"Embedding.Keras3/04-Show-vectors.ipynb" did not exist on "fbbb034ca462bd0ffba3055ad0ebb80d094916a8"
Newer
Older
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
"# <!-- TITLE --> [LOGR1] - Logistic regression, in pure Tensorflow\n",
"<!-- DESC --> Logistic Regression with Mini-Batch Gradient Descent using pure TensorFlow. \n",
"<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
" - A logistic regression has the objective of providing a probability of belonging to a class. \n",
" - Découvrir une implémentation 100% Tensorflow ..et apprendre à aimer Keras\n",
"\n",
"## What we're going to do :\n",
"\n",
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
"X contains characteristics \n",
"y contains the probability of membership (1 or 0) \n",
"\n",
"We'll look for a value of $\\theta$ such that the linear regression $\\theta^{T}X$ can be used to calculate our probability: \n",
"\n",
"$\\hat{p} = h_\\theta(X) = \\sigma(\\theta^T{X})$ \n",
"\n",
"Where $\\sigma$ is the logit function, typically a sigmoid (S) function: \n",
"\n",
"$\n",
"\\sigma(t) = \\dfrac{1}{1 + \\exp(-t)}\n",
"$ \n",
"\n",
"The predicted value $\\hat{y}$ will then be calculated as follows:\n",
"\n",
"$\n",
"\\hat{y} =\n",
"\\begin{cases}\n",
" 0 & \\text{if } \\hat{p} < 0.5 \\\\\n",
" 1 & \\text{if } \\hat{p} \\geq 0.5\n",
"\\end{cases}\n",
"$\n",
"\n",
"**Calculation of the cost of the regression:** \n",
"For a training observation x, the cost can be calculated as follows: \n",
"\n",
"$\n",
"c(\\theta) =\n",
"\\begin{cases}\n",
" -\\log(\\hat{p}) & \\text{if } y = 1 \\\\\n",
" -\\log(1 - \\hat{p}) & \\text{if } y = 0\n",
"\\end{cases}\n",
"$\n",
"\n",
"The regression cost function (log loss) over the whole training set can be written as follows: \n",
"\n",
"$\n",
"J(\\theta) = -\\dfrac{1}{m} \\sum_{i=1}^{m}{\\left[ y^{(i)} log\\left(\\hat{p}^{(i)}\\right) + (1 - y^{(i)}) log\\left(1 - \\hat{p}^{(i)}\\right)\\right]}\n",
"$\n",
"## Step 1 - Import and init"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"WARNING:tensorflow:From /home/pjluc/anaconda3/envs/fidle/lib/python3.7/site-packages/tensorflow_core/python/compat/v2_compat.py:65: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
"Instructions for updating:\n",
"non-resource variables are not supported in the long term\n"
]
},
{
"data": {
"text/html": [
"<style>\n",
"\n",
"div.warn { \n",
" background-color: #fcf2f2;\n",
" border-color: #dFb5b4;\n",
" border-left: 5px solid #dfb5b4;\n",
" padding: 0.5em;\n",
" font-weight: bold;\n",
" font-size: 1.1em;;\n",
" }\n",
"\n",
"\n",
"\n",
"div.nota { \n",
" background-color: #DAFFDE;\n",
" border-left: 5px solid #92CC99;\n",
" padding: 0.5em;\n",
" }\n",
"\n",
"div.todo:before { content:url(data:image/svg+xml;base64,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);\n",
" float:left;\n",
" margin-right:20px;\n",
" margin-top:-20px;\n",
" margin-bottom:20px;\n",
"}\n",
"div.todo{\n",
" font-weight: bold;\n",
" font-size: 1.1em;\n",
" margin-top:40px;\n",
"}\n",
"div.todo ul{\n",
" margin: 0.2em;\n",
"}\n",
"div.todo li{\n",
" margin-left:60px;\n",
" margin-top:0;\n",
" margin-bottom:0;\n",
"}\n",
"\n",
"\n",
"</style>\n",
"\n"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"FIDLE 2020 - Practical Work Module\n",
"Version : 0.6 DEV\n",
"Run time : Tuesday 8 December 2020, 19:02:06\n",
"TensorFlow version : 2.0.0\n",
"Keras version : 2.2.4-tf\n",
"Datasets dir : /home/pjluc/datasets/fidle\n",
"Update keras cache : False\n"
]
}
],
"source": [
"import numpy as np\n",
"import sklearn as sl\n",
"from sklearn import metrics\n",
"\n",
"import tensorflow.compat.v1 as tf\n",
"tf.disable_v2_behavior()\n",
"\n",
"import matplotlib\n",
"import matplotlib.pyplot as plt\n",
"import math\n",
"import random\n",
"import os\n",
"import sys\n",
"\n",
"sys.path.append('..')\n",
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1.1 - Usefull stuff"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"def vector_infos(name,V):\n",
" '''Displaying some information about a vector'''\n",
" with np.printoptions(precision=4, suppress=True):\n",
" print(\"{:16} : ndim={} shape={:10} Mean = {} Std = {}\".format( name,V.ndim, str(V.shape), V.mean(axis=0), V.std(axis=0)))\n",
"\n",
"def random_batch(X_train, y_train, batch_size):\n",
" '''Returning a data set for a batch'''\n",
" indices = np.random.randint(0, len(X_train), batch_size)\n",
" X_batch = X_train[indices]\n",
" y_batch = y_train[indices]\n",
" return X_batch, y_batch"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1.2 - Parameters"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"data_size = 1000 # Number of observations\n",
"data_cols = 2 # observation size\n",
"data_noise = 0.2\n",
"test_ratio = 0.2 # Ratio of data reserved for validation\n",
"random_seed = 123\n",
"\n",
"learning_rate = 0.01\n",
"n_epochs = 1000\n",
"batch_size = 50\n",
"\n",
"epsilon = 1e-7 # To avoid overflows on some calculations (log())\n",
"\n",
"learning_rate2 = 0.01 # Pour la version 2\n",
"n_epochs2 = 6000\n",
"batch_size2 = 50\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 2 - Data preparation\n",
"### 2.1 - Get some data\n",
"The data here are totally fabricated and represent the **examination results** (passed or failed) based on the students' **working** and **sleeping hours** . \n",
"X=(working hours, sleeping hours) y={result} where result=0 (failed) or 1 (passed)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"def do_i_have_it(hours_of_work, hours_of_sleep):\n",
" '''Returns the exam result based on work and sleep hours'''\n",
" hours_of_sleep_min = 5\n",
" hours_of_work_min = 4\n",
" hours_of_game_max = 3\n",
" # ---- Have to sleep and work\n",
" if hours_of_sleep < hours_of_sleep_min: return 0\n",
" if hours_of_work < hours_of_work_min: return 0\n",
" # ---- Gameboy is not good for you\n",
" hours_of_game = 24 - 10 - hours_of_sleep - hours_of_work + random.gauss(0,0.4)\n",
" if hours_of_game > hours_of_game_max: return 0\n",
" # ---- Fine, you got it\n",
" return 1\n",
"\n",
"def make_students_dataset(size, noise):\n",
" '''Fabrique un dataset pour <size> étudiants'''\n",
" x = []\n",
" y = []\n",
" for i in range(size):\n",
" w = random.gauss(5,1)\n",
" s = random.gauss(7,1.5)\n",
" r = do_i_have_it(w,s)\n",
" x.append([w,s])\n",
" y.append(r)\n",
" return (np.array(x), np.array(y))\n",
"\n",
"X_data,y_data=make_students_dataset(data_size,data_noise)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 2.2 - Show it"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAeMAAAFrCAYAAADilOGNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAABfuklEQVR4nO29e5gkWV3n/c2MrGxW1Jnu6WLA1emp6qrOqiwURFEQ1IZBGmetF3dV2NZBcF3X7VMMCuq6rsBMDzyP6/XVGergK66KjMyC6+uldKAUnAYBLwh467p0V2f1DCiXarsYUF+7qjLz/SPyZJ08eeIekRGZ+f08Tz3VWRkZceJEdnzjdzm/X6ndboMQQggh+VHOewCEEELIuEMxJoQQQnKGYkwIIYTkDMWYEEIIyRmKMSGEEJIzuYixEKIthGAaNyGEEAKgkvPxKciEEELGiZLtj3RTE0IIITlDMSaEEEJyhmJMCCGE5AzFmBBCCMkZijEhhBCSMxRjQgghJGcoxoQQQkjOUIwJIYSQnKEYE0IIITlDMSaEEEJyhmJMCCGE5AzFmBAytjR2G1iQC6jcV8GCXEBjt5H3kMiYQjEmhIwtiw8tYuPaBprtJjaubWDxocW8h0TGFIoxIWRs2by2iVa7BQBotVvYvLaZ84jIuEIxJoSMLbXjNZRL7m2wXCqjdryW6v7pBidhoRgTQsaWlbMrmDs+B6fkYO74HFbOrqS6f7rBSVgqeQ+AEELyYvroNC6Ki5ntn25wEhZaxoQQkhFZu8HJ6EAxJoSQjMjaDU5GB7qpCSEkI7J2g5PRgZYxIYQQkjMUY0IIISRnKMaEEEJIzlCMCSGEkJyhGBNCCCE5QzEmhIw9LFtJ8oZiTMYO3niJCctWkryhGJOxgzdeYsKylSRvKMZk7OCNl5iwbCXJG4oxyY5GA1hYACoV93ejGO5g3niJCctWkrxhOUySHYuLwMYG0Gq5vxcXgYv5lwZcObuCxYcWsXltE7XjNd54CctWktyhGJPs2Nx0hRhwf28Wwx3MGy/JksZuo+9hb/rodN7DIgWHbmqSHbUaUO58xcpl9zUhIw4TBEkcKMYkO1ZWgLk5wHHc3yt0B5PRhwmCJA50U5PsmJ4uRIyYkEFSO17DxrUNtNotJgiS0NAyJoSQFGFmNokDLWNCCEkRJgiSONAyJoQQQnKGYkwISR3W/yYkGhRjQkjqcHkPIdGgGBOSA6NuOXJ5DyHRoBiTsaMIQjjqliPrfxMSDYoxGTuKIISjbjlyeQ8h0eDSJjI8NBpus4nNTbe05sqKW1gkIkUQwlEvDMHlPYREg5YxGR5UF6hm87ALVAwG6UL1conTciSE6Ay/GBe0Zy7JgJS6QA1SCL1c4spyPHj9AS6Ki+zqQ8iYM/xinJK1RApA0INVSl2gBimERXCJ50kRkuUIGQaGX4wL2jOXxCDowWoIu0CNe1ZxEZLlCBkGhl+M2TN3dAh6sFJdoA4O3N8xkrcGzbjHhsfdM0BIWIZfjIfQWiIeDPjBahAu1HGPDY+7Z4CQsAy/GA+htUQ8GPCDFV2o2ePnGWA8mZBDSu12e+AHFUK0AUBKOfBjE6Ko3FdBs93svnZKDg5ef5DjiMaLBbnQs9Z67vgc1yaTcaBk++PwW8aExIQu1HxhPJmQQyjGZGwZ9+SqvOHD0GBhWKDYUIzJ2DLuyVV5w4ehwcIciWLD2tSEkFxg/erBwrBAsaFlTEYauuYIcWFYoNhQjMlIQ9dcMtJ8mMniwYgPW+FhWKDYcGkTGWm4fCkZaS4/ymIpE5dHkSGES5vI+GG65qaOTsWypIbRAktjzGnGGbOIWTIOSkYFijEZaUzXHIBYbuthdHenMeY044xe+0ry0JDG+IbxQYuMHhRjMtKYy5e2d7etllTQDXkYLbA0xpxmnNFrX2EfGmzXKI3xZf2gNapiP6rnlReMGZOxwivGGBR7HMbY5LCMOWxcP+r5NHYbWHxoEZvXNlE7XsPK2RXrWvKs8wqG5TpEZVTPawAwZkyIlyUVZEUOYybqsIx56uiU72tFVEs/rMWb9ZKfYfSqhGFUzysvWPSDjBVehSZqx2s9T/nmDTmtAhVhrbU0GLWiGkHXyCSsWKycXem7JmkydXQKW9e3el6PAlGvB/GHljEZLhoNYGEBqFTc34104lSDsiIHkQg2bLG87d1t39eKqNcorMXLsqjxGBbPy7DAmPGo02gAi4vA5iZQq7k9goe55/PCArCxAbRaQLns9j2+ODzW3yDWPQ9bLC+r8Q7SC+EH17oTA8aMx5LFRVe8mk3392Lxl+T4srnpCjHg/t4crjjVIEoSDlssLysLqygWL8tQkjBQjEedIRevPmo11yIG3N+14bqxDcK1V+Sbv82FnpdoDsqdT3cuCQPd1KPOkLt1+yiI270oLlAbSceW5bkVyYVepLGQscLqpqYYjzoFEa9Royg38iyEUz+3EkqYcCbQbDVT2X+R4qdFGgsZKxgzHkump11L+ODA/T3mQhzXNWl+zi8uG/UYSdylZx48g7WdNTTbTaztrOHMg2dCf9YL/dzaaGOvuZda9neWLvSo81hkdz4ZPyjGZKyIu7TI/JxTdjxv5FGPkWS5k75+1fY6DrpI6aSRDJZl/DTqPDKWS4oEi36QsSJuprH5uYPmAeYm56yFIqIeo2jZz3oRDKfs4KB1kFphhywLkUSdx1ErikKGG1rGZKyI65o0Pzc3OeeZARz1GEncpTPHZnxfx0HPbl5fWk9sPQ4qa5luZzLMUIzJWBHXNRnlc1GPkcRdunrXKuqTdTglB/XJOlbvWg392TCksexoUO0n6XYmwwyzqQkpMEVeQhUWZi0fMgrXkySG2dSEDBuDsiqzxHQfTx2dCuW2HmRW+qAYhetJsoFiTEiBKVpyVxxM9zGAUIKUVVZ6nqI9CteTZAPFmKRHRh2VCnfMAVI7XkNJ82o5ZSdXiy+OkOlx55WzK9i6vhVKkLLKSs/TOmWSGfGCYkzSI25TiiSCOiSNMOJaYytnVzDhTHRfH7QOcnVtJhUy2/ZeguSUHd/Xtv2EEbo8rVMmmREvKMYkPeI2pUgiqBk0wrAJZ1LXpk3Ewuxz+ug0mq3D5Ke8XZtJhcy2vZcgHTR7k7z2mnu+cx9W6PK0TovSSYoUD4oxSY+4HZWSCGoGXZxswpnUIrSJ2PzyfLeU5frOuuc+i+TaTDoW8/P1ybqnIM1NzvVVAvOb+7BCR+uUFBGKMYmP6V5eXna7QjmO+3sl5E0urKDa3NkrK73HXF5OHEO2CWdSi9AUoWa7ib3mXvf9Ntqe+yySeCQdS5z12jpJ64ADtE5JMaEYk/iY7uWlpXhNKUxB9RJxmzvbbISxtJQ4hmyz/pJahLoIVcr2KrTDkMyTVMiifF5tW5+sp1YHPC2yysgehuVZJBtY9IPEp1JxRU/hOK4o5nm8FMZkK8wAILViDWYRDMUjL38Ep28/3ff3orRrzAu/Qhl5FRTJ6pqM+7UeE6xFP9gogsSnVnOtz1YrtXht4uOlMCavBgJeN8WoVZVqx2vdG66iXCpj6eEl6zHGfW2qX0MHfS7LpTKcsoPKfZXMq1tldU3G/VrrjFu1MrqpSXzCupcHebxBjwnxW/fpqBuv7qacfWAWsw/M9lh+WSZwDaOL1HT/7zf3B+KyziqprkjJenkzbtXKKMYkPma8NmyMOO3j6Yldi4uuAA9qTIjfus8WC9VvQFvXt/r6E2eZwJXmzW9Qwj59dBorZ1dQO17DXnMPbbhhtyysSv2c9pp7mD46nXpSXZGS9fJm3LwEFGPSz7BVtcq48EeQsMS1ZkyrbmNnA2s7az3uax2n5GSS/avOTz920pufLuxrO2uYX57PTJDVsXSysCr1c2rsNlB1qqlnZDPT+5Bx8xJQjEk/Q1LVqksGhT90gizGuNaMuvHWjtdw0DpAC3YRBrK9GWUhZrpVA7gFO7JyM5rHArLxIARZasPo5i8y4+YliJXAJYSYADAP4CYAjwNYl1LupzkwkiMZi1toGg33QWBz003EWlnpdzs3Gm58WGVQZ5BIFnQT9kswAoITUWxi4pQcTB2dcj9/vYFKuYLNa5tYkAupJ7JkIWa14zWs7az1HScsjd0Gzjx4puumnzk2g9W7Vq3nbSZxJc1A9rpe5nHMhxX1UNNqt7oPbStnV8YqCSlNgv5fjRqRLGMhxC1CiLcA+CyAjwG40Pn9WSHEW4QQx1MfIRk8GVS1ikUYC31xEdjXngMrldSTtpK6y4Isa1tVqoPXH+Dy3Zdx+e7LmJucw0HrILNEFtvxk7pIV86uoOpUu69LKHXnzc+CVO+dvP9kT7x86/qW53mnbUF5Xa+g49ge2sYtCYnEJ7QYCyFuBfDnAL4XwB6A9wN4Z+f3Xufvf9bZjgwTaVXSSpswFvrmJqCvlW82YyVt+QlE0pu9eZNe21nrOUacm3yaeB0/SY3u6aPTWF9aR32yDqfkYH5yvme9tpdA2VzmCq/zTjvO6jXfQcexPbTZ9mXO4YWrF+jeJuGLfggh/heA7wHw8wDulVJ+TnvviwGcB/ADAH5FSvmfA/bFoh9FYmGhd23u3JybiZw3YcaV0tizLLYw+8BsX1Z0lGPoYwOAqlPF+tK6VXTSXJtpmxMAiefJr1CHV0EUAF2LPWvifhdsc6+7rr3msFKuuDkDxvHGbZ3tGGEt+hHFTf0tAP5ESvkaXYgBQEr5OSnlqwF8EAD9MMNGUWLEJn5rhpU1v7npWvRKiGNa8TbrNUtLpdVuYWPHbgHqNHYb2Gvu9cR095v7nu7OuN2hbGRRoxvwd/vr7+nMHJsZWAJP0oQ83XK27cucQ/366nNKF/d4ESWB64sAfCBgmz8B8Iz4wyG5MOhKWmFR64ptqHhyq+W6qRNa87aqWPpNMIlFtr27bf17xQn+76fEVMevqYRf3FJPLApzPl4JS35JTGGwJTX5vTdoazDNxCHbvsx5NS1jNafjts523IliGW8AeErANk8BwG+MiR6TnZ11f4q0hjeHqlWJSdma1y0YnTRugl7Wnt6n2AtbprOfCIaNW+p4Wc42qy6NZCm/2Kv5HgDr2JIsI9I/e+SNR+CcdwYaqzXncPWuVeucjts623EnSsz45QAkgGdLKf/G8v7T4bqpz0kpfz1gX+MVM9bjmjpFis/aCLO0KC8yjHOnHT9WsT99qU/Y/ZrxYsCNnXpZjGHjlvpxi9ycYEEuYH1nvVtZC3Bd1oB7rnHGbJtTr33kGbfN+tiMSeeGNWYcRYy/AcBrALwIwK/DzaL+NIBbAXwjgJcBeBeA/9v8rJTy/ca+xkuMzU5COll3OkpCWMHTRXvKXRuL7e1sBTzDB4UoN6mstk3ymaj7sCVNOSUn1Rt03PPwS+gyCTtmr33aOj4V+UElKaN8bgUnsRi3ALS1HekftP2ti5Syx/c3dmI8rJZx2HaERTi/nKz4Qd3QsrRibJYikK61GHeeVJlOG+VSOfSYvcYS9Lm8WjT6kdZ3oYjnNiYkzqa+r/NzvvNzX8Df9J/xRo/Jzsy4P8MQnw1b/EOP3+oMMjM7YQlPW7ekMPHIQSXZZJlZGzVe7jcWr1hu3Hkyi4coZo7NBI45TCy86lRRRtkz/p00bptFicy0vguMSReL0JZxmoydZTyshLU2i2AZh7XiDWzxXJ0gSyuMxZeGJTMoKybM+fiNxevztr+HzZwOKo0Z5ZhRvRZJr10WnpO0vguMGedGYsuYjBthWyQWwfKPWcLTr+ITEGzFhaleNb88j/Wd9USWTNpWTJQM6ihj8bKAbfsNa+FNH53G5bsvo31PG+172rh89+Ue0fAacxpei6TVvcKOIYoFndZ3gR2iikVkMRZCTAghXiSEeLUQ4nXa358ghHiSEIICP4wkaZuoi/bly+5PnH7CYcdg2y7m8izb0iGdoJud1w1NF5o0+uwOqv4y4HZYUq0Pzzx4pk8YdAtKFa1Q26jmFgr12jZPabn4va5BGqKV1M0cdgxRXM9+80+Gl6iNIl4E4CqAPwDwswDu1d5+OoBPAnhpOkPLiWHr5auTZOxFaJsYNAZ1fidPAmtrvduFteLVrjo3WVtW7cyxGcwcm0kkfF4iHyQKXjf/QdVfXnxoMbBBw/TRaTfW2hEZ5e6MStYxyygPMF7znjQ+G3YMUR5M0pp/UiyiNIr4agC/Azdj+tUA3q6/L6X8MwDbAP59iuMbPEUQpbgkGXsRSmIGjUGdn07Msdrc0/XJOq686kq3W1Ic4bOJfEkLEVXKFSzfuRw4rrA3/7iWm5cQ2kTA62828TCrjTWuZ9eAI+jcozzAeM17Uus97BiiPpiwOtfoEcUyfh2AfwHw1VLK+wFctmzzYQBPS2NguVEEUYpLkrEXoW1i0Bi8ksRijNW0XJ2Sk4rFaRN5vWDFQesASw8vhRpXmJusLiJrO2uYX54PJcheQuiUnb5tp45O9Ymel3iYf684FavIxV1zrY/jzINnAh9cwjysNHYbWNtZs877oDKOgx5MzPOYOjrFTOgRI4oYPwfA70gpP+WzzccRXDKz2BRBlOKSZOxFKIkZNIaKpZZzzLF63WSTxgiDYtBBAquPq4QSnLLjOxbzeHvNPZy8/2RgmUcvi22/ud+3LYA+0fMSD/PvzVYztSYI5me2rm8FPricefAM1nbWemLgtv2aqIeStGP1XgRZ0Oa5AxjIuMjgiNIo4gsBXAvY5gsw7BnaKyv9y3mGhSRj92vKkCVRinWYVcwcJ9aYzU5IqrsOcHjTi9pUQVl6QdWiyqVyV2BtFqG+3McpO90GAl5jqR2vWZdl7TX3ACByo4sJZ6L7WcBt17i9u93X0UoJsikaZmMEc2lPkiYI5meAw8IfXtah2brSfK32a3LQPLCeT16Y5769u51ZgQ4uecqHKML59wAWArZ5OoAhyniyEDERqFAM49ijxLlNy99xYiWr6Z2QyqUyqk61e7OJG4sLWiKlqJQr2G/ue1qEuoXkZVXqeBXFUCjxDNsMwWxecdA6sD5ghC32sXznMirlSvfcVbw8jvvXzNQ+cdOJwGVlXujb2FzzLbQyax4Rx/syyAIdbN2YD1HE+F0Azgghnmt7UwjxzQC+DsDvpzEwMgCKkDkeJc6tu7ErFWB/P1aymp/gxr3pBbmnVXJYs9UMvcwpzFimj05jfWkd9cm67/haaIW6sZrH9Donc+zmDfzMg2ewIBfwvLc+r2tp6/HyNNy/E85E4LIyE9VkQt/moHWAifJE37ZZCVEcsRuUuxxgclheRBHjnwDwWQB/KIT4SQB1ABBC/LvO69+Eu7Tp59IeJMmIImSOR4lz65Z/s+n2MQYORTzkw4WfyMW96Xm1SQR6k8OiiH3YsShr+sqrrqA+We8rEakIc2M1j+lFULGPretbfZ4C/fi2GGmQxWhmanv1ibY9GDklB/XJOlbvWrWOt9VuoX1Pu2fuguYrbn5BHLEbZIEOlsnMh9BiLKX8ewAvBPAPAH4EwHfALev1e53XnwTwIillUFyZFIUiZI7HTRyziXjIhws/kYt70zPrHavlTEnEPupY1PaX7r5kdV2HubGax7TtxzZ28wYOoE8Q9YQ0W/3vIIsxrEiY29Un66ELgkQRItt4wwh00cVukFY4OSRSspWU8qMAagC+FcBPAvhluJbwdwCYl1L+bdoDHHqK4Ar2ogiZ42Hj3OY8Li/3i3jIh4ssrAx9n+tL65ifnM/lZtbYbWB+eb4nCQuAtRmCTTjMv/3qi3+1K8hVp4pHXv6Idd7MG/jMsZk+T8GEM9GNQW9d38LW9a0eIQuyGMOKRJjtwmaD+10723jDuKDDHCOLBhNhYZnMfGCjiKwJ2xM4DwbZdjDpscx5VI0h9H0Vea47ZN1y0dZy0KuRgG0sAPraC9Yn65Ezam0ZuaceOOWZba56EQ9Tf13b/G1e20yliQN7DY806TaKEEIcFUJ8WfzxjAlFcAV7Mcjs66TxaXMe9/b69xXB5R3W8vDbrrHbwOwDsyidL6F0voTZB2YDLRgv6y+JJaR/1rbMycsNahuLLd4aJ5HJZl15xdWVq1a3GKePTmOvuZeLZeiFeY2W71zus3DTckFn0WCCFJuotam/UAjxs0KIT8Fdc7ytvfe1QoiHhRDPSHuQQ00RXMGDxuaaN8V0bS2ay16fRx21r4XOqjvj4SJpzWG/7cLUce47DY+bdZLlJH7Zw1Wn6ulqtY3FJpitdgsbO8HLtkz8xMtW/1sX8KpTRWO3kfvyGr/uW0sPL/U9cKQRb23sNnqWW6XVYIIUmyi1qW8C8Kdw61L/A4B19Jrbfwvg6wGcTXOAQ08RKlsNGpsVbHsIiWId6/NYrQIlw9PjYW0nrTnst11QHWfbg0DYdn9rO2t9VriXBWSzZstw1083W82eddU6trGov5m00IpsdZlzr8RLJZht7257FpXY2NnomY84DwNpELX7Vpx4q63Mp14JrVKuJGowQet5OIhiGf843KIfr5BSPgPuUqYuUsp/AfA+AHekN7wRYBgLcSTF5pq3PYREcdnr87i+DszP977favU3kYD3zSpuZq6+ne0zteO17s3v5P0nu6UYN65tYH55HqceOAUAuHT3pb7sXhO9dKOfBWTLHp6bnOsmS3lZTDbhUH+zLY+KanX5dYYKsuYqTsXztSkuF65eiBwuiHMOOmlmQdvKfOr1zJutZqIGE7Seh4MoYvwfAKxKKX/dZ5tHAfzbZEMiQ02j4VqvilLJtYqnp4F6PR2XvRLmqrHsxlK72rxZqaYHGzsbqJQriTJzV86udItIAG5BCVXO0rbGVvUJtt0Qbcffur7VEwv2soBsY0xSuKGx27DGdqMWf/DrDBU0NrMSmP7aVmDEDBfYalDHwVadS5/nNKxOrzKf6nfStegs4jEcRKlN/aUAfitgm38CcFP84ZChZ3HRtV4VExOHVnHadb/NWtXma/TWeq4dr2GvuedmqaKFg9ZBYJaqX23i6aPTuHx3b/My1QHID9sN0cvyse3LvEHbxqhnJgOuqDR2G6HcposPLWK/1d8wwqypDaBHCGeOzWD1rtXuMcy5V58xs6a9PAxe25jiYi7jAuw1qOOgalQryij3ZEfrWc9R64ArzHNV/YrNebMRpnZ2mPkm+RPFMv48gCcFbDOF4GYSZJTRXdSAK5DKNZ/EZW9LCgtRq1p3xa6cXQnV6ScJYVyAXjdE3cr2wq8Os26ZrZxd6daEBtxuTGHdk15zYtbUtiWwqTKYlfsq3WYSZvw0yTpgoN/i9qvNnZS5ybmeY81N9sbTbbF+W0ETP8xzXb1rNdV1viziMRxEsYw/DOBbhBBfJKX8vPmmEOIpAO4Ea1OPN7Va71rftLLHVVKYig0vLvZa2o7j1qputw/fN9YY28QobSshTKKRrfjG4kOLaFxvdJOubJnR9cl6nxXk1WVq+uh0j2u3jXboBw+vTlBhampvXd/q1rT2shTDWHN+2yiLe31n3dMyDvNgEwYv615heiCAXqs8jLWcdmco2xpvrlEuPlEs418AcAuAh4UQPdkznde/CeAJAO5Pb3gkN+JWDssqe9yWFBZUq9rchUfXozQxE490VHKVae10BbXjOq8dr6E+We+W1AS8lyiFbXoBAM12M5TVZouF6xW19GVQNrKOTyrxmnD6mzsA7ly9ZfEtqR7Ly0r1yj5XpLWWPApM2BpOotSmXgVwL4DnAPg7AD8GAEKIa53XXwfgx6SUH0p/mGTgxC3SkVb2uPkwMDXln/wVsJ7btnazPllPvdSfmXgEoNsRSFlxF65e6Lkpm8t4Nq9tYuXsCk4eO9ndx2033WY9XpimFzp6Gcq1nTVropOKhbfvaaN9TxuX777cEwtW57F853Io0c4Km0UM9HaHyhol1vXJum9BE2BwIsmEreEkam3q++AuXfo9ALsAmgDaAB4G8AIp5U+nPkIyeBoNt5DGICuHmeJ75kzvwwDgb3EHWOSLDy2GXruZBNMarTpVnLj5RPdvjd0GXvi2F3aXPK3trMEpO30CppJ49M/Zbt5e/YLDEjbRyTaepYeXrKI9qPikV6w4DQGKasXqcVlbQRNgcCJZ9EYUxA5rU5N+FhZcMdap17Ot82zWldaTwABXZA+i1/hVVO6rpFIzOIioNZkV9ck6Nq9tYuroFAC3PaBX3FgvkuFXw1h/z4/2Pd73AP18zPGEmUPbfKTljbhw9QLOPHimayGXUEIb7VRqOWdRG3pQ9aaznHOSCunWpiYodkemJNis4Kwrh5kxYSDVMqJB1kKceJ7tM0E1mb16Hqsbp9qvl3ib7s2gCmGmECuXuSIo0cmr1GbYOczSNXv69tO48dobaN/TxpVXXUm1U5ZtXpPGfAeV1cyuS8MJLeMkDEGXoFjkcV7mMaen3aIeKXWUCrIWTKtFX+upW6v6Z8NaOuax/3nvn/Ho44/GPhfdIg1rGav3bNnBfjdr06Ogjh80h2HGXmSCOloVtZMSreKhwGoZe4qxEKIFII5St6WUvkumRkaMVRs/RUJXamEYZGvFtI6Z8PM20bGh34Tjur79XL/qGLqgVZ1qX7LSIy9/BKdvP+178417Y9Y/55Qd7Df3Q7l/g+bQ6/NJBCQr8QkTbijigwVbLw4FVjH2E833I54Yjw9Zram1EUdsLlxwE6H29lwrc3UVOH06+FgqI3qQJD2mbR2yx/4au42+ylFTR6fQ2G0ExlfN+tZhKxt5iUbpfP//S3MMB63+G/7z3vo8XHnVlcAKYWEsdVPA9LXL7VYbE84Emq1mYDUo25pbhW5Nm3itlQ5D3M8GzUFQVbOiJkYxk3p4oZs6CYO0IP1cx17jOHLEFWJFtQrcuJHN+PLG9FIAwJUr1uuxIBf6ilqcuOkEnlh9YmApS93aCLqhmxbmQeugz2KpvqFqLT0ZBlsREBM9yanqVLF61yqWHl7ytZ5sFq4Sfj/U+epzGMYajpMY5jXWsJ+NY0HarjeAQrmFaRkPBUzgSp1BdmSyFb1QeK0J3jPWYe7t5ZNsNohEtwgtGm3WwqOPP+q5XtRrqYpZanPxoUXP5KW95p61RWKQJe5HGKtHzzbea+7hzINnAq0nm8U3vzwfmLCk5uPKq66gPlmHU3IwfXQae809337SOlEsTnPteAml0J+NY0HaEqOSJqilXQiEpS+Hl8SWsRDiONw+xv8C4D1SysDA28hYxoPEzzL2il2blrEizaSsMN6BtJKz/I7VaAAnT/Zu7xHDt1nGgLvEJ24M0maReMWEFVWnittuui2WKJdQwvzkfKAb2mt5lJ/11Nht4OT9J62fi2pleSXG2ebfKzEszL4Bdz7Xl9b7Pmu7prp7O4q3wyTpkjlasmNJMstYCHFOCPHnQohj2t++CsA6gP8Dt/DHh4QQT0w6UmLBr6iFV/Wp1VX7vtIs4hGmUpdp1W9txavu5XesCC0aV86uBC7xabfb2Lq+hVMPnPK1WJRlY2txqLKwvVAWq7JkzDEpVLUw3eKcn5z3tHq8rE7gsKymWU1LP7/po9OoT9b7Phs1/qg6WOnzsnV9C+s769bzC7sUxzbngHffX5v16mVBRrV0kxbYYIyXKKK4qV8KN1P6uva3nwZwFMCvwhXjZwL4r+kNj3Txc4l7CfXp070CpUgz2czPfa4wHxbUtn6fiXOskHWxp49OY+OVG11hq0/WsXqX++Ci14lW/YfXdtY83bS23sWAWwf60c8GL1/a3t3uuj71MXm5xoMwBVBnojzRLWsZVN1r5exKT4WrKC5gJZY26xo4bDihiOpOtc25nxDaBM9rLW5UcUzqFma1LKKI0rVpFsAfqBcd9/Q3AvhlKeX3d/725wC+E8DPpjlIEoBfJrLqbLSxcejOTqOXsCJMRrnZx3hvz3UrR81C9zqW6b6+dCnQ7a1uxkqITj1wCrXjNWuhDMC1Ym2Zul7bAwiVmNVsN7EgF7rWqp+L0qt3rpkd7sWJm09g6eGlPve1V3/l9aV1345FXng9oJh4ucjVMfX13U7ZwUHzAHOTc24tb/TOuV950yhZ0FEzpsN2XPJyfy/fudyN68cpZ0pGhyiW8S0APqO9fk7n929rf/sTACeSDoqkiBLqZtPNpE472SyMNWpa9aur8To7eR0rblML9Lsl9TrRJjYrKUx1LR1lieus7azh+W99fmAij5fVZvYV9mLr+lakpCm/Sk564pHZCUpvfGGi+g97WZL69dCbWuw199CC+xBi64zl5aIGEOiWN7f1s3TjJlzp57W2s4aT95/EglzA9618X3fp2iAbXCgG1UmKBBNFjK8DOK69/kYALQB6l6Y23DaKZFyIk1EeNwvd63NhXOUemAJ30DzA3PE5lI3/Gl5uWvPmbXYt0lGiZ7OkHn380W7zCDNWqW6YfUuAOpnEYeKMaiw2kZw+Om21Kv1u1F6iqcTS68Gk1W6h3W7DK3HUz9OgPt9sNfuaRDhlx1NIwjbdUNv6lZKMmz1tO6+NaxvYur6Va8yY7RaLQxQxXgewKIS4RQhxM9wY8oellJ/TtrkdwKfSGx4ZGvKs0x3QPtH3o0aXJeX+vPyqy6ESpsybt961aProNE7cdKJn2zCuXvOm7OX2PWgedM/BxkR5oifurD8o6Dz2+GPWzy8+tIj1nXVr3NxLNFtttyezmgNdBBX7rf2ulWve/M3rYaIeaNaX1nsEea+5h5P3n8SRNx4JtNaTiF7chCvbean95BkzZgJZcYgixr8A4CkAPgHg4wCeDKC7NkkI4QB4LoC/TnOAZEhI4CpOTMjELetHLT1/lUgkLbj/2OOP4eOf+ziAQ/es2odfgwbzpmwTvnKpjLnJue456KIPuEVMNl65gct3X+6O/y2Lb+m2W9RR8XCTzWubPclW+nZ+otlqt7BydgUHrz9As9X0tHRb7RY2dnofMrxaEVadKso4rK89fXTa2jtaJd3p1vr6znpPjDlKMppJ3IQr2/esXCpj5thMruuCmUBWHEKLsZTy9+BmSl8EsAngh6WUD2qbvACui9pjPQ0ZaRK4ihNb1cp9femS+/rUKXc/Fy5477dzzOnJU7i4DMzsatZxAgvBr9CHvs/Vu1aty4eA/uxim/CZxUeeWH1iz031idUn9j08fN/K9/XVuFZ4xcNN1nbWsCAXsHzncldEbIQRbQB98V/9Aejy3Ze7DxM3XnsDzXuaPQ9FQftWmNnblXLFWogkDHGzp20FUeaOz2H1rtVcOyzp5+NVoIUMBpbDJOmQpNNTWl2izP1UKm582bbfhQW0NzZQarXQLAGbx4GnLiXvh+vVLMErc9iska2WHplZxfvN/W6nJ307r+M6JQdTR6d69u2X5GUr6NHYbWB+ed4q4HqBDVsRFVX8oluJbMeNJZv7KpfKaL4+uEGHPiYz2zpM8pqJasYRdK2jFAEZho5JfmNkAZKBwXKYJEOCXMWm9atbrWtr8a1qHdM639vz3u/mJkqd95w2cOoaMOFMJHYX6tZaCaXAzGFlfehubCWCKqGrsdvAJ//pk74JSKa70Sk7PSLlJ1iqEIiJWt5ks+B1l7W5Jll3dyqLsHlPEzdee6Mv8crmNvdDj2NvXd/CY48/hkde/ohbxtTjdqbHrc1EtiAvSJQEp2FIhvIbI+PH+UIxHieyTLIKypA2Y8pnzhy+1gmTgOV1HmYiV7XqndhVq6HZeT5VlnGz1Qx2FwbMoe72m5+cx/rSumftar+lSroFqZbj+N0oTffpfjNc84mqU0Wz1ex5MNBRYupXkUsX7aCHGZV05vU6CFsc+3lvfR4AN+nOdJs7JQfrS+u+Ge9+cdIoAjUMYmaOUdVIBxg/zhuK8TiRZ5KVn9WqCJuA5XUepnXut555ZQXbt1ZxUAI2jgMv/s6QST3asZvra7j4rJM98TWvpC+bReJ18wtzE1fFQryOa7M4Z47N9FQdmzk2g4PWge9yKvXwsHzncmBFLr1i2ZkHz/SJe2O30RcjbqEVKT7pdY385tQv4z3ICxJFoGzbXrh6AUfeeASl8yUceeMRXLh6IdR5xsW8bheuXuh5bSvRqns42GQiPyjGUchz+U5c9DGn5Q6OQ5DVWq+HX3fslSxmJnK94AXu70uX+vc7PY3y2jqe9qY6nvZKB0dOedd69jq20wZq1xDKJWmzmrxufrXjNZTsYaUebMdVN2Nb9S8zWWh7d9vTkjMfHpYeXuqxfs2lXmbhka3rW31jW3xo0dqbOYpL16tCld+cmgLltZTLhtpfGWVUyhVsXtv0fHiwHdvWNQvIrtiGed3OPHim57UN3cORZzLZuMMEriiklWjkR9o9kvUx26jXs+3DrDDPa3kZWFqKd55B1yGt62S7FouLaK6vwWm77u2N48BTl4K79XglxwT1yHXKjmcGtELvduRXFrN9T+//db+EnaBuRGYile2Y5me8ktts25rH0Lstre+s92VI+/VN1pPQyiVXVG29pf2Im9xUOt//UNW+p51ZspTfHAOH3xUmauUKE7gSk2T5TljSdiXrY7YxKHe1GVM+fTp+L+igZLG0rpPlWjz24DI2jqPr3l48624a5OKO0iVIt1Bsa2lN9M96CbFtXbOfWzLIPWtW4LJhVsXyWork5f61zY0ZMwbg61aNE3+3ETcebCasqddZxZfN62Ymr6mHmkG4o1lqMxqeYiyEuC6E+G/a69cLIb5hMMMqKAkqPYUmbcE3x1yvuyKmGLS7Og2CksXSuk7GtThYX8PsH5zBU5eAiXtci3j7mHc2cs+QY3YJCruW1u+GrnelMse0fOcynLLTrbClYppBN+ygspWAW2tZdz97FfTwEgTb3Jgxz5ljM75uVduc2AQqiLjJTat3rXYFuOpUu9chq2Qp87rZ4uODckcPQ3Z5kfD7X34zeutM3wvgdIZjKT4JKj2FJm3Bt415EA8VeZLWddLmSWVc21zGtob2oQ8RcFO2rVWdOTbTLR5hfta0gGeOzfjecL1imkFNIlRdbD/MBwSvgh5e4zPnZuroVF9byqCscTP+rgRRCVSlXMHGzkag5RbXmjx9+2nceO0NtO9p48Zrb+D07acT7c+GboGqXs1qXk/ffjq3OPAwZJcXCT8x/jSALx3UQIaCuA0OopC24NvGPIiHijxJ6zp15sl0SQPA1HXg75aB/fPA9JPnYyf1Bd2UzSYHwGFy1I9//Y93s6ZV+z1V2cvs0+yF+XChv/ZyM3olYilsDxdxsorNuQH621KqQih++5ifnO/Ox/rSelegasdrbuzYo062TtrWZNiOWGHcu0W1QLlUKhqeCVxCiN8G8CIAvwbgk3At4wudHz/aUso3+G0wtAlc40ScRLK0k8+y3q9+iJCViUoo4dTjFfzV/fs40jQyMUImiwVVajLfN/sPA/ETkUyOvPFInyDXJ+vdZClboo9flTH18GCem3mcqlPFjdfeiDRWr+OaiWlx9xeUhBeHOFW5oiZ3DeI84jAMFclywprA5SfGMwB+F8B8xAO1pZS+PiyK8RAQJyM5zmfCCK0tIzxCFniYm4LfDdD8/Mfu30N106OileO4FrkPC3KhJyNYuU6XHl7qZlDvN/e7pTl10fUjzk34wtULPa7qoP3Xjtd6Sl+WUMKEM4Fmq+l7w/XKKvZDn3evrPITN53A1R+8Gjh2G4Mo/xjmGEEPX3Ez9UlhiZZNLaXcAvDlAGZxGCv+NQDPC/h5fkoDJlFIew10lEQydew465jDZI/bMsIjZIGHceP5xbdMl2J1a9t+oHIZezNTni5G5X5c21nrqyL1wre9sFv+cq+5132/1W5hv7nfV8Tj5G4JFztu8r9bdl/HcQOqmKZXw4fuqXVKbJprVVWVsUt3u2u7Tz1wyupa9coq9sNsumFjwpno/juqe3cQWcVh4qbm99MpO5HcuyzWMRr4FoaVUrYAXAFwRQgBAFellO8bxMBIRJSotVqHQpVkDXSt1mvl+iV5qWObNJuuSPtZsGFEXx+LwtzWx8IOc0M011763gBt43EcoFbD4kv3uvtRwq/WE3s1XQD6Y6EK3TLW+b23t1G75hYembsGvPsdEyivxb8J6+dvY+74nNsTGPb3dZe2ft6K1btWuxa4nlXsR5iM7e3dwwejoDEoTEv00t2XMnOfhvlemd/Pg+YB5ibn+taee6EeFk3oJh4uQq8zllKWpZT3ZTmYsSapZWuK2sZGsv1FSfLyW8scZMHaMrvNuVhedsegYz4g+FjY1kQS4xh/+GU/js03V7B/Hth8cwXvera90pM+N23HwdaTq5j9gTIW7q+h8YEVvBf2qlbmetewmOtiFac6Qgy4v2d2mqFutF7Wo63frkJ1dJqbnOtbZqWv//V74PHKKvYjzLIuJW6N3QbWdtb66i7bLORBJjyFsVrN7+fcpLudcll71Q0PoqiJXcROrApcQogvBfCVcJc/PQ7go1LKT0T4PGPGJkmrRkVpH5j12G01p73iqB5Vrqxz4Rdfdpze45bL3SYUVgvhuYuJ58sWqwPQ87dKuYJmq4nb/rGJlYfc8pmbnczs7WPRp7oEt83j3y27FrGqBObM10Nd36D4Ypi2jrZ2iVlUdTLHYuPKq654tnEE7JW5ipbwZMbGD5oHqDjJk/PSPE9a2akSLYHLhhDiNgC/BOCbLG//EYD/KqW8GmI/FGOTSqW3g1GIRKAeTKHa3Ey2vyiYx/7nfwYe1ZaczMwAly+H31+cuThyxG0+oahWgRs37AIOACdP+u8vxDFtN7tLd1/qubGqRCxTPDePA1/xSgfNdhNT1xFKqJ2S001k+oZtYPVB4EgTuOEAT3jPI25VswDSuEGbgq5uyjYBDyLoJq8fy0TvwRyl1GZRE578zjWN65TkPIs6Z0NKsnKYQognA/gggBcCeBTA2wD8VOf3dufvH+hsR6K6nZMW4jDX1g6ysId57ImJ4M8obPOkj71UcoUxaB7NVozqtc19feZM/+f92i16jHvjzU5fApVZylIlYtUMt/L8dafrnlx5yBXqShuo7wBX7nf3OXX98HD1yTou3X2p6+qWDwMTLfd/9UQLbp1vNTSfRKawaz/99qELZqvdwqOffdTalzloP0CvK3VtZw0n7+/tguUVN545NoPlO5cx+8AsSudLvvWYzQ5XRU148ouRx0nOCzrPKAlvLOCRPVFqU78OwL8F8KMAZqWUr5BS/piU8hUAagD+G4AvAfDa1Ec5jEStMZ12IY48C3tsb/e/9no4sc2TPvaJCVfkg+bR6+HDjKWvrQFbFrenX7tFG4uLOPmZA1S0BCrzZqcL3+ZxdPsno1xGqXZYI1gX6lLnZ+6aay0Dh+U29ZifKe56MptfrDCsEPntwyxEst/aD939yYxb2gRI384WN646VTSue7uwValNr31mVQ4yaS1mvxh5nAeGoPOMElNmAY/sCe2mFkJcBbAhpXyRzzbvBjAnpbw9YF+j76ZO6nYeZmzxb8AeBw6ap7Dz2Gi4Fq8S2pkZV2D1+LMfUXMnQoxLd8HegSmsvB3usigz3u3RWeugBDztTfXujfjk/Yeudd3tbca403BFB+3Dr/CH7sJ0zjs9GdhllNG8p+kZfzaPF7Sd1+f8ziGr+GdSV67tXLN0CUf5njBmnCqJuzY9GcBHArb5SGc7Mur1n/2wWeVeS5iC5insPE5P97qaVaxYjcWPmf6ORkCApaONqw2g3Wzi4pNKOHPfbHc73TJZff1lVDcu20t02sZYLqMyX+9aNH29gc8CjVsnrJZ8GlZM0D7090sooepUrdZ2xeldPaleK6vMhjqeLgBRxh10DlllGSd15arvi6o77teTOY0OSFG+J+x1nD1RxPhxACcCtrmtsx0ZpvrPaRcMsdWG9hLVoHmKu8RKCb4aS73eG4euVt191uuuBW3h3ANn8M571/Cv9zTxkVet4cQtJw/nxxBQ5Vr+uTdtdW/uoW+gaoxXrhx21TLO1byxbx8DFgTQ2LnUJ+5pxETNfSzfudxzLst3LnffV4U/bDdqswWkem26p8so9wmQLpo2SihhonyYnzBzbKbnXM3Y9l5zD43dxsDaF8Z15dqEz/YAkVSgixo7H1eiuKl/C8C/A/B8KeWHLO9/LYD3AfgDKeW3Bexr9N3Uw0TSZVVhGEB9ad/ziHH8i08qdV3BbXR8S+Z+DXf1QQl4wr2uuy8tt6VfOUjgsJZ0ltZK3HPx+pxZ79tWUtPmClcZ5QfNA0wfc893e3e7pziG7k5VAuy3/CwtN3CWrlybSzmL5WRkICRb2iSEeAaADwFwAPxvAI/AbSDxZLjlMs8CaAF4jpTS151NMS4YYeKygxDTpKQ1xs5+2mtr9v81+vwsLKC5vtZdsrRxHHjJvXVrQwVbTC5sg4pyqeyZaTuIG3HcOLTX+fWtrbWsqQ16AAizztucM3P52bDEP23nGrWGNSkMyWLGUsqPAvh2uG7o7wLwFgC/D+CXAbwMwOcAvCRIiEkBCROXjZod7kfabnFFWq0TO+daAroVpHseWR3ncMwrK2iemkGzBOyX3Sznj92/BzQaodyWpvvx3ANnunPzznvXcOIfD92pTsmx1nTOeqmJ2b+4hPB1sL1ijeYSMJvbOMiNarqb13bW+qpw6XiN+7HHH0s9Hps2trlghvNoEbkClxDiiQBeDOAZAG6CK84fA/A7Usp/DrkPWsZFIoxFmVZ2eKMBzM8fFugoldzXuls8byvcPFcT25hnZ3uXTM3MoPEXq24ZzEsbeNc7Kji508T+zBQWvxN4L7atHXr+bhlY+Ee3ipmytJ+6dGj9qliqV3ZxlIIbYZl9YLZnCdFEeQIbr9xI7RhpuMDDotzcXl2xhsndq3fcUvW+w5QZJbmTvAJXWlCMC4qfCKYVV1bdnXRMYc8ihh1F4G3HD6poVrL8/+r839qbm4VzaavHla0E1hSCG+dbqGjaclAGnnCP0+dO9Vvyo1emioPpWrYdI24P4TDHC+s21j/nV/QjKsPi7jXj7pVypdtwJIuHMpIaiZc2kVHHzxWdVna4V1cmc5uorRiDiOJmt51rgqVq5ctbPQU66jtuda1uhx7N/dianek5TmWubs1SVm5eW+vDpC5r03XuRxpLbsIumzGPBaD7ufpkvcdlG6ZFo77tMLp7dTd9G+2ezl9b17dSW7JFBgPFWCdqLDOr2Gde+IlgWvFYXdQAd4mRKexZrNG2VeLyuma2cw16GDlxwvP15i1GzBludS3VoaenV/LD0SqB2YTD9jcv0bT93YzFmujVrQbZGcjrWI3dRk9nK6fkuqF19CVQ6rV6AFq9azX1JT5ZrAs2CepqxZKVwwXFWCdqklKaSU1FYBCFSnRRq9eB9fV+Yc9ijbb5EAC4gnzypP+DlHrgOnXKfX2pf10vgP563Nrr17yyt6hICW6il7Wgw7sX0fjASuiHnpWzKz3iaK61VXgJme3vZmLQzLGZ7hrg+mS9pxfxIGsWex3LbDG439rv1gRXfMkXfUnP6xM3n+g+AJ2+/XTqBS0G8ZCiJ3XZPAHDYuETF8aMdaImKWVd8nLQiUx5J05lhVkq08QvLh02fu3zXWjsNtCqz2Pq03vumuVyGSVtP5l2xOlc04P1tZ6OUCouGtR5KiiGO8huPl7H8uvY5EXWceFBti+0tZpkzLjQMGYcSFTLMGtLMq7lHcd9nlSIi+yyX1z0H49fXNrPda+fs+McJnGVy8DUVPe96ecuYuahVbfnsOO4QmxU1zKX6IRxJ3eH4ecS7XyHVEML5R5XVpNteUyU0odRqzglcd96HSvIXWtj6uhUqmMzSXPZUZCVrXsGyiW3ktnluy9TiIcMWsY6UQUpa0syruUdxpozx77nro2NncE8iCpecQlaqqTGu7LSfz31RhPmeZnnrI4TcT69luiUUML85HxfAYyTuyW8+x0TmNlxj3XmpXt4Dxp269RSIUw1nzCLb9SO17B85zKWHl7qscIA9FheSayuLCzpMM0kJsoTPQlOM8dmcPnu3h7baY4tzWpcUZt2DEs2+BjDpU1DR1yBCyPi5r7NrkZRXe5F7lJlE839/cNOTdWqG7u2Ca9NoNUDl985R5gPPzGxuZP1jk36UinzM9ZzD/gOeVW1MsdWdapYX1qPLDBpV/Ly2qZcKveIr4ntuEUVtTiVyIZhnfQYk8xNLYS4XQhxZ6foh/pbRQhxXgjx10KIDwkh/n0aIyUd4iYyhXGfm+5Xta3fZ5IeMy/MedSFGDgsQOLXaMKWUOV3zhHmw2+Zks2dbPYyrl073L5cKmPq6FTX3XrmpXvYm53u+Q75uWNtSVK2pKy95l6spKS47tswCVG6e92WBR503LBjG0SmtE5QKIANH0aDKIGWewC8DcAN7W+vBfA6AF8O4FkA3imEeFZ6wys4WcdJ4y4nCiPipljMzCTLYC5Clyqv62HOo5n5DLjWr01A1T4dBzhypHff+jlPT7uirt5fXj58r1JxrdOA74gZ+6w61e6NVb/hXr21inZnnM0SsHn8cB/KWlTC9YftLRw5u4WF+2tulva0vQOQbQxKkLxEKShz2iZaXsKhbzv7wCxmH5j1XW4VdGxbRniQYIUVtTgdlJIIeFAMn+0NR4MojSI2APyNlPIlnddlAJ8CcA3AC+E2jHgPgFUp5UsD9jUabuoix0mDGMXM6biZz4Armpcu+ceMFbZ9ex07wnfEdMW+69nLuO2upf5rpDKkN9awecthhjTgVt+yVaTyyz7W3bE2dzCAvmzdMO7QKO5Tr7h53E5LfqUik8Zz43RQoiuZaCTOpr4VwKPa66cDOA5gWUr5CSnlXwL4XQDPjDvCocMr07bImcWKtIp4+NFouDWbSyX3Z3Y227kIW7nL5jKu1XrnRAnx2lp/PN22b+PYBxtuRvTBxlrvmNbWPL8X00encfFFKzh4Rw0XX7WJ2551xp5N3xnnY9eu4KlLh0IMoCsw5VIZU9fd+PL+eeBv3tTCjc11LMiFHiE5uVvCxpud7pimd9FjZQGuJbi9u42ZYzOYOTYDp+SgUq5gY2fD18qLYs2a/Y27U935XFRX7NLDSzhouQ8YB60DLD18GFRPugbY5j0IOtes1mMP2mVOsiOKGE+gt5DQczqv/1j72ycAPCWFcQ0HXnHBUSsGEpfFxd61vVtb3nORxgOMWdhD766ks7LiuuUVMzP9bnV1DW3YYsDasZslt+pWs910f5vPwX7fC/27s7fnWzVs+uh0T8EPwF2yo4Rr5SE30Usta/q9h9p95S3f/Y4JnPzMgeeYdOFq7DZQdaqoHa+59bTR8hWzKPFhr+VJKgYe1ZL1E78kwmhW+5o+Oh2qg1JWHZYGWQGNZEsUMf4EgK/QXt8J4JqUcl3725PgtlIcD7zipFnUVh5GbOftNRdhH2D8RHtlxf27Yn/fez/V6mEVsNVORSl9v/o1ND9ji4lr34WNTnENwP29cRzu53S8vhe24+qEeLhTMcT6dacn0evUtd7ylk7JwcxOEyWf76pXQpdNzEwrbfnO5dDWrG756ha47qKOIjh+4hdVGPXzml+ex5XrV7qfrTrVriDnkWQV58GC1nQxqQRv0uX3AbxaCPEzAP4VwDcB+FVjmzn0urJHG+XWNKnVeuOERcosTkqUWHOt1t+hyWsuwj7A6DFcJUzqGkxP98aC2237fmz7AHr/Vqm4n4+yVrszFy+RC3j02gbQbuHRW8p4yb2d+KAePwbc/c7OAtvbh5/Xvzs2jLnZ3t3uebv7utFAyXHQbjZRQn+iV1eEavD9rpqxUCVceny32W5iQS5gr7nnVhtruxbz0sNLoeOi6gHCRuW+SmTBUa0mzdh30Hs2lPXZarfQ1L5f+lj8xh/m/TioPtNqTGH7TOvnox5uGL/OnyiW8U8B2AbwGgD/A8An4WZYAwCEECcAfB2A96c5wKGkCJnFWeFnwZpW6/JyrzsYOCyGodNo9FqOfg8wQaIdZ1nX5mb/3w4Ogq+hx1yYVtC7nr3cL8SAa7lvbfV+Xv/uVKt99bSbJeDisWbXovG08hYXgf19lODGkvbLh9Z6j3UW8F21WXTqbzob1zawdX3LLpqd70W7UsHWU45g9gedSBZZHBevX4ax33tBjTN08u7wtPjQYjcuDgATzkQoi3uQ9cRJeCIV/RBC/BsAd3Revk9K+XntvQUAL4CbTe3bd21ksqmHibSyp/2KWcTNKF5YcItumEU4bOML2leY87TtA+gXy3rdf55sc6EyspV13WwezpGf+1l9vlY7HLvRQ7kNYG3SFdVHbyl3xdEaT7VU3jpyb7pZvLaa0OVSub9KmON013arIiVf8crwY0mzmlUQXkVP9L9VyhUcNA9QcSpotpqZj8mLuEVKmNmdO8kqcAkhvhvAp6WUq4EbB++LYjxo0lqG5bcfL6EOqkYVpXpXGg8Vb3878F3fdfj6N34DeNazDrOnFUFibxPZarW/qEgYVGUwtT/L64u3tLwrbZlo16lZAq4cBarOBE7suNYyZmbcWHnU+uPa3J956R7+qH2l2yFpojyBEzefwPbuNjbe7ODkZw4O49EaByVg4p7iVLjSCds4Q3f1pi1oYR8+4orqIB9uiJXES5t+BcCL0hkLSZUwmchpJZX5uTW9XMRBruMo1bvSWJL1Pd/T/1rtV3eX2+ZJd03v7/cmjAGuG95PiEulw0SwmRn3R82lLuwWV/lrXjkT3l2rXSdnvo5Tx2ZwuxJiwD+z3QvDLb/ydtc1qmi2m6g6VRy8/qA3MUxDxa7zdvHaUDFYhYrB2tzaWbp6w2ZIx00KY5GQYhJFjD8VcXsyKM6ccS26ZtP9feZM/zZxy1WaQg94i6FupbZah/HhoBh61jF28xxU+UuF/jponvSHmnb70A3thy7A8/OuC/7gALh82f1R65p1Ydet8s5cv/nu1cDqVd1YrPnQsr3dP66oD2TGA111axvNVm9C09rOmnt8cx6rVbQdB41bJ7B41t1WJXzlmd3b2G1g9oFZlM6XcPL+k9hrHn4X/GKwXnHsNM4lrNBTVEeLKOL6bgDP61TeIkXC7NNr69sbV/CirJmenu5NOlJuzSBrNusCJOY5mFS1xuzmPC0v9wr51FS/WNvEuCNAfQLsdX6dhKsulUrfNfK6+YaypIyHijaArUknmlgYa6kvHmv2WJLdU3losX8e19dROjjAt75uFo/e4u5DuUujrJVNW7gXH1rsqSym02w1PQXOyypNY91vVmuSSbGJEjO+FcCfAbgA4EeklNf8P+G7L8aM06RkCUGk1Y0rajemQXdvUoKvJ0wFdVcqldz61Ht7rmCurgKnT9v3b2ZBT0wAJ070Lkeane0vl+nXstFG2HmzxMwrD54KTuRpNIAzZ9DuPKhdPgbc+bISjpyaDx/rtJThvHqs1I0Z+x5fnaYlJgsgdCJS2slHtiQ0IFy5zzD7ixMXZ0x35LHGjKOsM34IwOMAvhvAfxRCXIXrujbv+m0p5R0gg2NmptcaNpcTJSHqmulBrLE2k6hUwpRyN5vrj80xRUleM4tw7O+7Aq4L5dycPTs77DnUaq7Frfc/9po3yxrp2pJ9LXAP09PA5cuY6BGLNpwosc6OB+MJxj6AwyzqMJW2/NYtB30+7Vht7XjN2rpSFfKIs7+w5+JFFmuSSfGJ4nI+DbcCVwnAEbjlAr6x83fzhwyS1VV3GY5ZUSoNgty2ZrLYINZYmyUjTS+AmXiVZExTU/1/M2OtcfZvc52H2YclES9KIo/VBeqXAGh5L05HpO5U+axbjjr+Ekpwyk4il/XK2ZW+kqIllLqVteLsj+0MSRwirTNOi0zd1KPYjahIRFkTnBW2rks6YazfsN+T2dn+GHy9nrw7Vxi3tG2MumUcY4ma1QX6XJ99WpayNT7gsbZ5AOjjd8qOWyM7BZd1Gu7lLKHreqRIvLRpOBilJg1pdn9Ka1+bm72W6N7e4OfYkqnbkzAVxjoN+z2xZSGnYe2HyW63jTHICg+4zj1JYC/qCPHamveyN0s3KgC5ZfHq42+2mqm5rIueNMWGEKPP6InxKDVpSPPBIq19eZWXHBSNRm83o+lp1zJvNoEbN8JnZAd9T5So6dZruexaxWl4AcKIqk0kg6z5KNfZqzNVs4m9uVmcuW8WF481u12nVDeqoghBmgJadPcyS1iOPlGyqb8h7E6llL71qTN1U6dVaaoIpJmZnNa+Gg13qY5Klko6x1HDCoOoJGa+rwgqj5kmCwv9TTZU4pjfuUe5zj7uflW2cvEssPIQULvmFutYPAs8dksxXLjj5LplCcuRIrGb+gKAR0L+5McoNWmIW6gjy30pS1QljCWd46gWu82ijeOC9/uemFYp4G6XxRpoHf08TCEG+it06da8lyXvd53N74SG03YFePsY8NQlt3zlU5