├── 82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg
├── Extract_Skin_from_an_Image_and_Find_the_Dominant_Colors_Tone.ipynb
├── Extract_Skin_from_an_Image_and_Find_the_Dominant_Colors_Tone.py
├── Human-Hands-Front-Back-Image-From-Wikipedia.jpg
├── README.md
├── skin.jpg
└── skin_2.jpg


/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/f1988037f6f365ece3aec00c0e7ae52e2b3f9938/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg


--------------------------------------------------------------------------------
/Extract_Skin_from_an_Image_and_Find_the_Dominant_Colors_Tone.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "name": "Extract Skin from an Image and Find the Dominant Colors/Tone",
  7 |       "version": "0.3.2",
  8 |       "provenance": [],
  9 |       "collapsed_sections": []
 10 |     },
 11 |     "kernelspec": {
 12 |       "name": "python3",
 13 |       "display_name": "Python 3"
 14 |     }
 15 |   },
 16 |   "cells": [
 17 |     {
 18 |       "metadata": {
 19 |         "id": "RxymLTBIdt39",
 20 |         "colab_type": "text"
 21 |       },
 22 |       "cell_type": "markdown",
 23 |       "source": [
 24 |         "# How to extract Skin from and Image and Find the Dominant Colours/Tone\n",
 25 |         "\n",
 26 |         "This the online notebook containing the explaination of the code for the article found at https://goo.gl/bpkVn3 \n",
 27 |         "\n",
 28 |         "If you are interested in testing out with code create a copy of this notebook and go to  menu on top \"Runtime -> Run All\" or download the notebook in my github -link here-\n",
 29 |         "\n"
 30 |       ]
 31 |     },
 32 |     {
 33 |       "metadata": {
 34 |         "id": "5_UVMQ6PXD2y",
 35 |         "colab_type": "code",
 36 |         "colab": {}
 37 |       },
 38 |       "cell_type": "code",
 39 |       "source": [
 40 |         "!pip install imutils"
 41 |       ],
 42 |       "execution_count": 0,
 43 |       "outputs": []
 44 |     },
 45 |     {
 46 |       "metadata": {
 47 |         "id": "9NiJ0QKzeRxC",
 48 |         "colab_type": "text"
 49 |       },
 50 |       "cell_type": "markdown",
 51 |       "source": [
 52 |         "## Section One : Importing Libraries\n",
 53 |         "\n",
 54 |         " - **numpy ** : OpenCV uses Numpy for numerical operation. Hence Numpy is used to align input with the respective data type\n",
 55 |         " \n",
 56 |         " - **cv2** : OpenCV used for image processing\n",
 57 |         " \n",
 58 |         " - **Counter** : Useful for counting labels\n",
 59 |         " \n",
 60 |         " - **imutils** :  Useful utilities for image processing\n",
 61 |         " \n",
 62 |         " - **pprin**t :  Library to pretty print data\n",
 63 |         " \n",
 64 |         " - **matplotlib** :  Normally used as a graph plotting lirbary , but we will use it show inline images since \"cv2.imshow\" doesn't work on collab"
 65 |       ]
 66 |     },
 67 |     {
 68 |       "metadata": {
 69 |         "id": "AuXA7RfbXOut",
 70 |         "colab_type": "code",
 71 |         "colab": {}
 72 |       },
 73 |       "cell_type": "code",
 74 |       "source": [
 75 |         "import numpy as np\n",
 76 |         "import cv2\n",
 77 |         "from sklearn.cluster import KMeans\n",
 78 |         "from collections import Counter\n",
 79 |         "import imutils\n",
 80 |         "import pprint\n",
 81 |         "from matplotlib import pyplot as plt"
 82 |       ],
 83 |       "execution_count": 0,
 84 |       "outputs": []
 85 |     },
 86 |     {
 87 |       "metadata": {
 88 |         "id": "ZujqTVQEjxaM",
 89 |         "colab_type": "text"
 90 |       },
 91 |       "cell_type": "markdown",
 92 |       "source": [
 93 |         "## Section Two.1 : Function to Extract Skin Color\n",
 94 |         "\n",
 95 |         "The ***extractSkin*** function takes an 8 bit 3 channel image in the BGR colorspace (as mentioned in the article this is how OpenCV reads color images) and returns the extracted image in same colorspace. \n",
 96 |         "\n",
 97 |         "The function works by using the** HSV colorspace** and uses thresholding (Thresholding is a process of filtering out pixel based on specified thresdhold parameter) to extracts pixel that corresponds to the skin color range,\n",
 98 |         "\n",
 99 |         "\n"
100 |       ]
101 |     },
102 |     {
103 |       "metadata": {
104 |         "id": "2kGCu9pCXYbN",
105 |         "colab_type": "code",
106 |         "colab": {}
107 |       },
108 |       "cell_type": "code",
109 |       "source": [
110 |         "def extractSkin(image):\n",
111 |         "  # Taking a copy of the image\n",
112 |         "  img =  image.copy()\n",
113 |         "  # Converting from BGR Colours Space to HSV\n",
114 |         "  img =  cv2.cvtColor(img,cv2.COLOR_BGR2HSV)\n",
115 |         "  \n",
116 |         "  # Defining HSV Threadholds\n",
117 |         "  lower_threshold = np.array([0, 48, 80], dtype=np.uint8)\n",
118 |         "  upper_threshold = np.array([20, 255, 255], dtype=np.uint8)\n",
119 |         "  \n",
120 |         "  # Single Channel mask,denoting presence of colours in the about threshold\n",
121 |         "  skinMask = cv2.inRange(img,lower_threshold,upper_threshold)\n",
122 |         "  \n",
123 |         "  # Cleaning up mask using Gaussian Filter\n",
124 |         "  skinMask = cv2.GaussianBlur(skinMask,(3,3),0)\n",
125 |         "  \n",
126 |         "  # Extracting skin from the threshold mask\n",
127 |         "  skin  =  cv2.bitwise_and(img,img,mask=skinMask)\n",
128 |         "  \n",
129 |         "  # Return the Skin image\n",
130 |         "  return cv2.cvtColor(skin,cv2.COLOR_HSV2BGR)\n"
131 |       ],
132 |       "execution_count": 0,
133 |       "outputs": []
134 |     },
135 |     {
136 |       "metadata": {
137 |         "id": "El9UHDb7mvGR",
138 |         "colab_type": "text"
139 |       },
140 |       "cell_type": "markdown",
141 |       "source": [
142 |         "## Section Two.2 :  Function to remove black pixels from extracted image\n",
143 |         "\n",
144 |         "The ***removeBlack*** function is more sort of the utility function to remove out the black pixel from the skin extracted. Since OpenCV by default doesn't handle transparent images and replaces those with zeros(black in color word).\n",
145 |         "\n",
146 |         "This function is useful when thresholding is used in the image."
147 |       ]
148 |     },
149 |     {
150 |       "metadata": {
151 |         "id": "ueCsY8mECI6Q",
152 |         "colab_type": "code",
153 |         "colab": {}
154 |       },
155 |       "cell_type": "code",
156 |       "source": [
157 |         "def removeBlack(estimator_labels, estimator_cluster):\n",
158 |         "  \n",
159 |         "  \n",
160 |         "  # Check for black\n",
161 |         "  hasBlack = False\n",
162 |         "  \n",
163 |         "  # Get the total number of occurance for each color\n",
164 |         "  occurance_counter = Counter(estimator_labels)\n",
165 |         "\n",
166 |         "  \n",
167 |         "  # Quick lambda function to compare to lists\n",
168 |         "  compare = lambda x, y: Counter(x) == Counter(y)\n",
169 |         "   \n",
170 |         "  # Loop through the most common occuring color\n",
171 |         "  for x in occurance_counter.most_common(len(estimator_cluster)):\n",
172 |         "    \n",
173 |         "    # Quick List comprehension to convert each of RBG Numbers to int\n",
174 |         "    color = [int(i) for i in estimator_cluster[x[0]].tolist() ]\n",
175 |         "    \n",
176 |         "  \n",
177 |         "    \n",
178 |         "    # Check if the color is [0,0,0] that if it is black \n",
179 |         "    if compare(color , [0,0,0]) == True:\n",
180 |         "      # delete the occurance\n",
181 |         "      del occurance_counter[x[0]]\n",
182 |         "      # remove the cluster \n",
183 |         "      hasBlack = True\n",
184 |         "      estimator_cluster = np.delete(estimator_cluster,x[0],0)\n",
185 |         "      break\n",
186 |         "      \n",
187 |         "   \n",
188 |         "  return (occurance_counter,estimator_cluster,hasBlack)\n",
189 |         "    \n",
190 |         "    "
191 |       ],
192 |       "execution_count": 0,
193 |       "outputs": []
194 |     },
195 |     {
196 |       "metadata": {
197 |         "id": "gOMlAU75mtix",
198 |         "colab_type": "text"
199 |       },
200 |       "cell_type": "markdown",
201 |       "source": [
202 |         "## Section Two.3 : Extract Colour Information\n",
203 |         "\n",
204 |         "The ***getColorInfomation*** function does all the heavy lifiting to make sense of prediction that came from the clustering.\n",
205 |         "\n",
206 |         "Taking the prediction labels (***estimator_labels***) and the cluster centroids(***estimator_cluster***) as the input and returns an array of dictionaries of the extracted colours.\n",
207 |         "\n",
208 |         "The function also takes an optional parameter (***hasThresholding***) to indicate whether a mask was used. This passed from the ***extractDominantColor*** function\n"
209 |       ]
210 |     },
211 |     {
212 |       "metadata": {
213 |         "id": "ywktI_ISyoFj",
214 |         "colab_type": "code",
215 |         "colab": {}
216 |       },
217 |       "cell_type": "code",
218 |       "source": [
219 |         "def getColorInformation(estimator_labels, estimator_cluster,hasThresholding=False):\n",
220 |         "  \n",
221 |         "  # Variable to keep count of the occurance of each color predicted\n",
222 |         "  occurance_counter = None\n",
223 |         "  \n",
224 |         "  # Output list variable to return\n",
225 |         "  colorInformation = []\n",
226 |         "  \n",
227 |         "  \n",
228 |         "  #Check for Black\n",
229 |         "  hasBlack =False\n",
230 |         "  \n",
231 |         "  # If a mask has be applied, remove th black\n",
232 |         "  if hasThresholding == True:\n",
233 |         "    \n",
234 |         "    (occurance,cluster,black) = removeBlack(estimator_labels,estimator_cluster)\n",
235 |         "    occurance_counter =  occurance\n",
236 |         "    estimator_cluster = cluster\n",
237 |         "    hasBlack = black\n",
238 |         "    \n",
239 |         "  else:\n",
240 |         "    occurance_counter = Counter(estimator_labels)\n",
241 |         " \n",
242 |         "  # Get the total sum of all the predicted occurances\n",
243 |         "  totalOccurance = sum(occurance_counter.values()) \n",
244 |         "  \n",
245 |         " \n",
246 |         "  # Loop through all the predicted colors\n",
247 |         "  for x in occurance_counter.most_common(len(estimator_cluster)):\n",
248 |         "    \n",
249 |         "    index = (int(x[0]))\n",
250 |         "    \n",
251 |         "    # Quick fix for index out of bound when there is no threshold\n",
252 |         "    index =  (index-1) if ((hasThresholding & hasBlack)& (int(index) !=0)) else index\n",
253 |         "    \n",
254 |         "    # Get the color number into a list\n",
255 |         "    color = estimator_cluster[index].tolist()\n",
256 |         "    \n",
257 |         "    # Get the percentage of each color\n",
258 |         "    color_percentage= (x[1]/totalOccurance)\n",
259 |         "    \n",
260 |         "    #make the dictionay of the information\n",
261 |         "    colorInfo = {\"cluster_index\":index , \"color\": color , \"color_percentage\" : color_percentage }\n",
262 |         "    \n",
263 |         "    # Add the dictionary to the list\n",
264 |         "    colorInformation.append(colorInfo)\n",
265 |         "    \n",
266 |         "      \n",
267 |         "  return colorInformation "
268 |       ],
269 |       "execution_count": 0,
270 |       "outputs": []
271 |     },
272 |     {
273 |       "metadata": {
274 |         "id": "OyaNX8GHBsHN",
275 |         "colab_type": "text"
276 |       },
277 |       "cell_type": "markdown",
278 |       "source": [
279 |         "## Section Two.4 : Putting it All together\n",
280 |         "\n",
281 |         "The ***extractDominantColor*** is the function that call the above function to output the information.\n",
282 |         "\n",
283 |         "The function take an 8 bit 3 channel BGR image as the input , the number of colors to be extracted. This does all the super heavy lifting by sparkling some magic power of machine learning.\n",
284 |         "\n",
285 |         "\n",
286 |         "As mention in the article , An unsupervised clustering algorithm, ***KMeans Clustering*** is used to cluster the pixel data based on their RGB values.\n",
287 |         "\n",
288 |         "\n",
289 |         "The function also takes an optional parameter (***hasThresholding***) to indicate whether a thresholding mask was used. This passed to the ***getColorInformation*** function\n",
290 |         "\n",
291 |         "\n"
292 |       ]
293 |     },
294 |     {
295 |       "metadata": {
296 |         "id": "dwMgm-9k-pq6",
297 |         "colab_type": "code",
298 |         "colab": {}
299 |       },
300 |       "cell_type": "code",
301 |       "source": [
302 |         "\n",
303 |         "def extractDominantColor(image,number_of_colors=5,hasThresholding=False):\n",
304 |         "  \n",
305 |         "  # Quick Fix Increase cluster counter to neglect the black(Read Article) \n",
306 |         "  if hasThresholding == True:\n",
307 |         "    number_of_colors +=1\n",
308 |         "  \n",
309 |         "  # Taking Copy of the image\n",
310 |         "  img = image.copy()\n",
311 |         "  \n",
312 |         "  # Convert Image into RGB Colours Space\n",
313 |         "  img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)\n",
314 |         "  \n",
315 |         "  # Reshape Image\n",
316 |         "  img = img.reshape((img.shape[0]*img.shape[1]) , 3)\n",
317 |         "  \n",
318 |         "  #Initiate KMeans Object\n",
319 |         "  estimator = KMeans(n_clusters=number_of_colors, random_state=0)\n",
320 |         "  \n",
321 |         "  # Fit the image\n",
322 |         "  estimator.fit(img)\n",
323 |         "  \n",
324 |         "  # Get Colour Information\n",
325 |         "  colorInformation = getColorInformation(estimator.labels_,estimator.cluster_centers_,hasThresholding)\n",
326 |         "  return colorInformation\n",
327 |         "  \n",
328 |         "  "
329 |       ],
330 |       "execution_count": 0,
331 |       "outputs": []
332 |     },
333 |     {
334 |       "metadata": {
335 |         "id": "jMgmvD0GI1Uh",
336 |         "colab_type": "text"
337 |       },
338 |       "cell_type": "markdown",
339 |       "source": [
340 |         "## Section Two.4.1 : Putting it All together: Making a Visually Representation\n",
341 |         "\n",
342 |         "The ***plotColorBar*** function gives a visually representation of the extracted color information. \n",
343 |         "\n",
344 |         "Taking the color information (***colorInformation***) as input  and returns\n",
345 |         " ***500x100 8 bit 3 channel BGR colorspace image***"
346 |       ]
347 |     },
348 |     {
349 |       "metadata": {
350 |         "id": "fZtHRM0qn-SH",
351 |         "colab_type": "code",
352 |         "colab": {}
353 |       },
354 |       "cell_type": "code",
355 |       "source": [
356 |         "def plotColorBar(colorInformation):\n",
357 |         "  #Create a 500x100 black image\n",
358 |         "  color_bar = np.zeros((100,500,3), dtype=\"uint8\")\n",
359 |         "  \n",
360 |         "  top_x = 0\n",
361 |         "  for x in colorInformation:    \n",
362 |         "    bottom_x = top_x + (x[\"color_percentage\"] * color_bar.shape[1])\n",
363 |         "\n",
364 |         "    color = tuple(map(int,(x['color'])))\n",
365 |         "  \n",
366 |         "    cv2.rectangle(color_bar , (int(top_x),0) , (int(bottom_x),color_bar.shape[0]) ,color , -1)\n",
367 |         "    top_x = bottom_x\n",
368 |         "  return color_bar"
369 |       ],
370 |       "execution_count": 0,
371 |       "outputs": []
372 |     },
373 |     {
374 |       "metadata": {
375 |         "id": "f3xdAIqTOwuU",
376 |         "colab_type": "text"
377 |       },
378 |       "cell_type": "markdown",
379 |       "source": [
380 |         "## Section Two.4.2 : Putting it All together: Pretty Print\n",
381 |         "\n",
382 |         "The function makes print out the color information in a readable manner "
383 |       ]
384 |     },
385 |     {
386 |       "metadata": {
387 |         "id": "YV3vAwHG-B8l",
388 |         "colab_type": "code",
389 |         "colab": {}
390 |       },
391 |       "cell_type": "code",
392 |       "source": [
393 |         "def prety_print_data(color_info):\n",
394 |         "  for x in color_info:\n",
395 |         "    print(pprint.pformat(x))\n",
396 |         "    print()"
397 |       ],
398 |       "execution_count": 0,
399 |       "outputs": []
400 |     },
401 |     {
402 |       "metadata": {
403 |         "id": "-uHKuSb2PM7V",
404 |         "colab_type": "text"
405 |       },
406 |       "cell_type": "markdown",
407 |       "source": [
408 |         "## Section Three: Baking the Pie\n",
409 |         "The below lines of code, is the implementation of the above defined function."
410 |       ]
411 |     },
412 |     {
413 |       "metadata": {
414 |         "id": "U7d_APvx9n4Z",
415 |         "colab_type": "code",
416 |         "colab": {
417 |           "base_uri": "https://localhost:8080/",
418 |           "height": 1175
419 |         },
420 |         "outputId": "2648c476-c9d7-4dbc-d104-d02baabc5878"
421 |       },
422 |       "cell_type": "code",
423 |       "source": [
424 |         "\n",
425 |         "\n",
426 |         "'''\n",
427 |         "Skin Image Primary : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg\n",
428 |         "Skin Image One     : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/skin.jpg\n",
429 |         "Skin Image Two     : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/skin_2.jpg\n",
430 |         "Skin Image Three   : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/Human-Hands-Front-Back-Image-From-Wikipedia.jpg\n",
431 |         "\n",
432 |         "'''\n",
433 |         "\n",
434 |         "\n",
435 |         "# Get Image from URL. If you want to upload an image file and use that comment the below code and replace with  image=cv2.imread(\"FILE_NAME\")\n",
436 |         "image =  imutils.url_to_image(\"https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg\")\n",
437 |         "\n",
438 |         "# Resize image to a width of 250\n",
439 |         "image = imutils.resize(image,width=250)\n",
440 |         "\n",
441 |         "#Show image\n",
442 |         "plt.imshow(cv2.cvtColor(image,cv2.COLOR_BGR2RGB))\n",
443 |         "plt.show()\n",
444 |         "\n",
445 |         "\n",
446 |         "# Apply Skin Mask\n",
447 |         "skin = extractSkin(image)\n",
448 |         "\n",
449 |         "plt.imshow(cv2.cvtColor(skin,cv2.COLOR_BGR2RGB))\n",
450 |         "plt.show()\n",
451 |         "\n",
452 |         "\n",
453 |         "\n",
454 |         "# Find the dominant color. Default is 1 , pass the parameter 'number_of_colors=N' where N is the specified number of colors \n",
455 |         "dominantColors = extractDominantColor(skin,hasThresholding=True)\n",
456 |         "\n",
457 |         "\n",
458 |         "\n",
459 |         "\n",
460 |         "#Show in the dominant color information\n",
461 |         "print(\"Color Information\")\n",
462 |         "prety_print_data(dominantColors)\n",
463 |         "\n",
464 |         "\n",
465 |         "#Show in the dominant color as bar\n",
466 |         "print(\"Color Bar\")\n",
467 |         "colour_bar = plotColorBar(dominantColors)\n",
468 |         "plt.axis(\"off\")\n",
469 |         "plt.imshow(colour_bar)\n",
470 |         "plt.show()\n",
471 |         "\n",
472 |         "\n"
473 |       ],
474 |       "execution_count": 9,
475 |       "outputs": [
476 |         {
477 |           "output_type": "display_data",
478 |           "data": {
