Social media insight with Watson

Social media insights with Watson Developer Cloud _ DSX.ipynb

{
    "cells": [
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "<div><img src=\"https://www.ibm.com/blogs/bluemix/wp-content/uploads/2017/02/NLU.png\", width=270, height=270, align = 'right'> \n\n<img src=\"https://upload.wikimedia.org/wikipedia/commons/thumb/5/51/IBM_logo.svg/640px-IBM_logo.svg.png\", width = 90, height = 90, align = 'right', style=\"margin:0px 25px\"></div>\n\n# Extract Insights from Social Media with Watson Developer Cloud and Data Science Experience\n\nThis notebook shows you how you can use the Watson APIs and Data Science Experience to analyze and visualize data from social media to get customer insights. Brand managers can use this information to obtain enhanced insights about customer preferences enabling them to make more accurate marketing decisions. For a more detailed description of the problem solved, solution architecture, data collection and creating/accessing the services involved, please consult <a href=\"https://www.ibm.com/developerworks/library/cc-cognitive-watson-extract-insights-spark-dsx/index.html\" target=\"_blank\" rel=\"noopener no referrer\">this tutorial</a>.\n\nThis notebook runs on Python with Spark 2.1.\n___________"
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "## Table of contents\n\n1.  [Load the required libraries](#loadlibraries)\n2.  [Load data from Db2 Warehouse on Cloud](#loaddata)\n3.  [Perform some exploratory data analysis](#exploredata)\n4.  [Take a data sample](#takesample)\n5.  [Read credentials for NLU, Personality Insights, and Twitter](#getcredentials)\n6.  [Enrich the data with Watson NLU](#enrichnlu)\n7.  [Visualize sentiment and keywords](#sentiment)\n8.  [Enrich data with Watson Personality Insights](#enrichpi)\n9.  [Spark machine learning for user segmentation](#sparkml)\n10. [Visualize user segmentation](#clusters)"
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "__________________\n\n<a id=\"loadlibraries\"></a>\n## Step 1: Load the required libraries\n\n- <a href=\"https://github.com/watson-developer-cloud/python-sdk\" target=\"_blank\" rel=\"noopener no referrer\">watson-developer-cloud</a> is the Python SDK for Watson Developer Cloud services \n\n- <a href=\"http://www.tweepy.org/\" target=\"_blank\" rel=\"noopener no referrer\">tweepy</a> is a Python library for accessing Twitter API\n\n- <a href=\"https://github.com/amueller/word_cloud/\" target=\"_blank\" rel=\"noopener no referrer\">wordcloud</a> is a Python library for generating Word Clouds \n\n- <a href=\"https://pypi.python.org/pypi/plotly\" target=\"_blank\" rel=\"noopener no referrer\">plotly</a> is a Python library for making plots and charts"
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "!pip install --upgrade watson-developer-cloud\n!pip install tweepy==3.3.0 --upgrade --force-reinstall\n!pip install --upgrade plotly\n!pip install --upgrade --force-reinstall wordcloud", 
            "execution_count": null
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "________________\n\n<a id=\"loaddata\"> </a>\n## Step 2: Load data from Db2 Warehouse on Cloud\n\nThe first step is to load the data. This notebook assumes you have tweets already in a Db2 Warehouse on Cloud database. For more details about how to collect relevant tweets into a Db2 Warehouse on Cloud database, check out <a href=\"https://github.com/joe4k/twitterstreams\" target=\"_blank\" rel=\"noopener no referrer\">this</a> github repository.\n\nIf you already know how to create data connections, you can skip this step.\n\n1. Click **Data Services** tab, then select **Connections**.\n2. Click **Create new connection** (+ sign)\n3. Provide a connection name (dashdbsingers) and select **Data Service** for **Service Category**.\n4. Select the Db2 Warehouse on Cloud service instance where the tweets are stored.\n5. Select **BLUDB** database and press **Create**."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Next you need to make the data connection you created accessible to your project. \n1. Navigate to your project.\n2. Click the **Find and Add Data** icon (top right).\n3. Click the **Connections** tab.\n4. Select the `dashdbsingers` connection.\n5. Click **Apply** ==> This makes that data connection accessible to your project."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Lastly, you need to add code to your notebook to read the data from Db2 Warehouse on Cloud.\n1. Create a new cell in your notebook.\n2. Click the **Find and Add Data** icon.\n3. Click the **Connections** tab.\n4. Find `dashdbsingers` connection, then click **Insert to code** under `dashdbsingers` connection.\nThis adds code into the notebook to load data from Db2 Warehouse on Cloud database. \n\n**Note:** You would select a specific table to load. In this  case, you load **DSX_CLOUDANT_SINGERS_TWEETS**."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "Row(COORDINATES_TYPE=None, CREATED_AT=u'Fri Jul 07 19:05:19 +0000 2017', FAVORITE_COUNT=0, ID=883401558540431360, LANG=u'en', PLACE_BOUNDING_BOX_TYPE=None, PLACE_COUNTRY=None, PLACE_COUNTRY_CODE=None, PLACE_FULL_NAME=None, PLACE_ID=None, PLACE_NAME=None, PLACE_PLACE_TYPE=None, PLACE_URL=None, TEXT=u'@lilly_malone7 @_HGlobal @justinbieber Because your boyfriend is a girlfriend', TEXT_CLEAN=u'   Because your boyfriend is a girlfriend', USER_FAVOURITES_COUNT=5, USER_FOLLOWERS_COUNT=7, USER_FRIENDS_COUNT=35, USER_ID=29899524, USER_LOCATION=u'43112', USER_SCREEN_NAME=u'btkb8', USER_STATUSES_COUNT=2, _ID=u'93b64fff28cb8ce9a9a0be5b5232af43', _REV=u'1-feb69621bcb84206c30850a4476f58d8')"
                    }, 
                    "metadata": {}, 
                    "execution_count": 2
                }
            ], 
            "source": "# This cell is where code is inserted to load data from your dashDB instance with tweets to analyze.\n", 
            "execution_count": 2
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# copy data into brandTweetsDF dataframe for processing\nbrandTweetsDF = data_df_2", 
            "execution_count": 3
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "___________\n\n<a id=\"exploredata\"> </a>\n## Step 3: Run some exploratory data analysis"
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "scrolled": true
            }, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  COORDINATES_TYPE                      CREATED_AT  FAVORITE_COUNT  \\\n0             None  Fri Jul 07 19:40:02 +0000 2017               0   \n1             None  Fri Jul 07 19:53:49 +0000 2017               0   \n\n                   ID LANG PLACE_BOUNDING_BOX_TYPE PLACE_COUNTRY  \\\n0  883410299310800897   en                    None          None   \n1  883413767937306626   en                    None          None   \n\n  PLACE_COUNTRY_CODE PLACE_FULL_NAME PLACE_ID  \\\n0               None            None     None   \n1               None            None     None   \n\n                  ...                  \\\n0                 ...                   \n1                 ...                   \n\n                                          TEXT_CLEAN USER_FAVOURITES_COUNT  \\\n0  RT : Justin Bieber makes Battersea boy's dream...                  5164   \n1     AUSTIN PLEASE \\nDON'T INGORE ME JUST ONE TI...                  9855   \n\n  USER_FOLLOWERS_COUNT USER_FRIENDS_COUNT             USER_ID  USER_LOCATION  \\\n0                19286                497          1586523338           None   \n1                  564                937  745476605410050049           None   \n\n   USER_SCREEN_NAME  USER_STATUSES_COUNT                               _ID  \\\n0         hairzflip                76597  cfa16f640a7126ea962f62101c6ba79f   \n1         dramzwiad                10212  bd6785d82008850602db30ce57607136   \n\n                                 _REV  \n0  1-30bc128851926975750b1186b0c7603b  \n1  1-75c95c9bd1bcf4bb8ad950226f570eb0  \n\n[2 rows x 24 columns]", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>COORDINATES_TYPE</th>\n      <th>CREATED_AT</th>\n      <th>FAVORITE_COUNT</th>\n      <th>ID</th>\n      <th>LANG</th>\n      <th>PLACE_BOUNDING_BOX_TYPE</th>\n      <th>PLACE_COUNTRY</th>\n      <th>PLACE_COUNTRY_CODE</th>\n      <th>PLACE_FULL_NAME</th>\n      <th>PLACE_ID</th>\n      <th>...</th>\n      <th>TEXT_CLEAN</th>\n      <th>USER_FAVOURITES_COUNT</th>\n      <th>USER_FOLLOWERS_COUNT</th>\n      <th>USER_FRIENDS_COUNT</th>\n      <th>USER_ID</th>\n      <th>USER_LOCATION</th>\n      <th>USER_SCREEN_NAME</th>\n      <th>USER_STATUSES_COUNT</th>\n      <th>_ID</th>\n      <th>_REV</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:40:02 +0000 2017</td>\n      <td>0</td>\n      <td>883410299310800897</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>RT : Justin Bieber makes Battersea boy's dream...</td>\n      <td>5164</td>\n      <td>19286</td>\n      <td>497</td>\n      <td>1586523338</td>\n      <td>None</td>\n      <td>hairzflip</td>\n      <td>76597</td>\n      <td>cfa16f640a7126ea962f62101c6ba79f</td>\n      <td>1-30bc128851926975750b1186b0c7603b</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:53:49 +0000 2017</td>\n      <td>0</td>\n      <td>883413767937306626</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>AUSTIN PLEASE \\nDON'T INGORE ME JUST ONE TI...</td>\n      <td>9855</td>\n      <td>564</td>\n      <td>937</td>\n      <td>745476605410050049</td>\n      <td>None</td>\n      <td>dramzwiad</td>\n      <td>10212</td>\n      <td>bd6785d82008850602db30ce57607136</td>\n      <td>1-75c95c9bd1bcf4bb8ad950226f570eb0</td>\n    </tr>\n  </tbody>\n</table>\n<p>2 rows \u00d7 24 columns</p>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 4
                }
            ], 
            "source": "# Return top 2 rows of Spark DataFrame\nbrandTweetsDF.limit(2).toPandas()", 
            "execution_count": 4
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "root\n |-- COORDINATES_TYPE: string (nullable = true)\n |-- CREATED_AT: string (nullable = true)\n |-- FAVORITE_COUNT: long (nullable = true)\n |-- ID: long (nullable = true)\n |-- LANG: string (nullable = true)\n |-- PLACE_BOUNDING_BOX_TYPE: string (nullable = true)\n |-- PLACE_COUNTRY: string (nullable = true)\n |-- PLACE_COUNTRY_CODE: string (nullable = true)\n |-- PLACE_FULL_NAME: string (nullable = true)\n |-- PLACE_ID: string (nullable = true)\n |-- PLACE_NAME: string (nullable = true)\n |-- PLACE_PLACE_TYPE: string (nullable = true)\n |-- PLACE_URL: string (nullable = true)\n |-- TEXT: string (nullable = true)\n |-- TEXT_CLEAN: string (nullable = true)\n |-- USER_FAVOURITES_COUNT: long (nullable = true)\n |-- USER_FOLLOWERS_COUNT: long (nullable = true)\n |-- USER_FRIENDS_COUNT: long (nullable = true)\n |-- USER_ID: long (nullable = true)\n |-- USER_LOCATION: string (nullable = true)\n |-- USER_SCREEN_NAME: string (nullable = true)\n |-- USER_STATUSES_COUNT: long (nullable = true)\n |-- _ID: string (nullable = false)\n |-- _REV: string (nullable = true)\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "# Print the schema of the loaded data\nbrandTweetsDF.printSchema()", 
            "execution_count": 5
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Drop unneeded columns\nbrandTweetsDF = brandTweetsDF.drop('_ID','_REV')", 
            "execution_count": 6
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Extract day from the `CREATED_AT` field. This is useful to plot tweet trends over time."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "          DAY\n0  2017-07-10\n1  2017-07-10\n2  2017-07-10\n3  2017-07-10\n4  2017-07-10", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>DAY</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>2017-07-10</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2017-07-10</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2017-07-10</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2017-07-10</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2017-07-10</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 7
                }
            ], 
            "source": "import datetime\nfrom datetime import date\nfrom dateutil import parser\n\ndef getDay(date):\n    print 'input date: ', date\n    day = parser.parse(str(date))\n    day = day.date()\n    return day\n\n# Add a field for the day the tweet was created (ignoring hour/minute/second)\nfrom pyspark.sql.functions import udf\nfrom pyspark.sql.types import DateType\n\nudfGetDay = udf(getDay, DateType())\n\nbrandTweetsDF = brandTweetsDF.withColumn('DAY',udfGetDay('CREATED_AT'))\n\n# Verify added field is as expected\nbrandTweetsDF.select(\"DAY\").limit(5).toPandas()", 
            "execution_count": 7
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "____________\n\n<a id=\"takesample\"> </a>\n## Step 4: Take a data sample\nFor purposes of this tutorial, we will work with a small sample of the data. In practice, you want to use large data sets for our analysis to capture as many insights as we can. However, to illustrate the approach, we can work with a small data set.\n\nFurthermore, we want to restrict the number of API calls to the free plan of <a href=\"https://www.ibm.com/watson/developercloud/\" target=\"_blank\" rel=\"noopener no referrer\">Watson Developer Cloud</a> services so users can run through the notebook successfully even if they only have access to free tier of the services."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Number of records:  198070  Sample size:  950  Fraction:  0.00479628414197  Seed:  98\nNumber of records to send to NLU: 936\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "## Take a sample of the data\n## Limit to 1000 records as Watson NLU allows 1000 free calls per day\nimport random\n\nnum_records = brandTweetsDF.count()\nsample_num_records = 950\nfraction = float(sample_num_records)/float(num_records)\n\nseed = random.randint(1, 100)\nprint 'Number of records: ', num_records, ' Sample size: ', sample_num_records, ' Fraction: ', fraction, ' Seed: ', seed\nbrandTweetsSampleDF = brandTweetsDF.sample(False, fraction, seed)\n\n\n## Alternative Stratified Sampling approach\n## Returns RDD with length of 2, first col is the key (day) and second col is the original row for the key\n## Take only the actual data (column 1)\n## If you'd like to use this approach, uncomment the following 4 lines\n\n#fractionList = brandTweetsDF.rdd.map(lambda x: x['DAY']).distinct().map(lambda x: (x,fraction)).collectAsMap()\n#keybyday = brandTweetsDF.rdd.keyBy(lambda x: x['DAY'])\n#brandTweetsDFrdd = keybyday.sampleByKey(False,fractionList).map(lambda x: x[1])\n#brandTweetsSampleDF = spark.createDataFrame(brandTweetsDFrdd,brandTweetsDF.schema)\n\n\nprint 'Number of records to send to NLU:', brandTweetsSampleDF.count()", 
            "execution_count": 8
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "+----------+------------------+\n|       DAY|NUM_TWEETS_PER_DAY|\n+----------+------------------+\n|2017-07-06|               122|\n|2017-07-10|               132|\n|2017-07-11|                78|\n|2017-07-12|                32|\n|2017-07-09|               253|\n|2017-07-05|                18|\n|2017-07-07|               130|\n|2017-07-13|                52|\n|2017-07-08|               119|\n+----------+------------------+\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "# plot number of tweets per day\nfrom pyspark.sql import functions as F\nbrandTweetsSampleDFperDay = brandTweetsSampleDF.groupBy('DAY')\\\n                              .agg(F.count('ID')\\\n                              .alias('NUM_TWEETS_PER_DAY'))\nbrandTweetsSampleDFperDay.show()", 
            "execution_count": 9
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "____________\n\n<a id=\"getcredentials\"> </a>\n## Step 5: Read the credentials for NLU, Personality Insights, and Twitter\nUpload a json file (for example, `sample_creds.json`) which has the credentials for NLU, Personality Insights and Twitter to your Object Storage instance.\n"