fcmthFceGGLXoxCi0Di265k+REWdp0H/qXMQHAzQCeCbdj0wqAjyYfVgK82hoOIyphR7cai7CvNOc4aljBtnTKr62ijSBr3PZA0Gy6YudlGaeROKifh46tmYUptDaXc9CDkvmdUBXTOvWsVcvF6euudXzqGnD11gqqDy9HO6+sCDnneh3p9Z11zC/P59rgIQ5c7jQGtNvtVH7OnTv3inPnzn3+3LlzTw2xbfvcuXNtQtr1ertdLrfbblqY+1Ovt9tXrti3v3LFfd9x3N+PPNL7WcB9L+wxy2X3tY7j9O/Ta9ugfZrj9Tovr+Pqn/Pbl/nZoDmwoe3/Rm2m/cLzM23nvNO+/ORqu+U3XxH3Hen6+s1X0HXs4Jx32rgXfT/l8+V2fTnGuXhw5fqVdn253nbOO+36cr195brP2IeMUT63HLDqYqpLm4QQfwjgX6WU/1fAdqzARVyUdRPULckLW2w1aOlRUFzVtk+vbYP2GSXGHWVb0yrUrNrUcyXSyDcwz02VTjWt2ihzEHJcerzVJM0lTFnEdQcVFw86DmPWqTKQpU1/DSB0ohdJiTSXQA2asN2SvLBtF+SCt8XP9Tnc23NLXpr4xWDN4iDqte52Vm50L6LkO+ix9vV14NFHezPMIy558sUr3yDKPs1wxNaWPVcgStgiZB6EHm+tOtXQGdhRY81ZZDwPaklT0HGYzZ09aYvxlyFaHJqkQdJlS0UQ8ygJZvp4HeewNnfYpUc20dPnsNEAnvhE1+F75UqyZDWzxaL5WidKwwyze5RqMqHWWydZ8mTi9ZAQZZ+2ZDGb6Eb5HoR8eNETvdaX1kMnQtkEyk+gs1irPCgRDDpO0ddhjwKpiLEQwhFC/GcA3w7gL9PYJ4lA0rXVRRDzuFbhwYHbvMErK9o2LpvomXO4tuZuH0YgG43+Kl3q2Ga7xqgNJLzQRUtHjd085yTfEa85sO3T63tgXt+ZGbvoLi8fPrBUKsDyMi5cvYAjbzyC0vkSjrzxCC5cveA/Lr9TidB20CZQfhZkFhnPgxLBoOMwmzt7oqwz9rrDVgDc2vm9B+AOKeWHAvbFmHGaJF17mzQmOOi13WHHGzVeGzX27PdZfU2wIuzchMkSNptlmMcyj5fFNTL3WakcNu1Qc+BVKvXCBbfvttk5yzLOI2e3evoMV50qbrz2RrKxhzk9S5x089rmQMtmFiVmTFIlccy43NmJ+bMP4G8B/D8AnhEkxCQDkq6tTroGOY2qZ1Gs67DjjTKuZctynSRx62azXxzVtVHn6jjAkSP95xzGU6FbhV6Cp5+zxeL0Jcz1ML93BwfhS6UuLR0+QB0cuK8B6zXThRhA3+ueYae4pthmDQ7aXRvFkh+G4xAfvNKss/zh0qaCEWU5iY2QS0xS20fY8UbZZ73ev6wo7HnYjuN3bNtyLn2bOMuUgvYZ9RrFuaa2OfQau9c5Wo5bfUO1Z0lS9Q1V7yEs19vl8+VMli6121ziQ1Ih+6VNYaGbesRIo+BFmqU/44zLVhryypVw52E7DuB9bK8ylOqcjxzpjTVXq27dbT9s+9TLd0ad3zjXI0qpVC+3uWUuL5Qfw5kHz2CvuYeqU8XqXas4ffvpw8NqLlbdhQwUr/sSIfBwU8fKfBZCTACYg1t963EA61LK/dhDI8NNGhW5bJW1Bjku8/hzc+EfKLyO43Xsqan+hC/9nE3RU2uI/cZjG79+/KjzG+d6qFKpYSq92SrCdYS4vbmJK5MOvvmbNlB99yJWzq7gxndq+31oqefhRq+wpcOsXzJMRMqmFkJ8sRDiFwF8FsBfwa1X/TEAnxVC/KIQ4uaUx0fGhbxriud9fP2YqvSlTlCGe9D4o55f3PkIyHDuxnQfPIWFJaCxc+nwoWF+HlhbQ6nZxMlP7eFv39TCjUvrbsayTxxdz3pWMOuXJCGPeuZRsqm/GMAHASwA+DxcEf4kgKcAeDqALwawBuDrpJSfC9gX3dQkG9JwmWc9JrP+tOkCbjSAkyd795GG237QWK7FwrsX7ZWcLBnpbQBrk8DTXung4D54us1TrQ5VxO8PGTgZVxxLnE39Y3CF+M0ATkgpT0spz0opTwM4AWAZQL2zHSH5kGYP6KzGVKn4Z4NPT7vxXn2bqan8C7NExXItPItLWDLSS3A7R9WO13wz6GOvgbVljBfx+0MGTh4Vx6KI8X8A8GdSyiUp5Wf1N6SUj0sp7wbwpwC+LcXxERKNNJZZpY05pmYz2AVsuomBdERikNXWLNfCc2mQVk5U+eqaJeDqrVVXXH3c5pGX5ag5OHnStcb1OS3i94cMnDwqjkUR49vgxoj9eB/ckpikKBSh1OUgSbMHtE6jAczOuqU3SyX332Hn0jamoMpRZux1ezsdkRik5Wc57zBWrPLhOfN1zHxw3RXXGNW2PLG1m1Rzqo+5VHLFf1z+75AueVQciyLG/wLgSQHbTHa2I0Vh1N1u5sPG8nI2iViLi70Z0Ftb4ecyjeSwtB4yBmn5mee9vIzp5y7i4qs2cfCOGi6+SKvytL3d+1nHSS66XuhzoFBzqo95YsIV/1H9v0M8yaMIShQx/jCA7xBCzNreFEKcBPCSznakKIy628182FhaimdBBXkQbPMWdi7TsOrSyvbOynNgwzzvpSXvB8NBjstW11vNqT5mvYraqPzfGTdP2RARRYx/GsAXAviwEOINQojnCyHmhRDPE0KchyvCXwjgZ7IYKInJIG9yeeD3sBHlxhPkQbDN2yDn0hQ2IN5NNc8lXH7XapDj0o9Vr7vFXWwPSaP4f2fUPWVDTKQKXEKI7wfwCwAmzP3ArVH9g1LKN4fYD5c2DYpRX6rh1wAhSnMEx+lv6qAvpWk03MYGylU9M+M2N8hrLgfdnCMO5ndPFS8p8ph1RvH/ThaV7vwYxTlMjnVpU+RymEKI2wC8DMBXArgJbgWujwF4UEr5aMh9UIyzYty+/H7nG+XGE6cEZZ4M+qYaB/OBYXranddx+W4WkUE/xA3DQ+PgsYoxG0WMGmk0bciasI0eBtnAIk5zhjwZhutszqma1zjXchAk/b4NA4M+x2H7fzUYrLoYqRwmGQKGIWHLL26lx3nn55PFt6LEIYctPphWjDXLhB5bolSRY5XjEE9Nc4lYGIbt/1WO+DaK6LikIyOlfCzecEhismi4kDZ+DwzqhqiKYyjSfrAw3dvLy262b1CDgzj7zsIdm0ZzDqB3vpUA2Zo4xM0AV/vJ8lqmRVEfZIc59GT7LhErQZbxVQDbEX+YK58neTc8CIPf07JtDahtuzD4WTppLYmKetykpG3J2gQoi/FXq9EspDyW4BTVihtmi33QlvgQEyTGj1l+Pgc3AG177zEAH89qsCQEw/Dl93tgMCsgVavxHyz8LJ0sraAs9227MScRLlOApqbcEpFpV/va33fHF/Za6p9dW3NDFlHOK86cpPEgm8VDRFEtdpIuXsFkr59z587de+7cuWbUzxn7YAIXsZNmgolfklOWCVBZ7tuWEJPkeOZ8z8z0J13FHX+S5B1b8leUcQwiwc32Xc3iuMOQrEeikFoCV7S1UIREIU3L3s/SydKdn9W+Gw13nwrlSk1iOdlqYJvs7aVjdUdx+9q2jXJeg7AmbV6KLI47DKEnkhhmUxM7ecTs0j6mn7Bn6c7Pat+Li667V1GpuDfmKKIXNMfmvqpVd5s48cokIrKy4h5bUSpFF/Os47824c3iuMMQeiKJoRgTO3kkjQxzosog2Nx0HbaKZtO9MUcRvaA5Nvd1cNArOGtr4R+WkojI9DSwvu6Wq3QcN2YcVcyztiZtwksrlsSEYkzs5JE0UsRElTSs9bQsfi+rK4roBc2xua+5ufzWCicV86ytSZvw0oolMaEYEzt5LPMo4tKSNKz1tCx+L6srithHnWP9mDpFeVjKEyW8ly65r0+dCvewxc5JxALFeFRJ+h8+D3dbEV18aVjraVn8XlZXFLGPOsf6Mev1wT4sDYtoRX3YYjiGWPBtFCGEaHq+6U1bShlU2YuNIrKGBdrTIY15zPpaJG0aEbbC06ArQQ3Ldzjq/A9Dkw+SJdZGEUGWcSnGD63tIpBH/HVYLBlFmPFGtSRt+8za4k/q3g9rqQ06HlrEHAIbUee/iOEYkj9eC5Cz/GHRjwGQR6GAYStOMCoFGpIWSkmjs04W3YCG5fsU9dzHoTsU8cOqi5H7GacB3dQDII/i8sPmfstivGnuc1DXsKiu+GFukECIN7Hc1GRYCXIpZuFSHjb3WxbjTXOfg0r0ScONnoVLmcuEyBhBMR5XsrjRmzf15eVix5CziOWmuc8sBM72EJaG6A3bg5hi2PIcyMhCN/W4MgiX8rBkwxaVLOYvq2syrC5lfkfJ4KGbmmjkVbs3L4poAQWNKQvLPatrMqwu5SJ9R8lYQzEeV/Kq3ZsXRSy0EDSmLASuSNckbeI8cI3yfJChgmI8ruRVuzcvimgB5TGmNNZNF5U4D1xF+o6SsYZiTLKjSK7LIlpAccaUVByjXpMiehS8iPNwU6TvKBlrKMZkPCiiBRRnTIMWxyJ6FLwo4gMXISHxrSFNyMigLKAiEWdMgxJHlR2tZ9wXXeBWVvozugkZEijGhAwTtVrvUpysxFFZ4DpF8Sh4UcQHLkJCQjc1IcNEFNd2kviyboED7vEYUyUkMyjGhAwTURKOksSXBx1/HaasbUIygGJMyKiSJL486IS3YcraJiQDKMYkmKysFn2/R464N35aRekxNeX/2g/2LiZkoFCMSTBZWS36fvf23JswraLxJIxbnK5sMsJQjEkwWVktZpJQ2vsfdpKKz/a2/+siEcYtTlc2GWEoxiSYJO5OP3RrSFH0tayDJKn4DFMRjDBucbqyyQhDMSb5oVtD1ephC7sir2UdJEnFp4hVx5IwTA8XhESEYkyCycrdqVtDN264FiDXsh6SVHxGre7yoB8uGKMmA4RiTIKhRZIPWYvPsInNoB8uGKMmA4RiTIIZNXfnsJC1+FBs/GGMmgwQijEJZtTcncSFYuNP1h6hYfNMkEyhGBMyrjD84E/WHiF6JogGuzYRMq6w5aA/WXeBomeCaNAyJsMNXX3xYfghX+iZIBoUY3LIMAobXX1kWGFiJNGgm5ocooRNrxFd9GbtdPWRYSVrNzgZKmgZk0OGUdjo6iOEjAAUY3LIMAobXX3ZMIwhi2GBc0ssUIzJIcMobExCygbG4rODc0ssUIzJIRQ2ohjGkMWwkObc0soeGSjGZDzhTcyfIocshv3apTm3tLJHBooxGU+CbmLDfsNPSpFDFsMuQGnOLT0YI0Op3W4P/KBCiDYASCkHfmxCALgi22wevnYc1z2vWFg4XOal+ixzGUoxCLp24wS/p8NIyfZHWsZkPAlyFdLiKC5FdqEPmiJ7MEgkKMZkPAm6ifGGX1woQIcw6XJkoBiT8SToJhb3hj/useZBQAEiIwjLYRJiI26pwmEsKUoIyR1axoSkCWPNhJAYUIwJSRPGmgkhMaAYE5ImTC4iNphLQAJgzJiQNGFbPGKDuQQkAFrGhBCSNcwlIAFQjAkhJGuYS0ACoBgTQkjWMJeABMCYMSGEZA1zCUgAtIwJIYSQnKEYE0IIITlDMSaEEEJyhmJMCCGE5AzFmBBCCMkZijEhhBCSMxRjQgghJGcoxoQQQkjOUIwJIYSQnKEYE0IIITlDMSaEEEJyhmJMCCGE5AzFmBBCCMkZijEhhBCSMxRjQgghJGcoxoQQQkjOUIwJIYSQnKEYE0IIITlDMSaEEEJyhmJMCCGE5AzFmBBCCMkZijEhhBCSMxRjQgghJGcoxoQQQkjOUIwJIYSQnKEYE0IIITlDMSaEEEJyhmJMSBo0GsDCAlCpuL8bjbxHRAgZIijGhKTB4iKwsQE0m+7vxcW8R0QIGSIoxoSkweYm0Gq5/2613NeEEBISijEhaVCrAeXOf6dy2X1NCCEhoRgTkgYrK8DcHOA47u+VlbxHRAgZIip5D4CQkWB6Grh4Me9REEKGFFrGhBBCSM5QjAkhhJCcoRgTQgghOUMxJoQQQnKGYkwIIYTkDMWYEEIIyRmKMSGEEJIzFGNCCCEkZyjGhBBCSM5QjAkhhJCcybUcphAiz8MTQgghg6YtpSyZf6RlTAghhORMqd1u5z0GQgghZKyhZUwIIYTkDMWYEEIIyRn2MyaEeCKEmAXwUwCeDeBWAI9LKW/OdVAREULcDmAbwFullK/IdzSE2KEYk7FACNEGAFsWo7bNVQAnAExJKa8OZmTFRQjhAPgdADMA3gbgEwD+Nc8xETKqUIwJIV5MAagDeIuU8r/kPRhCRhnGjAkhXnxJ5/c/5DoKQsYAWsaEhEAIcQeAHwHwNQC+AMBjAP5fAD8hpXzc2PYqAEgpb7fs514A9wB4npTygvb3NoD3AfiPAN4I4JsBPBnA90opf00IcWvn+IsAvhTAPoBPA/hTAPdJKRshz+OrAPwPAF8P4CYAnwLwBwDeIKX8pDEexT1CiHs6/z4vpbzXY99fCOA6gA9LKZ+j/f3fANgFcATAd0sp36a9JwAsd87zV7S/zwJ4HYA7AEwCuAbgPZ1xXjaOey86cwr3AeIHACwAuGa7BtrnygB+HsDdAH4bwHdKKemGJ7lAy5iQAIQQ3w/gjwA8B24M9efhis6PAviQEOLmlA51DMCfAXgWXKF/E4BPCyG+AMAHAfwQgEcBvBnA/wLwtwBeDNeVHOY8vgXAh+AK+nsA/ByATQDnAPxlJ9FJcR7AWzv/fl/n9XkAF7z2L6X8JwB/AeBrhBBfpL31HLhCDLjiqvP8zu/3auN8JoC/BHAXgA8D+Bm48/JdnXF+tccQfgjAr8B9UHoTgHd5jVUI8QQA74QrxMsAvp1CTPKEljEZKzpWlBc3W7Y/AeB+AP8E4GuklBvaexKukP0UgDRiql8ON1HqP0kpD7TjLAI4CeDnpZSvNsZXxaHQedKxWn8N7v/501LKP9He+1EA/xPALwF4IQBIKe8VQpwG8HIAF7ysYQt/DFd8vwGuxQ24AtwE8H5oYtyxTE8DaEgpH+38rQTg1wF8MYC7pJS/oW3/UgD/G8CDQoi6lLJlHPv5AJ4tpfyY3wCFEMcA/G5nnP9dSvmTIc+NkMygZUzGjXt8fm6ybH8XgCqAN+lC3OHHAXwewMuEEIGCGII9AD+sC7HB/2f+QUq5J6X8fIh9vxjALQDeoQtxh58FcBXANwkhboswXhvKwtUt4DsAfATAbwH4UiHEqc7fn94Z03u1bb8OwByAP9WFGACklO8A8AEANQDPtRz7l0II8Qm4XoavBfAyCjEpCrSMyVgRcmmTzjM6v//Ysq9dIcTH4FqBcwD+OuHwrkopP2P5+/sA/D2A/y6EeAaAh+EKyl9JKZsh9+13HgdCiPcDuB3AV8J188blT+E+NNwBAEKImzrH/int2HcAuIRDF7U+Js9xan9/bmec7zfe+4uAsdU643sigG+WUr43YHtCBgYtY0L8UdbyJz3eV3+/OYVjfcr2Rynl5+DGkX8VwFcB+AW4MdVPCSHOCyEmQux7IOchpdyDa71+uRDiSXDd0A6A90op1+FmZiur+Q4AbfQKb5JxWudP4xSApwBoAPhowLaEDBSKMSH+qEzpJ3u8/xRjOwBowdvrdLPPsTy7tkgpPyGl/F4ATwLwVACvAvCPAF7f+QkiznnE5Y8BlOBavncAuAHXkgeARwA8r+PW/3oAFw1vQJJxBnW9WYGbSf50AO8VQhwP2J6QgUExJsQfFYM8bb7RyaJ+OtyqVOvaW7sAbvWwWL0ygUMhpWxLKS9KKR8A8E2dP39riI/6nUcFhzHYNCxGPW78fAAf1DKV3ws3a/wcXHex6Sr2HKfx91jjlFL+BIBXw3VzP9JZMkZI7lCMCfHnQbhreu8WQswY770Bbtbvg1LKG9rf/wKuZfw9+sZCiFfAzeCNhBDiqcayI4USkn8JsZvfgbsc66wQ4lnGez8IYBrAe6SUSeLFio8A+CzcpLEF9Aqu+vePdX6bseEPwl1u9VwhxLfrb3RefwPcePMH4g5OSvnzcB8GFgC8TwjxJf6fICR7mMBFiA9SyqtCiB+Euxb1o0KIdwLYAfCNcJsnbMBdb6zzAFwhfnOnWMjHATwNbqbw7wP4lojDeAGAnxNCfKhzvM/ALfzxYrgu8Z8OcR7/JIT4TwB+E64A/SbcRK2vgruc6VMAvj/iuLyO1RJCvK8zPkATYynlY0KIK3CXajXhJqfpn20LIV4Od133O4QQvwv3nGtwPQCfh1s4xFzWFHWMvyiE+Fe467XfL4R4fkoPIoTEgpYxIQFIKSWAM3ALT3wbgNfAjd3+NNx1rdeN7dfgCugH4RbY+C9wly09G67VGJVVuIVGngBX4H4IroX4RwC+Xkr5f0Keh1pb+3DnfH4YwDyAXwTwVWGreIVECfDn4Cab2d77iFm9rDPOPwfwTABvhztnPwL3QeYhAM/svJ8YKeWvwV26dgKuIE+nsV9C4lBqt4NyHgghhBCSJbSMCSGEkJyhGBNCCCE5QzEmhBBCcoZiTAghhOQMxZgQQgjJGYoxIYQQkjMUY0IIISRnKMaEEEJIzlCMCSGEkJyhGBNCCCE58/8D1yNzy1dliMEAAAAASUVORK5CYII=\n",
"text/plain": [
"<Figure size 576x432 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Dataset X : ndim=2 shape=(1000, 2) Mean = [5.0032 6.9835] Std = [1.0171 1.5288]\n",
"Dataset y : ndim=1 shape=(1000,) Mean = 0.626 Std = 0.4838636171484688\n"
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
]
}
],
"source": [
"fig, ax = plt.subplots(1, 1)\n",
"fig.set_size_inches(8,6)\n",
"ax.plot(X_data[y_data == 1, 0], X_data[y_data == 1, 1], 'go', markersize=4, label=\"y=1 (positive)\")\n",
"ax.plot(X_data[y_data == 0, 0], X_data[y_data == 0, 1], 'ro', markersize=4, label=\"y=0 (negative)\")\n",
"plt.tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
"plt.xlabel('Hours of work')\n",
"plt.ylabel('Hours of sleep')\n",
"plt.show()\n",
"\n",
"vector_infos('Dataset X',X_data)\n",
"vector_infos('Dataset y',y_data)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 2.3 - Preparation of data\n",
"\n",
"We're going to:\n",
"- normalize the data\n",
"- add a column of 1 for bias\n",
"- Transform y_moons into a vector\n",
"- split the data to have : :\n",
" - a training set\n",
" - a test set"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"X_scaled : ndim=2 shape=(1000, 2) Mean = [-0. 0.] Std = [1. 1.]\n",
"X_train : ndim=2 shape=(800, 3) Mean = [ 1. -0.0116 -0.0056] Std = [0. 1.0066 1.0169]\n",
"y_train : ndim=2 shape=(800, 1) Mean = [0.6162] Std = [0.4863]\n",
"X_test : ndim=2 shape=(200, 3) Mean = [1. 0.0466 0.0223] Std = [0. 0.9717 0.9288]\n",
"y_test : ndim=2 shape=(200, 1) Mean = [0.665] Std = [0.472]\n"
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
]
}
],
"source": [
"# ----- Normalisation des données\n",
"scaler = sl.preprocessing.StandardScaler()\n",
"X_scaled = scaler.fit_transform(X_data)\n",
"\n",
"# ----- Ajout de la colonne de 1\n",
"X_scaled_1 = np.c_[np.ones((data_size, 1)), X_scaled]\n",
"\n",
"# ----- Verticalisation de y_moons\n",
"y_data_v = y_data.reshape(-1,1)\n",
"\n",
"# ----- Partage des données\n",
"test_size = int(data_size * test_ratio)\n",
"X_train = X_scaled_1[:-test_size]\n",
"X_test = X_scaled_1[-test_size:]\n",
"y_train = y_data_v[:-test_size]\n",
"y_test = y_data_v[-test_size:]\n",
"\n",
"vector_infos('X_scaled',X_scaled)\n",
"vector_infos('X_train',X_train)\n",
"vector_infos('y_train',y_train)\n",
"vector_infos('X_test',X_test)\n",
"vector_infos('y_test',y_test)\n",
"\n",
"y_train_h = y_train.reshape(-1,) # nécessaire pour la visu."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 2.4 - Have a look"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
"#### Train data :"
],
"text/plain": [
"<IPython.core.display.Markdown object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 576x432 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"#### Test data :"
],
"text/plain": [
"<IPython.core.display.Markdown object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 576x432 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pwk.display_md(\"#### Train data :\")\n",
"fig, axs = plt.subplots()\n",
"fig.set_size_inches(8,6)\n",
"axs.plot(X_train[y_train_h == 1, 1], X_train[y_train_h == 1, 2], 'o', color='green', markersize=4, label=\"Train / Positifs\")\n",
"axs.plot(X_train[y_train_h == 0, 1], X_train[y_train_h == 0, 2], 'o', color='red', markersize=4, label=\"Train / Négatifs\")\n",
"plt.tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
"plt.xlabel('Hours of work')\n",
"plt.ylabel('Hours of sleep')\n",
"plt.show()\n",
"\n",
"pwk.display_md(\"#### Test data :\")\n",
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
"fig, axs = plt.subplots()\n",
"fig.set_size_inches(8,6)\n",
"axs.plot(X_test[:, 1], X_test[:, 2], 'o',color='gray', markersize=4, label=\"A classer !\")\n",
"plt.tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
"plt.xlabel('Hours of work')\n",
"plt.ylabel('Hours of sleep')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 3 - Logistic model #1\n",
"### 3.1 - Build model"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"tf.reset_default_graph()\n",
"\n",
"X = tf.placeholder(tf.float32, shape=(None, data_cols + 1), name=\"X\")\n",
"y = tf.placeholder(tf.float32, shape=(None, 1), name=\"y\")\n",
"\n",
"initializer = tf.random_uniform([data_cols + 1, 1], -1.0, 1.0, seed=random_seed)\n",
"theta = tf.Variable(initializer, name=\"theta\")\n",
"\n",
"logits = tf.matmul(X, theta, name=\"logits\")\n",
"\n",
"#y_proba = tf.sigmoid(logits)\n",
"y_proba = 1 / (1 + tf.exp(-logits))\n",
"\n",
"#loss = tf.losses.log_loss(y, y_proba)\n",
"loss = -tf.reduce_mean(y * tf.log(y_proba + epsilon) + (1 - y) * tf.log(1 - y_proba + epsilon))\n",
"\n",
"optimizer = tf.train.GradientDescentOptimizer(learning_rate=learning_rate)\n",
"training_op = optimizer.minimize(loss)\n",
"\n",
"init = tf.global_variables_initializer()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3.2 - Training"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch: 0 \tLoss: 1.3204266\n",
"Epoch: 100 \tLoss: 0.25756028\n",
"Epoch: 200 \tLoss: 0.21319771\n",
"Epoch: 300 \tLoss: 0.1966115\n",
"Epoch: 400 \tLoss: 0.18770824\n",
"Epoch: 500 \tLoss: 0.18241489\n",
"Epoch: 600 \tLoss: 0.1788917\n",
"Epoch: 700 \tLoss: 0.17645508\n",
"Epoch: 800 \tLoss: 0.17437568\n",
"Epoch: 900 \tLoss: 0.17308258\n",
"Epoch: 1000 \tLoss: 0.17199597\n"
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
]
}
],
"source": [
"nb_batches = int(np.ceil(data_size / batch_size))\n",
"\n",
"with tf.Session() as sess:\n",
" sess.run(init)\n",
"\n",
" for epoch in range(n_epochs+1):\n",
" for batch_index in range(nb_batches):\n",
" X_batch, y_batch = random_batch(X_train, y_train, batch_size)\n",
" sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
" \n",
" loss_val = loss.eval({X: X_test, y: y_test})\n",
" \n",
" if epoch % 100 == 0:\n",
" print(\"Epoch:\", epoch, \"\\tLoss:\", loss_val)\n",
"\n",
" y_proba_val = y_proba.eval(feed_dict={X: X_test, y: y_test})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"### 3.3 - Evaluation\n",
"\n",
"Accuracy = Ability to avoid false positives = $\\frac{Tp}{Tp+Fp}$ \n",
"Recall = Ability to find the right positives = $\\frac{Tp}{Tp+Fn}$ \n",
"Avec : \n",
"$T_p$ (true positive) Correct positive answer \n",
"$F_p$ (false positive) False positive answer \n",
"$T_n$ (true negative) Correct negative answer \n",
"$F_n$ (false negative) Wrong negative answer "
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Accuracy = 0.948 Recall = 0.955\n"
"image/png": "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\n",
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
"text/plain": [
"<Figure size 1008x720 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"def show_results(y_proba_val):\n",
" y_pred = (y_proba_val >= 0.5)\n",
"\n",
