5875.000000   \n",
+       "mean      ...          0.310960      0.017156      0.020144       0.027481   \n",
+       "std       ...          0.230254      0.013237      0.016664       0.019986   \n",
+       "min       ...          0.026000      0.001610      0.001940       0.002490   \n",
+       "25%       ...          0.175000      0.009280      0.010790       0.015665   \n",
+       "50%       ...          0.253000      0.013700      0.015940       0.022710   \n",
+       "75%       ...          0.365000      0.020575      0.023755       0.032715   \n",
+       "max       ...          2.107000      0.162670      0.167020       0.275460   \n",
+       "\n",
+       "       Shimmer:DDA          NHR          HNR         RPDE          DFA  \\\n",
+       "count  5875.000000  5875.000000  5875.000000  5875.000000  5875.000000   \n",
+       "mean      0.051467     0.032120    21.679495     0.541473     0.653240   \n",
+       "std       0.039711     0.059692     4.291096     0.100986     0.070902   \n",
+       "min       0.004840     0.000286     1.659000     0.151020     0.514040   \n",
+       "25%       0.027830     0.010955    19.406000     0.469785     0.596180   \n",
+       "50%       0.041110     0.018448    21.920000     0.542250     0.643600   \n",
+       "75%       0.061735     0.031463    24.444000     0.614045     0.711335   \n",
+       "max       0.488020     0.748260    37.875000     0.966080     0.865600   \n",
+       "\n",
+       "               PPE  \n",
+       "count  5875.000000  \n",
+       "mean      0.219589  \n",
+       "std       0.091498  \n",
+       "min       0.021983  \n",
+       "25%       0.156340  \n",
+       "50%       0.205500  \n",
+       "75%       0.264490  \n",
+       "max       0.731730  \n",
+       "\n",
+       "[8 rows x 22 columns]"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.describe()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Need to drop a few varaibles\n",
+    "df.drop(df.columns[[0, 4]], axis=1, inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The columns in the dataframe are:\n",
+      "age\n",
+      "sex\n",
+      "test_time\n",
+      "total_UPDRS\n",
+      "Jitter(%)\n",
+      "Jitter(Abs)\n",
+      "Jitter:RAP\n",
+      "Jitter:PPQ5\n",
+      "Jitter:DDP\n",
+      "Shimmer\n",
+      "Shimmer(dB)\n",
+      "Shimmer:APQ3\n",
+      "Shimmer:APQ5\n",
+      "Shimmer:APQ11\n",
+      "Shimmer:DDA\n",
+      "NHR\n",
+      "HNR\n",
+      "RPDE\n",
+      "DFA\n",
+      "PPE\n",
+      "\n",
+      "There are 20 columns in the dataframe\n",
+      "There are 5875 rows in the dataframe\n"
+     ]
+    }
+   ],
+   "source": [
+    "features = df.columns[:-1]\n",
+    "features.drop('total_UPDRS')\n",
+    "\n",
+    "print \"The columns in the dataframe are:\"\n",
+    "for x in df.columns:\n",
+    "    print x\n",
+    "\n",
+    "print \"\\nThere are %d columns in the dataframe\" % len(df.columns)\n",
+    "print \"There are %d rows in the dataframe\" % df.shape[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Coefficients: \n",
+      "age                       0.3089\n",
+      "sex                       -2.7607\n",
+      "test_time                 0.0194\n",
+      "total_UPDRS               65.4101\n",
+      "Jitter(%)                 -68735.2267\n",
+      "Jitter(Abs)               -24007.3116\n",
+      "Jitter:RAP                -269.0664\n",
+      "Jitter:PPQ5               8228.2380\n",
+      "Jitter:DDP                123.1713\n",
+      "Shimmer                   -7.0602\n",
+      "Shimmer(dB)               46522.3337\n",
+      "Shimmer:APQ3              -121.0717\n",
+      "Shimmer:APQ5              81.0045\n",
+      "Shimmer:APQ11             -15565.9268\n",
+      "Shimmer:DDA               -10.6561\n",
+      "NHR                       -0.6229\n",
+      "HNR                       2.2856\n",
+      "RPDE                      -29.7971\n",
+      "DFA                       19.1736\n",
+      "Mean squared error: 91.50\n",
+      "\n",
+      "                            OLS Regression Results                            \n",
+      "==============================================================================\n",
+      "Dep. Variable:            total_UPDRS   R-squared:                       0.173\n",
+      "Model:                            OLS   Adj. R-squared:                  0.169\n",
+      "Method:                 Least Squares   F-statistic:                     45.04\n",
+      "Date:                Sun, 06 May 2018   Prob (F-statistic):          7.30e-153\n",
+      "Time:                        15:25:04   Log-Likelihood:                -15212.\n",
+      "No. Observations:                4112   AIC:                         3.046e+04\n",
+      "Df Residuals:                    4092   BIC:                         3.059e+04\n",
+      "Df Model:                          19                                         \n",
+      "Covariance Type:            nonrobust                                         \n",
+      "=================================================================================\n",
+      "                    coef    std err          t      P>|t|      [0.025      0.975]\n",
+      "---------------------------------------------------------------------------------\n",
+      "const            38.6877      3.906      9.904      0.000      31.030      46.346\n",
+      "age               0.3089      0.018     16.907      0.000       0.273       0.345\n",
+      "sex              -2.7607      0.380     -7.262      0.000      -3.506      -2.015\n",
+      "test_time         0.0194      0.003      6.717      0.000       0.014       0.025\n",
+      "Jitter(%)        65.4101    245.077      0.267      0.790    -415.075     545.895\n",
+      "Jitter(Abs)   -6.874e+04   1.16e+04     -5.936      0.000   -9.14e+04    -4.6e+04\n",
+      "Jitter:RAP    -2.401e+04   5.62e+04     -0.427      0.669   -1.34e+05    8.62e+04\n",
+      "Jitter:PPQ5    -269.0664    218.762     -1.230      0.219    -697.959     159.826\n",
+      "Jitter:DDP     8228.2380   1.87e+04      0.439      0.661   -2.85e+04     4.5e+04\n",
+      "Shimmer         123.1713     80.066      1.538      0.124     -33.802     280.145\n",
+      "Shimmer(dB)      -7.0602      5.756     -1.227      0.220     -18.346       4.225\n",
+      "Shimmer:APQ3   4.652e+04   5.68e+04      0.819      0.413   -6.49e+04    1.58e+05\n",
+      "Shimmer:APQ5   -121.0717     67.885     -1.783      0.075    -254.163      12.019\n",
+      "Shimmer:APQ11    81.0045     30.589      2.648      0.008      21.034     140.975\n",
+      "Shimmer:DDA   -1.557e+04   1.89e+04     -0.822      0.411   -5.27e+04    2.16e+04\n",
+      "NHR             -10.6561      7.423     -1.435      0.151     -25.210       3.898\n",
+      "HNR              -0.6229      0.084     -7.415      0.000      -0.788      -0.458\n",
+      "RPDE              2.2856      2.195      1.041      0.298      -2.018       6.589\n",
+      "DFA             -29.7971      2.796    -10.659      0.000     -35.278     -24.316\n",
+      "PPE              19.1736      3.576      5.361      0.000      12.162      26.185\n",
+      "==============================================================================\n",
+      "Omnibus:                      107.115   Durbin-Watson:                   2.030\n",
+      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              101.062\n",
+      "Skew:                           0.341   Prob(JB):                     1.13e-22\n",
+      "Kurtosis:                       2.647   Cond. No.                     4.86e+07\n",
+      "==============================================================================\n",
+      "\n",
+      "Warnings:\n",
+      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+      "[2] The condition number is large, 4.86e+07. This might indicate that there are\n",
+      "strong multicollinearity or other numerical problems.\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Start with Linear Regression\n",
+    "from sklearn import model_selection\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.metrics import mean_squared_error\n",
+    "import statsmodels.api as sm\n",
+    "\n",
+    "# Split features from output\n",
+    "X = df.loc[:, df.columns != 'total_UPDRS']\n",
+    "y = df['total_UPDRS']\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1075)\n",
+    "\n",
+    "# Do a regular regression\n",
+    "lm_model = LinearRegression()\n",
+    "lm_results = lm_model.fit(X_train, y_train)\n",
+    "\n",
+    "# Make predictions\n",
+    "y_pred = lm_model.predict(X_test)\n",
+    "\n",
+    "# See the results\n",
+    "# The coefficients are\n",
+    "print \"Coefficients: \\n\", \n",
+    "\n",
+    "for i in range(len(features)):\n",
+    "    print \"%-25s %.4f\" % (features[i], lm_results.coef_[i])\n",
+    "\n",
+    "# The mean squared error\n",
+    "print \"Mean squared error: %.2f\\n\" % mean_squared_error(y_test, y_pred)\n",
+    "\n",
+    "X_train_with_constant = sm.add_constant(X_train)\n",
+    "lm_est = sm.OLS(y_train, X_train_with_constant)\n",
+    "lm_est_results = lm_est.fit()\n",
+    "print(lm_est_results.summary())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Predicted Vs. Actual')"
+      ]
+     },
+     "execution_count": 37,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.figure(figsize=(12,8))\n",
+    "plt.scatter(y_test, y_pred)\n",
+    "plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'k--')\n",
+    "plt.xlabel('Actual')\n",
+    "plt.ylabel('Predicted')\n",
+    "plt.title('Predicted Vs. Actual')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "const                 652.54\n",
+       "age                     1.09\n",
+       "sex                     1.33\n",
+       "test_time               1.01\n",
+       "Jitter(%)              83.21\n",
+       "Jitter(Abs)             7.01\n",
+       "Jitter:RAP        1339656.73\n",
+       "Jitter:PPQ5            29.44\n",
+       "Jitter:DDP        1339934.64\n",
+       "Shimmer               185.63\n",
+       "Shimmer(dB)            75.79\n",
+       "Shimmer:APQ3     24534291.60\n",
+       "Shimmer:APQ5           55.35\n",
+       "Shimmer:APQ11          16.64\n",
+       "Shimmer:DDA      24533893.18\n",
+       "NHR                     8.38\n",
+       "HNR                     5.47\n",
+       "RPDE                    2.13\n",
+       "DFA                     1.67\n",
+       "PPE                     4.54\n",
+       "Name: VIF, dtype: float64"
+      ]
+     },
+     "execution_count": 39,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# https://stackoverflow.com/questions/42658379/variance-inflation-factor-in-python\n",
+    "\n",
+    "# One recommendation is that if VIF is greater than 5,\n",
+    "# then the explanatory variable given by exog_idx is highly collinear with the other explanatory variables,\n",
+    "# and the parameter estimates will have large standard errors because of this.\n",
+    "\n",
+    "from statsmodels.regression.linear_model import OLS\n",
+    "from statsmodels.tools.tools import add_constant\n",
+    "\n",
+    "\n",
+    "def variance_inflation_factors(exog_df):\n",
+    "    '''\n",
+    "    Parameters\n",
+    "    ----------\n",
+    "    exog_df : dataframe, (nobs, k_vars)\n",
+    "        design matrix with all explanatory variables, as for example used in\n",
+    "        regression.\n",
+    "\n",
+    "    Returns\n",
+    "    -------\n",
+    "    vif : Series\n",
+    "        variance inflation factors\n",
+    "    '''\n",
+    "    exog_df = add_constant(exog_df)\n",
+    "    vifs = pd.Series(\n",
+    "        [\n",
+    "            1 /\n",
+    "            (1. - OLS(exog_df[col].values, exog_df.\n",
+    "                      loc[:, exog_df.columns != col].values).fit().rsquared)\n",
+    "            for col in exog_df\n",
+    "        ],\n",
+    "        index=exog_df.columns,\n",
+    "        name='VIF')\n",
+    "    return vifs.round(2)\n",
+    "\n",
+    "\n",
+    "variance_inflation_factors(X_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Dropping variables \n",
+    "# They mean roughly the same thing\n",
+    "df.drop(df.columns[[4, 5, 6, 7, 9, 10, 11, 12, 13, 14]], axis=1, inplace=True)\n",
+    "features = df.columns[:-1]\n",
+    "features.drop('total_UPDRS')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 91,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.preprocessing import StandardScaler\n",
+    "# Because in linear regression the value of the coefficients is \n",
+    "# partially determined by the scale of the feature, \n",
+    "# and in regularized models all coefficients are summed together, \n",
+    "# we must make sure to standardize the feature prior to training.\n",
+    "scaler = StandardScaler()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Split features from output\n",
+    "X = df.loc[:, df.columns != 'total_UPDRS']\n",
+    "y = df['total_UPDRS']\n",
+    "\n",
+    "# Standardize\n",
+    "X = scaler.fit_transform(X)\n",
+    "X = pd.DataFrame(X, columns=features)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 98,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Mean squared error: 91.94\n",
+      "\n",
+      "                            OLS Regression Results                            \n",
+      "==============================================================================\n",
+      "Dep. Variable:            total_UPDRS   R-squared:                       0.155\n",
+      "Model:                            OLS   Adj. R-squared:                  0.153\n",
+      "Method:                 Least Squares   F-statistic:                     83.46\n",
+      "Date:                Sun, 06 May 2018   Prob (F-statistic):          8.20e-143\n",
+      "Time:                        15:55:37   Log-Likelihood:                -15257.\n",
+      "No. Observations:                4112   AIC:                         3.053e+04\n",
+      "Df Residuals:                    4102   BIC:                         3.060e+04\n",
+      "Df Model:                           9                                         \n",
+      "Covariance Type:            nonrobust                                         \n",
+      "===============================================================================\n",
+      "                  coef    std err          t      P>|t|      [0.025      0.975]\n",
+      "-------------------------------------------------------------------------------\n",
+      "const          29.0781      0.154    188.284      0.000      28.775      29.381\n",
+      "age             2.7246      0.161     16.942      0.000       2.409       3.040\n",
+      "sex            -0.8111      0.164     -4.941      0.000      -1.133      -0.489\n",
+      "test_time       0.9755      0.155      6.282      0.000       0.671       1.280\n",
+      "total_UPDRS     0.4932      0.285      1.728      0.084      -0.066       1.053\n",
+      "Jitter:DDP     -1.7263      0.307     -5.630      0.000      -2.327      -1.125\n",
+      "NHR            -1.8208      0.307     -5.935      0.000      -2.422      -1.219\n",
+      "HNR             0.2603      0.212      1.227      0.220      -0.156       0.676\n",
+      "RPDE           -2.2937      0.191    -12.033      0.000      -2.667      -1.920\n",
+      "DFA             1.2991      0.279      4.661      0.000       0.753       1.846\n",
+      "==============================================================================\n",
+      "Omnibus:                      114.603   Durbin-Watson:                   2.027\n",
+      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              110.402\n",
+      "Skew:                           0.362   Prob(JB):                     1.06e-24\n",
+      "Kurtosis:                       2.655   Cond. No.                         5.14\n",
+      "==============================================================================\n",
+      "\n",
+      "Warnings:\n",
+      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
+     ]
+    }
+   ],
+   "source": [
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1075)\n",
+    "\n",
+    "# Do a regular regression\n",
+    "lm_model = LinearRegression()\n",
+    "lm_results = lm_model.fit(X_train, y_train)\n",
+    "\n",
+    "# Make predictions\n",
+    "y_pred = lm_model.predict(X_test)\n",
+    "\n",
+    "# See the results\n",
+    "# The coefficients are\n",
+    "# print \"Coefficients: \\n\", \n",
+    "\n",
+    "# for i in range(len(features)):\n",
+    "#     print \"%-25s %.4f\" % (features[i], lm_results.coef_[i])\n",
+    "\n",
+    "# The mean squared error\n",
+    "print \"Mean squared error: %.2f\\n\" % mean_squared_error(y_test, y_pred)\n",
+    "\n",
+    "# Get better output\n",
+    "X_train_with_constant = sm.add_constant(X_train)\n",
+    "lm_est = sm.OLS(y_train, X_train_with_constant)\n",
+    "lm_est_results = lm_est.fit()\n",
+    "print(lm_est_results.summary())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 99,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "const          1.00\n",
+       "age            1.07\n",
+       "sex            1.13\n",
+       "test_time      1.01\n",
+       "total_UPDRS    3.47\n",
+       "Jitter:DDP     3.94\n",
+       "NHR            3.89\n",
+       "HNR            1.91\n",
+       "RPDE           1.51\n",
+       "DFA            3.23\n",
+       "Name: VIF, dtype: float64"
+      ]
+     },
+     "execution_count": 99,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Much better\n",
+    "variance_inflation_factors(X_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 101,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Predicted Vs. Actual')"
+      ]
+     },
+     "execution_count": 101,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(12,8))\n",
+    "plt.scatter(y_test, y_pred)\n",
+    "plt.xlabel('Actual')\n",
+    "plt.ylabel('Predicted')\n",
+    "plt.title('Predicted Vs. Actual')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 103,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ -90.13602729  -89.61880662  -95.78205543  -97.22438294 -105.78236928\n",
+      "  -97.77936522 -102.08968664  -93.94836455  -96.73739183  -93.88355773]\n",
+      "-96.29820075315617\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Use kfolds for model selection\n",
+    "# Provides train/test indices to split data in train/test sets. \n",
+    "# Split dataset into k consecutive folds (without shuffling by default).\n",
+    "\n",
+    "kfold = model_selection.KFold(n_splits = 10, random_state = 1075, shuffle=True)\n",
+    "\n",
+    "# Specify the model\n",
+    "lm_model = LinearRegression()\n",
+    "\n",
+    "# Will return 10 results due to the number of folds, and use mean squared error as the decision metric\n",
+    "# From the documentation\n",
+    "# All scorer objects follow the convention that higher return values are better than \n",
+    "# lower return values. Thus metrics which measure the distance between the model and the data, \n",
+    "# like metrics.mean_squared_error, are available as neg_mean_squared_error which return the \n",
+    "# negated value of the metric.\n",
+    "\n",
+    "results = model_selection.cross_val_score(lm_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "# To make the results easier to read\n",
+    "print results\n",
+    "print results.mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 104,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ -90.13759839  -89.61928894  -95.78049334  -97.22606236 -105.78074201\n",
+      "  -97.78021632 -102.09000541  -93.94664581  -96.73695773  -93.88320851]\n",
+      "-96.29812188316944\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Ridge Regression\n",
+    "from sklearn.linear_model import Ridge\n",
+    "\n",
+    "ridge_model = Ridge()\n",
+    "results = model_selection.cross_val_score(ridge_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "print results\n",
+    "print results.mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 138,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "GridSearchCV(cv=None, error_score='raise',\n",
+      "       estimator=Ridge(alpha=1.0, copy_X=True, fit_intercept=True, max_iter=None,\n",
+      "   normalize=False, random_state=None, solver='auto', tol=0.001),\n",
+      "       fit_params=None, iid=True, n_jobs=1,\n",
+      "       param_grid={'alpha': array([0.   , 0.002, ..., 0.998, 1.   ])},\n",
+      "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
+      "       scoring=None, verbose=0)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Let's do some tuning to the ridge\n",
+    "# The main tuning parameter for the Ridge model is alpha -\n",
+    "# a regularization parameter that measures how flexible our model is.\n",
+    "# The higher the regularization the less prone our model will be to overfit.\n",
+    "# However it will also lose flexibility and might not capture all of the signal in the data.\n",
+    "\n",
+    "from sklearn.linear_model import RidgeCV\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "\n",
+    "alphas = np.linspace(0, 1, 500)\n",
+    "param_grid = {\"alpha\": alphas}\n",
+    "\n",
+    "ridge_model = Ridge()\n",
+    "grid = GridSearchCV(ridge_model, param_grid=param_grid)\n",
+    "grid.fit(X, y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 142,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'mean_fit_time': array([0.00277774, 0.00228047, 0.00206439, 0.00188494, 0.00185831,\n",
+       "        0.00209594, 0.00537467, 0.00377599, 0.00361196, 0.00240707,\n",
+       "        0.00174332, 0.001755  , 0.00178806, 0.00181238, 0.00181325,\n",
+       "        0.00180459, 0.00186626, 0.00199167, 0.00164175, 0.00167004,\n",
+       "        0.00160829, 0.00168633, 0.00158572, 0.00153899, 0.00163873,\n",
+       "        0.00155592, 0.00150228, 0.00179132, 0.00185108, 0.0017124 ,\n",
+       "        0.00200701, 0.00195058, 0.00228143, 0.00203371, 0.00204198,\n",
+       "        0.00172901, 0.00182398, 0.00159637, 0.00182247, 0.00180332,\n",
+       "        0.00163706, 0.00177439, 0.00171836, 0.00173092, 0.00306543,\n",
+       "        0.0020074 , 0.00214267, 0.00218765, 0.00302132, 0.00216309,\n",
+       "        0.00276025, 0.00261641, 0.00299462, 0.00201035, 0.00208608,\n",
+       "        0.00175166, 0.00174038, 0.00227666, 0.00196902, 0.00217597,\n",
+       "        0.00185029, 0.00197132, 0.00199231, 0.00162299, 0.00162609,\n",
+       "        0.00181031, 0.00174729, 0.00186865, 0.00190536, 0.00155369,\n",
+       "        0.00166861, 0.00166241, 0.00186435, 0.00178432, 0.00178027,\n",
+       "        0.00208831, 0.00223001, 0.00182104, 0.00221459, 0.00170366,\n",
+       "        0.00197061, 0.00351866, 0.00187564, 0.00219973, 0.00210341,\n",
+       "        0.00199898, 0.00159764, 0.00153891, 0.00168721, 0.00163468,\n",
+       "        0.00171471, 0.00159971, 0.00157269, 0.00160758, 0.00177924,\n",
+       "        0.00156363, 0.00154098, 0.00190608, 0.00184536, 0.00188692,\n",
+       "        0.00188899, 0.00190528, 0.00174356, 0.00178496, 0.00154201,\n",
+       "        0.00220776, 0.00267005, 0.00422796, 0.00303523, 0.00496761,\n",
+       "        0.00248027, 0.00404326, 0.0041484 , 0.00378331, 0.004366  ,\n",
+       "        0.00194105, 0.00157237, 0.00180999, 0.00168228, 0.00234032,\n",
+       "        0.00348798, 0.00173839, 0.00182207, 0.00166972, 0.001755  ,\n",
+       "        0.00167735, 0.00165407, 0.00162307, 0.0018096 , 0.00162903,\n",
+       "        0.00184663, 0.0017213 , 0.00157873, 0.00152   , 0.00154829,\n",
+       "        0.00155528, 0.00173338, 0.00188629, 0.00188526, 0.00168133,\n",
+       "        0.00173736, 0.00172106, 0.00177574, 0.00163698, 0.00154829,\n",
+       "        0.00165534, 0.00172106, 0.00165733, 0.00187667, 0.00179537,\n",
+       "        0.00182263, 0.00177193, 0.00202235, 0.0021143 , 0.00212773,\n",
+       "        0.00154734, 0.00162697, 0.00155433, 0.00167966, 0.00180364,\n",
+       "        0.00151364, 0.00153963, 0.00189328, 0.00161338, 0.00262539,\n",
+       "        0.00177097, 0.00209737, 0.00185108, 0.00229629, 0.00188001,\n",
+       "        0.00202958, 0.00199024, 0.00216834, 0.00182033, 0.00193294,\n",
+       "        0.00224845, 0.00214863, 0.00210238, 0.00201559, 0.00256499,\n",
+       "        0.00554204, 0.00493789, 0.00353479, 0.00464765, 0.00329415,\n",
+       "        0.00290298, 0.00197093, 0.00177455, 0.00172567, 0.00178711,\n",
+       "        0.00167759, 0.00170636, 0.00167569, 0.00189177, 0.00165431,\n",
+       "        0.00163833, 0.00234437, 0.00156879, 0.00179474, 0.00158564,\n",
+       "        0.00165598, 0.00166909, 0.00159836, 0.00182192, 0.00167338,\n",
+       "        0.00261633, 0.00165995, 0.00149894, 0.00181198, 0.00159804,\n",
+       "        0.00155695, 0.00171296, 0.00229836, 0.00173561, 0.00155767,\n",
+       "        0.00188335, 0.00160265, 0.00162498, 0.00213377, 0.00176732,\n",
+       "        0.00168268, 0.00214402, 0.0016733 , 0.00176469, 0.00163333,\n",
+       "        0.00182366, 0.00158779, 0.00161433, 0.00177797, 0.00199962,\n",
+       "        0.00203292, 0.00243576, 0.00266075, 0.00212661, 0.0019323 ,\n",
+       "        0.00202497, 0.00252914, 0.00192595, 0.00206693, 0.00210865,\n",
+       "        0.0020109 , 0.00216667, 0.00199699, 0.00291133, 0.00188168,\n",
+       "        0.00215443, 0.0020473 , 0.00170763, 0.00188931, 0.00157332,\n",
+       "        0.00166965, 0.00196997, 0.0018367 , 0.00229375, 0.00205096,\n",
+       "        0.00181739, 0.00161831, 0.00162896, 0.00255092, 0.00180658,\n",
+       "        0.00211803, 0.00168967, 0.00168331, 0.00179338, 0.00159192,\n",
+       "        0.00169492, 0.0017883 , 0.00175333, 0.00172122, 0.00175905,\n",
+       "        0.00151698, 0.00203466, 0.00181707, 0.00176541, 0.00168403,\n",
+       "        0.00181365, 0.00177336, 0.001707  , 0.0016106 , 0.00194645,\n",
+       "        0.00166798, 0.0016164 , 0.0016396 , 0.00180666, 0.00158501,\n",
+       "        0.00175444, 0.00149639, 0.00164763, 0.00240596, 0.00185132,\n",
+       "        0.00169738, 0.0017    , 0.00157531, 0.00160336, 0.00174292,\n",
+       "        0.00154328, 0.00175103, 0.00174061, 0.00163968, 0.001683  ,\n",
+       "        0.00157142, 0.00170875, 0.00178965, 0.00162029, 0.00200907,\n",
+       "        0.00180968, 0.00164326, 0.00164898, 0.00157634, 0.00187047,\n",
+       "        0.00163706, 0.00182668, 0.00175301, 0.001887  , 0.00169873,\n",
+       "        0.00166202, 0.0016319 , 0.00164262, 0.001628  , 0.00172242,\n",
+       "        0.00168435, 0.00177828, 0.0016737 , 0.00173108, 0.00158358,\n",
+       "        0.00163301, 0.00170604, 0.00170795, 0.00196465, 0.0017306 ,\n",
+       "        0.00163937, 0.0016857 , 0.00169992, 0.00189296, 0.00152032,\n",
+       "        0.00171733, 0.00158564, 0.00186133, 0.00182541, 0.00165304,\n",
+       "        0.00173227, 0.00164461, 0.00161195, 0.00184965, 0.00166432,\n",
+       "        0.00161131, 0.00190822, 0.00165399, 0.00164262, 0.00165598,\n",
+       "        0.00170159, 0.00230598, 0.00175635, 0.0016187 , 0.00169865,\n",
+       "        0.0016923 , 0.00163873, 0.00168125, 0.00177375, 0.00181103,\n",
+       "        0.00179561, 0.00166305, 0.00158731, 0.00161974, 0.00168737,\n",
+       "        0.00156291, 0.00213703, 0.00201472, 0.00161139, 0.00161473,\n",
+       "        0.00163905, 0.00169365, 0.00164866, 0.00155044, 0.00188645,\n",
+       "        0.00156371, 0.00158644, 0.0018057 , 0.00167298, 0.00165494,\n",
+       "        0.00199676, 0.00183153, 0.00157539, 0.00168268, 0.00162729,\n",
+       "        0.00180626, 0.00160702, 0.00156434, 0.00166607, 0.00176803,\n",
+       "        0.00165232, 0.00161425, 0.00169301, 0.00185831, 0.00175103,\n",
+       "        0.00162959, 0.00160074, 0.00185132, 0.00166035, 0.00162037,\n",
+       "        0.00178671, 0.00158834, 0.00161004, 0.00170032, 0.00161926,\n",
+       "        0.00166607, 0.00149473, 0.00160138, 0.00161068, 0.00172663,\n",
+       "        0.00215371, 0.00190179, 0.00243139, 0.00209936, 0.0023543 ,\n",
+       "        0.00263643, 0.00286468, 0.00269771, 0.00255831, 0.00239889,\n",
+       "        0.00240596, 0.00222969, 0.00201734, 0.00242337, 0.00211358,\n",
+       "        0.00205437, 0.00222071, 0.00215777, 0.00193866, 0.00191665,\n",
+       "        0.00259296, 0.00168459, 0.00177368, 0.00165296, 0.00168737,\n",
+       "        0.00157134, 0.00160241, 0.00165049, 0.00174228, 0.00156212,\n",
+       "        0.00185895, 0.00154066, 0.00168633, 0.00224829, 0.0017066 ,\n",
+       "        0.00168395, 0.00156196, 0.00153438, 0.00186459, 0.00160376,\n",
+       "        0.00170302, 0.00160273, 0.00150029, 0.00176096, 0.00169102,\n",
+       "        0.00206598, 0.00151563, 0.00158302, 0.00158429, 0.00169802,\n",
+       "        0.00167068, 0.00152342, 0.00164342, 0.00169007, 0.00183606,\n",
+       "        0.001556  , 0.00184401, 0.00179338, 0.00161608, 0.00168363,\n",
+       "        0.00160638, 0.001659  , 0.0015316 , 0.00169698, 0.0015974 ,\n",
+       "        0.00174864, 0.00173402, 0.00160368, 0.00163897, 0.00154169,\n",
+       "        0.00158731, 0.00203037, 0.00166464, 0.00158858, 0.00196266,\n",
+       "        0.00191975, 0.00168737, 0.00160925, 0.00165137, 0.00159931,\n",
+       "        0.00181802, 0.00153502, 0.00155497, 0.00166663, 0.00172297,\n",
+       "        0.00200669, 0.00168967, 0.00150108, 0.00157158, 0.00174769]),\n",
+       " 'mean_score_time': array([0.00071931, 0.00066495, 0.00049496, 0.00054065, 0.00057975,\n",
+       "        0.00105405, 0.00115538, 0.00093293, 0.00077136, 0.00083558,\n",
+       "        0.00058103, 0.00049941, 0.00052897, 0.00044099, 0.00058039,\n",
+       "        0.00049241, 0.00045037, 0.00054304, 0.00042367, 0.00045172,\n",
+       "        0.0005068 , 0.0004216 , 0.00043472, 0.00040627, 0.00042137,\n",
+       "        0.00041302, 0.00041699, 0.00040634, 0.00056028, 0.00051657,\n",
+       "        0.00044862, 0.00044839, 0.00054455, 0.0006632 , 0.00044099,\n",
+       "        0.00041668, 0.00042899, 0.00042995, 0.00046563, 0.00042836,\n",
+       "        0.00043289, 0.00045625, 0.00044839, 0.00041803, 0.00046062,\n",
+       "        0.00094366, 0.00080474, 0.0014437 , 0.00053787, 0.00050306,\n",
+       "        0.00075301, 0.00101296, 0.00054097, 0.00046436, 0.00045204,\n",
+       "        0.00045904, 0.00063856, 0.00061973, 0.00043837, 0.00052039,\n",
+       "        0.00065271, 0.00047867, 0.0005486 , 0.00042359, 0.0004673 ,\n",
+       "        0.00047294, 0.00043066, 0.00058166, 0.00045999, 0.00040531,\n",
+       "        0.00043233, 0.00045164, 0.000422  , 0.00042502, 0.00042868,\n",
+       "        0.00051204, 0.00055027, 0.00067425, 0.00045768, 0.00043599,\n",
+       "        0.00084178, 0.00111604, 0.00042327, 0.0005873 , 0.00088231,\n",
+       "        0.00061464, 0.00050632, 0.00051602, 0.00052277, 0.00041827,\n",
+       "        0.00041358, 0.00052834, 0.00044465, 0.00041167, 0.00045029,\n",
+       "        0.00042741, 0.00042868, 0.00048399, 0.00048494, 0.00045943,\n",
+       "        0.00052937, 0.00048868, 0.00049639, 0.00057936, 0.00063531,\n",
+       "        0.000717  , 0.00102727, 0.00160472, 0.00079346, 0.00068863,\n",
+       "        0.00118899, 0.00111135, 0.00129429, 0.00094175, 0.00119567,\n",
+       "        0.0004487 , 0.0006067 , 0.0005993 , 0.00054034, 0.00048669,\n",
+       "        0.00063864, 0.00049297, 0.00054995, 0.00046833, 0.00043829,\n",
+       "        0.00042335, 0.00056465, 0.00047731, 0.00055273, 0.00042987,\n",
+       "        0.00058794, 0.00058595, 0.00047557, 0.00040611, 0.00040936,\n",
+       "        0.0004104 , 0.00063189, 0.00043408, 0.00046213, 0.00042709,\n",
+       "        0.00048129, 0.0004793 , 0.00050195, 0.00042764, 0.00041032,\n",
+       "        0.00040865, 0.00049869, 0.00041866, 0.00043837, 0.0004433 ,\n",
+       "        0.00059557, 0.00045204, 0.00055599, 0.00047334, 0.00046968,\n",
+       "        0.00048701, 0.00044998, 0.00040174, 0.00042129, 0.00058722,\n",
+       "        0.00050465, 0.00040944, 0.00060074, 0.00054034, 0.00076532,\n",
+       "        0.00052373, 0.00046031, 0.0004793 , 0.00048796, 0.00047604,\n",
+       "        0.00059533, 0.00067027, 0.0006663 , 0.00048868, 0.00055973,\n",
+       "        0.00052802, 0.00087007, 0.00065764, 0.00067639, 0.00048598,\n",
+       "        0.00216564, 0.0009141 , 0.00117461, 0.00105135, 0.0010163 ,\n",
+       "        0.0007383 , 0.00071303, 0.0006063 , 0.00048868, 0.00044592,\n",
+       "        0.0004096 , 0.00043066, 0.00048399, 0.00045689, 0.00041135,\n",
+       "        0.00047032, 0.00045967, 0.00049027, 0.00048661, 0.00044203,\n",
+       "        0.00043567, 0.00042828, 0.00047406, 0.00069865, 0.00049702,\n",
+       "        0.00052762, 0.00041771, 0.00052905, 0.0004317 , 0.00050767,\n",
+       "        0.00045864, 0.00049742, 0.00043805, 0.00045268, 0.00045808,\n",
+       "        0.00045228, 0.00042399, 0.00041397, 0.00046062, 0.00061464,\n",
+       "        0.00041731, 0.00044505, 0.00042971, 0.00042804, 0.00046245,\n",
+       "        0.00045737, 0.00050529, 0.00040929, 0.00042534, 0.00085139,\n",
+       "        0.00055607, 0.00056601, 0.000537  , 0.00069173, 0.00056577,\n",
+       "        0.00061067, 0.00053326, 0.00070063, 0.00063038, 0.00050465,\n",
+       "        0.00057205, 0.0005544 , 0.00064198, 0.000518  , 0.00071629,\n",
+       "        0.00052134, 0.00044735, 0.00055369, 0.0009853 , 0.00051602,\n",
+       "        0.00041231, 0.00053636, 0.00063372, 0.00053867, 0.00059064,\n",
+       "        0.00049766, 0.00042597, 0.00041628, 0.00054669, 0.00044473,\n",
+       "        0.00060066, 0.00044195, 0.00047803, 0.00043726, 0.00054471,\n",
+       "        0.00053136, 0.00042764, 0.00050998, 0.00052643, 0.0004576 ,\n",
+       "        0.00049806, 0.00054439, 0.00056998, 0.00047461, 0.0004216 ,\n",
+       "        0.00045371, 0.00045236, 0.00042931, 0.00047533, 0.00042295,\n",
+       "        0.00042025, 0.00045856, 0.00044401, 0.00046198, 0.00041866,\n",
+       "        0.00047199, 0.00044767, 0.00042597, 0.00058198, 0.00044036,\n",
+       "        0.00042828, 0.00042367, 0.00053501, 0.00041461, 0.00048192,\n",
+       "        0.00049798, 0.00042629, 0.00045204, 0.00044735, 0.00041437,\n",
+       "        0.0004309 , 0.00043162, 0.00045999, 0.00044131, 0.00045228,\n",
+       "        0.00054034, 0.00047874, 0.00041835, 0.00043472, 0.00043424,\n",
+       "        0.00045673, 0.00045609, 0.00048939, 0.00058667, 0.00050298,\n",
+       "        0.00042701, 0.00042033, 0.00047231, 0.00049933, 0.00045069,\n",
+       "        0.00041397, 0.00042208, 0.00045729, 0.00043758, 0.00041739,\n",
+       "        0.00048192, 0.00042597, 0.00041135, 0.00047867, 0.0004344 ,\n",
+       "        0.00041795, 0.0004189 , 0.00056545, 0.00043933, 0.00042129,\n",
+       "        0.00045602, 0.00041032, 0.00046666, 0.00049559, 0.0004406 ,\n",
+       "        0.0004286 , 0.00042097, 0.00040809, 0.00045737, 0.00055766,\n",
+       "        0.00041334, 0.0005765 , 0.00042001, 0.00043774, 0.00058031,\n",
+       "        0.00046341, 0.00056569, 0.0004971 , 0.00041064, 0.00043297,\n",
+       "        0.00042709, 0.00050124, 0.00041278, 0.0005606 , 0.00048304,\n",
+       "        0.0004727 , 0.00041405, 0.00041477, 0.00051626, 0.00042828,\n",
+       "        0.00043464, 0.00054471, 0.00048169, 0.00043066, 0.00049535,\n",
+       "        0.00044322, 0.00049798, 0.00043941, 0.00051697, 0.00044425,\n",
+       "        0.00047334, 0.00041199, 0.00046825, 0.00042534, 0.00042804,\n",
+       "        0.00051936, 0.00043511, 0.00049329, 0.00050131, 0.00043535,\n",
+       "        0.00043631, 0.00043503, 0.00047771, 0.00041731, 0.00042367,\n",
+       "        0.00043829, 0.00051705, 0.00045371, 0.00044155, 0.00050608,\n",
+       "        0.00041302, 0.00042097, 0.00052436, 0.00042431, 0.00045292,\n",
+       "        0.00048828, 0.00042764, 0.00045896, 0.00049734, 0.00040897,\n",
+       "        0.00042462, 0.00040364, 0.00041707, 0.00056036, 0.00050902,\n",
+       "        0.00077367, 0.00048033, 0.0006636 , 0.00076334, 0.00084186,\n",
+       "        0.00065819, 0.00078495, 0.00066868, 0.00075229, 0.00070445,\n",
+       "        0.00054232, 0.00060431, 0.00056895, 0.00057467, 0.00071907,\n",
+       "        0.00052802, 0.00067226, 0.00049861, 0.00055369, 0.00047938,\n",
+       "        0.00052373, 0.00042502, 0.0004456 , 0.00043003, 0.000434  ,\n",
+       "        0.00049686, 0.00040666, 0.00044163, 0.00046102, 0.00042256,\n",
+       "        0.00050934, 0.00044338, 0.0004344 , 0.0005494 , 0.00045164,\n",
+       "        0.00046341, 0.00042502, 0.00040793, 0.00046802, 0.000422  ,\n",
+       "        0.00047398, 0.00041167, 0.00050402, 0.0004797 , 0.00046301,\n",
+       "        0.00058166, 0.00041199, 0.00054963, 0.00051061, 0.00041501,\n",
+       "        0.00055997, 0.00041167, 0.0004197 , 0.00070461, 0.00043456,\n",
+       "        0.00042661, 0.00058071, 0.0004646 , 0.00046428, 0.00041334,\n",
+       "        0.00042065, 0.00049806, 0.00059668, 0.00048574, 0.00052651,\n",
+       "        0.00051665, 0.00043098, 0.00049472, 0.00042439, 0.00051196,\n",
+       "        0.0004073 , 0.00044004, 0.00043805, 0.00051602, 0.0004437 ,\n",
+       "        0.00043162, 0.00044926, 0.00042311, 0.00043766, 0.00041676,\n",
+       "        0.00045037, 0.00041699, 0.0004104 , 0.00042613, 0.00050672,\n",
+       "        0.00042836, 0.00044163, 0.00046722, 0.00040634, 0.00057499]),\n",
+       " 'mean_test_score': array([-0.45823751, -0.45823509, -0.45823267, -0.45823025, -0.45822783,\n",
+       "        -0.45822541, -0.45822299, -0.45822057, -0.45821815, -0.45821573,\n",
+       "        -0.45821331, -0.45821089, -0.45820847, -0.45820605, -0.45820363,\n",
+       "        -0.45820121, -0.45819879, -0.45819637, -0.45819395, -0.45819153,\n",
+       "        -0.45818911, -0.45818669, -0.45818427, -0.45818185, -0.45817943,\n",
+       "        -0.45817702, -0.4581746 , -0.45817218, -0.45816976, -0.45816734,\n",
+       "        -0.45816492, -0.4581625 , -0.45816009, -0.45815767, -0.45815525,\n",
+       "        -0.45815283, -0.45815041, -0.458148  , -0.45814558, -0.45814316,\n",
+       "        -0.45814074, -0.45813832, -0.45813591, -0.45813349, -0.45813107,\n",
+       "        -0.45812865, -0.45812624, -0.45812382, -0.4581214 , -0.45811899,\n",
+       "        -0.45811657, -0.45811415, -0.45811174, -0.45810932, -0.4581069 ,\n",
+       "        -0.45810449, -0.45810207, -0.45809965, -0.45809724, -0.45809482,\n",
+       "        -0.4580924 , -0.45808999, -0.45808757, -0.45808516, -0.45808274,\n",
+       "        -0.45808033, -0.45807791, -0.45807549, -0.45807308, -0.45807066,\n",
+       "        -0.45806825, -0.45806583, -0.45806342, -0.458061  , -0.45805859,\n",
+       "        -0.45805617, -0.45805376, -0.45805134, -0.45804893, -0.45804651,\n",
+       "        -0.4580441 , -0.45804168, -0.45803927, -0.45803685, -0.45803444,\n",
+       "        -0.45803203, -0.45802961, -0.4580272 , -0.45802478, -0.45802237,\n",
+       "        -0.45801996, -0.45801754, -0.45801513, -0.45801271, -0.4580103 ,\n",
+       "        -0.45800789, -0.45800547, -0.45800306, -0.45800065, -0.45799823,\n",
+       "        -0.45799582, -0.45799341, -0.457991  , -0.45798858, -0.45798617,\n",
+       "        -0.45798376, -0.45798134, -0.45797893, -0.45797652, -0.45797411,\n",
+       "        -0.45797169, -0.45796928, -0.45796687, -0.45796446, -0.45796205,\n",
+       "        -0.45795963, -0.45795722, -0.45795481, -0.4579524 , -0.45794999,\n",
+       "        -0.45794758, -0.45794516, -0.45794275, -0.45794034, -0.45793793,\n",
+       "        -0.45793552, -0.45793311, -0.4579307 , -0.45792829, -0.45792587,\n",
+       "        -0.45792346, -0.45792105, -0.45791864, -0.45791623, -0.45791382,\n",
+       "        -0.45791141, -0.457909  , -0.45790659, -0.45790418, -0.45790177,\n",
+       "        -0.45789936, -0.45789695, -0.45789454, -0.45789213, -0.45788972,\n",
+       "        -0.45788731, -0.4578849 , -0.45788249, -0.45788008, -0.45787767,\n",
+       "        -0.45787526, -0.45787285, -0.45787044, -0.45786804, -0.45786563,\n",
+       "        -0.45786322, -0.45786081, -0.4578584 , -0.45785599, -0.45785358,\n",
+       "        -0.45785117, -0.45784877, -0.45784636, -0.45784395, -0.45784154,\n",
+       "        -0.45783913, -0.45783672, -0.45783432, -0.45783191, -0.4578295 ,\n",
+       "        -0.45782709, -0.45782469, -0.45782228, -0.45781987, -0.45781746,\n",
+       "        -0.45781505, -0.45781265, -0.45781024, -0.45780783, -0.45780543,\n",
+       "        -0.45780302, -0.45780061, -0.45779821, -0.4577958 , -0.45779339,\n",
+       "        -0.45779098, -0.45778858, -0.45778617, -0.45778377, -0.45778136,\n",
+       "        -0.45777895, -0.45777655, -0.45777414, -0.45777173, -0.45776933,\n",
+       "        -0.45776692, -0.45776452, -0.45776211, -0.45775971, -0.4577573 ,\n",
+       "        -0.45775489, -0.45775249, -0.45775008, -0.45774768, -0.45774527,\n",
+       "        -0.45774287, -0.45774046, -0.45773806, -0.45773565, -0.45773325,\n",
+       "        -0.45773084, -0.45772844, -0.45772603, -0.45772363, -0.45772122,\n",
+       "        -0.45771882, -0.45771642, -0.45771401, -0.45771161, -0.4577092 ,\n",
+       "        -0.4577068 , -0.4577044 , -0.45770199, -0.45769959, -0.45769718,\n",
+       "        -0.45769478, -0.45769238, -0.45768997, -0.45768757, -0.45768517,\n",
+       "        -0.45768276, -0.45768036, -0.45767796, -0.45767555, -0.45767315,\n",
+       "        -0.45767075, -0.45766835, -0.45766594, -0.45766354, -0.45766114,\n",
+       "        -0.45765874, -0.45765633, -0.45765393, -0.45765153, -0.45764913,\n",
+       "        -0.45764672, -0.45764432, -0.45764192, -0.45763952, -0.45763712,\n",
+       "        -0.45763471, -0.45763231, -0.45762991, -0.45762751, -0.45762511,\n",
+       "        -0.45762271, -0.45762031, -0.4576179 , -0.4576155 , -0.4576131 ,\n",
+       "        -0.4576107 , -0.4576083 , -0.4576059 , -0.4576035 , -0.4576011 ,\n",
+       "        -0.4575987 , -0.4575963 , -0.4575939 , -0.4575915 , -0.4575891 ,\n",
+       "        -0.4575867 , -0.4575843 , -0.4575819 , -0.4575795 , -0.4575771 ,\n",
+       "        -0.4575747 , -0.4575723 , -0.4575699 , -0.4575675 , -0.4575651 ,\n",
+       "        -0.4575627 , -0.4575603 , -0.4575579 , -0.4575555 , -0.4575531 ,\n",
+       "        -0.4575507 , -0.4575483 , -0.4575459 , -0.4575435 , -0.45754111,\n",
+       "        -0.45753871, -0.45753631, -0.45753391, -0.45753151, -0.45752911,\n",
+       "        -0.45752671, -0.45752432, -0.45752192, -0.45751952, -0.45751712,\n",
+       "        -0.45751472, -0.45751233, -0.45750993, -0.45750753, -0.45750513,\n",
+       "        -0.45750274, -0.45750034, -0.45749794, -0.45749554, -0.45749315,\n",
+       "        -0.45749075, -0.45748835, -0.45748595, -0.45748356, -0.45748116,\n",
+       "        -0.45747876, -0.45747637, -0.45747397, -0.45747157, -0.45746918,\n",
+       "        -0.45746678, -0.45746438, -0.45746199, -0.45745959, -0.4574572 ,\n",
+       "        -0.4574548 , -0.4574524 , -0.45745001, -0.45744761, -0.45744522,\n",
+       "        -0.45744282, -0.45744042, -0.45743803, -0.45743563, -0.45743324,\n",
+       "        -0.45743084, -0.45742845, -0.45742605, -0.45742366, -0.45742126,\n",
+       "        -0.45741887, -0.45741647, -0.45741408, -0.45741168, -0.45740929,\n",
+       "        -0.45740689, -0.4574045 , -0.4574021 , -0.45739971, -0.45739732,\n",
+       "        -0.45739492, -0.45739253, -0.45739013, -0.45738774, -0.45738535,\n",
+       "        -0.45738295, -0.45738056, -0.45737816, -0.45737577, -0.45737338,\n",
+       "        -0.45737098, -0.45736859, -0.4573662 , -0.4573638 , -0.45736141,\n",
+       "        -0.45735902, -0.45735662, -0.45735423, -0.45735184, -0.45734945,\n",
+       "        -0.45734705, -0.45734466, -0.45734227, -0.45733988, -0.45733748,\n",
+       "        -0.45733509, -0.4573327 , -0.45733031, -0.45732791, -0.45732552,\n",
+       "        -0.45732313, -0.45732074, -0.45731835, -0.45731595, -0.45731356,\n",
+       "        -0.45731117, -0.45730878, -0.45730639, -0.457304  , -0.45730161,\n",
+       "        -0.45729921, -0.45729682, -0.45729443, -0.45729204, -0.45728965,\n",
+       "        -0.45728726, -0.45728487, -0.45728248, -0.45728009, -0.4572777 ,\n",
+       "        -0.45727531, -0.45727292, -0.45727053, -0.45726814, -0.45726575,\n",
+       "        -0.45726336, -0.45726097, -0.45725858, -0.45725619, -0.4572538 ,\n",
+       "        -0.45725141, -0.45724902, -0.45724663, -0.45724424, -0.45724185,\n",
+       "        -0.45723946, -0.45723707, -0.45723468, -0.45723229, -0.4572299 ,\n",
+       "        -0.45722751, -0.45722512, -0.45722274, -0.45722035, -0.45721796,\n",
+       "        -0.45721557, -0.45721318, -0.45721079, -0.4572084 , -0.45720602,\n",
+       "        -0.45720363, -0.45720124, -0.45719885, -0.45719646, -0.45719408,\n",
+       "        -0.45719169, -0.4571893 , -0.45718691, -0.45718452, -0.45718214,\n",
+       "        -0.45717975, -0.45717736, -0.45717498, -0.45717259, -0.4571702 ,\n",
+       "        -0.45716781, -0.45716543, -0.45716304, -0.45716065, -0.45715827,\n",
+       "        -0.45715588, -0.45715349, -0.45715111, -0.45714872, -0.45714633,\n",
+       "        -0.45714395, -0.45714156, -0.45713917, -0.45713679, -0.4571344 ,\n",
+       "        -0.45713202, -0.45712963, -0.45712724, -0.45712486, -0.45712247,\n",
+       "        -0.45712009, -0.4571177 , -0.45711532, -0.45711293, -0.45711055,\n",
+       "        -0.45710816, -0.45710578, -0.45710339, -0.45710101, -0.45709862,\n",
+       "        -0.45709624, -0.45709385, -0.45709147, -0.45708908, -0.4570867 ,\n",
+       "        -0.45708431, -0.45708193, -0.45707954, -0.45707716, -0.45707478,\n",
+       "        -0.45707239, -0.45707001, -0.45706762, -0.45706524, -0.45706286,\n",
+       "        -0.45706047, -0.45705809, -0.45705571, -0.45705332, -0.45705094,\n",
+       "        -0.45704856, -0.45704617, -0.45704379, -0.45704141, -0.45703902]),\n",
+       " 'mean_train_score': array([0.20161897, 0.20161897, 0.20161897, 0.20161897, 0.20161897,\n",
+       "        0.20161897, 0.20161897, 0.20161897, 0.20161897, 0.20161897,\n",
+       "        0.20161897, 0.20161897, 0.20161897, 0.20161897, 0.20161897,\n",
+       "        0.20161897, 0.20161897, 0.20161897, 0.20161897, 0.20161897,\n",
+       "        0.20161897, 0.20161897, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161896, 0.20161896,\n",
+       "        0.20161896, 0.20161896, 0.20161896, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161895, 0.20161895, 0.20161895,\n",
+       "        0.20161895, 0.20161895, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161894, 0.20161894, 0.20161894,\n",
+       "        0.20161894, 0.20161894, 0.20161893, 0.20161893, 0.20161893,\n",
+       "        0.20161893, 0.20161893, 0.20161893, 0.20161893, 0.20161893,\n",
+       "        0.20161893, 0.20161893, 0.20161893, 0.20161893, 0.20161893,\n",
+       "        0.20161893, 0.20161893, 0.20161893, 0.20161893, 0.20161893,\n",
+       "        0.20161893, 0.20161893, 0.20161893, 0.20161893, 0.20161893,\n",
+       "        0.20161893, 0.20161893, 0.20161892, 0.20161892, 0.20161892,\n",
+       "        0.20161892, 0.20161892, 0.20161892, 0.20161892, 0.20161892,\n",
+       "        0.20161892, 0.20161892, 0.20161892, 0.20161892, 0.20161892,\n",
+       "        0.20161892, 0.20161892, 0.20161892, 0.20161892, 0.20161892,\n",
+       "        0.20161892, 0.20161892, 0.20161892, 0.20161892, 0.20161891,\n",
+       "        0.20161891, 0.20161891, 0.20161891, 0.20161891, 0.20161891,\n",
+       "        0.20161891, 0.20161891, 0.20161891, 0.20161891, 0.20161891,\n",
+       "        0.20161891, 0.20161891, 0.20161891, 0.20161891, 0.20161891,\n",
+       "        0.20161891, 0.20161891, 0.20161891, 0.20161891, 0.20161891,\n",
+       "        0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 ,\n",
+       "        0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 ,\n",
+       "        0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 , 0.2016189 ,\n",
+       "        0.2016189 , 0.2016189 , 0.2016189 , 0.20161889, 0.20161889,\n",
+       "        0.20161889, 0.20161889, 0.20161889, 0.20161889, 0.20161889,\n",
+       "        0.20161889, 0.20161889, 0.20161889, 0.20161889, 0.20161889,\n",
+       "        0.20161889, 0.20161889, 0.20161889, 0.20161889, 0.20161889,\n",
+       "        0.20161889, 0.20161888, 0.20161888, 0.20161888, 0.20161888,\n",
+       "        0.20161888, 0.20161888, 0.20161888, 0.20161888, 0.20161888,\n",
+       "        0.20161888, 0.20161888, 0.20161888, 0.20161888, 0.20161888,\n",
+       "        0.20161888, 0.20161888, 0.20161887, 0.20161887, 0.20161887,\n",
+       "        0.20161887, 0.20161887, 0.20161887, 0.20161887, 0.20161887,\n",
+       "        0.20161887, 0.20161887, 0.20161887, 0.20161887, 0.20161887,\n",
+       "        0.20161887, 0.20161887, 0.20161886, 0.20161886, 0.20161886,\n",
+       "        0.20161886, 0.20161886, 0.20161886, 0.20161886, 0.20161886,\n",
+       "        0.20161886, 0.20161886, 0.20161886, 0.20161886, 0.20161886,\n",
+       "        0.20161886, 0.20161886, 0.20161885, 0.20161885, 0.20161885,\n",
+       "        0.20161885, 0.20161885, 0.20161885, 0.20161885, 0.20161885,\n",
+       "        0.20161885, 0.20161885, 0.20161885, 0.20161885, 0.20161885,\n",
+       "        0.20161885, 0.20161884, 0.20161884, 0.20161884, 0.20161884,\n",
+       "        0.20161884, 0.20161884, 0.20161884, 0.20161884, 0.20161884,\n",
+       "        0.20161884, 0.20161884, 0.20161884, 0.20161884, 0.20161884,\n",
+       "        0.20161883, 0.20161883, 0.20161883, 0.20161883, 0.20161883,\n",
+       "        0.20161883, 0.20161883, 0.20161883, 0.20161883, 0.20161883,\n",
+       "        0.20161883, 0.20161883, 0.20161883, 0.20161882, 0.20161882,\n",
+       "        0.20161882, 0.20161882, 0.20161882, 0.20161882, 0.20161882,\n",
+       "        0.20161882, 0.20161882, 0.20161882, 0.20161882, 0.20161882,\n",
+       "        0.20161881, 0.20161881, 0.20161881, 0.20161881, 0.20161881,\n",
+       "        0.20161881, 0.20161881, 0.20161881, 0.20161881, 0.20161881,\n",
+       "        0.20161881, 0.20161881, 0.2016188 , 0.2016188 , 0.2016188 ,\n",
+       "        0.2016188 , 0.2016188 , 0.2016188 , 0.2016188 , 0.2016188 ,\n",
+       "        0.2016188 , 0.2016188 , 0.2016188 , 0.2016188 , 0.20161879,\n",
+       "        0.20161879, 0.20161879, 0.20161879, 0.20161879, 0.20161879,\n",
+       "        0.20161879, 0.20161879, 0.20161879, 0.20161879, 0.20161879,\n",
+       "        0.20161878, 0.20161878, 0.20161878, 0.20161878, 0.20161878,\n",
+       "        0.20161878, 0.20161878, 0.20161878, 0.20161878, 0.20161878,\n",
+       "        0.20161878, 0.20161878, 0.20161877, 0.20161877, 0.20161877,\n",
+       "        0.20161877, 0.20161877, 0.20161877, 0.20161877, 0.20161877,\n",
+       "        0.20161877, 0.20161877, 0.20161876, 0.20161876, 0.20161876,\n",
+       "        0.20161876, 0.20161876, 0.20161876, 0.20161876, 0.20161876,\n",
+       "        0.20161876, 0.20161876, 0.20161876, 0.20161875, 0.20161875,\n",
+       "        0.20161875, 0.20161875, 0.20161875, 0.20161875, 0.20161875,\n",
+       "        0.20161875, 0.20161875, 0.20161875, 0.20161874, 0.20161874,\n",
+       "        0.20161874, 0.20161874, 0.20161874, 0.20161874, 0.20161874,\n",
+       "        0.20161874, 0.20161874, 0.20161874, 0.20161873, 0.20161873,\n",
+       "        0.20161873, 0.20161873, 0.20161873, 0.20161873, 0.20161873,\n",
+       "        0.20161873, 0.20161873, 0.20161873, 0.20161872, 0.20161872,\n",
+       "        0.20161872, 0.20161872, 0.20161872, 0.20161872, 0.20161872,\n",
+       "        0.20161872, 0.20161872, 0.20161872, 0.20161871, 0.20161871,\n",
+       "        0.20161871, 0.20161871, 0.20161871, 0.20161871, 0.20161871,\n",
+       "        0.20161871, 0.20161871, 0.2016187 , 0.2016187 , 0.2016187 ,\n",
+       "        0.2016187 , 0.2016187 , 0.2016187 , 0.2016187 , 0.2016187 ,\n",
+       "        0.2016187 , 0.2016187 , 0.20161869, 0.20161869, 0.20161869]),\n",
+       " 'param_alpha': masked_array(data=[0.0, 0.002004008016032064, 0.004008016032064128,\n",
+       "                    0.0060120240480961915, 0.008016032064128256,\n",
+       "                    0.01002004008016032, 0.012024048096192383,\n",
+       "                    0.014028056112224447, 0.01603206412825651,\n",
+       "                    0.018036072144288574, 0.02004008016032064,\n",
+       "                    0.022044088176352703, 0.024048096192384766,\n",
+       "                    0.026052104208416832, 0.028056112224448895,\n",
+       "                    0.03006012024048096, 0.03206412825651302,\n",
+       "                    0.03406813627254509, 0.03607214428857715,\n",
+       "                    0.038076152304609215, 0.04008016032064128,\n",
+       "                    0.04208416833667334, 0.044088176352705406,\n",
+       "                    0.04609218436873747, 0.04809619238476953,\n",
+       "                    0.0501002004008016, 0.052104208416833664,\n",
+       "                    0.05410821643286573, 0.05611222444889779,\n",
+       "                    0.058116232464929855, 0.06012024048096192,\n",
+       "                    0.06212424849699398, 0.06412825651302605,\n",
+       "                    0.06613226452905811, 0.06813627254509018,\n",
+       "                    0.07014028056112225, 0.0721442885771543,\n",
+       "                    0.07414829659318636, 0.07615230460921843,\n",
+       "                    0.0781563126252505, 0.08016032064128256,\n",
+       "                    0.08216432865731463, 0.08416833667334668,\n",
+       "                    0.08617234468937875, 0.08817635270541081,\n",
+       "                    0.09018036072144288, 0.09218436873747494,\n",
+       "                    0.09418837675350701, 0.09619238476953906,\n",
+       "                    0.09819639278557113, 0.1002004008016032,\n",
+       "                    0.10220440881763526, 0.10420841683366733,\n",
+       "                    0.1062124248496994, 0.10821643286573146,\n",
+       "                    0.11022044088176351, 0.11222444889779558,\n",
+       "                    0.11422845691382764, 0.11623246492985971,\n",
+       "                    0.11823647294589178, 0.12024048096192384,\n",
+       "                    0.1222444889779559, 0.12424849699398796,\n",
+       "                    0.12625250501002003, 0.1282565130260521,\n",
+       "                    0.13026052104208416, 0.13226452905811623,\n",
+       "                    0.1342685370741483, 0.13627254509018036,\n",
+       "                    0.13827655310621242, 0.1402805611222445,\n",
+       "                    0.14228456913827653, 0.1442885771543086,\n",
+       "                    0.14629258517034066, 0.14829659318637273,\n",
+       "                    0.1503006012024048, 0.15230460921843686,\n",
+       "                    0.15430861723446893, 0.156312625250501,\n",
+       "                    0.15831663326653306, 0.16032064128256512,\n",
+       "                    0.1623246492985972, 0.16432865731462926,\n",
+       "                    0.16633266533066132, 0.16833667334669336,\n",
+       "                    0.17034068136272543, 0.1723446893787575,\n",
+       "                    0.17434869739478956, 0.17635270541082163,\n",
+       "                    0.1783567134268537, 0.18036072144288576,\n",
+       "                    0.18236472945891782, 0.1843687374749499,\n",
+       "                    0.18637274549098196, 0.18837675350701402,\n",
+       "                    0.1903807615230461, 0.19238476953907813,\n",
+       "                    0.1943887775551102, 0.19639278557114226,\n",
+       "                    0.19839679358717432, 0.2004008016032064,\n",
+       "                    0.20240480961923846, 0.20440881763527052,\n",
+       "                    0.2064128256513026, 0.20841683366733466,\n",
+       "                    0.21042084168336672, 0.2124248496993988,\n",
+       "                    0.21442885771543085, 0.21643286573146292,\n",
+       "                    0.21843687374749496, 0.22044088176352702,\n",
+       "                    0.2224448897795591, 0.22444889779559116,\n",
+       "                    0.22645290581162322, 0.2284569138276553,\n",
+       "                    0.23046092184368736, 0.23246492985971942,\n",
+       "                    0.2344689378757515, 0.23647294589178355,\n",
+       "                    0.23847695390781562, 0.24048096192384769,\n",
+       "                    0.24248496993987975, 0.2444889779559118,\n",
+       "                    0.24649298597194386, 0.24849699398797592,\n",
+       "                    0.250501002004008, 0.25250501002004005,\n",
+       "                    0.2545090180360721, 0.2565130260521042,\n",
+       "                    0.25851703406813625, 0.2605210420841683,\n",
+       "                    0.2625250501002004, 0.26452905811623245,\n",
+       "                    0.2665330661322645, 0.2685370741482966,\n",
+       "                    0.27054108216432865, 0.2725450901803607,\n",
+       "                    0.2745490981963928, 0.27655310621242485,\n",
+       "                    0.2785571142284569, 0.280561122244489,\n",
+       "                    0.282565130260521, 0.28456913827655306,\n",
+       "                    0.2865731462925851, 0.2885771543086172,\n",
+       "                    0.29058116232464926, 0.2925851703406813,\n",
+       "                    0.2945891783567134, 0.29659318637274545,\n",
+       "                    0.2985971943887775, 0.3006012024048096,\n",
+       "                    0.30260521042084165, 0.3046092184368737,\n",
+       "                    0.3066132264529058, 0.30861723446893785,\n",
+       "                    0.3106212424849699, 0.312625250501002,\n",
+       "                    0.31462925851703405, 0.3166332665330661,\n",
+       "                    0.3186372745490982, 0.32064128256513025,\n",
+       "                    0.3226452905811623, 0.3246492985971944,\n",
+       "                    0.32665330661322645, 0.3286573146292585,\n",
+       "                    0.3306613226452906, 0.33266533066132264,\n",
+       "                    0.33466933867735466, 0.3366733466933867,\n",
+       "                    0.3386773547094188, 0.34068136272545085,\n",
+       "                    0.3426853707414829, 0.344689378757515,\n",
+       "                    0.34669338677354705, 0.3486973947895791,\n",
+       "                    0.3507014028056112, 0.35270541082164325,\n",
+       "                    0.3547094188376753, 0.3567134268537074,\n",
+       "                    0.35871743486973945, 0.3607214428857715,\n",
+       "                    0.3627254509018036, 0.36472945891783565,\n",
+       "                    0.3667334669338677, 0.3687374749498998,\n",
+       "                    0.37074148296593185, 0.3727454909819639,\n",
+       "                    0.374749498997996, 0.37675350701402804,\n",
+       "                    0.3787575150300601, 0.3807615230460922,\n",
+       "                    0.38276553106212424, 0.38476953907815625,\n",
+       "                    0.3867735470941883, 0.3887775551102204,\n",
+       "                    0.39078156312625245, 0.3927855711422845,\n",
+       "                    0.3947895791583166, 0.39679358717434865,\n",
+       "                    0.3987975951903807, 0.4008016032064128,\n",
+       "                    0.40280561122244485, 0.4048096192384769,\n",
+       "                    0.406813627254509, 0.40881763527054105,\n",
+       "                    0.4108216432865731, 0.4128256513026052,\n",
+       "                    0.41482965931863724, 0.4168336673346693,\n",
+       "                    0.4188376753507014, 0.42084168336673344,\n",
+       "                    0.4228456913827655, 0.4248496993987976,\n",
+       "                    0.42685370741482964, 0.4288577154308617,\n",
+       "                    0.4308617234468938, 0.43286573146292584,\n",
+       "                    0.4348697394789579, 0.4368737474949899,\n",
+       "                    0.438877755511022, 0.44088176352705405,\n",
+       "                    0.4428857715430861, 0.4448897795591182,\n",
+       "                    0.44689378757515025, 0.4488977955911823,\n",
+       "                    0.4509018036072144, 0.45290581162324645,\n",
+       "                    0.4549098196392785, 0.4569138276553106,\n",
+       "                    0.45891783567134264, 0.4609218436873747,\n",
+       "                    0.4629258517034068, 0.46492985971943884,\n",
+       "                    0.4669338677354709, 0.468937875751503,\n",
+       "                    0.47094188376753504, 0.4729458917835671,\n",
+       "                    0.4749498997995992, 0.47695390781563124,\n",
+       "                    0.4789579158316633, 0.48096192384769537,\n",
+       "                    0.48296593186372744, 0.4849699398797595,\n",
+       "                    0.48697394789579157, 0.4889779559118236,\n",
+       "                    0.49098196392785565, 0.4929859719438877,\n",
+       "                    0.4949899799599198, 0.49699398797595185,\n",
+       "                    0.4989979959919839, 0.501002004008016,\n",
+       "                    0.503006012024048, 0.5050100200400801,\n",
+       "                    0.5070140280561122, 0.5090180360721442,\n",
+       "                    0.5110220440881763, 0.5130260521042084,\n",
+       "                    0.5150300601202404, 0.5170340681362725,\n",
+       "                    0.5190380761523046, 0.5210420841683366,\n",
+       "                    0.5230460921843687, 0.5250501002004008,\n",
+       "                    0.5270541082164328, 0.5290581162324649,\n",
+       "                    0.531062124248497, 0.533066132264529,\n",
+       "                    0.5350701402805611, 0.5370741482965932,\n",
+       "                    0.5390781563126252, 0.5410821643286573,\n",
+       "                    0.5430861723446894, 0.5450901803607214,\n",
+       "                    0.5470941883767535, 0.5490981963927856,\n",
+       "                    0.5511022044088176, 0.5531062124248497,\n",
+       "                    0.5551102204408818, 0.5571142284569138,\n",
+       "                    0.5591182364729459, 0.561122244488978,\n",
+       "                    0.56312625250501, 0.565130260521042, 0.567134268537074,\n",
+       "                    0.5691382765531061, 0.5711422845691382,\n",
+       "                    0.5731462925851702, 0.5751503006012023,\n",
+       "                    0.5771543086172344, 0.5791583166332664,\n",