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "1. Click the **Find and Add Data** icon. \n2. Click **Files** tab.\n3. Either drop your file in the box or click **browse** to select the file with credentials information from your local disk.\n4. After you have selected the file, press **Open** to upload the file to Object Storage.\nHere is the format for the credentials file:\n```\n{\n\t\"nlu_username\": \"YOUR NLU username\",\n\t\"nlu_password\": \"YOUR NLU password\",\n\t\"nlu_version\": \"NLU version\",\n\t\"twitter_consumer_key\": \"YOUR Twitter App consumer key\",\n\t\"twitter_consumer_secret\": \"YOUR Twitter App consumer secret\",\n\t\"twitter_access_token\": \"YOUR Twitter App access token\",\n\t\"twitter_access_token_secret\": \"YOUR Twitter App access token secret\",\n\t\"pi_username\": \"YOUR Personality Insights username\",\n\t\"pi_password\": \"YOUR Personality Insights password\",\n\t\"pi_version\": \"Personality Insights version\"\n}\n```\nAfter the file has been uploaded, you should see the `sample_creds.json` file under **Files** tab under the **Find and Add Data** column."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Next, read the `sample_creds.json` credentials file and set the credentials for NLU, Personality Insights, and Twitter.\n1. Create a new cell in your notebook.\n2. If not open, click the **Find and Add Data** icon.\n3. Click the **Files** tab.\n4. Find the ```sample_creds.json``` file and click **Insert to code** under the file name \n5. Choose **Insert Credentials** ==> this inserts code into your notebook for setting the credentials to read that file from Object Storage."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Read credentials for NLU, Personality Insights, and Twitter from sample_creds.json\n## This cell gets populated with inserted code for reading sample_creds.json file from your Object Storage instance\n", 
            "execution_count": 10
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from io import BytesIO  \nimport requests  \nimport json \n\ndef get_data(credentials):  \n    \"\"\"This functions returns a StringIO object containing\n    the file content from Bluemix Object Storage V3.\"\"\"\n\n    url1 = ''.join(['https://identity.open.softlayer.com', '/v3/auth/tokens'])\n    data = {'auth': {'identity': {'methods': ['password'],\n            'password': {'user': {'name': credentials['username'],'domain': {'id': credentials['domain_id']},\n            'password': credentials['password']}}}}}\n    headers1 = {'Content-Type': 'application/json'}\n    resp1 = requests.post(url=url1, data=json.dumps(data), headers=headers1)\n    resp1_body = resp1.json()\n    for e1 in resp1_body['token']['catalog']:\n        if(e1['type']=='object-store'):\n            for e2 in e1['endpoints']:\n                        if(e2['interface']=='public'and e2['region']=='dallas'):\n                            url2 = ''.join([e2['url'],'/', credentials['container'], '/', credentials['filename']])\n    s_subject_token = resp1.headers['x-subject-token']\n    headers2 = {'X-Auth-Token': s_subject_token, 'accept': 'application/json'}\n    resp2 = requests.get(url=url2, headers=headers2)\n    return json.loads(resp2.content)\n", 
            "execution_count": 11
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# Note that we need to reference the credentials object returned by Insert to code (in this example, it is credentials_2)\ncredentials_json = get_data(credentials_4)\n\n## You can print the credentials if you'd like to confirm they're captured correctly\n## To print, uncomment the following lines\n#print 'Credentials for NLU'\n#print 'NLU username: ', credentials_json['nlu_username']\n#print 'NLU password: ', credentials_json['nlu_password']\n#print 'NLU version: ', credentials_json['nlu_version']\n\n#print 'Credentials for Twitter '\n#print 'Twitter consumer key: ', credentials_json['twitter_consumer_key']\n#print 'Twitter consumer secret: ', credentials_json['twitter_consumer_secret']\n#print 'Twitter access token: ', credentials_json['twitter_access_token']\n#print 'Twitter access token secret: ', credentials_json['twitter_access_token_secret']\n\n#print 'Credentials for Personality Insights'\n#print 'Personality Insights username: ', credentials_json['pi_username']\n#print 'Personality Insights password: ', credentials_json['pi_password']\n#print 'Personality Insights version: ', credentials_json['pi_version']", 
            "execution_count": 12
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "__________\n\n<a id=\"enrichnlu\"> </a>\n## Step 6: Enrich the data with  <a href=\"https://www.ibm.com/watson/developercloud/natural-language-understanding.html\" target=\"_blank\" rel=\"noopener no referrer\">Watson Natural Language Understanding (NLU)</a>\n\nWatson NLU allows us to extract sentiment and keywords from text.  In our case, the text is user tweets."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "import watson_developer_cloud\nimport watson_developer_cloud.natural_language_understanding.features.v1 as features\n\n## Define credentials for NLU service\nnlu_username=credentials_json['nlu_username']\nnlu_password=credentials_json['nlu_password']\nnlu_version=credentials_json['nlu_version']\nnlu = watson_developer_cloud.NaturalLanguageUnderstandingV1(version = nlu_version,\n                                                            username = nlu_username,\n                                                            password = nlu_password)\n\n## Send text to NLU and extract Sentiment and Keywords\n## Make sure text is utf-8 encoded\ndef enrichNLU(text):\n    utf8text = text.encode(\"utf-8\")\n    try:\n        result = nlu.analyze(text = utf8text, features = [features.Sentiment(),features.Keywords()])\n        sentiment = result['sentiment']['document']['score']\n        sentiment_label = result['sentiment']['document']['label']\n        keywords = list(result['keywords'])\n    except Exception:\n        result = None\n        sentiment = 0.0\n        sentiment_label = None\n        keywords = None\n    return sentiment, sentiment_label, keywords", 
            "execution_count": 13
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "We'll convert our dataframe to a Pandas dataframe and enrich the tweets with the results of Watson NLU."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# Create a Pandas frame and augment with Sentiment analysis and Keywords using Watson NLU\nbrandTweetsSamplePandasDF = brandTweetsSampleDF.toPandas()", 
            "execution_count": 14
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "**The next cell sends the records to the Watson API, so be aware of the number of calls that are being made. If you are on the free plan which is limited to 1000 API calls per day, you can only run this cell once. After that the server responds with a notice indicating you have reached the maximum number of API calls for the day.**  This could take 60+ seconds so some patience is required.  "
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## This calls the enrichNLU function which accesses the Watson NLU API\nbrandTweetsSamplePandasDF['SENTIMENT'],brandTweetsSamplePandasDF['SENTIMENT_LABEL'],\\\nbrandTweetsSamplePandasDF['KEYWORDS'] = zip(*brandTweetsSamplePandasDF['TEXT_CLEAN'].map(enrichNLU))", 
            "execution_count": 15
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Rows x Columns for brandTweetsSampleDF: (936, 26)\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }, 
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  COORDINATES_TYPE                      CREATED_AT  FAVORITE_COUNT  \\\n0             None  Fri Jul 07 19:43:27 +0000 2017               0   \n1             None  Fri Jul 07 19:27:36 +0000 2017               0   \n2             None  Fri Jul 07 20:00:41 +0000 2017               0   \n3             None  Fri Jul 07 23:15:10 +0000 2017               0   \n4             None  Sat Jul 08 00:05:03 +0000 2017               0   \n5             None  Sat Jul 08 01:32:00 +0000 2017               0   \n6             None  Sat Jul 08 01:54:11 +0000 2017               0   \n7             None  Sat Jul 08 02:52:38 +0000 2017               0   \n8             None  Sat Jul 08 04:11:43 +0000 2017               0   \n9             None  Sat Jul 08 06:37:17 +0000 2017               0   \n\n                   ID LANG PLACE_BOUNDING_BOX_TYPE PLACE_COUNTRY  \\\n0  883411155414392833   en                    None          None   \n1  883407168535502848   en                    None          None   \n2  883415494698598400   en                    None          None   \n3  883464437167824897   en                    None          None   \n4  883476992930217990   en                    None          None   \n5  883498871892193280   en                    None          None   \n6  883504454041034754   en                    None          None   \n7  883519164756963328   en                    None          None   \n8  883539065764810752   en                    None          None   \n9  883575699524681728   en                    None          None   \n\n  PLACE_COUNTRY_CODE PLACE_FULL_NAME PLACE_ID  \\\n0               None            None     None   \n1               None            None     None   \n2               None            None     None   \n3               None            None     None   \n4               None            None     None   \n5               None            None     None   \n6               None            None     None   \n7               None            None     None   \n8               None            None     None   \n9               None            None     None   \n\n                         ...                         USER_FOLLOWERS_COUNT  \\\n0                        ...                                          257   \n1                        ...                                            2   \n2                        ...                                           27   \n3                        ...                                          340   \n4                        ...                                         2277   \n5                        ...                                         1096   \n6                        ...                                         4251   \n7                        ...                                         1091   \n8                        ...                                          187   \n9                        ...                                           36   \n\n  USER_FRIENDS_COUNT             USER_ID                USER_LOCATION  \\\n0                238          1939518271        Aboard the Death Star   \n1                 74  882967133092794368                 bieberlandia   \n2                 22  864861587055812608               United Kingdom   \n3                247  754786084622446593                  Hawaii, USA   \n4               3328          1041634262                    Hong Kong   \n5               1068  741541482658680832       Rio de Janeiro, Brasil   \n6               3815          1138417584                          Arg   \n7                530          3054646307  En alg\u00fan lugar del mundo...   \n8                260          2998732628           Nuevo Le\u00f3n, M\u00e9xico   \n9                 72  803167039753953280                       Ranchi   \n\n  USER_SCREEN_NAME  USER_STATUSES_COUNT         DAY  SENTIMENT  \\\n0    ___Robertt___                 7656  2017-07-07   0.000000   \n1   goddesdemetria                   16  2017-07-07   0.996150   \n2   hannahhewittxo                  426  2017-07-07   0.000000   \n3  ParadiseTVNEWS7                24015  2017-07-07   0.000000   \n4        jizzclone                20524  2017-07-08  -0.484947   \n5    RaaiTeixeira1               173003  2017-07-08   0.000000   \n6    Queensofevery               205280  2017-07-08   0.881953   \n7    My_purpose946                14413  2017-07-08   0.000000   \n8     brenda199939                 3519  2017-07-08   0.000000   \n9  Prashan33972844                   53  2017-07-08   0.671932   \n\n   SENTIMENT_LABEL                                           KEYWORDS  \n0          neutral  [{u'relevance': 0.978919, u'text': u'new album...  \n1         positive                                                 []  \n2          neutral  [{u'relevance': 0.981626, u'text': u'unlocks'}...  \n3          neutral      [{u'relevance': 0.903695, u'text': u'\ud83c\udf3a\ud83c\udfddOMG'}]  \n4         negative  [{u'relevance': 0.951402, u'text': u'Shallow s...  \n5          neutral  [{u'relevance': 0.962964, u'text': u'Purpose T...  \n6         positive  [{u'relevance': 0.954656, u'text': u'head  hah...  \n7             None                                               None  \n8          neutral  [{u'relevance': 0.966035, u'text': u'SOCIAL'},...  \n9         positive  [{u'relevance': 0.909219, u'text': u'stylish P...  \n\n[10 rows x 26 columns]", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>COORDINATES_TYPE</th>\n      <th>CREATED_AT</th>\n      <th>FAVORITE_COUNT</th>\n      <th>ID</th>\n      <th>LANG</th>\n      <th>PLACE_BOUNDING_BOX_TYPE</th>\n      <th>PLACE_COUNTRY</th>\n      <th>PLACE_COUNTRY_CODE</th>\n      <th>PLACE_FULL_NAME</th>\n      <th>PLACE_ID</th>\n      <th>...</th>\n      <th>USER_FOLLOWERS_COUNT</th>\n      <th>USER_FRIENDS_COUNT</th>\n      <th>USER_ID</th>\n      <th>USER_LOCATION</th>\n      <th>USER_SCREEN_NAME</th>\n      <th>USER_STATUSES_COUNT</th>\n      <th>DAY</th>\n      <th>SENTIMENT</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>KEYWORDS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:43:27 +0000 2017</td>\n      <td>0</td>\n      <td>883411155414392833</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>257</td>\n      <td>238</td>\n      <td>1939518271</td>\n      <td>Aboard the Death Star</td>\n      <td>___Robertt___</td>\n      <td>7656</td>\n      <td>2017-07-07</td>\n      <td>0.000000</td>\n      <td>neutral</td>\n      <td>[{u'relevance': 0.978919, u'text': u'new album...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:27:36 +0000 2017</td>\n      <td>0</td>\n      <td>883407168535502848</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>2</td>\n      <td>74</td>\n      <td>882967133092794368</td>\n      <td>bieberlandia</td>\n      <td>goddesdemetria</td>\n      <td>16</td>\n      <td>2017-07-07</td>\n      <td>0.996150</td>\n      <td>positive</td>\n      <td>[]</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>None</td>\n      <td>Fri Jul 07 20:00:41 +0000 2017</td>\n      <td>0</td>\n      <td>883415494698598400</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>27</td>\n      <td>22</td>\n      <td>864861587055812608</td>\n      <td>United Kingdom</td>\n      <td>hannahhewittxo</td>\n      <td>426</td>\n      <td>2017-07-07</td>\n      <td>0.000000</td>\n      <td>neutral</td>\n      <td>[{u'relevance': 0.981626, u'text': u'unlocks'}...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>None</td>\n      <td>Fri Jul 07 23:15:10 +0000 2017</td>\n      <td>0</td>\n      <td>883464437167824897</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>340</td>\n      <td>247</td>\n      <td>754786084622446593</td>\n      <td>Hawaii, USA</td>\n      <td>ParadiseTVNEWS7</td>\n      <td>24015</td>\n      <td>2017-07-07</td>\n      <td>0.000000</td>\n      <td>neutral</td>\n      <td>[{u'relevance': 0.903695, u'text': u'\ud83c\udf3a\ud83c\udfddOMG'}]</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>None</td>\n      <td>Sat Jul 08 00:05:03 +0000 2017</td>\n      <td>0</td>\n      <td>883476992930217990</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>2277</td>\n      <td>3328</td>\n      <td>1041634262</td>\n      <td>Hong Kong</td>\n      <td>jizzclone</td>\n      <td>20524</td>\n      <td>2017-07-08</td>\n      <td>-0.484947</td>\n      <td>negative</td>\n      <td>[{u'relevance': 0.951402, u'text': u'Shallow s...