" precision = metrics.precision_score(y_test, y_pred)\n",
" recall = metrics.recall_score(y_test, y_pred)\n",
"\n",
" print(\"Accuracy = {:5.3f} Recall = {:5.3f}\".format(precision, recall))\n",
"\n",
" y_pred_1d = y_pred.reshape(-1)\n",
" y_test_1d = y_test.reshape(-1)\n",
"\n",
" X_pred_positives = X_test[ y_pred_1d == True] # items prédits positifs\n",
" X_real_positives = X_test[ y_test_1d == 1 ] # items réellement positifs\n",
" X_pred_negatives = X_test[ y_pred_1d == False] # items prédits négatifs\n",
" X_real_negatives = X_test[ y_test_1d == 0 ] # items réellement négatifs\n",
"\n",
" fig, axs = plt.subplots(2, 2)\n",
" fig.subplots_adjust(wspace=.1,hspace=0.2)\n",
" fig.set_size_inches(14,10)\n",
" \n",
" axs[0,0].plot(X_pred_positives[:,1], X_pred_positives[:,2], 'o',color='lightgreen', markersize=10, label=\"Prédits positifs\")\n",
" axs[0,0].plot(X_real_positives[:,1], X_real_positives[:,2], 'o',color='green', markersize=4, label=\"Réels positifs\")\n",
" axs[0,0].legend()\n",
" axs[0,0].tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
" axs[0,0].set_xlabel('$x_1$')\n",
" axs[0,0].set_ylabel('$x_2$')\n",
"\n",
"\n",
" axs[0,1].plot(X_pred_negatives[:,1], X_pred_negatives[:,2], 'o',color='lightsalmon', markersize=10, label=\"Prédits négatifs\")\n",
" axs[0,1].plot(X_real_negatives[:,1], X_real_negatives[:,2], 'o',color='red', markersize=4, label=\"Réels négatifs\")\n",
" axs[0,1].legend()\n",
" axs[0,1].tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
" axs[0,1].set_xlabel('$x_1$')\n",
" axs[0,1].set_ylabel('$x_2$')\n",
" \n",
" axs[1,0].plot(X_pred_positives[:,1], X_pred_positives[:,2], 'o',color='lightgreen', markersize=10, label=\"Prédits positifs\")\n",
" axs[1,0].plot(X_pred_negatives[:,1], X_pred_negatives[:,2], 'o',color='lightsalmon', markersize=10, label=\"Prédits négatifs\")\n",
" axs[1,0].plot(X_real_positives[:,1], X_real_positives[:,2], 'o',color='green', markersize=4, label=\"Réels positifs\")\n",
" axs[1,0].plot(X_real_negatives[:,1], X_real_negatives[:,2], 'o',color='red', markersize=4, label=\"Réels négatifs\")\n",
" axs[1,0].tick_params(axis='both', which='both', bottom=False, left=False, labelbottom=False, labelleft=False)\n",
" axs[1,0].set_xlabel('$x_1$')\n",
" axs[1,0].set_ylabel('$x_2$')\n",
"\n",
" axs[1,1].pie([precision,1-precision], explode=[0,0.1], labels=[\"\",\"Errors\"], \n",
" autopct='%1.1f%%', shadow=False, startangle=70, colors=[\"lightsteelblue\",\"coral\"])\n",
" axs[1,1].axis('equal')\n",
"\n",
" plt.show()\n",
"\n",
"show_results(y_proba_val)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Step 4 - Bending the space to a model #2 ;-)\n",
"\n",
"We're going to increase the characteristics of our observations, with : ${x_1}^2$, ${x_2}^2$, ${x_1}^3$ et ${x_2}^3$ \n",
"\n",
"$\n",
"X=\n",
"\\begin{bmatrix}1 & x_{11} & x_{12} \\\\\n",
"\\vdots & \\dots\\\\\n",
"1 & x_{m1} & x_{m2} \\end{bmatrix}\n",
"\\text{et }\n",
"X_{ng}=\\begin{bmatrix}1 & x_{11} & x_{12} & x_{11}^2 & x_{12}^2& x_{11}^3 & x_{12}^3 \\\\\n",
"\\vdots & & & \\dots \\\\\n",
"1 & x_{m1} & x_{m2} & x_{m1}^2 & x_{m2}^2& x_{m1}^3 & x_{m2}^3 \\end{bmatrix}\n",
"$\n",
"\n",
"Note : `sklearn.preprocessing.PolynomialFeatures` can do that for us, but we'll do it ourselves:\n",
"### 4.1 - Extend data"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"X_train_enhanced = np.c_[X_train,\n",
" X_train[:, 1] ** 2,\n",
" X_train[:, 2] ** 2,\n",
" X_train[:, 1] ** 3,\n",
" X_train[:, 2] ** 3]\n",
"X_test_enhanced = np.c_[X_test,\n",
" X_test[:, 1] ** 2,\n",
" X_test[:, 2] ** 2,\n",
" X_test[:, 1] ** 3,\n",
" X_test[:, 2] ** 3]\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4.2 - A more readable version of our model. Yes it is.\n",
"...and with Tensorboard tracking and checkpoint recording."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"def logistic_regression(X, y, initializer=None, seed=42, learning_rate=0.01):\n",
"\n",
" n_inputs_including_bias = int(X.get_shape()[1])\n",
" \n",
" with tf.name_scope(\"logistic_regression\"):\n",
" \n",
" # ----- Construction du modèle\n",
" with tf.name_scope(\"model\"):\n",
" if initializer is None:\n",
" initializer = tf.random_uniform([n_inputs_including_bias, 1], -1.0, 1.0, seed=seed)\n",
" theta = tf.Variable(initializer, name=\"theta\")\n",
" logits = tf.matmul(X, theta, name=\"logits\")\n",
" y_proba = tf.sigmoid(logits)\n",
" \n",
" with tf.name_scope(\"train\"):\n",
" loss = tf.losses.log_loss(y, y_proba, scope=\"loss\")\n",
" optimizer = tf.train.GradientDescentOptimizer(learning_rate=learning_rate2)\n",
" training_op = optimizer.minimize(loss)\n",
" loss_summary = tf.summary.scalar('log_loss', loss)\n",
" \n",
" with tf.name_scope(\"init\"):\n",
" init = tf.global_variables_initializer()\n",
" \n",
" with tf.name_scope(\"save\"):\n",
" saver = tf.train.Saver(max_to_keep=4)\n",
" \n",
" return y_proba, loss, training_op, loss_summary, init, saver\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4.3 - Build the model"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"tf.reset_default_graph()\n",
"\n",
"log_dir = './run/logs'\n",
"chk_dir = './run/models'\n",
"os.makedirs(log_dir, mode=0o750, exist_ok=True)\n",
"os.makedirs(chk_dir, mode=0o750, exist_ok=True)\n",
"\n",
"X = tf.placeholder(tf.float32, shape=(None, data_cols + 1 + 4), name=\"X\")\n",
"y = tf.placeholder(tf.float32, shape=(None, 1), name=\"y\")\n",
"\n",
"# Build model\n",
"y_proba, loss, training_op, loss_summary, init, saver = logistic_regression(X, y)\n",
"\n",
"# Save model\n",
"file_writer = tf.summary.FileWriter(log_dir, tf.get_default_graph())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4.4 - Train the model"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch: 0 Loss: 0.6296 checkpoint: ./run/models/model-ckpt-0\n",
"Epoch: 500 Loss: 0.1132 checkpoint: ./run/models/model-ckpt-500\n",
"Epoch: 1000 Loss: 0.0973 checkpoint: ./run/models/model-ckpt-1000\n",
"Epoch: 1500 Loss: 0.0896 checkpoint: ./run/models/model-ckpt-1500\n",
"Epoch: 2000 Loss: 0.0858 checkpoint: ./run/models/model-ckpt-2000\n",
"WARNING:tensorflow:From /home/pjluc/anaconda3/envs/fidle/lib/python3.7/site-packages/tensorflow_core/python/training/saver.py:963: remove_checkpoint (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.\n",
"Instructions for updating:\n",
"Use standard file APIs to delete files with this prefix.\n",
"Epoch: 2500 Loss: 0.0834 checkpoint: ./run/models/model-ckpt-2500\n",
"Epoch: 3000 Loss: 0.0817 checkpoint: ./run/models/model-ckpt-3000\n",
"Epoch: 3500 Loss: 0.0805 checkpoint: ./run/models/model-ckpt-3500\n",
"Epoch: 4000 Loss: 0.0792 checkpoint: ./run/models/model-ckpt-4000\n",
"Epoch: 4500 Loss: 0.0786 checkpoint: ./run/models/model-ckpt-4500\n",
"Epoch: 5000 Loss: 0.0778 checkpoint: ./run/models/model-ckpt-5000\n",
"Epoch: 5500 Loss: 0.0772 checkpoint: ./run/models/model-ckpt-5500\n"
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
]
}
],
"source": [
"n_batches = int(np.ceil(data_size / batch_size2))\n",
"\n",
"model_file = chk_dir + \"/model-ckpt\"\n",
"model_final = chk_dir + \"/model-final\"\n",
"\n",
"with tf.Session() as sess:\n",
" \n",
" sess.run(init)\n",
"\n",
" for epoch in range(n_epochs2):\n",
" \n",
" for batch_index in range(n_batches):\n",
" # get a batch\n",
" X_batch, y_batch = random_batch(X_train_enhanced, y_train, batch_size)\n",
" # train\n",
" sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
" \n",
" # Calculation of logistic loss and logs\n",
" loss_val, summary_str = sess.run([loss, loss_summary], feed_dict={X: X_test_enhanced, y: y_test})\n",
" # Logging\n",
" file_writer.add_summary(summary_str, epoch)\n",
" \n",
" if epoch % 500 == 0:\n",
" print('Epoch: {:6d} Loss: {:8.4f} checkpoint: {}-{}'.format(epoch,loss_val,model_file,epoch))\n",
" # Save checkpoint\n",
" saver.save(sess, model_file, global_step=epoch)\n",
"\n",
" # Save the final model\n",
" saver.save(sess, model_final)\n",
" # Evaluation with test data\n",
" y_proba_val2 = y_proba.eval(feed_dict={X: X_test_enhanced, y: y_test})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4.5 - Evaluation"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Accuracy = 0.978 Recall = 0.992\n"
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAzIAAAJECAYAAADNORFFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAADHJklEQVR4nOzde3xU9Zk/8M+ZSwYQQ2IuoqWWYIWIEQMFRdMK2HppIaJt2S6C1bbiyvyksnRbRa211DVut6tY6uhK62ohWouthBjabjUFayr1AqgEggpJXbwQEhJiRCeTmfP74+TMnEzmcs7Muc583r54Jcwt35mMzHnOc/kKoiiCiIiIiIjISVxWL4CIiIiIiEgrBjJEREREROQ4DGSIiIiIiMhxGMgQEREREZHjWBLI+P1+0e/3c8oAERHpjp8xRET5wWPxz+cHDRGR8wlWLyAJfsYQETlf0s8YlpYREREREZHjMJAhIiIiIiLHYSBDRERERESOw0CGiIiIiIgcx+pmfyLKY6FQCIcOHcInn3xi9VJIhVGjRmHChAnwer1WL4WIaBh+njhfJp8xjgtkesO92BncibZgG0IIwQsvKn2VmOGbgSJ3kdXLIyINDh06hBNPPBETJ06EINh18BUBgCiK6O7uxqFDh1BRUWH1ckhPfd3A3hbg4G4gFAS8PmBSNTC1BigssXp1RKrw88TZMv2McVRpWUeoA/V99WgNtiKEEAAghBBag62o76tHR6jD2gUSkSaffPIJSkpK+KHjAIIgoKSkhGc7c82h/UDjOuDNl6UgBpC+vvmydPmh/dauj0glfp44W6afMY4JZHrDvWjqb8IgBhFBZNh1EUQwiEE09TehN9xrzQKJKCP80HEO/q5yTF83sP0JYDAEiMM/VyFGpMu3PyHdjsgB+G+Us2Xy+3NMadnO4M5oANN1rAvrG9fjSM8RlBWXYVntMpSOK0UEEewK7sK8MfMsXi0R6Y1lpUQ629sChMPS9509wNpNwAfdwPgSYOUioLxYun5vCzD7cmvXqsRSOMoW30M5wzEZmbZgWzSQWd+4Hp09nYiIEXT2dGJ943oAUmamLdhm5TKJyAB2Kiutr6/HO++8Y9rPIzLMwd2xTMzaTcD73UBElL6u3SRdLkak29kFS+EoWzZ6D9nl88Qu68iEYwIZ+eAFAI70HIEoigCk5qAjPUei1w1gwPS1EZFxjC4rdbvdqK6uRlVVFRYtWoTjx48nve2vfvUrHDlyBKeddlrC66+99lo89dRTAIDrrrsOe/fuBQDcfffdGa0tG++99x6+/vWvAwB2796NrVu3Rq/bsmUL7rnnHgDAkSNHcN5552H69On461//avo6yULyQRwgZWKGPlchitLfo7ezyecqS+EoWwa/h5z4eZJuHWo8+uijeO+996J/V65306ZNOPPMMzFvnjHVUo4JZLyIjWIrKy6L1tEJgoCy4rLodQUoMH1tRGQcZVlpMnJZaSZGjx6N3bt3Y8+ePSgoKMBDDz007PqwXHoD4Dvf+Q5Wrlyp6nF/+ctfYurUqQCsCWROPfXU6IdgfCBz+eWX45ZbbgEAPPfcc6isrMSuXbvwhS98wfR1koW8vtj340sAuT5dEKS/R29nk8/V+FK4Wx8Gvl0nfe3skS6XS+GIElG+h5LJ4j3kxM8TLetIJj6QUa73V7/6FQKBAP7yl79k9TOScUwgU+mrhGtouctql6G8uBwuwYXy4nIsq10GAHDBhUpfpZXLJCKdKctKk9GrrPQLX/gC3n77bWzbtg3z5s3DVVddhbPPPhvhcBjf//73MWvWLEybNg3//d//DUDKCN94442YOnUq5s+fj87OzuhjzZ07F6+88gpuueUWfPzxx6iursaSJUvw0UcfYf78+TjnnHNQVVWFJ598csQ65s6di5UrV+KCCy5AVVUVXnrpJQDA0aNHccUVV2DatGmYPXs2Xn/9dQDA9u3bUV1djerqakyfPh0ffvghOjo6UFVVhYGBAdxxxx148sknUV1djSeffBKPPvoobrzxRuzevRs/+MEPsHXrVlRXV+Ojjz7Ctddei6qqKpx99tm47777sn5NycYmVQPC0GHAykXAKSWAS5C+rlwkXS64pNvZgRNL4chelO+hZHR6D9np8+Tmm2/Gueeei8mTJ0cz78nWEYlE4Pf7cdZZZ2HBggX4yle+Ej0ptmbNGsyaNQtVVVW4/vrrIYoinnrqKbzyyitYsmQJqqur8fHHH0fXu2bNGrzwwgu44YYb8P3vfx+tra0499xzUV1djWnTpuGtt97K+nV2TLP/DN8M7AvuQwQRlI4rxeqlq0fcxgUXpvumW7A6IjKKsqw0lWzLSgcHB/GHP/wBl112GQDgpZdewp49e1BRUYGHH34Y48aNw8svv4xgMIiamhpccskl2LVrF/bv34833ngDhw8fxtSpU/Htb3972OPec889+MUvfoHdu3cDAH73u9/h1FNPRVNTEwDg2LFjCdfz0Ucf4W9/+xuef/55fPvb38aePXvwox/9CNOnT8fmzZvR3NyMb37zm9i9ezd+9rOf4YEHHkBNTQ36+/sxatSo6OMUFBRgzZo1eOWVV/CLX/wCgHT2DACqq6uHXffqq6/i3XffxZ49ewAAvb29Wb2mZHNTa4ADO4HBiNTYf/f1I2/jdku3swOnlcKR/SjfQylvl1ufJ4ODg3jppZewdetW/PjHP8azzz6LX/3qVwnX8eqrr6KjowNvvPEGOjs7ceaZZ0bXceONN+KOO+4AAFx99dV45pln8PWvfx2/+MUv8LOf/QwzZ84c9nPvuOMONDc3R69bsWIFbrrpJixZsgQDAwPDMlSZckxGpshdhPlj58MDTzQzI3PBBQ88mD92PqcXEeUYZVlpKpmWlcpnt2bOnInTTjsN3/nOdwAA5557bnRTrv/93//Fr3/9a1RXV+O8885Dd3c33nrrLTz//PNYvHgx3G43Tj31VFx00UVpf97ZZ5+NZ599FjfffDP++te/Yty4cQlvt3jxYgDAhRdeiL6+PvT29uKFF17A1VdfDQC46KKL0N3djWPHjqGmpgarVq3Cz3/+c/T29sLjyewc1aRJk3Dw4EGsWLECf/zjH1FYWJjR45BDFJYAcxYDHm8sMyMTXNLlcxbbZ4qT00rhyH6U76GUt8utz5OvfvWrAIDPfe5z6OjoSLmOF154AYsWLYLL5cL48eOH9bb85S9/wXnnnYezzz4bzc3NaG1t1fT6nH/++bj77rvxH//xH/jHP/6B0aNHa7p/Io7JyADARO9ELClcgl3BXWgLtmEAAyhAASp9lZjum84ghigHVfoq0RpsTVlelk1ZqVzTHO+EE06Ifi+KItatW4dLL7102G22bt2qee795MmT8eqrr2Lr1q1YvXo1LrnkkugZLqX4xxUEITrkJP7yW265BfPnz8fWrVsxe/ZsPPvss8OyMmoVFxfjtddew5/+9Cc88MAD+O1vf4tHHnlE8+PkvFwa3TphClC7QvF8BqSDODs+n0nV0mQpMSKVvsWPiwbsVQpH9qN8DyWTxXvIrp8nPp8UwLndbgwODqZch5zdiffJJ5/A7/fjlVdewac//WnceeedmjevvOqqq3DeeeehqakJl156KX75y1+qCthScUxGRlbkLsK8MfOwvHg5rim8BlN8U7AvuA+P9T2GQE8AzcebuSkmUQ6Z4ZsxIgsbz+iy0ksvvRQPPvggQiGpzO3NN9/ERx99hAsvvBC/+c1vEA6H8f777ydtZvR6vdH7vvfeexgzZgyWLl2Kf/u3f8POnTsT3keudX7hhRcwbtw4jBs3DhdeeCHq6+sBANu2bUNpaSkKCwtx4MABnH322bj55psxc+ZMtLUN7xc68cQT8eGHH6Z9nl1dXYhEIvja176Gn/zkJ0nXltdsNLpVN4Ul0j4xV90BXHOX9HX25fYKYgApsHK7pe/lUrhHVktfy4uly+1UCkf2o3wPJWPwe8iKzxMt6/j85z+P3/3ud4hEIjh8+DC2bdsGANGgpbS0FP39/dG+GUD9Z8zBgwcxadIkfPe738Xll18e7fPMhqMyMkodoQ409TchMvQfENtXYl9wH+aPnY+J3onWLpKIsiaXlcb//w5IAYwLLsPLSq+77jp0dHRgxowZEEURZWVl2Lx5M6688ko0Nzfj7LPPxuTJkzFnzpyE97/++usxbdo0zJgxA9/85jfx/e9/Hy6XC16vFw8++GDC+xQXF+OCCy5AX19fNCty55134lvf+hamTZuGMWPG4LHHHgMArF27Fn/5y1/gdrsxdepUfPnLX8b7778ffax58+bhnnvuQXV1NVavHtlfKHv33XfxrW99C5GI9BrX1dVl9HrlLOXo1nhiROo12f6ElOGwWxCQC+RSuO1PSJOllGfVBZd0AGqnUjiyHxu8h6z4PNGyjq997Wt47rnnUFVVhcmTJ+O8887DuHHjUFRUhGXLluHss8/GxIkTMWvWrOhjXXvttbjhhhswevRovPjii0l/5pNPPomNGzfC6/Vi/PjxCbNHWgmJShWM5vf7RQAIBAIZ3b833Iv6vnoMYjDpbTzwYEnhEpabEdnYvn37cOaZZ6q6bW+4N2/KSufOnZuwcdIOkvzOtNVDGCzbz5ikdmyJlaV09owsbSovlg6GJs+SMhpkjGGlfTYuhSNTafk84Xsotf7+fowdOxbd3d0499xz0dLSgvHjx5vys7V+xjgyI6PcV6LrWBfWN67HkZ4jKCsuw7LaZSgdVxrdV2LeGGM24CEic8llpfx/miyTaPyvqBj/e/f1sdGtDGSMI5fCaX2Nc6m3ibKT6XsoTyxYsAC9vb0YGBjAD3/4Q9OCmEw4MpBR7iuxvnE9Ons6IYoiOns6sb5xPVYvXR3dV4IHPUTkNHJNMtkMx/8616H9I8uJ5N6mAzulcqIJU6xdI5FNOOkzyHHN/sDwfSWO9ByJTvIRRRFHeo5Er8t2XwkiIqIojv91JmVvU/y0KjEiXb79Cel2ROQojgxklPtKlBWXRcfVCYKAsuKy6HWZ7itBREQ0wqTq2H4rKxcBp5QALkH6yvG/9rW3RcrEAFJv060PA9+uk7529kiXh8PS7YjIURwZyFT6KqPjWJfVLkN5cTlcggvlxeVYVrsMQHb7ShAREY3A8b/OlKi3KaLobQJivU1E5CiODGSU+0qUjivF6qWrce+Ke7F66WqUjisFYPy+EkSUHz788EM8+OCDCTejpDwjj271eGOZGZngki7n+F/7YW8T2YRVnyfvvvsuNmzYYOrPNIsjAxl5XwkPPCM2ynPBBQ88hu8rQUS5we12o7q6GlVVVaitrUVvb2/0uoGBAfj9fsyZMyftjsuPPvoobrzxRoNXO9wdd9yBZ599FoC0l8zx48ej133lK1+JPpef//znOPPMM7FkyRJT15eTJkyR9omZPGuoZ0aQvk6eJV3uhIbxvm5plPTja4DHbpO+7thivx4RvdbJ3iYyiR0/T3p7e7Fq1Sp88YtfzPgxOjo68Pjjj0f//sorr+C73/0uACAYDOJLX/oSqquroxs5m8mRU8sAYKJ3IpYULsmbfSWIyBijR4/G7t27AQDXXHMNHnjgAdx2220AgIKCAlufxVqzZk30+7Vr12Lp0qUYM2YMAGDr1q3R6wKBAP7whz+goqLC9DXmJCePbnXK9C491zmpOrb/z8pFI/f/AdjbRLqw4+dJUVFR1gGGHMhcddVVAICZM2dG9znbtWsXQqFQ9HmbzZEZGZm8r8Ty4uW4qfgmLC9ejnlj5jGIIcpRB3sO4qzAWfCs8eCswFk42HNQ18c///zz8e677wIADhw4gMsuuwyf+9zn8IUvfAFtbW0AgCNHjuBrX/saZs2ahVmzZqGlZWSD8KZNm1BVVYVzzjkHF1544Yjrt23bhgsvvBBXXnklpk6dihtuuAGRiHSw9sQTT+Dss89GVVUVbr75ZgBAOBzGtddei6qqKpx99tm47777AEi7KT/11FP4+c9/jvfeew/z5s3DvHnSyPmJEyeiq6sLN9xwAw4ePIjLL78c9913H7Zv347q6mpUV1dj+vTp+PDDD3V9DcnGnDK9S+91sreJEjl4EDjrLMDjkb4edO7nydy5c/H1r38dlZWVWLJkSbR07dVXX8WcOXPwuc99Dpdeeinef/99AMDLL7+MadOm4fzzz8f3v/99VFVVAZACli984QuYMWMGZsyYgb/97W8AgFtuuQV//etfUV1djfvuuw/btm3DggUL0NnZiaVLl2L37t2orq7GgQMHcMstt2Dq1KmYNm0a/u3f/k3X1zQRx2ZkiCj/1D5Ri7auNkTECNq62lD7RC1a/a26PHY4HMZzzz2H73znOwCA66+/Hg899BDOOOMM/P3vf4ff70dzczNuuukm/Ou//is+//nP45133sGll16Kffv2DXusNWvW4E9/+hM+9alPDSstUHrppZewd+9efOYzn8Fll12G3//+97jgggtw880349VXX0VxcTEuueQSbN68GZ/+9Kfx7rvvYs+ePQAw4jG/+93v4t5778Vf/vIXlJaWDrvuoYcewh//+MfodbW1tXjggQdQU1OD/v5+jBo1SpfXjxwgfnpXfGaivDg2vcvKbFMm60y32eWcxSMzPICUiXG72duUj2prgbY2IBKRvtbWAq3O/DzZtWsXWltbceqpp6KmpgYtLS0477zzsGLFCjQ0NKCsrAxPPvkkbrvtNjzyyCP41re+hYcffhgXXHABbrnllujjlJeX489//jNGjRqFt956C4sXL8Yrr7yCe+65Bz/72c/wzDPPAIjtM1NeXo5f/vKX0euOHj2Kp59+Gm1tbRAEIel69cRAhogcY3/XfkSGDkIiYgT7u/Zn/Zgff/wxqqur0dHRgc997nO4+OKL0d/fj7/97W9YtGhR9HbBoNQw/Oyzz2Lv3r3Ry/v6+kZkNWpqanDttdfin/7pn/DVr3414c8999xzMWnSJADA4sWL8cILL8Dr9WLu3LkoK5PGyC9ZsgTPP/88fvjDH+LgwYNYsWIF5s+fj0suuSTj51tTU4NVq1ZhyZIl+OpXv4oJEyZk/FjkMImmd4mK6V13Xx+b3mVlIKN1nWrL0GpXKIKdAaknRhnsUH7Zv18KYgDp635nf57I/5bLP7+oqAh79uzBxRdfDEAKrk455RT09vbiww8/xAUXXAAAuOqqq6IBSigUwo033ojdu3fD7XbjzTff1PT8CwsLMWrUKFx33XWYP38+FixYoOn+mWAgQ0SOMaV0SjQj4xJcmFKafS2/XNN87NgxLFiwAA888ACuvfZaFBUVJaz5jUQiePHFFzF69Oikj/nQQw/h73//O5qamlBdXY3du3ejpGT4gVJ8s6cgCEkn2RQXF+O1117Dn/70JzzwwAP47W9/i0ceeUT7k4VUIjB//nxs3boVs2fPxrPPPovKSo6qzwtOmd6lZZ3KMrR4YgQYjEjX165wdm8T6W/KlFhGxuWS/p4lqz5PfL7YQAu3243BwUGIooizzjoLL7744rDb9vT0JP1Z9913H04++WS89tpriEQimjP2Ho8HL730Ep577jn85je/wS9+8Qs0NzdregytHN0jQ0T5pXFxIypLK+EW3KgsrUTj4kbdHnvcuHH4+c9/jp/97GcYPXo0KioqsGmTtMeEKIp47bXXAACXXHIJfvGLX0Tvl+jD6cCBAzjvvPOwZs0alJaW4v/+7/9G3Oall15Ce3s7IpEInnzySXz+85/Heeedh+3bt6OrqwvhcBhPPPEE5syZg66uLkQiEXzta1/DT37yE+zcuXPE45144omq+l0OHDiAs88+GzfffDNmzpwZrdWmPOCU6V1a1snNLilTjY1AZaVUWlhZKf1dJ2Z/niQyZcoUHDlyJBrIhEIhtLa2ori4GCeeeCJ27NgBAPjNb34Tvc+xY8dwyimnwOVyYcOGDQgP/b+l9vOlv78fx44dw1e+8hWsXbvWlAEADGSIstAb7kXz8WYEegK4v+d+BHoCaD7ejN5wr9VLy0mTiieh1d+KwTsG0epvxaTiSbo+/vTp03HOOefgN7/5Derr6/GrX/0K55xzDs466yw0NDQAkEYZv/LKK5g2bRqmTp2Khx56aMTjfP/734827F944YU455xzRtzm/PPPxy233IKqqipUVFTgyiuvxCmnnIK6ujrMmzcP55xzDmbMmIGFCxfi3Xffxdy5c1FdXY1rr70WdXV1Ix7v+uuvx5e//OVos38ya9eujTaOjh49Gl/+8pczfLXIcSZVx/a/WbkIOKUEcAnSV7tM7+rrBsaMi/093Tq52SVlatIkqSdmcFD6Osm5nyeJFBQU4KmnnsLNN9+Mc845B9XV1dHm/V/96le4/vrrcf7550MURYwbJ/0/5/f78dhjj2H27Nl48803ccIJJwAApk2bBo/Hg3POOSc6bCaRDz/8EAsWLMC0adMwZ86clLfVi2DFJm9+v18EpJGgRE7VEepAU38TIkP/yVxD/80fOx8TvROtW6AD7Nu3D2eeeabVyzDdtm3bhjVOOkmS31nqTRFMxs+YJPq6gcZ1icuwZB5vrAzLbHKvy+AggDTHJvI6n743dtm366QgRuYSpOlkAAABuOYuvVdMNpKvnyeZ6O/vx9ixYwEA99xzD95//33cf//9Fq9KovUzhj0yRBnoDfeiqb8JgxgccZ0c2DT1N2FJ4RKOAycie7Dz9K5UvS5K8ev0+mI9NeNLYoMBEpWhpZtsRpQnmpqaUFdXh8HBQXzmM5/Bo48+avWSMsZAhigDO4M7o1mYrmNdWN+4Hkd6jqCsuAzLapehdFwpIohgV3AX5o1JXepD+Wfu3LmYO3eu1cugfGTX6V1qRi4DQGEpcNHS2DrVbnZZPlHKRtl9I1AiE3zjG9/AN77xDauXoQsGMkQZaAu2RQOZ9Y3r0dnTCVEU0dnTifWN67F66WpEEEFbsI2BTBqiKI6Y4EX2ZEUpMhnAjtO71IxcBoDjx4YHW1NrpEBkMBLb7DKeywV8cAAIj8ygJ5xsRo7FzxNny+Qzhs3+RBkIIVb+cKTnSPR/PlEUcaTnSPS6AVg8xtTmRo0ahe7ubh4gO4Aoiuju7uYGmmSMTEdDy+VyHm9skIFMcEmXj58U2y+Ek81yFj9PnC3TzxhmZIgy4IU3GsyUFZdFMzKCIKCsuCx6uwJYPMbU5iZMmIBDhw7hyJEj6W9Mlhs1ahQ30CRjaOl1iZeuXO6ZB5yxEShlhZ8nzpfJZwwDGYv0hnuxM7gTbcE2hBCCF15U+ioxwzfDNs3hTlijVSp9lWgNtiKCCJbVLhvRIwNI08sqfdxoMBWv14uKigqrl0FEVlPb65JsNHSqcjmnbARKWeHnSX5iIGOBRGN7QwihNdiKfcF9thjb64Q1WmmGbwb2BfchgghKx5Vi9dLVI27jggvTfdMtWB0RkcOo6XVxu6XbaZVNtoeIbI09MiZTju1V7j0CSGN7BzGIpv4mSzdUdMIarVbkLsL8sfPhgQeuuP+NXHDBAw/mj52f95krIiJV1PS6ZDoa2gkbgRJRRpiRMZkTxvY6YY12MNE7EUsKl2BXcBfagm0YwAAKUIBKXyWm+6YziCEi0sKo0dBGZnuIyFIMZEzmhLG9TlijXRS5izBvzLy8fx2IiHRhxGhoO28ESkRZYSBjMieM7XXCGomIiFSz60agRJQVBjImc8LYXieskYiISBM7bgRKRFlhs7/JKn2V0ebwZbXLUF5cDpfgQnlxuW3G9jphjURERESU35iRMZkTxvbaaY3cy4aIiIiIEmEgYzJ5bG/8Hi2AFBy44DJ1bG+yQGHO6DnY/vF2S9fIvWyIiCilvm5F30tQ2jOGfS9EeYOBjAXsMrY3ZaCAfZg7Zi46w52WrFG5l008eb1N/U1YUriEmRkionx0aP/ISWShIPDmy9K45TmLpSZ/IspZDGQsYvXYXjWBwrbj27CkcIkla+ReNkRElFRftxTEDIZGXidGpD1jtj8hTSpjZoYoZ7HZP0/FBwp1G+uwat0q1G2sQ9exLgCIBgpWSLSXTUSMRPeykdfXFmyzZH1ERGShvS1SJgYAOnuAWx8Gvl0nfe3skS4Ph6XbEVHOYiCTp+weKHAvGyKiPNbXDezYAjy+BnjsNunrji3S5YDUEyOXk63dBLzfDURE6evaTdLlYkS6HRHlLJaW5Sm7Bwrcy4aIKE+p6X0JBWO3/6AbGPoMgyhKf5eFeLKLKJcxI5OnvPBGvy8rLoMgCABgm0CBe9kQEeUhZe+LGBl+nRiRLt/+BOBRfDaNLwGGPsMgCNLfZV6e7CLKZczI5KlKXyVag62IIIJltctGNNMD1gYKdtrLhoiITBLf+7J2k5RhGV8CrFwElBdL1xeWAn1dUnCzctHI2wGA4JJGMRNRzmIgk6fsHijYbb8du+FGoUSUkxL1voiK3pe7r5eu/6gXcLul6WTlxdLl8dxuaT8ZIspZLC3LU3Kg4IEnWsIlc8EFDzyWBwryfjtVvqpoiVsBClDlq8KSwiV5uxlmR6gD9X31aA22RvuI5P1/6vvq0RHqsHaBRESZUtv7MhiSemU8XinzoiS4pMvnLOboZaIcx4xMHrPLxpypWL3fjt1wo1AiymleXyyYGV8Sy8gk6n2ZMEXaJ2Zvi5TJCQ1Il0+qljIxDGKIch4DmRykpeyIgYKzcKNQohzR1604AA9KB/BaD8D1eAy7mVQtTSdT2/tSWALMvlz6Q0R5h4FMjukIdYzoK5HLjvYF92H+2Pl5W5KVCxLt/yOKYnT/n9VLV0f3/2EgQ2RTasYLT5hi/GPY0dQaaf3sfSEiFdgjk0OUZUfK5nhAKjsaxCCa+pvQG+61ZoGUNbvv/0NEaagdL9zXnfj+ej2GXRWWsPeFiFTLi4xMvkx4YtlR7uNGoUQOp3a88N6W5OVSejyGnbH3hYhUyvmMTD5NeEpUdhQRI9GyIwDRsiNyJm4USuRwicYLRxTjhQHp+oO7jX0Mu5N7X666A7jmLunr7MsZxBDRMDmdkcm3CU8sO8p9dt//h4jSUDteOJTi32k9HoOIKAfkdEYmvtSqbmMdVq1bhbqNdeg61gUA0VKrXOCFN/p9WXEZBEEAAJYd5RAn7P9DRCl4fbHvx5dIY4WBxOOFjXwMIqIckNOBTL6VWrHsKD9wo1AiB5tUHWtiX7kIOKUEcAnS10TjhY16DCKiHJDTpWX5VmrFsqP8wf1/iBxKj/HCHFFMRAQgxzMy+VZqxbIjIiKb02O8MEcUExEByPGMTKWvEq3BVkQQwbLaZSPGEQPaS63sPspZLjvaFdyFtmAbBjCAAhSg0leJ6b7ptlgjEVFe02O8MEcUExFBkMutzOT3+0UACAQChv6c3nAv6vvqE04tk3ngUT21rCPUgab+pujEM5lr6L/5Y+ezP4GI8pFg9QKUzPqMISIiUyT9jMnpjIxcapUu+FATxOTbKGcz2D27RURERET2ldOBDKBfqVX8KOf4MrXScaXRUc5swE4vUXZL3qh0X3Afs1sJMPAjIiIiisn5QAbQZ8JTolHOoihGRzmvXro6OsqZgUxqzG5px8CPiIiIaLi8CGT0kG+jnI3E7JY2DPyIiIiIRsrp8ct6yrdRzkbKt41KsxUf+NVtrMOqdatQt7EOXce6ACAa+BERERHlC2ZkVDJilLMRnNBHweyWNnYqa7Ty/eWE9zYRERGZh4GMSjN8M7AvuA8RRFA6rhSrl64ecRsXXJjum27B6iRO6aPwwhsNZsqKy6IH5rme3cr0QNwugZ+V7y+nvLeJiIjIPCwtU0ke5eyBB664l80FFzzwqB7lbARlH4VyzDQglR0NYhBN/U3oDfdasj6lSl9l9DVcVrsM5cXlcAkulBeX2yq7pdQb7kXz8WYEegK4v+d+BHoCaD7erPr17Ah1oL6vHq3B1mhgIh+I1/fVoyPUkfS+dihrtPL95aT3NhEREZmHGRkN9BrlbAQnNdA7IbullG02INtmfTuUNVr5/nLSe5uIiIjMw4yMRvIo5+XFy3FT8U1YXrwc88bMs7xG30kN9HbPbinpkQ3Itll/hm9G9HWSA797V9yL1UtXo3RcKQDjAz8r319Oem8TERGReRjI5Ai79FGoJWe3qnxV0ZKoAhSgyleFJYVLbNPvoMfEsGwPxO0Q+Fn5/nLae5uIiIjMkRelZfkw7ciKBvpsX1c9Nio1mh4Tw/Q4ELe6rNHKAQ35OhyCiIiIUsv5jEw2TdZOYnYDfb68rnoEIXo161tZ1mjlgAYnDocgIiIi4+V0RiafdkQ3s4E+n15XPbIBdmjWT0RLRs3KAQ1OGw5BRERE5sjpQCafph3JfRTx07UA6SDPBZdufRT59LrqEYTY8UBc6yQ2M99f8az82URERGRfOV1alm/TjsxqoM+n11WPiWF2aNZXynQSm5UDGpwyHIKIiIjMk9MZmXycdmRGA30+va56ZQOsbtZXyiajZuWABicMhyAiIiLz5HQgw2lHxsi311WvIMQuB+J6TGKj/JiGSOQIfd3A3hbg4G4gFAS8PmBSNTC1BigssXp1RGSgnA5k7Npk7XT5+LraJQjRQz5l1IyitceIiFTIJCA5tB/Y/gQQDgPiUMY8FATefBk4sBOYsxiYMMWsZ0BEJsvpQMaOTda5gK+rs+VbRk1v+TS1j8g0mQQkfd3SfQZDIx9PjACDEen62hXMzBDlqJxu9rdbk3Wu4OvqbNyXJTvxPUZ1G+uwat0q1G2sQ9exLgCI9hgRkQrKgEQcPoBECkhC0vV93cOv29siBT4A0NkD3Pow8O066Wtnj3R5OCzdjohyUk5kZFLVqtupyTqX8HV1LmbUssMeIyKdxQckazcBH3QD40uAlYuA8uJYQDL78tj9Du6OBT5rNwHvdwOiKH1duwm4+3rp+oO7h9+PiHKG4wMZtbXqudLfYCe51DeST7gvS3bYY0Sks0wDklAw9v0HQ/cBpK8fKLI3If6/SJSrHB3IsFadAE6PygQzapljjxGRzjINSLy+2H3Hl8QCIEGQ/h69Hf9fJMpVjg5k8mmH+VyXaTDC6VGZY0YtM/k4tY/IUJkEJH3dwJhxwLFO6e8rF40sSQMAwSVNPiOinOToQIa16rkh02CEGTmywgzfDOwN7gWApD1GAgT2GBGpNalamk4mRtQFJMoJZ7LyYqkELZ7bLY1vJqKc5OhAxkm16ix/SiybYESvjBx/N6SVCDGr64lIYWqNNGJ5MJI+IEk1cnkYAfB4pLHNHL1MlLMcHcg4pVad5U/JZROM6JGR4++GtNoZ3Bn9Ptl7FgBLWonUKiyRAo74fWQAKRPjdscCkh1b0k84A4BxZcBFSxnEEOU4R+8j44T9MJQZB+V0KEDKOAxiEE39TegN91qzQIslCkYiYiQajACIBiPxss3I2eV30xvuRfPxZgR6Ari/534EegJoPt6ct+8Ju8vmPUtESUyYIm1cOXmW1DMDQfo6eZZ0ubwZZqIJZxHFhDPZ8WMMYojygKMzMk7YD4MDCVLLJhjJNiNnh98NM0LO46SSViJHKSyRxiun2vOFI5eJSMHRGRkn7DDPs7epueGOfl9WXAZBEABAVTCSbUbO6t+N0RkhZnqM4YU3+r3W9yyR5fq6pfKsx9cAj90mfd2xRbrcCby+2PfjS6TJZgBHLhPlKUdnZABr98NQ0ySu5extoCeQV03mveHeYQfwyUbZChCiwUj8ay7LJCNn9Zl1IzNCzPQYh+OXybGU077k8qxQUJoYdmCn1Icil3DZldYJZ0SU0xwfyADW7Ieh9kBRbflTsvvbhRGTvdQ2TQPAdN/0hK95MvIO9XPHzE26bquHRRg1PpxjqY3lhJJWohFSTfsSI9LEsO1PSP0odu4t0TLhjIhynqNLy6yipSRITflTqvvbQUeoA/V99WgNtkYP/OWgq76vHh2hjowety3YFh1Tm6y0C4iVnyV7zeMVoABVvirMGT0H245vS7ruT3k+ZemwCKMyQvGZnrqNdVi1bhXqNtah61gXAEQzPaSdE0paiUbY2zJ82tetDwPfrpO+dvZIl4fD0u3sTJ5w5vFKmRclwSVdzpHLRHkjJzIyZtNSEpTu7K3dBwAYeXZf7YH8IAZVveYuuFDlq8K8MfPQG+5FfV99ynUfGjwEAVJ9tRVn1o3KCHGjWONZWdJKlJFE075ExbSvu6+Xrj+4O3WzvV76uqWg6eBuqbzN65PKwabWpA9C5Aln0fsPSD0xau9PRDmDgUwGtB4ozh87P2lJlN0PNI3s49ByIK/1NVe77tM8p+HQ4KERvxu5NM3IM+tG9VpY3fuTL6woaSXKmJ2mfenRq6NmwlkuyiYAJMpBLC3LgNYDRfnsbZWvasTZdbsfaBo52UvL1DGtr7nadb83+N6I341cmrakcImhfUozfDOiz1/OCN274l6sXro62h+USUaIU7WIaAS7TPtS9uqIcWXCYkS6fPsTzpmiZqZD+4HGdVLAJwemcgDYuE66nijPMCOTgUxKgpRnbwM9AUubzLUw8uy+lqbpfcF9ml4zLetOdGbdiOEG8eRei0TZumwyQpyqRUQj2GXaV3yvTvw6yotjvTr5lm1JRc2whm2PA5d/l5kZyivMyGQg2/1Lsr2/mYw8u6+laVrra5bNuo0abpBIomxdthkhozI9RORgU2ukaV5AbNrXI6ulr+XF0uVmTPtK1KsTUfTqALFeHYpRNaxhEHipybo1ElmAGZkMZDt+1UnjW40+u6+2aVrra5bpuq0YXax3r4VRmR4icjB52ld8bwogZWLcbnOmfdmpV8dJ1AxrAIB390vZG2ZlKE8wkMlAtgeKTjrQNCPoUnMgr/U1y3TdRg43MBOnahHRCHaY9uX1xYKZ8SWxA3Kze3W0srrJPlUA+F6XlJWRM2ssy6M8Isi9A2by+/0iAAQCAdN/tp56w71ZHShme3+zJNuIUhlAmLV5p5bXLJN1K/uX6jbWDevFKS8ujwZEBSjA8uLlhj5XIgcRrF6Akm0/Y6w+GLaDHVtivTrJemQEFzB5ln0OxhNNWQOGZ7LSTVnL1uNrYsHMrQ9LwYvSqaWxrIzXB1x1h7HrITJX0s8YZmSykG1JkFPGt9rp7L6W1yyTdXN0sXZmDEYgcjw9Rg7ngqk10vMdjMR6deKZ0aujlpom++1PSJkuI4JROfgNK37+ykXADx4cfjuW5VGeYiBDqpgZdPWGe7Hj4x14K/RWNJMiQMAZBWfg/FHnazo41rpuozapzFWJsl7yYIR9wX2mZuuIbMvqg2E7sUuvjlpWTllLlgkqL5YyME4pyyMyUEaBjN/vLwJQA6AHwIuBQEBUXHcCgO8FAoE1uqyQbMGss+4doQ480/8MwggPu1yEiDcH3sTbA2+jdmytYQfHaoYEAFJGJtATyOvMgxWDEbLBzJFz5NxnDEcOD2eHXh211DTZy1PW9PzdpQp+AWtHaBPZiOZAxu/3nwXgWQBlkMY37/T7/V8LBAL/GLrJWAA/AuCcDxlKyayz7vKBcXwQoxRBBM/0P4OlhUsNOfhUMyRAlu+ZBycNRmDmyDly8jPGqoNhOysskZ6r3Z+vVVPW1AS/di/LIzJBJvvI1AF4EcA4AJ8CcBBAi9/vP0PPhZE9KM+6KxvmASmoGMQgmvqb0Bvuzfpn7QzujJ7d7zrWhbseuwsrf74SK3++Ej957CfoOiY1N4YRxq7grqx/XiKp9rZJRO/XwEnagm3R98T6xvXo7OlERIygs6cT6xvXA5Ben7Zgm5XLNPU9TLrIvc8Yjhx2Lq8v9v34EqmMCzC+nEvNfjtKggvweO1VlkdkgkxKy2YDmBcIBD4C8BGAf/L7/fcC2Ob3++cBOKbnAp3O6eUsZp51Vx7wrm9cHw1cAKD7WDfWN66PZkjagm2GneVPNCRAZvfMg5mcMhjBSZkjApCLnzFOHTlMQPlEaW8WwNxyLrXBL5Cf0++IhmSSkfEBGDazORAIrALwWwDbAZypw7ocpzfci+bjzQj0BHB/z/0I9ATQ8GEDNvZtNGWHeKOYedY9/sA4npkHx/KQgOXFy+GFN3q5nTMPZlO+LmXFZRCGzlTabTCCUzJHFJV7nzGTqqWDXUA6+D2lBHAJ0lf2NthXXzfwwYHY3+VyrkdWS1/lfVtcLv3LuVRngoZGLc++XFsQ09ctjcJ+fA3w2G3S1x1bpMuJHCSTQGY/gJnxFwYCgX+F9EHTkO2inKYj1IH6vvoRAUvHYAfCCDu6nMXMs+7xB8bxrDo4dkrmwWyVvspo+d2y2mUoLy6HS3ChvLg8OhjBBRcqfZUZPX6ikwPNx5s1/z/D35/j5N5nzNQaqXcBSH4wzN4G+9nbAkSGPr87e6T9W75dJ33t7Indbvzp+mdCjAx+D+0HGtdJo7/lzI88CrxxnXQ9kUNkUlr2NIDFAH4df0UgELjJ7/d7AOTNToGpJjfJnFzOYuY44kpfJd4IvgFAOjB+aPND0fKyknElw6aGZXpwnAmOZE5MzWAEF1yY7puu+bH1bM7n789xrP2MMWLTSqeNHCaJmiENANDZEbuPXu8fo/bb4ShwyjGCfHbSTLbddTkDzcebsSe4ByLEpAGLk3eIbz7eHB1HnOz5ueBCla8q64CsN9yL+r76lEEhALjhNmxqWSJmvgZOkyjgAKQAxgVXRtPA1LwPPPCoHuvM358pku66bIWMP2OM3sF92EGujUcO24ERAaVWj90W+/7bdVKzvcwlSFk1AIAAXHOX/u8fI96PO7ZImRcxknwamuACJs+y/0Q5yidJP2PSZmT8fv/dgUDgVn3Xkzv2BfdBHCrnluvvRVGM1t+vXrra0eUsRp51jydPDEu0j4zyZy0Yu8DUQQlmvgZOk2gwQgEKUOmrxHTf9Ix+T3o35/P3Z2+2+Ywx40y1U0YOWy3RAbxc+nRgZ/YBpVpahjQY8f4xYr8dtaPAD+zi+5QcQU2PzC1+v/9Bw1fiUMqzxp1HO4cFLJ1HOwHYuxE6nVTjiF1wwQMP5o+dr1tgMdE7EUsLl2KKd8qwn+eCC1MKpuDqwqtN3+/D7NfAaZSDEW4qvgnLi5dj3ph5Gb8eejfn8/dne/b4jInftyNRP4S8aSUZRxkQiMP7S6WAICRdb0ZT+qRqRE8Ep+tTMer9Iwe/V90hZX0yaexXUjsNbXDA2F6ZgweBs84CPB7p68GDxv0symlqAplfA/gXv9//+FBt8gh+v7/G7/f/Td+lOY/L5Ur4dyMaoc0kn3Wv8lVFA68CFKDKV4UlhUt0DyyK3EW4bOxlWFG8AjcV34Sbim/CiuIVuOyEyyw72DT7NchnRjTn8/dna/b4jFGzb4e8aSUZx04B5UmnIjpAL92QBqe8f9ROQwOA5g3AewdgiNpaoK1N+l22tUl/J8pA2tKyQCBwrd/v7wbwrwDGDe2w/AkA+P3+yQDuAbDQ2GU6QyTu7JH891woZ5HPuudzDwFfA3MY1ZzP35892eYzhptWGkttz4va0qeDu40tferrBl5+Jv3tZi2Q1u+U98+k6liPjLwvzvvdgNsFvN8lBYxyr4woAs8+Cly0VP9Svv37YxPhIhHp70QZUDV+ORAIfA/A7QC+DOB//X7/ZL/f/wCANwBcAeBVAJcYtUg7U44MLi8uH1ZCVl5cnvA+diln0Wu8bS7ga2EfRo91JvuxxWeMVTu45wMt437tEhCoyQwJLuDoe9L3Tnn/JBoFfkoJEI5IySdlBgmQAh4jSvmmTJH23wGkr1NM6HminKR6H5lAIHA3gBUAPg9gH4AbABwAsCgQCJwbCASeM2aJ9lbpq4QwVEOb7KALQPTAzC7lLMn2vnHSZp164WthLzN8M6L/v8jZzHtX3IvVS1ejdFwpAGdlM0kdyz9juGmlMbT0vMQfLFsZEGgtFXPK+0ceBa6UKmAEjCnla2wEKiuloKqyUvo7UQZU7SPj9/sFAFcD+N7QRQKA9wF8IRAI5PU2sPJEpEEMJi0h0zIq1gyp9r6Rx+g29TfZas3Z6g33YmdwJ9qCbQghBC+8qPRV4rPez+bda2GGZK/3DN+MtK+j3Jyfbqwzfx+5wxafMUbt25Hv4jMbicb9JjpQ7uwBBsOxA+yyolhAABgfEGjNDGXz/jF71PSEKYCnQGroB1JPZAOMKeWbNAlobdXv8Shvpc3I+P3+KyCl9/8HwHhI9crfG/r+Ob/fn7h+Kk84cSJS/Hjbuo11WLVuFeo21kU3oJTH2+aCVBmXzf2bo6Oe8+G1MIMeGS425+cP23zGyGeqPd7YmXWZ4JIu56aV2mnJbCgb4dduAo70St8LAuBxxxrsAeMDSq2lYpm+f7SU3enp9OnpM0hKVvf2ECWhJiPzewARSJNlbg8EAu8CgN/v/wDAowBa/H7/xYFAoMOoRWYq3VnhbM4aKxmxl4aREo23jd/7Rh5v6/TG6HTZJ6Vcfy3MoGe2j835ecM+nzFG7NuR7zRlNkR1twWM/10kaopXZpKAkaViWt8/avae2fY4cPl39X++ajJISlb39hAloSaQ+TOA7wcCgdeVFwYCgSf8fv8xAJsAvOD3+y8NBAK2yRMm2nFcPiu8L7gPM0fNxCufvJL0eq07ktvpoCtdgGbEeFsz1p0JNZsryuz0WjiV3ptZUl6w12cMN63Ul5ZNJQGVt1VkS4ySaamYlvePqrK7QeClJuBL38z+OcWvc85iacSyKKa+bSa9PQcPSiOV9++XGvkbG6VyMj0Y+djkOGlLywKBwKXxHzCK67YCuBTAWADbdV5bUukmTCnPCsefdY8ggkEMYscnO1Je39TfZNnEqmwmaKkp61FOWrPLZp1GNdyr2VxRZpfXwsn03syScp8dP2NIR1qa4O3UMJ9tqWFfN7BjC/D4GuCx26SvO7YMH2igpuwOAN7db8wGoBOmAF/61sjnFy+T3jAj94nhHjSkoKrZP5VAIPCC3++fA+CPOqwnrXSZlvlj5+Ng6KDqs/B2O2uc7vlN8EzAu4PvJi2VU1PWM6lgEt4eeBsRRLCsdtmI5w+YO95Wzbqf6X8GSwuXas7MqM0+AbDFa+F0ds32kXOZ/RlDOtOa2bDTwIVMSw0P7ZdKxsLhWKAi970c2CkFQBOmpC67e69LytLIfUF7W4zJEp56urRPTPx6ASnAcbsz6w0zcp8Y7kFDClkHMgAQCARe8/v9n9fjsVJRe6Au/x1I3vcgs1NfhJrn1zHYEb0shBD2BPdoCuAiiEAQBbjgQgQRW2zWqaYcKYwwth/fjoUnatsXT+3mikBubFxqNaM2s6T8ZtZnDBlAzmyoPVDWcluz1q+l1FBN38v2J6QAKb7s7r2u4beXNwAFjN0A1IjesClTpGxJJKL/PjFGPjY5ji6BDAAEAoEDej1WMmrr75XlYunOwtvprLGWfg6ZCDFaCidAUNXE3x5qt9V4WzXDBwCgY7ADveFeTeuq9FWiNdiaMvsk7wOkfP0Ae4z6NaJvyEhqXm9muCgTZnzGkEG0HCg7feCClnHTk6qB/X+XbrtyEfCDB4c/lpkbgOrdG9bYOLKPRSmbkdPpHpvyim6BjBnUTttSSncW3k5njdU8v1QZC1Ex8SVdgGanSWupypEOHz2MrmNd0SBOa7mfvM9PquyTG27Ujq3FgdABy18LJTVllHYbQ6zm9WaGiygPaTlQdvLAhUR9L6Ki7+Xu62Pjphf8v1ggU14MnFqafhiCU6TaJ0Zt6V0mj015x1GBjJZ+B7l0KtlZYZmdzhqne35dx7pQt6EO4Yh0tqfzaCzAUQYxgLoATc2kNTMyAvHlSIePHh52vTIro7XcT+3miqd5T8Np3tNsM0lLzzHGZmZ1uJklURJmb3pI1tAybrqwBPjUFKmZH1A/5tnJtJTe8f8LUsFRgYza+nsvvBAhpjwrLLPTWeN0z2994/poEANIZWVaAzgtAZqcEVBme0II4Y3gG9gT3IOLRl+EqlFVWT/vSl8l3gi+AUAKLO967K5h12db7qc2+2SnMi69xhhbkdWxU7aPyBayPQNNzqF13PS584EtB6Qxy3YYcKBWpoH57udiQUy60jsnZuTIdGnHL9tJpa8SrqElL6tdhvLicrgEF8qLy4cdqJ/pOxPzx86HB57o7WUuuOCBB7NHzU55vRVnjdM9v/igBcCIAE6+vxyg3bviXqxeujpamqU2QHsn9A629G/BIAZHZHsAKYh67uPn0BrMPr07wzcj+n3puFKcfNLJuo9BlrNPy4uX46bim7C8eDnmjZkX/R0bNf45U3qMMVYzhtyoMePpXm+ivKE8Ay0O//9QOgMdkq43YrwumU/rCOnCEmDuVZmPec6EmtHQqRzaDzSukwJxOWiTA/PGddL1ye7X/lrs78lGTsuld0QqOCojo6X+vshdlPas8JSCKbY6a5zu+cWXXbld7mGlcmf6zkSFtyLrsp6OUAe29G+JBjCpBg88e/xZ/F/o/zB79OyMX7MidxEmeiZGJ7KZXe6nZxmXXvQYY8zNKSV2yrRRHtLS/J3vZ6Bzofwuk400zRxwkG12MNPSMPl+SulK74hUcFQgo7X+Pl0PiJoeETOlen5A4gN8OaAQIKgO4BLpDfdix8c78FborRE/d33jenQe7YQIqfm+bkMdVl8dy/LsD+3HgdCBrMqU5oyZg0N9hzCIQdPL/cw64NdyQK3HGGO1wzHMHjNulESv76c8n8KhwUOOGphAOUZL83c+BzK5Un6nddy08n5GDTiQA8QDu4DBJAGC2v6UTAPzRPeLKKo9nD7cgCzjqNIyIFZ/X+Wrih7EFaAAVb4qLClc4tiDkt5wL5qPN2Nr/9ZoOVd82VuycjEAuGj0RSMCOLVlPR2hDmzs24j9of3DgpiuY12o21iHw0cPDysvC0fC0fImWbZlSnIQZ0W5nxm70WstXVNbRpkqQ5VPm1Mme307BjssKa0jitLS/J2vcq38Ts6wTJ4lZZUgSF8nz5IuNzMgU5aBKYOYzh7g1oeBb9dJXzt7pMvlICSZRIG5mtKwRPdTSlZ6R5SGozIyMrtlUrKVqCFbHPoPiDXwJyJAwBfHfBFn+c7K6GfLZVVhhEdcJx/UJ3L46GGsWrdK16xFpk3i2ZYOGX3An0npmh5jjPNlc8pUr68sn0vryGJam7/zUS6W39lhhHSqMrBMs4OZBubJ7gdIPUTKEjw7Djcg23JkIJNL1ByEpQpiLhp9UcZBDCCVVck/u+tYFx7a/BC6jnWludfQuhRZC73KlLQGqXpM5TL6gD+T0rV0ZZQCBEzwTMDjfY8nDd7yZXNKNa9vPpTWkU1NqpbOhouR/BivmwmW3xkjVYD4fheihRZasoOZBuZq7wfoP9yAcprjSsucQi4VC/QEcH/P/Qj0BNB8vHlEGUv8QVjdxjqsWrcKdRvr0gYUIkRs/3h7VqUxynKp9Y3rk/5MAQLcLjdccVNVrCxT0msql5oyLgA41XNqRuvMtHQtWRnlaZ7TIEDAO4PvpCxTm+GbodsUOztT8/rmemkd2djUGukMMxBr/n5ktfS1vFi6PN/PQLP8zhipysDcbimIALRlB7VOZdNyPwCoOMcZvVBkG8zIGEBLliBZQ7bcVC+KIsqKy/D1uV/HU9ue0r00Jr6sKhGX4Br2M+s21tmiTEmvJn01ZVwA0DHYgft77je1dC0+Q9Ub7kV9X73qMrV82JxSzeuby6V1ZHOZNn/nE5bfGSNVgBiOSEGE1uxgJlPZ1N7P4wWqv6jpKRIxkNGZ1n6IZAdhAKKbX3b2dOKhzQ8hIkZ0L42JL6tSjncGgJNPOnnEgb1dypT0mspV5C7CBM+E6PjnVOOmAWtL17QGb/mwOaWa19cu71nKU2aO13Uip5bf2X1cdKoA8ZSSzDbfVBuYA9LeNMrX5uQK4IODQCTCgJ50w0BGZ1oPNJMdhCmJooiwGB72d71KYyp9lXgj+AYA6WBP2SNTMq5kWGmVzOzxyMno2aT/7uC70e+TBUVKWvaX0bNXJZPgLdeGY8RT8/ra5T1rJO6XY3N2aP4G7HnwnelZfis5YVy0mgBRpiWYkAPz3c8BHW/Enr8A4NNTgeMfJn5t3nsbcLmAU88AOjsY0JMuGMjoTOuBZrKDMEEQEIlEIEI6s+wSXNGMjJ6lMXJZlbx/y+3X3J70thPcE/B++H2IEG1RpqRnpiNdUGR06ZraA2q7jlTW4yA608dQWxqo5ITSOi2vhx5DLygP2PXg22nld5luCmk2NQEiAHgKgNOnawsm+rqA/9s7/LJIRAps2l9LfB8xIpW0HT5o/WtDOYPN/jrTeqCZtCH76tUoPynWdH7DFTdktJdIOnIfhRvupLdxwYWFYxfia4Vfw9LCpbbZw0ePvVZkXnij35cVl0EYaoKUg6Js9pfRc4+cdOuUmdn3oXWPHL0fI93r64YbEz0TbfGeVUvL66HX0AvKcXbfq8VOe6+kEz8NLJP9WMwgB4geb6zRXia4pMu/+E1gyY+kTKHqICbNe0lm59eGcoYQX8ZkBr/fLwJAIBAw/WcbLdATiB54xDfFlxeXR88WF6AAy4uXA4idTQ0jPGzjSTU88KQtbVKjN9yLHR/vwFuht6IHQy64cEbBGZg9arYtz1qnanyXqX19mo83RzNjibIvd//6bkQU/0C7BBfuXXFv9O83Fd+kar3