+       "                    0.5811623246492985, 0.5831663326653306,\n",
+       "                    0.5851703406813626, 0.5871743486973947,\n",
+       "                    0.5891783567134268, 0.5911823647294588,\n",
+       "                    0.5931863727454909, 0.595190380761523,\n",
+       "                    0.597194388777555, 0.5991983967935871,\n",
+       "                    0.6012024048096192, 0.6032064128256512,\n",
+       "                    0.6052104208416833, 0.6072144288577154,\n",
+       "                    0.6092184368737474, 0.6112224448897795,\n",
+       "                    0.6132264529058116, 0.6152304609218436,\n",
+       "                    0.6172344689378757, 0.6192384769539078,\n",
+       "                    0.6212424849699398, 0.6232464929859719,\n",
+       "                    0.625250501002004, 0.627254509018036,\n",
+       "                    0.6292585170340681, 0.6312625250501002,\n",
+       "                    0.6332665330661322, 0.6352705410821643,\n",
+       "                    0.6372745490981964, 0.6392785571142284,\n",
+       "                    0.6412825651302605, 0.6432865731462926,\n",
+       "                    0.6452905811623246, 0.6472945891783567,\n",
+       "                    0.6492985971943888, 0.6513026052104208,\n",
+       "                    0.6533066132264529, 0.655310621242485,\n",
+       "                    0.657314629258517, 0.6593186372745491,\n",
+       "                    0.6613226452905812, 0.6633266533066132,\n",
+       "                    0.6653306613226453, 0.6673346693386772,\n",
+       "                    0.6693386773547093, 0.6713426853707414,\n",
+       "                    0.6733466933867734, 0.6753507014028055,\n",
+       "                    0.6773547094188376, 0.6793587174348696,\n",
+       "                    0.6813627254509017, 0.6833667334669338,\n",
+       "                    0.6853707414829658, 0.6873747494989979,\n",
+       "                    0.68937875751503, 0.691382765531062,\n",
+       "                    0.6933867735470941, 0.6953907815631262,\n",
+       "                    0.6973947895791582, 0.6993987975951903,\n",
+       "                    0.7014028056112224, 0.7034068136272544,\n",
+       "                    0.7054108216432865, 0.7074148296593186,\n",
+       "                    0.7094188376753506, 0.7114228456913827,\n",
+       "                    0.7134268537074148, 0.7154308617234468,\n",
+       "                    0.7174348697394789, 0.719438877755511,\n",
+       "                    0.721442885771543, 0.7234468937875751,\n",
+       "                    0.7254509018036072, 0.7274549098196392,\n",
+       "                    0.7294589178356713, 0.7314629258517034,\n",
+       "                    0.7334669338677354, 0.7354709418837675,\n",
+       "                    0.7374749498997996, 0.7394789579158316,\n",
+       "                    0.7414829659318637, 0.7434869739478958,\n",
+       "                    0.7454909819639278, 0.7474949899799599,\n",
+       "                    0.749498997995992, 0.751503006012024,\n",
+       "                    0.7535070140280561, 0.7555110220440882,\n",
+       "                    0.7575150300601202, 0.7595190380761523,\n",
+       "                    0.7615230460921844, 0.7635270541082164,\n",
+       "                    0.7655310621242485, 0.7675350701402806,\n",
+       "                    0.7695390781563125, 0.7715430861723446,\n",
+       "                    0.7735470941883766, 0.7755511022044087,\n",
+       "                    0.7775551102204408, 0.7795591182364728,\n",
+       "                    0.7815631262525049, 0.783567134268537,\n",
+       "                    0.785571142284569, 0.7875751503006011,\n",
+       "                    0.7895791583166332, 0.7915831663326652,\n",
+       "                    0.7935871743486973, 0.7955911823647294,\n",
+       "                    0.7975951903807614, 0.7995991983967935,\n",
+       "                    0.8016032064128256, 0.8036072144288576,\n",
+       "                    0.8056112224448897, 0.8076152304609218,\n",
+       "                    0.8096192384769538, 0.8116232464929859,\n",
+       "                    0.813627254509018, 0.81563126252505,\n",
+       "                    0.8176352705410821, 0.8196392785571142,\n",
+       "                    0.8216432865731462, 0.8236472945891783,\n",
+       "                    0.8256513026052104, 0.8276553106212424,\n",
+       "                    0.8296593186372745, 0.8316633266533066,\n",
+       "                    0.8336673346693386, 0.8356713426853707,\n",
+       "                    0.8376753507014028, 0.8396793587174348,\n",
+       "                    0.8416833667334669, 0.843687374749499,\n",
+       "                    0.845691382765531, 0.8476953907815631,\n",
+       "                    0.8496993987975952, 0.8517034068136272,\n",
+       "                    0.8537074148296593, 0.8557114228456913,\n",
+       "                    0.8577154308617234, 0.8597194388777555,\n",
+       "                    0.8617234468937875, 0.8637274549098196,\n",
+       "                    0.8657314629258517, 0.8677354709418837,\n",
+       "                    0.8697394789579158, 0.8717434869739479,\n",
+       "                    0.8737474949899798, 0.8757515030060119,\n",
+       "                    0.877755511022044, 0.879759519038076,\n",
+       "                    0.8817635270541081, 0.8837675350701402,\n",
+       "                    0.8857715430861722, 0.8877755511022043,\n",
+       "                    0.8897795591182364, 0.8917835671342684,\n",
+       "                    0.8937875751503005, 0.8957915831663326,\n",
+       "                    0.8977955911823646, 0.8997995991983967,\n",
+       "                    0.9018036072144288, 0.9038076152304608,\n",
+       "                    0.9058116232464929, 0.907815631262525,\n",
+       "                    0.909819639278557, 0.9118236472945891,\n",
+       "                    0.9138276553106212, 0.9158316633266532,\n",
+       "                    0.9178356713426853, 0.9198396793587174,\n",
+       "                    0.9218436873747494, 0.9238476953907815,\n",
+       "                    0.9258517034068136, 0.9278557114228456,\n",
+       "                    0.9298597194388777, 0.9318637274549098,\n",
+       "                    0.9338677354709418, 0.9358717434869739,\n",
+       "                    0.937875751503006, 0.939879759519038,\n",
+       "                    0.9418837675350701, 0.9438877755511021,\n",
+       "                    0.9458917835671342, 0.9478957915831663,\n",
+       "                    0.9498997995991983, 0.9519038076152304,\n",
+       "                    0.9539078156312625, 0.9559118236472945,\n",
+       "                    0.9579158316633266, 0.9599198396793587,\n",
+       "                    0.9619238476953907, 0.9639278557114228,\n",
+       "                    0.9659318637274549, 0.9679358717434869,\n",
+       "                    0.969939879759519, 0.9719438877755511,\n",
+       "                    0.9739478957915831, 0.9759519038076151,\n",
+       "                    0.9779559118236472, 0.9799599198396792,\n",
+       "                    0.9819639278557113, 0.9839679358717434,\n",
+       "                    0.9859719438877754, 0.9879759519038075,\n",
+       "                    0.9899799599198396, 0.9919839679358716,\n",
+       "                    0.9939879759519037, 0.9959919839679358,\n",
+       "                    0.9979959919839678, 1.0],\n",
+       "              mask=[False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False, False, False, False, False,\n",
+       "                    False, False, False, False],\n",
+       "        fill_value='?',\n",
+       "             dtype=object),\n",
+       " 'params': [{'alpha': 0.0},\n",
+       "  {'alpha': 0.002004008016032064},\n",
+       "  {'alpha': 0.004008016032064128},\n",
+       "  {'alpha': 0.0060120240480961915},\n",
+       "  {'alpha': 0.008016032064128256},\n",
+       "  {'alpha': 0.01002004008016032},\n",
+       "  {'alpha': 0.012024048096192383},\n",
+       "  {'alpha': 0.014028056112224447},\n",
+       "  {'alpha': 0.01603206412825651},\n",
+       "  {'alpha': 0.018036072144288574},\n",
+       "  {'alpha': 0.02004008016032064},\n",
+       "  {'alpha': 0.022044088176352703},\n",
+       "  {'alpha': 0.024048096192384766},\n",
+       "  {'alpha': 0.026052104208416832},\n",
+       "  {'alpha': 0.028056112224448895},\n",
+       "  {'alpha': 0.03006012024048096},\n",
+       "  {'alpha': 0.03206412825651302},\n",
+       "  {'alpha': 0.03406813627254509},\n",
+       "  {'alpha': 0.03607214428857715},\n",
+       "  {'alpha': 0.038076152304609215},\n",
+       "  {'alpha': 0.04008016032064128},\n",
+       "  {'alpha': 0.04208416833667334},\n",
+       "  {'alpha': 0.044088176352705406},\n",
+       "  {'alpha': 0.04609218436873747},\n",
+       "  {'alpha': 0.04809619238476953},\n",
+       "  {'alpha': 0.0501002004008016},\n",
+       "  {'alpha': 0.052104208416833664},\n",
+       "  {'alpha': 0.05410821643286573},\n",
+       "  {'alpha': 0.05611222444889779},\n",
+       "  {'alpha': 0.058116232464929855},\n",
+       "  {'alpha': 0.06012024048096192},\n",
+       "  {'alpha': 0.06212424849699398},\n",
+       "  {'alpha': 0.06412825651302605},\n",
+       "  {'alpha': 0.06613226452905811},\n",
+       "  {'alpha': 0.06813627254509018},\n",
+       "  {'alpha': 0.07014028056112225},\n",
+       "  {'alpha': 0.0721442885771543},\n",
+       "  {'alpha': 0.07414829659318636},\n",
+       "  {'alpha': 0.07615230460921843},\n",
+       "  {'alpha': 0.0781563126252505},\n",
+       "  {'alpha': 0.08016032064128256},\n",
+       "  {'alpha': 0.08216432865731463},\n",
+       "  {'alpha': 0.08416833667334668},\n",
+       "  {'alpha': 0.08617234468937875},\n",
+       "  {'alpha': 0.08817635270541081},\n",
+       "  {'alpha': 0.09018036072144288},\n",
+       "  {'alpha': 0.09218436873747494},\n",
+       "  {'alpha': 0.09418837675350701},\n",
+       "  {'alpha': 0.09619238476953906},\n",
+       "  {'alpha': 0.09819639278557113},\n",
+       "  {'alpha': 0.1002004008016032},\n",
+       "  {'alpha': 0.10220440881763526},\n",
+       "  {'alpha': 0.10420841683366733},\n",
+       "  {'alpha': 0.1062124248496994},\n",
+       "  {'alpha': 0.10821643286573146},\n",
+       "  {'alpha': 0.11022044088176351},\n",
+       "  {'alpha': 0.11222444889779558},\n",
+       "  {'alpha': 0.11422845691382764},\n",
+       "  {'alpha': 0.11623246492985971},\n",
+       "  {'alpha': 0.11823647294589178},\n",
+       "  {'alpha': 0.12024048096192384},\n",
+       "  {'alpha': 0.1222444889779559},\n",
+       "  {'alpha': 0.12424849699398796},\n",
+       "  {'alpha': 0.12625250501002003},\n",
+       "  {'alpha': 0.1282565130260521},\n",
+       "  {'alpha': 0.13026052104208416},\n",
+       "  {'alpha': 0.13226452905811623},\n",
+       "  {'alpha': 0.1342685370741483},\n",
+       "  {'alpha': 0.13627254509018036},\n",
+       "  {'alpha': 0.13827655310621242},\n",
+       "  {'alpha': 0.1402805611222445},\n",
+       "  {'alpha': 0.14228456913827653},\n",
+       "  {'alpha': 0.1442885771543086},\n",
+       "  {'alpha': 0.14629258517034066},\n",
+       "  {'alpha': 0.14829659318637273},\n",
+       "  {'alpha': 0.1503006012024048},\n",
+       "  {'alpha': 0.15230460921843686},\n",
+       "  {'alpha': 0.15430861723446893},\n",
+       "  {'alpha': 0.156312625250501},\n",
+       "  {'alpha': 0.15831663326653306},\n",
+       "  {'alpha': 0.16032064128256512},\n",
+       "  {'alpha': 0.1623246492985972},\n",
+       "  {'alpha': 0.16432865731462926},\n",
+       "  {'alpha': 0.16633266533066132},\n",
+       "  {'alpha': 0.16833667334669336},\n",
+       "  {'alpha': 0.17034068136272543},\n",
+       "  {'alpha': 0.1723446893787575},\n",
+       "  {'alpha': 0.17434869739478956},\n",
+       "  {'alpha': 0.17635270541082163},\n",
+       "  {'alpha': 0.1783567134268537},\n",
+       "  {'alpha': 0.18036072144288576},\n",
+       "  {'alpha': 0.18236472945891782},\n",
+       "  {'alpha': 0.1843687374749499},\n",
+       "  {'alpha': 0.18637274549098196},\n",
+       "  {'alpha': 0.18837675350701402},\n",
+       "  {'alpha': 0.1903807615230461},\n",
+       "  {'alpha': 0.19238476953907813},\n",
+       "  {'alpha': 0.1943887775551102},\n",
+       "  {'alpha': 0.19639278557114226},\n",
+       "  {'alpha': 0.19839679358717432},\n",
+       "  {'alpha': 0.2004008016032064},\n",
+       "  {'alpha': 0.20240480961923846},\n",
+       "  {'alpha': 0.20440881763527052},\n",
+       "  {'alpha': 0.2064128256513026},\n",
+       "  {'alpha': 0.20841683366733466},\n",
+       "  {'alpha': 0.21042084168336672},\n",
+       "  {'alpha': 0.2124248496993988},\n",
+       "  {'alpha': 0.21442885771543085},\n",
+       "  {'alpha': 0.21643286573146292},\n",
+       "  {'alpha': 0.21843687374749496},\n",
+       "  {'alpha': 0.22044088176352702},\n",
+       "  {'alpha': 0.2224448897795591},\n",
+       "  {'alpha': 0.22444889779559116},\n",
+       "  {'alpha': 0.22645290581162322},\n",
+       "  {'alpha': 0.2284569138276553},\n",
+       "  {'alpha': 0.23046092184368736},\n",
+       "  {'alpha': 0.23246492985971942},\n",
+       "  {'alpha': 0.2344689378757515},\n",
+       "  {'alpha': 0.23647294589178355},\n",
+       "  {'alpha': 0.23847695390781562},\n",
+       "  {'alpha': 0.24048096192384769},\n",
+       "  {'alpha': 0.24248496993987975},\n",
+       "  {'alpha': 0.2444889779559118},\n",
+       "  {'alpha': 0.24649298597194386},\n",
+       "  {'alpha': 0.24849699398797592},\n",
+       "  {'alpha': 0.250501002004008},\n",
+       "  {'alpha': 0.25250501002004005},\n",
+       "  {'alpha': 0.2545090180360721},\n",
+       "  {'alpha': 0.2565130260521042},\n",
+       "  {'alpha': 0.25851703406813625},\n",
+       "  {'alpha': 0.2605210420841683},\n",
+       "  {'alpha': 0.2625250501002004},\n",
+       "  {'alpha': 0.26452905811623245},\n",
+       "  {'alpha': 0.2665330661322645},\n",
+       "  {'alpha': 0.2685370741482966},\n",
+       "  {'alpha': 0.27054108216432865},\n",
+       "  {'alpha': 0.2725450901803607},\n",
+       "  {'alpha': 0.2745490981963928},\n",
+       "  {'alpha': 0.27655310621242485},\n",
+       "  {'alpha': 0.2785571142284569},\n",
+       "  {'alpha': 0.280561122244489},\n",
+       "  {'alpha': 0.282565130260521},\n",
+       "  {'alpha': 0.28456913827655306},\n",
+       "  {'alpha': 0.2865731462925851},\n",
+       "  {'alpha': 0.2885771543086172},\n",
+       "  {'alpha': 0.29058116232464926},\n",
+       "  {'alpha': 0.2925851703406813},\n",
+       "  {'alpha': 0.2945891783567134},\n",
+       "  {'alpha': 0.29659318637274545},\n",
+       "  {'alpha': 0.2985971943887775},\n",
+       "  {'alpha': 0.3006012024048096},\n",
+       "  {'alpha': 0.30260521042084165},\n",
+       "  {'alpha': 0.3046092184368737},\n",
+       "  {'alpha': 0.3066132264529058},\n",
+       "  {'alpha': 0.30861723446893785},\n",
+       "  {'alpha': 0.3106212424849699},\n",
+       "  {'alpha': 0.312625250501002},\n",
+       "  {'alpha': 0.31462925851703405},\n",
+       "  {'alpha': 0.3166332665330661},\n",
+       "  {'alpha': 0.3186372745490982},\n",
+       "  {'alpha': 0.32064128256513025},\n",
+       "  {'alpha': 0.3226452905811623},\n",
+       "  {'alpha': 0.3246492985971944},\n",
+       "  {'alpha': 0.32665330661322645},\n",
+       "  {'alpha': 0.3286573146292585},\n",
+       "  {'alpha': 0.3306613226452906},\n",
+       "  {'alpha': 0.33266533066132264},\n",
+       "  {'alpha': 0.33466933867735466},\n",
+       "  {'alpha': 0.3366733466933867},\n",
+       "  {'alpha': 0.3386773547094188},\n",
+       "  {'alpha': 0.34068136272545085},\n",
+       "  {'alpha': 0.3426853707414829},\n",
+       "  {'alpha': 0.344689378757515},\n",
+       "  {'alpha': 0.34669338677354705},\n",
+       "  {'alpha': 0.3486973947895791},\n",
+       "  {'alpha': 0.3507014028056112},\n",
+       "  {'alpha': 0.35270541082164325},\n",
+       "  {'alpha': 0.3547094188376753},\n",
+       "  {'alpha': 0.3567134268537074},\n",
+       "  {'alpha': 0.35871743486973945},\n",
+       "  {'alpha': 0.3607214428857715},\n",
+       "  {'alpha': 0.3627254509018036},\n",
+       "  {'alpha': 0.36472945891783565},\n",
+       "  {'alpha': 0.3667334669338677},\n",
+       "  {'alpha': 0.3687374749498998},\n",
+       "  {'alpha': 0.37074148296593185},\n",
+       "  {'alpha': 0.3727454909819639},\n",
+       "  {'alpha': 0.374749498997996},\n",
+       "  {'alpha': 0.37675350701402804},\n",
+       "  {'alpha': 0.3787575150300601},\n",
+       "  {'alpha': 0.3807615230460922},\n",
+       "  {'alpha': 0.38276553106212424},\n",
+       "  {'alpha': 0.38476953907815625},\n",
+       "  {'alpha': 0.3867735470941883},\n",
+       "  {'alpha': 0.3887775551102204},\n",
+       "  {'alpha': 0.39078156312625245},\n",
+       "  {'alpha': 0.3927855711422845},\n",
+       "  {'alpha': 0.3947895791583166},\n",
+       "  {'alpha': 0.39679358717434865},\n",
+       "  {'alpha': 0.3987975951903807},\n",
+       "  {'alpha': 0.4008016032064128},\n",
+       "  {'alpha': 0.40280561122244485},\n",
+       "  {'alpha': 0.4048096192384769},\n",
+       "  {'alpha': 0.406813627254509},\n",
+       "  {'alpha': 0.40881763527054105},\n",
+       "  {'alpha': 0.4108216432865731},\n",
+       "  {'alpha': 0.4128256513026052},\n",
+       "  {'alpha': 0.41482965931863724},\n",
+       "  {'alpha': 0.4168336673346693},\n",
+       "  {'alpha': 0.4188376753507014},\n",
+       "  {'alpha': 0.42084168336673344},\n",
+       "  {'alpha': 0.4228456913827655},\n",
+       "  {'alpha': 0.4248496993987976},\n",
+       "  {'alpha': 0.42685370741482964},\n",
+       "  {'alpha': 0.4288577154308617},\n",
+       "  {'alpha': 0.4308617234468938},\n",
+       "  {'alpha': 0.43286573146292584},\n",
+       "  {'alpha': 0.4348697394789579},\n",
+       "  {'alpha': 0.4368737474949899},\n",
+       "  {'alpha': 0.438877755511022},\n",
+       "  {'alpha': 0.44088176352705405},\n",
+       "  {'alpha': 0.4428857715430861},\n",
+       "  {'alpha': 0.4448897795591182},\n",
+       "  {'alpha': 0.44689378757515025},\n",
+       "  {'alpha': 0.4488977955911823},\n",
+       "  {'alpha': 0.4509018036072144},\n",
+       "  {'alpha': 0.45290581162324645},\n",
+       "  {'alpha': 0.4549098196392785},\n",
+       "  {'alpha': 0.4569138276553106},\n",
+       "  {'alpha': 0.45891783567134264},\n",
+       "  {'alpha': 0.4609218436873747},\n",
+       "  {'alpha': 0.4629258517034068},\n",
+       "  {'alpha': 0.46492985971943884},\n",
+       "  {'alpha': 0.4669338677354709},\n",
+       "  {'alpha': 0.468937875751503},\n",
+       "  {'alpha': 0.47094188376753504},\n",
+       "  {'alpha': 0.4729458917835671},\n",
+       "  {'alpha': 0.4749498997995992},\n",
+       "  {'alpha': 0.47695390781563124},\n",
+       "  {'alpha': 0.4789579158316633},\n",
+       "  {'alpha': 0.48096192384769537},\n",
+       "  {'alpha': 0.48296593186372744},\n",
+       "  {'alpha': 0.4849699398797595},\n",
+       "  {'alpha': 0.48697394789579157},\n",
+       "  {'alpha': 0.4889779559118236},\n",
+       "  {'alpha': 0.49098196392785565},\n",
+       "  {'alpha': 0.4929859719438877},\n",
+       "  {'alpha': 0.4949899799599198},\n",
+       "  {'alpha': 0.49699398797595185},\n",
+       "  {'alpha': 0.4989979959919839},\n",
+       "  {'alpha': 0.501002004008016},\n",
+       "  {'alpha': 0.503006012024048},\n",
+       "  {'alpha': 0.5050100200400801},\n",
+       "  {'alpha': 0.5070140280561122},\n",
+       "  {'alpha': 0.5090180360721442},\n",
+       "  {'alpha': 0.5110220440881763},\n",
+       "  {'alpha': 0.5130260521042084},\n",
+       "  {'alpha': 0.5150300601202404},\n",
+       "  {'alpha': 0.5170340681362725},\n",
+       "  {'alpha': 0.5190380761523046},\n",
+       "  {'alpha': 0.5210420841683366},\n",
+       "  {'alpha': 0.5230460921843687},\n",
+       "  {'alpha': 0.5250501002004008},\n",
+       "  {'alpha': 0.5270541082164328},\n",
+       "  {'alpha': 0.5290581162324649},\n",
+       "  {'alpha': 0.531062124248497},\n",
+       "  {'alpha': 0.533066132264529},\n",
+       "  {'alpha': 0.5350701402805611},\n",
+       "  {'alpha': 0.5370741482965932},\n",
+       "  {'alpha': 0.5390781563126252},\n",
+       "  {'alpha': 0.5410821643286573},\n",
+       "  {'alpha': 0.5430861723446894},\n",
+       "  {'alpha': 0.5450901803607214},\n",
+       "  {'alpha': 0.5470941883767535},\n",
+       "  {'alpha': 0.5490981963927856},\n",
+       "  {'alpha': 0.5511022044088176},\n",
+       "  {'alpha': 0.5531062124248497},\n",
+       "  {'alpha': 0.5551102204408818},\n",
+       "  {'alpha': 0.5571142284569138},\n",
+       "  {'alpha': 0.5591182364729459},\n",
+       "  {'alpha': 0.561122244488978},\n",
+       "  {'alpha': 0.56312625250501},\n",
+       "  {'alpha': 0.565130260521042},\n",
+       "  {'alpha': 0.567134268537074},\n",
+       "  {'alpha': 0.5691382765531061},\n",
+       "  {'alpha': 0.5711422845691382},\n",
+       "  {'alpha': 0.5731462925851702},\n",
+       "  {'alpha': 0.5751503006012023},\n",
+       "  {'alpha': 0.5771543086172344},\n",
+       "  {'alpha': 0.5791583166332664},\n",
+       "  {'alpha': 0.5811623246492985},\n",
+       "  {'alpha': 0.5831663326653306},\n",
+       "  {'alpha': 0.5851703406813626},\n",
+       "  {'alpha': 0.5871743486973947},\n",
+       "  {'alpha': 0.5891783567134268},\n",
+       "  {'alpha': 0.5911823647294588},\n",
+       "  {'alpha': 0.5931863727454909},\n",
+       "  {'alpha': 0.595190380761523},\n",
+       "  {'alpha': 0.597194388777555},\n",
+       "  {'alpha': 0.5991983967935871},\n",
+       "  {'alpha': 0.6012024048096192},\n",
+       "  {'alpha': 0.6032064128256512},\n",
+       "  {'alpha': 0.6052104208416833},\n",
+       "  {'alpha': 0.6072144288577154},\n",
+       "  {'alpha': 0.6092184368737474},\n",
+       "  {'alpha': 0.6112224448897795},\n",
+       "  {'alpha': 0.6132264529058116},\n",
+       "  {'alpha': 0.6152304609218436},\n",
+       "  {'alpha': 0.6172344689378757},\n",
+       "  {'alpha': 0.6192384769539078},\n",
+       "  {'alpha': 0.6212424849699398},\n",
+       "  {'alpha': 0.6232464929859719},\n",
+       "  {'alpha': 0.625250501002004},\n",
+       "  {'alpha': 0.627254509018036},\n",
+       "  {'alpha': 0.6292585170340681},\n",
+       "  {'alpha': 0.6312625250501002},\n",
+       "  {'alpha': 0.6332665330661322},\n",
+       "  {'alpha': 0.6352705410821643},\n",
+       "  {'alpha': 0.6372745490981964},\n",
+       "  {'alpha': 0.6392785571142284},\n",
+       "  {'alpha': 0.6412825651302605},\n",
+       "  {'alpha': 0.6432865731462926},\n",
+       "  {'alpha': 0.6452905811623246},\n",
+       "  {'alpha': 0.6472945891783567},\n",
+       "  {'alpha': 0.6492985971943888},\n",
+       "  {'alpha': 0.6513026052104208},\n",
+       "  {'alpha': 0.6533066132264529},\n",
+       "  {'alpha': 0.655310621242485},\n",
+       "  {'alpha': 0.657314629258517},\n",
+       "  {'alpha': 0.6593186372745491},\n",
+       "  {'alpha': 0.6613226452905812},\n",
+       "  {'alpha': 0.6633266533066132},\n",
+       "  {'alpha': 0.6653306613226453},\n",
+       "  {'alpha': 0.6673346693386772},\n",
+       "  {'alpha': 0.6693386773547093},\n",
+       "  {'alpha': 0.6713426853707414},\n",
+       "  {'alpha': 0.6733466933867734},\n",
+       "  {'alpha': 0.6753507014028055},\n",
+       "  {'alpha': 0.6773547094188376},\n",
+       "  {'alpha': 0.6793587174348696},\n",
+       "  {'alpha': 0.6813627254509017},\n",
+       "  {'alpha': 0.6833667334669338},\n",
+       "  {'alpha': 0.6853707414829658},\n",
+       "  {'alpha': 0.6873747494989979},\n",
+       "  {'alpha': 0.68937875751503},\n",
+       "  {'alpha': 0.691382765531062},\n",
+       "  {'alpha': 0.6933867735470941},\n",
+       "  {'alpha': 0.6953907815631262},\n",
+       "  {'alpha': 0.6973947895791582},\n",
+       "  {'alpha': 0.6993987975951903},\n",
+       "  {'alpha': 0.7014028056112224},\n",
+       "  {'alpha': 0.7034068136272544},\n",
+       "  {'alpha': 0.7054108216432865},\n",
+       "  {'alpha': 0.7074148296593186},\n",
+       "  {'alpha': 0.7094188376753506},\n",
+       "  {'alpha': 0.7114228456913827},\n",
+       "  {'alpha': 0.7134268537074148},\n",
+       "  {'alpha': 0.7154308617234468},\n",
+       "  {'alpha': 0.7174348697394789},\n",
+       "  {'alpha': 0.719438877755511},\n",
+       "  {'alpha': 0.721442885771543},\n",
+       "  {'alpha': 0.7234468937875751},\n",
+       "  {'alpha': 0.7254509018036072},\n",
+       "  {'alpha': 0.7274549098196392},\n",
+       "  {'alpha': 0.7294589178356713},\n",
+       "  {'alpha': 0.7314629258517034},\n",
+       "  {'alpha': 0.7334669338677354},\n",
+       "  {'alpha': 0.7354709418837675},\n",
+       "  {'alpha': 0.7374749498997996},\n",
+       "  {'alpha': 0.7394789579158316},\n",
+       "  {'alpha': 0.7414829659318637},\n",
+       "  {'alpha': 0.7434869739478958},\n",
+       "  {'alpha': 0.7454909819639278},\n",
+       "  {'alpha': 0.7474949899799599},\n",
+       "  {'alpha': 0.749498997995992},\n",
+       "  {'alpha': 0.751503006012024},\n",
+       "  {'alpha': 0.7535070140280561},\n",
+       "  {'alpha': 0.7555110220440882},\n",
+       "  {'alpha': 0.7575150300601202},\n",
+       "  {'alpha': 0.7595190380761523},\n",
+       "  {'alpha': 0.7615230460921844},\n",
+       "  {'alpha': 0.7635270541082164},\n",
+       "  {'alpha': 0.7655310621242485},\n",
+       "  {'alpha': 0.7675350701402806},\n",
+       "  {'alpha': 0.7695390781563125},\n",
+       "  {'alpha': 0.7715430861723446},\n",
+       "  {'alpha': 0.7735470941883766},\n",
+       "  {'alpha': 0.7755511022044087},\n",
+       "  {'alpha': 0.7775551102204408},\n",
+       "  {'alpha': 0.7795591182364728},\n",
+       "  {'alpha': 0.7815631262525049},\n",
+       "  {'alpha': 0.783567134268537},\n",
+       "  {'alpha': 0.785571142284569},\n",
+       "  {'alpha': 0.7875751503006011},\n",
+       "  {'alpha': 0.7895791583166332},\n",
+       "  {'alpha': 0.7915831663326652},\n",
+       "  {'alpha': 0.7935871743486973},\n",
+       "  {'alpha': 0.7955911823647294},\n",
+       "  {'alpha': 0.7975951903807614},\n",
+       "  {'alpha': 0.7995991983967935},\n",
+       "  {'alpha': 0.8016032064128256},\n",
+       "  {'alpha': 0.8036072144288576},\n",
+       "  {'alpha': 0.8056112224448897},\n",
+       "  {'alpha': 0.8076152304609218},\n",
+       "  {'alpha': 0.8096192384769538},\n",
+       "  {'alpha': 0.8116232464929859},\n",
+       "  {'alpha': 0.813627254509018},\n",
+       "  {'alpha': 0.81563126252505},\n",
+       "  {'alpha': 0.8176352705410821},\n",
+       "  {'alpha': 0.8196392785571142},\n",
+       "  {'alpha': 0.8216432865731462},\n",
+       "  {'alpha': 0.8236472945891783},\n",
+       "  {'alpha': 0.8256513026052104},\n",
+       "  {'alpha': 0.8276553106212424},\n",
+       "  {'alpha': 0.8296593186372745},\n",
+       "  {'alpha': 0.8316633266533066},\n",
+       "  {'alpha': 0.8336673346693386},\n",
+       "  {'alpha': 0.8356713426853707},\n",
+       "  {'alpha': 0.8376753507014028},\n",
+       "  {'alpha': 0.8396793587174348},\n",
+       "  {'alpha': 0.8416833667334669},\n",
+       "  {'alpha': 0.843687374749499},\n",
+       "  {'alpha': 0.845691382765531},\n",
+       "  {'alpha': 0.8476953907815631},\n",
+       "  {'alpha': 0.8496993987975952},\n",
+       "  {'alpha': 0.8517034068136272},\n",
+       "  {'alpha': 0.8537074148296593},\n",
+       "  {'alpha': 0.8557114228456913},\n",
+       "  {'alpha': 0.8577154308617234},\n",
+       "  {'alpha': 0.8597194388777555},\n",
+       "  {'alpha': 0.8617234468937875},\n",
+       "  {'alpha': 0.8637274549098196},\n",
+       "  {'alpha': 0.8657314629258517},\n",
+       "  {'alpha': 0.8677354709418837},\n",
+       "  {'alpha': 0.8697394789579158},\n",
+       "  {'alpha': 0.8717434869739479},\n",
+       "  {'alpha': 0.8737474949899798},\n",
+       "  {'alpha': 0.8757515030060119},\n",
+       "  {'alpha': 0.877755511022044},\n",
+       "  {'alpha': 0.879759519038076},\n",
+       "  {'alpha': 0.8817635270541081},\n",
+       "  {'alpha': 0.8837675350701402},\n",
+       "  {'alpha': 0.8857715430861722},\n",
+       "  {'alpha': 0.8877755511022043},\n",
+       "  {'alpha': 0.8897795591182364},\n",
+       "  {'alpha': 0.8917835671342684},\n",
+       "  {'alpha': 0.8937875751503005},\n",
+       "  {'alpha': 0.8957915831663326},\n",
+       "  {'alpha': 0.8977955911823646},\n",
+       "  {'alpha': 0.8997995991983967},\n",
+       "  {'alpha': 0.9018036072144288},\n",
+       "  {'alpha': 0.9038076152304608},\n",
+       "  {'alpha': 0.9058116232464929},\n",
+       "  {'alpha': 0.907815631262525},\n",
+       "  {'alpha': 0.909819639278557},\n",
+       "  {'alpha': 0.9118236472945891},\n",
+       "  {'alpha': 0.9138276553106212},\n",
+       "  {'alpha': 0.9158316633266532},\n",
+       "  {'alpha': 0.9178356713426853},\n",
+       "  {'alpha': 0.9198396793587174},\n",
+       "  {'alpha': 0.9218436873747494},\n",
+       "  {'alpha': 0.9238476953907815},\n",
+       "  {'alpha': 0.9258517034068136},\n",
+       "  {'alpha': 0.9278557114228456},\n",
+       "  {'alpha': 0.9298597194388777},\n",
+       "  {'alpha': 0.9318637274549098},\n",
+       "  {'alpha': 0.9338677354709418},\n",
+       "  {'alpha': 0.9358717434869739},\n",
+       "  {'alpha': 0.937875751503006},\n",
+       "  {'alpha': 0.939879759519038},\n",
+       "  {'alpha': 0.9418837675350701},\n",
+       "  {'alpha': 0.9438877755511021},\n",
+       "  {'alpha': 0.9458917835671342},\n",
+       "  {'alpha': 0.9478957915831663},\n",
+       "  {'alpha': 0.9498997995991983},\n",
+       "  {'alpha': 0.9519038076152304},\n",
+       "  {'alpha': 0.9539078156312625},\n",
+       "  {'alpha': 0.9559118236472945},\n",
+       "  {'alpha': 0.9579158316633266},\n",
+       "  {'alpha': 0.9599198396793587},\n",
+       "  {'alpha': 0.9619238476953907},\n",
+       "  {'alpha': 0.9639278557114228},\n",
+       "  {'alpha': 0.9659318637274549},\n",
+       "  {'alpha': 0.9679358717434869},\n",
+       "  {'alpha': 0.969939879759519},\n",
+       "  {'alpha': 0.9719438877755511},\n",
+       "  {'alpha': 0.9739478957915831},\n",
+       "  {'alpha': 0.9759519038076151},\n",
+       "  {'alpha': 0.9779559118236472},\n",
+       "  {'alpha': 0.9799599198396792},\n",
+       "  {'alpha': 0.9819639278557113},\n",
+       "  {'alpha': 0.9839679358717434},\n",
+       "  {'alpha': 0.9859719438877754},\n",
+       "  {'alpha': 0.9879759519038075},\n",
+       "  {'alpha': 0.9899799599198396},\n",
+       "  {'alpha': 0.9919839679358716},\n",
+       "  {'alpha': 0.9939879759519037},\n",
+       "  {'alpha': 0.9959919839679358},\n",
+       "  {'alpha': 0.9979959919839678},\n",
+       "  {'alpha': 1.0}],\n",
+       " 'rank_test_score': array([500, 499, 498, 497, 496, 495, 494, 493, 492, 491, 490, 489, 488,\n",
+       "        487, 486, 485, 484, 483, 482, 481, 480, 479, 478, 477, 476, 475,\n",
+       "        474, 473, 472, 471, 470, 469, 468, 467, 466, 465, 464, 463, 462,\n",
+       "        461, 460, 459, 458, 457, 456, 455, 454, 453, 452, 451, 450, 449,\n",
+       "        448, 447, 446, 445, 444, 443, 442, 441, 440, 439, 438, 437, 436,\n",
+       "        435, 434, 433, 432, 431, 430, 429, 428, 427, 426, 425, 424, 423,\n",
+       "        422, 421, 420, 419, 418, 417, 416, 415, 414, 413, 412, 411, 410,\n",
+       "        409, 408, 407, 406, 405, 404, 403, 402, 401, 400, 399, 398, 397,\n",
+       "        396, 395, 394, 393, 392, 391, 390, 389, 388, 387, 386, 385, 384,\n",
+       "        383, 382, 381, 380, 379, 378, 377, 376, 375, 374, 373, 372, 371,\n",
+       "        370, 369, 368, 367, 366, 365, 364, 363, 362, 361, 360, 359, 358,\n",
+       "        357, 356, 355, 354, 353, 352, 351, 350, 349, 348, 347, 346, 345,\n",
+       "        344, 343, 342, 341, 340, 339, 338, 337, 336, 335, 334, 333, 332,\n",
+       "        331, 330, 329, 328, 327, 326, 325, 324, 323, 322, 321, 320, 319,\n",
+       "        318, 317, 316, 315, 314, 313, 312, 311, 310, 309, 308, 307, 306,\n",
+       "        305, 304, 303, 302, 301, 300, 299, 298, 297, 296, 295, 294, 293,\n",
+       "        292, 291, 290, 289, 288, 287, 286, 285, 284, 283, 282, 281, 280,\n",
+       "        279, 278, 277, 276, 275, 274, 273, 272, 271, 270, 269, 268, 267,\n",
+       "        266, 265, 264, 263, 262, 261, 260, 259, 258, 257, 256, 255, 254,\n",
+       "        253, 252, 251, 250, 249, 248, 247, 246, 245, 244, 243, 242, 241,\n",
+       "        240, 239, 238, 237, 236, 235, 234, 233, 232, 231, 230, 229, 228,\n",
+       "        227, 226, 225, 224, 223, 222, 221, 220, 219, 218, 217, 216, 215,\n",
+       "        214, 213, 212, 211, 210, 209, 208, 207, 206, 205, 204, 203, 202,\n",
+       "        201, 200, 199, 198, 197, 196, 195, 194, 193, 192, 191, 190, 189,\n",
+       "        188, 187, 186, 185, 184, 183, 182, 181, 180, 179, 178, 177, 176,\n",
+       "        175, 174, 173, 172, 171, 170, 169, 168, 167, 166, 165, 164, 163,\n",
+       "        162, 161, 160, 159, 158, 157, 156, 155, 154, 153, 152, 151, 150,\n",
+       "        149, 148, 147, 146, 145, 144, 143, 142, 141, 140, 139, 138, 137,\n",
+       "        136, 135, 134, 133, 132, 131, 130, 129, 128, 127, 126, 125, 124,\n",
+       "        123, 122, 121, 120, 119, 118, 117, 116, 115, 114, 113, 112, 111,\n",
+       "        110, 109, 108, 107, 106, 105, 104, 103, 102, 101, 100,  99,  98,\n",
+       "         97,  96,  95,  94,  93,  92,  91,  90,  89,  88,  87,  86,  85,\n",
+       "         84,  83,  82,  81,  80,  79,  78,  77,  76,  75,  74,  73,  72,\n",
+       "         71,  70,  69,  68,  67,  66,  65,  64,  63,  62,  61,  60,  59,\n",
+       "         58,  57,  56,  55,  54,  53,  52,  51,  50,  49,  48,  47,  46,\n",
+       "         45,  44,  43,  42,  41,  40,  39,  38,  37,  36,  35,  34,  33,\n",
+       "         32,  31,  30,  29,  28,  27,  26,  25,  24,  23,  22,  21,  20,\n",
+       "         19,  18,  17,  16,  15,  14,  13,  12,  11,  10,   9,   8,   7,\n",
+       "          6,   5,   4,   3,   2,   1], dtype=int32),\n",
+       " 'split0_test_score': array([-0.0257959 , -0.02579573, -0.02579556, -0.02579539, -0.02579522,\n",
+       "        -0.02579505, -0.02579488, -0.02579471, -0.02579454, -0.02579437,\n",
+       "        -0.0257942 , -0.02579403, -0.02579385, -0.02579368, -0.02579351,\n",
+       "        -0.02579334, -0.02579317, -0.025793  , -0.02579283, -0.02579266,\n",
+       "        -0.02579249, -0.02579232, -0.02579215, -0.02579198, -0.0257918 ,\n",
+       "        -0.02579163, -0.02579146, -0.02579129, -0.02579112, -0.02579095,\n",
+       "        -0.02579078, -0.02579061, -0.02579044, -0.02579027, -0.0257901 ,\n",
+       "        -0.02578993, -0.02578976, -0.02578958, -0.02578941, -0.02578924,\n",
+       "        -0.02578907, -0.0257889 , -0.02578873, -0.02578856, -0.02578839,\n",
+       "        -0.02578822, -0.02578805, -0.02578788, -0.02578771, -0.02578754,\n",
+       "        -0.02578736, -0.02578719, -0.02578702, -0.02578685, -0.02578668,\n",
+       "        -0.02578651, -0.02578634, -0.02578617, -0.025786  , -0.02578583,\n",
+       "        -0.02578566, -0.02578549, -0.02578532, -0.02578514, -0.02578497,\n",
+       "        -0.0257848 , -0.02578463, -0.02578446, -0.02578429, -0.02578412,\n",
+       "        -0.02578395, -0.02578378, -0.02578361, -0.02578344, -0.02578327,\n",
+       "        -0.0257831 , -0.02578292, -0.02578275, -0.02578258, -0.02578241,\n",
+       "        -0.02578224, -0.02578207, -0.0257819 , -0.02578173, -0.02578156,\n",
+       "        -0.02578139, -0.02578122, -0.02578105, -0.02578088, -0.02578071,\n",
+       "        -0.02578053, -0.02578036, -0.02578019, -0.02578002, -0.02577985,\n",
+       "        -0.02577968, -0.02577951, -0.02577934, -0.02577917, -0.025779  ,\n",
+       "        -0.02577883, -0.02577866, -0.02577849, -0.02577832, -0.02577814,\n",
+       "        -0.02577797, -0.0257778 , -0.02577763, -0.02577746, -0.02577729,\n",
+       "        -0.02577712, -0.02577695, -0.02577678, -0.02577661, -0.02577644,\n",
+       "        -0.02577627, -0.0257761 , -0.02577593, -0.02577575, -0.02577558,\n",
+       "        -0.02577541, -0.02577524, -0.02577507, -0.0257749 , -0.02577473,\n",
+       "        -0.02577456, -0.02577439, -0.02577422, -0.02577405, -0.02577388,\n",
+       "        -0.02577371, -0.02577354, -0.02577336, -0.02577319, -0.02577302,\n",
+       "        -0.02577285, -0.02577268, -0.02577251, -0.02577234, -0.02577217,\n",
+       "        -0.025772  , -0.02577183, -0.02577166, -0.02577149, -0.02577132,\n",
+       "        -0.02577115, -0.02577098, -0.0257708 , -0.02577063, -0.02577046,\n",
+       "        -0.02577029, -0.02577012, -0.02576995, -0.02576978, -0.02576961,\n",
+       "        -0.02576944, -0.02576927, -0.0257691 , -0.02576893, -0.02576876,\n",
+       "        -0.02576859, -0.02576842, -0.02576824, -0.02576807, -0.0257679 ,\n",
+       "        -0.02576773, -0.02576756, -0.02576739, -0.02576722, -0.02576705,\n",
+       "        -0.02576688, -0.02576671, -0.02576654, -0.02576637, -0.0257662 ,\n",
+       "        -0.02576603, -0.02576586, -0.02576568, -0.02576551, -0.02576534,\n",
+       "        -0.02576517, -0.025765  , -0.02576483, -0.02576466, -0.02576449,\n",
+       "        -0.02576432, -0.02576415, -0.02576398, -0.02576381, -0.02576364,\n",
+       "        -0.02576347, -0.0257633 , -0.02576313, -0.02576295, -0.02576278,\n",
+       "        -0.02576261, -0.02576244, -0.02576227, -0.0257621 , -0.02576193,\n",
+       "        -0.02576176, -0.02576159, -0.02576142, -0.02576125, -0.02576108,\n",
+       "        -0.02576091, -0.02576074, -0.02576057, -0.0257604 , -0.02576022,\n",
+       "        -0.02576005, -0.02575988, -0.02575971, -0.02575954, -0.02575937,\n",
+       "        -0.0257592 , -0.02575903, -0.02575886, -0.02575869, -0.02575852,\n",
+       "        -0.02575835, -0.02575818, -0.02575801, -0.02575784, -0.02575767,\n",
+       "        -0.02575749, -0.02575732, -0.02575715, -0.02575698, -0.02575681,\n",
+       "        -0.02575664, -0.02575647, -0.0257563 , -0.02575613, -0.02575596,\n",
+       "        -0.02575579, -0.02575562, -0.02575545, -0.02575528, -0.02575511,\n",
+       "        -0.02575494, -0.02575477, -0.02575459, -0.02575442, -0.02575425,\n",
+       "        -0.02575408, -0.02575391, -0.02575374, -0.02575357, -0.0257534 ,\n",
+       "        -0.02575323, -0.02575306, -0.02575289, -0.02575272, -0.02575255,\n",
+       "        -0.02575238, -0.02575221, -0.02575204, -0.02575187, -0.0257517 ,\n",
+       "        -0.02575152, -0.02575135, -0.02575118, -0.02575101, -0.02575084,\n",
+       "        -0.02575067, -0.0257505 , -0.02575033, -0.02575016, -0.02574999,\n",
+       "        -0.02574982, -0.02574965, -0.02574948, -0.02574931, -0.02574914,\n",
+       "        -0.02574897, -0.0257488 , -0.02574863, -0.02574845, -0.02574828,\n",
+       "        -0.02574811, -0.02574794, -0.02574777, -0.0257476 , -0.02574743,\n",
+       "        -0.02574726, -0.02574709, -0.02574692, -0.02574675, -0.02574658,\n",
+       "        -0.02574641, -0.02574624, -0.02574607, -0.0257459 , -0.02574573,\n",
+       "        -0.02574556, -0.02574538, -0.02574521, -0.02574504, -0.02574487,\n",
+       "        -0.0257447 , -0.02574453, -0.02574436, -0.02574419, -0.02574402,\n",
+       "        -0.02574385, -0.02574368, -0.02574351, -0.02574334, -0.02574317,\n",
+       "        -0.025743  , -0.02574283, -0.02574266, -0.02574249, -0.02574232,\n",
+       "        -0.02574214, -0.02574197, -0.0257418 , -0.02574163, -0.02574146,\n",
+       "        -0.02574129, -0.02574112, -0.02574095, -0.02574078, -0.02574061,\n",
+       "        -0.02574044, -0.02574027, -0.0257401 , -0.02573993, -0.02573976,\n",
+       "        -0.02573959, -0.02573942, -0.02573925, -0.02573908, -0.02573891,\n",
+       "        -0.02573873, -0.02573856, -0.02573839, -0.02573822, -0.02573805,\n",
+       "        -0.02573788, -0.02573771, -0.02573754, -0.02573737, -0.0257372 ,\n",
+       "        -0.02573703, -0.02573686, -0.02573669, -0.02573652, -0.02573635,\n",
+       "        -0.02573618, -0.02573601, -0.02573584, -0.02573567, -0.0257355 ,\n",
+       "        -0.02573532, -0.02573515, -0.02573498, -0.02573481, -0.02573464,\n",
+       "        -0.02573447, -0.0257343 , -0.02573413, -0.02573396, -0.02573379,\n",
+       "        -0.02573362, -0.02573345, -0.02573328, -0.02573311, -0.02573294,\n",
+       "        -0.02573277, -0.0257326 , -0.02573243, -0.02573226, -0.02573209,\n",
+       "        -0.02573192, -0.02573175, -0.02573157, -0.0257314 , -0.02573123,\n",
+       "        -0.02573106, -0.02573089, -0.02573072, -0.02573055, -0.02573038,\n",
+       "        -0.02573021, -0.02573004, -0.02572987, -0.0257297 , -0.02572953,\n",
+       "        -0.02572936, -0.02572919, -0.02572902, -0.02572885, -0.02572868,\n",
+       "        -0.02572851, -0.02572834, -0.02572817, -0.025728  , -0.02572782,\n",
+       "        -0.02572765, -0.02572748, -0.02572731, -0.02572714, -0.02572697,\n",
+       "        -0.0257268 , -0.02572663, -0.02572646, -0.02572629, -0.02572612,\n",
+       "        -0.02572595, -0.02572578, -0.02572561, -0.02572544, -0.02572527,\n",
+       "        -0.0257251 , -0.02572493, -0.02572476, -0.02572459, -0.02572442,\n",
+       "        -0.02572425, -0.02572408, -0.02572391, -0.02572373, -0.02572356,\n",
+       "        -0.02572339, -0.02572322, -0.02572305, -0.02572288, -0.02572271,\n",
+       "        -0.02572254, -0.02572237, -0.0257222 , -0.02572203, -0.02572186,\n",
+       "        -0.02572169, -0.02572152, -0.02572135, -0.02572118, -0.02572101,\n",
+       "        -0.02572084, -0.02572067, -0.0257205 , -0.02572033, -0.02572016,\n",
+       "        -0.02571999, -0.02571982, -0.02571965, -0.02571947, -0.0257193 ,\n",
+       "        -0.02571913, -0.02571896, -0.02571879, -0.02571862, -0.02571845,\n",
+       "        -0.02571828, -0.02571811, -0.02571794, -0.02571777, -0.0257176 ,\n",
+       "        -0.02571743, -0.02571726, -0.02571709, -0.02571692, -0.02571675,\n",
+       "        -0.02571658, -0.02571641, -0.02571624, -0.02571607, -0.0257159 ,\n",
+       "        -0.02571573, -0.02571556, -0.02571539, -0.02571522, -0.02571504,\n",
+       "        -0.02571487, -0.0257147 , -0.02571453, -0.02571436, -0.02571419,\n",
+       "        -0.02571402, -0.02571385, -0.02571368, -0.02571351, -0.02571334,\n",
+       "        -0.02571317, -0.025713  , -0.02571283, -0.02571266, -0.02571249,\n",
+       "        -0.02571232, -0.02571215, -0.02571198, -0.02571181, -0.02571164,\n",
+       "        -0.02571147, -0.0257113 , -0.02571113, -0.02571096, -0.02571079]),\n",
+       " 'split0_train_score': array([0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277406, 0.18277406, 0.18277406,\n",
+       "        0.18277406, 0.18277406, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277405, 0.18277405, 0.18277405, 0.18277405, 0.18277405,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277404, 0.18277404, 0.18277404, 0.18277404,\n",
+       "        0.18277404, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277403,\n",
+       "        0.18277403, 0.18277403, 0.18277403, 0.18277403, 0.18277402,\n",
+       "        0.18277402, 0.18277402, 0.18277402, 0.18277402, 0.18277402,\n",
+       "        0.18277402, 0.18277402, 0.18277402, 0.18277402, 0.18277402]),\n",
+       " 'split1_test_score': array([-0.17045855, -0.17045855, -0.17045854, -0.17045853, -0.17045852,\n",
+       "        -0.17045852, -0.17045851, -0.1704585 , -0.17045849, -0.17045849,\n",
+       "        -0.17045848, -0.17045847, -0.17045846, -0.17045845, -0.17045845,\n",
+       "        -0.17045844, -0.17045843, -0.17045842, -0.17045842, -0.17045841,\n",
+       "        -0.1704584 , -0.17045839, -0.17045839, -0.17045838, -0.17045837,\n",
+       "        -0.17045836, -0.17045836, -0.17045835, -0.17045834, -0.17045833,\n",
+       "        -0.17045833, -0.17045832, -0.17045831, -0.1704583 , -0.1704583 ,\n",
+       "        -0.17045829, -0.17045828, -0.17045827, -0.17045826, -0.17045826,\n",
+       "        -0.17045825, -0.17045824, -0.17045823, -0.17045823, -0.17045822,\n",
+       "        -0.17045821, -0.1704582 , -0.1704582 , -0.17045819, -0.17045818,\n",
+       "        -0.17045817, -0.17045817, -0.17045816, -0.17045815, -0.17045814,\n",
+       "        -0.17045814, -0.17045813, -0.17045812, -0.17045811, -0.1704581 ,\n",
+       "        -0.1704581 , -0.17045809, -0.17045808, -0.17045807, -0.17045807,\n",
+       "        -0.17045806, -0.17045805, -0.17045804, -0.17045804, -0.17045803,\n",
+       "        -0.17045802, -0.17045801, -0.17045801, -0.170458  , -0.17045799,\n",
+       "        -0.17045798, -0.17045798, -0.17045797, -0.17045796, -0.17045795,\n",
+       "        -0.17045795, -0.17045794, -0.17045793, -0.17045792, -0.17045791,\n",
+       "        -0.17045791, -0.1704579 , -0.17045789, -0.17045788, -0.17045788,\n",
+       "        -0.17045787, -0.17045786, -0.17045785, -0.17045785, -0.17045784,\n",
+       "        -0.17045783, -0.17045782, -0.17045782, -0.17045781, -0.1704578 ,\n",
+       "        -0.17045779, -0.17045779, -0.17045778, -0.17045777, -0.17045776,\n",
+       "        -0.17045776, -0.17045775, -0.17045774, -0.17045773, -0.17045773,\n",
+       "        -0.17045772, -0.17045771, -0.1704577 , -0.17045769, -0.17045769,\n",
+       "        -0.17045768, -0.17045767, -0.17045766, -0.17045766, -0.17045765,\n",
+       "        -0.17045764, -0.17045763, -0.17045763, -0.17045762, -0.17045761,\n",
+       "        -0.1704576 , -0.1704576 , -0.17045759, -0.17045758, -0.17045757,\n",
+       "        -0.17045757, -0.17045756, -0.17045755, -0.17045754, -0.17045754,\n",
+       "        -0.17045753, -0.17045752, -0.17045751, -0.17045751, -0.1704575 ,\n",
+       "        -0.17045749, -0.17045748, -0.17045747, -0.17045747, -0.17045746,\n",
+       "        -0.17045745, -0.17045744, -0.17045744, -0.17045743, -0.17045742,\n",
+       "        -0.17045741, -0.17045741, -0.1704574 , -0.17045739, -0.17045738,\n",
+       "        -0.17045738, -0.17045737, -0.17045736, -0.17045735, -0.17045735,\n",
+       "        -0.17045734, -0.17045733, -0.17045732, -0.17045732, -0.17045731,\n",
+       "        -0.1704573 , -0.17045729, -0.17045729, -0.17045728, -0.17045727,\n",
+       "        -0.17045726, -0.17045726, -0.17045725, -0.17045724, -0.17045723,\n",
+       "        -0.17045722, -0.17045722, -0.17045721, -0.1704572 , -0.17045719,\n",
+       "        -0.17045719, -0.17045718, -0.17045717, -0.17045716, -0.17045716,\n",
+       "        -0.17045715, -0.17045714, -0.17045713, -0.17045713, -0.17045712,\n",
+       "        -0.17045711, -0.1704571 , -0.1704571 , -0.17045709, -0.17045708,\n",
+       "        -0.17045707, -0.17045707, -0.17045706, -0.17045705, -0.17045704,\n",
+       "        -0.17045704, -0.17045703, -0.17045702, -0.17045701, -0.17045701,\n",
+       "        -0.170457  , -0.17045699, -0.17045698, -0.17045698, -0.17045697,\n",
+       "        -0.17045696, -0.17045695, -0.17045694, -0.17045694, -0.17045693,\n",
+       "        -0.17045692, -0.17045691, -0.17045691, -0.1704569 , -0.17045689,\n",
+       "        -0.17045688, -0.17045688, -0.17045687, -0.17045686, -0.17045685,\n",
+       "        -0.17045685, -0.17045684, -0.17045683, -0.17045682, -0.17045682,\n",
+       "        -0.17045681, -0.1704568 , -0.17045679, -0.17045679, -0.17045678,\n",
+       "        -0.17045677, -0.17045676, -0.17045676, -0.17045675, -0.17045674,\n",
+       "        -0.17045673, -0.17045673, -0.17045672, -0.17045671, -0.1704567 ,\n",
+       "        -0.1704567 , -0.17045669, -0.17045668, -0.17045667, -0.17045667,\n",
+       "        -0.17045666, -0.17045665, -0.17045664, -0.17045664, -0.17045663,\n",
+       "        -0.17045662, -0.17045661, -0.1704566 , -0.1704566 , -0.17045659,\n",
+       "        -0.17045658, -0.17045657, -0.17045657, -0.17045656, -0.17045655,\n",
+       "        -0.17045654, -0.17045654, -0.17045653, -0.17045652, -0.17045651,\n",
+       "        -0.17045651, -0.1704565 , -0.17045649, -0.17045648, -0.17045648,\n",
+       "        -0.17045647, -0.17045646, -0.17045645, -0.17045645, -0.17045644,\n",
+       "        -0.17045643, -0.17045642, -0.17045642, -0.17045641, -0.1704564 ,\n",
+       "        -0.17045639, -0.17045639, -0.17045638, -0.17045637, -0.17045636,\n",
+       "        -0.17045636, -0.17045635, -0.17045634, -0.17045633, -0.17045633,\n",
+       "        -0.17045632, -0.17045631, -0.1704563 , -0.1704563 , -0.17045629,\n",
+       "        -0.17045628, -0.17045627, -0.17045627, -0.17045626, -0.17045625,\n",
+       "        -0.17045624, -0.17045624, -0.17045623, -0.17045622, -0.17045621,\n",
+       "        -0.17045621, -0.1704562 , -0.17045619, -0.17045618, -0.17045617,\n",
+       "        -0.17045617, -0.17045616, -0.17045615, -0.17045614, -0.17045614,\n",
+       "        -0.17045613, -0.17045612, -0.17045611, -0.17045611, -0.1704561 ,\n",
+       "        -0.17045609, -0.17045608, -0.17045608, -0.17045607, -0.17045606,\n",
+       "        -0.17045605, -0.17045605, -0.17045604, -0.17045603, -0.17045602,\n",
+       "        -0.17045602, -0.17045601, -0.170456  , -0.17045599, -0.17045599,\n",
+       "        -0.17045598, -0.17045597, -0.17045596, -0.17045596, -0.17045595,\n",
+       "        -0.17045594, -0.17045593, -0.17045593, -0.17045592, -0.17045591,\n",
+       "        -0.1704559 , -0.1704559 , -0.17045589, -0.17045588, -0.17045587,\n",
+       "        -0.17045587, -0.17045586, -0.17045585, -0.17045584, -0.17045584,\n",
+       "        -0.17045583, -0.17045582, -0.17045581, -0.17045581, -0.1704558 ,\n",
+       "        -0.17045579, -0.17045578, -0.17045578, -0.17045577, -0.17045576,\n",
+       "        -0.17045575, -0.17045575, -0.17045574, -0.17045573, -0.17045572,\n",
+       "        -0.17045572, -0.17045571, -0.1704557 , -0.17045569, -0.17045569,\n",
+       "        -0.17045568, -0.17045567, -0.17045566, -0.17045566, -0.17045565,\n",
+       "        -0.17045564, -0.17045563, -0.17045563, -0.17045562, -0.17045561,\n",
+       "        -0.1704556 , -0.1704556 , -0.17045559, -0.17045558, -0.17045557,\n",
+       "        -0.17045557, -0.17045556, -0.17045555, -0.17045554, -0.17045554,\n",
+       "        -0.17045553, -0.17045552, -0.17045551, -0.17045551, -0.1704555 ,\n",
+       "        -0.17045549, -0.17045548, -0.17045548, -0.17045547, -0.17045546,\n",
+       "        -0.17045545, -0.17045545, -0.17045544, -0.17045543, -0.17045542,\n",
+       "        -0.17045542, -0.17045541, -0.1704554 , -0.17045539, -0.17045539,\n",
+       "        -0.17045538, -0.17045537, -0.17045536, -0.17045536, -0.17045535,\n",
+       "        -0.17045534, -0.17045533, -0.17045533, -0.17045532, -0.17045531,\n",
+       "        -0.1704553 , -0.1704553 , -0.17045529, -0.17045528, -0.17045527,\n",
+       "        -0.17045527, -0.17045526, -0.17045525, -0.17045524, -0.17045524,\n",
+       "        -0.17045523, -0.17045522, -0.17045521, -0.17045521, -0.1704552 ,\n",
+       "        -0.17045519, -0.17045518, -0.17045518, -0.17045517, -0.17045516,\n",
+       "        -0.17045515, -0.17045515, -0.17045514, -0.17045513, -0.17045512,\n",
+       "        -0.17045512, -0.17045511, -0.1704551 , -0.17045509, -0.17045509,\n",
+       "        -0.17045508, -0.17045507, -0.17045506, -0.17045506, -0.17045505,\n",
+       "        -0.17045504, -0.17045503, -0.17045503, -0.17045502, -0.17045501,\n",
+       "        -0.170455  , -0.170455  , -0.17045499, -0.17045498, -0.17045497,\n",
+       "        -0.17045497, -0.17045496, -0.17045495, -0.17045494, -0.17045494,\n",
+       "        -0.17045493, -0.17045492, -0.17045491, -0.17045491, -0.1704549 ,\n",
+       "        -0.17045489, -0.17045488, -0.17045488, -0.17045487, -0.17045486,\n",
+       "        -0.17045485, -0.17045485, -0.17045484, -0.17045483, -0.17045482,\n",
+       "        -0.17045482, -0.17045481, -0.1704548 , -0.17045479, -0.17045479]),\n",
+       " 'split1_train_score': array([0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595238, 0.08595238, 0.08595238,\n",
+       "        0.08595238, 0.08595238, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237,\n",
+       "        0.08595237, 0.08595237, 0.08595237, 0.08595237, 0.08595237]),\n",
+       " 'split2_test_score': array([-1.17867893, -1.17867184, -1.17866476, -1.17865767, -1.17865059,\n",
+       "        -1.17864351, -1.17863642, -1.17862934, -1.17862226, -1.17861517,\n",
+       "        -1.17860809, -1.17860101, -1.17859393, -1.17858684, -1.17857976,\n",
+       "        -1.17857268, -1.1785656 , -1.17855852, -1.17855144, -1.17854436,\n",
+       "        -1.17853728, -1.1785302 , -1.17852312, -1.17851604, -1.17850896,\n",
+       "        -1.17850188, -1.1784948 , -1.17848772, -1.17848064, -1.17847356,\n",
+       "        -1.17846649, -1.17845941, -1.17845233, -1.17844525, -1.17843818,\n",
+       "        -1.1784311 , -1.17842402, -1.17841695, -1.17840987, -1.17840279,\n",
+       "        -1.17839572, -1.17838864, -1.17838157, -1.17837449, -1.17836742,\n",
+       "        -1.17836034, -1.17835327, -1.17834619, -1.17833912, -1.17833205,\n",
+       "        -1.17832497, -1.1783179 , -1.17831083, -1.17830375, -1.17829668,\n",
+       "        -1.17828961, -1.17828254, -1.17827546, -1.17826839, -1.17826132,\n",
+       "        -1.17825425, -1.17824718, -1.17824011, -1.17823304, -1.17822597,\n",
+       "        -1.1782189 , -1.17821183, -1.17820476, -1.17819769, -1.17819062,\n",
+       "        -1.17818355, -1.17817648, -1.17816941, -1.17816234, -1.17815528,\n",
+       "        -1.17814821, -1.17814114, -1.17813407, -1.17812701, -1.17811994,\n",
+       "        -1.17811287, -1.17810581, -1.17809874, -1.17809167, -1.17808461,\n",
+       "        -1.17807754, -1.17807048, -1.17806341, -1.17805635, -1.17804928,\n",
+       "        -1.17804222, -1.17803515, -1.17802809, -1.17802103, -1.17801396,\n",
+       "        -1.1780069 , -1.17799984, -1.17799277, -1.17798571, -1.17797865,\n",
+       "        -1.17797159, -1.17796453, -1.17795746, -1.1779504 , -1.17794334,\n",
+       "        -1.17793628, -1.17792922, -1.17792216, -1.1779151 , -1.17790804,\n",
+       "        -1.17790098, -1.17789392, -1.17788686, -1.1778798 , -1.17787274,\n",
+       "        -1.17786568, -1.17785862, -1.17785157, -1.17784451, -1.17783745,\n",
+       "        -1.17783039, -1.17782333, -1.17781628, -1.17780922, -1.17780216,\n",
+       "        -1.17779511, -1.17778805, -1.177781  , -1.17777394, -1.17776688,\n",
+       "        -1.17775983, -1.17775277, -1.17774572, -1.17773866, -1.17773161,\n",
+       "        -1.17772456, -1.1777175 , -1.17771045, -1.17770339, -1.17769634,\n",
+       "        -1.17768929, -1.17768223, -1.17767518, -1.17766813, -1.17766108,\n",
+       "        -1.17765403, -1.17764697, -1.17763992, -1.17763287, -1.17762582,\n",
+       "        -1.17761877, -1.17761172, -1.17760467, -1.17759762, -1.17759057,\n",
+       "        -1.17758352, -1.17757647, -1.17756942, -1.17756237, -1.17755532,\n",
+       "        -1.17754827, -1.17754123, -1.17753418, -1.17752713, -1.17752008,\n",
+       "        -1.17751303, -1.17750599, -1.17749894, -1.17749189, -1.17748485,\n",
+       "        -1.1774778 , -1.17747075, -1.17746371, -1.17745666, -1.17744962,\n",
+       "        -1.17744257, -1.17743553, -1.17742848, -1.17742144, -1.17741439,\n",
+       "        -1.17740735, -1.17740031, -1.17739326, -1.17738622, -1.17737918,\n",
+       "        -1.17737213, -1.17736509, -1.17735805, -1.17735101, -1.17734396,\n",
+       "        -1.17733692, -1.17732988, -1.17732284, -1.1773158 , -1.17730876,\n",
+       "        -1.17730172, -1.17729468, -1.17728764, -1.1772806 , -1.17727356,\n",
+       "        -1.17726652, -1.17725948, -1.17725244, -1.1772454 , -1.17723836,\n",
+       "        -1.17723132, -1.17722428, -1.17721725, -1.17721021, -1.17720317,\n",
+       "        -1.17719613, -1.1771891 , -1.17718206, -1.17717502, -1.17716799,\n",
+       "        -1.17716095, -1.17715392, -1.17714688, -1.17713984, -1.17713281,\n",
+       "        -1.17712577, -1.17711874, -1.1771117 , -1.17710467, -1.17709764,\n",
+       "        -1.1770906 , -1.17708357, -1.17707654, -1.1770695 , -1.17706247,\n",
+       "        -1.17705544, -1.1770484 , -1.17704137, -1.17703434, -1.17702731,\n",
+       "        -1.17702028, -1.17701324, -1.17700621, -1.17699918, -1.17699215,\n",
+       "        -1.17698512, -1.17697809, -1.17697106, -1.17696403, -1.176957  ,\n",
+       "        -1.17694997, -1.17694294, -1.17693591, -1.17692889, -1.17692186,\n",
+       "        -1.17691483, -1.1769078 , -1.17690077, -1.17689375, -1.17688672,\n",
+       "        -1.17687969, -1.17687266, -1.17686564, -1.17685861, -1.17685158,\n",
+       "        -1.17684456, -1.17683753, -1.17683051, -1.17682348, -1.17681646,\n",
+       "        -1.17680943, -1.17680241, -1.17679538, -1.17678836, -1.17678134,\n",
+       "        -1.17677431, -1.17676729, -1.17676026, -1.17675324, -1.17674622,\n",
+       "        -1.1767392 , -1.17673217, -1.17672515, -1.17671813, -1.17671111,\n",
+       "        -1.17670409, -1.17669707, -1.17669004, -1.17668302, -1.176676  ,\n",
+       "        -1.17666898, -1.17666196, -1.17665494, -1.17664792, -1.1766409 ,\n",
+       "        -1.17663388, -1.17662687, -1.17661985, -1.17661283, -1.17660581,\n",
+       "        -1.17659879, -1.17659177, -1.17658476, -1.17657774, -1.17657072,\n",
+       "        -1.1765637 , -1.17655669, -1.17654967, -1.17654265, -1.17653564,\n",
+       "        -1.17652862, -1.17652161, -1.17651459, -1.17650758, -1.17650056,\n",
+       "        -1.17649355, -1.17648653, -1.17647952, -1.1764725 , -1.17646549,\n",
+       "        -1.17645848, -1.17645146, -1.17644445, -1.17643744, -1.17643042,\n",
+       "        -1.17642341, -1.1764164 , -1.17640939, -1.17640238, -1.17639536,\n",
+       "        -1.17638835, -1.17638134, -1.17637433, -1.17636732, -1.17636031,\n",
+       "        -1.1763533 , -1.17634629, -1.17633928, -1.17633227, -1.17632526,\n",
+       "        -1.17631825, -1.17631124, -1.17630423, -1.17629723, -1.17629022,\n",
+       "        -1.17628321, -1.1762762 , -1.17626919, -1.17626219, -1.17625518,\n",
+       "        -1.17624817, -1.17624117, -1.17623416, -1.17622715, -1.17622015,\n",
+       "        -1.17621314, -1.17620614, -1.17619913, -1.17619213, -1.17618512,\n",
+       "        -1.17617812, -1.17617111, -1.17616411, -1.1761571 , -1.1761501 ,\n",
+       "        -1.1761431 , -1.17613609, -1.17612909, -1.17612209, -1.17611508,\n",
+       "        -1.17610808, -1.17610108, -1.17609408, -1.17608708, -1.17608007,\n",
+       "        -1.17607307, -1.17606607, -1.17605907, -1.17605207, -1.17604507,\n",
+       "        -1.17603807, -1.17603107, -1.17602407, -1.17601707, -1.17601007,\n",
+       "        -1.17600307, -1.17599607, -1.17598908, -1.17598208, -1.17597508,\n",
+       "        -1.17596808, -1.17596108, -1.17595409, -1.17594709, -1.17594009,\n",
+       "        -1.1759331 , -1.1759261 , -1.1759191 , -1.17591211, -1.17590511,\n",
+       "        -1.17589811, -1.17589112, -1.17588412, -1.17587713, -1.17587013,\n",
+       "        -1.17586314, -1.17585615, -1.17584915, -1.17584216, -1.17583516,\n",
+       "        -1.17582817, -1.17582118, -1.17581418, -1.17580719, -1.1758002 ,\n",
+       "        -1.17579321, -1.17578621, -1.17577922, -1.17577223, -1.17576524,\n",
+       "        -1.17575825, -1.17575126, -1.17574427, -1.17573728, -1.17573029,\n",
+       "        -1.1757233 , -1.17571631, -1.17570932, -1.17570233, -1.17569534,\n",
+       "        -1.17568835, -1.17568136, -1.17567437, -1.17566738, -1.1756604 ,\n",
+       "        -1.17565341, -1.17564642, -1.17563943, -1.17563245, -1.17562546,\n",
+       "        -1.17561847, -1.17561149, -1.1756045 , -1.17559751, -1.17559053,\n",
+       "        -1.17558354, -1.17557656, -1.17556957, -1.17556259, -1.1755556 ,\n",
+       "        -1.17554862, -1.17554163, -1.17553465, -1.17552766, -1.17552068,\n",
+       "        -1.1755137 , -1.17550671, -1.17549973, -1.17549275, -1.17548577,\n",
+       "        -1.17547878, -1.1754718 , -1.17546482, -1.17545784, -1.17545086,\n",
+       "        -1.17544388, -1.1754369 , -1.17542991, -1.17542293, -1.17541595,\n",
+       "        -1.17540897, -1.17540199, -1.17539501, -1.17538804, -1.17538106,\n",
+       "        -1.17537408, -1.1753671 , -1.17536012, -1.17535314, -1.17534616,\n",
+       "        -1.17533919, -1.17533221, -1.17532523, -1.17531825, -1.17531128,\n",
+       "        -1.1753043 , -1.17529732, -1.17529035, -1.17528337, -1.1752764 ,\n",
+       "        -1.17526942, -1.17526245, -1.17525547, -1.1752485 , -1.17524152,\n",
+       "        -1.17523455, -1.17522757, -1.1752206 , -1.17521362, -1.17520665,\n",
+       "        -1.17519968, -1.1751927 , -1.17518573, -1.17517876, -1.17517179]),\n",
+       " 'split2_train_score': array([0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613045, 0.33613045, 0.33613045,\n",
+       "        0.33613045, 0.33613045, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613044, 0.33613044,\n",
+       "        0.33613044, 0.33613044, 0.33613044, 0.33613043, 0.33613043,\n",
+       "        0.33613043, 0.33613043, 0.33613043, 0.33613043, 0.33613043,\n",
+       "        0.33613043, 0.33613043, 0.33613043, 0.33613043, 0.33613043,\n",
+       "        0.33613043, 0.33613043, 0.33613043, 0.33613043, 0.33613043,\n",
+       "        0.33613043, 0.33613043, 0.33613043, 0.33613043, 0.33613042,\n",
+       "        0.33613042, 0.33613042, 0.33613042, 0.33613042, 0.33613042,\n",
+       "        0.33613042, 0.33613042, 0.33613042, 0.33613042, 0.33613042,\n",
+       "        0.33613042, 0.33613042, 0.33613042, 0.33613042, 0.33613042,\n",
+       "        0.33613041, 0.33613041, 0.33613041, 0.33613041, 0.33613041,\n",
+       "        0.33613041, 0.33613041, 0.33613041, 0.33613041, 0.33613041,\n",
+       "        0.33613041, 0.33613041, 0.33613041, 0.33613041, 0.33613041,\n",
+       "        0.3361304 , 0.3361304 , 0.3361304 , 0.3361304 , 0.3361304 ,\n",
+       "        0.3361304 , 0.3361304 , 0.3361304 , 0.3361304 , 0.3361304 ,\n",
+       "        0.3361304 , 0.3361304 , 0.33613039, 0.33613039, 0.33613039,\n",
+       "        0.33613039, 0.33613039, 0.33613039, 0.33613039, 0.33613039,\n",
+       "        0.33613039, 0.33613039, 0.33613039, 0.33613039, 0.33613038,\n",
+       "        0.33613038, 0.33613038, 0.33613038, 0.33613038, 0.33613038,\n",
+       "        0.33613038, 0.33613038, 0.33613038, 0.33613038, 0.33613038,\n",
+       "        0.33613037, 0.33613037, 0.33613037, 0.33613037, 0.33613037,\n",
+       "        0.33613037, 0.33613037, 0.33613037, 0.33613037, 0.33613037,\n",
+       "        0.33613036, 0.33613036, 0.33613036, 0.33613036, 0.33613036,\n",
+       "        0.33613036, 0.33613036, 0.33613036, 0.33613036, 0.33613036,\n",
+       "        0.33613035, 0.33613035, 0.33613035, 0.33613035, 0.33613035,\n",