</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>None</td>\n      <td>Sat Jul 08 01:32:00 +0000 2017</td>\n      <td>0</td>\n      <td>883498871892193280</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>1096</td>\n      <td>1068</td>\n      <td>741541482658680832</td>\n      <td>Rio de Janeiro, Brasil</td>\n      <td>RaaiTeixeira1</td>\n      <td>173003</td>\n      <td>2017-07-08</td>\n      <td>0.000000</td>\n      <td>neutral</td>\n      <td>[{u'relevance': 0.962964, u'text': u'Purpose T...</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>None</td>\n      <td>Sat Jul 08 01:54:11 +0000 2017</td>\n      <td>0</td>\n      <td>883504454041034754</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>4251</td>\n      <td>3815</td>\n      <td>1138417584</td>\n      <td>Arg</td>\n      <td>Queensofevery</td>\n      <td>205280</td>\n      <td>2017-07-08</td>\n      <td>0.881953</td>\n      <td>positive</td>\n      <td>[{u'relevance': 0.954656, u'text': u'head  hah...</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>None</td>\n      <td>Sat Jul 08 02:52:38 +0000 2017</td>\n      <td>0</td>\n      <td>883519164756963328</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>1091</td>\n      <td>530</td>\n      <td>3054646307</td>\n      <td>En alg\u00fan lugar del mundo...</td>\n      <td>My_purpose946</td>\n      <td>14413</td>\n      <td>2017-07-08</td>\n      <td>0.000000</td>\n      <td>None</td>\n      <td>None</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>None</td>\n      <td>Sat Jul 08 04:11:43 +0000 2017</td>\n      <td>0</td>\n      <td>883539065764810752</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>187</td>\n      <td>260</td>\n      <td>2998732628</td>\n      <td>Nuevo Le\u00f3n, M\u00e9xico</td>\n      <td>brenda199939</td>\n      <td>3519</td>\n      <td>2017-07-08</td>\n      <td>0.000000</td>\n      <td>neutral</td>\n      <td>[{u'relevance': 0.966035, u'text': u'SOCIAL'},...</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>None</td>\n      <td>Sat Jul 08 06:37:17 +0000 2017</td>\n      <td>0</td>\n      <td>883575699524681728</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>36</td>\n      <td>72</td>\n      <td>803167039753953280</td>\n      <td>Ranchi</td>\n      <td>Prashan33972844</td>\n      <td>53</td>\n      <td>2017-07-08</td>\n      <td>0.671932</td>\n      <td>positive</td>\n      <td>[{u'relevance': 0.909219, u'text': u'stylish P...</td>\n    </tr>\n  </tbody>\n</table>\n<p>10 rows \u00d7 26 columns</p>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 16
                }
            ], 
            "source": "# view top two records to verify Sentiment and Keywords enrichments are applied as expected\nprint 'Rows x Columns for brandTweetsSampleDF:', brandTweetsSamplePandasDF.shape\nbrandTweetsSamplePandasDF[:10] ", 
            "execution_count": 16
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from pyspark.sql.types import StringType\nfrom pyspark.sql.types import FloatType\nfrom pyspark.sql.types import ArrayType\nfrom pyspark.sql.types import StructType\nfrom pyspark.sql.types import StructField\n\nschema = brandTweetsSampleDF.schema\nschema1 = StructType([\n    StructField(\"relevance\", FloatType(), True),\n    StructField(\"text\", StringType(), True),                            \n])\n\nkeywordschema = StructType.fromJson(schema1.jsonValue())\nadded_fields = [StructField(\"SENTIMENT\", FloatType(), True),StructField(\"SENTIMENT_LABEL\",StringType(),True),\\\n                StructField(\"KEYWORDS\",ArrayType(keywordschema),True)] \n\nnewfields = StructType(schema.fields + added_fields)", 
            "execution_count": 17
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "root\n |-- COORDINATES_TYPE: string (nullable = true)\n |-- CREATED_AT: string (nullable = true)\n |-- FAVORITE_COUNT: long (nullable = true)\n |-- ID: long (nullable = true)\n |-- LANG: string (nullable = true)\n |-- PLACE_BOUNDING_BOX_TYPE: string (nullable = true)\n |-- PLACE_COUNTRY: string (nullable = true)\n |-- PLACE_COUNTRY_CODE: string (nullable = true)\n |-- PLACE_FULL_NAME: string (nullable = true)\n |-- PLACE_ID: string (nullable = true)\n |-- PLACE_NAME: string (nullable = true)\n |-- PLACE_PLACE_TYPE: string (nullable = true)\n |-- PLACE_URL: string (nullable = true)\n |-- TEXT: string (nullable = true)\n |-- TEXT_CLEAN: string (nullable = true)\n |-- USER_FAVOURITES_COUNT: long (nullable = true)\n |-- USER_FOLLOWERS_COUNT: long (nullable = true)\n |-- USER_FRIENDS_COUNT: long (nullable = true)\n |-- USER_ID: long (nullable = true)\n |-- USER_LOCATION: string (nullable = true)\n |-- USER_SCREEN_NAME: string (nullable = true)\n |-- USER_STATUSES_COUNT: long (nullable = true)\n |-- DAY: date (nullable = true)\n |-- SENTIMENT: float (nullable = true)\n |-- SENTIMENT_LABEL: string (nullable = true)\n |-- KEYWORDS: array (nullable = true)\n |    |-- element: struct (containsNull = true)\n |    |    |-- relevance: float (nullable = true)\n |    |    |-- text: string (nullable = true)\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "## Push the enriched data back out to Spark to continue working within the Spark API\nenrichedBrandTweetsDF = spark.createDataFrame(brandTweetsSamplePandasDF,newfields)\n\n# Print the schema of the Spark DataFrame to verify we have all the expected fields\nenrichedBrandTweetsDF.printSchema()", 
            "execution_count": 18
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "____________\n\n<a id=\"sentiment\"> </a>\n## Step 7: Visualize sentiment and keywords\n\nTwitter trends, sentiment, and keywords give a brand manager a view of a consumers' perceptions about the brand. These insights can be very useful to the brand manager. \n\nIn this step we visualize some of the data we received from NLU. Let's look at tweet trends and sentiment for **katyperry** and **taylorswift**. In addition, let's show the main keywords tweeted for both musicians.\n\nIn this section:<br/>\n7.1 [Compare sentiment between brand1, brand2 and brand3](#compare)<br/>\n7.2 [View sentiment over time for brands](#view)<br/>\n7.3 [Top keywords](#keywords)\n\n"
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# Separate tweets by brand\n# To do so, check if the tweet includes which brand and add a column to represent that\nbrandList=['katyperry','justinbieber','taylorswift']\ndef addBrand(text):\n    for brand in brandList:\n        if brand in text.lower():\n            return brand\n    return None", 
            "execution_count": 19
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  COORDINATES_TYPE                      CREATED_AT  FAVORITE_COUNT  \\\n0             None  Fri Jul 07 19:43:27 +0000 2017               0   \n1             None  Fri Jul 07 19:27:36 +0000 2017               0   \n\n                   ID LANG PLACE_BOUNDING_BOX_TYPE PLACE_COUNTRY  \\\n0  883411155414392833   en                    None          None   \n1  883407168535502848   en                    None          None   \n\n  PLACE_COUNTRY_CODE PLACE_FULL_NAME PLACE_ID      ...       \\\n0               None            None     None      ...        \n1               None            None     None      ...        \n\n  USER_FRIENDS_COUNT             USER_ID          USER_LOCATION  \\\n0                238          1939518271  Aboard the Death Star   \n1                 74  882967133092794368           bieberlandia   \n\n  USER_SCREEN_NAME USER_STATUSES_COUNT         DAY  SENTIMENT  \\\n0    ___Robertt___                7656  2017-07-07    0.00000   \n1   goddesdemetria                  16  2017-07-07    0.99615   \n\n   SENTIMENT_LABEL                                           KEYWORDS  \\\n0          neutral  [(0.978919029236, new album), (0.843757987022,...   \n1         positive                                                 []   \n\n          BRAND  \n0  justinbieber  \n1          None  \n\n[2 rows x 27 columns]", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>COORDINATES_TYPE</th>\n      <th>CREATED_AT</th>\n      <th>FAVORITE_COUNT</th>\n      <th>ID</th>\n      <th>LANG</th>\n      <th>PLACE_BOUNDING_BOX_TYPE</th>\n      <th>PLACE_COUNTRY</th>\n      <th>PLACE_COUNTRY_CODE</th>\n      <th>PLACE_FULL_NAME</th>\n      <th>PLACE_ID</th>\n      <th>...</th>\n      <th>USER_FRIENDS_COUNT</th>\n      <th>USER_ID</th>\n      <th>USER_LOCATION</th>\n      <th>USER_SCREEN_NAME</th>\n      <th>USER_STATUSES_COUNT</th>\n      <th>DAY</th>\n      <th>SENTIMENT</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>KEYWORDS</th>\n      <th>BRAND</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:43:27 +0000 2017</td>\n      <td>0</td>\n      <td>883411155414392833</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>238</td>\n      <td>1939518271</td>\n      <td>Aboard the Death Star</td>\n      <td>___Robertt___</td>\n      <td>7656</td>\n      <td>2017-07-07</td>\n      <td>0.00000</td>\n      <td>neutral</td>\n      <td>[(0.978919029236, new album), (0.843757987022,...</td>\n      <td>justinbieber</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>None</td>\n      <td>Fri Jul 07 19:27:36 +0000 2017</td>\n      <td>0</td>\n      <td>883407168535502848</td>\n      <td>en</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>None</td>\n      <td>...</td>\n      <td>74</td>\n      <td>882967133092794368</td>\n      <td>bieberlandia</td>\n      <td>goddesdemetria</td>\n      <td>16</td>\n      <td>2017-07-07</td>\n      <td>0.99615</td>\n      <td>positive</td>\n      <td>[]</td>\n      <td>None</td>\n    </tr>\n  </tbody>\n</table>\n<p>2 rows \u00d7 27 columns</p>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 20
                }
            ], 
            "source": "from pyspark.sql.functions import udf\nfrom pyspark.sql.types import StringType\n\nudfAddBrand = udf(addBrand, StringType())\n\n# For purposes of separating tweets by brand, we need to run it against original TEXT and not the TEXT_CLEAN\n# This is because in several cases, the brand is referenced with a handle\nenrichedBrandsDF = enrichedBrandTweetsDF.withColumn('BRAND',udfAddBrand('TEXT'))\n\n# view top records to verify brand column extracted as expected\nenrichedBrandsDF.limit(2).toPandas()", 
            "execution_count": 20
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "\nTo visualize the difference in sentiment between brands, we'll create a separate dataframe for each brand.\n"
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from pyspark.sql.functions import col\n\n## Create one DF for katyperry, one for taylorswift\nbrand1TweetsDF = enrichedBrandsDF.where(col('BRAND') == 'katyperry')\nbrand2TweetsDF = enrichedBrandsDF.where(col('BRAND') == 'justinbieber')\nbrand3TweetsDF = enrichedBrandsDF.where(col('BRAND') == 'taylorswift')\n", 
            "execution_count": 21
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Number of tweets for katyperry:  119\nNumber of tweets for justinbieber:  681\nNumber of tweets for taylorswift:  74\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "# Print out number of tweets for each brand in our sample\nprint \"Number of tweets for katyperry: \", brand1TweetsDF.count()\nprint \"Number of tweets for justinbieber: \", brand2TweetsDF.count()\nprint \"Number of tweets for taylorswift: \", brand3TweetsDF.count()\n", 
            "execution_count": 22
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "[Row(COORDINATES_TYPE=None, CREATED_AT=u'Sat Jul 08 06:37:17 +0000 2017', FAVORITE_COUNT=0, ID=883575699524681728, LANG=u'en', PLACE_BOUNDING_BOX_TYPE=None, PLACE_COUNTRY=None, PLACE_COUNTRY_CODE=None, PLACE_FULL_NAME=None, PLACE_ID=None, PLACE_NAME=None, PLACE_PLACE_TYPE=None, PLACE_URL=None, TEXT=u'@katyperry @ArianaGrande Planning to buy shoes under \\u20b92000?\\nI have found the most stylish PUMA shoes\\n\\nClick here\\nhttps://t.co/rpON3Zt9eU', TEXT_CLEAN=u'  Planning to buy shoes under \\u20b92000?\\nI have found the most stylish PUMA shoes\\n\\nClick here\\n', USER_FAVOURITES_COUNT=314, USER_FOLLOWERS_COUNT=36, USER_FRIENDS_COUNT=72, USER_ID=803167039753953280, USER_LOCATION=u'Ranchi', USER_SCREEN_NAME=u'Prashan33972844', USER_STATUSES_COUNT=53, DAY=datetime.date(2017, 7, 8), SENTIMENT=0.6719319820404053, SENTIMENT_LABEL=u'positive', KEYWORDS=[Row(relevance=0.9092190265655518, text=u'stylish PUMA shoes'), Row(relevance=0.35054001212120056, text=u'Click')], BRAND=u'katyperry'),\n Row(COORDINATES_TYPE=None, CREATED_AT=u'Sun Jul 09 02:35:45 +0000 2017', FAVORITE_COUNT=0, ID=883877302416924672, LANG=u'en', PLACE_BOUNDING_BOX_TYPE=None, PLACE_COUNTRY=None, PLACE_COUNTRY_CODE=None, PLACE_FULL_NAME=None, PLACE_ID=None, PLACE_NAME=None, PLACE_PLACE_TYPE=None, PLACE_URL=None, TEXT=u'RT @portalkatyperry: \"Swish Swish\" feat. @NICKIMINAJ acaba de se tornar o 18\\xba single de @KatyPerry a atingir o primeiro lugar na Billboa\\u2026 ', TEXT_CLEAN=u'RT : \"Swish Swish\" feat.  acaba de se tornar o 18\\xba single de @KatyPerry a atingir o primeiro lugar na Billboa\\u2026 ', USER_FAVOURITES_COUNT=7358, USER_FOLLOWERS_COUNT=722, USER_FRIENDS_COUNT=465, USER_ID=1633001821, USER_LOCATION=u'Jeffersonian Anthropology Unit', USER_SCREEN_NAME=u'Camolesi_', USER_STATUSES_COUNT=132928, DAY=datetime.date(2017, 7, 9), SENTIMENT=0.8227199912071228, SENTIMENT_LABEL=u'positive', KEYWORDS=[], BRAND=u'katyperry')]"
                    }, 
                    "metadata": {}, 
                    "execution_count": 23
                }
            ], 
            "source": "brand1TweetsDF.head(2)", 
            "execution_count": 23
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Now, find the number of tweets with different sentiment labels (Positive, Negative, Neutral) for **katyperry**, **justinbieber**, and **taylorswift**."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "+---------------+----------+\n|SENTIMENT_LABEL|NUM_TWEETS|\n+---------------+----------+\n|       positive|        45|\n|        neutral|        50|\n|       negative|        16|\n+---------------+----------+\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "from pyspark.sql import functions as F\nfrom pyspark.sql.functions import col\n\n## First for brand1\nbrand1TweetsDF = brand1TweetsDF.where(col('SENTIMENT_LABEL').isNotNull())\nbrand1SentimentDF = brand1TweetsDF.groupBy('SENTIMENT_LABEL')\\\n                              .agg(F.count('ID')\\\n                              .alias('NUM_TWEETS'))\n\n## Take a look\nbrand1SentimentDF.show()", 
            "execution_count": 24
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "+---------------+----------+\n|SENTIMENT_LABEL|NUM_TWEETS|\n+---------------+----------+\n|       positive|       201|\n|        neutral|       195|\n|       negative|       163|\n+---------------+----------+\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "## Now for brand2\nbrand2TweetsDF = brand2TweetsDF.where(col('SENTIMENT_LABEL').isNotNull())\nbrand2SentimentDF = brand2TweetsDF.groupBy('SENTIMENT_LABEL')\\\n                                .agg(F.count('ID')\\\n                                .alias('NUM_TWEETS'))\n\nbrand2SentimentDF.show()", 
            "execution_count": 25
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "+---------------+----------+\n|SENTIMENT_LABEL|NUM_TWEETS|\n+---------------+----------+\n|       positive|        23|\n|        neutral|        21|\n|       negative|        27|\n+---------------+----------+\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "## Now for brand3\nbrand3TweetsDF = brand3TweetsDF.where(col('SENTIMENT_LABEL').isNotNull())\nbrand3SentimentDF = brand3TweetsDF.groupBy('SENTIMENT_LABEL')\\\n                                .agg(F.count('ID')\\\n                                .alias('NUM_TWEETS'))\n\nbrand3SentimentDF.show()", 
            "execution_count": 26
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "### 7.1 Compare sentiment between brand1, brand2 and brand3<a id=\"compare\"> </a>"