Z9qqkW2fpuFK44EKVr8qUUjI9fgd6/R71eH3tQOvrYZP3hGDUA2cilz9jMrZjS6zMKNleLYJLChysLn+zu8fXxHpPbn04ee+J1wdcdYd165QNKyfMopxLfhz5fQQkfy8BznhtyCmSfsawtExnmfRDKBuy9wb3pjyAkeldGlPkLsJlYy/DZbgs68cyi55TudKVJ+kx9UqPXhU9y9T0oMfkOL2mz+VKL5DW10OvoReU47hXi36cNi5aj/6sRGWJQPL3EuCM14Ycj4FMCpnU7Gd6oKk8CIs/s+yFF4WuQvRF+hBCSPOZ5lxuAtZrKleqoAiwz9Qru41U1uMgOt8PxOP//1RS83rYtW+KbMZpB992lm/jolP1BKV6L+XDa0OWYyCTRKbNs3ocaOp5ZjkfmoD1er0SBUUyO2Q/ZHYaqazlILo33JtwbUYciDsleE/0/6eSmtdDz6EXlMPy7eDbSE4dF52p+J4g5fMtLQKO9CZ+L+XDa0OWYyCTgNa9YOJZcaAZf+Dmggvi0H+JaBkfnE/koKjCW5Gyb0mAADfclk29Uhu8pTugz/aAX+1BNADU99UnDJz1PhB3SvCe6t8ZmZrXQ8/x3pTD8u3g20hOHBedjVRliWVFiTfWBPLjtSHLMZBJQI+afTPr9RMduMWf3c2m9yDfqDnABIDasbU4zXuaSavSLt0B/cxRM/HKJ69kdcCv5iBaJk/Pig+c9TwQz/YkhJnU/Duj5vWwW98U2VS+HXwbyWnjorOVqiyxqxd4JNnIewFQngjMxdeGLMdAJgEn1eyrPei2+/OwEzUHmAIEHAgdsG0go+aAfscnOxLeV8sBf7KDaC2Bs54H4noNDjCDmn9n1LweduubIpvKt4Nvo8njovWYBmZ3assSgdh7adYC4Oh7uf/akOUYyCTgpObZZAduJxWeBBEievp6hpWnAPZ8HmZRU0rlpEA2GTUH9LJsJ4XJB9HKoEnL66bngbiTfndq/51RSvZ62Klvimwsnw6+zaDHNDAnUFOWCAAuN3DGzOHvJTu9NsPGUAelAI3vfcdjIJOAk5pnkx24dR3rit6ms6cTLsGFCCK2fR56ShaslLvKsf3j7WlLqZwUyCaj5oBelu0Bv3wQ/VjfY9HLtL5ueh2IO+l3p6W/CEDa1yNXxk+TwfLl4Jv0o6Ys0eOVgmS7BgSJxkeHglKAdmCnlI2006arpBoDmQSc1Dyb7MBNSRRFRMQIyk8qt+3z0EuyvpA9wT2qBx84KZBNRsvZfj0O+IvcRVm/bnociDvpd6f23xmzNjclIkrI6WWJqcZHixEpQNv+hL0DMUqKgUwCTmqeTXbgpiQIAsqLy239PPSQqi9EGcSkK6UyO5A1YlSwlrP9eh3w2+EEgB3WoJaT/p0hoiw5vazJyWWJqcZHr1wkZZnCYel2zFQ6jpDoDL7R/H6/CACBQMD0n61Wsv0dlDXqdhjh2ny8OXrglqpHJr4vwm7PI14mB/fJXov451+3sW7YgbsyyCtAARYXLkZ9X33KAQoeeHSZfGXU+0ztawEkD+y0ZgN6w72mvW52XoMWTvl3RgXB6gUoOeEzhvJIorImYHg2g2VNxnl8TWxYwa0PDx9WcEpJrFTO6wOuusO6dVIqST9jmJFJwinNs2rO6sbzwoszfWfa6nkoZboPiNq+kHSlVGZNgTJyVLCW94Ve2QAzX7dUQa6TJng55d8ZIsoQy5qsl2p89AfdittZ3ztJ2jGQScEJzbOpDtyUnHKGN5uDe7V9IWpKqcw4wDRyVLCaA/pE+8gor8/kgN/o101tkOuk4MAJ/84QUYZY1mQ9teOjvdb3TpJ2DGQcIN0Z6EQHbi64AEgH/3Y+iIuXzcG92r4Qtb0TRh9gGj0qWM0B/ZSCKbof8Bv1umkNchkcEJHlDu6OlZOt3RQ7iH6/W/r73ddL1x/czUDGKGrGRwsu6XbkOAxkdGJEwzag/gx0rhy4ZXNwr3aXebs0VpsxKjjd+8JJ7xsnbXZJRA5hdBM+y5qsp2Z8tNst3Y4ch4GMDjLt6UjHyB4Ku8rm4D6TfiHAut4JO4wKNioAN4KTNrskIgcwY28RljVZz+njoykll9ULcDplsBHfnxJBBIMYRFN/E3rDvZofO/4MdN3GOqxatwp1G+uiG17KZ6BzhRfe6PdlxWUQBGlQhZqDe7kvxANPtLROjTO8Z2BJ4RJTe4c6Qh3DAtRltctQXlwOl+BCeXG5KaOCO0IdqO+rR2uwNRpQyQF4fV89OkIdhvzcTDlps0sisjllE74Y11sqRqTLtz8h3S4bk6qlg2VAKmM6pQRwDU3LYlmTeeTx0ZNnScElBOnr5FnS5Zwa51jMyKiQ6qy1keUu+XgGOtt9QJR9IXuCe9L+blxwwefymZp9kINf5d42Zpe7OTHbZ4cMFhFZTK9SMLOa8FnWZB+FJdLvUv59yu+lZx5w5t4+BIAZmbTSnbWWS5mAWLARESPRYANANNjQKh/PQM/wzYhmU+SD+3tX3IvVS1dH9z5Jd3Av93244Y5epvfvJlO94V409jdGA4hkmTYAECAYVu7mxGxfpa8y+t6wIoNFRBY7tB9oXCeVfsnlWnIpWOM66Xq1EjXhRxRN+ECsCT8bclmTxxvLzMgEl3Q5y5rMp+d7iSzFQCYFNWVjyjPaegcbasusAGkDxEzK1+wmVXmYCy544FF9cG+3QFAOio9GjkYvSxZgAdLGjUaVuyXK9lkd5KWjR5BLRA6ldymYmU34LGuyF7PKCskULC1LQU3ZmJLe5S5qp3AByHqwgJ3otQ+InUqRkpVypQqwlIGY3uwW5KnhtM0uiUhHepeCmd2EH1/WRNbh3j45hYFMCmp3ipdl0tORipYpXHbta8iUHmOBs+230VOyoFgQBMitMqkCLC3TxdTc1k5BnhZO2+ySiHSi934savYWAYDyifo/F7IW9/bJKQxkUlB71lqmd8N2qjPQ3EcjPTWBoFmlSMmCYkEU4Ha5IYpiwgCrN9yL7ce3o2OwY9jjJRvvrXYUuJ2CPK2ctPcNEelE71KwqTXA268C4RRN+ADwwQGpxIg9LLnDjnv7GL2fUQ5jj0wKantUPEP/ZdvTkYh8BrrKVzXscqf0NVhJz36bbCUNiiFCFMWEvR5lrjJpDHJcECOLH++tZRQ4+02IyFG8vtj340ukEjAg81KwwhJg/Omxv3f2ALc+DHy7Tvra2SNdHolIB5iUO/R+L2WLgweywkAmBbVTkqb6pkaDDbkUpwAFqPJV6bI/iXwGWskpfQ1WUwaCRvxu1NIyuMEDD+aOmYvtH28f1lOTbrqY1klkEzwTEq7V7CCPiCgtI/Zj6eyIfW/k5DKyFzvt7cPBA1ljaVkKWkqTzCh3cWpfg9XsUIqkdnBDiasEC8YuSNhTc/jo4ejtEu0lJEJUte/Q3uBe7A3uHZG1kZ3mOQ1zxsxhEENE9mHEfix2LDEi49lpbx8OHsgaMzIp2Kk0CeA+Gk6mppTLAw8WjF2AIndRwp4apURZOLU9XfLY8GSBzKHBQzo8YyJylL5uYMcW4PE1wGO3SV93bLHPmWAj9mOxW4kRmcNOe/uYtZ9RDmNGJg07TUmyU/M6aaN1dHCyoESWKAsnQlSVsZNxYAQRAZBq8Lc/IZ35lQ+q5Br9Azulgzo77HUi78cSbYoekIKMTJui1UwuM6vEiMyl93spU8wKZo2BjAp2KE2S18F9NJxLS1CcrIxQligLJ0JUve8QkLr8rC3YZvn7nYhMoKzRjydGpPKb7U9IB312mJ6k534sdioxIvPZYW8fs/czykEMZBzGThki0k5tUJyup0a5GasyC6d23yGAAyOICPldoy+XGMVnowApE+N2m1dipBbH9OYWZgWzxkDGgeySISLjqN0M1Q33sCxcuoydCBFhSActHBhBRHm/OaBdSozUcEoJIKnHrGDWGMgQ2VCqMkLZRM/EEdPF0mXsdgZ3OnYjTCJS0OvMPGv07VFilI7TSgBJHSdmBW2GgQzZXm+4FzuDO9EWbEMIIXjhRaWvEjN8M3K6lC7TMsJUGTsOjCDKAXqemWeNvjPkcwlgrnNSVtCGGMiQrXWEOkZkJUIIoTXYin3BfZg/dr5pm1paQe8yQg6MIHI4vc/Ms0bfGXKxBJD9PjFOyAraFAMZsq3ecC+a+puG7W4vkw/Cm/qbsKRwCQ+8NeDACCIH0/vMPGv0nSHXSgDZ70M6YSBDtpVod3vueaIPDowgcii9z8yzRt8ZcqkE0Mh+H2Z58o4r/U2IrJFod/uIGInueQIguudJvugN96L5eDMCPQHc33M/Aj0BNB9vRm+41+qlEZEZjDgzL9foT541tNu9IH2dPEu6nGfGrTepOrYL/cpFwCklgEuQvjqtBDA+q3jrw8C366SvnT3S5XJWUYtD+4HGdVJWR/7/RM7yNK6Trqecw4wM2Vay3e3zdc+TfO8XIiIYd2aeNfr2lkslgEb0+3CqW95iRoZsywtv9Puy4jIIggAAebnnibJfKH4UcwQRDGIQTf1NzMwQ5bpcOjNP6sklgB5v7PcvE1zS5U4pATQiq2hUlodsjxkZm8rXkcNK6Xa3B/JnzxM1/UKDGERjfyNqx9bmzXuEKO/k0pl50saJY3oT9ay4XEBk6IScXlnFXJzqlg77gQAwkLEllhBJuOdJTKJ+IVEUo/1C8mtzNHIU9X31efMeIco7bM7Pb04qAUw2mUxJr5HfuTbVLR1OfYtiIGMzHDkcwz1PYtT2CwGIlpkp3yPM8BHlECeemaf8kqpnRUmvrGIuTXVLh/1AwzCQsZl8HTmc6kC7dmwtth/fjqORo9HbF7mKMGfMHJzmPc3CVZvHC280mCkrLotmZACpZ2jVulVJ3yPM8BHlICedmaf8o2a/o0QyzSrm08aueu8l5XBs9reZfBw53BHqQH1fPVqDrdGDdflAe0PfBmzp34KeSM+w+xyNHMXm/s3Y88keK5ZsukpfJVxD/7suq12G8uJyuAQX3C43IpFI0vcIhwQQEZHpEvWsRBQ9KzLBpc/I76k1UgAExLI8j6yWvspBU670jql5beV+oDzAjIzN5NvI4XSldKmIEPHcx8+hbaANXzrhSzldIpWsX2jVulUQkfw9kq8ZPiIispDanhVRBK66I/ufl0+9Y/nWD5QGAxmbSVZClKsjh9UcaKe77t3wu45scNfSt6LsF1IGfeneI2qGBMjZGwYyRESkCyt6VvKldyyf+oFUYGmZzSQrISovLs/JkcNqSumi1x2Vrjt89DDqNtSh61hX9HqnlUilKqer76tHR6hjxH0meidiSeESnOQ6KXpZuvdIvmX4iIjIBqza70juHbvqDuCau6Svsy/PnSAG4F5ScZiRsZl8Gzms9kD7SM+RaAkVAIQjYdz12F04+aSTsy6R0mOil5bHyGYyXZG7CLVja1HfV49BDKZ9j+wL7svpDB+nsRER2RD3OzIOX9thmJGxGbmEyANPNDMjc8EFDzw5NXLYC2/0+7LiMgiCAAAjDrSV3ytlOwQhk8xIto8RX05Xt7EOq9atQt3GWJZJDsoS0fIeyeUMnx6/OyIiMoDcs+LxxrIHMsElXZ4rPStm42s7DAMZG5JLiKp8VdEz5QUoQJWvCksKlziqDyQdNQfa8nVul3vE/bMpkdJjolcmj6HHZLpk75EzCs7A6d7TsbV/K+7vuR9twbZoJkvO3ty74l6sXro62mPkxAwfp7EREdmc3LMyeZY+k8kohq9tFEvLbKrIXYR5Y+blTAN2shKgz3o/m7aUDhg6CL96NdY3rsfho4ejl2dTIqXHRK9MHkOvvpX490iy/WIEDGW5IAwrz3PypqKcxkZE5ADc78g4fG0BMCNDJkhVAtTY34iZo2YmLZNSXiYHOrdfcztOPunkrEuk9MiMZPIYasvptARlqTIUyuBF/tlOz/Dl435LRERENBwzMmQoNY3tr3zyCmrH1uJA6ADagm0YwAAKUIBKXyWm+6bj3cF38ezxZ6P302sIgh6ZkUweo9JXidZgKyKIYFntshHZBPm5aAnK1GQoBAg403dmTmQoOI2NiIiIGMiQoZQH2MlEEMGB0IGkpXRF7iJ8yvMpbD++HR2DHSOuz7RESo89ezJ5DCMm0+XbfjH5tt8SERERjcTSMjKU8gA7GTUlQEXuIiw8cSGuKbwG03zTdBmCoMdEr0wew4jJdPmWobDbNLbecC+ajzcj0BPA/T33I9ATQPPxZg4bIOfq6wZ2bAEeXwM8dpv0dccW6XIiIptgRoYMpTzATkXtAbaeQxDUZEYiiOD14OvYF9yXcH+STLMr8tSxXcFdCcvptDbf51uGwk77LSUbstAabMW+4D7MHzvfkX1IlMcO7Qe2PwGEw4A4dCIqFATefFnav2LO4ryaikRE9sWMDBlK2dieihUH2KkyI/GS7U+STXZFDsqWFy/HTcU3YXnxcswbMy+jCWJ2y1AYzS77LXEMNOWcvm4piBkMxYIYmRiRLt/+BDMzRGQLzMiQoZSN7clYeYCdKDOSjHzGvam/CUsKl0QPkvXOrmTCThkKs9jhdecYaMo5e1ukTAwAdPYAazcBH3QD40uAlYukncTDYel2eT72lYisx0Amxx3sOYjaJ2qxv2s/ppROQePiRkwqnqTLYyfbG0ZZfqU8wE5G6wG2mp+rhbJcrfl4czTw0nJgavW+P3KGIr7ECXD2fjHp6PG6Z/N+yrchC5QHDu6OZWLWbgLe7wZEUfq6dhNw9/XS9Qd3M5AhIssxkMlxtU/Uoq2rDRExgrauNtQ+UYtWf2vWj6u2L0DvA2yj+xGcfGBqVIZC78DRTrJ9P+XbkAXKA6Fg7PsPhoIYQPr6gaKcLMT3NBFZj4FMjtvftR+RobNrETGC/V37s35MNXvDKMuv9DrA1vpzM+H0A1O9M0O53Miux/sp34YsUB7w+mLBzPiSWEZGEKS/R2/H9zQRWY/N/jluSukUuATp1+wSXJhSmv2kGbV7w+wK7or+XY/G9kx+rlbK4QRlxWUQBAEA8vLANNcb2eP7W+o21mHVulWo21iHrmNdANK/n/JtyALlgUnVwNBnBlYuAk4pAVyC9HXlIulywSXdjojIYgxkclzj4kZUllbCLbhRWVqJxsWNWT+mXnvD2PHn8sA0Ro8DfTtLVEYYESPRMkIg/ftphm9G9P0iD1m4d8W9WL10NUrHlQLIvSELlOOm1gBut/R9ebHUE/PIaulrebF0udst3Y6IyGIsLctxk4on6dITo6T33jB2+rlWTv+yWy+Kk/uF1NCjjDBfhyxQDisskfaJid9HBpAyMW63dH1hSfLHICIyCQMZ0kzZF5CK3uVXZvxcqw5M7diL4vR+oXT06m+xwxhoIl1NmALUrpBGLB/cLTX2ewukcrKpNQxiiCi1vm7Fvx9BqffOoH8/GMiQZlbtDWPWzzX7wNSMIQaZyPVGduX7aVntshGjtgH17yerx28T6a6wRBqvzBHLRKTFof0jM7qhIPDmy8CBnVJGd0L2/doyBjKkmRF7w9jt55p5YGrXTRX1PNC3o3zcRJSIiMgwfd1SEDOYoHpGjACDEen62hW6ZWbY7E+ayeVXHniijc4yF1zwwGNI+ZVVP9doapvOXw++jubjzaZNCcv1RvZcfT8RERFZYm+LlIkBgM4e4NaHgW/XSV87e6TLw2HpdjphIEMZkcuvqnxV0dKiAhSgyleFJYVLDOvnsOrnGkltLwoAtAZbUd9Xj45Qh+HryocD/Vx8PxEREVni4O5YOdnaTdI+VBFR+rp2k3S5GJFupxOWllHGrOoLyLV+BLW9KID5PTP50Miea+8nIiIiS8ib6QLAB0Ob6QLS1w+6FbfTb0gQAxkii6XqRfn63K+jbmOdpT0zPNAnIiKitLy+WDAzvkTKxIgiIAjS36O3029IEEvLiCyWqhflqW1PZbRRIxEREZGpJlVL+00BwMpFwCklgEuQvq5cJF0uuKTb6YQZGSKLpdq7Jhf3byEiIqIcNLVGGrE8GAHKi4G7rx95G7dbup1OmJEhsgFl07lSWXEZBEEAgJzZv4WIiIhyUGGJtE+MxxvLzMgEl3T5nMW6borJQIbIJuRelLN9Z0dLzZbVLkN5cTlcggvlxeU5sX8LERER5agJU6R9YibPknpmIEhfJ8+SLtdxM0yApWVEtsONGomIiMixCkuA2ZdLfwzGjAyRzeTD/i1ERERE2WJGhsiG8mH/FrJOb7gXO4M70RZsQwgheOFFpa8SM3wz+N4iIspWX7e0e/3B3dI4Yq9PmtQ1tUbX/hBiIENkW9y/hYzQEeoYMSEvhBBag63YF9yH+WPnY6J3orWLJCJyqkP7ge1PAOFwbJf7UBB482Vpotecxbr3ieQzlpYREeWJ3nAvmvqbMIjBYWO+AWlvokEMoqm/Cb3hXmsWSETkZH3dUhAzGIoFMTIxIl2+/QnpdqQLZmSIiPLEzuDOaADTdawL6xvX40jPEZQVl2FZ7TKUjitFBBHsCu5iJpCISKu9LVImBgA6e4C1m4APuqVd7VcukvZWCYel25nQCJ8PmJEhIsoTbcG2aCCzvnE9Ons6EREj6OzpxPrG9QCkzExbsM3KZRIROdPB3bFMzNpNwPvdQESUvq7dJF0uRqTbkS4YyBAR5YkQQtHvj/QcgSiKAABRFHGk50j0ugEMmL42IiLHCwVj33/QDQz9GwtRlP4evR3/jdULAxkiojzhhTf6fVlxGQRBAAAIgoCy4rLodQUoMH1tRESO5/XFvh9fAgz9GwtBkP4evR3/jdULAxkiojxR6auM7k20rHYZyovL4RJcKC8ux7LaZQCkvYoqfZVWLpOIyJkmVQPC0KH1ykXAKSWAS5C+rlwkXS64pNuRLtjsT0SUJ2b4ZmBfcB8iiKB0XClWL1094jYuuDDdN92C1REROdzUGmnE8mBEauy/+/qRt3G7pduRLpiRISLKE0XuIswfOx8eeKKZGZkLLnjgwfyx87kpJhFRJgpLpH1iPN5YZkYmuKTL5yzmppg6YkaGiCiPTPROxJLCJdgV3IW2YBsGMIACFKDSV4npvukMYoiIsjFhClC7QhqxfHC31NjvLZDKyabWMIjRGQMZIqI8U+Quwrwx87hXDBGREQpLpH1iuFeM4VhaRkREREREjsNAhoiIiIiIHIeBDBEREREROY6lPTJ+v9/KH09ERPoQA4GAYPUi4vEzhogoJyT9jGFGhoiIiIiIHEcQRdHqNRAREREREWnCjAwRERERETkOAxkiIiIiInIcBjJEREREROQ4DGSIiIiIiMhxGMgQEREREZHjWLqPDJHT+f3+/wVwMYCvBQKB3ysuFwD8D4BrAPxHIBC4xaIlEhGRQ/Ezhig1ZmSIsvN9ABEAd/n9frfi8p9B+oBZzw8YIiLKED9jiFJgIEOUhUAg8BqADQDOBHA1APj9/lsBrALwWwA3WLc6IiJyMn7GEKXGDTGJsuT3+ycAeAvAYUhnydYB+BOAywOBwICVayMiImfjZwxRcgxkiHTg9/vrAMjp/b8BuDgQCByPu82FAP4NwOcAnArgW4FA4FEz10lERM7DzxiixFhaRqSPI4rvvxP/ATNkLIA9AG4C8LEpqyIiolzAzxiiBDi1jChLfr9/MaR0/wcAxkP6EFkef7tAILAVwNah+zxq4hKJiMih+BlDlBwzMkRZ8Pv9XwHwGIBWANMAtAG4zu/3V1q6MCIicjx+xhClxkCGKEN+v//zAJ4CcAjAJYFA4AiAH0LKdN5j5dqIiMjZ+BlDlB4DGaIM+P3+cwA8A+AYpKbL9wEgEAg8BeAVAAv9fv8XLFwiERE5FD9jiNRhIEOkkd/v/yyk0ZcigEsDgcCBuJusHvr6n6YujIiIHI+fMUTqsdmfSKNAIPA2pIbLZNc/C0Awb0VERJQr+BlDpB73kSEyid/vHwvgs0N//RukGuctAI4GAoF3LFsYERE5Hj9jKB+xtIzIPDMB7Br6MxrAj4e+X2PlooiIKCfwM4byDjMyRERERETkOMzIEBERERGR4zCQISIiIiIix2EgQ0REREREjsNAhoiIiIiIHIeBDBEREREROQ4DGSIiIiIichwGMkRERERE5DgMZIiIiIiIyHEYyBARERERkeMwkCEiIiIiIsdhIENERERERI7DQIaIiIiIiByHgQwRERERETkOAxkiIiIiInIcBjJEREREROQ4DGSIiIiIiMhxGMgQEREREZHjMJAhIiIiIiLHYSBDRERERESOw0CGiIiIiIgch4EMERERERE5DgMZIiIiIiJyHAYyRERERETkOAxkiIiIiIjIcRjIEBERERGR4zCQISIiIiLbEQRhriAIYoo/g1avkazlsXoBREREREQpPAFga4LLI2YvhOzFkkDG7/eLABAIBAQrfj4REREROcZOURQ3ar2TIAgniqL4odbr9PoZZDyrS8tE/uEf/uEf/nH8HyIiywiCMHGo1OxOQRC+IQjCq4IgfAxg3dD1oiAIjwqC8EVBEF4QBKEfQKPi/lcIgtAiCEL/0J8WQRAWJvg5HYIgbBMEYbogCH8SBOEYgNeHrhs19PP3C4JwXBCEXkEQ3hAE4T9NehnyEkvLiIiIiMjOxgiCUJrg8gFRFPsUf78CwHcBPAjgIQDK62YC+BqA9QAeky8UBMEP4AEAbQDugnRy5loAmwVB+BdRFB+O+5mnAWgGsAnA7wCMHbr8AQDfBvBrAPcBcAM4A8BF2p4qacFAhoiIiIjs7MdDf+I1AVig+PtZAKaJorgvwW3PAnCxKIrPyhcIglAM4KcADgA4Tw6KBEF4EMAuAP8lCMJvRVHsVTxOBYBloij+Mu7xrwTwB1EUr9H0zCgrDGSIiIiIyM4ehpQBiXck7u9NSYIYAHhNGcQMuRjACQB+rszsiKLYJwjCOkiZlS8BeEpxn6MA/ifB4x8DcJYgCFWiKO5J/lRITwxkiIiIiMjO3koQhCTypsbrKoa+tia4Tg5GJsVdfkAUxXCC268EsAHAG4IgHATwF0h9OI2iKHK6mkGsbvYnIiIiItLDcY3XZTI9N+HPEEWxAcBEAFdD6qH5IoDNALYJglCQwc8hFZiRIUv0hnuxM7gTbcE2hBCCF15U+ioxwzcDRe4iq5dHREREue/A0NezADwXd93Uoa8H1T6YKIpHAWwEsFEQBAHAPQB+AGAhEpfGUZaYkSHTdYQ6UN9Xj9ZgK0IIAQBCCKE12Ir6vnp0hDqsXSARERHlgz8D+AjACkEQTpQvHPp+BYD+odukJAiCWxCEIuVloiiKkAYGAMBJei2YhmNGhkzVG+5FU38TBjE44rrI0H9N/U1YUriEmRkiIiICgBmCICxNct3mTB9UFMVeQRB+AGl08t8FQXh06KprAXwWwL+IonhMxUOdCOB9QRC2QApeOiH13ywH0APFnjWkLwYyZKqdwZ2IQOp56zrWhfWN63Gk5wjKisuwrHYZSseVIoIIdgV3Yd6YeRavloiIiGxg8dCfRM4AEpwdVUkUxYAgCO8D+D6AHw1d/BqAK0VR3KzyYY4DWAupL+ZLkPaWeR/AFgB1oii+l+n6KDUGMmSqtmBbNJBZ37genT2dEEURnT2dWN+4HquXrkYEEbQF2xjIEBER5YI7rxRw59Oi1ruJorgN6hvyk95OFMWUjyGK4tMAnlaxnolJLh8AsDrd/Ul/7JEhU8k9MQBwpOcIpBJSQBRFHOmJjYMfwIDpayMiIiKd3XmlG8A+3Hnlw7jzyvOtXg7lFmZkyFReeKPBTFlxWTQjIwgCyorLorcrACcVEhER5YCLAUwZ+rMMd165H8BDAB7BnU/3pbwnURrMyJCpKn2VcA297ZbVLkN5cTlcggvlxeVYVrsMAOCCC5W+SiuXSURERPq4Ou7vUwDcB+AQ7rxyHe688gwL1kQ5QpBLe8zk9/tFAAgEAqb/bLJWb7gX9X31CaeWyQQIuGLsFTjNe5qJKyOiLGSyqRwRxWloaXcB+BSkiVdlkKZhFQ59PRHAGABexR8PpJPS/QB6FX+OJfp+YU3Fh6Y8EdmdV54I4ANI604mAqABwH/izqdfNGVdlDNYWkamKnIXYf7Y+Wjqb0IYYYhIHEg39jdi/tj5mOidaO4CiYiIDNTQ0j4OwKShPxVx338GgM/An/0hgLahP/sUX99eWFOR8eSvFL6G1EEMIAViVwK4Ende+QKAn+DOp//XgLVQDmJGhizxTugdbO7fnDSQAQAPPNxPhsgZmJEhSqChpX00gHMB1Az9ORdAqaWLSiwEaZd7OcjZA2D7wpqKQ1k96p1XPgfgogzu+TyA23Dn0y9k9fMp5zEjQ5Z4O/Q2BAgQIXI/GSIiygkNLe0nQwpYPj/0dTqkEjC78wKoHPoT1dDS/haAZvnPwpqKLtWPeOeVEwDMzXA9FwJ4HndeOQV3Pv1Who9BeYCBDFmC+8kQEZHTNbS0nwBgAYCvQApcTrd2Rbo7Y+jPvwAQG1ra30AssNm+sKYi1dSxJchuqNQmBjGUDgMZsgT3kyEiIidSBC+LIAUwo61dkWkEANOG/qwEEG5oaX8FwO8B/GZhTcU7cbePn1amxSCA27O4P+UJBjJkCe4nQ0RETtHQ0j4Gw4OXdA3s+cAN4LyhP/c0tLS/COAJAJsW/nnVpwCclcVjP8JsDKnBQIYsUemrRGuwFRFEsKx22YgeGYD7yRARkXUaWtp9AGoB/BOA+WDwkooA4IKhPzMgjXzO1McAfqzHoij3MZAhS8zwzcC+4D5EEEHpuFKsXrp6xG1ccGG6b7oFqyMionzV0NL+aQDLASyDPSeM2Zp7MFgPoD6Lh1iHO59+T6/1UG5jIEOWUO4nExn6T+Ya+m/+2PkcvUxERKZoaGm/EMB3AVwBqWyKtHtn/l9uLQBwcob37wVwj37LoVzHQIYsM9E7EUsKl2BXcBfagm0YwAAKUIBKXyWm+6YziCEiIkM1tLS7AHwVwM0AZlq8nFxQL0DMpsn/P3Dn0z26rYZyHgMZslSRuwjzxszjiGUH6A33YmdwJ9qCbQghBC+8qPRVYoZvBoNOInKUof6XawD8G6TxwqSDkqNv/w7AXzO8+3sA7tdxOZQHGMgQUVodoY4RZYAhhNAabMW+4D7MHzsfE70TrV0kEVEaQxmYbwH4CYBTLF5Ornn1868GpiHzcdQ/wZ1Pf6zngij3ZbNRERHlgd5wL5r6mzCIwWG9TAAQQQSDGERTfxN6w73WLJCISIWGlvZ5AF4F8EswiDHCBmS+d8zbW+fe1drQ0r60oaVd0HNRlNsYyBBRSjuDO6MBTNexLtRtrMOqdatQt7EOXce6AEgBza7gLiuXSUSUUENL+xkNLe2bIe1GX23tanLW4Dl7N20HMDfD+/8w5B3zU0jB0MtDQSdRWiwto7zFng912oJt0UBmfeP66OalnT2dWN+4HquXrkYEEbQF29jrRES20dDSXgTgDgA3AvBau5qc96eJ7754GaT9ZLTa1fCln30CYPbQ3z8HoLmhpb0JwPcX1lTs02uRlHsYyFBesqLnw6mBUwih6PdHeo5AFEUAgCiKONJzJHrdAAZMXxsRUbyGlnYPgBsA3AmgxNrV5I0NkIJGzcIuz+0QXP+Z4Kr5AC5uaGm/C0DdwpqKwWwWSLmJpWWUd6zo+egIdaC+rx6twdZoYCAHTvV99egIdej2s/TmVZzILCsugyBIJ9wEQUBZcVn0ugIUmL42IiKlhpb28wG8DmAdGMSYpe+S53/8DoCpGdx32zNf/OnJKe5bAGANpHKz6gzXRzmMgQzlHbN7PpzeLF/pq4Rr6J+KZbXLUF5cDpfgQnlxOZbVLgMgbWJa6au0cplElMcaWto9DS3tP4E0+vdMq9eTZ54aHTz2jUzu+EnB2B9CypylUw3gpYaW9h83tLSzTJCiWFpGecfsno/4wGl943oc6TmCsuIyLKtdhtJxpdHAyY49JjN8M7AvuA8RRFA6rhSrl64ecRsXXJjum27B6ogo3zW0tE8GsBHALKvXko88oY8fB/B4Bnfd/Kc5a2YBOE3l7b2QyteuaGhp/9bCmoqdGfxMyjHMyFDeMbvnI1HgFBEj0cAJQDRwsqMidxHmj50PDzzRzIzMBRc88GD+2Pm27vMhotzU0NK+HMAuMIixyj++su12H4ByjfcL9xR++t8B3JrBz5wG4O8NLe13NbS0s6Y5zzEjQ3nHC280mCkrLotmZIzq+TAycDJrgMBE70QsKVyCXcFdaAu2YQADKEABKn2VmO6bziCGiEzV0NJ+MoBHAHzF6rXkuXoBYiZ7x2x4/rx/rQVQmuHP9QC4DVJ25pvMzuQvBjKUdyp9lWgNtiKCCJbVLhtR6gVk3/OhDDCU9AyczJ68VuQuwrwx82xZ/kZE+aOhpX0hgPUAytLdloxV1r3/95D6krQI/uPUc+/P4H6JnAWgpaGlfdnCmoqNOjweOQwDGco7Rvd8JAowZHoFTsoBAvHkn9vU34QlhUuYLSGinDDU5L0OwL9YvRYCALxywc7/PgfAaI33e3D3Wf/8LQBjdVrHKAAbhqaa3bywpiKs0+OSAzCQobwj93wkCjZcQ/9l2vORKsAAoFvg5PQBAkREWjS0tBcD+B0A/oNmHxsAaC0r+3D3mV//NYAdBqznewDObmhp/+eFNRU9Bjw+2RCb/SkvyT0fVb6qaElXAQpQ5avCksIlGZdkqRntrJRps7zTBwgQEanV0NJ+OoAXwSDGTgarW3/zPIA5Gu/3X/+YcMG/AoZtPHYJpD1nzjLo8clmmJGhvGVEz4ea0c6ybJrlzZ68RkRkhYaW9hoAm5F5UzgZ44+fee+lLwMQNNznyPOzvrsVxmRjlE4HsGNoCMDTBv8sshgDGSIdqQ0wAGB58fKMf47Zk9eMYNbENSJypoaW9qsgTSbzWb0WGmED1G1kqfTvPUUTfwhzqoHGAvhdQ0v7GgA/XlhTIZrwM8kCLC0j0pEXsQ2Hy4rLIAjSySq9A4xKX2V0T5dltctQXlwOl+BCeXG5bpPXjNQR6kB9Xz1ag63RgEyeuFbfV4+OUIe1CyQiSzW0tP8IQD0YxNjRsUue//EhAGdquE/HHy+8cxeAWoPWlIgA4EcAfjs0KIJyEDMyRDoyY7QzYPzkNSNx4hoRJdPQ0u4D8CsAS6xeCyX11OjgsW9ovM+Pgr7CuwxZTXpfB1DQ0NK+aGFNBeutcwwDGaIMJCuL+qz3s6YEGEZOXjMaJ64RUSINLe0nAGiC9gZyMpEn9PHjAH6j4S57Gr/4024AXzBoSWpcDuDphpb2ry6sqQhauA7SGUvLiDRKVRbV2N+ImaNmwgNPtPRLlumEsmSMmrxmNE5cI6J4Q0HMH8Agxu7+8ZVtt4+Ghs1IRQi3RVyefzdwTWp9BcCWhpb2UVYvhPTDjAzZlh2bwdWURb3yySuoHVuLA6EDaAu2YQADWU0oS8WIyWtG48Q1IlIaCmK2wtoz9qTORgGilr1j/rbl4v8aC+Acoxak0SUAnmloab98YU3FcasXQ9ljIEO21BHqGFE2JWc99gX3Yf7Y+ZZkHNSWRR0IHXBcgGGWXJi4RkT6UJSTXWj1Wii9su623wNoUXv7kGfUbZB6nuzkiwCaGlraFyysqfjI6sVQdlhaRrajzHooez8AKesxiEE09TehN9xr+tpYFpU9p09cIyJ9DDX2N4DlZE7x8gU7H54OQG1p1h+2zrv7LACTDFxTpuYC+ENDS/tYqxdC2WFGhmzHzs3gLIvKnpMnrhGRPhpa2t2QGsa/aPVaSLUNANSWlYkfnlD+Y0ibmdrVFwD8b0NL+6ULayo+tHoxlBlmZMh27Jz1MGufmFwmT1wzYyACEdlPQ0u7AGmjyyssXgrFOXL4Pdy+4ircuORirFh6KRp/+z/yVaHpex5/AcCF9a//H6Y92IxpDzbjgl89j9c+OCbd96MgPv/I86gKPIfAy+0vNF9wy5cAjL/7lutxtOuwRc8orfMBbGpoaeeJfYfiL45sJ9OshxnDAczaJybXyRPXdgV3GT4QgYhsZy2Ab1q9CBrJ7fbgWzfeitOnVOHj4/343rcvR/Wsz+PTFWf88bT3X/kyAKGieAy2X/t5FI8uwB/eOozrn9mNv183B0/sOYRrzjkN36j6VOhTa/9cuPFWfP+lF57DpMln4aTSk61+aqlcCiAA4HqrF0LaMZAh28mkGdys4QBWlUXZcYJbphI9l7N9ZzvyuRCRNg0t7d8F8F2r10GJnVRajpNKywEAo8eMxYSJn0V31wf4dMUZGwCsAYALPl0Svf3sCcU41PcxAMDrcuHjwTC6jw9sOLH01EvDg4PjGjc9gtv/45fmPxHtljW0tLcvrKmos3ohpA1Ly6zW1w3s2AI8vgZ47Dbp644t0uV5SmszuJnDAawoi0q1b019Xz06Qh26/Syj5dJzISJtGlraLwTwX1avg9Q5/P4hHHyzFWecWX3skud//C6AEaUGv9r1D3z5s1K25aqzJ+APbx0On/Xf22f/y/fWlP3h6Y2Yd9lX4Rs12uylZ+rfG1ra/9nqRZA2gly2Yya/3y8CQCAQMP1n28qh/cD2J4BwGBAVB+CCC3C7gTmLgQlTrFufRXrDvajvq0+4V4vMAw+WFC5BkbsIzcebo+VeyYYDuOBCla9Kt+EAveFeU8qitL4WdpZLz4VGEKxeANlbQ0v7BACvAii3ei2U3sfHP8JtNy7Gomv8OH/OZb9c+OdVHwNYobzNX9qPwL/1NbzwrQtRMiZaIXF3w8X3lvf3HbvuP3+0Arf8+4N4ZN1d6P/wGBb+83WorJph+nPR6BMAn19YU/Gq1QshdVhaZpW+bimIGQyNvE6MAIMR6fraFUBhycjbmLG+vS3Awd1AKAh4fcCkamBqjeHrkbMe8aVigJT1cME1LOuRaDiAKIrR4QCrl66ODgfQc8qZOPRf/Pd6svMEN62c9lxyqZyPyEpDY5Z/BwYxjjA4GMJ/3O7HnEsux/lzLoMn9PHjAJ5U3ub1w8dwXeMu/GHJBcog5uiL1df9HsDfn3z051j0zf+Hvz7biNOnVOHCiy/H3bf8C+5a97jZT0erUQA2N7S0z1xYU2HbCQUUw0DGKntbpEwMAHT2AGs3AR90A+NLgJWLgPJi6fq9LcDsy81dW6JMUSgIvPkycGCnKZkiLc3gZo9ENnOzTquCNK3UHPQb+Vz0DjrsuiErkUM9AOBcqxdB6YmiiF/U3YIJnzkdC//5OgDo+Mq2208AEG1QfefYcXz1yZew4crPYXLJsG1Y7uksm7r6vf9rdx/t6kTV9PPQ/tZeFPhGQRAEhAaCJj+bjE0A8LuGlvaLFtZUcC8Fm2MgY5WDu2NBwtpNwPvdgChKX9duAu6+Xrr+4G5zAxkbZYqK3EWYN2Ze2oNaM3eKV/bjxJMPepv6m3Qrj7Jy3xq1wYHag36jnoveQYfZv2OiXNbQ0n4DgO9YvQ5SZ9/rr2Dbn57GZ06fgpXXzkdf79HR/3nBqT8YCEv/tt4wswJrtu9H98cD8De9BgDwuFx45fq57/655tYWAD/d+PB/Yen13wMAfOHiWtStvgHPbHoUi69badGzykgNpAB8mdULodQYyFglpDgz8cFQEANIXz9QNPqHTD4ZYOdMURJmjkQ2uzzKzCBNSW1woOWg34jnYkTQ4bQSOCK7amhpvwDA/Vavg9Sbes4sbH7hYPTvJx/Z+5XZu3/ZorzNLy+fjl9ePmIy54+PjyldAwA/+MkvohcWFZfiPx56yrgFG+u6hpb25xbWVPzG6oVQcpxaZhWvL/b9+BJgaGNFCIL09+jtTN5YMVGmKKLIFAGxTJHBesO9aD7ejEBPAPf33I9ATwDNx5tHTB+b4ZsRnSAmj0S+d8W9WL10NUrHlQLQbySy2Zt1ap3gpgctU+DiD/rrNtZh1bpVqNtYh65jXdH77AruMuS5aPn5atl5Q1Yip2hoaT8FwFMAdwd2sJdm7/7ldEh9I6nsf+aiun8A+KIJazLbAw0t7adavQhKjhkZq0yqlnpOxIiU6YjPfADS9LJJ1eauyyaZIi3lQlqHA2RD7/KodOVbVuxboyUjoaXvZXHhYt2fixF9N1aW8xHlgoaWdgFAPYBTrF4LZWUD1G1c+sOw2/fvRi/GIicB+BWAL1u9EEos9wIZC6dtaTK1RmqcH4xI5Vp3J9hQ1u2Wbmcmry8WzIwvifXumJgpyqRcyKyd4vUsj1IbrJkVpMm0BAdaDvqzCTiTBXxGBB1WlfMR5ZDrAbDu0tlCM96ofwHAz9Pc7pWGL/1MBDDThDVZ5bKGlvYbFtZUPGT1Qmik3CotO7QfaFwnZTrkg3F52lbjOul6uygskaZ/ebxS5kVJcEmXz1lsfvA1qTq2npWLgFNKAJcgfTUpU5RpuZA8HGB58XLcVHwTlhcvx7wx83Q9yNerPEpL+ZYcpFX5qqIHzwUoQJWvCksKl+g+PUtLcOCFN/r3suIyCEMlkskO+jN5Lqk20VRS8/PVsKKcjyhXDO0X81Or10FZ++OnP3h1PtLsERV2eW6D4LrLpDVZ6WcNLe2nW70IGil3MjI2mral2oQp0nqiGaQBKdNhZQbJBpkiO48c1qvUS2tDudoJbnrQkpGY4puiedCClueSLjunpNegByvK+YhyyIMACq1eBGXt1wDSBSjPPfPFn34GQD7s3H0CgF83tLR/YWFNRSTtrck0uRPIOHDaFgApWJl9uX3WJGeK4veRAaRMjNudOFOkY0mfnXsU9OrHsXOwpmUK3HTfdEMP+tUEfDK9fr5ZPVfccJNyTUNL+1UAFli9Dspa76Xbf/Qe0gQoH/vG3QHgt+YsyRYuAPADAPdYvRCKyZ3SMhtN23I8OVM0edbQdDVB+jp5lnR5/GaYOpf0aS1XMpsepV52Dta0TIGTD/o98ETvI3PBBQ88WR30q5kgpvx5ev18o8v5UpXL1ffVoyPUkdXjE5mtoaW9FBy1nCs2jRr4cHGa2/zufy/8UQ2AT5mxIBv5cUNL+zSrF0ExuZORscm0rZyhNlOkQ0lf/Jlp5QGp0fvCZCrbUi87N5RrzUgYOWhBbcAHAFW+Kl1/vlHlfNxwk3LUzwGUpr0V2V7BwEePA9iU4ibhrqJJdQD+16Ql2UkBgA0NLe2fW1hTMfIfcTJd7gQyNpi2lZeyLOlLNLlLeeCcqz0KZm7imQmtwYlRB/1aAj6zeoiyxQ03Kdc0tLTXAkh3Bp+cof2y7XeciNRB6aMts278KqTRxPloGoB/AfCA1QuhXApk7LovSyJOGRGtRqKSPlFR0nf39bGSvrhAJtWZ6WSMGjlsNjMayrPtwTBzwEAydg/4MmHn/igirRpa2gshNfhTbtgoQLw6xfWfHPj0Fx4A8FezFmRTP2poad+wsKaiz+qF5LvcCWRsMG1LlUP7RzbSy/0kB3ZKjfTxPShmyDS4yqKkT20jtwsuRBAxZF8Yq2TaUK42ONGyoaid5eIEMTv3RxFl4IfIvz6JnHXykdanAfwtxU0e2FN55XWQpnjlszIAtwC41eqF5Lvcafa3674sSsp+EuU0MGConyQkXd/Xnfj+RsmmWd/ri30/vkQq5QNUlfSpbeT2wGPYvjBW0tpQrrZBXMseNXZn9DABK9h9mAWRWg0t7Z8CcKPV6yDd/H327l99DsCoJNcfe/Wsq+oBLDNxTXa2sqGl/dNWLyLf5U5GBrDnvixKdhwRrbZZf97VwDutIzM2EyqBjjcyKumz6sy0ncbeqi3f0tIgrrUHw06vRyJGDhOwwiTvJOwPSScHkpXLAUCFt8KqJRKpdTuSH/SS82wA8M0U1//noVNn/hugOBuT30YD+Hekfs3IYLkVyAD225dFKYt+EsOoCq4GgWf/B4AwshzO5ZKyLyI0l/RZMbnLqSVXWoITLT0YFd4KQ14PvYMjO/Tr6EUURHU3TLmfNpG1GlraKwB8x+p1kG5Cn3v91y0A1iW5/oNt5616FsBPTFyTEyxtaGm/b2FNxS6rF5Kvcqe0zAnsOCJa1f474tCfBOVw4aEMgdujuaSv0lcZLRdaVrsM5cXlcAkulBeXG9LI7eSSKzVleBFEsCe4R1Omy4jXI5s9UnrDvWg+3oxATwD399yPQE8Azcebbfk7yVT7QHv0+1QllcrbEdnQj8Ez87nkDxMO716A5KdQ7jpWOOFHKa7PVwKA/7J6Efks9zIydmanEdFyc7/a4ApInrERReAzVUDBKE0lfUY1cifLBgxEBtJmNQYxiMb+RtSOrTWkZCnTTIXa4CQ+IEmV6QIQLVXTawxwNnukODVbphWb/cnpGlrapwJYYvU6SFe/BnB3kusO/mHOT/YC+IWJ63GSeQ0t7QsW1lQ8Y/VC8hEzMmaaVB3LWqxcBJxSArgE6auZI6KVzf1KqZr1gRQZmwhwqE0qh7vqDuCau6Svsy9P2ZdkRCN3qmzA/tB+VcMFjkaOGrK7ejaZCkFxEqy4sHjYdfF/V0qW6YqXKsvTFmxT+xRHlMDVbazDqnWrULexDl3HuqKPuSs4PAvv5GyZVmz2pxywBjx+yCW9l27/0QcAJie5/o6BghPuMnNBDvTThpZ2t9WLyEfMyJjJDiOiUzX3J2vWl6UshwsCO7ZoHt+sZyN3umyAUrpd4uUDZ712V892N3cBAkSI0e/jr5MlyqwkynTF0yszkOkeKfm0SWQu7o1D+aOhpf1zAL5q9TpIV78dNfDhPye57rUtX/rZhwAuMHNBDnQmpP8vNlm9kHzDQMZM8ojo+H1kACkT43YbPyI6XXN/ouBKlqocDohtSApo2htHr0ZutfvSAHElVxAgCAJWrVtl2IFztgfqykDsaN/RYdcp/54seJAl+9l6DVvItGwqnzaJzMW9cSiv3AX2SeSUgoH+xwE8lei6iOC6XRRcdSYvyalWgoGM6ZgaNps8InryrKE9WATp6+RZ0uVGb4apprlfJriGSuGGPrOSlcPJLN4bR+2+NMDwkiuXy4VwJJx1WVW2a0v189SWI6XLNCX72alK0Ma6xqou6cq0bCqf+kZycW8cyg8NLe3nA7jM6nWQrtov2/6jcQBKE1z318Yv/ewkAFUmr8mpLmhoaZ9p9SLyDTMyVrByRLSW5v7Js4DTzgL+skEKSJKVw8ks3htH7cEwgGFnwletWxW93KgD52wP1NWUIwHpm/uT/exkmQEA6In0oL6vXlWzfaZlU1aM4rZSru2NQ3ljpdULIN1tFCBeneiKAe+YHwJ41NzlON5NABK+nmQMBjL5RvXkNF8s8EhVDqf8u8V746g9GI5nxoFztgfqasqRgNQbLKb72cmIEFX3DGVaNpWPfSO5tDcO5b6GlvZTwd6YnHPK4defBvCDBFc1/mHuXdUAJpq6IOf7p4aW9h8srKl43+qF5AuWluWbTCanpSqHU7J4bxy1+9JMKZiCk1wnRe9nxh422e6Zk6ocSdnsLwcP9664F6uXrh7WF5TqZytpmTamZZ2pyqZm+GZEb5/sObBvhMhSN4AnP3PNjnNff3QmAF/c5ZFjY0/9CYDbLFiT0xUAWG71IvKJIJeZmMnv94sAEAgETP/Zea+vWxq9nGhqmczjlQIXNUMHHl8Ty/Dc+vDwDM8pJbFSNK9PGslsoN5wL+r76hNOBpN54MGSQmn7A7W31WtqmR4/rzfci13BXdgX3DesXC0ZF1zRQCeMsKq11m2sG5a1KS8uj2ZWClCA5cXp/42W16mlbCrRPjLyc3DBpWkfmUz366GMsfk7hzW0tBcAeAfAyVavhXT1/xb+edU/A/hC3OUbGi6+9wCAO81fUk7oBHDawpqKYNpbUtZ4diXf6D05bVJ1bFpZsvHNZuyNg1g2IN3BsHwgq+W2eq4tUTCjtsG7yF2ECm8FWoOtaX+mMnjojfSmfK7KNaXr4Wk+3pw2IMikbEqvvhGzN9Zk0ER54EowiMk1oZmvP/YiRm5yOXBo/Iz7AGy3YE25ohzAYrC/yBQMZPKRXCq2t2Vo35cBwFugat+XEeywN46CloNhKxqu5b1g1F4eT96PJl12xQ03Fhcujj6HIndRyuf6eN/jqvuLjAoI5HVm0zeS7X49WpkdNBFZ5DqrF0C62/qpw68twMhs6n+/evbSbwI40YI15ZKbwEDGFCwto+wd2p8+w2P0WGkVrDxzrldpWfPxZrwRfAOA1Mvy0OaHoj0sJeNKsPyK5dGekmm+aaoDgubjzdFm+3R78GhZr9nUPA8XXKjyVWXdZK/X7zRHsLQsRzW0tE8EcBD8Heeary/886o6AGcoLuvfM/nyCw98Zu6LGNk3Q9rNWVhT8bzVi8h1zMhQ9vTM8OgoPnCJZ+aZc712rlfuM7O+cX00iAGA7mPdwzbA1LJ5ZKppY3LzfybrNZseG2uqDXj1+p0S2dy3wCAm1/Rctu2HH2B4EAMA9x34zNzvgkGMXpYCYCBjME4to/T6uoEdW6TG/sduk77u2DJ8k0t5b5yr7gCuuUv6Ovtyy4KYjlAHNvZtxBvBN1I2xUcQiY4WVrvpYyay3RBTFr8fTbxM98BJNW0sm/WaLdv9ejpCHajvq0drsDX6WHLAW99Xj45QR/S2ev1OieyqoaXdBeBaq9dBuvutL/TR4rjLulo+d0MjuAeKnr7a0NLuTX8zygYDGUrt0H5pytmbL8emk4WC0t8b10nX20xvuBeN/Y0J+0iyGS2cDb12rvci9m9ioj1gstkDR+4ZqvIN38Q5m/WaLf71EQTpRLKa/XqU/TXKgQhA4oBXr98pkY3NAXCa1YsgfRUM9D8O4J/jLq7rOmnybQDcFiwpV5UA+JLVi8h1DGQoub5uqfdlMDS89wWQ/j4Ykq5XZmZs4MVPXhxW8qMMXB7c/KAlZ86zOcBWUu4zE9+7UjKuZNi+MJnsgSM32+u1XrNls19PfKlYuoDXqa8RkQbG7mJMVjj45e13FEM6yJa986cv/PBlAAstWlMuiw8YSWfskcl3fd2K3pagtN+L3Nuyt0Vq4AeAzp6Ro5XLi6Xr97ZIZWQ28dbAW9HvH9z8ILqPSYHW4aOHh93OzDPneu1cL/eyDGIQpeNKcfs1tye8nRvurDaP1Gu9ZkvV6yNLtrGm1v4ap75GRBossHoBpLuNGFk+ducno4p/YsVi8sAVDS3tPu4pYxwGMvks0bQxuWzswE5AROzytZtim12+3y39/e7rpesP7rZVIKMcZSwHMUqCICQcLazXmfNEzeIVBRVwwZXRAbaS3MvyTP8zSUcwu+DCgrELVE/KSrTeSd5J0Y00s1mv2bTuJaSktVQsm6CJyO4aWtqnAPis1esgfZ1y+PXNAG5WXLT3mYvuOQypjJD0VwjgIgB/sHohuYqlZflKTdlYWNEk/8FQEANIXz9QBAghZ/UAaCk30ipZs/jbA28jggjccI9opnfBpXpDTEDqZVlauBRTvFOGPZYLLkwpmIKrC69WPYEt2XrfCr0Vfcxs12s2Za+PHJzKz2EQg9javxXNx5tHDHfQWiqWakCC3V8jIhXmW70A0t2L577+6CwoppKJEG4PuwvusnBN+cA+Z3pzEDMy+UpN2ZjS+JJYRkYQpL/LvPbqAZAzH4CUTVCOKDbyzHm6zRgBqeTrDO8ZaA+1Z7UBZ5G7CJeNvQyX4TLNa0w1kjp+vR54cHrB6WgfyG69ZpN7fSq8Fao3rMykVMyKTVWJTMKystyzAcPLyv6+5eL/8gFg2thYlze0tPsX1lSYv3FjHmAgk68O7k5fNqa0ctHIYAeQNr2cVK378pKVZwmigIOhgyn39zjDewb2h6RpajdccUPCA1JZunIjLdTsKyJChM/lw/Li5Vn9rEwk2oVelmofFJ9gzXqzlS6wjCCCpv6m6IaVmZaKyUET94qhXNHQ0j4OwOetXgfpauDc3Y/sAPCAfMGgu+A2AA9Zt6S8cSqAmQBetnohuYiBTL4KKfrOUpWNycqLRwY3AOB2S4MBdJTogDuEEN4ceHPY7ZKdWZ89ejbeDr2NMMJJD0gBqZToTN+Zw86cq90MMdFtlTLdjFENLWtU3ifZQb3R67WK1g0rs+mvIcoxlwKKWkvKBVtPObKnFrHNTf+36aJ7zgD7oMxyORjIGIKBjCzV9K50mzpmc1+reH2xYCZV2ZjbI12mHAgASJkYtxuYszjj55isyVwOQtRIdGa9yF2EBWMXJD1wl3sX4vtIkgVQbwTfwBvBNwAgGjSUu8qx/ePtCbMbgHH7inSEOkY0+str3BvciwVjFyTsj0l3UJ+L+6BonUIGsFSMaAjLynLPBgD/MfS9+NHokh8DeMrC9eSbC61eQK5iIAOkn941ZzEwYYr+97XSpGppjWIkddnYZz8XG8V8cLfU2O8tyDpQSxY0yCVhQOIDbgBpz6wD2g9I02UsZHLQkE5ZcVn0wFmv6WjyRp+JAicACCOMxv5GXF149Yjnl+6g3oj1Jlq/1kxSNjLdsJKlYpTPGlraXQC+bPU6SFdHL9t2eydi2ZdNz37+tjkATrFwTflmZkNLu2dhTUXqgwzSjIGMcnpXPDECDEak62tXjDxoz+a+VptaIwVag5H0ZWOFJdJ4ZZ1GLKsNGhIdcANQXQKl5YBUTRlSIslua8S+IvEbfSYL6HZ8sgOXnTB8CEC6g/pbv3mrofugJAtcE5UG6sULb/R5mxGoEeWImQAS/4NHTvVbX+j44qHvBw+XVP4UwLNWLigPjQFwDoBXrV5IrmEgk82mjw7dMBKAFJzMWTwymwToUjaWitqgIdlZdCNKoFJlLB7a/BDcbnfCNSYrWUrWmyNCxL7gPrwefF1zRiLVRp8Pbn4QP7zmh9HbxQcy6Q7qrZzmFl8aKN8n2+wNN6wkysgsqxdA+ioY6H8cwNNDf/3VjhnX/xOAIutWlLdmg4GM7riPTKLpXRHF9C4gtumjnve1gwlTpGzR5FlSzwwE6evkWdLlBpXEJQoaImJkWNYFSLyXh5r9PTKRKmPRdawLh48eltZ4dPgaU5UsJSJCHLZfyxvBN/Drvl9jzyd70q4x1Uafyr8nKj2r9FVG9zpZVrss4V468QQImDN6ju7T3Oo21mHVulWo21gXHY0tlwYCyfe2aQ22or6vHh2hDlU/d4ZvRvQ5y4HavSvujQaaADesJEpghtULIF0d+PL2O0oAlAD4eH/Flx4CsMLiNeWr861eQC5iRkbt9K5Emz5mc1+70LlsTA21vQvJzqIbcWY9WcYinojha0xVsiRzw51yeIEIEc99/BzaBtrwpRO+ZEi/iJrRwols/3g7xrrHZlX2paXpfrpvuubsTTJmTCEzu++HyAQMZHLLRsT2jvl522e/shzAaAvXk88YyBiAgYza6V2JNn3M5r55TG3vQrIDbiNKoJKVIUXEkdkN5RpTlSydUXAGBiIDaB9sj94+VSndu+F3Ud9Xn7RfJN1Gn8rbxUt1UJ+MCBGDGFQVOKQ6oNfSdK91ZHI6ek8hS7ehqNF9P0RGamhpLwBwltXrIP186oNdDQBWA+h9edo1mwDssHhJ+WxSQ0t72cKaitSlG6SJkOiss9H8fr8IAIFAwPSfPcKOLbHpXcn6XASXVG4Vn7XI5r55rPl4czRo0NJYn4jyzHo2B43vhN7B5v7Nw8q3AKBuYx0OHz0c/bvb5cbqq1enXaMcTCgDhq5jXajbUIdwJJadSfR4HngSBg5/7P9jdKpbqtdtincKZo+enTCw+Kz3szgQOhA9qFcGR8ke0wUXqnxVSQOHZBttyr8bEWI0I1W3sW5Y4FpeXB4NTAtQMKz0Lt1tzd6kM9WGookk+z3mKCH9TcjuGlraZ4A1/Lnkbwv/vOrXkDa9XN1w8b3TAfyTxWvKdwsX1lRssXoRuYQZGS3Tu/S8bx5TU+bkgQenF5yO9oH26Fn0ioIKQATaQ+267u8hH6Amkijjogw6BAjDgh8XXNHL4g921zeuHxbEAEA4EsZdj92Fk086OW3GQc1Gn264McEzAfV99SknhMmPHegJaN5rRUlNI78w9J8IMWkGCwAGMTjsNbPT3japnqcemSMim2BZWW7ZAKms7L3nLrh5O4C7LV4PSeVlDGR0xEAmm+ldFk7+cjK1vQsTvROBE/T5mclKnz7r/WzKUdCp+knO952Pj/DRiJKlYCQ4LHPy0OaHhpWBJaImcFCz0efcMXOx7fg21T0mme61ItM6tjrV6xkf+NlpZHKq5xkOh9Hd160pACSyKQYyuWPg3N2P/B3AgwCW959w8o/BzKkdzLZ6AbmGgQwQm96VyaaP2dw3j5m5g3qqPUz2BGPTwuIPUL8+9+t4attTCQ/MJ7gn4Nwx5wLAiAPVQE+sZHJ94/q0QQygPnBI97pp7TFR268ESKVts0fPHva7UdPIL0KEZ+ifGrVlWUDq/iOzRyanep7K8lyrM0dEWWIgkzuaTjmy53IAbzfN/fe3AVxs9YIIADDN6gXkGgYysmymd1kw+SsXmLGDerrSJyXlAerho4fxwO8fiF6nPDAHgM5wZ9KfGZ/lSMTtcg8rM9OScUj1ummZEDZvzDxU+irxRvANAFLgIGePRFHEYHgQXce6osHb/tB+HAgdGNaPpDajM4hBXFN4DXYFd2FPcI+qLE42e9tonSaW7vapnicg/f7MzhylWzMnqpEWDS3tbvAgK5dsAPBTAD8c9I7+d6sXQ1EnNbS0j11YU9Fv9UJyBfeRoZymZg8TmfIANZ6WM+1eeKPfJxrHfPJJJ+O/bvwv3H7N7Tj5pJNH7OeSTcZBa6nYDF/sBGzpuFK43e7oPj1H+44O2zMHQHSKWW+4N+FzTbXHjxyAueGOXp5qH6F4LrjggSftyGSt+9CouX2q51kyriThvjxGZo7SrfnvH/9dl714KK9UgmN5c8XRL2+7vQtAX8OXfjYA4FyrF0TDTLR6AbmEgQzlNLWbbwLDD1DjaTnTrjx4je8TKRlXEj3QNWKTRi2BBYARAUF88HP46OGUm1eq2Wgz/oBebbClXGcBClDlq8KSwiUpp9MpM3DxGbcIIiMCMbW3n+SdlPR5Lr9iuambbapZ845Pdqh+DYiGnG71Akg3TxaEji8Ouzy3Q3DdZfViaITPWL2AXMLSMnK0bEqC4g+a5Z4M5bhlWaoD8/g1eOCJTukqHVeK26+5XdVz0WOTxmT74aTqMUm3Gagy8IsvTVMzgS7+gF5tX04mI5a19gipvT2E2D4+6TYU1WuzzXjy+2xPcE90Ul66AQucqEYanGL1AkgfvmDfEwBueuaLPz0FwJlWr4dGmGj1AnIJMzLkWNmWBMWXfckHqMqSr5NPOhm3X3N70jPtidaQbAKakgsuTPRM1JxxSGeGb0Y0c6A245Msq6KUrDRNnkDngWfERpzJSsEyyeKopSYDF0EEe4J70BvuVX379oH2pM9TyQuvLr/HeMr3mXLcd7osY6rn1BZs0219lBPGW70A0sXblz1/Z8knBWPXArjT4rVQYszI6IgZGXIkNfuXNPU34XTv6Xgr9FbKDEX8XjCpzrjL+7Q83vd4wp3dZfF7y8gHyy64cEbBGZg9arYhDddqR1srf3ayrEr8hpTJSuu0TqBTk8URIWJvcC9eD74OL7yoKKiAIAo4GDqYsnFdbQYuggjq++qHvX/S9RTJz3P78e3oGOwYsWZ5888Kb4XumZhk7/V0WUY77cVDtseMTG7YCODEP81ZMwnAp61eDCU00eoF5BIGMuRIepYEueFG7dha7PpkV8IDVFm5qxxHI0fxzuA7I3oPkq3BBReqfFWmlvBoDSzk4OeZ/mcQRmySWqrALz5bomUCXapgSxlUygfuIYTw5sCbwx4jfoNPOfuhZZx0fGCgdt+aQ4OHEj6vRPv06CHVe/2kwpOie9gkeo522ouHbI+BTA445fDrm48UfxYAnrN6LZQUMzI6YmkZOZIeJUHK0qfTvKdh4YkLcU3hNZjmmzas5GuabxquHHsljkaOJmygTrcGK0p45MBiefFyXFN4Dab4pmBfcB8e63sMgZ4A/vDRH/DH/j8i0BPA/T33Y2v/VowVxkbvn673ItsmdjnYqvJVRV9rZRmgMqOVTKLG9WRlaycVnoRwOJx0Yl387ZOVuamZgqcchqCHVO91EWLCNWt5TkRDWFrmfC3nvv7om3+b6f86AG5iZ18MZHSUUUbG7/cXAagB0APgxUAgICquOwHA9wKBwBpdVkiUgJYxw1oyFMkyC83HmxOeFT+p8CSIENF9rDt6WzuV8CTbDDRRhuOYeCz692R70ABSBkuPbEP8a918vDnaA5IokJLXlapxXU2ZXPzzkakZVqB1nx49pHqv9/T14N4V9ya9bzZ78VDeYUbG+TY0XHzviQD+1eqFUEonN7S0j15YU/Gx1QvJBZoDGb/ffxaAZwGUQcro7PT7/V8LBAL/GLrJWAA/AsBAhgyjZfIVkP3mm8kOYBOd2bdLCU+q3op00m1uqeZna92MMV2QACBt4KAsW1Pb/wIAHnhU9RRp3adHD1rK5eQ1zxw1E6988orqPinKbw0t7QKAk61eB2UlCOC3kI69TrB4LZTeBABvWb2IXJBJRqYOwIsArgZQCOB+AC1+v39eIBDgL4VMkcmY4WwkO4BNxKwSnnTBQro+IiBxhgPIrrciWRYoUU+LkpogQU3gIGfgNvRtiP78dM9nceHihBm7072n4+3Q29jav3XEcAez+k/UvNcBRPux5CzjlIIpqvukKO+dBLBpyuGaABQBuN7idZA6Y6xeQK7IJJCZDWBeIBD4CMBHAP7J7/ffC2Cb3++fB+BYynsT6SCT/UuykW6vFUDKxJQXlw9bi1ElPGqChUwzHEDyRv90gZnaaXKJmuHVZB7UBg5F7iKc5TtLdbCbKGPXEepAY3/jiKyGzIwAGlD3XvfAM+I1zTYLSXmFZWXOtwFSNoYBqTP4rF5ArsgkkPEBwztxA4HAKr/fLwDYDmCxHgsjg/V1A3tbgIO7gVAQ8PqASdXA1BqgMEGPoNbbG0zNmOG5Y+ZqLm9Sis94yJQHsHKPTE9fz4gz5AIEQ0p41AYLakqrkmU4Mg0OtW5IqaQm86AlcMgm2FVTlmdW/0kmI7WJNCpLfxOysW4A/wfgKqsXQqoxkNFJJoHMfgAzAexVXhgIBP7V7/e7ADTosTAy0KH9wPYngHAYEIcOikJB4M2XgQM7gTmLgQlTMr+9SVI18Ze5yrDt+DbN5U2yRBkPWbqd3QEpiLli7BU4zXta1s8zntpgQTnKWGuGI57aA+ZsmuHVBB5aAodsAgA1r3Gmr1EmtI7UJtKIB1XO9iSkzS85idY5+P+cTjJ50z+NJFmXQCBwE6TNmIRsFkUG6uuWgpLBUCwokYkR6fLtT0i3y+T2JityF2G6bzqm+KbACy8GMIC9wb1o/rg54ajkRCN74ynPxicqKUpFHul8+djLDQliAPWjpwUICUcRyz08yUbzChBQ4ioZNoJa7W712TTDy4FHolHZqcwcNTPpgXyiMc9qno+a11hJy2uUKeVI7ZuKb8Ly4uXRAQdEWeKecs7WDmCB1YsgTRjI6ETzP16BQKAOUsN/suv/H4D/l82iyEB7W6TMCgB09gBrNwEfdAPjS4CVi4DyYun6vS3A7Mu1395kiTInynIgreVNgLaz8fIBdwQR086Qa9m9Xp7GJWc45Odz96/vTppdcMONBWMXaH4O74TeGfb3TJrhE2Ue0nnlk1cwpWBK0vVm0iui9jUGgJuKb1L9uEQ2xUDGud4GcIXViyDN2Mukk7SnPf1+/91mLIRMcnB3LLOydhPwfjcQEaWvazdJl4sR6XaZ3N5EajInmWxUqfZsvBderChegRXFK0w9Q67cOLKsuAyCICVAEwUL88fOH3bfdNkFeYNQrc+hI9SBzf2bo3/vOtaFwfBgNAAoKSwZ1j+UqhlemXk423d2NFg0awNKQNtrTJQDGMg4Vw+kff3IWZiR0Yma+o1b/H7/g4avhMwRCsa+/6AbkCdviaL09+jtBqRyMS23N5maXdYz2etD7dn4EELoCHXo+pzUSLZ7faKxz/FlTumyC5mURskBpaiYAbK+cT2O9h0FIB38u93uYZmfvcG9uL/nfgR6Amg+3ozecC96w71oPt6MQE8gep3c/C8/ptagNFNaXmOiHOC2egGUsTOtXgBlhIGMTtSchfk1gH/x+/3jAHwzEAiMGOPj9/trAPxnIBC4QO8Fks68vlhwMr5EyqyIIiAI0t9lbg/QuG74fVPd3mv+mWk1jeWZlDdp2YAw2ShhI2mdxqVl89BMnkeiUrzDRw9Hr08UMMnlf/IAhtZga3Q4gXI4g5JZG1AC5o/3JrJY8o2xyO7GWr0AyggDGZ2kzcgEAoFrAdwH4J8BNPj9/lHydX6/f7Lf7/89gOcBnGfUIklHk6oBYejXvnIRcEoJ4BKkrysXSZcLLiAyKDXyK6W6/aRqs55BlJrMSSZn0xOdjRcEAS7Bhc6jnYaXNaWTqileHjagLA8zOruQKKBUSjcRTe5vCiOccriCmSVeWl9jIodLPmeciIzAQEYnqupiA4HA9/x+fzeAuwD8r9/vvw7ATQCuA+AF8AqA1PNoyR6m1kgjkwcjUqP+3Uk2AZbPzyVr8Fdyu6XHNZmaTEMmZ9MTnY2v21gnPT7UjRI2mpZxvEZnF5IFlDJlwASkHqCQ6jqzNqCUceQx5REGMkTmYhZUJ6ob/AKBwN1+v/8YgHUA9g1dvB/ADwOBwO+MWBwZoLBE2vclfl8YQMqsuN1S6Vh46HNNbvAXFQ3+yuDH45Uez4JNMdVsoBhPzV4f8tn4hv7YlkhmljWppXYal9EbKqYKKMuLy6OBU6KyM