+       "        0.33613035, 0.33613035, 0.33613035, 0.33613035, 0.33613034,\n",
+       "        0.33613034, 0.33613034, 0.33613034, 0.33613034, 0.33613034,\n",
+       "        0.33613034, 0.33613034, 0.33613033, 0.33613033, 0.33613033,\n",
+       "        0.33613033, 0.33613033, 0.33613033, 0.33613033, 0.33613033,\n",
+       "        0.33613032, 0.33613032, 0.33613032, 0.33613032, 0.33613032,\n",
+       "        0.33613032, 0.33613032, 0.33613032, 0.33613031, 0.33613031,\n",
+       "        0.33613031, 0.33613031, 0.33613031, 0.33613031, 0.33613031,\n",
+       "        0.33613031, 0.3361303 , 0.3361303 , 0.3361303 , 0.3361303 ,\n",
+       "        0.3361303 , 0.3361303 , 0.3361303 , 0.33613029, 0.33613029,\n",
+       "        0.33613029, 0.33613029, 0.33613029, 0.33613029, 0.33613029,\n",
+       "        0.33613028, 0.33613028, 0.33613028, 0.33613028, 0.33613028,\n",
+       "        0.33613028, 0.33613028, 0.33613027, 0.33613027, 0.33613027,\n",
+       "        0.33613027, 0.33613027, 0.33613027, 0.33613027, 0.33613026,\n",
+       "        0.33613026, 0.33613026, 0.33613026, 0.33613026, 0.33613026,\n",
+       "        0.33613026, 0.33613025, 0.33613025, 0.33613025, 0.33613025,\n",
+       "        0.33613025, 0.33613025, 0.33613024, 0.33613024, 0.33613024,\n",
+       "        0.33613024, 0.33613024, 0.33613024, 0.33613023, 0.33613023,\n",
+       "        0.33613023, 0.33613023, 0.33613023, 0.33613023, 0.33613022,\n",
+       "        0.33613022, 0.33613022, 0.33613022, 0.33613022, 0.33613022,\n",
+       "        0.33613021, 0.33613021, 0.33613021, 0.33613021, 0.33613021,\n",
+       "        0.33613021, 0.3361302 , 0.3361302 , 0.3361302 , 0.3361302 ,\n",
+       "        0.3361302 , 0.3361302 , 0.33613019, 0.33613019, 0.33613019,\n",
+       "        0.33613019, 0.33613019, 0.33613018, 0.33613018, 0.33613018,\n",
+       "        0.33613018, 0.33613018, 0.33613018, 0.33613017, 0.33613017,\n",
+       "        0.33613017, 0.33613017, 0.33613017, 0.33613016, 0.33613016,\n",
+       "        0.33613016, 0.33613016, 0.33613016, 0.33613016, 0.33613015,\n",
+       "        0.33613015, 0.33613015, 0.33613015, 0.33613015, 0.33613014,\n",
+       "        0.33613014, 0.33613014, 0.33613014, 0.33613014, 0.33613013,\n",
+       "        0.33613013, 0.33613013, 0.33613013, 0.33613013, 0.33613012,\n",
+       "        0.33613012, 0.33613012, 0.33613012, 0.33613012, 0.33613011,\n",
+       "        0.33613011, 0.33613011, 0.33613011, 0.33613011, 0.3361301 ,\n",
+       "        0.3361301 , 0.3361301 , 0.3361301 , 0.3361301 , 0.33613009,\n",
+       "        0.33613009, 0.33613009, 0.33613009, 0.33613008, 0.33613008,\n",
+       "        0.33613008, 0.33613008, 0.33613008, 0.33613007, 0.33613007,\n",
+       "        0.33613007, 0.33613007, 0.33613007, 0.33613006, 0.33613006,\n",
+       "        0.33613006, 0.33613006, 0.33613005, 0.33613005, 0.33613005,\n",
+       "        0.33613005, 0.33613005, 0.33613004, 0.33613004, 0.33613004,\n",
+       "        0.33613004, 0.33613003, 0.33613003, 0.33613003, 0.33613003,\n",
+       "        0.33613003, 0.33613002, 0.33613002, 0.33613002, 0.33613002,\n",
+       "        0.33613001, 0.33613001, 0.33613001, 0.33613001, 0.33613   ,\n",
+       "        0.33613   , 0.33613   , 0.33613   , 0.33613   , 0.33612999,\n",
+       "        0.33612999, 0.33612999, 0.33612999, 0.33612998, 0.33612998,\n",
+       "        0.33612998, 0.33612998, 0.33612997, 0.33612997, 0.33612997,\n",
+       "        0.33612997, 0.33612996, 0.33612996, 0.33612996, 0.33612996,\n",
+       "        0.33612995, 0.33612995, 0.33612995, 0.33612995, 0.33612994,\n",
+       "        0.33612994, 0.33612994, 0.33612994, 0.33612993, 0.33612993,\n",
+       "        0.33612993, 0.33612993, 0.33612992, 0.33612992, 0.33612992,\n",
+       "        0.33612992, 0.33612991, 0.33612991, 0.33612991, 0.33612991,\n",
+       "        0.3361299 , 0.3361299 , 0.3361299 , 0.3361299 , 0.33612989,\n",
+       "        0.33612989, 0.33612989, 0.33612989, 0.33612988, 0.33612988,\n",
+       "        0.33612988, 0.33612987, 0.33612987, 0.33612987, 0.33612987,\n",
+       "        0.33612986, 0.33612986, 0.33612986, 0.33612986, 0.33612985,\n",
+       "        0.33612985, 0.33612985, 0.33612985, 0.33612984, 0.33612984,\n",
+       "        0.33612984, 0.33612983, 0.33612983, 0.33612983, 0.33612983,\n",
+       "        0.33612982, 0.33612982, 0.33612982, 0.33612982, 0.33612981,\n",
+       "        0.33612981, 0.33612981, 0.3361298 , 0.3361298 , 0.3361298 ,\n",
+       "        0.3361298 , 0.33612979, 0.33612979, 0.33612979, 0.33612978,\n",
+       "        0.33612978, 0.33612978, 0.33612978, 0.33612977, 0.33612977,\n",
+       "        0.33612977, 0.33612976, 0.33612976, 0.33612976, 0.33612976,\n",
+       "        0.33612975, 0.33612975, 0.33612975, 0.33612974, 0.33612974,\n",
+       "        0.33612974, 0.33612973, 0.33612973, 0.33612973, 0.33612973,\n",
+       "        0.33612972, 0.33612972, 0.33612972, 0.33612971, 0.33612971,\n",
+       "        0.33612971, 0.3361297 , 0.3361297 , 0.3361297 , 0.3361297 ,\n",
+       "        0.33612969, 0.33612969, 0.33612969, 0.33612968, 0.33612968]),\n",
+       " 'std_fit_time': array([5.55963272e-04, 2.46042737e-04, 3.17509183e-04, 3.64186701e-04,\n",
+       "        1.51571266e-04, 4.83683354e-04, 2.71353525e-03, 2.16780569e-03,\n",
+       "        2.49788081e-03, 4.81798233e-04, 2.43217881e-04, 1.74737994e-04,\n",
+       "        1.06882694e-04, 1.82960677e-04, 2.68991482e-04, 7.24740194e-05,\n",
+       "        1.89952072e-04, 6.16624392e-04, 1.39431594e-04, 1.38174806e-04,\n",
+       "        1.76988325e-04, 1.01840183e-04, 1.03278092e-04, 6.87712595e-05,\n",
+       "        9.66232453e-05, 8.90398967e-05, 4.21131175e-06, 2.76788420e-04,\n",
+       "        3.05187019e-04, 1.61458068e-04, 5.81170542e-04, 4.76812773e-04,\n",
+       "        5.13134057e-04, 3.37445992e-04, 4.17778490e-04, 1.34613841e-04,\n",
+       "        3.51270453e-04, 9.62977596e-05, 3.23025589e-04, 2.43720573e-04,\n",
+       "        9.51749794e-05, 2.56531399e-04, 1.68498506e-04, 1.58267681e-04,\n",
+       "        9.49249340e-04, 4.88454497e-04, 2.44930119e-04, 6.40425273e-04,\n",
+       "        1.24444969e-03, 8.17359821e-04, 2.84541851e-04, 4.83603629e-04,\n",
+       "        3.51682697e-04, 4.62701159e-04, 1.16375026e-04, 1.85877318e-04,\n",
+       "        1.34835678e-04, 8.87145527e-04, 3.12602693e-04, 5.37931857e-04,\n",
+       "        4.65051042e-04, 3.30312591e-04, 2.98657273e-05, 1.17289718e-04,\n",
+       "        1.23422758e-04, 1.96700481e-04, 2.13237960e-04, 3.96867012e-04,\n",
+       "        1.84566405e-04, 1.07245319e-04, 9.00709785e-05, 1.28799935e-04,\n",
+       "        3.96955805e-04, 1.93315081e-04, 4.13634041e-04, 1.54861631e-04,\n",
+       "        6.55104795e-04, 2.02414519e-04, 8.36322994e-04, 1.23782851e-04,\n",
+       "        5.55460588e-04, 4.30649619e-04, 1.27164672e-04, 6.61403716e-04,\n",
+       "        1.46534980e-04, 1.39680780e-04, 5.15781921e-05, 5.14838165e-05,\n",
+       "        2.20747738e-04, 1.02712453e-04, 1.78165638e-04, 3.09677090e-05,\n",
+       "        1.02857163e-04, 1.51682359e-04, 3.14147660e-04, 7.62843417e-05,\n",
+       "        7.75954705e-05, 4.09001326e-04, 6.97299961e-05, 3.11140681e-04,\n",
+       "        3.37615020e-04, 2.81476287e-04, 1.94647805e-04, 1.53108089e-04,\n",
+       "        8.07891557e-06, 6.64920423e-04, 1.17652537e-03, 3.76063137e-04,\n",
+       "        1.54753077e-03, 8.62747958e-04, 5.74503921e-04, 2.32474653e-03,\n",
+       "        3.44033138e-03, 2.64164287e-03, 2.28627342e-03, 3.64111010e-04,\n",
+       "        6.63977551e-05, 2.13586974e-04, 2.25474489e-04, 5.94001345e-04,\n",
+       "        1.46338402e-03, 1.54801707e-04, 2.84410237e-04, 1.47415493e-04,\n",
+       "        1.55552852e-04, 1.05678406e-04, 1.55160170e-04, 1.51321667e-04,\n",
+       "        3.18698197e-04, 8.12046198e-05, 3.35645567e-04, 8.84770316e-05,\n",
+       "        1.11618153e-04, 5.55886380e-05, 3.72304690e-05, 8.45684588e-05,\n",
+       "        2.46582474e-04, 3.55713933e-04, 2.94584934e-04, 1.00985397e-04,\n",
+       "        2.10654924e-04, 2.70196620e-04, 8.75417268e-05, 4.49682975e-05,\n",
+       "        8.05127928e-05, 2.32263508e-04, 1.64581059e-04, 1.64363900e-04,\n",
+       "        3.33757418e-04, 5.69191035e-05, 2.88235167e-04, 1.66266605e-04,\n",
+       "        5.63963390e-04, 5.39973221e-04, 3.88400883e-04, 7.48879803e-05,\n",
+       "        7.78007231e-05, 1.00959376e-04, 1.25394725e-04, 2.33280447e-04,\n",
+       "        2.87827834e-05, 9.85251296e-05, 4.43989468e-04, 9.68836419e-05,\n",
+       "        1.93203344e-04, 3.74768273e-04, 3.88655388e-04, 1.20677019e-04,\n",
+       "        3.01555763e-04, 2.42744097e-04, 3.47700419e-04, 9.70857186e-05,\n",
+       "        3.46054655e-04, 2.57940256e-04, 3.00897744e-04, 1.55442616e-04,\n",
+       "        2.78834988e-04, 2.75127809e-04, 2.25797458e-04, 7.60756422e-04,\n",
+       "        1.56655631e-03, 1.15820135e-03, 6.28950772e-04, 1.10086152e-03,\n",
+       "        1.06914700e-03, 1.18275996e-03, 3.11173199e-04, 6.59648673e-05,\n",
+       "        1.48859137e-04, 5.67046512e-05, 2.69688711e-04, 2.13187365e-04,\n",
+       "        1.49712862e-04, 1.60724226e-04, 1.30294584e-04, 1.17804569e-04,\n",
+       "        6.48927985e-04, 7.60202665e-05, 2.92541399e-04, 1.75579956e-04,\n",
+       "        1.45895851e-04, 1.47047170e-04, 9.72282152e-05, 4.15713694e-04,\n",
+       "        1.53970580e-04, 3.57669051e-04, 1.58887256e-04, 3.24114837e-05,\n",
+       "        2.41551792e-04, 1.03400085e-04, 8.70090268e-05, 1.41608906e-04,\n",
+       "        1.00279956e-03, 4.65603806e-05, 7.38980513e-05, 3.14751797e-04,\n",
+       "        1.10261999e-04, 1.31290848e-04, 5.14449337e-04, 1.47997884e-04,\n",
+       "        1.91452082e-04, 9.21426116e-04, 1.37827157e-04, 2.71533351e-04,\n",
+       "        1.00842009e-04, 3.74608321e-04, 1.17165584e-04, 6.19386812e-05,\n",
+       "        6.40423054e-05, 3.73269750e-04, 1.08022357e-04, 6.09518215e-04,\n",
+       "        9.09879165e-04, 2.37833125e-04, 2.80420951e-04, 2.45585050e-04,\n",
+       "        9.03609560e-05, 2.81240426e-04, 3.23614753e-04, 4.67001661e-04,\n",
+       "        3.39440874e-04, 4.88660552e-04, 2.00681135e-04, 6.55455907e-04,\n",
+       "        2.63195562e-04, 4.18468612e-04, 4.30064603e-04, 1.70792186e-04,\n",
+       "        4.07233559e-04, 7.21115376e-05, 1.70946066e-04, 1.48179827e-04,\n",
+       "        3.25913130e-04, 4.87792074e-04, 5.62237125e-04, 2.98087071e-04,\n",
+       "        8.49950995e-05, 1.18982152e-04, 7.58528683e-04, 1.81019985e-04,\n",
+       "        6.07804649e-04, 9.52101444e-05, 1.32850480e-05, 1.81200107e-04,\n",
+       "        8.22383985e-05, 2.20811218e-04, 1.53353916e-04, 8.84120473e-05,\n",
+       "        1.85228000e-04, 3.85049081e-05, 6.36996660e-05, 1.38730251e-04,\n",
+       "        1.77147447e-04, 3.59426497e-05, 6.08577401e-06, 3.63290461e-04,\n",
+       "        2.11199839e-04, 1.70036228e-04, 6.17482129e-05, 4.27794468e-04,\n",
+       "        8.33315284e-05, 6.89235449e-05, 1.34460610e-04, 2.12426301e-04,\n",
+       "        8.76790605e-05, 1.99327185e-04, 2.71952828e-05, 1.71013370e-04,\n",
+       "        4.08025024e-04, 1.56908168e-04, 1.79331807e-04, 1.71057240e-04,\n",
+       "        5.97352610e-05, 1.09505180e-04, 2.25054202e-04, 9.28059951e-05,\n",
+       "        2.34015524e-04, 1.82794073e-04, 8.65205077e-05, 1.97538486e-04,\n",
+       "        8.56014564e-05, 7.56377071e-05, 2.27411029e-04, 8.17607727e-05,\n",
+       "        3.08713569e-04, 1.54664190e-04, 1.45161785e-04, 1.39701398e-04,\n",
+       "        1.38910875e-04, 2.88181718e-04, 1.82296155e-04, 1.50239316e-04,\n",
+       "        2.43098917e-04, 1.91162379e-04, 1.36745197e-04, 1.28700254e-04,\n",
+       "        1.11049040e-04, 1.43461742e-04, 1.00128297e-04, 1.93111429e-04,\n",
+       "        2.44339950e-05, 9.97567065e-05, 8.21159644e-05, 8.16142546e-05,\n",
+       "        1.20763345e-04, 1.22890279e-04, 1.42133516e-04, 1.91220652e-04,\n",
+       "        3.86004727e-04, 1.78060250e-04, 4.09483476e-05, 1.13005768e-04,\n",
+       "        1.76724733e-04, 4.91029236e-04, 5.16517343e-05, 2.05320436e-04,\n",
+       "        1.03904253e-04, 2.86421664e-04, 2.29687261e-04, 8.54176130e-06,\n",
+       "        1.42525174e-04, 1.83797170e-04, 1.18055963e-04, 2.84153807e-04,\n",
+       "        1.19290377e-04, 1.18013477e-04, 3.51403069e-05, 1.32188906e-04,\n",
+       "        1.54243165e-04, 1.61809045e-04, 8.66749094e-05, 4.82285408e-04,\n",
+       "        1.48074936e-04, 1.83954968e-04, 7.02080533e-05, 1.58462226e-04,\n",
+       "        1.02341358e-04, 9.81567357e-05, 2.99617782e-04, 6.86802780e-05,\n",
+       "        2.64498190e-04, 9.85193600e-05, 1.07285241e-04, 1.88175991e-04,\n",
+       "        1.27756225e-04, 1.99060081e-05, 6.01626860e-04, 4.94208890e-04,\n",
+       "        5.86365387e-05, 1.18371777e-04, 1.21151728e-04, 1.08367182e-04,\n",
+       "        7.67078201e-05, 7.03266108e-05, 1.18361532e-04, 7.20863961e-05,\n",
+       "        9.74291235e-05, 1.42614218e-04, 1.13199541e-04, 9.25367864e-05,\n",
+       "        5.66364560e-04, 3.98536796e-04, 8.83360051e-05, 1.59962620e-04,\n",
+       "        1.39473895e-04, 9.79586092e-05, 4.17227215e-05, 9.09052104e-05,\n",
+       "        1.26034986e-04, 3.20483370e-04, 7.71487877e-05, 9.39152005e-05,\n",
+       "        1.61140978e-04, 4.35830264e-04, 3.50653909e-04, 1.24138184e-04,\n",
+       "        9.59938063e-05, 1.17674162e-04, 1.76597636e-04, 1.40248816e-04,\n",
+       "        2.41971798e-04, 1.67424966e-04, 7.59676572e-05, 2.08051356e-04,\n",
+       "        1.13852853e-04, 1.29337336e-04, 1.29777014e-05, 1.60912269e-04,\n",
+       "        1.15436839e-04, 1.42454253e-04, 4.16375476e-04, 2.82277484e-05,\n",
+       "        3.93574315e-04, 3.50739274e-04, 1.65842626e-04, 4.01340884e-04,\n",
+       "        1.53810394e-04, 4.93451242e-04, 5.01772063e-04, 5.46846724e-04,\n",
+       "        6.86263748e-04, 2.60296316e-04, 3.54591424e-04, 5.55395567e-04,\n",
+       "        2.02704966e-04, 3.27970521e-04, 3.03884180e-04, 2.55533880e-04,\n",
+       "        3.33332556e-04, 1.10930505e-04, 3.11277002e-04, 1.48829009e-04,\n",
+       "        3.10841285e-04, 3.67602999e-05, 6.07932574e-05, 8.40281094e-05,\n",
+       "        1.46495494e-04, 1.07308727e-04, 2.23202947e-04, 1.20305268e-04,\n",
+       "        3.39603199e-04, 7.52015688e-05, 5.97915897e-05, 5.98644697e-04,\n",
+       "        3.49823243e-05, 8.90045647e-05, 8.19647645e-05, 7.12054648e-05,\n",
+       "        5.37768959e-04, 9.63203847e-05, 1.23999209e-04, 9.37637625e-05,\n",
+       "        1.30708092e-05, 2.35721673e-04, 2.21588604e-04, 3.56937395e-04,\n",
+       "        6.17474969e-05, 5.56778710e-05, 1.01276179e-04, 3.02684632e-04,\n",
+       "        1.07055576e-04, 3.93640234e-05, 2.02791318e-05, 3.77707673e-05,\n",
+       "        4.07336218e-04, 7.40517343e-05, 2.08540751e-04, 2.29321743e-04,\n",
+       "        3.96694573e-05, 1.63751084e-04, 1.01954046e-04, 1.47862753e-04,\n",
+       "        2.29411043e-05, 3.33856185e-04, 1.49950440e-04, 7.76079230e-05,\n",
+       "        2.19555163e-04, 8.40894213e-05, 1.36456590e-04, 9.22046845e-05,\n",
+       "        1.48652959e-04, 2.30849602e-04, 1.34147409e-04, 1.19514234e-04,\n",
+       "        2.56927484e-04, 2.88137706e-04, 8.12818167e-05, 1.00658456e-04,\n",
+       "        3.23531659e-05, 1.11145174e-04, 5.16843725e-05, 5.58280860e-05,\n",
+       "        1.08075083e-04, 1.17645229e-04, 1.54491601e-04, 6.02354032e-04,\n",
+       "        6.11341067e-05, 4.73043130e-05, 1.18461382e-04, 1.13332255e-04]),\n",
+       " 'std_score_time': array([1.60080396e-04, 2.12944588e-04, 6.75472557e-06, 1.11102262e-04,\n",
+       "        1.09229772e-04, 4.83776031e-04, 2.67171467e-04, 3.26921227e-04,\n",
+       "        3.74141089e-04, 3.13354119e-04, 1.09668918e-04, 8.60820580e-05,\n",
+       "        6.73917667e-05, 3.23490661e-05, 2.04299482e-04, 2.28354168e-05,\n",
+       "        3.37112967e-05, 1.83906139e-04, 1.38734212e-05, 3.19138703e-05,\n",
+       "        1.18326468e-04, 1.53112668e-05, 2.24859044e-05, 6.11889412e-06,\n",
+       "        9.80387196e-06, 7.95364126e-06, 1.61890721e-05, 3.33975178e-06,\n",
+       "        1.09420535e-04, 6.48511196e-05, 3.31126319e-05, 4.58079523e-05,\n",
+       "        1.15602447e-04, 5.93039025e-05, 2.45562135e-05, 2.41837884e-06,\n",
+       "        1.70890673e-05, 2.89449276e-05, 6.45713928e-05, 1.55202173e-05,\n",
+       "        5.08129354e-06, 1.71499411e-05, 4.75549283e-05, 9.16043220e-06,\n",
+       "        4.97676536e-05, 2.95714464e-04, 3.13453754e-04, 6.36589582e-04,\n",
+       "        1.16522716e-04, 7.27243108e-05, 4.10880572e-04, 7.58370146e-04,\n",
+       "        9.50655530e-05, 4.88648881e-05, 2.01666981e-05, 3.85035958e-05,\n",
+       "        1.99819362e-04, 2.32270143e-04, 2.19287496e-05, 1.45477061e-04,\n",
+       "        2.07583444e-04, 1.36688698e-05, 1.86165373e-04, 1.18768726e-05,\n",
+       "        7.82284263e-05, 5.45856489e-05, 3.17095714e-06, 1.14132102e-04,\n",
+       "        9.66763738e-06, 1.73024663e-06, 7.41103060e-06, 4.91143559e-05,\n",
+       "        1.45428819e-05, 1.48680106e-05, 2.93342066e-05, 1.03998307e-04,\n",
+       "        1.37763446e-04, 3.51182231e-04, 5.18505446e-05, 1.47541530e-05,\n",
+       "        1.10589447e-04, 2.17648176e-04, 7.94410642e-06, 8.89837704e-05,\n",
+       "        2.76045566e-04, 1.10261999e-04, 9.87943866e-05, 8.43036618e-05,\n",
+       "        1.30826328e-04, 3.86568276e-06, 8.87893609e-06, 1.64557878e-04,\n",
+       "        4.75416451e-05, 5.18830577e-06, 3.21657993e-05, 3.09101410e-05,\n",
+       "        3.00716312e-05, 9.32288797e-05, 5.24835059e-05, 1.65215646e-05,\n",
+       "        8.69724340e-05, 6.77560080e-05, 1.02361475e-04, 7.53231673e-05,\n",
+       "        3.18018590e-04, 2.15964714e-04, 3.70379475e-04, 1.22599619e-03,\n",
+       "        3.08932156e-04, 1.75104051e-04, 5.16474996e-04, 2.79763950e-04,\n",
+       "        9.43439044e-04, 5.32944653e-04, 6.20369755e-04, 4.05414419e-05,\n",
+       "        1.29346614e-04, 7.41913092e-05, 3.56844645e-05, 5.48004411e-05,\n",
+       "        1.00747453e-04, 8.50929087e-05, 1.50842209e-04, 3.46663857e-05,\n",
+       "        2.26955898e-05, 1.18896285e-05, 2.04739085e-04, 8.47492256e-05,\n",
+       "        1.07574493e-04, 2.42226706e-05, 2.33338677e-04, 1.64301637e-04,\n",
+       "        6.97200320e-05, 5.08129354e-06, 1.27652315e-06, 6.95466991e-06,\n",
+       "        1.78092597e-04, 2.05283181e-05, 3.51656404e-05, 9.05223432e-06,\n",
+       "        3.83502458e-05, 8.75188530e-05, 8.96978591e-05, 1.46109075e-05,\n",
+       "        7.76478716e-06, 6.57562879e-06, 9.79132079e-05, 9.08705342e-06,\n",
+       "        3.05084705e-05, 1.78483375e-05, 2.44513413e-04, 2.28318209e-05,\n",
+       "        1.43698578e-04, 7.33726049e-05, 8.96107153e-06, 8.88121919e-05,\n",
+       "        2.42784058e-05, 2.36022352e-06, 1.18795312e-05, 1.97077822e-04,\n",
+       "        1.24237255e-04, 1.79826554e-06, 2.49734126e-04, 1.75163339e-04,\n",
+       "        2.31913842e-04, 8.78887190e-05, 5.07170113e-05, 2.37880205e-05,\n",
+       "        6.45024956e-05, 5.91231074e-05, 1.56870849e-04, 3.04330854e-04,\n",
+       "        1.76344706e-04, 4.79098099e-05, 1.60793649e-04, 7.38328958e-05,\n",
+       "        2.82666540e-05, 1.38075214e-04, 8.37572307e-05, 4.75334077e-05,\n",
+       "        1.12431706e-03, 2.08757489e-04, 8.07132928e-04, 4.24629729e-04,\n",
+       "        7.32978706e-04, 1.80322780e-04, 3.74491986e-04, 2.52253692e-04,\n",
+       "        9.71664837e-05, 2.97544943e-05, 3.70892267e-06, 1.62700180e-05,\n",
+       "        8.70775242e-05, 2.98904599e-05, 8.22534500e-06, 4.08080560e-05,\n",
+       "        2.15932308e-05, 1.10032061e-04, 9.42717634e-05, 5.10772508e-05,\n",
+       "        3.48854172e-05, 1.93558430e-05, 9.48617030e-05, 3.87329765e-04,\n",
+       "        8.05072229e-05, 4.04417698e-05, 9.01167563e-06, 1.44258937e-04,\n",
+       "        1.26554135e-05, 1.17936385e-04, 3.83156451e-05, 1.00260107e-04,\n",
+       "        2.21531220e-05, 4.64381348e-05, 6.03587867e-05, 3.65649437e-05,\n",
+       "        1.27240987e-05, 1.06291793e-05, 3.25159590e-06, 2.53415286e-04,\n",
+       "        5.33948804e-06, 4.53515061e-05, 1.99554436e-05, 1.71001514e-05,\n",
+       "        6.70034921e-05, 2.31125331e-05, 1.26480301e-04, 5.59592779e-06,\n",
+       "        6.54095915e-06, 3.45499590e-04, 8.98502452e-05, 1.39758577e-04,\n",
+       "        1.03670206e-04, 1.99274230e-04, 1.12840435e-04, 1.25357900e-04,\n",
+       "        7.21103114e-05, 2.38780251e-04, 1.37894595e-04, 9.15234103e-05,\n",
+       "        1.83119990e-04, 1.21782090e-04, 2.00239941e-04, 3.47390042e-05,\n",
+       "        1.11535791e-04, 4.78837005e-05, 6.55060801e-06, 1.95387901e-04,\n",
+       "        7.35388244e-04, 1.37176740e-04, 1.95961444e-06, 7.84259022e-05,\n",
+       "        1.97598906e-04, 1.47429374e-04, 1.40506892e-04, 9.01815627e-05,\n",
+       "        1.79541837e-05, 6.84390073e-06, 5.09349736e-05, 2.06846343e-05,\n",
+       "        2.47251626e-04, 2.17134048e-05, 6.36040119e-05, 2.08975845e-05,\n",
+       "        1.00954121e-04, 1.71695407e-04, 1.10812642e-05, 1.21817352e-04,\n",
+       "        1.24926330e-04, 5.47402458e-05, 1.26625631e-04, 1.20481117e-04,\n",
+       "        1.72737547e-04, 1.76647985e-05, 8.90876224e-06, 5.22526649e-05,\n",
+       "        4.37073273e-05, 1.22506840e-05, 8.32519837e-05, 1.13292736e-05,\n",
+       "        1.65837599e-05, 4.54891708e-05, 2.75524508e-05, 4.90603156e-05,\n",
+       "        1.01824553e-05, 6.38203186e-05, 5.68858047e-05, 2.27772599e-05,\n",
+       "        1.77032605e-04, 2.24577984e-05, 1.57396395e-05, 1.54990415e-05,\n",
+       "        1.77563107e-04, 8.77518574e-06, 7.90712221e-05, 8.03170665e-05,\n",
+       "        1.46052868e-05, 4.77222446e-05, 3.32149019e-05, 7.88585444e-06,\n",
+       "        1.27716619e-05, 1.66464816e-05, 4.17381593e-05, 3.01773005e-05,\n",
+       "        9.51486303e-06, 1.00464760e-04, 8.87386991e-05, 1.07514767e-05,\n",
+       "        1.91372890e-05, 2.46578555e-05, 1.81681980e-05, 3.19878013e-05,\n",
+       "        1.13028233e-04, 1.58113327e-04, 1.00358143e-04, 1.18938775e-05,\n",
+       "        8.28883258e-06, 8.95241923e-05, 1.19370405e-04, 2.57258938e-05,\n",
+       "        7.87703940e-06, 7.80211418e-06, 3.12361703e-05, 1.54025701e-05,\n",
+       "        1.45550371e-05, 7.02080533e-05, 1.84486620e-05, 3.42932526e-06,\n",
+       "        7.65976556e-05, 1.31498167e-05, 5.63192931e-06, 1.06321499e-05,\n",
+       "        2.07045119e-04, 4.44015088e-05, 2.24274043e-05, 6.80773998e-05,\n",
+       "        1.11879046e-05, 4.11870348e-05, 4.98431069e-05, 2.77447937e-05,\n",
+       "        1.93460513e-05, 8.07422352e-06, 3.74113792e-06, 2.75552014e-05,\n",
+       "        1.00532382e-04, 5.39947756e-06, 1.98401603e-04, 1.68501205e-05,\n",
+       "        2.21776274e-05, 1.71415166e-04, 6.84562625e-05, 1.35677047e-04,\n",
+       "        8.37051833e-05, 8.94696405e-06, 9.73339773e-06, 1.69376037e-05,\n",
+       "        8.59097570e-05, 6.51871257e-06, 2.13038293e-04, 2.60963992e-05,\n",
+       "        8.79163100e-05, 9.97249815e-06, 1.01830756e-05, 1.10427417e-04,\n",
+       "        2.73051445e-05, 3.63278778e-05, 1.48155274e-04, 5.79248666e-05,\n",
+       "        3.23139032e-05, 1.22124679e-04, 4.34750684e-05, 7.88951379e-05,\n",
+       "        4.50154650e-05, 1.44257273e-04, 2.68234735e-05, 6.94640985e-05,\n",
+       "        7.81586383e-06, 6.72461564e-05, 1.12234137e-05, 1.34551096e-05,\n",
+       "        1.35664199e-04, 2.12286275e-05, 1.12137022e-04, 1.29884574e-04,\n",
+       "        2.40420835e-05, 8.72972376e-06, 2.55470325e-05, 9.85251296e-05,\n",
+       "        9.26327689e-06, 1.82240816e-05, 2.47437699e-05, 1.60552133e-04,\n",
+       "        3.08608577e-05, 2.53786097e-05, 8.93108770e-05, 1.11935485e-05,\n",
+       "        1.38342139e-05, 4.52787508e-05, 1.99402457e-05, 5.53354861e-05,\n",
+       "        1.03354385e-04, 2.85713497e-05, 5.73940426e-05, 1.05832250e-04,\n",
+       "        2.13544032e-06, 6.77992463e-06, 1.27652315e-06, 6.64822654e-06,\n",
+       "        1.32795843e-04, 3.87917115e-05, 1.06641266e-04, 2.75696379e-05,\n",
+       "        1.09025236e-04, 3.98656698e-04, 1.53818606e-04, 1.25262337e-04,\n",
+       "        1.94451006e-04, 1.83568922e-04, 1.75711455e-04, 3.84459764e-05,\n",
+       "        7.41103060e-05, 1.26350800e-04, 1.11666184e-04, 9.86012567e-05,\n",
+       "        3.37079242e-05, 7.92301764e-05, 1.80909313e-04, 5.22139712e-05,\n",
+       "        7.36990698e-05, 2.50678615e-05, 9.40873410e-05, 2.34914568e-05,\n",
+       "        4.69109381e-05, 1.27864891e-05, 2.32188634e-05, 1.17437709e-04,\n",
+       "        1.79826554e-06, 4.15812452e-05, 4.76385273e-05, 2.09142007e-05,\n",
+       "        1.29202096e-04, 4.99657441e-05, 1.50146928e-05, 1.57837941e-04,\n",
+       "        5.83862172e-05, 6.98546998e-05, 2.34106596e-05, 5.05012349e-06,\n",
+       "        4.46604946e-05, 1.03394160e-05, 9.12062476e-05, 4.94906027e-06,\n",
+       "        1.06960963e-04, 5.18091125e-05, 6.23773358e-05, 1.20723381e-04,\n",
+       "        5.56423539e-06, 1.96194622e-04, 1.41016253e-04, 8.64539067e-06,\n",
+       "        1.45103949e-04, 1.05558388e-05, 1.64886554e-05, 2.17590739e-04,\n",
+       "        3.92439846e-05, 1.70346511e-05, 2.30125217e-04, 4.10672494e-05,\n",
+       "        5.91900500e-05, 6.34490129e-06, 1.04874067e-05, 6.90396426e-05,\n",
+       "        2.60537974e-04, 9.72690014e-05, 1.48141205e-04, 7.73897551e-05,\n",
+       "        1.80615078e-05, 9.66418077e-05, 1.45454874e-05, 1.06964033e-04,\n",
+       "        3.86568276e-06, 2.06130599e-05, 3.34943928e-05, 8.51294191e-05,\n",
+       "        2.07364779e-05, 2.23269236e-05, 5.06901050e-05, 1.13882140e-05,\n",
+       "        3.48245319e-05, 1.72355362e-05, 3.33995980e-05, 1.61890721e-05,\n",
+       "        7.37771859e-06, 1.49201696e-05, 6.08916674e-05, 2.81161006e-05,\n",
+       "        1.19722111e-05, 8.67442532e-05, 2.37888170e-06, 7.19593283e-05]),\n",
+       " 'std_test_score': array([0.5127766 , 0.51277333, 0.51277006, 0.5127668 , 0.51276353,\n",
+       "        0.51276026, 0.51275699, 0.51275373, 0.51275046, 0.51274719,\n",
+       "        0.51274393, 0.51274066, 0.51273739, 0.51273413, 0.51273086,\n",
+       "        0.51272759, 0.51272433, 0.51272106, 0.51271779, 0.51271453,\n",
+       "        0.51271126, 0.512708  , 0.51270473, 0.51270146, 0.5126982 ,\n",
+       "        0.51269493, 0.51269167, 0.5126884 , 0.51268514, 0.51268187,\n",
+       "        0.51267861, 0.51267534, 0.51267208, 0.51266881, 0.51266555,\n",
+       "        0.51266229, 0.51265902, 0.51265576, 0.51265249, 0.51264923,\n",
+       "        0.51264597, 0.5126427 , 0.51263944, 0.51263618, 0.51263291,\n",
+       "        0.51262965, 0.51262639, 0.51262312, 0.51261986, 0.5126166 ,\n",
+       "        0.51261333, 0.51261007, 0.51260681, 0.51260355, 0.51260029,\n",
+       "        0.51259702, 0.51259376, 0.5125905 , 0.51258724, 0.51258398,\n",
+       "        0.51258071, 0.51257745, 0.51257419, 0.51257093, 0.51256767,\n",
+       "        0.51256441, 0.51256115, 0.51255789, 0.51255462, 0.51255136,\n",
+       "        0.5125481 , 0.51254484, 0.51254158, 0.51253832, 0.51253506,\n",
+       "        0.5125318 , 0.51252854, 0.51252528, 0.51252202, 0.51251876,\n",
+       "        0.5125155 , 0.51251224, 0.51250899, 0.51250573, 0.51250247,\n",
+       "        0.51249921, 0.51249595, 0.51249269, 0.51248943, 0.51248617,\n",
+       "        0.51248291, 0.51247966, 0.5124764 , 0.51247314, 0.51246988,\n",
+       "        0.51246662, 0.51246337, 0.51246011, 0.51245685, 0.51245359,\n",
+       "        0.51245034, 0.51244708, 0.51244382, 0.51244056, 0.51243731,\n",
+       "        0.51243405, 0.51243079, 0.51242754, 0.51242428, 0.51242102,\n",
+       "        0.51241777, 0.51241451, 0.51241126, 0.512408  , 0.51240474,\n",
+       "        0.51240149, 0.51239823, 0.51239498, 0.51239172, 0.51238847,\n",
+       "        0.51238521, 0.51238196, 0.5123787 , 0.51237545, 0.51237219,\n",
+       "        0.51236894, 0.51236568, 0.51236243, 0.51235917, 0.51235592,\n",
+       "        0.51235266, 0.51234941, 0.51234616, 0.5123429 , 0.51233965,\n",
+       "        0.5123364 , 0.51233314, 0.51232989, 0.51232664, 0.51232338,\n",
+       "        0.51232013, 0.51231688, 0.51231362, 0.51231037, 0.51230712,\n",
+       "        0.51230386, 0.51230061, 0.51229736, 0.51229411, 0.51229086,\n",
+       "        0.5122876 , 0.51228435, 0.5122811 , 0.51227785, 0.5122746 ,\n",
+       "        0.51227134, 0.51226809, 0.51226484, 0.51226159, 0.51225834,\n",
+       "        0.51225509, 0.51225184, 0.51224859, 0.51224534, 0.51224209,\n",
+       "        0.51223884, 0.51223558, 0.51223233, 0.51222908, 0.51222583,\n",
+       "        0.51222258, 0.51221933, 0.51221608, 0.51221284, 0.51220959,\n",
+       "        0.51220634, 0.51220309, 0.51219984, 0.51219659, 0.51219334,\n",
+       "        0.51219009, 0.51218684, 0.51218359, 0.51218034, 0.5121771 ,\n",
+       "        0.51217385, 0.5121706 , 0.51216735, 0.5121641 , 0.51216086,\n",
+       "        0.51215761, 0.51215436, 0.51215111, 0.51214787, 0.51214462,\n",
+       "        0.51214137, 0.51213812, 0.51213488, 0.51213163, 0.51212838,\n",
+       "        0.51212514, 0.51212189, 0.51211864, 0.5121154 , 0.51211215,\n",
+       "        0.5121089 , 0.51210566, 0.51210241, 0.51209917, 0.51209592,\n",
+       "        0.51209267, 0.51208943, 0.51208618, 0.51208294, 0.51207969,\n",
+       "        0.51207645, 0.5120732 , 0.51206996, 0.51206671, 0.51206347,\n",
+       "        0.51206022, 0.51205698, 0.51205373, 0.51205049, 0.51204725,\n",
+       "        0.512044  , 0.51204076, 0.51203751, 0.51203427, 0.51203103,\n",
+       "        0.51202778, 0.51202454, 0.5120213 , 0.51201805, 0.51201481,\n",
+       "        0.51201157, 0.51200832, 0.51200508, 0.51200184, 0.51199859,\n",
+       "        0.51199535, 0.51199211, 0.51198887, 0.51198563, 0.51198238,\n",
+       "        0.51197914, 0.5119759 , 0.51197266, 0.51196942, 0.51196617,\n",
+       "        0.51196293, 0.51195969, 0.51195645, 0.51195321, 0.51194997,\n",
+       "        0.51194673, 0.51194349, 0.51194025, 0.511937  , 0.51193376,\n",
+       "        0.51193052, 0.51192728, 0.51192404, 0.5119208 , 0.51191756,\n",
+       "        0.51191432, 0.51191108, 0.51190784, 0.5119046 , 0.51190136,\n",
+       "        0.51189813, 0.51189489, 0.51189165, 0.51188841, 0.51188517,\n",
+       "        0.51188193, 0.51187869, 0.51187545, 0.51187221, 0.51186898,\n",
+       "        0.51186574, 0.5118625 , 0.51185926, 0.51185602, 0.51185279,\n",
+       "        0.51184955, 0.51184631, 0.51184307, 0.51183983, 0.5118366 ,\n",
+       "        0.51183336, 0.51183012, 0.51182689, 0.51182365, 0.51182041,\n",
+       "        0.51181718, 0.51181394, 0.5118107 , 0.51180747, 0.51180423,\n",
+       "        0.51180099, 0.51179776, 0.51179452, 0.51179129, 0.51178805,\n",
+       "        0.51178481, 0.51178158, 0.51177834, 0.51177511, 0.51177187,\n",
+       "        0.51176864, 0.5117654 , 0.51176217, 0.51175893, 0.5117557 ,\n",
+       "        0.51175246, 0.51174923, 0.51174599, 0.51174276, 0.51173953,\n",
+       "        0.51173629, 0.51173306, 0.51172982, 0.51172659, 0.51172336,\n",
+       "        0.51172012, 0.51171689, 0.51171366, 0.51171042, 0.51170719,\n",
+       "        0.51170396, 0.51170072, 0.51169749, 0.51169426, 0.51169102,\n",
+       "        0.51168779, 0.51168456, 0.51168133, 0.5116781 , 0.51167486,\n",
+       "        0.51167163, 0.5116684 , 0.51166517, 0.51166194, 0.5116587 ,\n",
+       "        0.51165547, 0.51165224, 0.51164901, 0.51164578, 0.51164255,\n",
+       "        0.51163932, 0.51163609, 0.51163286, 0.51162963, 0.51162639,\n",
+       "        0.51162316, 0.51161993, 0.5116167 , 0.51161347, 0.51161024,\n",
+       "        0.51160701, 0.51160378, 0.51160055, 0.51159732, 0.5115941 ,\n",
+       "        0.51159087, 0.51158764, 0.51158441, 0.51158118, 0.51157795,\n",
+       "        0.51157472, 0.51157149, 0.51156826, 0.51156503, 0.51156181,\n",
+       "        0.51155858, 0.51155535, 0.51155212, 0.51154889, 0.51154567,\n",
+       "        0.51154244, 0.51153921, 0.51153598, 0.51153276, 0.51152953,\n",
+       "        0.5115263 , 0.51152307, 0.51151985, 0.51151662, 0.51151339,\n",
+       "        0.51151017, 0.51150694, 0.51150371, 0.51150049, 0.51149726,\n",
+       "        0.51149403, 0.51149081, 0.51148758, 0.51148436, 0.51148113,\n",
+       "        0.5114779 , 0.51147468, 0.51147145, 0.51146823, 0.511465  ,\n",
+       "        0.51146178, 0.51145855, 0.51145533, 0.5114521 , 0.51144888,\n",
+       "        0.51144565, 0.51144243, 0.5114392 , 0.51143598, 0.51143276,\n",
+       "        0.51142953, 0.51142631, 0.51142308, 0.51141986, 0.51141664,\n",
+       "        0.51141341, 0.51141019, 0.51140697, 0.51140374, 0.51140052,\n",
+       "        0.5113973 , 0.51139407, 0.51139085, 0.51138763, 0.5113844 ,\n",
+       "        0.51138118, 0.51137796, 0.51137474, 0.51137152, 0.51136829,\n",
+       "        0.51136507, 0.51136185, 0.51135863, 0.51135541, 0.51135218,\n",
+       "        0.51134896, 0.51134574, 0.51134252, 0.5113393 , 0.51133608,\n",
+       "        0.51133286, 0.51132964, 0.51132641, 0.51132319, 0.51131997,\n",
+       "        0.51131675, 0.51131353, 0.51131031, 0.51130709, 0.51130387,\n",
+       "        0.51130065, 0.51129743, 0.51129421, 0.51129099, 0.51128777,\n",
+       "        0.51128455, 0.51128133, 0.51127812, 0.5112749 , 0.51127168,\n",
+       "        0.51126846, 0.51126524, 0.51126202, 0.5112588 , 0.51125558,\n",
+       "        0.51125237, 0.51124915, 0.51124593, 0.51124271, 0.51123949,\n",
+       "        0.51123628, 0.51123306, 0.51122984, 0.51122662, 0.51122341,\n",
+       "        0.51122019, 0.51121697, 0.51121375, 0.51121054, 0.51120732,\n",
+       "        0.5112041 , 0.51120089, 0.51119767, 0.51119445, 0.51119124,\n",
+       "        0.51118802, 0.5111848 , 0.51118159, 0.51117837, 0.51117516,\n",
+       "        0.51117194, 0.51116873, 0.51116551, 0.51116229, 0.51115908]),\n",
+       " 'std_train_score': array([0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300037, 0.10300037, 0.10300037,\n",
+       "        0.10300037, 0.10300037, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300036, 0.10300036, 0.10300036, 0.10300036,\n",
+       "        0.10300036, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300035,\n",
+       "        0.10300035, 0.10300035, 0.10300035, 0.10300035, 0.10300034,\n",
+       "        0.10300034, 0.10300034, 0.10300034, 0.10300034, 0.10300034,\n",
+       "        0.10300034, 0.10300034, 0.10300034, 0.10300034, 0.10300034,\n",
+       "        0.10300034, 0.10300034, 0.10300034, 0.10300034, 0.10300034,\n",
+       "        0.10300034, 0.10300034, 0.10300034, 0.10300034, 0.10300034,\n",
+       "        0.10300034, 0.10300034, 0.10300034, 0.10300034, 0.10300033,\n",
+       "        0.10300033, 0.10300033, 0.10300033, 0.10300033, 0.10300033,\n",
+       "        0.10300033, 0.10300033, 0.10300033, 0.10300033, 0.10300033,\n",
+       "        0.10300033, 0.10300033, 0.10300033, 0.10300033, 0.10300033,\n",
+       "        0.10300033, 0.10300033, 0.10300033, 0.10300033, 0.10300033,\n",
+       "        0.10300033, 0.10300033, 0.10300032, 0.10300032, 0.10300032,\n",
+       "        0.10300032, 0.10300032, 0.10300032, 0.10300032, 0.10300032,\n",
+       "        0.10300032, 0.10300032, 0.10300032, 0.10300032, 0.10300032,\n",
+       "        0.10300032, 0.10300032, 0.10300032, 0.10300032, 0.10300032,\n",
+       "        0.10300032, 0.10300031, 0.10300031, 0.10300031, 0.10300031,\n",
+       "        0.10300031, 0.10300031, 0.10300031, 0.10300031, 0.10300031,\n",
+       "        0.10300031, 0.10300031, 0.10300031, 0.10300031, 0.10300031,\n",
+       "        0.10300031, 0.10300031, 0.10300031, 0.10300031, 0.1030003 ,\n",
+       "        0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 ,\n",
+       "        0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 ,\n",
+       "        0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 , 0.1030003 ,\n",
+       "        0.1030003 , 0.10300029, 0.10300029, 0.10300029, 0.10300029,\n",
+       "        0.10300029, 0.10300029, 0.10300029, 0.10300029, 0.10300029,\n",
+       "        0.10300029, 0.10300029, 0.10300029, 0.10300029, 0.10300029,\n",
+       "        0.10300029, 0.10300029, 0.10300028, 0.10300028, 0.10300028,\n",
+       "        0.10300028, 0.10300028, 0.10300028, 0.10300028, 0.10300028,\n",
+       "        0.10300028, 0.10300028, 0.10300028, 0.10300028, 0.10300028,\n",
+       "        0.10300028, 0.10300027, 0.10300027, 0.10300027, 0.10300027,\n",
+       "        0.10300027, 0.10300027, 0.10300027, 0.10300027, 0.10300027,\n",
+       "        0.10300027, 0.10300027, 0.10300027, 0.10300027, 0.10300027,\n",
+       "        0.10300026, 0.10300026, 0.10300026, 0.10300026, 0.10300026,\n",
+       "        0.10300026, 0.10300026, 0.10300026, 0.10300026, 0.10300026,\n",
+       "        0.10300026, 0.10300026, 0.10300026, 0.10300025, 0.10300025,\n",
+       "        0.10300025, 0.10300025, 0.10300025, 0.10300025, 0.10300025,\n",
+       "        0.10300025, 0.10300025, 0.10300025, 0.10300025, 0.10300025,\n",
+       "        0.10300025, 0.10300024, 0.10300024, 0.10300024, 0.10300024,\n",
+       "        0.10300024, 0.10300024, 0.10300024, 0.10300024, 0.10300024,\n",
+       "        0.10300024, 0.10300024, 0.10300024, 0.10300023, 0.10300023,\n",
+       "        0.10300023, 0.10300023, 0.10300023, 0.10300023, 0.10300023,\n",
+       "        0.10300023, 0.10300023, 0.10300023, 0.10300023, 0.10300023,\n",
+       "        0.10300022, 0.10300022, 0.10300022, 0.10300022, 0.10300022,\n",
+       "        0.10300022, 0.10300022, 0.10300022, 0.10300022, 0.10300022,\n",
+       "        0.10300022, 0.10300021, 0.10300021, 0.10300021, 0.10300021,\n",
+       "        0.10300021, 0.10300021, 0.10300021, 0.10300021, 0.10300021,\n",
+       "        0.10300021, 0.10300021, 0.1030002 , 0.1030002 , 0.1030002 ,\n",
+       "        0.1030002 , 0.1030002 , 0.1030002 , 0.1030002 , 0.1030002 ,\n",
+       "        0.1030002 , 0.1030002 , 0.1030002 , 0.10300019, 0.10300019,\n",
+       "        0.10300019, 0.10300019, 0.10300019, 0.10300019, 0.10300019,\n",
+       "        0.10300019, 0.10300019, 0.10300019, 0.10300018, 0.10300018,\n",
+       "        0.10300018, 0.10300018, 0.10300018, 0.10300018, 0.10300018,\n",
+       "        0.10300018, 0.10300018, 0.10300018, 0.10300017, 0.10300017,\n",
+       "        0.10300017, 0.10300017, 0.10300017, 0.10300017, 0.10300017,\n",
+       "        0.10300017, 0.10300017, 0.10300017, 0.10300016, 0.10300016,\n",
+       "        0.10300016, 0.10300016, 0.10300016, 0.10300016, 0.10300016,\n",
+       "        0.10300016, 0.10300016, 0.10300016, 0.10300015, 0.10300015,\n",
+       "        0.10300015, 0.10300015, 0.10300015, 0.10300015, 0.10300015,\n",
+       "        0.10300015, 0.10300015, 0.10300014, 0.10300014, 0.10300014,\n",
+       "        0.10300014, 0.10300014, 0.10300014, 0.10300014, 0.10300014,\n",
+       "        0.10300014, 0.10300013, 0.10300013, 0.10300013, 0.10300013,\n",
+       "        0.10300013, 0.10300013, 0.10300013, 0.10300013, 0.10300013,\n",
+       "        0.10300012, 0.10300012, 0.10300012, 0.10300012, 0.10300012,\n",
+       "        0.10300012, 0.10300012, 0.10300012, 0.10300012, 0.10300011,\n",
+       "        0.10300011, 0.10300011, 0.10300011, 0.10300011, 0.10300011,\n",
+       "        0.10300011, 0.10300011, 0.10300011, 0.1030001 , 0.1030001 ,\n",
+       "        0.1030001 , 0.1030001 , 0.1030001 , 0.1030001 , 0.1030001 ,\n",
+       "        0.1030001 , 0.10300009, 0.10300009, 0.10300009, 0.10300009,\n",
+       "        0.10300009, 0.10300009, 0.10300009, 0.10300009, 0.10300008,\n",
+       "        0.10300008, 0.10300008, 0.10300008, 0.10300008, 0.10300008,\n",
+       "        0.10300008, 0.10300008, 0.10300008, 0.10300007, 0.10300007,\n",
+       "        0.10300007, 0.10300007, 0.10300007, 0.10300007, 0.10300007,\n",
+       "        0.10300007, 0.10300006, 0.10300006, 0.10300006, 0.10300006,\n",
+       "        0.10300006, 0.10300006, 0.10300006, 0.10300005, 0.10300005,\n",
+       "        0.10300005, 0.10300005, 0.10300005, 0.10300005, 0.10300005,\n",
+       "        0.10300005, 0.10300004, 0.10300004, 0.10300004, 0.10300004])}"
+      ]
+     },
+     "execution_count": 142,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "grid.cv_results_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 116,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "10.805747561666841\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "cv_ridge = pd.Series(cv_ridge, index = alphas)\n",
+    "cv_ridge.plot(title = \"Validation Parameter for Alpha in the Ridge Model\")\n",
+    "plt.xlabel(\"alpha\")\n",
+    "plt.ylabel(\"rmse\")\n",
+    "\n",
+    "print cv_ridge.min()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 117,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[10.05908179  9.79339047 10.06594626 10.4227384  10.4483333  10.13015445\n",
+      " 10.33094621  9.84658728 10.22185448 10.06682664]\n",
+      "10.138585928774962\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Lasso Regression\n",
+    "from sklearn.linear_model import Lasso\n",
+    "\n",
+    "lasso_model = Lasso()\n",
+    "results = model_selection.cross_val_score(lasso_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "print np.sqrt(-results)\n",
+    "print np.sqrt(-results).mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 143,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train = pd.read_csv(\"train.csv\")\n",
+    "test = pd.read_csv(\"test.csv\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 147,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from scipy.stats import skew\n",
+    "from scipy.stats.stats import pearsonr\n",
+    "\n",
+    "all_data = pd.concat((train.loc[:,'MSSubClass':'SaleCondition'],\n",
+    "                      test.loc[:,'MSSubClass':'SaleCondition']))\n",
+    "\n",
+    "#log transform the target:\n",
+    "train[\"SalePrice\"] = np.log1p(train[\"SalePrice\"])\n",
+    "\n",
+    "#log transform skewed numeric features:\n",
+    "numeric_feats = all_data.dtypes[all_data.dtypes != \"object\"].index\n",
+    "\n",
+    "skewed_feats = train[numeric_feats].apply(lambda x: skew(x.dropna())) #compute skewness\n",
+    "skewed_feats = skewed_feats[skewed_feats > 0.75]\n",
+    "skewed_feats = skewed_feats.index\n",
+    "\n",
+    "all_data[skewed_feats] = np.log1p(all_data[skewed_feats])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 148,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "all_data = pd.get_dummies(all_data)\n",
+    "all_data = all_data.fillna(all_data.mean())\n",
+    "#creating matrices for sklearn:\n",
+    "X_train = all_data[:train.shape[0]]\n",
+    "X_test = all_data[train.shape[0]:]\n",
+    "y = train.SalePrice"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 149,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.linear_model import Ridge, RidgeCV, ElasticNet, LassoCV, LassoLarsCV\n",
+    "from sklearn.model_selection import cross_val_score\n",
+    "\n",
+    "def rmse_cv(model):\n",
+    "    rmse= np.sqrt(-cross_val_score(model, X_train, y, scoring=\"neg_mean_squared_error\", cv = 5))\n",
+    "    return(rmse)\n",
+    "\n",
+    "model_ridge = Ridge()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 150,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "alphas = [0.05, 0.1, 0.3, 1, 3, 5, 10, 15, 30, 50, 75]\n",
+    "cv_ridge = [rmse_cv(Ridge(alpha = alpha)).mean() \n",
+    "            for alpha in alphas]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 151,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0,0.5,'rmse')"
+      ]
+     },
+     "execution_count": 151,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "cv_ridge = pd.Series(cv_ridge, index = alphas)\n",
+    "cv_ridge.plot(title = \"Validation - Just Do It\")\n",
+    "plt.xlabel(\"alpha\")\n",
+    "plt.ylabel(\"rmse\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 154,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Ridge(alpha=1.0, copy_X=True, fit_intercept=True, max_iter=None,\n",
+       "   normalize=False, random_state=None, solver='auto', tol=0.001)"
+      ]
+     },
+     "execution_count": 154,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model_ridge.fit(X=X_train, y=train.SalePrice)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 172,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "GridSearchCV(cv=None, error_score='raise',\n",
+       "       estimator=Ridge(alpha=1.0, copy_X=True, fit_intercept=True, max_iter=None,\n",
+       "   normalize=False, random_state=None, solver='auto', tol=0.001),\n",
+       "       fit_params=None, iid=True, n_jobs=1,\n",
+       "       param_grid={'alpha': array([1.00000e-02, 5.00601e-02, ..., 1.99599e+01, 2.00000e+01])},\n",
+       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
+       "       scoring=None, verbose=0)"
+      ]
+     },
+     "execution_count": 172,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "alphas = np.linspace(.01, 20, 500)\n",
+    "param_grid = {\"alpha\": alphas}\n",
+    "\n",
+    "ridge_model = Ridge()\n",
+    "grid = GridSearchCV(ridge_model, param_grid=param_grid)\n",
+    "grid.fit(X=X_train, y=train.SalePrice)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 173,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Ridge(alpha=4.496733466933867, copy_X=True, fit_intercept=True, max_iter=None,\n",
+       "   normalize=False, random_state=None, solver='auto', tol=0.001)"
+      ]
+     },
+     "execution_count": 173,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "grid.best_estimator_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 177,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "url = \"http://archive.ics.uci.edu/ml/machine-learning-databases/parkinsons/telemonitoring/parkinsons_updrs.data\"\n",
+    "content = requests.get(url).content\n",
+    "df = pd.read_csv(io.StringIO(content.decode('utf-8')), sep=',')\n",
+    "df.drop(df.columns[[0, 4]], axis=1, inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 225,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "GridSearchCV(cv=None, error_score='raise',\n",
+       "       estimator=Ridge(alpha=1.0, copy_X=True, fit_intercept=True, max_iter=None,\n",
+       "   normalize=True, random_state=None, solver='auto', tol=0.001),\n",
+       "       fit_params=None, iid=True, n_jobs=1,\n",
+       "       param_grid={'alpha': array([1.0000e-02, 3.0002e-02, ..., 9.9980e+01, 1.0000e+02])},\n",
+       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
+       "       scoring='neg_mean_squared_error', verbose=0)"
+      ]
+     },
+     "execution_count": 225,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# df.drop(df.columns[[4, 5, 6, 7, 9, 10, 11, 12, 13, 14]], axis=1, inplace=True)\n",
+    "\n",
+    "features = df.columns\n",
+    "features = features.drop('total_UPDRS')\n",
+    "scaler = StandardScaler()\n",
+    "\n",
+    "# Other variables are y_sale_price and X_train\n",
+    "# y_sale_price = train.SalePrice\n",
+    "\n",
+    "# Split features from output\n",
+    "X = df.loc[:, df.columns != 'total_UPDRS']\n",
+    "y = df['total_UPDRS']\n",
+    "\n",
+    "# Standardize\n",
+    "# X = scaler.fit_transform(X)\n",
+    "# X = pd.DataFrame(X, columns=features)\n",
+    "\n",
+    "alphas = np.linspace(.01, 100, 5000)\n",
+    "param_grid = {\"alpha\": alphas}\n",
+    "\n",
+    "ridge_model = Ridge(normalize=True)\n",
+    "grid = GridSearchCV(ridge_model, param_grid=param_grid, scoring='neg_mean_squared_error')\n",
+    "grid.fit(X=X, y=y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 226,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Ridge(alpha=1.250124024804961, copy_X=True, fit_intercept=True, max_iter=None,\n",
+       "   normalize=True, random_state=None, solver='auto', tol=0.001)"
+      ]
+     },
+     "execution_count": 226,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "grid.best_estimator_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/.ipynb_checkpoints/ridge-model-checkpoint.ipynb b/.ipynb_checkpoints/ridge-model-checkpoint.ipynb
new file mode 100644
index 0000000..ab56be9
--- /dev/null
+++ b/.ipynb_checkpoints/ridge-model-checkpoint.ipynb
@@ -0,0 +1,847 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 200,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10b1dc790>"
+      ]
+     },
+     "execution_count": 200,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "import numpy as np\n",
+    "import random as random\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Set seed\n",
+    "random.seed(1075)\n",
+    "\n",
+    "# Create random curve\n",
+    "length = 125\n",
+    "x = np.linspace(0, 10, length)\n",
+    "y = .5 * np.cos(x)\n",
+    "y_scatter = []\n",
+    "\n",
+    "# Create random scattering around the true function\n",
+    "for i, j in enumerate(y):\n",
+    "    y_scatter.append( y[i] + random.uniform(-.3, .3))   \n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 201,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10b3b0590>"
+      ]
+     },
+     "execution_count": 201,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Try fitting a simple linear function with a polynomial order\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "from numpy import vander\n",
+    "\n",
+    "# We need to fit a polynomial function to our data points using the vander function\n",
+    "# If we just used linear regression we obtain a poor fit because our data was generated  \n",
+    "# using a nonlinear function. To get linear estimates of our coefficients from non-linear inputs\n",
+    "# we increase the number of dimensions of our data. I'm taking a guess our data is polynomial of\n",
+    "# degree 3, so I'm using 4 (3 + 1) as the parameter in vander to return that many columns\n",
+    "# If I wanted to test polynomial of degree 10, I would put 11 as the parameter.\n",
+    "\n",
+    "lm_model_3 = LinearRegression()\n",
+    "lm_model_3.fit(vander(x, 4), y_scatter)\n",
+    "degree_3 = lm_model_3.coef_.size - 1\n",
+    "y_pred_3 = lm_model_3.predict(np.vander(x, degree_3 + 1))\n",
+    "\n",
+    "lm_model_8 = LinearRegression()\n",
+    "lm_model_8.fit(vander(x, 9), y_scatter)\n",
+    "degree_8 = lm_model_8.coef_.size - 1\n",
+    "y_pred_8 = lm_model_8.predict(np.vander(x, degree_8 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_3)\n",
+    "plt.plot(x, y_pred_8)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 3', 'Pred. Deg. 8', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 239,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10f011390>"
+      ]
+     },
+     "execution_count": 239,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "lm_model_30 = LinearRegression()\n",
+    "lm_model_30.fit(vander(x, 31), y_scatter)\n",
+    "degree_30 = lm_model_30.coef_.size - 1\n",
+    "y_pred_30 = lm_model_30.predict(np.vander(x, degree_30 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_30)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 30', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 210,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Looks like we could be overfitting our data if we increase the polynomial degree too much\n",
+    "# How do we find the right polynomial number?\n",
+    "\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn import metrics\n",
+    "import pandas as pd\n",
+    "\n",
+    "# Get the RMSE for each predicted values\n",
+    "def get_rmse(y, y_pred):\n",
+    "    return np.sqrt(metrics.mean_squared_error(y, y_pred))\n",
+    "\n",
+    "# Create dataframe to evaluate\n",
+    "rmse_df = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 211,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 211,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtoAAAG5CAYAAACwZpNaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl8VNX9//HXmclGICEkAZJh3yFAJkJEEVBRVFCDtu6KtdYWrVVbW+vSWttqfxVbW7VqVWrt17rh3hIrouCKqCyagGxCQNnCFhK2kJDl/P64kxgxkG0mdyZ5Px+PeWTmzr3nfgKC7xzOnI+x1iIiIiIiIsHlcbsAEREREZG2SEFbRERERCQEFLRFREREREJAQVtEREREJAQUtEVEREREQkBBW0REREQkBBS0RURagTHGa4zZb4zp7XYtIiLSOhS0RUTqEQjFNY9qY8zBOq8va+p41toqa20na+3GJtYRb4zZa4w5sZ73HjTGzGpqLYFrvQ18jxc1Z9zA2EuMMRcf5f0Rxhh72P33G2PObu49RUTCUZTbBYiIhCNrbaea58aYL4EfWmvnHel8Y0yUtbYyBHWUGmNeBL4HvF/nftHAxYHjzRm3Cqj7PW4Gpllr321RwY1XXvfX+EiMMQbwBOo94rFGjBOS3x8RkaPRjLaISDMYY/5gjHneGPOcMWYfMM0YM9YY87ExpsQYU2iM+VsgEGOMiQrM4vYNvH468P4cY8w+Y8xHxph+R7jdk8D5xpi4OsemAJXAm4HxfmWM2RqY/V5tjDk5CN9jlDHmd8aYDcaYXcaYp4wxiYH3OhljXjDG7DbGFAe+787GmPuBY4D/C8xS39OM+75kjLnfGDMPOAAce4RjKcaYWYHa1htjflFnjOuMMW8ZYx41xhQDN7X010NEpKkUtEVEmu87wLNAZ+B5nOD7UyAVGAdMBq4+yvWXAr8BkoGNwF1HOO8DYDdwbp1jlwPPWGurjDHDA/cZZa1NxAnhTVqicgS3ACcDY4FegWN/CXy9GrCAD+gK3AAcstb+DPgM+H5gqcwtzbz3NOA2nFn3z45w7B+BGvrg/FrfcNiSl4nAJzi/Hw80sw4RkWZT0BYRab4F1tpca221tfagtXaxtfYTa22ltXY9MBM46SjXv2StXWKtrQCeAbLqO8laa4F/E1gmYoxJAnJwZrrBCfhxwPDAEokNgfu31DXALdbabdbag8CdOMtVACpwAnb/wPe7KHBOY8UGZv7rPup+UPSFwK9ntbW2/PBjgBfnB49brLUHrLVfAH/D+QGkxhfW2n8F1sc3pTYRkaDQGm0RkebbVPeFMWYozozvaCAe5+/YT45y/bY6z0ups2a6Hv8Gfm2M6Q6cA6yy1i4HsNauCSybuBMYZoyZC9xord125OGOzhjjBXoAbxpjbJ23ogNBfybQDXjVGBOPE/rvCITgxii31iYd5f1NDRxLx5nNrnvsq0DNRxtDRKTVaEZbRKT57GGvHwM+BwYGlnDcAZig3MiZof4IuAxn1vbJw95/2lo7DuiHM9t7dwvvVwUUAidaa5PqPOKstSXW2jJr7e3W2iE4SzQuBS6oubwl9z7KGHWPFeL8P6xXnWO9gS0NjCEi0moUtEVEgicB2AMcMMYM4+jrs5vjSZw14MfhrA0HwBgzzBgz0RgTCxwMPBo7s3w0jwL3GGN6BO7TvWYLPmPMaYH7eoC9OMtXau65HegfhPsfkbW2FPgvcLcxpqMxZiBwPfB0KO8rItIUCtoiIsHzC+AKYB/O7PbzQR7/RZwP9s211u6oczwW+BOwC2c5Shfg1wDGmCuMMfnNvN/dOFsKvmeM2QsswNlRBJyZ5Fyc7zUfJ/S+HHjvL8BVgXXXR5pZj61nH+3pTazvR0A0zgc/5wGPEPxfcxGRZjPOZ2xERERERCSYNKMtIiIiIhICCtoiIiIiIiGgoC0iIiIiEgIK2iIiIiIiIdBmGtakpqbavn37ul2GiIiIiLRxS5cu3WWt7drQeW0maPft25clS5a4XYaIiIiItHHGmK8ac56WjoiIiIiIhICCtoiIiIhICChoi4iIiIiEQJtZoy0iIiISiSoqKti8eTNlZWVulyKHiYuLo2fPnkRHRzfregVtERERERdt3ryZhIQE+vbtizHG7XIkwFpLUVERmzdvpl+/fs0aQ0tHRERERFxUVlZGSkqKQnaYMcaQkpLSon9pUNAWERERcZlCdnhq6e+LgraIiIiISAgoaIuIiIi0Y0VFRWRlZZGVlUVaWho9evSofX3o0KFGjXHllVeyZs2ao57z8MMP88wzzwSjZMaPH8+QIUPIzMxk6NChXH/99ezZs+eo11RXVzNjxoyg3L+xjLW2VW8YKtnZ2VadIUVERCTSrFq1imHDhrldBgC/+93v6NSpEzfddNM3jltrsdbi8YTHHO348eN56KGHan8YuPnmm1m+fDnz588/4jWVlZWkpqZSUlLSpHvV9/tjjFlqrc1u6Nrw+NUSERERkbCybt06MjIyuOyyyxg+fDiFhYVMnz6d7Oxshg8fzp133ll77vjx48nLy6OyspKkpCRuvfVW/H4/Y8eOZceOHQDcfvvt3H///bXn33rrrYwZM4YhQ4awcOFCAA4cOMB5551HRkYG559/PtnZ2eTl5R21zpiYGO69917Wrl3LihUrAMjJyWH06NEMHz6cxx9/HIBbb72Vffv2kZWVxfe+970jnhdM2t5PREREJEz8PncFK7fuDeqYGb5EfpszvFnXrl69mn//+99kZzuTtzNmzCA5OZnKykomTpzI+eefT0ZGxjeu2bNnDyeddBIzZszg5z//OU888QS33nrrt8a21rJo0SJmz57NnXfeyRtvvMGDDz5IWloaL7/8Mvn5+YwaNapRdUZFRZGZmcnq1asZPnw4Tz75JMnJyZSWlpKdnc15553HjBkzePzxx78R3Os7r0uXLs36taqPZrRFREREpF4DBgyoDdkAzz33HKNGjWLUqFGsWrWKlStXfuuaDh06MGXKFABGjx7Nl19+We/Y3/3ud791zoIFC7j44osB8Pv9DB/e+B8Q6i6Hvu+++2pn1Ddv3kxBQUG91zT2vObSjLaIiIhImGjuzHOodOzYsfb52rVreeCBB1i0aBFJSUlMmzat3j2mY2Jiap97vV4qKyvrHTs2NrbBcxqrsrKSzz//nGHDhjFv3jzef/99Pv74Yzp06MD48ePrrbOx57WEgnYLrduxn6pqi9cDHmPwGIPXY/B4DF5j8ASOO8+d97zGYAy1zz0e7Z0pIiIi4W3v3r0kJCSQmJhIYWEhc+fOZfLkyUG9x7hx43jhhReYMGECy5cvr3fG/HCHDh3itttuY+DAgWRkZLBq1SqSk5Pp0KEDK1asYPHixYCzvAScUB4VFcWePXvqPS+YFLRb6Mr/W8Sm3QdbPM6RArjXY/AYvg7wps6xwHk1xz0evr7O1BP26xz3BO5T97gxBq+Hw+5T8/yb9zOm5vnXxz2196T2h41v/pDh1DGqdxd6JccH4VdfREREWsuoUaPIyMhg6NCh9OnTh3HjxgX9Htdffz3f+973yMjIqH107ty53nMvuugiYmNjKS8v5/TTT+eVV14B4KyzzmLmzJlkZGQwZMgQjjvuuNprrrrqKjIzM8nOzmbmzJlHPC9YtL1fC727ZgcHyquoCmx7U1XtPKqtpaoaqm3N83qOV1uq6n61OM8bcbza8vU51V/fo9ryrftVW75xTt3jNfeoqq5n3KMcb8l/NiN7dCb3+vHB+00QERGJYOG0vZ/bKisrqaysJC4ujrVr13L66aezdu3a2tloN7Rkez/NaLfQyUO6uV2CK2x9Ab7uDwGHBXsbCOqvfraFv81fy/qd++nftZPb34aIiIiEkf3793PqqadSWVmJtZbHHnvM1ZDdUpFbubjKWTriLBFpikvH9ObBt9eSm1/ITycNClF1IiIiEomSkpJYunSp22UEjbb3k1aV1jmOY/smMzt/C21l2ZKIiIhIfRS0pdVN9fso2HmA1dv2uV2KiIiISMgoaEurmzIiDa/HkJu/1e1SREREREJGQVtaXUqnWMYNTCV32VYtHxEREZE2S0FbXJGTmc6m3QfJ21TidikiIiLtWlFREVlZWWRlZZGWlkaPHj1qXx86dKjR4zzxxBNs27at3vemTZtGv3798Pv9DB48mCuuuIKtWxv+l+2//vWvQe/W2JoUtMUVZ4xII8brITe/0O1SRERE2rWUlBTy8vLIy8vjmmuu4cYbb6x9XbedekOOFrQB7rvvPvLz81m9ejUjR47klFNOoaKi4qhjKmiLNENiXDQnD+nKa8u2UlWt5SMiIiLh6Mknn2TMmDFkZWVx7bXXUl1dTWVlJZdffjkjR45kxIgR/O1vf+P5558nLy+Piy66qMGZcI/Hw0033URycjJvvvkmANOnTyc7O5vhw4dz5513Ak4w37FjBxMmTGDSpElHPC+caR9tcU2O38ebK7ezaMNuxg5IcbscERER9825FbYtD+6YaSNhyowmX/b555/z6quvsnDhQqKiopg+fTqzZs1iwIAB7Nq1i+XLnTpLSkpISkriwQcf5KGHHiIrK6tR448aNYrVq1dz1llnMWPGDJKTk6msrGTixImcf/753HjjjfzlL3/hgw8+ICkpCaDe8zIyMpr8vbUWzWiLa04d1o34GC+5y7T7iIiISLiZN28eixcvJjs7m6ysLN577z0KCgoYOHAga9as4YYbbmDu3Ll07ty5WePX3RDhueeeY9SoUYwaNYpVq1axcuXKeq9p7HnhQjPa4pr4mCgmDevOnOWF/H7qcKK9+rlPRETauWbMPIeKtZYf/OAH3HXXXd96b9myZcyZM4eHH36Yl19+mZkzZzZ5/Ly8PM466yzWrl3LAw88wKJFi0hKSmLatGn1rstu7HnhRMlGXJXj91FcWsGCdbvcLkVERETqmDRpEi+88AK7djn/jy4qKmLjxo3s3LkTay0XXHABd955J59++ikACQkJ7NvXcDM6ay333XcfRUVFnHbaaezdu5eEhAQSExMpLCxk7ty5tefWHfNo54UrzWiLq04cnEpiXBS5+VuZOKSb2+WIiIhIwMiRI/ntb3/LpEmTqK6uJjo6mkcffRSv18tVV12FtRZjDPfccw8AV155JT/84Q/p0KEDixYt+taOJTfeeCO//e1vOXjwIGPHjuXtt98mOjqaUaNGkZGRwdChQ+nTpw/jxo2rvWb69OlMmjSJXr168dZbbx3xvHBl2krDkOzsbLtkyRK3y5BmuPmlfF5fvo0lt08iLtrrdjkiIiKtatWqVQwbNsztMuQI6vv9McYstdZmN3Stlo6I63L8PvaXV/Lump1ulyIiIiISNAra4rqx/VNI6RhDbr52HxEREZG2Q0FbXBfl9XDmyHTmr97O/vJKt8sRERFpdW1lKW9b09LfFwVtCQtTs3yUVVQzf9V2t0sRERFpVXFxcRQVFSlshxlrLUVFRcTFxTV7DO06ImFhdO8upHeOY3beVs7J6uF2OSIiIq2mZ8+ebN68mZ079VmlcBMXF0fPnj2bfb2CtoQFj8dwdmY6/7fwS0pKD5EUH9PwRSIiIm1AdHQ0/fr1c7sMCQEtHZGwMdXfg4oqy9wV29wuRURERKTFFLQlbIzokUjflHhy8wvdLkVERESkxRS0JWwYY8jx+1hYsIsd+8rcLkdERESkRRS0Jazk+H1UW5izXMtHREREJLIpaEtYGdw9gaFpCWpeIyIiIhFPQVvCTo7fx5KvitlSctDtUkRERESaTUFbws7ZmekAvKZZbREREYlgCtoSdvqkdMTfK4ncZQraIiIiErkUtCUs5WSm8/mWvazfud/tUkRERESaRUFbwtLZmT6MQXtqi4iISMRS0JawlNY5jmP7JjM7fwvWWrfLEREREWkyBW0JW1P9Pgp2HmD1tn1ulyIiIiLSZAraEramjEjD6zHaU1tEREQikoK2hK2UTrGMG5hK7rKtWj4iIiIiESekQdsYM9kYs8YYs84Yc+tRzjvPGGONMdl1jt0WuG6NMeaMUNYp4Wuq38em3QfJ21TidikiIiIiTRKyoG2M8QIPA1OADOASY0xGPeclAD8FPqlzLAO4GBgOTAb+HhhP2pnTh3cnxuvR7iMiIiIScUI5oz0GWGetXW+tPQTMAs6p57y7gHuAsjrHzgFmWWvLrbUbgHWB8aSdSYyL5uQhXXlt2VaqqrV8RERERCJHKIN2D2BTndebA8dqGWNGAb2stf9r6rWB66cbY5YYY5bs3LkzOFVL2Mnx+9ixr5xFG3a7XYqIiIhIo7n2YUhjjAf4K/CL5o5hrZ1prc221mZ37do1eMVJWDl1WDfiY7xqyS4iIiIRJZRBewvQq87rnoFjNRKAEcC7xpgvgeOB2YEPRDZ0rbQj8TFRTBrWnTnLC6moqna7HBEREZFGCWXQXgwMMsb0M8bE4Hy4cXbNm9baPdbaVGttX2ttX+BjYKq1dkngvIuNMbHGmH7AIGBRCGuVMJfj91FcWsGCdbvcLkVERESkUUIWtK21lcB1wFxgFfCCtXaFMeZOY8zUBq5dAbwArATeAH5ira0KVa0S/k4cnEpiXJSa14iIiEjEiArl4Nba14HXDzt2xxHOPfmw1/8P+H8hK04iSmyUl8kj0nh9+TbKKqqIi9ZujyIiIhLe1BlSIkaO38f+8kreXaMdZkRERCT8KWhLxBjbP4XUTjFaPiIiIiIRQUFbIkaU18OZI9OZv3o7+8sr3S5HRERE5KgUtCWi5Ph9lFVUM3/VdrdLERERETkqBW2JKKN7dyG9cxyz87R8RERERMKbgrZEFI/HcHZmOu+v3UlJ6SG3yxERERE5IgVtiThT/T2oqLLMXbHN7VJEREREjkhBWyLOiB6J9E2JJze/0O1SRERERI5IQVsijjGGHL+PhQW72LGvzO1yREREROqloC0RaarfR7WFOcu1fERERETCk4K2RKRB3RMYmpag5jUiIiISthS0JWLl+H0s+aqYLSUH3S5FRERE5FsUtCVinZ2ZDsBrmtUWERGRMKSgLRGrT0pH/L2SyF2moC0iIiLhR0FbIlpOZjqfb9nL+p373S5FRERE5BsUtCWinZ3pwxi0p7aIiIiEHQVtiWhpneMY0zeZ2flbsNa6XY6IiIhILQVtiXg5fh8FOw+wets+t0sRERERqaWgLRFvyog0vB6jPbVFREQkrChoS8RL6RTLuIGp5C7bquUjIiIiEjYUtKVNmOr3sWn3QfI2lbhdioiIiAigoC1txOnDuxPj9Wj3EREREQkbCtrSJiTGRXPykK68tmwrVdVaPiIiIiLuU9CWNmNqlo8d+8pZtGG326WIiIiIKGhL23HK0G7Ex3jVkl1ERETCgoK2tBnxMVFMGtadOcsLqaiqdrscERERaecUtKVNyfH7KC6tYMG6XW6XIiIiIu2cgra0KScOTiUxLkrNa0RERMR1CtrSpsRGeZk8Io03V2ynrKLK7XJERESkHVPQljYnx+9jf3kl767Z4XYpIiIi0o4paEubM7Z/CqmdYtS8RkRERFyloC1tTpTXw5kj05m/ejv7yyvdLkdERETaqSi3C4h4/zgVSosgPhk6JEN8Sp3nXQJfD3svuoPbVbd5OX4f//7oK+av2s45WT3cLkdERETaIQXtlup/EhR/BQd3w4GdsHON8/zQ/iNfE9WhnjBeN6DX/drFeS+uMxjTet9XhBvduwvpneOYnbdVQVtERERcoaDdUqfeUf/xynI4WAylu53gXbrbmfmueV73ve0rnPfKSsAeodGK8QZCdz1hvN6AHvjqjQ7d9x7GPB5Djt/Hvz7cQEnpIZLiY9wuSURERNoZBe1QiYqFhDTn0VjV1U7YPmpAD3wt2Qhb85z3qsqPPGZs4hECekqdGfPD3ovp2CZmz3Myfcx8fz1zV2zjomN7u12OiIiItDMK2uHE4/l6hjplQOOusRYqSr8ZwmsDej3HdhdAaTGU7znymN7Yepav1LcGvc7XuCSn/jAyokcifVPiyc0vVNAWERGRVqegHemMcWagYzpCUq/GX1dV8e2Z82/NoBc7X2vWnZfuBnukJjDmyEtbataZ1xzzZUFsQlC+/aMxxlk+8vA769ixr4xuCXEhv6eIiIhIDQXt9sobDZ26OY/GshbK9zpBvCaEH2mJy94tsO1z53VF6TfH6dIPfjgPOqYG93uqx1S/jwffXsec5du44oS+Ib+fiIiISA0FbWk8Y5zdT+I6Q3ITrqs4+HUg3/UF/OdaeO5iuCI35FsdDuqewNC0BHLztypoi4iISKsKr0W10jZFd4DOPSBtJIw4D777D9i8BF7+IVQfaSlK8OT4fSz5qpgtJQdDfi8RERGRGgra0voypsIZf4TVr8Gbt4f8djmZPgBey98a8nuJiIiI1FDQFneMvRaOuwY+/jt8/EhIb9U7JR5/ryRylyloi4iISOtR0Bb3nPFHGHo2vHEbrMoN6a1yMtP5fMte1u88SsdOERERkSBS0Bb3eLzOeu0eo5312psWh+xWZ2f6MAZy8wtDdg8RERGRuhS0xV0x8XDJLKeD5nMXwe71IblNWuc4xvRNZnb+Fqy1IbmHiIiISF0K2uK+Tl3hspfBVsPT58OBopDcJsfvo2DnAVZv2xeS8UVERETqUtCW8JA60JnZ3rMZZl3i7L0dZFNGpOH1GGZr9xERERFpBQraEj56Hw/ffQw2fQKvXgPV1UEdPqVTLOMHppKbv1XLR0RERCTkFLQlvAz/Dpx2F6z8D8y7I+jD5/h9bC4+SN6mkqCPLSIiIlKXgraEnxOuh2N/BAsfhEX/COrQpw/vTozXo91HREREJOQUtCX8GANT7oHBU2DOzbD69aANnRgXzclDuvLasq1UVWv5iIiIiISOgraEJ48Xzv8npPvhpR/AlqVBG3pqlo8d+8pZtGF30MYUEREROZyCtoSvmI5w6QvO9n/PXgTFXwZl2FOGdiM+xquW7CIiIhJSCtoS3jp1c/bYrqpw9tgubfksdHxMFJOGdWfO8kIqqoK7s4mIiIhIDQVtCX9dB8PFz0LJVzDrMqgoa/GQU/0+iksrWLBuVxAKFBEREfk2BW2JDH3HwbmPwMaF8N9rW7zH9oTBqSTGRZGr5jUiIiISIlFuFyDSaCPPhz2bYN7voHMvOO33zR4qNsrL5BFpvL58G2UVVcRFe4NXp4iIiAia0ZZIM+5nkP0D+PB+WPzPFg2V4/exv7ySd9fsCFJxIiIiIl8LadA2xkw2xqwxxqwzxtxaz/vXGGOWG2PyjDELjDEZgeN9jTEHA8fzjDGPhrJOiSDGwJQ/w6Az4PWbYM0bzR5qbP8UUjvFqHmNiIiIhETIgrYxxgs8DEwBMoBLaoJ0Hc9aa0daa7OAPwF/rfNegbU2K/C4JlR1SgTyRsH5T0DaSHjpStj6WbOGifJ6OHNkOvNXb2d/eWWQixQREZH2LpQz2mOAddba9dbaQ8As4Jy6J1hr99Z52RFQqz5pnNhOzh7b8SmBPba/atYwOX4fZRXVzF+1PcgFioiISHsXyqDdA9hU5/XmwLFvMMb8xBhTgDOjfUOdt/oZYz4zxrxnjJlQ3w2MMdONMUuMMUt27twZzNolEiSkwWUvOdv9PXMBHCxu8hCje3fB1zmO2XnafURERESCy/UPQ1prH7bWDgBuAW4PHC4EeltrjwF+DjxrjEms59qZ1tpsa212165dW69oCR/dhsLFT8Pu9fD85VBZ3qTLPR7D2X4f76/dSUnpoRAVKSIiIu1RKIP2FqBXndc9A8eOZBZwLoC1ttxaWxR4vhQoAAaHqE6JdP1OhHP/Dl9+AP+9DmzTViDlZPqoqLLMXbEtRAWKiIhIexTKoL0YGGSM6WeMiQEuBmbXPcEYM6jOy7OAtYHjXQMfpsQY0x8YBKwPYa0S6TIvhFN+A8tfgLf/0KRLR/RIpG9KPLPVvEZERESCKGQNa6y1lcaY64C5gBd4wlq7whhzJ7DEWjsbuM4YMwmoAIqBKwKXnwjcaYypAKqBa6y1u0NVq7QRE37htGn/4F5I6gWjv9+oy4wxTPX7eOiddezYV0a3hLjQ1ikiIiLtgrFN/Gf2cJWdnW2XLFnidhnitqoKeO5iKHgHLn0eBp3WqMvWbt/Hafe9z++nDueKE/qGtkYRERGJaMaYpdba7IbOc/3DkCJB5Y2GC/4PumfAi9+HwvxGXTaoewJD0xLI1fIRERERCRIFbWl7YhPg0hchLgmeuRBKNjV8Dc6e2ku+KmZLycEQFygiIiLtgYK2tE2J6XDZi1BRGthju6TBS3IyfQC8plltERERCQIFbWm7umfARU9B0Vp44XKoPPo+2b1T4vH3SiJ3mYK2iIiItJyCtrRt/U+GqQ/Bhvdh9vUN7rGdk5nO51v2sn7n/lYpT0RERNouBW1p+7IugZN/Bctmwbt3H/XUszN9GAO5+YWtVJyIiIi0VQra0j6cdDNkTYP37oFPnzriaWmd4xjTN5nZ+VtoK1tfioiIiDsUtKV9MAZy7of+E+G1n8G6+Uc8Ncfvo2DnAVZv29eKBYqIiEhbo6At7Yc3Gi78N3QdCi9cAduW13valBFpeD1GLdlFRESkRRS0pX2JS4RLX3D22n7mQtiz5VunpHSKZfzAVHLzt2r5iIiIiDSbgra0P517wGUvQPk+Z4/tsj3fOiXH72Nz8UHyNjW8/7aIiIhIfRS0pX1KGwkXPgm71jjLSKoqvvH26cO7ExPl0e4jIiIi0mwK2tJ+DTwVch6A9e9A7s++scd2Ylw0E4d05bVlW6mq1vIRERERaToFbWnfjpkGJ90CeU/De3/6xls5fh879pWzaMNul4oTERGRSKagLXLybeC/BN79I+Q9W3v4lKHdiI/xqiW7iIiINIuCtogxkPM36Hei06Z9/bsAxMdEMWlYd+YsL6SiqtrdGkVERCTiKGiLAETFwIVPQcogeP5y2L4SgKl+H8WlFSxYt8vlAkVERCTSKGiL1OiQBJe9CNHx8Mz5sHcrEwankhgXRa6a14iIiEgTKWiL1JXUy9lju2wPPHMhsVWlTB6RxpsrtlNWUeV2dSIiIhJBFLRFDpfuhwuehB0r4YUrOGd4RM6zAAAgAElEQVRkN/aXV/Lumh1uVyYiIiIRREFbpD6DJsHZf4WC+Yxd/UdSO0areY2IiIg0SZTbBYiErdHfh5KNeD74CzPS47hu9SnsL6+kU6z+2IiIiEjDNKMtcjSn/AZGXsCkwpmcUfU+81Zud7siERERiRAK2iJHYwyc8zC2z3jujZnJFx+/7nZFIiIiEiEUtEUaEhWLufhp9sT15Jptd7B34zK3KxIREZEIoKAt0hgdurDrnGcpIwbvcxfCvm1uVyQiIiJhTkFbpJGGDM3gN/G/wVu2G569EMr3u12SiIiIhDEFbZFGMsYw5JgJ/OTQDdhty+GlK6Gq0u2yREREJEwpaIs0QY7fx/yqY/h46G2w9k2Y80uw1u2yREREJAwpaIs0waDuCQxNS+Avu8fDuJ/Bkifgw/vdLktERETCkIK2SBPl+H0s+aqYLdk3w4jzYN7vYPlLbpclIiIiYUZBW6SJcjJ9ALy2bBuc83fofQL858fw5YcuVyYiIiLhREFbpIl6p8Tj75VE7rKtEB0HFz8DSX1g1qWw8wu3yxMREZEwoaAt0gxT/T4+37KX9Tv3Q3wyTHsJvNHwzHmwf4fb5YmIiEgYUNAWaYazRqZjDOTmFzoHuvSFS5+H/TudPbYPHXC1PhEREXGfgrZIM6R1jmNM32Rm52/B1mzv12M0nP8EFObDS1dBdZW7RYqIiIirFLRFminH76Ng5wFWb9v39cGhZ8KUP8EXc2DOLdpjW0REpB1T0BZppjNHpuP1GGbnb/3mG2N+BGOvg8X/gI8ecqc4ERERcZ2CtkgzJXeMYfzAVHLzt369fKTGaXdBxjnw5u2w4lV3ChQRERFXKWiLtECO38fm4oPkbSr55hseD3xnJvQ6Hl65GjZ+7E6BIiIi4hoFbZEWOH14d2KiPN9ePgLOHtuXPAede8JzF8Ouda1foIiIiLhGQVukBRLjopk4pCv/W1ZIVXU9H3ys2WPbeAN7bO9s/SJFRETEFQraIi2U4/exY185izbsrv+E5P5wySzYt82Z2T5U2roFioiIiCsUtEVa6NSh3YmP8Tot2Y+k17Fw3uOwZSm88iPtsS0iItIOKGiLtFCHGC+nZXRnzvJCKqqqj3zisByYfDesfg3m/rr1ChQRERFXKGiLBEFOpo/i0goWrNt19BOP/zEcfy188gh89PfWKU5ERERcoaAtEgQTBqeSGBdFbn27jxzu9D84s9tzfwUr/xv64kRERMQVCtoiQRAb5WXyiDTeXLGdsooG1l97vPDdf0DPbHhlOmxa1DpFioiISKs6atA2xpxS53m/w977bqiKEolEU/092F9eybtrdjR8cnQHZyeShHRnJ5KigtAXKCIiIq2qoRnte+s8f/mw924Pci0iEe34/smkdoohN7+wcRd0TIVpL4O18Mz5cKAotAWKiIhIq2ooaJsjPK/vtUi7FuX1cObIdOav3s7+8srGXZQywJnZ3rPFmdmuOBjaIkVERKTVNBS07RGe1/dapN2b6vdRVlHNvJXbG39R7+PguzNh82JnzXb1UbYIFBERkYjRUNDub4yZbYzJrfO85nW/Bq4VaXdG9e6Cr3Nc43YfqWv4uc5uJKtmw1u/CU1xIiIi0qqiGnj/nDrP7z3svcNfi7R7Ho/hbL+Pf324gZLSQyTFxzT+4rE/gZKN8NFDkNQbjrs6dIWKiIhIyB11Rtta+17dB7AQ2AusCrwWkcPkZPqoqLLMXbGtaRca43SOHHIWzLkFVv8vNAWKiIhIq2hoe79HjTHDA887A/nAv4HPjDGXtEJ9IhFnRI9E+qV2ZHZTl4+As8f2eY9Dj1Hw0lWweWnwCxQREZFW0dAa7QnW2hWB51cCX1hrRwKjgZtDWplIhDLGkJOZzkcFRezYV9b0AWLi4ZLnoVM3ePZC2L0h+EWKiIhIyDUUtA/VeX4a8B8Aa20T/01cpH3J8fuotjBneTP/qHTqGthjuwoeOQGeuxQ+fQr2N6IZjoiIiISFhj4MWWKMORvYAowDrgIwxkQBHUJcm0jEGtQ9gaFpCczO38oVJ/Rt3iCpg+D7r8OSf8KaObDmf4BxWrcPngxDzoRuw5y13SIiIhJ2GprRvhq4DvgX8LM6M9mnAg1+UssYM9kYs8YYs84Yc2s9719jjFlujMkzxiwwxmTUee+2wHVrjDFnNP5bEgkPOX4fS78qZnNxafMH6Z4BZ/0FblwBV38AJ98G1ZXw9l3wyFh4INP54GTBO1B5qOHxREREpNUYa0PTd8YY4wW+wFlyshlYDFxirV1Z55xEa+3ewPOpwLXW2smBwP0cMAbwAfOAwdbaqiPdLzs72y5ZsiQk34tIc2wsKuXEP7/DbVOGcvVJA4I7+N5C+OIN57H+Xagsg9hEGHiqM9M9cBLEJwf3niIiIgKAMWaptTa7ofOOunTEGPO3o71vrb3hKG+PAdZZa9cHxpqFsy93bdCuCdkBHfm62+Q5wCxrbTmwwRizLjDeR0erRySc9E6Jx98ridxlW4MftBPTIftK53Go1Anba16HL+bCilfBeKH3WBgSWGKSEuT7i4iISIMaWqN9DfA58AKwFWjKYtAewKY6rzcDxx1+kjHmJ8DPgRjglDrXfnzYtT3quXY6MB2gd+/eTShNpHVM9fu467WVrN+5n/5dO4XmJjHxMPRM51FdDVs/ddZ0f/EGvHm780gZBEOmOI+eY8Db0B99ERERaamG1minAzOBM4DLgWjgv9baJ621TwajAGvtw9baAcAtwO1NvHamtTbbWpvdtWvXYJQjElRnjUzHGMjNL2ydG3o8zoclT/0N/PhD+OkymPJn6NwTPn4E/jUF7h0Ir1ztzHyX7W14TBEREWmWo05rWWuLgEeBR40xPYGLgZXGmFustU81MPYWoFed1z0Dx45kFvBIM68VCUtpneMY0zeZ2flbuOHUgZjW3iGkSx84brrzKNsLBfNhzRuwdi4smwWeaOg73lleMmSy0/pdREREgqKhGW0AjDGjgJ8C04A5QGPa1S0GBhlj+hljYnBC+uzDxh1U5+VZwNrA89nAxcaYWGNMP2AQsKgxtYqEm6lZPgp2HmD1tn3uFhKXCMO/A999DG5aB1fOgeOvgT2bYc4v4f6R8Mg4mH+X05GyutrdekVERCJcQx+GvBMnAK/CmXG+zVpb2ZiBrbWVxpjrgLmAF3jCWrsiMOYSa+1s4DpjzCSgAigGrghcu8IY8wLOBycrgZ8cbccRkXA2ZUQ6d/x3BbPztzIsPdHtchzeKOhzgvM4/Q+wax18McdZ273gr/DBvdCpOww+AwZPgf4nO2vBRUREpNGOur2fMaYa2ADUbARcc7IBrLU2M7TlNZ6295NwdsUTiyjYuZ8Pbp7Y+stHmqp0N6yb5+xism4+lO+FqDgnbA+e7DwS092uUkRExDVB2d4P6BekekTatRy/j5tezCdvUwnH9O7idjlHF58MmRc6j8pD8NWHzg4ma153vgL4Rn29i0n3EepOKSIiUo9mNawxxnhwms88E/ySmkcz2hLO9pZVkP2HeVx2XG9+mzPc7XKax1rYserrwL15CWAhsWcgdE+GvhMgKtbtSkVEREKqsTPaDS0dSQR+grOH9WzgLZyW7L8A8q215wSn3JZT0JZwd/VTS/hsYwkf3XYqXk8bmAHev8NpkPPFG1DwNlSUQkwnGHCKE7wHnQEdU9yuUkREJOiCtXTkKZwPKX4E/BD4Fc767HOttXktrlKkHcnx+5i7YjuLNuxm7IA2EEA7dYNRlzuPioOw4YOvZ7tXzQbjcZrj1CwxSR2sJSYiItKuNBS0+1trRwIYYx4HCoHe1tqykFcm0sacOrQ78TFecpdtbRtBu67oDjD4dOdhLRTmOft1r3kd5v3WeST3d3YwGTIFeh8P3mi3qxYREQmphoJ2Rc0Ta22VMWazQrZI83SI8XJaRnfmLC/k91OHE+1t1Db2kccY8B3jPCbeBnu2BD5MOQcWPw4fPwxxnWHQ6c4OJgMnQYckt6sWEREJuoaCtt8YU9Oj2QAdAq9rtvcLk02BRSJDTqaP/+ZtZcG6XUwc0s3tclpH5x5w7FXOo3w/rH/Hme3+4g1Y/iJ4Ant618x2J2uzIxERaRsaasHuba1CRNqDCYNTSYyLIjd/a/sJ2nXFdoJhOc6jugq2LHWWl6x5A+be5jy6DnUC9+Ap0DMbPPprSEREIlNDM9oiEkSxUV6mjEjnf8sLKauoIi66HYdIjxd6jXEek34Huzd8vV/3wgdhwX0Qn+p0pxwyBfpPdIK6iIhIhFDQFmllOX4fzy/ZxLtrdjB5hDos1kruB8f/2HkcLHG6U37xBqx+DfKeAW8s9DvR2a978BRnSYqIiEgYa1bDmnCkfbQlUlRWVXP83fM5rl8KD182yu1ywl9VBWz8+OvZ7t3rneNpmTDkTCd4p2dp60AREWk1wdpHW0SCLMrr4cyR6bywZBP7yyvpFKs/hkfljYZ+E5zH6X+AXWu/3q/7/T/BezMgwff1EpN+J0F0nNtVi4iI0Eb3FxMJb1P9Psoqqpm3crvbpUQWY6DrYBj/M/jBG3DTOjj3UedDk8tfhGcvhMcmQIV2IRUREfcpaIu4YFTvLvg6x5Gbv9XtUiJbxxTIugQuegpuXg/nPgK7voBFM92uTEREREFbxA0ej+Fsv4/31+6kpPSQ2+W0DVGxkHUpDDwNPrgXSne7XZGIiLRzCtoiLpnq91FRZZm7YpvbpbQtk34HZXthwV/drkRERNo5BW0Rlwz3JdIvtSOztXwkuNJGODPbn8yEko1uVyMiIu2YgraIS4wx5GSm81FBETv26cN7QTXxV84HJ9/+f25XIiIi7ZiCtoiLcvw+qi3MWa7lI0HVuSccdw0sex4Kl7ldjYiItFMK2iIuGtQ9gaFpCVo+Egrjb4QOSTDvt25XIiIi7ZSCtojLcvw+ln5VzObiUrdLaVs6JMGJv4SCt52HiIhIK1PQFnFZTqYPgP8tK3S5kjbo2B9CUm946w6orna7GhERaWcUtEVc1jslnqxeSeQu0/KRoIuKhVPugG3Lnc6RIiIirUhBWyQM5Ph9fL5lL+t37ne7lLZnxHmQ7oe3/6DW7CIi0qoUtEXCwFkj0zEGcvO1fCToPB447U7YsxEW/8PtakREpB1R0BYJA2md4xjTN5nZ+Vuw1rpdTtvT/2QYOAnevxcOFrtdjYiItBMK2iJhYmqWj4KdB1hVuM/tUtqmSb+Hsj3wgVqzi4hI61DQFgkTU0ak4/UYfSgyVNJGgP8S+OQxKNnkdjUiItIOKGiLhInkjjGMH5hKbv5WLR8JlVN+7Xx9R63ZRUQk9BS0RcLIVL+PzcUHydtU4nYpbVPnnnD8NZA/y9nyT0REJIQUtEXCyGnDuxMT5VFL9lAa/3Ona+Rbas0uIiKhpaAtEkYS46KZOKQr/1tWSFW1lo+ERIckmHATFMyHgnfcrkZERNowBW2RMJPj97FjXzmLNux2u5S2a8yP1JpdRERCTkFbJMycOrQ78TFe7T4SSlGxcMpvYNsy+Pwlt6sREZE2SkFbJMx0iPFyWkZ35iwvpKJKs60hM+J8SMuE+XdBZbnb1YiISBukoC0ShnIyfRSXVrBg3S63S2m76rZmX6TW7CIiEnwK2iJhaMLgVBLjosjV7iOhNWAiDDgV3v+zWrOLiEjQKWiLhKHYKC9TRqTz5ortlFVUuV1O23ZaoDX7gvvcrkRERNoYBW2RMJXj97G/vJJ31+xwu5S2LW0k+C+Gjx9Va3YREQkqBW2RMHV8/2RSO8Xw8DsFzM7fys59+sBeyEysac3+R3frEBGRNiXK7QJEpH5RXg8/P20Id7++ihue+wyAwd07ccKAVE4YkMJx/VPo3CHa5SrbiKRecNzVsPBBGHutM8stIiLSQsbattF9Ljs72y5ZssTtMkSCrrKqmhVb97KwoIiFBbtY/OVuyiqq8RgY0aMzYwekcMKAVI7t24X4GP3s3GwHi+GBLOiZDdNedrsaEREJY8aYpdba7AbPU9AWiSzllVXkbSxhYUERHxUU8dmmYiqqLNFeQ1avpNoZ76zeScRGed0uN7IsfBDevB2+91/of7Lb1YiISJhS0BZpJ0oPVbLky+JA8N7F8i17qLYQF+3h2L7JtTPeI3yJRHn1sYyjqiiDh46F+C7wo3edvbZFREQOo6At0k7tOVjBJ+uLame812zfB0BCXBTH9UvhhAEpnDAwhcHdEvB4jMvVhqH85+HV6fDdxyHzArerERGRMKSgLSIA7NxXzse1wXsXXxaVApDSMYbjBzjBe9yAVPqkxGOMgjfV1TDzRGdv7euWQFSs2xWJiEiYUdAWkXptLi7lo8Bs94cFu9i+19k20Nc5jrGB9d0nDEwhvXMHlyt1UcHb8NR34Iw/wtifuF2NiIiEGQVtEWmQtZYNuw7ULjNZWLCL4tIKAPqldgys705hbP8UUjq1s5ndp74DWz+DG/KgQ5Lb1YiISBhR0BaRJquutqzeto+FBbv4qKCITzbsZn95JQBD0xJqdzQZ0z+ZxLg2vod34TJ47EQY91OnTbuIiEiAgraItFhlVTXLt+ypnfFe/OVuyiudPbxH9kxylpkMSCG7TzIdYtrgVoKvXA0rXoUbPoXOPd2uRkREwoSCtogEXVlFFZ9tLOGjgl0sLCgib1MJldWWGK+HY3oH9vAemIK/ZxIxUW1ga7ySjfBgNow8H879u9vViIhImFDQFpGQO1BeyeIvdwfWdxfx+dY9WAvxMV6O7ZscmPFOJcOXiDdStxJ883ZY+BBcswDSRrhdjYiIhAEFbRFpdSWlh/h4/e7aGe+1O/YDkBgXxfH9a/bwTmVQt06Rs5VgbWv2Y2HaS25XIyIiYaCxQTuqNYoRkfYhKT6GySPSmDwiDYAd+8pqtxJcWFDEmyu3A5DaKbZ2ffcJA1LpldwhfIN3hy4w4Rfw1m9g/XvQ/yS3KxIRkQihGW0RaTWbdpfWbiO4sKCIHfucPbx7JHWo3b/7hAGpdE+Mc7nSw1SUwUPZEJ8CP3pHrdlFRNo5LR0RkbBmraVg5wE+KtjFh+uK+Gh9EXsOOnt4D+jasXYrweP7p9ClY4zL1QL5s+DVq+G8fzofjhQRkXZLQVtEIkp1tWVl4d7aGe9FG3Zz4FAVxsCwtMTaGe9j+yaT4MYe3tXVzr7a5XvhusVqzS4i0o4paItIRKuoqmbZ5j0sXOcsM1m6sZhDldV4PQZ/z861M96j+nQhLrqV9vCubc1+N4y9tnXuKSIiYUdBW0TalLKKKj79qpiFgRnv/M17qKq2xER5GN27S+2Md2bPJKK9IVxD/e9zoTBPrdlFRNqxsAjaxpjJwAOAF3jcWjvjsPd/DvwQqAR2Aj+w1n4VeK8KWB44daO1durR7qWgLdK+7C+vZPGG3XwYmPFeWbgXgP5dO/LKj08gKT5E67oL8+Gxk2D8z2DS70JzDxERCWuuB21jjBf4AjgN2AwsBi6x1q6sc85E4BNrbakx5sfAydbaiwLv7bfWdmrs/RS0Rdq33QcO8c7qHdz6yjLGD0zln1cciydUTXJemQ4r/wvXL1VrdhGRdqixQTuUe1SNAdZZa9dbaw8Bs4Bz6p5grX3HWlsaePkxoP9jiUizJHeM4bzRPbnj7AzeWbOTh95ZF7qbnXI72Gp45+7Q3UNERCJeKIN2D2BTndebA8eO5CpgTp3XccaYJcaYj40x59Z3gTFmeuCcJTt37mx5xSIS8aYd34fvHNOD++Z9wXtfhOjvhaTeMGY65D0D21eE5h4iIhLxwqLrgjFmGpAN/LnO4T6BKflLgfuNMQMOv85aO9Nam22tze7atWsrVSsi4cwYwx+/M5Ih3RP46azP2Fxc2vBFzTHhFxCXCPN+F5rxRUQk4oUyaG8BetV53TNw7BuMMZOAXwNTrbXlNcettVsCX9cD7wLHhLBWEWlDOsR4eWTaaKqqLNc+8ynllVXBv0l8shO2174JG94P/vgiIhLxQhm0FwODjDH9jDExwMXA7LonGGOOAR7DCdk76hzvYoyJDTxPBcYBKxERaaR+qR2590I/yzbv4fe5IfrrY8zV0LkXvHWH09BGRESkjpAFbWttJXAdMBdYBbxgrV1hjLnTGFOzVd+fgU7Ai8aYPGNMTRAfBiwxxuQD7wAz6u5WIiLSGGcMT+Oakwbw7CcbeWnp5uDfIDoOJv4atn4GK14J/vgiIhLR1LBGRNq0yqpqLv/nIj7dWMwr157AcF/n4N6gusrZV1ut2UVE2o1w2N5PRMR1UV4PD156DF3iY/jx05+yp7QiuDfweOG030PJV7DkieCOLSIiEU1BW0TavNROsTx82SgK9xzk5y/kUV0d5H/JG3gq9D8Z3vsTlO0J7tgiIhKxFLRFpF0Y3acLt5+VwfzVO3jkvYLg3+C0O+Hgblhwf/DHFhGRiKSgLSLtxvfG9mGq38df3lzDgrW7gjt4uh9GXggf/x32fGsnUxERaYcUtEWk3TDGMOO8kQzs1okbZn3G1pKDwb1BTWv2d/8Y3HFFRCQiKWiLSLsSHxPFo9NGc6iymh8Hu5lNlz6B1uzPwnbtSCoi0t4paItIu9O/ayfuvSCT/E0l3PVakAPxhF9AbIJas4uIiIK2iLRPk0ekc/WJ/Xn644288mkQm9nEJ8P4n8PaubDhg+CNKyIiEUdBW0TarV+eMYTj+iXzq1eXs6pwb/AGPu5qSOwJb/1GrdlFRNoxBW0RabdqmtkkxkXz46eXsudgkJrZRHdwPhi59TNY+WpwxhQRkYijoC0i7Vq3hDj+ftkoNhcf5KYX84PXzCbzQug+AubfCZWHgjOmiIhEFAVtEWn3svsm86szh/HWyu08+n6QmtnUtGYv/lKt2UVE2ikFbRER4MpxfTk7M517567hw3VBamYz4FTodxK8d49as4uItEMK2iIiOM1s7jkvk/5dO3HDc59RuCcIzWyM+bo1+4cPtHw8ERGJKAraIiIBHWOdZjZlFVVc+8ynHKoMwo4hviwYeQF89HfYu7Xl44mISMRQ0BYRqWNgt078+QI/n20s4f/9L0jNbE65HWwVvKPW7CIi7YmCtojIYc4cmc4Px/fjyY++4j+fbWn5gF36wrE/grxnYMeqlo8nIiIRQUFbRKQet0wZypi+ydz2ynLWbNvX8gFPvAli1JpdRKQ9UdAWEalHtNfDQ5ceQ6e4KK55eil7y1rYzCY+GSbcCF+8odbsIiLthIK2iMgRdEuM4+FLR7Fxdym/fDEfa1vYzOa4ayCxB7x1B7R0LBERCXsK2iIiRzGmXzK3TRnK3BXbmfn++pYNVtua/VNYodbsIiJtnYK2iEgDrhrfjzNHpnHPG6v5qKCoZYNlXgTdhsP836s1u4hIG6egLSLSAGMMfzrfT7/Ujlz/3Kds21PW/ME8XqeJTfGXsPRfQatRRETCj4K2iEgjdIqN4rHLR1N6qIqfPNvCZjYDT4V+JwZas+8NXpEiIhJWFLRFRBppYLcE7jkvk6VfFXP3nBbsh13Tmr20SK3ZRUTaMAVtEZEmyPH7uHJcX/714ZfMzm9BS3XfMTDifPjoYbVmFxFpoxS0RUSa6FdnDiO7TxdufXkZa7e3oJnNqb+B6kp49+7gFSciImFDQVtEpImivR4evmwU8TFRXP30UvY1t5lNl74w5kfw2dNqzS4i0gYpaIuINEP3xDgeuvQYvioq5eaXljW/mc2Jv1RrdhGRNkpBW0SkmY7vn8Itk4cw5/NtPP7BhuYNUrc1+5cLglugiIi4SkFbRKQFfjShP5OHpzHjjdV8sr6ZzWzUml1EpE1S0BYRaQFjDH++IJM+yfFc99xn7NjbjGY20R1g4q9hy1JY+Z/gFykiIq5Q0BYRaaGEuGgevXw0+8sq+cmzn1JR1YxmNv6Lndbs89SaXUSkrVDQFhEJgsHdE5hx3kgWf1nMjDmrmz6Axwun/R6KN8DS/wt6fSIi0voUtEVEguScrB58/4S+/HPBBl5b1owmNAMnQd8J8N4MtWYXEWkDFLRFRILoV2cOY1TvJG5+aRnrdjSxmU3d1uwL/xaaAkVEpNUoaIuIBFFMlNPMpkO0l6ufWsr+8sqmDdBjFIw4DxY+BHsLQ1OkiIi0CgVtEZEgS+/c4f+3d+/xcdV1/sdfn5nJvW1aenELaUGlXFq69BIKKhSlpepPRdD9AZXuQ3dd94eIBdmfrj8VEdB1ld1VQfHyAFdQRBFQFBXBLiByb9MCvXCXe4WUXmjTJJPMfH5/nJPMZDJJZ0JOziR5Px+PecyZ7/meM5+c5pG+5zvfmS+XrVzIX7a18a83DGExmxN6lmb/t2gKFBGREaGgLSISgbcePI1Pv/MwfvvwVn549zPlHbzfG+GofwqXZh/CBytFRKQiKGiLiETkzOPfxIq5b+Crv9vCg89sL+/gpZ+G6glaml1EZBRT0BYRiYiZ8R+nHknTlDo+cU0Lr+wuYzGbhqlw7Kfg8d/DM3dHV6SIiERGQVtEJEKTaqv47qrFvNbRxdk/XV/eYjbHfBwm7q+l2UVERikFbRGRiB0+cxJf/cB8HvjLdi75w2OlH1hVByd8Hl5cC5tviq5AERGJhIK2iMgIOGVhE39/zIH84E9P8/tHyvjaviNXwoy5sOZCyHRFV6CIiAw7BW0RkRHyhfcezoJZk/n09Q/zVOue0g5KJGH5hbD9aS3NLiIyyihoi4iMkJpUksvPWER1KsGZP15HW6mL2cw5MVia/Q4tzS4iMpooaIuIjKD9J9dx6ekLeap1D5+98ZHSFrMxgxMvhL3b4J7Loi9SRESGhYK2iMgIO3bONP5lxaH85qGXuOqeZ0o76IDFMO8DcK+WZhcRGS0UtEVEYvDx49/M8sNn8OXfbmHdsyUuZrPs/OADkXd8NdriRERkWChoi4jEIJEw/vPUBRwwpY6zrmmhdXfnvg/a703h0uw/htYyviZQRERioddp030AABY5SURBVKAtIhKTxroqvnvGYna1d/HJa1voLmUxGy3NLiIyaihoi4jEaO7+k/jKyfO57+ntXHJrCaPUDVPh2HPhsd/Bs/dEX6CIiAyZgraISMw+uLiJDx09m+/f+TS3bPzrvg84Wkuzi4iMBgraIiIV4IL3zeXIpkb+7y8e4ul9LWZTXQ/v+By88CBs+fXIFCgiImVT0BYRqQA1qSSXr1pMVdL4+E9a2Jvex2I2Cz4E0w+HP2ppdhGRSqWgLSJSIQ6YXMe3Tl/I46/s5nP7WswmkQwWsdn+lJZmFxGpUAraIiIVZOkh0zlv+SH8asNL/Pi+ZwfvPGcFHHhssDR75+6RKVBEREqmoC0iUmE+8Y6DOeGwGVx882ZantsxcEczOPGiYGn2uy8duQJFRKQkCtoiIhUmkTC+ceoC/qaxlrN+0sK2PYMsZtO0GOadEizNvruEbywREZERE2nQNrN3mdljZvakmX22yP7zzGyzmT1sZmvM7MC8fR82syfC24ejrFNEpNI01geL2ezYm2b1tevJZAeZr73si1qaXUSkAkUWtM0sCXwHeDcwF1hpZnMLuq0Hmt39b4Hrga+Hx+4HXAAcDSwBLjCzKVHVKiJSiY44oJGLTz6Ce556lf8cbDGb/d4ER30UWrQ0u4hIJYlyRHsJ8KS7P+3uaeBnwPvzO7j77e6+N3x4H9AUbr8TuM3dt7v7DuA24F0R1ioiUpFObZ7FyiWzuPyOp7h10yBTQ5Z+Gqrqg6/7ExGRihBl0D4AeD7v8Qth20A+Cvy+nGPN7J/NbK2ZrW1tbX2d5YqIVKYL3jeP+Qc08i/XPcQz29qKd2qYFi7N/lt49t6RLVBERIqqiA9DmtkqoBm4pJzj3P0H7t7s7s3Tp0+PpjgRkZjVViW5/IxFJJPGmT9ZR3s6U7zjMWfBxJlw2/laml1EpAJEGbRfBGblPW4K2/ows+XA54GT3L2znGNFRMaLWfvV883TFvDYy7v5/C8HWMymz9Lsvxn5IkVEpI8og/aDwBwze6OZVQOnA7/O72BmC4HvE4TsV/J2/QFYYWZTwg9BrgjbRETGrbcfOoNzls3hxvUvcs39zxXvdOSHYPph8McvaWl2EZGYRRa03b0bOJsgIG8BrnP3TWZ2kZmdFHa7BJgA/MLMNpjZr8NjtwMXE4T1B4GLwjYRkXFt9QlzePuh07noN5vZ8PzO/h2SKViupdlFRCqBFX37cRRqbm72tWvXxl2GiEjkdu5N855L/4y7c/Pq49ivobpvB3f40Xtg2+Owej3UTIynUBGRMcrM1rl78776VcSHIUVEpHST66v53qrFbGsbYDEbMzjxYmhrhXsui6dIERFR0BYRGY3mNzVy0Unz+POT2/jGbY/379C0GOaeDPdoaXYRkbgoaIuIjFKnL5nNqc1NfPv2J1mz5eX+HZZ9ETKdcMe/j3xxIiKioC0iMppd9P4jmLf/JD718w089+revjunvhmaPwotV0NrkVFvERGJlIK2iMgoVluV5HurFmMWLGbT0VWwmM3xnwmWZl+jpdlFREaagraIyCjXs5jN5q2v8YVfbey7mE3DNDj2HHj0ZnjuvviKFBEZhxS0RUTGgHccNoPVJxzM9ete4NoHnu+785izYMLfwK1aml1EZCQpaIuIjBHnLD+E4+ZM40u/3sRD+YvZVDeES7M/EIxsi4jIiFDQFhEZI5IJ49LTFzJ9Yg1nXdPCjrZ0bueCM2DaoVqaXURkBCloi4iMIVMaqrn8jEW07u7knJ9vyC1mk0zBiRfCq09Cy1XxFikiMk4oaIuIjDFHzprMl06ax58eb+Vba57I7TjkXTD7rcH3anfujq9AEZFxQkFbRGQMWrlkFh9c1MSla57g9kdfCRrNYEXP0uzfjrdAEZFxQEFbRGQMMjO+fPIRHD5zEuf+fAPPbw8Xs2lqDpdmvwx2F1lNUkREho2CtojIGFVXneR7qxaRde+7mE3P0ux3aml2EZEoKWiLiIxhB05t4JunLWDTS69xwU2bgsapb4bmf4R1V8G2JwY/gYiIDFkq7gJERCRayw5/A2e/42C+ffuTLDpwMqcdNRuWfgY2/BRuOhsOeWcwf9sSQHhviVybhWMy/dpsgDYboF+iyPPY4M/dp40Snrvcc4Z9S+oX3ouIlEhBW0RkHPjUiYew4fmdnH/TJubObGR+03RYdgHc8q/wvJZmL5klobYR6qZA3eTgvnZyaY+rauOuXkRGmPkYWY63ubnZ165dG3cZIiIV69U9nbzvsj+TSBg3f/JYJtdXQ3caPAt4cO/ZYJn23jbv29anb2GbD9Cv8JzZvn2L9evTTgnPXe45i/UdqHZybZk0tO+Ejp3QviPYbt8RPt4Z9BlIqra0YF43GWrz9zdCIhnVr4WIDIGZrXP35n3104i2iMg4MXVCDd85YxGnfv9ezv35Bn744aNIpKrjLmvsyGah87W84F0YxPMf74Kdz8HWh4PHXW2Dn7umMQzeZY6iVzdouotIjBS0RUTGkYWzp/DF983j/F9t5LL/eZJzls+Ju6SxI5HIheFydadLCOd5j3e9mHuc7R6kplR5wbx3RH0y6EWYVAJ36O6EdBuk94S3ttz9zAUweVbcVQ5IQVtEZJxZdfRs1j+7g2+ueZz5TZM4/pAZJBMa9YxVqhomzAhu5XAPwkYpo+jtO2DPy9D6KLTvgs5dg5+7qqEgjJc4N71mUvCiQ8an7nRBIG4LVqLt2S5nX3oPdO4Bzwz8fCd/DxasHLmfr0wK2iIi44yZ8ZVT5rN562v844+Cz7ZUJY3aqmR4S1CbCrbrqpLUVCV699XlbdemEtSEfXqP63OfDM+ToK462XvOmlSChIL98DCDmgnBjTJH9TLduakupYyiv/pU7nF3xyA1JYJQXjs5uE/VQqomuCWrw/ua4MVFn/vaIm35x1QX9CncF543qWhTskxXLth2FowUlxWW8x5nu0p//uoJwfSm/Pv6aTDloP7tPds1BcdMnh3Z5RkO+m0UERmH6qqTXP3RJdy0/iXa0t10dGXp6MrQ2Z2hoytLezpDR3eGjq4Muzu6ad3dSUdXsK+nvaMrO+Tnr04lwoCe6BPIawtCe0+Ir+kT/vv3K3yR0BPsa6oS1KQSmOYp95dMQf1+wa1cXe3FPxRaGM47dgWhvLsj2M6kg2kAPffdncHiSd2dDPpB0nJYcvBQn6otHtD79Sn2YiDvmH7HFQn+qdrh+yBrprv/aG+x0d9+YblwX96xmXTpz19V3zf41kwI3sFobOobhPP39wvLeftSdePinQ8FbRGRcWrGxFo+tvRNQz7e3enszuYCeFcQztvTuUDeGe5r78r069eR7h/c27sy7Nyb7tMehP4s6e6hBXszqOkN9rlR9dwoe5GwXmQ0vqYqOEd9dYq66iT1vbcU9dXBi4JxM1JfVRfcJs0cnvO5B3PN80N4pjOYhtATxPu15e3LpMNAX9hWeN8RnmuA4N9z3u4OYgn+iSro2psXiPNGkQd7F6FQqi4XemsmBve1k4J/r+qJAwTihoH3VdXrm2+GSEFbRESGxCw33WQkZLO5YN8nuPcG9bwg35Xr19kVBPWePu1due3OriyvtqXzRvCzve3pTPnBvrYq0Ru866uT1FWnqK9K0lCT2+4J6Q01qTC497SlaMjbzg/ytVVjfFTeDJJVwa0S7DP4h2F80OBfEO6LBv+e84bBP9sdvICpnhDM1+8zXaJgxDj/cU1BQFYorhgK2iIiMiokEkZdGESnjMDzZbKeC/A9AT8dhPe96Qx7O7uD+64M7elu2jp79gXt7ekMbelg39ZdXeHj3L7ubOkjpmb0hvLeEfSeMF+VoqEmt92zr6G6cPQ9/wVA7rGm1hRRacFfRi0FbRERkSKSCaOhJkVDTTT/Vaa7s/3Cd09I3xtuFw/1Gdo6c/u2t7UHQT/vHGVkeBLGwIG8Khx5r05SX5UbpW+oSfZOo8kF+/5TaqpTY38OrshgFLRFRERiUJ1KUJ1K0Fg/vKOmPXPne8N6b3DvO9q+t084z9DeFewLRua72dMZfAi2reAc5UgYJIqMlg80gG4U2TFg3yJtZZy3WN+BxvWLjfgP+B7A6zxvMmEkLNiXMEiaBduJYDthhoXXNZmwXL/87aL9ctt9zt97nJEM+xQ9Z2KQc1runH3PbyQTuXMmLXhnqlidxc7fe87w58/Vmdt38IwJTJ9YM9C/RuwUtEVERMaQ/Lnz+zUM76Iz7k5HV7YgfBeMwudtd3Rl8YIPFfoAo+3FmgfuW2RHWeft3xpVXQP1LXq8O1mHbM991nPb7kXaB+mXDba7s1nSmbA9O9D5ix9b9PwD1eRe1s86XP7r1CP5wKKmkX/iEiloi4iISEnMcvPkRQr5YIG9MLwP9kIh7wVBJhsE+Kw7Gffcc2SDxwfPmBD3jz0oBW0RERERed16pockB55YM+7oUwoiIiIiIhFQ0BYRERERiYCCtoiIiIhIBBS0RUREREQioKAtIiIiIhIBBW0RERERkQgoaIuIiIiIREBBW0REREQkAgraIiIiIiIRUNAWEREREYmAgraIiIiISAQUtEVEREREIqCgLSIiIiISAQVtEREREZEIKGiLiIiIiETA3D3uGoaFmbUCz8b09NOAbTE992ik61UeXa/y6HqVR9erPLpe5dH1Kp+uWXniul4Huvv0fXUaM0E7Tma21t2b465jtND1Ko+uV3l0vcqj61UeXa/y6HqVT9esPJV+vTR1REREREQkAgraIiIiIiIRUNAeHj+Iu4BRRterPLpe5dH1Ko+uV3l0vcqj61U+XbPyVPT10hxtEREREZEIaERbRERERCQCCtoiIiIiIhFQ0H4dzOyHZvaKmW2Mu5bRwMxmmdntZrbZzDaZ2Tlx11TJzKzWzB4ws4fC63Vh3DVVOjNLmtl6M7s57lpGAzN7xsweMbMNZrY27noqnZlNNrPrzexRM9tiZm+Ju6ZKZWaHhr9XPbfXzOzcuOuqZGb2qfBv/UYzu9bMauOuqZKZ2TnhtdpUyb9bmqP9OpjZUmAPcLW7HxF3PZXOzGYCM929xcwmAuuAk919c8ylVSQzM6DB3feYWRXwZ+Acd78v5tIqlpmdBzQDk9z9vXHXU+nM7Bmg2d21OEYJzOwq4C53v8LMqoF6d98Zd12VzsySwIvA0e4e18JyFc3MDiD4Gz/X3dvN7Drgd+7+o3grq0xmdgTwM2AJkAZuAc509ydjLawIjWi/Du7+J2B73HWMFu6+1d1bwu3dwBbggHirqlwe2BM+rApvemU8ADNrAt4DXBF3LTL2mFkjsBS4EsDd0wrZJVsGPKWQvU8poM7MUkA98FLM9VSyw4H73X2vu3cDdwIfiLmmohS0JRZmdhCwELg/3koqWzgVYgPwCnCbu+t6DeybwGeAbNyFjCIO3Gpm68zsn+MupsK9EWgF/jucnnSFmTXEXdQocTpwbdxFVDJ3fxH4D+A5YCuwy91vjbeqirYROM7MpppZPfC/gFkx11SUgraMODObANwAnOvur8VdTyVz94y7LwCagCXh22VSwMzeC7zi7uvirmWUOdbdFwHvBj4RToeT4lLAIuC77r4QaAM+G29JlS+cYnMS8Iu4a6lkZjYFeD/BC7r9gQYzWxVvVZXL3bcAXwNuJZg2sgHIxFrUABS0ZUSFc41vAK5x9xvjrme0CN+ivh14V9y1VKi3ASeFc45/BpxgZj+Jt6TKF46i4e6vAL8kmO8oxb0AvJD3rtL1BMFbBvduoMXdX467kAq3HPiLu7e6exdwI/DWmGuqaO5+pbsvdvelwA7g8bhrKkZBW0ZM+OG+K4Et7v5fcddT6cxsuplNDrfrgBOBR+OtqjK5+/9z9yZ3P4jgber/cXeNBg3CzBrCDyUTToFYQfB2rBTh7n8FnjezQ8OmZYA+yL1vK9G0kVI8BxxjZvXh/5XLCD7HJAMwsxnh/WyC+dk/jbei4lJxFzCamdm1wNuBaWb2AnCBu18Zb1UV7W3A3wOPhPOOAT7n7r+LsaZKNhO4KvzEfgK4zt31tXUyXN4A/DL4P50U8FN3vyXekireJ4FrwukQTwP/EHM9FS18AXci8H/irqXSufv9ZnY90AJ0A+up8KXFK8ANZjYV6AI+UakfTtbX+4mIiIiIREBTR0REREREIqCgLSIiIiISAQVtEREREZEIKGiLiIiIiERAQVtEREREJAIK2iIiw8DMMma2wcw2mtkvwmWBB+u/Z6RqK3jeZjO7dB993m5m/b5KMmzfFS5B/piZ/SlclVNERIpQ0BYRGR7t7r7A3Y8A0sCZcRdUjLuvdffVr+MUd7n7Qnc/FFgNfNvMlr3euiyg/5NEZEzRHzURkeF3F3AwgJmdF45ybzSzcws7mtnVZnZy3uNrzOz9ZvYRM7vRzG4xsyfM7Ot5fVaa2SPhOb+W177HzC4xs01m9kczW2Jmd5jZ02Z2Utind7Q63H9vOEJ9T96qhyVx9w3ARcDZ4fmmm9kNZvZgeHtbXvttYV1XmNmzZjbNzA4KR8avJliVcpaZrQhragnfGZgQnmOxmd1pZuvM7A9mNrOcWkVE4qCgLSIyjMwsBbybYAXUxQSrBx4NHAN8zMwWFhxyJfCR8NhG4K3Ab8N9C4DTgPnAaWY2y8z2B74GnBDuPyovqDcQLD8/D9gNfJlgZb5TCAJxoUeB49x9IfBF4N+G8CO3AIeF298CvuHuRwEfBK4I2y/Iq+t6YHbe8XOAy8N9bcAXgOXuvghYC5xnZlXAZcDfufti4IfAV4ZQq4jIiNIS7CIiw6POzDaE23cRBOiPA7909zYAM7sROI5geWUA3P1OM7vczKYThNMb3L07XBp9jbvvCo/dDBwITAXucPfWsP0aYCnwK4IpKz3LqD8CdLp7l5k9AhxUpOZG4CozmwM4UDWEn9vytpcDc8PaASaFI9LHEoR93P0WM9uRd8yz7n5fuH0MMBe4OzxHNXAvcChwBHBb2J4Etg6hVhGREaWgLSIyPNrdfUF+Q17g3JergVXA6QQj4D0687Yz7Ptvdpe7e7id7Tne3bPhSHuhi4Hb3f0UMzsIuKPUgvMsBLaE2wngGHfvyO+wj+vQlt8VuM3dVxYcPx/Y5O5vGUJ9IiKx0dQREZHo3AWcbGb1ZtZAMKp7V5F+PwLOBXD3zfs45wPA8eEc5ySwErhziPU1Ai+G2x8p92Az+1vgfOA7YdOtwCfz9ve88LgbODVsWwFMGeCU9wFvM7Oe+e0NZnYI8Bgw3czeErZXmdm8cusVERlpCtoiIhFx9xaCEP0AcD9whbuvL9LvZYJR4f8u4Zxbgc8CtwMPAevc/aYhlvh14Ktmtp7S3+E8rufr/QgC9mp3XxPuWw00m9nD4VSXnm9euRBYYWYbgf8N/JVgDnnhz9ZKEPivNbOHCaaNHObuaeDvgK+Z2UPABoK57CIiFc1y7zKKiEgcwu/cfgRY1DMneywxsxogE849fwvw3cJpNiIiY5HmaIuIxMjMlhN8cPIbYzFkh2YD14Xfk50GPhZzPSIiI0Ij2iIiIiIiEdAcbRERERGRCChoi4iIiIhEQEFbRERERCQCCtoiIiIiIhFQ0BYRERERicD/BxaNOdGR5mUzAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Plot the two RMSE of the training and the test data\n",
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df.degree, rmse_df.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df.degree, rmse_df.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 214,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10ce9cb10>"
+      ]
+     },
+     "execution_count": 214,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# It can observed our model starts to overfit ... as we would expect with a small dataset ... around 5\n",
+    "# Here there is no real increase or trade off between the error terms after 5\n",
+    "lm_model_5 = LinearRegression()\n",
+    "lm_model_5.fit(vander(x, 6), y_scatter)\n",
+    "degree_5 = lm_model_5.coef_.size - 1\n",
+    "y_pred_5 = lm_model_5.predict(np.vander(x, degree_5 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_5)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 5', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 332,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Scatter Curve To Estimate')"
+      ]
+     },
+     "execution_count": 332,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Let's try a different set of data\n",
+    "def curve(x_val):\n",
+    "    return (x_val ** 2) + 3.0\n",
+    "\n",
+    "y_scatter_curve = curve(x)\n",
+    "noise = 20 * (np.random.random(length) - 4.0)\n",
+    "y_scatter_curve = y_scatter_curve + noise\n",
+    "\n",
+    "plt.figure(figsize=(12, 7)) \n",
+    "plt.scatter(x, y_scatter_curve)\n",
+    "plt.title(\"Scatter Curve To Estimate\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 333,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "rmse_df_curve = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter_curve,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df_curve.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 334,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 334,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df_curve.degree, rmse_df_curve.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df_curve.degree, rmse_df_curve.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 335,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x110236a10>"
+      ]
+     },
+     "execution_count": 335,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Here we see a divergence ... so we'll stick to Poly level 2\n",
+    "# It can observed our model starts to overfit ... as we would expect .. around 5\n",
+    "# Here there is no real increase or trade off between the error terms after 5\n",
+    "# Mostly because we're extimating a cos wave, and it will be roughly the same \n",
+    "lm_model_2 = LinearRegression()\n",
+    "lm_model_2.fit(vander(x, 3), y_scatter_curve)\n",
+    "degree_2 = lm_model_2.coef_.size - 1\n",
+    "y_pred_2 = lm_model_2.predict(np.vander(x, degree_2 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y_pred_2)\n",
+    "plt.scatter(x, y_scatter_curve)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['Pred. Deg. 2', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 338,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Scatter With More Noise')"
+      ]
+     },
+     "execution_count": 338,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Add some more bumps to the data\n",
+    "\n",
+    "# Make a copy\n",
+    "y_scatter_curve_noise = y_scatter_curve.copy()\n",
+    "\n",
+    "for i, j in enumerate(y_scatter_curve_noise):\n",
+    "    if i < 25:\n",
+    "        y_scatter_curve_noise[i] = j + (5 * (np.random.random() + 5.0))\n",
+    "    if i > 100:\n",
+    "        y_scatter_curve_noise[i] = j + (5 * (np.random.random() - 5.0))\n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.scatter(x, y_scatter_curve_noise)\n",
+    "plt.title(\"Scatter With More Noise\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 385,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create a dataframe to captire the rmse\n",
+    "rmse_df_curve_noise = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter_curve_noise,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df_curve_noise.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 387,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 387,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "rmse_df_curve_noise\n",
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df_curve_noise.degree, rmse_df_curve_noise.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df_curve_noise.degree, rmse_df_curve_noise.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 419,
+   "metadata": {
+    "code_folding": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Create graphs of the different values of alpha you're testing\n",
+    "from sklearn.linear_model import Ridge\n",
+    "\n",
+    "# Set the expected values of lambda\n",
+    "alphas = np.linspace(.00001, 2, 500)\n",
+    "alphas_to_display = [alphas[0], alphas[100], alphas[200], alphas[300], alphas[400], alphas[499]]\n",
+    "ridge_error_df = pd.DataFrame(columns=[\"alpha\", \"rss\", \"intercept\", \"coef\"])\n",
+    "\n",
+    "def ridge_model_comparison(alphas, poly_degree, X_values, y_values):\n",
+    "    # Set local variables\n",
+    "    count = 0\n",
+    "    subplot = 1\n",
+    "    fig = plt.figure(figsize=(12, 8))\n",
+    "    \n",
+    "    # Construct your model to evaluate\n",
+    "    for i in alphas:\n",
+    "        ridge_model = Ridge(alpha=i, normalize=True)\n",
+    "        ridge_model.fit(vander(x, poly_degree + 1), y_values)\n",
+    "        ridge_degree = ridge_model.coef_.size - 1\n",
+    "        y_pred = ridge_model.predict(np.vander(x, ridge_degree + 1))\n",
+    "\n",
+    "        # Only display certain models\n",
+    "        if i in alphas_to_display:\n",
+    "            plt.subplot(230 + subplot)\n",
+    "            plt.tight_layout()\n",
+    "            plt.plot(X_values, y_pred)\n",
+    "            plt.plot(X_values, y_values, '.')\n",
+    "            plt.title('Plot for alpha: %.3g on Poly. Deg %d ' % (i, poly_degree))\n",
+    "            subplot = subplot + 1\n",
+    "\n",
+    "        # Fill dataframe\n",
+    "        rss = sum((y_pred - y_values)**2)\n",
+    "        intercept = ridge_model.intercept_\n",
+    "        coef = ridge_model.coef_\n",
+    "\n",
+    "        # Add error data to the dataframe\n",
+    "        # alpha, rss, intercept, coef\n",
+    "        ridge_error_df.loc[count] = [i, rss, intercept, coef]\n",
+    "        count = count + 1\n",
+    "\n",
+    "# Run the model\n",
+    "ridge_model_comparison(alphas=alphas, poly_degree=4, X_values=x, y_values=y_scatter_curve_noise);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 420,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Try running the model for a Lasso regression\n",
+    "from sklearn.linear_model import Lasso\n",
+    "\n",
+    "def lasso_model_comparison(alphas, poly_degree, X_values, y_values):\n",
+    "    lasso_error_df = pd.DataFrame(columns=[\"alpha\", \"rss\", \"intercept\", \"coef\"])\n",
+    "    # Set local variables\n",
+    "    count = 0\n",
+    "    subplot = 1\n",
+    "    fig = plt.figure(figsize=(12, 8))\n",
+    "    \n",
+    "    # Construct your model to evaluate\n",
+    "    for i in alphas:\n",
+    "        lasso_model = Lasso(alpha=i, normalize=True)\n",
+    "        lasso_model.fit(vander(x, poly_degree + 1), y_values)\n",
+    "        lasso_degree = lasso_model.coef_.size - 1\n",
+    "        y_pred = lasso_model.predict(np.vander(x, lasso_degree + 1))\n",
+    "\n",
+    "        # Only display certain models\n",
+    "        if i in alphas_to_display:\n",
+    "            plt.subplot(230 + subplot)\n",
+    "            plt.tight_layout()\n",
+    "            plt.plot(X_values, y_pred)\n",
+    "            plt.plot(X_values, y_values, '.')\n",
+    "            plt.title('Plot for alpha: %.3g on Poly. Deg %d ' % (i, poly_degree))\n",
+    "            subplot = subplot + 1\n",
+    "\n",
+    "        # Fill dataframe\n",
+    "        rss = sum((y_pred - y_values)**2)\n",
+    "        intercept = lasso_model.intercept_\n",
+    "        coef = lasso_model.coef_\n",
+    "\n",
+    "        # Add error data to the dataframe\n",
+    "        # alpha, rss, intercept, coef\n",
+    "        lasso_error_df.loc[count] = [i, rss, intercept, coef]\n",
+    "        count = count + 1\n",
+    "\n",
+    "# Run the function\n",
+    "lasso_model_comparison(alphas=alphas, poly_degree=4, X_values=x, y_values=y_scatter_curve_noise);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 422,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1e-05\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "51.552909041605204"
+      ]
+     },
+     "execution_count": 422,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Let's try RidgeCV to search for the correct alpha\n",
+    "from sklearn.metrics import mean_squared_error\n",
+    "\n",
+    "alphas = np.linspace(.00001, 2, 500)\n",
+    "\n",
+    "# Do the searching for us\n",
+    "ridgecv = RidgeCV(alphas = alphas, scoring = 'neg_mean_squared_error', normalize = True)\n",
+    "ridgecv.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "print ridgecv.alpha_\n",
+    "\n",
+    "lm_ridge = Ridge(alpha = ridgecv.alpha_)\n",
+    "lm_ridge.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "mean_squared_error(y_scatter_curve_noise, lm_ridge.predict(vander(x, 6)))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 423,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.733473266533066\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "80.92995386478391"
+      ]
+     },
+     "execution_count": 423,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Let's try RidgeCV to search for the correct alpha\n",
+    "from sklearn.linear_model import LassoCV\n",
+    "\n",
+    "# Do the searching for us\n",
+    "lassocv = LassoCV(alphas = alphas, normalize = True)\n",
+    "lassocv.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "print lassocv.alpha_\n",
+    "\n",
+    "lm_lasso = Lasso(alpha = lassocv.alpha_)\n",
+    "lm_lasso.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "mean_squared_error(y_scatter_curve_noise, lm_lasso.predict(vander(x, 6)))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 424,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "51.552909040751885"
+      ]
+     },
+     "execution_count": 424,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Do the searching for us\n",
+    "final_lm_model = LinearRegression()\n",
+    "final_lm_model.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "final_lm_degree = final_lm_model.coef_.size - 1\n",
+    "final_lm_y_pred = final_lm_model.predict(np.vander(x, final_lm_degree + 1))\n",
+    "\n",
+    "mean_squared_error(y_scatter_curve_noise, final_lm_y_pred)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Ridge at 5 = 51.552909041605204 with alpha 1e-05 <- Tied\n",
+    "# Lasso at 5 = 80.92995386478393\n",
+    "# Linear at 5 = 51.552909040751885 <- Tied\n",
+    "\n",
+    "# Ridge at 2 = 90\n",
+    "# Lasso at 2 = 90"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 434,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10fb9d410>"
+      ]
+     },
+     "execution_count": 434,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Let's plot the comparison\n",
+    "\n",
+    "final_ridge_model = Ridge(alpha=1e-05, normalize=True)\n",
+    "final_ridge_model.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "final_ridge_degree = final_ridge_model.coef_.size - 1\n",
+    "final_y_pred = final_ridge_model.predict(np.vander(x, final_ridge_degree + 1))\n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.scatter(x, y_scatter_curve_noise)\n",
+    "plt.plot(x, final_y_pred)\n",
+    "plt.plot(x, final_lm_y_pred)\n",
+    "plt.plot(x, lm_lasso.predict(vander(x, 6)))\n",
+    "plt.title(\"Linear vs Ridge Model - Poly Deg 5\")\n",
+    "plt.legend(['Ridge Model alpha = 1e-05', 'Linear Model', 'Lasso Model', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.12"
+  },
+  "varInspector": {
+   "cols": {
+    "lenName": 16,
+    "lenType": 16,
+    "lenVar": 40
+   },
+   "kernels_config": {
+    "python": {
+     "delete_cmd_postfix": "",
+     "delete_cmd_prefix": "del ",
+     "library": "var_list.py",
+     "varRefreshCmd": "print(var_dic_list())"
+    },
+    "r": {
+     "delete_cmd_postfix": ") ",
+     "delete_cmd_prefix": "rm(",
+     "library": "var_list.r",
+     "varRefreshCmd": "cat(var_dic_list()) "
+    }
+   },
+   "types_to_exclude": [
+    "module",
+    "function",
+    "builtin_function_or_method",
+    "instance",
+    "_Feature"
+   ],
+   "window_display": false
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/.ipynb_checkpoints/wine-checkpoint.ipynb b/.ipynb_checkpoints/wine-checkpoint.ipynb
new file mode 100644
index 0000000..8367310
--- /dev/null
+++ b/.ipynb_checkpoints/wine-checkpoint.ipynb
@@ -0,0 +1,1034 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Data is from here\n",
+    "# Using the white wine dataset\n",
+    "#http://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import io\n",
+    "import requests\n",
+    "\n",
+    "# Define URL\n",
+    "url = \"http://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-white.csv\"\n",
+    "content = requests.get(url).content\n",
+    "df = pd.read_csv(io.StringIO(content.decode('utf-8')), sep=';')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "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>fixed acidity</th>\n",
+       "      <th>volatile acidity</th>\n",
+       "      <th>citric acid</th>\n",
+       "      <th>residual sugar</th>\n",
+       "      <th>chlorides</th>\n",
+       "      <th>free sulfur dioxide</th>\n",
+       "      <th>total sulfur dioxide</th>\n",
+       "      <th>density</th>\n",
+       "      <th>pH</th>\n",
+       "      <th>sulphates</th>\n",
+       "      <th>alcohol</th>\n",
+       "      <th>quality</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>0.36</td>\n",
+       "      <td>20.7</td>\n",
+       "      <td>0.045</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>170.0</td>\n",
+       "      <td>1.0010</td>\n",
+       "      <td>3.00</td>\n",
+       "      <td>0.45</td>\n",
+       "      <td>8.8</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>6.3</td>\n",
+       "      <td>0.30</td>\n",
+       "      <td>0.34</td>\n",
+       "      <td>1.6</td>\n",
+       "      <td>0.049</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>132.0</td>\n",
+       "      <td>0.9940</td>\n",
+       "      <td>3.30</td>\n",
+       "      <td>0.49</td>\n",
+       "      <td>9.5</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>8.1</td>\n",
+       "      <td>0.28</td>\n",
+       "      <td>0.40</td>\n",
+       "      <td>6.9</td>\n",
+       "      <td>0.050</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>97.0</td>\n",
+       "      <td>0.9951</td>\n",
+       "      <td>3.26</td>\n",
+       "      <td>0.44</td>\n",
+       "      <td>10.1</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>7.2</td>\n",
+       "      <td>0.23</td>\n",
+       "      <td>0.32</td>\n",
+       "      <td>8.5</td>\n",
+       "      <td>0.058</td>\n",
+       "      <td>47.0</td>\n",
+       "      <td>186.0</td>\n",
+       "      <td>0.9956</td>\n",
+       "      <td>3.19</td>\n",
+       "      <td>0.40</td>\n",
+       "      <td>9.9</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>7.2</td>\n",
+       "      <td>0.23</td>\n",
+       "      <td>0.32</td>\n",
+       "      <td>8.5</td>\n",
+       "      <td>0.058</td>\n",
+       "      <td>47.0</td>\n",
+       "      <td>186.0</td>\n",
+       "      <td>0.9956</td>\n",
+       "      <td>3.19</td>\n",
+       "      <td>0.40</td>\n",
+       "      <td>9.9</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   fixed acidity  volatile acidity  citric acid  residual sugar  chlorides  \\\n",
+       "0            7.0              0.27         0.36            20.7      0.045   \n",
+       "1            6.3              0.30         0.34             1.6      0.049   \n",
+       "2            8.1              0.28         0.40             6.9      0.050   \n",
+       "3            7.2              0.23         0.32             8.5      0.058   \n",
+       "4            7.2              0.23         0.32             8.5      0.058   \n",
+       "\n",
+       "   free sulfur dioxide  total sulfur dioxide  density    pH  sulphates  \\\n",
+       "0                 45.0                 170.0   1.0010  3.00       0.45   \n",
+       "1                 14.0                 132.0   0.9940  3.30       0.49   \n",
+       "2                 30.0                  97.0   0.9951  3.26       0.44   \n",
+       "3                 47.0                 186.0   0.9956  3.19       0.40   \n",
+       "4                 47.0                 186.0   0.9956  3.19       0.40   \n",
+       "\n",
+       "   alcohol  quality  \n",
+       "0      8.8        6  \n",
+       "1      9.5        6  \n",
+       "2     10.1        6  \n",
+       "3      9.9        6  \n",
+       "4      9.9        6  "
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "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>fixed acidity</th>\n",
+       "      <th>volatile acidity</th>\n",
+       "      <th>citric acid</th>\n",
+       "      <th>residual sugar</th>\n",
+       "      <th>chlorides</th>\n",
+       "      <th>free sulfur dioxide</th>\n",
+       "      <th>total sulfur dioxide</th>\n",
+       "      <th>density</th>\n",
+       "      <th>pH</th>\n",
+       "      <th>sulphates</th>\n",
+       "      <th>alcohol</th>\n",
+       "      <th>quality</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>count</th>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "      <td>4898.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>mean</th>\n",
+       "      <td>6.854788</td>\n",
+       "      <td>0.278241</td>\n",
+       "      <td>0.334192</td>\n",
+       "      <td>6.391415</td>\n",
+       "      <td>0.045772</td>\n",
+       "      <td>35.308085</td>\n",
+       "      <td>138.360657</td>\n",
+       "      <td>0.994027</td>\n",
+       "      <td>3.188267</td>\n",
+       "      <td>0.489847</td>\n",
+       "      <td>10.514267</td>\n",
+       "      <td>5.877909</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>std</th>\n",
+       "      <td>0.843868</td>\n",
+       "      <td>0.100795</td>\n",
+       "      <td>0.121020</td>\n",
+       "      <td>5.072058</td>\n",
+       "      <td>0.021848</td>\n",
+       "      <td>17.007137</td>\n",
+       "      <td>42.498065</td>\n",
+       "      <td>0.002991</td>\n",
+       "      <td>0.151001</td>\n",
+       "      <td>0.114126</td>\n",
+       "      <td>1.230621</td>\n",
+       "      <td>0.885639</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>min</th>\n",
+       "      <td>3.800000</td>\n",
+       "      <td>0.080000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.600000</td>\n",
+       "      <td>0.009000</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>9.000000</td>\n",
+       "      <td>0.987110</td>\n",
+       "      <td>2.720000</td>\n",
+       "      <td>0.220000</td>\n",
+       "      <td>8.000000</td>\n",
+       "      <td>3.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25%</th>\n",
+       "      <td>6.300000</td>\n",
+       "      <td>0.210000</td>\n",
+       "      <td>0.270000</td>\n",
+       "      <td>1.700000</td>\n",
+       "      <td>0.036000</td>\n",
+       "      <td>23.000000</td>\n",
+       "      <td>108.000000</td>\n",
+       "      <td>0.991723</td>\n",
+       "      <td>3.090000</td>\n",
+       "      <td>0.410000</td>\n",
+       "      <td>9.500000</td>\n",
+       "      <td>5.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>50%</th>\n",
+       "      <td>6.800000</td>\n",
+       "      <td>0.260000</td>\n",
+       "      <td>0.320000</td>\n",
+       "      <td>5.200000</td>\n",
+       "      <td>0.043000</td>\n",
+       "      <td>34.000000</td>\n",
+       "      <td>134.000000</td>\n",
+       "      <td>0.993740</td>\n",
+       "      <td>3.180000</td>\n",
+       "      <td>0.470000</td>\n",
+       "      <td>10.400000</td>\n",
+       "      <td>6.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>75%</th>\n",
+       "      <td>7.300000</td>\n",
+       "      <td>0.320000</td>\n",
+       "      <td>0.390000</td>\n",
+       "      <td>9.900000</td>\n",
+       "      <td>0.050000</td>\n",
+       "      <td>46.000000</td>\n",
+       "      <td>167.000000</td>\n",
+       "      <td>0.996100</td>\n",
+       "      <td>3.280000</td>\n",
+       "      <td>0.550000</td>\n",
+       "      <td>11.400000</td>\n",
+       "      <td>6.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>max</th>\n",
+       "      <td>14.200000</td>\n",
+       "      <td>1.100000</td>\n",
+       "      <td>1.660000</td>\n",
+       "      <td>65.800000</td>\n",
+       "      <td>0.346000</td>\n",
+       "      <td>289.000000</td>\n",
+       "      <td>440.000000</td>\n",
+       "      <td>1.038980</td>\n",
+       "      <td>3.820000</td>\n",
+       "      <td>1.080000</td>\n",
+       "      <td>14.200000</td>\n",
+       "      <td>9.000000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       fixed acidity  volatile acidity  citric acid  residual sugar  \\\n",
+       "count    4898.000000       4898.000000  4898.000000     4898.000000   \n",
+       "mean        6.854788          0.278241     0.334192        6.391415   \n",
+       "std         0.843868          0.100795     0.121020        5.072058   \n",
+       "min         3.800000          0.080000     0.000000        0.600000   \n",
+       "25%         6.300000          0.210000     0.270000        1.700000   \n",
+       "50%         6.800000          0.260000     0.320000        5.200000   \n",
+       "75%         7.300000          0.320000     0.390000        9.900000   \n",
+       "max        14.200000          1.100000     1.660000       65.800000   \n",
+       "\n",
+       "         chlorides  free sulfur dioxide  total sulfur dioxide      density  \\\n",
+       "count  4898.000000          4898.000000           4898.000000  4898.000000   \n",
+       "mean      0.045772            35.308085            138.360657     0.994027   \n",
+       "std       0.021848            17.007137             42.498065     0.002991   \n",
+       "min       0.009000             2.000000              9.000000     0.987110   \n",
+       "25%       0.036000            23.000000            108.000000     0.991723   \n",
+       "50%       0.043000            34.000000            134.000000     0.993740   \n",
+       "75%       0.050000            46.000000            167.000000     0.996100   \n",
+       "max       0.346000           289.000000            440.000000     1.038980   \n",
+       "\n",
+       "                pH    sulphates      alcohol      quality  \n",
+       "count  4898.000000  4898.000000  4898.000000  4898.000000  \n",
+       "mean      3.188267     0.489847    10.514267     5.877909  \n",
+       "std       0.151001     0.114126     1.230621     0.885639  \n",
+       "min       2.720000     0.220000     8.000000     3.000000  \n",
+       "25%       3.090000     0.410000     9.500000     5.000000  \n",
+       "50%       3.180000     0.470000    10.400000     6.000000  \n",
+       "75%       3.280000     0.550000    11.400000     6.000000  \n",
+       "max       3.820000     1.080000    14.200000     9.000000  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.describe()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The columns in the dataframe are:\n",
+      "fixed acidity\n",
+      "volatile acidity\n",
+      "citric acid\n",
+      "residual sugar\n",
+      "chlorides\n",
+      "free sulfur dioxide\n",
+      "total sulfur dioxide\n",
+      "density\n",
+      "pH\n",
+      "sulphates\n",
+      "alcohol\n",
+      "quality\n",
+      "\n",
+      "There are 12 columns in the dataframe\n",
+      "There are 4898 rows in the dataframe\n"
+     ]
+    }
+   ],
+   "source": [
+    "features = df.columns[:-1]\n",
+    "features.remove('ala@ala.com')\n",
+    "\n",
+    "print \"The columns in the dataframe are:\"\n",
+    "for x in df.columns:\n",
+    "    print x\n",
+    "\n",
+    "print \"\\nThere are %d columns in the dataframe\" % len(df.columns)\n",
+    "print \"There are %d rows in the dataframe\" % df.shape[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Coefficients: \n",
+      "fixed acidity             0.0774\n",
+      "volatile acidity          -1.7876\n",
+      "citric acid               -0.0644\n",
+      "residual sugar            0.0790\n",
+      "chlorides                 -0.3791\n",
+      "free sulfur dioxide       0.0038\n",
+      "total sulfur dioxide      -0.0002\n",
+      "density                   -138.8416\n",
+      "pH                        0.5825\n",
+      "sulphates                 0.6213\n",
+      "alcohol                   0.2154\n",
+      "Mean squared error: 0.55\n",
+      "\n",
+      "                            OLS Regression Results                            \n",
+      "==============================================================================\n",
+      "Dep. Variable:                quality   R-squared:                       0.276\n",
+      "Model:                            OLS   Adj. R-squared:                  0.273\n",
+      "Method:                 Least Squares   F-statistic:                     118.2\n",
+      "Date:                Sun, 06 May 2018   Prob (F-statistic):          1.37e-229\n",
+      "Time:                        14:12:11   Log-Likelihood:                -3898.6\n",
+      "No. Observations:                3428   AIC:                             7821.\n",
+      "Df Residuals:                    3416   BIC:                             7895.\n",
+      "Df Model:                          11                                         \n",
+      "Covariance Type:            nonrobust                                         \n",
+      "========================================================================================\n",
+      "                           coef    std err          t      P>|t|      [0.025      0.975]\n",
+      "----------------------------------------------------------------------------------------\n",
+      "const                  138.8628     21.284      6.524      0.000      97.132     180.594\n",
+      "fixed acidity            0.0774      0.025      3.126      0.002       0.029       0.126\n",
+      "volatile acidity        -1.7876      0.135    -13.205      0.000      -2.053      -1.522\n",
+      "citric acid             -0.0644      0.114     -0.564      0.573      -0.288       0.159\n",
+      "residual sugar           0.0790      0.009      9.049      0.000       0.062       0.096\n",
+      "chlorides               -0.3791      0.692     -0.548      0.584      -1.736       0.978\n",
+      "free sulfur dioxide      0.0038      0.001      3.795      0.000       0.002       0.006\n",
+      "total sulfur dioxide    -0.0002      0.000     -0.473      0.636      -0.001       0.001\n",
+      "density               -138.8416     21.602     -6.427      0.000    -181.197     -96.487\n",
+      "pH                       0.5825      0.124      4.690      0.000       0.339       0.826\n",
+      "sulphates                0.6213      0.121      5.143      0.000       0.384       0.858\n",
+      "alcohol                  0.2154      0.028      7.789      0.000       0.161       0.270\n",
+      "==============================================================================\n",
+      "Omnibus:                       91.209   Durbin-Watson:                   2.014\n",
+      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              219.152\n",
+      "Skew:                           0.039   Prob(JB):                     2.58e-48\n",
+      "Kurtosis:                       4.236   Cond. No.                     3.52e+05\n",
+      "==============================================================================\n",
+      "\n",
+      "Warnings:\n",
+      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+      "[2] The condition number is large, 3.52e+05. This might indicate that there are\n",
+      "strong multicollinearity or other numerical problems.\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Start with Linear Regression\n",
+    "from sklearn import model_selection\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.metrics import mean_squared_error\n",
+    "import statsmodels.api as sm\n",
+    "\n",
+    "# Split features from output\n",
+    "X = df.iloc[:,:11]\n",
+    "y = df.iloc[:,11:]\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1075)\n",
+    "\n",
+    "# Do a regular regression\n",
+    "lm_model = LinearRegression()\n",
+    "lm_results = lm_model.fit(X_train, y_train)\n",
+    "\n",
+    "# Make predictions\n",
+    "y_pred = lm_model.predict(X_test)\n",
+    "\n",
+    "# See the results\n",
+    "# The coefficients are\n",
+    "print \"Coefficients: \\n\", \n",
+    "for i in range(len(features)):\n",
+    "    print \"%-25s %.4f\" % (features[i], lm_results.coef_[0][i])\n",
+    "\n",
+    "# The mean squared error\n",
+    "print \"Mean squared error: %.2f\\n\" % mean_squared_error(y_test, y_pred)\n",
+    "\n",
+    "X_train_with_constant = sm.add_constant(X_train)\n",
+    "lm_est = sm.OLS(y_train, X_train_with_constant)\n",
+    "lm_est_results = lm_est.fit()\n",
+    "print(lm_est_results.summary())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Predicted Vs. Actual')"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.figure(figsize=(12,8))\n",
+    "plt.scatter(y_test, y_pred)\n",
+    "plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'k--')\n",
+    "plt.xlabel('Actual')\n",
+    "plt.ylabel('Predicted')\n",
+    "plt.title('Predicted Vs. Actual')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "const                   2718099.405\n",
+       "fixed acidity                 2.585\n",
+       "volatile acidity              1.142\n",
+       "citric acid                   1.159\n",
+       "residual sugar               12.018\n",
+       "chlorides                     1.244\n",
+       "free sulfur dioxide           1.771\n",
+       "total sulfur dioxide          2.190\n",
+       "density                      25.506\n",
+       "pH                            2.135\n",
+       "sulphates                     1.133\n",
+       "alcohol                       6.835\n",
+       "Name: VIF, dtype: float64"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# https://stackoverflow.com/questions/42658379/variance-inflation-factor-in-python\n",
+    "\n",
+    "# One recommendation is that if VIF is greater than 5,\n",
+    "# then the explanatory variable given by exog_idx is highly collinear with the other explanatory variables,\n",
+    "# and the parameter estimates will have large standard errors because of this.\n",
+    "\n",
+    "from statsmodels.regression.linear_model import OLS\n",
+    "from statsmodels.tools.tools import add_constant\n",
+    "\n",
+    "\n",
+    "def variance_inflation_factors(exog_df):\n",
+    "    '''\n",
+    "    Parameters\n",
+    "    ----------\n",
+    "    exog_df : dataframe, (nobs, k_vars)\n",
+    "        design matrix with all explanatory variables, as for example used in\n",
+    "        regression.\n",
+    "\n",
+    "    Returns\n",
+    "    -------\n",
+    "    vif : Series\n",
+    "        variance inflation factors\n",
+    "    '''\n",
+    "    exog_df = add_constant(exog_df)\n",
+    "    vifs = pd.Series(\n",
+    "        [\n",
+    "            1 /\n",
+    "            (1. - OLS(exog_df[col].values, exog_df.\n",
+    "                      loc[:, exog_df.columns != col].values).fit().rsquared)\n",
+    "            for col in exog_df\n",
+    "        ],\n",
+    "        index=exog_df.columns,\n",
+    "        name='VIF')\n",
+    "    return vifs.round(3)\n",
+    "\n",
+    "\n",
+    "variance_inflation_factors(X_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Drop density, residual sugar, and alcohol\n",
+    "df = df.drop(df.columns[[3, 7, 10]], axis=1, inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Coefficients: \n",
+      "fixed acidity             -0.0642\n",
+      "volatile acidity          -1.2098\n",
+      "citric acid               0.1460\n",
+      "chlorides                 -7.4823\n",
+      "free sulfur dioxide       0.0069\n",
+      "total sulfur dioxide      -0.0042\n",
+      "pH                        0.1964\n",
+      "sulphates                 0.4813\n",
+      "Mean squared error: 0.68\n",
+      "\n",
+      "                            OLS Regression Results                            \n",
+      "==============================================================================\n",
+      "Dep. Variable:                quality   R-squared:                       0.110\n",
+      "Model:                            OLS   Adj. R-squared:                  0.108\n",
+      "Method:                 Least Squares   F-statistic:                     53.06\n",
+      "Date:                Sun, 06 May 2018   Prob (F-statistic):           1.50e-81\n",
+      "Time:                        14:39:44   Log-Likelihood:                -4250.7\n",
+      "No. Observations:                3428   AIC:                             8519.\n",
+      "Df Residuals:                    3419   BIC:                             8575.\n",
+      "Df Model:                           8                                         \n",
+      "Covariance Type:            nonrobust                                         \n",
+      "========================================================================================\n",
+      "                           coef    std err          t      P>|t|      [0.025      0.975]\n",
+      "----------------------------------------------------------------------------------------\n",
+      "const                    6.4275      0.417     15.406      0.000       5.610       7.245\n",
+      "fixed acidity           -0.0642      0.020     -3.244      0.001      -0.103      -0.025\n",
+      "volatile acidity        -1.2098      0.146     -8.290      0.000      -1.496      -0.924\n",
+      "citric acid              0.1460      0.126      1.163      0.245      -0.100       0.392\n",
+      "chlorides               -7.4823      0.710    -10.532      0.000      -8.875      -6.089\n",
+      "free sulfur dioxide      0.0069      0.001      6.290      0.000       0.005       0.009\n",
+      "total sulfur dioxide    -0.0042      0.000     -9.457      0.000      -0.005      -0.003\n",
+      "pH                       0.1964      0.106      1.847      0.065      -0.012       0.405\n",
+      "sulphates                0.4813      0.129      3.738      0.000       0.229       0.734\n",
+      "==============================================================================\n",
+      "Omnibus:                       49.308   Durbin-Watson:                   2.035\n",
+      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               70.178\n",
+      "Skew:                           0.169   Prob(JB):                     5.77e-16\n",
+      "Kurtosis:                       3.614   Cond. No.                     7.47e+03\n",
+      "==============================================================================\n",
+      "\n",
+      "Warnings:\n",
+      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+      "[2] The condition number is large, 7.47e+03. This might indicate that there are\n",
+      "strong multicollinearity or other numerical problems.\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Split features from output with the dropped variables, and set up new features array\n",
+    "X = df.iloc[:,:8]\n",
+    "y = df.iloc[:,8:]\n",
+    "features = df.columns[:-1]\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1075)\n",
+    "\n",
+    "# Do a regular regression\n",
+    "lm_model = LinearRegression()\n",
+    "lm_results = lm_model.fit(X_train, y_train)\n",
+    "\n",
+    "# Make predictions\n",
+    "y_pred = lm_model.predict(X_test)\n",
+    "\n",
+    "# See the results\n",
+    "# The coefficients are\n",
+    "print \"Coefficients: \\n\", \n",
+    "for i in range(len(features)):\n",
+    "    print \"%-25s %.4f\" % (features[i], lm_results.coef_[0][i])\n",
+    "\n",
+    "# The mean squared error\n",
+    "print \"Mean squared error: %.2f\\n\" % mean_squared_error(y_test, y_pred)\n",
+    "\n",
+    "X_train_with_constant = sm.add_constant(X_train)\n",
+    "lm_est = sm.OLS(y_train, X_train_with_constant)\n",
+    "lm_est_results = lm_est.fit()\n",
+    "print(lm_est_results.summary())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Predicted Vs. Actual')"
+      ]
+     },
+     "execution_count": 41,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(12,8))\n",
+    "plt.scatter(y_test, y_pred)\n",
+    "plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'k--')\n",
+    "plt.xlabel('Actual')\n",
+    "plt.ylabel('Predicted')\n",
+    "plt.title('Predicted Vs. Actual')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.80088373 0.84084977 0.8249501  0.82096358 0.86438663 0.81813768\n",
+      " 0.84735348 0.8418644  0.84227562 0.83253705]\n",
+      "0.8334202052931252\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Use kfolds for model selection\n",
+    "# Provides train/test indices to split data in train/test sets. \n",
+    "# Split dataset into k consecutive folds (without shuffling by default).\n",
+    "\n",
+    "kfold = model_selection.KFold(n_splits = 10, random_state = 1075, shuffle=True)\n",
+    "\n",
+    "# Specify the model\n",
+    "lm_model = LinearRegression()\n",
+    "\n",
+    "# Will return 10 results due to the number of folds, and use mean squared error as the decision metric\n",
+    "# From the documentation\n",
+    "# All scorer objects follow the convention that higher return values are better than \n",
+    "# lower return values. Thus metrics which measure the distance between the model and the data, \n",
+    "# like metrics.mean_squared_error, are available as neg_mean_squared_error which return the \n",
+    "# negated value of the metric.\n",
+    "\n",
+    "results = model_selection.cross_val_score(lm_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "# To make the results easier to read\n",
+    "print np.sqrt(-results)\n",
+    "print np.sqrt(-results).mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.80017932 0.83923217 0.82710583 0.82259813 0.86718906 0.8188416\n",
+      " 0.84719948 0.84360279 0.84524847 0.83592347]\n",
+      "0.8347120319723343\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Ridge Regression\n",
+    "from sklearn.linear_model import Ridge\n",
+    "\n",
+    "ridge_model = Ridge()\n",
+    "results = model_selection.cross_val_score(ridge_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "print np.sqrt(-results)\n",
+    "print np.sqrt(-results).mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Let's do some tuning to the ridge\n",
+    "# The main tuning parameter for the Ridge model is alpha - \n",
+    "# a regularization parameter that measures how flexible our model is. \n",
+    "# The higher the regularization the less prone our model will be to overfit. \n",
+    "# However it will also lose flexibility and might not capture all of the signal in the data.\n",
+    "\n",
+    "from sklearn.linear_model import RidgeCV\n",
+    "\n",
+    "alphas = [0.01, 0.05, 0.1, 0.2, 0.5, 1, 3, 5, 10, 15, 30, 50, 75, 100, 100000]\n",
+    "\n",
+    "cv_ridge = [np.sqrt(-model_selection.cross_val_score(Ridge(alpha = i), \n",
+    "                                                     X, y, \n",
+    "                                                     scoring=\"neg_mean_squared_error\", \n",
+    "                                                     cv = 5)).mean() \n",
+    "            for i in alphas]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.8378836709445743\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "cv_ridge = pd.Series(cv_ridge, index = alphas)\n",
+    "cv_ridge.plot(title = \"Validation Parameter for Alpha in the Ridge Model\")\n",
+    "plt.xlabel(\"alpha\")\n",
+    "plt.ylabel(\"rmse\")\n",
+    "\n",
+    "print cv_ridge.min()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.84017466 0.88959902 0.8562799  0.85179222 0.89417767 0.84409864\n",
+      " 0.87073258 0.87436353 0.87855796 0.87884362]\n",
+      "0.8678619814529853\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Lasso Regression\n",
+    "from sklearn.linear_model import Lasso\n",
+    "\n",
+    "lasso_model = Lasso()\n",
+    "results = model_selection.cross_val_score(lasso_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "print np.sqrt(-results)\n",
+    "print np.sqrt(-results).mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0.83906536 0.88332668 0.84953578 0.85374235 0.88910242 0.83850046\n",
+      " 0.86674667 0.87148712 0.87401569 0.87393539]\n",
+      "0.8639457918179743\n"
+     ]
+    }
+   ],
+   "source": [
+    "# ElasticNet Regression\n",
+    "from sklearn.linear_model import ElasticNet\n",
+    "\n",
+    "elastic_model = ElasticNet()\n",
+    "\n",
+    "results = model_selection.cross_val_score(elastic_model, X, y, cv=kfold, scoring='neg_mean_squared_error')\n",
+    "\n",
+    "print np.sqrt(-results)\n",
+    "print np.sqrt(-results).mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 102,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " rank_test_score 1\n",
+      "split6_test_score -0.5999194008883704\n",
+      "split7_train_score -0.5598891359137271\n",
+      "split0_train_score -0.5661486529622491\n",
+      "split2_test_score -0.5334411125387225\n",
+      "mean_fit_time 0.0028867721557617188\n",
+      "split3_train_score -0.5677727397132855\n",
+      "split6_train_score -0.5592125084531778\n",
+      "split9_test_score -0.5868137067705974\n",
+      "std_test_score 0.033399110107492205\n",
+      "params {}\n",
+      "std_fit_time 0.0005464605843794133\n",
+      "std_score_time 0.00019166495524480326\n",
+      "split8_test_score -0.5777099003087975\n",
+      "std_train_score 0.00362599414889879\n",
+      "split4_test_score -0.6237216236476324\n",
+      "split1_train_score -0.5615830687212218\n",
+      "split2_train_score -0.5666420434740032\n",
+      "split4_train_score -0.556799593457149\n",
+      "mean_score_time 0.000434112548828125\n",
+      "split9_train_score -0.5620925207173124\n",
+      "split5_test_score -0.5277563964303906\n",
+      "mean_train_score -0.5629048171941304\n",
+      "split8_train_score -0.5616180588990682\n",
+      "split7_test_score -0.5938554327648441\n",
+      "split0_test_score -0.5373080471216056\n",
+      "mean_test_score -0.5683411696447962\n",
+      "split3_test_score -0.523100725451225\n",
+      "split5_train_score -0.5672898496301106\n",
+      "split1_test_score -0.5798421694396347\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "\n",
+    "# Set up a grid search to show more detailed output \n",
+    "gs = GridSearchCV(lm_model, \n",
+    "                  param_grid={}, \n",
+    "                  cv=kfold, \n",
+    "                  scoring='neg_mean_squared_error', \n",
+    "                  return_train_score=True)\n",
+    "gs.fit(X, y)\n",
+    "\n",
+    "# Show results\n",
+    "for k in gs.cv_results_.keys():\n",
+    "    print k, gs.cv_results_[k][0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/readme.md b/readme.md
new file mode 100644
index 0000000..b7e786b
--- /dev/null
+++ b/readme.md
@@ -0,0 +1,9 @@
+# What is Ridge Regression — Applications in Python
+
+Jupyter notebook for the Medium article "What is Ridge Regression — Applications in Pythons"
+
+Notebook contains all code used for creating the article. The use of the Lasso regression was left out, to keep the length of the article manageable.
