{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "### Modelo de análisis de sentimientos" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# importamos bilbiotecas necesarias para el analisis de sentimientos\n", "import nltk, random\n", "from nltk.corpus import movie_reviews\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "#importamos librerias necesarias para la interpretacion del modelo\n", "import lime\n", "from lime.lime_text import LimeTextExplainer\n", "from sklearn.pipeline import Pipeline\n", "nltk.download('movie_reviews')" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2000\n", "['neg', 'pos']\n", "['plot', ':', 'two', 'teen', 'couples', 'go', 'to', ...]\n", "['neg/cv000_29416.txt', 'neg/cv001_19502.txt', 'neg/cv002_17424.txt', 'neg/cv003_12683.txt', 'neg/cv004_12641.txt', 'neg/cv005_29357.txt', 'neg/cv006_17022.txt', 'neg/cv007_4992.txt', 'neg/cv008_29326.txt', 'neg/cv009_29417.txt']\n" ] } ], "source": [ "# Usaremos el corpus nlkt.corpus.movie_reviews como datos de entrenamiento\n", "print(len(movie_reviews.fileids()))\n", "# catergorias\n", "print(movie_reviews.categories())\n", "# palabras\n", "print(movie_reviews.words()[:100])\n", "# documentos\n", "print(movie_reviews.fileids()[:10])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Reorganiza los datos del corpus como una lista de tuplas, donde el primer elemento son los tokens de palabras de los documentos, y el segundo elemento es la etiqueta de los documentos (es decir, etiquetas de sentimiento)." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('plot : two teen couples go to a church party , drink and then drive . they get into an accident . one of the guys dies , but his girlfriend continues to see him in her life , and has nightmares . what \\' s the deal ? watch the movie and \" sorta \" find out . . . critique : a mind - fuck movie for the teen generation that touches on a very cool idea , but presents it in a very bad package . which is what makes this review an even harder one to write , since i generally applaud films which attempt to break the mold , mess with your head and such ( lost highway & memento ) , but there are good and bad ways of making all types of films , and these folks just didn \\' t snag this one correctly . they seem to have taken this pretty neat concept , but executed it terribly . so what are the problems with the movie ? well , its main problem is that it \\' s simply too jumbled . it starts off \" normal \" but then downshifts into this \" fantasy \" world in which you , as an audience member , have no idea what \\' s going on . there are dreams , there are characters coming back from the dead , there are others who look like the dead , there are strange apparitions , there are disappearances , there are a looooot of chase scenes , there are tons of weird things that happen , and most of it is simply not explained . now i personally don \\' t mind trying to unravel a film every now and then , but when all it does is give me the same clue over and over again , i get kind of fed up after a while , which is this film \\' s biggest problem . it \\' s obviously got this big secret to hide , but it seems to want to hide it completely until its final five minutes . and do they make things entertaining , thrilling or even engaging , in the meantime ? not really . the sad part is that the arrow and i both dig on flicks like this , so we actually figured most of it out by the half - way point , so all of the strangeness after that did start to make a little bit of sense , but it still didn \\' t the make the film all that more entertaining . i guess the bottom line with movies like this is that you should always make sure that the audience is \" into it \" even before they are given the secret password to enter your world of understanding . i mean , showing melissa sagemiller running away from visions for about 20 minutes throughout the movie is just plain lazy ! ! okay , we get it . . . there are people chasing her and we don \\' t know who they are . do we really need to see it over and over again ? how about giving us different scenes offering further insight into all of the strangeness going down in the movie ? apparently , the studio took this film away from its director and chopped it up themselves , and it shows . there might \\' ve been a pretty decent teen mind - fuck movie in here somewhere , but i guess \" the suits \" decided that turning it into a music video with little edge , would make more sense . the actors are pretty good for the most part , although wes bentley just seemed to be playing the exact same character that he did in american beauty , only in a new neighborhood . but my biggest kudos go out to sagemiller , who holds her own throughout the entire film , and actually has you feeling her character \\' s unraveling . overall , the film doesn \\' t stick because it doesn \\' t entertain , it \\' s confusing , it rarely excites and it feels pretty redundant for most of its runtime , despite a pretty cool ending and explanation to all of the craziness that came before it . oh , and by the way , this is not a horror or teen slasher flick . . . it \\' s just packaged to look that way because someone is apparently assuming that the genre is still hot with the kids . it also wrapped production two years ago and has been sitting on the shelves ever since . whatever . . . skip it ! where \\' s joblo coming from ? a nightmare of elm street 3 ( 7 / 10 ) - blair witch 2 ( 7 / 10 ) - the crow ( 9 / 10 ) - the crow : salvation ( 4 / 10 ) - lost highway ( 10 / 10 ) - memento ( 10 / 10 ) - the others ( 9 / 10 ) - stir of echoes ( 8 / 10 )',\n", " 'neg'),\n", " ('the happy bastard \\' s quick movie review damn that y2k bug . it \\' s got a head start in this movie starring jamie lee curtis and another baldwin brother ( william this time ) in a story regarding a crew of a tugboat that comes across a deserted russian tech ship that has a strangeness to it when they kick the power back on . little do they know the power within . . . going for the gore and bringing on a few action sequences here and there , virus still feels very empty , like a movie going for all flash and no substance . we don \\' t know why the crew was really out in the middle of nowhere , we don \\' t know the origin of what took over the ship ( just that a big pink flashy thing hit the mir ) , and , of course , we don \\' t know why donald sutherland is stumbling around drunkenly throughout . here , it \\' s just \" hey , let \\' s chase these people around with some robots \" . the acting is below average , even from the likes of curtis . you \\' re more likely to get a kick out of her work in halloween h20 . sutherland is wasted and baldwin , well , he \\' s acting like a baldwin , of course . the real star here are stan winston \\' s robot design , some schnazzy cgi , and the occasional good gore shot , like picking into someone \\' s brain . so , if robots and body parts really turn you on , here \\' s your movie . otherwise , it \\' s pretty much a sunken ship of a movie .',\n", " 'neg'),\n", " (\"it is movies like these that make a jaded movie viewer thankful for the invention of the timex indiglo watch . based on the late 1960 ' s television show by the same name , the mod squad tells the tale of three reformed criminals under the employ of the police to go undercover . however , things go wrong as evidence gets stolen and they are immediately under suspicion . of course , the ads make it seem like so much more . quick cuts , cool music , claire dane ' s nice hair and cute outfits , car chases , stuff blowing up , and the like . sounds like a cool movie , does it not ? after the first fifteen minutes , it quickly becomes apparent that it is not . the mod squad is certainly a slick looking production , complete with nice hair and costumes , but that simply isn ' t enough . the film is best described as a cross between an hour - long cop show and a music video , both stretched out into the span of an hour and a half . and with it comes every single clich ? . it doesn ' t really matter that the film is based on a television show , as most of the plot elements have been recycled from everything we ' ve already seen . the characters and acting is nothing spectacular , sometimes even bordering on wooden . claire danes and omar epps deliver their lines as if they are bored , which really transfers onto the audience . the only one to escape relatively unscathed is giovanni ribisi , who plays the resident crazy man , ultimately being the only thing worth watching . unfortunately , even he ' s not enough to save this convoluted mess , as all the characters don ' t do much apart from occupying screen time . with the young cast , cool clothes , nice hair , and hip soundtrack , it appears that the film is geared towards the teenage mindset . despite an american ' r ' rating ( which the content does not justify ) , the film is way too juvenile for the older mindset . information on the characters is literally spoon - fed to the audience ( would it be that hard to show us instead of telling us ? ) , dialogue is poorly written , and the plot is extremely predictable . the way the film progresses , you likely won ' t even care if the heroes are in any jeopardy , because you ' ll know they aren ' t . basing the show on a 1960 ' s television show that nobody remembers is of questionable wisdom , especially when one considers the target audience and the fact that the number of memorable films based on television shows can be counted on one hand ( even one that ' s missing a finger or two ) . the number of times that i checked my watch ( six ) is a clear indication that this film is not one of them . it is clear that the film is nothing more than an attempt to cash in on the teenage spending dollar , judging from the rash of really awful teen - flicks that we ' ve been seeing as of late . avoid this film at all costs .\",\n", " 'neg'),\n", " ('\" quest for camelot \" is warner bros . \\' first feature - length , fully - animated attempt to steal clout from disney \\' s cartoon empire , but the mouse has no reason to be worried . the only other recent challenger to their throne was last fall \\' s promising , if flawed , 20th century fox production \" anastasia , \" but disney \\' s \" hercules , \" with its lively cast and colorful palate , had her beat hands - down when it came time to crown 1997 \\' s best piece of animation . this year , it \\' s no contest , as \" quest for camelot \" is pretty much dead on arrival . even the magic kingdom at its most mediocre -- that \\' d be \" pocahontas \" for those of you keeping score -- isn \\' t nearly as dull as this . the story revolves around the adventures of free - spirited kayley ( voiced by jessalyn gilsig ) , the early - teen daughter of a belated knight from king arthur \\' s round table . kayley \\' s only dream is to follow in her father \\' s footsteps , and she gets her chance when evil warlord ruber ( gary oldman ) , an ex - round table member - gone - bad , steals arthur \\' s magical sword excalibur and accidentally loses it in a dangerous , booby - trapped forest . with the help of hunky , blind timberland - dweller garrett ( carey elwes ) and a two - headed dragon ( eric idle and don rickles ) that \\' s always arguing with itself , kayley just might be able to break the medieval sexist mold and prove her worth as a fighter on arthur \\' s side . \" quest for camelot \" is missing pure showmanship , an essential element if it \\' s ever expected to climb to the high ranks of disney . there \\' s nothing here that differentiates \" quest \" from something you \\' d see on any given saturday morning cartoon -- subpar animation , instantly forgettable songs , poorly - integrated computerized footage . ( compare kayley and garrett \\' s run - in with the angry ogre to herc \\' s battle with the hydra . i rest my case . ) even the characters stink -- none of them are remotely interesting , so much that the film becomes a race to see which one can out - bland the others . in the end , it \\' s a tie -- they all win . that dragon \\' s comedy shtick is awfully cloying , but at least it shows signs of a pulse . at least fans of the early -\\' 90s tgif television line - up will be thrilled to find jaleel \" urkel \" white and bronson \" balki \" pinchot sharing the same footage . a few scenes are nicely realized ( though i \\' m at a loss to recall enough to be specific ) , and the actors providing the voice talent are enthusiastic ( though most are paired up with singers who don \\' t sound a thing like them for their big musical moments -- jane seymour and celine dion ? ? ? ) . but one must strain through too much of this mess to find the good . aside from the fact that children will probably be as bored watching this as adults , \" quest for camelot \" \\' s most grievous error is its complete lack of personality . and personality , we learn from this mess , goes a very long way .',\n", " 'neg'),\n", " ('synopsis : a mentally unstable man undergoing psychotherapy saves a boy from a potentially fatal accident and then falls in love with the boy \\' s mother , a fledgling restauranteur . unsuccessfully attempting to gain the woman \\' s favor , he takes pictures of her and kills a number of people in his way . comments : stalked is yet another in a seemingly endless string of spurned - psychos - getting - their - revenge type movies which are a stable category in the 1990s film industry , both theatrical and direct - to - video . their proliferation may be due in part to the fact that they \\' re typically inexpensive to produce ( no special effects , no big name stars ) and serve as vehicles to flash nudity ( allowing them to frequent late - night cable television ) . stalked wavers slightly from the norm in one respect : the psycho never actually has an affair ; on the contrary , he \\' s rejected rather quickly ( the psycho typically is an ex - lover , ex - wife , or ex - husband ) . other than that , stalked is just another redundant entry doomed to collect dust on video shelves and viewed after midnight on cable . stalked does not provide much suspense , though that is what it sets out to do . interspersed throughout the opening credits , for instance , a serious - sounding narrator spouts statistics about stalkers and ponders what may cause a man to stalk ( it \\' s implicitly implied that all stalkers are men ) while pictures of a boy are shown on the screen . after these credits , a snapshot of actor jay underwood appears . the narrator states that \" this is the story of daryl gleason \" and tells the audience that he is the stalker . of course , really , this is the story of restauranteur brooke daniels . if the movie was meant to be about daryl , then it should have been called stalker not stalked . okay . so we know who the stalker is even before the movie starts ; no guesswork required . stalked proceeds , then , as it begins : obvious , obvious , obvious . the opening sequence , contrived quite a bit , brings daryl and brooke ( the victim ) together . daryl obsesses over brooke , follows her around , and tries to woo her . ultimately rejected by her , his plans become more and more desperate and elaborate . these plans include the all - time , psycho - in - love , cliche : the murdered pet . for some reason , this genre \\' s films require a dead pet to be found by the victim stalked . stalked is no exception ( it \\' s a cat this time -- found in the shower ) . events like these lead to the inevitable showdown between stalker and stalked , where only one survives ( guess who it invariably always is and you \\' ll guess the conclusion to this turkey ) . stalked \\' s cast is uniformly adequate : not anything to write home about but also not all that bad either . jay underwood , as the stalker , turns toward melodrama a bit too much . he overdoes it , in other words , but he still manages to be creepy enough to pass as the type of stalker the story demands . maryam d \\' abo , about the only actor close to being a star here ( she played the bond chick in the living daylights ) , is equally adequate as the \" stalked \" of the title , even though she seems too ditzy at times to be a strong , independent business - owner . brooke ( d \\' abo ) needs to be ditzy , however , for the plot to proceed . toward the end , for example , brooke has her suspicions about daryl . to ensure he won \\' t use it as another excuse to see her , brooke decides to return a toolbox he had left at her place to his house . does she just leave the toolbox at the door when no one answers ? of course not . she tries the door , opens it , and wanders around the house . when daryl returns , he enters the house , of course , so our heroine is in danger . somehow , even though her car is parked at the front of the house , right by the front door , daryl is oblivious to her presence inside . the whole episode places an incredible strain on the audience \\' s suspension of disbelief and questions the validity of either character \\' s intelligence . stalked receives two stars because , even though it is highly derivative and somewhat boring , it is not so bad that it cannot be watched . rated r mostly for several murder scenes and brief nudity in a strip bar , it is not as offensive as many other thrillers in this genre are . if you \\' re in the mood for a good suspense film , though , stake out something else .',\n", " 'neg')]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "documents = [(' '.join((movie_reviews.words(fileid))), category)\n", " for category in movie_reviews.categories()\n", " for fileid in movie_reviews.fileids(category)]\n", "documents[:5]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Estadísticas descriptivas importantes:\n", " - Tamaño del corpus (número de documentos)\n", " - Tamaño del corpus (número de palabras)\n", " - Distribución de las dos clases\n", "- Es importante informar siempre de las estadísticas descriptivas necesarias anteriormente al presentar sus clasificadores." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Numero de reseñas/documentos: 2000\n", "Tamaño del corpus (palabras): 7808520\n", "Tamaño maximo de documento (palabras): 15097\n", "Tamaño mínimo de documento (palabras): 91\n", "Tamaño promedio de documentos (palabras): 3904.26\n", "Muestra de texto de Doc 1:\n", "------------------------------\n", "plot : two teen couples go to a church party , drink and then drive . they get into an accident . one of the guys dies , but his girlfriend continues to see him in her life , and has nightmares . what ' s the deal ? watch the movie and \" sorta \" find out . . . critique : a mind - fuck movie for the teen generation that touches on a very cool idea , but presents it in a very bad package . which is what makes this review an even harder one to write , since i generally applaud films which attempt to break the mold , mess with your head and such ( lost highway & memento ) , but there are good and bad ways of making all types of films , and these folks just didn ' t snag this one correctly . they seem to have taken this pretty neat concept , but executed it terribly . so what are the problems with the movie ? well , its main problem is that it ' s simply too jumbled . it starts off \" normal \" but then downshifts into this \" fantasy \" world in which you , as an audience member , have no idea what ' s going on . there are dreams , there are characters coming back from the dead , there are others who look like the dead , there are strange apparitions , there are disappearances , there are a looooot of chase scenes , there are tons of weird things that happen , and most of it is simply not explained . now i personally don ' t mind trying to unravel a film every now and then , but when all it does is give me the same clue over and over again , i get kind of fed up after a while , which is this film ' s biggest problem . it ' s obviously got this big secret to hide , but it seems to want to hide it completely until its final five minutes . and do they make things entertaining , thrilling or even engaging , in the meantime ? not really . the sad part is that the arrow and i both dig on flicks like this , so we actually figured most of it out by the half - way point , so all of the strangeness after that did start to make a little bit of sense , but it still didn ' t the make the film all that more entertaining . i guess the bottom line with movies like this is that you should always make sure that the audience is \" into it \" even before they are given the secret password to enter your world of understanding . i mean , showing melissa sagemiller running away from visions for about 20 minutes throughout the movie is just plain lazy ! ! okay , we get it . . . there are people chasing her and we don ' t know who they are . do we really need to see it over and over again ? how about giving us different scenes offering further insight into all of the strangeness going down in the movie ? apparently , the studio took this film away from its director and chopped it up themselves , and it shows . there might ' ve been a pretty decent teen mind - fuck movie in here somewhere , but i guess \" the suits \" decided that turning it into a music video with little edge , would make more sense . the actors are pretty good for the most part , although wes bentley just seemed to be playing the exact same character that he did in american beauty , only in a new neighborhood . but my biggest kudos go out to sagemiller , who holds her own throughout the entire film , and actually has you feeling her character ' s unraveling . overall , the film doesn ' t stick because it doesn ' t entertain , it ' s confusing , it rarely excites and it feels pretty redundant for most of its runtime , despite a pretty cool ending and explanation to all of the craziness that came before it . oh , and by the way , this is not a horror or teen slasher flick . . . it ' s just packaged to look that way because someone is apparently assuming that the genre is still hot with the kids . it also wrapped production two years ago and has been sitting on the shelves ever since . whatever . . . skip it ! where ' s joblo coming from ? a nightmare of elm street 3 ( 7 / 10 ) - blair witch 2 ( 7 / 10 ) - the crow ( 9 / 10 ) - the crow : salvation ( 4 / 10 ) - lost highway ( 10 / 10 ) - memento ( 10 / 10 ) - the others ( 9 / 10 ) - stir of echoes ( 8 / 10 ) neg\n" ] } ], "source": [ "print('Numero de reseñas/documentos: {}'.format(len(documents)))\n", "print('Tamaño del corpus (palabras): {}'.format(np.sum([len(d) for (d,l) in documents])))\n", "print('Tamaño maximo de documento (palabras): {}'.format(np.max([len(d) for (d,l) in documents])))\n", "print('Tamaño mínimo de documento (palabras): {}'.format(np.min([len(d) for (d,l) in documents])))\n", "print('Tamaño promedio de documentos (palabras): {}'.format(np.mean([len(d) for (d,l) in documents])))\n", "print('Muestra de texto de Doc 1:')\n", "print('-'*30)\n", "print(' '.join(documents[0][:50])) # first 50 words of the first document" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Counter({'neg': 1000, 'pos': 1000})\n" ] } ], "source": [ "## Revisar la distribución de las clases\n", "from collections import Counter\n", "sentiment_distr = Counter([label for (words, label) in documents])\n", "print(sentiment_distr)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Shuffle aleatorio**\n", "\n", "Mezclar los datos de entrenamiento aleatoriza el orden en el que las muestras se presentan al modelo durante el entrenamiento. Esto evita que el modelo aprenda patrones o secuencias en los datos que podrían no generalizarse bien a datos no vistos." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "import random\n", "\n", "random.seed(123)\n", "random.shuffle(documents)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**División de entrenamiento y prueba**\n", "\n", "- Dividimos todo el conjunto de datos en dos partes: conjunto de entrenamiento y conjunto de pruebas.\n", "- La proporción de conjuntos de entrenamiento y pruebas puede depender del tamaño del corpus.\n", "- En la división de entrenamiento y prueba, asegurate de que la distribución de las clases sea proporcional." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split\n", "train, test = train_test_split(documents, test_size = 0.33, random_state=42)" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Counter({'neg': 674, 'pos': 666})\n", "Counter({'pos': 334, 'neg': 326})\n" ] } ], "source": [ "## Distribución de clases en los conjuntos de entrenamiento y prueba\n", "print(Counter([label for (text, label) in train]))\n", "print(Counter([label for (text, label) in test]))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Dado que en la mayoría de los pasos de aprendizaje automático, los conjuntos de características y las etiquetas suelen estar separados en dos unidades, dividimos nuestros datos de entrenamiento en `X_train` y `y_train`, que representan las características (X) y las etiquetas (y) en el entrenamiento.\n", "Asimismo, dividimos nuestros datos de prueba en `X_test` y `y_test`, que representan las características (X) y las etiquetas (y) en las pruebas." ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [], "source": [ "X_train = [text for (text, label) in train]\n", "X_test = [text for (text, label) in test]\n", "y_train = [label for (text, label) in train]\n", "y_test = [label for (text, label) in test]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Vectorización** de textos usando `TFidfVectorizer`" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [], "source": [ "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n", "\n", "tfidf_vec = TfidfVectorizer(min_df = 10, token_pattern = r'[a-zA-Z]+')\n", "X_train_bow = tfidf_vec.fit_transform(X_train) # fit train\n", "X_test_bow = tfidf_vec.transform(X_test) # transform test" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(1340, 6155)\n", "(660, 6155)\n" ] } ], "source": [ "print(X_train_bow.shape)\n", "print(X_test_bow.shape)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Modelo de clasificación**" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
SVC(C=8.0, kernel='linear')
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "SVC(C=8.0, kernel='linear')" ] }, "execution_count": 45, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn import svm\n", "\n", "model_svm = svm.SVC(C=8.0, kernel='linear')\n", "model_svm.fit(X_train_bow, y_train)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Validación Cruzada**" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([0.84701493, 0.84328358, 0.82462687, 0.88059701, 0.81716418])" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.model_selection import cross_val_score\n", "model_svm_acc = cross_val_score(estimator=model_svm, X=X_train_bow, y=y_train, cv=5, n_jobs=-1)\n", "model_svm_acc" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Predicción del modelo**" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['pos', 'pos', 'pos', 'pos', 'pos', 'pos', 'neg', 'pos', 'neg',\n", " 'pos'], dtype='#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}
Pipeline(steps=[('vectorizer',\n",
       "                 TfidfVectorizer(min_df=10, token_pattern='[a-zA-Z]+')),\n",
       "                ('clf', SVC(C=1, kernel='linear', probability=True))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "Pipeline(steps=[('vectorizer',\n", " TfidfVectorizer(min_df=10, token_pattern='[a-zA-Z]+')),\n", " ('clf', SVC(C=1, kernel='linear', probability=True))])" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "## define the pipeline\n", "pipeline = Pipeline([\n", " ('vectorizer',tfidf_vec),\n", " ('clf', svm.SVC(C=1, kernel='linear', probability=True))])\n", "\n", "## Refit model based on optimal parameter settings\n", "pipeline.fit(X_train, y_train)" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ID Reseña-210:\n", "--------------------------------------------------\n", "Texto Reseña:\n", " quiz show , an almost perfectly accurate true story , is based upon the events of the popular television show of the mid - 50 ' s , \" twenty - one \" . on this trivial game show , contestants were placed in isolation booths and then answered questions corresponding to a category of their choice , on which they wagered an amount of points on . the game went on until a player reached twenty one\n", "points on felt they had earned enough points to win . but , after ratings began to fall when players were struggling to break the zero mark , the producers decided to fix the game by giving the answers to a contestant before the game began . quiz show illustrates the true stories of two particular contestants , herbie stempel and charles van doren . stempel ( john turturro ) , a former g . i . and\n", "your jewish man raising a family . stempel has been the reigning champion on \" twenty - one \" for many weeks and has accumulated thousands of dollars . in his mind , he is the best thing on television and the people love him . although , in the mind of the show ' s producers , herbie stempel is getting old . dan enright ( david paymer ) , in particular , feels that the people are tired of seeing a\n", "\" jewish guy from queens with bad teeth \" and that the kids need someone better to look up to . therefore , they need to find another contestant whom would be a worthy role model and the people will look up to and cheer to win . someone who can defeat stempel , even if they have to resort to cheating . enter charles van doren ( ralph fiennes ) , a well - educated professor from a widely recognized\n", "family . van doren had decided to try out for the game show \" tic tac dough \" because his friends thought he would be good at that sort of thing . but when albert freedman ( hank azaria ) , enright ' s assistant , spots van doren , the two decide that they have found their soon - to - be - ruler of the \" twenty - one \" kingdom . van doren is not too keen on the idea of receiving the answers ahead\n", "of time , so enright tells stempel that he is going to give the wrong answer , on purpose , in order to lose the game . after stempel loses the \" throne \" to van doren , he starts to feel cheated ( which he should ) . meanwhile , on his own , dick goodwin ( rob morrow ) , a harvard law graduate , has decided to start an investigation on \" twenty - one \" to try and find out if there have been any\n", "wrongdoings . his investigation yields shocking results and leads to a trial for enright and the others involved . quiz show is an extremely well done movie , and robert redford ' s direction is especially superb . the performances turned in by john turturro , ralph fiennes , and rob morrow are very good , although it seems that turturro stands out more than any . quiz show is also very precise\n", "when it comes to explicating the true events that inspired the film . definitely a film you should not miss .\n", "--------------------------------------------------\n", "Probabilidad(positivo) = 0.6900658842165507\n", "Probabilidad(negativo) = 0.30993411578344937\n", "Clase predicha: ['pos']\n", "Clase positiva: pos\n" ] } ], "source": [ "import textwrap\n", "reviews_test = X_test\n", "sentiments_test = y_test\n", "# Elegimos una reseña aleatoria para predecir\n", "idx = 210\n", "text_sample = reviews_test[idx]\n", "class_names = ['negative', 'positive']\n", "\n", "print('ID Reseña-{}:'.format(idx))\n", "print('-'*50)\n", "print('Texto Reseña:\\n', textwrap.fill(text_sample,400))\n", "print('-'*50)\n", "print('Probabilidad(positivo) =', pipeline.predict_proba([text_sample])[0,1])\n", "print('Probabilidad(negativo) =', pipeline.predict_proba([text_sample])[0,0])\n", "print('Clase predicha: %s' % pipeline.predict([text_sample]))\n", "print('Clase positiva: %s' % sentiments_test[idx])" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", "
\n", " \n", " \n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib\n", "matplotlib.rcParams['figure.dpi']=300\n", "%matplotlib inline\n", "\n", "\n", "explainer = LimeTextExplainer(class_names=class_names)\n", "explanation = explainer.explain_instance(text_sample,\n", " pipeline.predict_proba,\n", " num_features=20)\n", "explanation.show_in_notebook(text=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Coeficientes del modelo e importancia de las características**\n", "- Otra forma de evaluar la importancia de las características es mirar sus coeficientes correspondientes.\n", " - Los pesos positivos implican una contribución positiva de la característica a la predicción; los pesos negativos implican una contribución negativa de la característica a la predicción.\n", " - Los valores absolutos de los pesos indican los tamaños de efecto de las características.\n", "- No todos los modelos de ML proporcionan \"coeficientes\"." ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [], "source": [ "## Extraer coeficientes de modelo del pipeline\n", "importances = pipeline.named_steps['clf'].coef_.toarray().flatten()\n", "## Seleccionar los 10 pesos positivos/negativos principales\n", "top_indices_pos = np.argsort(importances)[::-1][:10]\n", "top_indices_neg = np.argsort(importances)[:10]\n", "## Obtener los nombres de las características de tfidf\n", "feature_names = np.array(tfidf_vec.get_feature_names_out()) # List indexing is different from array" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
CARACTERISTICAIMPORTANCIASENTIMIENTO
0and2.771421pos
1great2.133551pos
2well1.612445pos
3most1.436231pos
4true1.429615pos
5very1.401382pos
6fun1.331665pos
7quite1.291350pos
8as1.290924pos
9is1.272521pos
10bad-3.400709neg
11plot-2.472051neg
12only-1.661845neg
13worst-1.655213neg
14have-1.625395neg
15stupid-1.519736neg
16boring-1.416260neg
17nothing-1.406122neg
18lame-1.383713neg
19dull-1.377941neg
\n", "
" ], "text/plain": [ " CARACTERISTICA IMPORTANCIA SENTIMIENTO\n", "0 and 2.771421 pos\n", "1 great 2.133551 pos\n", "2 well 1.612445 pos\n", "3 most 1.436231 pos\n", "4 true 1.429615 pos\n", "5 very 1.401382 pos\n", "6 fun 1.331665 pos\n", "7 quite 1.291350 pos\n", "8 as 1.290924 pos\n", "9 is 1.272521 pos\n", "10 bad -3.400709 neg\n", "11 plot -2.472051 neg\n", "12 only -1.661845 neg\n", "13 worst -1.655213 neg\n", "14 have -1.625395 neg\n", "15 stupid -1.519736 neg\n", "16 boring -1.416260 neg\n", "17 nothing -1.406122 neg\n", "18 lame -1.383713 neg\n", "19 dull -1.377941 neg" ] }, "execution_count": 58, "metadata": {}, "output_type": "execute_result" } ], "source": [ "feature_importance_df = pd.DataFrame({'CARACTERISTICA': feature_names[np.concatenate((top_indices_pos, top_indices_neg))],\n", " 'IMPORTANCIA': importances[np.concatenate((top_indices_pos, top_indices_neg))],\n", " 'SENTIMIENTO': ['pos' for _ in range(len(top_indices_pos))]+['neg' for _ in range(len(top_indices_neg))]})\n", "feature_importance_df" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0.5, 1.0, 'Importancia de características: Palabras