479 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFKCAYAAADMuCxnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsvWuQJddVLvjtnZnnWVVd/ZZbb8u2\n/MAy9jVxsZDAAi4Ox+WO7XlgXwWPHwSXCTwmCBwhEwoDNkQQYAcOwBAGwsg/ZsaB5orAEDEE0pi5\n5kKMsPEDY8CyLNlSSy2p1a2ud51zMvdjfqy1dubJqpJb1dXdp6vX1z9O1zmZ5+RjZ+b+1vrWt0yM\nMUKhUCgUCsVlhb3cG6BQKBQKhUIfyAqFQqFQzAT0gaxQKBQKxQxAH8gKhUKhUMwA9IGsUCgUCsUM\nQB/ICoVCoVDMAPK9/sLf+I3fwFe/+lUYY3Dvvffitttu2+ufUCgUCoVi32FPH8hf+MIX8OSTT+L+\n++/H448/jnvvvRf333//Xv6EQqFQKBT7Ensasn744Yfxwz/8wwCAW265BSsrK1hfX9/Ln1AoFAqF\nYl9iTx/IZ8+excGDB9Pfhw4dwpkzZ/byJxQKhUKh2Je4qKIudeVUKBQKheL8sKcP5GPHjuHs2bPp\n7+effx5Hjx7dy59QKBQKhWJfYk8fyN/3fd+HBx98EADwr//6rzh27Bjm5ub28icUCoVCodiX2FOV\n9Zve9Ca87nWvw3ve8x4YY/Crv/qre/n1CoVCoVDsWxhtv6hQKBQKxeWHOnUpFAqFQjED0AeyQqFQ\nKBQzgD23zrwaEUIAABhjpl4VCsX+QzvLp9e7Yq+gDFmhUCgUihmAMuQLgPceALCysgIAsJbmNwsL\nC+n/CoVif0CY8draGoD6+l9YWAAAZFl2eTZMsW+gTw2FQqFQKGYAypB3AckZTyYTAIBzDgDQ7XYv\n2zYpFIq9RzNfXJbl1Ot2yygUFwJ9IL9ExBjTgzjP6fANh0MAQL/fBwANVysU+wze+/Tg7XQ6AOrr\nXUPVir2CPjkUCoVCoZgBqFPXS8RoNEJRFADqkLXMmCWU3el0tBRiRvFShrssq+VsVy8kHRVC2DIe\n5HqXVJWOj9nD+Vzvct5moZxNGbJCoVAoFDMAzSF/B8isSWbKxpgtOWRZRmbMitlECAHj8RjA1tmv\nMSblApusCKgZUJZlyoKuMshYsNZuuc5lvIjIS0Wds4Wm3kfOmZxD0fl0Oh1UVTX1mdzXJRJ6KaEM\nWaFQKBSKGYDmkHeAHBZhVM2ZssyExRhAZlvCnmSGpZgNyHmaTCbp3Mh7MnPO83xLrrgdHSmK4rLM\nmhUXH3KuZVy0X4uiSOOgPYZEQ2KM0QjKDEDOZVPvI+dOIOfQObfjMnIuL2XkQx/IDcihaIY62jfg\n8XicQlVyIY5GIwDAYDAAoBfmrKD5IAboImyXrsjfsows1/ysmYqQcy/L6Hm+ctG89clYaU6+gDoc\n7b1PZU5yvbfLHHUsXF7IdSrnLMuydI7lfMrfEqb23m85fzIW5Fo3xqT1L/Y51pC1QqFQKBQzAGXI\nDTTD1MKA2q483W53i+hHZk8y6+r3+zpbvoyQ8yLphibjbYYXm3ixy0DOt8ycAWXK+wnOualzC9TX\nsoQrm+HMdtmjGgJdXrSjXE2mK+dmp+vdGDMVGZX1gGkhr5zbi82UdQQpFAqFQjEDUPVRA81Zcpsx\nCSOy1m6ZTUtOSQU/lxfb5YwBTOWNd5rZvtiMt/k9wr7lt9o5acWVAzl3ZVmi1+sBqK9lYURy3YcQ\n0rlvX+8aHbl8iDFuiYRJVHMwGJzX9d4Wc8q5bzLmNgu/WGYwypAVCoVCoZgBXJU55PYuy6ynyXra\nxeHN2XR79izrNeXxOmu+tPDeb2HGgmaet6muBDClwmyvt13eqZ1PbpoJFEWBGKOe+xlFu4ytqaCW\n67ytCZBlrLXpehdGJk1lNHd86SHncnNzM91325oeoL5O5boXyPltRs12yjM759L67XPd7Xb39HrX\nkaRQKBQKxQzgqmbIwqhk5ivvZ1mGzc1NAHVuUGZGzrk0I5NZ007KXcXFR1NRLeymbXnZVM+3mVAz\nTyRs6MVUs22W1TQYGQwGcM6pMcyMoX3O2gYfWZale4EsK2NH2HC/30/Xu6yn9ceXD5LHb3oLtDU8\nTfV8+x7d1JuIfmCnNprNyJjcI5qRlL3Uj1yVdw65yHbyLs7zPB1kuQibJ7t90SouPdqTquZDsC24\naHoNty+6ppCjvbxcqNstv1N423uvD+QZQzu90D7PzbBlO8Qp94HJZJImaopLj7ZzYjNd2L7em/fz\nncRXzcmYkK+msVMT2xmDtEPgewUNWSsUCoVCMQPY91P5dkR+OxMAgcyCxuPxFh/TZpH4TqENxaWD\nRDMkbOi9T4x2pxDidxLftCMmguYYan+3fKd2+potNG1Phc3IeRXW20xbyGfCkiQk2k5/vPQN4RcT\nEGNrXLaHaazfNKbVm3fLwlcn2tdZr9fb0Xv+fARXxpgdI1rbZXMvdppCGbJCoVAoFDOAXTPkj3zk\nI/jSl74E5xx+9md/Fq9//etxzz33wHuPo0eP4qMf/ehMmSU0cw9t1ttkWQDNoNuWmc3+mYrLj/as\nuPnehX5nG01D+vasud1fVaMnlxcv1jygzW6a+pD2eJK/d7JePF94Q/cUEzIYT/eUcuVJ+v31Vfpb\nGtn05tBbvAYAkM0fpO2xFLXTUUU4nxKzl3qudlq+KQJsawzkObDXkbFdPZD/4R/+Ad/85jdx//33\nY2lpCe9617vwlre8BXfffTfe/va342Mf+xgeeOAB3H333Xu6sQqFQqFQ7FfsquxJTBgGgwG897j9\n9tsxHA7x13/91+h0OvjKV76C++67Dx//+Mcvxja/JDTl8cD0bEfQbsU3Ho+3vKc2ebOF7QzlZfbc\nNneQMdDMN73Yd0o0pd1EoNPpTFkpNl/lczUGubyQyJacF2NMYjqiMWi3WKyqaotCV6JoFxrxCJEV\nv5tLWH3sn2mbVp+j7xb2LMYUNoPLKYftBocAAAs33AoAyIdHAVy9JiTta7NZ4tgub2xer+dzvHa6\n3ptK7HYDChlfL2bHuxvsiiFnWZbEDw888AC+//u/H3//93+fHmKHDx/GmTNn9mwjLwTtUoXz8ZsW\nBx7F7EIugu1Kk9o433KVdphSsJ3oY6cLXR/GlxfbpZTagqz2Q7b5+V6npKzhmuXhURx6ww/t6Xdf\nTdjp2txumfO5JzTRXl6ebZcDF6Sy/uxnP4sHHngA9913H37kR34kvX85vUZk5tJWVspFmGXZjo0B\nmk3qZfn2Bao33NlAO1fYtMVs5/2adYkvNiGT5dqz4PYMvLmsoJmHVFxaxBi31A83I1vy2U6Wt82o\nWbtu9aWd04hQ/xcAENaJmJx99G8x2Fij7XAcZbHM4iNtR24NjOQmC3pITIbHAQCLt34ffaHJYOTL\nr6Lx1o5INZu8tB/STbOg9mdtrUCMcUtUpR0daTaVabfc3OtrftcP5L/7u7/DH/7hH+KTn/wk5ufn\nMRgMMB6P0ev1cPr0aRw7dmwvt/O8IQdMbrztB7Jzbstn7Q4v1lp135pxtB+awM4z42YXmLYDm6Ap\n4pNx0GZW1toU3rpU/VEVO6PpstYuZWxOvL5TOVuM8QIfxPJ9AcHwxD/QfWjlia/R9oxHiE4eIvxA\n9fxABk8YjYHNmFHnvG0Vk4RIy0aTwaTH/tUj9WpPrl9MRNm8NndyXGyK/trMun3umz75F5s97yoh\nsba2ho985CP4oz/6IywuLgIAbr/9djz44IMAgIceegh33nnn3m2lQqFQKBT7HLtiyH/1V3+FpaUl\n/MIv/EJ67zd/8zfxwQ9+EPfffz9OnDiBd77znXu2kS8F7Z6YbU9aYKtkvV0GtdcdPBSzgU6nk5iT\njI9maLMt6GnniZssWpnx5UfT9rQp5AGmuzTJuZLrvR2y7vV6e1Ku1hwJ46VT9J+Vs7QdzqffZ51X\nClVnErq2FobZs5FtlW/VYQZga2Ss6WXdvhabHdyanb2A7a0329d7M6Wxk63mXmNXD+R3v/vdePe7\n373l/U996lMXvEEKhUKhUFyNuKKtM5uJfYBmODLLkXyAzJxlFlSWZWLLbab8UtV5isuHJjsSiBbg\nfAQX7XM/Go12VNc3mfFOZvWCiJikPbKE2B6GRobIMjuCuXrygBcKKSlLTLNhyiLnvNkwAiB2I9d7\nu0vbTsxo19sHgyDlOWefBQCYQGPSuwDLpzx6zoPKAAki0oqAF1Ea34tY3BXFUpPW3JPtvZLQbuoi\nY6Asyy3nsXltNvseN9EshZT12uy72ef+UkXCrr4zq1AoFArFDOKKZMjtIvF2uQpQs902U+50OmmG\nLDMhzQdeeWjPmJtmLm39wHYz5nbZVLM8ot2eT3A+pU3EhrnhRHoPU78ZTL2MzojPH9LaUpSzwmCa\nDQLkem/2Lt6preaeNwqIBhb0u+XSC/ReoN80MQec1ELx/QqinOYcaLTp/1LSFLui6rWNt6/e+1Q7\nJ5xl2ZbrXbDddd9up9tsuyrPkbai+lI+F/R+oFAoFArFDOCKY8ghhMSMheFux242NjYAbLW+nEwm\nW0w/lBlfuRCWMxgMEnNqsmZgmilv13y8+T3AtB0n8NIiKBEAmLmF0TIAoFo7R+tzLWlx4Chi/6Bs\nwPns5lWLEEJiwuPxGPPz81vYMIB07rdTW7c1IherAYhBRLlMuWM7IROQwLXGcB7gOmRhyFJPbJm1\nWQvEKHluVhEXO+shdjJgijHuW4vNdk64LMv0Xlsv1Fxe0LS8lM+FGQt7vhzMWHDFPJBl8DUvsHaC\nv2n60RZ5yEXYTOIr9hfkhtsseQHqB/N38rJul0Kd14SteU80FIacLD8BANj89lcBAIXj7+WSllXT\nQe/EKwEACze8BgDgLffohcfVZPjwneCcSzfatvimSg8ymwR58mAWXEqhZogRo+eok5Nlsw/Dz2Bf\nlckIJAZWd/GDOYxJYOSCQ849k8O8+G2LqKv+DetoPE/OkSf2ZLROy7BQsNMbIh/ShK8YHKAVc55U\nIrYEYlcmmg9muW63867fKS3RFHDJ+udrsXsxoU8mhUKhUChmAFccQ/bebyn0lxBFp9PZMkNuz4yU\nHe9ftC0vZaYs42U8Hm/xpW6WT7TDYS/GjEWwJczFxgrGFBifOYnNx/8JANCviMlUgS8zZjCFnaB6\n6t8AAEsbxG4O3PpmWmabRhZXI5rXezvE3Ga94/F4yocemO72dKkQfIW4RCHryOmJzLFvvpsgVMyM\nHTH7SmwxK7bu9SUMpByPlgGLusS/ulp6GivfpPGVjSkt18nF8IRTJbnFRk5sL/aIKR+45d/Tb+Q9\nRC61u5IZssAYs0XM2RR+tS0z277mIYQd/a4vB/TppFAoFArFDOCKmY43xTdtq7SmLVq7J24736zY\n/5BxIeKMZk/sdj6ymWN6KeItEeREZi6jpWcwOHwjNh/7ErolMePSS/MCySdymU5ukbNdYjjzLQDA\nOs/kF17xJtV5Yfp6b9sdiiZA/i7Lcovpf7sE5mIilWGeexrG01gL0kCCWXB0E0RpgMHjIJacO15b\nBQBk0cFZYmuBc8mmQ39PVp4BAKz/82fRk45HojVIHjP0tw05BqIf8yQuGz9P46x/4jWo4zr7Y6C1\nO3U1c8rbdf8CMNUje5ZEvcqQFQqFQqGYAVxxtLFpnbddPrCdX9bSJkXTMrGZmwTq8dE0kGhjWkgt\n5g78ygrXtW99HYPDNyKvJpjwZ5FVtJm0+WPWEl1EzJkBZTSDH51+lLbn2A3oLVAPXMMmEftBFftS\n0SzpaZ+rNlMeDocXzQ5zezBDjfwbzHg3zzyJTBTUnDtGxSYgk60M2XAu2aR+3VWKooRAZztjtfby\nY18GAHS8wzjQPS5ntTWYTWc5R3tQIlrOPRes2l57nhbFq/ftONquv33TPAaYZsbN92cFV8wDuRmW\nkBrjtminGaKY1QOuuPSQm3On09niaSt48dKm0FiQS294wnfuW/8MACi47tTFLHkVw/NNIIjnMn9H\nCIDjm6fcRAyN5fUnH0Hvuw7TYnzDj1ZuzlcPmgI7uamORiPMz89v6UXd9DO+WDXGTYjDVuCxgBGF\nnMPyWWTixrVJEzU3Yd/qaOD5ARy5DC6U3JOX65MdDHKRC/Kze/T8U/TnORKLeWORi991au8svaCl\na1QBcD9lmUD60PjtRq/o/YT2Ndztdqd6XjeXmdXngoasFQqFQqGYAVwxDFlgjEkmAMKQm11+vlM3\nHsXVi92OCelbG02WwtBrT38dAODXxLOYWVOo+95KT1sJVTaNlYQZS7jRssjLLz0FPybGhf4hAFcX\nM27DGJPKUuQ6l9LGptnPpWDGgig+5Bw5WX7qGwCA3JcwEg0RTZeI0/IccULbWDFrDsyYhdkWJiAl\nJthRZOPZb9J3R/bft31kgYWJEvpOqi4OTwfAGv7MTm9P2BgjP8Bdza6CW+SsM+I2lCErFAqFQjED\nuOIYMlDPdiRPfCnzR4qrEJYuEx+BcontCpm5mIrzgSzCCc7BJqYy3VcVUgpjG+IwHrKZod8wvsLa\nGcoXzt3Afte1/cie7taVBmHKc3NzU+9fevbDJhOsG5i88DQAoAglKsnjssAq5zKmajxBDNOCQOmd\nLGc3eofA0ZhcjEE4P11It6IQETgHLZ7pMYnEOE9qIyKX3tkB/b7cGf3mJrIFirxofd3s4eq+whUK\nhUKhmBFckQxZoDaYikuBlO4tV7H0BFleFiWzEm4YYIT9eg8nZXkZXV7RTeeQg/EAK6dNnO7PbODg\nV8/y/wWi8r46x/us2d9Knnj99OMAgDyQojqvKsTUQIKtGaXRSVagZPbrOeebAiauVlYHLpPyJXet\nytlKU859NYJnC86kbSilOxmbX9gIy9HCWEokhrerXL8aUsdXLK7OK1yhUCgUihnDFc2QFYpLgch1\nn+ce/xfk68RerbS5bSmpEWs7TTGuyJmtOK5PjjHC7MBTonfwGysA6v7JmuubLURWPI+efwIA0GPv\nA7iYwiDBiFkR1xpHg3xA6uZyQirxyKYfSWNQRZiK64aZ9UbOm7NEAdF72InUFHNzCt4ecN46OsB4\nZtascTAdVniPRujyeIbdviZfcfmgD2SFIkFMBMzU32tPkYtWOHMSOd9wXai7xQC1UYgPIf1fSpvE\nHSjL6aboo0fgsGdmpx+2BhGh3OT/i1/2/unOsx/g2IM6jpboDXkg+6ZpCL0VuXwpRqCUTkPip8xl\nXI7D2oW3CGP6roI9sIO4f4hQ0PkkJIxjfrAbXkYesL6eREZ5IIvgK/fpYZ9xOFwG1pTDdTLDCZha\niF+DsY3w6v7yxr6c0JC1QqFQKBQzAGXICgUjJIZMf0/OUTnL6BSZgBQxpD63IqhJnXuEzcaQDEGQ\nPK2ZKfPvdHp9VGJlKKUwRsRdJtlk19jyhuIyouQe1hmHjL2YgZjaYlWyDCEnNhw3NgBm1pMJW/5y\ndCRFrK1Bx7JpSEn2wCFSiVIuERlXoeSOUhmHqo2EoCeS4rAIju1WvWwTl0q5NQROpYT+PC3OY3G0\nSn7Xg4PX4sxX/pp/lzeuS8tijkrxBseuRzFYpPXTXos9pfK83UKPnEKhUCgUMwBlyAoFQ3LHfp2Y\nwspjXwQAFKHuqZpKoJoNJ5rfYepmECZjkQ2kTIVWLssKRZeZk5EvlGWBWHT4/2LRyPNmTdHNBKrR\nMgAg8yLckgYOHhmfu8AWq2K1Gn2JwIzaMlOu1lj01yXh1oZz6PDYkVI5I92KhH1G1+gIxTlpHkKi\nVTA+JCtXw0JCK2YiWYnVZ6h0b6H/BgDA+BkyualOPwYAGHzff0Zvg/LjGbPdrCLG7ifnAADl2rOo\njt0KAOgfv5F+Ix2hCB2su4MyZIVCoVAoZgAXxJDH4zF+9Ed/FD/3cz+Ht7zlLbjnnnvgvcfRo0fx\n0Y9+dMdWdwrF5cd0r2EAAOf2zn7zHwEARUXWiDGwOrqhoBYDCFFZBzGE4H/N3zDCmJltGGtQMQMS\n21dhNzbLkA8pX5fsQET1raTjsiFGIAZimdU62afalLKVlpyobTFlXEjP4+CS4jlycwy3QeOtBN0n\nOwcXMWZGO5TcbVLxs0VrVQJRGLJ8JiyaBkgRY21Qwgw7bUfPoZicBgBsfu2/0We8PYUkswHECf2e\nY0vXwP27M0PXQm4niM8T097kcTk8diNvh2K3uCCG/IlPfAIHDhwAAPze7/0e7r77bnz605/GjTfe\niAceeGBPNlChUCgUiqsBu2bIjz/+OB577DG89a1vBQB8/vOfx4c//GEAwF133YX77rsPd999955s\npEKx15C8ros047fRYemJfwYAmJJUtHW+mF5DNEJOklWmFRUrsxXrQ7JGTKptpgxiZ5g1rB9LqUGV\npvEmohgeoeWldVxqQK/NUy41kuLdVTj96H8HAJTPfBkA0PPMHsF6gLxbS6Z5oGSsI/ABALddxJhV\n92yBGXJ6HY/WMHjZNbQ6W12OC8ov52vUcKSqKhTMoiWXnKxEmb3GmCV7TdE6yBg2MLBcD1+usEVr\nELV3w6y1EsvPMe8P6xoy0UdkiBIZOEc5aDe/AAAo+ofVy2aX2PUD+bd+67fwy7/8y/jMZz4DgHqV\nSoj68OHDOHPmzN5soUJxEWB46BfJ76DAkVd/L//xvduucz647n/4Xy9swxQzheTbUhR42et+mP4v\nr/sYi2//2cu9CVcldvVA/sxnPoPv/u7vxvXXX7/t5zFq3aRitlGLpOk/q898A+uniCFnYvYfmLX6\nuqF8KMWViZXXntWrEbj+Xe/Dyb/4eLJNTDaKzKItt3EsijzlrgMrqPOcPsu6XQxeewcAYHj0OgDN\nJuuqwbzUECXz0re/gupZypnmY9IWrJ39NgCg2zkMAMiyHry0TZReI+y45dfXEJbIEnV05gUAwHiT\ncsgjXrY7P4fsMNX2Ll5/AgCw8KYfBQCsf/Vv6DfWl2DZ4SvlpHmc5UyI8uE80KHWtHaRUopZn+uZ\nDyzCe8oZV+dIMW08M1227Tzyn/4Lzv3F79P6GX83R3dst8ffN4DpsP6BdRDm+HcBAPpHbtBE8i6x\nqwfy5z73OTz11FP43Oc+h+eeew6dTgeDwQDj8Ri9Xg+nT5/GsWPH9npbFYo9gwivRssUClw99U/o\ncYmKjxQmdPy3Ad1kQwiNUJyIuKTnMa/rSkQ7PSFND1T+PnjAWmmEXEwtM3ERhw4c4d9V68zLBzqH\nG89RKdDk9CMoxAqVH3wLczRhql54CgDgAWTZgP7PxhyBQ7yVd+nhXnF6Y5Mf1o4naj2bISvoXI8z\neshdd92r6e8n/wUAUEzW4SYURpb+70GEW5I+8R6ZGM3IZFDmiM7Ds2BLfEykw5Q8kGm5aZ/tXELV\nHKa2RQeBS/cMC75sR0LnWva0W+zqgfw7v/M76f8f//jHce211+IrX/kKHnzwQbzjHe/AQw89hDvv\nvHPPNlKhUCgUiv2OPTMGed/73ocPfOADuP/++3HixAm8853v3KuvViguCCGZKsTERN1oFQCw/CiV\nOHWiR+VYnMNsF6lhBP/pHcAh6sSaxfSDw9QmhMQ8Uh9fIyVWUq5iUPBnnstKTEbflx84gqzTly8A\nUIu7FBcRnMOQFEK1SRqYtac4jRFjahRhpSMTh2yzIzfR3+tnMV5+hr7OE3vdjDSmbAxJaFXK+vxZ\n0aPzbbsZbI++89CrvpffI8bdPUBhcXf2ScQOjRlfibCQvpcTLDCIiFJix6JFGZ8xTIARiRZF0Jga\nUUj3KNTWnRLIEeFXlAYpvgfL6yNSGBuWrx8oP94tLviB/L73vS/9/1Of+tSFfp1CoVAoFFcl1DpT\nse8hpUnRZvDMAs594/8DAHQcMWUfTbIplJxaMncQkwZfJRGXlJOkHFvcyglkfUGWWEtI+eEsl5Im\n+qx3+LqGiEt5xqVCajLIDRte+PbX6A1phWkCLGsA6paIfO7YISTvHUJ2hBhuySwUbI+5cvYcwhqJ\nsTY3ubkDJ3YH88SCO4MOskMk5jryqjcCADxzzWSjWnTh2XrTCiWOwry5hM7lsDxOM9FF8BjOjIGT\ndpExJZH5ZZyORxzTb4Dzw577KWexm/Y9csMKuT4KznsrP949VLapUCgUCsUMQBmyYt9Dmj04eCw9\nRaYObkymCGCFa5xMAG7i7ispaWox5OBgmB1JqYm0T6zNMmOyMEwQ0xBhvjamRhMZf18wRHe6B08o\nM74MkPM3WiZbSb9CrTe70jTTefhU4sbREbGjTErmCpMohhx0PjNDTHNuvgvH5zozc7R8n7UCnBO2\nc4s4eOv30Po9MtnIxGCkPwQArFcOc0Naf1xSGZUEZ1KjElclRb+R3HhqBVrbv6bqPFlWSrYAxJIi\nSanLKFcD5FZaLEZEPjZJKpEou2K30AeyYt/Di2PWxjImz1OJSg7xoOa7SdaFk9Bd5FDcmDvcSFkI\nYqrLhLgjsepFeh1Ha1I/ZLlRWXGXkAcz6nKpjMVhpaUb8GDh0AXvr+LFIZOpVFYWDSquzV16nDt8\n8VNO/KqjC8jE07yS8jU56xwyLkeIEo7e5PrjTe5rXE7g2SsdnLoQMVWHa9Bx4BgOv5xqecW/PJOx\nNDjKfzuUE1q/6FGNseMSK6knRuUBniykunnu9mSqDIZ7K3vpLCUTyNCYCIobGP9pszn+H49358Ct\nmxGN1CjzBCNGqFXX7qAha4VCoVAoZgDKkBX7HobDdWvPPYaMRV0SFg5JnBWTmEvCyWAfYjDz8K6E\nE1bE8b6C+9+KSYN3VTL9SOIs6fbEXxdQs2bPQpji4MsAYLuAt2KPIcy4gnRpmuCFf/s7AEDBqYyc\nGWXq4uV83fe4HPN6xD49Rzn8ZAzPoV7Hoqg4pr8rdtcCgJx9y23BTP0AhaMP3fbvgeFhXkp80Gkb\n54/R+FgyGXpSi8TGIsWARGGlMP8wqhkyh5hz02DxzOhjxREbcf6S1AyAin83LzilE8R/m8PuvUa3\nqZbnu9UBvGsoQ1YoFAqFYgagDFmx7xFYpDV64Ql0Dc3wXZDOOMx6nK/FORD2ygxK0sxZ43IRcZcw\nKc6/2RgQ2VyiXb5khHEbAwhr83MgAAAgAElEQVSjzshUYXjkWv5IOzpdbMQo54XO4fITn4ddepI+\nbPUxFoEfQgTY41wYcuAexZ7HQjXeQLlOOeM4Eg9rFke5gM6QcqycwkW3T9GVsHgzAODILW9GVhdg\n0Xo8dkbrS7wdJQybyYioasIRHNsnphw2x7VVJlteBhZuVa5K3cNyKXsa0f5UqEVdHe7cJEI2G4R9\nSwmgT73HGoabvOlKkXcLZcgKhUKhUMwAlCEr9g3qLmPMZPiv1VOPAKDcWupfXPnmoshhEXLJKdJ7\nNhCDKbp0mVRukvJkYIP+jihVJX+GWOfixE6zbYGZFXUpVo9YzeD4DbxsgPY9vrgQY4/Np6mP7+bp\nb6PLA8E0TGAAIDp5DSnS4tMrl0ExiwzrY2BEn1XMkB0bx2SdHODGER0eZ24wDwA48ro30/d0hohG\nIi98a66WAQCn//GvAADFxgpij/LMQVi00OEJ55RtD15U/MKUvajGq1TSJL2eUzmXND8BELmrE5w0\nP2GbTqZwWbQwhrs9CeWXiAGKhm0s/65yv/OCPpAV+xAchl5/HgAw4gbqWfCoKhFo8Y1WomvRpg5Q\n9VN3OvSW50V9g+UblufOO0bqT4ND5JCi93IT4xC2dMXJMlRcs7lw3S30fbleipcK41UK/y4//XkA\nQG9S1dFWEXHxhM14mpT5agO+ZA9ofmZXJT/IOG1RlSVKfiiK+KkW9tnUKUkmePEICbUOv/IN9Ldp\nPGR5g5595Au0TkmOcjF4bKxR+0Z78CD9Po+zglMik2odAw6PTyoSbHVzGm8hWkQvbUH5lUViWVGP\nwcD7JKJFEZc5ec5mOXJxA9uk47nx6Jdouw4cRf/wNbRgdy7tG6AeXt8JOm1RKBQKhWIGoNNyxT4C\nl4oww117+hsAgII71wQXEQL79UoojZ26qtLBSxlHErBM/w0TUHDJikT5cm4E7zfJNcnGAMNMw3FP\nWRFzSZjaG8AOydRh4cQraD0Vwlw0CNuUEPO5b/53AEB3wmFqX9XpDmbIhmlwEu1VFeBYCBimvc4d\nf693LrFmSVvYWHOeFCmZ53PPrlzokoCqCDExULd2CgAwOcndyHicmrmjcE99m35/fY1WX6T1JeDs\nqjE2lsjopDhA+1VylyZrF2GTMQnva0GhZy8uKAAsR3wCe3sjF79t9qvu9DAR8xQpv1qhDllx43ls\ncJ/x7gnq59xZPEbrGeWALwY9OgqFQqFQzACUISv2DYTkjHl2Pl6l147M9p1LJS/IKTfmOFdogdTv\nVlhB3RdZ2HSWcmkdFugYZlB5n1iKiQY5M40onXLYI7vgPHGwAxx4OeUNbWfAW68M+UIgPa+TDkBE\nWt6i4vfWnvoKvbdEftWROzlVEcjBuVIeRCGIJSqxvzKPsJHLlDhPHDhMUsnfpUtCKenOJOVwmSuQ\ndej/5fAIAODEza8BUNtkxhhTP+ZzjzwMABjwePM5MdO40MHwVbcCADaXKJe8dpZe3Trtjw8BRYc2\nZI7FaQtDjuRk48T+wazZiWXmuC57smOKJLmC8tT5ArH6wNEfHz0yEUiKEC2nV2sLdCLlrv3zj9M+\n8zkoFo/zLxiYlLhXXijQI6FQKBQKxQxAGbJi3yByOcvK048CALqOZulifwjYZGsZZC7KloIGgGc2\nYdPMndfiRSMCDBrmHgAyzimLE0Pe6yaz/oJzc5V0jxLLwyM3YXj8Zkz9iOKCIMzYSglOFFVwBFih\nvHGKNAWB/56s0is2RvAgRpgNiRHGPiuYQWplayZJZS0KZMe2mIGjLKH0qQGFWKxaLr7rZg62R4rj\nwS3/jn5r/pq09QAAExFHtE2bp5/g9bjsiAehNSaporuHuSzvADUkmWywKcnGCIYbo0ivY8emJt1e\nB8g5ClCJRoKNQoTeo25K0e3x/nN0xwu79iE1oJAKgdxyBCEWyHL+bkN57nJMGos8Ui5ZO5ptD2XI\nCoVCoVDMAJQhK/YFYoxYefrfAADZ+jMAgMBsKYmkjQGiKGV9Wg8gRbSRXFYM9fIAgrCdPIPJhDZL\n3o+ZMqtg0ekgExMGbhzhmeWUzL5OvPaNqQ+z4sIhzRTpD4qKlMw0N1bPojpHvY3jGqmAbWcRAJAd\nY+XvxhLCCuVhV17gGt/wHABgcIwtTYsCcruMzDbBNenJMGTi4SacT+VBlw1ovOXDPso5YrInbn4d\nvcfbL1aWARGrz3yLfjdyDtcu8n5Nj8mp/We1d6cQJXYXGNLvdzrcJIJz0JgEgMdjHFPOWXLY8LXK\n2rH+oZij3HMpqnFmyLb0yQhEavAlkhRiB5GvPWNp/aLob9luxVboXUFxRSK5DHGxR1g5B3eaDEBS\nA3leRsLTERFGXI3koZmMqk26IQW++UjZUy7lKoiQ279tORGJ2MsWfZiMb8rS3b0gwdc1r7udlunN\na2hqV4gADCKfXzmvNlQYnaNJ2MozX6VF1zhkazwsl5/1uLyoXD9H63UohBy6czCLdCvss9iv4gf0\nmUcp/TF37fHkIR1KOq9+wq/szuUqh4pLoDJ+oA379EDtDiOqg2QC0z1EwiaZAEqKw0Vg/SSJubr8\nIIORhx7fqkOozaP5OFguw4psApKFElkmwi1++HJY2Y0rWAk1s+lH5aR7Vd3tqXP8evo5ecpyeBtj\n6YLlYFKhFZd/sdAxVvX2VqD96PFxkH2miYWGrdvQ+4JCoVAoFDMAZciKKxLS09byzP3sM19A5lnE\nlbQp4iXNpgi5SWYOEp32lZQ2VTUzlt9IIWspjTIpPChMW5YRlmPyIgmLKkPspH89lbf0Dh2tN0ui\ng2pbfd4QSZ3wMr9G1qjLJ7+OsEYh5gw0BhKb9ZMU6ZDz2u1T+HRzida3zsFkxJZNTozOHCQWO8ho\n2dWTz2HuZYd4O+jcVyV3UGKDkcm4TFGRHhvGdHo8FvoDHLrpu2l93jbxdzbMdKuVJ1Et0zZlOaU3\nrJdwMvc3jrE2sBH7V2bGng1woq9gujT2ig6nTXgMR28SoxZRVrdH2zrZnKRjbYfc7YmjC4G9290a\nCdsyG9LYjVwOFr34w4c0vENOxy/rSXmf4sWgDFmhUCgUihmAMmTFFQmZjS8/zWYPG2dr0w9GWwAT\nQkhMAUnUJSKcMpVEifm+rG3F+jLG9KYwbcviroz7G0eTpWmuy6mbz9Ebb+MN6vA3ekSxQnxpu31V\nIpl1uBK26GL15D8DACbPUn7XRocc09qA4BtNP9I55+/hEzQ8QCx0/YWz8CvETCFjgKMbWYcYXu/4\nYawvnQUAdHhcjTmfurHGhhoxYMA9jnPO3VrOSW/0j+Gmm15J7xkpCWJzGWas5x77EvqBfs9nXKbE\nvY4dly9555JVZ2TLSutYnMVM2TiPYjDg7aDvKzfYwMZ26vAQN0bxXK6U9YdyyFFJiR6bhbgN+g0z\nGfFxAmy3Hs+0PdKQwyd9WD5Hx9iIiFHLnV4UypAVCoVCoZgBKENWzDAkGVzBR2Yc/Nb62acAAP4s\nGe1bV6RetokFJzUuIYSYPhPzfJmxm04XRpSk0t9WOirz7N6aPKmyxfYv42UkZ5nHCG8pbzd3AzeO\n6HNeMrGDDLUFw9WNOt0veV7xnmTLSAB+TCVMZ7/5BZx4w3/A5Ol/pWUD5zwN4KPkY6cbjAQfkYuy\n1/L4kDIlpnHduQNgIorlZ4kpu1Vmm30ad1Xw8Js0Lk6fpTzq5mg8tQ/9ToZOwdaqXRoDpsPq7Wte\niyIxUFZV85rlJpVjrZ98FAckMctNKth1Fa7RlzlK7nrEzJhzx5ZZdL83RMYrOj5EnpXhVbaKnBXc\n0rDCi+iirK0zpalGLEVVLeNcctpA4ONashK7w1EFC4uKS6v685SLjnKdQfFi2PUD+S//8i/xyU9+\nEnme4+d//udx66234p577oH3HkePHsVHP/pRdDqd7/xFCoVCoVAodvdAXlpawh/8wR/gz/7sz7C5\nuYmPf/zjePDBB3H33Xfj7W9/Oz72sY/hgQcewN13373X26u4qiBMIofwkPEqMZjNU/8EAMjZnGBi\nctgg7EjW5zpVfvW+rPOIYmYgvxRM7WzArCZucKtGaVRQdGplrKisWzJpbwI8K1SPXvuqbZep90wh\nCuNUiyvnjM/P6PTjWDr1NQBAZ8JRiSD14vQdIbp0HpOKXtprRqCqOJ/MSmXDuVbHymHjI6KnczR/\nkMxCVixZPb5wmhpRPH/qebhJrbYH6vr0QY/YYLcokHPDhtS4ga0nj9z46qRRKNM+EtN95l+o9rjj\nN+CYWYYxaxt4v2wmdcgurWc5IiTWrAPO6XrUOWDDOWRvmCpPNhE5P+2E0bL+wfZrApXxRVTyMRIr\nTYkQVSHCMGvOu8S4xdIzZnkywzGFkrKXgl09kB9++GG85S1vwdzcHObm5vDrv/7r+MEf/EF8+MMf\nBgDcdddduO+++/SBrNgT2ABMVqhz0/oTXwIAZHyTDTyEbQxJrGNEYCXGIBJarFwy9AjiuMVhyzhx\n6QZbibtRnzvs8IPZIqt9rhlSUiVwtoPFm0nEZbrzvJ6so4/hnZC8LjyFg89+64sAAHfmOXRZxCRl\naRKyNSldEMmNghai9URVFGrP5VBJ/FYGBk8CKgfPHZs8L5vz+Dh8lErViryHlbMUOl86R8YiuXjK\n8LZnxsKIg1uXjTHYne3gDa9EsOImRtu2fuoRAMDGk+Sx3c8KOJ4wSrmT5cmgiNRi5RBK2tbI8ehc\nDD6ChIUD4jyNvdihMHlnjo7h+ORpoMfh7z6VcVX80M3n69KkcpMmLTk/ZJ30h04pnhxgxzJJD0jY\n3xQFwD7umWWvd70EzgsmxviS01l//Md/jG9961tYXl7G6uoq3ve+9+EXf/EX8fDDNNM7efIk7rnn\nHvzpn/7pnm+wQqFQKBT7EbvOIS8vL+P3f//38cwzz+Anf/In0Xyu7+IZr7jqEROl9a2w5ebKGWx8\n+x8BALmj7jGph2sS71TwiSFwH2IOswlDRigRuTuT4fCeYSaQeZ+WFyFPKpdhYUtEQMF+wWAGleXM\nKroZjv3Qz+G5L38Wx277AXpPSl+UHkwj9Qyuuxc7FiadffTzAACzQlaYMZhabGfGeNkP/hec/uwf\nA6i7c3nvaz9oESN5CfnaOgrC73lm3DJOovfwG9LJSZah1zL5VUeUbJFZcinQ+soygLqj0/yBeQz7\n7AHNJjC9N/4HAMDLv/8dyTbSexJjPfbg/wkAyEYk6spinhi+ZfYr14Dsa3QVSi5XSvvMntQ5i7Sy\nYR/dG6jEKsQubz93szp3FhvPULQp5y5RgUPVw0UK11/7rl/EyQd+mz5j9pzxiZLweKfTQa+Q0D0b\ngyyy7/bBw8BRst5cvJb6ftdN0rSw58Wwq6Nz+PBhvPGNb0Se57jhhhswHA4xHA4x5qbWp0+fxjEx\nblcoFAqFQvEdsSuGfMcdd+CXfumX8DM/8zNYWVnB5uYm7rjjDjz44IN4xzvegYceegh33nnnXm+r\nYl/DwCUGyQIuNmvYfPIr6HhiUKWYa5gwtWw0td+BCLckvysG+caHeoYupIt/0vkKWSaMKbSWETac\n12VSkg/kba6YnRy88TUwWdv0Q5kxUJefJaFSqDBZIdHU2W+RSC+bEFM1TjpvOUQW7glbdCUxwiDl\nOjFOMUiA7CMBavphEw1nRimaAsk3OwdIDpmZceA8b8nvTzZLVJv0WYcbNSwukN2m5/Krjg3JlnMy\nR2zx5te8kfYHBuDtP/vIFwAAvVXuSsZM0wSTxILS1tmnfHdM25xxXncyouhOwdsKNrDpLBxE5Mx2\nJoJEMFNeOIQeXxcbT1PpYLHGzPr4CQjk2NR/s6DOJr/MOifOAjaTeoObWmmRmrBo1PR8sKsH8vHj\nx/G2t70NP/ZjPwYA+OAHP4jXv/71+MAHPoD7778fJ06cwDvf+c493VCFQqFQKPYzdp1Dfs973oP3\nvOc9U+996lOfuuANUly9kBIYv04WhevfJlvMIqzBS7OAKDSYZuXJaB8ZohE7S2bIzBiyATPkCWBA\ns3jHubAgy+R5MkZIStJETphlRAtrUzKMfosXCtzar7N4fNuetYqGtiQQw106+W/Y4P6/lvOqydiD\ny5CsL2GCsF4pU+JcvzRMAGA43e9ZgS0MGbmBF2bM7NfwqzROcJMJAjNh58RKNUwt48clqhHnp5kk\nyjjrsKI67/UQBqSqnnvV9wIAhkevo9+2GQyrvJeeoL7dxQbpIVY5711MPHpDYtg5N6eIKQggUZ96\n7GZu2r5V9BFxbhGZ9CPmPHMm5WWFBYakvD54000AgHEyOKkfByaNc7aNldx8lPajNoU6Au+/tfW4\nVx3R7qAZdoVCoVAoZgBqnam4TGAmJHWWsKhKyh+uPcH5RM4bT0yRaowlEynKZXkNCInVxLTkdDOB\nrOghcI4xlxRlYh4RQRi2vJeaGjCzMkUyDxGzD7EPnDt2IwDA2/zquqhaTChuyZdHSPMBv/kCAOD0\nY3R+4+oZZE5qjDkKwcw2l7aDISBGWsbzsiYxbanRDSmHLIww2aaGLNUYR1YMR24KUbJ63nuPIDW1\nrB+QqIgYYlRlSVEU1ErhPGerynlitfbAIeSv+HcAgJd/zw/x3stoCPCBmOjmMh2HYZfyzL2uNMZY\nxupZUm5Lvnt+kZszcG7aleP0mYxTKxEczlubvFtLLERPwcfFupDYbzWgvHJxkNXRrslwZZzzpnE1\ngu2RhiMUFjm3VIxsAiL8zgQDSItH3jWjOorzwlV171DMAlqhLI7/GVdi5Ul+EDsyYCglXAaT1hMh\njzwkxWMXQOr2VLsb8XfL5x51RycuAxFzA28ydKy4JHFZieEHMYdMUYX0bRK6HnNocfHYzVO/dbXA\nm+ljbtINmJ8I1Rjnnibji5VTjwEAMhZlmWqCyDfujMPRlkuSkplHDAh8w0+tgaWcTcqZYkgOUnKq\nDI+r4Fw6x0HSFG76AR9j3DKJc2lcSTevHI7Lr4ouvQ4PkqgLi1Q+NP+aO3DTD7yLvpNNYVKItyqx\n8hwJ2NyIJpqbbJ5RFPSwxWAegwGtV66xf/c5ekD3+vRA7A36COywlWc8hgsag/3FI/T9zsPJcZWQ\nt0w4QoA0RSs6XLaUrpv6Wgp8sA2nADLur+zZnKUbIxwLznrc0Ur0ZzksfJwWhSnODxqyVigUCoVi\nBqAMWXFpIdHGFLKmGfjS0/8EywYJUk4hc2wT6pCmY+aUfKoljGhNskKUiKZPvyHhzDwxpyzv8bcw\nK3AlKjaHmGwQg8lHxFI6LN6xnUIqS5Bz6Yg/eC193zwZQWQm4mrhyTHGFGJ2fKxGyyTIG63QufSb\nK+hw2dIhsSvlUKfv5HASqZBIM/9nwizajdZhJOXARituNJnajiyzdchb0hbSm3cyqQVeXqIsYeoV\nqEvkIpcmyWflSELYEcM5YrS9g8RoDfteD19LJZ4vv+M/IrJ3tHE0lk5/i6xelx77MspN2tdBzkKx\nEQ/UCf9GEeF5H7uHaTyBfaKXT50EAFTjMRYODvlY0TYXB6+hz6QTWQiIqXyL949DB9Y0DFHkUkqi\nxcYxTV6m0xGtgiNMduKRzWVy8OgzEZU1lr86roS9gzJkhUKhUChmAMqQFZcUIvqR3NqILRLLlafq\nrkgyqxerw+BTzlZm4cFPzyWbecBkSyAGH8mbwKQ8s0eLhWdd2AP03XNzzOCWlwAAk3OU+8vHy8hY\nyFNZWmb+ZdfzN3C+O4aGQ4kIamLzT/58NriDMNJEiCRHnnoaIRlcJi3VhCIIG2efwugsnT/DjFZ6\nSWds/Vg0GVnKN4uRRUTO+U9EykNaFi/lfT4HG31MzpEIauMZttNco79Dj/tMlzl8xnSPBU9e7E4n\nHo7LjapktSn7U48hGReiRUIlg5Beu/N9DA7Q72VsMXnoe/4TAOCaN3w/rZtFbDz7dQDAqa/9Pb23\nTtvaDyUGltnzkHLOZaS88PoSRRdsL0NxgM1GIltn9ugWvXCMWPDGs89i8gId68Gxw7QMW1Z6ESP6\nmErvAl9D0iEq+og8k7otYbS8bIMNZ1xH5tJJB//N5ii9ApEjHlJeKCVjvWEOx0Y5MQ1zuRZmY9zP\nKpQhKxQKhUIxA1CGrLi0kJ4Q3JN25RQZQ+Qxq9lqy7oymXGgZtZt44EQQnovE2WsfCaKUWOoVd82\nGxSirSmgZdZ26GUAgP5Res03z8CNKDc47hNLmuP2ejFGwNQMHKgZYXveS/x4RhpOSHmL7HrTvpRz\nv+M1tjB9lvKY5TL9XWCEQiIDRkqCJPIg/XsDAqRsbNo802ZZ+rV07vn7hDmbOQvHVqQD7i28uUzK\n40GX8/j9AoGtIaO0CfRi7FHCO1FXS16Vt01U/DHC8vIZs2fDxhpdZsW9RYvJYVLS33Dn/wgAWLzp\nuwAAbkTtGJ/+4mdRnSFF+ZzlvDDnlJF1YZh+S3VRNkfffaBD+zFa28Dys5R77wxovawjtpS00vyR\ng3Br9HvF/AEAQCXj3ov6vD7G0YiOgpDntaGH6C8kKOCa15T4jUp/Z2G4fMwcAgopC5RrmvtCV7FE\n3hGLWyheAvSBrLi0YAHJxtlTAIAOi1+izWCCKHvk4q8fpCKyia3Pmg3p08Ma2z+0Y4zp4d7+jNbH\n1GeyrDR7d3MvQ96n3+vzjW753/6Ov6GDl93+Lix97XPI5ugh3Z2n1w6/2i7fnJFTFyLUVr+XEjHG\nuodtFCcqCkOP1yhM71dXUfL/DXfYEi9pqf32sZecymKoS2YApK5Yoaoakyg5n/yZB9KEKD1AWHQn\nYdXMwnJ/as+h0cVbbgUArDz+TQBA7sbpYZ/Kb1OZXIDnEiZxdTMiLkslTRm8owkiVxShM2TR3wKF\ng+11N+O1d5EzYY9FVEuPU5ne6X+h8HQvlsi6FGovZWIhXtKhSikUw5MGqWm3/PAaHLDkgAVg6YUX\n+FjxNlp23IKHWaQHcbbAk0EveyPHIKTwsaR4ZPIRQkTOD1m5dnzqKIWEIBMs8dnmVIT4wqM3BPiB\n7tMTXlISjlI3QB2zVve684KGrBUKhUKhmAEoQ1ZcNMTm/8RLl80YxudIoCMMogK2dphpEBmbyQx/\n2lFJQtAUjubwntR6sGBJGJEBaiOKXLyxxTChFsLIDwujK2S7jK2FSRJ2TI5h7Le8ehpRQrzPUxh1\ng0tZ8iGxrc6R69E9dA1/gYT2OOzIoW86atOvL2n2PGXMMO1YhlBh4zRFKCanqeOPGxELzkX8Ex3y\nJIYTf2l2ypKevcEgCNvMphmQb0QuxPwlMIuuQ9d1CDS29tmnbc6QGzqOnQF/D7PW3o23AADWvv4Y\nioIYbuTzKhGIEGvxVsGsu0rRCf6N6GDYgarT5fXYoGP+ljcDAK5/6/8My8zy6c//33TsnnkEAHBA\naoRsqt6CkTC99FN2PkWBJSWTcVKlYjOUylXICnrv6FESbG2u03kZrZLwa9NNcO1NL+fleewlhzr+\nLe8RWFTnOO1gOuy3nRUw0p0piE+1GIHXV2w6Q1JmyOPCcLg/ZhaBUwZCmq0sW2SwWZd3dvsUkWJ7\nKENWKBQKhWIGoAxZcdERgdTJaYMFQRn7+noRBXmXKHV7Th1a+Umgrixq+ldLblIgOWCXjEFiYnue\nWYpNJSARMYm/hEFJnrpmdvKeWHDK1orVosmy1FMXXkqBxMaR9nm8dg7rzzwKABheQ0Kh3lEyGPG2\nt0XulWwoX8L8OTZkY8I6K+4v/cLJR1GskYGHSaYQtHQl3sc21JGGJJRig41UOmZTGYtp+Yi7KMfM\nIBe2JJGHkJKeSS9QdwqS/HSsF0nfyYxMSqO43Cc7fjjtW13qRt/nApBxeY9nL2sp0+HKHNgsQ8bl\nRSWXJB1+41sBANd/z48AAEI5xmN/83/R73Ef415H+v/KGK4SW02mI6l6yqRuUxnn1yu28nTstQ3n\nkgAtcqlWzufnANtiHugPUfQohyxiuSjMlMcZXAUrrFciOBLuyDuIU5a0gOf1Kxm3qCNJYvySSb67\nkB7jtWt5CixByhQNinw49fvaG/z8oAxZoVAoFIoZgDJkxUWDSXlAQGbKkyViFz4QA0jsxQOQPGy7\nl6qp2e5OamvvfbI9TLk6eRX1ZwiJkcXkKVgruVNJlpRzSK7U1ttTb5uU7jBzEDYeA2w+reT2zDZy\nZhuFGaELtoZ8ho7DuTPPAgAWrn8F8uECb7fYgopF4UtgyDEAnr577fTTAIAVtnHsVJPUJMDbaRVu\nOh6xLoES9iu9oyOzuMx7GGG9qXuWbAEreENAZIVwIdaXkv8HUoetlL+X32/si9nyyv1354iFDW+4\nDsuPkko8Y3tKJ6MuNiIFqZcvm2WwfWreH2C0SFaV17/l7QCAo6+i3HE1ohKrb/6//xX9dTpHBVt/\nNsce7XJI7D+NHd5oV5WpRKwacb6bWaxhphzLCuB8sjQ/8SWr35nhLhx+OTbW6BrqcHleJ2cTFQk8\nhJCsQ0VRnQXOG5sc4GYQm9zkwnJZWd4oL5R9ytN1xueec+2mUT4lSNdotCgkKoLpyIfixaEMWaFQ\nKBSKGYAyZMXFRwQctzR0Y6ptTcyB6yWzLEMItW0jr5a+QGbswkq2qyM2dSExfXdiKULfbPrWkJgM\nsyVbq7STP0ia8bMq1mZ1rjNROJPWB6gBQmzVU4sAWZhlGSxyMacYE4vtlJTTXfnG8wjMYHqHbwAA\nzB29EQCQ97mln7HJ+lOIpfTEFe/H1Wcfx/pJylOHc6SoRk7LTEYVwPnPUPC+8b7bpOy2qTFBauEn\nv5UU71WKNMj6psWErDVJ8SusPBcldEDDnKPubUz7JUzZTtWjA0DB46SUv+fmMTxOqvXR42Re4jhf\nb7qdpB7O+rTPls028j4bwLzslbj1rdQ2sXuEjnnJSvnHP3s/AGBQnkPocE9i0RjwmBQdg0Wo9QZy\nGIIoqoFqQlGRkBgx1/Yym62qCcCtJSG9mnlfXUF5Y9s/jk5JDHv0HCnkR3w+Fg4f52OWIy+4bzEr\nr+UcZp0+gkQG2G50cz50n/wAACAASURBVJ2YckdqltHIgWfTTTfk77zb50YqpJugV6kpDyj60gAD\nfDza141iO+gDWXERwQ+rCGxuUOjPGCnNEBMPCW3WftVyN5Nr1zccC9piGTSWyc30+ukuJO5LsMk0\nxKY7pvxt07ZkudyEpsuwQgzJb9ua6UlDmh8Ykx4ytdhlurSnMFndc1nKvjoU4rN5DutZ/PUs9Q+e\nPE8h52yOSmE6C4eQ98jzuOCHy9oSOTytnmFv6fE5WA6RG+5lW5UT3g+P3MpDjsuukumHhPcj5AzI\ndku5TjLWyDoI/BsxTE+YTOP+m0lolx/MQZoWI4dJNTNtMZS4rdXL1GYftUiPji9w4Bp6GIlAyj9G\nD2YfKrgefVefw61xgYRb2fW3AQBuuvMdwDwJxLBKYelH/ub/AAAM2DAl2LpTWDrZrWdMCD65gYm4\nLD18qzKVjYXWg1g6M6EsUyerUkxU5unBdvg62r/KT2B5YtA7Rj7qYx4fy8/Sth84ehjgHtIFG5VE\nPr6m6CBmEqqnn+X2zhhvrkNQbqzQ+jwuZf1M5n3RICTzFg6Ly/nodoBCyp5kcqqPmvOBhqwVCoVC\noZgB6LRFcfGQiIRHOSaG3DQfaKJpfSmMeDvf6hSyblgjAhRe3qmTjJQ2BR+2hD+b66RuN2E6bNr8\n/bYlYRIjyS7HuOU9YXhizuAmZbryJIyduguFDiyzm5zZhfFkDoElei2XT6IUK8S0H7R+j5lqtBm8\nRAg6xIgGJb26wQAlMz8RbDk+L1UlAqy84SFeC73o8DSEOt/BrpTMWMR8hc+VlE35KomFRFsmcfLU\n07ppiSomLswoJXROm0HneOFGYo3ZiEP3p59DtkBU0M+T//jhN90FADh2G71ak2Gdw/pP/rf/nfZ/\nncZrEC/NPEshaqT9YLqYIjABQULOiSFz6LmcJC/rUHKZEacrpBwrlB7gLldHX0mmJ51FSlNsrJEx\nSBivIxa0vLDg7mHyWt9gYeDK0vNYOHiYjw0t05mnqIAtuomFjUXPNyQW3uNxAgBFJgI0+t28K2NS\nysoiPF8DebLsZHOVxZthIQYtzP73Sai6LSbdayhDVigUCoViBqAMWXEJEBHGa/JfAPUMs9nsoV3S\nFFpWms3PJNcpM8oQAkJdtzT1KpaCxjRSz/IbzRlvi6G3G1EYg+RIubG+ObX+YMC5uoDEBtL6Yh4i\n+UVjajOJXOwspfHAJJW+BBY/SY4udWIKHjFyblJKolJplBiWABnXItkguXX+IC9g2LIzlYZx3txx\nzrMsRyhYGNQWu6X0uzG1kEcUYGI0IhthgCgMkplxxTlUa3N4Yb3C+NmkQn4isWDUeWlXtfLdBqh4\njYIP8dwrSZyVH1vAeIN+45of+p8AAAs3voE+47KwF779DZz8/F/Re1ziluXUuKHi3s++3ES/KyJB\nidLwdvG+V75M5UZSgxSdjOW6R7MpWUfBr5IvPviKl2P+hutovZbd5+EjlJPdXFrG0jJ1e5rnpiUj\nPnaDQ8SK3eoKjBiz5HzsCxak+QjPYwduWmOBUB/rYkjMvDuk9aKMyVQK5yFDVxpfjHNa58ixV6Tv\n2S/9j+UeMGbth1zb3W53x3V2A2XICoVCoVDMAJQhKy4eknk/YMfEKMUq00BavjUV1NP5w6TyncpL\n8ntuOs9sjUFoqbPFzyPCYydMsfGmNLjxWVLz+phMKST1K/lh7yQvCmRW9pvZoiiaRUAcAZkL21xs\nJfkYOI9YSfOGJGmll+l2HfzaMD1BvQs+1H9Y8YjMaqVs5H7DopJ2Qeww2f5wMk4/kqwvU8kXH3Nr\nEpMTK1LJk8qiASYxMCmRkm2vXImcS7xEde+8lMHVt6Z2pCQIYxcjjDrgkCIOIgQfHDqE7gH6rqVn\nqPf2+HkqvRuvUovD8XOPoS+mJ5C2g/QbnZzKjdxoDeNVyitL5KBgJbFESWh8gI8HH1eOitiI1Mwh\n9WVmBfR1r6V2ktmB+bSPlo9VzmYq4O0bXtNH/yDlg194jk12WFnOVUzoDYdwrCXo9clkJnWijiFF\nQXzJ44ptOoVVAyk9Dduh/HI+pLIyafqR97rwfJAdH/w+l4zZrF8PxH3wiIkxouToxkhMXficdzqd\nPc0nK0NWKBQKhWIGcOVPXxQzjxAcPOfrLCt/Ux1yQ0md8pDboK3ezbax0hR+stN8tTmTFcVuU3Ut\nDHSLSprnrS5EOCnB7TGFEPP9bLod5Hbb3rT/FGYZmZlmkivNM4TIytzU5J1ebFMRLoxYao1tvR8A\n5WvTMRNzh4b1pmyL5brq1L7CClM2KEvOl/UlUsD1pmLpaWpGmwTqkjttpvOTSQa/NtTryRhGWK+T\nemRmkbHOibdz+k3EMH3OpL2mzTy6OR2j/phUyJsTqlHGiKI2WbfeJzmuksuW2vZiMAfXIWY8XqUa\nXcPbnsZLCHCssk5tF6VmehLgJ7z8IrHu615ObRRjlwuCo0kHKTU9YaYuURJjDCwvf+TGVwIAyk3S\nZyydfJL3vsQ8K6/lakj7FUKtZZBGHDyGbN10EZkYx3AdstimmoLV17bOHYcOsefhkZtoN/ZN5pjg\nnEsMWcZgnxuc7LXaelcP5I2NDXzgAx/AysoKqqrCe9/7Xhw9ehQf+tCHAAC33norPvzhD+/ldl4S\nbCdpv9gy96sBrhwhiMik9Vn9kIrJM7p9w40hbFkvCWpsMwxYh8inFkodZ0wS5JgkCIqNRdK3Ti2T\nynxsB/McVlxfo7BnUdfr8LZ6RLl5pdj31NfCZllypEqlXtLb1lhkfON3XN4T0820Pi51KZcc1+kb\nr7FZLVITQY4IvoypS7xaYXn5jbzTRcWOUN7Rq+EQrRxD8acCGg9yMQFJpU717Tn5h9vaXMWlDkPS\nN5hzAdwRyWQRmThBtXyu02QOpvEIkIc272rI4Sw/cAzdRPs92o8FLkMbu4DNMR2jyYSWzSXmLT2D\njYHlMP9gngU+62SkkUK9lUsPZMlpyIRpNCnRP04PycM335S2GwBy7lJmENOk0PL+y8RC+n9bY+qy\nQB67nQMUwr7mVWQWs3LmSVg2jjFxegzF4BFlzEiawcvEog5ZW34gm46Ilni7ZHxlOQzYS/swTyxy\nEQFiX8CnSVVAh69JgZQ9blfmeCHY1QP5z//8z3HzzTfj/e9/P06fPo2f+qmfwtGjR3Hvvffitttu\nw/vf/3787d/+LX7gB35gzzZUoVAoFIr9jF09kA8ePIhvfOMbAIDV1VUsLi7i1KlTuO02sqK76667\n8PDDD8/sA1lCNtKPVCCznizLtphUpBCftcqWzxPCANzGBqyEJDEtPqpDzqFRrTR97E2MW1izbzHC\naOoZfwptTmuQ6NdT6G/a9CKinunWURHuW8vLzC8cwhoLgYSd5D0SvVRjth20DfbZCl9L2Y/h5fjH\nZIfozxDScRODkLojkrBpD2GCnsuXQiP0DgAhusY45e8TjRhC+v0AsYNksYp4U3d76HIY3m+yKQUz\n1SrU/aJNElw1WTMSC47Rp+5OqSRLRGYh1ko3L2KyaYZsG+VwCYlpy7kLSbiXDCmk/ZQBctBxHPHx\nd3zMczb96PYt5rh8Jd/IeTu4RIn3w1dl8uQWE5WcIyFug0vgSpeMQAJHFxx/z9zLb8LBG27i7TVT\nh0PEfzmAQo5bK7oiHZlCCDAieEuCSL4G2Jt77sQt8Mz4RVwmBjTeOUQJWac+17yveX2cpZwuy1vX\nkuV1CwPP/toDDo8ni1XqTo4rFXL9y/OhaVrULn/qdrspgrMXMHG7hMx54Kd/+qdx8uRJrK6u4hOf\n+AR+7dd+DZ/5zGcAAA8//DAeeOAB/PZv//aebahCoVAoFPsZu2LIf/EXf4ETJ07gT/7kT/DII4/g\nve99L+bna8n+Lp/xFxWyTc65qQ5Dzc8E3teNDoRlyGyp2+2mbjUCZczbQwwxVp75Btxpiqj41mdN\nsUntXDmdX0UIW2ao7SEWQki9JAS1zWaDPfJ7Wds8JEYYMamQvrWJ9dHrgYUj2NwkNtQbEruSmbIJ\nJW668z/j1N/fX2czG/1lm383BWTCPGTbg3+REi3Uy9Tdr6Y1DjVD9sliUowrUl9la5MHRCpbqojh\nC6MzRQ7wdRJ5H6WTkTTfsDZLO1CXrEleV64tk5icaUQjACCPpj7m/BteWMmA8r22002RKwC44V3/\nG07++e/Tb6YDbZBbiY6I8EwiWgFlRuVFvTlqIBE5VyrHczyeoJB8LK9fcl/ltSUSgtlqgmqNxFOO\nS19cKf23aUMmk3UYPn4j3o9rXv86AED/2PG0wVmKTkw3bMisTbV6JuXQ+VgzK86yLN2L5D5U8vry\n9+bSWURu5uACn7vkV+Lh5bxWNJbF+rLLx/mG/+XDePb/+QMAQMERIOnNHS0z7/4Ac9d9LwBgcIjM\nTITy0xC8cu6JW+8t9bMCoGuqbZkrIq/BYHD5c8hf/vKXcccddwAAXv3qV2MymdTKPQCnT5/GsWPH\n9mYLFQqFQqG4CrCrB/KNN96Ir371q3jb296GU6dOYTgc4tprr8UXv/hFvPnNb8ZDDz2En/iJn9jr\nbd0V2vlI59yWGY38LZOKpo2jfCazz6aRRZspK1pg9Wm5uZJaswlLCG2GHGOiLMnkQr6m8V5iz9uY\nhmx5L7E1Yaa27h+cVmosK00PmE4MDhCjGk+IEa2sLmHxANkTliWxC0gzhuSEEWpbzBdT6Jvp4yA7\nn9msbqLQCgMII4xZvR8mtBh1o0QqCCOWHKP8FXyyw6w3h8e5RI2sQcylvR4rbauNqe9DZtKxbZes\npfMbgVxynnVtFAAqmUq9onmfS24RWXAzAxs7ddmYbFu79yYMDDN9yWlLVMMVQ/S5tWK1Tsp4kQ94\nPgadogfHLS83mPkUHbaj5B7D5549hYxz0eOV5+j7OE+bSQOK4LE+ovdOfPfrAQC9Q0RMTGVqtxKB\nq/ts0+oubZxNw2Ire0sKe2b1UjI3Wqdc/2jpLDoD2mdnRF0tm+iT6t7yMctYPS6aCaChw5DrzdI1\nYHkshOIQeovX8tLMjFM1wZXFjiXi0MwZA/W9vtfrJYtMMQYRq9y9xq6eKO9+97tx77334sd//Mfh\nnMOHPvQhHD16FL/yK7+CEALe8IY34Pbbb9/rbd0V5CHbDD/Ig7T9QN3uAdt+WDvnGuFCqSGdLstQ\nMFioU47X0MnkwSs3Z3pJ3X4QU2mUPCzFjQsxpvekD3EKkTZCYz59KdJ6AGClrANuSxi7rm0FHIuo\nhlwnurpKN7jhkEpIxnGMde7rPDdP75UTDrXKT2JrsG67FE7ykE4Pqca2bwk/b6njqvsIS8ha1pdS\nlmjTxCBItVMUEU9IoXHLIh3xXO7YTjo+FVs/hazkZSreHhYYxbrDVkwzHSn5IqSJChqlSCLwCxGO\nBUWZhJzFBYzrkAuEpPuS8RG5lEfqsjNjEaQOO/XkpfPTG8xhwqHmuo6b950nBmE0Tv2hZRHvOHRv\n6MbbX1jEyjI/0OcojDsSX2PunAVvcPwNr6XjeOQQ7zS9uBjTfsi5Ejcuz/sVEFP7YFvIGJAQPAv8\nfIDt0jnyfE2VXBddrdNkBqVDOaH+2Dh8lI+Z9F4eI+PzGHnSE3nykWqe0SjlkrIp8T7na6R/6LpU\nknUlhafbCCGkSf5OqciyLFOouk3M9pqU7erbhsMhfvd3f3fL+5/+9KcveIMUCoVCobgase9iru2E\nfPKSbZQt7TSraToBtdmurNNM8Lf79ja/V9ky4Dism/lJCilK2dMW4RbqUHGGFkMMYapIH6hLiJph\nYWtbx1wiGA1xVR0iFw/mOmS9cIB8f5eZGbfTHVmWJVGXCGPm2bGnYpbUdNPaKWQdY0wlO6YVZg8x\nppKqtF6cZv5TvaP5PSnBSeG2UK+X144k9GJsrflw46nfiLmEL22KTKdQJgt7vJHfyhLr2xKmbGxr\nQiuUb02BwL7djrdDTFWQPMI9wG5XEJMMO+3vbLIsuUb5DkU3OgM6lxsbKyjEcKbVj7hpMOIkpisi\nO456eUdit+6gg7njFKI9+/VHAAB9T8dqo6Rzf+zVr8DgEHVgEuZvpWQNJnW9kpI9tNzisqke1NOQ\n9/M8T6VpG8vEjJ2UWPFgcFkXbkTbnUuvZVd3yMo4/Gpa3byyvNG5SI6t3NP4tSpo/44cugFXIjNu\nd5IryzKd69BKi0mYejweb2HEIuraa4asXtYKhUKhUMwA9h1DbjPjNktpzmjay8hMabtZajNf3J5J\n+VapytUu9kqF9RNmyKGUdr9JiCK5YB/rKIOwRWGfYi9JadXpnC+qqdZJiDHA5NPnrZ0vbrK4ZOPr\napayvELMeMA5wvq8CsPyOHSQGML6CrGTEdsn9jrCPk3qtNM2DPBJKIgUMUj54rZRCLbmnptRHjGT\nCGnGD/7NxnESrQPvY2zUhSWrTCde3MT2ROhkImCFibJhhGXrzDHbQ2bWIk/bxN+byTHnfQi19aXk\nedM1aQFIz2YpbZJxwt/jOxkMdzzK5JWZneQwfQiI3Dmp6FP55WSdDFzyPE++1CKA82WtJ0nHKrF2\n+n3xdbaR8rLjUKDP5/76N78JAPDco9Q96rrDxMqzXhdeeh1XxFonfFy6RTcJtnwrupN8zH3j+IlI\nzUxHFUajEaoxbVO1Tsx8wpGoaDm60Z2D4fNajWk7xJ4zWpNyxRnnhcUWMjaEeUlAmErLaJ3i4An6\njWzQ0ATgioHcsyfcIct7n/LD7dIm+bvX6yXB1zpf780y372EMmSFQqFQKGYA+4rKVVW1RQHdtMME\naEYkZg7yXjsv4JxDr9eb+kxgjEkzyrZcvommmcG+R0rDitkFM+SRdC2qS1WyFjuQnKtzLhnXtxlt\n8KEuexJWkWblzPS8R8WK57oJATMgmXdmed1Ll9ebP0ysx+RdrK6SilZmz3NzpNRdOkvmELk1WGPG\nITNZqQzyzC5ccGm7a5OMaSphYFLpj2kzZJiUl21HalL+CzF1cOpw3tCLzrtpfJJKmESZzn1vQwYb\nWTHd4a41zK6ka5KxSF2uxJ7TdpmBSd44zxHFZB/CQms/AjlAkp9NZTYN4l9wyY2Xcrgw4n2U3cnR\nFSMSOem8AV6sRG2OYUEMuVw/R8dFOhGFCMcK8pyjAJbZjbClcVkmlXfGx8hyx62MQyldN0IFYqJV\nQRGUY2z6sXH2efrNzMKL6Qers8FMeby6hqJDv4+e5G6lJEnyzQbw0sRAzhkdz01WUIcY03gyfI+K\n3Gu82qQKALtwCK5DOfQopV4dvm4GPYAbRmQFd5nisdMsoRPbVsN55YpLuw4eEoZsrwhm3NYUyb1a\n7uFlWSZdyHA4nFpWrs3xeJz+v7BAx7VtHLVXUIasUCgUCsUMYF8w5O1MP9q1xk3DcJkJCdrrdDqd\nxKLb6zdnRG323NyOpip738NMM1ths65c27Lodjal9B1mS0G+MBgbXeq7G2T26qf7IXsfUt4xz6cN\n+n1i2iaxxoMLxH7X1ygXXPSHmONxsbGxzq+Uo1s8dAQAsHzuLPq5sNy0R60drD+MrSRbc9+bCvLm\n39bYukXkTmMnNsYjv9XuFWyM2frdqZa1oRoV1sa5ykzqvL1LVpkCaQmIivO+nSIZcBgxx5Cccoqa\nmJpJNXLosl0pisB5VJ9E1hLVaNQYSw46r9XVANDr9rG2QTnjfj5dpz4JAfNsDCK3u5jXNpSyPWvn\nqG43sqJbcvJGGjCEgFyObUksfFSwne4i1RxPVpdRNHP4AAwzVW8mmDi6HjqS05f8uUSUApAXwpFo\n25zULMu2AnBSD84Hsse/H1bp+6vVlaQJGAnjZnObvA907HBq/6XZRHO0GW6lKPXH+ZB6HhfdBV6i\nOe5n9x7X1vnIPks0tNfrpetcmLL0Om4us5cNJF4MV/QDebuuHHLg2gdQwhB5nu94o2u+3wxxb/d9\nzeXl4dsssZKBcKlO5OVEcoCSkhGIqGs1LdN+AMUwPcHx3qeHbbpJi6PTZNw4nnzjbj5kAQRrU0jU\nseArGReIE1K0yet4eekF/j6ehGUWYw5HH1ykMPYKGy6sbbIbV9FFJvHa9LDlB0CjNKk9eWuWNAHT\nHcPaYekQQ8PzeXsYaxpKmtZvybshbPl9UegY41FwKDKkrldtAZlJD6Xk1iQTBRnvMSDnkLM4bUno\nPpWu/f/svVusZddZNTjmnGutvc85depyyi5f4sSxTW7EJiHNJUbiF0QgrPQLUi4PkZHSnX5KIuhW\nUBSiSIDygKLQEgodCWGJECUgQaKW2ry0aZoXXhqJ9t9pwt90GgghcRyXq+yqU+ey915rztkP8xvf\nnGvtc8qXlF0n5TkeatfZe13mmmvtvdb4vvGND3msxvHcyw3FBxgKmhiW5gMwx9NYNXxhWVsv18X2\ndgoL717bxayV0KS4UDHc3s63tUyK5Vv0haZgKUbg1KkkzNp7TtIeUuK1atjXGCn3AsCwRzL97TfT\n+6s4sA2ygsfnZw0aJ2LBw0V5yGp40p3axHxLbuANzVckLF98b6yWe4nnOhswnUuuYpjtY/n8M+k9\neYgKC4q8emxs7chxj8+5KwhGNHQDS++d3nlD+mDU//vkB1iZfpr+5pd9DOi6xfsI3bhuhuHTyZ/R\nioqKioqK1wB+pBnytE9qjPHYUDGXZTL/eijtNaf+ptcLO5as+jURqhYYDb2luZKIHAYJkzmYdZY4\nmTvEqAYFQRgUzRWM7VQo5rmd4VDWSy8WuZTKiNDIC2tjhdTmvIOXVASkT2yQ0NzVy8+gk/33wvoa\nUR85Ed0AQOQ+JqIOfbI1Bu4YkwxrbPHW0WFpY4zaatL0Q9cv+ilzmWCnDF1EcsZobVfDa1cERsE7\nFTix1RAZLmgZCafhazJUmqk0IgQ7ODyEaSSCxFA1ct9eIJW5kdkqa1aDEiWdOp9emT5FfwNcsyHv\nMZwu50naHNnVEqYfGz5wuxubjUZIQE9tjKM0AGDkd8G1LG8UMWjgPj1Wwr7jKq3X2sSsxF0SMXpE\nz2gM7TDZ0zvAswxQ5pqeLBvnkz96d34ndccCEMUjnZae7MHsjdXyLy/9mZuWqQgRss63YTak7Opa\nEnVZuRZaP1szx+D5iYXXdqQork0CuPm5u2Si5MWcvMjfNDxdim2nFsjl73tZ3lQuy2WOM2t5JVAZ\nckVFRUVFxQnAjzRDJiMt+xszEU8mPO0Te5Qt5lGIRb4PGJc4TYVe03xz27avKYY8hRFGaQpmSZAZ\nx7X+udnej+wiz2tcM18hW9N9mvyeBjGE5ZxVsdY+tsUe8+C5lGNzq5TnttaiEXMJI51/wqHYYVLM\nVDB9gn+pacdR511ZYCkFoyhrYpZR7EO7Pq3llI3aLsKM56HMX5N1Rs8csCzUOBUERZkjLsM8bwgh\nM/SGQi1hZlJe1sQeWLGBheRpHS1SVdWVxybRCI9scBIlUhF7adTA/uO+YG+MBkgJUCN5fOaCTTOD\nCeNcoSu+92qpOimHG+lNhOb6+cZo2wNyxK2haFPGPJdzcHD5kkxThz7SXEjG7CVK5HvYntGZtP+N\n86kT1FwYcgDg5fcrd1sai/fapi06hBESpeHcN8CsS3liPxONxnPPybFGWOmDrOdKrm9v8+0gSFen\n7mzKSzOqcRJ/147q6AekOeQ55u83WXBZHssS1WlJ1M3Q/1SGXFFRUVFRcQLwI82QCT7RlAXcx1lm\nrlYrfUo6DjFGZdpT4/HNzc21nAKVfKUJyUl8knylwSNeXEsGBcy9RmRVcZgwTC2BAXLzAcnJ0R4S\nMGv5epZllO8HybOtxL7w9NnEdvbEhD8Yg0DnTWEArfS7G4yFkxKgw2uJ8bTK0LV/YY6KUJE6yce9\n1AhMnoccOZiybM35WZZGFTnsePy+pp/0bG9pW1VMe1Vey7Fyfq3VDdDCVAMPwqxcWAED15fcsUwV\nvyMOUPtHmqE4bawR9AJg7pjfIWoGYox6HYh3Bxr5bCHfu6bbgFn2sv645O1wcYjT863R+Km0L5uR\nDNSKCNMP0mTj9OmzugznYykNG1Z7KcqysS1K82t7mh9W95TD9F2Iwwq9TM7WucRe29uFGfMc9DkK\npNoKngP+6bwq2e2aQp56DK/zv3EmlezZM6kN4+LSdzFIJGmeHWjSai7ra2gEcvY2MQIxJ5+7TX+z\ngXV9j1WTnPx9JVtmM4lXM2c8xcmf5YqKioqKitcAbgmGPM33HoWyOQTrzKaqOj4plS25+Fm5LPdH\nZjw1IXmtIbKrujSuH6Q5ujZQQH4KjZM8nOZSF73mqajGjawxDoPWwmoTAO6cObYY4cUKcHOWPt2/\nnBjMJu0tT+0gSA0o2S+kUUFYHmIQ20VHy0hPu08ZO4wa8/M4pmrrkV2C5izlb+ZCiyYYukyRUw92\nbCvKpZUlmczucjVyYauJ1A+CxhMBVPeSoc4RjNgnRp4H5pA5nrDWBlPZGs/TfAtelM4d1eZK7Jij\nAyLzj2RiJueX+X/LolxhZlabjliAeVmqi61ENxwV5oCRxhfR5BkBgDYscbifIiTz04mZspaWYxwO\n97ASE5soiu6tM2nZgec3eATJAftlYr3zedqnlxys22hx5Tvfk3mQvOYBjUYCTt2W6tvnp4RRy/U1\nsN0mvBqaaOSn0McAQFis0Eh+O9CuFEVUQ9b1Ep1iLtjINTU7dx6NkWu+YUcSmQ+b7X5nm6L83qDC\nfax5uNko88ZkxlO9EIC1HPLUQtMYo8z4JHhG3BJ3EP7Yl4YcR5WTAGNjELpxEeWNnf8/ypOaN/ST\nEOI4iehX6UFFvzTBFx1+xINZfgT8QoRfrcMgJTdTp65hGNZSEGEi8kqhJxEbbaVSDbOZXlfLJM7a\naHOXJ/64N3KDXYkfMFD6KI8f1ACT744MM8Z8I51i7b3iByP3OpbxlOVLvJ5pADG5MZpo8oMI75Vx\nfPM2plenLT5YdNIJKXgDS+9ry2McH3P5/VlzAXN06uowRPkOGZ5zM1rHmEZ/zNekSMaogQef3fjg\n1ToZe/AwRkQ6csJvSAAAIABJREFU+pAs303psLW3f4DOHf0ddNYgiuDr4PkfpDclHG2LB42mS2Ht\nTbl2NEwuF4zvF1gcSG/khjdAeWDjQ8XGHOdel0K8z/8g+VtD+jJvnTqFjdNyPS6kV7HrR8cehkHn\naPr75Qc+4DSFux1dvGTuCjGgF2c7yA2ZeSDjgnb6Uq94FRTmM3T6rh+T98Tr/ITciKfo+z6nOSbf\nxXCUOY6A73dddyJuxES9k1RUVFRUVJwA3BIMmU/jwzBoKKKZ+NUSpenH9DOGPmKMyn6nT6qLxUKf\nyF6pjh8/6lhjhjEqW1MDiF4YlZyv6NajG7Fgj0c9/SbkEKUVQ5ErV5IZwpkzKezGCPHeteexc+Ge\ntH8Rfu3vpbKntmkQySqmw1c2arU8iM+y3Ha2p4wjJlyifFrPBilj9mmNUbY7FWyVXaziJHQfyv0D\nCFghQEoAxX+4B6/XCMtjZZqlsBIcHc9k3LKwvLZawhPEDaZpx52Mgh80LJ2PJs9P1LA8y6bSElZY\n6HC4q/amVkK0DVsnB5aydEA/KReSPXnb6Rx3Go4Qi1YRMQXTYLa9w4NNi8hgaeaxPNjDrKWhCY/f\nlKvAOqA5nca6xVD+QYqmtbNWr3mu7aWvtM6PCWodmstyMD6uEAvhm7B3elHb8veMERiWvLGUMEd1\nVBymHbsyP2u375TNjK/3dangzUGZSix/t8vPQgj6G192dwJObnTzZI2moqKioqLiNYpbgiGXxhx8\nEqLgip07XgyLJeuaz+drrID55rZtX7PireNBYZX8JezLSH7RBcAWfYsBoBcXfop5ooc+jWuXIS3L\nyHaSZKuBFo2eQqEBQfJkM6Gty93UQGJrO4lpoj2F5596Ki1DTRfzkqHNZhYCG8Z/h+AzexVhDPv4\nlnnWKZdgVEAFaaHMB49z4bKxtLj+abkh2Xeeh4H5VY6VucbYohWLS4qqOvU9jIjs26vNIOQYyZ6K\n4Whvccn7U2QVAAQnbJNmKtqsSL53rlELTs5I7gpWRJnYaIHlU4FCmxZRryPJXUvPZC9ahfnGBSwo\nbFK7UJk75zVXbNlcQvbPMqRTZ87n0iNlxiLgku/9rGmzKC9KFID5cs31B52jrdvTMiy5i8tDLTGj\ntawRa1nIb1ZsHOwkwqfXO7uMzRwGKY+yjOIxfy7Xf+wHpVoGtEIV2A0MMo+OvYJZorZ1gUvBGTde\n7yYzY4qx+LtegoLbae/jsmsf7xEnScB1FCpDrqioqKioOAG4pahe13X6BMUnID4hTUucyv9PLTDL\n97g+mfdJyzmcRFAN6wtjD81tSq7PSeOIYWBO2GjJylFNQ0rFdtq29EeOjI4sYUFj/7ReI8+b166m\nnPLs1ByO3E/LJmTMxmour5mwRxXjGpMbEsixGbzwk/a4ECf9R5s5FNs+HjIOsY6MiFoDZSUXTvOO\nthP1etNlq05l+qOirNF70/zbUePRHH/WAmuuVeeKjM7kc6+b0mPN++Q8sH+y0XnNLKcXBXyw0ixA\nldzp52v/4CraU2dlPqTvL1s8RqPRmawuTi+53WBehqx7JbapLRsvhHhEiRuvXW7X6sbZonJjOymr\nr+zvwjqWZlHZL3ahkkt2zuQGGjS0aPhdinmeGu5foiEs6+tZnhe1v7S2syxU/YzKrKQF5+m7HwQA\n7Lz+Iazj5jJj5tKnTJeIMa6VsZYRU/6fkdKTjnp3qaioqKioOAG4pRgykOuGp0y5fIo6juWWCtOp\nHWZVVB+P42oUs/K4aMfHPBxNCORp30aLyKYUmr9Kn636vijkZ75OVM6ilO1XK7hubMjPpF8nLKNf\nHKBTZaqwCZNZMNkIV89a2nXTD22OsaYMj8pW1ez/iBrlaes7vSaLucpq0TFDLiMInAc7F2YsSmBv\nLYwfK7GPwtSSlPW3qSnDmK0p+ysbYnAeesllN/J3kTaeKobpHuJco8zUTiIPPHfGNTBdavrhgzBl\nz/Mh10DbIxxIFGTzDABguUo/bd1wqPXCNNAge+QJNsZgkNaUq0PWGlNRzfkpJ40v4+s+RKNNO1gc\nbCWy1s5nGMRm1MvPbreVxrrYf07mwCuj5dz7qT5jNQByrqlkj/SalXm1XaM2q6DRi+zzEBuYnUsK\n6jvvfTsAYL6VqhHKu8FJ+J3z3q/9jhMc32w2U/bMZcuc8tRH4iQc1/Vwy92Qp85alLlTJHFwcLAm\n9KIIgCiduqqA60WAXZb4QyVuR/rjGiMChV78AVZTAumy4wcNZ/OHTp3TgkO0knKQ7WydF49dCb+Z\nvQargxRmZB9lK/2MrYTynDHwFEypF3Qaog1BHUGCKmK05kT/NureJdumiIolJMNQCLbcaH32Ug6I\nWj6lvspFPJsOUnxo0DuAlS40jdXx58qo8c3ShFx+5c3kpleA4i7eQIyI1Fw062FsCtD4gGAtLF2v\n5KbJ+KnTh5DsY86x6twbq/PPMLYz7K7EfsJBxyY6QAwyLzO5ITW9BeQmu7iWHLe6jVMAgN63ADvA\nOd5c6VctN8vlIYb95L41Y0kkH85pZhUiLNMudAijIYe2dorZcY2njh2U5nOYQ3mgkvF4uT5b6TIW\nAPTcpggTKYhryj7Tco5UucaSM/avjgaGZWwQYxCk78+FH/s5nL4j3ZApcstYvz5uJhaLxVpnPYqy\n+NuwXC5HvQyA9c5OP0qoIeuKioqKiooTgBdF/771rW/hox/9KD784Q/j0UcfxdNPP41PfvKT8N7j\n9ttvx+c//3l0XYfHH38cX/7yl2GtxQc/+EF84AMfeKXHv4ZpSGJaEF5K4aedoMquH1W89VIwFghZ\nOxVehKKzDi0i6Scs5gjBwLVj4UWUkhrne/QSX6TgqpeerhtdemIOPmDzVAoBRmHIh9euyvCEtSAb\nUTiGpVnukgYq45XXiYezKQRKGrJmuY2MOQwejv2Tp9tTEVDIEQJhe30hvOL0OWFXKiYSg4zobGbv\nEwFcttCMUMHWuqxMof2pfRwvYiJiZH9qOUYN0+fjUIbOkCLLyIpSreyFTWYt+7ZWGf3at003rPEK\n5CispBvkGgiwiGJI0ghjX+6n0PPGqfNYhfR935cuTfNTadkNCWcuDq+peCtHdcjqJcoQQk5h8LPI\nKE/Q93k8uVuTlJzNN7G3l8rwHEPnjJKQlbc2h9G1s5d8X9Qe2Oj8efGDn86ecwYrGq1spH7Gd771\nZ9P8bJzR0P06TkY4t7QwPs4KmeDveomj7I5/VPCCd52DgwN89rOfxcMPP6zvfeELX8CHPvQh/Pmf\n/znuvfdefP3rX8fBwQG++MUv4k//9E/xla98BV/+8pdx5cqVV3TwFRUVFRUVtwpekCF3XYfHHnsM\njz32mL7393//9/jd3/1dAMAv/uIv4k/+5E9w33334aGHHsK2yPzf9a534cknn8R73vOeV2joLw6l\niTiQnqiaqXBCnj7L0qaTnvw/WRiXk2hXH3m69UUjCLLnIIb/LN1wLsJE6VATcs4XAHzXqLVikAYU\nS+nMtLtM+zizdQZWzCmuXRkzES8iIGOtCnoovqEoysNkwZiZMEkK0RA0p6ecc2rlWYiYVLjlaE9Z\nRGRYakfTirK5RDMeY5475nuN5n4tmzNMhGMuoqjXkvEXIi0tq5H1szGIrGKCiobyq9Yt6X4oHmLO\nk92wLJuHxEy6neaOuRlT5JdZXjS2anSFTSiE+TSbKT/MTiHDqodZSdSrpZAvfZcP96+i3UiNI06d\nFsHXIIz5SrJNta3RtCwmfX/VkKOxashhaAPL88rgQgw6fxqVkO5VzrZw3FY/Gr7OmWmsLqPXFXtY\nF9EOMmtjJVIgUQEvcz6YFt35+wAAd7zxp9L+JZ/qTVSdwAkhxGsotQu8TvnbTHEudT9bW1trJlA/\nytHNF7whN02zJmw6PDzUCTp//jyeffZZXLp0CTs7O7rMzs4Onn322Rs83JePUpU3xY9yiONEYNJm\n8Nzdbx+9vpbwhkf+u5s9hFsed/3Cj+Yc336zBwCgKTuWnVDwhnqUKGtzc3PtvVtJePtDH8lRJR3X\ne/8kYMpgeEIrK36Z0Jxr+vP5730TALD83pPpfT9oDmwY2B9W1NVyDmKIQBj3ZY2OOdQGh70YtUjO\nyC4SC7bCQLpmA4fPPw1ABbfKKONAZbVDEJbq1EaRDDNoEwNeD2R/jkpo61QVjInxgnUO9/3Kf4t/\n/+svrfXSJvNmWVWaLmHIRzzNa8MF+UyZsSYxc0mSmfy66hUc4tp3UFkool77zLfrMuzjG/0ag1IT\nFDXzGDCwz6ywRiv587xszqtqTwabc+tuUuI2NSaJVi8vAMCFd38IF/+Pr3G06VD7AeFA+hkPNNmQ\n89ptaJnRIBcoWzU22mwiRwrWfgLYycLZPH/KUCcRg2g1r6wPqSyls8DhbiIoh1eEqFAj0KSbjHUt\nHEuX2BNbGHIjr95EDKrVkKHJdbtoEyE6c987cP72e+UApDrBltfCyf6d43XbF+WOx5GmYRhOvB3m\nS8HLuiFvbm5isVhgPp/jmWeewYULF3DhwgVcunRJl7l48SLe+c533rCB3kjwxN0KJ/Akgk5devMd\n+qIfMn10pXwhUNQVAIbgWAPasoxjDrNI4UVzkF7RSqmIODMdXHsaM4aT+aMoQiW6JnmTa4OzriX/\nUDFUPuiPKm8WrAWNRVh+vSSKx6e6qOK9NB9e31c3oYkoq2ma7IE9jRDzzhRy/W9gSdNkHFO3s7S+\n7B/QEp4++NEi3I6DW++ek3Vn+h912tJ66vWblG6bns8qsssPO8aMv4vZG7zcIUP+4x/n6FoN4Q/i\nD81XO/SqN2PdMo9d3a9gNTRsJikErVdH1Icw3uzZZSmLkDodo/qW85wZq6kLOoQNcTxnzgJGRUxx\ntJ1BRWZWb9Y8nn47hadf95afBgB08w0A0xvYyb4Jl1Dv9OImfFxHp83NzVuKSL2s4MXP/dzP4Ykn\nngAA/PVf/zV+/ud/Hu94xzvwj//4j9jd3cX+/j6efPJJ/NRP/dQNHWxFRUVFRcWtihdkyN/85jfx\nuc99Dk899RSapsETTzyB3//938enPvUp/MVf/AXuvvtu/Oqv/iratsUnPvEJfOQjH4ExBh/72MdU\n4FVxq2PMCsgclLUV5QsMXYepe5MJgAhxjLyyFCj0u+pStdkmZt1LFxsKlWbbZ5O5B6CsTR27in3l\nvsFkIgxHWw0/m44iG2E0Not3Yqa/sq9xWLgUKhGZRdrR6/T/5TbKV+6CX9ZYvDlpmXzdPsaG/t/e\n517TGsaerF8ch7LGKRPxefmmZT9jCcszHGucbl1dxQohGuefpikqQMPozxHWGJEx8HLN5H0Js9y/\nBtvQQ1oiFW7sSx9CKPpRTxz5ptEJlC5radlceuPX5zwUJVLK/FIYOVAkRt/v2Os5spMjDzYLA3tJ\n05y5N0Ugz77ubTIelhLeGuk3Y8xaqHpq+nHU9+1HGS94Q37wwQfxla98Ze39L33pS2vvPfLII3jk\nkUduzMgqKioqKipeQ7h15GkVNw1BSpeiWEX2h5LvpfXjMCAKa9bUYhwL6wCDGOVyZO6X5WghAhRP\nibq/89lpGgD6wcOKn7Lm9opcZdrXgEhxS6S1YGY+2j1nnL7T5KmNVnOifCbP4huaZ4RcPmXHDJNP\n+95nJhUwZlsREZECKTsWh/ncLqkQbzEqwBy97DMMeW7VmETqbWyDYCkw43khU+e8lsyf/6f/N20t\nI9xMytFouGIYVRDGWpjEqHCLJDwWufgp1FSlYPoTY5OSzWv0QSIoRvbrhx44TGUxzLdHki7OFSKs\nofiqGY+b/ZkjELWP8DgSEzzz+R6uGZ97ihfDAC2pCpoLH7N6FKVq7J3saPBBQ5yNHdz95ncDAOZn\npH+xaifyfNxqYFUPX29VnHABfEVFRUVFxWsDlSFX/PAgu+qll+xuKj+aFQ/+VLSqSYjWLjM/aZU1\nGzWtSLCuyRZ6U4arvawt9HLmY6afsEfvEaU5Rc7B5pyh7eX/skx0ZE3MczYwkn90tIGkgYMclw9e\nhcWNyQwMSF2r0lid2i1Ok6SxSFRTpa51lrb4umaKLOvJX7TADJmFq+ekzFlAVPtEHQcmudNiq1ko\nTFVwZr25SxXz41KyZsj4y+2t592zNSSZ+lqtlTLjqQlLqbjVVK+WT4nV6mxLmzH0YqXasVxKWbAr\nqAkNSgpFO5AaSNAqU90/xgp3HwZdRqUGZOp+yAp4mXs32ScMEFn2xGtGrulTd6ZexXe98UGYGfPz\nEz51CzLjWyk//GJQGXJFRUVFRcUJQGXIFT80yE8W15LhQTckD3PDtoyxMKlgOzthAJ42hBZgjjL3\nzxVG1hl0s3GdamBuT41FgjZqMKquFmtFk5dpyCAhTFkZcrZLNRM7Svb6hbNAKwyIFpxT83tjYJhH\npGp7kjsNIWhNrvfj1p9AzLnA6SfF+zpHWK59lj5otb4152ATu4/DChD7SG0dqCw0FH+Paz7ZU1cb\nY8ArFVS2prl5jmP9OEr1dm59OmbBuQ45rqm9pwzZWjvqZV6+GtfCz5OlottPzSgWV5NfAvsRh2aO\nIqySXqQmPkhUA87BsDXkMdUEJgZAjVYkF8zez+0clFQ4bRzB+ciTNZikHl7MzwIA7nrgXQCAU+df\nl47LuFuSCVck1BtyxQ8P+YE8vPJ9AIBZSrhtP7lp+RD0944/lIN2DpKypWaOpfhUc3tGHR0Cpvct\n/q3lVDEiiMmIX4khSU/jBvbxDUVnHgqdKFQygKGYSoQjvOlruRLg+IOt962JcMz3sA2NNLje+EZW\n3uwo9BqGHLKNcXyT1xuPz0YUDPE2nYio3LhEKgSjjYuCDjabhWhzJy1/YrlQ7sE8NcmYlmGlN1mu\nNGkgzyMNsQg78gYq4j/e7IrPCH3QKY5flzzqbx6PPnDxhgw1imlF8DW4ZCqz3E0Pjt32+SyimmUT\nmXTM2ckNQ3r48cfc/C287p9e3jpG36tD2ODoEjcW1PVmA3PxoH79m34yLSMlUtoSGwE5mVNxq6GG\nrCsqKioqKk4AKkOu+KER+hT+XDz3XQDAjAxzQ9jGQY/Vkt2dEih2aSWsu1gcZn9omn3MaOjg1lhR\naeoAJJbiI/2xhSFLv9imz+U/XN6jDM0mO8PMJNmtSv50LO3xCHIcVvrvxkk41QSv9J3WhpHs9Sjf\najkulnMMfb/WtZh+0bZgzuScau048c9O2rAJkyy6NlkpuWF3pTwylk/ZovvV0SHS9P66CCu9ZHGU\nmodM+o6n9SnUGofH17poFZimCUIISv5zKVRxOLRAnaXQ9UzO5+Ez/wYAGBYLNJssb+I4uL0cUg+M\nYuhsTYxTELMVLL3SWWrlV2qTuiGWsL2ICPvuPADg3BsfwqkLiSEbjapw//yprhzqVkY9uxUVFRUV\nFScAlSFX/FAIIeDapZQ7boZUVuLFDGFFW8TWoKMRvuR3aSKylCYArrFou9TndsqSYHIjkClj4vt9\nv0IQEZeyYP4t+0QI2uABtCtUqpnz3Fpd5ISJiZDLOqcsU5pPKetjLthEwGpTYeaXZTxaopVzgDxG\nssambdEv/egzzocrjEI0d04jCT9ZBy6vx/0VdpDKMjlUbc7LYzXK1K/f0W3MKEujF46V0Yxxzjit\nrx25pmy+OM/Tc34Ua2aeOOdzhT17g8C+xdJ51cp1Fiw7hdlcjsd5UcYu8xqGfM2IgQ07O3F+Y4zK\n9BnNiJqT9wjCtnvRTbS3vwUAcMf9ki+eb2V7US3LY4OTitcCKkOuqKioqKg4AagMueIlgixBntmH\nFfae/VcAgPOiQl0mtbQRdbNFxCDP+N1mYic92aukyHz0aFb7ab025fq02QSMKntjpm0AoLnpoR9Y\n5ZTLjGjAwJydj7CdMFnP0hNyD6Ps15MdYVxahMEV7RuF+ZAJqUFG1HIr9v3VfKrOYW40oDl1qs+X\nK90HzSl0jHw/Rs1d83ysNYIIBSEmo9T1kc+fGefitdxoGFRdbiY9i3XMxqpqPfhJxIElQrbBalBJ\nfFq/yCFTiR4nOeRQHrt6dIzzzU5Lz6Ky1Zz3z8dq1alFmOwgefvNcwCAxd5zmJ2hJSOPY1wGFr2H\nUSU7lxnbY1qY1Pi4nAg96RG9SRT91OvfDgDYef2Pp09Ec2FNTKVTaeLG26l4TaAy5IqKioqKihOA\nypArXhJojM8n+d3v/hPag2fSZ2JR2EjOjux3NaxUldwz1ynsq52nmlCDrNZmjW1L8woUxg9kWSuy\ncD5TWhhhYsNKmLpYT8Yi5ahN6dXaUda2BlFytGRitNl0ajIRc72tMKggCm7vaJHoU8EyMutVgw6+\nb21uPkCGSUYIqMEJG9mrdaSZMu2c1lxDDMXckO0Vz99Z7i4TwKYGoogOvmiBSKasbyjUqlObZEg7\nS6nF7odlEdVYZ8iaK5Vlhjg224ixsKrkntZU1kUdMuuqQfbs4TQYIJESJyr87RStsf0eBrF9dWz3\nJ9eO1Rx9gJP6X9q3ZsMV5ovzNaRtJKPUxrsZzt/3MwCAU3e8kQOSffDIzLH13BWvDdQbcsVLAn/w\nVrtPAQCWz/6/aORGuBrkxyuw3CeF6Lr5Fjx/8KQshMUcKqYB0IkJgpv0QLU2OzotF3LTXiQxmHZ2\n8gPskH4YG4Y2+QNMYxAfNJys9hd0zAqAaVlmNXbjUs/iomwq0KhBQ9VHzBVfebPVkGfUciXemF1W\nRakoLDpZT8wq2D93NKZhXL6VHSRMsegk/HpESdR0We89go5tXII0XpPHRMFZGusgD0wh+txzeVKa\nlF75YHL0TZdjSds2a5/lUYxFf1MBmWwoLSNPMWGeOyk1PA65PngOfPE0p6VYskwzKWPzgLq0qdiu\nS25gd775p7Fx5m4Z7JpFV0UFgBqyrqioqKioOBGoDLnieMSs/VH+sroGALj07/8FAND6pRoeUPxD\ndtYI010Fr114XNPKImO/6DAMiML2yG3UfCMslNHSStBsCCM7kFKra1ex6snKyNoo3JJwch9yiJUM\nUqiubZz2IdbQ8iSsHUxmtGTIZNEU/KRw9LgTVZiG6dsWTsVUNBGhOGqFRuquVDdHRiaipGG5ytop\neWWJF8cOY+AZquYRFKVRGm614/Pgix6/jL6yAxH3kUVdWXDlZMyrhQjg1IQkqk2qsumyWxL/Oyl1\nU2GeNcqeLSME1NNpNMAiyHwOkSYquWSOoXqyZ6YgWKrWx4Z6M71WLEvnNNxtcvCBYfUo16ec+8EE\nzKRMapifBgDc9db/BADotnYKwdskmlFRIagMuaKioqKi4gSgMuSK64MP8z7lbC//y/8JAGgPUsec\n6KMyOWbbmm7cMCFGD6vGHhRnjUtYbNtlI32tcWJp1ApBzDJoAJG7BEnf240zQDduLuGEgeyvUt55\ntVqhGWijKGPU1K3R/LJTdkaWlfOyRhmxbIe9ik1mZIwYsJduI52qXFG2lDfJAeS5CiwzIvuclBs1\nTVM0zoijV1K9EDKbVyZGS88Qc36bHalok8n3jcm2nJbs2+n63CeZPg1estApW3A6Efnl60HOeYgI\nLB9rxjngbAFqso0l8+USnhiY320MvMRVNN9ddF2KgWVwnCKK9KTUaTYDBuoOeD60NZNsxqj+jfly\nMmPL+QkeOPMGAMDdb/7ZNDRhysaWbLgy44qjURlyRUVFRUXFCUBlyBXHIsYII/myK/+RcsbxyvfS\nh0PqLWttQ3IBy5yclIz4shkCDSuEVQSmm0WR7ZpGmRetCGmw32OOuEi56/4g7Zc9Zalq7Xur/X4h\n7y0ln2k3tgAAM7eB1TVhy5KPbJqZjNEV/WnNka/WWlVV24nCNjdg8MoOmSfncdBGMYRec5XaT9jn\nXtBk5uwZraVRMtEmRH2Sppp42sDB+pAToETZRzhO8pjM+eqiQcu3tOxIgwB5Pg5F9c71tOnHhLGX\nY9SSoBD0DxPH86nsN5p0LACMqKK1hafP5ixG87LynlyDxjSaV9dogvQ6tipWCMqEdcomzYqttZm9\ni9FMsImF+5DmfOv8A7jtbT+X1pMKA3OU/L6i4hhUhlxRUVFRUXECUBlyRTZs0CQZXyKuXfwPAMDB\nD/4FANAKExlcyvdG3yuj9ZJ3yy3syIptYdwwsXokM/EerD6Owl7JgPo4aK6SjOpwcSD7Yv1rByeG\nDzTL8MzxiXnHAAM/T8zFMQ+qh2xy/a+loQfzqjwON1Ixc444fv4d1X6RtpA0xGBeE7nVowzAUP0d\njbK9XPc7rn9FiAXzFGbNpgRUVvuIyCgCGa7kkEOIRQtBmRsyUjZ78EGNTVhTS7bHnPDqcJmLi1nz\nTTbLHDkKsw42AqG1agiZvZLtUsXOKEOImnuWXhBZ1c8ZGAZYcP5lGgfZR5PTwdrIIvdqzOOyPMcY\njZ/qedh8HK0ouQexwty4400AgPNv+hlYmZt1C8yKihdGvSFXIMoNQEuL5Af84PJ3ce07/xkA0Pp0\nA+y1i4+ESLsOK5pAiAjITm5axtiiZGRc3kIYk0K5AHR72uFptYBhuFZuwK38ci4XadmmsVjJ+vxV\n9fJKkwrrWv2BVcMH3iyKMG7UMKWUwBQhWqslTVIWQ1etTvY5DDDNuPxLvSWGbHCRH1p4A83dl1gm\npXbIjPQGGq8EhANxMYvp1W6eTbtQd7IIiKGI4R2JDwTR5pIhCrf0GpBjDWHNF9pK6H3JfSOqWEk7\nazEU73kuyge9aflRFuDRY1xvYAzhDzkFwHo4zytVBXG+6A7Fh6DcpUlLqSzng7sQIZjtEKSzV6N9\npWXTRXkez9XSJqHWbfc+CAA48/q3yTqtPrxV3VbFy0F9fKuoqKioqDgBqAy5Qo0TrFhM7j/zbwCA\nve/8X+hC6sCUw5YJZY/bKRlg+NYVpSxT28P1/rceRuw1G2FE/aGwv5CtM2lMbCVk7qT8Z3Ht+RyS\nFPMRhj3J8BZDDyv2mCwl6umJbQysS+u1zVyOkeFHbmaexVvCxtnRih2uhuUSndmQbTLEPGF0xmio\n1svGXTE/QQv1AAAgAElEQVQfDNmT9ZFxU8pVzjftNKPMA52QV6teQ8y0hcx+HLGg7QzTM3Ih2y3O\nGce6IvsuLEA1xkvWuUrlT5ahcGvVNITM3yk5z2YwGuHtZM557gafB8VdidCwaJWFaFhSJVENm+dz\neu3ZiciraRoENsGeeFHz3A9oYDYTM77zTe8CAGzflqwwvcTSKymu+GHxohjyt771LfzSL/0SvvrV\nrwIAnn76aXz4wx/Go48+ig9/+MN49tlnAQCPP/443ve+9+EDH/gAvva1r71yo66oqKioqLjF8IIM\n+eDgAJ/97Gfx8MMP63t/8Ad/gA9+8IN473vfiz/7sz/Dl770JXz84x/HF7/4RXz9619H27Z4//vf\nj1/+5V/G2bNnX9EDqHiJUO0QmarRkpvnv5dKmw6//48AgCYsMUizAC/GFTTUJ7NM20ivmvOVnQzK\nSIwmqK2KdhJUqOSDdnBS4wkZ17DsYch6VS0kZSUdRWYrGFm/F5OKIOOJRkxA+pUKiyii8jTYCAGt\nLKcWoJovV/9EfU8ZoTA8mn6YYDUfHNhPmXliyUc6a3WbuTmGbMfYLCQqypwAAJyDaLJQTO05yRDT\noq3rsCT7VwvNoOszh+00p888r+Td/aAlVKsFc8aybNkcQUuPZFpU3JaFU0bGxj7TKogL5XppGSG6\nuSd0DGtNHQIFWyrSimoIkkuqWDqGXMLEqATNaJj3dw5NU55jYKBRiJTSnbnnTTh//08AABoa2Aia\n6oRZcYPwggy56zo89thjuHDhgr7327/92/iVX/kVAMC5c+dw5coVfOMb38BDDz2E7e1tzOdzvOtd\n78KTTz75yo28oqKioqLiFsILMuSmafRJmdjcTD1svff48z//c3zsYx/DpUuXsLOzo8vs7OxoKLvi\n5IDslQ0YzGofz/7LPwAA+udT7tgxL2o6RGHGZICrFVW8OR9npoppskWbc41soqDmFjSwYJu7wcNK\nLq5fitmE9ggOOZ9K5bSwYM1XzzewoNGDEFyy+JWoi7sGONhPOfGGRVbMI7oGs410XTsqqaVJZNC/\ni+O04964ZF0RRsekamJHBTHNQLKxR2ReViMXTpW9DZnkUpYRMhpcp0zUalMFYchMtSMqMWTuV1Pa\n1mjOGMzhsmxK2J+FwULmympZ2/jY3ayDHxjVmGgDmLf1HlHOZzbWmMkyVhmpl17YTNq6oiyLxiiq\nYGZUgSzaZLMPttyE41wbNfuwxfJpHFJO5Qy8/BQugzDiu+4FAFx4Qypp6ja3ES3HPdZTVFTcKLxs\nUZf3Hp/85Cfx7ne/Gw8//DD+6q/+avT5cf1WK24u6J2soZHZKdzx9l+QP35hfYWKl4Qf/9CnbvYQ\nbnm8/X3//U3Zb70BV7zSeNk35N/6rd/Cvffei49//OMAgAsXLuDSpUv6+cWLF/HOd77zhx9hxQ2B\n5uFEHTxcfQ4A8Ny//wPMwdNcCADgSSHglFX4oi4UyD9OMWpFK5ywXj9R7iJEBLF45IPAIMyKOczQ\nDxgkj+jkWU5zlw0Q+ZlscylMkIzdWouVvNdLXnifDK9LrGf30g/Qzbdk/GmsrTCp+cYGtja20/JO\nRwkAaKQNpHOtsjzKgk3BCH/i0U/hm3/2e4W5BFXKVP62+r6dKH9R1G4zstCQtbIRh+SQeziYVaoL\nj4OwTzlGKzng0C+VmdOmc6CZikFumiAKcNuk9ZtZihIsrl0dW20CgGyPLNR1LYaeimdGA2T/nq0s\nC8tKWYZtDwGn7TC19aWorFUJ7Yfi4T7i7e//H/Bf/uf/EUBWqMMaPQ9qHcoIBlxukdnkCAUAmLaR\n4ziLzdvuAwCcFWbsTp1Lr6pGN1jTFFRU3GC8rBvy448/jrZt8eu//uv63jve8Q585jOfwe7uLpxz\nePLJJ/HpT3/6hg204njEyf/Ujxjph8oYoz/c176bHLf2v5/y+y4eIiKF4gaTw4QA0DmHXnvoZreq\nYpHCQzh7LvPHVN2rQkQjDkaB4Ub5AT7cTzeJxhgVFlEMxn7KC79CK9tcHOyl/XaptIgPAcMQ4Wbp\npsJCLN7Arl5LPthdN8fe1bT+6bPygztLx97MuuwaxVBzy7B0LpOJ2pVp7FqlpTURgKfpR5qX1o0N\nMWKMakChHZWa7Bplaccc2HcXsl56NU2+OdCQQztcmbzv5VK8vEWs5kCnrEP1FFeNnE1zt9hPqYB+\nsdLypNwsit2V6M0d1YjElcePfPONRUcnXl79im5eNpeWqYkKBXFR/85zPXZH005ZyF7p7FrFnQWX\nHyrzA5Jsb+eNAIB73vSzsI2UqulBU4hXdt6qUb+KVxYveEP+5je/ic997nN46qmn0DQNnnjiCVy+\nfBmz2Qy/9mu/BgB44IEH8Du/8zv4xCc+gY985CMwxuBjH/sYtre3X/EDqKioqKiouBXwgjfkBx98\nEF/5ylde1MYeeeQRPPLIIz/0oCpeKijMGZcUxdBjcek/cOrCA7j4f//vAAC7uppeaS1oZplxkAGI\nMUZfdOpR8qzGxHnvLIlieY1hWHvIZT9eQuXkGKyWYSmL9wOsdMhhb1yGJGfo1Ne6m58BAOwvdtOi\n3am0jnPwg5Q7yXi6uYRxD4RdH/ZwcsmvDhJ73D53Tg7HaTifbM0Is9QyGXhELiNsk+HgmFVZWt5E\nERKZv2dpjzHqx8xjNCGfQwruVHDF0iBhozYaBNn/UoxBWjkhPrIkqNGTthTBFMVqrnEYliky0c7S\nuT7YTVGERkLXQwzoJxEP9qLWLlS+V39uCt9ovUmiaaLPNpaTsiljYl6OAjaJLliWYZXRYS4jc87j\nSR+xrk5+0iQ6Yl0+R0xTzO58IwDgDvYsRou1cADD0yj/rKHqilcW1TqzoqKioqLiBKBaZ94C0JIg\nqXlZXrsCANj97v8Dv/sDnLrwAGYHPwAA9DSnYL617xHF/IP50OjJAg0yR5Z9URBTCFyclphA3pM8\nILsL+ZWKfNgRiusMtLA0jZYAzefa9ymtj6C5UrLXmTDja9IfuZnNEIQJ0k7SC9vaktTJc4cLzeN2\nk16/0Ub9v5kwIeZAY4ypCQWgPZdVuCYdpUyIGnGItJykiYrma23RAYodsmRnDbQTkxchHMVVlpGD\n1QqY5K6Zr7fFPqf9nPvDxIIHY1W8dU3y6yxrC70wXpjCJENy2cI6OWY7MkERIVpLjYHXdfWaYaMG\nOVRjTC6Fmgil1Pa06DsdaCqz1qfaZGohc8TSOwugt+laOX/PmwEA58TgI2rjiLB2zisqbgYqQ66o\nqKioqDgBqAz5xINlNlP1Z27QSkP/K9/757TG5e8AAKw/VBZxKA0PouTa2JIvlQSNzf+t5gh9NhCZ\nsJJsBJHHGOK40YEy5FVmpqq8FgbTztO4liuPVlrg+SzhTutbq6VLg6iJB+ZOhenv71+Di7RfpANG\nehlkO6dOn8b+kHLojZS8LA5SadR8e1t7FLeSy56y8hhjNrLQnLGwV/aC9gOguXy2/RvnlIP3Wgql\nbhucT2PR0wazZ3tAsnnmUINGQ5iX1TaQNp8XP1CdLPnmw3SsG5uncPVqysFvSH9o5oK9RCkCLJzk\nk7XVJFsbMhISozJhvTyzgkHnTK+Z61hMei2TyvaiAGCcU/V+0Po6+Q+jFCa3xVQ3FCnRGuwG7nnr\nu9Ox3nbPaHXtJW0scja6ouLmod6QTzDSDYCCmHHnoRh6XLv4bQDA3tP/mj7rU/jWeQnd5sAfjGF3\nIjpjpfedbbKMRW4KAxsjR1PsjzdJhnXlRzIEvdnyhkwxV5Abs18NWLEOmaFRulfJD2m3NYPvU/iU\nzeUdBVO2QVxJCFT6Hw8SVl+pZ7HB3rV0/DO5oQ5aPivLNA226K0uE9BKeH21fxV2i117pOsTQ/m8\nIcWooVntnSw3MPZXDr7Xm2wz4w1NjiPk8ieKuaw+cMmDhm+AnuIv2W+gJ7V4KBuHuEo31+hkH6xV\nFkFbGo+UhEnku5Ub7N7VKxpGXsmdqJNa6+GQdmd9rrVuOf70SQ5dD2oNxmtoWKWyMn2QK4Jw2RM8\n12BrvTEXYmmTCutsEU6W8zgRdxnbZNGhlGSFJqUpXv+2n8bmmdSVKWiJHhd2xb8VFTcfNWRdUVFR\nUVFxAlAZ8olGTDZVAKKwz4UKtv4JZj85bKnblQqEys43OXQIQMtlXEujjqUyEIb9SoEQy5IY9uQ4\ntGtUzMYPKn4iQ6ZPNZx2PsqlTdKX2LJMJsBKuRWXXYpJRYxL9AylyjZXLLWScQwIaGYp/L23lyIE\n3XbqXxttNiOhYCx6ltXIGBcLuM1Wj7tEaQOrcyNlOQxzM0zuV72yVDpSxYJhcxvcB4Vo2lfZBFhh\n0oNPpVmtCsjEuSwEDIdjgxT2svYSORhsLgVywqKvXk7ubDQlAYDeLmUcEt1os2FJmJbDTa4l74fs\nVDb1M6fgytgs6rLrDFlFXcqItV5K9tjrZ/Tt1muxDGvT3exMctq6560/CQCYt2eUvitBrvqtihOK\nypArKioqKipOACpDftWRuwIRBuNH98xHIuJ+yqs+/9Q/AQBWz38fANDGBQbm0PzYJ1oNQkLM+2MV\nSUuzCzI9r7aYA/Ox6kccshlDHIt1aHIRBp8ZIAVbhQEGAKBp0ahRAzsFifWl+EXb4LX8imVCLMM6\nvLoL06X1l8IauWx/kFj0YCKimEI0XWLfSzG3lj9hbVTzD4rBOhnjam8fXpZv09C0V7FaNoaCKTPR\nzlfJ+64ODtHNxPOZ4jYKr1jO5L16SbNrFTtsxcYg9HI+eKzSrsjQfhQDoojbGpmHEHk+xIClc2hl\nzvb3koCLQim/WuV+yHIN8W/ua77RZXEZc/HT0qQQ0bJUTo00xv7dIeTrSSMwjLIYqAiLJV36ytyy\nibq+iuNcZsYA4G2H0697OwDg7vsfSvulVsFEGNBvm+V0FRUnE5UhV1RUVFRUnABUhvyqgeyKRgpO\nC0QaJV6iJF6lXOHu9/8/LC5+DwBgIQYYzBWGbNsRJ6VJauyPnL9rpM9tL00mSHYConbcUS2rqpNb\n7SnMZrwsjxlW0rjAB22mEAqrTSCXqXhnlfHQqpFKXSjRDAhy3FaSfXtSprOKDn5JBinsk2piNgrw\nEbET9n0qsd/VNclBL4WFtR6t5FxRdAoCgGZrEzOhxotF2u+Mil0vJTQqPy9U56IeNzIf/eEB2vlZ\nWT69Z0TuHUQPYI1Rts3+xb30PA4bAyBzrvl65pslF9z4AHHIRBD214siu0NaZoYW157bQwnadcbV\nACu2osy9GjWXSX8vD1boNtN8sNkHy656yFhjgIm5q1KaFzLkQi3NnK9jAwqyV8BJfhwzeZWXrIee\n5VwxdQ/SCMPJPN/z5p/BqfOvS8vIhd0omzeAzEnNHVecdFSGXFFRUVFRcQJQGfKrhrFpvcWgfW69\n1N3uXfyP9PrMvwMA2uUuZqwXFSZIloMYiz6ztCbM6tW0E6NK36l9Y84JI9PlSXc5v1qhlUe2YSnG\nEcIEV4eJfRpn1UCDrfSYS2b7ROtcodIWttgLe5Q8Zb840PZ+UztFO2uxokUmGz70aTudtM2LDbCM\n3G8ax8Z2skxcXk1mINF4zOdiqaj11aIkbhs1JAnSeCLOE5tfFnl4N22bOND8RNTK1qgiPMrzLvOz\nQuoxDIP2WCazXMixd0M+V2pByudmqUf2Puj5I/tuNtNYSar39vfXro+B67jcc5nmGtpAg2w0BCyX\nKWLRyjzo9tges2myToD9ldnWkXneaPI2VXktizqHRvLccaKytlzIBu3VHIyw/50HAACvf+t/lcYx\n3851yBUVP8KoN+QbivKOdnR8TG9IqwNcu5TKlg6eST2KIWYPHX94MVNBimG3pGDWtjV9JRrXqDDp\nOBhjssuRhr7lzzBgtZDQtPwI80ZM8+UBQX2Y2Z+WN0T2Kk4hVz40yH7DODzfrxYYfOGIhSziCcFn\noZmEiFu5MetjhcsGEj39suWmxxKvYcie2o2ImRqKiLpWS7z6A+mORHGcy2HZ7LYlYjTZPXsPB5sl\ner2MdeC9ZeAD2KChYQ2Zi2NZvxzUO5o3ZMfORyLyCsV6lt2VZJ+HMp/Rh+J6SK/sbd0g35D1iuHf\nNCxxJq8n/tbzU1tpGCJWSx7ScrPVc51eeE00RbrCaDhaHopsdtiiyQf7NPOGHF2LYJLv9h1vfCsA\n4PU/8XDanpTQGfhiBioqfnRRnysrKioqKipOACpDvoFQpopcBqKGDcvEfnefSWHpxaXvwYlVpIv9\naDvZdnCVGWDMnWnKfU3/D6RQIoAsyEobTdtU4ReNMTLLiSyd6SkqWiJIOH1YMqSZWKe32YNZw58T\nZqyjilCrykDjCpYEkXl7aNkRoSVSJtIfBZ6MUpgVWaiLQJSxke1pZ6dTKXQ9XF5htZ9C352Es32T\nvbkDKZyE2g/20zmzYpYx2zyn7BB9wVZRlg0FLSnTw2dqQkxAbO8RWmHBErI10n2pX1zTmYvSQYrC\nr9hIKN2a1O8YQMfe1WKiohGR2Bf+4ZJSGOQ8uwZRSoAo6NMIMWPeLoKsk6kIlsW1m2mswzBoORvL\nnNSDmkzbZmbMCArD2dEaDVU7sVINTM3IdpqN83jdW34KALC1kywwWzGVyajsuOLWQGXIFRUVFRUV\nJwCVId8AZGbMso6AIOz3+e+nBhCLy6l8qe1TKYqNvmjYMN7OUdu+3n4JCriYgzxyXfUPpGWhRWR+\nmvaUZH/DoYqXyKSCTeyknaV84mq10pa12QBCGBATiTHqdgbmWmV7zL0Ow5C7QYxKVpKZydZWYmX7\nwloJFVkVTJ2sbbFY6DECwPb2Ng6kJ7BOR2njKMy0ldzz4SKx6dVBihJsbu6gF3EbxVxxYurivV+f\nf0Yl2FfYe+j8g1MkAirXYO9AjDy0VzO/piz/cXASDVhITl/NM3g83mLFUir2U7atzI/TUiKOYOL5\nAR+NWm6Cr+x0Jbnozc3NNdtUbofXgHN2ramEYWMRZ/TcgI1EXMqln73rfgDAnfe/A7ab6AUqKm5R\nVIZcUVFRUVFxAlAZ8g+BnCMVde7uRQDA3sXvYHGVjR9E2cpcshr2D7Cioh0i++USmbVdjyETZEer\nMmd83Jhp8BHJfJP9JQDtvwv2xvUGKzFVMDPpm0u7Tnmdb21r7hlgEwe2IpTjO1xqS0XmOL00M+gP\nl3oMbByhJT3MS7pOzU82NqSU6WBsemFHuXAyXdqESv7cOX3vUCw3O8lnOuTmHCR0M5Y9XUllU4d7\ne3AsQxP2TaY5yNz1Q4+4GtNNI8zSyPb9qoftpJ2mMlSJLiCrxdkHWJm1KMu7eYvlQtTmtPWclL4Z\n0yAEagFoQwkZs9FH8dzreRytMbbVnsJmYnuqJUmI2NhIc8RoAgX7TtXSWZlO21bQAtQaRLL203cA\nAO657ycAAFtn75BxWP2iNcX4KipuRVSGXFFRUVFRcQJQGfIRKEkp28upmtZEGKn1PLj8DABg//J3\n02f7lwEAbViiNWM7SWW6xfa02b0Rhqn5TDJFu8aQy5wn1dSDqHdHZh8AgnWwYD5Xmklo20K26xuy\nGldyyGR0wXW57pm1xmQ+DRsFBHTy/yD5VSjjzg0P+p55VTaySIu2m1tyDD2smFRQOZ1rY0O2XWSe\nVzpGUAnddq0qsKnmZVvKyLM3azFrU0vGsEgMmSwUjYMhi5e5GWTuGjLl5S6s5L5pFwohfUtRjy+W\nC2x0tJFMry0bHaiK3CNoba4wWlnfOJebQETWBEvNtLDqRb+AlXNPPXecuLoMFhikAUU3F1Uy1c0m\n5haTk7wuVc+hbeAkd2sb5uZtuQqsNWpUM5fIRWbacg1YB4juwDj5LkhO3G6cwe33/jgAYOeuHxsd\n6wiVFVe8RlBvyAVyz2CjP/yNhGP7/dSHeO/Zp7AUgVYzpJsBOxYxNOdhENmbeHpDPgpmfCPNjkjr\noUjCWpsdnXD0Tdsi6o3XyI2Y4c/BF4Ifhl2lfIQlRTBWjSd4k5zLzdch915eSfiU7txLCeeW/Zkp\ndKIRBEujdKzOwbVyc5Ayo17W6QYPZxk2lbFK2VW7MZPj8XmOpi5cRSibN7Ig+1jupVDrxsaG3sAs\nw8cy1k7C9cv9PRzspXM+b2dybHQXYyh7mdcT/3CWKHk+f8FkZymaf7BULAYNZ1twjtLCS3ngcV2b\nzTa4GXW4Su+vooeVMjTdXsubeNQbL6vf6JjFyXONWxPp5de8LzNJE8zEq3wlZW6wVsPq3qaHr9N3\nvwkAcMcDb4cV/3BT77oVFTVkXVFRUVFRcRJQGfIIQq36A/WVPrwk3ZaWUq7kIxwSUwn0PmZ1Sw5s\nI1tOX0+UNf6My7KsxBTPS2pywb7Ay9Vae6Zp6UoMAwwHJ+Fo79dLcijUotVkbgUV4ehJPPEhjsKw\nEYKGugcZN5nV4qD0u5aQuR2X5/C1bVv0ss2mEyGZsnEL37Pns4iApJ8xQ7ZD36NTH+b0HsOfg8+G\nI2R5XN9dS+c1HCzRbkjZmITXKSqzoLiswVxETJd+8CwA4Ozt51GidQ38itanEsqXcDC9pLvGKjN1\njueBvZcbWOlIxQjDgZiZMErgGqfMlGkLhunVLDMEdOJvrWI3zr0DooSf6TcelWE7fZ/bphBuWt7m\nnCuEdOkznuf5PJWpHQSPePYeAMAbHngXAGDrTJozY/J4pxGgiorXIipDrqioqKioOAG45RgyxS3Z\n/oI2lMIio8nLSH6133sOAHBNbC2Xz38fzh/qFtOyofiTOV4y2TBaNoRQPPGPx1OagChL9eM8MfO2\nvih7YsOG1fIwH8/Im7JgR/KfEILaJmrDAs8OQjLWWQMYyVWyPIabtRFOGFNLMZmyYWGBJueHo5Qy\nrThm6bk7rAY4lrpopyCKd/Kxd9LVR3dv8zqMFkRRg9mNxMCcbGc2b+Cl13MnjRrYrCKf+zznRrYz\nl2V3n7+CLVB8xIYHkq+WMbtZhwPp6tSJUcmeNKJoWb5jO2CWu1QBwCDWqFb7VBfGGTT7YG6962BF\nsHZ4TTQKzo7WNxiUpYK5ZzMWxDkYROaMyVolpw0T9ZrNgjHmkCHzMl/LITvHblGyUOO06bVT+pzO\nYe/S/Nx9/0M4/4a3pDFSwLZW4FdRUQG8SIb8rW99C7/0S7+Er371q6P3/+7v/g5vectb9O/HH38c\n73vf+/CBD3wAX/va127sSCsqKioqKm5hvCBDPjg4wGc/+1k8/PDDo/eXyyX++I//GLfffrsu98Uv\nfhFf//rX0bYt3v/+9+OXf/mXcfbs2Vdm5Mcg6jM3W9fR4lAYVr/EgeSFmScOi2RV2IbEfubw8Bgb\nLpB2eqWPZX54rKQu2e80hxyPYshhsgzV3jHn73pRABvNmhagHeY0Xe0DmMzWLCrzrGq1aFEakQA5\nL2lsBNRIRLYjjSd69v810MRybhsoDF/G023MYeUzMjg7UeeGIl+tw/dsdNCujY2q5k7yzcMwKEtT\nFkyGF1getspsT9Zj/ruZdWq12QlDZVOJQcvCBt0/TqXyqeeeuQQAaOeMkng0bj46RpqhRMllt7Mu\nW0ZSuSxstp3PsC+s20+U+pYGHW0D05C1jlkwj91aq+vxGtLvhDOqlGaenvPIc9c0TcGM2RpRTijZ\nuWthLUuZ0pxtXUi9iu940zvSdja2VZPAHtSVFldUHI0XvCF3XYfHHnsMjz322Oj9P/qjP8KHPvQh\nfP7znwcAfOMb38BDDz2E7e1tAMC73vUuPPnkk3jPe97zCgz7ePA7z/Bef5BclvaeSbXCh5e+r+VK\nLAliiDhoaDNoSI+uU1Fv0OVNI45ey5vvseVONt/opzf76U28aWf5Jkc1VcjOSlmYdbRwzLLYF9BG\n9BRwqXuTD7m+1E3dnqKKwOhGxptKEOHTEHPZVO4kJduR9/vgtRbWR97cKPLKpTRT72cnN6DguvRw\ngSxMmja939iY4/BQwunseMS51vrq3OOXHYdmmxtyfMDh8+khg0I4vjo9jqCOXPPtnfTZpXR9RT4w\nueMFSrwmnHPaP7iRMD3Lt1Z+KPpA8xqSw2E3rdlM68CtsWvbBtINmjdijse1coNtDFzDsHon42hG\nr8aY7I+tDw8yVfTWti3i9m0AgLvvfzsA4Mz5N6TPJA2iNfYAIsu4jpydioqKF7whN02TWYHg29/+\nNv75n/8Zv/Ebv6E35EuXLmFnZ0eX2dnZwbPPPnuDh/vCUFsB+UHottIPxs796RX3/+SrPqaTgDc9\n8us3ewivCfzq5//yZg/hhOGWk6lUVLxieFnflt/7vd/DZz7zmesu82I8mF8eorK1QEGMfuTR7yUD\njytP/SsAYNj9PgDA+WUxNrJfOmSVgq30nylbLQ068nvFmAoc5UE9ZY+jI5rsg+HD1WpV7FeYZMih\nyezORNaew79ADo+/6b2/jn/5X/+n9NlUZIbCLSqOmSFi7uJDv2uKxJbCOmOMOhFuwtashiqjsq1W\njm1PDDbUMMpm5kVez3Bq8DkqkaMH47IpAFhJv99Byo7YtaqoyYGl7zbZpuxsOfTwh2m9vauJ9c42\nJKwt6y8XKxWDmTaZXCyuppKkvUtP47/50yfwv3zqQ2g3NuT429G87j+XnNxOndlGdyaJnrY2aaQh\nB+Ea9NL1qpfQ9ULmavP0mXTs804ZvpFG0cHTAS11swrOYUYhnUQz3Fyuoa5DJ2VJ9JxuujQOnoMI\nr+Fw9bKWqIpvUhTs3L0/jrvue4sMm2HxKf8dXWmoqKg4Hi+57OmZZ57Bv/3bv+E3f/M38cEPfhAX\nL17Eo48+igsXLuDSpUu63MWLF3HhwoUbOtiKioqKiopbFS+ZId9xxx34m7/5G/37Pe95D7761a9i\nsVjgM5/5DHZ3d+Gcw5NPPolPf/rTN3CoWegUJDBtpfTm8Kp4Sj/9bQxXU5jcqp0l2bQ6dWSGLDlk\nCp9KtjXNAR/FkNdGeIyQ66jtTfcHZPENewSHELLnMJkG87uAWkPy2NZKq4r9UUyl+cQiB8uuTJwX\nekiB2SsAACAASURBVC5HHzQPS9tHklVb5Cp5FFMR0DDQk9qqMIo+191c+in37OdrNB/K9fuQ88w5\nR8qevLTizDaOLA3zwoytpSZAcsCrFYxEQ+jpTeXZRudwbZm2xS5Pg5SY0TClaRoEMS/ppKTJbic2\n/NyzMr+N1fyuoTEIy+NknRAjWrHeXLKMqxG/advAOYlGSCnSxtkzo3F0G3MEXgc9zWTY+zm933Wd\nXiuNiNM6KUNz3RxGctfUZ/FvI6wYMcLImLxLr2fuFsHWG5L1pds4rUYix6Oy4oqKF4sXvCF/85vf\nxOc+9zk89dRTaJoGTzzxBP7wD/9wTT09n8/xiU98Ah/5yEdgjMHHPvYxFXhVVFRUVFRUXB8veEN+\n8MEH8ZWvfOXYz//2b/9W///II4/gkUceuTEjm8AzIdofYP9Sygsf/ODb6bPDlPNrYw876fBjhCkH\nKReKWFc3azOCI8qWphaPL6ZHcWkMcr1cNEHR3LSfcblMbjiRldNq7A+WZMm2uY4Pyiy1nIW9eQvz\nkOALRgxoA4o4BAQqa2c0oBjPh3NuNKbpZ2l8mb2TaTP/72ZiExl6WJ0j9kO2XHitwQHHrJ2qvIdh\nVyJJZ2q50oqlQAaBJiw+G2gAwPJwiZnkfoNEE5pexmhpwRmzVaUo09t5+vv0ufTwGV1WLiOODVc2\nm8R058ZpLp6dpNh7OMIqs5+fSnnegT20PRtC2ELuIFGERtTf0qwh2gaQrlNqE6oWqTMEdlwSgqt9\nroUNu40zOHvX/QCA8/fcL9tJ47H5CtP8+Jpva0VFxUtGtc6sqKioqKg4Abi5NQnRi9N97ura08if\nrE/61u4/+1R6vfgdNIukpHbSc9UqY42IZsyypux3pPpc+yzBoFAaT9hnwDo7fjE54zWDkBg1Zzxl\nxtfDoAy1YNDqT7KeS8456ITcHKJgw4G5dHlh6XHM9cJqYCGt/OByA4opQw9UrbNvLgwMIxfKfscM\n11qrTJv71Fy098pM83zSTlIOzHu4mObR0gijFYbZ78n2DGJg/XVacUXjk36FRloHOpkkdyr1+D08\nTErquTFYyQ5tmzUNALB9OqVwWjRoZWxUJ8NKDlrY69CvNFfbMkIghxEjABk/leRNYE5d5i56tNrH\nOTHj1kkbQ+aCm0a346Qhhj5/G5utLpt0jKdvfz0A4Myd96bX2+6EJWs342vIXOevioqKl4+bekMO\nsLDI4csQA4ZrSal99eJ3AACryyk8TTOPNkYNm6oIqbj5Zp/qo1/zT9/1w8hmssy0tGn82frf1xN4\nAdLdSEKjx+EoAZkahCCJg4D1m3/uzGR1PjTkzhtx4QLF9Rl55uORDyFHIidmG6BwyWYxFh2p6FVs\n1Jgjrj08hDA2BlFXriPec262FnpXP2TOgQ+wZhyyZykOy8hsCBrGDsP44aVpGjUf2Tqdws8szeLN\nywagFeEYQ/+DPHzMxVu77ToNOdM/o5Mb4hDSA8N8tgW9Oco8NpwfY3NJk6UATTyx5cHAWKPLuBlv\nwHylm9cMzrIfs/SMlnD05rkL2Ln7PgDA9m13pv23dOqSYZhmJCA8Gq/tm/FxvcorKl4uasi6oqKi\noqLiBOCmMuToe6DpcHgxibP2v/8v6A/E1hLsfyss0jBEGAs7y3GXpSTG4jPGJLDGCGuIxzJbsjCE\nLFZRcRaXNevsV60aBdMwcYmXEqY2xqyXS8WCIU9Yo4qo9PhMZpBclqYhxXY9S8pIOlkLYx2CxK8b\nDWdLCY/LIit2YjLqMy0lSqIYCsavsYhcehZ1O7ln8roVJz/T0DWmkQuTTUy4f2GLhnN+eKjMmuVT\nwWfGrRaqamIiDFv2cHB1F/PtrdHYHF3s2FfYRrQt+w+nj3yQsLLYYwJGrVhNS99wmtxYLeuLIki0\nEt6ey3x6Z+Bm7EksYXZ2cpJuS9E0MFvJOe/861OZ0s6dqS9xM99AbtnEqEJOhXCMFUeDka3co3zs\nA15R8XJRGXJFRUVFRcUJwE1lyBf/8/+Gu376v8buvz4JIHVmaiadk7LVY8mKKW5hQjKzWZpkaF5Z\nc43Mx0XNvWKaH9bSoryM5jPZ7cnm/XrZ72KgAUViNPOuO9b0g0/XR5VWTZl7KSDTZQqGq/lU5mon\n+WJb5IchZUJlv4k0T1FLcDIrmvZ5BoKabIwNPkzX5HFydS1tEjgLT4bO6IGkjEs2b4XdLSV6wHxt\nwJDzw0P6zIngaBAGb1sHyGcsadLiIGGTFqfQr1KJnIrbtLGI0+iBFZONDcn9LuSY2/kWtFmCo9Vk\nWqcRxjnf2kQv+21blh0lZtx1iV0PwwquYXSDkQZaT+ZnZGukt7B02EInUYLZFtzslEwtG3Ck8XTb\n5wAAF+57CGdvT055zCsrKwe0R/Q6E64s7zjwu7S7m7rD8do9dy7Nufaorqh4magMuaKioqKi4gTg\npjJk2ycjfBMTo0ks9LhyodyXV/OOJLZFLlWJKZkkS6qOKHF6ISU0UOaVJS85ZJMSlrxQsdtomU+R\n2ZQBTXsFv2RMjgsoctU6D4wm5Ny2mnZM/BuGoVQ1j3PPmbFbxLVHNuakZf3Bw2ru2I/W17w1TM6R\nToxFtHeyD4CyRSnzkaYITUE8BlVn09giW4tO+w+zuQUvgdl8C/4waRRoD8oGCjZGnT9ucSa5weUy\nMeb5xgaWfSq560TVPNBgRIxCulmrKnWqq500brDSZ7lxLaKRVo805NC2mCazV0Y6tAwqrd+2G7mU\n6kxSSV+4/20AgM3bXpfeN3btWqvlSi8d5W8E+2UT0/aWFRU/LCpDrqioqKioOAG4uSprUJ2cjUHi\nhMnm11Ask2tPy2WstTmfrCYKsoc4Zo8ljrO5LP9fElQv7IRmEHguNbSgZwa8R3gR6aTpfteUyEeM\nyRbLqKlGfmP0fixsPrMBBxkZmam/bj3laG6xnqf2Iaw159Ax8ziL7VEtfdSc8xznOmS5PoZQKMBp\n7DHeDo8XxZ6Zr6c9pjcDZpupXrinIQhz4chsm/lqDpwqWviI1cAa78SOZnLSTZDmDi5iLo0ztHOD\nhE4okG9mm+h56TvOVY5K5Fx8+k8r0QAj5h/29F24/d43AwDOinLaTFtXVtZ2Q8DvzWKxwGw2G302\nF9vT61VVVFS8FNzcG7IIcvQVEcaPb8AsRdESp7heMsPXsvOQyrViFiZx2Rf6sYrWqKhLhWOynYVp\nsHkmiTiuSX9bOhmtZOwzxOwgpdFHCeMetetJqYmOL4SiHMWMx4NcEuV5jFqiJTcgZBFV0M8mJU7W\n6ENLvqHR0akoZdK5npTHRK8h+9bm9YBcYlVOd5AbaXTrYfIoojhGwymKGpYBrZWSJkOfaLmjhfzA\noKF3eW+Qux5vlsEaWJaoLGayGQlD2pWGjaXqKIvVeC0ag/lmOveGZUc6L+KU1W6pCxfFXBDhFcuX\nQhhg+GDEUH5k26UIp99KhrqTUcmZ+94OALjrDQ9oeZMaeYBDPz79UvHSwc5rXdflBzy5dsvPyvcr\nTgZKYrVmLHSEkFZJx000fKmPdhUVFRUVFScAN9c6U+0bKdiKsBPjChzxREMf5qOedqb8YCquQoxZ\nqDX5rOwCRdZoI2090+t8awv9KrEqei6f2Umh64O9a7KsyTuelgQdIaZRFju1wCyW0Sc8n8eukQGM\nl2fJ12guyNCn4WRrswnKZLsh5vW0sxSnSI8rKoNkJyXtzHTUUyjHocOiAMyvj18W6uYdBplzlm+1\nbizgctYg+jF7J2MJkP7S0cOJDeXZ21Ood/9aEmnt7T6nMzCTvsWxSeHtMMixDwvMNxMT7iVy4CYR\nkG7zFLRsjCxWGDKNUgwCGvnqaQTI0jPcIIgH99btqf/w3Q88lI556zRn7TrNlWqo+uVgzaKWgrqC\nNU1/b6brvpjoW8Urj/JcUog3PS9hJAQ+OnXIFMWreU4rQ66oqKioqDgBuLk55EkOOISQ85mTp5Uw\nESyVy5R/H1fCdL1ex0dtj6xzCWE3Yr6/2L2KnZ3bAABbd9wNANi7mkRdxgsTc612osrblteCt+r/\nw9HGICXFVTvIyd9ANgaZPtVPIwEvhPV5WRe3Tf9O1pkT05KjmnXkrgWjsb2op88Y4ObJCCOuUvcv\nrkUrzbRYGL3HcSz6dA632i10wqyvXL0ky6R1d87fwYZWWPq0DytlRptb5wEAu88+BS8NH3g9aD9i\nYb+27WBp1EKzDzPOLabzI9c8x96m42u2zuGOBx4EAJy9kDowBdlO5V6vHHg9Hh7K9TW5TruuWzP3\nYQSmvAYrbh6m95PFYqHnaNrIh+fSOadC0zD5/djfTyWSm5ubr5pwrzLkioqKioqKE4Cb+minKtzC\nkCEcw8RejIlHyX6Pa604yvPoMvKEJNvxgBr8b59Nqtpe7DHb1mH3uecAAGd3JFcqslhxboRDKOwk\nx8xQVd8hFopjO16W7DNEJam0egxhnbWGsM5Ij8OR86dJ13Ge2MBgmqzU/H2hPld1M/sia9tEKoeB\ngcfNA/HjcixjjJqnTM/ZMES17GQjCz9hKRFAFDtNL8rpaNJ52Tx3R9q1DTi8ls5d23DO0/r7u88j\nDqKa3RQFtqjnl1LqdPau12P36sW0viSPG7GsZJ2bc/NRi0zZsWxPcsgx5KoBsdPceX1SUN/xxrfB\nCDPnfFbt7iuDo0w/mDecNjPp+76WN51QTH/r2bjHOafvbW1tHbmO917ZcqdGQGP1/GKxeNVK3OoV\nVlFRUVFRcQJwcxky/1OofI/KKwMTA4hj2PJRDPloAwrWGI/HE+TxZECL7bMpb7gntcYc47nbLuBQ\nWkReuZLY1mlh0YdensxQGFlMjE00AhBCboE4VWDnhPPa8R+p6DymOcVROCq/q2uZsfnHUetRJU2m\naq09Niev7RSdLRpljMeaDUtsVkVr7psKb6N1x2x3OPTpKZZz6JwDQnrPCxNtT+3IZ2mse1eeQ6dS\nbsnxy5g3Nuc5wiH5YduM1dGHi0OclutisTiQZYTN9pwfAIYq61aPLR