+
+## Article
+
+The article can be found [here](https://medium.com/@rrfd/what-is-ridge-regression-applications-in-python-6ed3acbb2aaf)
diff --git a/ridge-model.ipynb b/ridge-model.ipynb
new file mode 100644
index 0000000..ab56be9
--- /dev/null
+++ b/ridge-model.ipynb
@@ -0,0 +1,847 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 200,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10b1dc790>"
+      ]
+     },
+     "execution_count": 200,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "import numpy as np\n",
+    "import random as random\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Set seed\n",
+    "random.seed(1075)\n",
+    "\n",
+    "# Create random curve\n",
+    "length = 125\n",
+    "x = np.linspace(0, 10, length)\n",
+    "y = .5 * np.cos(x)\n",
+    "y_scatter = []\n",
+    "\n",
+    "# Create random scattering around the true function\n",
+    "for i, j in enumerate(y):\n",
+    "    y_scatter.append( y[i] + random.uniform(-.3, .3))   \n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 201,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10b3b0590>"
+      ]
+     },
+     "execution_count": 201,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtUAAAGrCAYAAAAPc2tHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XlYlNXbwPHvYUcQUFwR9wVkHRERJXMXUyvTTE2zMtMWs9WiRbOyX/q2WWmZtmmLWZra5hq5YVqAmqi4L4CKiqIiKMuc948BAkVlGRjA+3NdXDLPPM859zNwDbdn7nOO0lojhBBCCCGEKD0rSwcghBBCCCFEVSdJtRBCCCGEEGUkSbUQQgghhBBlJEm1EEIIIYQQZSRJtRBCCCGEEGUkSbUQQgghhBBlJEm1EEKIKkcptVYpNcbScQghRB5JqoUQoghKqVuUUpuUUueUUmeUUlFKqQ5lbPMBpdTGK459pZSaWrZoC7UXoZRaX8TxOkqpTKWUnxn66KaU0kqpF0pwjVnvUwghKhtJqoUQ4gpKKRfgV+AjoDbQCHgNuGzJuIqilLK54tA3QGelVPMrjg8Ddmit48zQ7f3AGWCUGdoSQohqQZJqIYS4WhsArfUCrXWO1jpDa71Ka/1v3glKqYeVUruVUheUUruUUkG5xyOUUgcKHL8r93hbYDbQSSmVppRKVUqNBUYAz+ce+yX3XA+l1GKl1Cml1CGl1IQC/U5RSi1SSn2jlDoPPFAwcK11IhAJ3HfFPY0C5ue20UoptS53FP60UmphcV8YpZQTcDfwONBaKRV8xfN5I/ypSqmE3NH5a92nVkq1KnBt/mi2UqqWUurX3NfgbO73nsWNUwghKpok1UIIcbW9QI5Sap5S6jalVK2CTyqlhgBTMCWqLsAdQEru0weALoArptHtb5RSDbXWu4FHgL+01s5aazet9RzgW+D/co/drpSyAn4BtmMaIe8JPKWUCi8Qwp3AIsAt9/orzaNAUq2U8gIMwHe5h94AVgG1AE9MI/LFNQhIA34EVmIatc7rpymwPLe9url9bivqPovRjxXwJdAUaAJkADNLEKcQQlQoSaqFEOIKWuvzwC2ABuYCp5RSPyul6ueeMgZTgviPNtmvtT6Se+2PWutjWmuj1nohsA8IKUH3HYC6WuvXtdaZWuuDuTEMK3DOX1rrpbl9ZBTRxhKgvlKqc+7jUcByrfWp3MdZmJJVD631Ja31xiLauJb7gYVa6xxMSfowpZRt7nP3AmtyR/iztNYpWuttJWg7X+61i7XW6VrrC8CbQNfStCWEEBVBkmohhCiC1nq31voBrbUn4Ad4ADNyn26MaUT6KkqpUUqpbbnlD6m519YpQddNAY+863PbeAmoX+CchBvEno5pJHmUUkphKr2YX+CU5wEF/K2U2qmUGl2cwJRSjYHu/Dc6vgxwAPrnPr7m61JSSqkaSqlPlVJHcstc1gNuSilrc7QvhBDmJkm1EELcgNY6HvgKU4IMpqS25ZXn5ZY/zAXGA+5aazcgDlMCC6aR76uav+JxAnAotzwk76um1rrfda4pyjzgHqA3UBNTSUne/ZzQWj+stfYAxgEfF6xtvo77MP3d+EUpdQI4iCmpzisBKfJ1uU7M6UCNAo8bFPj+WcAL6Ki1dgFuzT2uEEKISkiSaiGEuIJSylsp9WzexLjcEdrhwObcUz4DnlNKtVcmrXITaidMyeOp3Ose5L9EHCAZ8FRK2V1xrEWBx38DF5RSLyilHJVS1kopv1Is57cBSAXmAN9rrTML3N+QApP+zubGbCxGm/djqhM3FPgaDPRTSrljGsHupZS6Ryllo5RyV0oZrnGfANuAe3PvsS+FyztqYqqjTlVK1QZeLe6NCyGEJUhSLYQQV7sAdAS2KKUuYkqm4zCNnqK1/hFTje93uecuBWprrXcB7wJ/YUoi/YGoAu1GAjuBE0qp07nHPgd8cks9lubWKg/AlLAeAk5jSuJdS3IDWmuNqeSjKYVLP8BUt71FKZUG/Aw8mVu7TW45yIgr21NKhea2NSt3pDvv62dgPzBca30U6Jf7Op3BlDQHFnWfuceeBG7HlPyPwPQ65pkBOObe/2ZgRUnuXwghKpoyve8KIYQQQgghSktGqoUQQgghhCgjSaqFEEIIIYQoI0mqhRBCCCGEKCNJqoUQQgghhCgjG0sHUBp16tTRzZo1s3QYQgghhBCiGouJiTmtta5bnHOrZFLdrFkzoqOjLR2GEEIIIYSoxpRSR4p7rpR/CCGEEEIIUUaSVAshhBBCCFFGklQLIYQQQghRRlWyploIIYQQwpKysrJITEzk0qVLlg5FmIGDgwOenp7Y2tqWug1JqoUQQgghSigxMZGaNWvSrFkzlFKWDkeUgdaalJQUEhMTad68eanbkfIPIYQQQogSunTpEu7u7pJQVwNKKdzd3cv8qYMk1UIIIYQQpSAJdfVhjp+lJNVCCCGEEEKUkSTVQgghhBBVTEpKCgaDAYPBQIMGDWjUqFH+48zMTLP1s2bNGlxdXfPbDg8PN1vbALGxsaxYsSL/8ZIlS3j77bfN2kdFkYmKQgghhBBVjLu7O9u2bQNgypQpODs789xzzxU6R2uN1horq7KNoXbv3p2lS5eWqY1riY2NJS4ujr59+wJw1113lUs/FUFGqoUQQgghqon9+/fj4+PDiBEj8PX1JSEhATc3t/znv//+e8aMGQNAcnIygwYNIjg4mJCQEDZv3lzsfkaOHFko0XZ2dgZMI9s9e/Zk0KBBeHl5MWrUqPxztmzZQqdOnQgMDKRjx45cvHiR119/nW+//RaDwcCiRYv47LPPeOqppwA4dOgQ3bt3JyAggN69e5OYmJjf95NPPknnzp1p0aIFS5YsKf0LZkYyUi2EEEIIUQav/bKTXcfOm7VNHw8XXr3dt1TXxsfHM3/+fIKDg8nOzr7meRMmTOD5558nNDSUw4cPM2DAAOLi4q46788//8RgMAAwbNgwIiIirtt/bGwsO3fupH79+oSGhrJ582YMBgPDhg1j8eLFBAUFce7cORwcHJg8eTJxcXHMmDEDgM8++yy/nccee4wxY8YwYsQI5syZw1NPPcWiRYsAOHnyJFFRUezYsYN77rmnUoxwS1IthBBCCFGNtGzZkuDg4Buet2bNGvbs2ZP/+OzZs2RkZODo6FjovJKWf4SGhuLh4QGAwWDg8OHD2Nvb06RJE4KCggBwdXW9YTtbtmzh119/BWDUqFFMmjQp/7mBAweilCIgIICkpKRix1aeJKkWQgghhCiD0o4olxcnJ6f8762srNBa5z8uuBaz1pq///4bOzu7EvdhY2OD0WgEICcnp9CIuL29ff731tbW1x0tL62CfRS8P0uSmmohhBBCiGrKysqKWrVqsW/fPoxGY6H64169ejFr1qz8x3kTH4ujWbNmxMTEAKYVO3Jycq57vo+PD0ePHiU2NhaA8+fPk5OTQ82aNblw4UKR14SGhvLDDz8A8M0333DrrbcWOz5LkKRaCCGEEKIamz59OuHh4XTu3BlPT8/847NmzSIqKoqAgAB8fHyYO3dusdscN24cq1evJjAwkK1btxYaOS6Kvb09CxYs4NFHHyUwMJA+ffpw+fJlevTowfbt22nXrl1+vXTB+ObMmUNAQAALFy7k/fffL9mNVzBVWYbMSyI4OFhHR0dbOox8S7cm8fbKPRxLzcDDzZGJ4V4MbNfI0mEJIYQQopzs3r2btm3bWjoMYUZF/UyVUjFa6xsXqCM11WW2dGsSL/60g4ws08ceSakZvPjTDgBJrIUQQgghbhJS/lFGb6/ck59Q58nIyuHtlXuucYUQQgghhKhuJKkuo2OpGSU6LoQQQgghqh9JqsvIw82xRMeFEEIIIUT1I0l1aexaBkseAWBiuBeOttaFnna0tWZiuJclIhNCCGFhS7cmETYtkuYRvxE2LZKlWyvHxhRCiPIlSXVpXDgB2xfAiR0MbNeItwb508jNEQU0cnPkrUH+MklRCCFuQnmT15NSM9D8N3ldEmshqj9JqkvD726wsoVtCwDTKh9RET04NK0/URE9JKEWQoiblExeFxXJ2toag8GAn58fQ4YMIT09vdRtrV27lgEDBhS7T19fXwIDA3n33Xfzd1Y0p2XLlhEQEIDBYCA4OJiNGzeavQ9zM0tSrZTqq5Tao5Tar5SKKOL595VS23K/9iqlUgs8l1PguZ/NEU+5c3IHr76w4wfIybJ0NEIIISoJmbwuKpKjoyPbtm0jLi4OOzs7Zs+eXeh5rbXZE968Pnfu3Mnq1atZvnw5r732mln7AOjZsyfbt29n27ZtfPHFF4wZM8bsfZhbmZNqpZQ1MAu4DfABhiulfAqeo7V+Wmtt0FobgI+Anwo8nZH3nNb6jrLGU2EMI+DiKdi/xtKRCCGEqADFqZWWyevCUrp06cL+/fs5fPgwXl5ejBo1Cj8/PxISEli1ahWdOnUiKCiIIUOGkJaWBsCKFSvw9vYmKCiIn3766QY9XK1evXrMmTOHmTNnorUmJyeHiRMn0qFDBwICAvj0008BMBqNPPbYY3h7e9O7d2/69et31e6JV3J2dkYpBcDFixfzv6/MzLH5SwiwX2t9EEAp9T1wJ7DrGucPB141Q7+W1aoXONWFbd+C122WjkYIIUQ5Ku5GXxPDvQqdBzJ5/aawPAJO7DBvmw384bZpxTo1Ozub5cuX07dvXwD27dvHvHnzCA0N5fTp00ydOpU1a9bg5OTE9OnTee+993j++ed5+OGHiYyMpFWrVgwdOrRUYbZo0YKcnBxOnjzJsmXLcHV15Z9//uHy5cuEhYXRp08fYmJiOHz4MLt27eLkyZO0bduW0aNH37DtJUuW8OKLL3Ly5El+++23UsVXkcxR/tEISCjwODH32FWUUk2B5kBkgcMOSqlopdRmpdTAa3WilBqbe170qVOnzBB2GVnbQsBQ2LMCLqZYOhohhBDlqLi10jJ5XVSkjIyM/JrjJk2a8NBDDwHQtGlTQkNDAdi8eTO7du0iLCwMg8HAvHnzOHLkCPHx8TRv3pzWrVujlGLkyJFljmfVqlXMnz8fg8FAx44dSUlJYd++fWzcuJEhQ4ZgZWVFgwYN6N69e7Hau+uuu4iPj2fp0qVMmjSpzPGVt4repnwYsEhrXfCdqanWOkkp1QKIVErt0FofuPJCrfUcYA5AcHCwrphwbyBwOPw1E+IWQcdxlo5GCCFEOSlJrfTAdo0kib7ZFHNE2dzy6puv5OTklP+91prevXuzYMGCQucUdV1pHDx4EGtra+rVq4fWmo8++ojw8PBC5/z+++9l6uPWW2/l4MGDnD59mjp16pSprfJkjpHqJKBxgceeuceKMgwo9FPVWifl/nsQWAu0M0NMFaOBHzQMNJWACCGEqLakVlpUVaGhoURFRbF//37AVJ+8d+9evL29OXz4MAcOmMYxr0y6i+PUqVM88sgjjB8/HqUU4eHhfPLJJ2RlmRZx2Lt3LxcvXiQsLIzFixdjNBpJTk5m7dq1N2x7//79aG0aQ42NjeXy5cu4u7uXOMaKZI6k+h+gtVKquVLKDlPifNUqHkopb6AW8FeBY7WUUva539cBwrh2LXblZBgBx7fDiThLRyKEEKKcyEZfoqqqW7cuX331FcOHDycgIIBOnToRHx+Pg4MDc+bMoX///gQFBVGvXr38a6Kjo6+52kZeyYmvry+9evWiT58+vPqqaarcmDFj8PHxISgoCD8/P8aNG0d2djaDBw/G09MTHx8fRo4cSVBQEK6urgBMnjyZn3++evG3xYsX4+fnh8Fg4PHHH2fhwoWVfrKiyvtfQJkaUaofMAOwBr7QWr+plHodiNZa/5x7zhTAQWsdUeC6zsCngBFTgj9Da/35jfoLDg7W0dHRZY7bLC6mwLtepvKP8DctHY0QQohysnRrEm+v3MOx1Aw83ByZGO4lZR43sd27d9O2bVtLh1FlpKWl4ezsTEpKCiEhIURFRdGgQQNLh1VIUT9TpVSM1jq4ONebpaZaa/078PsVxyZf8XhKEddtAvzNEYPF5K1Z/e9C6DXFNIFRCCFEtSO10kKU3oABA0hNTSUzM5NJkyZVuoTaHCp6omL1ZBgBu38xrVkty+sJIYQQQhRSnDrqqk62KTeHvDWrt35j6UiEEEIIIYQFyEi1OeStWb1lNqSdAue6lo5ICCFENSB13EJUHTJSbS5Bo8CYDdtLviSNEEIIcaW8XRyTUjPQ/LeLY1HbowshLE+SanOp6wWNQyF2PphhRRUhhBA3t+Lu4iiEqBwkqTanoFGQsg+ObrZ0JEIIIaq4kuziKG5O1tbWGAwG/Pz8GDJkCOnp6aVua+3atQwYMKDYffr6+hIYGMi7776L0Wgsdb/Xcu7cOW6//XYCAwPx9fXlyy+/NHsf5iZJtTn5DgS7mqbRaiGEEKIMZBdHcSN525THxcVhZ2fH7NmzCz2vtTZ7wpvX586dO1m9ejXLly/ntddeM2sfALNmzcLHx4ft27ezdu1ann32WTIzM83ejzlJUm1Odk7gfzfsXAKXzlk6GiGEEFWY7OIoSqJLly7s37+fw4cP4+XlxahRo/Dz8yMhIYFVq1bRqVMngoKCGDJkCGlpaQCsWLECb29vgoKC+Omnn0rcZ7169ZgzZw4zZ85Ea01OTg4TJ06kQ4cOBAQE8OmnnwJgNBp57LHH8Pb2pnfv3vTr149FixZdt22lFBcuXEBrTVpaGrVr18bGpnKvr1G5o6uKgkZBzJewYxF0eMjS0QghhKii8lb5kNU/Kr/pf08n/ky8Wdv0ru3NCyEvFOvc7Oxsli9fTt++fQHYt28f8+bNIzQ0lNOnTzN16lTWrFmDk5MT06dP57333uP555/n4YcfJjIyklatWjF06NBSxdmiRQtycnI4efIky5Ytw9XVlX/++YfLly8TFhZGnz59iImJ4fDhw+zatYuTJ0/Stm1bRo8efd12x48fzx133IGHhwcXLlxg4cKFWFlV7rFgSarNzaMd1Pc3lYBIUi2EEKIMZBdHcT0ZGRkYDAbANFL90EMPcezYMZo2bUpoaCgAmzdvZteuXYSFhQGQmZlJp06diI+Pp3nz5rRu3RqAkSNHMmfOnDLFs2rVKv7999/8Uehz586xb98+Nm7cyJAhQ7CysqJBgwZ07979hm2tXLkSg8FAZGQkBw4coHfv3nTp0gUXF5cyxVieJKk2N6VMo9XLJ8Lx7dAw0NIRCSGEEKIcFXdE2dzy6puv5OTklP+91prevXuzYEHhJX+Luq40Dh48iLW1NfXq1UNrzUcffUR4eHihc37//fcSt/vll18SERGBUopWrVrRvHlz4uPjCQkJMUvc5aFyj6NXVQFDwNoeYr8uty6Wbk0ibFokzSN+I2xapKxbKoQQQoirhIaGEhUVxf79+wG4ePEie/fuxdvbm8OHD3PgwAGAq5Lu4jh16hSPPPII48ePRylFeHg4n3zyCVlZWQDs3buXixcvEhYWxuLFizEajSQnJxdry/ImTZrwxx9/AJCcnMyePXto0aJFiWOsSJJUlwfHWuBzJ/z7A2SZf+kj2RBACCGEEMVRt25dvvrqK4YPH05AQEB+6YeDgwNz5syhf//+BAUFUa9evfxroqOjGTNmTJHt5ZWc+Pr60qtXL/r06cOrr74KwJgxY/Dx8SEoKAg/Pz/GjRtHdnY2gwcPxtPTEx8fH0aOHElQUBCurq4ATJ48mZ9//vmqfiZNmsSmTZvw9/enZ8+eTJ8+nTp16pTDK2Q+SlfBjUqCg4N1dHS0pcO4vkMbYN4AuGsOBJau+P9awqZFklTEOqVujrY42dvIhBYhhBCinO3evZu2bdtaOowqIy0tDWdnZ1JSUggJCSEqKooGDRpYOqxCivqZKqVitNbBxbleaqrLS7NboHYLiPnK7En1tRb+T83IIjXD9JFL3ug1IIm1EEIIISxqwIABpKamkpmZyaRJkypdQm0OklSXF6Wg/QOwejKc3A31zPe/WQ83xyJHqq+Ut52tJNVCCCGEsKTi1FFXdVJTXZ4MI8DaDqLNu7VmURsCXItsZyuEEEIIUf4kqS5PTnXAZyBsXwCZF83W7MB2jXhrkD+N3BxRQCM3R2rVsC3yXNnOVgghhBCi/En5R3kLHg07foC4xab1q83kyg0B8lYEycjKyT8m29kKIYQQQlQMSarLW5NQqNsWor8oMqleujXJLFvQyna2QgghhBCWI+Uf5U0p03blx7ZCUmyhp8y93vTAdo2IiujBoWn9iYroIQm1EMJiZIMqIcpfYmIid955J61bt6Zly5Y8+eSTZGZmAvDVV18xfvx4C0d4NWdn5yKPW1tbYzAY8PPzY8iQIaSnp1+3nc6dO9+wrxkzZtywHXOSpLoiBNwDtjVMo9UFvL1yT6FyDfhvxQ4hhKiqLLVBlSTy4maitWbQoEEMHDiQffv2sXfvXtLS0nj55ZfLrc/s7Oxyaztvy/W4uDjs7OyYPXv2dc/ftGnTDduUpLo6cnAF/yGmuuqM1PzD11qZQ1bsEEJUZZYYMDB3Ii8JujA3c/9ORUZG4uDgwIMPPgiYRnrff/99vvjii/xEMiEhgW7dutG6dWtee+01wLRNef/+/QkMDMTPz4+FCxcCEBMTQ9euXWnfvj3h4eEcP34cgG7duvHUU08RHBzMm2++SdOmTTEajfltNW7cmKysLA4cOEDfvn1p3749Xbp0IT4+HoBDhw7RqVMn/P39eeWVV4p1b126dMnfVv29997Dz88PPz8/ZsyYkX9O3oj32rVr6datG3fffTfe3t6MGDECrTUffvghx44do3v37nTv3p2cnBweeOAB/Pz88Pf35/333y/T618UqakuhT1n9rAucR1jA8YW/6LgByF2Hvy7EDqOA6693rSs2CGEqMosMWBwvUS+pKVwV078ls20RFmVx+/Uzp07ad++faFjLi4uNGnSJD8h/fvvv4mLi6NGjRp06NCB/v37c+TIETw8PPjtt98AOHfuHFlZWTzxxBMsW7aMunXrsnDhQl5++WW++ML0CXtmZiZ5O1nHxsaybt06unfvzq+//kp4eDi2traMHTuW2bNn07p1a7Zs2cJjjz1GZGQkTz75JI8++iijRo1i1qxZN7yv7Oxsli9fTt++fYmJieHLL79ky5YtaK3p2LEjXbt2pV27doWu2bp1Kzt37sTDw4OwsDCioqKYMGEC7733Hn/++Sd16tQhJiaGpKQk4uLiAEhNTS2q+zKRkepSWH1kNR9t/YiopKjiX+TRDjyCTCUguVvDF7XetKzYIYSo6q41MFCeAwbmTOSlNE+Ym6V+p3r37o27uzuOjo4MGjSIjRs34u/vz+rVq3nhhRfYsGEDrq6u7Nmzh7i4OHr37o3BYGDq1KkkJibmtzN06NBC3+eNbn///fcMHTqUtLQ0Nm3axJAhQzAYDIwbNy5/pDsqKorhw4cDcN99910z1oyMDAwGA8HBwTRp0oSHHnqIjRs3ctddd+Hk5ISzszODBg1iw4YNV10bEhKCp6cnVlZWGAwGDh8+fNU5LVq04ODBgzzxxBOsWLECFxeXUr2m1yNJdSk8HPAwzVya8cbmN0jPKkGtToeH4FQ8HP0LKHq96bcG+ctIiBCiSrPEgIE5E3kpzRPmVh6/Uz4+PsTExBQ6dv78eY4ePUqrVq0AUEoVel4pRZs2bYiNjc0vx3j99dfRWuPr68u2bdvYtm0bO3bsYNWqVfnXOTk55X9/xx13sGLFCs6cOUNMTAw9evTAaDTi5uaWf/22bdvYvXt3oX5vJK+metu2bXz00UfY2dkV+7Wwt7fP/97a2rrI2u9atWqxfft2unXrxuzZsxkzZkyx2y8uSapLwd7animdp5CUlsTH2z4u/oW+g0z11X/PzT9UWVbskPpBIYS5WGLAwJyJvCVG2kX1Vh6/Uz179iQ9PZ358+cDkJOTw7PPPssDDzxAjRo1AFi9ejVnzpwhIyODpUuXEhYWxrFjx6hRowYjR45k4sSJxMbG4uXlxalTp/jrL9OgX1ZWFjt37iyyX2dnZzp06MCTTz7JgAEDsLa2xsXFhebNm/Pjjz8CpkmU27dvByAsLIzvv/8egG+//bZE99ilSxeWLl1Keno6Fy9eZMmSJXTp0qXY19esWZMLFy4AcPr0aYxGI4MHD2bq1KnExsbe4OqSk6S6lNrXb8/dbe7m691fszOl6F+8q9jVgHb3we6f4fzx8g2wBCw1U18IUX1V9ICBORN5Kc0T5lYev1NKKZYsWcKPP/5I69atadOmDQ4ODvzvf//LPyckJITBgwcTEBDA4MGDCQ4OZseOHYSEhGAwGHjttdd45ZVXsLOzY9GiRbzwwgsEBgZiMBiuu7rG0KFD+eabbwqVhXz77bd8/vnnBAYG4uvry7JlywD44IMPmDVrFv7+/iQllSyvCAoK4oEHHiAkJISOHTsyZsyYq+qpr2fs2LH07duX7t27k5SURLdu3TAYDIwcOZK33nqrRLEUh9K59b1VSXBwsM4rmLek85nnGbh0IO6O7nzX/ztsrYreKryQM4fgw3bQ9Xno/lL5B1kMYdMii5ww2cjNkaiIHhaISAhxszLXhljVJQ5Ree3evZu2bdsW+3z5nar8ivqZKqVitNbBxbleVv8oAxc7F17q+BJPr32ar3d9zWi/0Te+qHZzaBNumrDY5Vmwsb/xNeVM6geFEJVBZVp1Y2C7RpLwCLOS36nqT8o/yqhX0170aNyDj7d9TML5hOJdFDIWLp6CnUvLN7hikvpBIURlIKtuCCGqMkmqzeClji9ha2XL5E2TyTYWY7ehFt3BvTX8/Wn5B1cMUj8ohKgM5FMzUdVUxRJaUTRz/CwlqTaD+k71iQiJIDo5mhkxM258gZWVabQ6KQYSY258fjmTpf2EEBXhRqsMyadmoipxcHAgJSVFEutqQGtNSkoKDg4OZWpHaqrN5M5Wd7IzZSfzds3D292bAS0GXP8Cw3D443XTaLXnnIoJ8jqk1ksIUZ6KUy89Mdyr0Dkgn5qJysvT05PExEROnTpl6VCEGTg4OODp6VmmNiSpNqOJHSay9+xepmyaQgvXFvi4+1z7ZPuaYLjXNGGxz1RwrldxgQohRAXwAENQAAAgAElEQVQrzjbief+ac4UEWXFBlBdbW1uaN29u6TBEJSJL6plZSkYKw34bBsD3/b/H3dH92ief3g8z20P3l01L7OWSPwJCiOqmecRvFPXXRgGHpvUvlz6vHB0H08i3lLcJIYqrJEvqSU21mbk7ujOj+wzOXjrLc+ueI8uYde2T67SClj1No9U5pvNkIxYhRHVkiXppWU1ECFGRzJJUK6X6KqX2KKX2K6Uiinj+AaXUKaXUttyvMQWeu18ptS/3635zxGNpvu6+vNrpVaKTo3lz85vXn8TQcRxcOG7aZRH5IyCEqJ4sscqQrCYihKhIZa6pVkpZA7OA3kAi8I9S6met9a4rTl2otR5/xbW1gVeBYEADMbnXni1rXJZ2e8vbOXjuIJ/t+AxbK1te6vgSy7Ydu7qsI7A31G4Jf80C30HyR0AIUS2VR730jXi4ORa5W6ysJiKEKA/mmKgYAuzXWh8EUEp9D9wJXJlUFyUcWK21PpN77WqgL7DADHGVm8UxiSyPO4GLgw0ujra4ONhQ08GW+q4OtG1Qk+Z1nLCxtmJCuwlk5WQxb9c8Dp1OJ2rLLWRkGYErZr6HPgq/PwcJf8sfASFEtVXRqwwVZzURrTXHzl1i74kLHDiVxrmMLC5ezuHi5WzSMrPJzDbi4mCLWw1batWwxbWGHfVq2uPT0AXPWo4opSrsfoQQlZs5kupGQMGtBBOBjkWcN1gpdSuwF3haa51wjWuLfMdVSo0FxgI0adLEDGGX3sXMbJJSM9idkcX5S1mkXc6mYIWHvY0VberXxLtBTdo3HcKQVtn8uP9bjLVPQ/LtmKbmFJj5/sy9EDkV/prJxPBpsqSUEEKYQVGj4xN6tKKGnTWvLN3BrmPn2ZucRtrl/zbtUgqc7GxwsrfGyd4GO2srLlzK5mx6JumZhUvzXB1t8fVwwdfDhaAmtbi1TV2c7GVRLSFuVmVe/UMpdTfQV2s9JvfxfUDHgqUeSil3IE1rfVkpNQ4YqrXuoZR6DnDQWk/NPW8SkKG1fud6fVa21T+MRk1aZjZJZzPYffx87tcFdh0/z5mLmSilsav7K3buUWSeCeNy8gDyEuv8me9rXoOoGfBELEuP2MnqH0IIYSaJZ9NZvSuZyPiTbDl4hswcI04Ol2njYU1jdysa1rKinivUrqmoU8OFWg5uuNi54GrvirOtM9ZWplrwy9k5nMvI4ljqJXYeO8fOY+fZmXSO3ScukJltxM7Gii6t6tDHtz692tbH3dnewncuhCirkqz+YY7/UicBjQs89sw9lk9rnVLg4WfA/xW4ttsV1641Q0wVyspK4eJgi0tDW9o2dMk/rrVm9/ELrNmdzEeRikwFdrWjQGVx+cSdgPV/ZR0hY2HTR7DlUwbeNk2SaCGEKIMco+bP+JPM33yIqIRtWNkfo5bbGRr7nCZDHeNC1ln2A/vTgDQKf2ZagI2yoYVbC7xre+NVywuv2l541/PG0Lhp/jlZOUZijpxl1c5kVu48wR/xJ7FSOwhrVYcHw5rRrU09rKykTESI6s4cI9U2mEo6emJKkv8B7tVa7yxwTkOt9fHc7+8CXtBah+ZOVIwBgnJPjQXa59VYX0tlG6kujqVbk4hYvB1jreXY11lLdlorMpLupXvrprx2hx9N3GvAT+Mg/ld4eic4ulk6ZCGEqHJOp11m3uadLIz7g/NWO7B13gvWFwGoYVODlm4tTV+uLXF3dKeGbQ2cbJ1wsnHC3sae9Kx0zmee59zlc5zPPM/J9JPsPbuXvWf2cjLjJABWygpDXQNdG3elm2c3mrs2z6+t1lqz6/h5VsadYGF0AsnnL9O8jhP3d2rK3cGNcZbyECGqlJKMVJtl8xelVD9gBmANfKG1flMp9ToQrbX+WSn1FnAHkA2cAR7VWsfnXjsaeCm3qTe11l/eqD9LJtVl2Zgl79pkvQHHhkuw1XU4f3gUOZnu3BHowYuGTBp83wd6vwFhE8r5ToQQovo4nZbOq2u+Z+3xX1COB1BKU8Paha6Nb+HWxl1oV68dDZ0aYqVKv5JsSkYKe87uYevJraxLWMfuM7sBaFyzMb2a9mJImyE0rvnfB7dZOUaWx53gy6hDbD2airO9Dfd3bsq4ri1xcbAt8z0LIcpfhSfVFc1SSbU5d+eKPhHN02ufJsdoJMTpaVZEO5Fj1Kyq/TZNSMb6qe1gLW+6QghxPQfOHOX19V8Qe2YlWKdhTx3uaDmAO7164eful18PXR5OXDzBuoR1/Jn4J5uPbcaojdzS6BaGeQ8jzCOsUN/bElKZu+Egv/17nNpOdkzo0Yp7OzbFzqZi92CTHXuFKBlJqstJ2LTIIpe7a+TmSFREjxK3l3A+gfGR4zl6/ijj/J7i8CEDp2N/5jPbd1ju9Sbd734UB9vy+4NwI/LmK4S4UmV5X9h/dj8vrX2H3ec2oQE3Anmk3QiG+/cu10T6WpIvJrN432IW7V3EqYxTNHJuxIi2I7jH6x7srf+bsLgj8Rz/+303fx1MoZl7DZ7v681tfg0qZGk+2bZdiJKTpLqcNI/4jaJerfwVPErhQuYFIjZEsD5xPT0a9+DBNhNpMC+c5Mt2PFXzPabfHUhI89plirs05M1XCHGlyvC+kJSWxNtbPuSPhOVoox21c7rz4i2jua1t2wrp/0ayjFn8efRPvov/jpjkGOrXqM8jgY8wsNVAbKxM9dRaa9buOcVby3ezNzmNW1rVYdpgfzxr1SjX2Mw9MCTEzaAkSXXFfu5UxV1rA5aybMxS064mH/X4iOeCn2N90nqe3zyaE7cMI9DqIL7Zcdzz6V9MWhpXaB3ViiDbpQshrmTJ94WUjBSm/jWVfosHsOboKvS5W5nQ5gvWPvR/lSahBrC1sqVPsz581fcrPu/zOfVr1Oe1v15j4LKBLD+0HKM2opSiu3c9fp/QhTfu9GXr0bOEv7+erzcfwWgsv4Eu2bFXiPIlSXUJTAz3wvGKcgxzbMxipay43/d+5vedj5Wy4oEji/msTgPe91zH6LDmfLPlCOHvr2fd3lNl6qck5M1XCHElS7wvaK1Zun8pA5bcwcI9P3LpbHuCrP6PNQ+8w9hbAir1UnUhDUP4pt83fNj9Q2ytbHl+/fPc9/t97D27FwAbayvu69SMlU/fSlDTWkxaGseIz7ZwNCW9XOIpj4EhIcR/JKkugYHtGvHWIH8auTmiMH1kZs6PPf3r+vPj7T/Ss0lPPqhpx9iL23nIL5lFj3TG0c6a+7/4m6m/7iIrx2iW/q5H3nyFEFeq6PeFhPMJPLzqYSZFTeL8+drYnpjI+z3fYN6oXjRwdSiXPs1NKUX3Jt1ZdPsipoZNJeFCAkN/GcoHsR9wKfsSAJ61ajB/dAjTBvkTl3SO8BnrWRSTaPZYymtgSAhhIjXVlZDWmqW7vmH639PAypYXOr/Kbc1uZ9ryPXy16TBBTdyYeW9QuSa4laF2UghRuVTU+0K2MZuvd33NrG0fk50DF4/3JaROP2YMDaJuzaq9S2HqpVTeiX6HZQeW0bhmYyZ3mkxow9D854+fy+CZhdv562AKwzo0ZsodvmadsF5ZJpoKUVXIRMVqIum3p3gl8XeiHezp1rgbr3Z6lc37MolYvANba8V7Qw1096pXbv3Lm68Q4krl/b5wMv0kz69/npjkGOwu+5OaMIAnu3Xg8e6tsK7EpR4lteX4Fl7/63WOXjjKPW3uYWKHiTjYmEbfc4ya91bvYdafB/D1cOGTEe1NG4QJISqcJNXVxbkkjB8E8rVPdz7MOIiTrROTO02mRY1QHvs2lvgTF3iiRyue7tWmUtcVCiFEcWxK2sSLG1/kQuZFMo4NxCm7Ix8Oa0fnlnUsHVq5uJR9iZlbZzJv1zxaubXina7v0NKtZf7zf+xO5umF2wB47x4DvXzqWypUIW5asvpHdeHaCKuAe7g/fgMLe35KA6cGPL32aebGv8n8Mf4Mae/JR5H7Gb8glktXzMgXQoiqItuYzYexH/LImkfQ2c6k7nscf7eeLJ9wa7VNqAEcbBx4rsNzfNLrE85cOsOwX4exZN8S8ga7eratz28TutDEvQZj5kfzydoDpvLArUmETYukecRvhE2LZOnWJAvfiRACZKS68ju1B2aFQNcXyOo6kbn/zmXOv3Nwd3Tn9c6vs/tgQ978fTeGxm58NioYd+eqXW8ohLi5pGSk8Ny654hOjsbDuit7dvZkULvmvDXIH3sby21+VdFOpZ/ixY0vsuX4Fm5rfhuvdnoVJ1snAC5l5TBx0b/8sv0YnVu6E3vkLJey/5uwLvNdhCg/MlJdndT1Aq/+8PccbLMu85jhMb7t9y3Ots48suYRTth+xwfDfdh17Dx3fbyJ/SfTLB2xEEIUy4HUA4z4fQQ7Tu/AI/NB9sTdxrO9/Hh3SOBNlVAD1K1Rl097fcqTQU+y6vAq7lt+H0lpphFoB1trPhhq4NFuLdl0IKVQQg2yh4AQlYUk1VXBLU9DxlmInQ+Abx1fFg5YyCifUfyw5wfmHpjAtOG1SM/MZvAnm9h8MMXCAQshxPVtPr6Z+36/j/SsDBxOP8GhI235YJiBJ3q2rpAtuysjaytrxviP4ZNen3Di4gnu/e1eYpNjAbCyUrzQ1/ua18oeAkJYniTVVUHjDtA0DP6aBdmZgKkWb2KHicztM5f07HRej32Ue/scpE5NW0Z98TeR8ckWDloIIYq2ZN8SHl39KLXs63LpyHjSzjdkwcMdudMg5QsAnTw68V2/73Cxc+GhVQ+xdP/S/OcayR4CQlRaklRXFbc8DecT4d+FhQ53bNiRn+74ie6Nu/Pl7pk08p5Hq4ZZjPs6huU7jlsoWCGEuJrWmg9jP2Typsn41A7i+O6HMGbV4vuxobRvWtvS4VUqzVyb8U2/bwiuH8ykqEm8G/0uRm0scgMXexurIjdwkQmNQlQsSaqrila9oKEBNrwLOdmFnnK1d+Xdru/yRtgbxJ/dRWrt6bRoksDj38WyZKv5d+USQoiSMmojb2x+g7k75tKt4e38Gz0YB2snfhgXincDF0uHVym52rvyca+PGeY1jK92fkXEhgj6B9TL39kXwEqBrbUV3g1rFro2b6OepNQMNJCUmsGLP+2QxFqIciRJdVWhFHR9Hs4egrhFRTytGNhqIItuX0QjZw+OOcykWev1PPNDLAv+PmqBgIUQwiTbmM2kqEn8uPdHbvMcQeTGLtR2cuSHRzrRoq6zpcOr1GytbHk59GWebv80yw8tZ8KfEwj3r01URA8OT+vPmme64mxvw7A5m4lLOpd/3dsr9xTa+RJkQqMQ5U2S6qrEqx/U94P174Cx6HWpm7g04evbvmZw68Gcsv6dhl5f89KyTXwVdaiCgxVCCMgyZhGxIYKfD/zMnU1Gs+zPQDzcavDDuE541pJdAotrtN9opnSawqZjmxi7aiznLpsS6BZ1nflhXCec7Gy4d+5mth49C1x74qJMaBSi/EhSXZUoBbc+Byn7YNfSa57mYOPAlM5TePOWN8m0OUytNjN5Y82vLPxHRqyFEOZ3rdrdyzmXeWbtM6w8vJKhLR/jpz/b0qR2Db4fG0p9FwcLR131DG4zmHe6vsPOlJ08sOIBTqWfAqCJew0WjgullpMdoz7/m7ikc9ecuCgTGoUoP5JUVzVt74Q6Xrmj1cbrnnpHyztY0P87PFxqUaPp57zyxxcs2yb1dEII87lW7e6PMYd4MvJJ1ias5UGvZ1i4pgX1atrz7ZiOsklVGfRu2ptZPWeRlJbEqOWjOJ5mmpDuWcv0nxUXR1vu+3wL94U2vWpCo6OtdZETGoUQ5iFJdVVjZQW3ToSTuyD+1xue3rpWaxb0/47Qhh1xaLiYF/58k5U7j1VAoEKIm0HRtbuZ/C/6ZaKORfGo34t8vaoRLg62fPtwKPVkhLrMOnl04vM+n3Pu8jlGrxzNiYsnAGjo6si3YzpiY23Fl5sO8VyfNjRyc0RhWopPdl0UonzJNuVVkTEHZnYAOycYt95UFnID2cZs3tr8f/ywbwE5ad580OMderdtWgHBCiGqs+YRv1H4r4gRB48fsXXdyljfZ5m/0hMF/DCuE83qOF11/dKtSby9cg/HUjPwcHNkYriXJH7FtOPUDsauHktth9p8Ef4F9Z3qA7DnxAWGzvmLmg42/DiuMw1c5T8yQpSWbFNe3VlZQ5dn4cS/sHdlsS6xsbJhUueXeC7oJayd9vL0hodYEb+znAMVQlR3hWt0NfYNfsbWdSu25/rx/Zom5Bg1347peM2EWpZ9Kz3/uv580usTTmecZsyqMfk11l4NajLvwRDOpGUy8vMtnLmYaeFIhbg5SFJdBRQ5CSjgHnBrAuv/D0rwacP9/sN5p8tMrGzPMzFqLH8c2FaOkQshqruCm5HY1V2JXa3N5Jzpiv3F3qSmZzJ/dAit69cs8lpZ9q3sDPUMzO49m+T0ZEavHM3pjNMABDZ247P7O5BwJp0Hvvyb9MzsG7QkhCgrSaoruWuO5Px7Ero8B0kxsG9VidoMb9mFWd2/AODp9WNZe/ifcohcCHEzGNiuEW8N8qdOoyjs66zF9mJnGnE3py5k8snI9vg1cr3mtbLsm3m0q9eOT3p9QnJ6MmNWjiH1UioAnVq6M/PeIOKSzjFhwVZyjFWv3FOIqkSS6kruuiM5hnuhVjOInFqi0WqALs38eL/LZxiznZiw9hH+OLzejFELIW4mNq5buezyC7c1u41ba49jX/JFpg0O4NY2da97nSz7Zj7t67dnZo+ZJFxI4PHIx0nPSgegt099ptzhy5rdJ5ny806Kmkcl25kLYR6SVFdy1x3JsbaFrhGm2urdv5S47V5t2jK14ydkX3bn6bUT+O3A72UNVwhxk/nnxD9MjppMhwYdqJV+H0u3HefZ3m24u73nDa8tWDqS52Zf9q0sCW5IwxCm3zqduNNxPLPuGbKMWQCM6tSMsbe24OvNR5i74eBV/UlduxDmIUl1JXfDkZyAe8C9Nfz5v2vusng9AwO8edZvBlkZjYnYGMFPe38qS7hCiJvIwXMHeerPp/Cs6Uln52f5dN1Rhoc0YXyPVsW6Pq90pCov+2bOUV5zJLi9mvZiUugkopKimBQ1CaM27WcQ0deb/v4N+d/v8fz27/H886WuXQjzsbF0AOL6JoZ78eJPOwq96RUaybGyhu4vwqLRsHMJ+N9d4j4eCvPhxPmpfHPwdV79awrWVtbc2epOc92CEKIaSslI4bE1j2FjZcODLafy3IIj9PCuxxt3+qKKscxnnoHtGlWpJLqgvCQ47/05LwkGrrqn4iwdeL0EtySv0d1t7ib1ciofxH5ALftaPN/heaysFO/eE0jy+Us8/cM26rnY06FZbalrF8KMZKS6kivWSI7PXVDPF9a+BTmlm+H9YngAXd0mkn2xJZOiJvHbwd/McwNCiGonIzuDCZETSMlI4cWg/+PVxcdpVdeZD4e3w8b65vmzUtxR3uKOQJszwX3I7yFGth3JN7u/4fO4zwFwsLVm7qhgPN0cGfd1DIln06WuXQgzunne/aqwge0aERXRg0PT+hMV0ePqEQsrK+j+EqTsh38XlqoPKyvF+0M60DxnPMaMFry04SVWHi7eGthCiJuH1pqXN77MjtM7mNzxTf5vWQY21lZ8dn8wzvZV98PP0pRxFDcJLm7ybc4EVynFxA4T6de8Hx/EfsCKQysAqOVkx9z7g8nKMfLw/Bgm9Gwlde1CmIkk1dWFd39oaIB10yC7dAv9O9pZ89l9nbFPeRirzGa8sP4F/jjyh5kDFUJUZXN3zGX1kdU8FfQ0P6x3JeFsOp+MCKJx7RqWDq3USlvLXNwk+FrJd1JqRqFEvrt3XbMmuFbKijfC3iCoXhAvb3yZbSdN+xK0rOvMzHuD2HPiPOv2nuJ/d/lV6bp2ISoLSaqrC6WgxyuQehS2fVPqZjzcHJl7X2cuHnkAu5xmPLfuOTYmbTRjoEKIqmpdwjpmbp3JgBYDOHqoAxv2nWbqQD86tnC3dGhlUtrJesVdveRaybeCQon84pgkBrdvZNYE187ajg+6f0ADpwZMiJxAwoUEALq2qcuLt7Xl9x0nSDibcf1PQ4UQxSJJdXXSqhc07gjr3oas0k8yCWpSi7fu6sDJfffhZOXJM2ufIe50nBkDFUJUNQfPHeSFDS/gXdsbX7vRfLXpCKPDmjO0Q5Ny77u811EubS1zcVcvKSr5VsCVK0ZnZOXwZ/wpsye4bg5uzOo5CyNGHv/jcc5dPgfAmC7NGRTUiPdW72VF3Iky9yPEzU6S6upEKeg5GS4cgy2zy9TU4PaejL3Fh6TdI7BXLjz+x+MknE8wU6BCiKrkQuYFnox8Entre8Z5v8HrP++nS+s6vNTPu9z7roh1lMtSy3zDOS8UnXxfa7uu8lp1o5lrM2Z0m0HChQSeXfssWTlZKKX4313+GBq78cwP24g/cb5c+hbiZiFJdXXT7BZoHQ4b3of0M2Vq6oW+3tzSojkn940iMyeHcWvGkZKRYqZAhRBVQY4xh4gNESReSGRKx+lMWpRI3Zr2fDisYlb6KI91lK8c+TZ3LXNRrky+G1lg1Y3gBsG81vk1tpzYwptb3kRrjYOtNXPua09NBxse+TqGcxlZ5db/jcjOjqKqk6S6Our1Klw+DxvfK1Mz1laKD4a1o469J8bjozmZfpLH//hv+1shRPX3yfZPWJ+4nuc7vMBna+D0xUxmj2xPLSe7Cum/uJP8ipuAFTXyXR61zDdiqd0k72h5Bw/7P8zifYv5Yc8PANRzcWDWvUEkns3guR+3YzReaxy9/MjOjqI6kKS6OqrvC4Z7YcscSC1byUZtJzs+HhHE2TMNaZQ5jvgz8Ty77lmyjaVbD1sIUXVsTNrIp/9+ysBWAzl6OJCo/SlMvdMPf0/XCouhuJP8ipuAXWvkuzxqma/HkrtJPm54nFs9b2Xa39OIPhENQHCz2rzUry2rdyXz6fqDN2jB/GRnR1EdmCWpVkr1VUrtUUrtV0pFFPH8M0qpXUqpf5VSfyilmhZ4LkcptS3362dzxCOAbi+a/l37VpmbCmzsxpQ7fNm+14NQl7FsTNrIO9HvlLldIYT5mesj9BMXT/DihhdpXas1ITUf4pN1Bxke0oR7OjQ2c8TXV5JJfsVJwCrTDoLFqccuD9ZW1kzrMg3Pmp48u+5ZjqeZti1/MKwZAwIa8vbKeDbtP10hseSpTD8XIUqrzEm1UsoamAXcBvgAw5VSPlecthUI1loHAIuA/yvwXIbW2pD7dUdZ4xG53BpDx7Gw7TtI3lnm5oaHNObu9p6s2NyUHg2G8O3ub1m8d7EZAhVCmIu5PkLPMmYxcd1EMnMyecr/DV5avIdAT1em3HHlW3v5M/ckvyq3g2BOFlxIhpPxkBQLR/6CA5EQ/zvsWga7fzF9v3cV7F8DhzbA8X/h7GHIOAvGnCKbrWlXkw96fEBmTiZP/vkkGdkZKKWYPjiAFnWdeWLBVk6cu1Rht1nlfi5CFEFpXbbaKaVUJ2CK1jo89/GLAFrrIodIlVLtgJla67Dcx2laa+eS9BkcHKyjo6PLFPdNIf0MfGiAxqEw4ocyN3cpK4dBH28i4Wwa7UOWsD3lHz7r8xnt67c3Q7BCiLIKmxZJUhGJZSM3R6IiehS7nXf+eYd5u+bxZth0Pv7VmeTzl/h1QpdrTq6raGW5z7z/eBQsNXC0tbbMhic5WXDmkCkBPnfUVK6XehTOJUL6aUhPgUvnyt6PYy1w8QQXD3BtBC6NoHZzqOvNuozjPLHuGW5rfhvTukxDKcX+kxe4c2YUXg1q8v3YTtjZlG38benWJN5euYdjqRl4uDkyMdzrqte6Uv1chChAKRWjtQ4uzrnm2FO2EVCwcDcR6Hid8x8Clhd47KCUigaygWla66VFXaSUGguMBWjSpPzXRa0WatSGW56BNa/C4Y2mlUHKwMHWmtkj29P/ow2cPjSERg2P8czaZ1jQfwEezh5mCloIUVrm+Ag98mgk83bNY5jXML7/szbxJ0wr/twz+68ikyFLmBjuVWQCVpxJfnnx3yjJMyutTYnz8W1wcjeciodTeyBlPxScn2Jla0p6XRuDRzuo4f7fl2MtsK0Btg5g42j619rONBKtc8BoNLWVfck0Uf3SObiU++/FU3A+yfSV+A9k/LcyVFdlxeMNmjLz0O/4pCZzv+8oWjVqz/S7Axj/3Vb+b0U8rwwo/ScUVybLeZ+eAIVec4v8XIQwM3OMVN8N9NVaj8l9fB/QUWs9vohzRwLjga5a68u5xxpprZOUUi2ASKCn1vrA9fqUkeoSyMqAj9qDc30Y8wdYlb2MfvmO4zz6bSzDOjuwPm0SHs4ezL9tPjVsq+42xUJUB2UdqU68kMg9v9xDY5fGtMqO4NvNxwo9X5lGDosz+mkx6Wfg6F+QGA3Htpq+LqWanlNWUMs0SkzdNlDHC2q3MJXsOdcHK+vrt20Omelw5mB+cq9P7ubptB2stcnhi+MnCbqcCXXaEGtsxfwTTRk8ZCRd2vmWqitzfXoihKWUZKS6wso/lFK9gI8wJdQnr9HWV8CvWutF1+uzKiTVleoNf9sCWPoI3DUHAoeapcnJy+KY/9cRnhtoZO7eV+jZpCfvdn0XpZRZ2hdClFxZPkLPMmbxwPIHOHTuEO/eMo97P95bZO2yJENFuJAMhzfAkU2mr1O7TcetbKCej2nU2cMADQ2mx7YOlo23CBcyLzD0lyFcyrzAD/XDqXNiJzrhb1TuqHZWXV9sW/eEVj2haRhY2xar3eYRvxX5e6SAQ9P6m+8GhCgnFV3+8Q/QWinVHEgChgH3XhFQO+BTTCPaJwscrwWka60vK6XqAGEUnsRYJRX3464KEzAU/v4U1kyBtgPAzqnMTb7Ury3Rh8/y2aoMHuo3nrk7P2T+rvnc73t/sduoVP/xEKIaKMtH6LO3z+bf0//yv7DpvLHkRIXv+FelZF82jUQfiIT9kSnBxEcAACAASURBVJBsen/HriY0DgH//2fvrsOqPN8Ajn9fukEFRbDAQhTsArsL2zlnzG12zk2dbrOYijlbt586dU7nxBmznTUDxULsbkJFuuO8vz9e3QxQ4gTxfK6LS3bOG8+ZcrjP89z3/XSH0h5KMG2YO/LQP8TSyJIFTRfRe09vJqQ84uePN6KPxJMb/vhuWkfLqGtUPr0CyW+xkori0h5cu4BTIzDIuGe5g41pujPVogBRyI9yPFMNIElSO2AhoA/8IsvyDEmSvIFzsiz/JUnSQcANCHl5yiNZljtKkuSBEmyrUDqRLJRlefWH7pfbZ6pz5XLXo9PwS2toPAGaTlTLJe+HxdFh8XEqOVjiWMGXf54cZU2bNVQvWv2D54qiFEHIPc6GnuWL/V/QuVxn5Ocfsf70QwqbGxEel/zOsQV2pjouDG7tUzpt3DsCKfFKDnSpelC2GTg3AXt30FfHXJXubLu9jcl+kxnoNpBRNUYBsONiEKM3XWRMI0dGOwcpXUdu7lVyt01swKUDVO+j/L94a7VSvNcLeZ1W0z90IbcH1bl2ucu3P9zcByPPK8UwavDqzXZgY3tOxH9PsioZXy9fCpsUfu95ufKDhyAUQFFJUXT7qxumBqZ84bSILzddZ2BDJyo7WItgKPKREkDe2AOPT4OsUrpoVGwL5Vooxd/GWWpelSdM8ZvC1ttbWdZ8GY1KNAJg/JZAfM8/Yf3ndWlQ3laZrb97BK5th+u7IDkGipSHGv2gai+wsPv3emJVUsjLRFCtY7k2YIx4CEtrQ+XO0PV/arvshD8v8ce5x/j0LMzcyyOoZV+L5c2Xo/+egptc+8FDEAoQWZb56uhXHH1ylIUNVzNy7XOcbM3xHeKBkYFewQyGokOUQPHKVnhyRnmsmBu4tFNSHuzd35mNzW8SUxPpu7cvwbHBbPbajKOFI/HJqXRaepKI+BT2jm6InaXxfyckx8HV7XDhV+XDh56BMntdfwSUrK27FyIIaiCCah3L1ctdh7zh+HwYcBhKqKe/dHxyKh2WnCAuKZVhXmHMuzCDYdWGMbTq0AzPybUfPAQhD8lp0Lvl1hamnZrGmBpfsd/PhStBUewe1ZAytjmvu8hTkmKUoPDSH0r7UWQlkK7SFSp3UXo6FzCPox/z0a6PcLZxZm2btRjqGXIzNIaOS09Qv2wR1vSvnX5h+vObSnAdsF5p51eyHniMgIrttNPZRBDULCtBtVq2KRfelN4OYLkioAZoMEZp27RvgtI7VQ3MjAxY/HF1wuOSORlQlg7OHVhxcQWnQ05neE56Ww9nts+sIAg53z3xftR9Zp+ZTf3i9Yl/7on//XCmdapScAJqlQruHYWtg2BeBfhrBEQHQ+NvYPgZGHoCGn5VIANqgJJWJZnqMZVLzy+xJGAJABXtLfmufSWO3nzOOr8H6Z9oVxFaz4Ax16DNbIgJhj/6wNJacG4NpL6bpy8I+YWYqS6ILqxXfoF0W61UqavJymP3mLHnOtM6l2db6HgikiL4s+Of2Jrapnt8gVxaFgQ1yclqT4oqhX57+vEk9gneNX9hwJrbtHMrzuKPq+X/tpjRwRDwmzKbGvUYjK2hSheo1htK1M73qR1Z5X3KG99bvvzU4ic8HT2RZZkv1p3jxJ0wdo5oQEV7y/dfIC0VbuwEvyUQdB5sSkOTCeD2UZ4v6hQKBpH+IbyfKg1WNoXY5zDirNoKbVQqmU/XnOHsg3CW93dk/KnPqWNfh2XNl2X6F7UItAUhc3JSl7AicAXLLy5nhucc5m01IU0ls2d0Q6xNM9d7OM9RpcGdQ3B+jdLBQ1Yp3Tpq9FPSEvJI2ztdSExNpNfuXoQnhrPFawt2ZnaExSbRZuFxipgbsWOEJyaGmUjrkGW4cxAO/wAhgUpRY9OJSls+NWxKJgiaItI/hPfT04d285RluWPqawuupycxr0dVTA31mb87ilHVx3A86Dibbm7K1Pk5Xc4WhIIkoz6/H+r/ezXsKj8H/kwH5w4cvVCcJxHxLPq4Wv4MqOPD4cRCWFQNNvZQtuj2HA2jAqDfDqjSTQTUH2BiYMK8xvOIT4ln4vGJpKnSsLUwZl4Pd24+jWHW3huZu5AkQfmWMOgf6PmbUsy45XNY2QQentLoaxAEbRFBdUFVso7SV/TUMqWwRE2KWZkwp3tVrgZH8/hBdRo6NmT+ufncibjzwXPn7r/5RnEnQEJKGnP3q298gpBfZKcuITE1kYknJlLEtAjVzT5j64UgRjUvT60y72+BmeeEXoG/RsKPrnBwChQqDT3WKnm+LaYq24ILmVbWpizf1v0W/1B/Vl9RtpJoUrEo/T3KsNbvAUduprtJcvokCSp5wdCT0HWl0v97TRslwI56oqFXIAjaIYLqgqzFNGV3xT1j1Va0CNDStRi965Zi1Yn7dHT8EnNDc745/g1JaUnvPS+jndrEDm6C8K7sFEQvurCI+1H3+braZKbvvE/1UjaMaFpOe4PWJJUKbu2HtR3gJ0+45AvuH8GQk9B/l9LF4z07/wnv17lcZ9o6tWXZxWUEPAsAYEJbFyoWs2ScbyAvYt///v4OPX3l72fEWaU49MZuWFILjs6GFPGeL+RNIqe6oDuzUgmqu/+iLIWqSXxyKu0XnyApJY3JPfUYd3wUfV37Mr72+AzPEW32BEFzToecZuCBgfSq+AnXrzbj3IMI9o7OB+3zUhLh8mbwWwphN8HSAeoOghqfglk+m4HXsdjkWLrvVIrbfb18sTSy5EZoNB2XnKRJRTt+7lsz+4WuEQ/h78lKj/DCzuC1GJwaqnH0gpA9IqdayLxanyubGez/TunVqiZmRgb8+FFVnsYkceBsIXq59GL9tfX4BflleI5osycImhGTHMOkk5MoY1UGu5QuHL8dxqQOrnk7oE6MUnruL3RTUj30jZR0gi8vKa1DC1hAvT0gCM9Zh3GasBvPWYc1UotiYWTBrIazCI0LZab/TABc7K0Y27oCB649Zcv5HKRvFCoNH61Tct1lFazrADtHK3/PgpBHiKC6oNPTh/Y/QkwI/KO+okWA6qUKMaJpObYGBFHZ5BPK2ZTj+5PfE5WU/ptkru7vLQh52Mh90wiNfcrVwHbM2H0X1+JW9KpTUtfDyp7Y53BwGiyoomxmZV8F+m6HIceVdAL9fFhw+QHaLPKuVrQag90Hs+veLnbf2w3AFw2cqeNUmGk7r/E4PD5nN3BuAkNPKbsxXvgVltVVtokXhDxApH8Iih3DIXCTkn9Y1EVtl01JU9FthR+PwuNZ2r8oI49+RhunNvg09FHbPQRByNjcYzv49f73JIU1Jvl5WwBMDPSY1c09b31gjXqi9Do+vw5SE8G1kzIj7VBN1yPTOW2nzqWqUvls32fcibzDlo5bcLRw5HF4PG0XHcfVwYrfB9ZDX08N/b6DzsOOkfDsKlT9BNrNAeMP9MUWBDUT6R9C1rWYBkYWsOtLpeBHTQz19VjQsxqJKWmsOpTMALcB7Lq3i0OPDqntHoIgpC8mOYb1d+aRllSU5LAW/z6emKrKO111Ih/DrjFKW7yzq5TajxFnlVQBEVAD2i/yNtAzwKehDzIy3x7/ljRVGiULmzHFy5Uz98NZfeKeem7kWBMGHYVG4+HSJvipITw5r55rC4IGiKC6AEo3987cFlpNh0en4MI6td6vrJ0F37ZTtrY1T2yDS2EXvE95E5EYodb7CEJ+lp2c2fnn5qPSiyIxuDvIb6ZF5PquOhEPlZzaxdWVXWBr9FX6S3deBrbldT26XCW7PctzooRlCb6r+x0Xnl1g1eVVAHSvWYJWrsWYt/8WN0Kj1XMjAyNo9h303w2qVPilFRybp2zoIwi5jAiqC5j35t5V7wNlGsLfUyAmVK337VuvNA3L2zJ7z21GVPme6ORoZvjPUOs9BCG3y24xWXZyZv2C/Pjz9p8YxTZFlVjqnec1GXDlSFQQ7PwSltSAixuh5qcw+iJ0WAA2774OQXdF3h2cO9DWqS0rAldwJewKkiTh09UNK1MDvtx0keRU9a16UtoDhpxQ0n4O/wDrOkJ0iPquLwhqIILqAua9G6xIEngtUvIV92bc+i47JEliTnd3DPQllu6PY4j7UPY/2M++B/vUeh9ByK1yUkyW1Y2RYpNjmXpqKk7WTria9njn+VzZVSf2GeydoMxMB/ymtMQbdRHazwfrEroeXa6mqyJvSZL4vt732JraMvH4RBJSEyhiYcysru7cCI1h8aHb6r2hqQ10Ww2df4LgAPi5ETzMuKOUIGibCKoLmA/m3hUpC43Hw7Udaq+4Lm5tylSvypx9EAGRTalSpAozTs8gLCFMrfcRhNwoJzuGZjVn9sfzP/I0/ikflR7LidtRNCxvm3u76iREwMGpsKgqnPkfuPeAkeehw49gnUvGmAd0ru7IyQnNuD+rPScnNNPa36+VkRXTG0znQfQDFl1YBEAL12J0q1GCFf/cJfBxpHpvKElQrRcMPAQmVrDOC06vUOsGZoKQXQa6HoCgWdsDgpi7/ybBkQk42JhiY2ZIRHzKO8e9sRTsORqu/Am7v4YyDZQ3rnSuNa51xSy/cXet4ci+q6HM//sOP3/2LWP9+jPj9AwWNF2Qo9cpCLldTorJHGxM0+3ukF4Kx5mQM/je8uWTin1ZsT8VJ1tz/te3FqZG+u8cq1PJ8eD/E5xcCInR4NYdGk8A23yyw2MBUq94PXpX6s2G6xtoXKIx9R3qM9nLFb+7YXztG8iukQ0wMVTzv7+ilWDgYdg2FPZNUDqFeC1SdgkWBB0RM9X5WHrLzbGJqRjqv9nq6J2lYH1D6LhE6V19+IcMr5WdPqiSJDGzixvmRvos2BvFYPchHHx0kIMPD+bw1QpC7paTYrLM5swmpCYw9dRUSlmWIvxJM0KiEpjXwz13BdRpKXBujZIzfWgalKyn5Mp2WyUC6jzsyxpf4mTtxKSTk4hOjsba1JDZ3dy58yyW+Qc01GnGxBp6/gbNJsHlLbCqJUQ+0sy9BCETRFCdj6W33JyikjE3MvjwUnCJWlBnkLKN+SP/HC1dv83O0pjpnd249CSKhOcNcCnswgz/GRluCiMI+UFOiskymzO7/OJyHsc8plOJ0fiee8qgRmWpWTqX7Cwoy3DtL1heT2ndaVMKPtsLvTcrG7gIeZqJgQkzG8wkLCEMH39lH4JGFezoXbcUq07c5+yDcM3cWE8PGo2FPn8qvcxXNhdt9wSdEZu/5GNOE3aT3t+uBNyf1f7DF0iKgeUeoG9IpeDvScA4+9dKx8jfA9h7OYSF/Wz5/swgOpXrxDSPadm6liDkBepIocrI1bCrfLLnE9qX6cTB442wMTNk58gGGBvkglnqR/7w9yR47A92LtB8ClRsq+THCvnKiosrWB64nPmN59OqTCviklJps+gYepLE3tENMTPSYNbp85uwobuy62bX/4FrR83dSygwxOYvAqCG3qXGltBpCYTfZYr51pxdKx3eHStTyNyIpfsS6VOpL1tvb8U/xD/b1xOE3C6zxWRZbb2Xokphst9kbE1siQluzYu4ZOb3qKb7gPrFXfijj9JbOOIheC1Wdm11aScC6nxqgPsAqhSpgvdpb8ISwjA3NmBu96o8fBHPrL03NHtzu4ow4LCy8rG5H5xc/EYBY3ZbWgpCZomgOh9TS+9S5yZQewA903bRwPBWzq71lkLmRszs4saN0BhUEa0oZVmKqX5TSUjN5ZtSCIIGZad+Yc2VNdyKuEUHx+HsvBjJsCZlcSthrb1Bvy0+XGmPt6wO3DkMTb6FUReUntP6oj4+PzPUM2RGwxkkpCTgfcobWZap51yEzzzL8Ouph/jd1XC3Jws7+HSn0s/670nKbpyqNLXVBQnC+4igOh9TW+/SFtOQCpXmJ6vVlLWW1NqWq6VrMbpUd+R/Rx/zafmxPIl9wvKLy3N0TUHIy7Jav3Av6h4/Bf5E0xIt2HjUChd7S0Y209GOg6nJcGq50mv6zM/KhlKjAqDJN6IrQwHibO3MqBqjOPL4CLvu7QJgfGsXyhQx45s/LxGXlKrZARiaQvc14PklnF8DWz5j4b4raqsLEoSMiJxqIXMenIS17aDOYGg3R62XjoxPpuWCYxQxN6JO7cPsuLuNje02Utm2slrvIwh5QVZqIVSyiv77+nM38i5VpZkcuBTH9uGeVHHU8iy1LMPNPXBgEoTfBeem0HoGFBM/wwVVmiqNz/Z/xp2IO2zrtI1i5sU4+yCcj34+Rd96pfHupKXiVL+lcOA7jqW5MThlDAmYvPF0TuqChIJB5FQL6lfGE+oOVWaf7h9X66VtzP5LAzGN7UQRkyJMOzWNVJWGZzMEIRfKSi3ElltbCHgWQDuHQey5GMuwpuW0H1A/vQq/doJNn4CePnziC323iYC6gNPX02e653RS5VSmnJqCLMvULlOYzzyctJMG8orHCOi4FE/9q6w3moUVsW88nZO6IEF4mwiqhcxrPhkKl4Udw5TOIGr0Kg1k9T8hfFJuJNfDr/P7jd/Veo+MiOIVITfJbC3E8/jnLDy/kBp2tdh2woFKxa0Y0VSLfZ7jXsCur+CnBhASCG3nwFA/qNBKFCEKAJSyKsWXNb7kZNBJtt5Wit3Hta6ovTSQV2r05XztH3GX7vGH0XTsUHZ5zGldkCC8TQTVQuYZmUHnFUov0D3j1X75KV6uFDI3wvdYYTwdGrI0YCmhcaFqv8/rRPGKkNtkthZi1plZJKUlYRz5EZFxKczr4Y6RgRbe0tNSlG2hl1SH82uh9gAlb7ruYGXjKEF4zccuH1PHvg5zz80lODYYUyN95vaoypOIBGbv03A3kNfUaf8ZZz1+oozeUzYbeVPVOkEtdUGC8DoRVAtZU6ouNBoHgRuVrczVyMbMCJ8ubtwMjaV46ieoZBWzzsxS6z3eps5NbQRBXT7Ueu/o46MceHiA5sV7c+CSimFNy1HZQQtpH3ePKDPT+yaAQw0YehLazQWzXLLBjJDr6El6eHsqXUAm+03WXRoI4Nm6B6af78TJJJYdFj50LpcLergL+YoIqoWsazQeStSGnWPYf/JsplInMpti0eJlGsj649F0cerPoUeHOPLoiMZeSnBk+u37MnpcEHQtPiWeGf4zcLIqyxF/V1zsLTWf9hHxUOk3vb4zpCTAxxuVvOmilTR7XyHXyEmanKOFI1/X+hr/EH+23N4C/JcGMuHPy8Qna7F+plRdZffF6BBY2wFinmrv3kK+J4JqIev0DaDrSlLSUilyYAQhkXHvTZ3IaorFFC9XbMyMOH6uMuVsyjHzzEziU+LVMvS3fzHYmKW/XC2KV4TcaknAEp7GPcUusQ/hcSrm9aiqubSP5Hg4MvNlv+lD0GwSDD8DLu1F3nQBoo40uR4VelDXvi7zz80nJDYEUyN9Zndz51F4PHP2aXllsFQ96LMFooNhXQeIfabd+wv5lgiqhewp7ISPNIBa0g2G6v/178PppU5kNcXCxsyI6Z2rcD0kDlfDzwmNC1VL7+r0fjHEJqZiqP9mcCCKV4Tc6krYFTbe2IhnMS8OBZoypLGzZrp9yDJc3wnL6sI/s5UgesQ5aDQWDE0+fL6Q570+AfH15sAcp8lJksRUj6moZBVTT01FlmXqOhehv0cZ1p16wNkH4Wp+BR9Q2gN6+yo1Quu8RGAtqIUIqoVsWxNThx1pHowx2EI16c6/j7+dOpGdFIs2VezxqurA5pP6tCzRid+u/8aNcKWoJbvLkOkF9ykqGXMjg5xvkCMIGpaqSsX7lDeFjAtzIaA+5YtaMKq5BjZ5CbsNv3VV0j2MLaD/buj+C1iLn4mC4u0JiLQM9rPIappcCcsSjKk5Br9gP7bd2QbA+DYVKVHIlPFbLpGQnPaBK6hZGU8lsI58BOu7QEKkdu8v5DsiqBayzcHGjEkpnxFKYRYZLsWS+JePm751XOb77r5uWsfKWJkYcuNaQ6yMrJh+ejpbLzzO9jJkRr8AohJS3lsUJgi5wcbrG7kefp1Scm+eR0vM61EVY4PsFVql+8E0KRb+ngLL68OTc9BmNgw+DmUaqPmVCLldehMQ6clOmlzPij2pVawWc8/OJTQuFDMjA2Z3c+d+WBzzD+igQLxMA6VG4PlN+P1jJeVJELJJBNVCto1rXZEUQytGJw/HUQpjtuH/MDXUeyd1IrN9d99W2NwI705VuBaUSjWLPgQ+D8Tn+PpsL0NmN7gXBF0LiQ1h6cWlVLGpx7GL9gxs5EzVkjbZuta7aVDxHN36P+IX1ISTC8GtB4w8D/WGKPUTQoGTmRno7KbJ6Ul6eHt4kyanMe3UNGRZxqOsLb3rlmL1yfucfxiRnSHnTNmm0G0lPDoNvv2VtpGCkA0iqBay7VU/3VDrasxN7Uk7/TP8UTXwnZnezPbdTU979+K0rWLPvlMlqFSoKgkWO0A/7p3jMvNLILvBvSDoms8ZH1SyzINbrXC2s2BMiwrZvtbrs5BlpSDWG/qwUH8hT5JM4fMD0GUFWBRV19CFPCijiQZ9SVJLmlxJq5KMrjGaE0En2HF3BwAT21XCwdqU8VsCSczELLnaVe4CHX6E2/thx3BQqbQ/BiHPU8s0hCRJbYBFgD6wSpblWW89bwz8CtQEXgA9ZVl+8PK5icAXQBowSpbl/eoYk6Adnas7Km+sclPYFIX7tXnwpAWUqJX+cdng3akKp++9IDaoI5LpDIzt9pMU2vWNYzIz2/zq/nP33yQ4MgEHG1PGta4o0j2EXO3Qo0MceXwEV5NenAk3Y8sQd0wMs99fNzgyAVMSGWmwnQH6u0nEmMkpn7IxrQV3StVV48iFvGpc64pM3Hr5jVVBU0N9tdab9HLpxYEHB5hzdg6eDp7Ymdnh09WNfr+cYeHB20xo6wIoKytae8+u9TnEv4DD09l8NY5v4nrhYGMmfk8ImZbjmWpJkvSBZUBbwBXoJUmS61uHfQFEyLJcDlgAzH55rivwMVAZaAMsf3k9Ia+RJOi8HKyKK8tn8eqr5LazNGZqx8pce2hOSYOWGNqcRc/k8b/PZ2W2+UObaghCbhKXEoePvw+OZs74B1Thc08napbOwUYrskwvy4scNB7HMIO/+EvlSdOk+fya1ppiNhbqG7iQp+VkdTGz9CQ9pnpMJSk1iemnpyPLMo0q2NGzVklWHr/HpSeROtnxdrtFL9ap2vFR2i6G6O8Uu+wKWaKO9I86wB1Zlu/JspwMbAI6vXVMJ2Ddy++3AM0lSZJePr5JluUkWZbvA3deXk/Ii0wLQY91EPsUtg1W6/JZx6oOtKhUlLu3PLEwsMGyxA4kVKJbh5CvLbu4jGfxz4h+0onSRSwZ2yoHqUov7sKG7sxMmUMs5nRPmszYlCG8wFqkQQnv0MYEhJO1E8OrD+fw48Psf6gsUn/bvhK2FkaM873EnH03tL7j7dwDt5ia/Ak70jz4xnAT7fVOi112hUxTR1DtCDx+7b+fvHws3WNkWU4FooAimTxXyEsca0DrmXD7AJxcoLbLSpLEjC5uGOmZYZPQHZXhE3z6x4nZZiHPy6hF5PUX19lwfQNOxs0JflqM2d3cMTXKxkJeSgIc8VG6ejzyh9YzudFxFyHW1UUbSUHn+rn2w7WIKz7+PkQkRmBtaohPVzduPo0hOCox3XM0ueNtcGQCMnqMTxnEGVVFfjRcQQ3plthlV8iUPFOoKEnSIEmSzkmSdO758+e6Ho7wPrUHQJVucHg63DmotssWszJhUgdXbtx1prRpNRZfWExYQpjari8I2pbR8vbWC4/wPuWNhYENly570q9+aeo5F8n6DW7/DcvrwT+zoFIHGHEW6g+nU83SIg1KyBUM9Azw9vAmOima2WdnA9DMpRhd3/NvUpMdm15dOwkjBiePIVguzEqj+dSyitLYPYX8Qx1BdRBQ8rX/LvHysXSPkSTJALBGKVjMzLkAyLL8P1mWa8myXMvOzk4NwxY0RpKg4xIo6gpbPleWndWkR80SNKpQlHs3W5KQmsDC8wvVdm1BULcPbVSU0W6jPifWcuXFFeQXXjhYFuabNi5Zu3HUE2Xzlg3dQc8Q+u1QNnCxKp7TlyQIalexcEUGug9k973d/PP4HwAme7liaWKA9NaxmU1Vyu4mYa93iYrAis9TxqOPzC9GcyBBB+3+hDxFHUH1WaC8JElOkiQZoRQe/vXWMX8Bn778vjtwWJZl+eXjH0uSZCxJkhNQHjijhjEJumZkrjTUl/Th916QGK2Wy0qShE9XN/RSi2GT2pwdd3dw8dlFtVxbENQpM0VW6S0pS/oxJFjsxN7QjeAgF2Z3c8fcOJONmtJS4OQiWFoHbh+E5pNh6ElwbqKW1yQImjLQbSDlbMrhfdqbmOQYbMyMmNu9KjJg9TK4zmyqUk4KHN8u0ky2duZyg+VYxj+BP/pCarI6Xq6QT+U4qH6ZIz0C2A9cBzbLsnxVkiRvSZI6vjxsNVBEkqQ7wFfAhJfnXgU2A9eAfcBwWZZ10KBS0IhCpeGjX+HFHdg6SG2Fi442pkxs58L92x5YGhRhpv9M0lTin42Qu2Q0C/16wVN6y9jGxXYh6aVy70ZretUpRYPytpm74YMT8FMD+HuyEkQP94eGX4OBcQ5ehSBoh6G+IT94/kBYQhjzz80HoE0Ve9q7FychJY39YxplOlUpMz977/N2kWbDlp2g0zJ4cBz2jsv6ixMKDLXkVMuyvEeW5QqyLJeVZXnGy8cmy7L818vvE2VZ7iHLcjlZluvIsnzvtXNnvDyvoizLe9UxHiEXcWoIbWbBrb1wdKbaLvtJnVJ4ODsQHdSW6+HX2XJri9quLQjqkFFh0+uPv70hkb75bQytAzGMaU5R0xJMbFfpwzeKfQZbB8Pa9pASD702Qa+NyodaQchDqthWoZ9rP/68/SdnQpRFa++OlbE0MWScbyCpaZmbmMnMz16WVe0JDcbA+bVwbk32ryPka3mmUFHIw+oMhOp94dhcuLpNLZeUJInZm2QUBQAAIABJREFU3dxRxbpjoXJhccBiIhLVl++W3Xw8QXglo2Kq1x9/Y6lZSsHCYQcmFCU8uAE+Xd2wMjHM+AaqNDizEpbUgit/QsOxMMwfKrZV90sRBK0ZVm0YJS1LMvXUVBJSEyhiYcy0jpUJfBLFqhP3M3WNzPzsZUuzSVCuBewZp3TSEYS3iKBa0DxJgvbzoWRd2DYEnpxTy2VLFjZjQttKPL3fltjkOBZdWKSW6+piwwEh/3l7FhrSL7J6tdQ8rmcwKoMwIh970b2GE00qvmer8KDzsLIZ7BkLDtVgqB80nwRGZpp4KYKgNaYGpkytP5XHMY9ZfnE5AB3ci9Omsj0//n2LO89iP3iNzP7sZZmePnRbBdYlYHNfiA7J2fWEfEcE1YJ2GBhDzw1gaQ8be0J45mYcPqRP3dLUdqyEKsqTrbe3cvn55RxfM6f5eIIAWduV7kHUA1ZdXoVpci0KSVWY1P7tTWlfSoiAXWNgZXOICVU6evTbAXYVNPtiBEGL6hSvQ/cK3fn12q9cCbuCJEn80LkKZkb6jNsSSJpKfu/5Gt0R0rSQUoSfFKsE1qlJOb+mkG9IShOOvKVWrVryuXPqme0UtCzsDqxuAWa28MUBMMvBlssvPQiLo83iA5g5z6eibUk2tN+AnpT9z4tOE3aT3k+FBNyf1T7b1xWE9MiyzMC/B3Ih9DIRN8ewsndTWrgWe/MglQoCNypFiAmRUHcwNJkIJla6GbQgoKzqzd1/k+DIBBxsTBnXuqLaep7HJMfQeXtnrE2s+aP9HxjqG7I9IIgv/7jI9+0rMaChs1ruk23X/lKC6hr9wGuxsiIr5EuSJJ2XZblWZo4VM9WCdtmWUz7lRz5U+uiq4VN+GVtzxreqRnRwG668uMK22znL29ZYPp4gpGPP/T34h/gTF9qKTm4V3w2oQy/DmrawYzgUKQeD/4E2PiKgFnRK02lylkaWTKo/idsRt1l9ZTUAnao50KJSMebuv8m95x9OA9Eo145KHcOFXyFgvW7HIuQaIqgWtK+0B3ReAQ9PwvZhamm119+jDFULNYVEJxacX0hUUvZ3v9JYPp4gvCU6OZq5Z+dilFoai+SGTPGq/N+TidGwdwL83Bhe3FZaen22D+zdNDIWUZwrZIU20uSalGxC2zJt+fnSz9yJuIMkSczsUgVjAz2++fMSqg+kgWhc02/BualSuBh6RbdjEXIFEVQLuuHWHZpPgStb4NC0HF9OT09ibveqpDztRFRyFMsClmX7WhrNxxOE1yy+sJjwxAgiHndkZhd3CpkbgSzDpc2wtBb4/wQ1P4UR56B6H9DTzFu2KM4VskojbevS8U2db7AwtGDKqSmkqdIoamXCZK/KnH0QwVq/B2q9V5bp6UPXlWBiA76fQlKMbscj6JwIqgXdaTAGan0OJxcqu8DlkLOdBWObNiU5vB6bbv7BzfDsz5i83fy/c3VHMZMnqNXl55fZfHMzKRH18XKpTevK9vD0mtJveutAsHKEgYegwwK11B68jyjOFbJKW2lyRUyLML72eC49v8Smm5sA6FbDkWYuRZmz/wb3w+LUer8ss7CD7qsh/J5SRJwH69QE9RFBtaA7kgTt5kHlrkoB1vm1Ob7kZ55OuJr0QE4zZarfdNRViCtm8gR1SlWl4n3KG0llhVlcB6a2KgX7v1N2RHx2DbwWwYBD4FhTK+PR1qyjkH9oM02ug3MHGjg2YNGFRQTHBiNJEj5d3TDS12Oc74e7gWhcmQZKKshlX7iwTrdjEXRKBNWCbunpQ5efoVxL2PklXNmao8vp60nM716f1LA2XHlxkTqL5qhlZlnM5Anq9MfNP7gRcYO44HasqR5EoV/qw6llUKMvjLwANftrLNUjPaI4V8gqbabJSZLEpHqTAPA+5Y0syxSzMmGKV2XOPYxgzUn1tGjNkQZfQ9lmsGe8UlwsFEiipZ6gdtlqs5QcD791gydnlW2Wy7fI0T2LWRtyQ5qJZBBF3L2xoDLG1FA/22/6os2eoC7P4p/RYasXcnQRfo+JolzCFXCoDu3mQwntzEy/7dVKzOsfHHPy8yIImrDx+kZ8zvgws8FMvMp6IcsyA9ad48SdMPaOboiznYVuBxj7HH5uCIZmMPgYGOt4PIJaiJZ6gs5kO03CyAw+2QRFXZRWew/9cnTPCw+jSQztiJ5hDMa2h4CczSyLmTxBXWb5/UBKagKbw8/gTLDS43bAYZ0F1CCKc4W84WOXj6lmV43ZZ2fzIuGF0g2kqxvGBnqM23JJ92kgFnbQbTVE3Id9E3Q7FkEnRFAtqFWO0iRMrKHPNmUL2N+6ZzqwTu+eAKrEUiRH1sKw8An0jJ4B2c8RFW32hBxTpXH86BT+DjrK4IhI9Ev3QG/keaW7hxZTPTKSXnGuIOQmepIe0zymEZ8Sz6wzswAoZmXC1I6VOZ9b0kDKeCpF+AHr4fpOXY9G0DLdv5ML+UqOC54s7KD/LrB2VALrByezfU+A5GdtQGWEcbG/ADnbM8tiJi9vyXWdWh75k7CyKTPubKZ4skRk4dmU6rtC4109BCG/cbZxZrD7YPY92MeRR0cA6FLdkRaVijJ3/03uPNPxpjAAjSdA8Wrw10iIDtH1aAQtEkG1oFZqSZOwtIdPXwbWG7rDgxPZuqcEyGkWJD1viYHFHUysr+VoZlnM5OUNuapTS3QIbB0Mv7TiZ1UYQYYGhEcNZWiP7tofiyDkE5+7fU75QuWZfno6MckxLzeFccPUSJ+vfQNJTcv5hmI5YmAE3VYpOwZvH6qWDc6EvEEE1YJaqS1NwrLYy8C6JKnruzNixuIMZx0zumfveqVwtDElJaIeaYn2mBTbRcvKhbL1utKT62ZDBSCXdGpJSYRj82BJTbi6lTv1BvCLmREpkTWY59UNazND7Y1FEPIZQz1DvD28CUsMY8H5BQAUtTLhh05VCHwcyc/H7ul4hIBteWg9A+4dUTZxEgoEEVQLaqXWNAnLYuytuZIHqUWYkzwdD73L6c46ZnTP6Z3dODmhGQ9mdaRfhTGk6UcwZv+PanmduWo2VHiDTnsuyzJc2wHLasPhH6BsU1TDTvNNYiiqNGPaOQ6kScWimh+HIORzVWyr0LdSX3xv+XI29CwAXlUdaO9enIUHb3EtOFrHIwRqfgYV28HBqfD0qq5HI2iBaKkn5Gqesw6TFBnKeqOZlJWC+TJlOHtU9XC0MeXkhGaZvo4syzReO5BwzrOq+SbqlcpZgaHnrMMEpROkZXVcgvrp7O8m5BLs/xYeHIeirtDGB5yb8Mf1P5l+ZiomkR9zZMg3WBgbaG4MglCAJKQm0HVHV/QkPf7s+CcmBiaExyXTasExbC2M+GtEA4wMdDx3GBcGy+uDuR0MOgIGxrodj5BloqWekG8ERyYQhjU9kycRKJdlqeESeusfzPKsoyRJLG87GQl9xhz0znHrJbEDXe6l9U4tMaGwYzj83EiZjWo3DwYfB+cmRCRGMOfsPFLjy7Cg/UARUAuCGpkamDLVYyqPYh6xPHA5AIXNjfDp6saN0BgWH7qt4xEC5rbQaSk8uwr/zNH1aAQNE0G1kKu9KkKMxoK+yRM5oqrGDMNfmGixS1lqz4Iq9qVo6dCHWP1LTDrgq5ZxZfZxQXu01qklJQH+mQuLa0DgH1B/OIy6AHUGgr4SPI8/PJ0kVTxt7YfToJxI+xAEdatbvC7dyndj3dV1XH2hpFi0dC1G95olWH70DgGPInQ8QqBCa6jWG04sgOAAXY9G0CCR/iHkam/v9GZAKvONVtJJ7zjUHQKtfbLU4zc5NRnPDe1JSE7h93ZbcHO0Vcu4QOxAV2CoVHDpDzg8HaKfQCUvaDENipR947DDD04y+p8hmMW14uiAOZga6WdwQUHIPzK7o262dt7NQHRyNJ23d6awSWF+7/A7hnqGRCem0GbBMUyM9Nk9sqHuf/4SIpQ0ENNCMOioSAPJQ0T6h5BvvD3rWMzGErnzCqg3XKmo3twXkuMyfT0jAyN+aPA9ktELhuycT1Lqu5vGZGdcom91AXH3CPyvEWwf8rKn+m7o+ds7AXVSWhITj01FlVyEpe3G6/4XuiBoQWYLuNVd6G1lZMV39b7jZsRN1lxZozxmYsjcHlW59zyOWXuv5/CVqYFpIfBaBM+uwbG5uh5NrpKfOmmJmWohb5JlJaje/y3Yu0GvTWDlkOnT+/w1nIsv/OhSdCE/tG+swYEK+cLTa/D3ZLjzN1iXghZToHLXDFdJRu+fxeHQDbQpMpm5HXpoebCCoBuZLRLWVDHx10e/5sjjI/h6+VLWRvmg673zGr+cvM+6z+vQuIJdtq+tNtuGKitdAw+DQzVdj0bn8sKqr5ipFvI/SYJ6Q5Vg+sVdWNkMgi9m+vR5zSdhoKfPlvvLOfsgXIMDFfK0yEfKL8EVHvDkDLT8AUacBbfuGQbUZ55c43DIJixS6jCzbTctD1gQdCezBdyaKvSeWHciZoZmTPGbQppKCdLGt6lI+aIWjPMNJCIuOUfXV4s2M5VOINuHQWouGI+O5Yp9BdRIBNVC3lahNXy+HyR9WNMWru/K1Gn25vYMrjoYA8trjNi+gZjEFA0PVMhT4sNh/3fK5i1X/gSPETDqIniOAkOTDE9LU6Ux6uD3yCpjlraZgqG+eIsVCo7MFnBrqtDb1tSWb2p/Q+DzQH6/8TsAJob6LOhZjYj4ZL7ffgWdr87/mwZyFY6JbiD5rZOWeMcX8j77KspSWtFK8EdvODwDVB/OlR7g1h9701LEmm9h6s7Mz3IL+VhSjNLRY1FVOL0c3D6Ckeeh1XQwK/zB0785sJo46TZtHQZQs0QpLQxYEHKPzLazVHfby9dzcn18zShvWZvFAYt5HPMYgCqO1oxpWYHdl0PYfjEX5OtWbAPuHyvdQJ5e0/VodCq/ddISQbWQP1gWU4rGqvVRPv3/1g3iXrz3FEN9Q7wbfI+eUTg7H25k/9VQLQ1WyHVSEsBvqRJMH5kOZRrAUD/ovAxsSmbqEuefPGJf8Cos5ArMbvWFhgcsCLlPZgu41Vno/XbRY3BkItcvtyRNBdNOTft3Znpwo7LULlOIyduvppvPrXWtZ4KxFez6UukoVEBpfV8BDROFikL+c+FX2D1WyVv7aB2UeH99wVdHvubgwyPoBY9j/4guFLPKeHlfyGdSkyFgvVKNHxMCzk2g2aQP/pt5W3KqioZrPyfe4CK/tNxEbUcXjQxXEIQ3ZVT0aOdwnkRrX6Z5TKNr+a4APA6Pp83CY1RxtGbjwHro60naHu6bLv6udBJq/yPULrgfxNXZXlETslKoKIJqIX8Kvqi024sOgdYzoM4gpbgxHaFxoXht60h8dBmqGo3ht8/roafrN1tBs1KTIXAjHJsPUY+gZF0lmHZq+M6hmXnDH75tPcei59DGoT9zW36trVchCAWe04TdpBfFSKho3ORPboXfYlunbRQzLwbAlvNPGOsbyLjWFRnetJx2B/s2WYZfO0JwIIw4A5b2uh2PkC7R/UMQHKrB4GNQrjnsHQ8be0Lss3QPtTe3Z1i1oeiZX+dM6HFWnbin5cEKWpOaDOfXKgWIO0crvaZ7b1GKXTMIqD/UT/fgjYccffEzlnolmdlslPZeiyAI78nJNcfbw5sUVQo/nP7h3zSQbjUc8arqwI9/39L9bouSBB0WQmoi7P1Gt2MR1EIE1UL+ZVpIabnXdg7c/0fZzerm3nQP7ePah3I25bAuuZu5By5xJShKy4MVNColEc6ufjeYHnAIyrfMcBXjQ+2eIuKSGXdoFnoG0SxqPhNDfUONvxRBEP7zvpzcUlalGFF9BP88+Yfd93cDIEkS0ztXwd7KhNGbLhKblKqLYf+nSFloNA6ubYdb+3U7FiHHRFAt5G+SBHUHK9vCWhWH3z+GnV++swujoZ4hk+pNIplwLO2PMur3AOKTdfxmK+RcUiz4LVEKEHd/BRZFMxVMv/K+dk+yLDNsy1ZSzE/SplQ3aouNHARB6z5U9NinUh/c7dyZdWYWYQlhAFibGrLw42o8iYhn8o4rOhz9S56jwc4Fdn+tvGcJeZYIqoWCoWglJZDyHK0s/6/wULacfk2NYjXoUq4LaVb/8DDmLt47C3arozwtPhyOzoaFVeDA92BXAfr9BQMOZiqYfuV97Z5+879HYNJKrAxsmdpwrDpHLwhCFnSu7sjJCc24P6s9Jyc0e6PmQV9Pnx88fiA+JZ6Z/jP/