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/plain": "<matplotlib.figure.Figure at 0x7fb34b708850>", 
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAABRgAAAGbCAYAAACvc/x4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XecFdX9//HXZ3dhYelVpNcFFnFV\nDIoGu1ETNbEQjQVNJMEWNUaNiSbX87VG40/FaNBgAVuIqLHGJJoIsWAhSJSOBKVI70pb9vz+OLNy\nWZftu3Pv3vfz8ZjH3il35jOzc+/nzpkz55j3HhEREREREREREZHqyIo7ABEREREREREREUlfKmAU\nERERERERERGRalMBo4iIiIiIiIiIiFSbChhFRERERERERESk2lTAKCIiIiIiIiIiItWmAkYRERER\nERERERGpNhUwpjELHjGzdWb2XsyxdDezzWaWHWccdcHMHjWzm+p5m5vNrHcdrPerfTGzI8xsSTnL\nejPrW9sxiIjUhHJf/cjg3FdXMcw0syMquewiMztmD/OGm9ncSq7nBjN7vAphikiaU46sH5maI+tq\nu3Ews7+a2XlJ4zeZ2WozWx5XTA1B7AWM0Y+oLdEHZkV0ojWPO65k5f3Qi9k3gWOBrt77oaVnmllj\nM7vTzJZEx3eRmd1dGxsufUy8959575t773fWxvqrGMseC8LM7GAz+6Ksc8rMppvZpXUfYdVFx3Jh\n3HFURzpf0ESJZnM07DCz7UnjY+sphj+Z2fX1sS2Jj3JfjSj3odyXyjGUddHkvR/kvX+jRsGF9fzb\ne9+/puupKeVLqUvKkTWiHIlyZF0p77imG+/9Cd778RAKw4GfAwXe+05m1jPa15yy3mtmY5Ny3vYo\nD5aM/7U+4jez28xsXH1sqypiL2CMnOS9bw4cABwIVPnHwp7++Q1cD2CR9/6LPcz/JeF4DgVaAEcA\n/6mf0FKD934qsAQ4PXm6me0DFABPxRGX1I2y7pBW5a5plGiaR99HTwC3l4x77y+szVhFUO6rLuW+\nCij3SW0r/V2jfCn1QDmyepQjK6Acmd7q6HPdHVjjvV9ZmYW99xcm5cBbgIlJOfCEOogvbaRKASMA\n3vulwF+BfQDMrJWZPWRmn5vZ0qjaanY073wze8vM7jKzNcAN0fQfm9lsM9tkZrPM7IBoemcze8bM\nVpnZ/8zsspLtRjWu/mxmE6L3zTSzA6N5jxFOuBejEulroulPm9lyM9tgZlPMbFDS+tqZ2YtmttHM\n3o/ifjNp/gAz+4eZrTWzuWb2/T0dkyjuF6JlF5jZj6PpFwDjgGFRXK6Mt38DeM57v8wHi7z3E0qt\nu1aOSelSfjN7I9rvt6NlXoyOyxNJx6VnZY6JhbuW95nZy1Es75pZn2jelGixGdF2zijjOIwHRpaa\nNhJ4xXu/JlrPHv+fpf4f5yf/L6NpX93JMbNcM/udmX1m4Y7rWDNrGs1rb2Yvmdn6aD//bWZlfgZL\nrXOP+7+H91ZqX6rCzL5pZosterzLzO6Jxjea2TQzGx5NPx74FXBG9P+YYWYjzGxaqfVdaWbPJ+3f\n2Oj/v8nMJptZj6RlKzo3/mBmr5jZF8CRZUy7MvpfZCe971Qzm1GN4/CumX0nen109H86Ohr/jplN\nTVp2dBTv2uh/1yVp3j5m9k8Lj6/MNrPvRdMvA04Dfh0dv6ej6b+28D24MVp+eFVjl9Sl3Pd1ptyn\n3BdT7isVwxtmNqqs42DBXWa2MvrffmThu/0nwNnANSXnQLT8VzV7yjvPknzDwmd5nYXHHZtE7z3C\nkh5fK+98jjQxs4nRdv5jZoWVeW8U4yQze9zMNgLnV/E4Kl9KrVCO/DpTjlSOjCFHlnVczaxNFP8q\nC9/TL5lZ12j5cq8By1j/j6PzeW10fncute+XmNl8YL4FZeXgXtGxzIre90czW5m0nsfM7Iro9Rtm\nNspCbv4H0Dnar0eBkn1dH00bVsVjNdHMLole94nivyAaH2RJj2Gb2Slm9t8o7n+bWUHSvG5m9ryF\nR7cXmtmF0fTvAVcC50XxvZd0DBdF58RCMxtRlbhrhfc+1gFYBBwTve4GzARujMafAx4AmgEdgfeA\n0dG884Ei4KdADtAUGAEsJXxxGtCXcBcnC5gG/AZoDPQGFgLHReu6AdgKfBvIBm4FppYVY9K0HxHu\n+uQCdwMfJs37UzTkEe6CLAbejOY1i8Z/GMW9P7CaUB23rOMzBbgfaALsB6wCjko6Bm+Wc2yvBz4D\nLgYGA5Y0r1aPCdAT8EBONP4GsADoA7QCZgHzgGOi/Z4APFKZYwI8Cqwh3GnLIdwp/1PStj3Qt5zj\n0C06V7ol7fsS4HuV/H8+Cty0p2OevH3gLuAFoG20vheBW6N5twJjgUbRMDz5f1LOOsvd/zLeW9l9\nOQJYUs56POEzdHz0/xmaNO8coF0Uz8+B5UCTpHPn8aRlc4G1wMCkadOB05Ji2gQcFi17D5X8vETv\n3QAcGv1fm+xh2izghKTtPwf8vILvpq+OVdK024E7otf/B3wCuKR5v41enwHMBvKj//VNwL+ieS2B\nzwkXoNmE76u1Sf/vPwHXJ22zkPDZ3IvwvdYb6FWX38sa6n5AuU+5T7mvrJhTJvcl/T9HJc376jgA\nxxHOpdaEz91AYO/S29vDZ/4GKj7PPo7+h22Bt8qKn8qdzzsItXQaAVcB/4teV/a934uWbVrOMStr\nf5UvNVR7QDlSOVI5sqyYUypHRuPtCDd78qL1Pw38JZpXmWvAku0eFf1/D4jedy8wpdR2/xEdw6aU\nn4M/A4ZEr+cSzuGBSfP2TzofR5W135Q6dyv4vrqBpGvfaNrFwNNJx/4TYHzSvInR64MJeW4I4TP1\nE8JnIica/wj4BeHzmB/Ff3j03tuAcUnbbAOsB/pE452Tj319DamSQDZHB+NTwpdlU8KPg20k/aAB\nfsCuHx3nA5+VWtffgMvL2MZBZSz7S3Z9gd0AvJY0rwDYUirGY8rZh9bRCdgqOhF2AP2T5t/ErgRy\nBvDvUu9/AEiUsd5uwE6gRdK0W4FHk45BeQkkG7iE8MN0G7AMOK8ujgllJ5DrkubfCfw1afwkoi+2\nio4J4csn+cPzbWBO0ni5CSRa5jXgV9HrYwmJuFFF/8+k7VeYQAhfbl8QfaijecOA/0Wv/w94vqJY\nS+9TRftfwXrK25cjqDiB/JLwudyngu2sAwqTzp3SX7J/AG6OXg+Kls9Niin5B0FzwnnfrZLnxoRS\n88ua9gvgieh1W+BLoiRUzj59daySpn0HeC/pHB8FvBGNvwt8O3r9L+DspPc1Inwv7AWcB/yj1HrH\nA7+IXpe+YBpESDxHUokkpyE9BpT7dvssl5qu3OeV+yra/wrWU9PcV5kCxqMIFwEHA1ml1vHV9so6\ndyp5nl1Yat8/KR0/lTufky/+swj5ZHgl3zsleX45x6ys/VW+1FDtAeVIUI5Ujvx6DCmVI/cwfz9g\nXdJ4RdeAJdt9iNDURsn7mhM+Mz2TtntU0vzycvBjhNp9nQgFjLcDFwK9CN8pWUnnY10VMA4CViTt\n54+BhdH4RODi6PUjJH0mommfEj6LhwPzS81zwB+i13sqYPwuUcWfOIZUaZfie97715InmNlgwo+M\nz82sZHIW4U5GieTXEL5wPylj/T0IVV7XJ03LBv6dNJ7cW9CXhEdKcrz3RaVXZqEa/s2EO2IdgOJo\nVntC8sspJ84ewEGlYskhfBBK6wys9d5vSpr2KaHdjAr50KDufcB9UTXsHwEPR1Voa/WY7MGKpNdb\nyhgvaVi3MsekdCxVbeh5POHR3VuAcwkFWjugwv/nhipsowPh7s20pHPWCMcV4A7CF9Dfo/kPeu9v\nq+S6K7X/tbgvJa4gFNZ9XGo7VwEXEM5RT6hl0L6c9YwHnrLQGPu5wJ+999uS5n/1GfHebzaztdG6\nK3NulP4eKGva48BsM2sGfJ/wg+XzcuLdkzeBQjNrD/SP9uvGaLwwmk8U91gzuy/pvUVA12jeYWXs\n07qyNui9n2lm1xL+rwMsNBx8pfd+RVnLS1pR7lPuU+4rX1y5r0Le+3+a2e8J51oPM3sWuMp7v7GS\nq6joPEv+/HxK+FyUVpnzOTm/Flt4vLokd1f6vdWgfCk1pRypHKkcWb7Yc6SZ5RFqaB5PKOACaGFm\n2dG5VtE1YInOJLUFGl0PrgG6EAquYfd8Vl4OngycTKiROoVQkHguofbtv733Jftfl2YBWdHjzt8k\nPEHwUwtNABxGKFiGcJ5/38yuTnpvY8J+Nwd6lvF53O17sYT3fp2ZnU0oXB1v4ZH2K733C2ptryoh\nVQoYy7KYcFelfTlfWL6M95TV9sBiwl2CftWMpfR2ziKUDB9DOOFbEX7sGOHOR8kPo3nR8t1KxTLZ\ne39sJba7DGhrZi2Skkh3QjX/qu2A91sIicSxq1p+bR6TmqjKMamuZ4H7zexI4FTCXYoS5f0/S/uC\nkCQAMLNOSfNWExLjIB/ai9lN9D/8OfBzC40I/9PM3vfev1793fqaquxLZYwAHjKzJd77ewAstGd0\nDXA0MDO6YEnextfODe/9VDPbTqg1cVY0JPvqM2KhR7e2hPO/MudGWefibtO890vN7B3C//5cwt20\nKvPebzCzjwlf3NO89zvM7INo/OOkC8vFwNXe+2dKr8PM9gX+7r0/qbL740MPZ+PNrDXhDt9NhDth\n0vAo9yn31SblvprZbb8JtSG+4r0fA4wxs47An4GrgV9TO+dJ8uenO+FzUVplzufk/JpF+IwuI3xe\nK3pvtfdD+VLqiHKkcmRtUo6suZ8TbiId5L1fbmb7ER6DNqjUNWCJZYTCNgCiSiHt2P28Ln19t6cc\nPJlQcLskev0m4TH0rdF4ZdToXPbe+6iA7xxgq/d+tZlNJjwCnUMogIRwnr/svb+z9Dqi83KO935w\nZWP03r8MvBwV/N5OuOaty8/Q16RUJy/JotpFfwfuNLOWZpZloYHMw8t52zjgKjMbYkFfC51FvAds\nMrNfmFlTM8u20AjoNyoZzgpCGxQlWhCS2xrCl8ktSXHvJHxZ3WBmeWY2gN0bkH0JyDezc82sUTR8\nw8wGlnEMFgNvA7eaWZPoh9YFhNpYFTKzKyw0Bt7UzHLM7Lwo9ul1cExqotLHpLqx+NCT2iRCNeRP\nvfcfJM3e4/+zDDOAQWa2n4UG129I2kYx8EfgruiLDjPrYmbHRa9PjM5JI9wt2smuO0i1pSr7UhnL\nCAWJl5vZRUnbKCL8WMoxs98QajCWWEG421L6+2UC8Htgh/f+zVLzvm2hI5nGwI2ER7oWU/Nzo/T2\nryG0N/NsNd5fYjJwKbsS1BulxiEksevNrD+AhQaQT4vm/QXY30LDyI3MrLGZHWxm+dH83c5nMysw\ns8PNLJfwA2ULtX/eSIpQ7lPuU+6rltrOfSU+BE6Nzum+hPMQgOh/dZCZNSJcXG5l137VxnlyiZl1\nNbO2wHWER6pKq8z5PMRCx2Y5hKcStgFTK/nemlK+lFqlHKkcqRxZLbWZI8s677cQOkNpCyTKeE95\n14AlngJ+GB3D3CjGd733i8pauLwc7L2fH8V0DqGQemMU92lUvoBxVbS+mpzPe8qBU7z3JYWDDxJq\nNh4YfT81N7OTLRQQlnQqd0X0Wc8xs30t6qQq2qde0blTcl59J3rvNkIzE/WeA1O2gDEyklBFdBah\nlH0SsPeeFvbeP02o/vskodOIvwBtoy/1EwltAvyPcCdhHKH0vjJuJfz4WW/h0dAJhKroS6PYppZa\n/tJo3csJ1bifIvyTS+5SfAs4k1B4sxz4LaEx07L8gNAGwDJCo8aJ0o8LlONLQtsWywn7fAmhUdWF\ndXBMqq0ax6S0Gwh3q9dbOT2uEapo9yD8/5JV9P9MjnUeoUrza8B8dj3iU+IXhMaLp1rodfE1wl0d\ngH7R+GbgHeB+7/2/Ktq5Kqr0vlSW9/4zQiHjtRZ60/wb8CrhDuynhC/05Mc8no7+rjGz/yRNf4zQ\nA2BZP4CeJCSktYRGbs+Jtl3TcyPZc4T//3Pe+y+r8f4SkwnJdMoexvHeP0VIpM9G58GHRHePvPfr\nCA0T/5DQVtQyQg2LRtHbHyT0HrrezP5EeKzmTsJn9HNCdflf1yB+SX3Kfcp9lXEDyn0laj33Re4C\nthN+xI8nNKJfoiXhonFdtO01hBoTEGrOFUT/m79Uc9tPEgpSFhIe77yp9AKVPJ+fJ7Rlto5Qg/9U\n7/2OWvgsVIbypdQF5UjlyMq4AeXIErWZI29g9+N6N+G7d3W03lfLeE9514AAROfvr4FnCN/ffQj/\n+z0pLwdDyDdrogL5knEj6THs8kTXijcDb0X7enBl3ldK6Zw3hZCXknPgW8BlhPZF1xOur88Ks/wO\nQtuah0T7uIpQI7HkUfiSTqPWmtnbhMenryV8VtYQOra6tBpx14jtKjyVumJmvwU6ee/PizsWkbhY\naOdlJXBAdGepZPqjhEZ1r6+HGD4h9DRY2R9hIlJNyn2SbizUvN8J9IhuromI1AnlSMkUe7oGlIYp\n1WswpiUzGxBVXzUzG0qotv5c3HGJxOwi4P24EouFR6488M84ti/S0Cn3SQOwD6FW/vKKFhQRqQrl\nSMlgsV4DSv1K5U5e0lkLQrX3zoTHau4kPKIikpHMbBGhWvr3Ytr+G4TGq8/19dNzmEgmUu6TtBXd\nhHoQ+IX3fnvc8YhIg6McKRkn7mtAqX96RFpERERERERERESqTY9Ii4iIiIiIiIiISLWpgFFERERE\nRERERESqTQWMIiIiIiIiIiIiUm0qYBQREREREREREZFqUwGjiIiIiIiIiIiIVJsKGEVERERERERE\nRKTaVMAoIiIiIiIiIiIi1ZYTdwAiItUxbdq0jjk5OeOAfdDNkqooBj4uKioaNWTIkJVxByMiIjWn\nnFhtyokiIg2McmK11TgnqoBRRNJSTk7OuE6dOg3s0KHDuqysLB93POmiuLjYVq1aVbB8+fJxwMlx\nxyMiIjWnnFg9yokiIg2PcmL11EZOVGmuiKSrfTp06LBRSaNqsrKyfIcOHTYQ7uiJiEjDoJxYDcqJ\nIiINknJiNdRGTlQBo4ikqywljeqJjpu+/0VEGg7lxGpSThQRaXCUE6uppjlRyVREJEa33357h9//\n/vftAMaMGdNu0aJFjUrmnXHGGT2mTZvWJL7oRERE6o9yooiISJCOOVFtMIpIg2DOhtTm+nzCT6vN\n9e3JNddcs6rk9eOPP95+v/3229KzZ88dABMnTvy0PmIQEZGGRTlRREQkUE6sP6rBKCJSTXPnzm3c\nq1evQSeffHKv3r17Dzr++ON7b9q0Kev5559vMXDgwIL8/PyCESNG9NyyZYsBXHzxxV369OkzKD8/\nv+AnP/lJV4Arr7yy829+85u9HnnkkTYff/xx3siRI3sPGDCgYPPmzTZ06ND+U6ZMybv99ts7jB49\numvJdseMGdNu5MiR3QHuv//+toMHDx44YMCAgrPOOqtHUVFRPAdDREQymnKiiIhIkKk5UQWMIiI1\nsGjRoiaXXnrpyoULF85s0aJF8Y033rjX6NGje02cOPGTefPmzSoqKuKOO+7osHz58uxXXnmlzfz5\n82fOmzdv1i233PJ58np++MMfrttnn32+nDBhwsI5c+bMat68+Vfthpxzzjnr/vrXv7YuGZ80aVLb\ns88+e+1//vOfJpMmTWr7wQcfzJkzZ86srKwsP3bs2Hb1uf8iIiIllBNFRESCTMyJKmAUEamBTp06\nbf/Wt771BcC55567ZvLkyS26du26bd99990GcP7556958803W7Rr125nbm5u8RlnnNFz/PjxrZs3\nb15c2W107ty5qFu3bttef/31ZsuXL8/+5JNPmhx77LGbX3311RYff/xxXmFh4cABAwYUvPnmmy0X\nLlyYW1f7KiIiUh7lRBERkSATc6LaYBQRqQEz2228ZcuWO9etW/e179ZGjRrx4Ycfzn7hhRdaTpo0\nqc0f/vCHjlOnTp1X2e2MGDFi7VNPPdVmwIABW0844YR1WVlZeO9txIgRa+67776ltbArIiIiNaKc\nKCIiEmRiTlQNRhGRGvj8888bv/baa80AnnjiibYHHHDAF0uXLm388ccf5wJMmDCh3fDhwzdt2LAh\na+3atdlnnHHGhrFjxy6eM2dOXul1NW/efOeGDRuyy9rO2Wefvf5vf/tb66effrrt2WefvRbg+OOP\n3/jSSy+1Wbp0aQ7AihUrsufNm9e47vZWRERkz5QTRUREgkzMiSpgFBGpgZ49e2699957O/bu3XvQ\n+vXrc66//vqVY8eOXTRixIg++fn5BVlZWVx11VWr1q9fn3388cf3y8/PLxg2bFj/G2+8cXHpdY0c\nOXL1T3/60x4ljfcmz+vQocPOvn37bl26dGnukUce+SXAkCFDtl5//fVLjz766Pz8/PyCo446Kn/x\n4sWN6mvfRUREkiknioiIBJmYE817X/FSIiIpZsaMGYsKCwtXxxnD3LlzG5944on95s+fPzPOOKpj\nxowZ7QsLC3vGHYeIiNSccmLNKCeKiDQcyok1U5OcqBqMIiIiIiIiIiIiUm0qYBQRqab+/ftvT8e7\nUiIiIrVNOVFERCTI1JyoAkYRERERERERERGpNhUwioiIiIiIiIiISLWpgFFERERERERERESqTQWM\nIiIiIiIiIiIiUm0qYBQRidncuXMbjx07tm113puXl7d/bccjIiISF+VEERGRIN1yYk59b1BE0o+Z\nXQh86b2fYGbnA3/33i+L5o0D/p/3flacMTrnhtTm+hKJxLTaXF955s+fnztx4sS2F1544drS83bs\n2EGjRo3qKxQREWkAlBNFREQC5cT6oxqMIlIh7/1Y7/2EaPR8oHPSvFFxFy7GZe7cuY179+496Mwz\nz+zRt2/fQYceemi/zZs328yZM3OHDx/eb9CgQQOHDBnSf/r06U0ATjvttJ6PPPJIm5L3l9xVuu66\n67p88MEHzQcMGFDgnOs4ZsyYdkcddVTfgw8+OP+QQw7pv2HDhqxhw4blFxQUDMzPzy94/PHHW8e1\nzyIiImVRThQREQkyNSeqgFGkgTOznmY2x8yeMLPZZjbJzPLM7Ggzm25mH5nZw2aWGy1/m5nNMrP/\nmtnvomk3mNlVZnY6cCDwhJl9aGZNzewNMzvQzC40szuStnu+mf0+en2Omb0XvecBM8uO41jUhc8+\n+6zJZZddtnLBggUzW7VqtXPChAltRo0a1eP+++//bObMmbPvuOOOJRdddFH38tZx8803Lz3wwAM3\nz5kzZ1YikVgJMHPmzLznn3/+k/fff39uXl5e8csvv7xg1qxZsydPnjzvV7/6Vdfi4uL62UEREZFK\nUk4UEREJMjEn6hFpkczQH7jAe/+WmT0MXAmMBo723s8zswnARWb2GHAKMMB7781stzsg3vtJZnYp\ncJX3/gMAMyuZ/QzwDnB1NH4GcLOZDYxeH+q932Fm9wNnAxNoALp06bLtkEMO2QKw//77f7lo0aLc\n6dOnNx8xYkSfkmW2b99ue15D2YYPH75xr7322glQXFxsV1xxRdepU6c2z8rKYuXKlY2XLFmS0717\n96La2xMREZGaUU4UEREJMjEnqoBRJDMs9t6/Fb1+HPg18D/v/bxo2njgEuD3wFbgITN7CXipshvw\n3q8ys4VmdjAwHxgAvBWtdwjwflQY2RRYWfNdSg2NGzf2Ja+zs7P9ihUrclq0aFE0Z86crz02npOT\n43fu3AnAzp072bFjxx4TSl5e3le3nh544IG2a9asyfnoo49m5+bm+i5dugzesmWLaqCLiEhKUU4U\nEREJMjEnKhmLZAZfanx9mQt5XwQMBSYBJwKvVnE7fwK+D5wGPOe994AB4733+0VDf+/9DVVcb9po\n2bJlcdeuXbc//PDDbQCKi4t55513mgL06NFj+7Rp0/IAnnzyydZFRUUG0KpVq52bN2/e42PjGzZs\nyG7fvv2O3Nxc/+KLL7ZYtmxZ4/rYFxERkZpQThQREQkyISeqgFEkM3Q3s2HR67OAD4CeZtY3mnYu\nMNnMmgOtvPevAD8DCstY1yagxR628xzwXeAHhMJGgNeB082sI4CZtTWzHjXdoVT21FNPLXzkkUfa\n9+/fv6Bfv36DnnnmmdYAP/3pT1e9/fbbLfr371/w9ttvN2vatGkxwNChQ7dkZ2f7/v37FzjnOpZe\n36hRo9bOmDGjWX5+fsH48ePb9erVa2t975OIiEh1KCeKiIgEDT0nWqhgJCINlZn1JNRE/IDwqPIs\nQoHiMOB3hKYS3gcuAtoCzwNNCDUPf+e9H29mNwCbvfe/M7PTgFuALdE6/srubTK+BBR473snxXAG\n8EvCTY0dwCXe+6k12a8ZM2YsKiwsXF2TdWSyGTNmtC8sLOwZdxwiIlJzyok1o5woItJwKCfWTE1y\notpgFMkMRd77c0pNex3Yv9S0zwmPSO8m+ZFm7/0zhA5dShxRatkTy3j/RGBilSIWERERERERkbSg\nR6RFRERERGJmZhea2cjo9flm1jlp3jgzK4gvOhEREZHyqQajSAPnvV8E7BN3HCIiIrJn3vuxSaPn\nAx8Dy6J5o+KISURERKSyVMAoIiIiIlIDSe0dTwMOAGYCIymjvWPv/TYzuw04GSgC/u69v6qkvWNg\nEXAg8ISZ7dbecTS9j/f+6mi75wMHfvjhh6xcubLtqlWr9vLeW15e3he9evX61MzqY/dFREREVMAo\nmcWc5QCdgPZAY6BROX+TX3tgI7ChjGGjT/iiet0RERGRGjJnBjQDWkZDizL+GqFzrqLob/Lr5Glb\ngdXACp/wG+p1R1JHf+AC7/1bZvYwcCUwGjjaez/PzCYAF5nZY8ApwADvvTez1skr8d5PMrNL2b0D\ntZLZzwDvAFdH42cANxcVFZ2ybt26ZgMHDpyTlZXlFy5c2H3VqlXtOnbsuKaud1pEpCGIcmJboCPQ\njtDpZaMKBiPkv62EDjC3Em4UbSJcO24ENviEV2/3khFUwCgNgjnLBvYCOgN7R39LD3sDHaiDtkfN\n2ZfsXui4EvgkGhZEfxf5hN9R29sWEREpzZztDfQGegDdgK7R326Ei6eWQHPqJiduJeTBlcCKpKFk\nfBEw2yf8+tredswWe+/fil5xhECnAAAgAElEQVQ/Dvwa+J/3fl40bTxwCfB7wkXoQ2b2EvBSZTfg\nvV9lZgvN7GBgPjAAeGvbtm1NsrOz82bNmjUQoLi4OKtRo0a6+SkiApiz9kB+NHQnXBN2jP6WvG4H\nZNfR9jcBS8oYlpa89gmvG0KS9lTAKGnHnHUGCqNh3+hvPvGez3nRsHc5y+w0Z5+xe6Fjyet5PuG3\n1XmUklZWr16dPW7cuLbXXnvtKoBFixY1uvDCC7u9+uqrC+OOTUTiZ86yCLXmDgD2A/oBfQgFi3kx\nhtaEcAHXvbyFzNlyYHbpwSf8sjqPsG74UuPrCResuy/kfZGZDQWOBk4HLgWOqsJ2/gR8H5gDPOe9\n92+//TZt2rRZ06NHj6XVCz31KSeKSHnMWTN2FSImD/2ANjGGBuGJgIHRUCZztpmQBz8itMH7EfCx\nT/jl9RKhpJVUzYkqYJSUZc5ygUHsKkTcNxraxxlXDWQDvaLhmFLztpuz6cBUwqNPU33Cf1rP8aU3\nsyG1uj7vp9Xq+qphzZo12Q899FDHksTRs2fPHXEnDRGJR1RTv4BQmDiEXYWKzeKMq4Y6RcORyRPN\n2QZC4dlMQk580yf8nPoPr8q6m9kw7/07wFnAB8BoM+vrvV8AnAtMNrPmQJ73/hUzewso63t9E+GC\ntCzPAdcB+wO/AMjNzd26YcOGdtu3b1/RuHHjIuVEEWnIopy4D3BQNAwl5Mhar5Vfj5oD34iGr5iz\n1YQCx5JCx3eBj3zCF9d7hOlKObHepPMHUBoYc9bcnJ1ozu42ZzMI7VdMAx4BriDc3U/XwsWKNCYk\nx8sJNRMWmbNl5uxZc3a1ORtuzprGG6KUNnfu3Ma9e/cedOaZZ/bo27fvoEMPPbTf5s2bbebMmbnD\nhw/vN2jQoIFDhgzpP3369CYAM2fOzC0sLByQn59fcNlll3XOy8vbH2DDhg1Zw4YNyy8oKBiYn59f\n8Pjjj7cG+PnPf9518eLFuQMGDCgYPXp017lz5zbu16/fIIDCwsIBH3zwQZOSWIYOHdp/ypQpeRs3\nbswaMWJEz8GDBw8cOHDgV+sSkfRizjqYsx+Ys/vM2VRCgdN/gUeBnwKHkt6Fi+VpRciJPwL+CMw2\nZ6vM2fNRThxmzhrHG2KZ5gKXmNlsQm2Zu4AfAk+b2UdAMTCWUHD4kpn9F3iT0FZjaY8CY83sQ7Pd\n87/3fh2hlksP7/17ADk5OTv23nvvpfPmzcv/6KOPCupm98qnnCgidcWcdTNnI8zZHeZsCqFtww+B\nBwi5Yh8abtlGe+AIQm33Bwj7vcacvWTOfmHODjFnjeIMUL4uU3OiajBKbKIOVw4m1OY7hnDnSV+O\nu+xNaAT+lGh8hzn7L6E2xz+A133CfxFXcBJ89tlnTR5//PGFhxxyyKff/va3e0+YMKHNY4891v7B\nBx/8dPDgwdv++c9/Nrvooou6T506dd6ll17a7eKLL145evTotbfffnuHknXk5eUVv/zyywvatm1b\n/Pnnn+ccdNBBA84666z1d95555ITTzyx6Zw5c2ZBSFQl7zn11FPXPvHEE20PPPDAZZ9++mmjlStX\nNjrssMO+vPTSS7sceeSRG59++ulFq1evzj7wwAMHnnzyyRtbtmypu5wiKSyqtT8cOBb4FqHmvroA\n3qU9odflk6PxrebsfUIB3ZvA5BTIiUXe+3NKTXudUNMw2eeE3zy78d7fkPT6GUKHLiWOKLXsiaXf\n36FDh3UdOnRYF43Wbm2NSlJOFJHaED3ufCRwfDT0iTeilNMa+E40AGwxZ+8CU4B/A2/7hP8yruAk\nyMScqAJGqVfmbB92FSgexp4f/5Gva0S4YBhCuIO1zZxNBl4GXvYJ/0mcwWWqLl26bDvkkEO2AOy/\n//5fLlq0KHf69OnNR4wY8dUPoe3btxvA9OnTm//9739fADBq1Kg1N9xwQ1eA4uJiu+KKK7pOnTq1\neVZWFitXrmy8ZMmScr+fR44cue7YY4/Nv+uuu5ZNmDChzUknnbQO4I033mj5t7/9rfWYMWM6AWzb\nts0WLFjQ+IADDlDvdSIpxpzty64CxeGAaqpXXhPCMRsejW81Z68BzwMv+oRfEVtkGUw5UUSqy5wN\nJhQmHgd8E8iNN6K00pRwI+qIaHyLOXudDM6Jy5cv75CVlVXcsWPH2DrPycScqAJGqVNRLcWjgRGE\nOyyd4o2oQcklXJR+C7jHnM0F/gI8C7zvE750Y/NSBxo3bvzVcc7OzvYrVqzIadGiRVHJ3aTKeOCB\nB9quWbMm56OPPpqdm5vru3TpMnjLli3lPubRq1evHa1bty569913mz777LNtx44d+ymA955JkyYt\nKCwsVKdBIikmajPqKOBM4NsoJ9amJsCJ0VAc1eR4Hni+Ptpv9N4vIjyil9GUE0WksqLHeo8jPK11\nHNAl3ogalKaUnRNf8Ak/O9bI6kmnTp1WxR1DJubEhtpOgcTInGWbs6PN2QOEx4BeBS5AF1J1rT+h\nsfd3gU/N2RhzdkR0QSv1pGXLlsVdu3bd/vDDD7cBKC4u5p133mkKsN9++21+9NFH2wA8/PDDbUve\ns2HDhuz27dvvyM3N9S+++GKLZcuWNQZo1arVzi+++GKP39OnnXba2ltuuaXTpk2bsg866KAtAEce\neeTGO++8c6/i4lDT/a233lKNKJGYRW0G3gssBf5OaC9KObHuZAHDgNsI7TfONWe3m7NDzZkeO69H\nyokiksycZUXXJw8Ay4EXCTlRhYt1JzknzjJn86K2LA+OOa49MrOeZjbHzJ4ws9lmNsnM8szsaDOb\nbmYfmdnDZpYbLX+bmc0ys/+a2e8AFi9e3Hnp0qV7rV69Ou4exL+SCTlRBYxSa8zZAebsHsIF1GvA\nT2i4nbKkum6ETgD+BSwxZ7eas94xx5QxnnrqqYWPPPJI+/79+xf069dv0DPPPNMa4N5771187733\n7pWfn1+wYMGCJs2bN98JMGrUqLUzZsxolp+fXzB+/Ph2vXr12grQqVOnnUOGDNncr1+/QaNHj+5a\nejvnnHPOupdffrntd7/73bUl02677bZlRUVFNmDAgIK+ffsOuv766/WDTSQG5mxw9N37P+BtQtMW\ne8UcVqbKB64mtNX4P3OWMGfdY44pYygniog5G2LO7gQ+I1yf/ARoW/67pI70A64C3jFnc8zZteas\nc9xBlaE/cL/3fiChU58rCR2gneG9H0x4GvciM2tHqAU7yHu/L3BT8krat2+/jhTS0HOieT1FKTUQ\nfRmdA5yLHg1KdR74J/Ag8Bef8NtjjqdGZsyYsaiwsHB13HFUxaZNm7KaNWtWnJWVxYMPPthm4sSJ\nbV9//fVY2s6cMWNG+8LCwp5xbFukoTJnPQg58QfAoJjDkfIVEzpMe4jwGLVyYj1TThRp2MxZT+B8\nQk7MjzMWqdBOwhMWj5ACOdHMegJTvPfdo/GjgF8D2d77w6JpRwOXAN8HpkXDS8BLH3744by2bds2\nzsrK2tmlS5cVs2fP7t+1a9fFLVq0+BIgeXzOnDn9OnfuvKxp06ZbZ82aVbDvvvt+tHz58g4rVqzY\nOycnpwiguLg4q02bNmu7deu2rK72uaHkRLXBKFUWPV50PHAZof0/1YRND0ZoD/NoYJU5Gw/80Sf8\nvHjDyhxvvfVW3uWXX97de0/Lli13Pvroo4vijklEas6cHUXIiSehnJgusghtfh0HrDZnjwEP+YSf\nGW9YmUM5UaThia4TjyE8SfUdlBPTRTZwQjSsNWdPAo/4hP9PjDGVrgm3Hmj3tYW8LzKzoYRr3NMJ\nT4xUWps2bdauXbu2TZMmTba2atVqnYWWVKxNmzZrevTosbSasVdZQ8mJKmCUSjNnuYSaij8DCmIO\nR2qmA6Fq/FVRT9R/BJ7xCa9eFevQ8ccfv3nu3LmVbtRXRFKXOcsj1Fb8KarBn+7aE37b/CxqCH8c\n8KRP+C/jDathU04UaTjMWQvgPEKNsgExhyM105ZQSHepOZsG/D/gzz7hi+o5ju5mNsx7/w5wFvAB\nMNrM+nrvFxDKJSabWXMgz3v/ipm9BSwENiWvKCsra+fOnTvL7Jegbdu262fPnr33li1btnfp0mUJ\nQMuWLTd+8sknfbdv376icePGRTt27MjeuXNndpMmTeqsZmdDyYkqYJQKmbMOwMXR0DHmcKT2HR4N\nY6JajXf6hK+3uzUiIukkegz6EkLnZWo/quE5KBp+a87uB8b4hI+9J0oRkVRkzgYQCqNGAi1iDkdq\n3xDgCeBWc3Y34em3zfW07bnAJWb2MDCL8KTIVOBpM8sB3gfGEn6LPW9mTQhP7F0JuOQVtWvXbvVn\nn33WY8mSJcUDBw7crRftRo0a7czNzd26devWpi1btvwSoFmzZlv33nvvpfPmzcv33