1z2n7vtnBKVN533PfjAID5\n5rYMp8l13DVr/IqC1+DBwYEyILIjXot8HYZB85BkzdPcY6nYrXj1MT13wzDoeV2zkZ3obUowSsLt\nNU2jEZRNMRZ6pXBTb8jTCxo4/qZyPYHSUTfb621Pl9fKKrkR0PnLRA0dWDm5G3PxLg4rnDkvN+v9\nJAC5fDndtDtxZLLOrI1XRWnI4flsAJLLrdKAxn8edRyjv3Nt1+izoy626Xsv1dubOFKwNYEal/gA\nuLFQbFriNQzDEePljT7og4zXG7AIt4p5psf47HQS3bFUa7mbBFwtstFK1Jsexxph5Msb5MHE8etB\nv2u/wPIwjWP73F0AgGu7qXtU08hYnQG/Vt6mL3Yvzm5nbk8+0/fc+xbMttM1ZFXwFYrXGrh6JZGv\ny3W3t+PMIebzuYZCy5BouUy9Gb96KH+reOPkuWPqa2Nj49hOesRR52xa9rRcLjWsTaHXxsbGaJ83\nCvWbX1FRUVFRcQJwIvT6R7G06d+jMp/rPPW8IMszBpEyGTZLEro03zoDADhcLvDc5acBAOdvS0zI\nr9L+Lz/7AzV86OQJqrHjsPLgo5pbZAtNLqMDyeFafVNKLRg5KA5B9V6FBefas90RgrbjPcG53XIr\nky2WPZev4/e9PpLxOgOClnYx5GwoIBN7SwusGZTkblZB55GdkEi4exFnLZces3Op/y/nqL92SZYV\nS07n0Mu5Z/9h9lVe+YBWxFRr+9fwZYMoFpcHi8SSTp1NpjD7u8lgJDYbcPP03rm77wcAnLkjlS/N\ntqR0C1btTe1E/BfVebviRmF6zR4cpHQDQ8/WWn1vGpJkGRSXA7AmAqrM+NXD9Fwul8u1qAbZa8me\nNX02YbRN0+h1cFxYezabaVSE557XxXw+v6GWqZUhV1RUVFRUnADcXOvMI8w+iOuVOx33RJoaFMS1\n90bbhcEgjJjin+3TSVBDlnNu5xyevZyerC5fTixrZyeVzrTzU7BSVuNWySqTdC0WlpV8EnNqljG2\nrgSi5j9pIbJ+XOsUORbLqA2mH8/f9YxBpiU5McZiNxM2jHXWej0h2drcM8cWy7z90YYgtpAwlc01\n+LdmoDXPT/MNKW06fUrLtg73pGexnBcvrNi4Gc5JCRRLpPh0a63F7tVdGe/YtpW2qc61aGQkKznU\n/UX67Nwb3gEAuPD2/4Szt6eoCjswaSOJSd76hf+quBHgtVeyGiDngruuUwbFZaY5wrZtR9dKxauL\n6W8LRVZN0+i543klVquVnrOpDSZ/G5fL5cj8pVyGMMaslUSRKS+Xyxsq9KpXVkVFRUVFxQnAibLO\nfDGK35IFH5VfLpc7cjvGYhhS+cL2uWTWf/VqUsoabesI3HZeVLR7iTVdfi7llLc2T2O53Jdtyf6n\n9VMFlBFPFNTAuqLzuGNIC639J390TN79qFKx6fpjhry+/3gddft0cGv7LeaeT7HcFc1EGpY6DWOr\n0bSMzePQ+RuzXtvO9fVA1NSNricmImLacdv5uxCNlEE0ZM25DeKZnfQUfe3K82lMhjl+2bdrsWrT\nNbNzz1sAAOfuSOVLg3yVNnbuBqRcSvPvR3aGrng1EEJYYzUsXyqvL7IcLkumzHVijJUZnwCQGZfn\nktEMorTAPM7WlOdyY2ND9QPX0wKUavtyHDf6mnhRN+Rvfetb+OhHP4oPf/jDePTRR9H3PT71qU/h\nO9/5Dra2tvCFL3wBZ86cweOPP44vf/nLsNbigx/8ID7wgQ9cd7tH1QwfV970Ykty6K40vRGy3MgH\nA9Ok8MPu88l1Kco0nBaP6uevXEYQD+szp9MP8OFuWnbRL+HkB5fiMGMmXZqMKUIPk4eOUNyY6bA1\ndXYShCOOV7dTvHfchfTSu89M5tYc/9AwSiGoE5WMe3Lzj7HoD91TYMUaZZnLEGGkL3au++Kfbv3h\ni92V5umH9PDgAJa+0rJaECFZN5MaX9tpiFv7ENtcusIweDtL2zTi3MX67I3zd+POt/x0eku8tfkg\n4dRr3MOptxZf64341cJRrk3T1Bi/EwxnrlYrXX5at1r9qk8GjvM4mM/na5+x7Ilh5ushxqjLcb1p\nORuXO+r1RuMFb+8HBwf47Gc/i4cffljf+8u//EucO3cOX//61/He974X//AP/4CDgwN88YtfxP/f\n3rnHVlHlcfw7cx+9XNtSbm9vV3aDdhHXGmpjI3/4ABUN4bGRRCLGTUNMIGq0lRiNbUgT/lIs1GR9\nxAgNRlL8w6T7D5sYMUSzElKbQJNG+g/bv5ZtjPRS6KVPemfO/jHzO/fM49IH0Jl2f59/2jt37sy5\nZ86dM7/H+f6+/vprdHV14cSJE7h+/fpdaTTDMAzDLDdmfeyLx+Po7OxEZ2en3PbTTz/h7bffBgC8\n/PLLAICenh7U1dWhrMyyRhoaGtDX14fNmzcXPbaf1TWXKkC3dGfTPiQAYVsp5FYuW5mCQRrFE3aC\nzk3ryej6NXJ5CsRXWFbSzWnLdUVWT0SDtMQKhhwpStlWlxAyIUm2lfaVIiBKgkExdTFd92h7z0cV\naK4WslTmcj/1icLSq1uJf0i3jU7LtpwWiRAmdNLAdteHti1WPRqBdnPG8ZamWOwedzjVQyZN62gU\nIm+7iqkeM1XBihRcyLStULO54LakEARZ0YatM33vA5aaVrw8hbw9DgxaOuNa8hCNxeGu7sQsHu6l\nTYlEQl5rtz41uadLSkqkG5usJPpMsSUxzOJSLFl3ttoGs103v3rIdN9StfXdcw6Ni/lW1JuNWSdk\nPz/80NAQfv75Zxw5cgTpdBoHDx5ENptFKpWS+6RSKQwPD9/y2H/a+DcAwJ+3Ny+k7cw8qN/VEnQT\n/i+4k2sSmflDE2lpaems+1Ic0v0/Ez5o0nTHi/2gjOq5Uuw3G8RveUGBESEEampq0NTUhC+++AJH\njx7Fww8/7NlnNv7zr5N44K/78e9//t3znvvJxm+JlDuFHSgILZD84bRpx4dtAYfR3Aj0mHVREyXW\n36mxEcfxtNg9SJbYC8Dt9yi5STfNQoyB2kGCHkJpo1wd5Ip9yFi2UTif4W/xi4hP7FRhLk/tuq6j\nflcL+v/Rbh0TTssdAHTDubSJME1TWpvFEEJIK970LDGjthvyHMJOqIuQ9WqLceiRGMwZZy1aki1V\nV3XJdtvJXKatSV2eXo1xOycgGnXGgEzNihGtSlfLGsWUB6Bi2IIvosQSiPnDfdaYFnYb4/EoIq4x\np+ob82QcHG7LmOKCah1cNUELKFg5ExMTcn96b743dWZxIKtVlT91T9J+1/xWqJrVQCFhi46r3mf9\n6irfSe/JglLE0uk0NmywklueeuopDA4OIpPJIJvNyn2uXLmCTCZzZ1rJMAzDMMucBVnImzZtwtmz\nZ7Fr1y4MDAygpqYG9fX1aGtrQy6XQyQSQV9fHw4cODCv46rxSI+1qEgMSqvRxwona8q0LaG8He/N\njdpWMEzEE9aTz8QEyeLZMQQ7K1fcnAby1pM2lSqmDNyIpuRPy4xnZwa1GnOQ8VVX4rBpChmz1F0x\naBl7VS1/93tKfMQpvKmEgjUfT4WP48Id35btUWxmt6WuSmp6nhDJOyAKf9VlZ2ojNbpgIo9CtSnn\n9zDhjdkIna4HWcO6jBVT4QmC4sYTE2NYsTJtf2eK7dsx7JIkKjI1AIAVK63CD6Ytk5mI2XFmmBC2\nSAiJSqjyi9RPHG9cXEzTlJYxQZZMPB6XVpU7hkyfMU1Tjs+5ZOYyi4/7fkrX0g/V0pW5IkU8fYZh\nyGPTmPGzjGkMEXcrxDHrhHzx4kW0t7djaGgI0WgUp0+fRkdHBz744AN0d3cjmUyivb0diUQC7777\nLvbu3QtN0/DWW2/JBC+GYRiGYW7NrBPy+vXr0dXV5dn+6aeferZt3boVW7dunfPJ3SX41Nq6bktM\n1hfWVAvKmbFrmiZuzlhPQpXpKgBAzLTiAZPjV+3P6Ijb9W5v3LSWZZEoQMyOK06NjUgrkaxwTbE0\npUSmOytaK7wma7OwbtZuoyPyTJ+jTfY/ZG35WMh+spi62+qVmd1eERV3rV0hhOxbCmDot0gc9Ivt\nU6EI95praocuCh4GWRyDim2g4A2QZQ8pdkzH0fWCZChJktpfWo9YT7PGTB4xlxymtOJtC9kQJnKj\nluiHnrTixKl77wMAlFb9EVE7t8CkMo4lXim9SduqoidkLsEXHGqBAYr5unMtDMOQ14pig26pRNM0\nOWa8RChW+lJFLQpB8WHPvdrtwUTxvIF8Pi/HzN0eJ6Goh0yoy57UbYDinvbRq1YncXJXZK/+FwCQ\nLLeSuZL3WG7I0dw1CMO6qBV2VrhuWBdtYtTKCo9oGkTetczGPofhdgEr5yd8U+HlaifXcX0+p97c\n3e4UVU93PovTiyl1qedTXe7UVmH6n0Od6KVQmeZ8T63zLL+vTsvRrB/CjB1SiMXigOHf5xFd9xQU\np4QvOl4kGoWY0R2fl25u2y0ttBKUp1cDAKrXrgcAxO0KX9P5wlK1eNyptEXfZ2pqSt7ceTnM4kPh\nAPe4TyQSnhsv/U7UJBx34pZbDIRZepimWaia53JL67ruSfjyE44pNsn6jZ27DWvBMQzDMEwICNRC\n9lu2VAxVcZJ8vPScPD1jW9oCACzrtyRu7TM1biVzwbaQy1auwsiwlQ2eLLVi3MaUVbUpCqpDrEEj\nP7JcvmSfQoPHaizYnl6rUR7GZyE7uVJV97Fjn4JWRcENTX2lWMgCTotStkuIQsPpnHD2HSC8eiCK\nN8JtAAqXha1Dky78gsgGJbL5XFfNacVK11EkomSBOZc9abqm6A47LWS6FtFoFFrCEvKAkXccR4tZ\n21f/pQFlmTX2MW2r2fbSaKbheVJ2eyfi8bhD/5gJBtKZJitYlT8krxvdU2KxmGdZC11DTuBaOtDv\nja4ZhR8ikYi8vnNZo0zQ5/28I4uVwOUHW8gMwzAMEwI0cbdUshmGYRiGmTNsITMMwzBMCOAJmWEY\nhmFCAE/IDMMwDBMCeEJmGIZhmBDAEzLDMAzDhACekBmGYRgmBPCEzDAMwzAhIFClrg8//BD9/f3Q\nNA0HDhzAI488EmRzlgW9vb3Yv38/1q1bBwB48MEHsW/fPrz//vswDANVVVU4cuQIqxQtkEuXLuHN\nN9/Eq6++isbGRvz222++fXvq1CmcOHECuq5j9+7deOmll4Ju+pLB3cetra0YGBhARYWlS7937148\n88wz3Me3weHDh3HhwgXk83m8/vrrqKur43EcBkRA9Pb2itdee00IIcTg4KDYvXt3UE1ZVvzyyy+i\nubnZsa21tVV89913QgghPv74Y/HNN98E0bQlz/j4uGhsbBRtbW2iq6tLCOHft+Pj42LLli0il8uJ\nyclJsWPHDnHt2rUgm75k8OvjlpYW8eOPP3r24z5eGD09PWLfvn1CCCFGRkbE008/zeM4JATmsu7p\n6cHzzz8PAFi7di1GR0cxNjYWVHOWNb29vXjuuecAAM8++yx6enoCbtHSJB6Po7OzE5lMRm7z69v+\n/n7U1dWhrKwMiUQCDQ0N6OvrC6rZSwq/PvaD+3jhbNiwAZ988gkAoLy8HJOTkzyOQ0JgE3I2m8Wq\nVavk61QqheHh4aCas6wYHBzEG2+8gVdeeQXnzp3D5OSkdFFXVlZyPy+QaDTqEaP369tsNouUXdoT\n4LE9H/z6GABOnjyJPXv24J133sHIyAj38W0QiURkDfju7m5s2rSJx3FICDSGrCJYUvuOcP/996Op\nqQnbtm3D5cuXsWfPHkfdae7nu0exvuU+vz127tyJiooK1NbW4tixY/j888/x6KOPOvbhPp4/Z86c\nQXd3N7766its2bJFbudxHByBWciZTAbZbFa+vnLlCqqqqoJqzrKhuroa27dvh6ZpWLNmDdLpNEZH\nR2W5sd9//31WdyAzd5LJpKdv/cY29/nCefzxx1FbWwsA2Lx5My5dusR9fJucPXsWX375JTo7hz3t\nigAAAXRJREFUO1FWVsbjOCQENiE/+eSTOH36NABgYGAAmUwGpaWlQTVn2XDq1CkcP34cADA8PIyr\nV6/ixRdflH39ww8/YOPGjUE2cVnxxBNPePq2vr4ev/76K3K5HMbHx9HX14fHHnss4JYuXZqbm3H5\n8mUAVsx+3bp13Me3wY0bN3D48GEcPXpUZq7zOA4HgZZf7OjowPnz56FpGg4ePIiHHnooqKYsG8bG\nxvDee+8hl8thZmYGTU1NqK2tRUtLC6anp7F69WocOnRoUYtuLxcuXryI9vZ2DA0NIRqNorq6Gh0d\nHWhtbfX07ffff4/jx49D0zQ0NjbihRdeCLr5SwK/Pm5sbMSxY8ewYsUKJJNJHDp0CJWVldzHC+Tb\nb7/FZ599hpqaGrnto48+QltbG4/jgOF6yAzDMAwTAlipi2EYhmFCAE/IDMMwDBMCeEJmGIZhmBDA\nEzLDMAzDhACekBmGYRgmBPCEzDAMwzAhgCdkhmEYhgkB/wObdqI8wI26qQAAAABJRU5ErkJggg==\n",
480 |             "text/plain": [
481 |               "<matplotlib.figure.Figure at 0x7f8aa3f050f0>"
482 |             ]
483 |           },
484 |           "metadata": {
485 |             "tags": []
486 |           }
487 |         },
488 |         {
489 |           "output_type": "display_data",
490 |           "data": {
491 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAFKCAYAAADMuCxnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsvWmsJdlVLvjtHRFnukPeHGuebHCZ\nZ2cZDDR4AsxjEA1qF+gJ/ErvPaSWaBB+RqiRbGQxGSQEWI0AgwCJZ368bqRq1WOSmqYMDabNw9jt\nAVPYuOwqu6asqls53PmecyJi790/1rd2nHOzxps3s25mre9HnrznnJh2xIkd31rf+pZLKSUYDAaD\nwWB4WeFf7h0wGAwGg8FgE7LBYDAYDIcCNiEbDAaDwXAIYBOywWAwGAyHADYhGwwGg8FwCGATssFg\nMBgMhwDlQa/wl3/5l/HZz34Wzjm8733vw1133XXQmzAYDAaD4ZrDgU7In/jEJ/Doo4/i3nvvxcMP\nP4z3ve99uPfeew9yEwaDwWAwXJM40An5Yx/7GL7jO74DAPDqV78aGxsb2N7exuLi4rN+3zmHBx54\nAKdPnz7I3TDsgY3xlYGN8+WHjfHlh43x5cdz+XEd6IR87tw5vO51r8t/Hzt2DGfPnn3OCfmBBx7A\n61//+ufcOcPBwcb4ysDG+fLDxvjyw8b48sE595yfHXgOeRYvdFJPnz6NlNLz7qDh0mFjfGVg43z5\nYWN8+WFj/PLhQFXWp06dwrlz5/LfzzzzDE6ePHmQmzAYDAaD4ZrEgU7Ib3nLW3D//fcDAD73uc/h\n1KlTzxmuNhgMBoPB0OFAQ9ZvfOMb8brXvQ7vfOc74ZzDz//8zx/k6g0Gg8FguGbhXs72i845y1dc\nAdgYXxnYOF9+2BhfftgYX34817RrTl0Gg8FgMBwC2IRsMBgMBsMhgE3IBoPBYDAcAtiEbDAYDAbD\nIYBNyAaDwWAwHALYhGwwGAwGwyGATcgGg8FgMBwC2IRsMBgMBsMhgE3IBoPBYDAcAtiEbDAYDAbD\nIYBNyAaDwWAwHALYhGwwGAwGwyGATcgGg8FgMBwC2IRsMBgMBsMhgE3IBoPBYDAcAtiEbDAYDAbD\nIYBNyAaDwWAwHALYhGwwGAwGwyGATcgGg8FgMBwC2IRsMBgMBsMhgE3IBoPBYDAcAtiEbDAYDAbD\nIYBNyAaDwWAwHALYhGwwGAwGwyGATcgGg8FgMBwC2IRsMBgMBsMhgE3IBoPBYDAcApT7XfDXfu3X\n8KlPfQpt2+JHf/RHcfr0abznPe9BCAEnT57EBz7wAfR6vYPcV4PBYLgqEUKCCzUAoF5/BADQ7GzK\n3xN5vxouYLByAwCgWDoqr4XdQ19J2NeE/I//+I/40pe+hHvvvRdra2v4/u//frzpTW/CPffcg+/5\nnu/Br//6r+O+++7DPffcc9D7azAYDAbDNQmXUkovdaEQAqbTKUajEUIIePOb34yFhQX85V/+JXq9\nHj7zmc/gQx/6ED74wQ8+/8adQ0oJzrl9H4DhhWFjfGVg43z5cbWNcYgNACDurmHzS58FALjNVQBA\n4Vv520nm0HmPtlwAALQjYcjLt74WAFAunJRliuKy7/PVNsZXI55r2t0XQy6KAqPRCABw33334Vu+\n5Vvw93//9zlEffz4cZw9e/YF1/PAAw88784ZDg42xlcGNs6XH1fjGPuFkzj2td9xyeu5Usd+NY7x\n1YLne9jZdw4ZAP76r/8a9913Hz70oQ/hu77ru/L7L/Zknj592p7GrgBsjK8MbJwvP66WMQ4xAgDi\nthCTc1/8O4yYM/a13HZjGQAALsprWXi4XsXPhgCA6eJ1AICVO98iyxaXdMt+UbhaxvhaxL7P7kc/\n+lH83u/9Hv7gD/4AS0tLGI1GmEwmGAwGWF1dxalTpw5yPw0Gg+GqgYsi1Np45J8BANVkjNQKUQlJ\nQtWoZdJ2jmHt5OG93JJ9KZ+lZiLfTfGK7Lfh5cW+yp62trbwa7/2a/j93/99rKysAADe/OY34/77\n7wcAfPjDH8bb3va2g9tLg8FgMBiuceyLIf/FX/wF1tbW8JM/+ZP5vV/5lV/Bz/zMz+Dee+/FjTfe\niLvvvvvAdtJgMBiuJkzWnpD/bJwHAPi6RYzKeuUj5yRUXYBM2QOu4P+DsOkikTNZBPkVgX2prA9s\n46ayviKwMb4ysHG+/Lhaxnj37EPy+oVPAgB8vQNECVUniWbnCbmCvHrv4fuSO3YDeZ0MTwAAlk9/\nq3znCtQlXy1jfDXjQFXWBoPBYLgYU7LgnQf/AQDgwi4AEXl5mXeRgkzMXue8Qm/OCeBn+Ybd6/MT\nmyBfCTDrTIPBYDAYDgGMIRsMBsMBwTNBXF+Q3DFol+lCBTRKkeU7icppJcOucB0zdkvyWX9B13x5\nd9xwKGBn2WAwGAyGQwBjyAbDASO2woDa8ToAoNm6AABwZETVkZNwQ7FGLK+AFaLhyqFefxIA4Jst\nAEBUxpsC0GotMQ1BVMzVyPXivUMaSM64qWW5kgYhV8Iy0/DywyZkg+EAEVLA5PzDAIBdNYWoJWyZ\neEPeLPoY3PDVAIAYpgAAX/Sv9K4aDhihbbH24H8HAHgqqh3n4zCtgSATsIq6dGKOYzH/aEODMhyR\n95Zkkq5689dFGyN8K9+fnn9KXsfbsl6WT/X6CygX5YGvGsn6UIoDmDf19KGGhawNBoPBYDgEMIZs\nMBwAIsU7k2cexe7D0tVnWI8BAE3Un5nQpSpO0Dz+OQDA2q6wm5YWiWU1uFK7bDhgxNAgrUnIOpH9\nFlOWMbU1Yk1mTIbcqC1mI9eJDzWclzpjTXugJ018Iv2u67XHsPFFub6KiVw7vZImIhUZt3fYqSTU\nnfrClI981Tcf5KEaLhOMIRsMBoPBcAhgDNlguAS0zBWOzz+O0YnbsfvQp9En82HvgJwz1K4+ripQ\n0iUinpV88zZbl8bY5gYDhqsDaom5s/owHPO7PNVAS/1AnCLF7AwiL1PmjsfSBapIDVovzDaSK7m+\nMOTp+hkAwPZn/xoDigOjRl5os+kCG1P0KoyC5IpjK+KyyapcZyE2KHx1IMdtOHgYQzYYDAaD4RDA\nHsUNhksBc8BbX/lXjE7cjrKZYMpSl8RyloLC1pQZc2eF6LwwqPHqgwCA3slbrtSeGw4Iynx3zz6K\nQllwo/limoBMpkhUWSt9dmTPbiLfjbGFG4bZr6CAfLb+8GcAAL3QYhIlmlK2Y+6AXEtFIdtyqUbq\n0RNbGffWM/JVfM1BHLLhMsEmZINhHwiN3EzPf/5jAICKApsWBVLLm3GUG2XLCdpT1CWOTVJXqkUo\nvSjhy+1Hv4DIm7gvLLR4NSAy5BwvnEUR+NS1o2I91hoHIFD4l3jtqHArcJk2ASXFYPoyXn1c/lx7\nim87lOr0pWFxxLlXFzxUQOj4GiKX4UOA4XDCQtYGg8FgMBwCGEM2GF4ClL1uPiZlS2H7vH4gLzEi\nRrITDV2rIQS6lmtOH4XJcrwXdhPWn0CYbF62/TccHALP+fq//iMAoEwNnJ5PGnA0jIH4Xom0I1GR\nZsprRcPa2obRpRx+VkeRnSe/KOsmw43FEEWQaIrX0HfOhaiQC/BOwtqO15XnfsSdyaUfuOGywRiy\nwWAwGAyHAMaQDYYXiSZGjM8/AQCYPvUlAIDTfGCSZ9vYtrnPrZo5BBXzKEP2gFP7RKaJC8eccmyw\ndfbJy3gUhoNCmO4AAKYX5Jqo0hQNE7uOSr6yL6KqZjpBYuQELJNSn+tEZptCjciQieaJMZWypYoM\nt0kJkdccGu0WxbI6JddVH6mmJeuAXtjcVtjdufQDN1w2GEM2GAwGg+EQwBiywfAikepNrD3yrwCA\nqlZF6/wr2hYtmVCfBgxd31t5PxYB8DR+0PQfH41daBA2zwEQRg4Albfn5sOEyHO9/phYWJZRFNXl\ntMkNRFKSiEdkYxHve6h5PoO+qvqeTBcBiIymBDLcXkXDGNXj12MEvfa8GoywfIoXU1UkeHKtpE6s\nJferNoZ8mGG/dIPBYDAYDgGMIRsML4DQCFs5++AnUG6fBQB4pgOVDefaYwCJTGUykTxgWchzb6ut\ncVNXf7wXKQaEHVFZZ/Ws4VAhMb87PvsIAGCg575NUJ1ApEo6RmGvAQ7l4gIAoG52ZT1UWWeNQQM4\nsuVA+9UUhOJGUqfUtPDazpPXJcB8M3smp7aGi9qkguy5lTaMzXh8CUduuNywCdlgeA6oR/HmV/5F\n/j77GErOpG3LMhIKt1SwE0LM/9eyp7Zl55+e3DBDCrk0qijmp2aHhEixkINNyIcRLbt4pfGavMEJ\nEsnl0rYs2PIq3EqoOXH7Qm67bZSJuZ3KxFoljzhmH+QJxYKLWhpF0WAI2eErTbkfjhPyiL2TI5Bo\nOqKTttNrMgRMt+WBr7+4/JzHmH238zXo5l6dL55zWcP+YSFrg8FgMBgOAYwhGwx70FK0Mz4ntoXj\nM18AAFQpIak9IeZLWBLNHVJKM0YNGs6mNSLX3+sP0Kj9YfYKYekLHGJ6roC24TCg3pZSpILsM5B9\nehczoXSlmnRQ2Le9DdQShp5O5LVifVxQE5HCo+foa92KUCymYwCAMkdkWtQMZxdJw9FkyGPNiXho\nI6hs5Rm1VGoLcTJvDqJh7fHman7v7Gf+b9muXsu9JXldlP7Kk+3zqEYrsg1jywcGY8gGg8FgMBwC\nGEM2GIhWy1HYGWfjoU8BACqyi5QSEnvPpvjs+d3ks/0HHMuVUqE5ZVmmbhpUAxHdJIp/wP61yQGp\nIquy5+VDBS1bW3tMNAXFHoOPEFoUpeRxI/tkJ2gkpEYko/YUdTW78ndJ846daYMe2XLS0iY1Gsk5\n4Da/p/oFtcVsmaN2IeTIjRrQeDa2QFlj84zYvo53JAe+8ZV/kv15+iEAwOit/x6D3XU5Rl6DxUR0\nDaEWq9h68yk0190p2+V+lMaULxn2izcYDAaD4RDgkhjyZDLB933f9+HHf/zH8aY3vQnvec97EELA\nyZMn8YEPfAC9Xu+g9tNguPygavXcFz8BAKhaUaMm/kwCYtcoQk0+tKlEtjGMM3lhshoqqR2ff10B\nNGQz/b6UtbTMA3pfoFyUfJ1prA8PYkyIZJnNtuRafZxXHCe0WRSQrwtlrbHNbRfTjlxn7ZYw5TrJ\nfbJ39CgmZL8LZNhotDGJtmycSi9H7hPQKaI12FKhuz41kqOtHl2/RVXL/u/+89/IZyyFqrLFK5DI\n3lsqwiNz4sVUruEyTZCeFqa9S8mDNtuwnPL+cUkM+Xd/93dx5IjUt/3Wb/0W7rnnHvzRH/0Rbrvt\nNtx3330HsoMGg8FgMLwSsG+G/PDDD+Ohhx7Ct33btwEAPv7xj+P9738/AODtb387PvShD+Gee+45\nkJ00GC432lBj7UHJGbtGFK5ZLN0q6wGSsiIyj2x7qDaIISHQ7F8ZrhJmXwpzKGasMGvWsFbMG8NF\nVAsn5fvO1NaHBaltsPrg3wEA6qc+AwAYtGSPapHaGwDaUpEXT6FK6hZArayX9elkpLFQI5lNjFau\n57r4Xn8EACg3pOFI07SopmS7mkPWbiZqDBIKhClz2HtafzrX1UHXG+e5HnAfu+tSG1h4R/ORVjUP\njPJUBRL9XtMFabTSLj13XbPhxWHfE/Kv/uqv4md/9mfxp3/6pwCA8XicQ9THjx/H2bNnX3AdDzzw\nAIAuvGK4fLAxfmGc+Jpv5v+++Xm/93y4+R0/djA7MwM7d/N4ucbjhtd/p/xHX69hrHzv/q9ju16f\nH+55HrT3NSH/6Z/+Kb72a78Wt9xyy7N+/mJPyOnTp6Vu05jAZYWN8XNDzfw3z3wB20/KA2JBZy0w\nt4eGzeHbBrFmzSfVq0nrOwHccve78dif/Fa2OezyiWTTTn5uVb/MTFubBpSlsKyi6mH0urcCAJZv\nuOPAjvNawZW+lgPP89qXP41mVRqLlDtSh7y19mUAQL84DgAoigEC2vnl2SQibG0hXtgAAIzPCTOd\njCWHPOa10D+yiOKY1Pau3HIjAGD5jd8LANj+rOR7i60L8Mz5ZjcusvGyJwrvcrgI9ES57Y9KSrEY\nsp55ZQWBDmHNuQsAABdV2S3X64n/6Udx4c8+KMtzhlD7V98TzUPRX4Dry++j7ct23XWvBwCMrrv9\nWcfS8MLY14T8kY98BI8//jg+8pGP4Omnn0av18NoNMJkMsFgMMDq6ipOnTp10PtqMBwYAg0TdtfO\nAAA2z/wTBhTUhCQ3s5bCGsfSlRiTOhh2hiB7eh6H2GTzEIVjaM8xlI065bAhCrmp6RwzDcCxlZMH\ncoyG/UMnua0nvwgAmK4+iIrnvOVEtLwohKShgUwAUBRy7QQ18mA4uWkbRJrHNHrtBQqn+KA28AUK\nhp0nhUxyN9/8NfL3IyKgqqZbaGnsod+NYf7BDylCHVn1ODSqndqAMKGdpi6nwrHQEanE9wJD8GWp\n17Bs01dNDtU7fub7VrRzqdjXhPwbv/Eb+f8f/OAHcdNNN+Ezn/kM7r//frzjHe/Ahz/8YbztbW87\nsJ00GAwGg+Fax4EZg7z73e/Ge9/7Xtx777248cYbcffddx/Uqg2GA4Myhpodlda/9EkAQC8GNGpz\nqCUnyn41vBxrIFBQ41iGUqiBgxo3RJAIZdYLTzadOwEBVVLbRLKMQtZXLi+jYLjR8PJhyq5eW+x5\nXCBmExevBjLsVVycvE3+3jiPyZaIr1IUJrnbyC3WIyFM5LqqtRFJlHNfDUS45XsefijM+NhrRMfg\nKerqr0jIuT1XIvVYhsfWxp7RGcoC4VJEYumRY/5ErV5TmAJjWbB1KgKbb0QBALGR72jI2jUMa2uD\nlNCH1ybeYNNlb2Wul4pLnpDf/e535///4R/+4aWuzmAwGAyGVyTMOtPwikKgkf6FL/wDAKDXiNAm\nwIvlIDozh2zuoK3oQptZhDLtANoGhouFRlGN/bm45vzcNMKVLIspVUAmbGNw4hYT4L2M0EYLT31O\nSpxAm0vnEnyrWgJticjzyxaL5egoCgqr6l0RfmFL8r0b5y4g8v+7fI1cbrQiEZHeqIfi+A0AgBOv\n+ToAQGAkRUuMfNlHKGR5T2KKMZk3BWRVU8Kz8USR5jUOhQNaihTBfLAKFOMMQ9bWjiAzDswPFyx/\nggu5L3Tk8hXz3ob9w7LwBoPBYDAcAhhDNrxiUMcW5x8RW8x2ck7epNo5TaYAVaah1nIS5nzVUjC2\ncNo0IKnKmrk5PtsmF7OqOkOVrqGzWlQOXKiCu5AMYH/lBnhvz8lXGtogYfe8qO7Dprz2NbzRtAja\n5lBZp9pisr2mTzWmqr5PLGNzwjQXlwZoKS4oikX5fp9agYp55oUVHL3zf5DlB2KyoQy3GC4AALab\nFovLsvzkvER3lOiqkQ3aOl+n2tAk5v1KF1ltagMM13YlfIlK7MQZIvJ3UmpOOaVuOUZ0nKtguDTY\nhGx4xSDtrGO6KiUqJW+OUUPNxQAtw3xRX3fEsUvDfSnGLHIBSz3A+sxWRTPOwTGUqSVSOhFD++am\nmPsnFyyFqp3cgEdHjh/MwRpeNGIImLYSml77snb4knPm1RM6JBQUSOkDGyJf6S1dTydIuwxHc0IL\nvIZiM0WYah9iPrxxAuwt8KJauQ7HXyW1vD03/x03klK4IgXUU9m3ipO2ls1rSRLaNosPI1290Ofk\nWU/z9wKvT3Xf6mr6AKj3tV7D1QKPueA2gx42EsVhfmBixEuFPYobDAaDwXAIYAzZcM1DOymtf+WT\nKFphKY6P/jEbeyTEWo1ACGXBWqaCFq36EdPfWj2oVbAV2jZ7VjuuyVUsPVEjBqTsUx1oDFIdFWem\nBBN0XWk0scb5z38UAFBNpNypVAFX00VH1IAjZGGfsNfA0G2oJwiNXF/thG5aZMrNRNlx58rmKexz\ny8I+j931TcDCfITElXJ91FviqrXmCwy0m1KiiItlUzTaEo9rdZcb0sWLEZm63kVyTLtMGWqnY5ia\ngQBAw2u/ZIya+q38WyowzKLF3FnKHDMvGcaQDQaDwWA4BDCGbLjmEckWxuceRZ9ewy3VKvpUH9oG\nLqg4h++R6SYm6YpYzPxitJvOfCmMT7HrCKXmDCS9rqd9Yh3AdYdS8m4Lx2+SbZQmjLlSqOlDvv7Q\nf4e/8Ii8qVGMpKxXhU4JYGevUDNPTIaq3b2a8TZq9jhWZhy2yKZDQG9JmGyk3qA/EvYbV8Sz/MSr\nvwHVHkGf9hjeXtX9q+H02mW50ZRRFRWJxfEYXptOUeOgpUkNYu5UVqpgbZssvmjydnsLkp8OvL41\nl+5yCWCAXs0zhpswXBqMIRsMBoPBcAhgDNlwzaIha9348j8DAFwaw0elv2Q+zAWXziMqI1a/B5oR\nVrQqbPyky5PRhKFHdkL/BnjErvxEmbEa8msbqKLMueKWytTR9bde4tEaXizaVs7dzmMPAgB2V7+C\nflILVDVzketDWWhqIiLZYmjmX0EldNyaAFQ1NztsHKHRlX6RJcu9PpXKoyUAwInXfb2sp7dw0b6m\niZQ2rX7iLwAA1fYG0khsNCM7SGReOmZv7WKIwAYnmcVS6e9ji8Tjj2yAEbRjWdMx5DSg60hDa1dV\nW5eyTJE8XCHfaaMajMh62tCgLCzSsx/YhGy45hBZjlJvPw0AGJ+XBupFCGj0JsQbZS4ZDlJDPL8i\nFVhx0i4rtAxrFwwPBop3NJSXQovEm5HWqaqAy9HkyFclGtZsLt/8VbK+0n6KlxsqPtq5INfF+hmp\nSR/UdT7XmrZIDE+7KCctxB2ERkqYWqqnmkYf6uS1qWvULG1qtUa51PSFh2M4umLdcaIr1/HXfK28\nX3QBy5oCqzMP/JV8VrPmOAbsbEv7Rr98TI9MvkMh1nSyhdGChMendBrrs444ugKp0bC8eqvThWvU\nlS2pY1nF1orqFKblfa4qUPI7xVgEZzsPii+8P3oKYXddVjSQmunC2/X9YmAha4PBYDAYDgHsscVw\nzUGZ0NbjEpKsaulcE0NCZA9ajcF5mho0ISAkZbn8irpwaZjbpVzmFMgySnZmUkbgkeDIdjNLIjNK\nhXZ4AvyiiGaWbyRD1lIWw2VDy3KlCw9JiVN/Sseq2CLN9BIGAKflThrenTZAw/MI9Thnf2QKv0Ib\n0Owy1M2IiZ85r/p/t3QEALD8WnHlQn85f6ehmKy58AQAYPq4sPie9h5ePon28a/I9jfEL7u/Iutj\n8AdtPcHOujDjaklYdM2yPO+OZie4VPNYexJ6Dr6LEHl1+FL3rr6E03sL9KuuhphqakfZ9Ja436Xx\nM9hZk65X/Zu+Zm48nF3nzwtjyAaDwWAwHAIYQzZcM1CLy8ma2GNONp8CAPRIeWPdZs9pMM/bkgl5\nJCQKYWKaL2nS/KIvCiQKs3qVsArn5TvlQFiKc0BZaP9jFc3Iaqog24zlCEfukLyh740O4MgNz4cJ\nIxXrj/x/AAC3JjnkNBUW2SSHkr18E32pY9CcKa1NiwjPfHKshZnGRnPHKpJqoRxT7SRVOFX4HugB\ng3pRzD9uvP21ADqbTACIzNVeePBjAIARZVmhYi53uYeFO+8EAOyeF0a6dV5yyu2OHE8IEVUlEZxF\nCrWWR2TB/QncWENAjORoG+Nx52XtaWTSLh0FAJRk9VF7MceAQgVw6u9eqmaiQi+JtiKsPiTHbCVR\nLwrGkA0Gg8FgOAQwhmy4ZpDIbDee+CIAoE+bzNzP2Dl4dGpTALkxgIsOQc099jzNq0o6pVljSzKf\nrI6m2UJvkHPQVSmspiFLSSyrwsnbsXD9Hdwleya+nKjbiMie1ztPfgEAEKfCcKcbm/Kl8S6Ck2ul\nWJCca+qtyGdONAK+XyO0bBRB+8iWDDkyFxvrgCLN6wd8Ia/9ooWn4nj0KilzKpZvuGh/01j2dffp\nR2Q5Xl/aUcn7ApE54P5JsVutVoRxT7d3uI4x3M4OV6ilSJH7UQEl1dXaASpoT+/uuo+TmtsQZu2V\nGWvnszZIEwt0Zjal9lBBhYIMPXkZs5rlW4bnh90NDAaDwWA4BDCGbLgmEGPE2mOfBQAU26LwjNAG\nEjMsI2lbO6poSQratoFL2clDXvy87aB3BVyhzFpeNKdckEGgV6FIymaEIQcqs+uR5ONu/Ddfh6LU\nxJ3hINDmntPCdOuxsN+tpz+HZk0Uy4kqYN8X9ltcd0oW3r2AuCF52I2z/E4U/cHo+psBAK6qcuPh\nrMhmoxFlzGEa0I4ZDeF+FSM2aVgcombd8I13vE7e23sMocaFh+UaHpGxt45MXdXfz9J7RBXMPW1q\nMuoDbCrRozq6pJIakzjTA3yXx8qVznSHaPusmV4Y8lCptVDmPw2A9kOm6lwF1DH18+/C8TdQUXNh\neH7YhGy4qlGrCcjGWbSrEqrWm5eKqWJSN63UtXxV56JsseW7HrJaoqGev14bsKc8WWvpCLyuWyZY\n3xvBZVGYqrmkrOX6179JvjNYusSjNswitjV2zj4KANg4I65sYPgWaPMkMuB5qGlk4UuZrGJ/CW5Z\nHpqGVF41mzJBn31QrqnFG6/LD2NxKusLrPsJEw1hNzk9UfTkulgYngAA9EcRzdFXAwB6x6+X7xTz\nD2VtKrD9+D/K9z1NOjz7IQfeqlPsJk4trdLrrREhVRFrFAVDyxQvJhpztNMGnqmZgkKxhqmdxHEC\ngN7JW2QTLNUDw/LgA0cMDRw7SOUeyfTCTnWCG9HdLslxDIZHYXhhWMjaYDAYDIZDAGPIhqsanhaH\n5858HAUZQszSK4b5vJoiAJEmH/oNtbdEbDpmrH2Mc89khum860xD9nRyUrmXK6vMuhuKXYY3iznC\n4Ngp7oc9B18Kau1hzRD06r/+A+KWhJiLxH7XLBeKsc6RDsfoRr8v4dPdNel97GML50Rw5SphdO6Y\nsNiRl7K0zceewuJN9JAmQ2147TRkytPdGonmGoNSttHr81oYjHDstjcAAJKbv+1qSmRn/RE0G8/I\ncZTCKL22KOY1GFNEUHtP9dmuWWLUUtQVGzhas1ZkwUG/Gxy8CrMgy/UZlp6yQxXQGdfo7yOyw1W7\nKSKtooh59khMv6RI/+oY8+8sMlRdLa7A8MKwO4PBYDAYDIcAxpANVyUCS07WHv0UAMDtnsdsURLQ\nsWDN+8Y4U7aheTfta5ym8LlxePXfAAAgAElEQVT/MRfTfDFzhym3cepKRtQOsdDOO66ASnraQnLF\nJ2+/i+vpzPsNLx461oEWjZuPSZ54+qRYo/rUonTaxUuXIVNOMbPL/BlL3haOCmvbPn8eYV2YKSqK\n/pKwvYJiqMF1x7F9Thh5j+d6MmXXqA011IgYDYQtlixX8pX8vTM6hdvveA2ArtFDBi+4Cw99CsOW\nBh6l7Gyf4qg2qD1ni1bL6Bo5Rs8GEr5lo5M6oBoJsy/78lpfkOXLogewCQVo/hGYyy6GXbepRhVa\n7F7Vss+zIxuHc/BDZcQUSNIYxA0DAn995YLljl8KjCEbDAaDwXAIYAzZcFUhMn+29YwY7IdzjwAA\nfFPCMcHbsWBtM4f8qp95NTnU1oj9AZwqSbW/rapIyWg8KhVV58YThRqLMC9YpoTgheUs3iqNI/zQ\nSj72izZENCxhOvfFj+PGr/0uTJ/4HADAscQJAALz8o7mFiqej01Cqfl+Ko/DlGyTEY/+wjICGfb6\nGWHK7SbZ5gJV121AoLXk6jnJ1e6SYWozk2FVoldoz2Naq/apI7jhdaiG8/2Op9z+ZEvKsrYfeRBH\nlD0zAqS6544VR6SGuetdMmM2T/GsOBiOFlBMGaUZy2vgtd0UmyhLidT0uK3AnuCYzPRD5m8gcaw0\nSqH6CgSX2zbWjCz1yLR9LNBQ1T1cPgLDi8e+J+Q///M/xx/8wR+gLEv8xE/8BO6880685z3vQQgB\nJ0+exAc+8AH0elZraTAYDAbDi8G+JuS1tTX8zu/8Dv7bf/tv2N3dxQc/+EHcf//9uOeee/A93/M9\n+PVf/3Xcd999uOeeew56fw2vUDTMU03WpTHA7hP/BAAoozzBT4sSvtZ8sGaPWWvMzEyIdS5Ojmpm\noCYgLTrJdMWc8S4ZGJm3q/rwhRoecDk3304uICKMmDu+SZoAlNac/SVD1b27T38Ja6wt7lEEXERt\nfyh/x9TCM3SRlfJe22um3PzBMQ/rmGttx1rjm5BYS7t0XJTwG6Ww8vNPrQIAnnlqFS3V1BolKWky\nMxoJ8ej3SpRqHNNjbW5f2OiJW++ca8UIINerr/6LNJLohV20vFa01ll1EXnZ2OY6Ym2R2JAxjwZU\nVCeg4X6oajwUHLx6N9fHt1o7z2vaD/t53wpGgOqJNs4Q9uy5P00CnFpu8hiVMaeyyHlpVxkpeynY\n153iYx/7GN70pjdhcXERi4uL+KVf+iV8+7d/O97//vcDAN7+9rfjQx/6kE3IhgNBCAHTdXHf2v7K\npwEAhfpUM1TsQ8zOXNqPWEOJcVduKqkN+YaSDQ80xDlp8s206dFdachyjm1Z3vs2909WKya3R4bR\nugord0h5izMDkJeMaSM+0+ce/n8BAO0zq+izh3XkBJAmFA+5GfeqyAlHexRPZvob6wTOawYaoqU5\nTNu2CHUnmgKAktfH8evF2KPqD7BxVibptQtiGqI25oXT9IXvyuA4ITcUNR297TVwDBFPOaHunJHQ\n+84jYj4yLEu02mM4G9fwAdJpyLrNXaYS+zuXnPQaPoi6mJCOLHM/pJyrtyjLTx57BuhzXUPxwG54\n7OVC13ms3h3zGGkoog+7mhrol4D2jFbRHX8KruwB3Keqmg/TG54f+5qQn3jiCUwmE/zYj/0YNjc3\n8e53vxvj8TiHqI8fP46zZ8++4HoeeOABAN2N03D5cLWP8fDozXOvhxU3f8f/fNF7V/vYvxw49Zq3\ny39ec/Fn13/nxWN8NWDvddC7WdT3K//+rpdjd54XN9/9EweyHrv2L4Z7Nv9TYt+xtPX1dfz2b/82\nnnzySfyn//Sf5gb+xZ6E06dPSwed59lBw6Xjah1jNUzYXX8GO498EgBQtsKgWvXfJVWNbQPVcumx\nBpZsRJanoK2RKvpaszzFbdNuEDGLfZyWc9B0BGRoKUZU9KVGod2eyCoGHqf+7bvw9Kf+Cqfe8G3y\nmYXrXhBBGS2FSeceFOtIR6OPFF0ntnMT3PDt/wtW//r3ZVmy2RBCtnGMLc+VGmFEn8V92t9ay6cc\nS4pSCAjb7HHMi0jXrR7OsUmoaRtZU/y0vbEOQCInALC0tISFAT2gj58EAAze+B0AgFd9y91ScgSg\nboRpP3T//wEAKHaFvBSuyOzd85rzaoLCkHEKDeopS7ooKIzsg6xstlgYon+LCAqjE3FZZLlSce4c\ndlYl7VMu0eiEIfeFIxKuv+kH/lc8dt//Jp8xZM0W3zk83uv1MKDpSe4lvUIB18oJ4JRYbx677Y0w\nvHjsq+zp+PHj+Lqv+zqUZYlbb70VCwsLWFhYwISqw9XVVZw6depAd9RgMBgMhmsZ+2LIb33rW/HT\nP/3T+JEf+RFsbGxgd3cXb33rW3H//ffjHe94Bz784Q/jbW9720Hvq+EVgoasabImLGn3sc+gF9hX\nFWSoXvOAZD9A1zhC7TELsmcoIwq5+0yuemIT13bcoNC+x+18hEctNIvSA9lWM38o+0wxztHbvibn\nCg3Pj9hOMVl7HABw7svs1DWR8+waHeAWicI9T3FdO6FFqhq1pNRZQ7ZqEUlG6XzX35oJ2jjWiAfX\n1waAwq9AMVWkBWbNvsDT3RrNtqyzxyjLypLkRwP3r1dG9BeEkU4XxHTkjq8RhujgEMh6Vz8n+fHB\nFruSae61nblm8+XN//BYU4woqHWYkhlXejFTyNZbPEqDGqCA2sGSKR89hgF1FztPPibL75BZX9/1\nZ9axyX9zN4pCf2QJ0c+LIJ02yygcLFC9P+zrznHdddfhu7/7u/GDP/iDAICf+ZmfwenTp/He974X\n9957L2688UbcfffdB7qjBoPBYDBcy9j3o/w73/lOvPOd75x77w//8A8veYcMr1y0mr/bkpza9iOf\nAQBUcQuBZvmqpFZSoDnIBI+kVpclGbL2Kh6p+hpwNMBv1TTfay64yP1tVQORu8tpL1rvlYQApZZU\nkXX1RNXaO3p915rR8KxoW2HBa4/+E3aeFIMXH2nNqOczsCFIrOGU7ebcL/vwqmVjSnBUJQc9h/wu\nSpevkcjPHF81L9tOp4hkwtmAQ1t4MoccJjUaahFSzXPP60wZc1kNEIfCjBdf+00AgIWTIkL0RYVQ\n63H/KwCg2pJc8ibLsao6YjCUEqKy1+d+MGqTVf0JLfe7IOP3vBYTe3Kn5SMogpbnyf4XHLNUFsCy\nMPuj5e0AkFONKXXTgTZP0UhUUmV3UjvZ7hqPZMi6H7PfM7w02J3DYDAYDIZDAEt2GQ4NGjKIrUfE\n9KNohTVNXdUxY31Ch74Koou5HlLNFKI2lUiaAx7kVoplmlsdXNFZZXrNQevyVGu7QSU0G4BLbO9H\nJrF43W0AgODsJ7UXkeys2ZX63Scf+HsAQNo6i6LdY83Iv8tWC3ojUtIaYZ4HzVBmg5CIQGaramvN\nuUYUCGMyQFUqUzlca+1xCIiqRdDltH0gmWFTT1FQxazssfTy2XBJ8rN+5SjKr/oGAMCrvlHU1Wnm\nFhvY8GJ3TcZhYSA1ygO9CMM6Ns+Lchu1HMfSinzHjYQ5t+04f6amNl7zxKx5dmUfTqsQWr2Wyabb\nCKftQReFhVfH2BoxzjBcRpd0rMOO5O09a/NjWaAcyD4lT12H2tAGAKGz4TS8eNjdw3AoEOopzn5Z\nHIuKVkJ5NUO/aSaQo17WWp6ipVGAlMjI93mjUg9jncsD4Cm40k45am4Q4NHjjUVFP84xxMkJBZMI\n0DREO0BNGFpcOfUqAEBVXH3lZZcDQUVBzQTnvyJ+AxtPfAkAULAEx7WTLMIqVLjFSXJ2so0MCQed\ntybzk3iKsTN8mXJS0H7I09B5mmsZHCdm7RiWkOfvLKZqcw0dH+aKCi0nwmogX1o4IqYbOCIGG0v/\n5i24/dt+QNbZF1OYbE5TT7D25EOy7gm9sEtOiOydjOEiRgNZZ71Dg5QLawCAwVQeTgfDYTbDKVXY\nSDeu4TExMWnbmPdfL8c0Y+KhweSqx7IlaFh8ppuZHj6PuWC5n4rn+iGCw4gBy6ZC0PUV2Sfc8NJg\nIWuDwWAwGA4BjCEbXla0ZEYXHvk0PA0SNHKmry51HsWt9rZNKiQRluGcywsoA9DCDQ15OlfAJfZ+\nLWQ5P+g6PDVOWMl0RxhMORGm3vMqnunlvsfqYxyO3iTrWzq53yG4qhFo36gdmcbrcg5X/0XC0mG8\ngR5FS8dU9OPor9wv0Wp/XU0hMOQ83RZG2I63c4i4KCnIG2v/YUHRczPdiMias4CrzqHqzJRDx74V\njmVrytj1O/WOWlcmLCyyN/IKOzkdE6+Fhdd9CwDgVW/5XiSyVddI+mX1y9Kve+2hT6Nm6HxUUFQ2\nVbcNbqNMCGSi/ePCdlHJdbr+lJQoNbsTLB9f4OHIPlcr18+uBi7GfPyatlFhnEfn8500qqyixRlS\nq0Yg2MN0K04Zvgm5/EpD1RUjGSbn2j+MIRsMBoPBcAhgDNnwsmK8dgYAUK8/kfN+Ci1rCbHNOduK\nfumxnX+WTCnmHriuc1WQVzVeSD6XkYQs/OJnRQ9+WawEFxdo+LB2AQAwXZeOP+V0A0WPAiWyvKUb\nbsXcRq5h5PKhqUQQds4+hmce+CgAwDXCaEHrykJLeVzKAqOkzEyFVy6hXJBxRGLEoi9WpOVAXsN4\niOlZEUHtPC1GGmlb/o7sbuTaEqHkOtVOs6aAqw5oaTXZsLtSaDT33J2z3D85NyBWJwz2TB4NMToq\n2yuOXAcAOPaN3wcAuP4N3yrLFgk7T34eAHCGwrW4I/s6TA1GvGYcLSvrTRmz7bUNOfZRiWpJcsg6\n1p42rssnhAXvrD6F6XmJKoyuk/V4Wlbq5e5SyMY1yvS903KwhFJv+9qERX8LM6VKBaNCWaHBzzQ3\n3RuWSIwWaQRDSwkHi2XuWmV4abj27yIGg8FgMFwFsMcYw8uChhaJq1/6BADJyWa2qo/6SnCLjjnn\n0qY9iaoYU1a06re1LV5UkwTncilTB7IDFMhZ54Kq1xM3AgCGp+S13D2Ldix55slIWNIiLRJTSoBz\nCCHknFpxlRuEqBJ9vCkRgrV//QcAQL32DACgcruomEvX0henfYlp7IE2ItKcwin9pDrYO59LcFT7\nq3+XPeZpfYGWt6nRUFjz7qYoj0cTqq/7JSI1AanuzEIAIEynuVFE1FdtoUlml1KCb/Xa0f2Qz/qL\nwmoHJzymK3cAAG59myipV+44LYe4K5GUJz71V2jOSivFRTLrqNdS6sORb3I3UHDdR2joMd7awfqT\nMra9RTnWgm0cHZtWLB0/inZLtlfx2lOXUY1KxK7wAEmNPXhpl1WRTXW04qDlmLez+WI/b8eJnpb7\nkSm7hCp1+XUASGTzTapR9qyxyn5gE7LhZcHOeQlV9+jalHyZG7ZnL2p+18FlUVd20WI5Sm5I77pJ\neu+rIqWUJ/cU58UqznVilFw2lX2qGZJbvB7lUCbiIW9065/7KJfq44a3fD/W/vlvUTAkufv0l+UY\nGQr3bOTui64R/MuNmiHmwFKcyZZMdmFjE099ViZgxzI09ZLWGvCQBvn/e8ezYjg0phaJE5GGRh0n\n6xiQ63LUMS1xkij0ZBQFfMU0AyenlVfdCQDYeEjKqMp2msVcmq2I2RM6zXSFoniJ4iO9wIqiRGBZ\nkeO2ekPWFi9LONjffAf+zdvFmXBwTDyf1x4WJ7lVhqcHqFH05BzXnPRLFZu5uqsJ1lI9XleencdG\nzmWXt7Xz57kcH+4Klh+lFu6o7FOxLHXHqeZ6vU6QMYePK+04xvXGAJQcRxW1hZwaQkZkyNlVfJjl\nNRBLPsz0R0DBUih9yNVzmBpz6tonru5HeIPBYDAYrhEYQzZcUSgTOfvFjwMASoaJGwBpjwgrZY7c\nmXxkpqzMWEOMzsHRXzeLunLXJzKzlHKf3IJMLHL7KaQcbs3lU+q7oAYMziPpvvFVy0O8o8nE5irS\ntpT+7LL7zQ6ZRMkQ487qV9AnywK/U17B8HZop9hZlQjFxmfFjKVlCqGM6ufdoFQxnDZVYomaI5Xy\nEYjqaVzNG6KoB7TzHsiMmKVIM/U1LkcstGSN5WRaouQ8SicRhV6f6yEzHNwuPX+3Pv8Qqp4InZRh\np6ARlITEaErFsGuDeReqGNrcH7s34jb6Iq5a+upvBADc8m3/Dp7s8Il//L8AANMnxZP6SNmxQS21\n02iPywY2AT5qhECOv6AHdEOjk6apUdB45uQJRlm25byMN4Sh7rYT3HT7q+X7/L34nMaZMVNhOVrL\ncifXU7/tKpc9ObL4xHM1S5Gj/vay8Q6PR721ywKRYXS9dL1+tyzhCwtZ7wfGkA0Gg8FgOAQwhmy4\notjZEIFQQV/fAGVIMxaYewVbmFGp5DIlfjdqP+SUhTgK7Uij5iMOKYtScllJfrxPmaU5ZVCap1Zm\nV0Q4aJenbjmgKwdxVZlLbhBoeuGFrUTmayfbF7B9RvKfC9eLUKilb3dZjnDQaHmszYYIhp76l4+j\nYkctp6YQHPMm5zxjfg+N9hjmeWBeMySfvaNdPe8j3uqYeY8ysyUKwNLM+dR0qkYnmAtW60WHruOR\nirv8QPK7peZSbziej017Hitta5ND4WjtSFtOjYRoZY4vPIqBRDFq+kEff+PbAQC3fON3yV7VEzz0\n/9wr29uU8qtBbz4/m0KT2ap6YesF65LPPurqid1wXFsahiC02a87keGWZPFHjok955HRAqrBEY6U\nnhc9eSqka+Ez22XeX8e36CGVur88j9xmM2M/redavcXVBMRz2ei7cjY9cwmai3aoqgUYXjqMIRsM\nBoPBcAhgDNlwRRCCPPFf+JJYCWrnm4LlMqEtoHQp7TWmd8+SQ85qa11/yAw3p3z1O8pgYtdoIIWc\n7Ztbz+y6dVvedznpjrxr/ozbVEYWYs415tIbbqskM6yaMfqgNeRTMg4Xzgrrmlw4g3LxCPebimNt\nPvASEEILcMw3ycY3aOPYqyc5QhD2dOrJHXs84Jko13GMzMWrFWUR4ozph3ZL4np8109Y2VZFptyG\nTkUfveaVMb/9mWNxe1/ZG9svSJ534dabsf6gqMS1cUWrFpoAkpZmJY2KaN9sMu3BAsYrYlV5y5v/\nRwDAyddI16ZmLN2XvvQ3/yeG20/LcdC0ZPbakwGKOdKjCmZV6rehRsVrppkw382GD8pC06QF9P8a\nQRpLDnnCaM/y8UXsbMm10gNz6z0a2QRl4zGvu+RyhUZ9ihKgyn+XFrFe+znP9DjO1yz/VrV3Yrcn\nV86eIcwda4RHVbz0a9ZgDNlgMBgMhkMBY8iGK4J2Ik/87VTqXDNzaDvVs+bYlA13bDTmJ3ZlJXvz\nzM77Tl2t5gXKcNUqJLksj45at8rte0f2gK6mVlmvfrdwBbz2ctR0dZpnyEVZzeTDtYaUX6VxQx09\nykRmyt68PSd1pxtffAaxFBYyOC49lpttGbNyKC39dD+fDZqn3njyQWw/Ksw40p4UrGWdThqATQwi\nDTgKJuU1GJBckSMEhebpqaT2lSqYm5zD1+W1HjibTySHSKaeWRebREgEQaMJqkbWHLayc9epgjme\nFQmptpioFpewcEqsJcc7jwIAWuZgXdXL10NBJbWnkln7+eKmr8ad3/r9AID+CbFCrWmG8vDfSN54\nNL2AyMYRrbLNoBacvM5SyFaVmT8mHUOgoeWotn9M2g6SyzfTKcD31II0NIy8LC7LNgbXo8cowPip\nx+W1Ogegs9d0qUSp55c5/myWUwyziUrJJhe7O8LCe1WnjM6NNyrNN/MD7QXdH6KgWl6vR2XNERHV\n0HLI+4FNyIYrgumOTCoqAtIbb+7MlOoswtIbqN7UcnQZs2KZeYQ2oMym1Xk24EK6vgJJG7WnHNeW\nv4si31iLPSFn3ckYUxaTeS3/AOa/W3STfn540HCfZ/9bXyDyAUUnsIY3Q19U8K18NmH/3OnqE7Lq\nJRH2bDz+IMq+3PCqvvyEd+ie9fhnPiLrHZ+Hj1ryQgclhjFjivlm7PLB8cFEe+OGlA9APajVWzyL\nvcpefqBRQY/LIWcNfTsUKh6KFLdpf2lX5etBz1meCNQQI1XZXU2tpyPbFOkDWHLAkRuk81JbcyL7\n0mPcZoN2IOsa9vkgwP7FxS3itHX72+4GlqQkDRsSDv7C3/zvAIDRWFILEWXntb7naTDL0WI3IYOl\nRHl82mkO9Uft58xQfprqMjXaHTn3Na8vtyLn+fjNYkjTtJNsMDO4/mYAwIQlbOtPPgUAOHLyWC6n\nqwZ8yNTJs99DKmT5cijb6HPIJ7vb+ZhqhuordptK+lCan7d8V/LGJ051OUu9fu5SZXhpsJC1wWAw\nGAyHAMaQDVcE9UQ62uQuOnsg1pfzIij9e5aQZHFJ7m2r4WWXTQu658w9oeeYulConw+NAp3tIuL8\nZ50NYETLshq1JHR7tC0pxrl1AjPGJIxSt3UD0EwiUcSDJtNweIZ0SxU80bYQa/Jarz+arRlV/qZM\nd6AmEb5UPw6gJ9sfTSU83Q5r1AyfKutVbwgtxamqqisJy6U7akc5E4nYc466sUrdVzB/PjVwEdom\nn0/fV4WUmlyQVbvQRRh0W2T6WoKDmJB4sMu33wIAKCbCRjeffhrFMjtHLQiLPv71UtJ06i559a7A\n9nmJQjz6t/9Vjn9LGGLUk1Z5ODWh1v7YWlqkEZ2UEKf13HvKhkM9zWOtrLnZ1bA0GfO0Bdhp7ORX\nvQoA0DsqoeqdrQ1+ZxsJjDDQcKZ/jJ2gzglD3jj/DJaPShQApQjfeiyb8mU/251OeA26RdnmoN+x\n2orst9mRa64c8ZrUK84lBJZ9ldm2lOYqK7fD29SyLxhDNhgMBoPhEMAeYwyXHSk0OPvQx+fey8xI\nS5JSzPnhruzoYjadS1eYUPT5u0DMypP55TT/5WY/ycspG+4+DWpJ6OfZn/MJWpG1s7s7t/xoJCxM\nPtfl/dzetLOlVsq+NQ+nYqA2AcqYyvm+tT4T9ZDNNVJOaitr7Vhtwf/7oKVI/KjowdEFIud6Kd5p\nmees691cptSJ3bTUrOPlyv6jRj6yoUd3DnTMNJjQTIUZ+qLMfalTM2/Cktl04fOx6fJtrflu7ldy\nyhlRcT8Wv0oEceWJZUx2ZQeu/7f/DgCwfNsb5DMKp84/8iAeox1myUEuemJd2dSSVw27uxgO5kuZ\nlLlrd7ImTJFqTbLOl/DFNGNMorl8Hit7Q+Doq1+Npdtu5rpp98nze/yEnIvdc+tYWxc9xtKimJiM\nWXo3Oi6lW+3mOpx22OrznKkgLSYEFR2qoYhelqEz1qlGwsz7el0zkhOzoK2F9pkoGMmZVDJmJ677\narjSuN5+YKNmMBgMBsMhgDFkw2VHBOAnwihDNvanccSMhDorr7Uxcs5LzhqFqDmFMhB51zsgKmPR\nUiTNp85ab+5BRzrDDLvU7eoW1UwEcI49ZGs1tGAzhJavwaHQNnjKvjOrmGHcWmZU0UCBqcrYhC5H\nqhGCWvv4duOQM7U6VHG+3CYklxPcXj0iC1XKJiSWO2kJTpu0qYP8HabTvE+lng8/kzuGjLmabuTW\nhqrWVmU0XE50J+bf1eG0iTVK3+Nxy2dqualKd2BGea0vPAfqwOlcrsZByopfwejYcfRX5K+1Jx8G\nAEyeEYY52ZRyocnTD2Go0QzmjLXpRW8o6ut2uonJuuRxK+ZOq57kXCOvzxS6a0XtKNtcVufgGNWJ\ntFaNVMrffOdrZJ+PLedj1vx0qeeulG0t3DDC8Kgw0fPPiCI8MG9N4TwGo0W0vIYGw2XuYxd9ahmB\nYd8JRPa9dqn7LWplnec+loui8g68JstBH4HMvCUbHp6QqISnitvw0mEM2WAwGAyGQwBjyIbLjhjb\nXIPqtXF86J7Y5TViT4/7OexteF5cZKXZscWLTf34/owkOsU4913nfVZZuz3SaZdNTIBWc4xsRoDc\nzlHbDT5L3jsbjXT73Nk5st5Wzf/LApEGIkFrdLWseub5Oeaa4Cbvv+wr9zm5bszIstKMNaLuiy+U\nzev7ZLwAajInPWcaHSi0eb0DWm08ofulOWDNs87sf37NjeyBlvLuYk8rwKgMO3YdD3yxpz4cOTww\nk7HmeNIO0ruAPnOcw4lYX+5OpEYZjNoUAzdzPXE8eFylmpEMltCSEU/WRYHtyIKzqj6GzIi1aUdk\ndttNEgKV3+6YWKPe/Cppo5j6jJJElxPuOarBOmKtTnDOwQ8lr3viFmHW9XgLALD2+KMcghpLNAkB\n9QOqx4gx5jHPBf6kyt51o1iwvj1qFIBj72iq0hQOBfPUsZRtLZy4Xb5zBVuJXmvY14S8s7OD9773\nvdjY2EDTNHjXu96FkydP4hd+4RcAAHfeeSfe//73H+R+Gq5itNNxduHaExXOXtAxFPBeS17mb7gp\nposm2fwNvZG2bZ5V4p4b/6zRiL61t3xK3tQFefPhd6OGbIselpaljGSbZTGV3jizR3aL5OeX3wtf\nlYiN9nVW0wyKmJxDQQOLtp3x4EY3nwHd5BTb+QeL7HRVlbnzUWi02xQDuaXvwsDZH2U+zF72+mim\n9FNuxRnK0dgkt41OyGFsnjoEDeHndEHqeujm2DUny+hyiFrjz3qdaFzb9btOQ3s9rfPDnHPdQ1Tq\nrhnZ5zJvwzuZ+IZ07Fpmj+BJE7FLh6wpxW6lejW36hrn4OnPPFqSMPCEDlf6UIS6RUvhFnSy5uu4\nbjC8Tiau43fcMburKDVM7jpho+8x9K6COoaTvXM5PZD4oNQ7IuKu64dybW6cfQSe/Zw1DJ1m+ojn\n9IZeO/oQ4bsrzPdV0Ef3LhXZqfiwKOAgoeneCSnRSqX1QL5U7GtC/pM/+RPccccd+Kmf+imsrq7i\nh3/4h3Hy5Em8733vw1133YWf+qmfwt/93d/hW7/1Ww96fw0Gg8FguCaxrwn56NGjePDBBwEAm5ub\nWFlZwZkzZ3DXXXcBAN7+9rfjYx/7mE3IBgBAu7OdQ3Aq0HLZKpEsB2HGXEJeg/ZkRbrIuzpk9son\nf5+kZAizpVTg8orUhQMG3KQAACAASURBVKVzY+UufJq7QmVGTctJfndp6Ri2tkQIpFaEamHZjHfy\n+lTkgz39mVXw44ITi825g+WfIWWrymxoUs6LwkII2a4wqH+wHo72q01tFzqc8ZUGgNiGTHOjlk1h\nOrc7Rb+PPhlY2KUpBbfZFF2/aJetLrs+udwajyd05V4aTXCqGIqdGwvPdaCRBhZG/EqciwxwlPiq\nzDt0aYFse9qFYkp2tBon7ZGsnZCE0fVLh0XtsbzNcVWGC5//Vp9wNYcpOR7tFkvgQotIhh216xTN\nPxa/6g4cvVlET7lcSkvGnHYDA6pSfydaDsehIlMW+1b9TFVt/A0NJYKxeP2rEcj4Q5B9S0H/DkiN\nlj1pukdD752oS8vpiorjqXq+gtGAnkMoJfQ+OnaDfNfPX++Gl459Tcjf+73fiz/+4z/Gd37nd2Jz\ncxO/+7u/i1/8xV/Mnx8/fhxnz559wfU88MADAC7ODxoOHi/3GC+euuVl3f5B4dQLfH7DW3/oiuzH\nKxm3vuNdL/cuXPO48bv+80te5uW+x1wt2KtRmcW+JuQ/+7M/w4033oj/8l/+C77whS/gXe96F5aW\nlvLnL/bEnD59Giml591Bw6Xj5R7jtTOfR/v0FwEAQUuIsmhHX7t+xnt7HiOlOfEX35pDjBFFnD/G\nnFbMPY99frNQAcuMuCzn28gOlBlHvh5ZOo5dGoIMFoRdTbRJRKpx+9v+Pc7893tz0wO/hzFovtw5\nnztIJa9NFDgO7XMr25SlxCZ03a+8in30mLVDVUBSEZEyc2Wo3uUSMTWgCLUIg1oyK1cW2U8zsX9v\n7v7E3Kt00eIY7TkvKXW5VzSao9ScOBkmXM5dpzFtJGlA4Shc8r0+KnYuAoBbf+A/47E//m3Zpp7D\nwqHUUqxChWcaZQioC1pDLhzhOM5rFSbTKSqOUY9jVO/Ked3aEDtKX0/QbHKM+JkyfzVemdbbcGT4\nY+aSr3/D6wEAw1OnchijyNEJlqGpE2bhs/uLNgTRMcvHVRTZ3rQkm655HNrhaffCWSTm6Vs2KtFr\nOLRtZ01by7WsY9enFeetP/h+PPVXMsbVgLloFdQ5Cg4HC1i8+ZsAAKPjt3Afu/Nk2B/2NSF/+tOf\nxlvf+lYAwGtf+1pMp9OsLgSA1dVVnDr1QlzCYDAYDAaDYl8T8m233YbPfvaz+O7v/m6cOXMGCwsL\nuOmmm/DJT34S3/AN34APf/jD+I//8T8e9L4arlLUO5soOlcHeeFnancZEzqlcrZo1PzyjNI4q3cv\nVknn9/bUTynjdb5ji1lQmildzBQ0cFujZTGFmDAfuLG1hpUjYk9Yk10o+6s0JxziTF764gYWGU6Z\npf4tL4X3zx0F0OhC5fPYzJo58EtyfA6I6u6gEQf9Sghw5fxPX/dR86LJO6RKHTxYltOwPZ+WzXgg\nqdnFnp3N/abhUeb+zfPfadvQpdJ5zmqaXKj5hk9Vlx/uKUvrrDvz/qsqO6vFuY1yEcPFI9x9litp\nbpy3v141QBvlHO/wvFYDWd/IiyHGhafOoKCmYLIl5VPNmIYcoLo4tthmf+sbv05aOw6OyvJuCiDb\nSXIcGA3JJVehzbn4nIq/qHnHjLZAGTaXGW9Irn98/hx6S0e5Cf0N6biEnKfXMrqCpU05goKZnuRZ\no8HSKBqUxOooBkfF5tOY8cFhXxPyD/3QD+F973sf/sN/+A9o2xa/8Au/gJMnT+Lnfu7nEGPEG97w\nBrz5zW8+6H01XGVQYcuZL3wUPa131Usu9yjmDQgp32C0HjJpN50YcwWu15CsiqBmbspBb9QaKdbQ\nsfYFDt3EoyUnXW0r0PLGssCylk2GKBfYgWcSJ9jeFpenRZY/1VoalLow6N70QOc01oWwcwmPzsj6\n6pEFV7mUKc9jM9W26mSlmqosjlLHriLf1XVi7kpgUu7b6wvt1ysr6rEEJzmHxrEUqeSxMQoWOfnF\nssjnT1MQapml+17NDEUnstNwdsre2UXFcz9Wr3I+6MyUTcWa+69lTBruRy/3StbdSAy1DkaLmG7J\ng4TPOQzdhpaDTbIvs+549p12EjofHlnBxvoFeY/Xw5iCLU+/awSH6+56nYzj8WOYRetSPkfaNSt3\nvcrdkrpj1X3VSbKgEC62EX4oDwCBT5X1tkzE+oCAJqC+QA3P0RM8ZP6mJhMU6rPN1EPSB4VeNx3k\nB708IauDmbwOj9+MotzX9GF4HuxrRBcWFvCbv/mbF73/R3/0R5e8QwaDwWAwvBJhjziGy4aW4b8i\n1nCxY0XAvMOWQsUmmazksG7Kn+31h86OW95f7AObBVxco+u2Gyl6CblLEbB8QoSJ62TGymzViKEo\niizqCheEZSwNxRyhmUjZk58JT+fyqWdjzMqA9oS3Y3RIfu9y86FeN/OOln+VuRRpxmiE7LnM8XC6\nc3nXaT7qKTdBFlwOu/XsYbRqcKzmHw4+s73ncuGdC9dnIRkZKqoc6m7rCbc/37UqhBaZ/lbqOMYQ\nsbqTFR4FIzChJ9GN3lDO5c7uBqpCUyFk+O2873eKEe2ePt1qRhJaOa/9YYXF628CAJz7/BcAAMMo\nofwdru/Una/G6ISGijUsrB7fXSoiZKY+L8grytke1PPQEHJZlkjc152tzbmx09babdFHOxbWXmoZ\nlvqjO9+FqHvzvt3FrLFH7uHNc0/BWFNJGufEsduePRVjuCTYiBoMBoPBcAhgDNlw2dBMhV0UYZrz\noFqqok/8uQRjJvcatBtO7BjE3v7JaOZdRFJou6d5wmVBjFK97vlTex6HttvG+oYwjtHiwty+qaVm\nihHHjklucJt+xuNtYSKDPtlnAlq6OSjLUqYZssjL5Vyn7lPuHxs6pra3fFDz3d65zCxj7rer0QB+\n1xVQiuzUanHGZ1vFbdrtqSiENQVlncnBM58Mze9WwqYnLI0qygKl6zQAcjjzPaBTitn+UYmulkQ5\nn3Iy3/fnb0WRZC30yy6PSgZXkK0VpbyGGJHI+iraR063Jd9bloXYqqLLiwbt1MVt+cJ3egMV/eln\nzDNPXA/DFTn3t3zD1wMAnn5IukfdfFREY8Wgn9et+okpL4F+2Yd2+FKGrNGdQo0+2s4zRUsAHY9Z\n2fV4XKNhVKPZkt/XtJbX5IWxu/5CtuxsdmU/tGtU8i6LywpaofY4nmnm2tPfq5blaTSiWpEogStG\nMBw8jCEbDAaDwXAIYAzZcNnQ0JgiiPklAKCIyqDIRNTDv2kzO1HrSGhJVBszW9QccrEnDxdjQKMm\nHa5jecAMq/ZF10uX+7h0XHJ+rupjc0NY1XRKu8NFYVtr56XMpfTA1lQYh5JNFZoG5ifb1M7Zcc4e\na3brRJrJz3Y5cH0nM7e9uWc1UfEOIKvqBWWJyubJfoCcM1aWE9OU41JmFXNkpyHnhtyG5rYBp4yO\n7NnTXtJp4tqXnVI3afcpbQ6hph1ltqrMZTYzpV5VX7YbkqrnaUOpX0kF+mpEkoeIuVgtX3IlFipZ\nT70j57BT9adc+lOyOYRfZg9mRkAm0xpFxeszsMMVc8AFWWM/jtFAmGhDo5FTrxfTj52zz3CbRc6v\nR72w2TFrsrmNSrs6aeMGNSiJ2m2pM1FBpaVQ8tkujVNiip1afSDnJY1lzBpanPrlo2iZS0+7UhXg\nSh7fYh/gfhQ97VvMiEHsSuh8qfso22hY8nX02I2y7J5olOFgYAzZYDAYDIZDAGPIhsuGlnaMs0h7\nFMMhN+J1uU5zr6Lapzb3cNV+u67pFNiyTERko4WynO81HHJTBZ9Z49G9bRRHi5kR7zAvvLMjzGPl\nmNRyrl84hyELa7M3xZ7jefZj3WNi4txMjTFzr1oX7Ir8lHxRX9kZw5O9jTNyPbWqzt2MIj3neSuu\nZprHuKz6XIHmoJWphqxIV6h9IwoqiHtVtmR0ygj5t88mLa5rQ6lt/1zHwnMUoVQLT25LR6HwXY0x\nE+Sp1zW3AIDBaICt3fMAgOGetpRTAEu0zITWWJedDSUAuCPA1rlnuL9qusF91P7OMaLUPDmEdY75\nWf+oRFmmGxsqRs5RGVcKUw2DKaZRNAo91U00au8pyzQJKBmN0HxzqzXLpdZ3F2hVOc0FB9x+3JDf\nW7O1mTUB41ZNS5hLTgk9vzh3/FpfPhuP0XrjxDx9uSCtI6vBMgyXDzYhGy4bmulG/n8uYZoRcQFd\ngDbElMU3s4IgQMo6cqiaN6HAm3JsNYzr8028bfjdbFxAUZHz+Ua5viE3cDVc6HmPCT2bjx6V0o4N\niry2dmSfq7KHQmdijVnrBDYzL3fHNn8c+dnD+06kticsHVPozEueY653DvN3z9lt5WXjzHq0+5RO\nngGVOi6p33Z21OA+FkBuTaye1mpYwvKjNkWUFFPpw4eK5IqkpVEz/+qEGlS0NyNm46tnKU7e57Lo\nekVz+YaT7dKKhI43NzfQZ0i2VS9sHms1WMhlUhqC95xkZs2wFpdl0t6+oN7k8t2aD3mu7RbQkH2i\n+FAj0XVq0U72lKjpg8awRNnKRBjohV0yZB29jG9vsIDBIidw7qO6s2lKIgXAVzrZczzohV0dE1cw\n7O5iuiZpFvUBj7syLs24wXDhOI97b6lXNx2ou52meJaP3Spjpw9whssCC1kbDAaDwXAIYAzZcOCo\naTu4+vm/BSAiHn0Yzw6R2fQj5g+c2gMybtmyvMMV/WwmoWwptuO5bXpo9yHAaTkM192Q6o0GQwQK\ntqAsbSDsaePC0+hx+03J3riFsr06byd3jlKLyFyOpTvikIPOsXtPvjPrvayscf6Z2AFd6H4+Cpz7\nGTt0hDYqQ9/DmFuH7JWc/alZihMbj5LHqAqrpI13g3aIKnL4OqgFZ6DfdV9KXnZ3dzNbVvMV7Xs9\nW5alUYkid93S8q0unaAh0gDZx2xt6hoUNCtJHNdIk5mGkRDf1LlXtIbZAwdvOCyznSd4XcyWsSkc\nmX5B1lmQVasIMVYRNa8d5ZUVZBy0XCi5gMR9i9C0Cxl3iAgUBBYUubU19/GYMNbe0WNdCD93dGLU\nR8v0UKCgjWVgJ6eyokiNF0E1XIKbSLplypSM52+hagYzERs1aOE2i+4iSiqKq8RgZUAxl+Hywhiy\nwWAwGAyHAMaQDQcOx5IRR1OFWShJzP2NtXuTS4i1lqp0VpUAENrYmXQQERczy5jmS6pAlrOyQLHW\n7g6WjkiObveC5NiKWvLE3vvM/FwgkxzTDrOnBh8um1oo/U3KkMnOn8XAs9vHmXxxwp4884wlaNrz\nnuZuU158xjwzm53Mj6fzPguttElHbgDRL7uxKvU7ZPzKjGLIvau1xEq7+gQ21CjRAo2c61jM9+/t\n8tbd8ScyOS1x8r5ASsLIteeyWo82ZLqzVpIq6CudllFpRKQPl7TkjQxXRVXeo2tLrREGfmdW3KXM\nesA+zLz2WgrhnAPKEZktIy4DHuPuhXMcjx6a3L+YrJWRnCI08GS0YMBleFJa1A5OCEOOKeWmFrnb\nEs9dzolXJaKfz/3mCIqe74FD/zpZZ+gzyiE6NBRVgid7zueKeeYw0ywisuypt3I9x2rGVtNw2WAM\n2WAwGAyGQwBjyIYDx2RTHscLZT/oSl/29s3V911y2fdRzfNjZw+Rn+ZVpe1UOTvzvlbp1GTayyvC\ndrbZJzZ6l607G9oMVmRbrfMomFcdb4sCu1KFbFCKlbIBRjZGSPO5y2drv/hs2CugzmyljXO5Zlkn\nWeuMMrszP9EvzVscAhcJsdFo20FX5XyhslXdZS1Dc0WRH9ej5kij5lfZjCBOAeZBXU+V7PPWqIVL\nXfmWNulQtbaP6kmRLTtVFaztIFNKM+VvLGvjvk4aicCU1RCOdp5VpcxSvjSejrE8WJgfIi0Dmxkr\nbQPp1OaTeetl9sSWBeRlyp7H9c4qAGC4QKX5zg6KXI/Gi3G6zuOZouF1tEALzuqklNPl0rG6y2nv\nLTmLQXPtAQwUZBbf1cDlXpz5Ghqyf7c/elLGbPVxtDzng9yulNde0bHgxjG6dMJyx1cSxpANBoPB\nYDgEMIZsOHC0teTUXMz0d6aJBGuLmXtU1uImNZzm9NRUIqiqtu0UsZnlcWNqNgEg9IThjsiAds4J\ngxmpxeLiMURaEFa5/pV1mtMJ2jGZPZ9TlYkUZDsJDugpk+V2VW2d87wp75ybq8Lt1NOxvJjF6mcp\nzbBMZXTzhwrnUibCXa2xtpXkYcWuJjlCm3zIARX9PqKjqpgK8lw7ncuS44wqWlc6r3Iu+osIbOvX\nU/auqmnmWV3R5d1dqS04iRBzwwufx4Q2mUkNOTxA5bYakwQIe6sK3R8H19d6aA4az3kVJhjTUnKw\nfIyfMbqiaufdbdQNW25WElVZOMI2itrWMgREqv5DLax3QCvRoM0uhj2sP/q47KuK1neo5g8Ri2zN\nODhC1TgjD63ap7qQoxF6XmOtzT/I3Kc1Sq1GULvSqPlyl5cNUXP7bLXI4e0fP46S4xcrLYwvOR7d\ndNAfSQ66Gg5huHKwCdlw4Gi0PERdtGLoomq8M/R6cuMLEypcegXalje8PU5dbdvMLA9+Nh/SSykh\ncPmSIi63ICUbNXsVD0sgZOcjOhCxbKeejKGVP2rG0PUW1m25rsuU9ovNTacudvHY60qWRUWuC8Hn\nLkPqMBXb/5+9d42V5DivBE9EZGbVvbdvP243u0mKFF96WBJb1HBkWxRsY0xIcK9mfxigSGO1NMAF\n95coyBjQkGWBgGwQhkFIizVk0LBNwBJBWYAtDhZL/1jQ67Wxlnc9nvH0rExqrKFkPWxRJPvF7vus\nqsyI2B/xnS8y697u5qvZl2ScP9W3Kh+RkdUVeb7HORratTZvn/buvTI6aVgcJmFY/u1aDe+zVaxZ\n3CvnMLBxqH3tMaei1Qt4M7RLLW3+8NumVmcrGLnn8qdVf+ZKi8m2zYYx6rXMWjkWMdWWY+9gqPWs\nDz+y2IpD1Pr6Bppey04fzhpE0afePP1cerMeD64LFqiaFNZeXJI5ota6iNX4doLJpngj15wrOYcs\n/nZxhAPXJDekF3+clL+wkI63tHcPFkQdbibHqYw8DFlqa3c9D+3hdXgq1KHSe6UPp0z18LsYvG4P\nKn9BQvI25AcssPUQ286596qb0lTZJRS8figh64KCgoKCgl2AwpALXnNsY4sx5mIs9UFOrCVKMU10\nNus60y1Jdk/tMbnIB8heyf3SJStM6qx4Fe/bl0KUjIaur7+IlSPXpvPP0ng21lM4s64qdd3JKhss\nZspj12tjgRG36Qmc8DHXzJVV6XWFqOFGFhgxJG+NVR9nsl7MRRcMMhtik5ROB+fHzNTdqRHnn9ZQ\nzCTCkkFJGFmdoSLH2Bs3D63MivS80bmihGlFlyApGApdB0ZCe7Iosk0OMdvY6CH719VtnVPfYyss\nj/oe9HIejxqgm2sXkjN5N9JIQ6OTJuFwkYEMscJoeWUwSDIVT1/jrQ2Magqa0LR4eF3W1aj2psEt\nTdP1N5upxage1/AS8qYWjKdULO+3DTksz3YnHpvXFYCa8qKgCIqwZ9v7OTdz32F6dNc94RlGZyRV\nFHt+4fXyVSh4/VEYckFBQUFBwS5AYcgFrznINCnW4EKvMEmYTCvshi42qU2FhUnM2TIhmQtXtHiJ\n4veejKhVuUL6505Xkzfu0nJqXYmjPXjxRz9K20jerabjj6l7OV9hDnN6lKGdwahpgohcaLcRGSZg\n5/O7ZKHMDffcnkwc5sLTB0NzCW2DGrgksSCoQ94DWolmQqVewzQKaPJJs0ey+kvLNfIwgyFxroSa\nsphqBoRKWB7FVCha4URkpap63tWMfGQWrnlhx8iHFCxRttQ2emxHAwr5XrFmYDw+jAnnWuaYP2y2\n8rAsFDP0VRbxkS697tm3okIpvGcUP/HisT1ydZYrDYwCCLun2YMPqOX6lw6nbdYlWhOnWyoT2rGF\nSQQ60DQyPRWsOEVU1dBkI7BmYWTRiaym+lNHRlKk+HDW5SgNPbBlPmAW0NEzW6IYnXw/3eJhbpXv\ndcHrisKQCwoKCgoKdgEKQy54zUGDAU++lTUflSE7ee3YCuStilXMi4dEAFGORcMIL201ITLnNlUp\nRVZgV8IE1kSoZLRnQVuYCCpyWmNBokHhiewnTCELq4IVqNhudfFnWuov6FUFaLuPbnMhLRHmGGPP\nnpI5Z85DN6w89k2TBUo0733+E2pGfocqax12GFYXoyf0qe06NKLgPBuTx6G59dziRelMbZdiK4+8\nNnWNdiKtQ05y/MIw+T3b2DiHes9+OYXIeopBSYgms3C9WGHflIo0BpbmJZLnnYlsaq0VzEFNMZhq\nVcMUfieM1bMwKrKwN+Xvzz6/Biv5dbaBzYSFW7k+t2h7BhoSuRD27dUIwmpRRARby2RaaJNpAqLK\nnUp+mCGQmFvSZiKEs/eq9wEAVt7+fhRcXhSGXFBQUFBQsAtQGHLBJUOfkLHKmjKBgaWyynYsoohU\nGOZ1hdnMuhatVqJSMjJ9dSuTjtPOZnDN0EyBaIQJtZMNNMwRssKVuUcT4ciwaReoOdth1Xc6h7za\nvD9AqUcaIzCvPBxPfz6Yz9SeWN8TfKBFJVkX2XDfNpB5zLEYBIgRhjdOq2ntDj3S8/vzmIHVvabX\nG63GHkOhkIio5d20C7TS96qX7HJfuTJrQ/OQShm2Bgy0kVnGUVUw7FkPkqfm3Mv4qqpD2Eq52pH0\nWk+n6Z437WaS6ARgaAjCMudeNXwn1dSzyXo6jmO+GTI/2yMG81GNgCxvStpKVlw3I3SUCRUhkWZh\nHwBgsnVGLrmDrYY90qyE1mDH1AOjufsg3wvKddqm0u8wwz0mpnNuxTFGVyTDiCuvuxkAMF5KIiBO\nevJ36qkveH1QFuSC1xxGdID1V9kHhI7CBgksVArSyuJDp6pEKsghP+SzYBFFe9rIcZZEY9fIOcxa\nhZm0mHQ+hR0tBRum0vLhDDwLisRTluFk26vG0paqLIelryygcWDImCFW+btrc3uSo8IVF3QJR8ag\nSl/sVLHqIhV14WAIMvd/ibpWbbRFJc7pXqt4SIg6bi082+F3VvfXlijI9ZkcxuZDEFdCXSQcrDwQ\ncdGEPszIzsFkFTOek6Ih1uUWr8AiLlGNYoGf89qSJVkOdPLQMKLfM6zKmk3WkuJWs5BEYdqQW6Ia\n+a6FIOpssjDW0wm6DVnQqbrFNjuZPOejhtUDv7u8r1zgQy4q06mWhR3jEcxErknU4rx8VIs6WDAR\nrdwkunA5eair5AE0Fc3xnsk5NEUjD2PR6MNsDNSnTv9/Dr/jw9h75VVy7FK4tdtQQtYFBQUFBQW7\nAC+JIT/zzDP45Cc/iXvuuQd33303nnvuOXzmM5+B9x5XXHEFvvCFL6BpGjzxxBN49NFHYa3FXXfd\nhTvvvPNSj79gF0JdaATRRG0NUSEJCaEZFaKwcOJHrPsJG3a+RSstKxVbVmaJDS80Iu4QIhb3pHBl\n9IkJba2dlQHJ8WJmhI6MTOUoozL6iNzCJBvJNlaLlqI6IFE0JBfjODes+lGxDTlcCF5DkqFN52yl\nCAkxwio7ErZIRtkv0CGzDcNwuEpoxpDPn5vOMA/6N0eNA5N9ZWerMMdiee0hBC2EU/WVmgyXtNzl\noi6mBxgxtgZGIgOMZig0RJz9jK16HUu6Qb4fobWI8j2o5Cdtup6Y8sLyIcy2Ujh6Q7yGx3vStgvC\nhieTVdQam2YxmnwHKC0afE5hcM6YftEoic0pAIb55XvbjBexLi5ijqFmYfhW5sxVBmK/DOOYymHa\ngI5X/QhSq3MkA5PjG8wYuRgfAQBc+Z4PpflZ2Kde0QW7Dxe9M5ubm3jwwQdx22236Xtf+tKX8IlP\nfAJf+9rXcN111+Hxxx/H5uYmHn74YXzlK1/BY489hkcffVQVkwoKCgoKCgoujIsy5KZp8Mgjj+CR\nRx7R9/7u7/4Ov/mbvwkA+Pmf/3n80R/9EW644QYcPXoUy8spd3Prrbfi+PHjuP322y/R0At2KzRn\n2FHaL7caWQp6BDEjMMKCbRYxIOMgI/Ejp+IOlEucrqdCmFWRKNy3Zx+syHGunU2fMc/r21ycFaUd\nRY2o6OzUy7sZZZI5N5he+g5I6b0s5ZnZZzbHkDwxtSP5UlWAZ+uNMESb88uQaEIww+dlJTbWKNPn\nfMY5puyi2VZ1xIKj0JMydbI/ryswJ29Dzs/PGx70jsviIc2jOjJeKTIzORc9XyRnTPbJJovXgiKT\nIxj6Hg1BJD/MiEY3bWHaxIJNldhvI8VUW5vnUIsf8h7xNp526Xuy8WJi0bbpEXLN5cs9ZFFUbdHN\nsisTsL3lLQZt9EPkzWLNgW0SAwZgJCri1UXL6TkofqKCICz+k/sbjc3CO5ZtTpzr9NrZBs3BGwAA\nR67/IACgXhxGnwp2Jy66IFdVlX5Aetja2kIj6jIHDx7EyZMncerUKaysrOg2KysrOHny5AWP/dRT\nTwEoVX2vBy7HHB+4+ubX/ZyXG28/9j9f7iG86XHVz78x5/iKy3jul/v/v/wmXzqY+fL8Hl51lfX5\nbtxLuaFHjx5FjPGCAyx49Xi95/j0v/wDAGD6o/+czt95fdKnD3IQUQMKHsQIwEveS5702dZiXI0t\nsT2cTMVEYCt5LluxrmvqBWyd+TGAZLkHQHOxsRXWYaxKEDo5B/OSMUZlq3wADcKAHFmfza1ElI+k\nML+1Djf8d/fiB09+Wbc3c4zQ1dkgQBmgGebbgZybZCW6nWvTgTHZmtEO99fxxe3/B6O2bwXNC0eW\nLnMbYe4++m0JrWx2wXxmpzKOneRzbcMccqV7kQgyL8rS9hhjnisKg7BCnczZmkG72eEPfQIn/sOf\n6jUCQGhbhA3xM5a8KvOkcbwAL/UK0hyEhpXLWmKfPbC3/T9RudAclYjqWTwUPEG0Wi+gLWJ1ro7e\nOpcICl/BSn+RGbVVDWdpOsJ2MKmglmiLryI6torxeyW1BZM6EaJ9178fBw9fn7ZxL9/PuPwmXz68\nogV5cXERk