fbx2mcKMbFaerReC2KHrNnsGRkoaSNRjOOqj27EIOSKCaqHgMDCGlt7Qf5eyWcz6zspuefH/7ag4puYYLI0sKF1xH5vOPmTP5RAdDljIsogHsGc8LKgMR2cqBYhfHIRPd4Jz40wH069ktLTcp14pfPyWoW/8jFmNp2JuaK7GFyEIgjo52zgzrNow/n74NwceHPj38ZHNylGzdCG+33aFx+HxOhwhULo+1OgHp1fAs+u6HYuQbSKoFgqeMg1g6EloOBYub4alteDSZpBlCpkU4quaX/E85QbOTjf45s9LPHqh4zdb4cOCLoBvf1hcHc79Aq6dYegp+OQPKFk725dNb2l5WsfK/BF4Dv3Ch2lWojWNSjZS28sQBEEz+lfuj2sRV2b4zyAiUSlQNNDXY2FPJW1r5O8BJKfquF9086lgbAl7ximdQYQ8R7TUEwq2p1fhr1EQdA6cGkFrH1TFXPl076fci7xP9J2vcC5cFN8hHhgZiM+guUpaKtzYCad/gsenlY0UavaHekPBykFjtx3nG8DuF99ibRnL7m47KGyi7LSY23utCkJBdzP8Jh/v/phWpVsxu9Hsfx/fczmEYRsuMLChE9+1d9XhCIGzq5Tc6u5roEpX3Y5FAERLPUHIvGKV4YsD0G4ehF6Gnxuit2sMk6qOIC41lto1/Ah8EsXsfTd0PVLhlfhwOP4jLHJXZqdjQ6G1D4y5Aq1+0GhAvT0giO33f0ff9AlTPL97I6D+UOs9QRB0q2LhigxyG8Se+3s49OjQv4+3cytO33qlWXn8PgevPdXhCIGan4G9m1ILIooW85wcBdWSJBWWJOlvSZJuv/yzUDrHVJMk6ZQkSVclSbokSVLP155bK0nSfUmSLr78EuXzgvbp6UOdgTAqAOoOgYDfqLCuG/2sXTn7Yj/tasWx+sR9/tb1m21BJsvw+AxsGwLzXeDQNKUV1ce/w8gLUH8YmFhrdAj3nsfy3a5DmBb9m2Ylm9G6dOt/n/tQ6z1BEHKHAe4DcCnsgvcp73/TQAC+a1+Jyg5WfO0bqNttzPX0lUme6CA4Pk934/iA7QFBeM46jNOE3XjOOiwmEF7K6Uz1BOCQLMvlgUMv//tt8UA/WZYrA22AhZIk2bz2/DhZlqu9/LqYw/EIQvaZFoI2PjDsNJRpwNCLe3BMk3mUvAR3BwPG6vrNtiBKjIYzK+GnBrC6JVzfBdX7wFA/pfjQpZ3yS0jTw0hJY/jG8+gV9cXCyIzv632P9FrR4/ta7wmCkHsY6hky3XM60cnR+Pj/12nDxFCfZZ/UIE0lM3LjBVLSdJhfXaoeVO0Ffksh7I7uxpEBsTKXsZwG1Z2AdS+/Xwd0fvsAWZZvybJ8++X3wcAzwC6H9xUEzbEtD59swrTfX0zGlocpkbQw+Ipuqv2M2eCv2zfbgkClgnv/wNZBMK8C7BkLkh54LYKvb0CHH5W0HS2avvsadxL3g8kDJtT9BjuzN9/C3td6TxCE3KVi4YoMdh/M3gd7Ofjw4L+Pl7E1x6erGxceRTLvgI5XmVpMA0NTZUfgXFb7JlbmMpbToLqYLMuveo6FAsXed7AkSXUAI+Duaw/PeJkWskCSJOP3nDtIkqRzkiSde/78eQ6HLQiZ4NQIj8+O0MGuFmvM9PjEaB0/PvuCI2unKjOognqF34MjPrC4KvzaEW7ug6ofw4DDypbzNfuDsYXWh7Ut4Akbzgdgbn+ABo4N8HL2eueY9+3qJghC7vOF2xdUKlyJH07/QHjif21Vvao60LtuKX7+5x6Hrusw5c+yGDSZCHcPwY3duhtHOsTKXMY+2P1DkqSDgH06T30HrJNl2ea1YyNkWX4nr/rlc8WBo8Cnsiyffu2xUJRA+3/AXVmWvT80aNH9Q9Cm8MRwOm7riJNxYbzvhuMUd5kUAwsMa32q7NZYqLSuh5h3RYfA1W1wZQsEnQckcG4C1XpDpQ7KTI0O3QiNpvOy41g7rUEyecL2TtuxN0/v7VB0/xAEbVDnz9mtiFv03NWT5qWaM6/xf/nLiSlpdF3ux5OIeHaNbEipImbqGn7WpKUoqW+pSTDcX9lrIRfwnHU43VRIRxtTTk5opoMRaVZWun/kqKWeJEk3gSayLIe8CpplWX5nakaSJCuUgHqmLMtbMrhWE2CsLMsdPnRfEVQL2rbjzg6+P/k9E2t/x8W/E2gU7kt7fX8kWQUuHaDmp+DcVCv5vXledAjc3A1Xt8ODE4AM9u7g1h2qdAPrEroeIQDRiSl0XHKCaKMjpNhsY2r9qXSr0E3XwxKEAutVLu/rqQemhvpvbEueVSsvrWRxwGLmNZ5H6zL/FR8/ehGP19ITONiYsnWoB6ZGOnpvv30QNnSDVjPAY4RuxvAWTfw95GbaDKrnAi9kWZ4lSdIEoLAsy+PfOsYI2AvslGV54VvPFX8ZkEvAAiBRluX0ih3fIIJqQdtkWWbggYFcfXGVlc3+oP+qmzgbR7He7RLGgb9CQgRYOSrFJdU+UTpTCP8Ju6P0lL6+S+kJDlCkvBJEu3VX8tizSRMzxLIsM3j9eY7cvYplucXUK16XZc2XvVGcKAiCdmlihjRVlUqfPX0Iig1iW6dt2Jra/vvckZvP+HztWbpUc2T+R1V19/P/Wzd4fFbpUGVeRDdjeEtBWpnTZlBdBNgMlAIeAh/JshwuSVItYIgsywMkSeoDrAGuvnZqf1mWL0qSdBilaFECLr4854ONGUVQLejCo+hHdP2rK/Ud6tOr1BT6/nKGVq7FWN6zMtKtvRCwQcl/k1VQqr6yq1+lDrlm5lWrkuOVWeg7f8Odg0q+NIBDdXBpDy5eYFcxy9uGv01TMyYrjt5l9r6rlK/2K/FyCNs6baOoWdEcjVUQhJxxmrCb9CIWCbg/q322r3sv8h49dvbAw8GDxc0WvxE8Lzp4mwUHb/FDp8r0rV8m2/fIkWfXYYUH1B4A7ebqZgwFmNaCal0RQbWgK+uurmPeuXnMajiLoCcuzNxzg2/buTCo0cuZ6ehgCNwEl33h2TXlMYcaUMlLSROxLZ/jQDJXSk2G4AB4eALuH4OHpyAtCQxMwakhlGuptL9T8wcMTcxc+d0No88qf6q4nue+ypfZDWfTzrldTocqCEIOaTKX99ervzL33Fy8PbzpUr7Lv4+rVDIDfj3H8dvP2TSoPjVLp1s2pnm7xsD5dUrLV7sKuhlDASWCakHQkDRVGv329uNhzEO2d9zOpK0P2X81lPVf1MWznO2bB79Kebj2FwRfUB6zdFAK8Zwbg1NjsCqe5THkimW3+HAliA46Dw9PKhuzpMQrz9m5QNnmUL4FlPIAQxONDUPdM1ePw+PpuPQEVtZhRBeeS7OSzZjXeJ5I+xCEXECTubwqWcUX+7/gevh1tnbcioPFfzuzRsWn4LX0BEmpaewc2YCilpp7T8tQ7HNYUgNKe8Anf2j//gWYCKoFQYPuRt6lx84eNCnZBO/6c+i6/CRPo5PYPtwTJ1vz9E+KfKykQtz7R5nJTXjZwqlwWXCsAY41lRnt4u7v7Xih9QIRWYaYUHh+HZ5eg5CLSiD9Kp0DoGhlKNMAynhCaU8wt834emqmzpmr2KRUui33IyQ6ljLuq4hKfsH2TtspZKKjmSlBEN6hyUmFoNgguu7oSmXbyqxqtQo96b+uw9dDoumy/CSuxa34fVA9jA10ULh4YiEcnAJ9t0PZptq/fwElgmpB0LBVl1ex6MIi5jeej4tlAzotO0FhcyO2DffEysTw/SerVPD0shJgP/aHoAsQE6w8J+lDkXJKmohtebCtoBT02ZQEczs85/yj/uVPWYa4MIh8CBEP/vsz7I6SwpIY+d+xVo5KXrRjTeXDQPFqYGqT0ZU1Tl0fMlQqmcG/nefwjWd0bXaJfUEbWNx0MU1LiV9cglCQbL29lSl+U/im9jf0ce3zxnN7LocwbMMFutZwZH4PHRQupiTCsjpgbKn07hfdprRCBNWCoGGpqlR67+lNaFwo2ztt50aQir6r/WlQ3pbVn9ZGXy+Lb7bRIUqKSNAFpSjlxW1lNliV+t8xkj4hKmueyjY8lwsRhTmxsgmxmBInm/JNx5qgpwdI/+VtyzKkJCipGclxyvdJ0RD3XPmKffmnKuXN8ZjZKsF90UpQ1PXln5W0OgudWeqYuZp/4CZLDt9hQAsVm4O+o2PZjkxvMF1DIxYEIbeSZZmRh0dyOuQ0m70242zt/MbzrwoXJ7Z1YXBjHXR5uroNfPtDxyVQo5/2718AiaBaELTg1cYBLUu3ZE6jOWzwf8h3264wqJEz37arlPMbpKVAxEMlwI56AjGh7PK7gGVyGEWlCCylBCxQvgykTGydrm8EhmbKLIe53WtftmBZXNnExqY02JTSyc6FurIzMJiRvwfQrVZhLjMFAz0DfL18MTfMIJVHEIR8LSwhjC47uuBg4cBv7X7DUO+/1UdZlhn5ewC7L4fwhacTe6+Eare+RZZhdUuICoJRF3S+QVZBkJWg2kDTgxGE/KpCoQoMch/E8ovLaVW6Fb3rtuBmaAz/O3aPCsUs6V4zh50u9A3Btpzy9VJq4SCGbL1MQvLr6Q56zO5UgY6VrJV2frIMyMqfkqS86Rqag774cX/b5SdRjNsSSK0yNki2W3n26Bm/tv1VBNSCUEBktNI1pf4Uxhwdw4qLKxhVY9S/x0uSxNzuVQl8HMmqE/f/fTwoMoGJWy8DaDawliRoMRXWtgf/n6HBl5q7l5Bleh8+RBCEjAxwG4CFFC+5AAAgAElEQVRrEVemnZpGWEIYkzq44lG2CN9uvYz/vRdqv1/n6o74dHXD0cYUCSWX2qerOx1rlVVmnC2KgmUxsLRXOotY2oOJtQio0/E4PJ7P152liLkxXRs+Zf/DfQyrNgx3O3ddD00QBC14VZMRFJmAzH+B8faAIFqUbkGXcl1YdXkV55+ef+M8UyN9UtLeXeVPSElj7v6bmh94mQZKm9ITPyobjwm5hkj/EIQcuhd5j492fUTd4nVZ2mwpUQkpdFvhx/OYJP4c6kH5Ypa6HmKukhtaAkbFp9DtJz+eRSey7NPSjDvVn4qFKvJL61/QF8U/glAgfKh7UHxKPN13didNlcaWjluwNPrvvVxTG9FkWuhl+KkheI6GltM0f78CLCvpH2KmWhByyNnGmTE1x3DsyTG23N6CjZkRaz+rg7GhPv3XnOVpdKKuh5gl2wOC8Jx1GKcJu/GcdZjtAUFqvXZGM0PakpSaxqD153j0Ip4Vfavx8/Xp6KGHT0MfEVALQgESnE5A/frjZoZm+DT04Wn8U3z8fd44xsEm/VzmjB5XO3s3cOsB/j8pm44JuYIIqgVBDXq59KJe8XrMPTuXR9GPKFnYjDX9axMZn8xna84Sm5T64YvkApoOeufuv/lG+zvQ4pIpSuu8sb6X8L8fztwe7gRE+xL4PJBJ9Se9sdmDIAj5X2YC46p2VRnsPpid93ay7/6+fx8f17oipoZvfgjXlyTGttLibofNvgNV2v/bu+/4qKr0j+OfkwZJCAklJBBAei8Cka5IEUQUsKw/FewCq2sFFGxrwYKCiruwKK4VwY6KBQFpIkUMvUsntNA76ef3RwJLSZ9yZ8L3/XrlleTOmXueceTmmXPPeQ7MHu69PiVPSqpF3CDABDCs3TCCAoJ48vcnSc9Mp1FcJP/p24L1Sce4/9PFpGUUoEKHwzyd9OY3MuRpI6at54fluxhydT2iK2xj3Ipx9KrZi+7Vu3ulfxHxHTklxqHBgTzere45x/o16UeT8k14ceGL7DmxB7hwfUtEiSAyrGX7Qe9cywAoUw3i74Gln8L+Dd7r1wWevBPqC5RUi7hJbHgsz7Z+lhX7VvDBqg8A6FAnmldvaMzcDfsZ+s1KMjN9ew2Dp5NeJ2+ZfjhvC2Nnb6Jv66rccFkpnpz7JDWjavJ066c93reI+J6cF35fuHFUUEAQr17+KhmZGQz5bQjp2fsH9G4Wx7yhndgyvAcrnu/Kjc0r89avf/Hln4neexFXPJ5V4WnGi97rs4h8YfqfpympFnGj7tW7071ad8YuG8uKfSsAuDm+Co91qcM3S3bw4o9r8OXFwZ5Oegs6MuRuX/6ZyAs/rKFbwxie6VGXoXOHcir9FCM7jCQ0SHVeRS5WZyfG84Z2ynXRdNXSVXmm9TMs2buEcSvGXfC4MYbhNzbm8trlefLblfyyao+nQ89SKhraPAhrJ8OOxfm3d5DT0/+8QUm1iJs90+YZYsJjeOK3JziaehSAhzvX4t721flo/lZGTvPdC4ink96Cjgy5048rdjFk0gpKBAUwdXUS7d59koSkBJ5p/Qw1oxzYEU1E/NJ1Na/juhrX8e6Kd/lzz58XPB4cGMDYvi1oUjmShz5bwqz1e70TWNsHIbQszH7FO/0VkdPT/7xBSbWIm5UOKc3rV7xO0okknp//PNZajDE806M+t7WqyphZmxgza6PTYebIG0lvQUeG3GHmuiQe/mwpBkhJzyQw/C9SSk0n8+hlZB5t4bF+RaR4err101SJqMLQuUM5nHz4gsdLlQjio7tbUicmgr+PX8z8jfuL3FeB5x+XiMgqrbfxV9j+R5H787TC3gkds2wMoxaP8mRIbqekWsQDmkQ34ZHmjzB923S+WP8FkHV78KVejbi+WRwjpq7nw3lb8jmLM7yZ9HrS/E37+funSwgMMGRaMEFHKFnpCzJTKnBi13XF6pajiHhHeHA4r1/xOgeTD/Ls/GdznM4XGRrM+HtbcUm5MO77JIGErQcL3U+h5x+37Afh0TDr5UL35S2FuROanJ7MhLUT2HXCv8oFKqkW8ZA7Gt7B5XGX8/qfr7Pu4DoAAgIMI25qQreGMbzwwxo+X7Td4SiLp0VbDtLv4wQuKRuWtfOZSSO08qeYgDSSd/YBG1KsbjmKiPc0KNeAgS0GMjtxNp+t+yzHNmXDQ/j0vlbElC7J3R/+yYodF45q56XQ849DwqH9Y7BlDmz9vVB9eUth7oRO3zadY6nHuKn2Td4P1AVKqkU8JMAE8HL7lylTogyD5wzmRNoJAIICA/jXrc24sm40QyetZPyCrY7GWdzM27ifOz9YRExkST69rxWVokpSIvZ7AkMTSd51M5mpFQAvbtIgIsVO3/p9uaLyFYxMGMnaA2tzbFMhoiQT7mtFZFgwt7+/iGWJBU+sizT/OP4eKBUDs14BH10QX9A7oV//9TVVI6pyWexlXo7QNUqqRTyoTMkyvHbFayQeS2TYwmFnbhWWCArknb4t6FI/hme/X8243zY5HGnxMGvdXu7+6E8uKRfGF/3bEFO6JB1abCAkKoGU/R1JP9YI8E7FEREpvowxvNTuJcqULMOgOYPOLEo/X6WoUD7r15rSoUH0eW8hCzcfKND5i1SJKTgULh8E2+ZljVj7qc2HN7Nk7xJurHMjxhinwykUJdUiHhYfG88DTR/gp80/MXHdxDPHSwYHMrZvc65tUpFXfl7HW9P/8ulye77ul1V76D8+gToxpfisX2uiI0qwOGkxv+x+lzoRLSmf1tNrFUdEpPgrU7IMb3R4g93Hd/Ps7znPrwaoUjaMrwa0JTayJHd+sIjZBagKUuRKTM3vhNJxPj1anZ9vNnxDUEAQvWr2cjqUQlNSLeIF/Zr0o2OVjoz4c8Q5pZiCAwN4+5Zm3NSiMm/P2MCrU9YpsS6Cyct38Y+JS2gUF8mE+1pTJjyEPSf2MHD2QOIi4vjw2reYP7SL3y++FBHfcmmFSxkYP5CZiTP5ZM0nubaLjSzJlwPaUDO6FP0+SWDKyt15nrfIlZiCS2aNVif+ARtnFOEVOSs1I5XJmybTsUpHyoWWczqcQjP++Ac8Pj7eJiQkOB2GSKEcTz3OrT/dytHUo3ze43Mqlqp45rHMTMvzP6zmkwXbuLVlFYb1akRQoD7zFsT7v2/hpZ/WcFm1snxw12WUKhFEcnoyd/9yN1uObmHiNROpEVXD6TBFpJiy1jJoziBmbp/JB90+oHlM81zbHjmVxt0fZs2vfv2mptzUorL7A0pPhX+3gPDy0G8m+NEUiilbpvDEb0/w7lXv0rZSW6fDAcAYs9haG1+QtvqrLeIlpUJK8Xant0nJSOHR2Y+SnJ585rGAAMMLPRvywJU1+WxRIvd8nMCx5DQHo/WuAtdjPUtGpuX5yasZ9uMaujaI4eO7W1KqRBCZNpOnfn+K1QdW80r7V5RQi4hHGWN4oe0LxJWK4/E5j3PgVO7zpk+X22tTsxyDv1rumWl/QSHQ4XHYtQQ2THfvuT3s67++Jq5UHK0rtnY6lCJRUi3iRTUia/Bq+1dZc2DNOQsXIevC/MTV9Rh+Q2PmbdzP395ZwM6LoOxboeuxAidT0xkwfjEfzd/Kfe2r858+LQgNyZp/OGrxKKZvm86g+EF0qtrJS69CRC5mESERvHnlmxxJPcKQuUPIyMzItW14iSA+vKslNzbPmvb3yOfL+CohsdADC3lqcgtEVoHfRvjN3OptR7exaM8ibqx9IwHGP9NT/4xaxI91rNqRB5o+wORNk89ZuHh6tPbJSSuJCg1m64ET9B4zr9D1Tf1NYeux7j2WzC3jFjJzXRIv9GzIM9c2IDAg6/bmF+u+4MPVH3JL3Vu4o8EdHo9dROS0umXr8nSrp/lj9x+8veTtPNuGBAUw8m9NeLxbXSYv38WQb1YUamAhX0Eh0P5R2LEItvxW9PN40TcbviHQBNK7Vm8AVuw4zNwN+xyOqnCUVIs4YEDTAXSs0pHX/3yd2YmzLxitPXAiFWshPSOTm99dwE8r8l7U4s8KU491WeJhrh8znw1Jxxl3ezx3tq125rHfdvzGK4teoUPlDgxpOcTvSjGJiP+7vvb1/F/d/+PD1R/y0+af8mxrjOEfHWtRNiyEzPMGk/Pc6KWgLu0LpWKzRqt9XFpGGt9v/J4OlTtQPrQ8E//Yzk1jF/Dqz+vIPP8/jg8LcjoAkYtRgAlg+OXDuWfqPTzx2xME7LmfU2kVzmmTkp5JmbAQqpUP5x8Tl/Dn1mo8eU09SgQF5nJW/1QpKjTHaS5n12O11vLJgm289NMaYkqX5Ku/t6FRXOSZx9ccWMPgOYOpW6Yur1/xOkEBurSJiDOGXDaEDYc28Nz856geWZ0G5Rqc8/h3S3cyYup6dh0+RaWoUA6eTM3xPAXd9fX88z3erW5WlZDgktDuYZj6FGxfCFV9d57yrMRZHEw+yHU1rmfwVyv4ZskOrqgTzdv/dykBAf4zQKKRahGHhAWHMbrzaMqWLMuxqHGY4AsXtyQdTeaL/m24p111Ppq/lZvfWUDiwZNui6EoCwTdLb96rMdT0nnos6U8N3k1V9SO5seH2p+TUCceTeTBGQ8SWSKSMZ3HEBYc5tX4RUTOFhwYzJtXvkmZkmV4ZNYj5yxczGkNSW4pY2xkyXz7yndNSou7IKwc/DbS1ZdVaAX9+2KtZfya8VQIjWXEdzBp6Q4e7VKbD++6jDLhIV6O2jVKqkUcVD60PGO7jCUwIJOwKh9C4IlzHq8UFUpIUAD/vK4B7/Rtweb9J+jxr7lMW73H5b6LskDQE/Kqx7p+zzF6jv6dn1fu5omr6/LeHfFEhf3vIrvnxB7um3YfaZlpjO08luiwaK/GLiKSk3Kh5RjVcRSHkg8xaM4g0jKzqjnltIbEQo6JtbXwV9KxPPvJd01KSDi0eRA2ToddS4v6cgqtMH9fFu5eyLJ9y9ib2Jako6l8eNdlPNqlzpm1Mv5ESbWIw6pHVufe2sMwwYcJq/wJmKyL7/m7Z13dKJafHrqcS8qF03/8Yp76diVHXSi7V9gFgp7Uu1kc84Z2OrM5S48mFRkzayPX/ft3jp5KZ8J9rXngylrn3Abcf2o/9027j6OpR3nnqneoVaaW1+MWEclNw3INeaHtCyxOWsxri14Dcp/SYeGcgYX7O9QkPTOTnqN/Z/yCrbnOKy7QmpTL7oOSkV4drS7o35f9x5IZ/OtIMtNKU73ElfzwYHuurHvuVEh/oomHIj7gkfZdOZZ2kC+2vUJo3ETKHO/HE90aXLB7VtVyYXx9fxvemPYX/527mZlr9/JS70Z0aRBT6D4Ls0DQm1bvOsITX69g9a6j9GhckRd6NaR8qRLntDmcfJh+0/qx9+Re3r3qXRqWa+hQtCIiuetRowfrDq7jo9UfUT2yOpWiKua4hiQuKpR5Q88tAXp3u2oM/HI5z36/mu+X7WL4jY2pVSHinDYFWZNCydLQ6n6YMxySVkOM56+XBfn78vPK3Tw95VvSY/7iigr9+HePDn6/6Zl/Ry9SjDzT8Raebv0UQRFrad16Cj2a5vxpvURQIE9dU59vH2hHVFgw932SwIMTl7D/eEqh+jvnoluA456Wkp7Bm9PW02v0PJKOpjC2T3PG9Gl+QUJ9LPUYA34dwPaj2/lXp3/RrEIzR+IVESmIR5s/eqba03VtDuW5huRsFUqXZPy9LRlxUxM27D3ONW//zqhf/yIl/X8jwPmtSTmj1QAIKQVz33DfC8tDXn9f9h5L5h8Tl/DAhCUElZtBmRLlGdVjgN8n1KCkWsSn3FLvFp647Ammb5vOU3OfIj0zPde2TatEMfnB9gy8qg7TVifR5c05fDhvyzkX3LwU+GLsYdZaflyxi65v/ca/Zm6kZ9NKTH/sCro3rnhB22Opx3jg1wf46+BfvHnlm36765aIXDwCAwIZfvlw6petz6Qdr/Hg1SVzXEOSE2MMf4uvwoxBHbi6USyjft1Aj3/9ztwN+7DW5rkm5RxhZbOmgayaBAc2efw15/T3pWRQAE0qR3LliNlMW72HPh3SSQ76i/5N7qVEYIlczuRfjNu3x/SC+Ph4m5CQ4HQYIh7z8eqPGZkwkmuqX8Mr7V8hMCDvMnobko7x3OTVzN90gMplQhnUtQ49m8blu9Aj11JMXrJg0wGGT1nL8h1HqBsTwVM96tOhTs6LDQ8mH+Tv0//OhkMbeO2K1+haravX4hQRcdX+U/vp81MfUjNTmXjNRCqWunDgID+z1u/l2e9WsePQKVrXKMvj3erS4pKyBXvysSQY1Qia9YVr3yp034V1+u/LzsOniAoNxgJHTqVxdcNYhnSvx8tLHmHjoY38cuMvlAzKv9KJU4wxi6218QVqq6RaxDf9d+V/eXvJ21xX4zqGtRuWb2JtrWXuhv289ss6Vu86Sr3YCAZ3rUunehV8rs7n8sTDvD1jAzPX7aViZEkGXlWHG5pXzvVDwJ4Te+g/vT+7j+/mzSvf5PLKl3s5YhER1208tJE7ptxBTHgMn3T/hIiQiPyfdJ6U9Aw+X5TIv2duZP/xFDrXq8CgrnVpUKl0/k+e/DAs/xweWwWlPLsgMC0jk19W7WHMrI2s23OMS6tE8UyP+sRXK8vSvUu5Y8odDI4fzJ0N7/RoHK5SUi3iEHeP/L67/F1GLxtN9+rdeandS4QE5l+zMzPT8uPK3bwxbT3bDpykevlw+ra+hJtaVCYyNLjIsbgqNT2Tn1fu5qP5W1mWeJiIkkH8o2Mt7mpbjZLBuX9g2HZ0G/2m9eNY6jFGdx5Ni5gWXoxaRMS9Fu5eyP3T76dFbAv+0/k/Bbqu5+Rkajofzd/KO7M3cTQ5nfa1ytO3dVU6148hOLf5yfs3wuh4uHwQdH7WhVeRu8MnU/lsUSKfLNjK7iPJVC8fzqCudejRuOKZnW4HTB/AuoPrmHLDFJ/fW0BJtYgDTtflPLuMUGhwYJ7z5Qri9Ih1q9hWvNXxrQKPbKRlZPLjil18smAbS7cfJjQ4kN7NKnFby0toFFfaa9t4b9l/gu+W7mTiou3sO5ZCjfLh3Nm2Gjc0jyOiZN5J/vqD6xkwfQCZNpN3rnrngp3J3M3p6TAicnGYvGkyT//+NJ2rdmZkh5Eu7QJ75FQan8zfymeLtrPrSDIVIkpwy2VVuKVl1ZwXDH7RF7b8Bo+thhKFHynPSUamJWHrQb5fvotvl+zkVFoGbWuW49721elY99y7pcv3Lafvz315rMVj3NPoHrf070leS6qNMWWBL4BqwFbgZmvtoRzaZQArs3/dbq3tmX28OvA5UA5YDNxurc15v86zKKkWX9Ru+MwCl0oqrMmbJvPcvOeoEVWDsV3GUiGscLftVu08wicLtvL9sl2kpGdSMbIknepVoEv9GNrULJfnSHFhZWZalu84zLQ1SUxfk8TGvccB6Fg3mjvbVuOK2tEFmo4ya/sshs4dSkRIBOO6jqNGZA23xZgTT30oEhHJyYS1Exi+aDg9a/ZkWLthBBjXakdkZFpmr9/LhD+2M2v9Xqwla2FgnWiurFeBppWjsqbY7UiA/3aGbq9Am38Uub+U9AzmbzrAtNV7mLY6iQMnUgkJCqD3pZW4u1116le8cDpKRmYGt/18G0knkvj5hp99fpQavJtUvw4ctNYON8YMBcpYa4fk0O64tbZUDse/BCZZaz83xrwDLLfWjs2vXyXV4ouqD/2JnP41GWDL8B4un3/+zvk8NvsxIktEMrbLWGpG1Sz0OQ6fTGXa6iRmrEti7ob9nEzNIDQ4kPhqZahfsTT1YiOoGxtBrQqlKBGUf6KdmWlJPHSSNbuOsmb3UdbsOsryHUfYfzyFwABDq+pluapBDFc1iKFymYJdPK21vLfyPUYvHU2Dcg0Y1XEUseGxhX6thVWYD0Ua0RYRd3hn+TuMWTaG2+rdxtCWQ912BzHx4Em+X7aTWev3sXT7ITItlAkLpm3N8tSLjaDP2vspnbwDHl5GUEj+iwTTMzLZvP8EK3ccYeXOrK81u45yKi2D8JBAOtWP4eqGsVxZN5rwErmPuo9fM57X/3ydER1GcHW1q93yWj3Nm0n1euBKa+1uY0xFYLa19oJ6XDkl1Sbr/5x9QKy1Nt0Y0wZ43lrbLb9+lVSLL/LkSPVpaw+s5YEZD5CSkcIbHd6gTaU2RT5XcloGf2w5yIy1SSzedogNScdJzcgEIDDAUCGiBKVKBFGqZBARJYMpVSKQ1PRMjpxKO/N1+GQaKen/e07N6HAaVoqkQ51oOtatQGRY4eZwn0w7ybPznmXatmn0qNGD59s877VV4QX9UKQRbRFxF2stIxNG8smaT+jfpD8PNXvI7X0cPpnKbxv2M3vdXhZtPciOQ6e4MmAZH4W8zhPpD7AosithIUGElwgkNCSI8JBA0jIsh06mZn2dSOXwqTROp4thIYE0rFSaRnGRXF67PG1rli/Q3c7dx3fT6/texMfEM6bzGK9NQXSVN5Pqw9baqOyfDXDo9O/ntUsHlgHpwHBr7XfGmPLAQmttrew2VYAp1tpGufTVH+gPULVq1Rbbtm0rctwinuCtZGvn8Z08OONBNh3exN+b/p0BTQbkWxmkINIzMtl64ARrdx9j/Z5jJB1N5nhKOsdT0jmWnM6x5DRCggKJDA0iMjT4zFeN6FI0rFSaOjERLk0j2Xl8J4/OepT1B9fzWIvHuKvhXRdcdD05QlzQD0Xe+PAkIhcPay3PL3ieSRsm8WjzR7m38b0e7e9ESjobk45R7aurSE3P4MUq73MqLYOTqRmcSM3gZEo6gQGGsuEhlAkPoWxYCGXCgqkeHU7juEiqly+Vb7nW81lreWjmQyzas4hve31LXCn/GYAoTFKd78x4Y8yvQE73Xp8++xdrrTXG5JahX2Kt3WmMqQHMNMasBI4UJMCzzj8OGAdZI9WFea6IN5xO7jw9LSCuVBwTrpnAy3+8zNjlY1mStIThVwynfGh5l84bFBhArQoR1KoQwXVN3RRsAVhr+WHzD7z6x6sYDGM6j8mxZN75H1p2Hj7Fk5Oylmq447/x493q5vih6PzNcHx1e3cR8U/GGP7Z+p+cSjvFqCWjSMtMY0CTAR4byQ0vEUTTqmWgy2D4dgD/jt8HdfKdJOCS6dumM2fHHAbHD/arhLqw8k2qrbVdcnvMGJNkjKl41vSPvbmcY2f2983GmNlAM+AbIMoYE2StTQcqAzuL8BpEfEbvZnFemQIQFhzGy+1f5rLYy3h54cv87Ye/8drlr9GyYkuP952Xwo4kH0w+yIsLXmTG9hk0r9Ccl9q/RJWIKjm2HTF1/TkJL8CptAxGTF3vlv/mBf1QVCkqNMeRaqe2dxcR/xcYEMgrl79CcGAwY5aNITk9mUeaP+LZKRKNboQZw2De2x5Nqo+mHuXVRa9Sv2x9+tTv47F+fEHRa7hkmQzcCQzP/v79+Q2MMWWAk9balOwpH+2A17NHtmcBN5FVASTH54tI7nrX6k3Dcg0ZNGcQ/ab345a6t/BgsweLtKGAqwo7kjw7cTbPzX+OY6nHGNhiIHc0uCPPaSzeGCEuyIeigo5oi0jx4Y3FyUEBQQxrN4wSgSV4f9X7JGckM+SyIZ5LrAODoc0DMPUp2LkE4pp7pJu3F7/NweSDjO482qXSgf7AtfotWcn0VcaYDUCX7N8xxsQbY/6b3aY+kGCMWQ7MImtO9Zrsx4YAA40xG8kqq/e+i/GIXHRql6nN5z0+5+Y6N/PZus/o+V1PpmyZgrdr0Oc1kny2xGOJDJ4zmIdmPkR0aDSfX/s5dze6O9954bmNBHt7hLh3szhevaExcVGhGLLmUmuRokjxdXrAYOfhU1j+N2Dw3VL331wPMAE82/pZbm9wOxPWTuCFBS+QkZmR/xOLqtntEBIBC//jkdMv3L2QL//6kj71+9CwXEOP9OFLtPmLSDGyev9qXlz4ImsOrKF1xdY83eppqkVW80rf+VXPOJJyhHErxjFx3USCA4K5u+Hd3Nf4PoIDC1YhRFU3RMQJTixOttby76X/5r2V79H1kq681P4lQoM8NIDwy5OwaBw8uhJKV3Lbabce2Uqfn/sQHRrNxB4T/aImdU4Ks1DR1ZFqEfEhDcs3ZOI1E3mq1VOs2r+K6ydfz7PznmXz4c0e7zu3EeOKUYGMXzOeayZdw/g147muxnX8eP2P3H/p/QVOqEEjxCLiDCcWJxtjeLj5wwyOH8z0bdO555d72Hdyn2c6azUAbGZWYu0mR1KO8NDMhwg0gYzuPNpvE+rC0ki1SDG1/9R+xq0Yx7cbviU5I5lOVTpxb+N7aRLdxCP9nT+SbIIPEFbuD8LLL+FUxnHaVGzDoPhB1C2rucci4j+cLqM5a/sshswdQumQ0ozuPJp6Zeu5v5Mv+sKWuTBwDYSEu3SqtMw07v/1fhYnLeb9ru/TPMYzc7W9RSPVIkL50PI81eoppt40lQFNBpCQlECfn/twx5Q7mLB2AntO7HFrf72bxfFi77pUiNlEaJUPKFVrBIFl5pF2vBYnt/Zn9ZJbWLvtgo1VRUR82uPd6hJ6Xg1+by5O7li1I590/wRjDHdMuYOZ22e6v5PW/4Dkw7D8M5dOY61l+B/D+WP3HzzX5jm/T6gLSyPVIheJk2kn+fqvr5m0YRKbjmwCoFG5RnS+pDPt49pTI7IGIYEhhTqntZYNhzewYNcCFuxaQEJSAikZKUSHRtOodDemL6zGqeT/JdKaAy0i/sgb1T/ys//Ufh6e+TCr9q+if5P+DGg6gOCAwu1amytr4b2OkHIM/vEnBBRtzHXC2gkMXzScuxvdzcAWA90Tm8O8tqOiU5RUi79z4gJ9dp8x5Y7Stsludqf9yaoDqwAINIFULV2VWlG1qBVVi+iwaIJMEMGBwQQFBBFsgjmYcpCdx3ay6/gudp7YSeLRRA6lHAKgRmQN2lZqS5tKbWhTqQ1Xvj5XOw+KiLhRcnoyLy18ie83fU/j8k9pb5UAABQNSURBVI0Zfvlwqpau6p6Tr/gKJt0Ht31ZpLrVX67/kpf/eJkOlTswquMoAkzxmAyhpFrEhzlRxSKvPlvXCWTp3qVsPLyRjYc2svHwRhKPJWJzrOUBQSaI2PBY4iLiiCsVx6XRl9KmUhtiw8/deDW/aiAiIlI0U7dO5YUFL5Cemc7QlkO5vtb1rtezzkiDUU2gfC2484eCPy0zg5EJI/l07ae0j2vPGx3eKFYLE926TbmIuJendwYsbJ/zmnWie/Xu5zyWnJ7MkZQjpNt00jLSSM9MJy0zjagSUVQIq5BvTWnQzoMicnHx5h3IbtW60TS6KU///jTPzX+OuTvm8mSrJ6kQVqHoJw0Mhpb9YMYLsGclxDbO9ykn0k4w5LchzNkxhz71+zA4fnCx3+AlL8VjbF7EjzhRnqmwfZYMKklMeAxxpeKoFlmNWmVqUb9cfSqWqlighBqcX9wjIuIt3twg5rTY8Fje6/oeA1sMZPaO2Vz77bWMXTaWk2kni37SFndBcBgsHJtv0z0n9nDnlDv5fefvPN3qaYa2HHpRJ9SgkWoRjzt/9CIqLJhDJ9MuaJfTCG5RRz5c6dNdTsfp9OIeERFPc+IOJGTtwHh3o7vpUrULby15i/8s/w9f/fUVDzV7iJ41exZ4EOSMsLLQ9BZYOgGuGgbh5S5okpaZxrcbvmXMsjGkZqQypvMY2sW1c9Mr8m+aUy3iQTnNZQ4OMGAgLeN///ZymlNd1LnXrvR5MfGF1fwiUjz4yhqSZXuXMSJhBCv2raBWVC1urXcr3at3JyIkouAn2bsW/tMaujwP7R87c9hay7Rt0/j30n+z7eg2mldozj/b/JOaUTWB4ntNVZ1qER+R0+hFWqYlPCQo350B8xr58FSfFwsnbtWKSPGV210/b68hubTCpXza/VNGdhgJwLCFw+j0ZSee/v1pEvYkUKCB1Ar1odrl8Of7kJHOqfRTzEmcw20/3cbgOYMJDghmdKfRfHT1R+ck1LqmavqHiEflNmf5yKk0lj3XtUjPzW/utSt9XiyculUrIsXT493q5nhnsWO9aNoNn+nV0VtjDN2qdaPrJV1ZtX8VkzZOYsqWKUzeNJmK4RVpVL4RDco1oH7Z+tQvV5+yJcsCWSPRpxenb2rQnYXzhrPw+xtZeiKRtMw0YsJiGNZuGNfVuO6CaSW6pmZRUi3iQa5UwCjqc1V1I39OLBYVkeIrpzUkHetF883inWeSzdOjt2e39yRjDI2jG9M4ujFPXPYE07dNZ3bibNYeWMv0bdPPtAsNCiUtM6vK0znKRlHv2A76NOxD64qtiY+Np0RgiRz70jU1i5JqEQ/KbfSiIBUwivpcV/q8WOiDh4i4W+9mcecky+2Gz/SZ0dvQoFB61uxJz5o9ATiScoT1B9ez9uBa9p3c979NvgKyvlcKr0TLrQmUnTMCrrkWovP++6FrahYl1SIe5EoFjKI+V1U38qcPHiLiab48ehtZIpKWFVvSsmLL3BtViIff34ZF70GPkXmeT9fULKr+ISIXpeK6Ul1EfEO74TNzHL2Niwpl3tBODkRUBN/+Hdb+AAPXQsnSeTYtrtdUbVMuIiIi4qCilkX1KTsXw3udoPsIaNXf6WgcoZJ6IiIiIg7q3SyOV29o7N+lTONaZH0tGgd+OAjrbZpTLSIiIuIB5y9e9Est+8O3A2DzbKjZ0elofJpGqkVERESE75bupN3wmVQf+hPths/M2ryl4fUQVj5rwaLkSUm1iIiIyEUu110RV+6HZn3hr1/g6C6nw/RpSqpFRERELnJ57YpIi7vAZsCS8c4E5yeUVIuIiIhc5PKsq122OtTsBEs+hoz0HNuJkmoRERGRi15uux+eOR5/DxzdCRumeTEq/6KkWkRERMQP5LiQ0E0e71aX0ODAc46dsytinauhVCws/tBtfRY3SqpFREREfFyuCwndlFjnW1c7MBia3wEbpsOhbW7ps7jRjooiIiIiPs4ntj0/nAhvN4H2A6Hzs97p02HaUVFERESkGMlzIaG3RFWB2l1h6XjISPNev35CSbWIiIiIj8t3IaG3xN8Dx5Ng/c/e7dcPKKkWEZ/kyQU5IiL+Jt+FhN5SqwtEVoGEDzzWhb9e/5VUi4jP8fSCHBERf5PvQkJvCQiE5nfC5tlwYJPbT+/P138tVBQRn+MTC3JERCRnx/bAmw2g7YNw1YtuPbWvXf+1UFFE/JpPLMgREZGcRcRm1a1e9pnbFyz68/U/yOkARETOVykqNMeRCq8vyBER8TPfLd3JiKnr2XX4FJWiQnm8W12Xpojker7mt8P6n+CvqVD/WrfF78/Xf41Ui4jP8ZkFOSIifsTd85HzPF+tq7J2WFw63q2vwZ+v/xqpFikG3D0y4XSfp8/j7dckIuLPRkxdz6m0jHOOnUrLYMTU9UW6fuZ7vktvg3mj4OhuKF3RpdhP8+frv5JqET93eiTh9IXv9EgC4LGLkDf67N0szi8uoiIivsLd85HzPV+zvvD7m7B8Ilw+qEh95MRfr/8uTf8wxpQ1xkw3xmzI/l4mhzYdjTHLzvpKNsb0zn7sI2PMlrMeu9SVeEQuRnmNJBSnPkVEJG+F2SCmILWg8z1fuZpwSXtYMh4yM4seeDHh6pzqocAMa21tYEb27+ew1s6y1l5qrb0U6AScBKad1eTx049ba5e5GI/IRceJldL+vDpbRKS4Kuh85ILOvS7Q+ZrfDoe2wLZ5bn0t/sjVpLoX8HH2zx8DvfNpfxMwxVp70sV+RSSbE1vX+sx2uSIickZBN4gp6N3GAp2vfk8oUdrtCxb9katzqmOstbuzf94DxOTT/hbgzfOOvWyM+SfZI93W2pScnmiM6Q/0B6hatWrRIxYpZh7vVvec+c3g+ZXSTvQpIiL5K8h85MLcbcz3fCFh0PhvsGwCdH8dQqMKFW9xku9ItTHmV2PMqhy+ep3dzmZtzZjr9ozGmIpAY2DqWYefBOoBlwFlgSG5Pd9aO85aG2+tjY+Ojs4vbJGLhhNb1/rMdrkiIlJobr/b2Px2SE+GlV+5EJX/y3ek2lrbJbfHjDFJxpiK1trd2Unz3jxOdTPwrbX2zNY7Z41ypxhjPgQGFzBuETmLp1dK51Y+T0m0iIj/cfvdxoqXQkzjrCkgLfu5KUr/4+qc6snAndk/3wl8n0fbW4HPzj6QnYhjjDFkzcde5WI8IuJm7t5MQEREnOX2u43GZI1W714Ou1e4NVZ/YrJmbRTxycaUA74EqgLbgJuttQeNMfHA362192W3qwbMA6pYazPPev5MIBowwLLs5xzPr9/4+HibkJBQ5LhFpODaDZ+Z45axcVGhzBvayYGIRETE55w8CG/Uhfh7oftwp6NxG2PMYmttfEHaurRQ0Vp7AOicw/EE4L6zft8KXPDxx1qrv8giPk7l80RE/JtXdt0NKwt1u8PKL+GqFyEoxL3n9wOuTv8QkWJO5fNERPyXV6fwNb0NTh6AjdPdf24/oKRaRPJU0M0ERETE93h1B9xanSE8GpZNdP+5/YCSahHJk8rniYj4L69O4QsMhib/B39NhRMH3H9+H+fq5i8ichFQ+TwREf9UKSo0x8XmHpvC1/RWWDAaVn0NrQZ4pg8fpZFqERERkWLK61P4YhtBbOOLcgqIkmoRERGRYsqRKXyX9oHdyyBpjef68EEu1al2iupUi4iIiPioE/uzala3vh+6vuR0NC4pTJ1qjVSLiIiIiPuEl4faXWHFl5CR7nQ0XqOkWkRERETc69Lb4HgSbJ7ldCReo6RaRERERNyrdjcILQvLJjgdidcoqRYRERER9woKgcY3wbqf4dQhp6PxCiXVIiIiIuJ+TW+FjBRYM9npSLxCSbWIiIiIuF+lZlCuVtaCxYuAkmoRERERcT9jsrYt3/Y7HE50OhqPU1ItIiIiIp7R+Kas76u+djYOL1BSLSIiIiKeUbYGVG4JK75yOhKPU1ItIiIiIp7T5GbYuxr2rHI6Eo9SUi0iIiIintPweggIgpXFe8GikmoRERER8Zzw8lCzM6z8GjIznY7GY5RUi4iIiIhnNbkZju6EbfOcjsRjlFSLiIiIiGfVvQZCSsGKL5yOxGOUVIuIiIiIZ4WEQb1rs3ZXTEt2OhqPUFItIiIiIp7X5GZIOQIbpjodiUcoqRYRERERz6veAcIrFNtty5VUi4iIiIjnBQZl7bC4YRqcOuR0NG6npFpEREREvKPRTZCRCut+cjoSt1NSLSIiIiLeEdccoi6BVd84HYnbKakWEREREe8wBhrdCJvnwPF9TkfjVkqqRURERMR7Gt0INgPWfu90JG6lpFpEREREvCemIZSvC6smOR2JWympFhERERHvOT0FZNt8OLLT6WjcRkm1iIiIiHhXoxsAC2u+czoSt1FSLSIiIiLeVb42xDYpVlVAlFSLiIiIiPc1uhF2LoaDW5yOxC2UVIuIiIiI9zW8Puv76uKxYFFJtYiIiIh4X5lLoHLLYlMFREm1iIiIiDij0Y2QtAr2rXc6Epe5lFQbY/5mjFltjMk0xsTn0e5qY8x6Y8xGY8zQs45XN8b8kX38C2NMiCvxiIiIiIgfadgbMMVitNrVkepVwA3Ab7k1MMYEAmOA7kAD4FZjTIPsh18D3rLW1gIOAfe6GI+IiIiI+IuIWKjWHlZ9DdY6HY1LXEqqrbVrrbX5jde3BDZaazdba1OBz4FexhgDdAK+zm73MdDblXhERERExM80vB4ObIS9a5yOxCXemFMdBySe9fuO7GPlgMPW2vTzjufIGNPfGJNgjEnYt2+fx4IVERERES+q3xNMAKz2741g8k2qjTG/GmNW5fDVyxsBnmatHWetjbfWxkdHR3uzaxERERHxlFLRcEm7rN0V/XgKSFB+Day1XVzsYydQ5azfK2cfOwBEGWOCskerTx8XERERkYtJg17w82DYtw4q1Hc6miLxxvSPP4Ha2ZU+QoBbgMnWWgvMAm7Kbncn8L0X4hERERERX1K/J2D8egqIqyX1rjfG7ADaAD8ZY6ZmH69kjPkZIHsU+kFgKrAW+NJauzr7FEOAgcaYjWTNsX7flXhERERExA9FxPxvCoifynf6R16std8C3+ZwfBdwzVm//wz8nEO7zWRVBxERERGRi1mDXjDlcdi7DirUczqaQtOOiiIiIiLivAbZU0DW+OdsYCXVIiIiIuK8iFio2kZJtYiIiIiISxr0gr2rYf8GpyMpNCXVIiIiIuIbGvTM+u6HVUCUVIuIiIiIbyhdCaq09sspIEqqRURERMR3NOgFSSvhwCanIykUJdUiIiIi4jsa9Mr6vvqCqs0+TUm1iIiIiPiOyDio3NLvNoJxafMXERERERG36z4cQss4HUWhKKkWEREREd8S18LpCApN0z9ERERERFykpFpERERExEVKqkVEREREXKSkWkRERETERUqqRURERERcpKRaRERERMRFSqpFRERERFykpFpERERExEVKqkVEREREXKSkWkRERETERUqqRURERERcpKRaRERERMRFSqpFRERERFykpFpERERExEVKqkVEREREXGSstU7HUGjGmH3ANofDKA/sdzgG8Sy9x8Wb3t/iT+9x8ab3t/jzhff4EmttdEEa+mVS7QuMMQnW2nin4xDP0XtcvOn9Lf70Hhdven+LP397jzX9Q0RERETERUqqRURERERcpKS66MY5HYB4nN7j4k3vb/Gn97h40/tb/PnVe6w51SIiIiIiLtJItYiIiIiIi5RUi4iIiIi4SEl1ERhjrjbGrDfGbDTGDHU6HnEfY0wVY8wsY8waY8xqY8wjTscknmGMCTTGLDXG/Oh0LOJexpgoY8zXxph1xpi1xpg2Tsck7mWMeSz7Gr3KGPOZMaak0zGJa4wxHxhj9hpjVp11rKwxZroxZkP29zJOxpgfJdWFZIwJBMYA3YEGwK3GmAbORiVulA4MstY2AFoD/9D7W2w9Aqx1OgjxiLeBX6y19YCm6H0uVowxccDDQLy1thEQCNzibFTiBh8BV593bCgww1pbG5iR/bvPUlJdeC2BjdbazdbaVOBzoJfDMYmbWGt3W2uXZP98jKw/xnHORiXuZoypDPQA/ut0LOJexphI4ArgfQBrbaq19rCzUYkHBAGhxpggIAzY5XA84iJr7W/AwfMO9wI+zv75Y6C3V4MqJCXVhRcHJJ71+w6UdBVLxphqQDPgD2cjEQ8YBTwBZDodiLhddWAf8GH29J7/GmPCnQ5K3MdauxMYCWwHdgNHrLXTnI1KPCTGWrs7++c9QIyTweRHSbVIDowxpYBvgEettUedjkfcxxhzLbDXWrvY6VjEI4KA5sBYa20z4AQ+fstYCid7Xm0vsj5AVQLCjTF9nY1KPM1m1YD26TrQSqoLbydQ5azfK2cfk2LCGBNMVkI9wVo7yel4xO3aAT2NMVvJmr7VyRjzqbMhiRvtAHZYa0/fYfqarCRbio8uwBZr7T5rbRowCWjrcEziGUnGmIoA2d/3OhxPnpRUF96fQG1jTHVjTAhZiyMmOxyTuIkxxpA1F3OttfZNp+MR97PWPmmtrWytrUbWv9+Z1lqNchUT1to9QKIxpm72oc7AGgdDEvfbDrQ2xoRlX7M7o8WoxdVk4M7sn+8EvncwlnwFOR2Av7HWphtjHgSmkrXi+ANr7WqHwxL3aQfcDqw0xizLPvaUtfZnB2MSkcJ5CJiQPfCxGbjb4XjEjay1fxhjvgaWkFWxaSl+tp21XMgY8xlwJVDeGLMDeA4YDnxpjLkX2Abc7FyE+dM25SIiIiIiLtL0DxERERERFympFhERERFxkZJqEREREREXKakWEREREXGRkmoRERERERcpqRYRERERcZGSahERERERF/0/DDYxxnSYzCIAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Try fitting a simple linear function with a polynomial order\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "from numpy import vander\n",
+    "\n",
+    "# We need to fit a polynomial function to our data points using the vander function\n",
+    "# If we just used linear regression we obtain a poor fit because our data was generated  \n",
+    "# using a nonlinear function. To get linear estimates of our coefficients from non-linear inputs\n",
+    "# we increase the number of dimensions of our data. I'm taking a guess our data is polynomial of\n",
+    "# degree 3, so I'm using 4 (3 + 1) as the parameter in vander to return that many columns\n",
+    "# If I wanted to test polynomial of degree 10, I would put 11 as the parameter.\n",
+    "\n",
+    "lm_model_3 = LinearRegression()\n",
+    "lm_model_3.fit(vander(x, 4), y_scatter)\n",
+    "degree_3 = lm_model_3.coef_.size - 1\n",
+    "y_pred_3 = lm_model_3.predict(np.vander(x, degree_3 + 1))\n",
+    "\n",
+    "lm_model_8 = LinearRegression()\n",
+    "lm_model_8.fit(vander(x, 9), y_scatter)\n",
+    "degree_8 = lm_model_8.coef_.size - 1\n",
+    "y_pred_8 = lm_model_8.predict(np.vander(x, degree_8 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_3)\n",
+    "plt.plot(x, y_pred_8)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 3', 'Pred. Deg. 8', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 239,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10f011390>"
+      ]
+     },
+     "execution_count": 239,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "lm_model_30 = LinearRegression()\n",
+    "lm_model_30.fit(vander(x, 31), y_scatter)\n",
+    "degree_30 = lm_model_30.coef_.size - 1\n",
+    "y_pred_30 = lm_model_30.predict(np.vander(x, degree_30 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_30)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 30', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 210,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Looks like we could be overfitting our data if we increase the polynomial degree too much\n",
+    "# How do we find the right polynomial number?\n",
+    "\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn import metrics\n",
+    "import pandas as pd\n",
+    "\n",
+    "# Get the RMSE for each predicted values\n",
+    "def get_rmse(y, y_pred):\n",
+    "    return np.sqrt(metrics.mean_squared_error(y, y_pred))\n",
+    "\n",
+    "# Create dataframe to evaluate\n",
+    "rmse_df = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 211,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 211,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtoAAAG5CAYAAACwZpNaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl8VNX9//HXmclGICEkAZJh3yFAJkJEEVBRVFCDtu6KtdYWrVVbW+vSWttqfxVbW7VqVWrt17rh3hIrouCKqCyagGxCQNnCFhK2kJDl/P64kxgxkG0mdyZ5Px+PeWTmzr3nfgKC7xzOnI+x1iIiIiIiIsHlcbsAEREREZG2SEFbRERERCQEFLRFREREREJAQVtEREREJAQUtEVEREREQkBBW0REREQkBBS0RURagTHGa4zZb4zp7XYtIiLSOhS0RUTqEQjFNY9qY8zBOq8va+p41toqa20na+3GJtYRb4zZa4w5sZ73HjTGzGpqLYFrvQ18jxc1Z9zA2EuMMRcf5f0Rxhh72P33G2PObu49RUTCUZTbBYiIhCNrbaea58aYL4EfWmvnHel8Y0yUtbYyBHWUGmNeBL4HvF/nftHAxYHjzRm3Cqj7PW4Gpllr321RwY1XXvfX+EiMMQbwBOo94rFGjBOS3x8RkaPRjLaISDMYY/5gjHneGPOcMWYfMM0YM9YY87ExpsQYU2iM+VsgEGOMiQrM4vYNvH468P4cY8w+Y8xHxph+R7jdk8D5xpi4OsemAJXAm4HxfmWM2RqY/V5tjDk5CN9jlDHmd8aYDcaYXcaYp4wxiYH3OhljXjDG7DbGFAe+787GmPuBY4D/C8xS39OM+75kjLnfGDMPOAAce4RjKcaYWYHa1htjflFnjOuMMW8ZYx41xhQDN7X010NEpKkUtEVEmu87wLNAZ+B5nOD7UyAVGAdMBq4+yvWXAr8BkoGNwF1HOO8DYDdwbp1jlwPPWGurjDHDA/cZZa1NxAnhTVqicgS3ACcDY4FegWN/CXy9GrCAD+gK3AAcstb+DPgM+H5gqcwtzbz3NOA2nFn3z45w7B+BGvrg/FrfcNiSl4nAJzi/Hw80sw4RkWZT0BYRab4F1tpca221tfagtXaxtfYTa22ltXY9MBM46SjXv2StXWKtrQCeAbLqO8laa4F/E1gmYoxJAnJwZrrBCfhxwPDAEokNgfu31DXALdbabdbag8CdOMtVACpwAnb/wPe7KHBOY8UGZv7rPup+UPSFwK9ntbW2/PBjgBfnB49brLUHrLVfAH/D+QGkxhfW2n8F1sc3pTYRkaDQGm0RkebbVPeFMWYozozvaCAe5+/YT45y/bY6z0ups2a6Hv8Gfm2M6Q6cA6yy1i4HsNauCSybuBMYZoyZC9xord125OGOzhjjBXoAbxpjbJ23ogNBfybQDXjVGBOPE/rvCITgxii31iYd5f1NDRxLx5nNrnvsq0DNRxtDRKTVaEZbRKT57GGvHwM+BwYGlnDcAZig3MiZof4IuAxn1vbJw95/2lo7DuiHM9t7dwvvVwUUAidaa5PqPOKstSXW2jJr7e3W2iE4SzQuBS6oubwl9z7KGHWPFeL8P6xXnWO9gS0NjCEi0moUtEVEgicB2AMcMMYM4+jrs5vjSZw14MfhrA0HwBgzzBgz0RgTCxwMPBo7s3w0jwL3GGN6BO7TvWYLPmPMaYH7eoC9OMtXau65HegfhPsfkbW2FPgvcLcxpqMxZiBwPfB0KO8rItIUCtoiIsHzC+AKYB/O7PbzQR7/RZwP9s211u6oczwW+BOwC2c5Shfg1wDGmCuMMfnNvN/dOFsKvmeM2QsswNlRBJyZ5Fyc7zUfJ/S+HHjvL8BVgXXXR5pZj61nH+3pTazvR0A0zgc/5wGPEPxfcxGRZjPOZ2xERERERCSYNKMtIiIiIhICCtoiIiIiIiGgoC0iIiIiEgIK2iIiIiIiIdBmGtakpqbavn37ul2GiIiIiLRxS5cu3WWt7drQeW0maPft25clS5a4XYaIiIiItHHGmK8ac56WjoiIiIiIhICCtoiIiIhICChoi4iIiIiEQJtZoy0iIiISiSoqKti8eTNlZWVulyKHiYuLo2fPnkRHRzfregVtERERERdt3ryZhIQE+vbtizHG7XIkwFpLUVERmzdvpl+/fs0aQ0tHRERERFxUVlZGSkqKQnaYMcaQkpLSon9pUNAWERERcZlCdnhq6e+LgraIiIiISAgoaIuIiIi0Y0VFRWRlZZGVlUVaWho9evSofX3o0KFGjXHllVeyZs2ao57z8MMP88wzzwSjZMaPH8+QIUPIzMxk6NChXH/99ezZs+eo11RXVzNjxoyg3L+xjLW2VW8YKtnZ2VadIUVERCTSrFq1imHDhrldBgC/+93v6NSpEzfddNM3jltrsdbi8YTHHO348eN56KGHan8YuPnmm1m+fDnz588/4jWVlZWkpqZSUlLSpHvV9/tjjFlqrc1u6Nrw+NUSERERkbCybt06MjIyuOyyyxg+fDiFhYVMnz6d7Oxshg8fzp133ll77vjx48nLy6OyspKkpCRuvfVW/H4/Y8eOZceOHQDcfvvt3H///bXn33rrrYwZM4YhQ4awcOFCAA4cOMB5551HRkYG559/PtnZ2eTl5R21zpiYGO69917Wrl3LihUrAMjJyWH06NEMHz6cxx9/HIBbb72Vffv2kZWVxfe+970jnhdM2t5PREREJEz8PncFK7fuDeqYGb5EfpszvFnXrl69mn//+99kZzuTtzNmzCA5OZnKykomTpzI+eefT0ZGxjeu2bNnDyeddBIzZszg5z//OU888QS33nrrt8a21rJo0SJmz57NnXfeyRtvvMGDDz5IWloaL7/8Mvn5+YwaNapRdUZFRZGZmcnq1asZPnw4Tz75JMnJyZSWlpKdnc15553HjBkzePzxx78R3Os7r0uXLs36taqPZrRFREREpF4DBgyoDdkAzz33HKNGjWLUqFGsWrWKlStXfuuaDh06MGXKFABGjx7Nl19+We/Y3/3ud791zoIFC7j44osB8Pv9DB/e+B8Q6i6Hvu+++2pn1Ddv3kxBQUG91zT2vObSjLaIiIhImGjuzHOodOzYsfb52rVreeCBB1i0aBFJSUlMmzat3j2mY2Jiap97vV4qKyvrHTs2NrbBcxqrsrKSzz//nGHDhjFv3jzef/99Pv74Yzp06MD48ePrrbOx57WEgnYLrduxn6pqi9cDHmPwGIPXY/B4DF5j8ASOO8+d97zGYAy1zz0e7Z0pIiIi4W3v3r0kJCSQmJhIYWEhc+fOZfLkyUG9x7hx43jhhReYMGECy5cvr3fG/HCHDh3itttuY+DAgWRkZLBq1SqSk5Pp0KEDK1asYPHixYCzvAScUB4VFcWePXvqPS+YFLRb6Mr/W8Sm3QdbPM6RArjXY/AYvg7wps6xwHk1xz0evr7O1BP26xz3BO5T97gxBq+Hw+5T8/yb9zOm5vnXxz2196T2h41v/pDh1DGqdxd6JccH4VdfREREWsuoUaPIyMhg6NCh9OnTh3HjxgX9Htdffz3f+973yMjIqH107ty53nMvuugiYmNjKS8v5/TTT+eVV14B4KyzzmLmzJlkZGQwZMgQjjvuuNprrrrqKjIzM8nOzmbmzJlHPC9YtL1fC727ZgcHyquoCmx7U1XtPKqtpaoaqm3N83qOV1uq6n61OM8bcbza8vU51V/fo9ryrftVW75xTt3jNfeoqq5n3KMcb8l/NiN7dCb3+vHB+00QERGJYOG0vZ/bKisrqaysJC4ujrVr13L66aezdu3a2tloN7Rkez/NaLfQyUO6uV2CK2x9Ab7uDwGHBXsbCOqvfraFv81fy/qd++nftZPb34aIiIiEkf3793PqqadSWVmJtZbHHnvM1ZDdUpFbubjKWTriLBFpikvH9ObBt9eSm1/ITycNClF1IiIiEomSkpJYunSp22UEjbb3k1aV1jmOY/smMzt/C21l2ZKIiIhIfRS0pdVN9fso2HmA1dv2uV2KiIiISMgoaEurmzIiDa/HkJu/1e1SREREREJGQVtaXUqnWMYNTCV32VYtHxEREZE2S0FbXJGTmc6m3QfJ21TidikiIiLtWlFREVlZWWRlZZGWlkaPHj1qXx86dKjR4zzxxBNs27at3vemTZtGv3798Pv9DB48mCuuuIKtWxv+l+2//vWvQe/W2JoUtMUVZ4xII8brITe/0O1SRERE2rWUlBTy8vLIy8vjmmuu4cYbb6x9XbedekOOFrQB7rvvPvLz81m9ejUjR47klFNOoaKi4qhjKmiLNENiXDQnD+nKa8u2UlWt5SMiIiLh6Mknn2TMmDFkZWVx7bXXUl1dTWVlJZdffjkjR45kxIgR/O1vf+P5558nLy+Piy66qMGZcI/Hw0033URycjJvvvkmANOnTyc7O5vhw4dz5513Ak4w37FjBxMmTGDSpElHPC+caR9tcU2O38ebK7ezaMNuxg5IcbscERER9825FbYtD+6YaSNhyowmX/b555/z6quvsnDhQqKiopg+fTqzZs1iwIAB7Nq1i+XLnTpLSkpISkriwQcf5KGHHiIrK6tR448aNYrVq1dz1llnMWPGDJKTk6msrGTixImcf/753HjjjfzlL3/hgw8+ICkpCaDe8zIyMpr8vbUWzWiLa04d1o34GC+5y7T7iIiISLiZN28eixcvJjs7m6ysLN577z0KCgoYOHAga9as4YYbbmDu3Ll07ty5WePX3RDhueeeY9SoUYwaNYpVq1axcuXKeq9p7HnhQjPa4pr4mCgmDevOnOWF/H7qcKK9+rlPRETauWbMPIeKtZYf/OAH3HXXXd96b9myZcyZM4eHH36Yl19+mZkzZzZ5/Ly8PM466yzWrl3LAw88wKJFi0hKSmLatGn1rstu7HnhRMlGXJXj91FcWsGCdbvcLkVERETqmDRpEi+88AK7djn/jy4qKmLjxo3s3LkTay0XXHABd955J59++ikACQkJ7NvXcDM6ay333XcfRUVFnHbaaezdu5eEhAQSExMpLCxk7ty5tefWHfNo54UrzWiLq04cnEpiXBS5+VuZOKSb2+WIiIhIwMiRI/ntb3/LpEmTqK6uJjo6mkcffRSv18tVV12FtRZjDPfccw8AV155JT/84Q/p0KEDixYt+taOJTfeeCO//e1vOXjwIGPHjuXtt98mOjqaUaNGkZGRwdChQ+nTpw/jxo2rvWb69OlMmjSJXr168dZbbx3xvHBl2krDkOzsbLtkyRK3y5BmuPmlfF5fvo0lt08iLtrrdjkiIiKtatWqVQwbNsztMuQI6vv9McYstdZmN3Stlo6I63L8PvaXV/Lump1ulyIiIiISNAra4rqx/VNI6RhDbr52HxEREZG2Q0FbXBfl9XDmyHTmr97O/vJKt8sRERFpdW1lKW9b09LfFwVtCQtTs3yUVVQzf9V2t0sRERFpVXFxcRQVFSlshxlrLUVFRcTFxTV7DO06ImFhdO8upHeOY3beVs7J6uF2OSIiIq2mZ8+ebN68mZ079VmlcBMXF0fPnj2bfb2CtoQFj8dwdmY6/7fwS0pKD5EUH9PwRSIiIm1AdHQ0/fr1c7sMCQEtHZGwMdXfg4oqy9wV29wuRURERKTFFLQlbIzokUjflHhy8wvdLkVERESkxRS0JWwYY8jx+1hYsIsd+8rcLkdERESkRRS0Jazk+H1UW5izXMtHREREJLIpaEtYGdw9gaFpCWpeIyIiIhFPQVvCTo7fx5KvitlSctDtUkRERESaTUFbws7ZmekAvKZZbREREYlgCtoSdvqkdMTfK4ncZQraIiIiErkUtCUs5WSm8/mWvazfud/tUkRERESaRUFbwtLZmT6MQXtqi4iISMRS0JawlNY5jmP7JjM7fwvWWrfLEREREWkyBW0JW1P9Pgp2HmD1tn1ulyIiIiLSZAraEramjEjD6zHaU1tEREQikoK2hK2UTrGMG5hK7rKtWj4iIiIiESekQdsYM9kYs8YYs84Yc+tRzjvPGGONMdl1jt0WuG6NMeaMUNYp4Wuq38em3QfJ21TidikiIiIiTRKyoG2M8QIPA1OADOASY0xGPeclAD8FPqlzLAO4GBgOTAb+HhhP2pnTh3cnxuvR7iMiIiIScUI5oz0GWGetXW+tPQTMAs6p57y7gHuAsjrHzgFmWWvLrbUbgHWB8aSdSYyL5uQhXXlt2VaqqrV8RERERCJHKIN2D2BTndebA8dqGWNGAb2stf9r6rWB66cbY5YYY5bs3LkzOFVL2Mnx+9ixr5xFG3a7XYqIiIhIo7n2YUhjjAf4K/CL5o5hrZ1prc221mZ37do1eMVJWDl1WDfiY7xqyS4iIiIRJZRBewvQq87rnoFjNRKAEcC7xpgvgeOB2YEPRDZ0rbQj8TFRTBrWnTnLC6moqna7HBEREZFGCWXQXgwMMsb0M8bE4Hy4cXbNm9baPdbaVGttX2ttX+BjYKq1dkngvIuNMbHGmH7AIGBRCGuVMJfj91FcWsGCdbvcLkVERESkUUIWtK21lcB1wFxgFfCCtXaFMeZOY8zUBq5dAbwArATeAH5ira0KVa0S/k4cnEpiXJSa14iIiEjEiArl4Nba14HXDzt2xxHOPfmw1/8P+H8hK04iSmyUl8kj0nh9+TbKKqqIi9ZujyIiIhLe1BlSIkaO38f+8kreXaMdZkRERCT8KWhLxBjbP4XUTjFaPiIiIiIRQUFbIkaU18OZI9OZv3o7+8sr3S5HRERE5KgUtCWi5Ph9lFVUM3/VdrdLERERETkqBW2JKKN7dyG9cxyz87R8RERERMKbgrZEFI/HcHZmOu+v3UlJ6SG3yxERERE5IgVtiThT/T2oqLLMXbHN7VJEREREjkhBWyLOiB6J9E2JJze/0O1SRERERI5IQVsijjGGHL+PhQW72LGvzO1yREREROqloC0RaarfR7WFOcu1fERERETCk4K2RKRB3RMYmpag5jUiIiISthS0JWLl+H0s+aqYLSUH3S5FRERE5FsUtCVinZ2ZDsBrmtUWERGRMKSgLRGrT0pH/L2SyF2moC0iIiLhR0FbIlpOZjqfb9nL+p373S5FRERE5BsUtCWinZ3pwxi0p7aIiIiEHQVtiWhpneMY0zeZ2flbsNa6XY6IiIhILQVtiXg5fh8FOw+wets+t0sRERERqaWgLRFvyog0vB6jPbVFREQkrChoS8RL6RTLuIGp5C7bquUjIiIiEjYUtKVNmOr3sWn3QfI2lbhdioiIiAigoC1txOnDuxPj9Wj3EREREQkbCtrSJiTGRXPykK68tmwrVdVaPiIiIiLuU9CWNmNqlo8d+8pZtGG326WIiIiIKGhL23HK0G7Ex3jVkl1ERETCgoK2tBnxMVFMGtadOcsLqaiqdrscERERaecUtKVNyfH7KC6tYMG6XW6XIiIiIu2cgra0KScOTiUxLkrNa0RERMR1CtrSpsRGeZk8Io03V2ynrKLK7XJERESkHVPQljYnx+9jf3kl767Z4XYpIiIi0o4paEubM7Z/CqmdYtS8RkRERFyloC1tTpTXw5kj05m/ejv7yyvdLkdERETaqSi3C4h4/zgVSosgPhk6JEN8Sp3nXQJfD3svuoPbVbd5OX4f//7oK+av2s45WT3cLkdERETaIQXtlup/EhR/BQd3w4GdsHON8/zQ/iNfE9WhnjBeN6DX/drFeS+uMxjTet9XhBvduwvpneOYnbdVQVtERERcoaDdUqfeUf/xynI4WAylu53gXbrbmfmueV73ve0rnPfKSsAeodGK8QZCdz1hvN6AHvjqjQ7d9x7GPB5Djt/Hvz7cQEnpIZLiY9wuSURERNoZBe1QiYqFhDTn0VjV1U7YPmpAD3wt2Qhb85z3qsqPPGZs4hECekqdGfPD3ovp2CZmz3Myfcx8fz1zV2zjomN7u12OiIiItDMK2uHE4/l6hjplQOOusRYqSr8ZwmsDej3HdhdAaTGU7znymN7Yepav1LcGvc7XuCSn/jAyokcifVPiyc0vVNAWERGRVqegHemMcWagYzpCUq/GX1dV8e2Z82/NoBc7X2vWnZfuBnukJjDmyEtbataZ1xzzZUFsQlC+/aMxxlk+8vA769ixr4xuCXEhv6eIiIhIDQXt9sobDZ26OY/GshbK9zpBvCaEH2mJy94tsO1z53VF6TfH6dIPfjgPOqYG93uqx1S/jwffXsec5du44oS+Ib+fiIiISA0FbWk8Y5zdT+I6Q3ITrqs4+HUg3/UF/OdaeO5iuCI35FsdDuqewNC0BHLztypoi4iISKsKr0W10jZFd4DOPSBtJIw4D777D9i8BF7+IVQfaSlK8OT4fSz5qpgtJQdDfi8RERGRGgra0voypsIZf4TVr8Gbt4f8djmZPgBey98a8nuJiIiI1FDQFneMvRaOuwY+/jt8/EhIb9U7JR5/ryRylyloi4iISOtR0Bb3nPFHGHo2vHEbrMoN6a1yMtP5fMte1u88SsdOERERkSBS0Bb3eLzOeu0eo5312psWh+xWZ2f6MAZy8wtDdg8RERGRuhS0xV0x8XDJLKeD5nMXwe71IblNWuc4xvRNZnb+Fqy1IbmHiIiISF0K2uK+Tl3hspfBVsPT58OBopDcJsfvo2DnAVZv2xeS8UVERETqUtCW8JA60JnZ3rMZZl3i7L0dZFNGpOH1GGZr9xERERFpBQraEj56Hw/ffQw2fQKvXgPV1UEdPqVTLOMHppKbv1XLR0RERCTkFLQlvAz/Dpx2F6z8D8y7I+jD5/h9bC4+SN6mkqCPLSIiIlKXgraEnxOuh2N/BAsfhEX/COrQpw/vTozXo91HREREJOQUtCX8GANT7oHBU2DOzbD69aANnRgXzclDuvLasq1UVWv5iIiIiISOgraEJ48Xzv8npPvhpR/AlqVBG3pqlo8d+8pZtGF30MYUEREROZyCtoSvmI5w6QvO9n/PXgTFXwZl2FOGdiM+xquW7CIiIhJSCtoS3jp1c/bYrqpw9tgubfksdHxMFJOGdWfO8kIqqoK7s4mIiIhIDQVtCX9dB8PFz0LJVzDrMqgoa/GQU/0+iksrWLBuVxAKFBEREfk2BW2JDH3HwbmPwMaF8N9rW7zH9oTBqSTGRZGr5jUiIiISIlFuFyDSaCPPhz2bYN7voHMvOO33zR4qNsrL5BFpvL58G2UVVcRFe4NXp4iIiAia0ZZIM+5nkP0D+PB+WPzPFg2V4/exv7ySd9fsCFJxIiIiIl8LadA2xkw2xqwxxqwzxtxaz/vXGGOWG2PyjDELjDEZgeN9jTEHA8fzjDGPhrJOiSDGwJQ/w6Az4PWbYM0bzR5qbP8UUjvFqHmNiIiIhETIgrYxxgs8DEwBMoBLaoJ0Hc9aa0daa7OAPwF/rfNegbU2K/C4JlR1SgTyRsH5T0DaSHjpStj6WbOGifJ6OHNkOvNXb2d/eWWQixQREZH2LpQz2mOAddba9dbaQ8As4Jy6J1hr99Z52RFQqz5pnNhOzh7b8SmBPba/atYwOX4fZRXVzF+1PcgFioiISHsXyqDdA9hU5/XmwLFvMMb8xBhTgDOjfUOdt/oZYz4zxrxnjJlQ3w2MMdONMUuMMUt27twZzNolEiSkwWUvOdv9PXMBHCxu8hCje3fB1zmO2XnafURERESCy/UPQ1prH7bWDgBuAW4PHC4EeltrjwF+DjxrjEms59qZ1tpsa212165dW69oCR/dhsLFT8Pu9fD85VBZ3qTLPR7D2X4f76/dSUnpoRAVKSIiIu1RKIP2FqBXndc9A8eOZBZwLoC1ttxaWxR4vhQoAAaHqE6JdP1OhHP/Dl9+AP+9DmzTViDlZPqoqLLMXbEtRAWKiIhIexTKoL0YGGSM6WeMiQEuBmbXPcEYM6jOy7OAtYHjXQMfpsQY0x8YBKwPYa0S6TIvhFN+A8tfgLf/0KRLR/RIpG9KPLPVvEZERESCKGQNa6y1lcaY64C5gBd4wlq7whhzJ7DEWjsbuM4YMwmoAIqBKwKXnwjcaYypAKqBa6y1u0NVq7QRE37htGn/4F5I6gWjv9+oy4wxTPX7eOiddezYV0a3hLjQ1ikiIiLtgrFN/Gf2cJWdnW2XLFnidhnitqoKeO5iKHgHLn0eBp3WqMvWbt/Hafe9z++nDueKE/qGtkYRERGJaMaYpdba7IbOc/3DkCJB5Y2GC/4PumfAi9+HwvxGXTaoewJD0xLI1fIRERERCRIFbWl7YhPg0hchLgmeuRBKNjV8Dc6e2ku+KmZLycEQFygiIiLtgYK2tE2J6XDZi1BRGthju6TBS3IyfQC8plltERERCQIFbWm7umfARU9B0Vp44XKoPPo+2b1T4vH3SiJ3mYK2iIiItJyCtrRt/U+GqQ/Bhvdh9vUN7rGdk5nO51v2sn7n/lYpT0RERNouBW1p+7IugZN/Bctmwbt3H/XUszN9GAO5+YWtVJyIiIi0VQra0j6cdDNkTYP37oFPnzriaWmd4xjTN5nZ+VtoK1tfioiIiDsUtKV9MAZy7of+E+G1n8G6+Uc8Ncfvo2DnAVZv29eKBYqIiEhbo6At7Yc3Gi78N3QdCi9cAduW13valBFpeD1GLdlFRESkRRS0pX2JS4RLX3D22n7mQtiz5VunpHSKZfzAVHLzt2r5iIiIiDSbgra0P517wGUvQPk+Z4/tsj3fOiXH72Nz8UHyNjW8/7aIiIhIfRS0pX1KGwkXPgm71jjLSKoqvvH26cO7ExPl0e4jIiIi0mwK2tJ+DTwVch6A9e9A7s++scd2Ylw0E4d05bVlW6mq1vIRERERaToFbWnfjpkGJ90CeU/De3/6xls5fh879pWzaMNul4oTERGRSKagLXLybeC/BN79I+Q9W3v4lKHdiI/xqiW7iIiINIuCtogxkPM36Hei06Z9/bsAxMdEMWlYd+YsL6SiqtrdGkVERCTiKGiLAETFwIVPQcogeP5y2L4SgKl+H8WlFSxYt8vlAkVERCTSKGiL1OiQBJe9CNHx8Mz5sHcrEwankhgXRa6a14iIiEgTKWiL1JXUy9lju2wPPHMhsVWlTB6RxpsrtlNWUeV2dSIiIhJBFLRFDpfuhwuehB0r4YUrOGd4RM6zAAAgAElEQVRkN/aXV/Lumh1uVyYiIiIRREFbpD6DJsHZf4WC+Yxd/UdSO0areY2IiIg0SZTbBYiErdHfh5KNeD74CzPS47hu9SnsL6+kU6z+2IiIiEjDNKMtcjSn/AZGXsCkwpmcUfU+81Zud7siERERiRAK2iJHYwyc8zC2z3jujZnJFx+/7nZFIiIiEiEUtEUaEhWLufhp9sT15Jptd7B34zK3KxIREZEIoKAt0hgdurDrnGcpIwbvcxfCvm1uVyQiIiJhTkFbpJGGDM3gN/G/wVu2G569EMr3u12SiIiIhDEFbZFGMsYw5JgJ/OTQDdhty+GlK6Gq0u2yREREJEwpaIs0QY7fx/yqY/h46G2w9k2Y80uw1u2yREREJAwpaIs0waDuCQxNS+Avu8fDuJ/Bkifgw/vdLktERETCkIK2SBPl+H0s+aqYLdk3w4jzYN7vYPlLbpclIiIiYUZBW6SJcjJ9ALy2bBuc83fofQL858fw5YcuVyYiIiLhREFbpIl6p8Tj75VE7rKtEB0HFz8DSX1g1qWw8wu3yxMREZEwoaAt0gxT/T4+37KX9Tv3Q3wyTHsJvNHwzHmwf4fb5YmIiEgYUNAWaYazRqZjDOTmFzoHuvSFS5+H/TudPbYPHXC1PhEREXGfgrZIM6R1jmNM32Rm52/B1mzv12M0nP8EFObDS1dBdZW7RYqIiIirFLRFminH76Ng5wFWb9v39cGhZ8KUP8EXc2DOLdpjW0REpB1T0BZppjNHpuP1GGbnb/3mG2N+BGOvg8X/gI8ecqc4ERERcZ2CtkgzJXeMYfzAVHLzt369fKTGaXdBxjnw5u2w4lV3ChQRERFXKWiLtECO38fm4oPkbSr55hseD3xnJvQ6Hl65GjZ+7E6BIiIi4hoFbZEWOH14d2KiPN9ePgLOHtuXPAede8JzF8Ouda1foIiIiLhGQVukBRLjopk4pCv/W1ZIVXU9H3ys2WPbeAN7bO9s/SJFRETEFQraIi2U4/exY185izbsrv+E5P5wySzYt82Z2T5U2roFioiIiCsUtEVa6NSh3YmP8Tot2Y+k17Fw3uOwZSm88iPtsS0iItIOKGiLtFCHGC+nZXRnzvJCKqqqj3zisByYfDesfg3m/rr1ChQRERFXKGiLBEFOpo/i0goWrNt19BOP/zEcfy188gh89PfWKU5ERERcoaAtEgQTBqeSGBdFbn27jxzu9D84s9tzfwUr/xv64kRERMQVCtoiQRAb5WXyiDTeXLGdsooG1l97vPDdf0DPbHhlOmxa1DpFioiISKs6atA2xpxS53m/w977bqiKEolEU/092F9eybtrdjR8cnQHZyeShHRnJ5KigtAXKCIiIq2qoRnte+s8f/mw924Pci0iEe34/smkdoohN7+wcRd0TIVpL4O18Mz5cKAotAWKiIhIq2ooaJsjPK/vtUi7FuX1cObIdOav3s7+8srGXZQywJnZ3rPFmdmuOBjaIkVERKTVNBS07RGe1/dapN2b6vdRVlHNvJXbG39R7+PguzNh82JnzXb1UbYIFBERkYjRUNDub4yZbYzJrfO85nW/Bq4VaXdG9e6Cr3Nc43YfqWv4uc5uJKtmw1u/CU1xIiIi0qqiGnj/nDrP7z3svcNfi7R7Ho/hbL+Pf324gZLSQyTFxzT+4rE/gZKN8NFDkNQbjrs6dIWKiIhIyB11Rtta+17dB7AQ2AusCrwWkcPkZPqoqLLMXbGtaRca43SOHHIWzLkFVv8vNAWKiIhIq2hoe79HjTHDA887A/nAv4HPjDGXtEJ9IhFnRI9E+qV2ZHZTl4+As8f2eY9Dj1Hw0lWweWnwCxQREZFW0dAa7QnW2hWB51cCX1hrRwKjgZtDWplIhDLGkJOZzkcFRezYV9b0AWLi4ZLnoVM3ePZC2L0h+EWKiIhIyDUUtA/VeX4a8B8Aa20T/01cpH3J8fuotjBneTP/qHTqGthjuwoeOQGeuxQ+fQr2N6IZjoiIiISFhj4MWWKMORvYAowDrgIwxkQBHUJcm0jEGtQ9gaFpCczO38oVJ/Rt3iCpg+D7r8OSf8KaObDmf4BxWrcPngxDzoRuw5y13SIiIhJ2GprRvhq4DvgX8LM6M9mnAg1+UssYM9kYs8YYs84Yc2s9719jjFlujMkzxiwwxmTUee+2wHVrjDFnNP5bEgkPOX4fS78qZnNxafMH6Z4BZ/0FblwBV38AJ98G1ZXw9l3wyFh4INP54GTBO1B5qOHxREREpNUYa0PTd8YY4wW+wFlyshlYDFxirV1Z55xEa+3ewPOpwLXW2smBwP0cMAbwAfOAwdbaqiPdLzs72y5ZsiQk34tIc2wsKuXEP7/DbVOGcvVJA4I7+N5C+OIN57H+Xagsg9hEGHiqM9M9cBLEJwf3niIiIgKAMWaptTa7ofOOunTEGPO3o71vrb3hKG+PAdZZa9cHxpqFsy93bdCuCdkBHfm62+Q5wCxrbTmwwRizLjDeR0erRySc9E6Jx98ridxlW4MftBPTIftK53Go1Anba16HL+bCilfBeKH3WBgSWGKSEuT7i4iISIMaWqN9DfA58AKwFWjKYtAewKY6rzcDxx1+kjHmJ8DPgRjglDrXfnzYtT3quXY6MB2gd+/eTShNpHVM9fu467WVrN+5n/5dO4XmJjHxMPRM51FdDVs/ddZ0f/EGvHm780gZBEOmOI+eY8Db0B99ERERaamG1minAzOBM4DLgWjgv9baJ621TwajAGvtw9baAcAtwO1NvHamtTbbWpvdtWvXYJQjElRnjUzHGMjNL2ydG3o8zoclT/0N/PhD+OkymPJn6NwTPn4E/jUF7h0Ir1ztzHyX7W14TBEREWmWo05rWWuLgEeBR40xPYGLgZXGmFustU81MPYWoFed1z0Dx45kFvBIM68VCUtpneMY0zeZ2flbuOHUgZjW3iGkSx84brrzKNsLBfNhzRuwdi4smwWeaOg73lleMmSy0/pdREREgqKhGW0AjDGjgJ8C04A5QGPa1S0GBhlj+hljYnBC+uzDxh1U5+VZwNrA89nAxcaYWGNMP2AQsKgxtYqEm6lZPgp2HmD1tn3uFhKXCMO/A999DG5aB1fOgeOvgT2bYc4v4f6R8Mg4mH+X05GyutrdekVERCJcQx+GvBMnAK/CmXG+zVpb2ZiBrbWVxpjrgLmAF3jCWrsiMOYSa+1s4DpjzCSgAigGrghcu8IY8wLOBycrgZ8cbccRkXA2ZUQ6d/x3BbPztzIsPdHtchzeKOhzgvM4/Q+wax18McdZ273gr/DBvdCpOww+AwZPgf4nO2vBRUREpNGOur2fMaYa2ADUbARcc7IBrLU2M7TlNZ6295NwdsUTiyjYuZ8Pbp7Y+stHmqp0N6yb5+xism4+lO+FqDgnbA+e7DwS092uUkRExDVB2d4P6BekekTatRy/j5tezCdvUwnH9O7idjlHF58MmRc6j8pD8NWHzg4ma153vgL4Rn29i0n3EepOKSIiUo9mNawxxnhwms88E/ySmkcz2hLO9pZVkP2HeVx2XG9+mzPc7XKax1rYserrwL15CWAhsWcgdE+GvhMgKtbtSkVEREKqsTPaDS0dSQR+grOH9WzgLZyW7L8A8q215wSn3JZT0JZwd/VTS/hsYwkf3XYqXk8bmAHev8NpkPPFG1DwNlSUQkwnGHCKE7wHnQEdU9yuUkREJOiCtXTkKZwPKX4E/BD4Fc767HOttXktrlKkHcnx+5i7YjuLNuxm7IA2EEA7dYNRlzuPioOw4YOvZ7tXzQbjcZrj1CwxSR2sJSYiItKuNBS0+1trRwIYYx4HCoHe1tqykFcm0sacOrQ78TFecpdtbRtBu67oDjD4dOdhLRTmOft1r3kd5v3WeST3d3YwGTIFeh8P3mi3qxYREQmphoJ2Rc0Ta22VMWazQrZI83SI8XJaRnfmLC/k91OHE+1t1Db2kccY8B3jPCbeBnu2BD5MOQcWPw4fPwxxnWHQ6c4OJgMnQYckt6sWEREJuoaCtt8YU9Oj2QAdAq9rtvcLk02BRSJDTqaP/+ZtZcG6XUwc0s3tclpH5x5w7FXOo3w/rH/Hme3+4g1Y/iJ4Ant618x2J2uzIxERaRsaasHuba1CRNqDCYNTSYyLIjd/a/sJ2nXFdoJhOc6jugq2LHWWl6x5A+be5jy6DnUC9+Ap0DMbPPprSEREIlNDM9oiEkSxUV6mjEjnf8sLKauoIi66HYdIjxd6jXEek34Huzd8vV/3wgdhwX0Qn+p0pxwyBfpPdIK6iIhIhFDQFmllOX4fzy/ZxLtrdjB5hDos1kruB8f/2HkcLHG6U37xBqx+DfKeAW8s9DvR2a978BRnSYqIiEgYa1bDmnCkfbQlUlRWVXP83fM5rl8KD182yu1ywl9VBWz8+OvZ7t3rneNpmTDkTCd4p2dp60AREWk1wdpHW0SCLMrr4cyR6bywZBP7yyvpFKs/hkfljYZ+E5zH6X+AXWu/3q/7/T/BezMgwff1EpN+J0F0nNtVi4iI0Eb3FxMJb1P9Psoqqpm3crvbpUQWY6DrYBj/M/jBG3DTOjj3UedDk8tfhGcvhMcmQIV2IRUREfcpaIu4YFTvLvg6x5Gbv9XtUiJbxxTIugQuegpuXg/nPgK7voBFM92uTEREREFbxA0ej+Fsv4/31+6kpPSQ2+W0DVGxkHUpDDwNPrgXSne7XZGIiLRzCtoiLpnq91FRZZm7YpvbpbQtk34HZXthwV/drkRERNo5BW0Rlwz3JdIvtSOztXwkuNJGODPbn8yEko1uVyMiIu2YgraIS4wx5GSm81FBETv26cN7QTXxV84HJ9/+f25XIiIi7ZiCtoiLcvw+qi3MWa7lI0HVuSccdw0sex4Kl7ldjYiItFMK2iIuGtQ9gaFpCVo+Egrjb4QOSTDvt25XIiIi7ZSCtojLcvw+ln5VzObiUrdLaVs6JMGJv4SCt52HiIhIK1PQFnFZTqYPgP8tK3S5kjbo2B9CUm946w6orna7GhERaWcUtEVc1jslnqxeSeQu0/KRoIuKhVPugG3Lnc6RIiIirUhBWyQM5Ph9fL5lL+t37ne7lLZnxHmQ7oe3/6DW7CIi0qoUtEXCwFkj0zEGcvO1fCToPB447U7YsxEW/8PtakREpB1R0BYJA2md4xjTN5nZ+Vuw1rpdTtvT/2QYOAnevxcOFrtdjYiItBMK2iJhYmqWj4KdB1hVuM/tUtqmSb+Hsj3wgVqzi4hI61DQFgkTU0ak4/UYfSgyVNJGgP8S+OQxKNnkdjUiItIOKGiLhInkjjGMH5hKbv5WLR8JlVN+7Xx9R63ZRUQk9BS0RcLIVL+PzcUHydtU4nYpbVPnnnD8NZA/y9nyT0REJIQUtEXCyGnDuxMT5VFL9lAa/3Ona+Rbas0uIiKhpaAtEkYS46KZOKQr/1tWSFW1lo+ERIckmHATFMyHgnfcrkZERNowBW2RMJPj97FjXzmLNux2u5S2a8yP1JpdRERCTkFbJMycOrQ78TFe7T4SSlGxcMpvYNsy+Pwlt6sREZE2SkFbJMx0iPFyWkZ35iwvpKJKs60hM+J8SMuE+XdBZbnb1YiISBukoC0ShnIyfRSXVrBg3S63S2m76rZmX6TW7CIiEnwK2iJhaMLgVBLjosjV7iOhNWAiDDgV3v+zWrOLiEjQKWiLhKHYKC9TRqTz5ortlFVUuV1O23ZaoDX7gvvcrkRERNoYBW2RMJXj97G/vJJ31+xwu5S2LW0k+C+Gjx9Va3YREQkqBW2RMHV8/2RSO8Xw8DsFzM7fys59+sBeyEysac3+R3frEBGRNiXK7QJEpH5RXg8/P20Id7++ihue+wyAwd07ccKAVE4YkMJx/VPo3CHa5SrbiKRecNzVsPBBGHutM8stIiLSQsbattF9Ljs72y5ZssTtMkSCrrKqmhVb97KwoIiFBbtY/OVuyiqq8RgY0aMzYwekcMKAVI7t24X4GP3s3GwHi+GBLOiZDdNedrsaEREJY8aYpdba7AbPU9AWiSzllVXkbSxhYUERHxUU8dmmYiqqLNFeQ1avpNoZ76zeScRGed0uN7IsfBDevB2+91/of7Lb1YiISJhS0BZpJ0oPVbLky+JA8N7F8i17qLYQF+3h2L7JtTPeI3yJRHn1sYyjqiiDh46F+C7wo3edvbZFREQOo6At0k7tOVjBJ+uLame812zfB0BCXBTH9UvhhAEpnDAwhcHdEvB4jMvVhqH85+HV6fDdxyHzArerERGRMKSgLSIA7NxXzse1wXsXXxaVApDSMYbjBzjBe9yAVPqkxGOMgjfV1TDzRGdv7euWQFSs2xWJiEiYUdAWkXptLi7lo8Bs94cFu9i+19k20Nc5jrGB9d0nDEwhvXMHlyt1UcHb8NR34Iw/wtifuF2NiIiEGQVtEWmQtZYNuw7ULjNZWLCL4tIKAPqldgys705hbP8UUjq1s5ndp74DWz+DG/KgQ5Lb1YiISBhR0BaRJquutqzeto+FBbv4qKCITzbsZn95JQBD0xJqdzQZ0z+ZxLg2vod34TJ47EQY91OnTbuIiEiAgraItFhlVTXLt+ypnfFe/OVuyiudPbxH9kxylpkMSCG7TzIdYtrgVoKvXA0rXoUbPoXOPd2uRkREwoSCtogEXVlFFZ9tLOGjgl0sLCgib1MJldWWGK+HY3oH9vAemIK/ZxIxUW1ga7ySjfBgNow8H879u9vViIhImFDQFpGQO1BeyeIvdwfWdxfx+dY9WAvxMV6O7ZscmPFOJcOXiDdStxJ883ZY+BBcswDSRrhdjYiIhAEFbRFpdSWlh/h4/e7aGe+1O/YDkBgXxfH9a/bwTmVQt06Rs5VgbWv2Y2HaS25XIyIiYaCxQTuqNYoRkfYhKT6GySPSmDwiDYAd+8pqtxJcWFDEmyu3A5DaKbZ2ffcJA1LpldwhfIN3hy4w4Rfw1m9g/XvQ/yS3KxIRkQihGW0RaTWbdpfWbiO4sKCIHfucPbx7JHWo3b/7hAGpdE+Mc7nSw1SUwUPZEJ8CP3pHrdlFRNo5LR0RkbBmraVg5wE+KtjFh+uK+Gh9EXsOOnt4D+jasXYrweP7p9ClY4zL1QL5s+DVq+G8fzofjhQRkXZLQVtEIkp1tWVl4d7aGe9FG3Zz4FAVxsCwtMTaGe9j+yaT4MYe3tXVzr7a5XvhusVqzS4i0o4paItIRKuoqmbZ5j0sXOcsM1m6sZhDldV4PQZ/z861M96j+nQhLrqV9vCubc1+N4y9tnXuKSIiYUdBW0TalLKKKj79qpiFgRnv/M17qKq2xER5GN27S+2Md2bPJKK9IVxD/e9zoTBPrdlFRNqxsAjaxpjJwAOAF3jcWjvjsPd/DvwQqAR2Aj+w1n4VeK8KWB44daO1durR7qWgLdK+7C+vZPGG3XwYmPFeWbgXgP5dO/LKj08gKT5E67oL8+Gxk2D8z2DS70JzDxERCWuuB21jjBf4AjgN2AwsBi6x1q6sc85E4BNrbakx5sfAydbaiwLv7bfWdmrs/RS0Rdq33QcO8c7qHdz6yjLGD0zln1cciydUTXJemQ4r/wvXL1VrdhGRdqixQTuUe1SNAdZZa9dbaw8Bs4Bz6p5grX3HWlsaePkxoP9jiUizJHeM4bzRPbnj7AzeWbOTh95ZF7qbnXI72Gp45+7Q3UNERCJeKIN2D2BTndebA8eO5CpgTp3XccaYJcaYj40x59Z3gTFmeuCcJTt37mx5xSIS8aYd34fvHNOD++Z9wXtfhOjvhaTeMGY65D0D21eE5h4iIhLxwqLrgjFmGpAN/LnO4T6BKflLgfuNMQMOv85aO9Nam22tze7atWsrVSsi4cwYwx+/M5Ih3RP46azP2Fxc2vBFzTHhFxCXCPN+F5rxRUQk4oUyaG8BetV53TNw7BuMMZOAXwNTrbXlNcettVsCX9cD7wLHhLBWEWlDOsR4eWTaaKqqLNc+8ynllVXBv0l8shO2174JG94P/vgiIhLxQhm0FwODjDH9jDExwMXA7LonGGOOAR7DCdk76hzvYoyJDTxPBcYBKxERaaR+qR2590I/yzbv4fe5IfrrY8zV0LkXvHWH09BGRESkjpAFbWttJXAdMBdYBbxgrV1hjLnTGFOzVd+fgU7Ai8aYPGNMTRAfBiwxxuQD7wAz6u5WIiLSGGcMT+Oakwbw7CcbeWnp5uDfIDoOJv4atn4GK14J/vgiIhLR1LBGRNq0yqpqLv/nIj7dWMwr157AcF/n4N6gusrZV1ut2UVE2o1w2N5PRMR1UV4PD156DF3iY/jx05+yp7QiuDfweOG030PJV7DkieCOLSIiEU1BW0TavNROsTx82SgK9xzk5y/kUV0d5H/JG3gq9D8Z3vsTlO0J7tgiIhKxFLRFpF0Y3acLt5+VwfzVO3jkvYLg3+C0O+Hgblhwf/DHFhGRiKSgLSLtxvfG9mGq38df3lzDgrW7gjt4uh9GXggf/x32fGsnUxERaYcUtEWk3TDGMOO8kQzs1okbZn3G1pKDwb1BTWv2d/8Y3HFFRCQiKWiLSLsSHxPFo9NGc6iymh8Hu5lNlz6B1uzPwnbtSCoi0t4paItIu9O/ayfuvSCT/E0l3PVakAPxhF9AbIJas4uIiIK2iLRPk0ekc/WJ/Xn644288mkQm9nEJ8P4n8PaubDhg+CNKyIiEUdBW0TarV+eMYTj+iXzq1eXs6pwb/AGPu5qSOwJb/1GrdlFRNoxBW0RabdqmtkkxkXz46eXsudgkJrZRHdwPhi59TNY+WpwxhQRkYijoC0i7Vq3hDj+ftkoNhcf5KYX84PXzCbzQug+AubfCZWHgjOmiIhEFAVtEWn3svsm86szh/HWyu08+n6QmtnUtGYv/lKt2UVE2ikFbRER4MpxfTk7M517567hw3VBamYz4FTodxK8d49as4uItEMK2iIiOM1s7jkvk/5dO3HDc59RuCcIzWyM+bo1+4cPtHw8ERGJKAraIiIBHWOdZjZlFVVc+8ynHKoMwo4hviwYeQF89HfYu7Xl44mISMRQ0BYRqWNgt078+QI/n20s4f/9L0jNbE65HWwVvKPW7CIi7YmCtojIYc4cmc4Px/fjyY++4j+fbWn5gF36wrE/grxnYMeqlo8nIiIRQUFbRKQet0wZypi+ydz2ynLWbNvX8gFPvAli1JpdRKQ9UdAWEalHtNfDQ5ceQ6e4KK55eil7y1rYzCY+GSbcCF+8odbsIiLthIK2iMgRdEuM4+FLR7Fxdym/fDEfa1vYzOa4ayCxB7x1B7R0LBERCXsK2iIiRzGmXzK3TRnK3BXbmfn++pYNVtua/VNYodbsIiJtnYK2iEgDrhrfjzNHpnHPG6v5qKCoZYNlXgTdhsP836s1u4hIG6egLSLSAGMMfzrfT7/Ujlz/3Kds21PW/ME8XqeJTfGXsPRfQatRRETCj4K2iEgjdIqN4rHLR1N6qIqfPNvCZjYDT4V+JwZas+8NXpEiIhJWFLRFRBppYLcE7jkvk6VfFXP3nBbsh13Tmr20SK3ZRUTaMAVtEZEmyPH7uHJcX/714ZfMzm9BS3XfMTDifPjoYbVmFxFpoxS0RUSa6FdnDiO7TxdufXkZa7e3oJnNqb+B6kp49+7gFSciImFDQVtEpImivR4evmwU8TFRXP30UvY1t5lNl74w5kfw2dNqzS4i0gYpaIuINEP3xDgeuvQYvioq5eaXljW/mc2Jv1RrdhGRNkpBW0SkmY7vn8Itk4cw5/NtPP7BhuYNUrc1+5cLglugiIi4SkFbRKQFfjShP5OHpzHjjdV8sr6ZzWzUml1EpE1S0BYRaQFjDH++IJM+yfFc99xn7NjbjGY20R1g4q9hy1JY+Z/gFykiIq5Q0BYRaaGEuGgevXw0+8sq+cmzn1JR1YxmNv6Lndbs89SaXUSkrVDQFhEJgsHdE5hx3kgWf1nMjDmrmz6Axwun/R6KN8DS/wt6fSIi0voUtEVEguScrB58/4S+/HPBBl5b1owmNAMnQd8J8N4MtWYXEWkDFLRFRILoV2cOY1TvJG5+aRnrdjSxmU3d1uwL/xaaAkVEpNUoaIuIBFFMlNPMpkO0l6ufWsr+8sqmDdBjFIw4DxY+BHsLQ1OkiIi0CgVtEZEgS+/c4f+3d+/xcdV1/sdfn5nJvW1aenELaUGlXFq69BIKKhSlpepPRdD9AZXuQ3dd94eIBdmfrj8VEdB1ld1VQfHyAFdQRBFQFBXBLiByb9MCvXCXe4WUXmjTJJPMfH5/nJPMZDJJZ0JOziR5Px+PecyZ7/meM5+c5pG+5zvfmS+XrVzIX7a18a83DGExmxN6lmb/t2gKFBGREaGgLSISgbcePI1Pv/MwfvvwVn549zPlHbzfG+GofwqXZh/CBytFRKQiKGiLiETkzOPfxIq5b+Crv9vCg89sL+/gpZ+G6glaml1EZBRT0BYRiYiZ8R+nHknTlDo+cU0Lr+wuYzGbhqlw7Kfg8d/DM3dHV6SIiERGQVtEJEKTaqv47qrFvNbRxdk/XV/eYjbHfBwm7q+l2UVERikFbRGRiB0+cxJf/cB8HvjLdi75w2OlH1hVByd8Hl5cC5tviq5AERGJhIK2iMgIOGVhE39/zIH84E9P8/tHyvjaviNXwoy5sOZCyHRFV6CIiAw7BW0RkRHyhfcezoJZk/n09Q/zVOue0g5KJGH5hbD9aS3NLiIyyihoi4iMkJpUksvPWER1KsGZP15HW6mL2cw5MVia/Q4tzS4iMpooaIuIjKD9J9dx6ekLeap1D5+98ZHSFrMxgxMvhL3b4J7Loi9SRESGhYK2iMgIO3bONP5lxaH85qGXuOqeZ0o76IDFMO8DcK+WZhcRGS0UtEVEYvDx49/M8sNn8OXfbmHdsyUuZrPs/OADkXd8NdriRERkWChoi4jEIJEw/vPUBRwwpY6zrmmhdXfnvg/a703h0uw/htYyviZQRERioddp030AABY5SURBVKAtIhKTxroqvnvGYna1d/HJa1voLmUxGy3NLiIyaihoi4jEaO7+k/jKyfO57+ntXHJrCaPUDVPh2HPhsd/Bs/dEX6CIiAyZgraISMw+uLiJDx09m+/f+TS3bPzrvg84Wkuzi4iMBgraIiIV4IL3zeXIpkb+7y8e4ul9LWZTXQ/v+By88CBs+fXIFCgiImVT0BYRqQA1qSSXr1pMVdL4+E9a2Jvex2I2Cz4E0w+HP2ppdhGRSqWgLSJSIQ6YXMe3Tl/I46/s5nP7WswmkQwWsdn+lJZmFxGpUAraIiIVZOkh0zlv+SH8asNL/Pi+ZwfvPGcFHHhssDR75+6RKVBEREqmoC0iUmE+8Y6DOeGwGVx882ZantsxcEczOPGiYGn2uy8duQJFRKQkCtoiIhUmkTC+ceoC/qaxlrN+0sK2PYMsZtO0GOadEizNvruEbywREZERE2nQNrN3mdljZvakmX22yP7zzGyzmT1sZmvM7MC8fR82syfC24ejrFNEpNI01geL2ezYm2b1tevJZAeZr73si1qaXUSkAkUWtM0sCXwHeDcwF1hpZnMLuq0Hmt39b4Hrga+Hx+4HXAAcDSwBLjCzKVHVKiJSiY44oJGLTz6Ce556lf8cbDGb/d4ER30UWrQ0u4hIJYlyRHsJ8KS7P+3uaeBnwPvzO7j77e6+N3x4H9AUbr8TuM3dt7v7DuA24F0R1ioiUpFObZ7FyiWzuPyOp7h10yBTQ5Z+Gqrqg6/7ExGRihBl0D4AeD7v8Qth20A+Cvy+nGPN7J/NbK2ZrW1tbX2d5YqIVKYL3jeP+Qc08i/XPcQz29qKd2qYFi7N/lt49t6RLVBERIqqiA9DmtkqoBm4pJzj3P0H7t7s7s3Tp0+PpjgRkZjVViW5/IxFJJPGmT9ZR3s6U7zjMWfBxJlw2/laml1EpAJEGbRfBGblPW4K2/ows+XA54GT3L2znGNFRMaLWfvV883TFvDYy7v5/C8HWMymz9Lsvxn5IkVEpI8og/aDwBwze6OZVQOnA7/O72BmC4HvE4TsV/J2/QFYYWZTwg9BrgjbRETGrbcfOoNzls3hxvUvcs39zxXvdOSHYPph8McvaWl2EZGYRRa03b0bOJsgIG8BrnP3TWZ2kZmdFHa7BJgA/MLMNpjZr8NjtwMXE4T1B4GLwjYRkXFt9QlzePuh07noN5vZ8PzO/h2SKViupdlFRCqBFX37cRRqbm72tWvXxl2GiEjkdu5N855L/4y7c/Pq49ivobpvB3f40Xtg2+Owej3UTIynUBGRMcrM1rl78776VcSHIUVEpHST66v53qrFbGsbYDEbMzjxYmhrhXsui6dIERFR0BYRGY3mNzVy0Unz+POT2/jGbY/379C0GOaeDPdoaXYRkbgoaIuIjFKnL5nNqc1NfPv2J1mz5eX+HZZ9ETKdcMe/j3xxIiKioC0iMppd9P4jmLf/JD718w089+revjunvhmaPwotV0NrkVFvERGJlIK2iMgoVluV5HurFmMWLGbT0VWwmM3xnwmWZl+jpdlFREaagraIyCjXs5jN5q2v8YVfbey7mE3DNDj2HHj0ZnjuvviKFBEZhxS0RUTGgHccNoPVJxzM9ete4NoHnu+785izYMLfwK1aml1EZCQpaIuIjBHnLD+E4+ZM40u/3sRD+YvZVDeES7M/EIxsi4jIiFDQFhEZI5IJ49LTFzJ9Yg1nXdPCjrZ0bueCM2DaoVqaXURkBCloi4iMIVMaqrn8jEW07u7knJ9vyC1mk0zBiRfCq09Cy1XxFikiMk4oaIuIjDFHzprMl06ax58eb+Vba57I7TjkXTD7rcH3anfujq9AEZFxQkFbRGQMWrlkFh9c1MSla57g9kdfCRrNYEXP0uzfjrdAEZFxQEFbRGQMMjO+fPIRHD5zEuf+fAPPbw8Xs2lqDpdmvwx2F1lNUkREho2CtojIGFVXneR7qxaRde+7mE3P0ux3aml2EZEoKWiLiIxhB05t4JunLWDTS69xwU2bgsapb4bmf4R1V8G2JwY/gYiIDFkq7gJERCRayw5/A2e/42C+ffuTLDpwMqcdNRuWfgY2/BRuOhsOeWcwf9sSQHhviVybhWMy/dpsgDYboF+iyPPY4M/dp40Snrvcc4Z9S+oX3ouIlEhBW0RkHPjUiYew4fmdnH/TJubObGR+03RYdgHc8q/wvJZmL5klobYR6qZA3eTgvnZyaY+rauOuXkRGmPkYWY63ubnZ165dG3cZIiIV69U9nbzvsj+TSBg3f/JYJtdXQ3caPAt4cO/ZYJn23jbv29anb2GbD9Cv8JzZvn2L9evTTgnPXe45i/UdqHZybZk0tO+Ejp3QviPYbt8RPt4Z9BlIqra0YF43GWrz9zdCIhnVr4WIDIGZrXP35n3104i2iMg4MXVCDd85YxGnfv9ezv35Bn744aNIpKrjLmvsyGah87W84F0YxPMf74Kdz8HWh4PHXW2Dn7umMQzeZY6iVzdouotIjBS0RUTGkYWzp/DF983j/F9t5LL/eZJzls+Ju6SxI5HIheFydadLCOd5j3e9mHuc7R6kplR5wbx3RH0y6EWYVAJ36O6EdBuk94S3ttz9zAUweVbcVQ5IQVtEZJxZdfRs1j+7g2+ueZz5TZM4/pAZJBMa9YxVqhomzAhu5XAPwkYpo+jtO2DPy9D6KLTvgs5dg5+7qqEgjJc4N71mUvCiQ8an7nRBIG4LVqLt2S5nX3oPdO4Bzwz8fCd/DxasHLmfr0wK2iIi44yZ8ZVT5rN562v844+Cz7ZUJY3aqmR4S1CbCrbrqpLUVCV699XlbdemEtSEfXqP63OfDM+ToK462XvOmlSChIL98DCDmgnBjTJH9TLduakupYyiv/pU7nF3xyA1JYJQXjs5uE/VQqomuCWrw/ua4MVFn/vaIm35x1QX9CncF543qWhTskxXLth2FowUlxWW8x5nu0p//uoJwfSm/Pv6aTDloP7tPds1BcdMnh3Z5RkO+m0UERmH6qqTXP3RJdy0/iXa0t10dGXp6MrQ2Z2hoytLezpDR3eGjq4Muzu6ad3dSUdXsK+nvaMrO+Tnr04lwoCe6BPIawtCe0+Ir+kT/vv3K3yR0BPsa6oS1KQSmOYp95dMQf1+wa1cXe3FPxRaGM47dgWhvLsj2M6kg2kAPffdncHiSd2dDPpB0nJYcvBQn6otHtD79Sn2YiDvmH7HFQn+qdrh+yBrprv/aG+x0d9+YblwX96xmXTpz19V3zf41kwI3sFobOobhPP39wvLeftSdePinQ8FbRGRcWrGxFo+tvRNQz7e3enszuYCeFcQztvTuUDeGe5r78r069eR7h/c27sy7Nyb7tMehP4s6e6hBXszqOkN9rlR9dwoe5GwXmQ0vqYqOEd9dYq66iT1vbcU9dXBi4JxM1JfVRfcJs0cnvO5B3PN80N4pjOYhtATxPu15e3LpMNAX9hWeN8RnmuA4N9z3u4OYgn+iSro2psXiPNGkQd7F6FQqi4XemsmBve1k4J/r+qJAwTihoH3VdXrm2+GSEFbRESGxCw33WQkZLO5YN8nuPcG9bwg35Xr19kVBPWePu1due3OriyvtqXzRvCzve3pTPnBvrYq0Ru866uT1FWnqK9K0lCT2+4J6Q01qTC497SlaMjbzg/ytVVjfFTeDJJVwa0S7DP4h2F80OBfEO6LBv+e84bBP9sdvICpnhDM1+8zXaJgxDj/cU1BQFYorhgK2iIiMiokEkZdGESnjMDzZbKeC/A9AT8dhPe96Qx7O7uD+64M7elu2jp79gXt7ekMbelg39ZdXeHj3L7ubOkjpmb0hvLeEfSeMF+VoqEmt92zr6G6cPQ9/wVA7rGm1hRRacFfRi0FbRERkSKSCaOhJkVDTTT/Vaa7s/3Cd09I3xtuFw/1Gdo6c/u2t7UHQT/vHGVkeBLGwIG8Khx5r05SX5UbpW+oSfZOo8kF+/5TaqpTY38OrshgFLRFRERiUJ1KUJ1K0Fg/vKOmPXPne8N6b3DvO9q+t084z9DeFewLRua72dMZfAi2reAc5UgYJIqMlg80gG4U2TFg3yJtZZy3WN+BxvWLjfgP+B7A6zxvMmEkLNiXMEiaBduJYDthhoXXNZmwXL/87aL9ctt9zt97nJEM+xQ9Z2KQc1runH3PbyQTuXMmLXhnqlidxc7fe87w58/Vmdt38IwJTJ9YM9C/RuwUtEVERMaQ/Lnz+zUM76Iz7k5HV7YgfBeMwudtd3Rl8YIPFfoAo+3FmgfuW2RHWeft3xpVXQP1LXq8O1mHbM991nPb7kXaB+mXDba7s1nSmbA9O9D5ix9b9PwD1eRe1s86XP7r1CP5wKKmkX/iEiloi4iISEnMcvPkRQr5YIG9MLwP9kIh7wVBJhsE+Kw7Gffcc2SDxwfPmBD3jz0oBW0RERERed16pockB55YM+7oUwoiIiIiIhFQ0BYRERERiYCCtoiIiIhIBBS0RUREREQioKAtIiIiIhIBBW0RERERkQgoaIuIiIiIREBBW0REREQkAgraIiIiIiIRUNAWEREREYmAgraIiIiISAQUtEVEREREIqCgLSIiIiISAQVtEREREZEIKGiLiIiIiETA3D3uGoaFmbUCz8b09NOAbTE992ik61UeXa/y6HqVR9erPLpe5dH1Kp+uWXniul4Huvv0fXUaM0E7Tma21t2b465jtND1Ko+uV3l0vcqj61UeXa/y6HqVT9esPJV+vTR1REREREQkAgraIiIiIiIRUNAeHj+Iu4BRRterPLpe5dH1Ko+uV3l0vcqj61U+XbPyVPT10hxtEREREZEIaERbRERERCQCCtoiIiIiIhFQ0H4dzOyHZvaKmW2Mu5bRwMxmmdntZrbZzDaZ2Tlx11TJzKzWzB4ws4fC63Vh3DVVOjNLmtl6M7s57lpGAzN7xsweMbMNZrY27noqnZlNNrPrzexRM9tiZm+Ju6ZKZWaHhr9XPbfXzOzcuOuqZGb2qfBv/UYzu9bMauOuqZKZ2TnhtdpUyb9bmqP9OpjZUmAPcLW7HxF3PZXOzGYCM929xcwmAuuAk919c8ylVSQzM6DB3feYWRXwZ+Acd78v5tIqlpmdBzQDk9z9vXHXU+nM7Bmg2d21OEYJzOwq4C53v8LMqoF6d98Zd12VzsySwIvA0e4e18JyFc3MDiD4Gz/X3dvN7Drgd+7+o3grq0xmdgTwM2AJkAZuAc509ydjLawIjWi/Du7+J2B73HWMFu6+1d1bwu3dwBbggHirqlwe2BM+rApvemU8ADNrAt4DXBF3LTL2mFkjsBS4EsDd0wrZJVsGPKWQvU8poM7MUkA98FLM9VSyw4H73X2vu3cDdwIfiLmmohS0JRZmdhCwELg/3koqWzgVYgPwCnCbu+t6DeybwGeAbNyFjCIO3Gpm68zsn+MupsK9EWgF/jucnnSFmTXEXdQocTpwbdxFVDJ3fxH4D+A5YCuwy91vjbeqirYROM7MpppZPfC/gFkx11SUgraMODObANwAnOvur8VdTyVz94y7LwCagCXh22VSwMzeC7zi7uvirmWUOdbdFwHvBj4RToeT4lLAIuC77r4QaAM+G29JlS+cYnMS8Iu4a6lkZjYFeD/BC7r9gQYzWxVvVZXL3bcAXwNuJZg2sgHIxFrUABS0ZUSFc41vAK5x9xvjrme0CN+ivh14V9y1VKi3ASeFc45/BpxgZj+Jt6TKF46i4e6vAL8kmO8oxb0AvJD3rtL1BMFbBvduoMXdX467kAq3HPiLu7e6exdwI/DWmGuqaO5+pbsvdvelwA7g8bhrKkZBW0ZM+OG+K4Et7v5fcddT6cxsuplNDrfrgBOBR+OtqjK5+/9z9yZ3P4jgber/cXeNBg3CzBrCDyUTToFYQfB2rBTh7n8FnjezQ8OmZYA+yL1vK9G0kVI8BxxjZvXh/5XLCD7HJAMwsxnh/WyC+dk/jbei4lJxFzCamdm1wNuBaWb2AnCBu18Zb1UV7W3A3wOPhPOOAT7n7r+LsaZKNhO4KvzEfgK4zt31tXUyXN4A/DL4P50U8FN3vyXekireJ4FrwukQTwP/EHM9FS18AXci8H/irqXSufv9ZnY90AJ0A+up8KXFK8ANZjYV6AI+UakfTtbX+4mIiIiIREBTR0REREREIqCgLSIiIiISAQVtEREREZEIKGiLiIiIiERAQVtEREREJAIK2iIiw8DMMma2wcw2mtkvwmWBB+u/Z6RqK3jeZjO7dB993m5m/b5KMmzfFS5B/piZ/SlclVNERIpQ0BYRGR7t7r7A3Y8A0sCZcRdUjLuvdffVr+MUd7n7Qnc/FFgNfNvMlr3euiyg/5NEZEzRHzURkeF3F3AwgJmdF45ybzSzcws7mtnVZnZy3uNrzOz9ZvYRM7vRzG4xsyfM7Ot5fVaa2SPhOb+W177HzC4xs01m9kczW2Jmd5jZ02Z2Utind7Q63H9vOEJ9T96qhyVx9w3ARcDZ4fmmm9kNZvZgeHtbXvttYV1XmNmzZjbNzA4KR8avJliVcpaZrQhragnfGZgQnmOxmd1pZuvM7A9mNrOcWkVE4qCgLSIyjMwsBbybYAXUxQSrBx4NHAN8zMwWFhxyJfCR8NhG4K3Ab8N9C4DTgPnAaWY2y8z2B74GnBDuPyovqDcQLD8/D9gNfJlgZb5TCAJxoUeB49x9IfBF4N+G8CO3AIeF298CvuHuRwEfBK4I2y/Iq+t6YHbe8XOAy8N9bcAXgOXuvghYC5xnZlXAZcDfufti4IfAV4ZQq4jIiNIS7CIiw6POzDaE23cRBOiPA7909zYAM7sROI5geWUA3P1OM7vczKYThNMb3L07XBp9jbvvCo/dDBwITAXucPfWsP0aYCnwK4IpKz3LqD8CdLp7l5k9AhxUpOZG4CozmwM4UDWEn9vytpcDc8PaASaFI9LHEoR93P0WM9uRd8yz7n5fuH0MMBe4OzxHNXAvcChwBHBb2J4Etg6hVhGREaWgLSIyPNrdfUF+Q17g3JergVXA6QQj4D0687Yz7Ptvdpe7e7id7Tne3bPhSHuhi4Hb3f0UMzsIuKPUgvMsBLaE2wngGHfvyO+wj+vQlt8VuM3dVxYcPx/Y5O5vGUJ9IiKx0dQREZHo3AWcbGb1ZtZAMKp7V5F+PwLOBXD3zfs45wPA8eEc5ySwErhziPU1Ai+G2x8p92Az+1vgfOA7YdOtwCfz9ve88LgbODVsWwFMGeCU9wFvM7Oe+e0NZnYI8Bgw3czeErZXmdm8cusVERlpCtoiIhFx9xaCEP0AcD9whbuvL9LvZYJR4f8u4Zxbgc8CtwMPAevc/aYhlvh14Ktmtp7S3+E8rufr/QgC9mp3XxPuWw00m9nD4VSXnm9euRBYYWYbgf8N/JVgDnnhz9ZKEPivNbOHCaaNHObuaeDvgK+Z2UPABoK57CIiFc1y7zKKiEgcwu/cfgRY1DMneywxsxogE849fwvw3cJpNiIiY5HmaIuIxMjMlhN8cPIbYzFkh2YD14Xfk50GPhZzPSIiI0Ij2iIiIiIiEdAcbRERERGRCChoi4iIiIhEQEFbRERERCQCCtoiIiIiIhFQ0BYRERERicD/BxaNOdGR5mUzAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Plot the two RMSE of the training and the test data\n",
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df.degree, rmse_df.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df.degree, rmse_df.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 214,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10ce9cb10>"
+      ]
+     },
+     "execution_count": 214,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# It can observed our model starts to overfit ... as we would expect with a small dataset ... around 5\n",
+    "# Here there is no real increase or trade off between the error terms after 5\n",
+    "lm_model_5 = LinearRegression()\n",
+    "lm_model_5.fit(vander(x, 6), y_scatter)\n",
+    "degree_5 = lm_model_5.coef_.size - 1\n",
+    "y_pred_5 = lm_model_5.predict(np.vander(x, degree_5 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y)\n",
+    "plt.plot(x, y_pred_5)\n",
+    "plt.scatter(x, y_scatter)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['True Function', 'Pred. Deg. 5', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 332,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Scatter Curve To Estimate')"
+      ]
+     },
+     "execution_count": 332,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Let's try a different set of data\n",
+    "def curve(x_val):\n",
+    "    return (x_val ** 2) + 3.0\n",
+    "\n",
+    "y_scatter_curve = curve(x)\n",
+    "noise = 20 * (np.random.random(length) - 4.0)\n",
+    "y_scatter_curve = y_scatter_curve + noise\n",
+    "\n",
+    "plt.figure(figsize=(12, 7)) \n",
+    "plt.scatter(x, y_scatter_curve)\n",
+    "plt.title(\"Scatter Curve To Estimate\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 333,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "rmse_df_curve = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter_curve,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df_curve.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 334,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 334,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df_curve.degree, rmse_df_curve.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df_curve.degree, rmse_df_curve.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 335,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x110236a10>"
+      ]
+     },
+     "execution_count": 335,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Here we see a divergence ... so we'll stick to Poly level 2\n",
+    "# It can observed our model starts to overfit ... as we would expect .. around 5\n",
+    "# Here there is no real increase or trade off between the error terms after 5\n",
+    "# Mostly because we're extimating a cos wave, and it will be roughly the same \n",
+    "lm_model_2 = LinearRegression()\n",
+    "lm_model_2.fit(vander(x, 3), y_scatter_curve)\n",
+    "degree_2 = lm_model_2.coef_.size - 1\n",
+    "y_pred_2 = lm_model_2.predict(np.vander(x, degree_2 + 1))\n",
+    "\n",
+    "# Plot side by size\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.plot(x, y_pred_2)\n",
+    "plt.scatter(x, y_scatter_curve)\n",
+    "plt.title(\"Scatter Vs. Actual\")\n",
+    "plt.legend(['Pred. Deg. 2', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 338,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Scatter With More Noise')"
+      ]
+     },
+     "execution_count": 338,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Add some more bumps to the data\n",
+    "\n",
+    "# Make a copy\n",
+    "y_scatter_curve_noise = y_scatter_curve.copy()\n",
+    "\n",
+    "for i, j in enumerate(y_scatter_curve_noise):\n",
+    "    if i < 25:\n",
+    "        y_scatter_curve_noise[i] = j + (5 * (np.random.random() + 5.0))\n",
+    "    if i > 100:\n",
+    "        y_scatter_curve_noise[i] = j + (5 * (np.random.random() - 5.0))\n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.scatter(x, y_scatter_curve_noise)\n",
+    "plt.title(\"Scatter With More Noise\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 385,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create a dataframe to captire the rmse\n",
+    "rmse_df_curve_noise = pd.DataFrame(columns=[\"degree\", \"rmse_train\", \"rmse_test\"])\n",
+    "\n",
+    "# Number of degress to test in our model\n",
+    "train_X, test_X, train_y, test_y = train_test_split(x, y_scatter_curve_noise,\n",
+    "                                                    test_size=0.33,\n",
+    "                                                    random_state=1075)\n",
+    "\n",
+    "# Get the rmse for each prediction\n",
+    "for i in range(1, 10):\n",
+    "    p = np.polyfit(train_X, train_y, deg=i)\n",
+    "    rmse_df_curve_noise.loc[i-1] = [i,\n",
+    "                        get_rmse(train_y, np.polyval(p, train_X)),\n",
+    "                        get_rmse(test_y, np.polyval(p, test_X))]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 387,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5,1,'Train Vs. Test Error')"
+      ]
+     },
+     "execution_count": 387,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "rmse_df_curve_noise\n",
+    "plt.figure(figsize=(12, 7))\n",
+    "plt.plot(rmse_df_curve_noise.degree, rmse_df_curve_noise.rmse_train, label='Training Data')\n",
+    "plt.plot(rmse_df_curve_noise.degree, rmse_df_curve_noise.rmse_test, label='Test Data')\n",
+    "plt.ylabel('RMSE')\n",
+    "plt.xlabel('Polynomial Degree')\n",
+    "plt.legend()\n",
+    "plt.title('Train Vs. Test Error')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 419,
+   "metadata": {
+    "code_folding": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Create graphs of the different values of alpha you're testing\n",
+    "from sklearn.linear_model import Ridge\n",
+    "\n",
+    "# Set the expected values of lambda\n",
+    "alphas = np.linspace(.00001, 2, 500)\n",
+    "alphas_to_display = [alphas[0], alphas[100], alphas[200], alphas[300], alphas[400], alphas[499]]\n",
+    "ridge_error_df = pd.DataFrame(columns=[\"alpha\", \"rss\", \"intercept\", \"coef\"])\n",
+    "\n",
+    "def ridge_model_comparison(alphas, poly_degree, X_values, y_values):\n",
+    "    # Set local variables\n",
+    "    count = 0\n",
+    "    subplot = 1\n",
+    "    fig = plt.figure(figsize=(12, 8))\n",
+    "    \n",
+    "    # Construct your model to evaluate\n",
+    "    for i in alphas:\n",
+    "        ridge_model = Ridge(alpha=i, normalize=True)\n",
+    "        ridge_model.fit(vander(x, poly_degree + 1), y_values)\n",
+    "        ridge_degree = ridge_model.coef_.size - 1\n",
+    "        y_pred = ridge_model.predict(np.vander(x, ridge_degree + 1))\n",
+    "\n",
+    "        # Only display certain models\n",
+    "        if i in alphas_to_display:\n",
+    "            plt.subplot(230 + subplot)\n",
+    "            plt.tight_layout()\n",
+    "            plt.plot(X_values, y_pred)\n",
+    "            plt.plot(X_values, y_values, '.')\n",
+    "            plt.title('Plot for alpha: %.3g on Poly. Deg %d ' % (i, poly_degree))\n",
+    "            subplot = subplot + 1\n",
+    "\n",
+    "        # Fill dataframe\n",
+    "        rss = sum((y_pred - y_values)**2)\n",
+    "        intercept = ridge_model.intercept_\n",
+    "        coef = ridge_model.coef_\n",
+    "\n",
+    "        # Add error data to the dataframe\n",
+    "        # alpha, rss, intercept, coef\n",
+    "        ridge_error_df.loc[count] = [i, rss, intercept, coef]\n",
+    "        count = count + 1\n",
+    "\n",
+    "# Run the model\n",
+    "ridge_model_comparison(alphas=alphas, poly_degree=4, X_values=x, y_values=y_scatter_curve_noise);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 420,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x576 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Try running the model for a Lasso regression\n",
+    "from sklearn.linear_model import Lasso\n",
+    "\n",
+    "def lasso_model_comparison(alphas, poly_degree, X_values, y_values):\n",
+    "    lasso_error_df = pd.DataFrame(columns=[\"alpha\", \"rss\", \"intercept\", \"coef\"])\n",
+    "    # Set local variables\n",
+    "    count = 0\n",
+    "    subplot = 1\n",
+    "    fig = plt.figure(figsize=(12, 8))\n",
+    "    \n",
+    "    # Construct your model to evaluate\n",
+    "    for i in alphas:\n",
+    "        lasso_model = Lasso(alpha=i, normalize=True)\n",
+    "        lasso_model.fit(vander(x, poly_degree + 1), y_values)\n",
+    "        lasso_degree = lasso_model.coef_.size - 1\n",
+    "        y_pred = lasso_model.predict(np.vander(x, lasso_degree + 1))\n",
+    "\n",
+    "        # Only display certain models\n",
+    "        if i in alphas_to_display:\n",
+    "            plt.subplot(230 + subplot)\n",
+    "            plt.tight_layout()\n",
+    "            plt.plot(X_values, y_pred)\n",
+    "            plt.plot(X_values, y_values, '.')\n",
+    "            plt.title('Plot for alpha: %.3g on Poly. Deg %d ' % (i, poly_degree))\n",
+    "            subplot = subplot + 1\n",
+    "\n",
+    "        # Fill dataframe\n",
+    "        rss = sum((y_pred - y_values)**2)\n",
+    "        intercept = lasso_model.intercept_\n",
+    "        coef = lasso_model.coef_\n",
+    "\n",
+    "        # Add error data to the dataframe\n",
+    "        # alpha, rss, intercept, coef\n",
+    "        lasso_error_df.loc[count] = [i, rss, intercept, coef]\n",
+    "        count = count + 1\n",
+    "\n",
+    "# Run the function\n",
+    "lasso_model_comparison(alphas=alphas, poly_degree=4, X_values=x, y_values=y_scatter_curve_noise);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 422,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1e-05\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "51.552909041605204"
+      ]
+     },
+     "execution_count": 422,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Let's try RidgeCV to search for the correct alpha\n",
+    "from sklearn.metrics import mean_squared_error\n",
+    "\n",
+    "alphas = np.linspace(.00001, 2, 500)\n",
+    "\n",
+    "# Do the searching for us\n",
+    "ridgecv = RidgeCV(alphas = alphas, scoring = 'neg_mean_squared_error', normalize = True)\n",
+    "ridgecv.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "print ridgecv.alpha_\n",
+    "\n",
+    "lm_ridge = Ridge(alpha = ridgecv.alpha_)\n",
+    "lm_ridge.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "mean_squared_error(y_scatter_curve_noise, lm_ridge.predict(vander(x, 6)))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 423,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.733473266533066\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "80.92995386478391"
+      ]
+     },
+     "execution_count": 423,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Let's try RidgeCV to search for the correct alpha\n",
+    "from sklearn.linear_model import LassoCV\n",
+    "\n",
+    "# Do the searching for us\n",
+    "lassocv = LassoCV(alphas = alphas, normalize = True)\n",
+    "lassocv.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "print lassocv.alpha_\n",
+    "\n",
+    "lm_lasso = Lasso(alpha = lassocv.alpha_)\n",
+    "lm_lasso.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "mean_squared_error(y_scatter_curve_noise, lm_lasso.predict(vander(x, 6)))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 424,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "51.552909040751885"
+      ]
+     },
+     "execution_count": 424,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Do the searching for us\n",
+    "final_lm_model = LinearRegression()\n",
+    "final_lm_model.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "final_lm_degree = final_lm_model.coef_.size - 1\n",
+    "final_lm_y_pred = final_lm_model.predict(np.vander(x, final_lm_degree + 1))\n",
+    "\n",
+    "mean_squared_error(y_scatter_curve_noise, final_lm_y_pred)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Ridge at 5 = 51.552909041605204 with alpha 1e-05 <- Tied\n",
+    "# Lasso at 5 = 80.92995386478393\n",
+    "# Linear at 5 = 51.552909040751885 <- Tied\n",
+    "\n",
+    "# Ridge at 2 = 90\n",
+    "# Lasso at 2 = 90"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 434,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x10fb9d410>"
+      ]
+     },
+     "execution_count": 434,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAssAAAGrCAYAAAA2FNocAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xdc1WX7wPHPDYIgCDhBXKgpykYBNRzkwrJMHKnZsGFTrZ9PprYebZots2xr6fNUWOZoWI5Hzb1Q3OJEZYiggsoQONy/P86RQEDWYcn1fr3OS8533Pf1/UKvrnOf63vfSmuNEEIIIYQQoiCLqg5ACCGEEEKI6kqSZSGEEEIIIYogybIQQgghhBBFkGRZCCGEEEKIIkiyLIQQQgghRBEkWRZCCCGEEKIIkiwLIcpMKdVTKRVV1XFUJqXUQaVUSBH7QpRSMZUcUpmVJl6l1HSl1H8rOqZC+v1OKfVmZfcrhBDXSbIshCiWUipaKdXvxu1a641aa/eqiKkiKKXclFJaKXXV9IpWSk3Ne4zW2lNrvb4KYtNKqfNKqTp5tlmZtlX7CfNN8aea7musUupDpZRlBfYXopTKyfO7jFFK/aSUCqzAPvNe41Wl1DcV1ZcQovJIsiyEqHHyJowVxElrbQ8MB15VSvWv4P5K6hJwZ573d5q21RS+pvvaF7gfGFfB/cWZ+qsPdAOOABuVUn0rsE9frbW96fV4BfYjhKgkkiwLIcrsxq/xTSOxLyil9imlUpRSi5RSNnn2362UilRKJSultiilfPLsm6qUOqGUuqKUOqSUCsuzb6xSarNS6iOl1AVg+g1xuCql0pVSDfNs81dKJZlGX29TSv1tiilJKbWoJNentd4FHAT8brjGfqafbU1lApeUUoeAfKOWSqnOSqk9pmv62XQ/3syzv8j7UYT/AA/lef8QsLCQe/GrUuqiUuq4Umpcnn3FxeuqlPpFKZWolDqllJpYkvtUWlrrI8BGwMvUbyel1HrTfTiolBpc2HlKqQNKqXvyvLcy/T79i+lPa61jtNavAd8A7+Zpo6NSarXpfkUppe7Ls6+RUuo3pdRlpdROpdSbSqlN5bt6IURNI8myEMLc7gMGAm0AH2AsGJNXYD7wJNAI+BL4VSlV13TeCaAn4AjMAP6rlGqWp92uwEnAGXgrb4da6zhgKzAsz+b7gcVa6yzgDWAV0ABoAXxSkgtRSnXDmNAdL+KQfwPtTK9Q4OE851oDS4HvgIbAj0DeDwDF3Y/CLAN6KaWclFINMN6v5TccEw7EAK4YR8bfVkr1KUG8FsBvwF6gOcbR3+eVUqE3iadMlFIeptj3KKWsTP2uApoCE4DvlVKFlfcsBB7I8/4uIF5rvacU3S8BOiul7JRSdsBq4AdT36OAz0zxAcwFUgEXjPfq4ULau9EGpdQ5pdQSpZRbKeISQlRTkiwLIcxtjtY6Tmt9EWMSdH1U9gngS631dq21QWu9ALiG8etxtNY/m87L0VovAo4BQXnajdNaf6K1ztZapxfS7w/AaACllMKY+Pxg2pcFtAZctdYZWuviRgeTlFLpGBPwzzAmqYW5D3hLa31Ra30WmJNnXzegjul+ZGmtlwA78uy/6f0oQgbGezrS9PrVtA3TdbcEgoEppuuMxDiSen00+mbxBgJNtNava60ztdYnga8x3kdz2a2UumS6hm+AbzFerz0w09TvWuB3TL/LG/wXuEsp5WB6/yDG0fbSiAMU4ATcDURrrb81/V3tAX4BRihjPfUw4N9a6zSt9SFgQTFt9wbcgI6mfn6vhJIhIUQFk2RZCGFu5/L8nIYxEQJjsvov01ftyUqpZKAlxhFQlFIP5SlJSMY4ots4T1tni+n3F6C7aTS6F5CD8at+gBcxJkg7TF/zP1pMW41Ncf8LCAGsijjO9Ya4Tt+wL1Zrnffhu7zH3vR+3MRCjMlvgRIM07kXtdZXboipeQnibQ243hDPSxhH8m9KKfWn+uehtjE3ObSz1rqB1rqd1voVrXXO9ZhMPxcWcy7TNwibgWFKKSeMNdvfFxffDZoDGkjGeM1db7jmMRhHkptg/LCT937d9G9Qa73BlPAnA89h/HalUynjE0JUM/KJVwhRWc5iHNV868YdSqnWGEcx+wJbtdYGpVQkxgT3upvO+KC1vqSUWoVxxLUTEH49UdVan8P0MJlSqgewRim1QWtdVHkFWmsD8KFSaijwDDC7kMPiMSa4B03vW92wr7lSSuVJmFtiLDeBm9yPYmwEmmG8H5swllRcFwc0VErVz5MwtwJiSxDvWeCU1rp9KeNBa31n8UcVKQ5oqZSyyJMwtwKOFnH8AuBxjP//2qq1ji3iuKKEAbu11qlKqbPA31rrAg9wmkaWszGW7VyPpWUp+9Lk/xsWQtRAMrIshCgpK6WUTZ5XaT9sfw08pZTqqozslFKDlFL1ATuMiUUigFLqEUwPf5XSDxhHXIfzTwkGSqkRSqkWpreXTH3lFDy9UDOBF1WeBxXz+AmYppRqYGp/Qp59WwEDMF4pVUcpdS/5y0pudj+KZEq87wEG3zBqjam0Ygvwjul35AM8hrF8obh4dwBXlFJTTA8CWiqlvFQFTrVmsh3jNxAvmh7YC8F4feFFHL8M6Ixx5PbGkfVCme5vc6XUvzEm2i+Zdv0OdFBKPWjq20opFaiU6mT6sLQEmK6UqqeU6kj+hytv7MNTKeVnum/2wAcYP6QcLkmMQojqS5JlIURJrQDS87yml+Zk08wS44BPMSasxzE9/GeqB/0AY4KZAHhj/Lq9tH4F2gPntNZ782wPBLYrpa6ajnnOVJNbEn+Y4i1smrMZGEsGTmF8QC23flZrnQkMxZisJmN8MO13jHXJN70fxdFaH9RaHyxi92iMdbNxGB8w/LfWek0J4jVgrOH1M+1PwlhX7FiSmMrKdJ/uwVhSkYSxRvwh04wZhR2fjrHkpg3GZPZmXE2/86vATox/VyFa61Wmtq4AAzDWZcdhLCF6F7j+kOV4jNd/DuO9+hHT768QzsAi4DLGB1HdgLtND5gKIWowdcPAhBBCiAqilNoOfKG1/raqY6nJlFKvAR201g8Ue7B5+30XcNFal2RWDCHELUJGloUQooIopXorpVxMZRgPY5xK76+qjqsmU8a5tB8DvqqEvjoqpXxMZRxBpn6XVnS/QojqRZJlIYSoOO4Y5y1OxjizxnCtdXzVhlRzKeMCK2eBP7XWGyqhy/oYSz1SMZZYfEDBea2FELc4KcMQQgghhBCiCDKyLIQQQgghRBGq1TzLjRs31m5ublUdhhBCCCGEuMVFREQkaa2bFHdctUqW3dzc2LVrV1WHIYQQQgghbnFKqdPFH2WGMgylVEul1Dql1CHTMrLPmbY3VEqtVkodM/3boLx9CSGEEEIIUZnMUbOcDfxLa+0BdAOeVUp5AFOB/5mWTv2f6b0QQgghhBA1RrmTZa11vNZ6t+nnKxiX9mwO3AssMB22ABhS3r6EEEIIIYSoTGatWVZKuQH+wHbAOc98oucwLgVa2DlPAE8AtGrVqsD+rKwsYmJiyMjIMGeoQpidjY0NLVq0wMrKqqpDEUIIIYSZmC1ZVkrZA78Az2utLyulcvdprbVSqtAJnbXWX2FaiSkgIKDAMTExMdSvXx83NzfytilEdaK15sKFC8TExNCmTZuqDkcIIYQQZmKWeZaVUlYYE+XvtdZLTJsTlFLNTPubAefL0nZGRgaNGjWSRFlUa0opGjVqJN+ACCGEELcYc8yGoYB5wGGt9Yd5dv0KPGz6+WHKsUSoJMqiJpC/UyGEEOLWY44yjGDgQWC/UirStO0lYCbwk1LqMeA0cJ8Z+hJCCCGEEKLSmGM2jE1aa6W19tFa+5leK7TWF7TWfbXW7bXW/bTWF80RcFWwtLTEz88PLy8v7rnnHpKTkwGIi4tj+PDhhZ4TEhJilgVW1q9fj1KKb775JndbZGQkSinef//9ErcTHR2Nl5dXuY+5kb29vVmOMZdPP/2U2267DaUUSUlJpT7/4sWL9O/fn/bt29O/f38uXboEGH8Pjo6O+Pn54efnx+uvv27u0IUQQghRDZmlZvlWZ2trS2RkJAcOHKBhw4bMnTsXAFdXVxYvXlzh/Xt5efHTTz/lvv/xxx/x9fWt8H5rouDgYNasWUPr1q3LdP7MmTPp27cvx44do2/fvsycOTN3X8+ePYmMjCQyMpLXXnvNXCELIYQQohqTZLmUunfvTmxsLJB/JDY9PZ1Ro0bRqVMnwsLCSE9Pzz1n3rx5dOjQgaCgIMaNG8f48eMBSExMZNiwYQQGBhIYGMjmzZsL7bN169ZkZGSQkJCA1pq//vqLO++8M3d/ZGQk3bp1w8fHh7CwsNzR0IiICHx9ffH19c1N8AEMBgOTJ08mMDAQHx8fvvzyy5te89WrV+nbty+dO3fG29ub5csLlp+vX7+eXr16MWjQINzd3XnqqafIycnJ3f/yyy/j6+tLt27dSEhIAOC3336ja9eu+Pv7069fv9zt5eHv74+bm1uB7ampqTz66KMEBQXh7+9f6DUALF++nIcfNpbaP/zwwyxbtqzcMQkhhBCi5jLrPMsVbcZvBzkUd9msbXq4OvDvezxLdKzBYOB///sfjz32WIF9n3/+OfXq1ePw4cPs27ePzp07A8ZSjTfeeIPdu3dTv359+vTpkzsq/Nxzz/F///d/9OjRgzNnzhAaGsrhw4cL7Xv48OH8/PPP+Pv707lzZ+rWrZu776GHHuKTTz6hd+/evPbaa8yYMYPZs2fzyCOP8Omnn9KrVy8mT56ce/y8efNwdHRk586dXLt2jeDgYAYMGFDkA2o2NjYsXboUBwcHkpKS6NatG4MHDy5w/I4dOzh06BCtW7dm4MCBLFmyhOHDh5Oamkq3bt146623ePHFF/n666955ZVX6NGjB9u2bcstM5k1axYffPBBvjajoqIYOXJkoXGtX78eJyenQvfd6K233qJPnz7Mnz+f5ORkgoKC6NevH3Z2dvmOS0hIoFmzZgC4uLjkS+C3bt2Kr68vrq6uvP/++3h6luzvRgghhBA1V41KlqtKeno6fn5+xMbG0qlTJ/r371/gmA0bNjBx4kQAfHx88PHxAYwJZO/evWnYsCEAI0aM4OjRowCsWbOGQ4cO5bZx+fJlrl69WmiN73333cfIkSM5cuQIo0ePZsuWLQCkpKSQnJxM7969AeNo6IgRI0hOTiY5OZlevXoB8OCDD/Lnn38CsGrVKvbt25dbQpKSksKxY8fo0KFDodevteall15iw4YNWFhYEBsbS0JCAi4uLvmOCwoKom3btgCMHj2aTZs2MXz4cKytrbn77rsB6NKlC6tXrwaMc2iPHDmS+Ph4MjMzC52f2N3dncjIyALbS2vVqlX8+uuvuXXeGRkZnDlzhk6dOhV5jlIq9wNB586dOX36NPb29qxYsYIhQ4Zw7NixcsclhBBCiOqtRiXLJR0BNrfrNctpaWmEhoYyd+7c3MS4PHJycti2bRs2NjbFHuvi4oKVlRWrV6/m448/zk2Wy0JrzSeffEJoaGi+7dHR0YUe//3335OYmEhERARWVla4ubkVOp/wjSPN199bWVnl/mxpaUl2djYAEyZMYNKkSQwePJj169czffr0Am2aa2RZa80vv/yCu7t7vu2PPPIIe/bswdXVlRUrVuDs7Ex8fDzNmjUjPj6epk2bAuDg4JB7zl133cUzzzxDUlISjRs3LlH/QgghRG21bE8s762MIi45HVcnWyaHujPEv3lVh1ViUrNcCvXq1WPOnDl88MEHuQnfdb169eKHH34A4MCBA+zbtw+AwMBA/v77by5dukR2dja//PJL7jkDBgzgk08+yX1f3Ajq66+/zrvvvoulpWXuNkdHRxo0aMDGjRsB+M9//kPv3r1xcnLCycmJTZs2AcaE97rQ0FA+//xzsrKyADh69CipqalF9puSkkLTpk2xsrJi3bp1nD59utDjduzYwalTp8jJyWHRokX06NHjpteTkpJC8+bG/1gWLFhQ6DHXR5YLe5U0UQbjNX/yySdobVwkcs+ePQB8++23REZGsmLFCgAGDx6cG8uCBQu49957ATh37lzuuTt27CAnJ4dGjRqVuH8hhBCiNlq2J5ZpS/YTm5yOBmKT05m2ZD/L9sRWdWglJslyKfn7++Pj48OPP/6Yb/vTTz/N1atX6dSpE6+99hpdunQBoHnz5rz00ksEBQURHByMm5sbjo6OAMyZM4ddu3bh4+ODh4cHX3zxxU37vv322xkyZEiB7QsWLGDy5Mn4+Pjkm6nh22+/5dlnn8XPzy830QN4/PHH8fDwoHPnznh5efHkk08WSP7zGjNmDLt27cLb25uFCxfSsWPHQo8LDAxk/PjxdOrUiTZt2hAWFnbT65k+fTojRoygS5cuZhuhnTNnDi1atCAmJgYfHx8ef/xxAF599VWysrLw8fHB09OTV199tdDzp06dyurVq2nfvj1r1qxh6tSpACxevBgvLy98fX2ZOHEi4eHhsgiJEEIIUYz3VkaRnmXIty09y8B7K6OqKKLSU3mTqKoWEBCgb5yb+PDhwzetK60JrtchZ2dnExYWxqOPPlpsIlnTrF+/nvfff5/ff/+9qkOpUrfC36sQQghhLm2m/kFhmaYCTs0cVNnh5I9BqQitdUBxx8nIciWYPn167qImbdq0KXR0WAghhBDiVuPqZFuq7dVRjXrAr6YqzUp7NVVISAghISFVHYYQQgghqpHJoe5MW7I/XymGrZUlk0Pdb3JW9SLJshBCCCGEqBDXZ72oybNhSLIshBBCCCEqzBD/5jUqOb6R1CwLIYQQQghRBEmWhRBCCCGEKIIkyyVQ2PLTX3zxBQsXLqzUOEJCQmjVqlW+OZOHDBlSaHw3M3bs2NylrstzjBBCCCFEmZ3dAWe2QTWaxrgwUrNcRk899VSFtq+1RmuNhUX+zzNOTk5s3ryZHj16kJycTHx8fIXGIYQQQghRIda+AclnYeKeqo7kpmRkuYymT5+eOyVcSEgIU6ZMISgoiA4dOuQuPW0wGJg8eTKBgYH4+Pjw5ZdfAsZFSvr27Uvnzp3x9vZm+fLlAERHR+Pu7s5DDz2El5cXZ8+eLdDvqFGjCA8PB2DJkiUMHTo0d5/WmsmTJ+Pl5YW3tzeLFi3K3T5+/Hjc3d3p168f58+fzz0nIiKC3r1706VLF0JDQyX5FkIIIUTFS4mBUxvBdxRU8xVxa9bI8p9T4dx+87bp4g13zix3M9nZ2ezYsYMVK1YwY8YM1qxZw7x583B0dGTnzp1cu3aN4OBgBgwYQMuWLVm6dCkODg4kJSXRrVs3Bg8eDMCxY8dYsGAB3bp1K7Sfvn37Mm7cOAwGA+Hh4Xz11Ve88cYbgDF5joyMZO/evSQlJREYGEivXr3YunUrUVFRHDp0iISEBDw8PHj00UfJyspiwoQJLF++nCZNmrBo0SJefvll5s+fX+77IYQQQghRpH0/ARp87qvqSIpVs5Llauz6CG+XLl2Ijo4GYNWqVezbty+39jclJYVjx47RokULXnrpJTZs2ICFhQWxsbEkJCQA0Lp16yITZQBLS0t69OhBeHg46enpuLm55e7btGkTo0ePxtLSEmdnZ3r37s3OnTvZsGFD7nZXV1f69OkDQFRUFAcOHKB///6AcSS8WbNm5r41QgghhBD/0Br2hkPLbtCwbVVHU6yalSybYQS4otStWxcwJrPZ2dmAsfzhk08+ITQ0NN+x3333HYmJiURERGBlZYWbmxsZGRkA2NnZFdvXqFGjCAsLY/r06eWKWWuNp6cnW7duLVc7QgghhBAlFh8JSVFw9+yqjqREpGa5AoWGhvL555+TlZUFwNGjR0lNTSUlJYWmTZtiZWXFunXrOH36dKna7dmzJ9OmTWP06NEFti9atAiDwUBiYiIbNmwgKCiIXr165W6Pj49n3bp1ALi7u5OYmJibLGdlZXHw4EEzXLkQQgghRBH2hoOlNXgOqepISqRmjSxXkbS0NFq0aJH7ftKkSSU67/HHHyc6OprOnTujtaZJkyYsW7aMMWPGcM899+Dt7U1AQAAdO3YsVTxKKV544YUC28PCwti6dSu+vr4opZg1axYuLi6EhYWxdu1aPDw8aNWqFd27dwfA2tqaxYsXM3HiRFJSUsjOzub555/H09OzVPEIIYQQQpSIIQv2Lwb3O8G2QVVHUyJKV6O57QICAvSuXbvybTt8+DCdOnWqooiEKB35exVCCCFuIuov+HEkjPoROt5VpaEopSK01gHFHSdlGEIIIYQQonLsC4d6jeC2flUdSYlJsiyEEEIIISpeejIcWQFew6GOdVVHU2JSsyyEEEIIISreoeVguAa+I1m2J5b3VkYRl5yOq5Mtk0PdGeLfvKojLJQky0IIIYQQouLtDYfGHViW4My0pftJzzIAEJuczrQlxkXnqmPCLGUYQgghhBCiYl2KhjNbwGck7606mpsoX5eeZeC9lVFVE1sxJFkWQgghhBAVa+8i478+I4lLTi/0kKK2VzVJlkvA3t6+yvoOCQmhVatW5J3ib8iQIaWOaezYsbnLbpfnGCGEEEKIUskxwO6F0DYEnFri6mRb6GFFba9qkizXAE5OTmzevBmA5ORk4uPjqzgiIYQQQogSOr4GLsdAwKMATA51x9bKMt8htlaWTA51r4roiiXJchn99ttvdO3aFX9/f/r160dCQgIAf//9N35+fvj5+eHv78+VK1eIj4+nV69e+Pn54eXlxcaNGwH48ccf8fb2xsvLiylTphTZ16hRowgPDwdgyZIlDB06NHef1prJkyfj5eWFt7c3ixYtyt0+fvx43N3d6devH+fPn889JyIigt69e9OlSxdCQ0Ml+RZCCCFExdk1H+ydwd24CMkQ/+a8M9Sb5k62KKC5ky3vDPWulg/3QQ2bDePdHe9y5OIRs7bZsWFHpgQVnagWpUePHmzbtg2lFN988w2zZs3igw8+4P3332fu3LkEBwdz9epVbGxs+OqrrwgNDeXll1/GYDCQlpZGXFwcU6ZMISIiggYNGjBgwACWLVvGkCEF10nv27cv48aNw2AwEB4ezldffcUbb7wBGJPnyMhI9u7dS1JSEoGBgfTq1YutW7cSFRXFoUOHSEhIwMPDg0cffZSsrCwmTJjA8uXLadKkCYsWLeLll19m/vz55b6XQgghhBD5JJ+FY6ugxySwtMrdPMS/ebVNjm9Uo5Ll6iQmJoaRI0cSHx9PZmYmbdq0ASA4OJhJkyYxZswYhg4dSosWLQgMDMxNVIcMGYKfnx9r164lJCSEJk2aADBmzBg2bNhQaLJsaWlJjx49CA8PJz09HTc3t9x9mzZtYvTo0VhaWuLs7Ezv3r3ZuXMnGzZsyN3u6upKnz59AIiKiuLAgQP0798fAIPBQLNmzSr4bgkhhBCiOqj0+Y13LwStocvDFddHBatRyXJZRoAryoQJE5g0aRKDBw9m/fr1TJ8+HYCpU6cyaNAgVqxYQXBwMCtXrqRXr15s2LCBP/74g7FjxzJp0iQcHR1L1d+oUaMICwvL7aestNZ4enqydevWcrUjhBBCiJpl2Z5Ypi2pxPmNDVnGZLl9f3BqZf72K4nULJdRSkoKzZsb/7AWLFiQu/3EiRN4e3szZcoUAgMDOXLkCKdPn8bZ2Zlx48bx+OOPs3v3boKCgvj7779JSkrCYDDw448/0rt37yL769mzJ9OmTWP06NEFti9atAiDwUBiYiIbNmwgKCiIXr165W6Pj49n3bp1ALi7u5OYmJibLGdlZXHw4EFz3x4hhBBCVDPvrYyq3PmNj/4FV8/lPthXU9WokeWqkpaWRosWLXLfT5o0ienTpzNixAgaNGhAnz59OHXqFACzZ89m3bp1WFhY4OnpyZ133kl4eDjvvfceVlZW2Nvbs3DhQpo1a8bMmTO544470FozaNAg7r333iJjUErxwgsvFNgeFhbG1q1b8fX1RSnFrFmzcHFxISwsjLVr1+Lh4UGrVq3o3r07ANbW1ixevJiJEyeSkpJCdnY2zz//PJ6enma+a0IIIYSoTip9fuNd88GhBbQfUDHtVxKVd/7eqhYQEKB37dqVb9vhw4fp1KlTFUUkROnI36sQQojqKnjmWmILSYybO9myeWof83Z28STM8YeQlyCk+pTR5qWUitBaBxR3nJRhCCGEEELUApU6v3HEAlCW0PlB87ddyaQMQwghhBCiFrj+EN+Ns2GAcdTZbDNkZF+DPf8F9zvBwdUcoVcpSZaFEEIIIWqJG+c3rpAZMg7/BmlJEPBIueOtDqQMQwghhBCiljL7DBlaw7bPoGFbaGvmOugqIiPLQgghhKi2Kn0RjVrG7DNknNkGsRFw1/tgcWuMyd4aVyGEEEKIW871EoHY5HQ0/5QILNsTW9Wh3TJcnWxLtb1YWz8F24bgN6YcUVUvkiyXQExMDPfeey/t27enXbt2PPfcc2RmZgLw3XffMX78+CqOsCB7e/tCt1taWuLn54eXlxcjRowgLS3tpu3cfvvtxfY1e/bsYtsRQgghSqvSF9Gohcw6Q8aFE3DkDwh8DKzrmSnCqifJcjG01gwdOpQhQ4Zw7Ngxjh49ytWrV3n55ZcrrM/s7OwKa9vW1pbIyEgOHDiAtbU1X3zxxU2P37JlS7FtSrIshBCiIlT6Ihq10BD/5rwz1JvmTrYojHMuvzPUu2ylLlvngqUVBI4ze5xV6ZZLlpftiSV45lraTP2D4Jlry/1Vzdq1a7GxseGRR4xPdFpaWvLRRx8xf/783ATx7NmzhISE0L59e2bMmAFAamoqgwYNwtfXFy8vLxYtWgRAREQEvXv3pkuXLoSGhhIfHw9ASEgIzz//PAEBAbz11lu0bt2anJyc3LZatmxJVlYWJ06cYODAgXTp0oWePXty5MgRAE6dOkX37t3x9vbmlVdeKdG19ezZk+PHjwPw4Ycf4uXlhZeXF7Nnz8495voI9fr16wkJCWH48OF07NiRMWPGoLVmzpw5xMXFcccdd3DHHXdgMBgYO3YsXl5eeHt789FHH5Xr/gshhKi9zF4iIAo1xL85m6f24dTMQWye2qdsiXLaRYj8AXzug/rO5g+yCt1SD/hVxPQnBw/KE2pZAAAgAElEQVQepEuXLvm2OTg40KpVq9xEc8eOHRw4cIB69eoRGBjIoEGDOH36NK6urvzxxx8ApKSkkJWVxYQJE1i+fDlNmjRh0aJFvPzyy8yfPx+AzMxMrq9guHv3bv7++2/uuOMOfv/9d0JDQ7GysuKJJ57giy++oH379mzfvp1nnnmGtWvX8txzz/H000/z0EMPMXfu3GKvKzs7mz///JOBAwcSERHBt99+y/bt29Fa07VrV3r37o2/v3++c/bs2cPBgwdxdXUlODiYzZs3M3HiRD788EPWrVtH48aNiYiIIDY2lgMHDgCQnJxcpvsuhBBCTA51z/f/dajARTRE+eycB9np0L36laaW1y01slxVtU39+/enUaNG2NraMnToUDZt2oS3tzerV69mypQpbNy4EUdHR6Kiojhw4AD9+/fHz8+PN998k5iYmNx2Ro4cme/n66PR4eHhjBw5kqtXr7JlyxZGjBiBn58fTz75ZO7I9ObNmxk9ejQADz5Y9Go56enp+Pn5ERAQQKtWrXjsscfYtGkTYWFh2NnZYW9vz9ChQ9m4cWOBc4OCgmjRogUWFhb4+fkRHR1d4Ji2bdty8uRJJkyYwF9//YWDg0OZ7qkQQghh1hIBUXGyMmDHV3BbP5bFOpj1G/7q4JYaWa6I2iYPDw8WL16cb9vly5c5c+YMt912G7t370YplW+/UooOHTqwe/duVqxYwSuvvELfvn0JCwvD09OTrVu3FtqXnZ1d7s+DBw/mpZde4uLFi0RERNCnTx9SU1NxcnIiMjKy0PNvjKMw12uWy6Ju3bq5P1taWhZaW92gQQP27t3LypUr+eKLL/jpp59yR86FEEKI0rpxEQ1RDe3/GVLPs7nJqFJ9w5+ZncPljCwa29ctsK86uaVGliuitqlv376kpaWxcOFCAAwGA//6178YO3Ys9eoZn/RcvXo1Fy9eJD09nWXLlhEcHExcXBz16tXjgQceYPLkyezevRt3d3cSExNzk+WsrCwOHjxYaL/29vYEBgby3HPPcffdd2NpaYmDgwNt2rTh559/BowPH+7duxeA4OBgwsPDAfj+++9LdY09e/Zk2bJlpKWlkZqaytKlS+nZs2eJz69fvz5XrlwBICkpiZycHIYNG8abb77J7t27SxWLEEIIIWoQrY3TxTl78eLuhiX+hj8tM5vHFuxkzNfbyczOqaxoy+SWSpbNOv2JiVKKpUuX8vPPP9O+fXs6dOiAjY0Nb7/9du4xQUFBDBs2DB8fH4YNG0ZAQAD79+8nKCgIPz8/ZsyYwSuvvIK1tTWLFy9mypQp+Pr64ufnd9PZJkaOHMl///vffOUZ33//PfPmzcPX1xdPT0+WL18OwMcff8zcuXPx9vYmNrZ0X3l07tyZsWPHEhQURNeuXXn88ccL1CvfzBNPPMHAgQO54447iI2NJSQkBD8/Px544AHeeeedUsUihBBCiBrk2GpIPALdxxOXklHoITd+w5+clsmYb7az+XgSj/Vog3Wd6p2OKq11VceQKyAgQF9/wO26w4cP06lTpxK3ISv9iKpU2r9XIYQQosbSGr7pB1cTYMJugt/fRGwhpa/NnWzZPNW49HXC5QwemreDU0mpzBntz0Avl8qOOpdSKkJrHVDccbdUzTJIbZMQQgghRKU4vgZid8Hds6GOdbGzl0QnpfLAvO1cSs3k20cCCb6tcVVFXiq3XLIshBBCCCEq+Nt2rWH9O+DYKndp6+ttF9bnwbgUHp6/E0NODj+M64ZvSyfzxFEJakSyrLUu0UwPQlSl6lTSJIQQonariLUn8jm+BmIj4J6PoY517ubCvuH/+2giz36/m/o2dQh/oju3Na1f/v4rUfWuqAZsbGy4cOGCJCKiWtNac+HCBWxsbKo6FCGEEKJi157QGta9bRxV9r3/pocu2nmGR7/bScuG9Vj6THCNS5ShBowst2jRgpiYGBITE6s6FCFuysbGhhYtWlR1GEIIIUSFrD2R69hqiNsN98zJN6qcl9aaD1Yd5dN1x+nVoQlz7/envo1V+fuuAtU+WbaysqJNmzZVHYYQQgghRI3h6mRb6MwU5Vl7AjDVKr8NTq1YTi9mzVxboD45MzuHKb/sY+meWEYGtOTNMC+sLKt9MUORqn2yLIQQQgghSqe4mSnK7NgqiNvDHr/XmbosqkBN9JWMLH7dG8fO6Eu8MKADz95xW41/7kySZSGEEEKIW8zNZqYos+szYDi14rnDnUjPysq3Oz3LwPRfD1HHUjFntD+DfV3LcwnVhiTLQgghhBC3ILOvPXHkD4jbA4M/4exPWYUeYtCaZU8F493