Clave')" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABC0AAANPCAYAAAD+FyXUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAABcSAAAXEgFnn9JSAAEAAElEQVR4nOzdd3gUZfv//c+EAAmEEFrovffeBBREsCAi0gQERUGwoiIiguUGvwregt42FBBBQDrSRbog0pvUID2hQwhJIKTO8wdP9pdN3U12sht8v46Dg5mdmWvO7M7OzJ5zFSMsLMwUAAAAAACAh/FydwAAAAAAAACpIWkBAAAAAAA8EkkLAAAAAADgkUhaAAAAAAAAj0TSAgAAAAAAeCSSFgAAAAAAwCORtAAAAAAAAB6JpAUAAAAAAPBIJC0AAAAAAIBHImkBAAAAAAA8EkkLAAAAAADgkUhaAAAAAAAAj0TSAgAAAAAAeCRvdwcAAACQEx09elRLliyxzVeuXFk9e/Z0X0AAANyDSFoAAAA4KSoqSgMGDNCxY8ckSQULFtS6devcHBUAAPcemocA8GgvvfSSAgICbP/Onj3r7pDcgvcB8CyjR4+2JSy8vb01Y8YMVa1a1c1ROaZTp0525xMgu3jitWzLli12MX366afuDgkewhOP138raloAAAA44bffftOPP/5om//000/Vtm1b9wUEANCFCxd0+PBhhYSEKDw8XLGxsSpYsKACAgJUsWJF1a1bV3nz5nV3mMgEkhYuUrduXQUHB9vmly9frjZt2rgxIgAA4GqXLl3Sq6++apsfOHCgBg0a5MaI8G/x0ksvac6cORmu5+fnp4IFC6pMmTJq0KCB2rVrpw4dOihXrlzZECWQvf755x9Nnz5dK1asyLAmRJ48edSgQQN169ZN3bt3V5EiRbIpSmQVzUMAJ9WtW9dWTaxu3bruDgcA3Obf1szANE299NJLun79uiSpbdu2GjduXLbGMHv2bLv3fPbs2dm6f3i+yMhInT9/Xjt27NAPP/ygp59+WvXr19fChQvdHRrgMpcuXdKgQYPUrFkzffvttw413YiJidHOnTs1YsQI1ahRQ++8845CQ0OzIVpkFUkLAAAAB3z77bfauHGjJKlKlSqaPn26vL2ptArPFxISooEDB2ro0KHuDgXIso0bN+q+++7TggULZJpmiuX+/v6qVKmSGjVqpEqVKsnPzy/FOrGxsZo8ebIaNmyoW7duZUfYyAKutAA82qRJkzRp0iR3hwHgX+7vv//WmDFjJEkBAQGaO3dujq1dsnLlSneHABfo1auXevfuneL1mzdv6vTp01q9erW2b99ut2zGjBkqUaKERo4cmV1hAi61dOlSvfDCC4qLi7N7vXbt2urXr586dOigypUrp9guODhYq1ev1vLly7V582bb6zdv3lRsbKzlcSNrSFoAAABk4Pjx43rzzTclSQ8++KCqVKni5ojwb1ehQoV0O4B944039Pvvv+vFF1/UzZs3ba9/8cUX6tOnj8qXL58NUQKuc/DgQQ0ePNguYVGgQAF98skn6tu3r7y80m5EULZsWQ0aNEiDBg3Stm3b9NFHH2nHjh3ZETZcgKQFAABABrp37+7uEACnPfzww5o2bZq6detmey0mJkZTp07V2LFj3RgZ4JzIyEgNGDBAd+7csb1WpEgRLVq0SA0aNHCqrJYtW+r333/XF198wfcgh6BPCwAAAOAe1b59+xQ1MjZs2OCeYIBM+uyzz3TixAnbvJeXl2bNmuV0wiKpN998UwsWLFDu3LldECGsRE0LDxUWFqYdO3bowoULCg0NVUBAgBo0aKDGjRtnuO3Fixe1c+dOnTt3TnFxcSpWrJiaNGmiGjVquDTGK1euaPv27Tp79qxiY2NVrFgxValSRc2bN0+3epaj4uPjtWvXLp0+fVrXrl2TaZoqUqSIypUrp+bNmytPnjwu+CvsHT9+XEePHtXly5cVHh6uIkWK6KmnnlLBggVdvq9EcXFx+ueffxQUFKTLly8rIiJCPj4+KlSokKpUqaKGDRta8rfeuXNH27dvV1BQkCIiIlSwYEGVLl1arVq1ctnfe+fOHe3evVshISG6fv26oqKi5Ofnp7Jly6pmzZqqVKmSS/aTnosXL+rYsWM6c+aMwsPDlZCQoEKFCql48eJq2rSpihYtankM6QkJCdHOnTt18eJFRUdHq2jRoqpXr57q168vwzBcvr8LFy5oz549unbtmm7cuCE/Pz/bOaJs2bIu358zrDhebt68qaNHj+rEiRO6ceOGoqOjVbBgQRUtWlQNGjRQxYoVLfhL7p4f9+7dq8uXL+vatWvKnz+/HnrooXSbFNy6dUvHjh3TP//8Y/v7CxQooMKFC6tOnTqqUaOGS48JT/h+pufUqVM6ePCgrl69qps3byogIECBgYFq0aKFihUr5vL9HThwQKdOndKlS5d0+/ZtFStWTL1793bZzWxYWJhtH+Hh4YqJiZGvr68CAgJUvnx51ahRw5K/y0o3btzQrl27dPnyZYWGhiohIcHWAV7t2rUVGBjodJnZcc6+Fz+LjDz88MPatGmTbf7YsWMyTTPNc0p0dLSCgoJ0/PhxXb16Vbdu3VL+/PlVqFAh1ahRQ3Xr1nXLEKruuKbHx8drz549OnLkiK5fv658+fKpVKlSatGihYoXL+7SfSUkJGjPnj06e/asLl26pJiYGJUtW1Y9evRId7uzZ8/q2LFjCg4OVkREhLy8vFSoUCGVLl1aTZo0ccl93ZUrV/T333/r7NmzCg8PV3x8vPLly6fChQurfPnyqlmzpmX9/ISHh2v69Ol2rw0ePFgtW7bMctnt27fPchlp8fR70Mw4fPiwgoKCdPXqVUVGRqpw4cIqWbKkWrRoYWk/TyQt3KRTp07aunWrbT4sLEySdPLkSX366adasWKFXfWnRFWrVtW4ceNS/YIdPHhQY8aM0fr165WQkJBieb169fT555+rWbNmWYrx6NGj+vDDD7V27dpUe+wtUaKEBg8erNdeey1TvaqfP39e48eP19KlS+3aYCbl5+enRx55RCNHjky1s53UbNmyRZ07d7bNjxgxQiNHjlR8fLymT5+un376SYcOHUqxXePGjTVp0qRUx0YPDg5O9wvaqlWrVDs8u379upYuXarVq1dr27ZtioiISLMMHx8fPf7443rzzTdVu3btDP7K/yf5eO4HDhxQ+fLlFRERof/+97/66aefUt2vt7e3nnrqKX344YcqXbq0w/tLavPmzfrqq6/0559/pnocJypTpowef/xxDRgwQNWrV3fq70hLfHy8Nm3apKVLl+qPP/7IcAisevXq6bXXXtNTTz2VrTdg+/bt0+jRo/XXX3+l+j0qX768hg0bpv79+2d5X7GxsZoxY4amTZumI0eOpLlerVq1NHToUPXs2dOShElaXHm8SHc7TFy8eLE2bNigQ4cOpXo+TFSqVCkNHDhQAwcOlL+/v8MxJ/3eJ/2er1+/Xt9++63++OMPxcfH223zySefpEhanDx5UgsXLtSGDRu0Z8+eFB2LJVW4cGH1799fr7zySpZ+UGX1/f700081fvz4VLdJ73xYtmxZHTx4MN3Ybt26pUmTJumXX37RqVOnUl3HMAw1btxYw4cP18MPP5xueYlmz56tV155xTb/7bffqm/fvrpz546+++47zZo1K9X9PfHEE3Z/U926dRUcHOzw3yPdvfb873//08aNG1McE8lVqFBBHTp00AsvvGD3oOHs2bOqX79+qtu88sordn9bcsuXL1ebNm3sXkvr+u6IuLg4zZkzR9OmTdOBAwfS/H4ZhqHatWure/fueuaZZ9K8Oc/Oc7YrPoucqkKFCnbz8fHxunHjhgoXLmx77fz581q8eLHWrl2rnTt3pnt+KFCggHr06KHXX389Rdmu5M5relxcnCZNmqRvvvlGly9fTrHcy8tLbdu21dixYx2+P0t+/kz8foaFhenLL7/U/PnzdeHCBbtt/P39UyQt7ty5ozVr1mjFihXavHmzLl26lOY+vby81Lx5c73xxhsOnzOTWrZsmSZNmqTt27ener+SyDAMVatWTY8++qgGDhyoMmXKOL2vtEyfPl3h4eG2+Tx58mjYsGEuK99VrDxe+/Tpo1WrVtnmZ8+erU6dOjkdY69evfT777/b5ufOnatHHnkk3W1CQ0P15ZdfauHChSmOz0S5cuVSmzZtNHLkSDVv3tzpuDJC0sKDrF69Ws8//7xu376d5jr//POPunfvrokTJ2rAgAG212fOnKlhw4YpJiYmzW3//vtvderUSdOnT8/UQS5Jv/76q4YMGaLo6Og017l06ZL+85//6Ndff9XChQudetIyffp0jRw5UlFRUemuFxkZqYULF2rp0qUaOXKk3nrrLYf3kVRYWJj69u1rd/NmtbCwMFWvXj3dHydJ3blzRwsXLtSvv/6qjz76SK+99lqm9338+HH17NlTZ86cSXOduLg4zZ8/X5s2bdLixYtVp04dh8u/du2aBg0aZBsSMCMhISH6/vvv9eeff+rPP/90eD/pGTBggJYtW+bw+n///bcGDRqkmTNnavr06XY3cFb58ssvNXbs2HRvms+ePavXX39d69ev15QpUzK9r/3792vAgAE6ffp0huseOXJEgwcP1vTp0zV79mzL3wsrjpfJkyfrnXfecTiGCxcuaMyYMfrpp580e/Zs1atXz+FtkzJNU++9955TI92sXr1aTz/9tMPrJ940zJgxQ9OnT9cDDzzgVIye8P1Mz7p16/TKK6+k+uMgKdM0tXv3bvXq1UudO3fW5MmT5evr6/T+QkJC1LNnz3QTeVlhmqbeffdd/fDDDw5vc+bMGU2ZMkUFCxbU6NGjLYkrK/bv36/nn38+zYRSUqZp6tChQzp06JAuXbqkcePGpbpedpyzrfwskifEJOeSQNkltRpDSe9DDh06pDZt2qT7ozSpiIgITZs2TXPmzNFXX32VYU2AzHLXNT08PFy9e/dO9/4wISFBGzZs0JYtW/TJJ59o0KBBmdrX4cOH1atXL4WEhDi8zSOPPKL9+/c7tG5CQoK2bdumbdu2qVu3bvrmm28cOmdGR0frxRdf1NKlSx3aj2maCgoKUlBQkCpVquSShy6Jli9fbjf/2GOPeWQtBSuP1969e9slLebMmeP077krV65o/fr1tvnixYvroYceSnebX375Re+++65d0ig1iQmbTZs2adCgQRo3bpxLHwaStPAQO3bs0LPPPmtLBuTJk0cVKlSQn5+fLly4YJdBNU1Tb7/9tmrXrq1mzZppwYIFev31120XGl9fX5UvX16+vr4KDg7WtWvXbNvGxsZq4MCB2rZtm9OZ8S1btmjQoEG2i5yXl5fKlSunQoUK6erVqylOtn///be6dOmilStXOnTRmDBhQqqd4RQoUEBly5ZVrly5dO7cObvaF7GxsRozZowuX76c5pO/tMTHx6tPnz7666+/bK/5+/urTJkyypUrl86fP6/Q0FCnynR0v8kTFl5eXipTpoz8/f3l4+OjyMhInT171i55Ex8fr/fff1/x8fF64403nN5vcHCwBg4caHcslSlTRsWKFdOdO3d08uRJu6TXlStX9PTTT2vbtm0qUKBAhuUfP35c3bt317lz51Is8/f3V8mSJeXn56fw8HAFBwen+wQnK1JLqBUpUkTFihWTn5+f7ty5oytXrujKlSt262zevFldunTRunXrlDdvXktik6Svv/5aH330UYrXCxYsqHLlyikuLk7nzp2zjRm+dOlSp2oAJLVmzRoNGDAgxfjjuXPnVvny5VWwYEFFRkbqzJkzdu/btm3b1KFDB61du9ayxIVVx0tqn3/+/PlVqlQpFShQQAkJCQoNDVVwcLDdzXlwcLA6deqkP/74I1PNIj7++GO7hEXevHlVrlw5+fn56fLly7p48aJDsfr4+Kh06dLy8/OTt7e3bty4obNnz9oluG7cuKFu3bppxYoVatGihUPxecr3My0///yz3nrrrRTnRh8fH5UvX94W26lTp+zei+XLl6tLly5avny5U9/byMhIPfXUUzp+/LjttSJFiqhkyZJKSEhQSEhIhjdoGfnkk09S/ZFcsGBBlSlTRvny5dPt27d148aNNJ9ceZJly5Zp8ODBqT5UKFasmAIDA5UnTx7duHFD586dS7eGU1LZcc6+1z6LzEjtHJT0/B4bG5siYZE7d26VLl1a/v7+yp07t8LDw3X27Fm7e4WoqCgNGjRIuXLl0lNPPeXyuN1xTTdNU88995xdwqJgwYIqW7asEhISdO7cOUVGRtqWxcbGavjw4cqVK5eef/55p/Z14cIFvfDCC3bxFy9eXCVKlNCdO3fSTGSk9r4UL15chQsXlp+fn27fvq0LFy7oxo0bdussWrRIERERmjdvXoY1Kl9//fVUExZFixZVyZIl5ePjo1u3bunatWsp3n9Xun37dooETWYfvlrNyuP1kUceUZEiRXT9+nVJd+/xQkNDnbpPmzdvnt11tkePHunWih83blyqCWc/Pz+VKVNG+fPnV2hoqM6cOWN3/pgyZYouX76sGTNmuKzmLkkLDzFo0CBFR0crMDBQo0ePVteuXe1+KO7atUtDhw61PRGKj4/XqFGj9P3332vo0KEyTVOVKlXSBx98oEceeUQ+Pj6S7p54169fr6FDh+r8+fOS7l5gPvrooxRtwzLyyiuvKC4uTt7e3nrttdf04osvqmTJkrblp0+f1hdffKGff/7Z9trRo0c1bNgw/fTTT+mWvX79en388cd2r1WqVEljxozRww8/bHtCkJCQoM2bN2v06NF2TTl++OEHNWjQINXxytMyc+ZM2xO9++67T++9955atmxplxXcuXOnSpUqpaFDh6pXr16SpBdffNF2sgkMDNTkyZPT3Ed6VaXr16+vzp07q3379qpZs6btM0uU2KfH999/ryVLlthe//jjj9WuXbs0qwqn5dVXX9WlS5fk4+OjV155RQMGDLCruhcREaGpU6fq//7v/2wntJCQEE2cOFEffvhhumUnPpFI+oPIMAx17dpVQ4YMUePGje3e17i4OB06dEirVq3S/Pnznfo7HFGkSBE9+eST6tChg5o0aZJqNj4kJETz58/Xl19+aftxcvDgQY0dOzbFsegqu3btSvFe1q5dWx9//LEeeOABW18w0dHRWrlypd5//32dP39eM2fOVLly5ZzaV1BQUIqERdOmTfXGG2/owQcftHvKEhUVpRUrVujjjz+2VWU8efKkXn75Zc2dOzezf26arD5e8ubNq0cffVQdO3ZUq1atVK5cuRQXzfDwcK1cuVKff/65Tp48Kenud+CFF17Qhg0bnLrIHj9+XNu2bZN0t1nP+++/r8cee0z58uWzrXPmzJkUySPpbsLyoYce0iOPPKL7779fFStWTPFkIioqSmvXrtXEiRNtN25xcXEaOHCgdu7cabef1Lj6/e7du7etHfGoUaN0+PBh27Kk56rkkp/jEm3ZskVvvPGG3Y/cDh066NVXX1XLli3t+vQJDw/XggUL9Omnn9oS8jt37tTo0aP13//+N933IamJEyfazv+dOnXS8OHD7fqRSUhI0B9//JGpGhzS3R+IX375pd1rzz77rIYMGaKaNWumWD88PFx79+7VmjVrtHDhwhTLixcvbntvN2zYoK+++sq27PXXX9eDDz6YZix169bN1N+Q1IEDB/Tiiy/aJbPy5cunIUOGqHfv3qpatard+rdu3dLu3bu1bNkyLViwIMPyrTxnu/qzyKmS15YqU6ZMqj9W7rvvPnXq1Elt27ZVtWrVUtTQiI2N1ebNm/X111/b9ZHxxhtvqGXLlnb3ha6S3df0uXPn2s6XtWrVsl2jE8+TMTEx+u233zRq1Ci7pMKIESPUrFkzp2qofvDBB7py5Yq8vLzUt29fvfbaa6pWrZpteWxsrN37nFSZMmXUpUsXdezYUfXr10/1nvPEiROaNWuWJk2aZPtBvWbNGk2ZMkUvvvhimnHt3btX8+bNs817e3vr1Vdf1XPPPZfqQ8/Q0FDt2LFDa9as0aJFixz74x20a9cuxcbG2r3WqFEjl+7Dlaw6XnPnzq3u3bvbErAxMTFauHBhup9jcsmbuvfp0yfNdefOnZsiYdG9e3cNHjxYjRo1srtvuH79umbMmKEJEybY7nWWLVumb775Jks1xJMywsLCHKsHhnQlbecqpd6GNKnk7Umluz/Sly1blmYbsOvXr6tFixa6evWq7bXq1asrKChIzZs31/z589PsaOf48eNq06aN7YSVO3duHTt2TEWKFHEqxty5c2v27Nnq2LFjmttNnz49RU2A+fPnp7lNVFSUGjZsaFcDoFmzZlq8eLH8/PxS3SY2NlbPPPOMXZssf39/7d27N83qYsn7tEj0/PPPa8KECQ7/SMlMm+akoqKidPDgQYf7FpGkWbNm6dVXX7XNd+/eXVOnTk13m+R9QUh3nxQsXLhQTZs2TXO7OXPm6KWXXrLNBwYG6vDhw+l2Rjd48GC7i5uvr69+/PFHPfbYY+nGKN1Nzvz1119pfl+c7dNi586dqlevXpo/kJI7e/asOnfubLtByZcvn44cOeLyzoRM01SrVq3sqqI/+OCDmjt3bpodrYaGhurRRx9VUFBQimXpvQ/x8fFq06aN3b6GDRumUaNGpdtJblhYmLp3767du3fbXstsm8n0WHm8HD16VEWLFnW4z4c7d+7omWee0bp162yvLV68ON0fgVLqCckmTZpo0aJFDnd4dubMGRmGke7xnFRCQoJef/11zZo1y/bal19+qeeeey7d7ax8v7PSN4J0t6PU5s2b287/hmHo888/1wsvvJDudiEhIercubNds6ctW7ak+QM9tSr8kvThhx/qzTffdDheR8//U6dO1dtvv22bHz58uEaNGuXQPmJiYnThwoU0a0Om1T+HM5z53GJiYtS8eXO797p8+fJatGhRuh3LJgoPD9fJkyfVsGHDVJdbfc628rOQsrd5SPLrYWLfXBk5c+aMmjdvbvcUuE+fPvruu+9s81evXtW1a9dSTeSkJXn/DG+99ZY++OADp/4GT7imp3V/eP/992vBggVpPvm+ceOGOnXqZHetbdy4sV31++RS6xPIy8tL33//vXr27Jnen2Zn69atatGihcPV7/fv368nn3zSdmyWKVNG+/fvT/Mp+wcffGCXHP3666/Vr18/h/YVGRmpsLAwl/Vpkfw77Ofnp+Dg4Gzpe8vTjtcDBw7YNQ1t2LChw00+k2/boEGDNBNiZ86cUatWrWwJiLx582ratGkZ3g8ePnxYnTt3ttVUz5s3rw4dOuSSjo0Z8tRDeHt7a/r06el+wYsUKZLiwhgUFKSAgABNnz493RvlatWq2dVCSC9zm55333033YSFJD333HMp2rF98803aa6/YMECu4RFoUKF9Msvv6SZsJDuJk9++uknu9EOwsPDM6zRkVzjxo313//+N1s7HfT19XUqYSFJzzzzjN3FbOnSpel23pmWr776Kt2EhXT3KWrSUWquXLmiAwcOpLn+yZMnUzxJmzRpkkM/iKT/13GPqzRr1szhi4V09+Y76YX59u3bWrx4scviSbRx40a7G5vAwEBNnz493ZFhChcurF9++cXp0WOWLl1qt6++ffvq/fffz3BUn4CAAM2cOdPuu5fedzczrD5eatas6dTF0cfHR5MnT7ZrgpM0KeAof39//fzzz0710F6hQgWHExbS3RvbCRMm2P2AyihWT/t+JvfTTz/Znf/ffffdDBMW0t0b7p9//tnumP7666+d2vcTTzzhVMLCGUmH5ZPkVFv3xOahnmLu3Ll2CYsCBQpo6dKlDiUspLvfjbQSFpL15+x76bPIjIsXL6pv374pqq0n70+nWLFiTiUsJGnkyJF2ozdk5tyZEXdd04sWLaqff/453aYlhQoV0uzZs+3W2bNnj12zY0e8/PLLTiUspLsdQDvTX0CDBg00ZswY23xISEi6vwOSfm/8/PzSfSKfXGLTAVdJ3sSlSJEi2Xrf7gyrj9f69evbdfq6b98+HTt2zKF9/fLLL3bz6X2mX331lV3t0C+//NKhB1i1a9fWt99+a5uPjo52qi+h9JC08BBdunRxqAO4Rx99NMVrL7zwgkPV8ZLfpP7999+OB6i71VOTPu1PzwcffGD3Q2vLli1pdv6YtDmJdDdT70jnOvny5UuR0Z8xY4ZD8SUaNWqUW4bsyozE5inS3aTT3r17ndq+UaNG6tKli0Prdu3a1W4+vc6evv/+e7tq3Y8++qiefPJJp2Jzt7Zt26pEiRK2+e3bt7t8H8lv5t5++22H+qqoXLmyXae7jkjat4KPj0+qfcWkpWTJknZPU7Zt22ZXuyurPPF4KVy4sDp06GCb37Fjh9NlDBw4UKVKlXJlWKnKmzev3ffzwIED6fY/4Ynvd6KEhAS7m5lSpUo5Xesh6TXxt99+y3BEiKTef/99h9d1VvLPxFXDprrD999/bzf/7rvvuv2HvDPnbKs/i759+yosLMzun7tFRERo//79GjdunFq1amXXhEuSOnbsqPvvv98l+0r6Y/vKlSsOdfpsNVdc099++22HalxWrFgxRT8Wye9r0+Pr62tXi8BK3bp1s7vndfR74+XlleFDDyslT1q4YvhWT+Ls8Zq8KXzyZERqYmNj7Zq75cmTR927d0913bCwMLvaJY0aNXKq+f2jjz5qV+sxeSeqmUXSwkM4ehNZpUqVFE9dHf0hWqtWLbt5Z3oplu6e7BztzKho0aJ2PwJM09SWLVtSrHfr1i3t27fPNu/t7e3UF+OJJ56wO3mFhISkOzJG8hjbtm3r8L7cLfkTWWeTTskTEelJnkBL7A8lNRs2bLCbT9q0JCdJ2meEs01+HJG0PXHu3Lmd6mndmarfoaGhds07Onbs6HRnmsmbRjj71Cg9nnq8JP1+nT9/3tbRlaPSuvhbIWmssbGxOnr0aJrreur7Ld09hyXtHPCpp55yulZR0mM1IiLC4e9ugwYNUvTD4EpJb0Al2TXPyUkuXrxoV2vLz8/PpSMCZIWj5+x75bNIzfjx4xUQEJDiX9myZdW2bVuNGzcuRYfiNWrUSLcvLmdl9d7EKlm5pufOndupmg/Jn1g7M9pShw4dXN4cNS358+e3q4no6PcmPDxcv/32m6WxpSdpp6fS3b/jXuPM8dqrVy+7Zj3z58/PMGG/Zs0au/uahx9+OM17w82bN9t1uOzMKGeJkl6bjx8/7vQ9VWroiNNDpFd1MqlcuXLJ39/f1gFZ7ty5HR4bOvnB6WzP6K1bt3Z6/ZUrV9rm9+7dm6I93N69e+2+aLVr13ZqCKO8efOqRYsWdn1b7Ny506GnQI0aNXJr5li62+Hdxo0btW7dOh0+fFhnzpxReHi4IiMjM+x53dkTgDOdFiWvXp/WsXL58mVbJ4bS3RvaVq1aORWXlU6cOKElS5Zo//79OnbsmG7cuKGIiIh0hwaWnH9vM3Lu3Dm7nqJr1aqlQoUKObx9vXr1VLBgQbuRc9Kybds2ux6ckzb1cVTSZleS7EZYyIrsPl6uX7+uJUuWaOfOnTp8+LCuXLmiiIiIDIdUTtw2vT5/kipQoIBq1KiRpVgjIiK0fPly/fXXXzp8+LAuXryoiIiIVDvvTC3W1Hj69zOx89JErjhWg4KC1KBBgwy3a9KkidP7cka7du3sOjAbPXq0rl27poEDB6b4Ee3JkicsW7Ro4dBoUpllxTn7XvksXKFnz5767LPPMvyRfOfOHf3+++/6448/dPjwYdtoGZGRkRkOierq62dS2XVNr1mzplPJ/rp169pdo0NCQnT58mUVL148w21dcS7av3+/VqxYoYMHDyooKEg3b95UREREitGYksvoe5P0Cf6LL76oESNGqF+/ftmWZEmUvLm4I9dFT2DV8VqsWDE99NBDWr16tSTp0qVL2rhxY7pDlzrTNMTV12bTNHX8+HG7pmSZQdLCQzjzQz1pb+aFChVyuHlD8h7mHblxTyp5TQ1n10+tBkTSzkslOdXjcqK6devaJS0crUHiTFtyK8yZM0f/+c9/7NpzO8PZKqjOtPN39FhJXg20bt26HtHcJigoSO+8847++OOPTG3v6uq9iSNyJHK23bB09/uU/EKSmuQJhg8//DDD0V8ykrxqZmZl1/Fy48YNjRkzRjNnzszwpi0tzhwDZcuWzXQCNCoqSp9//rm+++47p8/JidKK1VO/n4mSH6sDBgxwuilUco4eq1af/5s3b6527drZOkiLi4vT559/rokTJ6pp06a6//771bJlSzVu3NijqzonP4YcfcDiLCvP2ffKZ5EZhmGoatWqeuCBB/Tss89meI8VFxenH374QZ999plDSfLUWNE8Jruv6c7e7yZuk/QafebMGYeSFlk5F+3atUvDhw9PtwlvetJ7X5588klNmDDB1l9CZGSk3n//fY0ZM0b33Xef7r//fjVv3lyNGjXKcASrrEr+kCerw1FbLTuO1z59+tiSFtLd3xRpJS2uX7+uNWvW2OYDAwPtasMnl/za3L59+wzjyYgr7iNJWngIZzptScrZsaeTyihbnpyjTx0TJc9Sp3YBTP7FdHYfqW3j6MXJkf4ErJDaCACZkdpY0Omx4lhJXu3UFb0DZ9W6dev0zDPPpNvOPyMZZcGdlfzYz8xx7uhTn+SfiStk9uY1uew4XpL3xJ1Zzny/Mvvk+caNG3ryySfT7ejWEWnF6onfz6Tceaxmx/l/6tSp6t27t3bu3Gl7LSEhQTt27LD1m+Ll5aX69evroYceUs+ePS1tspIZ2XEMZcc5+174LFLTq1evFE1qDcNQvnz55O/vr1KlSjl8fkptJKXMcPbeJCPuuKa74hpt9blo5syZGjp0aIa1ctOT3meVO3duzZ07Vz179rT7ERsbG6s//vjD9oM8d+7catKkiR5++GF1797dpR1wJkqetLCyNk9WZdfx+sgjj6hw4cK2c/TKlSt18+bNVBOvCxcutBsytkePHmmOGiN57n0kSQs4zNlMavI2Z8nbpEkpq3hlJlvryH5Sk94X1koTJkxIkbAoUKCA7rvvPjVo0EClS5dWoUKF5OPjY9dh2JUrV5waizk7eFo7w1OnTqW4WBiGoYYNG6pZs2aqUKGCSpQoIR8fnxSJwlGjRqXorMxVkh/nSWtLOcrR99ZVCYaksnJTlJTVx0tsbKx69uyZImFRsWJFtWrVStWqVVPJkiWVP39++fj42NWOmDNnTqbbumf2XPLcc8+lSFiUKlVKrVu3Vo0aNVS6dGn5+fnJx8fHrobEhg0b7HobT4unfT+Tc+exmh3n/yJFimjVqlWaOXOmvvvuO/3zzz8p1klISNC+ffu0b98+/fe//9Vjjz2mTz75xO0dXSay+hjKrnP2vfBZpKZChQou65tr+PDhKRIWhQsXVuvWrVW3bl2VLl1a/v7+8vX1tTsfHTx40LJObd11TXfFNdrKe9GtW7fq9ddft3uglCtXLjVv3lxNmjRRuXLlVKxYMfn4+KR4YPXiiy/aNVdNT4UKFbRp0yZNnjxZU6dOTbUmc2xsrLZt26Zt27Zp7Nix6tWrl8aMGeNUDfKMJE8gRkRE6NSpU6pUqZLL9uEK2Xm8Jnakmdg/zZ07d7RkyRI9++yzKdZ1pmmI5Ln3kSQt4LDbt2879UQx+Q+11IYwTX6Sv337ttNxObIfT3H16lVNnDjR7rU33nhDw4YNy/C9Te0my92Sx+zudoYfffSR3cWiUaNG+u677xzqb8DK6o1ZbZolOf7eJr/Zeumll/Twww87vb+kXNXu2+rj5aefflJQUJBtvlixYvrmm28c+vszW40zs1avXm23Tz8/P33++efq0aNHhk04HO2d39O+n8klP1Y/+ugjh/qjSI+n/cD09va2NXs5cOCANm/erD///FM7duxItVbgqlWr9Oeff2rBggVq3rx59gecjNXHUHaes3P6Z2GlQ4cO2T1M8fb21kcffaRBgwZlWEvTmRF7nOWua7orrtFW3ouOGjXKLmHRoUMHTZw4MUUfP6lxdqjQfPny6Y033tDrr7+uXbt2afPmzdq6dat2796dIjETHx+vX375RRs3btSKFStUuXJlp/aVlqZNmyp37tx2tQX27t3rcUmL7D5e+/TpY9ep7pw5c1IkLY4cOWL3cKRevXoZ9oWY/Nr87bffqnTp0k7Hl1RW+/2SSFrACdevX3cqaZG8elFqVZaSd+aTmSpfybfJ7g6CnPHbb7/ZXQyfffZZffTRRw5t66p+BVwpeZU9Vw6N6axbt27Z9W0SGBioRYsWOdzhpZXvb/JjPzPHuaPV9ZJXay1RooTHjJJj9fGSfGzzmTNnqkWLFg5tm93fr+SxfvHFFw6PKONorJ70/UxN8mO1fPnyHnOsWqF+/fqqX7++XnvtNZmmqaNHj2rjxo1atmyZ3TC74eHh6t+/v/bs2eP2JLyVx5A7z9k58bOw0pIlS+x+BI8cOdLhIe6tOne68/hwxTXaqv5RTp06ZdeHRc2aNTV79myHR17KbJ8jXl5eat68uZo3b67hw4crPj5ef//9t9avX69ff/3VrpbAxYsX1b9/f23ZssUlHd7ny5dP9evXtxsZbdWqVdk6aldG3HG8NmjQQLVq1bKN8LR9+/YUNVCSDl0qZVzLQkp5ba5Ro0amOuN0NYY8hcOSDnvmiOTVnFJ7ApZ0iB/pbrbfWcmHBnIk0+wuSW+GJGngwIEOb5vesIbukjzLffDgQUufuqRn//79du0zu3Xr5vDFIioqyuGhcjMj+bHv7HfJmW2Sd+qVdPQId7PyeElISLC7oalbt67DCQsp+79fSc8FhQsX1lNPPeXwto7G6knfz9R48rFqNcMwVKtWLb3yyiv6/ffftXLlSrsbxcuXL2v+/PlujPCu5E9KM9vhX2o85ZydUz4LKyU9H3l5een55593eFurzp3uPD5ccY22qtZX8vvI/v37O5ywOHnypMv6HMmVK5caNmyot99+W1u3btWMGTPsntAfPnw4xZDbWfHEE0/Yza9YscKj+rZw1/GavE+bpE1B4uPj7c5duXPndujhiKdem0lawGFbtmxxav2tW7fazac25GbDhg3tqkIfPnzYqZNQdHR0ihN406ZNnYrTWUmzxs52Zpr8KZUzHX1t3rzZqX1lh8DAQLu/ITIyMsXnnl2y8t7+9ddfmR5lwhHlypVTYGCgbf7o0aNOdXR04MABh9sYtmnTxm7ek44bK4+X0NBQu8/Qmc//5s2bLv0x5oikx2ulSpUcHtUjPj7e4fcsO76fyZ+iOXNO9ORjNbu1atUqxSg/27dvT3XdrLznzrrvvvvs5rdt26aIiAiXlO2p52xnPot7RdLPomjRok4NyW3V99adx8eRI0ecukYfPHjQ7hpdpkwZh0YOyQxPvY/s0qWLXnnlFbvXXPm9efbZZ+1qe8fExOiLL75wWflZ5a7jtVevXnb9osybN892TVi/fr0uX75sW/bwww871Mmsp16bSVrAYYsXL3Y4Q3v16lWtXbvWNm8Yhlq3bp1ivfz589slM+Li4jR37lyHY1q+fLldVbcyZcpY3qY5aT8czrZ7TH5z6WiP1pcvX9by5cud2ld2ST4U0vfff++WODL73krSlClTXB1OCq1atbJNx8bGauHChQ5vm7wTpfSUKlXKbri2M2fO2FVZdDerjpesfP4zZ850eY/3GUkarzOxrly5UhcuXHB4fau/n8nb4TrTL1GTJk3sfhxt2bIlU0847xXJawZdu3Yt1fVc0UeOo4oXL243TGZkZKRmzpzpkrI9+Zzt6Gdxr8js+Wj//v3atWuXFSG59fiIi4tzqnZN8mt0ave7rpLZ98U0TU2dOtWKkGys/N4ULFhQzz33nN1r3333nd2IQJm1ZcuWLPfX467jNTAw0O46HxwcbHvInJmmIZLUtm1bu4EAFi9e7BHnQJIWcNjly5f1zTffOLTu2LFj7b6wbdq0UcWKFVNdt3///nbzEyZMcCjDHRUVpTFjxti9lvyEZoWkfWZcv37dqadOSZ+2S7Ib0zs9b7/9tsuH4XSVIUOG2D0lXrVqlZYuXZrtcWT2vf3tt9/sxrq2yjPPPGM3//nnnzs01vjJkyc1ffp0p/Y1dOhQu/mRI0dmuh2rq1l1vBQuXNjuacOuXbscenIREhKizz77LMv7d1bymjeOfD4REREaPXq0U/ux+vuZ/ImsM1Vcvb299fLLL9vmTdPUm2++6bHnOqs52j9TVt7zzEj6GUnSp59+6pJ9evI5Oyf1leUKST+LsLAwh5p8xMXFadiwYdkSk5T9x8eECRMcOi+fPn1a06ZNs3st+X2tK2X2fZk8ebJlI6Qlsvp7884779g1e0xISFDfvn2zNGz41KlT1a1bN7tOPjPDncdr8mTEnDlzFBYWplWrVtleK1asmDp27OhQeYGBgXr66adt87dv39awYcMsrdXnCJIWcMq4cePsalCkZsaMGfr555/tXkteZSypHj16qGTJkrb50NBQ9e3bN90ndrGxsXr++efthjb09/fPlqRF0qfY0t0OrByVvAfyjz/+OMMnkx9++KHH1rKQ7rbbTN6mbsiQIfrtt98c2j4+Pt7ppkepadiwoV27zuXLl2d40dixY4cGDx6c5X074sEHH1TNmjVt81euXNFzzz2X7g+0xO+Cs7UAunfvbnecnjp1St27d9f58+cdLiMhIUHLli3TBx984NS+M2LV8ZIrVy67pmGXL1/OcFjQK1euqGfPng4lj1wt6bkgNjY2ww55IyMj1adPnxTDuWbE6u9nVs6HkjR48GC7atQ7duxQv379nEqyxcTEaPbs2fryyy+d2reV3n77ba1cudLhm7yEhAR9/fXXdq+lNZJK8vf8t99+szTR06NHD7uqzhEREXryyScdbud88+ZN7du3L8Xr2XXOtvKzkKTZs2crICDA7l9Ok/ze5IMPPkh3iMK4uDi99NJL2rNnj2UxufuafvXqVT377LPpXn/DwsJSXKMbNmyYolmVKyWvzfDTTz/pxIkT6W6zYsUKp4elHThwoFPNAqKjo1PU5MvqaFDJFShQQD/99JPdiDZXr15V586d9csvvzg1rOaRI0f05JNPuuyhoDuP10cffdQumb18+fIUNUi7d+/u1PC6w4YNs6vVt3TpUr322mt2o6NkJDIyUpMmTUrxmzCzSFrAYeXKlVNsbKx69+6tMWPG6NKlS3bLz5w5o6FDh6Z4ytulS5d0hxz08fHRN998YzcM07Zt23T//fdr1apVdtnPhIQE/fHHH2rfvn2Km+7x48e7dFzotDz44IN288OGDdPbb7+tBQsWaP369dq0aZPtX/J28p07d7Zrk3fw4EF16tQpRb8cCQkJ2rp1qx5//HH973//kyRVr17dmj/IBcaNG2d3UxsVFaU+ffrohRde0K5du1JcSOLi4nTgwAF98sknatSokUaOHJnlGPLly6cuXbrY5hMSEtSzZ09NnTo1RfXpc+fO6aOPPlLnzp0VHh4uX1/fFJ3CupphGPryyy/t2qNv2LBBDz74oDZt2mT3HkVHR+vXX3/V/fffr2PHjklK2WltenLlyqWZM2fa9V6+e/du3XfffRo7dqzdsKBJhYaGauPGjRoxYoTq1q1r6zXf1aw6XpI+GZDu1vgaOXKkXZtO6e6PqJ9++kmtWrWyNUfI7u9X8kTC9OnT9eKLL6Z4gn379m0tWrRIrVu3tiUPnI3Vyu9n8vPhhAkTNHjwYP3yyy9at26d3fkwtfbN/v7+mjFjht3N3u+//64WLVrof//7X5pP9C9duqRVq1bp9ddfV82aNfXKK6+keVy7w/bt29W3b1/VrVtXI0eO1KZNm1JNxCSe65988km7p2L58uVLs8O0okWLql69erb5U6dOqX379vr222/122+/2b3nae3XGblz59ZPP/1k18nemTNn1KZNG40dOzbV5MWtW7f0xx9/6K233lLdunU1b968FOtk1znbys/iXtGzZ0+7a9PatWvVo0ePFE/mY2Nj9fvvv6tt27ZasGCBJOvOne68pidu+8cff6hdu3bauHGj3XkyJiZGS5cuVevWre2atHl7e9vu2axSsWJFuyRTZGSkHnvsMS1atChFbYHjx49r6NCh6t+/v2JiYlSsWDEVLlzYof2sWbNGTzzxhJo2baqxY8dq27ZtqTahiImJ0Zo1a9ShQwe75GSJEiWyPNx6aurXr6/vvvvO7gd4eHi4Xn75Zd1///2aPHlymsOCnz9/XtOmTVO3bt3UqlUrbdq0yWVxufN4zZMnj91IKpGRkfq///s/u3UcbRqSqEKFCvr222/tXps1a5buu+8+TZs2LcVvwETnzp3Tr7/+qoEDB6pWrVoaOXKkUw/M0sOQp3DYd999p65duyo2NlYTJ07Ul19+qfLly6tQoUK6evWqgoODU2xTvXp1TZgwIcOy27dvr9GjR2vs2LG2106cOKE+ffqoQIECKlu2rHLlyqXg4OBUbzYGDx6c4keAVR5++GHVqFHD9mMyJiZGU6dOTbWtYKtWrbRy5UrbfKFChfTOO+/YZbz37dunhx9+WEWLFlXZsmUVGxur4OBgu06dSpQooS+++EKPPfaYhX9Z5vn5+WnOnDl66qmnbE+CTdPUokWLtGjRIvn7+6tUqVLy8/NTeHi4goOD7U7iSdtMZ8WoUaP0+++/256cR0RE6O2339Z7772nKlWqyMfHJ9Vjdfz48Zo3b57TT7Gd1bx5c73//vv6z3/+Y3vt0KFDevLJJ1WoUCGVLVtWcXFxOnfunN345/3791dsbKxT8VWuXFm//PKL+vXrZ2tudfPmTU2YMEETJkxQoUKFVKJECeXPn1+RkZG6ceNGih/3VrHqeOnTp4+mTZtmqypqmqYmTZqk77//XpUqVVKhQoUUFhams2fP2t3c9ezZUxUrVtT48eMt/KvttWvXTo888ohdtdD58+dr/vz5to5bIyIidPbsWbsnG61bt1bPnj31+uuvO7wvK7+fderUUfv27bV+/XpJd2/U5s2bl+qP1LJly6YY7Um6++RwypQpGjJkiG2/ly5d0ocffqgPP/xQgYGBKlasmHx9fRUeHq7r1697VK/x6QkJCdGkSZM0adIkSXf7nClUqJDy5cunW7du6ezZs6n+EPjkk0/saiAmN3ToUL3wwgu2+YMHD6b63kp3n7ol71jNWXXq1NGUKVM0aNAg22d0+/Zt2/kkMDBQxYsXV+7cuRUaGqrg4GCHRqrJznO2VZ/FvaBatWp64YUX7NrWr1+/XuvXr1fJkiVVqlQpRUVFpbg21ahRQx988IHTP4gc5a5r+tNPP609e/Zo/fr1OnLkiLp27aqAgACVLVtW8fHxKd6HRJ988oldQtEqH3/8sR577DHbdezKlSt64YUXlC9fPlWuXFm5cuXS5cuXdfHiRds23t7emjRpkt58802nOhn9559/bN9zLy8v2/fGx8dH4eHhOnPmTIraKN7e3vr666/tEp2u1K1bN/n7++vFF1+0Gyr00KFDeueddyTd7QOjaNGiCggIUFhYmK5cuZJmc+4iRYrY9eGQWe68B+3Tp4/d9zfpfUPdunVVt25dp8vs2rWrLl++rFGjRtnO56dOndJbb72lt956S6VKlVKRIkWUJ08ehYeH68qVKw53Gp8Z1LSAw1q3bq0pU6bYqmUlJCTo9OnT2rt3b6oJizp16mjp0qUO134YNmyYvvzyyxQnuYiICB05ckQHDx5MkbDInTu3Pvjgg2z9sZErVy7NmjXLqZ6Bk3rttddSHer02rVr2rdvnw4dOmT3pS9btqx+/fVXlS5dOtMxZ4cqVapo/fr1uv/++1MsCw8P17Fjx7R7924dP37cso7jKlSooJkzZ8rf39/u9ZiYGB05ciTFsZorVy6NGzfO0vanyb355pt6//33U4wAcOPGDf399986cuSI3c1Qly5dHEr8paZVq1bauHFjqiPq3LhxQ0ePHtXu3bt17NixNBMWZcqUydS+M2LF8ZI7d27Nnj07xXfTNE2dPHlSu3fv1okTJ+wSFj169HC4rx5Xmzx5spo1a5bi9XPnzmn37t0KCgqyu/F44IEHNHv2bIdHGknKyu/npEmT1KRJE6djSqpLly5avXq1qlWrlmLZlStXdPjwYVt8qSUsDMPw+HOkJF24cEGHDx/Wrl27dOTIkRQ/kvPly6evvvoqw6aO3bp10/Dhw1OcR6z0+OOPa+XKlSmGw5PufkYHDx7U3r17debMGYeH1nXnOdtVn8W94pNPPkn1wcjFixe1Z8+eFNemOnXqaPHixXa1R13NXceHYRj66aef7DrUDAsL08GDB1O8D9LdH+njxo3Tiy++mKX9Oqpp06b69ttvUwx1evv2bR08eFD79++3S1j4+vpq6tSpeuihh7K034SEBIWEhOjgwYPatWuXgoKCUiQsChUqpFmzZqlDhw5Z2ldGOnTooK1bt+qpp56yq6md6ObNmzp58qT27NmjkydPppqw8PX11dChQ7V37167TvYzy53ns4YNG6ZoOpgoK0nFIUOGaNGiRaleXy9cuKCDBw9qz549+ueff1JNWOTKlctlSV+SFnDKk08+qY0bN6pjx46pniSku72Nv//++9q4caNKlCjhVPnPPfecdu/erX79+tlVbU/Oz89P3bt31/bt2/XWW285tQ9XqFKlirZs2aLJkyfrqaeeUrVq1eTv7+9we7HPP/9cU6dOVeXKldNcx9/fX6+//rq2bt1q1xeCJytWrJiWLVumBQsWqE2bNhlmritVqqQ33nhDs2bNclkMDzzwgDZs2KBOnTqleYwahqGHHnpI69ev15AhQ1y2b0cNGzZMa9euVcuWLdOMsVy5cvrqq680Y8aMLD0BKF++vNauXasFCxaoXbt28vHxSXd9wzBUv359vfXWW/rrr7/0ww8/ZHrfGbHieClTpozWr1+vl156KcUoC0nVqlVL06ZN05QpUxwe497V/P39tWLFCo0cOTLddvDly5fXhAkT9Ouvv6Z7XsyIVd/PwMBA/f7775o1a5aefvpp1a5dWwEBAU4ft/Xr19f27ds1depUtWjRIsPzqbe3t5o3b65Ro0Zp3759TndSaqW5c+fq888/18MPP+zQ8JFFihTRoEGDtHPnTodvYEeNGqW//vpLb7zxhu677z4VL17csiebiRo1aqTdu3fr888/T/MGOZGXl5eaNGmiTz/9VO+++26a61l9zs6Oz+JekJj0/eyzz9L9kZF4j7d+/XqVKlXK8rjcdU339/fXkiVLNHbs2BSdLCby8vJS27ZttWnTpmy/l+jZs6dWr16dbi0qb29vde3a1dbsyRkbN27U2LFj1bZtW/n5+WW4funSpfXmm29qz549euSRR5zaV2aVKlVK06ZN044dO/Tyyy879JAlT548atmypSZOnKijR4/qP//5T5auq8m58x40tRrnuXPnVs