mNmvnv37p/V\nZQFjQ6E2GGWPzNlAwgf0HKBJzOFI/dlGaH/jVp/wn8UdzJ7MmDFj4eDBg9dlZWXpS6yKiouL7aOP\nPmpTWFiojn9EKsmcHUH4cXsy4VEiyQxbgfHA73zCL4g7mD1RTqw+5USRqjNnxxI6zzqGUKgjmWE9\noT3/e3zC11l7hFEbjC9576v1hIhyYvXVNCeqTQT5GnN2jDl7BZgJjEKFi5kmF7gQWGDOxpmzXnEH\ntAcfr1q1qlVxcbF+1FRBcXGxrVq1qhXwcdyxiKQDc3asOZtK6PXyFFS4mGmaAKOBueZsojkbHHdA\ne6CcWA3KiSJVY86+Y87eIXQIciwqXMw0rYFrgEXmbLxyYsNSGzlRNRjlK+bsOOBWYP+4Y5GUUkSo\nGn+zT/j5cQdTYtq0aR1zcnLGEdo+082SyisGPi4qKho1ZMiQlXEHI5KqzNk3gZsITUiIlPDAC8CN\nPuGnxR1MCeXEalNOFKmEkW7kCY/x2E3AAXHHIinnL8B1PuFTpv1A5cRqq3FOVAGjYM72B24nVHEX\n2ZOdwJ+Am3zCz4k7GBGRumDODiQULB4XdyyS8l4BrvUJ/1HcgYiI1AXn3AnA/3n8kAd5cMHnfN4v\n7pgkJRUDjwG/SeUmtqTuqYAxg5mznoSLqLNQ9XapvGLgacD5hJ9d0cIiIukgesznRuC7ccciaWUn\nod3i633Cr4g7GBGR2uCcGw7cBhxSMm0ta98Zw5hh8UUlaWAb8AfCk2+r4w5G6p8KGDOQOWsLXEfo\nBTM35nAkfRUBvwcSPuE3xh2MiEh1mLN8Qm+DZ6CbbVJ9mwkX43f6hN8adzAiItXhnOsB3AGMKD3P\n4/0f+eOCZSxTLUapyCbgTkJOrK9epyUFqIAxg5izJoQeMH9JaKBVpDasAH4BTPAJfaGISHowZ82A\n3wA/AxrFHI40HJ8BvwKeVE4UkXThnGsGXAtcRTkdfK5l7dQxjDm43gKTdLeS8HTIH3zC74w7GKl7\nKmDMAOYsCziX8OHuFnM40nC9DVzqE3563IGIiJTHnJ0C3A10jzsWabDeB670Cf9m3IGIiOyJc86A\nswk1sLtUtLzH+3GMm7+Upfl1Hpw0JDOAS3zCvxV3IFK3VMDYwJmzAcBDJLWfIVKHioEHCD2JrYs7\nGBGRZOasF3Av8J24Y5GM8QzwM5/wi+MOREQkmXNuKHAPUKUaiarFKNXkgQnANT7hq9VDsaQ+FTA2\nUOYsB7gaSKB2FqX+rSa08znOJ3xx3MGISGYzZ7nANYQmQprGHI5kno3A1T7hH4w7EBER51xzQo3F\ni6lG28OqxSg1tI7QvNY4NSXS8KiAsQEyZ4XAw8ABccciGe8DQnX49+IOREQykzk7ltAhlS6EJG6v\nA6N8wi+KOxARyUzOuaOBcUDPmqxHtRilFkwBRvuEnxN3IFJ7VMDYgJizxsCvCXcE1GC9pIpi4LeE\n3qZ3xB2MiGQGc9YWuA84M+5YRJJ8QehI4T7V3BCR+uKca0no1XdUbawvqsU4bylL+9fG+iRjbQdu\nAW72CV8UdzBSc1lxByC1w5wdBEwHrkeFi5JasgiPJb5tzlSDSETqnDk7GvgvKlyU1NOM0A7oG+as\nb9zBiEjD55z7NjCTWipcBDDMTuf09bW1PslYjYEbgLeUExsG1WBMc+YsD7gJuBwVGEvq+5LQq+YD\ncQciIg1PVJP/FuBKqtGulEg920K4MXy32isWkdrmnGtD6MTl3LpYv2oxSi37ArjCJ/y4uAOR6lMB\nYxozZwcDjwN94o5FpIpeAC7wCb867kBEpGEwZwOBJ4H94o5FpIreAc72Cf+/uAMRkYbBOTcM+BPQ\nvS63s451U+/hHrXFKLXpWeDHPuHXxh2IVJ0KGNOUObsM+B16HFrS13Lghz7hX407EBFJb+bsYkJO\nVA/Rkq7WAef5hH8x7kBEJH055wy4GrgZyKnr7Xm8f4iH5i1hiWoxSm1aRsiJr8UdiFSNChjTjDlr\nRuj5S+1KSUPgCb27XuMTfmvcwYhIejFnHYGHgBPjjkWkFnjgduA6n/A74w5GRNKLc649MAE4oT63\nu451797DPQfV5zYlI3jgLuBXPuG3xR2MVI7a7Esj5mwA8B4qXJSGw4CfAtPM2b5xByMi6cOcfYvQ\nkYsKF6WhMOAXwOvmrFPcwYhI+nDOHQZ8SD0XLgK0pvXQrnSdW9/blQbPCG1qv2POesYbilSWajCm\nCXP2fUItjeZxxyJSR74ERvqEfybuQEQktZmzawmPf+lGqTRUy4EzfcJPjjsQEUldzrks4DogAWTH\nFcd61r93N3cPjWv70uCtBkb4hH8j7kCkfCpgTHHmrBFwB6GXaJGGzgO/8Ql/U9yBiEjqiZoJeRj4\nftyxiNSDnYRepn/rE/rBLiK7c861IHRulhI1+ccxbq7aYpQ6VARc6RP+3rgDkT1TAWMKM2edgT8D\nh8Ydi0g9ewr4kdplFJES5qwX8BdAzSlIpnmJUMN/XdyBiEhqcM71BF4E9ok3kl3Ws/7du7lbbTFK\nXXsYuMgn/Pa4A5Gv06NFKcqcHQFMR4WLkpl+AExWG1QiAqFtqV/xq0m55PaIOxaRGJxIaIOqV9yB\niEj8nHOHEtrlT5nCRYDWtD6oG93mxB2HNHg/Ilwn7h13IPJ1KmBMQebsLODvQMe4YxGJ0VDgfXO2\nf9yBiEh8nHPnA/9oTOMDLuXS+YYVxx2TSAz6EwoZh8QdiIjExzk3Engd6BB3LGU5jdM2xh2DZISD\ngQ/Mmdr9TDF6RDrFmLMrgP9H6DVJRNT5i0hGcs4ZcBtwTfL0T/hk8mM8dng8UYnEbjPwfZ/wf407\nEBGpP1FOvJXQ03xKe4iHZi9m8cC445CMsI1wnfjnuAORQDUYU4g5uw24CxUuiiTLA542Z9fHHYiI\n1A/nXGNCG8TXlJ7Xhz6HH8Zhb9Z/VCIpoTnwgjm7IO5ARKR+OOeaAs+SBoWLAKdx2qa4Y5CMkQs8\nZc4ujDsQCVSDMQWYsxzgj8D5MYcikuqeAM73CV8UdyAiUjecc3nAc8C39rSMx297jMfmLWTh4PqL\nTCTl/J9P+ETcQYhI3XHOtSR05nJY3LFUhWoxSgx+4xP+xriDyHQqYIyZOWtKqKVxYtyxiKSJZ4Ez\nfcLviDsQEald0YXUy8A3K1q2mOKV93DPzg1sUCPfkskeAX6iG28iDY9zrh3wNyDt2l5dz/r37uZu\ntY8n9e0e4Gc+oUKuuOgR6RiZszbAa6hwUaQqTgWeMWeN4w5ERGqPc6498E8qUbgIkEVWx0u4ZH0O\nOVvqNjKRlPZD4CVz1jzuQESk9jjnOgNTSMPCRYDWtB7ane6z445DMs7lwIToCVGJgQoYY2LOugJv\nAofEHYtIGjoJ+Is5axJ3ICJSc9GF1GSqeCHVmMYDL+TCD+smKpG0cRzwNxUyijQMzrnehOvEgrhj\nqYnTOG1z3DFIRjqHcJ3YNO5AMpEKGGNgzgYAb5PmSUMkZicQGrpX8hBJY865XsC/qWZObE/7Yady\n6hu1GpRI+jkEeMWcNYs7EBGpPudcASEn9oo7lppqRatvqBajxOQ7wD/MWau4A8k0KmCsZ+asD/Av\noFvcsYg0AMcCL+uCSiQ9OecGEi6ketdkPYMZfPgQhrxbO1GJpK3hhMel8+IORESqzjm3P+Gx6M5x\nx1Jb1KO0xOhQ4K+q3V+/VMBYj8xZF0Kbi53ijkWkATkSeNWctYg7EBGpPOfcIMJj0V1qui7D7ERO\nHNSZzvNrHplIWjsC1e4XSTtRzcW/A+3ijqU2taLV0B70mBV3HJKxhhFyoprVqicqYKwn5qw98A+g\nZ8yhiDRE3yS0P9Uy7kBEpGLOuZ6EC6kOtbVOw5pfwAVNm9FsdW2tUyRNHU1ofyo37kBEpGJRm4v/\nANrHHUtdOJVTv4g7BsloRwLPqoPQ+qECxnoQFXr8DRgYdywiDdgw4DVz1jruQERkz5xzHQkXUrX+\nCFg22V0v4ZKlWWTtqO11i6SZbwHP6YJKJLU550qecGswj0WX1opW31AtRonZCcCT5iw77kAaOhUw\n1rHoEZWXgAPijkUkA3yDUA1etTZEUpBzrhXhhlvfutpGHnmFP+JHU+tq/SJp5ARgkjlrFHcgIvJ1\nzrkOhMLFtO/QpSKncZpqMUrcTgMeNWcqA6tDOrh1KPpB9wyh0W0RqR/DCcnD4g5ERHZxzjUBXgD2\nq+ttdaXr8OM4bkpdb0ckDZwETNQFlUhqcc61JjQVMiDuWOpDS1qqFqOkgnOAP8QdREOmHxt1JPoh\n9wTh7rGI1K8zgVviDkJEAudcDvBn4LD62ubBHHzIAAZMr6/tiaSwU4D/F3cQIhI455oBL1MPN9xS\niWoxSor4iTlTTqwjKmCsOw8CI+IOQiSDXWvOfhJ3ECKZzjlnwEOEmlT1xrCcMzijZ3vaf1qf2xVJ\nUZebs4vjDkIk0znnsoA/AYfEHUt9a0nLb/Sk58y44xABfmbOLo07iIZIBYx1wJzdDlwQdxwiwv3m\nTLWIReJ1JzAyjg0b1mY0o4tyyd0Yx/ZFUswY5USR2N0BnBh3EHE5lVO3xB2DSORuc3Zc3EE0NCpg\nrGXm7Dzg6rjjEBEAsoE/m7P94w5EJBM55y4BfhZnDI1o1OdiLp5rWHGccYikgGxCe4yD4w5EJBM5\n50YBV8YdR5xa0vJA1WKUFFFynVgQdyANiQoYa5E5OxAYG3ccIrKb5sDL5qxb3IGIZBLn3GHA3XHH\nAdCKVt84i7PU6YsItABeMmed4g5EJJM4544E7o87jlSgWoySQloScmL7uANpKFTAWEvMWUfgWaBJ\n3LGIyNfsDbxizlrFHYhIJnDOdQOeBnLijqVEP/odcSiHvhV3HCIpoDvwgjnLizsQkUzgnOsHPAM0\nijuWVNCSlgf2opdqMUqq6AU8Z84axx1IQ6ACxlpgzkp6x1QNKZHUtQ/wjDnTjzuROuScawI8B3SM\nO5bSjuGYIXo0SwSAbwCPmTOLOxCRhsw51wZ4CWgTdyyp5BROUS1GSSXfBP4YdxANgQoYa8edwOFx\nByEiFToa+F3cQYg0cA8CQ+IOoiyGNRnJyPYtabk87lhEUsCpwC1xByHSUDnncoBJQH7csaSaqBbj\nx3HHIZJkpDn7ZdxBpDsVMNZQ1KnLZXHHISKVdpk5Oz3uIEQaIufcFcC5ccdRniyy9rqES9bmkLM1\n7lhEUsAvzNl34g5CpIG6BTgq7iBS1amcui3uGERKudmcfSvuINKZChhrQJ26iKSth8xZv7iDEGlI\nnHNHAXfEHUdl5JJbMJrR/4k7DpEUYMB4c9Yl7kBEGhLn3MnAVXHHkcpa0GKIajFKijFC8yHqCK2a\nzHsfdwxpyZx1AKahdheltGLCA4ItgLOBdYSHI74EOgOn8PVuD9YB9wHtovGuwElAEfAUsJHQWtLQ\naP4LwIHR+qS6/gsc5BNetZhEasg51wP4AEirXvg+5MPJf+EvauJEBKYAR/mE3xl3ICLpzjnXC/gP\n0DruWFLdJjZNu5M7U7JZFclorwPf8glfHHcg6UY1GKsh6tTlaVS4KGWZyu6X2P8ADgYuJ/QxPn0P\n72sDXBQNJ0XTFhD6eryIUBwGsBzwqHCx5vYFfh93ECLpLmpj6s+kWeEiQCGFw/djv/fijkMkBRwG\n/CbuIETSnXOuETARFS5WimoxSoo6GlB7jNWgAsbquQF16iJl2QDMBw6Ixj3wP6AgGt8PmFOF9WUD\nOwi1IksqG/8TOLLGkUpwgTk7M+4gRNLc9eyqX51WDMv6Lt8duDd7L4g7FpEUcL050y8MkZq5jfDc\nkVSS2mKUFOXM2TfjDiLdqICxiszZIcC1ccchKepV4FhC6w0QHotuQigoBGhJeNy5LOsJLXo+Anwa\nTesdTR8HHEQonNw7Wo/UlrHmrEfcQYikI+fcUOC6uOOoCcNajGJU4zzy1sYdi0jMsoAnomaARKSK\nnHMnAD+LO45004IWQ3rT+6O44xApJRt40py1jTuQdKICxiowZ82BCewqLhLZZS7QjOo9utyC8HPk\nQuA44BlgK+FMOz2aPojw+PUhhILMiVStNqTsSSvCBZU+1yJV4JzLAx7j663Kpp1ssrtfwiWfZpFV\nFHcsIjHbG5hgzqzCJUXkK865TsB4dlUzkCo4hVO2xx2DSBm6AY/GHUQ6UQFj1dwF9Ik7CElRiwmF\njHcROnX5H6EgcCtQ0mT6RsqufZgD5EWvOxPaY1xTapn3gUJgCaFW5AjgndoLP8MdCvw67iBE0szv\ngPy4g6gtzWi2/w/54dtxx5GxdhA6SPsDodOzf0XTPaGp9TGEVnOn7uH9H0bLjIleQ+go7bFofckt\nbb4ALKvF2Bue41HvtyJV9QCg2r/V1IIWQ/rQR7UYJRWdZM4ujzuIdKECxkoyZycDo+KOQ1LYMcDP\nCTURTwd6AadFf2dFy3wI9C/jvV8Q2lkEWBsNbZLmbwHmEQoYd7Dr3uiO2gtfuN6cHRp3ECLpIHoM\n7KK446ht3eh22DEcMyXuODJSDnAe4ay6kNDJ2WJC3twAXBoN+5Tx3i+BNwi/0n4cvd6COkqrmZvN\n2f5xByGSDpxzZwEnxyyYPMwAACAASURBVB1HujuFU3RlI6nqVnPWN+4g0oEKGCvBOdf2l/zyup70\nnBl3LJKGjiHUNLyHcMFT0gHMHEKHLRDaXPxDNPwZOJFdNRoBJgPDCZ/YPknLF9Zx7JklG/ijOWsc\ndyAiqcw51w54OO446sqhHHpIf/p/WPGSUqsMyI1e74wGI9TeP5xdv1ibl/HeTwi5MQ9oGr1egDpK\nq5lGwEPmLO2bQBCpS865vQh1p6WGmtP8ANVilBTVlHCdqCYQKqACxsq5O5fcoedx3oAf8IM3sslW\nGxFSvl7A2dHrtsBPgMuB77OrtbIBwFHR6wLgEnbV3Chdy/H4aJ0QfvKPjJY/qA5iz2wDgWviDkIk\nxT0IdIo7iLpiWM6ZnNm9LW0Xxx1Lxikm3Dy7g1BI2BVYB8wkPHz4OF9vPgS+3vxISYdq6iitpvYH\nro47CJEUdx/QLu4gGgrVYpQUdgR6orVC5r2veKkMFj0G9krytB3smD+BCTsXs3hATGGJSN3ZCgz2\nCb8g7kBEUo1z7hxCq3YN3g52zL+DOzptZ3uLuGPJOFsIHZmdQCgcPJLQwdksQhuMPyq1/FuE9hYP\nj8YnE27mJTd6sZNw5v6A0L7jBsJTAPolV5FtQKFP+LlxByKSapxzIwjPHkkteozH/vsJn+wbdxwi\nZdgAFPiEV0vOe6AajOVwzrUk3DPfTSMa9fsRP+pzOqe/oR4n5f+zd+dhVlT3vv/fq+cGmhkEBASZ\nQWxmAQU0ThgnVDQa4myARoziEDn33POr1Dm5ObkxiprEAY2aRGM08arEY4iJhMEBUWaZBAQEmWSe\nGuhh/f6ojbbajL33XruqPq/n6Sfdu6t2f3bc7Kr61netJZFTADzmOoRIpvF9vz7woOsc6ZJLbocx\njFnCV4NrJV0KgTYEw5zrEvSWk/jfTdVsf6hj8ZDqFlTTQmknKp9gqLSGhYlU4ft+Y4LuRUmyK7hC\n19fH4nCLo70C/Crx2Gt8tdhodfYTnNn9T+JnLY52NPXQv/sjUoHxyH5BsDT5txhM7mmcdvZ4xi9v\nTvPlac4lIql1vvHNda5DiGSY/wKaug6RTvWp3+9arp3mOkcs7CXoXITgoulToDFBh+GqxOOrqX4g\nYjuCeRhLE1+H5mQ8RAul1dSZBJO9iMhXfoVWjU6JOtTp1Z72C46+ZcwdbnG07gSLoo0hKBjOOcJz\n/As4pcrPWhztWAwzvhnuOkSmUoHxMHzfH8IxnEzlkddlJCNbD2PYVIM50v0BEQmXCcY39V2HEMkE\nvu8XE8FVo49FZzqfPYAB77nOEXm7gd8R9I9PJJg/sRNwFrAk8fjbfLVO6+fA64nvawGDE/tNJBgq\nrYXSku2/jW9idYNB5HB8378QuNZ1jigbxjBdVx/N4RZH65j4XwOczNc7/KtaD+zh6zfktDjasfq1\n8U0D1yEykeZgrIbv+1nAbKDH8ey3n/0fP8MztTezue3RtxaREHjSena06xAiLvm+b4AZfH1Gu1ix\n2NLneG7VGtZ0dZ1FxKHnrWevdx1CxCXf93OAhWgG15R7nucXrGCF5mI8kkqCCd22Af2A86v8rgJ4\nimA+41Oq2e93wJUEIwbWAxcn9nkV2EIw93EesAEVGKv3rPXsN2eFjj11MFbveo6zuAhQQMFpJZQ0\nu5iLpxlMZQpyiUh6jTS+GeA6hIhjNxDj4iKAwRTeyI0NiiiqbgZAkbj4gfGNLjMl7sai4mJaDGOY\n5mI8miyC8SV3E3T2Vz1L+R+CwuI3i4sQzEvcgWBGwaqygeEEQ667ESysNhCYTLD42tIkZg+/m4xv\n+rsOkWnUwfgNvu8XEszU07Imz7OPffN/y28bbGVr6+QkExFHFgK9rGd1kiOx4/t+PYJjooZGAvvZ\nv+iX/LJdOeUFrrOIOLIE6G49q+GLEjuJhV2WA5pCJ01e4IX5y1muCS2OxVQgl+CW8FSCzsPvUX1L\n2SsE04UY4CBB52Jfvt4BOZNgCHZdgrkdBxN0Pd6civCh9QEwwHoqqh2iDsZvG0cNi4sAtahVPJax\njc7n/OloBUqRMOtOcF9QJI5it7DLkRRQ0O2H/HC26xwiDnUBNCRM4ur/oOJiWg1jmEYFHs7hFkeb\nTbBYy3AOX+25iuDqZhxwAcGcxFWLi1oc7VidAYxwHSKTqIOxCt/3mxL8cyxK5vPuYc+cp3n6pB3s\nODmZzysiabMbaGs9u9V1EJF0SSzsMptgwIxUMYc5Uycx6WzXOUQcWQ+0t54tPeqWIhGROCbOQQ06\naacuxsPYCLzGVwuydAPOBnyCMnheYrsuicc/Bz4CLv/G88zlqzkYD5lMsNBaW4Ki4osEV0N9CEpq\nUtU6oJP17D7XQTJBjusAGeYnJLm4CFCHOr3u5M5d05k+41/8a1Cyn19EUq4IuB/4sesgImn0CCou\nVqsnPQevZvWHC1jQ13UWEQdaAHcCP3cdRCSNHkHFRSeGMcw+wAOuY2SeZgRzJX6Td5jtT058fVPP\nxFdVQ6t8n0swG7ccTkvgHoJRP7GnDsYE3/c7AR+T4qLrLnZ9+BRPtd7N7pNS+XdEJOlKgVOtZze6\nDiKSar7vnw+85TpHJrPYXU/wxBeb2NTOdRYRB3YA7axnt7kOIpJqvu8PB/7sOkec/ZE/zv+ET9TF\nKJlqN8Ex8QvXQVzTXZiv/II0dHTWpW7fu7k7/0zOfDfVf0tEkqoQ+HfXIUTS5D9dB8h0BlN3JCNz\nCinc7jqLiAP1gf/lOoRIqvm+n4U6k5y7nMs1F6NksiLg/3MdIhOowAj4vj8EuCxdf89g6p/P+Wfe\nyZ0za1M79lVukRAZaXyjleEl0nzf/y7Q33WOMMgm+5TbuX1VFllaZV7iaKyOiRID1wGdXYeIu9rU\n7tmRjvNc5xA5glHGN+1dh3At9gVG3/cN8EsXf7sBDfrfy71Z/ej3vou/LyLHLQ/dnZLoU/ficahD\nnV43cINGJUgc5aPPC4kw3/ez0Xlfxrj8W6uTiGSUXIIldmIt9gVGgrtSfVz9cYNp9F2+O2AsY9/T\nMCuRULhRd6ckqnzfHwb0dp0jbNrQZsh3+M4M1zlEHLje+OY01yFEUuT7QEfXISRQm9o91MUoGe57\nxjexnps71gXGxF2pn7rOAdCYxgPv476yHvSY5TqLiBxRDro7JRGU6OjXe/sEDWJQ/w50mO86h0ia\nZQH/7TqESLIlrhP/w3UO+Tp1MUqGywbudx3CpVgXGIErgbauQxySRVbTYQzrN5rR7+STv9N1HhE5\nrGuNb7q5DiGSZMOB012HCCuDyf0+3z+5AQ3Wuc4ikmaXGN90dx1CJMl+AHRwHUK+rja1e3Sik27m\nSSa70fimhesQrsS9wHi36wDVaUazs+7n/n2ncdpHrrOISLWy0LxTEiGJVTJ/4jpH2BlM4zGM2ZdL\n7l7XWUTSLCPPqUVOhO/7Oah7MWOpi1EyXB5wr+sQrsS2wOj7fn8yeJXMLLKaD2d4n9u4bUYeeXtc\n5xGRb7nS+Kan6xAiSXIN0NV1iCjIJbfjGMYsBKzrLCJp9H3jm+auQ4gkyQgg1vOoZbJa1CruRCfN\nxSiZbKTxTSPXIVyIbYERGOc6wLFoSctB93P/Dn2IimSkO10HEEmSUBwTw6IBDfpfwzXTXOcQSaM8\nYKzrECJJomNihlMXo2S42sBdrkO4YKyN3w123/dbA58STMIZChZr17Bm+gu80LeMslqu84gIAPuB\nltazW10HETlRvu/3Az5wnSOK/sbf3v+ADwa4ziGSJtuA1tazmiJAQsv3/SHAVNc55Oj+xJ/mLWVp\nD9c5RA5jB8ExcbfrIOkU1w7GHxGi4iKAwZg2tBlyP/dvbke7ha7ziAgABcBtrkOI1NAdrgNE1VCG\nFrei1VLXOUTSpCFws+sQIjUUy66jMLqMy4zrDCJHUB8Y4zpEusWug9H3/TrAOqCe6ywnymIrV7Jy\nxou82L+CinzXeURibjXQznq20nUQkePl+/5JwGcEwxslBSqpXP8QD+XuYU8T11lE0mAl0FHHRAkj\n3/fbAiuIbxNO6KiLUTLcWqCt9WyF6yDpEscPz1sIcXERwGCy2tN+yL/xb+tO4ZTFrvOIxFwb4BLX\nIURO0ChUXEypLLJajGXsxmyyD7rOIpIG7YBhrkOInKCxxPP6OLTUxSgZrhVwkesQ6RSrD1Df97OI\n0KIMOeS0u4mbOl7LtVN14SLi1O2uA4gcL9/3c4HRrnPEQQEF3X/ID2e5ziGSJve4DiByvBKj3G51\nnUOOTy1qFXem81zXOUSOYJTrAOkUqwIjcDlwqusQyWQwOZ3pfPZ4xq8+mZOXuc4jElPnG990dB1C\n5DhdBTR3HSIumtHsrIu5WCtLSxwMNL7p6zqEyHG6kZCPcoury7gsbjUNCZeLjG9auQ6RLnH7xzjO\ndYBUySW3423cdupVXDUti6xy13lEYsYQw0l8JfS0uEua9aHPWd3p/pHrHCJpoMVeJGzGug4gJ6YW\ntYq70EVdjJKpsolRd3RsFnnxfb8nMMd1jnQ4wIHFz/Js/kY2tnOdRSRGdgInW8/udR1E5Gh83+8F\nzHadI44sdufjPL5tM5vbus4ikkLbgObWs5rCRzKe7/v9gfdd55ATt49983/BL4pd5xA5jHVAmzgs\n9hKnDsYfuA6QLvnkdx3FqJaXcdk0g9EqfiLpUY8Yfc5I6MVqPphMYjD1RjKSAgp2us4ikkIN0QJo\nEh43uA4gNZPoYoxFM5GEUkvgu65DpEMsCoy+7xvgGtc50slg8nvRa8j93L+oCU1Wu84jEhMaJi0Z\nz/f9fGJ2TMw0OeS0vZ3bVxhM5O9kS6ypaCMZz/f9POB7rnNIzV3GZTmuM4gcQSxu7seiwAgMJKga\nx04BBd3HMOaki7hoOhCP8fAi7pxufNPVdQiRo7gYqO86RNwVUdT7eq5/x3UOkRT6rvFNY9chRI7i\nEoKOWwm5QgpP70pXzcUomeoi45vWrkOkWlwKjNe6DuCSwRSewRmD7+O++Q1puNZ1HpGIu9p1AJGj\nuN51AAmcyqlDhjBERUaJqlzgOtchRI5CnbYRcimXZrvOIHIYWcSgWzryBUbf97OA4a5zZILa1O5x\nB3c0OI/zprvOIhJh+ryRjOX7fgNiMgdMWJzN2f3a0W6B6xwiKaLijWQs3/cboWNipCS6GDUXo2Sq\nyDeiRL7ACAwBmrkOkSkMps5ZnDX4Hu6ZXY96G1znEYmg04xvOrkOIXIYVwJ5rkPIVwwmbwQjmten\n/nrXWURSoI+mDpEMdi1Bp61EyKVcqrkYJVP1jfow6TgUGCPfhnoiiijqfRd31dLQLJGUiPzdKQkt\nHRMzUBZZTcYwZncuuftcZxFJAXUxSqbSlCERVEjh6d3oNtt1DpHDiPRot0gXGH3fzwGucp0jUxlM\nvXM456xxjJtVRNFm13lEIiTSBw4Jp8RQsHNc55Dq5ZHXqYSS+WhBNome77sOIPJNvu+fDJzhOoek\nxiVcotEakqkifZ0Y6QIjcC6g1euOoh71+t3N3bkDGPCe6ywiEVFsfNPBdQiRb7gS0LChDNaQhgOG\nM1zzJEvUtDK+KXYdQuQbLnEdQFKnkMLu3eimuRglE/U3vmnpOkSqRL3AqKFgx8hgGlzIhQN/xI9m\n1qb2Ftd5RCIg0nenJJQ0dD8EutFtcB/6zHSdQyTJtJCGZJrLXAeQ1LqESzS/pmQiQ4RH2Ua2wOj7\nfh5whescYdOQhv3v5V50cSNSYyrmSMbwfb8OcLbrHHJ0BmMu5uLuJ3PyMtdZRJJIBUbJGL7v1wK+\n4zqHpFaii1FzMUomimwjSmQLjMCFQH3XIcLIYBpfwiX9b+f2dwsp3OE6j0hI9TS+OdV1CJGEc9BK\nmaFhMLVv4ZbaGlEgETLA+KaB6xAiCecDBa5DSOpdyqWai1Ey0UDjm+auQ6RClAuManuvoSY0OfM+\n7jtQTPGHrrOIhFRk705J6JzvOoAcn2yyW45l7Ppssg+6ziKSBNkEN/9FMoGuE2OigILup3Gauhgl\n02QBQ12HSIUoFxiHuA4QBVlknXQFV/Qdxah38snf5TqPSMhoSJhkigtcB5DjV0jh6bdy6weuc4gk\niY6J4pzv+wa42HUOSZ9LuCTfdQaRapzrOkAqRLLA6Pt+c0AruCZRc5qfdT/379E8FiLH5QzjG53U\niFO+77cGOrnOISemBS0GXcRF01znEEmCocY3kbz2kFDpB5zkOoSkTwEFp3Wnu65hJdNEch7YqB7k\n1b2YAllktbiaq3vfyq3T88jb4zqPSAgUEJzIirik4dEh149+Z+kGn0RAE6Cv6xASexqqH0MXc7Fu\n+EumaW5808V1iGRTgVGOWytaDb6f+7d3pOM811lEQmCw6wASexoeHXIGkz2c4e2a0GS16ywiNaRh\n0uKazstiKNHF+JHrHCLfELlh0iowygnJJrvVdVxXfCM3Tsshp9R1HpEMps8jccb3/SwiePISRwZT\nfxSjKvPJ3+k6i0gNXOQ6gMSX7/s5QH/XOcSNi7m40HUGkW+I3DDpyBUYfd9vAkSu1TQTGYxpS9sh\n4xm/8VROXeg6j0iGGmB8k+M6hMRWL6CR6xCSHDnknHo7ty83mErXWUROUE/jG13kiyu9gdquQ4gb\nBRR0UxejZJizozY3caReTILa3tMsh5y213N9txGMmJpN9gHXeUQyTB2CIo+IC5p/MWLqUrfPCEbM\ncJ1D5ATloGOiuDPIdQBx6xIu0Q0OySQNgJ6uQyRTFAuMGo7ogMFkdaDD2eMZv7Y1rZe4ziOSYfS5\nJK6c5TqAJF972g8ZxKB3XOcQOUFnuA4gsaUCY8zlk9/tdE5XF6NkkkhNZaQCoyRVLrntb+bmDtdw\nzdQssspc5xHJEOqsFld6uw4gqfEdvtO3LW0/dp1D5ARoDrwYMsa0McZ8/wT33VPTv+/7vkE33QTN\nxSgZJ1I3PiJVYPR9vyHQ3XWOuDOYnK50PXs84z89mZM/cZ1HJAOcFbX5NSTz+b5/MnCS6xySGgaT\nfz3XN61HvQ2us4gcJ3UwxlMboNoCozFpmau6G9AwDX9HMpy6GCXDRGrakKhd8A4CjOsQEsgjr9Nt\n3NbmCq6YZjAVrvOIOFQfON11CIkddS9GXBZZTccwZkcOOaWus4gch9bGN81ch5Bjk+g8XGKMecoY\ns8gY85YxptAY084YM9kYM9sYM8MY0zmx/XPGmOFV9j/UffhzYJAxZp4xZpwx5iZjzCRjzBTgbWNM\nHWPM28aYOcaYhcaYy5P8UtS9KF+6mItruc4gktDC+CYyDQFRKzBqGGKGMZi8YoqHjGf80pM46VPX\neUQcilT7u4SCCowxkE9+l9GMnus6h8hxUhdjuHQAfmOt7QbsAK4CJgJ3WGt7A/cCjx3lOcYDM6y1\nPay1ExKP9QKGW2uHAPuBK6y1vYBzgAeNMclsHOmTxOeSkMsnv2sxxR+6ziGSEJmFXqJWYOzrOoBU\nL5/8bqMZ3eJSLp1mMJWu84g4oOkbJN1UYIyJxjQeeCVXTnWdQ+Q4qMAYLqustfMS388mGO48EPiz\nMWYe8CTQ/ASe9x/W2m2J7w3wM2PMAuCfQLKn+eiRxOeSCPgu363tOoNIQmSGSUetwNjZdQA5PIMp\n6E3vIT/mxx83pvEa13lE0qyr6wASOyowxkh3ug/pTe8PXOcQOUYqMIbLgSrfVxDMZbgj0Y146KtL\n4vflJK4xjTFZQN4Rnndvle9HAE2A3tbaHsAmoCAZ4X3fzwFOS8ZzSXSoi1EyiDoYM43v+w0IDkqS\n4QopPP12bm8ylKHTAes6j0iadDn6JiLJ4ft+C0BznMWIwZhLuKRbC1osd51F5Bj01eJnobYLWGWM\nuRrABIoTv1vNVze4LgNyE9/vBoqO8Jz1gM3W2jJjzDnAKUnM2xnIT+LzSUSoi1EyhDoYM1An1wHk\n2BlMrf70H3wf981rQIN1rvOIpEFD45umrkNIbGiuqRgymDq3cmthLWptdZ1F5CiKCIbZSniNAG41\nxswHFgGHFmV5ChiSeHwAX3UpLgAqjDHzjTHjqnm+F4A+xpiFwA3A0iRm1UJ7Uq188rv2oIe6GMW1\ntsY39VyHSIYc1wGSSMOjQ6g2tXv+iB/tnsGMGVOYokUwJOq6AJtdh5BY0PDomMomu+VYxs77Jb+s\nW0ll7tH3EHGmA6AFADOctXY1VYYXW2t/WeXXQ6vZfhPQv8pD9yceLwO+843Nn6uy3xaCgmR1Geoc\nZ+xv0vBoOayLuKj2POYdfUOR1DEEw6SnOs5RY+pgFOcMpmgwgwfdwz0f1aXuBtd5RFJI8zBKunRz\nHUDcqUWtHrdwy/uuc4gcRXvXASQ2dEyUw1IXo2SISCxEpQKjZIwiivqMY1ytQQx6x3UWkRTRPIyS\nLm1dBxC3WtJy8AVcMN11DpEj6OA6gMSGCoxyRBdxUU27ZEVqKhI33aJUYNQQ6QgwmHrncu5Zd3HX\nrDrU+cJ1HpEkU4FR0kUFRmEAAwZ2pvNc1zlEDiMSF1OS2Xzfz0PHRDmKfPK7qItRHIvE51QkCoy+\n72cD7VznkOSpT/1+93BPdn/6v+c6i0gSaYi0pJzv+/WBBq5ziHsGk/M9vtemEY3WuM4iUg0VGCUd\nWhGRa15JLXUximNtXAdIhqgs8tIWyHMdQpLLYBoOZejAvvR9/7f8tuM+9jVynUmkhloY39S1nt3l\nOohEWiTugEpyGEyD0Yze9gAP7D7IwSLXeUSqaGt8k209W+E6iETaKa4DnIidO3fy6quvsmfPHowx\n9O7dm/79+7Nx40beeOMNDh48SP369bnyyispKCj41v6lpaVMmjSJzZs3Y4zh8ssvp1WrVvzjH/9g\n+fLlNGvWjCuvvBKA+fPns2/fPgYMqHaNndjIJ79LT3rOmsvcfq6zSCy1cR0gGaJyN0fDoyOsEY0G\n3Mu9lb3p/YHrLCJJoGHSkmoqMMrX5JLb7nZuX2Iwla6ziFSRB7R2HUIiL5QFxqysLC644ALGjh3L\nbbfdxqxZs9i8eTOTJk3ivPPOY8yYMXTu3Jn33qt+sNfkyZNp3749