8kETeskCAAAIABJREFU4/EYL7zwAg4fPozDhw/j1KlTus2JEyfwgQ984DUbaMEbB1xA\n+CMdfKshVSOLkmupsCVuTyGAru6RrjoSdvRxBNOlH1yzldyZIP61YZqKuzbXzmKkbkjyC9nmXlYg\ntf/YHCPF3D/yQwMXQA1rs5fT5vqZahiiZZGVcSb3NqufsbTttOzHtbpg+IFqdHoYyJHhYb+pwscc\nRucPt9XKs/T+nF+0vKmbsAWqpeuTgMdxyA5bXCTnFdQQoxaDsWhP+6k1BJx/2G3F+5OL3HRBnvsp\n4gNHWuC0AQtA1nnWYVSVLpbdNIWu2Vdsu1Y7j6iQxWtXt6Nocqiefd16f6iOZgF9QJPPWhaXcczN\ntjRB7gE3qppF96ou8IErbeFMhOE9yo3ZaVs+TPlc5MbFv91zIwDgbT/xkwCAZrwAa1/+Qlxw+fGK\nyu0+/OEP48knnwQA/Pmf/zl+9md/FrfccgueeuoprK6uYmNjA8ePH8cHP/jB13SwBQUFBQUFb1Zc\nlCE//fTTeOihh/Dss8+iqio8+eST+OIXv4jPfvaz+JM/+RNcffXV+MVf/EXUdY37778f9957L4wx\nuO+++7TAq+CtBQpaKFsKRot/grDE7I8sO8UA0EtXwoRU5wrtGowUsixK0U4rql487mh5PyyVmBgG\np8awsp+YfYPZquJyURUZDFk824zI4hD6fJoMridJBRYqzRX7YKjcZXssytrtIesctuU50vsV22V6\njHk+MdQnsTxOZrpsEwoazuY9UrlrLaoy28ZhVcSZNLzV81eaXpCwfJ/5aqGYXL/OXYQx2714ASDI\nkXcKnJr5tp0Y4WsKYAyV07qNNdguiWIoi5Y0hxbSBata6XT40vMqHe2NjbrSVJKTdiqYLqdbyHT5\nFbROj1qLGhf/L+SgfKdqbHZO8CXYHDpnC+C+6/4VAGD/296Tjiv+zgVvXFx0Qb755pvx2GOPbXv/\ny1/+8rb3jh07hmPHjr02IysoKCgoKHgLoUhnFrxm6KTY5eR3vgEAKnOJziP6oXQm2520XcgYzfEx\nrUpW3IQIiIiDHUkRV+d0PwBouwA7FRlOOUB0wpRndIbyiBVzq9Tt5eiDjkkFMEheyYiiU3aoshVM\nIVNO0ode7piFV4mJ1VLw5DufhUTivO41VJuYDI7beh2rzQxOJR7ntL5Dl3O+FDExwuRcjUDxE2RR\nCzkADwyFsmaZV2pqI8I1cwVwbKMSFqraoOgVbpHFhpxn3wZl7L1IwVwuPfa+Oxp9aERi1XGuW2DC\nQkKZB5JpFnLFmNvxmHDG8DVEgxgomyrzQOGUng44PYY5jcFzrqDtTZTerMjqmS8OUM1ReierOIzs\n0y4cxNXv/GkAwHh/Ev3YKcpS8MZEkWwpKCgoKCjYBSgMueC1QydesmvPAwBGPZbjtZ2GVb29ymWk\nvGpma9ISpB4PVfYmVnEMMlxxDrIWELamKg8dc8pyxs6rWElmYGShgBXWTWav0oxkr6hgxM3IkeVR\nRlLG7qPX6teKQiPCmmbCllzldq6CBpJ7lIy7k7GqDoDrVRfb3uQCyXkJvepq3+Vctjo9CNsLQYuB\nydCVLQ6MC3jsIePvS6Pqvw3vo+RQmb8eVB2zUrh3GXTYisMISm8Dvbac/veDv02MvcjLsH2qqpfU\nXaNdSxX6DedKxTuq/EvIaWUlN3PAzujotA5Bv580dWjz/pxGeoN7n/en13GvjS29AURte0pvzURU\nZs+RowCAq264GWaUW+0K3lwoDLmgoKCgoGAXoDDkgtcM09UTAIDGJw1zwzxlmOZ8n7yQHWSvXKsf\nsjdZdRdqoFkYflWZAw2BVcJtZtEqjiFMk/3NiKhU25Aesrl3mf3LhoIiZJ8tG00dUMtnjhW6Q+9i\nGKsCGFqtTRbOsfugzJK+zJo4j1EjBPPoC32wkt1AJEjnpDNhmp57JHOwjVz6DGgnch2aKJdzUIQk\nKvvWXDhYkS72hdH3thlKRcYLCEtkN0ujveoqFiLbaPU38rzp/KmYjFSKO5cr6+OQTZuqhl9Iwhtu\nI/UmT0SYo1lKfsShGgOB45cdHSvS2bdeaT85JUW9yLiSqZoYABo+SJTGi6Wnq8YMfGThGjOcXxiD\nzqQc+GSUZD6vese/BgDsOfg2udZSSf1mRlmQC141GEI8873/BAAwEwnRrr0IAPA+APUwtKqRY/kh\nHNUjTKkaNdcmA0T42VDAwncM/+XCmiD7eynuYkENi28QY160GQblNtYCtRRTgVU/svDQecdFOOo9\naZ/Q8M/QzfRHM7CVxtKBiaOPeo21KDl1usBm7eZ5vWwbeD0BgeHsRoqoGgkVdwyhmzwmXawZfwU8\n/0lhEckPGD6o9JyLVBXM5EVSodZNTBfI2zz+QPVpKGLStn2VsOFDiN6naDD/eLJNgSyEfGn6XCLX\nU+WFvBa5326WFujpOQlh763gqTktRWFe10gW23lAHsw8vztz6QKLXLxoVbxE5sx3aOR71FG/XFM0\nacvWjjEWDepr33Vr2qZO4ynh6bcGSsi6oKCgoKBgF6Aw5IJXjSDhz8mZfwEAjKyE1RaloGWrxWxL\nQqssEJIn/lrCupPJRD11yShHo3ScaOw2lpRFHdiCIvrLyIVjoU1MufK5kIsSiT4MqZyrKvWcZb+T\nev1qiDYgBNHSFgZEV59s+esBadeKrekfDnGgW02Wlf6iWUvXzrZdq+pEq8BHZn0slouqBS0fZDvk\nfEaT0wNWWm5iTz6yD2PQa9/a+bndmP4JBQyT97IGZJAsThukKWT+mWVQLk1RGD8XikfPn5nuVQOG\nvH17/syZJslJjvame7d14nsAgG5rC9USi8uGgiRRixAjQmDh2ZxeNSMJiPAacZHCK4q5YAYvUq4L\nEtVop+mzdnwIAHDguqPYcyQx5Hl50IK3BgpDLigoKCgo2AUoDLngVcF7j9XnEtOoutX0njDEGXOO\ndY1Gnv1aEWmI0s4xnSZ27WqDut4DIItbqCylcSpfGfywXYjvt+0sCzSEIQsOU2E5waPj/naOPvqo\ndkC24mfixjPKDIjtPJT1JJOpRbjEhCwSAjPMMXphqK7XNsRr5LCqukErc9JTsUz7aRsVEAJzxTJ8\ntgIpw8s5YD1fLw/JYjjmjA2ZKNlelaMSWR50iHSqIaPUfDUL2XrOVMOccdo/GyUML1bb05CvVf/e\ngTUrI2crkmptGARe/3gknyVJ3+BEZKZxqc4B0BthdDqYt+/ym0GiI3JuHj72rt+quAyP2yHI/W+l\n4Ks+/G4AwJEbkwSmHS+hsuUn+a2MwpALCgoKCgp2AcrjWMGrQzfD+sl/AgC4kJgdq5wNrfks0DFX\nuiexk1YYUJhRMMSjmm2k/STXhzpVwxpjekXNvSQpsrB/13Vqp5c7kISRSqtSGy2syiVKvppky9jM\nVqVzxQi7Maz+bm02zPCsKhYmRCWHmIU9oJXLw1wyYPRRuC9uAQDdZArL1io/rEpW9oiY89us1O21\nEgFJXFMznWS4zI33JSgpGan3SrZpWxVEUTlOtjQhzzNzrkFagBSqxFljRrbLuWIu1hgEjkmFXzge\nOUyPMvQlKgHASc1B8DmHrnlzm/+2nCypfue8NouptWiyegajfZKztUOKze9E7AKnKlftyzlC/884\nvFcawkBEi3T+Pde+DwCwcu175TpKK1NBQmHIBQUFBQUFuwCFIRe8IrDH99z3/z/UGy/IeyIZSdYl\nTGo2m8HSNpGm7iKsUS8uAUhiHkGkN9ljWxvJpfYZMquqZ8LCcym0Sl52s3QcL1XWZE+IgGfbrBou\nyN82ZrEPPZdIXU6FvVUVlNMq6RNWLdXi3ntNjJKIMi2owiPWZHMJ9lHLcawBgjBs52hNGAZzFqPJ\nghzD9uwessmFymvGHi3vRQbSieWyhIVW2kXdEzYxc7nb2Mvnaj8zLSYl1+6nAwMPAIi9KmurTFR6\ndNnH27OeDHOV00GNPPh3QDYLYe+3yIR2XoU4OOdtlD71pRStsdN1dG36zjipdodERXgd3gc4Wmwy\n8qHyoBQ1sZgrVkeQc3k3wsEbfwoAsOfI9TJX5ee3YIjyjSh4Wejkx3F69p/T64lvo5KFa9bJj5ds\nG+UHpxktwfMHT37MVJ+qp+nciCiDm/uhsrVFJ4vrdCI/ppPNtD91p72HpViG7MeQays/8qlwyw/G\nSOEHHwBDJStHZSsVfE7jCAGh32qDnufzDtLU+hDBkGbNBTWqyhN/1J0qXXWwVJ2SeLqlo1LXW5i4\n8PDpYT7G23NRmvc6jvFihVrpvlAQxTEszXEN9uDCOVTz6qbycBV9Vt1SfRaOwwORblHDCQy9WLW2\nSc09xAxGwVsl89HOegVk+tAiDz/yYBAWaj1uxRC3zDHth/vfT85ZkIe5ak6sw8eoKm90dPJ1UgO7\n8l0fxML+pLZl3HC/ggKihKwLCgoKCgp2AQpDLnh5mK0BAE794B8BALWfacuIsb2QKHJL0CwGLcYi\n+7UqSUjW0SG20hYirMlK+DD6qfrD2joVfBkRHTHrSf7Qr53ETFi0hpw9pR+FBbcxewQTFYuZKi3W\nMWwJmpOKDDZHeANDrcKIVL/aum1tWyrhyTB9VcHJszCZIUPxoZ2pgAYjzJ3MbyVFTN201SIwFQjh\n3z2mrDonmSLri8pyahhcjkP22EUl23Qg4n3WQjRj9S/69lIABuoLHNHNNPg9mBf4CLBQigyVW8p3\nwTij99NOqBMt+/DqnUGQG9NJiNjqPPRERmSOnRyQsqVtdAgz+a6MpNhPrlUlNF1WY7EsgOO0UhIT\nASOZq268FwBw1U/8HACgWVqBcYX/FFwY5RtSUFBQUFCwC1AYcsFLgp+lnO2p//YfAAD1VnLMiSFo\nGw3zq1VNGUgWVQXYmsIe0P2AzGZt1cAujIcnJcNspwjCsoxQlq7T3qR0zvF+oBmaS7gmbbsh+evZ\nbIZKqrrMSMZIzZAamodlPldVIZnyqwyMsFQjTN8Iy1K65DKT5MVVYzGAsDmHrE5WSr/JPj2CkWOr\nO9CQqVeVQ2BxHPebyyGHELVNR0/WUTwkZNEP8S/OjlS5jSuLjQwjIDkXHHRs0+mmnEqYLguvYhZ1\n4T1T8Q7ELPTCIiwt+GJS2PaMQJgvl++ACsdYeKR7T5/qAPl+dAbRDL2WtSiskgKuZqSTm3PXfjh3\ncPnaGFWgaxNohBGA5esAAFe/+6cBAJUwZVvYccFLQPmWFBQUFBQU7AIUhlxwQURp/3jxh/+Q/j6X\nDCTQJkZkXY2O+V2X2BYlEjW37DK7oth+YJ5Y6JJzlTInMipW7LZxAXGSZDnbreRpq60swlhbaxCN\nCCyIvd10K7VN2XFqrRrZBczW03szsXOshJUba/XptGdPK6/Z9EKtFO3wWTabOnQql2jZJmPoxyym\nF4iwFJBQjwux9gteZSzJPplnDmodHHWsLavMhS1WtpcDZQKUoMhFcnwQUC+U4h+9fKvh+bk/WS/n\nwGBrkuaTLJb5erZxodeyNJ++DyEqbTVzrU1drwqcLWpmxNYmqdRX0xCDbNLI/LRUsaPS9i9NpVtp\nraKWSRWAVjoEVIMz+1un67LZElKq5YOTlia5l0sHb8Kh93447SciJNbO3YOCggugMOSCgoKCgoJd\ngMKQC84L7z1Wn/8+AGDzhe8CAGphIp1LzDL6VhmtJ5OjbGIgc7aakmO+eZ59eu/12BSOIANqY6ds\nkZRqaysxdFbXGjuCMzM5r4y/6rFFAJ0x8E1iLq6mfaKMCzb3/zLHSdYYyJKc9i3nimHOVdDjaVXz\nnHFCDLk3VmU3WNGt/cQWxjKKIPt1Q9MM+JiT8cJwydIsRSq6XP1rNPeaq9B5b8gAOxFaAXuwuwBD\ndVAyczW5SPvMtqZ5AmS/llERGZ01Ua+fdQQ0GDEh5Lxsx6S69PFyf/hsPyn5eoqXkMz6tsvXPTfX\nxkYVrCFbpQ0lSbV1Ti/OsmqeXhe83TYHFWrJSXcihblw5TsBAAff+dOoigxmwatAWZALtsHLArB5\n8odY++fjAIBaQtSt+hlLwdJopHrSLDSy6i7ExdJpLMaGOc1hgbFB9ZDZvqQOT92WKlIxXFrLYjvd\nEs9j5zBj+JUPCBK67ibZw5g/sGoapdLFMS10AGLNYh0Jq+vlWA1Ds0WKC7E+lExbGFl4QKUtLohs\nq7JWW3CMLqC5jSmrX/n+bnkBCgFBwvERqWDNLiQBCi5esQuApBCMyoqxbcggipMVWobTpc1HbpSN\nUdvQuNgyqD/dSGmDaKMuqF4WVBakIbLoLjtkEbn9yKh/sJX6Ku0rkxvku6BKWKzT8johPG7Qhzh+\n0aiy1ld548OUPrB5KQSzI4RaQv6OxW2cqtzCxrD41KRCrUPX3wwA2Hfte2TfshgXvDqUkHVBQUFB\nQcEuQGHIBQqylY3nU3h6/Qf/BU1MDkyUAmRolW0cbTvL4VY9ztD3N/iQpQ01pDgs4okxwIiWdSVs\nrZ0kJmaDQUfHoJouTSKzKe9P1s9k/WNhhjosYdOTroWVliy2NrWie21MYtAAUFfp2JbtR/QVNmMN\n40IZuzA5kYrs2ikaK+IldnthU/ogqKsR6676Yor0As4OSGyzYbFbhrY/1TLnwsZnbadzXmHoShQR\ntcBLC71cZoJpW59dnuSiZ1IoxQhCDMjhA62lkmI/ME3g4BiCD2xNSuhCq9GDOJGpkcItFpeh67aF\nxYN4OPessrS1iT7b2c15u6ymnSvyqiqXmT1DKJZCKxT9cDBLIoP5zuRfvHwoSWFaR3pfUPDq8JIY\n8jPPPIOPfOQj+OpXvwoAeO6553DPPffg7rvvxj333IOTJ1NP6hNPPIE77rgDd955J77+9a9fulEX\nFBQUFBS8yXBRhry5uYkHH3wQt912m773O7/zO7jrrrvwsY99DH/8x3+ML3/5y/jUpz6Fhx9+GI8/\n/jjqusbHP/5xfPSjH8X+/fsv6QUUvHoEydm++M/fAgBs/Si1OFVxhk6YA80QqmpYmANgW7sSi7oo\n+WgDlEFZbeURtsM8ZQyIckz1oBVK101aGLJeR0op7UryDe7MDEbYbotEtzT/ZylUMlNmz0InL+1P\nsQqokbZjuxXmBUKMyVVgbFMSNuvU4cllqUbKP7JwjEYS1g6K2QAgMhcNm1k4231otqFuVEa359wH\n5o5lnmvXYNrm4ql0fhaLAUHyyVl0hHlrybt7rz8Osynz1WbwCgsV6SB4zTTGgAGMZ0sXE/b0ewZy\nbZuId6i3MFlsLgojw2atgfoTB8AxMSxMmcNyxiD3Vkm+m8OgQ5dz7JTTbbrAnHj63u275l04eOP7\nAQBVLREUWzJ+Ba8tLvqNapoGjzzyCA4fPqzvff7zn8cv/MIvAAAOHDiAs2fP4pvf/CaOHj2K5eVl\njMdj3HrrrTh+/PilG3lBQUFBQcGbCBdlyFVVqdg9sbi4CCA93X/ta1/Dfffdh1OnTmFlZUW3WVlZ\n0VB2we5Dy/zkbAMnvv036b0X/wkA4LTitlYxCye525nIUpK9Wud6ogySZzZD0Y6IqAYLzHmS8VI6\n0rcdrHzP2s05sQmELN8YhMFJrpItMG5hAZN2mMtupZVnJsyqqSI219N+VWQ1cHqpbI3RQsr9Opo7\nUMhD2G8EejaHwohF8tGzZQu5cjgIa3OB8po8Zxb2iJQX5f7GIYi8p1oCShW7aonYRiuVbRiaKrB9\nyCGoOcSM3tHMpcIgqm2k7CfCGLZh/hyYbFH8hRaVqnOazjGq4SXvz9YsbfnSvHWHOB22bRk5RzJ/\nFtbabnFG0rGrLGEZ9N7PjTkwAgI1wAha4U/6bFTsg4w4BhqCsJ0KYDxgGlKl9L4r3w4AOHzduwAA\nzeIyTLWAgoJLiVdc1OW9x2c+8xl86EMfwm233YY/+7M/G3we5/tadsBTTz31krcteHU47xyP9uDI\nzf9G/vg3O29T8JLx3v/hs5d7CG96vO+Of3dZzvtW+p16K13r6w3Vad8Br3hB/vVf/3Vcd911+NSn\nPgUAOHz4ME6dOqWfnzhxAh/4wAcueIyjR4+K0H6Rl7uU6M9xJ4ymO3cGAHDm+/8JZut5bggg59gQ\nKxVKYK7TMy86OHb6N60VWTlM8X34gCAsjRWuneQ3XcxMij20Ts4/E1nG6CKisE5KZk4l10gJTGst\nZsKQW2HhGxupQtyOEutZPf0cmmZJxp/GWkvycLywgKXFvXqsBJHXFBERZ5sszgFaNbLJ1eL9v/xZ\nPP3Hvw1bUfyEcpQi2lHlqm3mSMMc4waM5pBphMFpDFakSWFhZtIL7NNrEMETKxGEMJvASUU5e6e7\nSa9inrl4jXSkbSuRGZ2sndXvg9olgnl8yYWPGnQSMWHyWusIQs6f69dJ3qtlPhErZb9aUS7iH1oJ\n7bvB4vC+O/8d/uv/9r8AyFEJmCydyWie1Yp0q9EA7Q/nbnIuV+/H4qHrAQD7r06vbs+BNFbz1ssT\nl9/ky4dXtCA/8cQTqOsan/70p/W9W265BQ888ABWV1fhnMPx48fxuc997jUbaMErA3/wvbjxrP7g\n2wCAjR//FwCAi1uIUszU6aKQfjgbV6GlwAJVt2y/QWeuBYeKTJZiEWxFyYuaegPL69amhJCt0RBk\nJ8fhPpNuilp+6Ceb6+m8UljjKP7RRrhxeq8Vxa4god5za0kHu6nHWD+X9t+7P6VXnIRPq9Eo+x9T\nuKKiFjVDnQExDrWjnbRKDdq6VNkqbVM3cu09JTAKUPABJzZZC9p2Mg8alpaFUTtzap1438r9jfKA\nQ/cmB0ylFYvtW04O0G5twtK1iuFcWZAnEtJvJ62u2YSP4nUs1xMnER2Lw+jvzIcxjtkH1eLm4Vpq\nYtsOlq5Zup+EnnuuUXmueX+kWJADNCanSVQCjuNxsDXFPhiqluOtJGema975IVgJR7t6+P0uKHg9\ncdEF+emnn8ZDDz2EZ599FlVV4cknn8Tp06cxGo3wy7/8ywCAm266Cb/xG7+B+++/H/feey+MMbjv\nvvuwvLx8yS+goKCgoKDgzYCLLsg333wzHnvssZd0sGPHjuHYsWOvelAFrw1aP8PGye9hz+GbcOIf\n/i8AgG3PplcyPTQaUlQxBZfYQhs8+BVRTV/ts8nn8XIsymnSrci0Qd+nTCFbZlRemQx+1sFakR6k\nZrGwt5FrtIWqGSdxho3JOQBA1exJ+1QOvpPwrbCsZiwh2w0RBtlq4aQgabaV2N7ySmrLs8YhxGHx\nkxZskfkjqPsVW6NsRfasVkK9EjcyOmHDEm0wBqA6Jq/RiHRmdDGHsVkoJccOUsxkKyDI+afSdsXQ\nKpmycZVqRE7bFPp3ZMpNhU4iJrWEujfPrck0SrQkerQqpEG97TQsxzak2Ol8BKYrAkVUKAWaZS1j\nT4YyHS/k7cwc+2XZm0Gv7ywhsqisF60x1FaVsD5lQ62LsJYsPL2OjlwPADj8ruRZTCGYgoLLjbde\ngqSgoKCgoGAXokhnvokQpHVksvYiAOD0038Dv/o89hy+CaPN5wAArbAKN0pMqG07RKG/LMqKFJQw\npk+E02dkN8iiGcpUAlmRMCjJIQc/Q2BRmLBpthZ1wn5aU2teebwgUoRCyULX81YWsjSqEzNe20zX\nXNUNgjDBIMySJhlLe1Pq5MzWVI0bmhFFNuS6QlCxEuVjQrY49oiYowhkxsxjSpWZ8RFRIgXUEJmJ\nYIkabNgsHkL2SFaPBjCRrUCUiGQuWdqP4iwbR1DsgiIsdGKaTXNRGWVCxVO6MxbVKLUurm0kZqze\nyxTfgM0iGTKPlBblfbKd1XYtsvqKOViJBnRdl78zFEiJzPdalbNkvp1Q2dNeIpvyrdmoSuoaDHJL\nGYVfeoYYrUmFageveTcA4MBNSeCjtiVfXLC7UBhyQUFBQUHBLkBhyG9gkJV001Q5fOoHSfIynPpn\nAICNW8pKtkxiRBThpymCbRpVnAiav8ush+nMbDTAPKCwxggVy2A1bWzJjOV1Nsm5YlZZC4Oqpcp5\nagNq2d+TlzPP7Cys5AbZZkOzCbbQbGytwXm2IgnblP6rTq55z7692DiTcuiUAJ2spbkb79ur46/F\nRk99d9X7OGgy3WcvxDRnZLpdq2YWrDpnu41a+gXfE7ngRcrcdU49hRmpcHNyo9EHGK0BGFa0OzVO\nyAIl3H8qFe0LS0s4t5py8Asj8bWWdjhKpAZj4eo0D8zrem1FoihKVCasjUm0RuSfMfS+M+dvpWFb\nHVkv26dMXWmEInDOhQWrd7NxsHb+M/m+uEVc8+6UK1644loAQFUkLwt2KcqC/AaF72ZYff47AID1\n55LClhH1KhekuCkGDZsaK2FoajizQKdyuYiJCzx/OGNUD146EKkLDl2BQtCFR9tbZFuqcvlZq2pR\nbEEJXuLBslg1zQi+TeHT2KbFwcnXMzqHOBFday7I0m4zY0sMDNal73gkxWEdf6Nrqj9VWKK2umcx\nVNp/tnEOdon+wdKjLK+ODwoxajuOeidLIZiXbULsgFn6dzWWBU23ZVg4whs+/PDhQ0K8vgaoZmbY\n0iTFZl6K3oxDbNO1RpPC+1wQ+RCUPIMlLUCjLCnYWl89p2HkmVxP00gx14SeydlliQ8UXGv5cBRT\nLoFDSvvPpNdZw/95Ec6a4JyHrCbG0DfU2Srvr1reMh6tL5R/GVdndyhqntcpTXHte34Si/veJqct\nC3HB7kb5hhYUFBQUFOwCFIb8BkEnLS/TtRRyPfGP/y/MeirUImNgWFgLYkIEuZN2K7H1RVqLgp9q\nERFFPxh0tNbqMRlCjPyUzDnGHFLUsKuEqmUgHpWyZ8sWE2FkDLE6G1WcgcdhiDVioupbZOMzYcis\nBeoQUY3S/uurqbirWU7KW9GxKMtjTFGLQM1jYaFbU7iFWq+7j76IIEOrVj1+hSGKSpiftghIY6tF\nKYsuS57eyT5k1yspRDN0j3IeVkLCXZc+q5us2czjdZuJiZpGxhOZEmh1PlQAQ1jjuRdPAwAql4uZ\nWkxlHHKYSiExAAAgAElEQVQddEBqA0JPSzx9SG3yPBdGfZXpkCUvNIQyVr97phrOawg9feu5AjQi\notXCNbZCafEhC7isRZD5rPYnDeprfuJWAMC43rftvAUFuxXlm1pQUFBQULALUBjyLoYPHeJGKjp6\n8Tv/EQAwO/0sAKA2U3SRWsnCkmQ/ylyGCKW7JDJWGCJFKmwMur3mXF2vyMtlx6Z80MwUQ/Aqd5i9\nfSmVKDSpqVUGk85BTtisEzZpvUeUryPbhMjstlZXYYRRU+SC4hDtNOXLO0REYf/Vwki2pQSoXDui\n+u6SvTYyxNn6hnoj16n+DV6cmFSysUeVKVnJYizmfWcbW2gWKBMqBVJyZzxfu07ntZNWNddRQtMg\nyL87yaVrmxDzo5NWXaIqiY8wAhFEqCNWDrXM2caatDbRtWnaavSCQilB27DS5YwXR2jpRMXvBYYI\nIWpRXVQSLWOVaEvwRovKqKnt+4VfZL/CzFljQN/qaIIy4uwiNjyHtw32vu1mAMDVNx5NY63LT1vB\nGw+FIRcUFBQUFOwClMfIXYRAMYVZYoEv/uApTE78CwDARjFhoFexT7KEQF9oAfK3vG+sClBU0sLS\nCiMj2wgxINAflixJ8qOurtFK6422tzBHKSzW+qD0iIIcej0sSLZW3Y1qMYCgyAQZfNcGBMmZWrF7\nWl9PedJZdPBTOT9ZorZIyVhDRBQqTJej2ZqwabkEW3nUjXjasnxcxlUtLWEkhhX0AR7VZHlSgdxl\nZqcOSBRRmabXdrKBeuGAzNGwsj1I3toaiyiOUBwbvZuDbQBh/Yw8oOK9SvewigFxJseUlq5W5qEx\naZuRrbB2Zh19aARj2mlFuVZDM5crlzXdmqIZp7nqVLpTRFwcmX/Q9i3DGgNOa+xVS5Pt8jpCNqKg\nzzZGTDpjcByLEaLNLVBpv3SNTmRUr3nXT2HPoWvSe26exxcUvHFQGHJBQUFBQcEuQGHIlxFBDBe8\nsKxzz/43AMD6Cz8AANSzNYwC+0XTPn3RDGXC7BdVj10m/QxqkaicST6QaV1lX/035+BnLduE1b/Y\nTxIzplexqUyWbZQcLplYJSzY1lWu0mYqWvKjRiqy2+mmHidX3qZtbVNjtkWbxkbGJhKYzaIedyo9\nuNUonXdhTxrPdC2JYETjMWY+U6IKaqxRO2XbgT7MSAxxGqa6rXo+S75ZK8rF6jA4i1YqwKPPfccA\nECRJ33UelbB49sZOJLrQzHIOnWYdYULDaRHviNmS0IiQB/PmUVjs+sZG7/uRdifTNa6X369ofEEb\nRhpJBEwlUlNLvp/H4XGrqsq9xWSmLLZ2ZN42exPzflpWS1tUIuGqfcdz28BGWDJjsQkdrdwIALj2\nPR9M4xgvZyvGgoI3MMqC/DqDYekw28A5EfTYfP676UMRVWgkjukxypFVEYeIvaiw6iHH4d9E5Zyq\nNZ0PxhgttJrrbkH0LWaTNBYri+VMwqk8aeettlnxR7QaDdtTYF0WmWBbjBT2dLN0vHY6UUESDXuy\nMCh2eqwgwiC1CGLkhw+nBUX0cHbiZMR9u9kMYUEWE2nToWKXXag15N5KqNvJeAJVo+A0FMtWLytx\n+elUdLSd0fEzDN1xbelY1OV7vr0SMpfwcDttNTRurPgyq1a4LPrR6MLHBwsnC+yWLKLJu3mYQuC8\nVDC6YEZVv0ovUddBCz6utPLwNN6zR8YvhXiw6d4CWbxDXqjwVlUGli1NGo7OuttcrOkExe+Q1fHU\nCKJFfeT6nwAAXHvLhwEkf+uCgjcTSsi6oKCgoKBgF6Aw5EsMlTKcJvb74g//KwBgcupZuE7aUSR0\nrfuwSAuz7LbEIpm+hjQ/mmNCDBW3095x6ZbEAhvuYrOrTwTZu7TOxKm27nRTCWlKQY0X9oYQ9Jjs\nL3J0BdILghaF6fFUPCQdZ+ZNdjfi0IR1OQNEKfTyGmlN5yILdSEgOrpEcWjS9rQnMazu9AyztRT6\nbvYK2yMT8x6BtExawzZFitM24qM72q+a0aCzFC9RmHpsA3w9vI7YsVVMmGXXIRgRNqFr02IKvbeb\na3pz4kReyWYlZB2dUW9fFl614nql48MsC7OIlKdt5d67CjFIhKHjfZX2I0ZUmgio57PoQsvx6kWZ\nz67T7xoFPRzlLCseNyoztgwKSAg6WqOhaif3LlCoRbQ4q8VDeNu7/jUAYOlgksAszLjgzYrCkAsK\nCgoKCnYBCkO+BGh9hyBGD6d/+C0AwORUal+qO2FdodvGbOdSwPLevCPx+T+jSANzkPM55fRmGLwa\nGG1l4qsWKPktdTwKIoQRXGIn9TixpNlspgYB9DhmHjDSwQJR2S9buphjZO616zqAhJ7fSrK2qsbS\nUmKQG8JaCVdlsQkaHFTCOidSgEYGv7y8jM3N1cH+LKoK3iPInFBQY2ua7uFsPQ1sceEA2k3J0dK1\nKie80zi8780/k/LCcFsy5KACK0aUSeJWGmtd11hfZeRkWJRFg2ZjHZw4Uk22RFjEDQungneYMa+t\nBVsinNJU2cdZJStZiACZz9w+BtMMtmVdwuLiwjbZ1L5kJselcpgs9NLaAgNTDZPX0aXIxf6rUuHW\nlTfeoq1yaS52+E4XFLxJUBhyQUFBQUHBLkBhyK8BmCeenXsBAHDuu/8Rk3PPAwCc5HwrmgfIPhYe\n1kilMSUwmWSj9GXwO7LmeZAdzWazi2wJRIqJMF8cOmW/ZHAQ+UUfLGbMCS6Ib648wxl5f7y4V48F\nkcGkVaP6NU+myr69fMa2oVZYrKsqBI6tJXuXHPKo0Xz0wlJiUFORgyRstNm6TzauJVfJlh5TW40i\nbG0kZtmwBcdEFb5gA81IvIKn08Sqt9bW4WQks03J54rYBdvC2naW2adAvaTpKzyZwUprlpnKNqwy\nRrbDVDlLyQtbmY9m1GAqNol11fNqRk9eEk6vh8Ix9H+IzuijeBaTGQwZBjUgsprM+VpWpKtGK7Ag\n34stqb5n+xFNQ6zLlelW5p7HDc4gipVjtfcwAOCa698PAFg6cGU6Dhl0QcFbAIUhFxQUFBQU7AIU\nhvwy0QYPI3nQzdOJEb/w9N8AAMLmGQBA7aeozdCeTxkIZSaBbHZv+iId2T4xRrstT2zU3N2jkpxt\n1wlbC3PHcU4F/aOwMytsS+3+4DUPyNwg7RiDbZRdkdqr+L/mbgMaydGGCb0Qhel6Vm13PaMC098E\n9YJU7Latsqu2neuNhVfWybmppcd4Uywa61EFD7Iy7s6KcmHwdY3R3mTJGLZYJU3XDQsjEQaaH3Qy\ngErONZ2twnoqm0Be0zmmEgGYtFMsjCmCkl5r9i5LVADGI6h1oLBQzr2LasbgWaksERAadEzaLa38\n9sy9zpUadAbopGe7EdEQioDEEGDlvJpDZvU9W56rGk6qza3sV3HMKt6R7j8AjBf3yLG9XiOAZPgh\nOWgjupyUwrQL+3DF298LAFi5+p0Ach1CQcFbEeXbfx7MZAFjqHm2kXyIz/3TP2B6OhVoVV1aDJyE\nbBmo9DBa0LRtQd4JdriQqtBHiFkkQ6OgFFVwGqLmDx2h/rEhwsgPo1H1KCmukrB0CAGQxZWh80qd\ncpwOjQvquGZ7i9fRzOSH34my1HTCYiKGriM69REetkbpj7tzcLWEW0UerJWHh6bzcJaqX7KgywJd\ny2LThU5Dwypm1mUBCiDNq/oQyzmm0ga1sLCAjoIXhqHZdI5mJAvy5gY219dlHlKoli1q6tO8NdUF\ntBHvZ+kUg6c4StbJUj9mhmYjAuxIQsTy35N60VO5325U6XXwWY7XzvdnvtNFmw5fHJc3WRmLuuVc\n7Clm7Wqngip27tgqomWtnpcPfCNpw5pJOxqcRS33yrv08LX36ncBAI7c9D7YWly/bAlNFxSUkHVB\nQUFBQcEuQGHIPSija7ew+qN/BABsnf4RAMBOEzOyXYQTUYcg9IRhVMPSo+izUMQFmfHwQ25LLWgW\nw6SxidaxFANNJ7OeATLPP0Q0HoaDo0CJtutwK4No6JJElpTD1I6tKiw6YhGQhGgRo0opdnJsFplN\nJtlZioIVthm25/C4dd2gFYEU6huzJcYZl0U1pPWH23g5ULfVaoEWAwaWBVchC3Vkec+0v1tPxWFh\nc4p6UdrGxFmK52fY3zmHsbTgnHrhFABg/6EV9FFXTvW+vYSug7D7Tqh7U1sNwas5keHcVbBVavFi\nsSDD8k50vJ1zykwrsl9GBRjmjwbNIp2thMXW9BWOiI7RENGSZlSFRVZ1pbKiLISjeAjvr3NO8wP8\nOvG7O15I17AZPeK+5MT09neIwMe+g+l4xZmpoGCAwpALCgoKCgp2Ad6SDLkLFFxIT/PtWirGOvmd\n/wwAmJ59Ds5vDvYZFFdxf3Ul8txIPg45n0kZRO7b83uN2YYnHUediNLbvtf2VAmzpNiDMTEflO0x\nWgSVx+FbKVASFmuY2+Y2TQVWT7GVSdmzC3DCnmuhYIGGCYEGGDk/HOVcM7oBCZvvZp3mjMn0WSCk\nAiXGoqkaXkCaB7bHhADDAi3mhZuUj2QR0Gjs4KW4rZFWnE7clox4B5vYy8mTyUnOc/XMWSxJspeF\nVnp76Lk8brAp7U3NUtpvXYrDajJLO4KYEmlrUweye7lPJqqrEg0sKJbhqhq2kVai1RSVYX7VqpiL\n1/uiDFXYLP2rHYwyYmWtlSSzbf7OasHYaPhTMB4vKBPO7kyUX+UNcgDNLZgDNiLb6hJDvvq6ozh4\n3U8MzlVQULAzXhJDfuaZZ/CRj3wEX/3qVwfvf+Mb38C73/1u/fuJJ57AHXfcgTvvvBNf//rXX9uR\nFhQUFBQUvIlxUYa8ubmJBx98ELfddtvg/el0ij/8wz/EFVdcods9/PDDePzxx1HXNT7+8Y/jox/9\nKPbv339pRv4y4SWPGNspNn/8HQDA+okfAgDCJAk/1GJ7OI5eTQxyrpUGAfmYmTUL81HmHRHJGufG\nkSuoQ68VSpPHg+OEmPN3rVQym7kWKQCZUs+rbHZeR8A2o9zmIvlEGE3oBk+fW2HM1gI0s5BctApg\nSMVvsFElEWkYoSloGVazMIaVa+N+dk7EI8QAZ4dfR7WOdLXm5Dk2ylw2InPZdZ22CykLdrTyk/aw\ndprZXi0mGfQTHjeYSHV4I4IgbO3qZPJ8r9UMS6l96syJlEuuF3ghHpUcm+fiOaK81k2TIyhkofJa\nL46xsSZGEXOylBTWsE2V8+sSeZivDbDWajSC3yFGdpw1WkFupGq+kdYu3ruqcrkinvPIMXMOXA3L\nWgkxh1g6/A4AwJF33pI2XVguzLig4CXiogty0zR45JFH8Mgjjwze//3f/3184hOfwBe+8AUAwDe/\n+U0cPXoUy8vLAIBbb70Vx48fx+23334Jhn1xBGnzaaVd6ez3/wEAsHXyWW1XMqACEltXJCQXQ+6Z\nZNuS9rTuEFSIeSHO59/+HuSs6f3egs4fNR4ucvGpszcuQ7zaRhV14d626gtsF/R8GLH4hmFpjiOi\nYjEU1ZuoxmUCPPtKeY3zrk2zCFRDc/ks75yO0waPZiwLYKRS17AlyZp+z7WEQRnmdo26VWkP65zp\n/cJ4hK2ttNjzwUA1pDu2h+V7Z6S3dsTCJw9srcp2TCGwBYjX0QV04g893puKudypc+kcfGCqhufp\ngw8VzjkYubZKwvT1QnqdtZ0ukmFrmO5Q96tmlPvBwb50hpWpN52VyjhHrmbxnslKWlLcxgedqteO\nlvWx2eeUXqiuBdsg7k0FWlffdDMAYN/Bt6eP5DgFBQUvHRddkKuqyqxA8P3vfx/f/va38Su/8iu6\nIJ86dQorK7nidGVlBSdPnrzgsZ966ikAl1YwvtlzKI1HXnHjv7pk59rNeOcvfPpyD+EtgV/84p9e\n7iHsClzK/9PFYOLSo8zxpYMx8/0wGa+oqOu3f/u38cADD1xwm5dyQ48ePYoY4wUH+HIQfIt2XRjx\ns/8EAOjO/RgA4MI0j40sc05VKG8QM1udY7rGGP131Eaj4bUG77eR1p23nDuH/N1IiHHWtoPzAtDQ\nb4xAZH+PVnPlUDcAGujgnR/7NL77f/xu+kyLyvJIyIDYtqWtUnItABCkYIvuT9OOnsn5vCw0YnSA\nLVI2ZLbF8Ok6vYbJrgFEFnGJyhnDsqHLjFOL3Oq8HzGTeejEBzqIZ7FObNfBRmHP8pDJ+qSpb+E3\nE8tdX03foZGwVgp6TDdnWhRn6lRUNjmXoi3rZ57H//TlJ/G//9onUC9KyJsKVZJL2Dh1GgCwZ/8y\nmn1p/yUpQJsxvu9qtHKOVgrGJmtprhaX96WhjxudGxM455I28NLGVTmMxPWKCmFulFlxM16U07H9\nKUUweA8ishgL2byVL5SvUxTswNvfi6tufPdg/0uN1/L3omBnlDm+fHjZbU8vvPACvve97+FXf/VX\ncdddd+HEiRO4++67cfjwYZw6dUq3O3HiBA4fPvyaDragoKCgoODNipfNkI8cOYK/+Iu/0L9vv/12\nfPWrX8VkMsEDDzyA1dVVOOdw/PhxfO5zn3tNB0u0wiastN7QWenUt/4fdOfSQ4FVOUtheNpLFBHB\nvLAwTApx2Px8EuaKsrTFydqcu51DZs7bEeZYcBrbEGSPlEgMIeRCmrm9jImqXxzs3LE1l5uZPqUv\neY3Mixpj8rWSYYt3ceyC6h+zyI3cnzlcH0LPqUhyv/RlpnuUMyqKwVxws5C0j2et+PlWBpYtNJL7\nbEPOM3P+yOhYKEWGba1DBUpDcj9KVUoOeBZgZK6UKcuwFpoKayIIQvnHToRN6Atc1TWCfOco0GL3\npG3PnJT8c2U012sYMZDvq5F8bYhRPZenEnmwDVuSKjgZlJf88sJ+YfPCQpvFMQK/By0LzsT72bJI\nq9GigEpyx81YhEWaEYz4KbuK0YhhyxcQYaRQy4sE6L6rbwIAHLlWdKcX96Jyb8nOyYKCS4KL/m96\n+umn8dBDD+HZZ59FVVV48skn8bu/+7vbqqfH4zHuv/9+3HvvvTDG4L777tMCr4KCgoKCgoIL46IL\n8s0334zHHnvsvJ//5V/+pf772LFjOHbs2GszsvOgm25g/bnvAgA2n/8eAMBvpUrXOraa5yILVobM\nvLHtVSf7IZfttyr1GXH/daf88DySkAX0WECPPfePK3ma6jx+xn0Grfv3c78qWygtWcy9ahtWVGap\nUousoJY8cYgBoWPLDP2IyZAjgjgvxYXECGlKQNbp+mOak/BkxW+MOZ9Nph2EvbmGJg0d7LzgilVa\nnquxhQUHEdvg9XkfYOTbzALfTsalFd0GoIqmcZJXldFOJ1OMJJ8bJAddqbyl5KZ90CpkW7NNKf29\n90Bqg4rOqJkDB0T3rEWXcsBj6+BlXis6MTlKWFpl7+O9wtQZEWKlv7HU4wAMHaFo/pH2ibYCGrnn\nUuHuKAFajxCECVtJojN3TFbsFvdj/1U3AgAOXpNeq1HKO1fFCKKg4JKgSGcWFBQUFBTsAuzaBBD7\nX73k8TZOPgsAeP6bf4Nqkhixo/wfk6kxItrcp5v+geHf3fb8704V0UbV8pmjE0a5TX3jPDnjOUa8\nTSAkBM0ZzzPjC6EzZKgxM0jCbz/XfA7amiHTjrMWanYsL6Fjj3LQqmrMZK4pseiUB2u1OmMHzGmT\nK9uQ5zPSDpJiJBS9sDYzfbkOJ7Q6MVP2T8s10iKRFeHBw0nvuaV9ZCUMc5aqlF1lEBlBEdY5ixSM\nmaESS0Uye7cnpVy2NtP+Y2MwE0ZpG8qTptflfSmFU5sKtYqvNJyiNJwlqYSeTWGkyrsW1hn69Qfs\nUZZK8kokQFtW3IcONb/zwvRrGbuRvLOpHSDfL8de66Bel9qHDJeuce/h1D+878rr0uuhK+HkmAUF\nBa8PdsWC7FWZKqiu9BnRlZ6dSW1LFPOo+y1J4H55CaIPcZxbiLWAq7dc7dTSJP/IIWcuSFzXeuvx\ndhWuHY59nmuu61oLrc4H0z8/39OF0Oh1Z8/lrAsNJHcenj8vgDIfLUP5QR8y2DrD0LcPubFLJco8\nFaa46GxXpFJhEdl10EbBQjTkhTjtu92Rid8L5yodt94ifWCS43UBrEdSIQxZ2Bo6Ivmg4d+gCmSy\n6FWVKJsBS3tTwdn6hrgsUU0rRtTUfpar62Su6W5U17VqVts6vTYsEpMHhvFoKU0coEVzFGdBMCpa\nYiwL6OhCJYI2xsJUDKfLA9pcUZYdjeCMLNKi0d1JUdjiyhGsXH0DAGD50JXp/FK45qgnXlBQ8Lqj\nhKwLCgoKCgp2AS4rQ+7a1Kqx9cL3AQAbP/4O2k2yANE8jmSRDBn3hD1MHHwWQ8jSlnPyhUrQfNgm\nWsK/grLAmEUyqI/csfjFbCvQ6voC18i+sTtBw9TT6Xm30TH32n0u1EpFRkzno8yqbRYLUY9lDLaJ\nMarwBR2gIpXZTFRP3orMth22iFlrYRhhUEcnPxhPCF3P/YrnH7aVWWuVEfPOZYadtZqpGqdFTbqx\nRaCsJ5k1ZSClxWi2sZnZM4v0VC416zsHhpPlXjVyfZtbaxgvL8nphi1aaGRcDqilmIoFcF6+w/VC\nFs9g7MHIOUyvWC5Q3pTSrsLKx2zrqo22fzmX2DeL4yBiIBE1zJ6knHfwmqQvvXLVtWkOxwtwrkhb\nFhTsNhSGXFBQUFBQsAtwWRnyieP/J676qX+L1X86DgBwsUNlhkVZrEPJbT9eaZ5mSCm8AKMFRcwR\nKs8i+4o5j6pQy6HMRsMcW2NRlzdGW4b43qRlsY3kCMej84p+MG8ce9fE/Oo8cw8x6PCV2arXg++1\nZPG9OV9lH9TBif0+3fy0mKhFVHqgQDYdetvJJpQb7cjwRipSoYO0lHOkzqXTIj1D5x9JGQ8K0CR/\nOZXoAc0RAtos8iGmFpR17LSQywEScTGSj9baMmlRsovLaM+d6w8RUfqwgnGqoGp9kqxcWEqscyKT\nXi8sAoZ9U+mzKKGDSlqcxksLYEynZn5YRD+aZo+MeZa9hTXfz/+KTmfFSu63levCSKIEoyW4Oh3L\nSeFYEKGPRkwvDl9/M/YfPpK2eZ1kLQsKCl4dCkMuKCgoKCjYBbisDNl2SQjfIOWLPZCtCOctDXt2\niHyPUovMR5pedbTmKjGUSOwzsvlc8E4gUyaLdhMt7FVmGIR2cjJTA48wSMvWnTl5zhByYvulQKl6\nvgI7V7k8L/PZzdqexWN64S4dWTGMymBmxi7HsRZzpF3ngXlieK9SkSTEvAvcxlirIiZaI26Hldg+\nhBzFkArkVgRKKpfnSRkx255UpCJuq9imS2AU04nReAl+cz2NTY6tFdQQqVBkCdCR5KCnVa6knnYv\nAgAa+ayTtiknk9Y0NTz/vcj8rvgjy2vVVYgQNk+TiMi8sekJ+0vbGES6U/LEdb0AiHFFtS+x4MM3\nvgcAsHjFNencRdKyoOANh8KQCwoKCgoKdgEu62O0slfaCCIiKkPmq3yiydPch6z9tz0TAsKwepZE\nzpyfDWsl9Zx0Y38cRqu8DbxNTGd5b5JCxIvJ0IKKieg8Amf2Aux7Xl5z3vIs5ZmHfdVWzeKt9iGz\n75WiJ1mUxPfS45RYJFOl5GSWAt2Jr+t5OUeW0Yher7CdY9a8Zyx47x2PldTbpUXztSrTZfV62yFI\nFXKkHKcfjiNdr34z0m5kwZH91R6jpdQv3G7Id6fleAKsVFdbJp/l0I1UaSMAM8/q8vTeiBaFMeWd\njY0Yi7Uhe4zpQ8nRVeMltGy7pnQlGO2xg8pxAKglB25E6MQuX4Urrk+2h/uvTIy49A8XFLzxcXkX\nZCnQ0dcIGC6KfvjjHLOihP5bF285ng8h6zjzB44B5p5w18UCxdGZHH71+kQAAJhYi8X9BwAAa2dO\nDc41k2qekYmw3I/KWvLjGrRV6/wwuUprm8iG6V1I0EKz4QLCQiVv8sOG+iCrmhbnx+g1ani7Fzgx\nshiox7KsskZXXY8gLWE1RylRZNOxdSxfszpJ1cMWLWOAGNm2lLalCEfXedTyQBZkUYK6NUl42lp9\n2IhScNZ1LE6byr4GVgQwjDxURcP2szY/EDGrEIf33hqD8SjdezOStiN9gFyQ9/fonLGYi61IDK+H\n2KmAR/ai5uIdkTuSZAEW/+F9N7wXAHDVde8ohVoFBW9ClJB1QUFBQUHBLsBlZcgUq8h+vFG1ibWY\ny5NB5dYgMuNh6YuEeOf4r9aE2cw6w7wcJbftFXCxMImFYyGyrWUJ7SyFJ72IZOxbSaxpcz0VqQ2O\nrgM4PzOmVGJQLephWDedX65Z+5byHNFBidurwEc/HECGPic0YqxTBsuOHg1zmwg+s+VwtNwHZb1R\nWWoAHZkY1t3ezsWIh5aU9Vy0srwnw/T0Tm7QyZxDoilkz17mw1mj/tAqlEKmi4mco9Y2of2HDwIA\nNtbOAgDW107rZI3E6zhWS7IfxWEmGC+m91opxnJ+eB3Nwh6dSJWzNKIpzXRBCKjkvx6fiJkKgLEI\n4ti0dCgJelz9jqPpmpf2yhwWt6WCgjcjCkMuKCgoKCjYBbi8OeQ5j98QwjZpR2VQvW10/x2Od74W\npn7B1jYzibnxALkIbBqF3UjRzGT1HFYOHAIALF15NQBg/dzJdDwvuUpXw9nMZAfX2jvnfBQg53C5\nQd5W5SD5d98Xma8UBKFZh315z1tK5ndIsm+ba3m11uYIg+8x87njzB8zhO1RgAsNzI1THjXORFpV\nRkApTSC3Lam8psznxKd7t9QsoRFThrNnT8p1pL9XVq5UJ6xplxzGrBRzLS4msY3Vk8/q98JJLpqa\noiy2s02tBV/RstAqt2YBw+JDLWesktBHtWcFR266GQCw/3CSuiSrLygoeHOjMOSCgoKCgoJdgMvK\nkClByVdrDcKcX/G82MW8vORg2xDOu30/B52ZsTA6MS5gBbSPBkGEGpb3pfxwy0rixmH1zGkAwP6V\nxJQpjNGxyNj47Jvcs3QcjD+GnDuvyKCG47IhqAoJW5v6sxOVCUsl9wVarHSfHd/VgQ+2MTvwV1Zp\na2ZzrKIAABGYSURBVPV58LBqhpBAIwhHcQoHdCzTpsgH/aV796fryYr2RoXOG5XspICGp4SmsuGI\nKPKkXiIVUXK2i3sPAwCsCdhaTfaelLWEHG9jdQtRJCobMYGIkqud+uSHvP+qa7G6eiLtLzn0SiQr\neRxnF3oWmbwQ5t3ZIhVy9X+TctIr1yZWfOT698CNig9xQcFbEYUhFxQUFBQU7AJcXobMf5Al+V6u\nVT5S28B+7vU8loT9/PA8y+q/H3vnA6ApviA52w41lvenvOH66dPoH/DAoSuwJRaRZ88ltrVXtt0i\na4PPucG58yuHDzHnXHEeZhu351r7EQAzlyO+UPRAt5nLzfePSceFnfLw/ZwxkBmyNVZztfkWUdij\n0230ug1lOoXd90Rdsl3jMLqQ0rPCumWbbpIqp+NEcuuuAmLK/Xrp8a2XVuSzNJ71sy+i0bLmYc/1\nwtICOmlJdqOUs6VQCJn+1nQLe/en6uzJVvoO2FpMJmbMJUO/WEb7jyn6keagrZewRyQuj0hv8Xhx\nWc5Z+osLCt6quKwLMpWYLlSoRYQ5z+E++uHp8xUf7bx9+tsEajHL+zZQfApWwskL4/RD2cUW+w7K\nYr2eFoDTp0SpSwqGrOstQBw/w8vIesm9R4z0MveAEHuuVDteV5gPiw8VxyiU0YedX8T7/76A9/I8\nhgVb84EWLrbpL98GoDZznw1V0boQto1NH9RM1O2pK0avYyptpc/S/qM9adGkfvZ0LT1U1cbDiEJK\ndMPWudAFGAkVB9nP8b+HLMi228J0K+23vD8V9K2tJveoSkLg6XlDwuoiPtJKcde+I28DAFxz3bsx\nWk5jrIvmdEFBgaCErAsKCgoKCnYBdsXj+YClzYVdtXCpx6Lntad32v5CLC9K2DQz0fQ63pO0qbdm\nE5w59RwA4OCRxIT8NG10+tTzKvTQLCRGVam3rYS8g8+hWLYuqR9wDj5nz+WhxEkQwZHQ49AqCWrp\n3xthKBad49jDP3ttYDkqMCeK0me3fScqOdJ8SxUZaU8ME9hW/EW50PRXF4MaKpPZGsdxyakR9RpV\nCbRmLsFTDhpGJDQ5jW2T/jGdeIxWrkjHlBO3q6m1yYl6R3AWLV2ihJk6Ya8zH1BTVpMKmpwPeXVV\nrdKhm5PkULbnwH4AwMa5JDASq0W4hfTegatvBADsO5Lal0ZLe+Rwu+K/XUFBwS5DYcgFBQUFBQW7\nAJdXOpOFTjvkhy/U7nQ+MQljzDZmPL9thEEnLIc5x+W9qaBmYy3lAw+s7MfJaSoaOn0qsayVA1cC\nAOpmCVbaatwkSWWqqYMaW2R25dQLgrKQvbwwpS89c6Xzg+1dDdk8r8iYnPOeCxSoPOVOzlZuvhAs\n9HUs5YURCJ9b05jP3VZItt0ZS89Fr1+YHlOfd2lKf1tYZcY6D13eNsubSqsYC6WEbS7uGwOSM99a\nT2y1kby1l8kzdoQDK0dkv6GcpbUGq6viz815YzQClOKsUdFIRC51Yyv9feDt7wcAHL7557D/iqsA\nZP/igoKCgpeCwpALCgoKCgp2AS6vdOZcvncnH+KdGO88a96RCfI488e2Dl2b8n/LB5Owx7mziRmb\nHuM8dFCqaNdXAQCnz6Sc8tLiXkzXNwdH39FS8QIsNW8iTHSe4vIaBu4S3Gn7dvO59GwcsUNr09y5\nYozbctB6fvRJ+oVER3ZuNQPbukyA98N7TYOEShh7aH3vgqVanS1WpjcOM2S9bDuy9Ribq6cGx2SV\ntRf3jUOHrkJE2t7UrIqmDaLBvpVUFb32YmLYwfRamQDA1Zg1KZqy8rbkR3zgSskPL+9DjBGH3nbD\nBeapoKCg4Px4SQvyM888g09+8pO45557cPfdd6NtW3z2s5/FD3/4QywtLeFLX/oS9u3bhyeeeAKP\nPvoorLW46667cOedd17wuPOLpbH2vO1NL7UlJ2KofQ366MqPqvcGRn7EV0+/IPukadgrylsvnj2N\nAGlV2ZscdrZW07aTdprVoeiEJD1S/euwcyPdcfzav7tzoCKYHdq4+NDQ+2A+jNw/5/k+2xnbZ/d8\nDw2D81OJSv7WhxCXt7XcQfyQgxRaqe5zBZiu5RGHJzGu17esvWoAgGZhEQCwtbkJKx7JFuxnTsdp\nRqmYythG1be4WJtemJ7pgHos3sai3MXe6YVDV+HKd/9Uemucjlm/TL3wgoKCgvPhor8mm5ubePDB\nB3Hbbbfpe3/6p3+KAwcO4PHHH8fHPvYx/P3f/z02Nzfx8MMP4ytf+Qoee+wxPProozh79uwlHXxB\nQUFBQcGbBRdlyE3T4JFHHsEjjzyi7/3VX/0VPv3pTwMAfumXfgkA8Ld/+7c4evQolpdTSO/WW2/F\n8ePHcfvtt5/32NtCrSF7FV+I2e0U4uY+OdQtrE3YDRny8v4D8MKKpptSoCOs7ezZJCBRmYBmIRX9\nzMRdyBlpkzERNs45ObHNhyFWxByunW9F2qF9K/v/yvskiMZuC3n35+diTkkRL81Nid1TmNMRhzHK\nSHdygOJ4tD2JhWv8rOfvbFm4ptSaF0sdcweDVo4SB69Af25kIHJO32YFryiuTkZazawItRjX6L7q\nyiTHsYaFZ3ksLPjyojN91TveBwBo9q2gWdy780QUFBQUvEpcdEGuqmpgcQcAzz77LP76r/8aX/jC\nF3Do0CF8/vOfx6lTp7CysqLbrKys4OTJkxc89jU/9z8CAG78t59+JWMveBm45Y5fu9xDeFPgYvKk\nL0W+tODVoczxpUeZ40uHedvfPl5RUVeMETfccAM+9alP4fd+7/fwB3/wB3jve9+7bZuL4Z//76/i\nHf/9r+A7T/yv2z6bZ8gqPdnLMfPC+t7A2RVJWLC48OwV16Zzq2e0EGg8SvnHyfrpwfFMvQeLI9Ev\nXj8zuB4bAlyv2AiACnQElYXcPkbCM5fddZlRz7FpIlbVjsIoeuyXkL+01uKWO34N3/z3D6VjSptR\n/+7Ydjsj5TmZaz0fYgw0blLXKf2sJxdqhX3TWYvdV0a0m21VI8ySFCnzusxf+96w9FZT5lK+D3sP\nXY2NF1Oenw+QnHrWAxw4dKW2SVGgpA8vY4mjJBBz5XXpO70gUZ+LYegkVnApUOb40qPM8eXDK6pI\nOXToEH7yJ38SAPAzP/Mz+O53v4vDhw/jlGg6A8CJEydw+PDh12aUBQUFBQUFb3K8Iob8cz/3c/jG\nN76BO+64A9/61rdwww034JZbbsEDDzyA1dVVOOdw/PhxfO5zn3tZxw09g4Gd8ssAAGOUJanGRo8h\ne7ZCSc63k3zvqvjgmui1Mndzc0v2olexvLZTwEvueM6px/VbiVT0Q9gfWaDJbDfLLw73SYcbttUY\nEamYN17oHyf0qrVzk9DQKKFvZLRNXnSH9ilV4FQHp2FOPI17yNS1OtlcSKhFXn3c1rZEjm4kz4zY\naTVzzjJnmdBKvI7DXLU6nZwMDEyVmLCbk6a0orO5ubWGhX2U15RzUEpzvIT9V1wPANh/5XXnuaKC\ngoKCS4eLLshPP/00HnroITz77LOoqgpPPvkkvvjFL+K3fuu38Pjjj2NxcREPPfQQxuMx7r//ftx7\n770wxuC+++7TAq+CgoKCgoKCC+OiC/LNN9+Mxx57bNv7X/rSl7a9d+zYMRw7duwln3ye7Vibu3fn\nfZCDylOa7T293MZ7zGaJ8Rw8lMLldUwSmFtrKZwejUMj+cO1WWrLWlxM1bS1+NdONs6oKQTFIXJ+\n1fR6gVmBTBnHzFSVCc9XSZOaGqNSkdtFQcnKewxZ8qLkrP3j2vn0cq/SfLswyvBMET1hE6HqdphK\nHmC+ZzqEAFPpqPqn13/YaBBZwcx5ZfV2lU0vgjQuU87Sc8yV69tcpG3km2tdinb4rtP7x/2VoRvm\noiNWz6ZIiV1KBhArVyU2vOeKt2Hv3oPnv/CCgoKCS4xd4YdMRGzXtc5qXL0Fce44fcUuhjZPnf4X\nAMDi8n55FVWu1RcRfVLq2n8gFXpRm3pz9QQAwMGqLrUWN0g41++gqjVfwLajOpdcVqR5/Q6BXg0H\n80qt1YW0bVNLUNZenpceuTAutO28e1Z/EY/nCUj3BUrMvI8z54j3LMa8kM/pfreybV3XQCfjmTul\n67V46YPJXLGZqyrEdugyVcmDF9MXESPsvSIpsB256WYAwMLeQzteX0FBQcHrjSIzVFBQUFBQsAtw\nWRkyWc/55DL7CLHHfhjqlRDrtOX+BoiJ/Y5G6bPJ5ovpo6XUI728bz/OnExtTot7Uo7bz5LLTwX6\nEEcYYWvoi30AiC4XbxFZhITFWP9/e/cX0uQaxwH8+87p0dHCpm7ghRSxTruYkdRFSfbHECooCBoE\nLyNIVGQigbgRA+9S17oQvagNg7AuAq92ESXSTcQalDBqN+LdiDCX4tqYkvKci+l0usPp6Dk878b3\nc7eXR/zx9YEfz7PX5xFbh3vsONBia5s9fyWc/bndx2vmVuibtxzlXsoq2zZ+xy/Dto87Vptb2+Sb\nl/4KiLXCx3xC2b2OF/m70xvb0Zv3Om87ZAPYuuoq/5+ssmN2vOSm05UB5fljdOVbf/Pcuda6zfuM\nN14A21iF68v0UP7IfvWA9c2l9sbPlGef1//ZBKO5ITt+4wUwIiKt4AqZiIhIAxTBI1mIiIik4wqZ\niIhIA9iQiYiINIANmYiISAPYkImIiDSADZmIiEgD2JCJiIg0gA2ZiIhIA6Se1PXgwQNEo1EoioL7\n9++jsbFRZjklIRKJoLe3F1arFQBw7NgxtLe3o7+/H+vr66irq8PDhw9RUcGTqvZidnYW3d3duHPn\nDlRVxbdv3wpmGwqF8OzZM+h0OjgcDty6dUt26UVjZ8YejwexWAzV1dlz6e/evYsLFy4w433w+Xz4\n9OkT1tbW0NnZCbvdznmsBUKSSCQiOjo6hBBCzM3NCYfDIauUkvLhwwfR09OT98zj8YhXr14JIYR4\n9OiRePHihYzSil46nRaqqgqv1ysmJiaEEIWzTafToq2tTSSTSZHJZMS1a9fE0tKSzNKLRqGM3W63\nePv27a5xzHhvwuGwaG9vF0IIsbi4KM6fP895rBHStqzD4TAuX74MADh69CiWl5eRSqVklVPSIpEI\nWltbAQAXL15EOByWXFFxqqioQDAYhNlszj0rlG00GoXdbofRaERlZSWampowMzMjq+yiUijjQpjx\n3p0+fRojIyMAgIMHDyKTyXAea4S0hpxIJHBo4/pDADCZTFhYWJBVTkmZm5tDV1cXbt++jffv3yOT\nyeS2qGtqapjzHun1elRWVuY9K5RtIpGAyWTKjeHc/n2FMgaA58+fw+l04t69e1hcXGTG+1BWVgaD\nIXuP+OTkJFpaWjiPNULqd8jbCR6p/Z84fPgwXC4Xrly5gng8DqfTmXfvNHP+//xdtsx8f27cuIHq\n6mrYbDYEAgGMjY3h5MmTeWOY8b83PT2NyclJPH36FG1tbbnnnMfySFshm81mJBKJ3Ofv37+jrq5O\nVjklw2Kx4OrVq1AUBQ0NDaitrcXy8jJWVlYAAPPz8/+4HUi/z2Aw7Mq20Nxm5nt35swZ2Gw2AMCl\nS5cwOzvLjPfp3bt3ePz4MYLBIIxGI+exRkhryM3NzXjz5g0AIBaLwWw248CBA7LKKRmhUAjj4+MA\ngIWFBfz48QM3b97MZT01NYVz587JLLGknD17dle2J06cwOfPn5FMJpFOpzEzM4NTp05JrrR49fT0\nIB6PA8h+Z2+1WpnxPvz8+RM+nw9PnjzJvbnOeawNUq9f9Pv9+PjxIxRFwcDAAI4fPy6rlJKRSqXQ\n19eHZDKJX79+weVywWazwe12Y3V1FfX19RgcHER5ebnsUovOly9fMDw8jK9fv0Kv18NiscDv98Pj\n8ezK9vXr1xgfH4eiKFBVFdevX5ddflEolLGqqggEAqiqqoLBYMDg4CBqamqY8R69fPkSo6OjOHLk\nSO7Z0NAQvF4v57FkvA+ZiIhIA3hSFxERkQawIRMREWkAGzIREZEGsCETERFpABsyERGRBrAhExER\naQAbMhERkQb8BfLaftbUelUBAAAAAElFTkSuQmCC\n",
492 |             "text/plain": [
493 |               "<matplotlib.figure.Figure at 0x7f8aa201b908>"
494 |             ]
495 |           },
496 |           "metadata": {
497 |             "tags": []
498 |           }
499 |         },
500 |         {
501 |           "output_type": "stream",
502 |           "text": [
503 |             "Color Information\n",
504 |             "{'cluster_index': 4,\n",
505 |             " 'color': [222.96986073580376, 181.78320515485106, 149.09041779256452],\n",
506 |             " 'color_percentage': 0.32252457043524774}\n",
507 |             "\n",
508 |             "{'cluster_index': 2,\n",
509 |             " 'color': [231.1843409316113, 196.8800792864229, 168.19573835481265],\n",
510 |             " 'color_percentage': 0.2705087918700274}\n",
511 |             "\n",
512 |             "{'cluster_index': 0,\n",
513 |             " 'color': [211.4051658647802, 164.9225120283632, 130.32615852114876],\n",
514 |             " 'color_percentage': 0.26302065922310625}\n",
515 |             "\n",
516 |             "{'cluster_index': 3,\n",
517 |             " 'color': [195.0976775956313, 134.61407103825226, 100.6448087431703],\n",
518 |             " 'color_percentage': 0.09734572440997526}\n",
519 |             "\n",
520 |             "{'cluster_index': 1,\n",
521 |             " 'color': [248.32137733142207, 243.87087517934114, 238.32711621233653],\n",
522 |             " 'color_percentage': 0.04660025406164338}\n",
523 |             "\n",
524 |             "Color Bar\n"
525 |           ],
526 |           "name": "stdout"
527 |         },
528 |         {
529 |           "output_type": "display_data",
530 |           "data": {
531 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAB8CAYAAABE+eipAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ni7RS0QAAAABJRU5ErkJggg==\n",
532 |             "text/plain": [
533 |               "<matplotlib.figure.Figure at 0x7f8a9d31bc18>"
534 |             ]
535 |           },
536 |           "metadata": {
537 |             "tags": []
538 |           }
539 |         }
540 |       ]
541 |     },
542 |     {
543 |       "metadata": {
544 |         "id": "z8sQ_dXQ9-L7",
545 |         "colab_type": "code",
546 |         "colab": {}
547 |       },
548 |       "cell_type": "code",
549 |       "source": [
550 |         ""
551 |       ],
552 |       "execution_count": 0,
553 |       "outputs": []
554 |     }
555 |   ]
556 | }


--------------------------------------------------------------------------------
/Extract_Skin_from_an_Image_and_Find_the_Dominant_Colors_Tone.py:
--------------------------------------------------------------------------------
  1 | import numpy as np
  2 | import cv2
  3 | from sklearn.cluster import KMeans
  4 | from collections import Counter
  5 | import imutils
  6 | import pprint
  7 | from matplotlib import pyplot as plt
  8 | 
  9 | 
 10 | def extractSkin(image):
 11 |     # Taking a copy of the image
 12 |     img = image.copy()
 13 |     # Converting from BGR Colours Space to HSV
 14 |     img = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
 15 | 
 16 |     # Defining HSV Threadholds
 17 |     lower_threshold = np.array([0, 48, 80], dtype=np.uint8)
 18 |     upper_threshold = np.array([20, 255, 255], dtype=np.uint8)
 19 | 
 20 |     # Single Channel mask,denoting presence of colours in the about threshold
 21 |     skinMask = cv2.inRange(img, lower_threshold, upper_threshold)
 22 | 
 23 |     # Cleaning up mask using Gaussian Filter
 24 |     skinMask = cv2.GaussianBlur(skinMask, (3, 3), 0)
 25 | 
 26 |     # Extracting skin from the threshold mask
 27 |     skin = cv2.bitwise_and(img, img, mask=skinMask)
 28 | 
 29 |     # Return the Skin image
 30 |     return cv2.cvtColor(skin, cv2.COLOR_HSV2BGR)
 31 | 
 32 | 
 33 | def removeBlack(estimator_labels, estimator_cluster):
 34 | 
 35 |     # Check for black
 36 |     hasBlack = False
 37 | 
 38 |     # Get the total number of occurance for each color
 39 |     occurance_counter = Counter(estimator_labels)
 40 | 
 41 |     # Quick lambda function to compare to lists
 42 |     def compare(x, y): return Counter(x) == Counter(y)
 43 | 
 44 |     # Loop through the most common occuring color
 45 |     for x in occurance_counter.most_common(len(estimator_cluster)):
 46 | 
 47 |         # Quick List comprehension to convert each of RBG Numbers to int
 48 |         color = [int(i) for i in estimator_cluster[x[0]].tolist()]
 49 | 
 50 |         # Check if the color is [0,0,0] that if it is black
 51 |         if compare(color, [0, 0, 0]) == True:
 52 |             # delete the occurance
 53 |             del occurance_counter[x[0]]
 54 |             # remove the cluster
 55 |             hasBlack = True
 56 |             estimator_cluster = np.delete(estimator_cluster, x[0], 0)
 57 |             break
 58 | 
 59 |     return (occurance_counter, estimator_cluster, hasBlack)
 60 | 
 61 | 
 62 | def getColorInformation(estimator_labels, estimator_cluster, hasThresholding=False):
 63 | 
 64 |     # Variable to keep count of the occurance of each color predicted
 65 |     occurance_counter = None
 66 | 
 67 |     # Output list variable to return
 68 |     colorInformation = []
 69 | 
 70 |     # Check for Black
 71 |     hasBlack = False
 72 | 
 73 |     # If a mask has be applied, remove th black
 74 |     if hasThresholding == True:
 75 | 
 76 |         (occurance, cluster, black) = removeBlack(
 77 |             estimator_labels, estimator_cluster)
 78 |         occurance_counter = occurance
 79 |         estimator_cluster = cluster
 80 |         hasBlack = black
 81 | 
 82 |     else:
 83 |         occurance_counter = Counter(estimator_labels)
 84 | 
 85 |     # Get the total sum of all the predicted occurances
 86 |     totalOccurance = sum(occurance_counter.values())
 87 | 
 88 |     # Loop through all the predicted colors
 89 |     for x in occurance_counter.most_common(len(estimator_cluster)):
 90 | 
 91 |         index = (int(x[0]))
 92 | 
 93 |         # Quick fix for index out of bound when there is no threshold
 94 |         index = (index-1) if ((hasThresholding & hasBlack)
 95 |                               & (int(index) != 0)) else index
 96 | 
 97 |         # Get the color number into a list
 98 |         color = estimator_cluster[index].tolist()
 99 | 
100 |         # Get the percentage of each color
101 |         color_percentage = (x[1]/totalOccurance)
102 | 
103 |         # make the dictionay of the information
104 |         colorInfo = {"cluster_index": index, "color": color,
105 |                      "color_percentage": color_percentage}
106 | 
107 |         # Add the dictionary to the list
108 |         colorInformation.append(colorInfo)
109 | 
110 |     return colorInformation
111 | 
112 | 
113 | def extractDominantColor(image, number_of_colors=5, hasThresholding=False):
114 | 
115 |     # Quick Fix Increase cluster counter to neglect the black(Read Article)
116 |     if hasThresholding == True:
117 |         number_of_colors += 1
118 | 
119 |     # Taking Copy of the image
120 |     img = image.copy()
121 | 
122 |     # Convert Image into RGB Colours Space
123 |     img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
124 | 
125 |     # Reshape Image
126 |     img = img.reshape((img.shape[0]*img.shape[1]), 3)
127 | 
128 |     # Initiate KMeans Object
129 |     estimator = KMeans(n_clusters=number_of_colors, random_state=0)
130 | 
131 |     # Fit the image
132 |     estimator.fit(img)
133 | 
134 |     # Get Colour Information
135 |     colorInformation = getColorInformation(
136 |         estimator.labels_, estimator.cluster_centers_, hasThresholding)
137 |     return colorInformation
138 | 
139 | 
140 | def plotColorBar(colorInformation):
141 |     # Create a 500x100 black image
142 |     color_bar = np.zeros((100, 500, 3), dtype="uint8")
143 | 
144 |     top_x = 0
145 |     for x in colorInformation:
146 |         bottom_x = top_x + (x["color_percentage"] * color_bar.shape[1])
147 | 
148 |         color = tuple(map(int, (x['color'])))
149 | 
150 |         cv2.rectangle(color_bar, (int(top_x), 0),
151 |                       (int(bottom_x), color_bar.shape[0]), color, -1)
152 |         top_x = bottom_x
153 |     return color_bar
154 | 
155 | 
156 | """## Section Two.4.2 : Putting it All together: Pretty Print
157 | 
158 | The function makes print out the color information in a readable manner
159 | """
160 | 
161 | 
162 | def prety_print_data(color_info):
163 |     for x in color_info:
164 |         print(pprint.pformat(x))
165 |         print()
166 | 
167 | 
168 | """
169 | The below lines of code, is the implementation of the above defined function.
170 | """
171 | 
172 | '''
173 | Skin Image Primary : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg
174 | Skin Image One     : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/skin.jpg
175 | Skin Image Two     : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/skin_2.jpg
176 | Skin Image Three   : https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/Human-Hands-Front-Back-Image-From-Wikipedia.jpg
177 | 
178 | '''
179 | 
180 | 
181 | # Get Image from URL. If you want to upload an image file and use that comment the below code and replace with  image=cv2.imread("FILE_NAME")
182 | image = imutils.url_to_image(
183 |     "https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/master/82764696-open-palm-hand-gesture-of-male-hand_image_from_123rf.com.jpg")
184 | 
185 | # Resize image to a width of 250
186 | image = imutils.resize(image, width=250)
187 | 
188 | # Show image
189 | plt.subplot(3, 1, 1)
190 | plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
191 | plt.title("Original Image")
192 | # plt.show()
193 | 
194 | # Apply Skin Mask
195 | skin = extractSkin(image)
196 | 
197 | plt.subplot(3, 1, 2)
198 | plt.imshow(cv2.cvtColor(skin, cv2.COLOR_BGR2RGB))
199 | plt.title("Thresholded  Image")
200 | # plt.show()
201 | 
202 | # Find the dominant color. Default is 1 , pass the parameter 'number_of_colors=N' where N is the specified number of colors
203 | dominantColors = extractDominantColor(skin, hasThresholding=True)
204 | 
205 | # Show in the dominant color information
206 | print("Color Information")
207 | prety_print_data(dominantColors)
208 | 
209 | # Show in the dominant color as bar
210 | print("Color Bar")
211 | colour_bar = plotColorBar(dominantColors)
212 | plt.subplot(3, 1, 3)
213 | plt.axis("off")
214 | plt.imshow(colour_bar)
215 | plt.title("Color Bar")
216 | 
217 | plt.tight_layout()
218 | plt.show()
219 | 


--------------------------------------------------------------------------------
/Human-Hands-Front-Back-Image-From-Wikipedia.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/f1988037f6f365ece3aec00c0e7ae52e2b3f9938/Human-Hands-Front-Back-Image-From-Wikipedia.jpg


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # Skin-Extraction-from-Image-and-Finding-Dominant-Color
2 | Repository containing that code for the article Skin Segmentation and Dominant Tone/Color Extraction found at "https://medium.com/@mithushancj/skin-segmentation-and-dominant-tone-color-extraction-fe158d24badf". 
3 | 
4 | Project is an implementation of skin segmentation using OpenCV and dominant color extraction using SciKit-Learn. Read the article and notebook file of breakdown for the process.
5 | 
6 | ## Getting Started
7 | Provided you already have NumPy, SciKit-Learn, Matplotlib, OpenCV and imutils already installed clone the project and either run the python file or the notebook and run it locally.
8 | 


--------------------------------------------------------------------------------
/skin.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/f1988037f6f365ece3aec00c0e7ae52e2b3f9938/skin.jpg


--------------------------------------------------------------------------------
/skin_2.jpg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/octalpixel/Skin-Extraction-from-Image-and-Finding-Dominant-Color/f1988037f6f365ece3aec00c0e7ae52e2b3f9938/skin_2.jpg


--------------------------------------------------------------------------------