+ePbNUbtu2bbVv3z5NnDhR9erVyzBZnjdvXj3wwAP6+OOPdeTIEZclfo2wsDD3dgUKj9WpUydt3brVNp+8lsPly5e1fft2nTt3ztZWrnLlymrZsqVLnv7ExcVp9+7dOnXqlK5fv66EhAQVKVJE5cuXV/Pmzd32Q8OVTNPU4cOHtXfvXru/sXr16mrSpIlLqqu5U0REhHbs2KGLFy/a/r4CBQqoXLlyql27tmVP8RNdvXpVf/31l0JCQnT79m3lz59fFSpUULNmzbKl/xNHBAcHa+fOnbpw4YLte1SvXj3Vr18/zQteVty5c0e7d+9WSEiIrl+/bntfChUqpMqVK6t69eouvYA7w9XHy+3bt7V9+3adOHFCN2/elI+Pj0qWLKmGDRummzB0h5iYGO3cuVNBQUG6ceOGvL29VaJECdWtW1e1a9e2ZJ/u/n5mJDIyUrt27dKFCxcUGhqq6Oho+fn5qUiRIqpSpYqqVavmkqdjVjNNU6dOndKJEycUEhKiiIgIxcXFKX/+/CpWrJhq1aql6tWrZ6oGjbtdvHhRu3fv1tWrV3Xjxg3lzp1bBQsWVKVKlVS3bl2nO6W0+px9L38WrhQfH2+r+RkaGirDMFS0aFHVrl1bDRo0yNYaPkm565oeHx+vPXv26PDhwwoNDZWPj49KlSqlli1bOv1wzgrBwcG2c/mdO3fk7++vypUrq2nTpi65nsfHx+v48eM6deqULly4oIiICJmmKT8/P9uwyJUrV7bknsVZwcHBOnr0qIKDgxUeHq74+Hj5+/srICBAlStXVt26dbPt90NOuAfNjBs3bmjXrl26fPmyQkNDFRcXJz8/PwUGBqpq1aqqWrWqXWeprkLSAmnKKGkBAAAAAICVaB4CAAAAAAA8EkkLAAAAAADgkUhaAAAAAAAAj0TSAgAAAAAAeCSSFgAAAAAAwCORtAAAAAAAAB6JpAUAAAAAAPBIRlhYmOnuIAAAAAAAAJLzdncA96LY2FitXbtW+/fv14EDB3TmzBmFhoYqLCxMefPmVYkSJdSwYUN17dpVjz76qAzDcHfIAAAAAAB4HGpaWOD8+fOqXbu2Q+s2b95cM2bMUIkSJSyOCgAAAACAnIWkhQXOnz+v1q1bq1WrVmrYsKHKly+vwMBABQQEKDw8XIcOHdLPP/+sI0eOSJKqV6+uP/74Qz4+Pm6OHAAAAAAAz0HSwgIJCQmSJC+vtPs5jYuL0zPPPKPVq1dLkj7//HMNHDgwW+IDAAAAACAnYPQQC3h5eaWbsJAkb29vvfXWW7b5LVu2WB0WAAAAAAA5CkkLNypQoIBtOiIiwo2RAAAAAADgeUhauNH8+fNt01WrVnVjJAAAAAAAeB6GPM1GCQkJunr1qo4fP66ff/5ZCxYskCTlyZNHzz//vJujAwAAAADAs5C0sNj169dVuXLlNJf7+/tr6tSpql69eqbKT+z0EwAAAACApDLqazEnIGnhJoZh6OWXX9bQoUMVGBiY6XLOnj2ruLg4F0YGAAAAAMjpvL29VbFiRXeHkWUkLSwWEBCgv/76S5IUHx+v0NBQ7dq1S9OnT9f333+vEydO6IsvvlCpUqXcHCkAAAAAAJ7FCAsLM90dxL9RZGSk+vfvrw0bNigwMFDLly/PVBMRmocAAAAAAFJzLzQPIWnhRlevXlW9evUUFRWl++67T6tWrXJ3SAAAAAAAeIycn3bJwYoVK6YWLVpIkv766y9dunTJzREBAAAAAOA5SFq4WZEiRWzT586dc2MkAAAAAAB4FpIWbnb+/HnbtJ+fnxsjAQAAAADAs5C0cKOzZ89q9+7dkqT8+fPfE8PRAAAAAADgKiQtLDB//nxdv3493XWuXbumAQMGKDY2VpLUrVs3+fr6Zkd4AAAAAADkCIweYoFOnTppz5496tChg9q0aaPq1asrICBA8fHxunDhgrZu3ao5c+YoLCxMklSpUiWtW7dOhQsXdm/gAAAAAAB4EJIWFujUqZO2bt3q0LodO3bU119/reLFi1scFQAAAAAAOQtJCwuEhIRo3bp12rFjh4KCgnTlyhVdv35d8fHx8vf3V6VKldS4cWN169ZNTZo0cXe4AAAAAAB4JJIWAAAAAADAI9ERJwAAAAAA8EgkLQAAAAAAgEciaQEAAAAAADwSSQsAAAAAAOCRSFoAAAAAAACPRNICAAAAAAB4JJIWAAAAAADAI5G0AAAAAAAAHsnb3QEAAAAAAJCT1f5up7tDsHP45WbuDsFlqGkBAAAAAAA8EkkLAAAAAADgkUhaAAAAAAAAj0TSAgAAAAAAeCSSFgAAAAAAwCORtAAAAAAAAB6JpAUAAAAAAPBIJC0AAAAAAIBHImkBAAAAAAA8EkkLAAAAAADgkUhaAAAAAAAAj0TSAgAAAAAAeCSSFgAAAAAAwCN5TNIiODhYQ4cOVZ06dVS8eHHVqFFDAwcO1MGDB90dGgAAAAAAcANLkxaXL19W6dKlVbJkSTVt2lSmaaa63uHDh/XAAw9o5syZOn/+vGJiYnT58mUtXrxY7du316JFi6wMEwAAAAAAeCBLkxa//fabbt++rejoaPXp00eGYaRYxzRNDRo0SDdu3Eh1WWxsrF566SUdO3bMylABAAAAAICHsTRpsX37dtv0448/nuo6y5Yt09GjR20Jjd69e2vWrFmaNGmSqlevLkmKi4vTxx9/bGWoAAAAAADAw3hbWXhi7Qh/f39VrVo11XXmzZtnm37uuec0ceJE2/xDDz2k5s2bKzQ0VGvWrNHNmzdVsGBBK0MGAAAAAAAewtKaFiEhITIMQ5UqVUp1eVxcnDZv3mybf/XVV+2WFy1aVP369bOtu2/fPuuCBQAAAAAAHsXSpEVkZKSkuzUtUrNv3z7dunVLhmGoRo0aqSY3mjRpYps+efKkNYECAAAAAACPY2nSIiEhQZIUHx+f6vKkfV60bt061XWKFi1qmw4PD3dhdAAAAAAAwJNZmrRIrGFx6dKlVJcnbRrSqlWrVNeJiYmxTac2+ggAAAAAALg3WZq0qFKlikzT1KlTp3T58mW7ZTdv3tSWLVts82nVtLh27ZptOq1mJgAAAAAA4N5jadKiZcuWkiTTNPXpp5/aLZs4caKio6NlGIbq16+vIkWKpFrGoUOHbNPlypWzLlgAAAAAAOBRLB3ytG/fvvr666+VkJCgn3/+WUFBQWrZsqUOHTqkdevW2dbr379/mmVs3brVNl2zZk0rwwUAAAAAAB7E0qRFlSpV9Prrr+uLL76QYRjasWOHduzYYbdOzZo1bcOaJhcSEqJdu3bJMAyVKlVKpUuXtjJcAAAAAADgQSxtHiJJH3zwgYYNGyZvb2+Zpmn3r2HDhpo3b568vVPPnfz4448yTVOS1L59e6tDBQAAAAAAHsTSmhaJRo8ercGDB2vDhg26cOGC8ubNq4YNG9r6vEhLVFSUevfuLUlp1sYAAAAAAAD3JiMsLMx0dxAAAAAAAORUtb/b6e4Q7Bx+uZm7Q3AZy5uHAAAAAAAAZAZJCwAAAAAA4JFIWgAAAAAAAI+ULR1xJmWapo4cOaKLFy8qPDxc0dHRDm+b2CknAAAAAAC492Vb0mLPnj369ttvtWbNGt2+fdvp7Q3DIGkBAAAAAMC/SLYkLcaOHasvv/xSpmnKNNMfrMQwjAzXAQAAAAAA9z7LkxYTJ07UxIkTbfM+Pj6qWbOmTp8+rbCwMBmGofvuu0+RkZG6cOGCrl69Kulu8kKSGjdurLx581odJgAAAAAA8DCWdsR54cIFjR8/XtLdJMTgwYN14sQJbdiwQY0bN7att2LFCm3atEnHjx/Xvn37NGzYMPn6+kqSoqOj9c0332jFihVWhgoAAAAAADyMpUmLGTNmKCYmRoZhqEePHho3bpzy58+f7jYVKlTQ6NGj9eeff6p8+fI6dOiQnnjiCYWFhVkZKgAAAAAA8DCWJi3+/PNP2/S7777r1LYVK1bUokWLlDdvXoWEhDi9PQAAAAAAyNksTVqcOnVKklSmTBlVrFgxzfXi4uJSfb1SpUrq3bu3TNPU0qVLFR4ebkmcAAAAAADA81iatLhx44YMw1DZsmVTLMudO7dtOioqKs0y2rVrJ+lu3xZbt251fZAAAAAAAMAjWZq08PK6W3zSBEWiAgUK2KYvXbqUZhlFixa1TV+4cMGF0QEAAAAAAE9madKicOHCkpRqs44SJUrYpo8dO5ZmGaGhobZpmocAAAAAAPDvYWnSokqVKjJNU6dPn06xrE6dOrbptWvXpllG0mWFChVybYAAAAAAAMBjWZq0aNSokSTp5s2bOnPmjN2y9u3by9vbW5I0f/587dy5M8X2Gzdu1OzZs23zjRs3ti5YAAAAAADgUSxNWjzwwAO26d9//91uWZEiRdSlSxeZpqno6Gh17txZr7zyin788Uf9+OOPGjRokHr27Km4uDgZhqG6deuqbt26VoYLAAAAAAA8iBEWFmZaVXh8fLyqVaum0NBQ1a5dW3/++afd8suXL6tNmza6du2aTNOUYRh2y03zbmg+Pj5atWqVGjZsaFWoAAAAAABkSu3vUrYccKfDLzdzdwgu421l4bly5dKMGTNso35ERUXJ19fXtrx48eJasWKFnnvuOR09etSWpEiqRIkSmjJlCgkLAAAAAAD+ZSytaeGohIQErV69Whs3blRwcLDi4uJUvHhxtW7dWl27dpWPj4+7QwQAAAAAIFXUtLCOpTUtHOXl5aXHHntMjz32mLtDAQAAAAAAHsLSjjgBAAAAAAAyi6QFAAAAAADwSCQtAAAAAACAR7I0aREVFaUKFSqocOHCqlKlim7fvu3U9l9++aUKFy6swoUL69tvv7UoSgAAAAAA4IksTVqsXLlSN2/elCQNHDhQ+fLlc2r7gQMHytfXV6ZpatasWVaECAAAAAAAPJSlSYv169fbprt16+b09n5+fnr44YclSUFBQQoJCXFZbAAAAAAAwLNZmrT4+++/JUlFihRR1apVM1XGfffdl6I8AAAAAABw7/O2svDg4GAZhqFKlSpluoyk2wYHB7siLAAAAACAB6r93U53h2Dn8MvN3B3Cv57lHXFKkq+vb6bLSLrtrVu3shwTAAAAAADIGSytaVGwYEGFhobq+vXrmS4jNDTUNp0/f35XhJUtTp8+rXXr1umvv/7S4cOHdeHCBUVHR6tgwYKqXr26HnzwQfXr10+BgYHuDhUAAAAAAI9kadIiMDBQ169fV1BQkCIiIlSgQAGny9i1a5ddeTnBSy+9pDlz5qS67Nq1a7p27Zq2bt2q//3vf/rvf/+rXr16ZXOEAAAAAAB4PkubhyR2ohkXF6eZM2c6vX1UVJTmz59vm2/WLGe0J7pw4YKkuzVDunXrpv/9739auXKl/vjjD/3yyy/q2bOnDMNQeHi4hgwZosWLF7s5YgAAAAAAPI+lSYtHHnnENj1+/HgdOXLEqe1HjBihixcvyjAM1alTR6VLl3Z1iJYoUaKEPv30UwUFBenHH3/Us88+q1atWql+/fp67LHHNHnyZE2dOlWSZJqmhg8frtu3b7s5agAAAAAAPIulSYuHHnpIDRs2lCSFh4friSee0NKlSzPc7vr163r++ec1a9Ys22vvvPOOZXG62g8//KCXXnpJfn5+aa7TrVs3derUSdLdv3fTpk3ZFB0AAAAAADmDpX1aSNLEiRPVqVMnRUVF6fr16xowYICqV6+uxx9/XA0aNFCxYsWUJ08e3bx5U6dOndJff/2llStX6s6dOzJNU4Zh6IknntDjjz9udajZ7oEHHtDKlSslSSdPnnRzNAAAAAAAeBbLkxYNGjTQjz/+qBdeeEFRUVEyTVNBQUEKCgpKc5vEZIUktW3bVpMnT7Y6TLeIiYmxTXt5WVrpBQAAAACAHCdbfik/8sgjWr9+vRo0aCDpblLCNM0U04nzkpQvXz6NGjVKixYtUp48ebIjzGy3ZcsW23TNmjXdGAkAAAAAAJ7H8poWiWrUqKENGzZo8+bNWrRokbZt26YTJ07YlpumqQIFCqhJkyZq166d+vXrp4CAgOwKL9vt27dP69atkySVKlVKbdq0yVQ5CQkJrgwLAAAAAPD/y6m/txLjvhdq9Gdb0iLR/fffr/vvv1/S3URFWFiYoqOjVbBgQfn6+mZ3OG4RGRmpl19+WfHx8ZKkDz74QLlz585UWWfPnlVcXJwrwwMAAAAAKOf2PXjy5El5e3urYsWK7g4ly7I9aZGUYRgqVKiQO0PIdgkJCRo0aJCOHj0q6e4oIk8//bSbowIAAAAAwPO4NWnxb2Oapl5//XX99ttvkqQmTZroq6++ylKZ5cuXd0VoAAAAAOB+a3e7OwI7lStXdmzFnBp3DkDSIpuYpqlhw4Zp1qxZkqR69epp4cKFyp8/f5bKvRfaKAEAAACAJ8qpv7dyatypuXf+Eg/3zjvvaNq0aZKk2rVra8mSJfd0R6MAAAAAAGRVlmtadO7c2TZtGIaWLVuW6rKsSl52TjJ8+HBNmTJFklSrVi0tW7ZMhQsXdnNUAAAAAAB4tiwnLf78808ZhiHTNGUYRqrLsiq1snOKd955x5awqFmzppYtW6YiRYq4OSoAAAAAADyfS/q0ME0zU8vudSNGjNDkyZMl/b+ERdGiRd0cFQAAAAAAOUOWkxbffvttppbd60aOHKkffvhBklSjRg0tW7ZMxYoVc3NUAAAAAADkHFlOWvTp0ydTy+5lH374oSZNmiRJKlq0qMaPH6+rV6/q6tWraW4TEBCgUqVKZVeIAAAAAAB4PIY8tcDixYtt09euXVOXLl0y3KZ37962RAcAAAAAAGDIUwAAAAAA4KEsrWkxfvx4SVKlSpXUo0ePTJXx66+/6vjx45LudmyZExw8eNDdIQAAAAAAkONZmrQYN26cDMNQ+/btM520WLhwoVatWiXDMHJM0gIAAAAAAGQdzUMAAAAAAIBHImkBAAAAAAA8kscnLW7fvi1J8vX1dXMkAAAAAAAgO3l00sI0TR09elSSVLhwYTdHAwAAAAAAspOlHXFm1p07d3Tq1Cn98MMPunz5sgzDUJ06ddwdFgAAAAAAyEYuS1qkVRPCNE2tX78+yzUlunTpkqXtAQAAAABAzuKypIVpmjIMQ6ZpprosK9q0aaNevXplqQwAAAAAAJCzuLR5SFaTE0n5+Piodu3a6t69uwYOHCjDMFxWNgAAAAAA8HwuS1ocOHDAbt40TTVo0ECGYei+++7Td99951A5Xl5eyp8/vwoWLCgvL4/uJxQAAAAAAFjIZUmLcuXKpfq6aZry9fVNczkAAAAAAEBqLB09ZMSIEZKkSpUqWbkbAAAAAABwD7I0afHuu+9aWTwAAAAAALiH0WkEAAAAAADwSJbWtHBGQkKCFi1apPXr1+vq1asqVKiQWrVqpV69eilfvnzuDg8AAAAAAGQzS5MWt2/f1oABA5SQkCA/Pz9NmzYt1aFLb9y4oR49emjv3r12ry9evFj/+9//NGfOHNWsWdPKUAEAAAAAgIextHnI2rVrtWbNGq1fv17FixdPNWEhSa+++qr27Nkj0zRT/Dt79qyeeuophYWFWRkqAAAAAADwMJYmLf7880/bdNeuXVNdZ9euXVq1apUtoVGtWjUNGTJEffv2Vb58+WQYhi5fvqxx48ZZGSoAAAAAAPAwliYtDh06JEnKmzevmjZtmuo6M2fOtE23bt1amzdv1qeffqpvvvlGy5Ytk7e3t0zT1Ny5cxUTE2NluAAAAAAAwINYmrQ4e/asDMNQ5cqV5eWV+q5+//132/R7772nvHnz2uYbN26s7t27S5LCw8N14MABK8MFAAAAAAAexNKkRWI/FIUKFUp1eVBQkK5cuSLDMFSiRAm1bNkyxTrt2rWzTR89etSSOAEAAAAAgOexNGmR2JwjV65cqS7fvn27bbpNmzaprlO6dGnbNJ1xAgAAAADw72HpkKd+fn4KDw/X9evXU12+detW23SrVq0yLC8uLs5lsQEAAADAvar2dzvdHUIKh19u5u4QkANZWtOiTJkyMk1TJ06cUFRUlN2yuLg4rVu3zjbfunXrVMu4ceOGbTp//vzWBAoAAAAAADyOpUmLxo0bS5Kio6M1Y8YMu2ULFy7UjRs3ZBiGypQpo0qVKqVaRlBQkG26VKlS1gULAAAAAAA8iqXNQ7p3766ff/5ZkvThhx8qNDRULVq00KFDhzR+/Hjbej179kyzjJ07/1+1pmrVqlkXLAAAAAAA8CiWJi3atGmjDh06aO3atYqNjdXnn3+eYp2AgAC98sorqW4fERGhTZs2SZL8/f1VvXp1K8MFAAAAAAAexNLmIZI0depUtWjRQqZppvhXoEABTZ8+Pc0hUefPn6/o6GgZhpHm6CIAAAAAAODeZGlNC+luDYnffvtNy5cv15o1a3ThwgXlzZtXDRs2VL9+/VSiRIk0t126dKnKli0rSeratavVoQIAAAAAAA+S6aRFXFycvL0d37xz587q3LmzU/tYtmyZs2EBAAAAAIB7RKabh3z77bdavHixK2MBAAAAAACwyXTSIioqSgMHDtSECRNcGQ8AAAAAAICkLHbEaZqm/u///k8vv/yy4uLiXBUTAAAAAABA5pMWjz32mMqWLSvTNDV37lx17dpVYWFhLgwNAAAAAAD8m2W6I8569eppy5Ytev3117Vs2TJt3bpVHTt21Pz581WhQgVJUuHChV0VpwzD0PXr111WHgAAAAAA8GxZah5SsGBBzZgxQxMnTpSPj4/++ecfdejQwbbcNE3b/674BwAAAAAA/j2ylLRINGDAAK1bt07Vq1dPURuCZAMAAAAAAMiMTDcPSa5WrVratGmT3n33XdtrBw4ccFXxAAAAAADgX8ZlSQtJ8vHx0ZdffmmbL1eunCuLBwAAAIBsVfu7ne4OIYXDLzdzdwhAtnFp0gIAAAAA0uJpCQB+/AOez9KkxXvvvWebfvnll1WmTBkrdwcAAAAAAO4hliYtJk2aJMMwVKRIEf3f//2flbsCAAAA/jU8rcaCRK0FANZwyeghafHz85MkVa5cWYZhWLkrAAAAAABwj7E0aVG8eHFJImEBAAAAAACcZmnSon79+jJNU6dOnbJyNwAAAAAA4B5kadKiZ8+ekqSrV69q48aNVu4KAAAAAADcYyxNWjz88MN68MEHZZqmhg8frmvXrlm5OwAAAAAAcA+xdPQQSfrhhx/Us2dP7du3T+3atdMnn3yiTp06ycvL0nwJAAAAkCFG4QAAz2Zp0mL8+PGSpLZt2+rEiRMKCQnRs88+q6JFi6pJkyaqWLGi/Pz8HE5gjBgxwspwAQAAAACAB7E0aTFu3Di7kUMMw5Bpmrp69apWr17tdHkkLQAAAAAA+PewvHmIaZpOvZ4Whk0FAAAAAODfxdKkRe/eva0sHgAA4J6UU/tZ8LS46RsCAHI+S5MW3333nZXFAwAAAACAe5jlzUMAAADcxdOe/Es8/QcAwBmMOwoAAAAAADwSSQsAAAAAAOCR3NI8JCoqSocOHdK1a9cUHh6uhIQEOu0EAAAAAAB2si1pkZCQoMWLF2vq1Knas2eP4uPj7ZanlrR45513FBoaKkn67LPPVLhw4WyJFQAAAAAAuF+2JC3Onj2rQYMGaffu3ZIk0zTtlhuGkep2RYsW1ZQpU2QYhpo2barBgwdbHisAAAAAAPAMlvdpERwcrMcee0y7d++WaZoyTVP58uVT3bp1FRAQkO62zz77rLy87oa4ZMkSq0MFAAAAAAAexPKkxfPPP68LFy7INE2VLFlSP/74o86cOaPNmzercePG6W5bvHhxNWvWTKZpas+ePbp165bV4QIAAAAAAA9hadJixYoV2r17twzDULly5bRx40Y99dRTyp07t8NltGjRQpIUFxenw4cPWxUqAAAAAADwMJYmLZYtW2abnjhxoooXL+50GXXq1LFNnzhxwiVxAQAAAAAAz2dp0mLXrl2SpMDAQLVv3z5TZRQpUsQ2nTiSCAAAAAAAuPdZmrS4evWqDMNQ5cqVM12Gj4+PbTomJsYVYQEAAAAAgBzA8o44syosLMw2XbBgQfcFAgAAAAAAspWlSYtixYrJNE2dO3cu02UcPHjQNh0YGOiKsAAAAAAAQA7gbWXhtWvX1pkzZ3T+/HkFBQWpevXqTpeRtDPP5s2buzI8S4WHh+vvv//Wvn37tH//fu3fv1+nTp2SaZqSpAMHDqh8+fJujhIAAAAAAM9ladKiY8eOWrlypSRp3Lhx+umnn5zafsGCBTp06JAMw1C9evVyVE2LTp062dUSAQAAAAAAzrG0eUiPHj1UsmRJSdLSpUs1btw4h7ddv3693nzzTdv80KFDXR6flRJrVEiSv7+/WrdunakhXwEAAAAA+LeyNGnh6+ur//u//7P9gP/ss8/06KOPavny5bp582aK9SMjI7Vx40YNHDhQvXr10q1bt2QYhlq3bq2uXbtaGarLPfPMM5o6dar27Nmjs2fPasWKFapSpYq7wwIAAAAAIMewtHmIJHXt2lWnT5/W2LFjZRiGduzYoR07dkiSvLz+X86kZMmSio6Ots0nJjqqVKmiGTNmWB2myw0ZMsTdIQAAAAAAkKNly5Cnb731lqZPny5/f3+Zpmn7l5CQIMMwJEl37tyxWybd7RNjzZo1KlSoUHaECQAAAAAAPEi2JC0kqUuXLjpw4IA++ugj1axZU4ZhpEhSSHeblHTs2FFLlizRvHnzFBAQkF0hAgAAAAAAD2J585CkChYsqKFDh2ro0KEKCwvT0aNHFRoaqtu3b8vf31+BgYGqU6eOcufOnZ1hAQAAAAAAD5StSYukAgIC1LJlS3ft/p6RkJDg7hAAAIATcuq1OyfGnRNjlog7O+XEmCXizk45MWbp/8WdtB/JnMptSQu4xtmzZxUXF+fuMAAAgINOnjzp7hAyJSfGnRNjlog7O+XEmCXizk45MWbpbtze3t6qWLGiu0PJMkvTLvXr11f9+vX1yiuvZLqMN954Q/Xr11eDBg1cFxgAAAAAAPB4lta0OHfunAzD0JUrVzJdxtWrV23lIKXy5cu7OwQAADzX2t3ujiCFypUrZ7wScbtEToxZIu7slBNjlog7OzkUs5Rz484BaB6Sw90LbZQAAPg3yanX7pwYd06MWSLu7JQTY5aIOzvlxJilnBt3ajz+L7mXOhABAAAAAACO8/hMQGhoqCQpf/78bo4EAAAAAABkJ49OWly6dEkHDhyQYRgqU6aMu8MBAAAAAADZyGV9WsyZMyfNZRcuXEh3eVKmaerOnTs6deqUlixZoujoaBmGoebNm7sqVAAAAAAAkAO4LGnx8ssvpzrCh2maOnr0aJaGPfXy8tKzzz6blfAAAAAAAEAO49LRQ0zTdOp1R+TNm1fjx49X/fr1M12GO5w6dUrbtm2zey3p0K/Lli1T4cKFbfN+fn7q0qVLtsUHAAAAAICnc1nS4r777ktR02Lr1q0yDEMFCxZU7dq1HSrHy8tL+fLlU/HixVW/fn117txZxYoVc1WY2Wbbtm3p1i55//337ebLli1L0gIAAAAAgCRclrRYuXJlitcKFSokSWrSpIkWLFjgql0BAAAAAIB/AZc2D0lNVpqG5GR9+/ZV37593R0GAAAAAAA5lqVJiwMHDkiSfH19rdwNAAAAAAC4B1matChXrpyVxQMAAAAAgHuYl7sDAAAAAAAASI2lNS1M01S/fv108+ZN5c2bVz///LPy5cvn8PYzZszQwoULJUmvvvqqHn74YatCBQAAAAAAHsbSpMUff/yhlStXyjAM9evXz6mEhSQ9+OCDeuutt2SapgzDIGkBAAAAAMC/iKXNQ1avXm2b7t27t9Pbly1bVq1bt5Zpmtq2bZtu3rzpyvAAAAAAAIAHszRpsXfvXklSvnz51Lx580yV0a5dO0lSfHy89u3b57LYAAAAAACAZ7M0aXHy5EkZhqGqVavKMIxMlVGjRg3b9KlTp1wVGgAAAAAA8HCWJi3Cw8MlSQULFsx0GQEBAbZpmocAAAAAAPDvYWnSIrHjzcTkRWZERETYpr29Le03FAAAAAAAeBBLkxZFixaVaZo6ceKE4uPjM1XGkSNHbNPFihVzVWgAAAAAAMDDWZq0aNCggSQpMjLSbiQRZyxatMg2Xbt2bVeEBQAAAAAAcgBLkxbt27e3TX/44YdO90kxbdo0HTx4UIZhqGTJkqpbt66rQwQAAAAAAB7K0qRFt27dVKpUKUl3R/7o1q2bgoODHdr2xx9/1IgRI2zzL730kiUxAgAAAAAAz2Rp0iJv3rwaM2aMTNOUJO3Zs0ctW7bU8OHD9ccff9jVvEhISNA///yjGTNmqF27dho+fLji4uJkGIZq1KihF1980cpQAQAAAACAh7F8OI5u3brp2LFj+vzzz2UYhm7duqUff/xRP/74oyQpd+7cypMnj27dupXq9iVKlND8+fOVJ08eq0MFAAAAAAAexNKaFolGjRqlb775xjYEqmmatn8xMTGKjIy0zSdd/sADD2jTpk0qU6ZMdoQJAAAAAAA8SLYkLSSpb9++OnDggN5++21VqlQpzfV8fX3VsWNHLVq0SEuWLFFgYGB2hQgAAAAAADyI5c1DkipSpIhGjRqlUaNG6cqVKwoKCtKNGzcUExOjggULKjAwULVr15a3d7aGBQAAAAAAPJDbsgOBgYHUogAAAAAAAGnKtuYhAAAAAAAAziBpAQAAAAAAPBJJCwAAAAAA4JGytU+LhIQE7d27V3v27NHFixcVHh6u6Ohoh7Y1DEPffPONxRECAAAAAABPkS1JC9M09c033+j777/XxYsXM10OSQsAAAAAAP49LE9a3L59W927d9f27dsl3U1gZMQwjBTrGYZhSXwAAAAAAMAzWZ60ePXVV7Vt2zbbfIsWLdS0aVOtWLFCp0+flmEYeueddxQZGakLFy5o586dOn/+vKS7iYpu3bqpcuXKVocJAAAAAAA8jKVJi3379unXX3+VJOXNm1dTp07V448/Lkk6cuSITp8+LUl699137bbbtGmTRo0apSNHjuj333/X999/r8cee8zKUAEAAAAAgIexdPSQuXPnSrpbY2LYsGG2hEVG2rZtq02bNqljx46KiIjQwIEDdezYMStDBQAAAAAAHsbSpEViPxa5cuXSoEGDnNo2d+7c+vHHH1WiRAlFRUVp2LBhVoQIAAAAAAA8lKVJi/Pnz8swDFWpUkUBAQF2y5J2rBkTE5Pq9n5+furXr58kadu2bQoODrYsVgAAAAAA4FksTVrcvHlTkhQYGJhiWd68eW3Tt27dSrOMpk2b2qZ3797twugAAAAAAIAnszRp4ePjI0mKj49PsaxAgQK26cTRQlLj5+dnm7506ZILowMAAAAAAJ7M0qRFYGCgTNNUWFhYimUVKlSwTe/fvz/NMpI2CUmrGQkAAAAAALj3WJq0qF69uiTp1KlTMk3Tbln9+vVt00uWLEmzjPnz59umS5Qo4doAAQAAAACAx7I0adGiRQtJ0p07d3TgwAG7Ze3atZO/v78kacOGDfr666/tlickJOjTTz/V+vXrJd3tuLNVq1ZWhgsAAAAAADyIpUmLdu3a2aZXrFhhtyxv3rx68cUXbTUwPvzwQzVo0EADBgzQgAEDVK9ePf33v/+VdDdh8dhjj6lMmTJWhgsAAAAAADyIpUmLunXrqnLlyjJNU7NmzVJsbKzd8rffflvNmze3JS7OnTunpUuXaunSpbpw4YJtvbJly+qLL76wMlQAAAAAAOBhLE1aSNK6det0+PBhbdiwQV5e9rvLmzevFi1apOeff15eXl4yTTPFv06dOmnt2rUqWrSo1aECAAAAAAAP4m31DgICAhQQEJDm8vz582vChAkaPXq0Nm/erODgYMXFxal48eJq1aqVypUrZ3WIAAAAAADAA1metHBUoUKF1KVLF3eHAQAAAAAAPISlSYtVq1bZplu0aKHChQtbuTsAAAAAAHAPsTRp0bdvXxmGoXz58ikoKMjKXQEAAAAAgHuMpR1x5smTR6ZpqmrVqsqfP7+VuwIAAAAAAPcYS5MWxYsXl2EYKlCggJW7AQAAAAAA9yBLkxZVqlSRaZoKCQmxcjcAAAAAAOAeZGnS4sknn5QknTlzRsePH7dyVwAAAAAA4B5jadKie/fuqlixoiTp3XffVUJCgpW7AwAAAAAA9xBLkxb58uXTtGnT5O/vr02bNql37966dOmSlbsEAAAAAAD3CEuHPN26daskaezYsRo1apTWrl2rBg0aqGPHjmrRooUqVqwoPz8/eXk5ljtp1aqVleECAAAAAAAPYmnS4vHHH5dhGLZ50zQVHR2tFStWaMWKFU6VZRiGrl+/7uoQAQAAAACAh7I0aZHINE0ZhpEigQEAAAAAAJAWS5MWZcqUsUtUAAAAAAAAOMrSpMXBgwetLB4AAAAAANzDLB09BAAAAAAAILNIWgAAAAAAAI9E0gIAAAAAAHgkkhYAAAAAAMAjkbQAAAAAAAAeydLRQ5I7fvy4Vq1apT179ujixYsKDw9XdHS0Q9sahqH9+/dbGyAAAAAAAPAY2ZK0uHjxot544w2tXbvW7nXTNCXdTUgkfy2RYRgyTdNuHQAAAAAAcO+zPGlx+vRpdezYUdevX081ISGlnqhIlHwZAAAAAAD4d7A8adG/f39du3ZNklSwYEENGjRITZs21fjx47V3714ZhqHly5crMjJSFy5c0I4dO7Ry5UrdunVLhmHoueeeU7du3awOEwAAAAAAeBhLkxbLly/XoUOHZBiGihcvrtWrV6t8+fKSpMmTJ9vWa926tW36+eefV3h4uD788ENNnz5dM2bMUJEiRTR69GgrQwUAAAAAAB7G0tFDVq5caZseM2aMLWGREX9/f33xxRcaMWKETNPUxIkTtXz5cqvCBAAAAAAAHsjSpMWePXskSb6+vnryySed3n7EiBGqV6+eTNPUmDFjXBwdAAAAAADwZJYmLa5evSrDMFStWjXlzp3bblnSzjbv3LmT6vaGYahPnz6SpJMnTzLkKQAAAAAA/yKWJi1u3bolSQoICEixLF++fLbpmzdvpllGjRo1bNNBQUGuCw4AAAAAAHg0S5MW+fPnl5R6TYqCBQvapoODg9MsI1euXLbpK1euuDC67BMWFqYJEyaoXbt2qlixokqWLKmGDRvq9ddfp/YIAAAAAABpsDRpUaZMGZmmqatXr6ZYVq1aNdv0jh070izj6NGjtmkvL0vDtcSePXvUqlUrjR07Vvv27dONGzcUFRWl06dP6+eff1b79u312WefuTtMAAAAAAA8jqVZgFq1akmSzpw5o+joaLtlTZo0sU3PmjVL8fHxKbaPiYnRjz/+aJt3dPQRT3HmzBn17NlT58+fl2EYGjBggJYuXap169Zp3LhxKlGihOLj4/XJJ5/ohx9+cHe4AAAAAAB4FEuTFq1atZIkJSQk6M8//7Rb1qJFC5UtW1bS3b4qnnvuObtmIidPntTTTz9t68fC19dXbdq0sTJcl3vvvfd0/fp1SdIXX3yhL774Qg888ICaNGmiIUOGaN26dSpatKgk6T//+Y8uXrzoznABAAAAAPAoliYtHnroIdsoIcuWLUuxfPTo0TJNU5K0cuVK1a9fX1WrVlXVqlXVtGlTbdq0SdLdUURefPFFu34wPN2xY8e0atUqSVLLli313HPPpVinTJky+uCDDyRJt2/f1vfff5+dIQIAAAAA4NEsTVqULl1anTp1UoUKFbR7925FRUXZLe/Zs6cGDx5sS1yYpqlr167p2rVrMk3T9nqHDh00evRoK0N1uaVLl9qm+/fvn+Z6PXr0sI2kknQbAAAAAAD+7Szv2XLmzJnau3evtm7dKl9f3xTLx40bp2nTpqlOnToplpUrV06ffvqp5syZYzeKSE6wdetW2/T999+f5nq+vr62/j3OnDmjkJAQy2MDAAAAACAn8HZ3AJLUtWtXde3aVdeuXVNISIji4uJUvHhxW58XOdGxY8ckSf7+/ipdunS661avXl2bN2+WdLd/jzJlyji8n4SEhMwHCQAAsl1OvXbnxLhzYswScWennBizRNzZKSfGLP2/uHPiCJzJeUTSIlHRokVtHVPmZNHR0bpy5YokZZiwkGSXpEjaGakjzp49q7i4OOcCBAAAbnPy5El3h5ApOTHunBizRNzZKSfGLBF3dsqJMUt34/b29lbFihXdHUqWeVTS4l4RGRlpm86fP3+G6/v5+dmmIyIiMrXPEt99kKntrHLp5TEZruNpMUvEnZ1yYswScWennBizRNzZyZGYl3QolA2RuB5xZ5+cGLNE3NkpJ8YsEXd2y6lx5wQkLSyQtMPR3LlzZ7h+njx5bNN37txxal/ly5eXJN1yaivrVa5cOcN1PC1mibizU06MWSLu7JQTY5aIOzs5EjMAAMjZLE9afP/997p586YMw9Crr75qGynDEb///rv2798vSXriiSdUs2ZNi6J0raQdjsbGxma4fkxMjG3ax8fHqX15ahslT40rI8SdfXJizBJxZ6ecGLNE3NkpJ8YMAACcY2nS4vDhwxo5cqQMw9CDDz7oVMJCkgoVKqRx48bJMAydOHFCU6ZMsShS10ra3OPWrYyfSyVdp0CBApbEBAAAAABATmPpI4oVK1bYpvv37+/09s2aNVPNmjVlmqZWr17tUK0FT5A3b14VK1ZMknT+/PkM10/a+aYzI4cAAAAAAHAvszRpsW3bNklSrly51LFjx0yV8eijj0q6Wxth3759LovNajVq1JAkhYeHZ5i4CAoKSrEdAAAAAAD/dpYmLY4fPy7DMFS1alWn+2pIVK9ePbvycopWrVrZprds2ZLmelFRUdq9e7eku51qUtMCAAAAAIC7LE1ahIaGSpKKFi2a6TISm1kkLS8n6NKli216xowZaa63YMEC3b59O8U2AAAAAAD821matMiVK5ck+9ExnJV024SEhCzHlF1q1qypRx55RNLdZjLTp09PsU5ISIjGjh0r6e6II0OGDMnOEAEAAAAA8GiWjh5SpEgRBQcH6+zZs5ku48yZM7bpwoULuyCq7PPJJ59ox44dunHjht588039/fff6tKli/z8/LRnzx5NnDhRV69elSS9//77KlWqlJsjBgAAAADAc1iatKhataqCg4N1+fJl7d+/Xw0aNHC6jN9++802XalSJRdGZ71KlSpp/vz56t+/vy5evKhp06Zp2rRpdut4eXlp+PDhevnll90UJQAAAAAAnsnS5iHt2rWzTY8ZM8bp7bdv3641a9ZIkvz8/NSiRQuXxZZdmjZtqm3btmn06NFq0KCBChYsKB8fH5UvX17PPPOM1q9fr5EjR7o7TAAAAAAAPI6lSYsePXooX758kqRNmzbprbfeUnx8vEPbHjx4UP369ZMkGYahXr16ydvb0oohlgkICNDbb7+tTZs26ezZs7p06ZIOHDigb775Rg0bNnR3eAAAAAAAeCRLkxbFixfXSy+9JNM0JUnTp09X27ZttXTpUkVHR6e6zcmTJzV69Gg99NBDun79uiTJ399f77zzjpWhAgAAAAAAD2N51YX33ntP+/fv1/r162UYhg4fPqwBAwbI29tb1atXV7FixZQnTx7dvHlTp0+f1pUrVyTJlujIkyePfv75Z7uhTwEAAAAAwL3P8qSFl5eXZs6cqbfeektz586VdLe5R2xsrA4fPmy3bmKiwjAMSXdrakyePFn333+/1WECAAAAAAAPY2nzkES+vr6aNGmS5s2bpxYtWtiSE6Zp2v1LVLBgQb355pvaunUrCQsAAAAAAP6lsrVny44dO6pjx466ePGitm3bpqCgIN24cUMxMTEqWLCgAgMD1bRpUzVq1CjHdroJAAAAAABcwy2ZgZIlS+qpp55yx64BAAAAAEAOkS3NQwAAAAAAAJxF0gIAAAAAAHgkkhYAAAAAAMAjkbQAAAAAAAAeiaQFAAAAAADwSCQtAAAAAACARyJpAQAAAAAAPBJJCwAAAAAA4JFIWgAAAAAAAI9E0gIAAAAAAHgkkhYAAAAAAMAjkbQAAAAAAAAeiaQFAAAAAADwSN7u2OmJEyd04MABXbt2TeHh4UpISNCIESPcEQoAAAAAAPBQ2Za0iIyM1OTJk/Xjjz/q4sWLKZanlrR4/vnnFRISIsMw9NNPP6lUqVLZESoAAAAAAPAA2dI8ZN++fWrTpo0+/vhjXbx4UaZp2v1LS9OmTbVr1y7t2rVL8+bNy45QAQAAAACAh7A8aXHw4EF16dJFZ8+etSUpKlasqE6dOqlkyZLpbtu7d295e9+tDLJs2TKrQwUAAAAAAB7E0qRFfHy8Bg4cqIiICElSrVq1tHbtWu3du1ezZs1SrVq10t0+ICBAbdq0kWma+vvvvxUWFmZluAAAAAAAwINYmrRYsGCBjh8/LsMwVKdOHa1Zs0ZNmjRxqoymTZtKkkzT1OHDh60IEwAAAAAAeCBLkxYrV660TU+cOFH58+d3uoyaNWvapk+ePOmSuAAAAAAAgOezNGmxf/9+SVLp0qWdrmGRqFChQrbpmzdvuiIsAAAAAACQA1iatLh27ZoMw1CFChUyXUaePHls03FxcS6ICgAAAAAA5ASWJi1y584tKWvJhuvXr9umAwICshoSAAAAAADIISxNWgQGBso0TZ0+fTrTZSQ2MZGU4RCpAAAAAADg3mFp0qJhw4aSpCtXrmjv3r1Ob5+QkKDFixdLkry8vNSiRQuXxgcAAAAAADyXpUmLRx55xDY9ZswYp7f//vvvdfr0aRmGoebNm9M8BAAAAACAfxFLkxZPPvmkKleuLEnavHmzXn31VcXGxjq07c8//6wPP/zQNj9s2DBLYgQAAAAAAJ7J28rCc+XKpQkTJqh79+6Kj4/XL7/8oq1bt2rgwIF64IEHFBMTY1v32rVrunLlinbt2qXZs2dr9+7dMk1ThmHoqaee0oMPPmhlqAAAAAAAwMNYmrSQpAceeEBff/21XnvtNcXHx+vs2bN6//337dYxTVPVqlVL8ZokNW/eXN98843VYQIAAAAAAA9jafOQRE8//bRWrlypChUqyDRN2z9JMgxDhmHYvW6apry8vDRw4EAtW7ZMPj4+2REmAAAAAADwIJbXtEjUrFkz7d69W0uXLtW8efO0Y8cOhYWFpVivQoUKeuihhzRkyBBbfxgAAAAAAODfJ9uSFtLdYUu7du2qrl27SpLOnz+v0NBQ3b59W/7+/goMDFSRIkWyMyQAAAAAAOChsjVpkVzp0qVVunRpd4YAAAAAAAA8VLb0aQEAAAAAAOAskhYAAAAAAMAjkbQAAAAAAAAeKct9WnTu3NkVcWTIMAwtW7YsW/YFAAAAAADcL8tJiz///FOGYbgiljSZpmn5PgAAAAAAgGdxyeghpmk6vG5i8iG9bRxZBwAAAAAA3NuynLT49ttvM1wnMjJSn332mUJDQ2WapgoVKqQHHnhANWvWVOHChZUnTx5FRkbqzJkz2rt3r/bs2SPpbvKia9euat++fVbDBAAAAAAAOUyWkxZ9+vRJd/m5c+fUtWtXhYaGqmjRovroo4/Us2dPeXunveszZ85ozJgx+vXXX7VkyRLVqFFDw4cPz2qoAAAAAAAgB7F09JCYmBj1799fp06dUrly5bR+/Xr16dMn3YSFJFWoUEHTpk3TqFGjlJCQoE8//VS//vqrlaECAAAAAAAPY2nSYtGiRTpw4IAMw9D//vc/lS1b1qnt3377bd13330yTVMffvihRVECAAAAAABPZGnSYt68eZKkUqVK6YEHHshUGb1795YkhYSEaPv27S6LDQAAAAAAeDZLkxYnTpyQYRiqWLFipstIuu0///zjirAAAAAAAEAOYGnS4tq1a5KkiIiITJeRdNvr169nOSYAAAAAAJAzWJq0KFKkiEzT1JEjR3Tjxo1MlbFlyxbbdOHChV0VGgAAAAAA8HCWJi3q1KkjSYqLi9OYMWOc3v706dOaMWOGbb5evXouiw0AAAAAAHg2S5MWTz/9tG16xowZGjFihKKjox3adt++fercubNu3bolwzBUrVo1NWjQwKJIAQAAAACAp7E0adG1a1e1bt1apmlKkqZMmaLGjRtr/Pjx2rNnj27fvm23/rlz57RkyRL1799fDz30kC5cuGBb9tlnn1kZKgAAAAAA8DDeVu9g+vTpevLJJ3Xo0CEZhqHz589r/PjxGj9+vCQpT548ypMnj27dumVLbkiyTRuGof/+97+6//77rQ4VAAAAAAB4EEtrWkh3O+NcuXKlnnnmGdtrpmna/kVHRysiIkIJCQm2REXi/6VLl9bcuXP1wgsvWB0mAAAAAADwMJYnLSTJ399fX3/9tTZt2qTnnntOxYoVS3PdXLlyqWnTppo4caJ27typjh07ZkeIAAAAAADAw1jePCSpevXq6YsvvtAXX3yh4OBgnTp1SmFhYYqJiVGBAgUUGBioWrVqycfHJzvDAgAAAAAAHihbkxZJlS1bVmXLlnXX7gEAAAAAgIfLluYhAAAAAAAAziJpAQAAAAAAPBJJCwAAAAAA4JFIWgAAAAAAAI9E0gIAAAAAAHgkkhYAAAAAAMAjuW3I03tZfHy8goKCtG/fPu3fv1/79+/XoUOHFBUVJUkaMWKERo4c6eYoAQAAAADwbCQtLDB58mSSEgAAAAAAZBHNQyxgmqZt2tvbW7Vq1VLjxo3dGBEAAAAAADkPNS0sUK9ePX322Wdq0KCB6tatK19fX82ePVt79uxxd2gAAAAAAOQYJC0s0Lp1a7Vu3drdYQAAAAAAkKPRPAQAAAAAAHgkkhYAAAAAAMAjkbQAAAAAAAAeKdv7tLhy5Yr27NmjixcvKjw8XNHR0Q5vO2LECAsjy5kSEhLcHUKqPDWujBB39smJMUvEnZ1yYswScWennBgzAADZycsr59dTyLakxdKlS/X1119r7969mS6DpEVKZ8+eVVxcnEq4O5BkTp48meE6nhazRNzZKSfGLBF3dsqJMUvEnZ0ciRkAgH8rb29vVaxY0d1hZJnlSYv4+HgNGTJEixYtkiSZppnu+oZhpLpe4uuu8scffygqKirL5dSrV0+lSpVyQUQAAAAAACApy5MW7733nhYuXGibL1OmjBo3bqydO3fq4sWLMgxDTz/9tCIjI3XhwgUdOnRI0dHRtiRFhw4dVKRIEZfH9eqrryo4ODjL5fzwww/q1auXCyLKnPLly0uSbrktgtRVrlw5w3U8LWaJuLNTToxZIu7slBNjlog7OzkSMwAAyNksTVr8888/mjp1qqS7bWnGjh2rl19+WZLUvXt3Xbx4UZL03Xff2baJiorSggUL9Omnn+rSpUs6fPiwZsyYocaNG1sZao7lqW2UPDWujBB39smJMUvEnZ1yYswScWennBgzAABwjqVJi1mzZikhIUGGYWjw4MG2hEV6fH191b9/fz3xxBPq1q2b9u7dq549e2rLli0ubYaxbt06xcXFZbmcQoUKuSAaAAAAAACQnKVJi7/++kvS3f4oXnvtNae2DQgI0Jw5c9S0aVPduHFDw4YN05w5c1wWW/HixV1WFgAAAAAAcD1L61WeO3dOhmGoYsWKKlmyZJrrxcbGpvp6YGCg+vfvL9M0tW7dOl29etWqUAEAAAAAgIexNGlx48YNSUo1YZEnTx7b9O3bt9Ms47777pN0dxSSbdu2uThCAAAAAADgqSxNWuTOnVtS6sOVFihQwDad2CFnapL2GXHp0iUXRgcAAAAAADyZpX1aFC1aVMHBwbp582aKZWXLlrVNHzp0SDVq1Ei1jMuXL9um06uR4Wlmz55tN799+3bb9MGDB1Msf+ihh+hnAwAAAACAJCxNWlSrVk3nzp3TqVOnUiyrW7eubXrlypXq3r17qmUsW7bMNl20aFHXB2mRV155Jc1lq1at0qpVq+xeW758OUkLAAAAAACSsLR5SJMmTSRJt27d0rFjx+yWPfjgg/L19ZV09wf70qVLU2w/a9YsLV682DbfokULC6MFAAAAAACexNKaFu3atdO4ceMkSatXr7ZrAlKgQAH17dtXU6dOVUJCggYMGKBWrVqpUaNGku42p9i1a5eku31itGrVSlWqVLEyXJcKCwtzdwgAAAAAAORoliYtmjVrplKlSunChQuaMWOG3njjDbvl77//vjZt2qQTJ07IMAxt3bpVW7duTVFOQECAvvrqKytDBQAAAAAAHsbSpIUkLVmyRNevX5ckRUdHK2/evLZl/v7+WrlypV555RWtW7cu1e3r1aunKVOmqGLFilaHCgAAAAAAPIjlSYuqVauqatWqaS4PDAzUggULdOjQIW3cuFHBwcGKi4tT8eLF1bp1a7Vq1crqEAEAAAAAgAeyPGnhqDp16qhOnTruDgMAAAAAAHgIS0cPAQAAAAAAyCySFgAAAAAAwCORtAAAAAAAAB6JpAUAAAAAAPBIWe6Is3DhwrZpwzBsw5smX5ZVycsGAAAAAAD3tiwnLUzTlGEYMk3TqWUAAAAAAADpcUnzkPSSEiQsAAAAAABAZmS5psWBAwcytQwAAAAAACA9WU5alCtXLlPLAAAAAAAA0sPoIQAAAAAAwCORtAAAAAAAAB6JpAUAAAAAAPBIWe7TIiPff/+9bt68KcMw9OqrrypfvnwOb/v7779r//79kqQnnnhCNWvWtChKAAAAAADgaSxNWhw+fFgjR46UYRh68MEHnUpYSFKhQoU0btw4GYahEydOaMqUKRZFCgAAAAAAPI2lzUNWrFhhm+7fv7/T2zdr1kw1a9aUaZpavXq1YmNjXRkeAAAAAADwYJYmLbZt2yZJypUrlzp27JipMh599FFJ0q1bt7Rv3z6XxQYAAAAAADybpUmL48ePyzAMVa1aVT4+Ppkqo169enblAQAAAACAfwdLkxahoaGSpKJFi2a6jGLFiqUoDwAAAAAA3Pss7YgzV65ckqSYmJhMl5F024SEhCzHBACAuxX4Yo67Q8iUnBo3AADIuSytaVGkSBGZpqmzZ89muowzZ87YpgsXLuyCqAAAAAAAQE5gaU2LqlWrKjg4WJcvX9b+/fvVoEEDp8v47bffbNOVKlVyYXQAgJyOJ/8AAAD3NkuTFu3atdOGDRskSWPGjNHixYud2n779u1as2aNJMnPz08tWrRweYwAgLtIAAAAAMDTWNo8pEePHsqXL58kadOmTXrrrbcUHx/v0LYHDx5Uv379JEmGYahXr17y9rY0xwIAAAAAADyIpUmL4sWL66WXXpJpmpKk6dOnq23btlq6dKmio6NT3ebkyZMaPXq0HnroIV2/fl2S5O/vr3feecfKUAEAAAAAgIexvOrCe++9p/3792v9+vUyDEOHDx/WgAED5O3trerVq6tYsWLKkyePbt68qdOnT+vKlSuSZEt05MmTRz///LPd0KcAAAAAAODeZ3nSwsvLSzNnztRbb72luXPnSrrb3CM2NlaHDx+2WzcxUWEYhqS7NTUmT56s+++/3+owAQAAAACAh7G0eUgiX19fTZo0SfPmzVOLFi1syQnTNO3+JSpYsKDefPNNbd26lYQFAAAAAAD/Utnas2XHjh3VsWNHXbx4Udu2bVNQUJBu3LihmJgYFSxYUIGBgWratKkaNWpEp5sAAAAAAPzLuSUzULJkST311FPu2DUAAAAAAMghsqV5CAAAAAAAgLNogwEALlbgiznuDgEAAAC4J1DTAgAAAAAAeKRsrWlx584d/f333woKClJYWJiioqLsRg3JyIgRIyyMDgAAAAAAeJJsSVpcunRJH3/8sZYsWaLbt29nuhySFgAAAAAA/HtYnrTYvn27evXqpYiICKdqVSRnGIYLowIAAAAAAJ7O0qRFaGionn76aYWHh9uSDr6+vqpbt65KlSolX19fK3cPAAAAAAByMEuTFt99951u3rwpwzDk7e2t999/X88//7zy589v5W4BAAAAAMA9wNKkxZo1a2zTEyZMUL9+/azcHQAAAAAAuIdYOuTp2bNnJUlFixYlYQEAAAAAAJxiadIiLi5OhmGoWrVqVu4GAAAAAADcgyxtHlKqVCmdPHlScXFxVu4GwD2qwBdz3B0CAAAAADeytKZFixYtZJqmjh8/nqXhTgEAAAAAwL+PpUmLAQMGSJLCwsK0fPlyK3cFAAAAAADuMZY2D2nUqJFefvllfffdd3r33XfVqFEjlSlTxspdApbLqU0WcmrcAAAAAP69LK1pIUn/+c9/1K9fP128eFHt2rXTvHnz6OMCAAAAAABkyNKaFq+88optunDhwrp27ZpeeuklDR8+XA0bNlTx4sWVO3duh8oyDEPffPONVaECAAAAAAAPY2nS4pdffpFhGLZ5wzBkmqYiIiK0ZcsWp8sjaQEAAAAAwL+HpUkLSWmOGuLsaCJJkx8AAAAAAODeZ2nSYsSIEVYWDwAAAAAA7mGWJi3effddK4sHAAAAAAD3MMtHDwEAAAAAAMgMkhYAAAAAAMAjkbQAAAAAAAAeiaQFAAAAAADwSJYPeZqa8PBwXbp0SeHh4YqOjnZ4u1atWlkYFQAAAAAA8CTZlrQICQnR1KlTtWrVKp08eVKmaTq1/f/H3l2HRZW+/wN/HxCwUMLWNdbEwE7sWttVBARdxe5ev3bs6q66tq4diIIY2KKCimCuARYmiiIYIBKCgNT8/uA358NIDcjMmZH367r22mFO3eDEOfd5nvsWBAGfPn1SUXREREREREREpGnUkrRwdHTE3LlzERcXBwA5TlgQERERERERUf6j8qSFs7Mzpk6dqvCcqakpvnz5gvj4eAiCgAoVKiAmJgZRUVFISUkBkDqyAgDKlSsHHR2W3iAiIiIiIiLKb1SaDYiIiMDs2bMBpCYhevXqhbt37+LFixcK9SkePHiAgIAABAUF4fjx47CyshKTFmZmZrh8+TIePHigylCJiIiIiIiISMOoNGmxb98+xMTEQBAEdOzYEfv27UPlypUzXb9w4cJo164dtm/fDjc3N5iYmODixYvo1atXjgp2EhEREREREZH2U2nSwtvbW3y8aNGiHG3bokULHDx4EDo6Onj8+DEWL16cx9ERERERERERkSZTadLi6dOnAIDSpUujXr16ma4nr2PxrcaNG6N///6QyWRwdnZGfHy8SuIkIiIiIiIiIs2j8poWgiBkOCVEV1dXfCzvKpKRbt26AQBiYmJw9erVPI+RiIiIiIiIiDSTSpMW8hEUBQsWTLesWLFi4uPQ0NBM91G2bFnxcXBwcB5GR0RERERERESaTKUtT42MjBAaGoro6Oh0y0qUKCE+9vf3R5UqVTLcR9ptIyIi8j5IFYiJiYGnpyeuXLmC+/fv4+XLl4iKikLhwoVRvnx5NGvWDIMGDUKzZs2kDpWIiIiIiIhIY6k0afHzzz8jJCQEb968SbesTp064uNLly6ha9euGe7j8uXL4uO0ozM01YULFzB48OAM6298/vwZnz9/xpMnT+Do6Ahra2usX78ehQoVkiBSIiIiIiIiIs2m0ukhDRo0AACEhYXh/fv3Csvat28PQRAAAM7OzggICEi3vZ+fHxwcHMSfzc3NVRdsHvn06ZOYsGjWrBnmzp0LV1dXeHl5wc3NDfPmzRNHmRw6dAhDhw6VMlwiIiIiIiIijaXSkRbt2rXDli1bAADnz5/HkCFDxGUVKlRAp06dcOHCBURHR6NTp04YN24cGjZsCAD477//sGPHDsTGxkIQBFSpUgVNmjRRZbh5Qk9PD3Z2dpg2bRqqV6+ebrmFhQV+++03dOvWDa9fv4aHhweOHTuGfv36SRAtERERERERkeYSIiMjZara+devX1G9enVER0ejRYsWOHv2rMLyFy9eoH379oiNjYVMJhNHXsjJZKmh6erq4uDBg+jUqZOqQlU7Nzc3DBo0CADQq1cvODk5fdf+oqfZ5kVYecZwrUu262hazIBycRMREREREZF6qHSkhYGBAdauXYtXr15BEATExcUp1G+oVq0aDh8+jKFDh+Ljx49ikiKtQoUKYcOGDT9UwgJIHYUi9/LlSwkjISIiIiIiItJMKk1aAIClpWWWy1u2bAlfX1/s27cPXl5eCAoKQlJSEkqXLo3WrVvD3t4epUuXVnWYapeYmCg+1tFRaWkRIiIiIiIiIq2k8qSFMooWLYpx48Zh3LhxUoeiNmm7opiZmeV6PykpKXkRTp7T1Liyo61xExERERERfetHuEGuEUmL/CY5ORlr1qwRf7axscn1vgIDA5GUlIQyeRFYHlJmyoumxQxwqg4REREREf0YChQogCpVqkgdxndTadIiKCgIQGpdCnmbz5z69OkTYmNjAQA//fRTnsUmpRUrVuD+/fsAgLZt26JLly4SR0RERERERESkeVSatDA3N4cgCOjcuTMOHTqUq31MnjwZZ8+ehSAI+PTpU57F5u3tjbi4uO/ej7m5OcqVK6f0+seOHcPKlSsBAKampti6det3Hb9SpUoAgC/ftZe8V7Vq1WzX0bSYAeXiJiIiIqKMZVRYn4hy59vumvmVWqaHfO+Hlyo+/CZOnCiOBPke27ZtU3p6h6enJ8aOHQuZTIZChQph7969OUp4ZERT5yhpalzZ0da4iYiIiKSQnJyM2NhYxMXFISEhgUkLojwkCAL09fVRqFAhFC5cGLq6ulKHJAnWtFATb29vDBo0CF+/foWBgQGcnZ1hYWEhdVhERERERLmSmJiIjx8/Ql9fH0WLFoWBgQFvABHloZSUFHz9+hWxsbGIjo5GyZIloaenJ3VYaqfxSYukpCQAqUVE8tKFCxfEfX8PY2PjbNe5evUqbG1tERcXB319fezbtw8dO3b87mMTEREREUkhOTkZHz9+hKGhIQwNDaUOh+iHpKuri8KFC6Nw4cKIjo7Gx48fUbp06Xw34kLjkxZv374FgDz/MCxdunSe7i8zV69ehY2NDWJjY6Gvrw9HR0d07dpVLccmIiIiIlIF+bktExZE6mFoaCiOushv7zuNHr915coVPHr0CIIgoFq1alKHk2PXrl2DjY0Nvnz5Aj09PezZswfdu3eXOiwiIiIiou8SFxeHwoULSx0GUb5SuHDhPGkmoW3ybKTFhAkTMl32+PHjLJenJZPJEB8fj4CAADx8+FB8vm3btt8dozpdv34d1tbWYsLCwcEBPXr0kDosIiIiIqLvlpCQAAMDA6nDIMpXDAwMEB4eLnUYapdnSYv9+/dn2JJFJpPh/fv3cHFxyfW+ixcvjhEjRnxPeGp148YNhYTF7t270atXL6nDIiIiIiL6bjKZDDKZjEU3idRMR0dHfP/lp3aoeVrTIrMWR9/T+qhevXrYsGEDypQpk+t9qJOPjw+sra0RExMDAJg5cyaqVq2Kx48fZ7ld7dq11REeERERERERkdbIs6TFrFmz0j23YsUKCIKAKlWqwMrKSqn96OjooHDhwihdujTq16+PGjVq5FWIauHh4YHo6Gjx57///ht///13tttFRkaqMCoiIiIiIiIi7ZNnSYvZs2ene27FihUAgJ9//jnD5UREREREREREmVFpy9MKFSpAEASULFlSlYfRKHPmzMGcOXOkDoOIiIiIiIhI66k0aXHt2jXxcXJyMnR1dVV5OCIiIiIiIiL6gag0aVG5cmUAgK6uLu7du4fy5cur8nBERERERERE9ANRadJCEASkpKSgevXqTFgQEREREZEoKSkJbm5uOHHiBO7fv4/Q0FDExsaiUKFCKFWqFCpXrgxzc3O0bNkSFhYWKFq0qML248aNg4uLS46OWbduXVy9elXhuStXrqB3797iz9WrV8d///2X7ShxIyMjAECLFi1w7tw5AMCyZcvEun7fY9asWeKU8549e4oj2O/fv49KlSoprBsYGIj69esrPOfk5IRevXple5zHjx+jVatWCs8dP34c7du3T7eu/Pf96aef8PDhw3TLnZ2dMWHChGyP+a3Xr1+L+/72WACgp6eH27dvizfEM5P27xQWFoYCBQqk+7fNLQsLC7i5uWW4LCoqCkeOHMH58+fx5MkThIWFITk5GSVKlEC1atXQsWNHWFlZaU03TE2k0ubKJUuWzHc1LYiIiIiIKGtPnz5Fu3btMHToUBw9ehQvX75EdHQ0kpOTERMTg4CAAHh6emLdunWwsbHBggUL1Babv78/nJyc1HY8VXB2dlZqvX379qk4ku+XmJiIv/76S+owMrRlyxY0aNAA06dPx9mzZ/H69WvExMQgLi4OQUFBuHTpEhYsWIBGjRph6dKlSEpKkjpkraTyQpwhISH49OmTKg9DRERERERaIjg4GD179hSvERo0aICBAweibt26KFasGGJjYxEUFAQfHx+cP38eL1++zHaf//77Lxo1apTteoUKFVIqxhUrVsDa2lrp9eVGjhyJvn37ZrpcPqqhbNmyOHLkSKbr5famb6FChRAXF4fz588jNDQUpUqVynTdxMREHD58WGG7vDJixAiMGDFCqXWLFSuW7Tqurq6YNGkSzM3NcxRHo0aNcP369UyXT5gwAXfv3gUAHD16NNPREIULF1b4OTExERMnTsTBgwcBADo6Oujduzd69OiBypUro0CBAggODsaFCxdw+PBhxMbGYtWqVfD19YWjoyMMDQ1z9HvkdypNWnTv3h0+Pj54+vQpwsPDYWJiosrDERERERGRhvvzzz/FhMXEiROxZMkSCIKgsE6LFi1gZWWF5cuXw8fHB+/evctyn5UqVULt2rW/O7YSJUogLCwM7969w7Zt2zB16tQcbV+yZEmlEg4FChTIk3i/1aFDB3h7e+PLly84ePAgJk2alOm6Z8+eRVhYGACgV69eYgIjL5QoUSJPfj9TU1OEh4dDJpPhzz//hKura462L1KkSJZxpE1GVK1aNd3Um8zMnz9fTFiUL18eTk5OaNiwocI6jRs3Rt++fTF9+nQMGTIEDx8+hKenJ8aNG6f1I3nUTaXTQ2xtbWFoaIiUlBSsXLlSlYciIiIiIiINl5ycLNYGKFGiBBYvXpwuYfGtxo0b50ldAmUMHDgQFSpUAACsXbsWkZGRajluXilatKg40iO7KSLy5S1btsTPP/+s8thyo3r16uLvc+HCBVy5ckXiiIBLly5h27ZtAIDixYvDzc0tXcIirSpVquDkyZNiTY7Tp09j79696gj1h6HSpEW5cuWwcuVKyGQybNu2DWvXrlXl4YiIiIiISIOFhYXhy5cvACAOo9ckBQsWxOzZswGkFljUxuuXQYMGAUitG3Lnzp0M1/nw4QMuXLigsL6mmj9/vvg6+eOPPySOBvjnn3/Ex0uWLMm2QCgAGBsbY/369eLPa9asQXJysirC+yGpNGkBADY2NnB0dETRokWxZMkStGnTBtu3b8eDBw8QHR2t6sMTEREREZGGMDAwEB+/fPkSiYmJEkaTMVtbW9SqVQsAsH379mynpmgaCwsLceREZqMtDhw4gOTkZBQtWhT9+vVTZ3g5Vq1aNfz2228AgDt37uDkyZOSxeLv748bN24ASJ0KZGtrq/S27dq1Q7169QCkdkzx9vZWSYw/IpUmLUxMTGBiYgJ7e3vExMRAJpPBz88Ps2fPRvv27VGpUiVxnez+MzU1VWWoRERERESkYkZGRuKd6YiICEybNi1PC0DmBV1dXbFbSVxcHJYvXy5xRDlnZ2cHADhy5EiGf195MqNv374oUqSIWmPLjVmzZon1J5YuXSrZKIW001N++eUX6Onp5Wj7tNOcvm29S5lTadJCJpMp/F8QBHHOmkwmy/F/RERERESk3SZMmCA+dnJygpmZGUaPHo1du3bhzp07iI2NzfE+AwMD8fjx42z/U7ZGRc+ePdGsWTMAqRf4z58/z3FMUrK1tYWOjg4+f/6MU6dOKSy7efMm/P39AahuakhYWJhS/x7KjmIpU6YMxo4dCwB4/vy50i1d89r9+/fFx8p0q/lW2toXafdFWVP5JLJvExdERERERJR/jRo1Cq9fv8amTZsAAJGRkTh06BAOHToEIHWkQ7169dC5c2cMGjQIVapUyXafEydOVOrYmzZtUvpCffHixejRoweSk5Px559/alXHh/Lly6NDhw64ePEinJ2dYW1tLS6T/x5Vq1YVW7DmtV27dmHXrl3Zrmdra4stW7Yotc8pU6bAwcEBERERWL58OaysrHLckvZ7ybveAMiynWxm0m6Tdl+UNZUmLZg9IiIiIiKib/3111/o168fNm/ejHPnzimMrkhOTsa9e/dw7949rF27FiNGjMCSJUsU6mGoQ6tWrfDLL7/A3d0dp0+fxp07d9CkSRO1xvA9Bg0ahIsXL+Ly5ct48+YNKlasiC9fvuD48eMA/jeFRFsUL14c06dPx4IFC3LdkvZ7xcTEiI8NDQ1zvH3abT5//pwnMeUHKk1aVKxYUZW7JyIiIiIiLdWkSRPs3r0bX79+xd27d+Hj4wM/Pz/cuXNHnL6QnJyM7du3IzAwEAcOHMi0PeqpU6fQpk2bPI9x4cKFOH/+PFJSUrBo0SKxXas26NmzJ4yMjBAZGYn9+/dj9uzZOHHiBKKjo6Gjo5OjIpI5NWvWLMyZMyfP9ztq1Chs27YNwcHBWLt2Lezt7WFkZJTnx8lM0aJFxcdpExjKSrtNsWLF8iSm/EDl3UOIiIiIiIgyY2BggBYtWmDChAnYsmULbt++jVu3bil0tXB3dxdHCKhTnTp1xKkV165dw/nz59UeQ24ZGBjAysoKAODi4gKZTCZODenYsSPKlSsnZXi5InVL2rTNIUJCQnK8fdptTExM8iSm/IBJCyIiIiIi0ig1atSAg4MDevbsKT7n6uoqSSxz584Vp6b88ccfWlWrT16/IzAwEHv37hXbdQ4ePFjKsL6LlC1pzc3Nxcd3797N8fb37t0TH9evXz8vQsoXmLQgIiIiIiKNNGrUKPHxy5cvJYmhYsWKGD58OADAz88Phw8fliSO3GjQoAHq1KkDIHXKhkwmg7GxMbp37y5xZLknZUva1q1bi4/d3d2RmJiYo+3TdnJJuy/KmtqTFlFRUTh+/Djmz58POzs79OjRA507d4alpSVGjx6NTZs2KWSgiIiIiIgofypbtqz4WEdHuvutv//+u1hE8a+//kJCQoJkseSUfFRFfHw8AMDKykrtRU3zmlQtaWvWrInmzZsDAEJDQ+Hi4qL0tpcvX8aDBw8ApCbC2rVrp5IYf0Qqb3kqFxERgSVLluDQoUNZ9l6WD/uqU6cOZs6ciT59+qgrRCIiIiIi0iC+vr7i48qVK0sWh6mpKSZNmoS///4bgYGBcHBwkCyWnLK2tsbff/8tJlqUbfmq6dK2pF2yZInajjtz5kwMGDAAQGqh1nbt2qFSpUpZbhMZGanQ6WT69OkoUEBtl+JaTy3pymvXrqFly5bYs2cPvnz5AgCQyWRZ/ufn5wd7e3uMHj0aSUlJ6giTiIiIiIhUKD4+Hu3atcPx48ezHVr/8uVLhYvRvn37qjq8LE2YMAGlSpUCAKxcuVLSWHLC1NQUQUFBCAkJQUhIyA9TS0HekhZInXbx7NkztRy3c+fO4rSlyMhI9OzZM8uZAq9fv0bfvn0REBAAAOjevTvs7e3VEOmPQ+XpHV9fXwwcOBAxMTFiiyJBEFC/fn3Url0bpqam0NfXR3R0NF6/fo27d+/i48ePAFITG66urkhMTNSqbCYREREREWXs/v37sLe3h7GxMX755Rc0bdoU1apVg5GREZKTk/HmzRt4eXnh4MGDiIuLA5A6/1/eCSMjgYGBCp0dslKrVq1cTTUpUqQIZs6ciZkzZyIsLCzH2+cnYWFhePz4sVLr/vTTT+LUm5xK25JWnf8mf/31FyIiIuDq6org4GB07NgRffv2RY8ePVCpUiUUKFAAwcHBuHDhAg4fPiy+jtu2bYutW7eqLc4fhUqTFsnJyRg/fryYsNDT08O4ceMwZswYhflpaaWkpMDT0xN//vknHj58CJlMhhMnTuDw4cNZflAREREREZFm09XVRdmyZfH+/XtERETgwIEDOHDgQJbbWFlZYd26dVkmGiZOnKh0DK9fv4aRkZHS66dlb2+PzZs349WrV7naPr/YtWsXdu3apdS6Tk5O6NWrV66OU6dOHVhZWeHgwYO52j639PX1sWPHDjRo0ACrVq1CZGQkjh07hmPHjmW4fqFChTB27FjMnTsXenp6ao31R6DSpMWRI0fw7NkzCIIAQ0NDHDp0SCxckhkdHR107twZHTp0wLhx43D48GHIZDIsW7aMSQsiIiIiIi2mp6eHx48f4969e7h8+TJu376NFy9e4N27d/jy5Qv09fVRvHhxVK1aFc2aNYOlpSXq1q0rddgiPT09zJs3DyNHjpQ6FPr/5s6di2PHjqm9OKogCJg4cSIGDx4MV1dXXLhwAY8fP8anT5+QnJyMEiVKoGrVqujUqROsrKwyvWlP2RMiIyNV1mjY3t4eJ06cgCAI2LRpE2xtbXO0fUJCAlq3bg1/f38IgoBr167BzMxMRdFqt+hpOfvbqprh2uwr6WpazIBycRMRERHlZzKZDMHBwahQoYI4/ZuIVC+/vvdUWohTXpDE2NgY1tbWOd5eX19foUjJ/fv38ygyIiIiIiIiItJ0Kk1afPz4EYIgoEaNGtDV1c3VPurUqaOwPyIiIiIiIiLKH1SatJAXy0lJScn1Pr5nWyIiIiIiIiLSXipNWpQsWRIymQxPnz7Ntg9zZh48eCA+lvdFJiIiIiIiIqIfn0qTFo0aNQIAREdHw8nJKcfbx8XFYc+ePen2R0REREREREQ/PpUmLdL2212wYAG8vb2V3jYxMRFjxozB69evIQgCqlevjurVq6siTCIiIiIiIiLSQAVUufO+ffuiTp06ePz4Mb58+YIBAwZg+PDhGDt2LKpUqZLhNomJiThz5gyWLVuG58+fi8/PnTtXlaGSBNhelIiIiIiIiLKi0qSFIAjYsmULevXqhejoaCQlJWHHjh3YsWMHqlatitq1a8PExAR6enqIiYnB69ev4efnh5iYGIX92NjYoG/fvqoMlYiIiIiIiIg0jEqTFgBQr149uLq6YtiwYXj79i0AQCaT4eXLl3j58mW69WUyGQRBgEwmAwAMHToUq1evVnWYRERERERERKRhVFrTQq5p06a4fv06JkyYgOLFiwNITU5k9J98WfPmzXH48GGsW7cOurq66giTiIiIiIiIiDSIykdayBUrVgxLly7F/PnzcfXqVdy5cwcBAQGIiopCQkICihUrhhIlSqB+/fpo1aoVqlWrpq7QiIiIiIiIiEgDqS1pIVewYEF07twZnTt3VvehiYiIiIiIiEiLqGV6CBERERERERFRTjFpQUREREREREQaSe3TQwAgNjYW9+/fx8uXL8WaFoaGhihZsiTq16+PypUrSxEWEREREREREWkQtSYtzp49CwcHB1y6dAnJycmZrlemTBnY2dlh5MiRKFOmjBojJCIiIiIiIiJNoZbpIe/evYOlpSUGDRqECxcuICkpKct2p+/fv8eaNWvQrFkz7NmzRx0hEhEREREREZGGUflIi9evX6NPnz4IDg4WExNyRkZGMDExgYGBAaKjo/HhwwckJSWJy6OjozF9+nQEBQVhwYIFqg6ViIiIiIiIiDSISpMWycnJGDp0KIKCgiAIAgDA3NwcI0eORJcuXdJN/UhISMD9+/dx+PBh7Nu3D1+/foVMJsPatWthbm6Ovn37qjJcIiIiIiIiItIgKp0e4uzsjAcPHkAQBAiCgEWLFsHLywu//fZbhrUq9PX10bRpU/zzzz+4ceMGatWqBUEQIJPJMG/ePFWGSkREREREREQaRqVJiyNHjoiPp06diqlTp4ojLrJTuXJlHDt2DMbGxgBS62LcuHFDJXESERERERERkeZR6fSQp0+fAgAKFiyIadOm5Xj70qVLY9iwYVi9erW4v5YtW+ZpjERERERE9OOqs/mW1CGozaPxzaQOgSjPqXSkRVRUFARBgJmZGYoWLZqrfbRo0UJ8/Pnz57wKjYiIiIiIiIg0nEqTFqVLlwaQOtIitwwMDMTHJUuW/O6YiIiIiIiIiEg7qDRpYWZmBplMhhcvXuR6H8+fPxcf165dOy/CIiIiIiIiIiItoNKkha2tLQDg48ePOHXqVI63T0lJgaOjIwRBQM2aNdGgQYM8jpCIiIiIiIiINJVKkxZ9+vRB+/btIZPJMGPGDDx+/DhH28+aNQsPHz6Erq4uVqxYoaIoiYiIiIiISFnOzs4wMjKCkZERrly5AplMBhcXF/Tu3RvVq1dH6dKlYW5ujilTpig16j4yMhKrVq1C165dUbVqVZQsWRLVq1dHnz59sHXrVsTFxWW7j0uXLmHUqFFo1KgRypUrh5IlS6JGjRpo0aIF7O3t4ejoiE+fPuXFr09qptLuIYIgYPfu3bC1tcXNmzfxyy+/YOrUqRg+fLjYyjQjN27cwNKlS3Hjxg0ULFgQq1atQtu2bVUZKhEREREREeVQUlISbG1tce7cOYXn37x5A0dHRxw4cAAbN26EtbV1htt7enpi+PDhiIyMVHj+48eP+PjxIy5fvox///0XLi4uqFevXrrtU1JSMH78eBw4cCDdstDQUISGhuLp06c4fvw4BEHAkCFDcv/LkiRUmrSQj45o3bo1Xrx4gU+fPuGvv/7CihUr0KBBA5iZmcHExAR6enqIiYlBYGAg7t27h/fv30Mmk0EQBHTq1AnBwcHZjrSYNWuWKn8VIiIiIiIi+sbSpUvh4+ODZs2aYfTo0ahWrRoiIyPh5uYGBwcHfP36FWPHjkXp0qXRrl07hW1v3rwJa2trJCUlQRAE2NjYoH///ihVqhSCg4Ph7OyMs2fPIjg4GL169YK3tzcqV66ssI89e/aICYsaNWrA3t4ederUgZGREeLj4/H69WvcuXMHZ8+eVdefhPKYEBkZKVPVzo2NjSEIgsJzMlnq4b59XtnlmQkPD89FhD+O6Gm2UoegwHCti9QhEBEREZEKyGQyBAcHo0KFCjk+Z5dCnc23pA5BbR6Nb6aW4zg7O2PChAnizwMGDMD27duho6NYfcDd3R22trZISUlBpUqV4OvrC11dXQBAcnIymjZtioCAAADA9u3bMxyNsXLlSvz1118AgPbt2+P48eMKy7t3744bN26gQoUKuH79OooVK5ZhzCkpKfj8+TOMjIxy+2tLTtvee3lFpTUtgNQ/bNr/Mnte2eVZbUNERERERETqY2JigjVr1qRLWADAL7/8gsGDBwMAAgMDcebMGXHZ2bNnxYSFlZVVptNHfv/9dzRp0gQA4OXlBT8/P4XloaGhAID69etnmrAAAB0dHa1OWORnKp0eIu8eQkRERERERD+efv36ZZksGDp0KPbu3QsgtX5F7969AaQWzpQbNmxYptsLgoDhw4fjzp074nZ169YVl5crVw4vX77E9evX4e/vj+rVq3/X70OaR6VJi82bN6ty90RERERERCQh+SiIzJibm0NfXx8JCQkKoyQePXoEAChQoAAaN26c5T6aN28uPn748KHCMnt7e1y5cgURERGwsLBA165d0alTJzRp0gRmZmYoUECll7ykBvwXJCIiIiIiolwpVapUlsv19PRgbGyMkJAQhZaj8pqExYsXh4GBQZb7KF26tPg4IiJCYZmlpSXev3+Pv//+G7GxsTh9+jROnz4NAChatCgsLCwwYMAA9OvXjwkMLaXymhZEREREREREqjJx4kQ8fPgQq1atQq9evcRESkxMDNzd3TFq1Ci0bdsWr1+/ljZQyhUmLYiIiIiIiChX5IUwM5OYmCiOjjA1NRWfNzExAQBERUXh69evSh/D2Ng4w3VMTU0xcuRIODk54fnz5/D19cXKlStRu3ZtAMDjx48xdOjQ7H8h0jhMWhAREREREVGuyAtkZubBgwdISEgAAIUCmnXq1AEAJCUlwcfHJ8t93Lx5U3xcr149peL6+eefMWrUKHh5eYnb3L9/H/7+/kptT5pD7ZN6UlJSEBgYiKioKMTGxuaoZamFhYUKIyMiIiIiIqKcOHbsGBYtWpRpBxFHR0fxcceOHcXHnTp1wq5duwAAu3fvRqtWrTI9hoODQ4b7UIa+vj7atGkjFvAMCwtjhxEto5akRUpKClxcXHDgwAH4+PggPj4+x/sQBEGhcAsRERERERFJKzw8HNOnT8f27duho6M4kN/d3R1OTk4AgEqVKqFHjx7ism7duqFq1ap4+fIlXF1d8csvv8DKyird/tesWYNbt24BADp06CCO0JBzdnbGgAEDMi3mGR8fjytXrgBIvaasXLlyrn9XkobKkxbBwcEYOHAgHj9+DAA5GllBREREREREmqtJkyZwdXVFUFAQRo8ejapVqyIqKgqnT5+Gg4MDUlJSoKOjgw0bNkBXV1fcTkdHB5s3b0bPnj2RlJSE0aNH49KlS+jXrx9KliyJt2/fYv/+/XBzcwOQ2mVk7dq16Y4/YcIEzJ8/H7/88gtatmyJqlWrolixYvj8+TOeP3+OPXv2iKMs+vTpg7Jly6rnD0N5RqVJiy9fvqBXr14IDAxUeF5XVxfGxsYoVKiQKg9PREREREREKjRv3jzs2LEDZ86cUag9IWdgYIANGzagXbt26ZY1b94chw4dwvDhwxEZGYn9+/dj//796darUKECXFxcMh0lERERgQMHDuDAgQOZxtm5c2ds3LhR+V+MNIZKkxb//vsvAgMDIQgCZDIZrK2tMWLECDRs2BB6enqqPDQRERERERGpWIECBeDs7IwDBw5g//79ePLkCT5//ozSpUujQ4cOmDRpUpY1JDp27Ih79+5h586d8PDwwIsXLxAdHY3ixYvDzMwMPXv2xNChQzO94X3z5k1cuHABt27dwosXLxAWFoZPnz5BX18f5cqVQ6NGjWBlZYUuXbqo6k9AKiZERkaqbL5G69at8ejRIwiCgIULF2Lq1KmqOlS+Fz3NVuoQFBiudZE6BCIiIiJSAZlMhuDgYFSoUAGCIEgdDknA2dkZEyZMAACcOnUKbdq0kTii/CG/vvdUOtJCPi3E1NQUU6ZMUeWhNMqDBw9w/fp13Lt3D0+ePMGnT58QHh6O5ORkGBsbo1atWujcuTPs7OzE/sREREREREREpEilSYukpCQIgoBatWrlq0zQggUL4O3tneGyDx8+4MOHD/Dy8sLq1auxYcMG9O7dW80REhEREREREWk+lSYtfvrpJ7x48QIJCQmqPIzG0dPTQ/PmzdGkSRPUqlULpUuXRqlSpZCcnIzAwEC4ubnh2LFjiIiIgL29PY4dO4a2bdtKHTYRERERERGRRlFp0qJt27bw9/fHkydPkJiYmG+Kbx44cAAFCmT8p23cuDH69++PXr16YdiwYUhOTsbixYvh6emp5iiJiIiIiIiINJuOKnc+evRo6OrqIiYmBs7Ozqo8lEbJLGGRVr9+/VCtWjUAgK+vL758+aLqsIiIiIiIiIi0ikqTFjVq1MCCBQsgk8kwf/58XL16VZWH0zqGhobi4+joaAkjISIiIiIiUs6gQYMQGRmJyMhIdg4hlVNp0gIApkyZgkWLFiEuLg6//vorZsyYgXv37iElJUXVh9ZoT548wYMHDwAARkZGKFWqlMQREREREREREWkWlda0kJs6dSrq16+PgQMHwsHBAQ4ODjAwMICxsbHSdS4EQcC9e/dUG6iKff78Ge/evcPp06exZcsWJCcnAwBGjRoFHR2V54+IiIiIiIiItIpakhb//vsv1qxZg8TERACATCZDfHw8Pnz4oNT2MplMa1um2tjYwN3dPdPlI0aMwKxZs3K9f00dsaKpcRERERHR95HJZOJ/RKQ+8vddSkqK0tfHP8LNcZUnLWbPno3t27eLiYe0H275+YOuYcOGWLFiBZo1a/Zd+wkMDERSUhLK5FFceeXly5dSh0BEREREKqKrq4uvX79q7Y1FIm0kk8mQlJSEgIAApdYvUKAAqlSpouKoVE+lSYszZ85g27Zt4odZ8eLF8euvv6JRo0YoU6YMChUqpMrDZ8nb2xtxcXHfvR9zc3OUK1cu0+WrVq3CokWLAABfvnzB8+fP4erqikuXLmHcuHFYtGgR+vTp891xEBEREREREf1ohMjISJUNd7C0tISnpycEQYCFhQX27dsHIyMjVR0uR+rVq4egoKDv3s+2bdtgY2OT4+327NmDqVOnQhAE/Pnnn5g0aVKuji+fhvFlxqBcba8qRVbnnxa3RERERPmJTCbD27dvUaFCBY60IFIjmUyG4OBglC9fntND8srdu3dTD1KgAHbu3KkxCQtNYG9vD09PT5w8eRKLFy9Gz5498fPPP+d4P5r6ItTUuIiIiIjo+8infcv/IyL1EQQBOjo6+eq9p9KkRWxsLARBQL169VC6dGlVHirHLly4gKSkpO/ej7Gxca637d27N06ePInk5GQcP34c06dP/+54iIiIiIiIiH4UKk1alC5dGkFBQShSpIgqD5MrmpBEMTU1FR+/efNGwkiIiIiIiIiINI9Kx/DXqVMHMpkMgYGBqjyM1nr79q34uGjRohJGQkRERERERKR5VJq0sLW1BQAEBQWJ9S3of44cOSI+rlu3roSREBEREREREWkelSYtevfujS5dukAmk2H69On48uWLKg+nEe7cuYPbt29nuY5MJsOKFSvg5eUFADAxMUH37t3VEB0RERERERGR9lBpTQsA2L59OwYNGoTr16+jW7duWLduHRo3bqzqw0rm2bNnmDBhAszNzdG9e3eYm5ujdOnS0NfXR0REBB4+fIiDBw/i4cOHAABdXV2sXbsWxYsXlzhyIiIiIiIiIs2i0qTFihUrAAAtW7bE06dP4efnhy5duqBmzZpo2rQpSpUqBT09PaX3N2vWLFWFmucePHiABw8eZLlO+fLlsXbtWnTt2lVNURERERERERFpDyEyMlKmqp0bGxun6x8r7+ucG+Hh4XkRlkrFxcXB29sbV69exf379/H+/Xt8/PgRX758QeHChVG2bFnUrVsX3bp1Q69evVCoUKE8OW70NNs82U9eMVzrInUIRERERKQCMpkMwcHBqFChQq7P64ko5/Lre0/l00NksvQ5kYyey462/KMUKlQI3bp1Q7du3aQOhYiIiIgo39O0m3uqxBuH9CNSadJC3j2EiIiIiIiIiCinVJq02Lx5syp3T0REREREREQ/MJW2PCUiIiIiIiIiyi0mLYiIiIiIiEhpzs7OMDIygpGREa5cuQIAOH/+PGxtbVGrVi2UKlUKtWrVwtChQ3Hnzp1s9yeTyXDixAkMGzYM9erVQ9myZVG+fHk0bdoUU6dOhZ+fn1Jx3bp1C8OHD4eZmRlKlSoFMzMzDBw4EB4eHpnGTZpP5YU4iYiIiIiI6Mckk8kwdepU7NmzR+H5Dx8+4MSJEzh16hQ2bNiAwYMHZ7h9cHAwhgwZAl9f33TL/P394e/vD0dHR8yYMQPz5s3LtEHD8uXLsWLFCoWmD+/fv8f79+9x7tw5jB49GvXr18/9L0qSYdKCiIiIiIiIcmXZsmW4ceMG2rZti99++w3VqlVDXFwc3N3dsWnTJiQlJWHGjBmwsLBAlSpVFLYNCQlB165d8e7dO+jq6sLS0hKdO3dGpUqVoKurCz8/P+zYsQOPHj3CqlWrYGBggJkzZ6aLwcHBAcuXLwcAFClSBOPGjUOHDh1QqFAhPHnyBJs3b8b27dvRrFkztfxNKG8xaUFERERERES5cuPGDUydOhWLFy9WeL5Vq1b4+eefMWXKFHz9+hW7du3C0qVLFdYZP3483r17h1KlSuHYsWOoU6eOwvImTZpg0KBBGD58OE6dOoV//vkHNjY2qFixorhOeHg4FixYAAAwNDTEmTNnUK9ePXF5o0aNYGVlBWtra3h5eeXtL09qkSdJC3UMsxEEAffu3VP5cYiIiIiIiEg5derUwcKFCzNcNnjwYPz1118IDQ3F1atXFZb5+vri4sWLAFKndnybsJDT09PD+vXr4e7ujoSEBDg5OWHu3Lni8v379yMmJgYAMGPGDIWEhZy+vj42bdqEBg0aIDExMVe/J0knT5IWb968gSAICvOH8op8v5nNXSIiIiIiIiJpDBw4EDo6Gfd30NXVRYMGDeDh4YFXr14pLDt58iQAwMDAAL17987yGCYmJqhduzbu3buHmzdvKiy7dOmS+NjOzi7TfZQvXx4dO3aEu7t7lscizZNn00NUkbBQ5X6JiIiIiIjo+9SoUSPL5cbGxgCA6Ohohed9fHwAAF+/fkXJkiWVPl5ISIjCz48ePQIA/PTTTyhVqlSW2zZq1IhJCy2UJ0kLW1vbvNgNERERERERaZHChQtnuVw+CiMlJUXh+bCwsFwdLzY2VuHniIgIAFAq8ZFdUoM0U54kLTZv3pwXuyEiIiIiIqJ8ICkpCQBgZGSEM2fOKL2dvr6+qkIiDcXuIURERERERKRWJUqUgL+/P2JiYlC9enXo6enlaj/Gxsb48OEDPn78mO26oaGhuToGSSvjiilEREREREREKiLvQJmUlJSuuGZOyLuOBAUFZZu48PX1zfVxSDpMWhAREREREZFa9enTR3y8cePGXO+nQ4cO4mMXF5dM13v37p1CpxHSHkxaEBERERERkVq1atUK7du3BwC4u7vjzz//zLJzZEpKCk6cOIGnT58qPG9nZ4eiRYsCAFatWiV2E0krMTERkyZNQkJCQt79AqQ2rGlBREREREREardt2zZ07twZQUFBWLNmDdzd3TFo0CDUr18fhoaG+PLlC16/fg0fHx+cPn0aHz58wPHjx1GrVi1xHyYmJliyZAmmTZuGz58/45dffsG4cePQoUMHFCpUCE+ePMHmzZvh5+eHZs2a4datWwAAQRCk+rUph5i0ICIiIiIiIrUrXbo0Lly4gDFjxsDLywuPHj3C3LlzM11fV1c3wxarw4YNQ0hICFasWIGYmBisXLkSK1euVFhn9OjRqFevnpi0KFiwYN7+MqQyTFoQERERERGRJEqXLo3jx4/j6tWrOHz4MG7evIl3797hy5cvKFy4MMqVKwczMzO0bdsWvXr1QqlSpTLcz+zZs9GhQwds3boVN27cwKdPn2BqaooGDRpg+PDh6Nq1K/79919x/WLFiqnrV6TvJERGRmY+cYi0RvQ0W6lDUGC4NvMiOERERESkvWQyGYKDg1GhQgUOsSetMn78eOzfvx96enoIDg6GgYGB1CHlSH5977EQJxEREREREf3QoqOj4ebmBgBo2LCh1iUs8jMmLYiIiIiIiEhrpaSk4NWrV5kuT0hIwPjx4xEVFQUAGDx4sLpCozzAmhZERERERESktRISEtCkSRN06tQJXbt2hZmZGQwNDREdHY27d+/C0dER/v7+AIAmTZrAzs5O4ogpJ5i0ICIiIiIiIq2WnJwMDw8PeHh4ZLpOs2bN4OzsjAIFeBmsTfiv9YNg4UsiIiIiIsqPDAwM4OLiAk9PT9y6dQsfP35EeHg4BEGAqakpGjVqhP79+6NPnz7Q0WGFBG3D7iFERERERKS0/NrBgEhq+fW9xzQTEREREREREWkkJi2IiIiIiIiISCMxaUFEREREREREGolJCyIiIiIiIiLSSExaEBEREREREZFGYtKCiIiIiIiIiDQSkxZERERERKQ0QRAgCAJSUlKkDoUoX0lJSRHff/kJkxZERERERJQj+vr6+Pr1q9RhEOUrX79+hb6+vtRhqB2TFkRERERElCOFChVCbGys1GEQ5SuxsbEoVKiQ1GGoHZMWRERERESUI4ULF0ZCQgKio6OlDoUoX4iOjkZCQgIKFy4sdShqJ0RGRsqkDoKIiIiIiLRLYmIiPn78CH19fRQuXBgGBgbQ0eE9UaK8kpKSgq9fvyI2NhYJCQkoWbIk9PT0pA5L7Zi0ICIiIiKiXElOTkZsbCzi4uKQkJAAmYyXFkR5RRAE6Ovro1ChQihcuDB0dXWlDkkSTFoQEREREVGeYNKCKO/kty4hmSkgdQBERERERPRj4EUWEeU1TjojIiIiIiIiIo3EpAURERERERERaSQmLYiIiIiIiIhIIzFpQUREREREREQaiUkLIiIiIiIiItJITFoQERERERERkUZi0oKIiIiIiIiINBKTFkRERERERESkkZi0ICIiIiIiIiKNxKQFEREREREREWkkJi2IiIiIiIiISCMxaUFEREREREREGolJCyIiIiIiIiLSSExaEBEREREREZFGYtKCiIiIiIiIiDQSkxZEREREREREpJGYtCAiIiIiIiIijcSkBRERERERERFpJCYtiIiIiIiIiEgjMWlBRERERERERBqJSQsiIiIiIiIi0khMWhARERER/aBu3rwpdQhERN9FiIyMlEkdBBFpjt69e0MQBPz777+oWLGiUtsEBwdj3LhxEAQBJ0+eVHGEREREpCxjY2OYmZlh2LBhsLGxQbFixaQOKVcSEhJw9+5dfPjwAXFxcejRo4fW/i6Ud1xcXAAAPXv2VPr18PnzZ7i5uQEAbG1tVRYb5R0mLUhrXLt2DQDQqFEjFCpUSKlt4uLi4OvrCwCwsLBQWWyZmTJlCoYNG4YGDRqo/di5ZWxsDEEQcP36ddSqVUupbfz9/dGsWTMIgoDw8HAVR0hERDnl7u4OV1dXhIWFoUqVKhg+fDjq1q0rdVii27dvw8HBATdv3kRISAji4uJw7do1he+hmzdv4vnz5yhWrBj69u0rWawymQznz5/H1atX8erVK0RHRyM5OTnLbaRM6su/1wGgUKFC6NevH+zt7dGkSRNJ4smpsLAwLF26FAcPHsTXr1/F5789T9m3bx/279+P4sWL48CBA1KEKuKFtPp8z3mrjo4OPn36pOIIKS8waUFaw9jYGDo6OulOYrIi9YeS/IPU3Nwcw4cPh6WlJYoUKaL2OHLiR0laaNoJcO/evfN8nxzZkj8FBwfDyclJ4bV95MgR/Pzzz+I6z58/x/v371GkSBGVX5isWLFCJfudNWuWSvabnaSkJAQHBwMAypUrB319fYXlCQkJWLJkiUICYOzYsRg+fLgU4QJIvXgbPXo0ChQoAC8vLxgZGSksX7t2LZYsWaLwnJ6eHnbt2oVevXqpMdL0kpKSMGPGDOzbtw9AakIAQIbfQ15eXujXrx90dHRw9+5dpUcD5qWHDx9i5MiR8Pf3V3obmUwm6ffjH3/8gf379yM0NBQAxARG7dq1MWzYMFhbW8PQ0FCS2LLj7++Pfv364d27d+JrA8j49fH27VvUr18fKSkpOHnyJFq3bi1FyAA0+0JafhMwr0lxcxDQ/PPWH+3vLZUCUgdAqvWjvVHSfmGpY7u8IJPJ8ODBA0ydOhXz58+HlZUVhg4dCnNzc8liymsJCQkAkO7kXgpZnQB/Ky4uDpMnT1bLCfDVq1czjEFOflKrrJyurwr169fP8TaCIKBgwYIoVqwYqlWrhhYtWqBfv35qO2HWxpjTWr16Nf755x8kJiYC+N/rQP4elHvy5Ans7e2hp6eHx48fo0SJEiqLafny5Sp5LUqVtDhz5gzs7e1RrFgxPHnyJN3ykSNH4vTp0wBS//7+/v74/fff8e7dO8yfP1/d4QIALly4gLdv36Jz587pEhavXr3CX3/9BUDxuzAhIQHjx49Hy5YtYWpqqs5wFUyfPh1OTk6QyWQoXbo0mjdvnmkytn379qhUqRLevHkDNzc3jBs3Tq2xvnv3Dn379kVkZKT4tyxSpAiMjIygo6O5ZeIWLVqEefPmwc3NDY6OjvDy8oJMJsPjx48xc+ZMLFq0CP3794e9vT0aNWokdbiihIQE2Nra4u3btyhYsCBGjBiBNm3aYODAgRmuX758eVhYWODKlSu4cOGCpEmL76Hqc9ZevXrl+We2IAhaNWJB/jdWx/uWf++8waTFDy6/v1GyumBVh1u3bsHBwQEHDx5EeHg4oqOj4eDgAAcHBzRo0ADDhg2DpaUlChcuLEl8eUV+Ym9iYiJxJJp7AlyhQoUsX4dxcXH49OmT+JrV19eHsbExACAiIkK8KBUEAaampkpPkVKlN2/epHtOEIR0J1yZPXfnzh0cOHAA8+bNw+LFizFixAiVxgtoZ8xyf/31F1avXg2ZTAZ9fX3Url0b9+7dy3DdPn36oHTp0ggNDcXp06dhb2+v0tjy+iRbyoSc/IKua9eu6d5n169fx6lTpyAIAgoVKoRq1arh+fPniI+Px7p162BpaQkzMzO1x3zt2jUIgoB27dqlW+bg4IDk5GQYGBhg27Zt6NixIzw8PDBhwgTExMTA0dER06dPV3vMAHDjxg3s27cPgiBgzJgxWLJkCfT09MTPvoz06tULmzZtwrVr19SetFi7di0iIiIgCAKsrKwwbdo0Sf69c6NAgQLo27cv+vbti8DAQDg6OmL//v0ICQnBly9f4OTkBCcnJ9StWxfDhw/HgAEDULRoUUlj3rdvH16+fAl9fX0cP34czZs3z3abjh074vLly7h9+7YaIsxb6ryQlvJmniYICQkBALW9xvP73zsvMGmRD+TnN4p8yJdUUzKqV6+Ov//+G4sXL8aJEyfg6Ogojn65d+8epkyZgnnz5sHGxgZDhw6VZH5xZqNxfH19s01OJSQkwN/fH+vXr4cgCJLPj9bkE+CHDx9muszPzw+//fYbwsPD8dtvv4kjcQoUSP2ITkpKwoMHD7Bnzx44OzvD0NAQTk5OqF27tsriVUarVq0gCAJCQkLw4sUL8flKlSqhVKlSAIDQ0FAxUSAIAqpVq4YSJUogKioKL1++xNevXxETE4OZM2ciPDwcM2fOZMwZePToEVavXg0gdarR+vXrYWxsnOlrWxAE9O7dGzt37sTly5dVmrS4f/++yvYthTt37kAQhAxHFDo5OQEASpYsifPnz6NixYoIDAxEt27dEBISgr1792LZsmXqDlk8Aa9Tp066ZefOnYMgCBg0aJA4Dc7S0hL//fcfdu7ciYsXL0qWtHB0dAQAtGjRAsuXL1dqm8aNGwMAnj17prK4MnPx4kUIgoA+ffpg+/btaj9+XqlUqRIWLlwojr7Yu3cvLl26hJSUFDx8+BDTp0/HggULYGlpCXt7e8nqcp0+fRqCIMDe3l6phAUA8Tzk1atXqgxNJdR1Ib1p0yaV7l8qyia7P3/+jK1btwKAwrRKVcnu7+3g4IA7d+6gQIEC6NChAxo3bozSpUsDSH1N+Pj44NKlS0hOTkaTJk1UfhNCUzFp8YPL72+U/fv3A4Ak817T0tfXh5WVFaysrPDy5Us4ODjgwIED+PTpE6Kjo7Fr1y7s2rULjRs3xrBhw9C/f38ULFhQLbFlNBpHJpNh4sSJSu9DPkR90KBBeR1ejmjbCTAAfPr0CdbW1vj48SNcXFzQtWvXdOsUKFAAjRo1QqNGjdCzZ08MHjwY1tbWuHz5sqSjW9zc3MQLYkNDQ/z+++8YNGhQuqHm4eHhcHZ2xsqVK/Hp0yesXr0abdq0QUJCAo4ePYoFCxYgLCwMK1aswK+//orq1asz5m/s3LkTMpkMdevWhYODA3R1dbPdpnHjxti5c2eGUxzyktSfr3lNnuzO6N9UftE6fPhw8feuVKkShg8fjr///hvXr19Xa6xy8pi/fR2HhYXB399fvNBOq3Pnzti5cyeeP3+utji/dfPmTQiCgGHDhim9Tfny5QEAHz58UFVYmXr37h0AYMiQIWo/tiro6uqiT58+6NOnD968eYO9e/di//79eP/+PWJiYrB3717s3btXrMtlZWWl1lF+jx49AgD88ssvSm8jfw9ERESoJKac0sQLaTs7O5XuX9Uym+bZv39/8WZPZhISEvDx40ekpKRAEAR06dJFFSEqyOrvPXPmTPj4+MDCwgKbN2/O9Ps0MDAQ48ePx40bN9CwYUOV1ZLSZExa/OC0+Y2SWeHCCRMmZDudIiEhAQEBAfj06VOmQ2alUrVqVSxduhSLFi3CyZMn4ejoiKtXr0Imk+HOnTvw8fHBnDlzYGNjA3t7e7XcTc9oNE5ORugULVoUkydPlrSaO6B9J8AAsGXLFrx//x4jR47MMGHxrV9++QX29vbYuXMnNm/eLNkcegAICgrCkCFDkJycjLNnz2Y60sbExASTJk1Cx44d0a1bNwwdOhRXrlxB+fLlMXDgQDRs2BCdOnVCbGwsdu/erdI71doYM/C/4f+jRo1SKmEBAJUrVwaQWpyOlCdPAHxb8f/169cIDQ2FIAjpLqKaNm0qriOFuLg4ABBrncjJpw8VKFAg3Z1q+Q2KqKgo1QeYCfmd5Zwk/eQXzfHx8SqJKSuFCxdGQkICSpYsqfZjq1rFihUxf/58NG7cGNOnTxf/bdLW5Vq0aBEmT56MiRMnQk9PT+UxRUZGAoBKa/LkFW27kNZmGU3zlMlkYlJRWWZmZjm6QZfX3N3dsXPnTpiZmeHo0aNZ1oWrVKkSjh49inbt2mHHjh3o3LlzvnudMGmRT2nDGyWjwoUymQx3795Vanv5RXfZsmUxefLkPI/ve+np6cHS0hKWlpYICAiAo6MjnJycEB4ejs+fP2Pnzp3YuXMnmjVrhtGjR6Nfv34qmef97WicCRMmQBAEzJs3D2XLls10O/mc7jJlyqB+/foaUWNB206AgdSif4Ig5KiCf69evbBz506cOXNG0qTFli1bEBUVhZkzZyo1NahOnToYN24cVq5cic2bN4vFAWvWrAk7Ozvs2LEDV69eZcwZkJ+M5WQKlnxanPyClpSTkpICAIiOjlZ4Xv7dU7Ro0XQXKPIRT7GxsWqIML1ixYohIiIiXfJVPv2vbt266UbvyX9PdY3qy4h87n5O/m7yO+jKtpHMSzVr1sTNmzfx/v171KtXT+3HV5UPHz5g37592Ldvn9g5Rz6CsnXr1nj27Bk+fvyIyMhILFmyBCdOnMCJEydQvHhxlcZlaGiIiIgI8btdGUFBQQCQ5bRQVfhRLqS1gXyap5w8qd+gQYMsb2qmPW9t1aoVfv31V0k//xwcHCAIAiZMmKBUIXsDAwNMnDgRkyZNgoODA5MWlD9owxvl28KFQUFBEAQBZcqUyTJr/e2H0vDhwzU+Sx8UFIQ3b94gJiZGLAAoT7rcunULt27dwj///IP169crPa9TWd+OxpkwYQKA1N7iyraO0hTadgIM/O8E69uK/1mRrys/uZTK+fPnIQgCOnTooPQ2HTp0wMqVK+Hh4SEmAIDU4mk7duxQ+e+kjTED/7vATE5OVnob+UW3prdZ1jSmpqZ4//49AgMD0bJlS/F5b29vAECjRo3SJZDlSU+pPkdq1KiBmzdvwsPDAz179hSfP3nyJARBQKtWrdJtI09wSDlqoHTp0nj16hXevHmTYYwZ8fHxAZB6jqBuVlZW+O+//3D69GmlRsZpMplMBg8PD+zZswcXLlxAcnKyeN5hamqKwYMHw97eHpUrV0ZSUhLc3Nywbt063Lt3Dw8ePMDq1avx559/qjTGypUrIyIiAs+ePVP6vPPixYsAoPYCqT/KhbQ2cHNzU/hZnqDavHmzVp23yutBZVSLKDPyGxfK3sD9kTBpkU9pwxvl28KF8g+lo0ePatWHUmbCwsLg7OyMvXv3igWj5CcMFhYWGDRoEPz8/HDgwAGEh4fj2bNn6NOnD86dO4eGDRuqLK5Tp04BSB1ho2207QQY+N9815cvXyrdjvPly5eqDElp8jtIORllI1/3/fv3Cs/LR/V8+fIlj6LLmDbGDKQOjw4ODkZgYKA4FSE78s/5cuXKqTI0uLi4qGS/tra2KtlvdurUqYN3797h0KFDYmvFL1++iAmAjNooypOP8mKu6tapUyf8999/cHZ2Ru3atdG6dWvs3bsXAQEBEARBIZEhJ/+OLVOmjLrDFTVv3hwBAQE4ceJEpm0s00pMTMTevXszTcSo2tChQ+Hi4gIXFxf07t1bK+90vn37Fvv27YOTk5P4eSg/92jRogVGjBiBvn37Kkz/kHce6dOnD0aPHg1XV1e4ubmpPGnRvn17+Pr6Yvfu3Rg/fny2XTX8/f1x8OBBCIKAjh07qjS2b2nThbT88yqv/fTTTyrZb3YGDhwIQRBydPNHE8inIubkRpt8XU2p2aJOTFrkU9r4RpFnsbW9PeilS5ewZ88enDt3DomJieLJQrFixTBw4EAMHz4cNWvWFNdfuHAhdu/ejaVLlyI2NhbLly/HwYMHVRaftvY1B7TvBBhIrXFy//597NixA/379892fZlMJlatr1q1qqrDy5Kenh7i4uLw5MkTpavLy4tCfjtaSj6CQNXDjbUxZgBo0qQJgoODceHCBQwYMECpbZydnSEIQp6PzvrW+PHjVdJaW6qkRZ8+fXD+/Hl4eXlhyJAhsLCwwJEjRxAREQEdHZ0M36dSJz9HjhyJzZs3IzIyEnPmzFFY1rRpU4URI3LyUUdSdYYAAGtra7i4uMDd3R3Xrl3LsGOLXEpKCiZPnozXr19DEARJignq6uri0KFDGDRoEOzs7DBixAgMHDgQNWvW1IgpkplJSUnBuXPn4OjoiIsXLyIlJUU89zA0NISNjQ2GDx+e7egEeWcuV1dXlV34pjVixAhs2rQJr1+/xvz58/HXX39l+llz7949DB06FPHx8TA0NMRvv/2m8viyoskX0ubm5ir5zM6u45yqbNmyRZLjfq8SJUrg/fv38PT0VPoc1NPTE0D6osv5geobAZNGkk+XkL/4lSH1G8XNzQ2nT5/Wykr1ISEhWL16NRo0aABLS0ucOnUKCQkJkMlkaNCgATZs2ICnT59ixYoVCgkLIHVqzrhx4/DHH38A+F9hNSn5+flhyZIlmDJlCtatW5ej+aaqZG1tDQDiCXBW0p4AA9JV0/71118hk8lw8+ZNjB49Ot08+rRiYmIwZswYseCoMkkOVapZsyZkMhm2bNmCpKSkbNdPSkrC5s2bIQhCujtPgYGBAFT/+aKNMQNA3759IZPJcOTIETx9+jTb9ZcuXSpW3reyslJ1eOKUtrz8Tyq2traoV68eZDIZTp8+jTlz5uDOnTsAUi9EMqrsf/bsWQiCIHYjUjcjIyMcOnQI5cqVU/gb1qpVCzt37ky3/vPnz+Hr6wtA2kR1+/bt0alTJ8hkMgwcOBAbN25UqAEQHx+PN2/e4MCBA+jQoYN4F33AgAGS1JQwMTFB1apV8d9//yEpKQnbt29Hx44dUb58eZiYmGT5n5QXGXXr1sXgwYNx/vx5cRpIvXr1sHbtWjx58gQrV65UejqF/PdQ5vPze5UrVw6LFi2CTCbD1q1b0aZNG6xatUpcfuTIESxfvhx9+vRBp06d8ObNGwiCgOXLl0s2VUtuy5Yt2Lx5s6QjmbLyI31maysLCwvxfESZ1uH37t3Dli1bJL3RJiWOtMinLCwscPjwYWzZsgW9e/fOdlh6fn+j5NaFCxewZ88eeHh4ICkpSfxQL1SoEPr3748RI0YoPdVDfgfq48ePKosXSB1SPnv2bOjq6sLJySndXYIDBw5g4sSJ4hx7AFi/fj0OHjyIZs2aqTS27MhPgC9evIiBAwfi//7v/2BpaSkul58AX79+HVu2bMHDhw8lPQEGgLFjx8LJyQkvXryAq6srLl68iF9//RWNGzcWh5qHhobCx8cHx48fF0c6Va9eHWPGjJEkZjkrKyvcunULfn5+sLGxwaZNmzI9Qfvw4QMmTZoEPz8/CIKQ7kJa3ipS1cNotTFmIDVp0bBhQ9y9exe//vorVqxYoVC8VRAEJCcn4+bNm9i4cSPc3d0hCALat2+v8s9s+ZSyH4Wuri6OHTuGmTNn4vTp00hMTETBggVha2uLv//+O9363t7e4t3/nNRKyWtNmjTB3bt3cePGDYSGhqJcuXJo0aJFhkPqw8LC8H//938AUj83pbRz50707NkTjx8/xqJFi7Bo0SLxLvC3Q/zlif61a9dKEWq6CzNtuVCTT20rWLAgfv31V4wYMQJNmjTJ1b4MDQ3FUQTqMG7cOERHR2PFihV49OgRHj9+LB579erV4nrywqELFy7UqpaeL1++RFhYGCpXrix29FG1WbNmqeU4lLWxY8fiyJEjiIuLQ69evTBt2jQMHjw43TTD0NBQ7Nu3D+vWrUNsbCx0dXUxfvx4iaKWjhAZGakdn7iUp3x9fdGlSxfIZDIUKVJEqTdKTEwMdHV1cf78eZXWVFDWuXPn4OnpCT8/P/FCztjYGHXr1kXnzp01okiWsbGxWFgTSC2WNmzYMNja2uZ4SPmrV6/EAnDy6T2q8M8//2DZsmVo1apVujmaISEhaNiwYYbdCMqUKYPbt2+jaNGiKotNGZGRkeIJcHYnVfIT4NOnT0tarDA4OBj9+/eHv78/gMz7uqd9HR09elRs1yqV5ORkdOnSBXfv3oUgCNDT00P79u3RqFEjsbjfx48f4evrC29vb3F0UePGjeHh4SFeTMXHx6Nu3boIDw/H2rVrMXToUMacgeDgYHTt2hXv37+HIAgwMDBAfHy8OAQ5JiZGvPspk8lQsWJFXLhw4Ydsz6guX79+RUREBExNTTNt8fj69WuxGKuFhYXaLuZ+JLGxsZg3bx6cnZ3TtW2V09XVxaBBg7Bs2TLJpol+74XC5s2b8yiSnGncuDGGDRuGwYMHa+R0BWXcvHkTq1evxqVLl9KN8tDR0UHr1q0xe/bsDKdCSSE8PFw8h+rTp0+6c743b95g6NCh4h12QRDQt29fbNiwQfLzKG2VnJyMo0eP4uTJk7h//z7Cw8OznQYv5bQWAFi3bh3++OMPhe+NChUqKNy0StvRBwAWL16MKVOmqD9YiTFpkY9p6xvF19cXY8eOxYsXL7Jcr0aNGtiyZYukCRZjY2Po6emhZ8+eGD58ONq0aZPrfcXHx4tDlFU5nLdXr164fv065s+fj+nTpyssW758OVasWAFdXV0sWrQIHTt2hIeHB5YsWQIAWLZsmeR3/wHtOQFOKz4+HmvWrIGDgwPCwsIyXKdEiRIYPnw4pk2bpjHVxSMiIjBw4EDcunULQPYJl+bNm8PFxUWhHd3bt2/h4eEBIPXkTtXDqLUxZrnQ0FCMGzdOYWpf2sSoXPv27bFt2zbJCkMS5cb79+9x7tw5+Pj44OPHj0hOToapqSkaNGiAHj16aGWBaE0QFRUFQRCgr6+vMd8dufXlyxfcv39f4fVRv359jUvG7N27F1OmTMFPP/2EBw8eKCxLTk5GmzZt8PTpU4XPbkEQ0KVLF5XWLftRBQcHY/DgweLfWtlRUKq+EagMZ2dnzJ07F58/fxafk5+XpP09DA0NsWzZMgwaNEjtMWoCJi3yOW17o3h7e8Pa2lqhgKWurq6YwY6KilJoCaivr4/Dhw+jbdu2ksS7atUqDBkyRKsuHBo1aoTXr1/j4MGD6Sqjt2nTBo8ePcKAAQPEYpBA6hC3gwcPon379jh27Ji6Q86UNp4AJyUl4cGDB/Dz8xO/SE1MTFC3bl2Ym5tn2e5XKjKZDHv27MHOnTvx+PHjDNepXbs2Ro0ahaFDh2rEnWhtjDmtu3fv4uTJk/Dx8UFYWBiSkpLE13afPn005m4jSefNmzfincbsTuCzKoBJ2s/ExASCIODPP/8U25qTag0bNgzHjx/H6NGjsWLFCoVlrq6uGDVqFARBQN26ddGmTRtcuXJFnLJ6+PBhdOrUSaLItU9iYiI6duwIPz8/AEC9evVQtmxZeHh4QBAEWFtbIyIiAvfu3UNoaKhYgFg+tVOqEVBpRUVFwcXFRRxB/u35X8eOHXM1SvtHwqQFac0bJSoqCg0bNhSngvTr10+cl2lgYAAgdSivj48Pdu3ahaNHjwJILRrl6+sreVEmbVG5cmV8/vwZnp6eClXlo6KiUKVKFQDA/v370a1bN3HZ6dOn8dtvv6FMmTJipwXKn0JDQ/Ho0SOFKVt16tTR6MSdNsZM3y9tq9a0nUq+t4WrVF1PgNTisGvWrMGpU6cQGRmp1DZSD48m1StdujQSExNx7tw5yWtP5RctW7bEs2fPsG3btnT1kGxsbODh4YE6derg0qVL0NPTw9evX9G1a1c8fPgQVlZW2LZtm0SRK3r//j1CQkIQGxuLhg0bamSXHCcnJ0yaNAmCIGDdunUYMmQInjx5InYdlF/XyGQyHD16FP/3f/+Hr1+/Yvfu3RoxlZyUo3m37EjtihcvjrFjx2Ls2LFSh5KlXbt2iS3nNm/eDBsbm3TrGBgYoFWrVmjVqhW6deuGMWPGIDw8HLt27cK0adMkiFr7fPnyJcPn79+/D5lMBl1d3XR35cqVKwcgf/aNJkWlSpXSuot9bYxZG2j6xb+8Veu37VW/p4WrlK1ar1y5gsGDByM6OlprCkSam5tDR0cHR48ezbArS0Zev36Nvn37QhAEjeimpQ1KlSqFt2/fZlqXhfKe/EK5cuXKCs8nJyfj+vXrEAQBw4YNE/9NDAwMYG9vj+nTp+Pu3bvqDldBfHw8Nm/eDEdHR4XWttevX1coPH306FG4u7ujePHi+Oeff6QIFcD/ikG3adMGQ4YMyXQ9QRBgaWmJmjVromvXrhg9ejSuXbsmeX0wUg6TFqQ15NXwbWxsMkxYfMvKygqXLl0Se8AzaaGcIkWK4PPnzwgNDVV4/saNGwBSW0YaGhoqLJMXJ5R66sLcuXMxaNAg1KlTR9I4iEg7Lv4zu7jXlot+uYiICAwdOhSfP39GkSJFMGTIEBQvXhzLly+HIAjYsGEDIiIi4OvrizNnziAxMRGtWrWSvMtCUFAQBEFAQkKC0tskJiaKrS3VLe0FXG789NNPeRRJzjRu3Bhv377FkydPNKKQurI0PfGZFflNnG9riDx58gQxMTEQBCHdFJAaNWoAgELbX3X78OEDrKys8OjRo3T1Nr5Vt25djBgxQvy8luq1JZ9WM3DgQKXWr1u3Luzt7bFlyxbs2LEDixcvVm2AlCeYtCCt8fLlSwBQaGGZHUtLS7i4uGRbtPN79e7dO8/3KQgCTp48mef7zU7VqlVx9+5dXLp0SWHY3OnTpyEIQoZz5eVtWEuUKKG2ODOyZcsWbN26FXXr1sWgQYMwYMAAtRVIzE7ak920J67aehJMpAxNvvjPrFWrNrZwlY9E1NfXx7lz51C3bl08efIEy5cvBwAMHjxYXDcoKAj29va4fv062rVrh5kzZ0oVttYxNzf/rkScVNNwhgwZghMnTmDHjh0YOHBghm1wNZE2JD4zo6enh6SkpHQjUOUF1UuVKpVuFIZ8GnNOknh5KSUlBYMHDxbbfPfu3Rtt2rQR2yN/q0aNGmjcuDF8fX1x7tw5yZIW8lEtFStWFJ9LexMtPj4+XfKoa9eu2LJlCzw8PJi00BJMWpDWkBcLzUn7PvlFdHR0tEpikrt69Wqe3vWR9xuXQvv27eHr64u9e/eibdu2aN26NRwdHcUvsbS1LOQePXoEILXtqdRkMhn8/PwwZ84cLFiwAF26dIGdnR26desGXV1dyeKqX78+gPQnrtp6EvytyMhI3Lp1C69evUJ0dLRCQdzMSN0rXhtjDggIgIODA65evYrXr18jOjoaKSkpWW4j1etEmYv/2NhYPHv2DAcPHsSjR4/QqlUrzJ49Wy2ff5l1YVJldyZVuXjxIgRBwIABA1C3bt0s1/3pp59w5MgRtGrVCitWrEDHjh3RuHFjNUX6/eLj4wFArGWlbpqciMtMx44dMWzYMDg4OGD48OFYv3695HXKlKWNf28g9Xzo1atXePz4sUIx+IsXLwJAhrVF5Oe5Ut1scXV1hY+PD3R0dLB792707dsXADJNWgDAL7/8Ah8fH7EblxTk3xdpa9elbRv78ePHdDd45N1m3r59q/oAs/H582ccOnQIV69eVfp8JD9Oj2PSghAcHAw3Nzfcv38fnz59QlxcXJZfElKNADAyMkJYWBiCgoJgbm6u1Dbylq2q/nKuUKGCxnUXyK3hw4dj69atiIuLS9ctpmbNmhlWtJafMGd3sqxqFy5cwP79+3H06FFERkYiMTERZ8+exdmzZ2FqaooBAwbA1tZW6ddPXsrqPaWtJ2VA6pftwoULceDAgRzfHZIqAaCNMQOpd9PnzZsnxqzprxtlL/67du2KiRMn4o8//sCGDRtw9OhRrF27VsXR/Vj8/f0BINOicikpKQp3142MjDB27FgsWrQIDg4OWpW0kM/3l2JknzLvf3kizsvLC4mJiWjevDnatWunhugy5+LigqZNm+LOnTs4efIkPD090a1bN9SrVw/GxsbZJvSlGrGg6YnPrDRu3BgBAQHYs2cPfvvtNxQpUgQvXrwQO1q0b98+3TbyEcVSjVo9evSo2G1DnrDITr169QD87zNICqampnj37p1C8eGSJUuKo12ePn2aLmkhT1Z8/fpVnaGmc+nSJYwePVq8sZCTVq35DZMW+VhiYiLmz5+P3bt3K3WHEZB2BECtWrVw9epVuLi4oGfPnkpts3//fgCAmZmZKkPDw4cPVbp/dSpfvjx27tyJMWPGKIxQKVmyJHbs2JHu3z84OFisdyH1HcrGjRujcePGWLZsGc6ePQsXFxdcvHgRSUlJCAsLw7Zt27Bt2zbUrl0bgwYNgpWVldpODjI72ZX6zv33iIuLQ+/evfHw4cMcX0BL9TmijTEDgJeXF37//Xfx53LlyqFevXowMjLSmqHeWREEAYsXL4avry8cHR3RvXt3VnXPgaioKACpCXS5tCMRYmNjFe48Av+703vt2jU1RJgqsxoFZ86cybb4YEJCAvz9/bFv3z4IgiDJUPTZs2crve779+8xevRo3LhxA4MHD5a0Zfy30yyio6Ph6uoKV1fXbLeVcpqFNic+7ezscPjwYTx//hwtWrRAgwYNcP36dSQkJKBIkSLo169fum3k51LVqlVTd7gAUguuA1A6YQFALGItn6IhhRo1auDdu3fw9/dHmzZtAKRODzEzM8PDhw/h6uqKLl26KGwj/yySF5KXwvPnz2FnZ4evX79CJpNBT08PVatWhbGx8Q/xvZ7XmLTIx8aNG4ejR4+KHSFMTEzw8eNHCIKAcuXKISoqCjExMQBSv7RKlCiRbk6YOnXv3h1XrlzBmTNnsG7dOkydOjXL9detW4czZ85AEAT06NFDPUH+ILp37y7OUQwNDUW5cuXQo0ePDNvGPn36FNbW1gCADh06qDvUDOnr66Nv377o27cvwsLCcOjQIRw4cEBMLj1+/Bjz5s3DokWL0KlTJ9jZ2aF79+4qLSSa2cluTk6CNc3WrVvx4MEDAKm1UMaMGYMmTZpo9BeuNsYMABs3bgSQWih306ZNOTqp1CZDhw7FlStX4ODgwKRFDhgYGCA2NlbhjnnagskfPnxIdyEk71rwbdFlVcqoRoFMJsPSpUuV3of85smIESPyOrw8VbZsWRw6dAitW7fGjBkz0KhRI5XfQMnKt0laTR+plVOalvhs3749bGxscPDgQQQHB+Pt27fi33zu3LkwNjZWWD8hIUEsOC9VW1p54iEnF/Lyzxxlb36qQosWLXDp0iVcu3YNw4cPF5/v27cvHjx4gMOHD6N8+fIYMGAAYmNjsW/fPri5uUEQBHTs2FGyuNetW4f4+HgIgoAZM2Zg4sSJ4rQVSk+IjIz8sT61SCne3t749ddfxWFgy5Ytw/v372FhYaHQ0/jZs2fYunUrHB0dUaNGDRw4cCBd4SB1iY+PR5MmTcSqys2bN8ewYcPQrFkzlC5dGgAQEhKCW7duwcHBATdv3oRMJkP58uXh4+Mj2fxX0hyPHz/G/v374erqipCQEAD/u3tuYmKi8oKtP5p27drhwYMHMDc3x9mzZ1G4cGGpQ8qWNsYMAFWqVEFUVBQWLlyYbcJWm927dw8dOnRA6dKl8fTpU6nD0RpNmzbFy5cvcejQIXTu3Fl8/qeffsKXL1+wbds2WFlZKWzj4uKC8ePHo3Dhwmqb1/3thVpuVKpUCXPmzFGqi5gm2LFjB/7v//4Pv/32GzZs2CBJDFevXv2u7aUeRZkTR44cwciRI9GtW7fv7j7yvWQyGRwcHODq6ireALK3t0f//v3Trbt//35MmDABAHDr1i1Ur15d3eGK3zNHjhxRuAllbGwMQRDStTwFUjv7DRw4UNLP7AcPHqBdu3YwNDTE48ePxVFlMTExaN68Od69e5dhstTQ0BBXr15VKOCpTnXr1sW7d+9gb2+PNWvWSBKDNuFIi3zK2dkZQOoQtK1btwJIvRPzrZo1a2Lt2rVo0aIFxo0bBxsbG3h7e0sy4qJgwYJwcnJCnz59EB0djZs3b+LmzZuZri//QNq/fz8TFjkgrzLfu3dvrTkpVFbt2rWxdOlS/Pnnn7h48SJcXFxw9uxZxMfHSzq0UVsFBARAEARMmzZNay7+tTFm4H/V5KWeG69q8iKL31bcp6zVrl0bL1++xOPHjxWSFk2aNIGXlxd2796NAQMGiCfu8fHx2Lx5M4DUEUfqkrZGgUwmQ58+fSAIAjZu3IhKlSplup0gCChUqBDKlCkj6XDu3JBPY/H29pYsBm1KOnwv+es5u+lG6iAIAoYPH65w9z8zdnZ2krcgLl++PKKiohAQEKD0yNnLly8DkG5KC5Ba0NzJyQlJSUmIi4sTkxZFixbFkSNHMGTIkHQ1N8qWLYtdu3ZJlrAA/td5L6OpQpQekxb51J07d3I0vNLGxgZnz57FyZMnsWvXLjEbrG4NGjSAt7c3Jk+enO2dg7Zt22L9+vWSjQzRVmfOnAEApb5ktZUgCNDV1dXo6QAJCQliIsXExAT6+voSR5S5n3/+WeoQckzbYi5fvjxevHgh6RBcdfDw8ACQ+pon5VlYWODkyZO4dOkSJk+eLD5vY2MDLy8v3Lx5E927d0f//v0RGxuLw4cP4/HjxxAEAX369FFbnJldPDdu3DjdHdwfhXzaoXx0H6kWE5+517ZtWzx69AiOjo5KXR98+PABTk5OEARB8oR6ZrXuatWqhRs3buDy5ct49OgRkpOTUa1aNXTq1EnSKe9AapOAsLAwrenkIzUmLfIp+RzWGjVqiM+lvYBLTEwU57vKDRgwACdOnMDJkyclS1oAqcPXTp06hYcPH+LSpUvw8/MTv5yMjY1Rt25ddOrUCXXq1JEsxm9pUzsjU1NTfPr0SbLq1ar07NkzuLi44NChQ+LIIvkcU0340ggICMCOHTvg6emJFy9eiLEJgiB+yY4YMUKtd0az8tNPP+Hp06dimzZtoI0xA6n1Yl68eAFfX180adJE6nBUwtnZGf/++y8EQUDLli2lDker9OrVC7NmzcKVK1cQFBQkVsq3sbHBvn37cP36ddy6dStdW8Lq1atj/PjxUoQM4H+F/7Rt9ERO3L59GwC0amSXNmPiM/eGDRuG7du3w8/PD5s2bcryXP/du3ews7PD58+fUbBgQQwdOlSNkeaMrq4uOnTooDF11+Rq166Ny5cvIzAwUJKudtqGSYt8Sj7UOO2Fadov1PDwcLFOhJz8JCggIEANEWavXr16YqslTaZt7Yx+/vlnfPr0CR8+fPghPkQjIiLg6uoKFxcXMQkk/zeQf5HZ2dlJXqx1xYoVWL16NZKSkgAovk5kMhn8/f3h7++PnTt34vfff8+yb7q69OzZE0+fPsWVK1e0ZvixNsYMpBZOdnZ2xqZNmzBw4MAMi+JqGmWS2ykpKYiIiMD9+/cREhIiFoaeNGmSGiL8cZQrVw6vXr1CSkqKQgFOQRBw4MABzJ07F4cPHxbb++no6KBnz55Ys2aNpBfTUg7NVgd/f3+sWrUKgiBo1PnK1atXcfPmTXz48AFxcXGYP38+ypQpo7BOSkoKBEHQqtaKmpz4fPz4MTw9PTO90SZlkVa56tWrY+LEiVi/fj0WLFiAy5cvw9LSUlx+/fp1+Pr64vr16zh27Bji4uIgCAJmzZoldhEh5Q0ePBje3t44cuQIevfuLXU4Go+FOPOpGjVqICwsDGfPnkXz5s0BpLZEk7dLc3NzS/eB7+npCUtLS+jp6am12rg2e/78Odq1a5frdkanT59WQ5SK/v33XyxYsACDBw8WOxZom+TkZLi7u8PFxQUeHh5ITEwE8L9EQK1atWBrawsbG5t0yTkpzJo1Czt27BDjK1asGMzNzRUKzD548EAcHSAIAkaNGoUVK1ZIFjOQmty0sLBAfHw8vLy8spyTrim0MWa5U6dOYcSIEahduzY2bdqkUaPJMiIv3qYM+WvfwMAAq1evlrQ95I8qJiZGnGJUtWpVVqnPBWUKO8oTcXfv3sWZM2fE7gC7du2SfO76tWvXMH369HTz+78tsLhp0yYsWLAAhoaGePbsmWTD6L8n8Xn+/HlJ2uJ+682bN5g6dSq8vLyyXK9jx45Ys2aN5Ik8mUyGyZMni9M+sloPSJ1KvHr1anWFp7T3798jJCQEsbGxaNiwIQoVKiR1SBkaOHAgPDw8sHbtWo0eraIJmLTIp7p164Zbt25h8+bNGDhwoPh8w4YNERgYiOnTp2P+/PkK28ydOxdbtmxhVfccGD9+PFxcXCAIAqZPn64V7Yzi4+PRtm1bBAQEwNHRMdN5gpro/v37cHFxwZEjR9KNbDE2NoalpSXs7Ow04kRGzsvLC/369YMgCChVqhSWLFmCX3/9Nd30rKSkJBw9ehQLFy5ESEgIBEHAsWPHJJ9HevfuXdjY2EBPTw8rVqxAz549Nf7OnDbGLHfv3j1YW1sjLCwMZmZmqFGjRrZ3ygVBwL///qumCP9HmU4ROjo6KFq0KCpXroy2bdti+PDhqFKlihqiI3VKe7Fva2ub4fO5kXZf6pCTRBzwv+8fTbgJ4ObmhmHDhiEpKUlhJF9GXSG+fPmCGjVqIC4uDg4ODpK1WNb2xKefnx969+6NqKiobEfZCoIAIyMjnD59GrVr11ZThJk7cuQIVq1alen5fpUqVTBnzpx0XYmkJC8w7OjoiKCgIPH5b1/fR48ehbu7O4oXL45//vlHilABAEFBQYiPj8eUKVPw33//oVu3bhg4cCBq1qyp1Ag4+Qj4/IJJi3xq3rx52LJlC0aMGIGVK1eKz8+aNQvbt29HkSJFsHfvXnTs2BEymQyurq6YNGkSEhIS0Lt3bzg6OkoYvfbQ1nZGQUFBGDRoEB49egQrKyvY2tqiXr16OT5hUzd5fPKTgwIFCqBTp06ws7ND9+7d0yUCNMGgQYNw5swZmJiY4NKlS9neZQkMDESHDh0QGRmJ7t27i52ApCAfzvjhwwe8ePECgiDA0NAQ1apVU+pC+uTJk+oIU4E2xiz35s0bTJkyBd7e3pDJZDl6L7I7DklJ/tksCIKYUE77fG58uy91yEnLVkEQUL9+fYwbNw7W1tYqjCp7Hz9+ROPGjREdHY2qVavir7/+goWFBX766adMW1mOGDECx44dw9ChQ7F27VpJ4tbmxGd8fDyaNm2K4OBgAKmdfIYPH47mzZuLoyhDQ0Nx8+ZNODg4iPVmKlasiFu3bmlM1zt/f3/4+Pjg48ePSE5OhqmpKRo0aKBR052A1O90KysrPHr0KNuk3PPnz9G8eXMIgoCLFy9KdiMr7edfTr/Tpfj8kxprWuRTnTt3xubNm3HmzBmsWLFCnKowfvx4ODo6IjY2FgMGDICRkRESEhIQGxsLmUwGHR0djBs3TqWx1a9fP8/3KVVRS21sZ5S2eJVMJsOhQ4dw6NAhpbbVhA9RmUyGOnXqiNM/NL2gqLyTz+TJk5UaFlqpUiVMnjwZf/zxB+7cuaOGCDN39epV8UtWniz6/Plztq3mcvrlnJe0MWYg9eS2Z8+eePv2rXhCpmx9HCJNkNnrVZtex2lbtmZGV1cXRYsWRaVKlTSm9sz27dsRHR2NUqVK4ezZsyhZsmS227Rq1QpHjx4Vi6VKQZs7gDg6OiI4OBiCIGDevHmYMWNGunWqVKmCKlWqYODAgVizZg2WLFmCoKAg7N27F6NGjZIg6vSqV6+O6tWrSx1GllJSUjB48GD4+flBEAT07t0bbdq0ybT2V40aNdC4cWP4+vri3Llzko6+/bZ+GWWOSYt8qm3bthgzZgySk5Px5s0bsS1opUqVsGXLFowbNw5fv35V+MIQBAFLlixBixYtVBrbmzdv8nyfUl1saGM7o28/NLXpQ3Ts2LGwtbXVqgKikZGRAFJPEJVlYWEBAIiKilJFSEqrUKGCRo+8yYg2xgwA69evF0+Au3btivHjx8Pc3BxGRkZa+ftQ/pLZxb4ySQAp9O7dW5xWlTaZLK+BU65cOejq6koVXo5dvHgRgiBg7NixSiUsgP91lwsMDFRlaD+sM2fOQBAEdO/ePcOExbemT58OHx8fnDlzBqdPn5YkabFx40a0adMGDRo0UPuxv4erqyt8fHygo6OD3bt3i9OZsipY/ssvv8DHxyddRyV1mjVrlmTH1kZMWuRTurq6WL58eYbL+vXrhwYNGsDJyUmhp7GdnZ1ahoO1atXqhzkJ18Z2RuqeI5yXli1bJnUIOVayZEm8e/cuRyfA8nWlHkXy8OFDSY+fG9oYMwC4u7tDEAT06tULe/fulTocohzJrFOPpnbwkY/Iio2NVXje3NwcOjo6uHbtWrrpFJrs1atXAJCjm07ymy3R0dEqiUkZW7duBZA61VZTXyuZefLkCQDkqLaGfLqofFt1W7hwoVhbw8LCAu3atUPbtm3FBJamOnr0KARBgLW1tdL1V+TXM98WpVWn2bNnS3ZsbcSkBWWoSpUqWLBggSTHdnNzk+S4qqCN7Yw2b94sdQj5StOmTXHixAncvXsXjRo1UmobHx8fAECzZs1UGRppkLdv3wIAhg0bJnEk6f1IU/qI0spopKE2jT6UkydfctJBQd4eV8raCnPmzIEgCHBwcJAshtySj6KUd+VTRvny5QFIO4pSJpMhIiICbm5u4vl4mTJl0Lp1azGJoWkFIOVTmHJSMFbeopX1nrQHkxZEKjRgwAC4urri5MmTcHR0ZDsjSmfkyJE4ceIE1q1bB0tLy2y7y0RGRmL9+vXQ0dHRmDmvpHrFixdHaGgoTE1NpQ4lnZxM6UtbKDe754ikUrRoUXz58gVhYWFSh5InTExMEBISgqCgIKWH/svvQEv5mVOsWDFER0eLU5i1SbFixRAeHo7Q0FClt5HXQTM0NFRVWFny8fGBt7c3vL29ceXKFfGC/v3793B1dYWrqysAiEVP5UkMqb+X5HGWK1dO6W3kI1aTk5NVEhPlPSYtCDKZDGfOnIGnpyf8/PzEQoqmpqaoW7cuOnbsiB49evAkMheCgoKwZMkSfP78GdOmTYO7uzvbGalJcHAw3NzccP/+fXz69AlxcXFZ3iGTqjOEhYUF5s+fjyVLlqBHjx7YsGEDmjRpkuG6t2/fxuTJk/Hu3TssXrwYLVu2VHO0JJU6deogNDRUI6eaZTel78OHDwgICIBMJoNMJkPFihXF6vkhISFi0kMQBFStWlVcRiSVypUr49GjR9i3bx8sLCzEYuXayszMDCEhIbh7967Soz5PnDgBQRCUHgGoCuXLl8fTp08lnaKSW1WrVkV4eDiOHTuGzp07K7XNsWPHAADVqlVTZWiZ+vnnn/Hzzz+LI/oePHgAb29vXL58GTdu3MCXL18ApE43ev36tThV0czMDO3atcPff/8tSdyFCxdGVFRUjgrBf/jwAUDOOgKpw5s3bxAeHi42QMiKvL5ZfsGkRT53+fJlTJkyRaHQkvxNEhAQgNu3b8PBwQGVK1fG+vXr0aZNG6lCVaAtiRZzc3OFdkbnzp3DuXPnlNpWEzpxpJWcnCwOdzQyMtLYImSJiYmYP38+du/erXQGXcrOECtWrAAANGrUCL6+vujatStq1qyJJk2aiMMXQ0NDcefOHTx79kxcNy4uTtw2Iyzw9GMZMmQIPD09sX//fo2bapbVlD5vb2/Y29ujcOHCmDZtGoYMGZKuEODHjx/h6OiIdevWITw8HKtXr0bbtm1VHTZpABcXl+/aXlU1mLp16wY/Pz+4urrC29sbVapUUWiZPWHCBKVuPKQlZcvkrl274tKlS9i1axemTJmSbXHwM2fOwMPDA4IgoEePHmqKMr0uXbrg6dOnuHLlitbVtOjSpQtu3bqFAwcOoGPHjrC0tMxy/SNHjsDFxUUstqwJzM3NYW5ujkmTJiEpKQm3b98WR2L4+PggMTERAPD48WM8efJEsqRF+fLlERUVhYCAAHTo0EGpbS5fvgxAugRRWoGBgVizZg1OnTolnmdnR9OuEdRBiIyM1L7JeZQnDh8+jHHjxiElJUVMVOjr64tZx4iICCQkJIjr6+rqYuvWrRgwYIAk8cpllWhJe+GpCYmW78ngCoIg+Vy7N2/eYPv27fD09MSzZ88U/s41a9ZEly5dMHLkSI0aETJy5EgcPXoUMpkMurq6MDExwcePHyEIAsqVK4eoqCjExMQASP09SpQogYIFCwJIvaugbmn7dANZJ1ByklzJy9eO/CL525Pu77l4VvUJvDbGnJ1Ro0bhyJEjmD17dpZV0TXFmzdv0KZNGyQlJcHNzS3bYel3795Fr169oK+vjytXruRoLjhpp28//3JClSftMTEx6Nq1q1gQMe3Nh7Q/K0v+2S3Vd3pcXBwaNGiAjx8/okWLFti9ezfKlCkj/v2vX7+OWrVq4evXr9ixYweWLl2KhIQEVKlSBbdv35ZspMnbt2/RsmVLCIIALy8vVKlSRZI4ciMyMhINGzYU61NYWVlh2LBhaNy4sZgAS0xMhI+PD3bv3g1XV1fIZDIYGxvj7t27Gt117v79+zhz5gy2bt2K6OhoyV/fc+fOxZYtW1CvXj0xGQEg3etb7sOHD2jevDmio6MxZ84czJw5U4qwAQBXrlzB4MGDxb+jsjThGkHdmLTIp16+fAkLCwt8/foVOjo6GDx4MIYOHQpzc3MUKJA6ACcpKQkPHjzAnj174OzsjJSUFBQsWBBXr15F1apVJYlb2xItmXVoUZaUlYW3bNmCP/74Q/x7ZjbnXF9fH4sXL8bYsWPVHuO3vL298euvv4pVpJctW4b379/DwsJC4QP+2bNn2Lp1KxwdHVGjRg0cOHBAsjmzqhiamNdfZmkvLNLuN7cXHOo4wdHGmLNy7do1JCcn4++//8atW7dgbm6OAQMGKD3VTIphpLNnz8a2bdswY8YMzJ8/X6ltlixZgjVr1mDs2LFa2Q2IckaTE/tfvnzBzp07cfHiRbx//x5fv35FUFAQBEFAmTJlxHOlnJAiMS535coVWFpaIikpCXp6emjZsiW8vb0hCAI6duyI+Ph43Lt3TxyWbmBggLNnz6Jhw4aSxQwAnp6esLe3h4GBARYtWoQBAwaINxo0naenJ2xsbBRGfRYoUEBMSERFRSEpKQlA6ndMgQIFcPjwYbRv316KcDMVEBCgUOsiIiICwP/OC3V1ddGwYUOcP39ekvj8/f3RsmVLpKSkYMmSJZgwYQKAjJMW7969g52dHe7fv4+CBQvi/v374qhWdYuIiEDjxo0RERGBIkWKYMiQIShevDiWL18OQRCwYcMGREREwNfXF2fOnEFiYiJatWoFOzs7ABD/n18waZFPTZ06FY6OjtDT04OTk1O2Q9Hc3d0xePBgJCcnY+jQoVi7dq2aIv0fbU20aKONGzdi0aJF4hdS4cKFUa9ePXGueWhoKB4+fCjObxQEAX/88QcmTZokWcwAMHr0aBw+fBjVq1cXe28/efJEnHP/7QnuwYMHMW7cOFSvXh3e3t6SnAjlpIhhTlSsWDHP9lWvXj3xQj/tSXfa53NDlSfw2hhzVjT1jnRWmjRpgoCAAJw+fRqtWrVSapvr16+jZ8+eqFatGm7fvq3iCElqV69ezXad2NhYPHv2DAcPHsSjR4/QqlUrzJ49G4IgqH3KQGZ3brWFt7c3Ro4cibCwsAw/T+Tf+aampnB0dJR8zrx8ZFxwcDBev34NQRCgp6eHqlWrZjtNVeqRcXI3b97EuHHjxLazmalatSo2b96sEV3BQkJCxCSFt7c33r17B0Dx5lWtWrXEYpytW7dGsWLFpAoXALB48WKsX78egiCgS5cusLS0xJgxYyAIAlatWoWCBQvi+vXrOHbsGOLi4gCktnidOnWqZDGvWrUKf/31F/T19XHx4kXUrVs303PWoKAg2Nvb4+7du5KPDpEKkxb5VP369REUFISRI0fin3/+UWqbmTNnYufOnahYsaLYXkidtDHRoo0CAgLQvHlzJCcno1ixYli4cCFsbW3TtUqLj4+Hi4sL/vzzT0RGRkJPTw83b96UdPhmo0aN8Pr1ayxbtgxjxowBkHXSAgDs7e1x8uRJhew8kabR5DvSmSlfvjzi4uLg6empdMeCe/fuoUOHDihcuLDY5pUISL1g+uOPP7BhwwbJvtO1PWkBANHR0XBwcMCpU6dw79498U6/IAioU6cO+vTpgzFjxkh+EQpkPH1SLqskrtQj476VkpKCs2fP4tKlS3j06JEYl4mJCerUqYNOnTqhW7duktdhmzlzJq5cuYLnz58DUPx7//TTT2jXrp3YMUSq0QmZkclkmDx5MpycnLJ9bQDA8OHDsXr1anWFl6Hu3bvj5s2bsLW1xaZNmwBkfc4aGRmJVq1aITQ0FO7u7mjcuLEUYUuGhTjzqZCQEABAz549ld6md+/e2Llzp7itul26dAmCIMDe3l6pIkW//PIL7O3tsXPnTnh6eqohQuW9fv1aHF5nbGysUe28tm/fjqSkJBQtWhRnzpxB7dq1M1yvYMGCGDZsGFq0aIGuXbviy5cv2LFjh2SFmACIrcVq1KghPpd2Lm5iYqJCMTUgtS3tiRMncPLkSSYtSGPJT2i0iZ6eHuLi4vDo0SOlkxaPHj0StyVKSxAELF68GL6+vnB0dET37t3VXrBQ/j7MSWtFTWNoaIjJkydj8uTJSElJQWRkJJKSkmBiYpKrKS+qVKFCBckv5POCjo4OevbsmaNzbins3LlTbEFdokQJtGnTRkxUaNJ5akYEQcDGjRvRvn17rFq1Ck+fPs1wvSpVqmDOnDmwsrJSc4TpydsKZ/Y5lpKSonAOa2RkhLFjx2LRokVwcHBg0oLyh2LFiiEsLAxGRkZKbyOfgydV9l0bEy1p3bx5E5s2bYK3t3e69l2Ghobo2LEjxo0bJ/nQwMuXL0MQBIwbNy7ThEVaZmZmGDt2LFatWgUvLy/VB5gFef2NEiVKiM+lne8fHh6erp2ivIhoQECAGiL8sZw5cwZA6lSUunXrShzNj00b566amZnhv//+w6ZNm2BlZQV9ff0s1//69Ss2bdoEQRBgZmampihJ2wwdOhRXrlyBg4OD2pMW2vg+BAA/P78MP6N1dHRgYmIiQUTKefjwodQh5EuCIKB69eqoVasWatWqpVHF1rNjaWkJS0tL+Pv7w8fHBx8/fkRycjJMTU3RoEED1KtXT+oQRfIirWmLThsYGIiPY2NjUbRoUYVt5NcI165dU0OEmkW7G09TrsmHNb58+VLpbeTrSnUyKU+WaFOiBUgdijZnzhx0794dp0+fxufPnyGTyRT++/z5M06cOIFu3bph7ty5ksUKpM4dBaB02ygA6Nixo8K2UpG/NmJjY8XnTE1NxTs1GSUm5HP95V8epLxBgwZh8ODB8PPzkzoU0kDW1tYAgKdPn2LAgAFZfj4EBQVhwIABYreGgQMHqiVG0j7y+lR3796VOBLtIb9jvm3bNo2ZMkGap2XLltDT04NMJsONGzewfPlydO/eHZUrV4aNjQ02bdokjobTJFu3bsXWrVsVauRUr14dAwcOxKRJkzB16lT89ttvGpWwAP6XoEhbm8XQ0FB8/OHDh3TbyEchykcW5yccaZFP2dra4sqVK9ixYwf69++f7foymQzbt2+HIAiS3WmoVasWrl69ipcvX6J+/fpKbSN1ogUA5s2bh61bt4o/V6pUCc2aNRPv+IeEhODWrVtiC9etW7dCEAT89ddfksQrH63wbQ2LrMjXlffslsrPP/+MsLAwvHr1Cs2bNweQOtKiUqVKCAwMxMWLF9GyZUuFbS5cuAAAkt1t+p4pKYIg4N9//83DaHKmaNGi+PLli9bO7SbVGjp0KA4cOIBbt27h6tWraNSoEdq1a4fGjRuL86FDQ0Ph4+MDb29vcW59s2bNMGTIEClDJw0WHx8PAOIUS1LOw4cPMWfOHCxcuBCdO3eGnZ0dunXrlmUxS6n17t1b/J5Ttrh0cHAwxo0bpzGFOAMCAuDg4ICrV6/i9evXiI6ORkpKSpbbSFU8+cyZM4iLi8ONGzfEIpwPHz5ETEwMPDw8xO4gpqamYhHOtm3bSj51ZM6cORAEAQ4ODpLGkVNly5bFy5cvERYWJj5XsmRJ8dzq7t27qFatmsI28noj+RGTFvmUra0tTpw4AXd3d4wetfNutgAAUE9JREFUPRpr1qxJNwRJLiYmBtOnT8fNmzfRs2dP2NjYqDnaVNqYaLl3756YhKhQoQLWrFmDzp07Z7juxYsXMX36dLx58wZbt26FtbW10smZvFSiRAm8e/cOjx8/Vnoe+uPHjwGkfpFJqXHjxrh16xZ8fHwU7tR27doV27dvx9atW9GqVSt07NgRMpkMrq6u2L17NwRBEJMc6rZ///7vmrMrZdKiTJkyePnyJb5+/SpZDLn1+fNnHDp0CFevXsWrV68QHR2t0JYuI4Ig4N69e+oJ8Aego6ODQ4cOYdCgQbh27RoSExNx8eJFXLx4Md268uJorVu3zraQGuVvHh4eAKRLNGujYcOG4dixY4iMjERCQgLOnj2Ls2fPwtTUFFZWVhg4cCDMzc2lDjOdq1evQhAEhdGT2YmLixO3k9quXbswb968TFvHa6JChQqhY8eO4gjayMhIXL58GZcvX4a3tzdevHiBsLAwHD16FMeOHQOQOs22ffv2WL9+vSQxFytWDNHR0ZInT3Kqdu3aePnyJR4/fqxwbdCkSRN4eXlh9+7dGDBggPhajo+Px+bNmwEgX3ZEZPeQH1xWc54SExOxdOlS+Pr6wsTEBP369UOTJk0U7oDduXMHx44dQ3h4OJo0aYJ58+ahQIECkrXBGjhwINzd3WFlZaVUouXw4cPo2bMnnJyc1BxpKnnHkxIlSuDy5csoW7Zsluu/f/8e7dq1Q1hYGIYMGYJ169apJ9A0hg4dipMnT6JevXq4dOlStndhkpKS0KFDBzx69Ai9e/eGo6OjmiJN79KlS+jfvz/KlSuHhw8figWMAgMD0bx5c/HEwcjICAkJCWI/eh0dHbi5uaFFixZqjzmnXSHkRbLkpLzbOHPmTOzatUsj2t3mxKVLlzB69GjxTpayJ5KaVI3+8+fPePDgAT59+oS4uLhsfwdbW1s1RZYxJycnbN++PdM56vXq1cOYMWMwaNAgNUdG2sTZ2RnTpk1DUlISfv31V+zevVvqkLSGPFmxf/9+eHp6KnQNAVIvoAYNGgRra2vJb0DI5aZbi7+/P5o1ayb557WXlxf69esn/lyuXDnUq1cPRkZGCsUVMyO/ONU07969E0dhnD17Fp8/fwYg7fdjq1at8PTpU5w8eVLtbZC/x/bt2zFr1iy0b99eTAABwIEDB8TRQs2aNUP//v0RGxuLw4cP4/HjxxAEAXPnzsXvv/8uYfTqx6TFD+7bdlGZkbeHUmaZVMPW5HfqcptoyYwqEzBNmzbFy5cvc3RRt3HjRixcuBDVqlXD7du3VRZbZk6cOAF7e3sIgoBffvkFmzdvzvTCOiIiAuPGjYO7uzsEQYCjo6PYV10KycnJmDdvHpKTkzFhwgSFrPuxY8cwbty4dKMCBEHA0qVLMX78eDVHq7zY2Fg8e/YM+/fvh4ODA5o3bw5nZ+cc1XdRhWfPnqFdu3YwNjbG1atXNeZENyvPnz9Hu3bt8PXrV8hkMujp6aFq1aowNjZW6kTy9OnTaogycz4+Pli6dCkuX76co2SLFJ/ZGQkLC4Ofn59Cy7+6desqFM+l/EOZ6XEpKSmIiIjA/fv3ERISAplMBl1dXZw/fx4NGzZUQ5Q/no8fP+LQoUM4cOCAWJNIfp5XoEABdOrUCXZ2dujevbukHUVyk7R49OgRWrduDQMDgwxrAqiLpaUlPD09UaRIEWzatAl9+/aVLJa8EhkZiStXriiMugCkbzG7aNEibNy4ETNnzsScOXMkiSE33r17hzp16kBXVxd3794VC57KZDL06tUL169fT3dtJpPJUKNGDVy6dEmh0Hx+wKTFDy6nd3GVIdUHU0b9upVNtGRG1SfzP/30E758+QJ3d3c0bdpUqW1u376Nrl27omjRoggKClJZbFnp0aMHbty4AUEQYGhoiL59+2ZYh+PEiROIjo6GTCaDhYWF5Bd02Xn16hWcnJzw6NEjJCcno1q1arCzs9O44kxZOXfuHAYPHoxGjRrh7Nmzks9HdnFxwZQpU1C5cmWsWrUKbdu2lTSe7IwfPx4uLi4QBAHTp0/HxIkTJU/+KOvo0aMYO3YskpKScjTMWOo7jkSZUfbGCvC/EVEGBgZYvXo1R+XkET8/P+zfvx9HjhwRi/vJ/01MTEwwYMAALF++XJLYcpO0cHV1xahRo1C2bFlx6qoUqlSpgqioKCxcuBBTp06VLI7vER8fn66+Rdp6HPL3ZMGCBdG8eXMcP35ckjjfvn2Lli1bQhAEeHl5oUqVKpLEkRuRkZFISUmBoaGhQqvv6OhozJ07F4cPHxZvtuno6KBHjx5Ys2YNSpYsKVXIkmHS4ge3f/9+lexXihoR2piAKVeuHOLj43HhwgU0atRIqW18fX3RqVMnFCpUCO/evVNZbFmJiIhA3759xaHcWSWHAMDc3BwnTpzQmos/bTdx4kTs378fK1euxIgRIySLQ36X9O7du3jy5AkEQUC5cuVQp04dGBsbZ5lQkaqIaN26dfHu3TvY29tjzZo1aj9+br19+xZNmzZFXFwcypYti8mTJ6Nw4cKYMmUKBEHA8ePHERERAV9fX7i4uODjx4+wsLDA//3f/0FHR0erhsxS/qHM97qOjg6KFi2KypUro23bthg+fLhWXZRoi+TkZFy4cAEuLi44d+6ceKGkzqTnt1Oae/XqJX5XVKpUKcttExIS4O/vj/Xr1+P9+/fo3LkzDh06pMpws1S+fHnExcXh4sWLWjMiKDk5WSyM7O3tjTt37mRYj0NXVxcNGzZE27Zt0bZtW7Ro0UKhVacUPD09YW9vDwMDAyxatAgDBgxAwYIFJY0pL8TExODFixdITk5G1apV8/V5NpMWpDW0MQFTv359BAUFYdWqVRg+fLhS2+zatQu///47KlWqJGnRv4SEBKxduxY7d+5UqGycVsmSJTFq1ChMmTIF+vr6ao4w//Lw8ICNjQ2aNWsGd3d3yeLIyV3SjEhx97906dJITEzEiRMn0KZNG7UfP7f++OMPrFu3DkWKFMF///2HChUq4MmTJ2jVqlW6i4ro6GiMHDkS58+fx9ChQ7F27VoJIycibSKTyXDs2DFMnz4dUVFRak1aZDSiFsj8xklG5CNtHRwcJJ2S0axZM7x48QIeHh5o0qSJZHHkRMWKFRETEyP+nDZRYWZmJnYMsbCwQLFixaQIMUPyacnBwcF4/fo1BEEQp34aGRllewNFE7rMUPbYPYS0hlQdQL5HixYt8ObNG2zYsAHW1taZFg6Vi4mJwcaNGyXtZiGnr6+PWbNmYcaMGXjw4AH8/PzEoo/GxsaoW7cu6tevL/n0hPyoTJkyADSj9ZU2VENPq3jx4ggLC0Px4sWlDiVHLl++DEEQ8Ntvv6FChQpZrmtoaIi9e/eidevWcHR0RJ8+fdChQwc1RUpE2uj58+dwcXHBoUOH8P79e8niyOg7JSffM0WLFsXkyZMlryHRoUMHvHjxAr6+vlqTtIiOjhYfV6pUSaGtqSZPR0jbLUZerDwhIUEcAZoZZaeSk2Zg0oJIhezs7HDo0CG8efMGffv2xfbt2zNtU/TixQuMGTMGgYGBkrZp/VaBAgXQqFEjpae3aILXr19j9+7dCu0sNbUvem7IR77Ex8dLGoeUnUtyq3bt2rh8+TICAwM1ssVfZl69egUgtUq6XNqTreTkZIUEooGBAcaMGYOZM2di3759TFoQUTqRkZFwdXWFi4sL7t69C+B/CQJdXV106NBBrZ2HNm3apPDzhAkTIAgC5s2bl2X3NUEQUKhQIZQpUwb169dHoUKFVB1qtsaNGwdnZ2ds2rQJAwcO1KiRCZmxtLQUExXZTcfRJBUqVGDyIR/g9BAiFZO3EBUEQZxbnlFRy6tXryIlJQUymQy//vorHBwcJIl348aNaNOmDRo0aCDJ8b/X/v37MWPGDHE+rja2s8zOsGHDcPz4cfz888/w8fGROhytIi/S1rdvX+zZs0fqcJRWsmRJJCcnw9PTU3xvvnr1Co0aNYIgCHj9+nW6k+I7d+6gS5cuqFChQqatRokof0lOToaHhwdcXFzg4eGRrmZBrVq1YGtrC2tra3FUn1RyU4hTk5w6dQojRoxA7dq1sWnTJtSpU0fqkIi0FkdaEKnYtm3bEB8fDw8PDyQnJ+Py5cu4fPlyuvXkJwxdu3bF1q1b1R2maOHChRAEAUZGRrCwsBCHBtaoUUOymJTl4+ODSZMmQSaTQSaToVSpUmjQoIHS7Sw1nb+/P9auXYvjx49DEAR07txZ6pC0zoABA+Dq6oqTJ0/C0dERQ4cOlTokpRQqVAgxMTEKI4bSJinevn2bLmkhXzezmjREUqtfv36e71MQBEnrQWmqBw8ewMXFBa6uruKoQvl5h7GxMSwtLWFnZ6dRRSNPnToFAFp11z+t3r17w8PDA9bW1mjTpg3MzMxQo0aNbFtVSlWomkiTMWlBpGIFCxbEwYMH4eLigs2bN4s90b9lbm6O8ePHw8bGRs0RpieTyRAREQE3Nze4ubkBSK2j0Lp1azGJIe8nrUk2btyIlJQU6OnpYdWqVRg8eLDGJyuUOWlPSUlBVFSUQoEsY2NjrW2jpg5ZtQtesmQJPn/+jGnTpsHd3R0DBw5EzZo1lep5LtXrvkKFCnj69ClCQkLE50xNTVG8eHF8/vwZt2/fhpmZmcI2jx49AgCFNmpEmuTNmzfpnpPPSc/uucxwmHh6FhYWePLkCYD/JSoKFCiATp06wdbWFj169NDIzwlt73r05s0b/PHHHwgLC4NMJsOTJ0/Ef4fsMGlBpIjTQ4jULCwsDI8ePRKnIpiYmKBOnTooUaKExJGlCggIENtdXblyRWHKRNqTQXn7OXkSw9TUVIpwFZiZmSEkJARTpkzBokWLpA5HKblp5VutWjXs2rVL42oyPH78GJ6enhkWbe3UqVO6i2pVUqazSU6LcElZ92T06NFwdXXFnDlzMHPmTPF5W1tbnDt3DmZmZrh48aI4lzs8PBydOnVCYGAgmjRpAg8PD0niJspKz549IQgCPnz4gICAAPGCumLFigpTKOXJDUEQULVqVXFZZk6fPq3awLVM2u+Z2rVrw87ODtbW1hpdXFHbhYaGolOnTnj79m2uClZrY80oIlVi0oKIsvTgwQN4e3vj8uXLuHHjBr58+SIuS3vBZ2Zmhnbt2uHvv/+WIkwA/2tneebMGbRo0UKyOHJCftKeFV1dXRQtWlRMFHXp0kWjRpC8efMGU6dOhZeXV5brdezYEWvWrEHFihVVHlNukkHZkbLuyd69ezFlyhQ0b94c586dE58/e/Ys7OzsIAgCKlWqhJ49eyI2Nhbnzp3D+/fvIQgCli9fjtGjR0sSN1F2vL29YW9vj8TEREybNg1DhgxJdzH98eNHODo6Yt26dTAwMICDgwPatm0rUcTap1q1ahgwYADs7Ow0Ltn9o5o3bx42b94MQRDQpUsXjB8/Hubm5jAyMuJooDyWm2lmgiCgYMGCKFasGKpVq4YWLVqgX79+MDQ0VEGElBeYtCAipSUlJeH27dviSAwfHx8kJiaKy6UuZikfaZG2WCGplp+fH3r37o2oqKhs7ybJa6WcPn0atWvXVmlcy5cvV8l+Z8+erZL9Zic8PFx8TXt5eeHnn38Wl8kLs6Y9EZb/W7Ro0QKnT59ma2LSSG/evEGbNm2QlJQENze3bD+37969i169ekFfXx9XrlzJtv0vpUpKSkKBApwRrk5NmjRBQEAAevXqhb1790odzg8to5sUyk4zS/u9WaRIESxevBgjRoxQTaD0XZi0IKJcuX//Ps6cOYOtW7ciOjpaHGovZdLC1tYW7u7u2LFjBywtLSWLI7+Ij49H06ZNERwcDCD1JG348OFo3ry5OHw7NDQUN2/ehIODA27dugUgdej3rVu3YGBgIFnsP5KkpCSsXbsWDg4OeP/+PYDUkzhbW1vMmzdPqVodRFKYPXs2tm3bhhkzZmD+/PlKbbNkyRKsWbMGY8eOxbJly1QcIVHulC1bFl+/fsWRI0fYclrF5CNWQ0JC8OLFCwCpyYiKFSuiVKlSAFLPRd68eSOeq1arVg0lSpRAVFQUXr58KXacEwQh3TRM0gxMWhCRUr6tdSGfb5m2p3vDhg1x/vx5yWI8f/48rK2t0b59exw7dkyyOPKLbdu2Yfbs2RAEAfPmzcOMGTOyXH/NmjVYsmQJBEHAihUrMGrUKDVFmn9EREQgKSkJJUqU4BBk0njyu9GnT59Gq1atlNrm+vXr6NmzJ6pVq4bbt2+rOEKi3KlVqxZCQ0Ph5eXFKTlqcPnyZdjb2yMpKQm///47Bg0alK7WWnh4OJydnbFy5Uro6elhz549aNOmDRISEnD06FEsWLAAYWFh0NXVxY0bN1C9enWJfhvKiOZMiiYijRISEoJDhw5hwoQJqFu3Lpo0aYIZM2bgxIkTCA8Ph0wmQ82aNTF69Gg4OzsjICBA0oQFAHTp0gVDhw6Fl5cXlixZImksyjI2NoapqSmePn2q9Db+/v7idlI6c+YMBEFA9+7ds01YAMD06dPRo0cPyGQyyQrlBQX9v/buO6yp8/0f+DuIFERk1b1QxIULrajFgVKlat24FbdoW62jtLbFVmutA79t1aLWVVfBLe4FKCCICnUiDnCBi42AIBDy+4NfzieBQE4wEKzv13VxXQln3Tk5Ocm5z/PcTyxiY2MhlUpFLyOVSoXlKjpzc3NUr16dCQt6J8hbBmnSGkg+77Nnz8okJiJtsLW1BQA8fvxYx5H898XGxsLV1VWoaTZ79myVv48sLCwwa9YsnDx5Ejk5OZgwYQKePn0KAwMDjBo1CseOHYOxsTHy8/OxdetWHbwSKgk7uBGREnd3dwQHB+PevXsAoNT/r379+ujRo4cwYoi82V15CwkJKXaai4sL7t27h99//x1nzpzByJEjRQ9n6eDgoM0wRStNZfG3WU5b5EO3jR07VvQyY8eOxYkTJ0QP+6Ztbdq0gZ6eHkJCQtC8eXNRyzx48AD29vbQ09PT2eghRP9FlStXRlZWFiIjI0XXIeJQvvQucHV1RUBAALy9vTFgwABdh/Oftn79eqSlpcHd3R2tWrVSO7+trS1mzpwJT09PrFu3DkuXLgUANGvWDGPGjMGmTZtw4cKFsg6bNMSkBREp2bx5s1Cs6MMPP0S3bt2ERIWVlZWuwwMAfPbZZ6LuJEdGRuLHH38UtU5dDmf5rkpNTQUAjYrh1a1bFwCQlpZWFiGJUpGTRD4+PmWy3tGjR5fJeoneRosWLRAWFgYvLy8MHz4cBgYGJc7/5s0beHl5QSKRlOsQykSaGjRoEFxcXHDgwAGsXLkS33zzja5D+s86e/YsJBKJRrVDevbsCU9PT5w5c0ZIWgAFo5xt2rRJqNVFFQeTFkSkkkQigY2NDZo3b47mzZujfv36ug5Jia5bGeiSfNhZIyMjncZRrVo1JCcnIz4+XvQyCQkJAPBODitWHl0uPv/8c61vRyKRMGlBFdKIESMQFhaGO3fuwMXFBevWrSs2CRobG4vPP/8cUVFRkEgkGDVqVDlHSyReSEgIxo8fj9jYWCxfvhwnTpyAi4tLhW/5+S6SdxXT5DeRfF55FzW52rVrA/jf7yyqOJi0ICIlXbp0QUREBHJycnDx4kWEhYVh+fLlMDY2xscff4zu3bvD0dFR6K+pC15eXjrbdkUgb7YoH6FDV6ytrZGcnIxDhw7hk08+EbWMvEBqkyZNyjI0rXr16hWA8ksSvc8JOXq/TJgwAbt378bly5dx4cIFtG/fHj169ECHDh2Uqv5HREQgMDAQeXl5AAB7e3u4urrqMnSiEhVuEXrjxg3cuHFD1LJs+akZeTezqKgo0d3M5F1UCw8FLK93ZWpqqtUY6e0xaUFESk6cOIGsrCxcvHhRGC3k5s2byMjIwJkzZ4Rim5aWlujevbtQ36I8u46MGTOm3LalbStWrFD5/y1btuDDDz8scdmcnBzcv38fp06dgkQiQceOHcsiRNF69+6Ny5cvY/fu3ejVq5faYWYPHDgAHx8fSCQS9OnTp5yifHsnTpwAoFk3mNI6evRomW+DqKLQ09PD3r17MXbsWISEhCA3Nxf+/v7w9/cvMq88mde1a1fs2rWLxWapwmMCunw0a9YMly9fxvr16zF8+PAiiYjC8vLysG7dOkgkkiK1reSFU3Vd6JyK4pCnRKRWamoqgoKCEBQUhMDAQGEcbOB/Tebr168PR0dHrF69WldhvhPMzc2VfmzLf9Ro8gNcJpNBX18fZ8+eFX1XoSykpqbCzs5OqE8xfPhwTJo0CR06dBCK5OXm5iIiIgJbt27F/v37IZPJYG5ujqtXr5bLnYwvvvhC6bm3tzckEgn69eundvvyJJH87tiECRPw+++/l1msRO+zXbt2YePGjbh586bK6a1bt4abm5tGhX+JdMXb2/utln+Xb86Ut82bN8Pd3V2oa+Hl5YVatWqpnPfFixeYNWsW/Pz8IJFI4OnpiSlTpgjT3d3dsXnzZgwaNAjbtm0rp1dAYjBpQUQae/bsmdAK4+TJk0LzeYlEguTkZB1HV7GZm5u/1fIffPAB7O3t4e7ujm7dumkpqtILCAjAyJEjlYYQ1dfXFxICaWlpQpNuebJl3759cHR0LJf4tJUkqlq1KgIDA9G4cWOtx0hE/5OYmIhbt24J3yUWFhZo1aqV2pZoRPR+kkql6N27N65evQqJRILKlSvD0dER7du3R/Xq1QEU1NP6999/ERgYiJycHMhkMnTo0AFnzpyBnp4eACA7OxutWrVCcnIyfv/9d0yYMEGXL4sKYdKCiDSSmpqK4ODgIq0uZDIZkxYiPHnyRHgsk8nQrl07SCQSHDx4sMQLYolEAiMjI5ibm6NSpUrlEapoly5dwsyZM/Hw4cMS57O2tsa6detgb29fTpEVTRLJkxVimu0aGRmhVq1a+PjjjzF79mw0bdq0TGIkIiKi0ktJScGoUaNw+fJlAMXfmJB/93fq1Ak+Pj5KvxGePn2KM2fOAAAGDhzILiIVDJMWRFSi7OzsIvUt8vPzhenyLwBDQ0N06tQJvr6+Oor03SRvCRAaGlqkb+W7JD8/HydPnsS5c+cQGRmpdJfU1tYWTk5O+PTTT3XeD/2/sr+JiIjof2QyGbZt24bNmzfj9u3bKudp2bIlpk2bhgkTJuj89whphkkLIlIilUqFSu2BgYEIDw9HTk4OAOW705UqVYKdnR26d++O7t27o3Pnzvjggw90FTaRKK1bt4ZEIoGvry+7ehAREf0HxcfHIzIyEikpKQAKbljY2toKoxLRu4dJCyJS0qBBA2RkZAjPFRMVLVq0EEYMcXBwQLVq1XQR4n+OvOWKvF9lYZs2bcL+/fuRmJiIRo0aYcaMGaKHGCUiIiIiepcxaUFEShT79zVs2FBpWFN5QSPSnjNnzmDUqFEwMjJCZGQkzMzMlKZ/88032Lx5c5HlCle81jWZTIYTJ04gICAAt27dEsaYt7S0RKtWrdCrVy/069ePzTGJiIiISCNMWhCRkqlTpwqJioYNG+o6nP+87777Dhs2bMCAAQOwY8cOpWk3btxAjx49IJFIIJPJYGZmhtTUVAAFo4hcvnwZDRo00EHUyoKCgvDVV18J45sDqkfpsLKywurVqyvEqCdyt2/fFhItis1IW7VqBScnJ7Ro0ULHERIRERG935i0ICLSoV69euHatWtYsWIFpk2bpjTt66+/xpYtW2BiYoLDhw/Dzs4OV65cgYuLC9LT0zFv3jx4eHjoKPIC+/btw8yZM5Gfny8kKgwMDIQWOykpKUJNFKCgFsqGDRvg4uKik3jlnjx5gjlz5uD8+fMlzterVy/89ttvFSI5RERERPQ+Ut2BmoiIykVCQgIAoFmzZkWmnTlzBhKJBBMnToSdnR0AoGPHjpg4cSJkMhkCAwPLNdbCYmJiMGvWLEilUkgkEri6usLf3x9Pnz7FnTt3cOfOHTx9+hT+/v4YP3489PT0IJVKMWvWLMTExOgs7lu3bqFHjx44f/48ZDJZiX8BAQFwdHQsthI5EREREZUtJi2IiHRIsUuComfPniE2NhYA0L9/f6Vp3bt3BwCdXvgDwNq1a/HmzRtUrlwZPj4+WL16Ndq3bw99fX1hHn19fbRv3x5r1qyBt7c39PX18ebNG/z55586iTk7OxujR49GamoqZDIZPvroI6xbtw4RERGIi4tDXFwc/v33X6xfvx729vaQyWRISUnB6NGj8ebNG53ETERERPQ+Y9KCiEiH5F0nsrKylP5/9epVAAW1Kzp06KA07cMPPwQApKenl0OExTt37pzQEqRPnz5q53d2dhZaiQQEBJRDhEVt374dcXFxkEgk8PDwwNmzZzF69Gg0btwYxsbGMDY2RqNGjTBq1CicPn0aCxcuBADExsYWqTlCRERERGWPSQsiIh2St7B49uyZ0v+Dg4MBAO3atVNquQAAubm5AABjY+NyiLB4L1++BFC0JUhJBgwYoLRseTtx4gQkEgn69u2L+fPnq51/3rx56NevH2QyGY4dO1YOERIRERGRIiYtiIh0qHnz5gCAw4cPC//Ly8vD0aNHIZFI4ODgUGSZp0+fAgBq1KhRPkEWo1q1agBQZJjWkpiamiotW96ioqIAAGPHjhW9jHxe+bJEREREVH6YtCAi0qG+fftCJpPh8OHD+OGHH3Dy5ElMmjRJaHkxcODAIstcv34dAFCnTp1yjbUwecJFk9oa8nl1NZSofMjYevXqiV6mbt26AIC0tLSyCImIiIiISsCkBRGRDk2YMAENGjSATCbD+vXrMXbsWBw/fhwA0KdPH7Rp06bIMqdOnYJEIkH79u3LO1wlo0ePhkwmw6ZNm0TNL5PJsHHjRkgkEowZM6aMo1NN3sIjPj5e9DLyEV5MTEzKJCYiIiIiKh6TFkREOmRkZARfX1906NBBaahNJycnbNiwocj8//77r9BNoUePHuUdrpLRo0fD2dkZYWFhmD59OjIyMoqdNyMjA25ubrh06RL69euHkSNHlmOk/2NtbQ0AOHTokOhl5PM2adKkTGIiIiIiouLpq5+FiIjKUqNGjeDn54dHjx4hPj4ederUKbb7gr6+Pry8vABAZb2LshASElLstBkzZiAxMRH79+9HQEAAhgwZgo8++kiotxEfH4/w8HAcOnQIycnJ6NixI6ZNm4aQkJByi19R7969cfnyZezevRu9evXCsGHDSpz/wIED8PHxgUQiETVCChERERFplyQ1NVWm6yCIiKjiMjc3h0QiUTufTCYrdr7C0yQSCZKSkrQWo1ipqamws7MT6lMMHz4ckyZNQocOHVC5cmUABaOzREREYOvWrdi/fz9kMhnMzc1x9epVoZAoEREREZUPJi2IiKhE8mFZtUkikSA5OVnr6xUjICAAI0eOhFQqFf6nr68vJCTS0tKQl5cHoCDZoq+vj3379sHR0VEX4RIRERG915i0ICKiEnl7e5fJenVVjBMALl26hJkzZ+Lhw4clzmdtbY1169bB3t6+nCIjIiIiIkVMWhAR0XspPz8fJ0+exLlz5xAZGSm0/LCwsICtrS2cnJzw6aefiuoaQ0RERERlg0kLIiIiIiIiIqqQOOQpEREREREREVVITFoQERERERERUYWkr+sAiIiIytujR4+wdetWXLhwAQ8fPkR6ejry8/NLXEZXw7QSERERvc+YtCAioveKt7c35s+fjzdv3gAoGNaUiIiIiComJi2IiOi9ERERgVmzZkEmk0Emk6FGjRpo164dzM3NoafHHpNEREREFQ2TFkRE9N5Yu3Yt8vPzUblyZaxatQrjxo1jsoKIiIioAmPSgoiI3huXLl2CRCLBF198AVdXV12HQ0RERERq8PYSERG9N5KTkwEAzs7OOo6EiIiIiMRg0oKIiN4bFhYWAABDQ0MdR0JEREREYjBpQURE74127doBAGJiYnQbCBERERGJwqQFERG9NyZPngyZTIZdu3bpOhQiIiIiEoFJCyIiem/07t0bEyZMwPnz57FkyRJdh0NEREREakhSU1Nlug6CiIioPISEhEAmk+HXX39FWFgYbG1tMXLkSDRr1gxVqlRRu7yDg0M5RElEREREckxaEBHRe8Pc3BwSiaRUy0okEiQlJWk5IiIiIiIqib6uAyAiIipPMhlz9URERETvCiYtiIjoveHl5aXrEIiIiIhIA+weQkREREREREQVEkcPISIiIiIiIqIKiUkLIiIiIiIiIqqQmLQgIiIiIiIiogqJSQsiIiIiIiIiqpCYtCAiIiIiIiKiColJCyIiIiIiIiKqkPR1HQAREZE2BQUFISQkBABgYWEBNzc3HUdERERERKXFpAUREf1nPH78GOPHj0daWhoMDAxw8OBBXYdERERERG+B3UOIiOg/IS8vD1OnTkVaWhoAYPXq1ejatauOo/qfZcuWwczMDGZmZpg5c6auw6H3XHBwsHA8tm7dWtfhEBERFYstLYhI5+7cuQM/Pz8EBgbi8ePHSEpKQlpaGkxMTGBhYQFbW1t07NgRAwYMgJWVla7DpQpq6dKluHLlCgDA3d0do0eP1nFERERERPS22NKCiHTm+vXrGDFiBDp37gwPDw+cPXsW9+7dQ1JSEvLy8pCSkoKYmBgcOXIECxcuhJ2dHfr164fg4GBdh/6f0bp1a+Fu67u8X8+fP48//vgDAODi4oLvv/++zLbFO9REVFZmzpwpnF+WLVum63DKBM+hRKQpJi2ISCfWrl2Lnj174syZM0r/Nzc3h62tLbp164a2bduidu3awjSZTIbQ0FAMGDAAq1evLu+QqYJKSEiAm5sbZDIZOnfuDC8vL0gkEl2HRURERERawO4hRFTu3N3dsWnTJuG5gYEBJk+ejFGjRqFt27ZFLjgfPXqEU6dO4a+//sLDhw8BAM+fPy/XmKlikslkmDFjBl6+fAkrKyv8888/+OCDD3QdlkrfffcdvvvuO12HQQQA6NatG1JTU3UdBhERkVpMWhBRudqxY4dSwqJJkybYu3cvGjduXOwyVlZWmDFjBqZMmYJt27bhp59+Ko9Q6R0gkUhw4MABXYdBRERERGWESQsiKjcPHz7Et99+Kzy3srLCqVOn8OGHH4pavnLlypg2bRq6dOmCkJCQsgqTiIiIiIgqCElqaqpM10EQ0fth7ty5+PvvvwEAenp6OHbsGD7++GOtbyc6Ohp+fn4IDQ3F3bt38fz5c7x+/RomJiaoVasW7O3tMWTIEDg6Oopan5mZmfD4+vXraNiwIZKTk7Fnzx4cOXIET548wcuXL5GXl4egoCC0adNGmF8mk+Hy5cs4d+4cwsPDce/ePSQmJiI3Nxfm5uZo0KABPv74Y4wbNw5NmzbV+LWmpaVh37598Pf3x+3bt5GcnIzs7GyYmZmhSZMm6NSpE/r27YtOnToJywQHB2PAgAGit+Hg4IDjx48XOz0lJQV79uyBn58f7t69i8TEROjr66NmzZro0qULXFxc0KNHD7Xb6d+/v5CM8vLywtixY5Gbm4sjR45g//79iIqKwsuXL5GVlYVff/0Vn3/+eZHXU79+fdy8ebPYbeTn5+PUqVPw9fXFtWvX8OLFC2RmZsLQ0BDm5uawsrJC27Zt0a1bN/Tq1QsGBgbCsjNnzoSPj4+ofab4GuSWLVuGFStWAABGjx6N9evXq12HVCrF0aNHcebMGYSHhyMhIQHp6ekwNjaGlZUV7Ozs8Mknn8DZ2RmVK1dWuY7s7GycP38egYGBuHbtGh48eIDU1FTo6enB3NwczZs3R48ePTB+/HhYWFiIfn3Xrl2Dt7c3rly5gkePHiE9PR36+vowMTFBgwYNYGtri86dO6Nv374wNzcXvV5VHj9+jLZt2wrP5d0a4uLisH37dpw6dQpxcXF4/fo16tevD2dnZ8yZMwfVq1dXWk9ubi58fHxw4MAB3Lx5E+np6bC0tETHjh3h5uYmeojcxMREnD17FhcuXEBkZCQeP36M9PR0GBkZwdLSEnZ2dujduzeGDx9e7PtS2Nscm6Uh9nOj6nOZk5ODgwcPYs+ePbh37x4SEhJgamoKe3t7zJw5U+V+DA8Px5YtWxAaGornz5/D0NAQNjY2GDp0KKZPn652P6mK482bN9i/fz/27t2L6OhoJCYmwtLSEq1bt8aoUaMwePBgjerbvHnzBnv37sXJkydx69YtJCQkQF9fH9WrV0fHjh0xcOBA9OvXT+06i9u30dHR8Pb2hp+fH549e4akpCSYmJjgyZMnSt8zYqjq2vPq1SsEBAQgKCgIN2/exMOHD5GWlgYDAwOYm5ujdevW6NWrF8aMGQNjY2O121A853377bdC97Zz585h586duHr1qvBeNmnSBP369cP06dNRtWpVtesTo/A5tLBr167h4MGDCAoKwrNnz5CSkgIzMzNYWVnByckJrq6uqFOnjqhtJSUlwcfHBwEBAYiKikJKSgry8vJQpUoV1KlTB9bW1ujYsSN69+4NW1tb0a+BiLSDLS2IqFzIfxDI9e7du0wSFo6Ojrh27ZrKaSkpKUhJSUFUVBS2b9+Obt26YevWrUUubNQJCAjAjBkzEB8fX+J89+7dw5AhQ/D06VOV0+Pj4xEfH4/w8HD8+eefmDp1Kn799Vfo64s7NW/YsAHLly9X+eM1ISEBCQkJuHjxIv744w/MmDEDy5cvF7VeTWzcuBG//vqryhjS09MRHR2NnTt3ok+fPvjrr780unh9+PAhJk+ejKtXr2ol1qdPn2LChAkIDw8vMi0zMxOZmZmIi4vDhQsX4OXlhSlTpuD//u//tLLt0jh//jzc3d1x//79ItPS0tJw/fp1XL9+Hdu2bUOrVq1w4cKFIvPt378fc+fORXp6usptZGVl4dmzZwgICICnpyeWL1+OcePGlRhXTk4O5s6di3/++afItLy8PGRnZyMhIQERERHYsWMHmjVrhkuXLol81eIdPnwYX375ZZHXFh0djejoaOzduxfHjh1Ds2bNABTUxhk7diwiIyOV5n/x4gWOHj2Ko0ePYsGCBViwYEGJ212wYAE2b96MvLy8ItPS09ORnp6OR48e4dChQ1i+fDm2bduG9u3bl7jOd+nYjIuLw8SJE4vEmpCQgOPHj+P48eNYsmQJZs2aBaAgGfP9999jw4YNSvPn5OQgIiICERER2LdvH3x9fTW6cH/8+DFcXV1x/fp1pf8/e/YMz549w+nTp/H3339j69atolrzBQUFYdasWXj8+HGRaenp6Xjw4AH27NmDjh07YsOGDbC2thYdK1BQfPrnn39Gbm6uRstpsv6lS5ciOzu7yLTc3FzhGDp58iSWL1+ODRs2oHfv3hptIyMjA3PnzsW+ffuU/p+dnY3w8HCEh4dj69atOHjwYKmS8GIlJyfj66+/xsGDB4tMk3/3XblyBWvWrMEPP/yAL774osT1+fr6Ys6cOcUmgl69eoU7d+7g+PHjWLRoEfz9/dGhQwdtvRwiEoFJCyIqFwEBAUo/plxdXctkO4p3C/X19WFtbQ0LCwtUrlwZSUlJuHfvnvCjMTg4GJ9++inOnz8PExMTUesPDw/HjBkzhHU0bdoUNWvWRHJycpGLy6SkJKWEhbGxMRo3bgxTU1PIZDI8f/4cDx8+hEwmQ35+PjZu3IikpCRs2bKlxBjy8/Px5ZdfwtvbW+n/5ubmaNy4MapUqSLEk5OTA6DgIldxPicnJwBASEiI8L60b99eZVJB1V0lmUyG+fPnY+vWrUr/b9SoEerUqYO8vDzcuXNH2O6ZM2fQt29fnDx5UlTiIiUlBQMHDkRsbCwAoG7dumjYsCHevHmj8iJenaysLAwcOBAxMTHC/6pUqQIbGxuYmZkhJycHycnJiImJES5G8/PzldbRsmVLODk5ISUlBf/++y8AwNDQEA4ODiq3KfYOnyrbt2/HvHnzIJVKhf8ZGhqiadOmMDMzQ0ZGBqKjo/Hq1SsAyu+voocPHypd1H/44Ydo0KABTExM8ObNGzx8+BAvX74EUHBB8uWXXyI3NxeTJk0qNrYvv/wSe/fuFZ5XqlQJNjY2QvIvNTUVMTExeP36NYCi+1Eb/P39MWnSJOTn58PQ0BAtW7ZElSpVEBMTIxTpTUhIwODBgxEWFoacnBx89tlniIuLAwBYW1ujTp06SElJQWRkJGSygkany5cvR7NmzTBkyJBitx0VFSUcIxKJBPXr10etWrVgZGSEjIwM3Lt3T9jnT548Qf/+/XH69GmlFliKtHFslpeMjAwMGTJE+Aw2adIEtWvXLrIfFy5ciPr162Pw4MH4+uuvhfOEubk5mjZtCj09PURGRgrH77Vr1zBt2rQiF8Pq4njw4AEAoF69emjYsCHS09Nx+/ZtYT8FBQVhyJAhOHbsGExNTYtd34kTJzBx4kThfAkAlpaWsLGxEc5lGRkZAIArV66gb9++OHz4MFq0aCEq3rVr12LhwoUACgpPt2jRAiYmJnjx4gUSEhIAQDgn3759WziGGzVqVGK9J0X37t1T+o6tXbs26tatC2NjY2RlZeH+/ftISUkBUPDdNGrUKPj4+KBPnz6i1i+VSuHq6oqAgAAAQI0aNYTEjeJ7GRcXh6FDhyIsLKxIiwttnEOfPHmCoUOHIjo6WvjfBx98gObNm8PU1BTJycmIioqCVCrF69ev8cMPPyAhIQGLFi1SuY2AgABMnjxZ6TNVu3ZtNGzYEIaGhsjMzERsbCxevHghTNfV54/ofcakBRGVi4sXLwqPJRIJunXrVibbMTU1xfDhwzFgwADY29sXaUKdkZEBb29v/PLLL3j16hViYmKwaNEi0Xct58yZg9zcXAwdOhSLFi1CgwYNhGmJiYlFttewYUO4urqib9++aNGiRZFmxU+fPsXatWvx119/QSaT4cCBA/jss89KvGhasWKFUsKibdu2WLx4Mbp37w49vf+NZJ2Tk4PAwED8888/qFSpkvD/Vq1aCcUrW7duLSQGFi9eLPp9Wbt2rVLCYtKkSZg7d67S/pBKpTh06BDc3d2RkpKCO3fuYPbs2di5c6fa9a9YsQLp6elo27YtPD09YW9vL0zLzs5GcnKyqDjltmzZIlwUmpiYYOXKlRg2bFiR9+vNmzcICQnB/v37i0ybNWsWZs2apdT0u3r16lovBBoYGIi5c+cKP4xr1qwJDw8PDBs2DFWqVBHmk8lkuH79Ovbu3Ytz586pXJdEIkGnTp0watQo9OnTB3Xr1i0yz/Xr17F48WLhYuT777+Hk5OT0nspd+3aNaWExbx58zB79uwid8jz8/Nx48YNHDlyROmzry3Tpk2DRCLB999/jy+++EJo6i6TybB7927Mnj0bubm5eP78Of766y+Eh4cjLi4O3bt3h6enp9D6AgDu37+P8ePH486dOwCAH3/8EYMGDVL6LCmqXLkyBg8ejCFDhqBnz56oVq2a0nSpVIozZ87Aw8MDMTExyMrKgpubG0JDQ1V2K9DGsVleVqxYgeTkZHTv3h2rVq1SupseHR2N8ePHIyoqCgDwyy+/QCqVYuvWrbC0tISnpycGDRoknIuys7OxcOFCoTDz2bNnERgYKKor2cqVK5GUlISGDRtizZo1SsskJibip59+EloC3bx5EwsXLsSaNWtUruvx48dwc3MTEhaWlpZYuXIlBg8eLMSalZWFzZs3Y8mSJcjJyUF8fDwmTZqE8+fPw9DQsMRYExMTsXjxYujr6+Obb77BzJkzlZLk8hGx5OcRxS4UI0aMED3ikJ6eHj755BO4uLigd+/esLS0LDJPSEgIPDw8cPXqVUilUnz++ee4du1asd05FG3duhXJycmwsbHBypUr4ejoKBzPubm5+OOPP7B06VIABYmLP//8s0irpbc9h+bk5GDcuHFCwsLS0hI//fQTRowYofQ+JCUlYdmyZdi8eTMA4I8//hC6Shbm4eEhnGvt7e3h6emp1BVN7uXLlzh9+jS2bdsmKlYi0i7WtCCictGzZ0+hmb+NjQ2uXLlSJtvJzMwU1Vc3IiICzs7OyMvLg6GhIW7fvl1sf/7CF2QTJkzA6tWr1W4jKysLH3zwQbEXP4q8vLzwww8/ACho8SC/gCwsMjIS3bp1E35kOTs7Y8eOHWqH+czIyFD5w1QxaXH06FFRSYuYmBh07txZaG2yZs2aElvOREVFoXfv3sKdylOnTqFz585F5lPssw4Abdq0wYkTJ9T+oBbTN3/gwIEICgoSFa+cVCpVSvZosj1VxNS0yMnJgZ2dndBCx8rKCkePHkX9+vVLXHdx76/Yz0N+fj5cXV1x7NgxAMDs2bPx888/F5lv5cqV+PXXXwEUXFBt3LhR7bqL24+aKFzTAgDWrVuHMWPGqJxfcV8bGBggJycHPXv2xN69e1XWTrh//z66dOki3KEv6bMgdp8mJyejZ8+eQneDvXv3qryrrc1jUxOlqWkBoMT9GBMTg06dOgn70cDAAJUrV4afn1+xrRL69u0rJLZKqvVSOI6aNWvC398f9erVUzm/Yg0liUSCwMBAla1dXF1dceTIEQAFSaOTJ0+iVatWKtfp6+uLiRMnCs9//PFHzJs3r8h8quoGbdy4ESNGjFC5XkXF1ZFQR+xxmZ2djc8++0zo3rNq1SpMnTpVbSxAQcuas2fPFtta7uuvvxYSBSUdU6U9hy5fvlzo5li3bl2cOnWqxHPjihUrsGzZMiH2K1euKCUO4+LihPe6atWquHXrlqguStr4/BGRZtT/kiYi0oLExEThcXE/MrVBzI82AOjQoQOGDh0KoOBHXHFJgsJq1KghXLSpY2RkJCphAQCff/65sF/+/fdfpaaoitasWSMkLGrVqoW//vpLbcICgKg7aWKtW7dOSFi4uLiovchq0aIF5s+fLzyX/6hVZ/Xq1VqL+9mzZ8JjVQkTVXTxo3TPnj1CwkIikWDTpk1qExZA8e+v2M+Dnp4elixZIjw/ceKEyvkqyn6UFxMsjuKFZU5ODvT19bF27dpiiz3a2NgoJSnCwsKKXbfYfWphYaF03Ff0fSqGvr4+1qxZU+x+tLa2Rvfu3YXnOTk5mD9/fondKCZPniw8Lmm/F7Z48eISv0t++eUX1K5dG0BBCxxVd8ifPn2qVGT4m2++KTZhAUBoYSO3detWpS5cxXF2dhaVsHgbYo9LQ0NDoasKUPxxqcoff/xRYvc+xdoRsbGxSsf228rKylJKkq5du1btufGbb74RWlVFR0cXaZGmGF/z5s1F11RhwoKo/DFpQUTlQt6XFkCR5tS60rFjR+FxRESEqGVcXFxE/zjUhEQiUSrspSqe3NxcHD16VHg+Y8YMjSvOv638/HylfufqCpzJKf5gDw4OVjt/mzZtYGdnp3mAxVBM7Ny6dUtr69U2xWbSTk5OSsdoWWvUqJHQpDwmJkZlUbqKsh/VJcrk/fnlevXqpTZZqvj5u3fv3tsF+P+JOcdUlH0qRq9evdReKBYuOjp+/PgS5//oo4+Ex48ePVKqK1EcS0tLuLi4lDiPsbGx0sgT8lZEik6dOiUkHYyMjDBhwgS125aPWgQU3KkXUyhYzHrLk+I+l9eWUMfGxkbt6DqNGjVCrVq1hOfa+hwBBXWR5N0CW7RogV69eqldRiKRKB0n8hZNcoqfPXlXLiKqmFjTgojKxZs3b4THYloGvK28vDwEBwfj6tWriI6ORnp6OrKysoRCcQCEYmeFH5dE7J3QwuRD0d28eROxsbFIT08v8uP89u3bJcZz7do1obghUNCsvLzdunVLKLhWrVo10YmFunXrwtTUFGlpaXj58iWeP38u3AVVpbT7uTh2dnbCqBHu7u4wNDRE3759NRoOsazl5eUpdZvS9vsbGxuL8+fPIzIyEomJiUhPTy8yAkZmZiaAgjvTL168KJIUU3y/t23bhsaNG2Py5MlKtTbKg+JFV3Fq1qwptFoRM7/ixVZxhU0Lu3btGsLCwnD37l2kpKQgMzNTqUif4kVQceeYd+HYlNN0P1pZWakdnalmzZrCY5lMhlevXqkd7aNHjx6iRlnq3bs3Vq1aBaBgtKbY2FilpIviCCidO3cWlVDv2LEjLCwshAvo8PBwtfulS5cuaterTffu3cOFCxcQFRWFpKQkZGRkFNsiJDU1Fa9fv1b7GVasK1SSOnXqCC0FVSU+S0uxNo6YuidyLVu2FB4XHmmmWbNmqFKlCl6/fo2UlBSMHz8enp6eaNSo0dsHTERaxaQFEZULU1NToYuI/KK3LEilUmzcuBG//fabUJVdDLExafpjJi0tDb/88gt27dql0V0cVfHcvXtXeFytWjXRVeW1STGxIpVKMWzYMNHLKiauEhMTS0xaaPtH47Rp07B7927k5eUhKSkJY8aMQe3ateHk5AQHBwd06dIFVlZWWt2mpuLi4oSkAQCVxeBKIzo6Gt999x38/PyUknbqqDoGBw0ahCVLluDp06fIz8+Hh4cHli9fDkdHR3Tt2hWdO3dGmzZtRHeLKq0aNWqoncfIyKjU8ysmB1Xx9/eHh4eHUHRSjOLOMe/CsSlXFvu98MWyun0PQPSoHYXne/DggVLSQl4EE1C+uFWnZcuWwhDD8hFMimNqaqrRUM9v499//8V3332n8RDDr169Upu0UEwulUSTz5EmFL97/Pz8RH/3KLbyVOymChR0lZk8eTL+/PNPYb3t27dH+/bt4ejoiC5dusDe3r7CtA4lep8xaUFE5cLc3Fz4waDNuy+K8vLyMHnyZKGomiYUL6hLInZoVKCg2vhnn31WqiE6VcWj+ONL3Z3IsqI4akdmZib8/f1LtR51SSJt1uAAChIA69evx6xZs4RhAZ8/f45du3Zh165dAAoKwvXv3x+urq4aXcBoS+HPhbo71GJcvHgRLi4uSskQsVQdg0ZGRti9ezdGjx4tDB+akZGBY8eOCc3vzczM4OTkhDFjxgjDOGqbpqNnFFeDoTglJXcUi+ZqorhuD+/CsSmn6X7UdH6g5H0vV1zR5MKqVauGypUrCzV4Cn/GFFvUiF0nAKWROdR9n2nynfE2fH19MWXKFFE1NgoT8/1XVu+lWIrfPdHR0UpDnoql6nvnxx9/RGxsLA4fPgygIOaIiAihO1elSpXw0UcfYejQoRgzZky5vZ9EpIw1LYioXCjeKYyKitLqjxk5Ly8vpYRFhw4dsGrVKpw7dw7379/H8+fPkZKSgtTUVKSmpsLLy0vjbWjSZPvLL78UEhZ6enoYNGgQNm7ciNDQUDx69AgvX74UYklNTcXo0aNLXJ/iRU95dLFRRVt3ztSNc18Wd+qHDx+Oy5cvY/r06SoTArGxsdiwYQMcHBwwe/bscu/fXPjC4W3f4/T0dEyYMEFIWFStWhXTp0/H7t27ERERgdjYWCQmJiodg2KKfrZu3RphYWFYtGiR0pCXcqmpqThw4ACGDRuGvn374smTJ2/1OiqS8PBweHh4CM/r1auHH374AceOHcOtW7fw9OlTJCcnC/uzcHP04lT0Y7Oi0eQCWjHBVThxpPiZ0yQRVtI6CyuPbj5PnjzBjBkzhISFpaUl5s6di0OHDuHatWuIi4tDUlKS0mf9XaON7x5V3zsGBgbYvn079u7di969exc5tqRSKS5duoRvv/0Wbdu2xcGDB986DiLSHFtaEFG56NKlC86ePQug4O7W/fv3VV7wlJZUKlUahnTq1Knw9PQs8QejfAjOsnDz5k3h9QLAli1blKrOlyYeU1NT4bHYPvfapthMtiyHri0rDRo0wMqVK7FixQpERkYiJCQEISEhCA4OFlqyyGQy7NixA0lJSfjnn3/KLTbF9xcoeI8V7+hqateuXYiPjxfW7efnBxsbmxKXEfuZqFq1KubMmYM5c+YgNjZW2I9BQUHCEJ9AQUuPAQMGIDAwsNyLxpaF33//XUi4dujQAb6+viXeedXkHFORj82KRux+zc/PV2plVPi9Ujyfpaeni96+4ryFP7e6sH79eqGVjpWVFU6fPl1idw5NXmtFofhe/fTTT5g7d65W19+nTx/06dMHmZmZuHTpEkJDQxESEoLLly8LyaDk5GRMmTIFBgYG+Oyzz7S6fSIqGVtaEFG5KFx1XHGUBG24fv260Hy0SpUq+Pnnn9Xe4dLmcGyFKQ6t5uDgoDZhISYexf7hL1++LFWT/7elGIMmNUMqGolEglatWsHNzQ07duxAdHQ09u7dC1tbW2Ge48ePKxV/K2uFLzJiYmLean2Kx+CMGTPUJiwyMzNLlQyrX78+Ro0ahbVr1+L69esIDAzEgAEDhOmPHz/Gpk2bNF5vRSOTyXD+/Hnh+aJFi9Q2FS/NOaYiHpsVjWJiTJP5CrdiUexm9+jRI9HbV5xXV131FCl+1t3d3dXWnyjL776yUl7fPcbGxujVqxc8PDxw8uRJ3L9/H4sXLxZqfshkMqUhY4mofDBpQUTlwt7eXqkv9s6dO7XaxDk2NlZ4LK8Irs7ly5e1tv2S4ik8BKAqr1+/VjvcoeLwiVKpFGFhYaUP8P9T7IYhpsuOYpX81NTUUtXrqIgqVaqEPn364MiRI0qtG1TV7NB0n4llbm4Oa2tr4XloaOhbrU9ecwIQdwyGh4er7bYjRtu2bbFjxw5069ZN+F9AQMBbr1fX5KODyInZp5oWRFRFk2PzfSF2mE7FYWb19fWL1ANRLHarOJJISQqf99q0aSNqObFKc37R9LOujePybZTmNSp+95RnCz8LCwt89dVX8PT0FP738OFDpSKuRFT2mLQgonIza9Ys4fGzZ8/w888/l3pdhe9CFx66Uczy2rjoL46m8ezfv19tMbTatWsrVcP/+++/SxWbImNjY+GxvHlxSerVq4fmzZsLz/9rTdQtLS3RqVMn4bmqO3qa7jNN9OrVS3js4+MjukCsKvLig2Jp872USCTo27ev8FzeTeVdpun+lEql2LNnj9a2L+bYfF+Eh4eLam2xf/9+4bGdnZ3SyBYA8PHHHwuPHzx4ICoZcvDgQaG7gJ6entaHZ1ZMuIs9v2h6bHp7e2s0v7aV5hz6ySefCI/Dw8Nx584drcdVkv79+ys9/y+c04jeJUxaEFG5GTFiBBwcHITnGzZs0LjZeFZWFtzc3Iosp9gcNioqqsRm7jKZDN9++22ZFANVFY+65Ehqaip+/fVXUeudPn268PjYsWM4ffp06QL8/xSb3Kobuk/u888/Fx5v3LgRN27ceKsYyoMm77Vif29VQxUq7rOkpCSt1heZNm2a0K3pxYsX+OWXX0q9Lk2OwbCwMKULvOJocz++aywsLJSK9Knbp2vWrFFqcVWc93mflpZMJsPSpUtLnOfy5ctK58cxY8YUmcfR0RH16tUTni9atKjE9yM9PV3pjruzs7PooUDFUlyf2HOyJp/1AwcOlGnCXozSnEPbt28vJIhkMhnmzZuncbKmsNJ+9oD3+/NHpAtMWhBRualUqRK2bNki9CuWyWRwd3fH7Nmz8fLlyxKXlUql2LVrFzp27Kjy7mWHDh2Eu2jZ2dlYsGCByqbuOTk5mDNnDvz8/LTwioqnWMPjypUr2Llzp8r54uPjMWzYMLx48ULUekePHq3U0mHSpElqh3i9ffs29u3bp3KaYvPof/75R9SPx9GjRwtNol+/fo2hQ4eKav7/8OFD/PDDD1i1apXaebXNwcEBe/bsUdsl6cSJEwgJCVFarrA6deooHcMbNmzQWpxNmzbFuHHjhOdr167F0qVLS2y5k5iYiHXr1hX5v+IxuGnTJty8eVPl8hERERg7dqyoriFTpkzBb7/9pvYu/8OHD7Flyxbhuar9+K6pXLmyUkuHhQsXFjsKw44dO7BkyRJR69Xmsfk+2bt3L37//XeV06KjozFhwgThorRevXoYMWJEkfkqVaqkVNAxKCgI33zzjcrP26tXrzBu3Dg8f/4cQEEri3nz5mnjpShRPCefO3dOVIsCxc/6ypUrlbqLKDp9+jS+/PLLtw/yLZX2HPrzzz9DX79gDIHQ0FCMHDlSbYsHqVSKgIAADB06tEh3xr1792Lq1KlqW9jk5eVh0aJFwvNatWqhSZMmomImIu3g6CFEVK5q1aqFY8eOYcSIEULz3h07dmDfvn1wdHQU7nyZm5vj1atXePnyJcLCwnD27FkkJiYWu14jIyNMnDgR69evB1DQtP7+/ftwdXWFtbU1srOzcfPmTezYsQMPHjyAvr4+Ro4cWWbdGxwcHNCmTRuhFcKsWbNw7tw5DBw4EDVr1kRKSgpCQ0OxY8cOvHr1CnXq1EGrVq1w5syZEtdraGiIv//+G3369EF6ejpev34NV1dXdOnSBYMGDYKNjQ2MjIyQnJyMW7duwc/PD+Hh4Rg9ejSGDx9eZH3Dhg3DmjVrIJPJcPPmTbRs2RJt2rSBmZmZcMe/ZcuWSsM8Vq5cGTt37oSTkxMSExORmJiIoUOHokuXLujbty+aN28OExMTvH79GvHx8bh58yYuXLgg7Is5c+ZoaS+Ld/v2bbi5uWH+/Pn45JNP0KFDB1hbW8PMzAxSqRSPHz/G6dOncezYMaXRIRS7ayhycXERjrVly5Zh586daNasmdIwpW5ubujRo4fGsa5YsQIRERG4ffs2AMDT0xMHDhzA8OHD0a5dO5iZmSEjIwN3795FcHAwAgICUKtWLaUWMAAwceJErF69GllZWcjIyICzszMmTZqE7t27w8TEBM+fP8fp06dx4MABSKVSODk54e7du8Ve8AAFBWAPHjyIpUuXomvXrujUqRNsbW1hYWGBSpUq4eXLlwgNDYWPj48wwkO1atWUWgi9y2bOnIkLFy4AKDimunbtimnTpqFdu3aQSCR48OAB9u3bJyQXJkyYgO3bt5e4Tm0fm++DIUOGwNfXF4sXL4afnx9GjRqFRo0aIT09HUFBQdi+fbswRKZEIsH//d//KXVJUDR58mQcP35cSLxu2rQJoaGhcHV1RfPmzZGXl4erV6/i77//xtOnT4Xl5syZo1RnSFt69OiB6tWrIyEhAZmZmcJ3SY0aNVCpUiVhPsUuHm5ubvDx8YFMJsPz58/RvXt3TJ06FZ07d4aBgQFiY2Nx+PBhnDp1CoC447KsleYcam9vj2XLlsHd3R1AQa2ctm3bYvDgwejatSvq1q0LAwMDpKWl4dGjR7h27Rr8/f2F3w6FW1bk5eVh//792L9/Pxo3bgwnJye0bdsWtWvXhrGxMdLT0xEZGQkfHx/cvXtXWG7u3LllMiw3ERWPSQsiKnfNmjWDn58fvvrqK5w4cQJAQbePkydP4uTJk2qX79mzJyZOnFjk/x4eHggNDcX169cBFPR7VVVcrVKlSli+fDmMjIzKLGkhkUiwadMmODs7C3djDx48qHKMdzMzM2zfvh1bt24Vte4WLVrgxIkTGDlypFAF/uLFi6UaTaBNmzaYP3++0PohMzOzyHpUtb5o2LAh/Pz8MHbsWERGRr5VDOUpIyMDvr6+8PX1LXE+GxsbbN++XekiQdGCBQsQGBgoJBbi4uKKXOwX7gMtVpUqVXD8+HGMGTNG2J8PHjzAihUrNFpPrVq1sGbNGri5uSE/Px+vX7+Gl5cXvLy8iszbsmVLbNq0SXSSRSqVIjAwEIGBgSXOV7VqVezcuRN169bVKPaKqn///pg0aZJQTyYuLg4//fSTynmHDx+OuXPnir441Nax+T745JNP0LRpU6xYsUIYGlYViUSCVatWwdnZudh1SSQS7Ny5E+PGjRNG4YiMjMS3335b7DJubm5lNoKEgYEB1q5di4kTJyI7OxtSqRRXr14tcZm2bdvixx9/xOLFiwEUDM25cuVKlfN27doVK1eu1HnSorTn0GnTpsHU1BSzZ89GdnY2srKy4OPjAx8fn7eK58GDB6K640ybNu0/k4QlepcwTUhEOlG9enV4e3vDz88PgwcPVhqDXZVq1aph7NixOHXqFA4dOqTURULO2NgYx48fx/jx44UmpIW1bt0avr6+mDp1qlZeR0maNWsGf3//YptxV6pUCc7OzggODtb4jl3r1q1x8eJFzJs3r8S+tQYGBnB2doarq2ux83h4eODIkSMYMWIEbGxsULVqVbXDxQKAlZUVzp07h99++01p1AtVDA0N0bNnT6xZswbz589Xu25t8/T0hJOTU7F3W+WqV6+Or7/+GufPn1fq616Yqakp/P39sXLlSvTs2RO1atWCoaGh1uI1NzfHsWPHsGbNGjRq1KjY+SQSCezs7LBgwQKV04cPH449e/YUu46qVati+vTp8Pf3h4WFhdq45syZAxcXF7XDPBoZGWHkyJEIDQ0tVWuTiuy3337DkiVLYGpqqnJ6nTp1sGrVKmzatEnU50jbx+b74rvvvsPmzZuL3RdNmzaFr68vpkyZonZdxsbG2L9/P3777bcSE2y2trbYs2cPVqxYIeq9La1PP/0UFy5cwMyZM9G2bVuYmpqqTVLNnTsX69evR61atVROt7CwwIIFC3D48GGl1gy68jbn0BEjRuDKlSuYPHmy2mGHa9eujXHjxuHo0aNFhn3u1q0bvvrqK7Rs2VLt+9m+fXt4e3vD09OzTN97IlJNkpqaWnaV6IiIRJLfTXr8+DESExORnp6OqlWrwtLSEq1atUKzZs00ao4ZHx+P4OBgxMXFQU9PDzVr1kSbNm1UJjvKw7179xAWFobExERUqVIFtWvXRufOnbVSxE0qlSIiIgJ3795FUlISZDIZzMzM0KRJE7Rv317txZC2PHr0CBEREUhISEB6ejqMjIzw4YcfwsbGBra2tlq9qC+tvLw8REZGIiYmBs+fP8fr16/xwQcfwMLCAra2tmjdunWxCS9dunfvHq5evYqEhARkZ2fDxMQEVlZWsLOzUypqVxypVIrw8HDcuHEDr169grm5OerVq4euXbuKGh5YlZiYGKE7SXp6OiQSCczMzGBjYwM7OztUrVq1VOt9V2RmZuLChQuIjo5GTk4OqlevDmtra3Tq1KlUTcff1WOzvPTv319oUeHl5YWxY8cCAPLz83Hx4kXExMQgMTERlpaWaNOmDezs7Eq1HZlMhuvXr+PWrVtISkqCnp4eatSoAXt7+xITiBVFTk4OLl68iKioKGRmZqJ69epo0KABHBwclArJ/lfk5eUhIiIC9+/fR3JyMnJzc1G1alVhpCt1CXW51NRU3LhxA48ePUJKSgpycnJgbGyMevXqoV27dmjQoEEZvxIiKgmTFkRERERUoRWXtCAiov8+dg8hIiIiIiIiogqJSQsiIiIiIiIiqpCYtCAiIiIiIiKiColJCyIiIiIiIiKqkJi0ICIiIiIiIqIKiaOHEBEREREREVGFxJYWRERERERERFQhMWlBRERERERERBUSkxZEREREREREVCExaUFEREREREREFRKTFkRERERERERUITFpQUREREREREQVEpMWRERERERERFQhMWlBRERERERERBUSkxZEREREREREVCExaUFEREREREREFRKTFkRERERERERUITFpQUREREREREQVEpMWRERERERERFQhMWlBRERERERERBXS/wO7Grv0CNJwgQAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "\n", "\n", "## Barplot (sorted by importance values)\n", "plt.figure(figsize=(7,5), dpi=150)\n", "plt.style.use('fivethirtyeight')\n", "sns.barplot(x = 'CARACTERISTICA', y = 'IMPORTANCIA', data = feature_importance_df,\n", " hue = 'SENTIMIENTO', dodge = False,\n", " order = feature_importance_df.sort_values('IMPORTANCIA').CARACTERISTICA)\n", "plt.xlabel(\"Características mas importantes\")\n", "plt.ylabel(\"Importancia de características\")\n", "plt.xticks(rotation=90)\n", "plt.title(\"Importancia de características: Palabras Clave\", color=\"black\")" ] } ], "metadata": { "kernelspec": { "display_name": "3.12.3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.3" } }, "nbformat": 4, "nbformat_minor": 2 }