d9xxB6NHj6Zx48bs37+fDRs2\nMGbMGLKzs9m0aRNlZWXMmzePfv1UUwO4iIt0I05cqWV8c5LrEDUVlQKjFniJuCyymlzKpWeMYcy7\nBRTsdJ1HpAb0eSWppgKjfEs96vW7juu06ItkGi30IqkWyiJ2UVERLVq0ACA/P58mTZqwe/dutm7d\nyimnBDXTdu3asXjx4m/tu3//ftasWUOvXr0AyMnJobCwEGMMFRUVWGspKysjKyuL9957j379+pGd\nnZ2+F5fB8sjr0pOes1znkNgK/Tm8CowSKk1peuaP+XHp6Zz+kessIico9HemJOOF/uREUqMjHc8+\nkzPfdZ1DpArNwyipFsoOxqq2b9/Ohg0bOPnkk2nSpAlLly4FYNGiReza9e1Zd7Zv306tWrV47bXX\neOKJJ3j99dc5ePAg+fn5dOjQgSeeeIKioiIKCgpYt24dXbpocE1VF3FRXdcZJLZCfw4flQJjG9cB\nJH2yyGp2JVf2GcnIGfnkay47CZvGrgNI5IX+5ERS5zzO692GNotc5xBJ0CKNkmqh7GA85MCBA7z8\n8ssMHTqUgoICLr/8cj788EOefPJJDh48WG3nYWVlJRs2bKBv376MHj2avLw83nnnHQDOOussSkpK\nuPDCC5kyZQrnnHMOs2fP5uWXX2batGnpfnkZKY+8zr3opS5GcaGN6wA1FZUCY0PXAST9WtBi0I/5\n8e4udJnjOovIcWjiOoBEngqMclgGU3ADNzSqS92NrrOIoGOipF5oOxgrKip4+eWX6d69O127dgWg\nSZMm3HDDDYwaNYrTTjuNBg0afGu/unXrUrduXVq2bAlA165d2bBhw9e2OfRz48aNWbx4Mddccw3b\nt29n69atKX5V4TCUoepiFBdCfw4flQJjfdcBxI1ssk++hmt63sIt03PJ3es6j8gx0MWUpFor1wEk\ns2WR1ex2bt+WQ85+11kk9tQkICnj+74hpMdEay2vv/46jRs3ZuDAgV8+vmfPHiDoUpw+fTp9+vT5\n1r5FRUXUq1ePLVu2APDpp5/SpMnXTz8PdS9WVFRQWRms/2WMoaysLFUvKVTUxSiOnOw6QE3luA6Q\nJN++dSOxYTCmNa0Hj2f8mj/xpxXLWV7sOpPIEWiItKSM7/s5QB3XOSTz5ZPfdRSj3v0NvznTdRaJ\nNRUYJZXqA/muQ5yIzz77jAULFtC0aVMef/xxAM4991y2bdvGrFlB3atLly707NkTgF27djFp0iR+\n8IMfAHDRRRfxyiuvUFFRQYMGDRg2bNiXz71kyRJatGhB3bpBk16zZs147LHHOOmkk2jWrFk6X2ZG\nG8rQenPQQDlJq9DXtYy11nWGGkncmSoDtPSVYLGVq1g144/88Yxyygtc5xGpxkrrWU1qLynh+34j\nYIvrHBIec5k77XVeH+I6h8TWMuvZzq5DSDT5vt8W+NR1DgmvSUyaNYc5/VznkNhYYj3b1XWImojC\nEOkiVFyUBIPJOpVTh4xn/Ia2tNUk9pKJ1MEoqaQpQ+S49KDHoB700DAwcaWR6wASaTomSo0MZWg9\nINwdWRImof/MikKBMfT/EST5cshpewM3dL6O66Zmk33QdR6RKuoZ3+S6DiGRVc91AAkXg8m6nMu7\nNKf5CtdZJJYaGN8Y1yEksnRMlBrJI69Tb3rrJpykS+iHSEehwBj6/wiSGgaT3YlOZ49n/JpWtFri\nOo9IFepilFTRTTc5bgZTdBu35RVSuN11FomdbECrtUqq6JgoNXYhF9ZHXYySHgXGN6GcN/aQKBQY\ndeCQI8olt8Mt3NLhaq6emkWWlkaTTKCVpCVVdEyUE5JNduuxjF2dRVa56ywSO1roRVJFHYxSY4ku\nxg9c55DYCPW5fBQKjOpglKMymJxudDt7PONXNKf5ctd5JPbUwSipoospOWG1qd3zJm56z3UOiR0V\nGCVVQn2hLpnjYAW46gAAIABJREFUQi5sgLoYJT1CXd+KQoFRBw45ZnnkdRnJyFOGMWyqwVS4ziOx\npUntJVV0TJQaaU3rwedx3nTXOSRWdEyUVNFNN0mKPPI69aGP5mKUdAj1uXwUCoyhrvBK+hlMXg96\nnH0/9y9tStNVrvNILGmRF0kVXUxJjZ3JmQM60nGe6xwSG0WuA0hk6ZgoSaMuRkkTFRgdC/V/AHGn\ngIJuJZQ0u5iLpxlMpes8Eis5rgNIZGmxBKkxg8m9jutaNaThWtdZJBayXQeQyMpzHUCiI5fcjn3p\nq7kYJdVC/bkVhQKjLqbkhBlMYV/6DrmP+xY2otFnrvNIbOhiSlIlCsd1yQAG06iEkv155O1xnUUi\nT8dEEQmFC7igIepilNQK9TExChcimkdPaqwWtYrHMrbRBVwwHR00JPVCfeCQjKZubEmaXHI7jGHM\nInRclNTSMVFSxbgOINGiLkZJg1AfE6NQYDzoOoBEg8HUHsjAwfdy79z61P/cdR6JNA2RllRRgVGS\nqj71z7iWa7Xoi6RSqC+mJKOpwChJdwEXNEI33iR1Ql2ji8JFrgqMklR1qNPrTu7ctXXNa5M7Lppf\ny3UeiZ5NdSh1nUEiSwVGSbrOdB7y049av1l382d1XGeR6NlQh/2uM4iIHKtcm9P+ySmN/3bgwBYd\nEyXp1heFu76lAqNINQymbrMWl5w96vnFq/LKyrq4ziOR8yfXASSyVGCUlLj/b593zamgjescEknP\nug4gkaUORkkqU1lZMfLJJ99vtmnLd11nkcj6jesANRHq9ssEFRglJcpzcwueGD26yMJ211kkclQE\nklTRvMSSdA22bVuXU1HRxnUOiSwdEyVVVGCUpMkuLz94x6OPfths06azXGeRSAv1uXwUCowHXAeQ\n6NreqFHLV4YP/9Tq5FeSK9QHDslo+qySpCueN2+V6wwSafrcEpGMlnvw4N67JkxY2GDHjv6us0jk\nhfo6MQoFRnUwSkotOu203gu7d9cE95JMupiSVNF7S5Ku26JFUZhSRzJXqC+mJKNpIQ6pscJ9+3bc\n/eCDn9bZu7e36ywSC6E+JqrAKHIMXr3yyiE76tWb5TqHRMY+1wEkslRglOSy1jbaurWD6xgSaTom\nSqrsdR1Awq3O7t1fjHvooc0FBw50d51FYqPcdYCaUIFR5FgYY54oKelUnp292nUUiYSdrgNIZIX6\nrqdknpbr1n1ioLHrHBJpu1wHkMjSe0tOWINt29bdNWHCvtzy8o6us0ishHoKQBUYRY7RgYKCes/c\nemuZ1d1QqTmd8Eqq6PNJkqrn3LkbXWeQyNNNN0kVvbfkhDTduPHTsb/6VXZ2ZeUprrNI7Gx1HaAm\nolBgDHWFV8JlQ4sWHd664IL5rnNI6OmEV1Jli+sAEi0dPvmkjusMEnk6Jkqq6IauHLfWa9YsGf3E\nE/WyrG3uOovEkgqMjqmDUdJq5sCBA1e1aTPNdQ4JNZ3wSqqE+qREMkt2efmBOnv2dHWdQyJPx0RJ\nFb235Lh0WLZs/k3PPnuygUaus0hshbpZIAoFRnUwSto9f/31Z5YWFqqTUU6UujUkVUJ9UiKZpcPy\n5YsNFLrOIZGnY6KkigqMcsxOnz//w+tefLGjgbqus0hs7cPa/a5D1EQUCozbXQeQ+KnMzs55vKSk\neaUxm1xnkdCxwG7XISSyVGCUpOk5d64uziXV9mOtRiNJqugzTI5J//fee2/Yq6/20E01cSz0I5Gi\nUGDc4DqAxNPuunWb/nHEiM0WylxnkVDZg7WVrkNIZIX+xEQyxymrVzdxnUEiTwUgSSV1x8pRnfuP\nf0y/4K23+hvIdZ1FYi/05/FRKDB+AVS4DiHxtLJ9++4f9O//nuscEirqupaU8TxvJ7rpIUlQuG/f\n9ryDBzu7ziGRt8N1AIm00F+sS2pd9tprU896993BJhp1EQm/0I9ECv0/JM/zKgENUxVn/j506JDN\nTZq86zqHhMY61wEk8ra5DiDhd9rChUt1wSVpoGOipIzneVuAUM9nJqlz3QsvTOs5b97ZrnOIVBH6\nmyI5rgMkyQaghesQEl9PjRzZ675f/GJZXllZJ9dZJOOtdR1AIm8LcJLrECeirKyMZ599loqKCior\nK+natSvnnHPOl79/8803mTt3Lv/+7//+rX3Ly8t54403WL9+PcYYhg4dStu2bSkvL+fFF19k165d\n9O3bl379+gEwadIk+vTpQ4sWOn2ozukLFpS7ziCx8JnrABJ564D2rkNI5jCVlZU3P/PMu63WrRvi\nOovIN4S+wBiVO9PrXQeQeCvPzS2cOGpUgdVcL3J0upiSVAvt8IqcnBxuvPFGSkpKGD16NCtWrGDt\n2qAm//nnn7N//+EbUebMmQPAmDFjuP7663nrrbeorKxkxYoVtG7dmpKSEhYsWADAxo0bsdaquHgE\nzTdsOMV1BokF3XSTVNN7TL6UVVFRXvLYY++3WrdukOssItVQgTFDqMAozm1t3PiU14cN+8QGqwSL\nHI5OdCXVQrv4mTGG/Px8ACoqKqioqMAYQ2VlJf/4xz84//zzD7vvF198Qdu2bQGoU6cOBQUFrF+/\nnuzsbMrKyqisrMTa4ON5ypQpX+uMlK9rtGXL2uzKytauc0gs6KabpJrOuwSAnLKy/T965JE5TbZs\nOdN1FpHD2Ow6QE1FpcC4xnUAEYD5PXr0Xdy16zTXOSSj6URXUm256wA1UVlZyeOPP84DDzxAu3bt\naNmyJbNmzaJTp04UFRUddr+TTjqJZcuWUVFRwfbt21m/fj27du3i1FNPZceOHTz99NOcccYZLF26\nlObNm1O3bt00vqpwKZ43b7XrDBIbOiZKquk9JuQdOLB73EMPLa23a1c/11lEjmCF6wA1FZU5GFe5\nDiByyF+uvnpIq4ce+rDu7t19XWeRjKQTXUm1UBcYs7KyKCkpobS0lJdeeonVq1ezaNEibrrppiPu\n17NnT7Zs2cLEiROpX78+rVq1whhDdnY2w4cPB4KuyD/84Q9cd911TJ48mZ07d1JcXEznzlosuapu\nixfnus4gsaEORkk1nXfFXK29e7f96JFHNuYfPNjDdRaRo1jmOkBNRaWDUQVGyRzGmCdKSjpUZGXp\npFmqoxNdSbVQFxgPKSwspE2bNqxevZpt27bx6KOPMmHCBMrKynjkkUe+tX12djZDhw6lpKSE6667\njv3799OoUaOvbfPhhx9SXFzMunXrKCgo4Oqrr+b9999P10sKBVNZWdlg27aOrnNIbOiYKKmm91iM\n1d25c8NdEyZszz94sKvrLCJHcYAIjMxVgVEkBUpr1ar/3M03l1oodZ1FMsp+rA393BqS8UJbYNy7\ndy+lpcHHZllZGZ9++inNmzfnvvvuY9y4cYwbN47c3FzuvPPOb+178OBBDh48CMDKlSvJysqiadOm\nX/6+tLSUTz75hOLiYsrKyjDGfPl35Cut1q5dZqCh6xwSC9uxdo/rEBJ5KjDGVKMtW9b86JFHKnLL\ny9u5ziJyDFZgbaXrEDUViSHSnudt9n1/L1DbdRaRQ9a1atVpyrnnvnvu229rImE5JPTzakjm8zxv\nq+/72whhkWj37t289tprXy7I0q1bNzp16nTY7ZcuXcr69ev5zne+w969e3n++ecxxlBUVMSVV175\ntW2nTZvGoEGDyMrKol27dsyaNYuPP/6YPn36pPplhUrPuXM3A11c55BYWOk6gMTCSqCS6DTWyDFo\n8fnny299+ul6WdY2PfrWIhkh9MOjAcyhFRXDzvf9RYBanyXj3Pzb305vvXbtYNc5JCP8CWuvcx1C\nos/3/ZnAGa5zSPjc+8ADc2rv3dvLdQ6Jheew9mbXIST6fN9fAaiLLSbafvrpx9f//vctDdR3nUXk\nOPwca//NdYiaitKdnE9cBxCpzu9uumlAaUHBQtc5JCN87DqAxEZoh0mLOzllZaW19u7t5jqHxIbO\njSRd9F6LiS6LF8+5/ve/b6viooRQJDoYo1RgnO06gEh1KrOzc58oKWlSacwXrrOIcyowSrqowCjH\nreOyZYsN5LvOIbGhY6Kki95rMdDro48+uPrll7sZTZsm4RSJhrkoFRg/dB1A5HB21avX7E/XXrve\nQrnrLOKUTnAlXSJxkiLp1XPePC24IemkY6Kki95rEXfW9OnvXPLGG310k0xCTB2MGeYj1wFEjmR5\np07FH/Xp867rHOLMPuBT1yEkNiJxkiLp1XrNmpNcZ5DY2Ia1612HkNjQEOkIu/Bvf5v2nSlTzjSQ\n7TqLyAnairVbXYdIhsgUGD3P2wqscp1D5EjevOSSIVsaNXrPdQ5xYjFRWVVLwuBj4IDrEBIetfbu\n3ZpbVnb4JbtFkksdZZJOnwAHXYeQ5LvqL3+Z1v+DD4YYMK6ziNRAZBoDIlNgTFAXo2S8iaNGFZfl\n5Gh+tPjRxZSkjed5ZcAC1zkkPLovWLBMF2iSRjomStp4nldOhC7gBbDWXv+730077eOPh7iOIpIE\nH7gOkCxRKzBqHkbJeGV5ebWfGjky18Iu11kkrea7DiCxo2OiHLPTFyyodJ1BYkU3QCTd9J6LCFNZ\nWTHyySffPXXVKhUXJSoiM8IxagVGdTBKKHzRtGmbv1566RILGjIbH5p/M4aMMW2MMd8/wX1ruuCG\nCoxyzE7atKmN6wwSKzNdB5DYed91AKm57PLyg2N/9asPm2/ceJbrLCJJFJnrxKgVGGejgo2ExNze\nvc/4pFOnaa5zSFrsA+a6DiFOtAGqLTAaY3JS/LdVYJRj0njz5jXZlZUtXeeQ2NiJFt2Q9HvHdQCp\nmdyDB/fd+fDDCxtu397fdRaRJFqNtRtch0iWSBUYPc/bRTCJr0govPS97w3eXafObNc5JOVmYW25\n6xBy7BKdh0uMMU8ZYxYZY94yxhQaY9oZYyYbY2YbY2YYYzontn/OGDO8yv6Hug9/Dgwyxswzxowz\nxtxkjJlkjJkCvG2MqWOMedsYM8cYs9AYc3kSX8ZiYEcSn08iqse8eZ+5ziCx8j7Waki+pNtCND1R\naBWUlu4c99BDK4v27OntOotIkkVmeDRErMCYoI4NCQ2blZX1RElJ24qsrHWus0hKRabtPWY6AL+x\n1nYjKNRdBUwE7rDW9gbuBR47ynOMB2ZYa3tYayckHusFDLfWDgH2A1dYa3sB5wAPGmOSstCG53kW\nDUOUY9B1yZJc1xkkVtRJJmnneV4lGiYdSrX37Pli3EMPbSzcv7+76ywiKRCp68QoFhg1D6OEyr7a\ntRv+/sYbd9ug0CDRFKkDR4ysstbOS3w/m2C480Dgz8aYecCTQPMTeN5/WGu3Jb43wM+MMQuAfwIn\nAyfVKPXX6b0nR2QqKyvrb9/exXUOiRUVGMUVvfdCpv727Z/fNWHC3ryysk6us4ikiDoYM5w6GCV0\nPjvllC7ThgzJ2PfuLUBT4LRqfvcgQYVkyxH23wW0BMYmfj4ADE08X9X2r5HAnJqGzTwW3TEPqwNV\nvq8AGgI7Et2Ih74OFWbKSRxTjTFZQN4Rnndvle9HAE2A3tbaHsAmoCBZL4CInbRI8p2yZs0SA/Vc\n55DYKANmuQ4hsaUCY4g03bRp1R2PPmpyKirauM4ikiK7idicxFEsMM7l6xeFIqEw7ZxzBn1+8skz\nXOeozk3A5GoeXwu8BbQ+yv7/AQyu8vPfgbOABcAfEo/NJ6jg9KpJ0My0CGs1D1407AJWGWOuBjCB\n4sTvVgOH5gW6DDg05HQ3UHSE56wHbLbWlhljzgFOSXLmDwgu6EWq1XPu3CPdHxJJtjlYW+o6hMTW\nLHRMDIWWn322dPTjjxdlWdvCdRaRFPoAaytch0imyBUYPc8rBf7lOofIiXj25pvPOJCXt8h1jm8a\nTNC69U3jgF8QdDAezmyClqwLqjyWS7CschlfLfv+H8B/1ThpRtLd8mgZAdxqjJkPLAIOLcryFDAk\n8fgAvupSXABUGGPmG2PGVfN8LwB9jDELgRuApckM63neXjRMWo6g/YoV6l6UdNIxUZzxPG8fQTOK\nZLD2y5cvuOWZZ5obaOw6i0iKRW6kUY7rACnyBsEITJFQqcjJyXuipKTBjx55ZEumH1RfJ5gsrvgI\n21QC9wDPE0wud8j5BJ2L/YH7gEkEnYsRvUX5d9cB5PhZa1dTZVYAa+0vq/z6W8cXa+0mgrf0Ifcn\nHi8DvvONzZ+rst8WgoJkdRnqHGfsw3kTODtJzyURknvw4L7Cffu6us4hsfK26wASe/8C+rkOIdXr\nPn/+R1e8+mo3A4Wus4ikQeRuukWugzHhr64DiJyoHQ0atHj5e99ba4MRwxlpH/Az4D+Pst1jwHcJ\n5l+sKgf4I8Et5KuBhwkKkXcDwwkKjhFxkK/XVkVc+B/XASQzdVq6dLE58nyhIsm0D40yEvf+5jqA\nVO+MmTPfv+LVV4tVXJSY2A1Mcx0i2SJZYPQ87zOCYWkiobS0S5ee83r2zMj5GAFWAqsIuhfbAOsI\nOhA3fmO794FfJ7a5F/g9MP4b2zxGMC50JsFkdC8RLBwTEdOxdo/rEBJvnuctJpgjUuRresybt/fo\nW4kkzdtYu991CIm9dwnmVJYMcs6UKTMunDz5DPPVHNYiUfd3rD3oOkSyRbLAmPCG6wAiNTHp8svP\n3tagwUzXOarTHdhMULFYTdChOAdo9o3tXgA+S2zzS4JC4s+r/H47wT/UGwjaGrII5nOM0Ozv6hyT\nTKGODfmW1p991tx1BokVHRPFOc/zytHokoxy6euvTx08ffogE+3ahMg3ve46QCpE+R+xhklL6D05\nenS38uzsT13nuI5gkrhlBMXE3x5h24+A247xef8T+HeCD6ILgRkExcvrTzhpxnnTdQCRBL0X5Wvq\n7N79RU55eQfXOSRW9DkkmULvxQxx7R//OK3X3Llnu84hkmblRPSmW1QXeQGYRdBk1dR1EJETdTA/\nv+ipH/7wi9FPPLHbQJGrHC8e5ferq3zfB3i6mm1uSnxVNaHK9wXAW8eZK8OtwNpPXIcQSZgC7Cf4\npybC6QsWfGKgiescEhsLsHat6xAiCW8AlmDgjLhgrb35mWdmtF67dojrKCIOzMDa7a5DpEJkOxg9\nz6tEd6ckAjY3a3bq3y666GPXOeS46fNHMobnefuAqa5zSObovnChdZ1BYkVTF0nG8DxvE/CB6xxx\nlVVRUV7y2GPvtV67drDrLCKORGhN06+LbIExQcOkJRI+POOMASvatZvqOoccl0i2vUuoqegtX2q6\nadOprjNIrOiYKJkmshf4mSynrGz/HY8+OqfpF1+c6TqLiEORnH8Rol9gfAuI3Mo8Ek9/HDFi0N5a\ntea6ziHHZAvBkFSRTKIOIgGg6aZNq7KsbeE6h8TGJiAjF62TWHvNdYC4yTtwYPe4CROW1t+5s5/r\nLCIOLcTaVa5DpEqkC4ye5+1BQ8IkImxWVvbjY8a0qjRmvessclR/wdpy1yFEqvI8bxW6yBegx9y5\nmgtP0uklrK10HUKkKs/zlgDzXeeIi1p79267+8EH19bat6+H6ywijkW2exEiXmBM0DBpiYy9deo0\n/sMNN2y36szNdC+4DiByGL93HUDc67JkSb7rDBIrz7sOIHIYf3AdIA7q7ty58a4JE7blHzzY1XUW\nkQygAmPIvQpUuA4hkiyr27bt9u6ZZ6oLKXOtAd51HULkMF5CNyhizVRWVtTbubOL6xwSG59g7Yeu\nQ4gcxh8BddemUKMtWz770SOPlOWWl7d3nUUkA6wBZrsOkUqRLzB6nvc58DfXOUSS6e3zzx+8oVmz\nd1znkGr9CWu1OqtkJM/ztqHFFmKt7apViw3UdZ1DYkMd/ZKxPM/bgObMTplm69evGPOb3+RnV1a2\ncp1FJEM8E/XrxMgXGBMmug4gkmzP3HprnwN5eUtc55Bv0cWUZDoNk46xnnPnbnOdQWJFx0TJdBrC\nnwJtVq1aNHLixEZZ1p7kOotIhqgEnnEdItXiUmB8E1jnOoRIMpXn5hY8OXp0kQVdLGaOj7F2oesQ\nIkfxP8BW1yHEjXYrV9Z3nUFiYybWrnQdQuQoXgH2uQ4RJZ2XLJl7w+9+d4qBBq6ziGSQv2Nt5GtS\nsSgwep5XQQyqxRI/2xs2bPnK8OGrrOaPyRTq1JCM53leGcFcjBIzeQcO7CkoLdUk+5Iu6gyTjOd5\n3h4ivuhCOvWcPfuDa156qauBOq6ziGSYp10HSIdYFBgTnkZFGImgRaed1nth9+7TXecQyoDnXIcQ\nOUYaJh1DXZYsWWIg13UOiYUD6EaGhIdWk06CM995591L//rX3gbyXWcRyTCbgL+6DpEOsSkwep63\nFpjsOodIKrx65ZVDdtSrN8t1jpj7f1i70XUIkWPhed4HwCeuc0h6Fc+bV+o6g8TGy1i7xXUIkWP0\nFrDadYgwu3Dy5Gnn/vOfAw3kuM4ikoF+h7VlrkOkQ2wKjAla7EWiyRjzRElJp/Ls7NWuo8TYb1wH\nEDlOv3UdQNKr5bp1LVxnkNj4tesAIscqMZ3WY65zhNUVr7wyrf/MmUMMGNdZRDJULIZHQ/wKjG8A\n612HEEmFAwUF9Z659dYyC3tdZ4mhhVg7w3UIkeM0EX1exEbRrl2bcsvL27vOIbHwIdZqVIWEzdNo\nsZfj9oPf/37a6QsXDnGdQySDTcfa5a5DpEusCoxa7EWibkOLFh3euuCC+a5zxJC6FyV0PM/bATzr\nOoekR/H8+StcZ5DY+JXrACLHy/O87WixvmNmKisrf/jkkzPaffqpiosiRxab7kWIWYExQYu9SKTN\nHDhw4Ko2baa5zhEjO9FKmRJeD6NjYiyctnCh6wgSD5uBl12HEDlBKo4fg+zy8oO3//rXH7TYsGGQ\n6ywiGW4H8BfXIdIpdgVGz/PWEEzkKxJZz19//Zn7CgvVyZgev8NaDTOVUPI8byUwyXUOSb0mX3zR\nznUGiYWnsPaA6xAiJ8LzvIXAVNc5MlnuwYP77nz44QWNtm0b4DqLSAj8BmtjtcBeXFd5ehwY6jqE\nC5WVlUycOJGioiJGjBjBq6++ypo1a8jPzwdg2LBhNG/e/Fv7zZs3j+nTpwMwePBgevToQXl5OS++\n+CK7du2ib9++9OvXD4BJkybRp08fWrTQXPKuVGZn5zw+ZkyLcQ89tDHL2mau80SYRZOCS/g9BAxz\nHUJSp9mGDSuzrFWBUVKtHHjCdQiRGvoVcLbrEJkof//+nXc+/PCawv37+7jOIhICewlGCsVK7DoY\nE/4KxLK7a+bMmTRu3Phrj51//vmUlJRQUlJSbXFx3759TJ06ldtuu40f/vCHTJ06ldLSUlasWEHr\n1q0pKSlhwYIFAGzcuBFrrYqLGWBPUVGTP44Y8YWFMtdZImwS1i5zHUKkJjzPmwF86DqHpE6PuXM/\nd51BYuH/Ye061yFEauh1YI3rEJmm9p49X9z94IMbCvfvP911FpGQmIi1W1yHSLdYFhg9z7PAT1zn\nSLedO3eyfPlyevXqdVz7rVy5knbt2lGrVi0KCwtp164dK1asIDs7m7KyMiorK7HWAjBlyhTOOeec\nVMSXE7CyffvuH/Tv/57rHBH2M9cBRJJkgusAkjqdly4tcJ1BYkHHRAm9xKKgD7rOkUnqb9++/q4J\nE/bklZV1dp1FJCQOAL90HcKFWBYYATzPew2Y6zpHOk2ePJnzzz8fY8zXHp8yZQqPPfYYkydPpry8\n/Fv77dq1i7p16375c926ddm1axennnoqO3bs4Omnn+aMM85g6dKlNG/e/Gvbint/Hzp0yOYmTd51\nnSOC3sbaWa5DiCTJn4G1rkNI8mVVVJTX3bWrq+scEnlvYG0sRwdJJE0E1rsOkQmabN686o5HHyWn\noqKt6ywiIfIc1sbyMyS2BcaEn7gOkC7Lli2jdu3a3xq6fN555zF27FhGjhxJaWkp77zzzjE/Z3Z2\nNsOHD2f06NF069aNmTNnMnDgQCZPnsxLL73E0qVLk/0y5AQ9NXJkr4O5uRrKm1zq1JDI8DyvHHjE\ndQ5JvlM//XSxgTquc0jk/R/XAUSSxfO8A8D/dZ3DtZZr1y4b/fjjdbKs1dxXIseunBh/fsS6wOh5\n3iRgtusc6bB27VqWLVvGhAkT+Mtf/sKqVat45ZVXKCoqwhhDTk4OPXr04PPPvz1N06GOxUO+2dEI\n8OGHH1JcXMy6desoKCjg6quv5v3330/565JjU56bWzhx1KhCCztcZ4mId7B2iusQIkn2OLDRdQhJ\nrp5z5253nUEi722snek6hEiSTQQ2uA7hSrsVKxbc8tvfNsuytonrLCIh8yLWrnIdwpVYFxgTfuI6\nQDqcd9553HPPPYwbN47hw4fTtm1brrrqKnbv3g2AtZalS5fStGnTb+3brl07Vq5cSWlpKaWlpV/O\nyXhIaWkpn3zyCcXFxZSVlX05BLusTGuLZJKtjRu3fu2KK5bbYOVjqRnfdQCRZPM8bx/wX65zSHK1\n/fTThq4zpMNa4BygK9CNr9px5wMDgO7ApcCu49gX4H7gdOCGKo89TwyXhTwyz3UAkWTzPG8/Me1C\nOm3hwo9GPP98ewP1XGcRCRkL/LfrEC7FvsDoed4bQGznUXvllVd47LHHeOyxx9i3bx+DBw8G4PPP\nP+f1118HoFatWgwePJiJEycyceJEhgwZQq1atb58jmnTpjFo0CCysrJo164da9as4fHHH6e4uNjJ\na5LDW1Bc3Hdx167TXOcIufew9p+uQxyJMaa+MWZMlZ9bGGP+4jKThMZTwErXISQ58vfv31Wwf38X\n1znSIYdgVYbFwEzgN4nvbwN+DiwErgAeOI59dwJzgAVAXuI5SoFngdtT91LC5h9Yq3meJaqeJGZd\njH0/+OD9K1955XQDtY6+tYh8w6tYu8R1CJfModV/48z3/YuAN13nEEkLa+24hx76qO7u3X1dRwmp\n80NQYGwDvGGtPc1xFAkh3/e/D7zgOofUXM85c2ZdNmlSP9c5XLgcGAsMJ5gbxBB0Kl5IUDw8ln37\nE3Q9/gvWDcVyAAAgAElEQVQYAfx/BKshdQeGpSR1KA3EWs2JI5Hl+/5dwATXOdLh7ClTZgyePn2g\ngWzXWURCqjfWznEdwqXYdzACeJ73N4Kb1iLRZ4x5oqSkQ0VW1meuo4TQ35JRXDTGtDHGLDHGPGWM\nWWSMecsYU2iMaWeMmWyMmW2MmWGM6ZzYvp0xZqYxZqEx5qfGmD2Jx+sYY942xsxJ/O7yxJ/4OdDO\nGDPPGPNA4u99nNhnpjGmW5UsU40xfYwxtY0xzxhjZhlj5lZ5LomfFwlGlkrIFc+bt991BhdWA3OB\nMwiGPL+eePxYlkqvum8R8F2gJ9CcYKzgB6i4WMUbKi5KDMSii/GSv/512pDp0wepuChywv4c9+Ii\nqMBY1U9cBxBJl9Jateo/e8stpTYY7SXHpgK4N4nP1wH4jbW2G0GDzVUEE4rfYa3tnfhbjyW2fQR4\nxFrbHVhX5Tn2A1dYa3sRTCH2oAkmQR0PrLTW9rDW3veNv/sScA2AMaY50Nxa+xHw78AUa22/xHM9\nYIypncTXKyHheZ4F/pfrHFJzJ3/+eUvXGdJtD8GH6cNAXeAZgg/S3sBugqHOx7ovwI+BeQRDqP8D\n+E/gaYIP0Z8mP36YlJHcY6JIRvI8rxT4365zpNL3/vSnab1nzx7iOodIiO0HvnnNFUsqMCZ4nvd3\nQHdhJTY+b9my05Rzz439XZbj8DTWHm1k3fFYZa2dl/h+NtAGGAj82Rgzj+COefPE7wcQNN8A/LHK\ncxjgZ8aYBcA/gZOBk47yd18mGDUIwTXyobkZLwDGJ/72VKAAaH3cr0oiwfO8N4EZrnPIiau3Y8eG\nnIqKU13nSKcyggLhCODKxGOdgbcIPmSvA9pVv2u1+1Y1l2Dm9k4EH8YvE0xWujxJ2UPocaxd5jqE\nSJo8RzAla7RYa2965plpnZcuVXFRpGZ+ibVrXIfIBCowfp06NiRW3hk06MzPWrWa7jpHCOwmmH4r\nmQ5U+b4CaAjsSHQdHvo62uIMI4AmQG9rbQ9gE0Fh8LCstZ8DW40xpwPfI+hohKBYeVWVv93axnyS\nYuHfXAeQE1c8f/6nrjOkkwVuBboAd1d5fHPifysJOg5HH8e+Vf0HwRLrZQQf2BCcRO+rUerQ2gb4\nrkOIpIvneZXAXa5zJFNWRUX56Mcff++Uzz5TcVGkZtYTTE8lqMD4NZ7nTUUT20vM/O6mmwaUFhQs\ndJ0jw/0cazcffbMa2QWsMsZcDWACh5Zin0nQXANwbZV96gGbrbVlxphzgFMSj+8mmD7scF4iGPlX\nz1q7IPHY34E7EkOsMcb0rOkLknDzPO9d4K+uc8iJ6fbxx64jpNW7wB+AKUCPxNebBBOKdiToZGwB\n3JzYfj3B/IpH2veQ14A+if3rJ37fnWA8VDGx5GPtNtchRNLJ87wZfDXqI9Syy8sP3PHoo7NP2rz5\nTNdZRCJgPNbudR0iU6jA+G13A9tdhxBJl8rs7NwnSkqaVBrzhessGWot6Vs9cARwqzFmPrCIYDFT\nCO6a350YCt0e2Jl4/AWgjzFmIXADsBTAWrsVeNcY87Ex5oFq/s5fCAqVL1d57L+AXGCBMWZR4meR\ncQR1FAmZxlu2dHSdIZ3OIuhEXEAwZ+I8ggLincAnia+fE7RqQ1AsfPMo+x4yjK9P1P1LYCGxvSO9\njK/mBxaJm/sI+TEx78CBPeMeemhx/Z07z3CdRSQCPgCedx0ikxhrresMGcf3/VHAE65ziKRTh2XL\n5l/34ovdDOS4zpJhrsdapwcOY0wtoNRaa40x1wLXWWu1yrOkhe/7/xsVnEOlxeefL//hU091cJ1D\nIulSrH3DdQgRV3zf/xkhnUKkcN++7Xc+/PD6/IMHu7nOIhIBFhiAtR+4DpJJ1MFYvYlowReJmeWd\nOhV/1Lfvu65zZJhpZEaTSm9gXqKDcQxwj+M8Ei+/INEdK+HQc+7c9a4zSCT9U8VFEX4GbHAd4ngV\n7dq1adxDD21RcVEkaV5QcfHbVGCshud5FhgFlLvOIpJOb1588ZAtjRq95zpHhjgAjCID2ryttTOs\ntcXW2tOttYOttStcZ5L48DzvINWvjSEZquOyZbVdZ5DIKQVKXIcQcc3zvD0E81iHRsOtW9fe+fDD\nB3LLy9XZLpIce4H7XYfIRCowHobneQtJ37xrIhlj4qhRxWU5Octd58gAP8XaZa5DiGQCz/OmAc+5\nziFHl11efrBo9+6jrUAvcrx+gm5uiQDged7zwN9c5zgWzTZsWHn7r3+dm11Z2dp1FqnefqAfwaJh\n3QAv8fjbQC+ChcXOAg73AfzfBBO0dyJYsRHgi8Q+pxEsVHbI5QSLnEmN/TfW6v/KaqjAeGQ/Ada4\nDiGSTmV5ebWfGjky1361kEgcfQz8X9chRDLMfcBW1yHkyNqtXLnYgDoYJZlmAw+6DiGSYUYBu12H\nOJJTVq9ePPLJJxtkWdvMdRY5vHxgCjCfYJGxycBMgpbxFxKPfR/4aTX7Lgb+RLAy5GSCeZQqgBcJ\nhp7MAh5ObPtXoCfBImdSIwsJpg+SaqjAeASe5+0D7nCdQyTdvmjatM1fL7tsmQ0mr42bSuCHWFvm\nOohIJvE8bwshGxYWRz3nzo3zzSFJvnLgVqytcB1EJJN4nreWDD4mdlq6dN6Nzz3XykBD11nkyAxQ\nJ/F9WeLLJL52JR7fSfWFwdeBawmKlG0JOhlnAbnAPoL5nrIJPsgfJoPfsOFRDtys68TDU4HxKDzP\n+yvwquscIun2/7d37+F2jnf+x9+3OItS1SpapShKnQ+lVLVURzvUoBTVGkzV1E/59ddpq9N7ntbo\n+cBQlJ6GmlJjaJWqw0SJqkMTISIREpXUMWgkIpLs7++PZ2VsBtlZe691r8P7dV3rsvbaaz/PZ++r\n6b2e73Pf33vcttvuOGWTTW4onaOAs4i4pXQIqUP9BLixdAi9svWnTVuzdAb1lG8RcWfpEFKHOod6\nQ8COsvW4cbce/ItfbJpg1dJZNDSLqJdCvwHYC9gJOA/YB3gTcD7w+Zf5uZnAmwd9/abGa4dSFx/3\nAr4I/AD4GLBya+L3k28ScUfpEJ3MAuPQ/B9gTukQUrtddPDB735m9Oh++j/RGcAXSoeQOtWgTdCe\nK51F/9uK8+b9dYX58zctnUM9YwrwldIhpE7VGBOPpt4EqSPsctNNY/e9/PJtE6xYOouGbhT1UugZ\n1DMQ76beDOLKxmtHAictxfFWA34D3E7dx/HXwIHAMY3//mGkgveXiUBVOkSns8A4BDnnGcCXSueQ\n2i2WWWaZsz/1qQ0WLbPMjNJZ2iCol0Z3dD8dqbSc8yTcOa8jbT5x4qRUX6dIw7V4TPRmgvQqcs5T\ngS+XzgGw19VX/37Pa6/dJcGypbOoOasDe1DvIHQn9UxGgIOBm1/m/esCDw36ekbjtcG+CpxM3Zdx\nV+Bn1BtNaKksXhr9fOkgnc4C49CdTt07Veorz66yyho/+8Qnnonen7F0GhH+G5eGIOd8OvXNcXWQ\nrcaP94OvRsp3ifh96RBSl/ge8MeSAT586aU37PKHP7w71a371EUeB55uPJ8HXANsRt13cUrj9cWv\nvdS+1Ju8zAemAfdR70i92H3URcf3UPdkXIb6fyAdM+W2e5xCxG2lQ3QDC4xD1JgC/zHqtgZSX3lo\nvfU2u+E977m9dI4WupOXb20i6ZUdCTxSOoResPbDD69XOoN6wu3YLkQaspzzIuq2d7OX9N5WOOz8\n88dsNWHC7iXOreF7mHrW4pbADtR9Ez8EnAscAGxF3YPxW433/4oXpsxuDnwEeDvwAeBMXryM4WTg\nXxvPPwqc1TjHCa35VXrVH3nhz6glSBH9uEls86qq2hUYg0uQ1IeOPvfcG9edOXO30jlG2DxgOyIm\nlQ4idZuqqvYCrsYZE8Wt/tRTM0847bSXroySltYzwLZETC0dROo2VVUdTD2hrC3SwMDAUeedN3bd\nv/yl1z6bS51iLrC1Y+LQOYNxKeWcbwL+uXQOqYSfHHnkTvOXX35i6Rwj7CSLi1Jzcs7XAN8tnUOw\n9fjx00pnUE84zgspqTk554uoJ5613DKLFi34xzPO+KPFRamlTnJMXDoWGJvzdeoZG1JfWbTsssuf\nddxxawQ8UTrLCLmMiLNLh5C63BeBP5UO0e82v/tuP9NpuM4n4oLSIaQudwL1JsAts+yCBfNO+P73\n73zdk0/u3MrzSH3ul0T8sHSIbuOH0SbYj1H97K+rr772RQcf/FDAotJZhmkmcHTpEFK3yzk/T93a\nZ27pLH0rIl43a9bbSsdQV7sPOK50CKnb5ZznUW/6+2wrjr/Cc8/NPum7353ymmee2b4Vx5cE1DcJ\njiwdohtZYGxSzvlx6guqbi+ySEtt8mabbTNum21uKp1jGBYAhxAxq3QQqRfknKcA/6d0jn71phkz\npiRYs3QOda36JkHEnNJBpF6Qc76HFoyJq8yZ88RJ3/nOzJXmzdtqpI8t6X88BXyYCG+cN8EC4zDk\nnG/khU2cpL7y6/322/3J1772ltI5mnQ8Ed1cIJU6Ts75x8B5pXP0o23GjXM3bw3Hp4m4o3QIqZfk\nnH8EXDhSx1vt6acf/sz3vvfM8gsWbDZSx5T0vwwAhxFxf+kg3coC4/B9Dfsxqk+dc+yxmy8cNeqB\n0jmW0g+JOKd0CKlHHQf8vnSIfrPxlCmjS2dQ1zqTiLZsSiH1oWMYgR7Fr3/ssenHn376omUXLdpg\nBDJJemVfJuKq0iG6mQXGYRrUj/EvpbNI7fb8Ciuseu4xxxDwTOksQzQW+HTpEFKvyjkvAA4A3NG4\nTUYtXDh/9Jw5by+dQ13peuAzpUNIvSrn/CywL/Bws8dYd8aMyceeddYqowYG3jRyySS9jP8CTi0d\nottZYBwBjX6MHwHml84itdtjb3zjW6/cZ5+JpXMMwQzgACIWlA4i9bKc8xPA39I9Nx662sb33XdP\ngpVK51DXuR84iIiFpYNIvSznPBP4MPDc0v7shlOn3nXUeee9cZmI1498MkmDTAI+TkSUDtLtLDCO\nkJzzWOAw6nX7Ul+5fccd3zl1o41uKJ3jVTwH7E/Eo6WDSP0g5zwROBTHxJbbZty42aUzqOs8A+xL\nxJOlg0j9IOd8K/D3S/Mzm9999x2HXXDBhglWa1EsSbXZ1Ju6eGN8BFhgHEE55//EXTTVpy489NBd\n56688rjSOV7BPxBxe+kQUj/JOV8BfKF0jl73lunTndmipTEAHErEPaWDSP0k5/wfwClDee/2t956\nywGXXPKOBCu3OJbU7wI4nIgppYP0CguMIyznfCau3VcfimWWGXXWcce9eSClTutH+iUizi8dQupH\nOedvAj8rnaNXrfTss08t//zzm5bOoa7yRSKuKB1C6lNfBi59tTfsPmbMTftceeUOCZZvUyapn51I\nxK9Lh+glFhhbIOd8MvDT0jmkdps7evSa5x9xxFPROf1IzyDiX0uHkPrcJ4GbS4foRVvcdde9yc9y\nGrrTiPhG6RBSvxq0OejLrqrZ54orbth9zJh3JRjV3mRSX/oqEaeVDtFr/FDaOscAV5YOIbXb9A02\n2PymXXe9tXQO4GLghNIhpH6Xc55PvYvmpNJZes2WEya4QYeG6gLgxNIhpH7X2Fn6b4DJg18/6KKL\nbtjh9tt3T5DKJJP6yplEfLl0iF5kgbFFcs4LgYOATii0SG11/Z577vbwG994Y8EI1wEfI8INJqQO\nkHOeBewFTC8cpaes/fDDbymdQV3hSuBId8eUOkPO+Qng/cAMIuLjP/nJDW+fNGn30rmkPnEhcHzp\nEL3KAmMLNe5QfRCwaaj6zo+POmqH+csvX2LG0jjqHaOfL3BuSa8g5zyTusj4SOksvWCNWbMeGjUw\nsF7pHOp4Y4EDiXC2q9RBcs5/BvY+6rzzrl7/wQctLkrtcSXwcW+4tU7yb9t6VVWtD/wBeGPZJFJ7\nvfbJJ2ccf/rpKydYo02nfADYhYhH23Q+SUupqqotgTHAawtH6WrvvfbaG3e76abdSudQR7sLeDcR\nT5cOIukVpLQd9cqb1UpHkXrcTcD7iZhXOkgvcwZjG+Scp1P32phdOIrUVk+tscabLjnwwGkB7Viq\nPB14n8VFqbPlnCcAe+OYOCxvv+eeZUtnUEebBuxtcVHqcBF3APsAc0tHkXrYncCHLC62ngXGNsk5\njwf2w8FDfeaeLbbY7q4tt2x1P8YHgN2JmN7i80gaATnn26hvvM0pnaUbpYGBgTWefHKT0jnUsR4A\n9iDi4dJBJA1BxM3Um6E9VzqK1IOmUt9w+2vpIP3AAmMb5ZzH4KwN9aH/2n//dz+92mqt2vBoKvAe\nIv7couNLaoGc883UfYqfLZ2l27z5oYcmt7H1hLrLZOpl0Q+WDiJpKURcD/wdYA9xaeRMBfZ0hVv7\nWGBss5zzWOB9wJOls0htk1I6+1Of2mThqFHTR/jIU6iLiw+N8HEltUHO+ffA3+JMxqWyzbhxj5XO\noI40kXo2/8zSQSQ1IeIq6htvjonS8E0AdvOGW3tZYCwg53w7sAfgBYL6xvwVV1ztR0cdtTBGrk3A\nZOriohdSUhfLOV9PfePtidJZusXG9923aukM6jjjqMdEZ2lI3SziWuC9wKzSUaQudjP1DbdHSgfp\nNxYYC2k0ud8dsDiivvHIOutsdPXee985AoeaRH0hZX8pqQfknG8FdgVsdbAEyy5YMG/luXM3L51D\nHeVW6k3OLNJLvSDiNuox0RU60tL7LbCXm5yVYYGxoJzzvdSDx5TSWaR2+ePOO+8ybYMNbhjGIW6h\n7i/lHSmph+ScJwO7APeUztLJ3jZ58j0JViidQx1jLPWF1FOlg0gaQRH3Au8C7i0dReoiFwP7EmF/\n70IsMBaWc55OPXjcVjiK1DYXHH74u55daaVmZjJeBrzXWRpSb8o5zwR2A/5QOkun2mb8eHtzabHL\ngPcT4eaBUi+qe4zvhteJ0lD8EPgoEQtKB+lnFhg7QM75CeqejFeXziK1w8CoUcueddxx6wyktDSz\nEM8EDiBiXqtySSov5/wksCdwVeksnWi9Bx9cq3QGdYTvU4+JztKQell9U/29wLWlo0gd7OtEfJKI\ngdJB+p0Fxg6Rc55LvZPmBaWzSO0wZ9VVX//zww9/PGBJd5kC+BwRn3bQkPpDzvlZYF8cE19k5blz\nZy23YMEmpXOoqEXA8USc6Jgo9YmIOdS7S19SOorUgf6JiC+UDqGaBcYOknNeABwBfKN0FqkdHthw\nw3fcsvPOr7YUcj5wKBHfalcmSZ0h57yQekz8ZuksneIdEyZMSZBK51Axc4EPE3FG6SCS2izieeBg\n4N9KR5E6xBzgQCL8nNhBUkSUzqCXUVXVgcCPgVVLZ5Fa7dgf/GDsWo899q6XvPwUsD8Rw9kQRlIP\nqKrqYOoxceXSWUo65pxzblzn4Yd3K51DRTwMfIiIP5UOIqmwlD4OnA2sWDqKVMhU6htuE0sH0Ys5\ng/FlpJTWTykd2uTPjkjz9ZzzJcCOuJum+sB5xxyz7fPLLTd50Et3AdtbXJQEkHO+CHgncH/pLCWt\n9eijG5TOoCImADtZXJQEQMTPqDcJfbB0FKmAq4AdLC52JguML2994GULjCmlZdsVIud8L7AT9Xbr\nUs9auNxyK/3wk59cKeBp4CJgZyIeKJ1LUufIOd8F7AD8tnSWEtZ8/PEHRw0MvKl0DrXdz6nHxIdK\nB5HUQeobDtvh5i/qL1+jns3/dOkgenk9VWBszDyclFI6N6U0MaX0u5TSSimlDVNKv00p3ZFSujGl\ntGnj/T9NKR046OcXzz78OrBbSml8SunElNInUkq/SildD1yXUhqdUroupfSnlNJdKaX9WvU75Zzn\n5JwPBk5kyZthSF1r1pprrnPJQQd9gYhDiJhbOo+kzpNzfoq60f2p1BtA9Y2tx437c+kMaqsF1Ju5\nHO5O0ZJeVsQs4APYv1+9b3G/xS+6wVln66kCY8PGwJkRsTn1bKgDgB8Cx0fEdsBngR8s4RifB26M\niK0j4nuN17YFDoyI3YHngP0jYltgD+A7KaWWNl3POX+/ca6HW3keqZCZwB4HXXzx2aWDSOpsOeeB\nnPPJ1OP7M6XztMvbJ01arnQGtc0M4D1u5iJpiSIWEfF54EDqIozUa6YC7yTiP0sH0ZL1YoFxWkSM\nbzy/g3q58y7AL1NK44FzgLWbOO41EfFk43kCTk0pTaCelr4usNawUg9BznksdaHz960+l9RG1wDb\n5JxvKh1EUvfIOf8XdRuRyUt6b7dLAwMDqz/11Galc6gtrgK2JuLm0kEkdZG6+LIjMKV0FGkE/Rb7\nLXaVXiwwzh/0fBGwBvB0Yzbi4sfiD+kLafwNUkrLAMu/ynEHL9k8DHg9sF1EbA08Spt28co5PwK8\nD/hOO84ntdAA8C/AB3LOjxfOIqkL5ZwnUfegOqd0llZ6y4MPTkqwWukcaqmFwBeADzaWPUrS0omY\nRN2r+KLSUaRhmgd8BtjHfovdpRcLjC81G5iWUjoIINW2anxvOvWFCcC+wOLlR88Aq77KMVcDHouI\nBSmlPYC3jHjqV5FzXphz/ixwEPXvJ3Wbh4C9c85Vztk+GpKalnOem3M+FtiHHm0jss24cU+UzqCW\nmgTsQsTXieir3qKSRljEbCIOAQ4GHDvUjW4DtiXiNMfE7tMPBUaoZxwelVK6E5gILN6U5Vxg98br\nO/PCLMUJwKKU0p0ppRNf5ng/B7ZPKd0FHAHc29L0ryDnfAnwduBXJc4vNSGoe6BunnN21ztJIybn\nfBXwDuCXpbOMtI2mTnX2Ym8aAL5LfSF1W+kwknpIxMXAFsBlpaNIQ7QA+DL1Dbci9RUNX7Io3Buq\nqvoI8G/AG0pnkV7BFODonPONpYNI6m1VVR0KnAG8tnSW4Vru+eef/cKppy6bXr2Ni7rPA8CRRNhX\nW1JrpXQ49XXi6qWjSK9gInAEEX8qHUTD0y8zGHtezvliYDPgZ6WzSC+xCPgGsJXFRUntkHO+kHo2\n4zWlswzXJvfee4/FxZ5zDrCVxUVJbRFxAbA59SZSUicZAL4NbGdxsTc4g7EHVVX1fuoPr+sXjiLd\nCRyVc76jdBBJ/aeqqgQcB3wdGF04TlMO//d/v2HDBx7YvXQOjYgZwNFEXF06iKQ+ldJR1K0ZXlM6\nivreA8AniHACSg9xBmMPyjn/jrrnxvep7wpI7TYf+BKwvcVFSaXknCPnfCawKfCL0nma8eaHHlq7\ndAYN2wLgm8Bm3VZcTCmtnlI6btDX66SULimZSdIwRPyIHpnhr661APgO9Ux+i4s9xhmMPa6qqh2B\nH1EXHKV2uJl61qLNeSV1lKqqdqfuzdgVY+Iqc+Y8/n+//e01E6TSWdS064BPd2vD+pTS+sAVEdEV\n/2YkLYWUPky9PHXD0lHUN64ETiRiSukgag1nMPa4nPOtwLbUOzI9XziOetts4ARgN4uLkjpRzvkG\nYBvgM8BfC8dZoi3vvPM+i4tdawbwESL2bGVxMaW0fkppUkrp3JTSxJTS71JKK6WUNkwp/TaldEdK\n6caU0qaN92+YUrolpXRXSumUlNKcxuujU0rXpZT+1Pjefo1TfB3YMKU0PqX0rcb57m78zC0ppc0H\nZRmTUto+pbRKSunHKaVbU0rjBh1LUieJuAx4O/A56s/xUqtMBj5IxActLvY2ZzD2kaqq3gpUwKFY\nXNbIeQ44E/haznlW6TCSNBRVVb2BegOqj9OhRbx/OPvsm9Z+5JFdS+fQUllA3d/sq0TMbfXJGjMM\npwLbR8T4lNLFwK+AI4FjI+K+lNJOwNci4r0ppSuAn0fEf6SUjgW+HRGjU0rLAitHxOyU0prALcDG\nwFsYNINx8IzGlNKJwOoRkVNKawNjImKTlNKpwD0RcUFKaXXgVmCbaMPfQ1KTUnoDcApwFF4nauTM\nBr4CnE7EgtJh1HoWGPtQVVVbUA8g3lHWcCwCfgJUOecZpcNIUjOqqnon9bLp7Upneal/rqq/LBOx\nTukcGpIALgVOJmJyu07aKPhdExEbN77+J2A54GTqGSOLrRARm6WUZgFrRcTClNJrgL80CozLAd8D\n3k3dv3sTYANgRV65wLgu8LuI2DyldALwhog4OaV0e+PnFjbOvQawd0RMatkfQtLISGlr6j7+bi6m\n4Rigvk78IhGPlQ6j9vHuRB/KOd+dc/4wsDPw36XzqOsEcAmwec75GIuLkrpZzvkWYEfgEKBjCiBv\nePTRaRYXu8bVwA5EHNjO4uIg8wc9X0Rd0Hs6IrYe9NhsCcc4DHg9sF1EbA08Sl0kfEURMROYlVLa\nEjgYuKjxrQQcMOjc61lclLpExHgi3gMcAEwrnEbdaSywIxFHW1zsPxYY+1jO+Zac83uBvYDbSudR\nV7gG2CHnfFDOucRFlCSNuJzzQM75IurNXw4FiveR3XrcuIdKZ9ASjQV2J+IDRNxROswgs4FpKaWD\nAFJtq8b3bqEuHEBdVF9sNeCxiFiQUtqDemk0wDPAqq9yrouo+7etFhETGq9dDRyfUkqN828z3F9I\nUptFXApsBpwE/KVwGnWHscDfELFrh42JaiOXSOt/VFX1d8BXqZv9SoPdCnwh53x96SCS1GpVVS0D\nfJR6g7S3lchwwve+98fV//rXnUqcW0s0HvgSEb8pHeSluzynlD4LjAZ+BpwFrE29ZPoXEfGVlNLG\nwAXASsBvgcMiYt1G38VfN372duCdwN9ExPSU0oXAlsBV1D2XB59vLWAm8NWIqBqvrUS9xHIX6skM\n0yLiQ63+W0hqkZSWBz5BfTPBHaf1UtcBpxAxpnQQlWeBUS/SuKj6GPAvwPpFw6gTTAK+lHO+tHQQ\nSWq3qqpGUS8d/Wdgo3adNw0MLPrnr3xlboLXtOucGpJbgW8Cl9KlH6BTSisD8yIiUkqHAB+NCHty\nS1qylEZRz3z+PPWMf/W3K4B/JeKW0kHUOSww6mVVVbU89TKxT9OBje/VUkG9FPoM4Dc554HCeSSp\nqONU8mEAAAngSURBVKqqlgUOB/4J2LTV53vr/fff9bHzz39Hq8+jIQnqmX3fJuLG0mGGK6W0G/X4\nnoCngb+PiKllU0nqKnX7g32BL1L3MFb/WLyh2SlEjC8dRp3HAqOWqKqqnYB/BD4CrFA4jlpnNvBT\n4Myc85TCWSSpI1VVtSf1mPi3wKhWnOOASy65YYu773YHz7KeA/4d+A4RjomS9HJS2pO60LhH6Shq\nqYXUPXdPJeKe0mHUuSwwasiqqno9cBRwLC80/1b3m0jdU+n8nPOc0mEkqRtUVbUe8CngaGDNkTz2\n577xjTtXmjdvqyW/Uy3wOHXvwjOIeLx0GEnqCiltDxxD3b/41TaGUneZAfwQOI+Ih0uHUeezwKil\n1ujT+CHqGRx7US+zUXdZBFwOnJFz/u/SYSSpW1VVtQJ1T6p/BHYY7vGWnz9/zue/9rUVUr0xh9pj\nEfXOxz8Cfk3EgsJ5JKk7pbQK9aq3o4B3FU6j5gTwO+qbbVcQsahwHnURC4walqqq3kY9g+MTwOpl\n02gIHgfOBc7OOT9UOowk9ZKqqnYEjgP+jiZncGw5fvxt+1922bALlRqSB4AfAz8lYmbpMJLUU1La\nlLrQeATwhsJptGQPUrfL+ikR08tGUbeywKgRUVXVytR3qw6gntVor8bO8TT1Ll+XAlfmnOcXziNJ\nPa2qqhWBD1LPbPwgsNJQf/aIn/70hg2mT7f/YuvMox4PfwSM6dbdoCWpa6S0HHXf4qOBvYFlygbS\nIIvHxB8D/+2YqOGywKgRV1XVaGAfYH/qCyv7cLTf48Bl1APGdTlnl3tJUgGNMXFf6mLj3sDyr/b+\nL55yytTlFi7cqB3Z+sgzwG9o3GgjYm7hPJLUn1J6E3Ag9bi4G7Bs2UB96WnqMfEy4LdE2INfI8YC\no1qq0ZvqfdTLxfYFXl82UU+bQX3xdClwU87ZfhmS1EGqqlqd+ubbIcB7ecmF1aqzZz960ne/u1aJ\nbD3oSeBXwH8C1xDh7H1J6iQpvZZ6Usq+wAeA15QN1NP+TN1//3LgBiIWFs6jHmWBUW1TVdUoYFfq\ni6v9gfXKJuoJU6kvni4Fbss5+w9akrpAVVVrAHtSX1TtDayz6403jn3fddfZFL95k4FrqWdljPEC\nSpK6RL2M+j3AftTLqb1OHL47qQuKlxExrnQY9QcLjCqmqqrtgPcDOzcea5ZN1BUeBG5uPMbknO8u\nnEeSNAKqqnrHIRdeuPMmU6Z8mHrZ2OjSmbrAQ8B1wPXA9W7UIkk9IqWtqQuNewDvZCl6GfexB4Gx\njceVbtSiEiwwqmNUVbURsAsvFBy3AEYVDVXWAuBPvFBQvDnn/JeykSRJLZfSssAO1BdWOzWer100\nU2eYQT0eXg9cR8TUwnkkSa1Wz27cjvrm227U14uvK5qpvIXUMxTH/s/Dm2zqABYY1bEajfF34oWC\n4zuBNYqGaq3HgT/wQkHxtpzzc2UjSZI6QkrrUhcaFz+2B15bNFNrPQqMA+4AbgVuI+LhspEkSR0h\npQ2prxN3bDy2AVYsmqm1nqIeCxcXFP/ohmXqRBYY1TWqqkrAJsCWwFsHPTag7tPRDbuQPQdMAx4A\n7h/0mJxzdiaGJGnoUtoI2ArYGHjboEe3bKgWwEzqfsL3A/cBE4BxRDxSMpgkqYvUM//fyovHwsVj\n47pAKhduyBZSXyNOHvS4F5hMxOMlg0lDZYFRPaGxgcx6vLjwOLgA2c5p9E/y4uLh4GLiTDdikSS1\nVEqr88KF1cbUy6vXeslj5TYkmU09O3/xY3ExcfHjfiKcqS9Jap2UVqYeCwffjFuXemXc6xr/XbXF\nKRZRXyPOAp5oPB6nHgsXFxPvJ2JBi3N0tFR/fjk0In7Q+Hod4PSIOLBsMg2VBUb1haqqVgZWazxe\n85LnqwArvMLjeWDOUjzm5pwH2vV7SZLUlJRG80KxcQ3qBvorNh6Dny9+BPWY+Dwwf9DzxV8/xwsX\nTI8DTxAxv32/kCRJTar7PK7Bi4uOi5+Pph4DFz94ydeDX1vI/y4kPgE8hYWXJUoprQ9cERFbFI6i\nJllglCRJkiRJ0itqFACvAm6i3mxnJrAfsA5wJnWLlmeBYyLi3lT3yvw59YSey4HPRMToVN/kvJy6\nl/RywJci4vKU0i8ax5sMXNM45hURsUVK6RbgqIiY2MgyBvgsMAn4N+oNYpcD/iUiLm/xn0KvYJnS\nASRJkiRJktTxNgbOjIjNgaeBA4AfAsdHxHbURb8fNN57GnBaRLwDmDHoGM8B+0fEtsAewHdSSgn4\nPHB/RGwdEf/vJee9CPgIQEppbWDtiLgdOBm4PiJ2bBzrWymlVUb8t9aQWGCUJEmSJEnSkkyLiPGN\n53cA61PPZvxlSmk8cA5172eAnYFfNp5fOOgYCTg1pTQBuJa6J+ZaSzjvxcDiXowfAS5pPH8/8PnG\nucdQt3VZb6l/K42Ibth1V5IkSZIkSWUN7q+8iLow+HREbL0UxziMejn1dhGxIKU0nbow+IoiYmZK\naVZKaUvgYODYxrcScEBETF6K86tFnMEoSZIkSZKkpTUbmJZSOggg1bZqfO8W6iXUAIcM+pnVgMca\nxcU9gLc0Xn+GV9/R+yLgc8BqETGh8drVwPGNJdaklLYZ7i+k5llglCRJkiRJUjMOA45KKd0JTKTe\nqAXgM8BJjaXQGwF/bbz+c2D7lNJdwBHAvQARMQsYm1K6O6X0rZc5zyXUhcqLB732VerNXSaklCY2\nvlYh7iItSZIkSZKkEZNSWhmYFxGRUjoE+GhE7Lekn1P3sgejJEmSJEmSRtJ2wBmN5ctPA39fOI9a\nzBmMkiRJkiRJkppmD0ZJkiRJkiRJTbPAKEmSJEmSJKlpFhglSZIkSZIkNc0CoyRJkiRJkqSmWWCU\nJEmSJEmS1DQLjJIkSZIkSZKaZoFRkiRJkiRJUtMsMEqSJEmSJElqmgVGSZIkSZIkSU2zwChJkiRJ\nkiSpaRYYJUmSJEmSJDXNAqMkSZIkSZKkpllglCRJkiRJktQ0C4ySJEmSJEmSmmaBUZIkSZIkSVLT\nLDBKkiRJkiRJapoFRkmSJEmSJElNs8AoSZIkSZIkqWkWGCVJkiRJkiQ1zQKjJEmSJEmSpKZZYJQk\nSZIkSZLUNAuMkiRJkiRJkppmgVGSJEmSJElS0ywwSpIkSZIkSWqaBUZJkiRJkiRJTbPAKEmSJEmS\nJKlpFhglSZIkSZIkNc0CoyRJkiRJkqSmWWCUJEmSJEmS1DQLjJIkSZIkSZKaZoFRkiRJkiRJUtMs\nMEqSJEmSJElqmgVGSZIkSZIkSU2zwChJkiRJkiSpaRYYJUmSJEmSJDXNAqMkSZIkSZKkpllglCRJ\nkiRJktQ0C4ySJEmSJEmSmmaBUZIkSZIkSVLT/j8aThI7BISyKwAAAABJRU5ErkJggg==\n"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "## Pull aggregated back to driver and convert to Pandas dataframe for plotting\nbrand1SentimentDF = brand1SentimentDF.toPandas()\nbrand2SentimentDF = brand2SentimentDF.toPandas()\nbrand3SentimentDF = brand3SentimentDF.toPandas()\n\nimport matplotlib.pyplot as plt\n\n# Plot sentiment\n%matplotlib inline\nplot1_labels = brand1SentimentDF['SENTIMENT_LABEL']\nplot1_values = brand1SentimentDF['NUM_TWEETS']\nplot1_colors = ['green', 'gray', 'red']\n\nplot2_labels = brand2SentimentDF['SENTIMENT_LABEL']\nplot2_values = brand2SentimentDF['NUM_TWEETS']\nplot2_colors = ['green', 'gray', 'red']\n\nplot3_labels = brand3SentimentDF['SENTIMENT_LABEL']\nplot3_values = brand3SentimentDF['NUM_TWEETS']\nplot3_colors = ['green', 'gray', 'red']\n\nfig, axes = plt.subplots(nrows = 1, ncols = 3, figsize = (23, 10))\naxes[0].pie(plot1_values,  labels = plot1_labels, colors = plot1_colors, autopct = '%1.1f%%')\naxes[0].set_title('Percentage of Sentiment Values in all katyperry Tweets')\naxes[0].set_aspect('equal')\naxes[0].legend(loc = \"upper right\", labels=plot1_labels)\n\naxes[1].pie(plot2_values,  labels = plot2_labels, colors = plot2_colors, autopct = '%1.1f%%')\naxes[1].set_title('Percentage of Sentiment Values in all justinbieber Tweets')\naxes[1].set_aspect('equal')\naxes[1].legend(loc = \"upper right\", labels = plot2_labels)\n\naxes[2].pie(plot3_values,  labels = plot3_labels, colors = plot3_colors, autopct = '%1.1f%%')\naxes[2].set_title('Percentage of Sentiment Values in all taylorswift Tweets')\naxes[2].set_aspect('equal')\naxes[2].legend(loc = \"upper right\", labels = plot3_labels)\n\nfig.subplots_adjust(hspace = 1)\nplt.show()", 
            "execution_count": 27
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "### 7.2 View sentiment over time for brands<a id=\"view\"> </a>"
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "          DAY SENTIMENT_LABEL  NUM_TWEETS\n0  2017-07-05        positive           2\n3  2017-07-06        positive           2", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>DAY</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>NUM_TWEETS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>2017-07-05</td>\n      <td>positive</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2017-07-06</td>\n      <td>positive</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 28
                }
            ], 
            "source": "## Get aggregated sentiment for each day\nbrand1OverTimeDF = brand1TweetsDF\\\n                    .groupBy('DAY', 'SENTIMENT_LABEL')\\\n                    .agg(F.count('TEXT_CLEAN').alias('NUM_TWEETS'))\\\n                    .orderBy('DAY', ascending = True)\n\n## Get total tweets each day\nbrand1TweetsPerDayDF = brand1OverTimeDF.groupBy('DAY')\\\n                        .agg(F.sum('NUM_TWEETS').alias('NUM_TWEETS'))\\\n                        .orderBy('DAY', ascending = True)\n        \n## Convert back to Pandas\nbrand1OverTimeDF = brand1OverTimeDF.toPandas()\nbrand1TweetsPerDayDF = brand1TweetsPerDayDF.toPandas()\n\n## Identify rows with positive sentiment for each day\npositive1Index = brand1OverTimeDF['SENTIMENT_LABEL'] == 'positive'\npositive1TweetsDF = brand1OverTimeDF[positive1Index]\n\n## Identify rows with negative sentiment for each day\nnegative1Index = brand1OverTimeDF['SENTIMENT_LABEL'] == 'negative'\nnegative1TweetsDF = brand1OverTimeDF[negative1Index]\n\n## Check results\npositive1TweetsDF[:2]", 
            "execution_count": 28