2VQCCQvEQSSB6UCBMNKvDKZeEbkQAxkiMz1odULyBWqAhm/3y8AuBrA94YuEgC8D+ALgUDAmg1EnKqvWxpVfHC31Kvi9UllWVNrzAsGJkwBalco1jEg9bjI63j63thtUzX4A9LjWBDEAOoyDQIEeOBBCCFNZ9MneifCA0+0lyOTXhs7MTK7oDagTFR2Fh8UphtGkIjIE1tE2WIgQ2SuPqsXkCvSBjJ+v/8KSCVlZwIIArgHwBEAPwPwnN/vvyQQCHQmfwSKOrR/ZCYkFATefFkq95qzWAoyzFBYIu37kmjvF7UDAbw+y4IYQH2mIdMNCs/0nZn2AN1Jk6uMyi6oCSiBxGVn8UFhqoAxVdlZsj2BiEgVHlQRmYv/z+lEzfjl3wOohDS97IxAIHBrIBC4D8BSSMFNi9/vn2jcEnNEX7cUxAyGhpdzAdLfB0PS9X02SHCpHQhgQYN/vEx3bldjhm9GtNFbPkC/d8W9WL10dfQAmpOrUjfGKymb9WXx/TPJhhEA5o5fJsozH1i9AKI8w0BGJ2pKy/4M4PuBQOB15YWBQOCJoZ6ZTQBe8Pv9lwYCgVYjFpkT9rZImRggeQN9OCzdLlGWJFtaStrUDASwqME/EaMyDUb3luilN9yLncGdaAu2IYQQvPCi0leJGb4Zpq0tWenaWNdY9ER6IEJMmNVS7i0DJO+DAezZp0SUI963egFEeeaY1QvIFUJ8qYdWfr//8wCeATAYCARK091+6D4iAAQCgax+tqM8viZWrnXrw8PLtU4piQULXh9w1R36/uxEJW1ArLk/UUlbJvdxggx6lHrDvbadXNUR6jCstE4PveFe1PfVR3uNknHBNWIfmXjRyXEJBgkUoADLi5frvn5STUh/E7Kzhpb2owCK096QiPRw0sKaih6rF5EL1JSWpRQIBF4AMAeI272OhpODGCB1A31I5zPLmZa0yQMBJs+SDvYhSF8nz5Iud2IQc2i/tMnn/pdiv49QENj/d2DzfcCbryS8m5zxWV68HDcV34Tlxcsxb8w8y4OY3nAvmvqbMIjBEQf/EUQwiEE09TehN9xrzQKhruxsomciri68GksLl44oETzJdRKEoWNkI/bAIaIolpcRmSPEIEY/qscvpxIIBF4bysxQMqob6HWegJVNSVuqgQBOowzoEhFF4MWnpe8nzzRvXVnYGdwZDWCSNcLLG3Za2QifqOxMDmoiiODdwXexM7gTM3wzRpQIKjM6RuyBQ0RR70PqeyUiY6Ufx0mqZZ2RkQUCgQN6PVZOsqqB/uDuWCZG3hMmotgTBpCuP7hb359rN3tbgMGh8qbOHqm879t10tdOxYmRHQ32GLigQluwLRrI2L0RXs5qfXnsl+EZOn8irz2EEFqDrajvq0dHqGPE/ZJldFxwwQOPLfqUiHIAMzJE5uCkXx3pkpEhFaxqoLeqpE0NM/fUObgb0Y10124C3uuSvn+vSwpm7r5e+r2IEXzS2oy/nVNiafO8GiFFNacTGuGVpXDx5B6fLf1bcMXYK3Ca97TodUbugUNEUWz4JzLH4fQ3IbUYyJh1MF1YIjXIp2ug1/sA3qqStnTM3FOnr3tkQKc0GJaCm6HgUjj4GlorTx6RMdgX3Gd587ySF95oMOOEDTvVlMKJELG5fzMuH3v5sNfZqMl0RBTFQIbIHG9ZvYBcoltpmSPJzd9vvjy8+fvNl6XLD+3X9+dZ0UBvxz1hzNxTR/4dK41PECwqgpuCwZGTs+zSPN8b7kXz8WYEegLDMjJOaIRXUwoHACJENPU34Z3QO9Hnen/P/Qj0BNB8vNnS158oh7G0jMgc3KpER/mbkUnV/C1GpBKw7U9IAYbemRkzG+jtuCeMWXvqJPsdr1wklZMNDq0hLjs14BFs2TyfbNQykHr/Fbs0wqcqhTt89DDqNtZFX+cwwtjcvxkCBNtnxohyxLtWL4AoTzCQ0VH+ZmTiD6YTNX/LB9NOJpe0ebyxzIxMcEmXG1HSlopZAwiS/Y7XbgK+98/AqaUjslNhAWib4LNd83yqUcupRBDBWwPmZrGVWSNlJsWjOG9SVlwGQRi+9YjydRaH/rNrZowoB71h9QKI8gQDGR3lb0Ym0cG0qDiYvvv62MG008cPyyVt0V6gAaknxqjG+nTMGkCQ6ne84U8Js1MRF7Dr9DE4st1ezfPp+ksAJMwgAcDfPvkbRgujUTWqKuFj94Z7sTO4U9Vwg3S3TZQ1kjMpIsToppfLapdhfeN6HD4a63mMf52TPVerM2NEuWhhTUV3Q0v7OwBOS3tjIsrUBwtrKo5avYhckr+BjJ2neRnBLnvCxPe9GDmAQO3vGAAEF0IuEU0zC3HsBE/S5vlxHw1i5oEg8O4a4yetKSTqLxFFcVgWo/NoJ0QMlWltqMPqq1dHg5nmj5sxwTthRGCSKvCIL+FKd9u5Y+Zi2/FtSaeSKcmlcHUb65IOKUj2XFcvXR3NjJkVyGgJ9ogcbCcYyBAZidkYnTmvtKyvG9ixBXh8DfDYbdLXHVu0N4Z7fbHvx5dIB9GA9dO8cp2yVK+zR+pTkQOMsqLYAAIg+wEEan/HADB5Fl65ZDb+7+TRABI3z3/mcBBLth3F1H8cN2c4hEK6UctHeo5AlMdLAwhHwqjbUIeuY9KYaREidgV3DXvMVOVq8SVcam773PHnEIZUytd1rAt1G+uwat0q1G2MrUPA8HKyZEMKUj1XmVmZsY5QB+r76tEabI3+HlLtfUPkYLvS34SIssBARmfOysjoObJ3UrV0PzEiHTzHN5wD5k/zygfKvpe1m4AjvdL3ggB43FKjv2xqTXbjsdX+jifPAmZfjjPDvdjZ9w9EEBnRPD/uo0HMf+UovGEAioABgGHDIZRZAKVk2SJlmRYgBTNyBgPAiAyGmnHIcgmXsl8l1ehkWbJMiggRHngQRhgixJRDClI9V8CcsdJq9r5p6m/CksIlzMxQLthp9QKIchwDGZ05J5DRe8qYHad5WcHMTSkBbeVefV3ZBa4af8fyLvKJJoPNOPAxXPJfk01aGwwBu58DLvwnjS/KSKkmlMn9JfE9MnUb6hCOhIfdNlUGI125mrKESxnIJLtt/M9NlkkZxCAWjl2Ipv6maECTTLLnatZYaS3BHvt1KAcwkCEy1l6rF5BrnFNapveUMTtO8zKb2fvoAOrLvTwF2e81k8HvWN5FvspXFT3jX4ACTD00ALd8vJ1s0hoAtL+W9euWbkKZnMW4d8W9WL1U6oMpHVeK1Vevhtvljj1FpM5gpCtXkw1gQPVtZcqpZIkyKfLrXOyKZeASlaMleq6AFMiYMVZazd43Vk2yI9LbwpqK98D9ZIiMxIyMzpwTyBgxsteKDSrNkq6XqK8b2Pa4OZtSKpVPjH2faoPOE4r0CVwz+B3Lu8gvL16Om4pvwvLi5fAMKjId6TJJzRuA9w6ofUVGUGYBCju6sPT8OqwoW4Wl59ehsKMr6f3kYObkk06Wek5OivWcCBBGZDC88Ea/TxV4uOBKOjo5UYM+oG6DziJ3ET6MfBi9T6pNMuPXMn/sfFNKubQEe5lKNq6a46XJIuyTITJG+8Kaih6rF5FrnFNaZtSUMbtM89KTml6ith1AeKju38hNKZX6uoEPFAf4ycq9XC7go179xmPr8Tv2+mLvQeWkNUAKqG99OPa6iSLw7KPARUszCoaVWYDLF69H8VudcEVEFL/VicsXr8fGF2NlXAUowEmuk/BBRDqJmqznxA33iAxGpa8SrcFWRBBJWsIFSBmHRKOT428rN/Kn6n2Jz6RoyfQUoACVvkpM9003rR/FC290jUb062iZGkdkkl0Avmz1Iohy0J+tXkAuck5GhlPG1FH2EiXLtGx7HHj3zdjlRm5KqbS3BYgMrSlZlgUAxp8ODCoCUjuMx55UHStPkzNJSvElZmIk44yW8uC++O0jcEWk5+6KiCh+e/jB/fLi5fjGuG/gi6O/OGIiGCAFF8kyGDN8M+Aa+icgWQlX9OkM/Zfqtm64cdHoi+CBJ/q4smSZFLVZoQIUYHnxcswbM8/UpvpKX2X0uajJMmmhZWrc/2/v3sOkrM/7j79nD25ERckioiENDxKgaCJaj31U6mXUGNTHaLFBmxhtJDqJlZikV9VfbDUpmsZ6YfE3v0YSgzGRekh1WdB4wEjsKEZEjKJgAkuMiBzWhRWEPc38/vjuMLO7M7Nz/j7PzOfVa69ddp/dvWcw9Lnnvr/3LVJBr9gOQKRKPWM7gGoUnIqMpozlZvBZorSVlpQJTFs74L2UdqVyJgq5LCEF2LoxcwUkW+JazsEF6QYHXHmbSf4A4nF6N2/nmE330DpmJhMac6topdtPkqh8AHRMPHRfRSZWF6JjYvoqwNEfO5qR9SNZ/tFyPogld22NqhvF9BHT+YvGoashUocbpE7lynSoPZEohQgNuPGu6/+/RPVgXOM4Xu16lbVda+mmO2slJZeq0HCJQjl3vBzXdBxvdb2VdpJd6uMfXO3KJSYNEhCf+l8gRpBe6BTxvxiwzHYQ1Sg4/1BNdc2EKUjeSN57g3mfGNlbC1PGhpPLWaJU6T5XrgpXPu2B6Sog6c7SJBLXTIML1r0Ej82Dt1cWF3ticECqlMpgXwjWjYa1Pe2cvzW3ilam/SSpU7wWL7qKjk+PIVZfR8enx7B4Ufqb+409G2nd1cqO2I4Bv2NHbAetu1oz7jpJHLpPraBkOquSGJ08eBDC0U1Hc9nIy/a1QKU7Y5SpkpJLVSjbwf5y73hJJHv5VJlyjUmDBMSPPNfZCrxsOw6RKvOq5zofDH+Z5Cs4FZnEjeTgsx9gbmjr66t/ylgu8hlvnLhmsFIupUyVT5Uln9HJ2UZzg/lv5cVHzceTji88/nGTwTnGTCaDfZXB3s3bWTcazp8FMeKs6xm+opVtP0mqzvGjB5yJSUi9uS9218kh9YcMqLBkO6vSQw9njDijZBWCbCOvUys96eKu1I6XRLKXS5Upn5gqMUhApEBLgJNsByFSRXQ+pkyCk8hAcgLVvvahbnPTW869J0GTa7KQMPiaw5uHLqUslXzaA7MlronrTjjPXLdi8fDtdAArHoOxTnH/nUw7E/78pkma+hOsYzbdw9qedmLEqSPE5MbhK1qDJ5NdMGsBo/64jY6Jh7J40VV0jh+d9vvS3dzn8rOGa1Eq16H2XFqs8kkUMj2H5W7NSlSZhvs5+cRU7kECIkVYCnzfdhAiVUTnY8okOK1lCYkJVJfeDJf/wLw/+QIlMQm5tGSlynbNJyaX9nnNtz1w3GSTrKRbmBiPw8tLTEtZpna697YPHCQQj+e+ZyiTRIKVGDYBtI6ZyZTGZuoJMaWxmdYxw5/ZSjuZrC+2bzJZQh11Wdu4cv1Zw7UoleNQez5tX/m0o6V73H5pzconpnIOEhAphuc6rwKbbMchUiX2YM6eSRkEqyIjw8upJav/r72vN/s1J84obWz5tgd2tptkJZ4mkSGe3HWT2lKW2k4H0Ns3cJBALuOahzNuMnzuCjNiOR5jQuMo1nwiS+tbGrlOJjtodzdfffdTKQMMtsGEDwdUIHP9WdlalAo91J5JJdq+/NialU9MpX7ORUpsKZDmHzYRydPznut0DX+ZFCJ4FRnJLpdt9n9zqXkb7ppyVLnyWVA5eAJbpqWYdSmPIV37XDmmsB1xpNkTk+05zHJmK3XscMfEQ4nVmQpP6mSyT23p4rLnOoYOMHj7ZTPY4N11Of8syN6iVMih9mwGt1jd9ovbuH7+9dz2i9vYvtNMyeullxV7V+T089LJZ3RzpeQTU6mfc5ESW2I7AJEqobayMqrtRKaz3ZyveOBWuO8m837F4tJvs6+0XJKFAjbel0yu7YG5TGCLxwZ2ns2ZCQ31A39OuaawFfEcprYVpZtMdvDuXmas3EljXzzzPqD+PTXD/SzIrUUpcVZluKlkucilxQpgXfe6gieL+bE1K9+YSvmci5TYMmCv7SBEqsATtgOoZqF42rad8gqHw3GASCRS8d+9z7vrhm9xKufNvAzvvpuSH6fubAFzpufelFacUF3y7zHjgf8QTD6x8NayYvbUDPreeON+vPGJBl458mPsPGBoh+fZr+xk8qYuc0uc6fGE6mDSCew44XR+2fnLrBPQGmgoenpXPu7quGvfx9fPv55Yyv/G6kJ13HntnUXHtqNvh+8etx9jqpCh21gl8FqibY8D59qOQyTAXvVc5zjbQVSz2qzIpI7rHebV7qoVhGpUY1Py45SdLUPHNTfByV7yz5kGCTQ0FD6FLdOemkFtXrl+b6inm6P/tIfLnuvA2TJwbLSzpYcpiSQGslejNqz2ZYtSphYrgFg8NqDFLDHFK19+fNx+jEmkCGovEynOfbYDqHa1mcjkevai2AlXflXMTXkl5bMUc9LxcMoXB0wTSwoNe2Ylo852+O1DsOznhSW+WZLmUDxOY1+cGSt3Mnp3/1LI3XXMWLlz4Mvbwy0PxX8tSularFKVarKY3x63X2MSKdCjQJ/tIEQCqgf4pe0gql1ttpY9cGvyBv7Ge4buUUlMuGpsMuc3qklnu0lWMi2PBHPTf/619kdaFxLrgBauIvcMJdoPU3//MG1eQ9rWVixO7s7J5XvTXf/e9uTPC8h/o+larIZrMbtu1HUVjVFKSq1lVaol2tYKnGc7DpEAavFc50LbQVS72qzI9KRMwcvh1e6qEqRqVC4T2AZXWUq1Zyi1kpJqmDavIXIdWJD43nTXp0pXjfKhRItVKr9MFhORvPzUdgAiAbXQdgC1oDb3yDQ2JZOZwZvtyzXhyi/S3SjHU26s585O3lgXu2+lFBKTwUpVZcnV4IQvUUlJHTiQS+Kbb9Kc6XowbXWpO3+y7Knxg/GN45ncOJl1PaZV8arzrxqy6R609NGGHX07WNW1irVda+mhh0YamdI0heOajtP5HRlsCbAFOMx2ICIBsh2zi0nKrDYTmQnTku07c2YObfcBX7/aXZQgVqMSVZZKJlaZEr5UuSS++SbNuV4PhZ35qbCT9z+Z9T3r6aVXSx99YmPPRpbuWrpvMSmYRZ5rutbwVtdbzDhwhs7xyD6e6/S2RNvuB75jOxaRAHnAc50sffFSKrWZyEx1Yf0q6I1l2Wxfple7ixnhWwp+r0bZfn4SslVGwFRHckl8c0maCcGIgwee3YIs1wPOMYEYD55oMRt84wwmgamjTlO8KmhH3w6W7lqadjx04u9n6a6l1TgeWopzD/BtdBZKJFcLbQdQK2ozkUmcvRhuj0ypb5zT7a5JTAtbv6oyu2v8XI3yw/OTkC3hSz1sn5Ap8c0laSYOnduHTkTLdH1DI0w7s6CHZUNiiterXa+ytmst3XSzH/sxpWkKxzYdqxvmClrVtWpfMrl95/YhrX6jDx69bxz2GSPOsByt+IXnOn9oibY9DZxtOxaRAPi95zr57xSQgtTmYX+o/GZ7v+yumeqam27IvG/FxtkLvzw/CbmMfk7INtp5ZDMcdfrwv2/wY04n04CDADik/hDOGHEG14y6hutGXcc1o67hjBFnKImpsLVda/clMgtaF7C1YyuxeKxk47Clqs23HYBIQCywHUAtqc2KTEIlz15kOjyeOoY3MS2snPHYqkYNxy/PT0JOlRRMi9e0M9M/X53tsHoZtL02/O/L9JgTbLXYSVXpIdmyva1jG4nx+/F4nG0d2/Z9rRsfnZETv3gc2ABMsB2IiI9tB+61HUQtqd2KTKXlO4a3nCpdjcqFn54fyG3085lfgdMvSZ9YJJaOpiYxmcZdQ+bHDMldMYWMkRZJ0Ujjvo81Dlvy4blODLC0/E0kMOZ7rvOR7SBqSW1XZCrJb9PCbEwCy8Zvzw8UPvp5uB00g8ddg38es1S1KU1TWNO1hhgxjcOWQiwAbgQ+bjsQER/aBdxtO4hao0SmUvw+Lcw2vz4/hSR8heyg8dNjlqp1XNNxvNX1FjFiGoctefNcp7Ml2vZD4Ie2YxHxoQWe63xgO4hao9aySsnl8Hi17q7JRTU9P5na5FINTlaC/pglEBLjsBtooG7QP/911NFAg8Zhy3DmA5ttByHiM3uBO2wHUYtUkakUm7trgqCanp98d9BA8B+zBIbGYUsxPNfZ0xJt+zfUQiOS6see67xnO4hapESmUvw6Lcwvqun5yXcHDWD2zKUkPJV8zH5ZQioVkxiHrV0xUqAFwHeA8ZbjEPGDPcDttoOoVWotqyQ/Tgvzk2p5fvLdQXPKF2HyiXYec2K62tsvJ5OvxBLS1vnm6yIiKTzX6QZusR2HiE/8P8913rcdRK0KxQe3vVRAOByOA0QimuQoBfB7BaGz3SQBg6eWDZZtB00l5BJnQ6NJqPzwvIqfhWwHIJXVEm2rB94ANOJOatluYILnOlttB1KrVJGRYAlCBaHYHTSVMni6WrodN4klpCIiKTzX6QP+xXYcIpb9QEmMXUpkJDhS97OknqEB8+feHvP1zvb0319JQWiT89sSUhEJmoeB1baDELHkTeA/bAdR63TYX4Ij036WxASwMaOSFQQ/LPr029LRwfy4hFREAsNznXhLtO2fgKdsxyJiQdhznWF6yKXclMhI+ZT6LEu6CkI8pYIwd3ayguDX5MFP/LqEVEQCw3Odp1uibfcDX7YdS6m1PvQznm59kHg8zlkX/B0XXHIlP7r5Wja9swGA3bs6OeDAkcxbuHTI9y5+8Kc83foQoVCIT02YxLU3/oj9mpq4L3I7q15ajjNxKnO+Z17M/82vH2VX5w7Ov+SKij4+KcrPPddZbjsIUWuZlEs5zrKoglBa1bSEVERsmgNssR1EKf1pwzqebn2QHy14lHkLl7Iy+izv/bmN7946n3kLlzJv4VJOmf55Tpl+zpDvbd/2PkseuY87ftrCf97/a/piMZ5f1sruXZ2sfWMVd933BLFYHxvXr6Wray/PPvEI51709xYepRSoAzN+XHxAFRlJr5hqSupZlsHiMbP0cvmi/KdhqYJQWrksIY3HoHuv+TvV5DIRScNznQ9aom3XAg/ZjqVU3t24nklHTaPpY/sDcNSxJ7Hit09x0WVfByAejxP9zeN8/65fpP3+vr4+urv20lDfQHfXHj4++jDq6uro7ekhHo+brzU08tgD93De336VhobGij02KdoNnutssx2EGKrIyFDFVlPKNQ1LFYTSyjZdLdXG1/0zEU5EfMlznYeBR23HUSp/MWESb67+HZ07O+jau4dVLz7H9q2b9339zdde5pBRzRzxSWfI9zYfOpYLv/Q1rrr4VK648GRGHHAQx554GvuPOJBT/ubzfOuK8xhz+CcZccBB/OGt33PSaWdV8qFJcV4C7rEdhCQpkZGBSjEZrFzTsKa6Zts9JCsI995g3o8ZZT5fX2+uk9wkpquN/0zma/w2EU5E/OobwA7bQZTCJ8dP5It//3X+9Vtf4ZZvf5XxE6dQn/j/P8DzzyzmtM+lP4u5q3Mnv/vfZ/jxQ8u597EX2bt3D889+RgAF132deYtXMqV197EAz+5k0u/9i2ebn2Qf//eN3lo4d2VeGhSuD7gas91Kr+AUTJSIiMDlaKakvNZli7ykst+lumz1AKVr5HNsN/Hks+pdsqISAE819kMfNt2HKVy1nl/x533tjL3/z7IgSMP4fBx4wHo6+3lxeVPcuqZM9J+32sro4w5fBwHj2qmoaGRU04/h7WvvzLgmg1vrwHgiE86/ObX/8M/ff9u3ml7m/f+3FbWxyRFudtzndW2g5CBlMjIQKWopjQ2JT8e22zOsMDQsyyQ/yv8QdjPEkS2d8p0tsOKxfDArXDfTeb9isWqAIkEjOc69wLP2I6jFHZ0bAdg2/ubWLH8SU7vr8C8tjLKuE8dyegxh6f9vkMPO4K316yma+8e4vE4v3/lBcaNnzjgmkQ1pre3l1jM/NsbCoXo2ru3jI9IirAGuNF2EDKUDvvLQKWYDDZhGqx7yXw8Zybc8d/mVf14HHr7zMeJVrBCdr74fT9LENmcCPfuOtO21teXTKYSZ7LWrzJVNiWoIkEyG3gdOMB2IMX44U1hPuzcQUN9A7Ovv4UDRx4MwPPLlnDa584fcO0H27dw9+3/zM13/IxJR03jr8/4PNdfeT719Q04k6ZyzgVf2nftit8+xcQpn+Xjow8DYPJRx/KPX/k844+cgvPpv6zcA5Rc7QZmeq7zke1AZKhQPF75Vr9wOBwHiEQiFf/dMowHbk3e1N54z8DJYIc3JydbNTbBpTen/xmd7fDonck/F/pzpHJK8fdeiM52M0gg3YS7hIbG/CfcSaWFbAcg/tISbbsKHYqW6vBlz3XSj6cT69Ra5gd+aqspxWSwwTec2vnif7YmwpVrwp2IWOW5zgJgge04RIr0EyUx/qbWMtv81laTy26RXCaDaedLsJTq7z1f6c7mxFPO5sydnTybo1ZCkaD5JvBZ4CTbgYgU4PfAtbaDkOxUkbGpFKOOS61Uk8H8vPPFTxUwv7A1Ec7m2RwRKSvPdbqBi4EttmMRydMu4BLPdTR9wedUkbFpcFvNvIfNzdvY/pv9MaOSbTWVfDU6MRnszah5Jbyn21ROJkwzr8jncjNr6xX+4fitAuYnpfh7z5cqdyJVzXOdTS3RtpnAMkDr6yUoZnuuoy3QAaDD/sXqbE+58esyN2a53vjZOmBdKemSBjCv8NfXVz5p0MFy/1mx2CSR8VjmZD5UZ8Zrq7XMz3TYX7JqibZ9E5hvOw6RHPzYc52rbQchuVFFphjFvrpf7W01Nl7hz8avFbBa5tfKnYiUlOc6d7dE244HLrcdi0gWLwJzbAchuVMiU6jU8y2DxWPmxmz5ouyv7tdCW42fdr4E9WB5MVU/v0uczRmuchf0xykiAFcDRwN/ZTsQkTTWADN0LiZYlMgUqhSv7k+YlmyrmTNz6M8Aewfiq1EQK2ClOtPj52TIb5U7ESkLz3X2tkTbLgJeAsbajkckxZ+AczzX6bAdiORHiUyhSvHqvtpqKitoFbBSVP0gGAMO/FS5E5Gy8VznnZZo29nAcmCU7XhEgG3A2Z7rbLIdiORP45cLVYpX922NvK1Vfh4JnU4plkX6ccS3iNQ0z3VeB74A7LYdi9S8XcAXPNd523YgUhhVZApVqlf31VZTOUGrgJWi6leNAw783CYnIjnxXGdFS7TtQmAp4JMyuNSYbuBCz3VW2g5ECqdEplClPN+itprK8PvB8sE36KkKrfoFdcBBJkFokxORnHiu80xLtO1LwEPofkQqKwZc5rnOMtuBSHH0D0ehgvbqvhh+rYBl2rmTUGjVL4gDDjIp1ZkhEfENz3Ue7U9m/hvdk0jlfMNznUdsByHF0z8ahfL7q/uSmd8qYNlu0BMKrfoFbcBBNtXYJicieK7zKyUzUiFx4HrPdf7LdiBSGjrsX4zEq/uTTjA3jITM+0knmM+rxUVykcuh/kTV794bzPsx/cN+hqv6BW3AQTbp2uRiKW1ykGyTE5FA8VznV8AsoNd2LFK1+oArPdeZZzsQKR298lEsv726L8GTyzmWwXKt+lVTC2Q1tcmJyBCe6zzSEm3rAhYBB9iOR6pKF/Alz3Uesx2IlJYqMiK25XqDDuRd9aumEd+NTcmPxzab9jgIZpuciKTluU4rcBqgnR5SKokRy4/ZDkRKT4mMiG0536A3weU/gEtvNhXAXJOPammBrKY2ORHJyHOdV4GTgNWWQ5Hgew84zXOdZ20HIuWh1jIR20o5yjuTamiBrKY2ORHJynOdTS3RtlMxbWbn245HAul1YIbnOn+2HYiUjyoyIrZNdc0NOBR+qL8WVFObnIgMy3Od3cCFwF2WQ5HgeRo4VUlM9VMiI2KbbtBzVy1tciKSE891Yp7rzAG+iZk6JTKc/8RUYjptByLlF4onDhZXUDgcjgNEIpGK/24R3+ps99+iTpHchGwHINWvJdp2LvAgcJDtWMSXOoArPNdpsR2IVI7OyPjRgBvaLvOKs25oq181nGMRESkTz3WeaIm2nQD8AjjedjziKy8AszzXecd2IFJZai3zm3fXQet8c/g7MZa3p8v8uXW++bqIiEgN8lxnHfDXwL+hVjOBOHA7MF1JTG1SIuMnne2wfBH09iQXJCbEY+bzyxeZ60RERGqQ5zo9nuv8H+B0YIPteMSarcC5nuvc4LlOr+1gxA4lMn7yZhT6+l9g2toBN94DV95m3m/tMJ/v6zPXiYiI1DDPdV4ApgE/sxyKVN6zwDTPdZ60HYjYpUTGTzasTlZi5j0Mm9shFjfv5z1sPh+PmetERERqnOc6H3qucyVwEaB2heq3F7gJOMtznc22gxH7lMj4SeJMDJiFiImJcvG4+fO+67orG5eIiIiPea7zKPAZ4Ne2Y5GyeRw42nOduZ7rxIa9WmqCEhk/aWxKfjy2GUL9E01DIfPnfdftV9m4REREfM5znc2e65wLfAXYZDseKZmNwIWe68zwXGe97WDEX5TI+MmEacmFiHNmwuHNUBcy7+fMNJ8P1ZnrREREZAjPde4HJgH/CnxkNxopQhfwfWCqdsNIJtoj4ydTXVi/CnpjMGYUzJ099Jr6enOdiNinnU8ivuS5zkfALS3Rtp8Ac4Evo8WtQfIE8I+e6/zRdiDib6rI+MnIZpg+Cxoak5WZhFCd+fz0WbpBEvED7XwS8T3PdTZ5rnM5cALwvO14ZFh/Ar7ouc4XlMRILpTI+M24yXD+tTDphP4zMyHzftIJ5vPjJtuOUES080kkUDzXecVzndOBv0W7Z/xoI/ANYIrnOo/ZDUWCRK1lfjSyGU6+wLyJiP8M3vk072EzWXBs/3m2MaOSO5/0v2MR3/Bc51ct0bYlwNXAt4FPWg6p1r0F3A48oKWWUghVZERE8qWdTyKB5blOl+c6dwFHApcDayyHVItWAhcDR3mu83MlMVIoVWRERPKlnU8igee5Tg/w85Zo2/3ADOA7wHS7UVW95cBcz3Wesh2IVAclMiIi+WpsSiYzY5tNJSYe184nkQDyXCcOLAGWtETbPgNcC1wGjLAaWPXoAVqB//Bc5wXbwUh1UWuZiEi+tPNJpCp5rvO65zqzgXHAdzFnOKQwazDnkD7huc7FSmKkHFSRERHJl3Y+iVQ1z3U6gDuAO1qibUcDM4FLgClWA/O/bcDDwH2e6/zOdjBS/ZTIiIjkK7HzafkiM50sdQRzqM4kMdr5JFIVPNd5A3gD+Jf+pOYSTGKjpMbYCTwKLAKWea7TZzkeqSFKZERECpHY+fRm1Ewn6+k2Z2ImTDOVGCUxIlUnJam5uf88TSKpqaUlbzHgNeBZYBnwrOc6Xdm/RaQ8QvHEtJ0KCofDcYBIJFLx3y0iFdTZnnKj32UOyetGvxqFbAcgYlNLtG0C4Ka8HUV1/e9iLcnE5TnPdT6wHI8IoIqMiJTLu+uGtl71dMHbL5vzJdNnmaqGiEjAea6zAdgA3A/QEm0bBZxCMrE5EdjfWoD5iWMey3JM8vKs5zqb7YYkkp4SGREpvc52k8T09gz9WjxmDskvX2Ras1SZEZEq0z8s4PH+N1qibY3AsZiEZiIwof/Nwd6Y5y7gbUy15a2U9+s819ljKSaRvCiREZHSezNqKjEAWzvMtvv3282OlTkzzaSvvj5z3ckX2I1VRKTM+pdv/q7/bYCWaNthJBOb1ATnUOCglLf6HH5VH+bw/U5gR/9b6sfvkUxa2nQwX4JOiYyIlN6G1cl2snkPJxdGbm43f54723x9w2olMiJS0zzX2QJsAV7Mdl1LtG1/TPWmcdBbPfAhsMNznQ/LG62IvyiREZHS60kZYPN+fxID5v377SnXdVc2LhGRgOpv91LLl0iKOtsBiEgVamxKfjy2GUL9w3tCIfPnfdftV9m4REREpGookRGR0pswzSyGBHMm5vBmqAuZ93Nmms+H6sx1IiIiIgVQa5mIlN5U14xY7o2Zg/1zZw+9pr7eXCciIiJSAFVkRKT0RjabPTENjcnKTEKoznx++iyNXhYREZGCqSIjIuUxbrLZE/Nm1Ewn6+k2Z2ImTDOVGCUxIiIiUgQlMiJSPiObzXhljVgWERGRElNrmYiIiIiIBI4SGRERERERCRwlMiIiIiIiEjhWz8iEw2Gbv15EREojHolEQraDEBGR2qKKjIiIiIiIBE4oHo/bjkFERERERCQvqsiIiIiIiEjgKJEREREREZHAUSIjIiIiIiKBo0RGREREREQCR4mMiIiIiIgEjtU9MiJBFw6HnwLOAi6ORCL/k/L5EPAz4HLgh5FI5J8thSgiIiJSlVSRESnOd4EY8INwOFyf8vk7MEnMAiUxIiIiIqWnREakCJFI5DXgfuAvgS8DhMPhG4HrgYeAq+1FJyIiIlK9tBBTpEjhcHgc8AdgC6YSMx94ErggEol024xNREREpFopkREpgXA4fBuQaCF7ATgrEol8NOia04HvAH8FHAFcEYlEFlYyThEREZFqodYykdLYlvLxPwxOYvodCLwBXAfsqUhUIiIiIlVKU8tEihQOh2dhWsreB8ZiEpVrBl8XiUQeBx7v/56FFQxRREREpOqoIiNShHA4/AXgPmAN8FlgLfC1cDg8xWpgIiIiIlVOiYxIgcLh8KnAI8C7wNmRSGQb8D1MpfN2m7GJiIiIVDslMiIFCIfDxwBLgJ2Yg/2bASKRyCPASsALh8OnWQxRREREpKopkRHJUzgcnogZrxwHzolEIusHXXJD//sfVTQwERERkRqiw/4ieYpEIn/EHOrP9PVngFDlIhIRERGpPdojI1Ih4XD4QGBi/x9fwJyjWQx8EIlE3rEWmIiIiEgAqbVMpHKOB17tf9sfuKX/41ttBiUiIiISRKrIiIiIiIhI4KgiIyIiIiIigaNERkREREREAkeJjIiIiIiIBI4SGRERERERCRwlMiIiIiIiEjhKZEREREREJHCUyIiIiIiISOAokRERERERkcD5/yA19sTD55/OAAAAAElFTkSuQmCC\n",
"text/plain": [
"<Figure size 1008x720 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"show_results(y_proba_val2)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"End time is : Tuesday 8 December 2020, 19:03:13\n",
"This notebook ends here\n"
]
}
],
"source": [
"pwk.end()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---\n",
"<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
}
},
"nbformat": 4,
"nbformat_minor": 4
}