C0Xz9VrGaW0AihBBCCCEqR3YmrH4NGncA39FF1j67ONjcUokySLIshBBCCCGKs2s+XDwBA94ESysmh7pjUyd/GmlTx4Kpd3asogArjiTLQgghhBCiaGkXjbXKbUOg/QAAOjarn28qOFdHG2YO8zFv2Uc1ITXLQgghhBCiaBveh4wUGPAWKMXSPTFMW7IfBxsrfn6qO4FuDas6wgolybIQQgghhCjchROw4yvwf4BrjTvx5rID/Gfbabq2acgn9/vTtP6tv3KtJMtCCCGEEKJwa/4Nltac8ZvEM59v4UDsZZ7o1ZYXQ92pU4MXGikNSZaFEEIIIURB0Zvh8G8c7jSR4fOOUsfSgq8e7MIAT5eqjqxSSbIshBBCCCHyy8nB8Nc0rtRpQtgef7zdHPh4lH/5l8uugSp8/FwpNVApFaWUOq6UmlrR/QkhhBBCiPKJXfclluf2MiN9BOP6ePLjuG61MlGGCh5ZVkpZAnOB/kAMsFMp9avW+lBF9iuEEEIIIUrPkKNZuGobw7a+ToSFJyPGTuL29k2qOqwqVdEjy0HAca31Sa11JhAO3FvBfQohhBBCiFI6cyGN+77cSpPN07G1yKL9Y/NqfaIMFV+z3Bw4m+d9DNA17wFKqSeAJwBatWpVweEIIYQQQtRuWTlZXMm8kvu6fO0yq46c5ueI03RSp8BhH392Goa6dhyb0zHYWdkZX3XssLe2p7FtY+pY1J7H3qr8SrXWXwFfAQQEBOgqDkcIIYQQokbTWpOUnsSJlBOcSD5B7NVYzqWey30lpSehKZhyWThDFDCNxnBhC2zaUmj7FsqCxraNcanngrOdM652rrRzakeHBh1o69QW2zq3Vm1zRSfLsUDLPO9bmLYJIYQQQohyMuQYiL4czb7EfRxIOsDRS0c5kXKCK5lXco+xsbTBxc4FFzsXgpsH42LnQoO6DTgYk8ny3ZcwZNvwaHd3hqUtRe39AYth87Fw7YxGk56dTmpWqvGVncqVzCucTzufm3gfu3SMDTEbuGa4BoBC0bJ+S9wbutPFuQtdnLvQ3qk9lhaWVXWLyq2ik+WdQHulVBuMSfIo4P4K7lMIIYQQ1cyyPbG8tzKKuOR0XJ1smRzqzhD/5lUdVo1zzXCNvef3sv3cdiLPR3LwwkFSs1IBsLeyx72hO3e1uYu2jm1p69SWto5taWLbBKVUbhsxl9KYtmQ/G48l0b1tR94d5kOra1Hw9X+h88PQsXSPlxlyDMRcjeHYpWPGV/IxDiYdZPXp1QDUt6qPv7M/QS5BhLQMobVDa/PdkEqgtK7Yygel1F3AbMASmK+1fquoYwMCAvSuXbsqNB4hhBBCVK5le2KZtmQ/6VmG3G22Vpa8M9RbEuY8CvtAMdivGYcuHGJz7GZ2nNtB5PlIMnMysVAWdGzYEe/G3vg08cGrsRduDm5YqKLnbjDkaBZsieaDVVEATLurE/cHtcJCZ8M3feHKOXh2B9g6meV6zqWeIyIhgoiECHYl7OJUyikA2jq2pU+rPvRp2QfPxp43jbkiKaUitNYBxR5X0clyaUiyLIQQQtx6gmeuJTY5vcD25k62bJ7apwoiqn7yfaBQmdSxO0ZdhygcGh3javYlADo27EiQSxBdm3Wlc9PO2Fvbl7j9A7EpvLR0P/tiUghxb8KbQ7xo0aCecefat2DDLLhvIXhU3KRlcVfjWHd2HevOrGNXwi4M2kA7x3YsvXdpvpHvylLSZLnKH/ATQgghxK0trpBE+Wbba6NZq/aRZbsbG+e91LE7hrLIRhvqknHFg7f7T6ZH8x40sGlQ6nbTMrP5aPVR5m+OpkE9az4Z7c/dPs3+SU5Pb4WN74Pv/RWaKAO42rsyptMYxnQaQ8q1FDbEbCDlWkqVJMqlIcmyEEIIISqUq5NtoSPLtXVFuOuuGa7x99m/+Sv6Ly43XY+tRRY5WQ5kJQeRfcUDQ5obijrc025QqdvWWrPqUAKv/3aI2OR0Rge1YurAjjjWs/rnoIwUWPIEOLWCu2aZ8cqK51jXkXva3VOpfZaVJMtCCCGEqFCTQ90LrVmeHOpehVFVDa01B5IOsPzEclacWsGVzCs0smmEdXo3UhI9MKS3Ju+acWX5QBGdlMr03w6yPioRd+f6/PRkd4LaNCx44B//gsux8OhKqFu/HFd1a5NkWQghhBAV6vpDfLV5NoyLGRf59fivLDu+jBMpJ6hrWZd+rftxb7t7CXIJ4re954wfKCj7B4r0TAOfrT/Ol3+fxLqOBa/e7cFD3VtjZVnIA3T7foL9P8MdL0PLQHNc4i1LHvATQgghhKgAWmv2Ju5lUdQiVkavJCsnC98mvgy5bQihbqHUt84/mlvW6fW01vy+L56Zfx4hNjmdMP/mTLuzI00dbAo/4VI0fN4DXLxg7B9Qg+dALg95wE8IIYQQogpkZGfwx8k/WBS1iMMXD2NnZcfwDsO5r8N93NbgtiLPG+LfvNSj7ftjUnj994PsjL5Ep2YOfHifL13bNir6BEOWsU5ZKQj7stYmyqUhybIQQgghhBkkpScRfiScn6J+4tK1S7Rv0J5Xu73KoLaDsLOyy3dseRdpOX85g/dWRrF4dwyN7KyZOdSbEQEtsbQoZmaJv6bC2e0wfD40KPniILV5URlJloUQQgghyuFE8gkWHlrI7yd+Jysni94te/OQx0MEOAcUOi3ajYu0xCanM23JfoBiE9DUa9l8teEkX288SZYhhyd6tuXZPrfhYGN10/MAiFgAO7+B2yeC17ASX1954r0VSLIshBBCCFEGB5IO8NW+r1h3dh11Lesy5LYhPOjxIG6Objc9772VUflmBgFIzzLw3sqoIpPPbEMOP+2K4aM1R0m8co27vF14MbQjbo3tCj2+gDPbjbNftOsL/aaX7JxyxHsrkWRZCCGEEKKEtNbsStjF1/u+Zmv8VhysHXja92lGdxxd4kVDSrNIi9aa/x0+z7t/HeHY+at0ad2ALx7oQpfWpVig5HIc/PQgOLWE4fNKXadc2xeVkWRZCCGEEKIYWmt2nNvBZ5Gfsfv8bhrZNOL/uvwfI91HFqhHLk5JF2nZfvICs1ZGEXH6Em0a2/HFA50J9XQp3Yp3WRkQPgYyU+Gh5WBbsiQ7b42yhVIYCpk9rbYsKiPJshBCCCHETUQkRPDpnk/ZlbCLpvWaMi1oGkPbD8WmThFTsxWjuEVaDsal8N7KKNZHJeLsUJe3w7wZEdCi8PmSb0Zr+O05iNsNo36App1KdNqNNcqFJcq1aVEZSZaFEEIIIQqxN3Evn+75lG3x22hs25ipQVMZ3mE4dS3rlqvdohZp8WruyPgfdvP7vngcba2YemdHHu7uhq11Gad3W/Nv2BduXHikY8mXzC6sRhnAUilytJbZMIQQQghRu9TmacEKczL5JB/v/pi1Z9fS0KYhLwS8wEj3kWUeSS5M3jmVo5NSmfO/Y0z6KRIbK0ueCWnHk73b4WhbghkuirJpNmz+GAIfh16TS3VqUbXIOVpzambJk+5bhSTLQgghRC1W26cFy+tc6jk+3/s5y44vw7aOLeP9xvOgx4PUs6pXIf2dvZjGp2uPs3h3DHUsFI/1aMNTvdvRyL58I9fsXmgcVfYaBne+Z1yApBRKWlNdW0iyLIQQQtRitX1aMIArmVf4Zv83fH/4e3J0Dvd3vJ8nfJ4o8ewWpXX2Yhpz1x1ncUQMFkrxYLfWPBPSrujlqUvj0K/GOuV2fWHIF2BRyjpniq+prm0kWRZCCCFqsdo8LVh2Tja/HP2FuZFzuXTtEne3vZvx/uNpbl8xHxLOXEjj03XHWLI7FgsLxZiurXgqpB3NHM00Ynvyb/jlMWjeBUb+B+pYl6mZomqqa8uHpxtJsiyEEELUYrX1K/dNsZt4f+f7nEg5QRfnLkwOnIxnI88K6ev4+at8tv44yyPjqGOheKBba54OaYezOUaScztZA+EPQMN2cP9PYF266exulLemuraTZFkIIYSoxWrbV+6nUk7x7s532Ry7mZb1WzI7ZDZ9WvUp3dzFJXQwLoXP1p1gxYF4bOpYMvZ2N57s1dY85RZ5HfoVFj8KTTvCA0uhXkPztl/LSbIshBBC1GK15Sv3q5lX+XLfl/z30H+xqWPDCwEvcH/H+7GyLMeME0WIOH2Rz9ad4H9HzlO/bh2eCWnHo8Ftyv/gXmEif4Tlz0DzABjzM9g6mb+PWk6SZSGEEKKWu5W/cs/ROfx24jc+iviIixkXCWsfxkT/iTSybWTWfrTWrD+ayOfrT7Dj1EWc6lnxr/4deOh2t/JNAXczO76GFS9Am97GRUfq2ldMP7WcJMtCCCGEuCUduXiEN7e9yd7Evfg09uHTvp/i1djLrH1kG3L488A5Pl9/gkPxl2nmaMOrd3swOqgl9awrKM3SGv6eBevfBvdBMHw+WJm5tEPkkmRZCCGEELeUq5lXmRs5lx+O/IBTXSfeCH6Dwe0GY6FKP41aUdIzDSyOOMvXG09x5mIabZvYMWu4D0P8mmNdx3z9FJCZBsufhYNLwGcU3PspVEApifiHJMtCCCGEuCVorfkr+i/e2/keSelJ3Od+HxP8J+BY19Es7S/bE8vMP49w7nIGFgpyNPi1dOKluzoywMMFCwvzPySYT0oMhN8P8fug3wwIfq7UC46I0pNkWQghhBA13tnLZ3lj2xtsjd+KRyMP5vSZY9aSi282nuSdP49gyNGAMVG2trTg4e6tGejVLN+x5lw+/HpbzVIi+arubOrXycbq/kXQIbTc1yRKRpJlIYQQQtRYWTlZLDi4gC/2fkEdizpMC5rGSPeRWFpYmqX9PWcu8fXGk6zYf67AvkxDDu+vOkpY5xa528y5fLixrX0My1nJa9YLic1pzIPXXmRcqhdDynFNonQkWRZCCCFEjRR5PpIZW2dwPPk4/Vr1Y2rQVJztnMvdriFHs+ZwAt9sPMnO6Es42BSdLt240qE5lw//+q/tfMIc+lntYb3Bl4lZz3IZ+1q1FHl1IMmyEEIIIWqU1KxUZkfMZlHUIprWa8qcO+ZwR6s7yt1uWmY2iyNimL/pFNEX0mjuZMurd3swMrAloR9tKNFKh2ZbPjzqLxZkPEd9i3SmZz3EAsMANBZla0uUiyTLQgghhKgxNsZs5PVtr5OQmsD9ne5ngv8E7KzKt7RzwuUMFm6N5vvtZ0hOy8KvpRNzQzsS6ulMHUtjglrSlQ7LvXx4ZhqsegV2zeOShRv3ZzzNUd2y2LbMWSct8pNkWQghhBDV3qWMS8zaOYvfT/5OW8e2LLxzIX5N/crV5sG4FOZtOsVve+PIztEM8HBmXM+2dGndoMDy1yVd6bDMy4drDUf+gL+mQcoZ6D6ew40e5+zyKCimLXPWSYuCJFkWQgghRLW2Mnolb29/m8vXLvOU71OM8x6HtaV1mdrKydGsizrPvE2n2HLiAvWsLRnTtTWPBLvRutHNR6hLstJhmZYPv3AC/pwCx1dDUw8YuwLcghkM5FhaF9uWOeukRUGSLAshhBDViHyd/o8L6Rd4a/tbrD69Gs9Gnnw94Gs6NOhQprbSMrP5JSKGbzdHczIplWaONky7syOjglqZfTnqEi8fnpkGm2fDptlgaQ2hb0PQE/kWGSlJW2arkxaFkmRZCCGEqCZq+9fp/3xQSKOJyxFU46Vk6XSe7/w8D3s+TB2L0qct8SnpLNhymh93nCElPQvfFo7MGe3PnV4uWFlW4Ep7N5OVARHfwsYPIfU8eI+A/m+AQ7Pizy1EueukxU1JsiyEEEJUE9Xp6/TKHuG+/kEhIyeZus2Xke5wEJ3akok+7/GY9+2lbi/ybDLzNp1ixf54tNaEerrwWI82hdYjV5rsTNizEDZ8AFfiwK0njPwPtOpWrmbLXCctSkSSZSGEEKKaqC5fp1fFCPd7K6PIsomknssylMU1riXcSebFHiy4lMET3UvWhiFHs+rgOeZtOsWu05ewr1uHsbe7MfZ2N1o2rFchcZdIxmWI/B62fmZ8eK9lVwj7Atr2NkvzZaqTFiUmybIQQghRTVSXr9Mre4Q75VoKF+rNx9ZxL4b0FqTHjSAn07i4SEk+KFzOyOKnnWf5bks0MZfSadnQltfu9mBEQAvq25i3HrlUko7Djq+MiXLmVWOSfPdHcFtfMPPodonrpEWpSbIshBBCVBPV5ev0yhzh3hCzgelbpmPlcIFrif3JTAoB/lmq+mYfFM5cSOPbLaf4eVcMV69lE+TWkFcGedDfwxlLiyoqtcjKgKN/GRPkY6vAwgq8hkHXJ6B5l6qJSZSLJMtCCCFENVFdvk6vjBHutKw0Zu2cxS/HfqF9g/YMb/FvPjmRDtz8g4LWml2nLzFv4ylWHTqHhVLc7dOMx3q0xbuFo9niK5WcHDizBfYtgoPL4VoK2LtAyDTo8gjUL/8S3KLqSLIshBBCVCPV4ev0ih7h3nN+Dy9tfInYq7E86vUoz/o9i7WlNa62RT9UmGXIYcX+eOZtOsW+mBSc6lnxVO92PNTdDRdHG7PEVSrZ1yB6I0T9BVF/wuUYsLIDj8HgMxLa9AILy+LbEdWeJMtCCCFEJahJ8ycXNcINEDxzbZmvIcuQxWd7P2P+gfk0s2vGdwO/o7Nz53z93theSloWP+48w4It0cSnZNC2sR1vDvFiWOcW2FpXcjJ66TSc2gDHVsKJdcY65Dq20O4O6DcdOt4F1uVbeltUP0prXdUx5AoICNC7du2q6jCEEEIIs7pxdgkwjtS+M9S7zAlzVU3tVtZrOH7pONM2TePIxSMMbT+UFwNfxM6q6MTy9IVUvt0czU+7zpKWaeD2do14vGcbQjo0xaIy6pG1hosn4cxWiN5kfKWcNe6r7wruA6HDQOMIspXMZ1wTKaUitNYBxR0nI8tCCCFEBTP37BJVNbVbWa5Ba80PR37go4iPsLOy4+M7PqZPqz5FHhtx+hJfbzzJqkMJ1LFQ3OPrymM92uDpWoH1yFobE+H4vRC7G+L2GF8Zycb99RqDWzAEPwetg6Fpp1LNZlGTvlUQBUmyLIQQQlQwc88uURWLlxQVa2xyepGlGYlpiby6+VU2x22mV4tezLh9Bo1tGxdoI9uQw8qDCXy98SSRZ5NxtLXi6d7tePh2N5wdzFiPbMg2znN84SQkRcH5w8ZX4hFjSQWARR1o6gGeQ8DVH1p2gybuZZ7qrbavyngrkGRZCCGEqGDmnl2iKhYvKeoaFORuz5sI1m94hOlbp5ORncErXV/hPvf7Cqyct2jHGd7+8wgp6VkANLa35vV7PRnepQX1rMuQouTkQFoSJJ81JsXJZ4w/J5+GCyeM/+Zk/3N8vUbGxNjvfuNosbM3uHiZtayiOq3KKMpGkmUhhBCigpl7domqWLyksGtQwI1PPqVnp/P6thlk1dtKp4admNlrJm0d2+Y75vzlDF5aup81h8/n2341IxsHG6vCE+VrV+FqAlyJhyvnTK944+tyHFyONW4zZOY/z8YRnFqBizd43AuN2kHDttCoPdg3AfKWSZzH1WmrWcskqsuqjKLsJFkWQgghKpi550+uisVLCruGGxN2C5sYbF3DybS+wGNejzLebzxWlv+soHcs4QpfbTjJ8sg4Mg05gKY+6TRVl2iqknHOuUTsHysgwd6Y+F5N+Off62USedWxAQdX4wN3LbuBQzNwaG5Mjh1bglNLY7J8ExVdJlFdVmUUZSezYQghhBA1UHV4aCx45lpTIpiDdcONWDddhc62w/7yQ2yf9CRkpqKTzxIVdZjtkZFcToimhcVFfBzS4HIszuoS9iqjYMNW9aC+i3Fhj/rO//xbvxnYm/6t7ww2TuVeNvqfa8ivuZMtm6cW/iBiaVTETCjCPEo6G4Yky0IIIYQok2V7Ynln+V/QdBFpdvG4XXXigcR69G+QRoNrsai0C/mOz8ESXd8FS8fmrI2z5FSmE+d0AxJ0A87TgPPaCUsHV1ZPG1Rp19Bm6h8FSknAWGJyaqZ54qgOH2xEQTJ1nBBCCFHLmS1J0xqunofzByHhICRGQdIxnK6cwLqVNdeUYnriJe69GsM122bEZbqw6oovZ3KaEKeacNttHXn87l7YNGyRu6rd5T2xvL9kP+nZN4y4DvQ21+WXSGWUSVSHVRlF2UmyLIQQQtyCylyLe30xjutzDZ/bBwmHjLNMmFyza8oHTRrzYwNbOtZtxLudHqNp0yDmH9F8+vdpLmdk52vS9oQlLc7WYUjjf1bcM3cdd1lVRf23qFmkDEMIIYS4BZW4Fjc9GWJ2wpltxn/jIyEjxbivjo1xajVnT3D2AmdPTtjaMXnHmxy7dIwHOj3AAx2e4fttsfxn22muZGRTt44F17Jziu+3GpEyidpJyjCEEEIIM6tJSVVRU5NlJJ/jtbdm0CF9L92tjtFWn0WhQVka5xj2HGpcjMPV3zj3sGk2C601vxz7hXc3v0s9q3rM6PohkVHN6PPrJjINOQz0dOHpkHbc++nmUsVTHUiZhLgZSZaFEEKIEqhpK7Fdr8W1JYPuFocItjjI7RYH6GRxFrLgiqUtu7Pb84fqRuceA+kRMhCs7Qpt63LmZf6fvTsPi6psHzj+PQMDwyKrbIILIIq4K4r70qaZ2mKWaZa9pi1ve5lb+6KmLa9tZmX2s7JMTcsyLbM0NVdUXBAQReYpV3IAACAASURBVEVlk30ZGGbO749RcgEZYIBB7s91zYUyZ3mOMpz7PM/93M+r217ltxO/0dm7B+4F9/H8kkIU5SS3dw3koYGhhPq4XnLe8tojREMkwbIQQghRjst7kQtLShvOSmyZx1kQtoucmF/oySEclVL0qpbdpjbMNdzNVlN7DqrBGDHnEAdGO7H1pvID5f3p+5m6eSopBSm0UkazdWtXdPZ67uvdikkDgglwvzQIlhxgca2RYFkIIYS4THm9yBWxifQCVTVPxDv8I8T+DBlxdALyXFvxg34YPxd14EyTzhzPMZa7e3nXYFJNLD64mPejP8Be9SQ36SGS1BAeG9yKB/oG4+XiUO6xbGXinhDWIsGyEEIIcZl56+Ou6EWuSL2lF6gqnIk2B8iHf4SsJHPecau+EPkAhN1EE+9Q7gHuOb9LRZP+Lr+GjMIMHv19CrHZuzHkdkLJvotnB7ZnfO+WuOm0V+x/OckBFtcSCZaFEEKIy1jaW1wv6QWZxyHme4hZBpmJoLGHkEHQ/1loewu4eFe4a2UpEqqq8vnu9Xx06HVK1UK0OXfxRLexjOvVEmcHCRlE4yQ/+UIIIcRlKpqk5uGkxcXRvu7TC/Q5cHAl7P8OTu0AFGjVD/o9De2Gg5OnRYepKEXi1i7N2HDkLK/+/S5Z2vXYlfoxIXQWj/Xrh05rV8lRhbi2SZ1lIYQQ4jKX5yzD+dXl7uhYd+kFqgqn98CexXDwBzAUgk84dLobOt0F7kFWOIXKX3HpvP3Hdo5pPsXe+QSd3G/i46Gv4q5ztcJFiNrQkEoY2jKpsyyEEEJYqLzgY/YdHesnICnOM/cg7/kSUg+C1gU6jobu90OzbqAoNT7FhSD5fxviOZj9Dy6BK3C2V3mp12xuDRte82sQtaahlTC8FkjPshBCiEbNJnqRwZyLvPMz2PsVFOdCQBfoPgE63gmOTaxyClVV2RSfznu/x7M/+RzeLTZQ4vIX4Z7hvD3obVq6tbTKeUTtsXhlRlEp6VkWQgghLFBe5Ys6q5+sqpC0BbYvgLi1oLGD9rdD1CMQ1N2Kp1HZevQc7/4eR/TJbAK8C2jT9TvO6hMYGz6WZyOfxcGu/FJwwrZUNPnUJkoYXqMkWBZCCNGo1UvwYTKZg+Mt75rzkp29zdUsekwEt2ZWPdWOY+d45/d4dh7PJMBdx/jrs/k9/UPyTQrvDXqPG1reYNXz1ZfGkscrKyTWPQmWhRBCNGo1DT6qFKQZDXBgOWz5H2TEgWcruOVd6DIWtNYNdvadyuad3+L4OyED3yaOvDg8jGTNMlYkfE/Hph2ZN3Aega7XRjDZmPJ4ZYXEuifBshBCiEatJsGHxUGa0QD7lsLmtyHnJPh1gFGLIOI2sLPurTj2bC7v/BbPhthUvFwcmDmsHQPaw4vbphGbGcv9EffzZLcn0dpVvrhIQ1GvqTR1TFZIrHsSLAshhGjUahJ8VBqkGUsh5jvYNBeyT0Bgd7jlHQi70SpVLS6WlFHAu7/HsybmDK6O9jx7Yxse6BfM32d+5751r6C10/LBdR8wqPkgq57XFlSUMnM6u4i+czZec0GlrJBYtyRYFkII0ehVN/ioKEg7m10A+5fBpjmQecxc2WLYPAi7yepBckqOnvc3JvD9rlNo7TQ8MjCUhwaE4uhg5K1db7IifgWdfTozb8A8AlwDrHpuW1FRKo0CZd+/llMzRO2SYFkIIYSopvKCtP6aGF50XAarjoNfRxjzLbS92epBcnZhCQv+SuTLbUmYVJWxUS147LrW+DbRcTznOM9teI74rHge6PAAj3d9HK3m2km7uFx5qTQKcHlx3Gs1NUPULgmWhRBCiGq6OEhrryQxzX4p/e0OUuAUCEM/hw6jQKOx6jmLSows3nacBX8lkl9cyu1dAnn6xjY093IG4OdjP/PaP6/haOfIR9d/xICgAVY9vy0qL5WmvJ5mkBJroupqFCwrijIPGAGUAInAA6qqZp9/bzowETACT6iqur6GbRVCCCFqzJolxm7rGoiuKBV1w6vcbPqLbJpwoMM0Ot72DNg7WrXdpUYTK/Yk896GeFJzi7ku3Jfnh7Yl3N8NAH2pnjk757AyYSVdfbsyd8Bc/F38rdoGW3Z5Kk1Fi3dIiTVRVTXtWf4dmK6qaqmiKG8B04GpiqJEAGOA9kAzYIOiKG1UVTVe5VhCCCFErbJqiTGDHv75kKF/vwuKAfo+hUf/Z/DQuVu1zaqq8vvhVN5ad4TE9AK6tvDg/TFdiQrxLtvmWM4xntv0HAlZCUzsMJHHuj6GvaZxDx5LiTVhLTX6JKmq+ttFf90O3Hn+z7cC36mqWgwcVxTlKNAT+Kcm5xNCCCFqwiolxlQVYtfAbzMh+ySED4eb3gCvYKu3N/pkFrPXxrIrKYsQHxcWju/OTRF+KBflP19Iu9DZ6VhwwwL6BfazejsaIimxJqzFmo+d/wGWnf9zIObg+YLk898TQggh6k2NV+s7lwi/PAvH/gTfCLjvRwgZZLX2XZCUUcDc9UdYeyCFpq6OvHl7B+6ObI693b/5zxenXXTz7cbcAXPxc/GzelsaMimxJqyh0mBZUZQNQHlJTzNVVf3x/DYzgVLgm6o2QFGUycBkgBYtWlR1dyGEEMJi1V6tz6A3L0295T2w18HNcyFyotUXFMkqKGH+Hwl8vf0EDvYanrohjEn9Q3BxvPQ8F6ddTOo4iUe7PNro0y7qS2NZZrsxq/STparqVReNVxRlAjAcuF5V1QtVWk4DzS/aLOj898o7/qfApwCRkZGXV3kRQgghrKZaeaxH/4C1z5nrJXe4E4a8CU2sO3GuuNTIkm0n+GBjAvnFpdzdowVP3xiGbxMdcGlA1jTgICavFbhonfjkhk/oG9jXqm0RlmtMy2w3ZjWthjEUeB4YqKpq4UVv/QQsVRTlXcwT/MKAnTU5lxBCCFFTVcpjLciAddPgwHLwbg3jV0PoYKu2R1VVfjlwlrfWHeFUZhGD2vowY1g72vg1KdumLCArLcIx4Cf0HrtRC0J4sttr9A3sbNX2iKppTMtsN2Y1HbP5EHAEfj8/2WC7qqoPq6p6SFGU74HDmNMz/iuVMIQQQtiCSvNYVRUOroRfnwd9LgycBv2tXwpu36lsXv/5MHtOZBHu34SvJvakf5jPFdvNWx9HsXIW51bfoHFMozjjOkrSr2dh9jnu6/nvdpIOUPdqnAMvGoSaVsNofZX33gTerMnxhRBCiDqVcxp+eQbi10FgJNz6Ifi2q/Jhrha4ns0pYu66OFbtPU1TV0fm3NGR0ZHNsdOUv8JfmroF5+DVqCYHik49gLGgDXBpQCbpAPWj2jnwokGR2QBCCCGEqkL0Elg/E1QjDJkNUQ+Bxq7Kh6oocC0pNXE6u4iFmxMxqfDooFAeHdwaV8fyb8WFhkJm7ZiFrtmPlBYEoz9zD2qpW9n7Fwdkkg5QP6SWc+MgwbIQQojGLfcsrHkCEn6DVv1h5Ac1qplcUeA6/YcDGFWVWzoFMG1oeNny1OU5mnWU5zY9x7GcYwzyHcuGbZ1QS/99//KATNIB6ofUcm4cJFgWQgjROF3ITf7lWSgtNpeD6zEJNJrK972KigJUo6ry/UO96RnsdZUmqaw+uppZO2bhrHVm4Y0L6d2sN6v9r56PLOkA9UdqOV/7JFgWQgjR+BRmws9Pw+HVENQDbvsEmlY4DadKKgxc3XVXDZQLDYW8sf0N1hxbQ0//nszpPwcfZ/OEv8oCMkkHEKL2SLAshBCicTn2F6x62Fwa7vqXoc8TVltcpKTURGRLzyuCZSetHc8PDa9wv7jMOJ7b9Bwn807yaJdHmdxxMnZVyJeWdAAhao8Ey0IIIRqH0hLY+Dps+wCahsHYZRBgvTrFm+PTeXXNIRLTC4gIcCMjv5j0vOKrBq6qqrIiYQVv7XyLJg5N+Pymz+nh36Na55d0AOuSUnziAgmWhRBCXPvS42HlREiJgcj/wE1vgkPFE+yq4lRmIa//fJjfDqfSytuZLyZEcl24X6X75Zfk89o/r/Fr0q/0adaHWf1m4e3kbZU2iYpZEgRLKT5xMQmWhRBCXLtUFfZ9A2ungL0OxiyF8Fuscmi9wcgnmxJZ8FcidhqF54e2ZWK/YBztK0+fOHTuEFM2TeFM/hme7PYk/+nwHzRKzSYWispZGgRLKT5xMQmWhRBC1Ik6H9YuzjNXuohZBsED4PZPwS2gxodVVZUNsWm89vMhTmUWMbxTADNvaUeAe+WVJ1RVZemRpbyz+x28dF4sHrqYrr5da9wmYRlLg2ApxScuJsGyEEKIWlfnw9opB2D5BMg8BoNnQv9nq7XAyOVOnCvglZ8O8WdcOmG+riydFEWf0KYW7ZtTnMPL217mj5N/MChoEK/3fR0PnUeN2yQsZ2kQLKX4xMUkWBZCCFHr6mxYW1Vh9xewbjo4e8H9a6BVvxofVm8wsuCvRBZsSkSrUXjhlnbc36cVWjvLUif2pe3j+c3Pk16UzpTIKYyPGI+ilL+8tag9lgbBUopPXEyCZSGEELWuNoa1L0/rmHZ9c0acfAsOLIfWN8DtC8HFsl7fq/nzSBov/3SIk5mFjOzcjJm3tMPPTWfRvibVxBcHv+DDvR8S4BLA1zd/Tfum7WvcJlE9lgbBUopPXEyCZSGEELXO2sPal6d16HKO0u7nx1GVMyjXvQD9nq3xSnyns4t49adD/HY4lVAfF5Y+GEWf1pYH3xlFGcz4ewb/nP2Hoa2G8lLvl2ji0KRGbRI1U5UgWErxiQskWBZCCFHrrD2sfXFaxzDNduZqP0WPA09qX+b9AU/VqK0Go4lFW44zf0MCKirPD23Lg/1CcLC3PPjedmYbM/6eQYGhgFd6v8IdYXdI2oWNkCBYVJUEy0IIIWqdtYe1z2QXYU8p0+2/ZaL9r+wxhfHfkidILfbm/Rq0c+fxTF5YfYD41HxuaOfHKyMjCPK0vB6zwWTgw70f8sXBL2jt0ZrPbvqMMM+wGrRICFHfJFgWQghRJ6zZoxfhXsyLRXPppYllcekQZpWOw4A9gdVM68gsKGH22liW70km0MOJz+6L5MaIyhcWuVhyXjJTN08lJiOG0W1GM6XHFJzspXqCEA2dBMtCCCEaltPRLFemo1HO8VTJo6w2matdVCetQ1VVVuxJZtbaWPL0pTw0MIQnrw/D2aFqt8d1Set4ddurKCi8PfBthrQaUqX9hRC2S4JlIYQQDcfeb+Dnp3F29eXPqG/ZtcMOpZppHUfT8pix6iA7j2fSvaUnb97egXB/tyo1p9BQyNxdc1mZsJJOPp2YO2Auga6SDyvEtUSCZSGEELbPaID1M2HnQvNqfHcuZrBLU7YOrvqh9AYjH/95lAWbEnF2sGfOHR25K7I5Gk3VJuDFZcYxZfMUknKSmNRxEo90eQStRlv1Bokqq/PVIEWjJsGyEEII21aYCcvvh+Obodd/4cbXwK56t69tRzOYufogxzMKuL1rIDNvaUdTV8cqHePiJas9HD347KbPiAqIqlZ7RNXV+WqQotGTYFkIIYTtSjsC346B3NNw2wLoMvaqm1fU45hZUMKbv8SyMjqZlt7OfD0xin5hVV+wJFOfyUtbX2JT8iYGBg3k9b6v46nzrO7ViWqos9UghThPgmUhhBC2KX49rJgIWieY8As073nVzcvrcZy2MoZdSZmsPXCWPH0p/x0cyuPXhaHT2lW5OdvObGPmlpnkFucyrec0xoaPldrJ9aA2VoMU4mokWBZCCGFbVBW2vQ+/vwwBnWDMUlYfU5j3zcar5qiW1+OoLzXxzY6TdG3hwew7OlZ5Ah+AwWjg/b3v8+WhLwl1D+WTGz6hrVf1FlMRNWft1SCFqIwEy0IIIWxHaQn8/DTs+xoiboPbFrD6UJZFOapX61lc8XAf7Ko4gQ/geM5xpm6eSmxmLHe3vZtnI5+V2sn1zNqrQQpRGQmWhRBC2IbCTFg2Hk5sgYFTYeA00GiYt367RTmqFfU4Bno4VTlQVlWVlQkrmbtrLg52DswfPJ/rWlxXvesSVmXt1SCFqIwEy8ImSBkgIRq5jARYehfkJMMdn0Gnu8resiRHtajESBs/1yuC5er0OGbps3hl2ytsPLWRXgG9eLPfm/g6+1bpGKJ2WXM1SCEqI8GyqHdSBkiIRu74Zlh2L2i0cP8aaNHrkrcry1HdejSD6T8c4GRmIb1DvEnKKCAlV1+tB+9tZ7bxwpYXyC7O5rnI5xgfMR6NoqnZ9QkhGjQJlkW9kzJAQjRi+5bCT4+Dd2sYuww8W12xSUU5qv8dHMqU5ftZvieZVt7OLJ0URZ/QqpeDAyg2FjM/ej5fHf6KEPcQPr7hY8K9wqt7VRaTUTUhbJ8Ey6LeSRkgIRq2agV8qgp/zYZNb7FL04mJpx6nycJjTBmivWLf8nJUb2rvx3sbEsgsKOHhgaE8dUP1ysGBeSW+aX9P42j2Uca0HcMzkc/UySQ+GVUTomGQYFnUOykDJETDVa2Ar7TE3Jsc8x0/mAYyVT8RA/bkXmXfCzmqaXl6Xv7xEIu3JtG+mRuLJ/SgQ6B7tdpuUk18dfgr5kfPx83BjY+v/5j+Qf2rdazqkFE1IRoGCZZFvZMyQELUPWsN/1c54CvKMle8SPqbT+3HMiv/FkCpdF9VVVmxJ5nXfz6MvtTE1KHhPNg/GK1d9fKJUwpSeGHrC+w4u4PBzQfzSp9X8NJ5VetY1SWjakI0DBIsi3onZYCEqFvWHP6vUsCXfRK+vhMyj8HtnzL7W1eL9j2VWciMVQf4OyGDHq08mTOqE6E+5e9ribXH1vLGjjcoNZXycu+XGRU2ql5W4pNRNSEaBgmWhU2QMkBC1B1rDv9bHPCdjYFvRoOhCMavguD+NPt141X3NZlUvtp+grfWHUEBXr+1PeOiWqKpxuIiADnFObyx/Q3WJa2js09nZvWbRQu3FtU6ljXIqJoQDYMEy8JmySxxIWqHNYf/LQr4jv4B398HOg+YuB5821W6b2J6PlNXxLD7RBb9w5oy+46OBHk6V7l9F2w7vY0Xt75Ipj6TJ7o+wQMdHsBeU7+3QBlVE6JhaPTBstGk8tqaQ4T5NeHeXi3ruzniPJklLkTtqcnwf3kPsbPv6FhxwHehNJxPOIxbDm7Nyo5VXrD4zI1tSMnV8/z8GHT2Gubd2Yk7uwdVO02i0FDIu3veZVncMkLdQ/ng+g+I8I6o1rFqg4yqCWH7Gn2wrKoqyVlFLNl+Am8XB27uGFDfTRLILHEhalN1h/8reoidfUdHtk67bCloVYW/34aNb0DIILjrK9C5XXHMi4PFuJQ8pqzYT0xyDjdF+PHGbR3wddNV+zr3pe1j5paZnMo7xfiI8TzR9Ql09tU/nhCicWr0wbK9nYYPx3Zj3OfbefK7fXi6ONArxLu+m9XoySxxIWpPdYf/LX6INRlh7RTYvQg63Q0jPwR7hwqPazCaWPBXIh9sTKCJTsuHY7tyS8eAavcmlxhL+GjfR3x56EsCXAJYNGQRPfx7VOtYQgjR6INlACcHO76Y0IM7P/mHSf+3m2UP9Sai2ZU9IKLuyCxxIWpXdYb/LXqINejhhwchdg30fRKufwU0FZd3O3g6hykrYog9m8vIzs14eUQE3q6OVWrXxWLPxTJz60wSshIYFTaKKT2m4KJ1qfbxhBCi0QfLF+ff+bo5YqdRuH/xTn54pA/Nvao/mUTUjMwSF8L2VPoQW5QF346Fk9tgyGzo/WiFxyouNfLhxqMs+CsRTxcHFo7vzpD2/tVum8Fk4POYz/k05lM8dZ58dP1HDAgaUO3jCSHEBY06WL48/y41txhHew0lRhP3fbGTFQ/3rlEPh6g+mSUuhO256kNszmn45k7ISIBRi6DjnRUeZ/+pbKas2E98aj53dAvkpeEReDhXnKZRmbjMOF7Y+gJHMo8wPGQ403pOw92xeqv6CSHE5RRVVeu7DWUiIyPV3bt319n5+s4pv8ZnU1cH8otLCfNtwtJJUTTRaeusTUIIYcvKLekYVABf3Q76HBjztXlCXzn0BiPz/0hg4aZEfJo4MvuOjlwX7lftthhMBr448AWfxHyCm4MbL/V+ietbXG9Zm+XBW4hGT1GUPaqqRla2XaPuWa4o/+5cfgmLJkQyeckeJi/Zw+IHeqDT2tVx64QQwvZckeucvBu+GA0aO3jgFwjoXO5+e09mMWVFDEfT8hndPYgXhkfg7lT9joi4zDhe3PoisZmxDG01lBlRM/DUeV6xnZShFELUVMWzLhqBiiaLNfNw4rpwP94e3Zl/jp3jiW/3Umo01XHrhBDCxh3dAP83wlwS7j/ryw2U9QYjs3+NZdSCbRQUl/LlAz2YN7pztQNlg9HAgn0LGPPzGFILU3lv0HvMGziv3EAZrl7BQwghLNGoe5Yrm0R2W9dAsgtLeGXNYWasOsBbozpVu5RRZWSYUAjRoMQsh9UPg087uHclNLkynSL6ZBZTlu8nMb2Ae3q2YMaw8BqltcWei+WFrS8QnxXPsOBhTO85HQ+dx1X3kTKUQoiaatTBsiWTyCb0DSar0MD8PxLwdHZg+rB2Vm+HDBMKIRqUHQvh1+ehZT+4ZynoLp1MpzcYeff3eD7/+xgB7k58NbEn/cN8qn26YmMxn+z/hMUHF+Ol8+L9we8zuMVgi/aVMpRCiJpq1MEyWFZr9KkbwsguLGHh5mO4O2t5dFBrq7ahPlerkx5tIYTFVBX+nAWb50L4cHPVC+2lK+LtOZHFlBX7OZZewNioFswY1g5Xx+rfaqJTo3l528sk5SZxW+vbeC7yuSpVupAylEKImmr0wbIlFEXh5RHtySkyMHddHE0c7Rnfu5XVjl9fw4TSoy2EsNjFq/J1vReGzwe7f28heoORd36L4/Mtx2nm7sTXE6PoF9a02qcrMBTwvz3/47u47wh0DWThjQvp06xPlY8jZSiFEDUlwbKFNBqFeaM7U1Bi5MUfD+HiaM8d3YKscuz6Giaszx5tIYTtqXCkqbQEVj0Eh34wr8p3w6tw0fyNPScymbI8hmMZ1ulN3py8mTe2v0FKQQr3truXx7s+jrO2+otEVWe1QiGEuECC5SrQ2mn44J6uTPy/XUxZEYOzgz1DO1R/xakL6muYUCa+CCEuqGikya60kBFHpkLiH3Dja+Zg+Ty9wcjb6+NYtNU6vckZRRnM3TmXX5N+pbVHa5bcvIQuvl1qfG1CCFETEixXkU5rx6fjIxm/aAdPfLuXRRMiazRxBepvmFAmvghx7anuPITyRpocDDm0WjsOSICRH0K38WXvXdybPC6qBdNr0Jusqiqrj67m7d1vU1RaxH+7/JeJHSaitZMFoYQQ9a9Rr+BXEzmFBsZ8tp2kjAKWTOxJj1Ze9d2kKru8JwnMPdqz7+goQ5ZCNEA1+UwHT/uFi+8GPmTxlcMcgpWzOI75EtqNAK7sTZ57Zyf6tq5+b3JSThKvb3+dnSk76ebbjZf7vEyIe0i1jyeEEJaydAU/CZZrICO/mLsW/kNabjFfPxhFl+ZXr/dpi6QahhDXjr5zNpY7WhTo4cTWaddZvG8LJZWvtbPwUvJ4RnmeQ7qunMkuwtvVAUVRSM8rrnFucrGxmEUHFvH5gc/R2et4uvvTjAobhUZp1GtlCSHqkATLdSQlR89dC/8hu7CEpZN60SHQ8pJGQghRE5c/7JYXKAMowPE5t1R6rOk/HKBF6XG+cpiDPaU8WDqNA4RiMP57n1CARwaF8vzQ8Gq3e+fZnby+/XWScpMYFjyMKT2m0NSp+r3TQghRHZYGy/IIX0P+7jqWToqiiU7L+EU7iEvJq+8mCSEagQvB7ensIlTME/IqWl/UknkIt3UN5JNBJpY7voERDY87zuK4Y9tLAmUAFfhx35lqtTmjKIMZf89g4m8TMapGFt6wkLcGvCWBshDCpkmwbAVBns5882AUDvYaxn2+g8T0/PpukhDiGlfehDwVrgiYLa6sk7iRgdsn4ublS8BTf7Foyr1kFRrK3bSqFXOMJiPLjixj5OqR/Jr0K5M6TuKHkT/QJ7DqdZOFEKKuSTUMK2nV1IVvHuzFmE//Yexn2/n+od609Hap72ZdQvKThWgYLPmsVhSwqphzlKv0OT/8I6yYCD5t4d4f2H1Oy/OL/q5wc0t6qi9cQ4o+AdfAHzE5nCLKP4qZvWYS7B5c6f5CCGErJFi2ota+rnz9YBT3fLqdMZ9u57vJvWwmYJbV+oRoGCz9rFaUo2zJZL5LRH8Fa56AoB4Ujf6OdzallFW6eHRQKIu3JlW5BvzqvaeZvnoHJs9fcfLfQanRFTVlLLd0HE+wu3UWcxJCiLoiaRhWduRsHlo7DWdz9Fz39iY+//tYfTcJuPpqfdei1XtP03fORoKn/ULfORtZvfd0fTdJCItY+lmdMqQtTlq7S75X5cWMtn0APz0GIYPZM+ALhn0aw+dbjjO2ZwvWPz2A54eGM/uOjgR6OKFgDsQrK0NnNBl5c/NiNC3eQuuxE0NWbwoSn6UwqxNv/xZveduEEMJGSM9yOSxNV7h8u8HhPqzcc7rsRmdUVd78JRYFmNi/fuuGNqbV+qQXXTRkln5Wa7SYkarCxtfh73coDb+VOU7PsuiL/QR6OLH0wSj6XFQ3uSpLRcekxzBrxyz0HocwFQRTnDoSU3FApdcmhBC2TILly1gaaJW33TfbT3J5IT4VmLX2CDe196e5l3NdXEK5GtNqfVfrmZNgWdi6qnxWqxLIljGZ4NcpsOtz0sLuZszJuziWmcx9vVsydWg4LtWom5xRlMH86PmsProaHycfnLLGk5YSweXTDa/F3zdCiGufpGFcxtIh0IpmopfHqKqM+XQ7pzILrdnUKrHKkG0D0Zh60cW1p1Y/q0YDrJoMuz5nq+84og6OxICGbyf14rVbO1Q5UDYYDSw+uJjhq4bz87GfeaD9A6y5fQ0zBo7DSXvpsa7V3zdCiGuf9CxfxtJAqyqBl4+rI3l6YkAE4gAAIABJREFUA2M+3c7SSVH1MumvRkO2NsSSFJnG1Isurj219lk1FMHyCRC/joXae5l98mYm9AlmypC2VQ6SVVXl79N/M3fXXE7knmBg0ECm9JhCS7eW56/BpXauQQgh6oEEy5exNNCqaDuFS3uYnbR2zLylHa19Xbl30Q7uXmgOmEN8XK3c8spVa8jWhliaIjNlSNtLtgPp1RINi9U/q/pcjEvHoJzcxkuGB/jb5VaWTe5EVIh3lQ+VkJXA27vfZtuZbbRya8WCGxbQL7DfFds19N83QghxgQTLl7E00Kpou1HdA/nzSPq/vSk3hXFbQCYkr2dD52Ns23eA9I8zCfIowqEkG+wczC97R/PL0Q28QsC7tfnVNAw8W4Gdtq7+CWyWpbnI10ovuhBWUZBB/qKR6DKP8LThv/j0Hse6m9ri5GBX+b4XOVd0jo/3fcyKhBW4aF2YEjmFe8LvQSu/m4QQ1zgJli9jaaB11e3S4+H4JvPr961QlAlAU8WOm519OFLQhC3ZXnQL74mHowLGYigtBmMJFGXBkV+gMOPfk9k7QYteEDzA/AroAnYN47/OmguhVCUXWXq1hIC8tCSKFo3ETX+GF51mcN+E/xDZyqtKxyg2FvNN7Dd8FvMZRaVFjGk7hkc6P4KHzqOWWi2EELZFUdWKpqXVvcjISHX37t313YzqMejh8GrY+RmcPn8N7i0guL85wG3RG9yDQGNHYno+Yz/bTkmpia8mRtEh0P3K4xVlwblEyEiAs/vg+GZIO2x+z6EJhA6CDndCmyGgtc1c3MvTJsDc+15ZndaK9J2z0TqLMAjRCGzfuYOWa8fiqhawOuI9Rt9xFzqt5b3JJtXE2uNr+SD6A84UnGFA0ACejXyWEPf6LYMphBDWoijKHlVVIyvdToLlGso+Cbu/gOglUHgOvMOgx0Roe7M5faICSRkFjPt8B7lFBr78Tw+6t7Sgtyc/HZL+NvdYH1kLBWnmtI12I6DjndBqgE31OFs7uK0o+L4i9UVSLkQ1XQtLwmcXlvD5ih+5P/EZtBqFtJHf0KZr/yodY+fZnbyz5x0OnztMO692PBP5DL0CetVSi4UQon5IsFzb9Lmw6S3Y8QmoJmg7DHo8CCGDQFEq2xswT1C79/MdpOTo+ey+SPqFNa18pwtMRnNv84EVEPsTFOeCWxD0nATd7gPnqg211obgab+UW05PAY7PuaVax6xsIRioWe+1aLysPRJSH349cJaVq5fzXuksVEc3dBN/wsEv3OL947PimR89n83JmwlwCeDxro9zS8gtaBSpMiqEuPZIsFxbVBVivoffX4T8NOg2HgY8Dx7Nq3W4tDw99y3aybH0Aj4e140bIvyqfhCDHhLWw67PzQG0vRN0HgNRD4Ov5TdKa6uLtAlJzRDW0pB/ltLy9Lz84yH0h3/lE4f54N4cxwd+tPj30pn8M3y07yPWJK7BVevKg50eZFy7cTjaOdZyy4UQov5YGixLd0FVpByExcPMRf3dAmHSHzDyg2oHygC+TXR8N7kX7QKa8PDXe/hp/5mqH0Srg4hb4f418PBW6DgK9i2Fj6Ng6d1wZl+121cTdbEQiixAIqylIf4sqarKyj3J3PjuZpzifuBzx3dx8G+H4+TfLPq9lKnP5K2dbzF81XDWJ61nQvsJ/DrqV/7T4T8SKAshxHlWSXBVFOVZ4G3AR1XVDEVRFGA+MAwoBCaoqhptjXPVC1U15yX/OhUcm8CI96HreNBY51nDw9mBrx+MYuL/7ebJ7/aSU2RgfK+W5W5baU6lfwe49SO44VXYtQi2fwyfDoTw4TBoGvh3tEqbLVEXJdxkARJhLQ3tZyk5q5AZqw6yOT6dmT5beDBvAUrLfjBmKejcrrpvXkkeSw4v4avDX1FUWsRtrW/jkc6P4O/iX0etF0KIhqPGaRiKojQHPgfCge7ng+VhwOOYg+UoYL6qqlGVHcsm0zAMevjlWdj3NYTdBLcvrLV8YL3ByGNLo9kQm8YzN7bh8etao1yU/1ytnEp9DmxfAP98ZM5rjrgVBr8APm0qbEdDmuR0LeSZCtvQUH6WTCaVJf8kMXd9HKDyTdhmuiZ+bJ43cedi80hTBQoNhSw9spTFBxeTW5LLjS1v5LEujxHiIRUuhBCNT53lLCuKsgJ4HfgRiDwfLC8E/lJV9dvz28QBg1RVPXu1Y9lcsJx9Cr4fD2f2wsCpMHCa1XqTK2Iwmpi6MoYfok8zoU8rXhoegUZjDphrlFNZlAX/fGwOnA2F5omAA6deEfg3lIDhYg0puBe2zdZ/lo6m5TN1ZQx7TmQxMMybD72+p8n+RdD5Hhj5YYXVcIqNxayIX8FnMZ9xTn+O/oH9eazrY0R4R9TxFQghhO2wNFiuURqGoii3AqdVVd2vXFoBIhA4ddHfk89/74pgWVGUycBkgBYtWtSkOdZ1fDMsnwClJTDmWwgfVien1dppePvOzni7OPDZ38fJKixh3p2dcbDX1Cyn0skTrpsJPSfDn2/Czk9h/3cwaLq51N35VbgsXSXPlsgCJMJabPVnqaTUxCebEvlw41GcHe14d1Q7bj85C2X/cuj9GNz4erkP8heC5EUHFpFelE5P/578r+v/6OLbpR6uQgghGqZKg2VFUTYA5SWyzQRmADfVpAGqqn4KfArmnuWaHMtq4n6FZePNy06P+ca85HQd0mgUZgxrh5eLI2+tO0J2oYGPx3WzTk6lqw+M+J+5Z3n9DFg31VxFY9hcCL2uQU5yEuJaFn0yi2krY4hPzWdE52a8NKQlPmsnwdENcP3L0O/pK8pVFhuLWRm/kkUHFpFWlEZ3v+68NeAtevj3qKerEEKIhqvSYFlV1RvK+76iKB2BYOBCr3IQEK0oSk/gNHDxVOyg89+zfXHrzIGyf0cYvwqc6mdJV0VReGRQKF4uWmasOsiYT7fz8MAQZq09ckWKRLWqS/i1h/GrIX49rJ8OX90OHe6kg/swDuRcmfNoq5OchLhWFRSX8vZvcXy5LQl/Nx2L7o/k+pZac4Wb07thxHzoPuGSfYpKi1gZv5LFBxeTVpRGN99uzO4/mx7+PVAsrP8uhBDiUtVOw1BV9QDge+HviqIk8W/O8k/AY4qifId5gl9OZfnKNiF+vTlH2b9DvQbKF6zee5r3/ziK0aRy8HQOZ7KLeOqGMJb8c8I6OZWKAm2HmhdS2fIebHmXH5T1vOlwN/9XMgj1fGVBa5d7E0Jc3cYjqby4+hBncoq4r1dLnhvSlibFabD4VshMhNFfmifrnpdfks+yuGUsObyETH0mPfx78Gb/N4nyj5IgWQghashqi5JcFiwrwIfAUMyl4x5QVbXSmXv1OsEv/jdYNg58I+C+1eYc33pU3kQ7ABcHO5ZMjKJ7y1poX0YC/PIMHN/MIaUNz+gnku/exuYmOQlRVbY+ce+CtDw9r645zC8xZwnzdWXOqI50b+kF6fEUfjECtSiHSSXPcMItkilD2jKonQtLjyzlm9hvyC3JpW9gXyZ3nEw3v271fSlCCGHzZAW/qrCxQBkqrnxhp1Gw1yi8f09XhrSvhZqoF1YoXD/dvKT3wKnQ76myCYBCNDQNocKLyaSybPcpZq+NRV9q4onrWjN5QCgO9hpI3k3xklHkFqtMKJnKIbUVin02zj5bcPTchUEtZnDzwUzuNJkOTTvU96UIIUSDUSfVMK4JpcWw9lnwbWczgTJUPKHOaFLpGOjOw1/vYeawdkzsF2zdYVZFgc53Q+vrYe0U+PMNiP0Jbvu4Thc0EcJabL3CS3xqHjNXHWBXUha9QryYdXtHQnxczW8mbIDvx5NR2oR7SqaRrAWd9/fYu59flbMgktX3zCDUI7T+LqAWNJSRACFE4yDBsr0jG2+cjpdbEO0dXbGV/tOKKl8Eejjx7aRePLt8H2/8EsuxjAJeHdkerZ2V6z+7NIXRi6H97ebUjE8HQf/nYMBzYKeVm5loMGy1woveYOSDjQks3HQMV509c+/sxOjuQf8+/O5fBj8+iurbjhFpwygKWouLaxyqSYshqzcl5/pTUOpxTQbKF48EnM4uYvoPBwDkd4wQol40+mBZVVVmHf6C1MJUnO2d6e7XnaiAKHr696StV1s0Su0uQlKRKUPaljt0PGVIW5wc7Pjwnm7M845jwV+JnMos5KNx3XDT1UKoHzESWvUzL/W9aQ7E/8qGdm8w/fdCuZmJBsEWl7HeHJ/OC6sPcjKzkFHdgpgxLBxvV0fzm6oKW+dj2PAy61p1ZYmnFwanb9GUulKcdiMl2b3A6AKYH56vNbY+EiCEaHwkZxnI0mexK2UXO1N2sjNlJ8dzjgPg5uBGD/8e9PTvSU//noR6hNbpzHJLem+/33WKGasOENzUhS8m9KC5l3PtNSh2Dax5kuLCPGYbxvB/xpvKKmaAhSsJClHHbClnOTVXzxu/xLJm/xlCmrrwxu0d6BPa9N8NTCYyf32GFfErWebtQ5pqIMQ9hE5NRrJiky9FJf9+3mwt79pagqf9Qnl3JQU4PueWum6OEOIaJhP8aiCtMI2dKTvZlbKLHWd3cDrfXCLaW+dtDp4DzMFziyYtbKIs07bEDB7+ag/2dhoWjOtGVIh37Z0sL5U/5t7N9XZ7+dvYgSmGh0jBfD65mQlbVd9pQ6VGE19tP8E7v8VTYjTx6KBQHhkUiqO9Xdk2cWkH+Oa3x/nFkEGJRqFPQG/GRdxLv8B+aBRNvV9DXalocrM8jAshrE2CZStKzks2B84pO9h1dhdpRWkA+Dn70dO/Jz38exAVEEUz12b11sZj6fk8uGQ3J88V8uqt7RkX1bLWztV39h8MyP+FF+y/phQ7phseZK2pl8U3s8Zy0xcCYN+pbGauOsChM7n0D2vK67d2oFVTcxqFwWjgj5N/8F3s1+xJ34+TycQIz/aMGzSHEI+Qem55/bClkQAhxLVNguVaoqoqSblJZWkbu1J2kanPBCDQNbAseO7p3xM/F786bVuu3sAT3+7lr7h07u3VgpdH1MLEP/69mfmWnma+9iO6aBJZaRqMw4i5jOjRxqJ95UYornVZBSXM+y2Ob3eexMfVkZdGRHBLxwAURSG1IJXl8ctZmbCSjKIMAk0Kd+fkcMeA13Dvdn99N73eyQO1EKIuSLBcR1RV5Wj2UXO+89md7E7dTW5JLgAt3VoS6RdZFkD7OPvUenuMJpW5646wcPMxeoV48fG47ni5OFj9PBduZmnZecx0+Yn7jStRvELgzkXQrGuF+8kQq7jWmUwq3+8+xVvrjpCrL+W+3i155sY2ODto2HZmGyviV7ApeRMm1UT/pp25O3E3/Qry0dz9lXk1TSGEEHVCguV6YjQZic+KZ2fKTnan7GZP6h7yDHkAtHJrVRY4R/pH0tSpaSVHq75Ve5OZuvIAPq6OLLi3G52Cannp7qQt8MNkyE+D616APk+A5spebZm8I65lB0/n8MLqg+w7lU2PVp68dmsHvNyKWHV0FT8k/MDZgrN46by4tfWtjNYF0fzHZ0DnBuOWg1/7+m6+EEI0KhIs24gfok8yd+OfZJoO4+KehJ1zEsWmQgCC3YOJ9Is0B89+kVbveY5JzuaRr6NJzy/m9Vvbc3ePFlY9/hUKM2HNk+ZFTEKvg9sXgqvvJZtIz7K4FmUVlPD2+ZQLLxcHpgxpjadPAj8m/siW01swqSZ6BfRidJvRDG4+GG3M97DmCfAJh7Hfg3vtphhIWoMQQlxJgmUbUH5+Ljw21BkX9xPsTNlJdFo0BYYCwNzz3N2vO5H+kUT6ReLvUvPlrDMLSnji271sOZrBmB7NeWVke3Rau8p3rC5VhT2LYd10cHSDOxaaA+fzJGdZXEuMJpWlO0/yzm9x5OlLGREJXv772HByHVnFWfg6+TIidASjwkbR3K25+fPx1xxzzfKQwXDXEnPPci2Sz5wQQpRPgmUbYEkvaqmplLjMOHan7mZXyi6iU6PL0jYCXQOJ9Is0B9B+kQQ1CapWqTqjSeWd3+L4+K9EOgW58/G4bgR51mI9ZoDUQ7D8AciIh35PweCZYGdeNEV6uYQtqOnP4a6kTF7+8RCx6cmEBsej89jHifyjaDVaBjcfzG2tb6NPsz7Yac4/nJYWw4+PwYHvocu9MOJ/ZZ+J2iSjOUIIUT4Jlm1AdfJzL+Q870ndw+5Uc85zdnE2AL5OvnT36053v+508+tGqEdolVYYXH8ohee+349Go/D26M7cGFHL1TpKCmHdVIheAkE9zZP/PGo5FUQIC9Skt/V0dhFvrN3L70l/4Oq9H5MuARUTHZt2ZHjIcIYFD8NDd9kcgcJM+G4cnNwG170I/Z+FOqrRLvMEhBCifBIs2wBr9OiYVBPHc47/Gzyn7Cmr8+zm4EZX36508+tGN99utPduj7aSnqqkjAL+uzSaQ2dyebBfMM8PDcfBvpaX9D6wAtY8ZZ7wd+vH0G547Z5P2AxbHUWozmczq7CAVzasYMOp9SjOsSiaUgJcmjEidDjDQ4YT7B5c/snOJcI3oyEnGW5fAB1GWfNSKiU9y0IIUT4Jlq2sOjf92sgVVFWV5Lxk9qTtYW/aXqJTo0nKTQLA0c6RDk070M23G118u9DFtwtuDlfmQ+oNRmatjWXJPyfo3NyDD+/pWrvLZIM5YFjxHzi7D6IehhtfA3vH2j2nqFe2nCtraW+rvlTPttPb+GL/j+w/txU0xWhxY2jwEEaHj6CzT+erj+4kbYVl40DRwJhvoUWU1a+lMrb8/yCEEPVJgmUrqsnNpi561jKKMtibttf8St1LbGYsRtWIgkKoRyhdfLvQ1bcrXX26XpL3vPbAWaauiEFR4K1Rnbi5Y4BV23WF0mL4/WXYsQACOsOdi8E7tHbPKeqNLfdoXq1tvz/biy2nt7DhxAb+PPUXemMRqtEJV2NXHu5+J/d2GYy9xr7yk0QvgZ+fBq8QGLsMvELqrafdVnv4hRCiPkmwbEXWvunX9o2r0FDIgYwDRKdFsz9tP/vT95NvyAfAW+dNZ5/OdPLpRGefzjTRBDPl+1hiknO4KzKIl0e0x8XRgkCgJmJ/hh8fBZMJRs6v82FpUTdsOVf28gdgxS4fJ/c4IlqfJKlwH8XGYrQ0oSCrHa6l3Xhu4DDu6t4KO40FecYmI/z2Imz/CEKvhzu/ACcP6eEVQggbY2mwXMtR0bXhTDmB8tW+fzWX3zBPZxcx/YcDAFa7YTprnYkKiCIqwDzka1JNJGYnsi99H/vS9rE/fT8bT20EwF6xJ6xlGyJ9m7MqwYN/PjrCB3feRJcWnpe02arBfbvhENAJVkw0p2Yc3wxD54DWqUbXLWxLMw+nch8ym3nU7v+zJT+vt3ZpRqo+ic92r6XQfj92TidBUck2+tPSYTAHE1pg0Afz8IA2TB4QYvkDpD4XVk6EhN/M6UY3vQl25n3nrY+7JFAGKDIYmbc+ToJlIYSwYRIsW8CaN/36uGFqFA1hnmGEeYYxus1oALL0WcSkx7A/fT/70vdxMHczumZFZAP3/v4uLVzDuaVNFAW5zVi80UhRsQ6wYnDv0QIeWAsbX4et8+HULhj9Jfi0qeHVClsxZUjbcntSpwxpW2vnvNrD6JCOXuw8u5O/T//N5uTNnC04C27Q3iuc/oGTKcgM57ttRhKLSrm9ayBThrQlwL0Kn/HMY/DtWHO5xFvehR4TL3nbmg/dQggh6o6kYVjAmsOntjo0bTQZScxJZMfpvSyJ3szpojjsHNNAMbfWVOKFsSgIoz4Ikz4IP8dQtk0dZp2TJ/wOqx4Cgx6Gvwudx1jnuKLe1XWu7KUpUyY0jmexd03A2T0RjVMSBpMBJ3snegX0YkDQAPoG9GVnool56+NIziqif1hTpt0cTvtm7lU7ceKfsHyC+c93LYGQgZW07V+2kMMthBCNkaRhWNGFm7s1bvr1NTRdGTuNHW0829DGsw33tr+Ln/af4cWf9lDACTS6U9g5JWPnfBKtewwAuarCyNWfEOEdQYRXBBHeEYR7hePq4Fr1k4fdCA9vgZUPmoPmY5tg2DxwrMaxRL2pKDCuqxQDVVU5W3AKrUcids7HsHM5isbevDpmid6f/4SPpU+zPkT6R6LVaNmckMGDi49w6EwuEQFufDWxI/3DqrjkvKrC9o/htxfMS1ePWQpe5hJyl/97DA73YeWe03Xa0y6EEKLmpGe5jjWkST5puXoGzP0Tfamp7HuKXT4a3Wk8PVPpHVHI4YzDZXWfAVq6taSdVzvCvcJp59WOtl5t8XbytuyExlLYPBc2zQXv1jB6Mfh3tPZliVpQHz/XqqpyIvcE0WnR7E7Zzc6UnaQWpgJgMjTBWBhKaUEYxoIwmrn6lfXe7jmRxdx1R9hxPJMgTyeevakNt3YORGPJ5L2LGYrM9cNjvoN2I+G2BWUPeBX9e4zqHsifR9KlKoUQQtgAqYZhwxpSGadV0ck8vzIGg/HfnxOdvYY5ozqVtTmjKIPYc7EcPneY2MxYYs/FcqbgTNn2Pk4+tPVqS1vPtmW91y3dW6LVVLCAyvHNsHISFGXBkDehx4N1ttqZqJ66SDEwGA3EZcWxL20f0WnRRKdGc05/DgBPR096+PfAsbQNq7Y5UVTohTm56d+gvV2AG/PWx7EhNpWmro48fl1r7unZonqL8mSfgu/vgzPR5qXc+z9nXnTnPEm5EEII2ydpGDasLoema+r2bua6zLN/jSU1txgAXzcdbfyalG3T1Kkp/YP60z+of9n3copziM+K50jmkbLX9rPbKTWVAqDVaAlxDyHMM4zWHq3Lvga4BKAED4BHtsKqh2Htc3DsLxj5ATh71em1C8tZe/KaOaXiLAczDrI/fT8x6TEcPneYElMJAIGugfRp1qds9cpg9+Cy+uE9vS99GL2vd0v+OJLG09/vw9XBnuduasMDfYOrXyLx2F/mKi6lJea0i/Ar5xrIZD4hhLh2SM/yNaC8nmqwTo71xVRVZU3MWV5bc4isQgOTB4Tw5PVh6LR2Fu1vMBk4nnOc+Kx44jPjSchO4Gj2UVIKUsq2cbZ3JsQ9hBCPEELdQwg9e5jg3V/TTOeN/R2fQau+NboGUTtq0pOqqipnCs5wJPMIhzIOcTjzMIczDpNVnAWAg8aBCO+IsvrgnXw64e/iX2mbTpwrYP6GBFbvO41Oa8eEPq2YPCAED2eH6l2kqsKW98wVXJq2gbu/gaaty91UepaFEML2SRpGI1FebqRWo4DCJakT1swfzS4s4c1fYlm+J5mW3s68MrI9g9v6Vvt4uSW5JGYnkpCVQGJ2IsdyjnEs+9gludD2KrQwGGjlHkxwyI20dA+mhVsLWrq1xFvnXdarKOqHpTnL+SX5HM0+SmJ2InFZccRlxpGQlUCeIQ8AO8WO1h6tifCOoL13e9o3bU9bz7Zo7SpI2SnHiXMFfPTnUVZGn0Zrp3Bf71Y8NCAEb9caLK+uz4HVj8KRn6H9HeaRjqtMQG1IcxOEEKKxkmC5kaioB6s81u7V2nY0gxdWH+RYRgE3RvjRK8SLL7YkWa03O7ckl2PZxziec5ykrHiSEtZyvDCVU1otpRfFxs72zrRwa0HzJs0JahJEkGtQ2Z/9Xfwrzo0WVvXvCEch/l4G7urtSKBvPkm5SRzLPsbR7KNlE/DA/P/WxrMNbb3aln1t69kWnb2uWudPTM/noz+P8uO+M9hrFO7p2YJHB4Xi61a945VJOQDf3w9ZSXDTG9DrEYty6BvS3AQhhGiMJFhuJCqq21ye2qjlXFJqYtGW47y3IZ6Si6pmQC31pO1fRukvz3DWXsuJvg9zwqs5J3NPciLvBKfzTpOcn1yWFw3mBVl8nHxo5tqMAJeAsq9+zn74u/jj5+yHu6O79ExXUYmxhNSCVJLzkzmVd4rk/GSS88yvE7knKCwtLNvW0c6REPcQQj1CzS/3UFp7tCawSSAapRqT6y4Tn5rHR38eZc3+MzjYa7g3qiWTB4TUPEhWVdjzJca1z5NlcuHR4sc57d5Vgl4hhLhGSLDcSNRnz/LFes36g5Rcfd2cM/MY/DAZkndBpzHmmsw6N8C8uEp6Ubo5gMtL5kzBGc7km19nC86SUpCCUb10BUWdnQ4fZx98nHzKvjZ1aoq3kzdeOi+8nbzx1pn/7GBXzXzXBsKkmsgrySOjKIP0onTSC9PLvqYWpnI2/yxnC86WVaG4wF5jT5BrEIGugbR0a0lLt5a0cm9FsFswfi5+VgmKLxd9MouP/0xkQ2wqzg52jO/dkkn9Q2hak3SLC4rzzGXhDq5gi9qJJ4sf4RzmhUoknUIIIa4NUg2jkShvSeGKcpZrc/GD1HICZcDiQL5KvELggXWweZ65LvPJf+COz6BFFHYaO/xd/PF38aeHf48rdjWajGQUZZBSmEJqQSqphamkFKSQXpRORlEGcZlxbCnaQoGhoNxTO9s74+HogbujO546T9wd3Gni0OSSl6vWFRetC85aZ5y1zrjYu+Bk74TOXofOXoeDxqHWerKNJiPFxmKKSovQG/UUGAooNBSSb8inwFBAgaGAnOIccktyyS3OJackh+zibLL0WWTqM8nSZ13xMHHhun2dfWnm2oy2Xm3L/o0DXQNp3qQ5Pk4+2Gksm+hZE6qqsjkhg4//PMqO45l4OGt56oYw7u/dCk8XKz3IlKVdHGeh/Tjm5N+Myr/Bfm0vTy+EEMK2SLDcwFW0umB536vNm3tFKxMCPPHtXp4f2pYgT2frndDOHgZPh9Dr4IdJsHgo9H8WBjwP9hUHTXYaO/xc/PBz8YOrLNZWaCjknP4cmfpMzhWZv2bqM8kuziZbn012cTY5xTkk5yWTb8gntziXUrW04gNeREFBZ69Dq9Fir7G/5KudYoeiKGgUDRpFg3K+VrBJNaGimr+qKqVqKaWmUgxGA6VqKSXGEvRG/SUpKFdjr9jj5uiGm4Mbbo5uNHNtRsemHfHUeeKl88JL54Wvsy8+Tj7sSDDw/oZTHMwuItPDiZvtofvvAAAYA0lEQVSHtOW2znUbKJaUmvg55gyfbj7GkZQ8/N10vHBLO+7p2aL6JeAup6qwYyH8/hI4ecL9PzPnk+xy05ysWQJOcpuFEMK2SRqGsIryZv/r7DUMaOPDpvh0VBXGRrXgv4Nb49PECsPkF9PnwrppsO8b8O8Ety8EvwjrnqMSqqqiN+rJK8kjv8Tci1tYWljWm6s36ikuLUZv1KMv1VNsLMZgMmAwGsxfz78uBMMm1YQJEybVhAYNKKDhfACtKNgr9mjtLg22dXY6HO0d0dmZe7B1djpctC6XvFy1rrg5uuFs72xR73Z9V3XIKTLw7c6TLN56nNTcYsJ8XZnUP4RbuzbD0d6KPdn5aeZqF0d/hzZDYeSH4OpT6yXg6vvfV4j/b+/O46uu73yPv75ZgBCWyA4BJCigCCKKYt23FmsdtTq9Yzt2fHTs2Ol0m+XSq53be2emdXSqd+ntw/bR3lqn0/rQa62irQutW622iiKbyCabJCwJSwgJIev3/vELCJgDITnJSc55PR+P8yD8cjz5kF+xbz/5nM9XymXOLKvHpeqQba2u5/+8sI5fLC6nX34en7twEl+45BSGDkzzlorVT8NTX4WGGrjiv8JHvgw9MBqQzTK1L3h9VS3/8YdNPLa4nLrGFi48dTh/dfFkLp06Mv0jLGt/A0/+TTKn/LFvH3FiZHeHWfcxS1LmGJbV62zcWcf/+u1afrV8K4P6F/BXF0/m1gsmMbQojaG5tgp+/bfJPtyJH4Ebvp/MOKtTUm1b6Y7NKq2tkd+treLBP2zilbVV9MvP49pZY7ntojLOGDc0rV8LgMb98Pw/waIfwqjpcNMD7f5EojvHJHry+ytJOpJhWb3W6u01/I/frOW37+5gcP8C/uKCk/nLC8u6dmjE4WKEZY/As/8FWhrhyv8Gc7+Q013mzga+nuh87q5r5JeLy3nojc1s2rWf0UP6c8vck/n03Inp2WzRni2LkuPUd6+HuX8NV/0zFHZx1Vwn2FmWpMwxLKvXe3drDfe//B7PrNjGgIJ8Pn3eRG6/ZDJjhqYptNRsTdZ/rVsIE+bC9ffDiCnpee0+pCujBN01hhBj5M1Ne3jojc08u2I7jS2tzDn5JG69YBJXzxhDYX76V80B0NwAL98Nr30XhpQm/5uYfGn3fK0OcGZZkjLHsKw+473KWn7w8noWLK0gANeeOZbbLprMzPFp+NF7jLD8UXj269BUD5d/I5llzs+dRTBd7V6mcwyhal8DTywp59G3ynmvspbBAwq46ezxfGbuRKaOHtyp1+ywbcuSbnLluzD7szDvXw/t506Hzn6f3IYhSZlhWFafs2X3fh58bRP/7833qWts4byyYdx2URlXnT6a/Lwuvqlr3w54+u+TWeYxM+FPvgul56Sn8F4u03OxTS2tvLi6kl+8Vc5LayppaY3MnljCp8+dyLWzxjKwX+f/w6VDQbOpHl6+B/7wPSgeAdd9D6bO6+Kf6sN12CGWpL7FsKw+q+ZAE4++uYUHX9tERXU9pSVFfGbuRD41ZzyjBndhRCNGWPUUPPN1qN0B592ebM1IY3exN8rEXGyMkSVbqnlySQW/Xr6NXXWNjBzcnxvPLuVT54zn1FFd7yJ3KKBufCXZkLJnY9JN/ti3kh3KaebssST1PYZl9XnNLa0sXLmDn7++mT9u2EVBXuCj00fzmbkTufCUEeR1ttt8YC+8+G1Y9H9h8Bj4+Hfg9D85tC6ss9rrckLPHg6Tqq6e6nq+V7mPJ5du5cmlW3l/9376F+Rx1emjuemcUi6ZMpKCNM4iHzOgfm02/OabsORncFJZ8pOEbpxNznT3XpJ04gzLyiobqmp5eNH7PLa4nD37mygtKeKTs0v55NmlnDJyUOdetPwt+NXXYMc7cMqV8PF/6/QbANsLpKmOHc/Ej+a7ay42xsjaHbU8s2Ibz76zjbU7askLcOGpI7j+rFLmnTGawQPSvE+7TXsBNdDKp/Jf4TtDH4f6PXDBV+CyO6CwqFtqOMjOsiT1PYZlZaUDTS0sXLmdX75dwavrqmiNMGtCCTfOLuWamWNP/HTAlqakw/zy3cls60f+Bi6ZD/1PbEwgVVhqT18PUC2tkaVbqnlh1Q6eW7mdDVV1hADnTRrGNTPH8vEZYxg1pPvXsB39PT8jbORbhQ9ydt57MOF8+MR9yXx6D3BmWZL6HsOysl5lzQGeXLqVx5dUsGpbDXkB5kwaxjUzxnD1jLEntoKuthKe/2dY+nMYNCaZbZ3xp5DXsbGBVD+Gb09f/NF8bUMzr66r4vlVlby0upJddY3k5wXmliUB+WNnjO7aPHknHAyo/Zr28g8Fv+CW/OfZzRA2n30H51z3xS6P1XSmnkyP3EiSOs6wrJyyZvu+I0YBAGZPLOGq00dz6dSRnDFuSMeOSS5/C56ZD1vfhnGz4WPfZsGesuOGoGzrLLe0RpaXV/Pqup38ft1O3n5/D82tkSEDCrj8tFFc2fZ9TevpiyequYEVT9zHySvvpzju55cF11A875t84rzTM1eTJKnPMCwrZ62vquW5d7bz3DvbWVGxF4CRg/tz2dSRXDZtFOdPHnbs0wJbW2HFo/DCt6CmnBfjOdzVeDPrYxKQ2/vxeldnljPdlWxpjazaVsOPf7+B51Zu50BT66HPzSgdwsVTRnLp1JHMOfmktL5Jr1NihJVPJEdVV2+GU6+Cj/4LjD4js3VJkvoUw7IEVO47wCtrd/LymkpeWVtFzYFmAKaOHsTcsuGcP3k455UNa3/Wuame7//bfG5peoyBNPBoy2V8r/mTbGN4u93hzm7DyMS86979TSwrr2bZlmoWv7+HxZv2sK+h+UPPG1CQxz03ndl7xgk2vZqE5PI3YfSMJCSfemWmq5Ik9UGGZekozS2tLCvfy+sbdvHGxt0s3rSbusYkoI4/qYhZ40uYNWEoZ44vYUbpUAb1L6DsjqcpoYavFjzBn+c/TyTwcMsV/KD5et6455a01NXdmxR21Tawats+Vm+vYeXWGpZtqWbDzrpDn58yahDnlQ3jmRXb2LO/qdvq6JL3X4eX7kr2Jg8ak+zHPuszkJef2bokSX2WYVk6juaWVt7ZWsOijbtYVr6XZVuqKd/zQWidMKyIqpoGDjQnIwnj2MmXCxbwqfzf0Rry6D/383DR38Hg0V2qIx07eltbI9tqDrCxqo6NO2tZX1XH+qpaVm/fR9W+hkPPGzW4P7MmlHBW22Pm+KEMaVvt1it3BZcvTkLy+hegeCRc9Pcw53PdvgpOkpT9OhqWO3/OrNTHFeTnHQqNB+2qbWB5+V6Wl+9lbeU+Fm/aw/aaAwBsZQTfaP4832+5jjsH/oqr3/ghcdEDrBt3Hdum30bx2GmMHNyf4cX9GTSgoMNHdI8rKWq3szyupIgYI/samtld28iuukZ21zWyq7aBbXsPsLW6nq1769lanXzc0PzBnPHAfvmUjSjmkikjOX3sYE4fO4RpYwYz4hiz2seqo0fFCBtegte+CxtehqJhybjFuZ+HfsU9W4skKefZWZaO47G3tvCdhWuo3NfAkAEFzCgdSn5eoLlqPdfXPcYn816hkGaeaz2XHzZfy7J4KgDF/fIZUlTI4AEFFBXmU5iflzwK8pI3/wEtMbKt+gDrKvfRethfxQAU9cunsbmV5tYP/x0NIekSjyspYtzQIsaVDKBsxCDKRhQzeWQxowb379j2j8NkfFdwSzO8uyAJyduXw6DRcP4Xk5B8gnuvJUk6HscwpB7Q2hqpqSqn5fUfMnTlTylorGH70NksHv2nLCm+mL2NUHOgiQNNrTS1JI/GlkhTWxe4ID+QFwJ765uo2FNPY0srAwrzOLO0hOnjhlDUL59hA/sxrLgfwwb1Y3hx8vGowQPoV5D+rRQZ2cpRt4t3nr6fYat+xrhYyeZQyu5Zf83sa78ABSd4yIwkSR1kWJZ6WsM+ePs/YNGPYM+mZMb27L+Acz4HJRMyXV2vsWBJBfc+t5rRNcu5veglrox/pDA28kbrafy4+Rqebz2bAYWF7a7n89APSVK6GJalTGlthfUvwlsPwNrnkmunXAmzboZp10C/gZmtL4Oe/eNSlj37ANfxO6bnbWZfLOKJlov4WctHWRfHH/Hcw7dwZHxERJKUdQzLUm9QvQUW/zssewRqyqHfYJh+HZz5ZzDpotxYfdZQC6ufhuWP0LL+ZfJpZXlrGY+0XMGTLRdQR/tvIDx8C0d3r9eTJOUet2FIvUHJBLjym3D5P8Lm12D5I7DySVj6UDKmMfVqOO1amHxpdq1Dq61KuuprnoH1L0FzPQydyA+ar+OJlgsPnYZ4LIdv4dia4ijxVNclSUoXw7LUE/LyoOzi5HHNfbDmWVj1K1i5AJb8DAoHwilXJEc3l10CwyYnKy/6ipZm2LYUNv4O1i6ELYuACEMnwNmfhTNuhAlzefg7L7fbIQ7Jsw8pKsw/dAIi9KK1dm2cn5ak3GFYlnpaYRHMuDF5NDfCpt8nYwprnoXVv06eM6QUJrWF6/HnwvBTe9fIRlM97FgJm/+Q1L/5j9C4L/ncmDPhsjuS+ewxM48I/fPnTWt39vimc0p5aXVVyvCZ6p87PFD3lKPnpyuq67nz8RUABmZJykLOLEu9RYyw672kO7vx90kI3b8r+VxhMYydBePOgrFnwYgpSfe5qOTYr5mOmvZth90bknC8bVnSQa5cBbEtuA6fkoT6SW2PQSOP+ZKd7cr2lm6u89OSlB18g5/U17W2ws41UPF2ElC3LoXtK5L534MGDodhpyTBedCoZA560CgoHpF8rmBAsqs4v3/ya14+tDRBcwO0NCa/NtZC3U6oq0oe+3fB3vIkIO/eAE37D/t6Iz4I7GNnJV3vIWN7/nuTQb3yWHBJ0gnzDX5SL9XhDmleHow6PXnM/vPkWksz7FqXdKB3rf8g0G56FeoqkwDcVf2HwOAxSQAvuxSGlSUfjzwNhozrW7PU3aC3zU9LkrqXYVnqQV2ed80v+CBAHy3G5GCUuqqkU7x/FzQfaOsgH0i6yK3NkN8v6TIXDEg+7lecdKKLRyad48IB6fwjZ53eND8tSep+hmWpB927cM0RIQugvqmFexeu6fr8bQgwYEjyGH5K115LKR28T71hflqS1P0My1IPcl9wdrhhdqnhWJJyhGFZ6kEnMu/aW7Y/SJKUy/IyXYCUS+bPm0ZR4ZH7ktubdz0421xRXU/kg9nmBUsqerBaSZJkWJZ60A2zS7n7xpmUlhQRSHbz3n3jzA91jI812yxJknqOYxhSD+vIvGtPzDY75iFJ0vEZlqVeKNVs89CiQi6858UuB1yPbJYkqWMcw5B6ofZmmwvzAnWNzWmZY3bMQ5KkjjEsS71Qe7PNgwYU0NRy5EHLnQ24rrCTJKljujyGEUL4CvAloAV4Osb49bbrdwK3tV3/aoxxYVe/lpRLjp5tLrvj6Xaf15mA65HNkiR1TJc6yyGEy4HrgVkxxjOA+9quTwduBs4Arga+H0LIT/lCko4rVZDtTMDt6Ao7SZJyXVfHML4I3BNjbACIMVa2Xb8eeCTG2BBj3Ai8B5zXxa8l5bR0BtyOrrCTJCnXdXUMYypwcQjhLuAA8J9jjG8CpcDrhz2vvO3ah4QQbgduB5g4cWIXy5Gy18Egm651bx7ZLEnS8R03LIcQngfGtPOpf2z754cB5wPnAo+GECafSAExxh8BPwKYM2dOPM7TpZxmwJUkqWcdNyzHGK9K9bkQwheBx2OMEVgUQmgFRgAVwITDnjq+7ZokSZLUZ3R1ZnkBcDlACGEq0A/YCTwF3BxC6B9CKAOmAIu6+LUkSZKkHtXVmeWfAD8JIbwDNAK3tnWZV4YQHgXeBZqBL8UYW47xOpIkSVKv06WwHGNsBG5J8bm7gLu68vqSJElSJnX5UBJJSrcFSyrStvVDkqSuMCxL6lUWLKngzsdXUN+UTG5VVNdz5+MrAAzMkqQeZ1iWskxf78reu3DNoaB8UH1TC/cuXNOn/hySpOxgWJaySDZ0ZbdW15/QdUmSulNXV8dJ6kWO1ZXtK8aVFJ3QdUmSupNhWcoi2dCVnT9vGkWF+UdcKyrMZ/68aRmqSJKUywzLUhbJhq7sDbNLufvGmZSWFBGA0pIi7r5xZp8ZI5EkZRdnlqUsMn/etCNmlqFvdmVvmF1qOJYk9QqGZSmLHAyYfXkbhiRJvYlhWTmjr69U6yi7spIkpY9hWTkhG1aqSZKknmdYVk7woIv0ypUuvSRJhmXlhHSvVMvlsGiXXpKUS1wdp5yQzpVqB8NiRXU9kQ/C4oIlFV2ssm/IhoNPJEnqKMOyckI6D7rI9bCYDQefSJLUUYZl5YR0HnSR62ExGw4+kSSpo5xZVs5I10q1cSVFVLQTjHMlLGbLwSeSJHWEnWXpBKVzpKMv8jhqSVIusbMsnSBPyfPgE0lS7jAsS51gWJQkKTc4hiFJkiSlYFiWJEmSUnAMQzkrl0/hkyRJHWNYVk7yyGZJktQRhmXlpGOdwteXwrLdcUmSupdhWTkpG07hszsuSVL38w1+yknZcGTzsbrjkiQpPQzLyknZcApfNnTHJUnq7QzLyknZcGRzNnTHJUnq7ZxZVs7q66fwzZ837YiZZeh73XFJkno7w7LURx0M+m7DkCSp+xiWpT6sr3fHJUnq7ZxZliRJklIwLEuSJEkpOIYhHcYT8SRJ0uEMy1IbT8STJElHcwxDauOJeJIk6WiGZamNJ+JJkqSjGZalNp6IJ0mSjmZYltrMnzeNosL8I655Ip4kSbnNN/hJbTwRT5IkHc2wLB3GE/EkSdLhHMOQJEmSUjAsS5IkSSkYliVJkqQUDMuSJElSCoZlSZIkKQXDsiRJkpSCYVmSJElKwbAsSZIkpWBYliRJklIwLEuSJEkpGJYlSZKkFAzLkiRJUgqGZUmSJCmFEGPMdA2HhBCqgM0Z+vIjgJ0Z+trqOd7n7Oc9zg3e5+znPc4NmbzPJ8cYRx7vSb0qLGdSCOGtGOOcTNeh7uV9zn7e49zgfc5+3uPc0Bfus2MYkiRJUgqGZUmSJCkFw/IHfpTpAtQjvM/Zz3ucG7zP2c97nBt6/X12ZlmSJElKwc6yJEmSlIJhWZIkSUoh58NyCOHqEMKaEMJ7IYQ7Ml2P0i+EMCGE8FII4d0QwsoQwtcyXZO6RwghP4SwJITw60zXou4RQigJITwWQlgdQlgVQvhIpmtS+oUQ/q7t39fvhBAeDiEMyHRN6roQwk9CCJUhhHcOuzYshPDbEMK6tl9PymSN7cnpsBxCyAfuBz4OTAc+HUKYntmq1A2agX+IMU4Hzge+5H3OWl8DVmW6CHWr7wLPxRhPA2bh/c46IYRS4KvAnBjjDCAfuDmzVSlN/h24+qhrdwAvxBinAC+0/b5XyemwDJwHvBdj3BBjbAQeAa7PcE1Ksxjjthjj220f7yP5P9fSzFaldAshjAc+Afw407Woe4QQhgKXAA8AxBgbY4zVma1K3aQAKAohFAADga0ZrkdpEGN8Bdh91OXrgZ+2ffxT4IYeLaoDcj0slwJbDvt9OYaorBZCmATMBt7IbCXqBv8b+DrQmulC1G3KgCrgwbZxmx+HEIozXZTSK8ZYAdwHvA9sA/bGGH+T2arUjUbHGLe1fbwdGJ3JYtqT62FZOSSEMAj4JfC3McaaTNej9AkhXAtUxhgXZ7oWdasC4GzgBzHG2UAdvfBHtuqatpnV60n+42gcUBxCuCWzVaknxGSfca/baZzrYbkCmHDY78e3XVOWCSEUkgTlh2KMj2e6HqXdhcB1IYRNJONUV4QQfp7ZktQNyoHyGOPBnww9RhKelV2uAjbGGKtijE3A48AFGa5J3WdHCGEsQNuvlRmu50NyPSy/CUwJIZSFEPqRvIHgqQzXpDQLIQSSGcdVMcb/mel6lH4xxjtjjONjjJNI/h6/GGO0E5VlYozbgS0hhGltl64E3s1gSeoe7wPnhxAGtv37+0p8I2c2ewq4te3jW4EnM1hLuwoyXUAmxRibQwhfBhaSvNv2JzHGlRkuS+l3IfBZYEUIYWnbtW/EGJ/JYE2SOucrwENtDY4NwOcyXI/SLMb4RgjhMeBtkm1GS+gDRyLr+EIIDwOXASNCCOXAfwfuAR4NIdwGbAb+U+YqbJ/HXUuSJEkp5PoYhiRJkpSSYVmSJElKwbAsSZIkpWBYliRJklIwLEuSJEkpGJYlSZKkFAzLkiRJUgr/HzFSuqRkWkCyAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 864x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Let's plot the comparison\n",
+    "\n",
+    "final_ridge_model = Ridge(alpha=1e-05, normalize=True)\n",
+    "final_ridge_model.fit(vander(x, 6), y_scatter_curve_noise)\n",
+    "final_ridge_degree = final_ridge_model.coef_.size - 1\n",
+    "final_y_pred = final_ridge_model.predict(np.vander(x, final_ridge_degree + 1))\n",
+    "\n",
+    "plt.figure(figsize=(12, 7))    \n",
+    "plt.scatter(x, y_scatter_curve_noise)\n",
+    "plt.plot(x, final_y_pred)\n",
+    "plt.plot(x, final_lm_y_pred)\n",
+    "plt.plot(x, lm_lasso.predict(vander(x, 6)))\n",
+    "plt.title(\"Linear vs Ridge Model - Poly Deg 5\")\n",
+    "plt.legend(['Ridge Model alpha = 1e-05', 'Linear Model', 'Lasso Model', 'Observed Points'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.12"
+  },
+  "varInspector": {
+   "cols": {
+    "lenName": 16,
+    "lenType": 16,
+    "lenVar": 40
+   },
+   "kernels_config": {
+    "python": {
+     "delete_cmd_postfix": "",
+     "delete_cmd_prefix": "del ",
+     "library": "var_list.py",
+     "varRefreshCmd": "print(var_dic_list())"
+    },
+    "r": {
+     "delete_cmd_postfix": ") ",
+     "delete_cmd_prefix": "rm(",
+     "library": "var_list.r",
+     "varRefreshCmd": "cat(var_dic_list()) "
+    }
+   },
+   "types_to_exclude": [
+    "module",
+    "function",
+    "builtin_function_or_method",
+    "instance",
+    "_Feature"
+   ],
+   "window_display": false
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}