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Repeat for brand2\n## Get aggregated sentiment for each day\nbrand2OverTimeDF = brand2TweetsDF\\\n                    .groupBy('DAY', 'SENTIMENT_LABEL')\\\n                    .agg(F.count('TEXT_CLEAN').alias('NUM_TWEETS'))\\\n                    .orderBy('DAY', ascending = True)\n\n## Get total tweets each day\nbrand2TweetsPerDayDF = brand2OverTimeDF.groupBy('DAY')\\\n                        .agg(F.sum('NUM_TWEETS').alias('NUM_TWEETS'))\\\n                        .orderBy('DAY', ascending = True)\n        \n## Convert back to Pandas\nbrand2OverTimeDF = brand2OverTimeDF.toPandas()\nbrand2TweetsPerDayDF = brand2TweetsPerDayDF.toPandas()\n\n## Identify rows with positive sentiment for each day\npositive2Index = brand2OverTimeDF['SENTIMENT_LABEL'] == 'positive'\npositive2TweetsDF = brand2OverTimeDF[positive2Index]\n\n## Identify rows with negative sentiment for each day\nnegative2Index = brand2OverTimeDF['SENTIMENT_LABEL'] == 'negative'\nnegative2TweetsDF = brand2OverTimeDF[negative2Index]", 
            "execution_count": 29
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Repeat for brand3\n## Get aggregated sentiment for each day\nbrand3OverTimeDF = brand3TweetsDF\\\n                    .groupBy('DAY', 'SENTIMENT_LABEL')\\\n                    .agg(F.count('TEXT_CLEAN').alias('NUM_TWEETS'))\\\n                    .orderBy('DAY', ascending = True)\n\n## Get total tweets each day\nbrand3TweetsPerDayDF = brand3OverTimeDF.groupBy('DAY')\\\n                        .agg(F.sum('NUM_TWEETS').alias('NUM_TWEETS'))\\\n                        .orderBy('DAY', ascending = True)\n        \n## Convert back to Pandas\nbrand3OverTimeDF = brand3OverTimeDF.toPandas()\nbrand3TweetsPerDayDF = brand3TweetsPerDayDF.toPandas()\n\n## Identify rows with positive sentiment for each day\npositive3Index = brand3OverTimeDF['SENTIMENT_LABEL'] == 'positive'\npositive3TweetsDF = brand3OverTimeDF[positive3Index]\n\n## Identify rows with negative sentiment for each day\nnegative3Index = brand3OverTimeDF['SENTIMENT_LABEL'] == 'negative'\nnegative3TweetsDF = brand3OverTimeDF[negative3Index]", 
            "execution_count": 30
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "scrolled": true
            }, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/plain": "<matplotlib.figure.Figure at 0x7fb35d37ab90>", 
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAABJUAAAKDCAYAAACjaztiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "## Define values for matplotlib\nx = brand1TweetsPerDayDF['DAY']\ny1 = brand1TweetsPerDayDF['NUM_TWEETS']\npy1 = positive1TweetsDF['NUM_TWEETS']\nny1 = negative1TweetsDF['NUM_TWEETS']\n\ny2 = brand2TweetsPerDayDF['NUM_TWEETS']\npy2 = positive2TweetsDF['NUM_TWEETS']\nny2 = negative2TweetsDF['NUM_TWEETS']\n\ny3 = brand3TweetsPerDayDF['NUM_TWEETS']\npy3 = positive3TweetsDF['NUM_TWEETS']\nny3 = negative3TweetsDF['NUM_TWEETS']\n\nfig, axes = plt.subplots(nrows=4, ncols=1, figsize=(20, 10))\naxes[0].plot(range(len(y1)), y1, linewidth=2, color='purple')\naxes[0].plot(range(len(y2)), y2, linewidth=2, color='blue')\naxes[0].plot(range(len(y3)), y3, linewidth=2, color='yellow')\naxes[0].set_xticks(x.index.tolist())\naxes[0].set_xticklabels([date.strftime(\"%Y-%m-%d\") for date in x])\naxes[0].margins = 0\naxes[0].set_xlabel('Date/Time')\naxes[0].set_ylabel('Num of Tweets')\naxes[0].set_title('Number of Tweets Over Time for All Musicians')\naxes[0].set_xlim(0, len(y1))\naxes[0].legend(loc=\"upper right\", labels=['katyperry','justinbieber','taylorswift'])\n\naxes[1].plot(range(len(y1)), y1, linewidth=2, color='blue')\naxes[1].plot(range(len(py1)), py1, linewidth=2, color='green')\naxes[1].plot(range(len(ny1)), ny1, linewidth=2, color='red')\naxes[1].set_xticks(x.index.tolist())\naxes[1].set_xticklabels([date for date in x])\naxes[1].margins = 0\naxes[1].set_xlabel('Date/Time')\naxes[1].set_ylabel('Num of Tweets')\naxes[1].set_title('Number of Tweets Over Time for katyperry - All, Positive and Negative')\naxes[1].set_xlim(0, len(y1))\naxes[1].legend(loc=\"upper right\", labels=['All Tweets', 'Positive', 'Negative'])\n\naxes[2].plot(range(len(y2)), y2, linewidth=2, color='blue')\naxes[2].plot(range(len(py2)), py2, linewidth=2, color='green')\naxes[2].plot(range(len(ny2)), ny2, linewidth=2, color='red')\naxes[2].set_xticks(x.index.tolist())\naxes[2].set_xticklabels([date for date in x])\naxes[2].margins = 0\naxes[2].set_xlabel('Date/Time')\naxes[2].set_ylabel('Num of Tweets')\naxes[2].set_title('Number of Tweets Over Time for justinbieber - All, Positive and Negative')\naxes[2].set_xlim(0, len(y1))\naxes[2].legend(loc=\"upper right\", labels=['All Tweets', 'Positive', 'Negative'])\n\naxes[3].plot(range(len(y3)), y3, linewidth=2, color='blue')\naxes[3].plot(range(len(py3)), py3, linewidth=2, color='green')\naxes[3].plot(range(len(ny3)), ny3, linewidth=2, color='red')\naxes[3].set_xticks(x.index.tolist())\naxes[3].set_xticklabels([date for date in x])\naxes[3].margins = 0\naxes[3].set_xlabel('Date/Time')\naxes[3].set_ylabel('Num of Tweets')\naxes[3].set_title('Number of Tweets Over Time for taylorswift - All, Positive and Negative')\naxes[3].set_xlim(0, len(y1))\naxes[3].legend(loc=\"upper right\", labels=['All Tweets', 'Positive', 'Negative'])\n\n## Rotate x-axes for legibility.\nfor ax in fig.axes:\n    plt.sca(ax)\n    plt.xticks(rotation = 45)\n\nfig.subplots_adjust(hspace=1)\nplt.show()", 
            "execution_count": 31
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "### 7.3 Top keywords<a id=\"keywords\"> </a>\nExtract and display top keywords expressed in the tweets referring to the brands."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from pyspark.sql.functions import explode\n\n# Explode keywords\nbrand1KeywordsDF = brand1TweetsDF.select(explode('KEYWORDS').alias('TOPKEYWORDS'))\nbrand2KeywordsDF = brand2TweetsDF.select(explode('KEYWORDS').alias('TOPKEYWORDS'))\nbrand3KeywordsDF = brand3TweetsDF.select(explode('KEYWORDS').alias('TOPKEYWORDS'))\n\nbrand1TopKeywordsDF = brand1KeywordsDF.select('TOPKEYWORDS').rdd.map(lambda row: row[0]).toDF()\nbrand2TopKeywordsDF = brand2KeywordsDF.select('TOPKEYWORDS').rdd.map(lambda row: row[0]).toDF()\nbrand3TopKeywordsDF = brand3KeywordsDF.select('TOPKEYWORDS').rdd.map(lambda row: row[0]).toDF()", 
            "execution_count": 32
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# UDF to filter profanity words and other non-insightful words\nfilterList = ['the', 'eh', 'beep', <add your own terms here> ]\ndef filter_profanity(word):\n    if word in filterList:\n        return None\n    if \"http\" in word:\n        return None\n    return word\n\n# UDF to return lower case of word\ndef toLowerCase(word):\n    return word.lower()", 
            "execution_count": 33
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# Process extracted keywords to filter profanity and change to lower case\nudfLowerCase = udf(toLowerCase, StringType())\nbrand1TopKeywordsDF = brand1TopKeywordsDF.withColumn('TOPKEYWORDS',udfLowerCase('text'))\nbrand2TopKeywordsDF = brand2TopKeywordsDF.withColumn('TOPKEYWORDS',udfLowerCase('text'))\nbrand3TopKeywordsDF = brand3TopKeywordsDF.withColumn('TOPKEYWORDS',udfLowerCase('text'))\n\nudfFilterProfanity = udf(filter_profanity, StringType())\nbrand1TopKeywordsDF = brand1TopKeywordsDF.withColumn('TOPKEYWORDS',udfFilterProfanity('TOPKEYWORDS'))\nbrand2TopKeywordsDF = brand2TopKeywordsDF.withColumn('TOPKEYWORDS',udfFilterProfanity('TOPKEYWORDS'))\nbrand3TopKeywordsDF = brand3TopKeywordsDF.withColumn('TOPKEYWORDS',udfFilterProfanity('TOPKEYWORDS'))", 
            "execution_count": 34
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# Group by TOPKEYWORDS and computer average relevance per keyword and also number of tweets for each keyword\nbrand1KwdsNumDF = brand1TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.count('TOPKEYWORDS').alias('KWDSNUMTWEETS'))\nbrand2KwdsNumDF = brand2TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.count('TOPKEYWORDS').alias('KWDSNUMTWEETS'))\nbrand3KwdsNumDF = brand3TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.count('TOPKEYWORDS').alias('KWDSNUMTWEETS'))\n\nbrand1KwdsRelDF = brand1TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.avg('relevance').alias('KWDSAVGRELEVANCE'))\nbrand2KwdsRelDF = brand2TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.avg('relevance').alias('KWDSAVGRELEVANCE'))\nbrand3KwdsRelDF = brand3TopKeywordsDF.groupBy('TOPKEYWORDS').agg(F.avg('relevance').alias('KWDSAVGRELEVANCE'))\n", 
            "execution_count": 35
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "# join dataframes into one\nbrand1TweetsKeywordsDF = brand1KwdsNumDF.join(brand1KwdsRelDF,'TOPKEYWORDS','outer')\nbrand2TweetsKeywordsDF = brand2KwdsNumDF.join(brand2KwdsRelDF,'TOPKEYWORDS','outer')\nbrand3TweetsKeywordsDF = brand3KwdsNumDF.join(brand3KwdsRelDF,'TOPKEYWORDS','outer')\n\n# Define keyword score as product of number of tweets expressing that keyword and average relevance\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.withColumn('KEYWORD_SCORE',brand1TweetsKeywordsDF.KWDSNUMTWEETS * brand1TweetsKeywordsDF.KWDSAVGRELEVANCE)\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.withColumn('KEYWORD_SCORE',brand2TweetsKeywordsDF.KWDSNUMTWEETS * brand2TweetsKeywordsDF.KWDSAVGRELEVANCE)\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.withColumn('KEYWORD_SCORE',brand3TweetsKeywordsDF.KWDSNUMTWEETS * brand3TweetsKeywordsDF.KWDSAVGRELEVANCE)\n\n# Sort dataframe in descending order of KEYWORD_SCORE\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.orderBy('KEYWORD_SCORE',ascending=False)\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.orderBy('KEYWORD_SCORE',ascending=False)\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.orderBy('KEYWORD_SCORE',ascending=False)\n\n# Remove None keywords\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.where(col('TOPKEYWORDS').isNotNull())\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.where(col('TOPKEYWORDS').isNotNull())\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.where(col('TOPKEYWORDS').isNotNull())\n\n# Remove the brand name from the list of top keywords\n# Note we want to keey one brand name in another brand's list because that could be of interest\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"katy\")\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"katy perry\")\nbrand1TweetsKeywordsDF = brand1TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"katyperry\")\n\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"justin\")\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"justin bieber\")\nbrand2TweetsKeywordsDF = brand2TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"justinbieber\")\n\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"taylor\")\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"taylor swift\")\nbrand3TweetsKeywordsDF = brand3TweetsKeywordsDF.where(col('TOPKEYWORDS') != \"taylorswift\")", 
            "execution_count": 36
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Top Keywords from tweets mentioning katyperry\n+--------------------+-------------+------------------+------------------+\n|         TOPKEYWORDS|KWDSNUMTWEETS|  KWDSAVGRELEVANCE|     KEYWORD_SCORE|\n+--------------------+-------------+------------------+------------------+\n|         swish swish|            5|0.9588405966758728| 4.794202983379364|\n|         music video|            4|0.9115470051765442|3.6461880207061768|\n|                vote|            4| 0.884224995970726| 3.536899983882904|\n|billboard dance club|            4|0.8477847427129745| 3.391138970851898|\n|                 amp|            5|0.6252798080444336| 3.126399040222168|\n|              friend|            3|0.9811710119247437| 2.943513035774231|\n|               feels|            4|0.7342885136604309|2.9371540546417236|\n|               views|            4|0.6526602506637573|2.6106410026550293|\n|                time|            3|0.7708459893862406| 2.312537968158722|\n|               feat.|            5|0.4406650006771088| 2.203325003385544|\n|               video|            4|0.5388682447373867| 2.155472978949547|\n|                  ft|            3|0.6815443436304728|2.0446330308914185|\n|         lyric video|            3|0.6663476824760437| 1.999043047428131|\n|swish bishes\ud83c\udfc0 watch|            2|0.9922400116920471|1.9844800233840942|\n|       digital sales|            2|0.9742680191993713|1.9485360383987427|\n|                song|            3| 0.649350663026174|1.9480519890785217|\n|           attention|            2|0.9583365023136139|1.9166730046272278|\n|          dark horse|            2|0.9393815100193024|1.8787630200386047|\n|prismatic world tour|            2|0.9361480176448822|1.8722960352897644|\n|       billboard hot|            2|0.9280830025672913|1.8561660051345825|\n+--------------------+-------------+------------------+------------------+\nonly showing top 20 rows\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "print \"Top Keywords from tweets mentioning katyperry\"\nbrand1TweetsKeywordsDF.show()", 
            "execution_count": 37
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Top Keywords from tweets mentioning justinbieber\n+------------------+-------------+------------------+------------------+\n|       TOPKEYWORDS|KWDSNUMTWEETS|  KWDSAVGRELEVANCE|     KEYWORD_SCORE|\n+------------------+-------------+------------------+------------------+\n|            follow|           26|0.8963579627183768|23.305307030677795|\n|               amp|           15|0.7159108599026998|10.738662898540497|\n|     vip packages\u2026|           10|0.9964770078659058| 9.964770078659058|\n|         despacito|           15|0.6195692668358485| 9.293539002537727|\n|                ft|           11|0.8021693608977578| 8.823862969875336|\n|     bkstg presale|           10|0.8304200172424316| 8.304200172424316|\n|              love|           10|0.7182109951972961| 7.182109951972961|\n|       philippines|           10|0.6325240135192871| 6.325240135192871|\n|           retweet|           10|0.6111530959606171| 6.111530959606171|\n|              guys|            7|0.8500784380095345| 5.950549066066742|\n|              vote|            6|0.9529718259970347| 5.717830955982208|\n|             spree|            6|0.9464104970296224| 5.678462982177734|\n|              tats|            6|0.9360204935073853|5.6161229610443115|\n|official fan video|            6|0.9126629829406738| 5.475977897644043|\n|           tickets|           13|0.4207659157422873|5.4699569046497345|\n|          belieber|            7|0.7604727234159198| 5.323309063911438|\n|                uk|           10|0.5268328011035919| 5.268328011035919|\n|            summer|            6|0.8689666589101156| 5.213799953460693|\n|              sync|            8| 0.632955476641655|  5.06364381313324|\n|          saturday|            5|0.9722449779510498| 4.861224889755249|\n+------------------+-------------+------------------+------------------+\nonly showing top 20 rows\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "print \"Top Keywords from tweets mentioning justinbieber\"\nbrand2TweetsKeywordsDF.show()", 
            "execution_count": 38
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Top Keywords from tweets mentioning taylorsiwft\n+------------------+-------------+------------------+------------------+\n|       TOPKEYWORDS|KWDSNUMTWEETS|  KWDSAVGRELEVANCE|     KEYWORD_SCORE|\n+------------------+-------------+------------------+------------------+\n|              kiss|            6|0.9197408258914948|5.5184449553489685|\n|  sultry pop magic|            4|0.9216832369565964|3.6867329478263855|\n|      live forever|            3| 0.873779316743215|2.6213379502296448|\n|              time|            3|0.6947599848111471| 2.084279954433441|\n|      biggest fans|            2|0.8632540106773376|1.7265080213546753|\n|          playlist|            2| 0.836839497089386| 1.673678994178772|\n|               amp|            2|0.8306369781494141|1.6612739562988281|\n|              life|            2| 0.700468510389328| 1.400937020778656|\n|                ig|            2|0.5945299863815308|1.1890599727630615|\n|           witness|            1|0.9997289776802063|0.9997289776802063|\n| possible expenses|            1|0.9969090223312378|0.9969090223312378|\n|gym crushing bench|            1|0.9958959817886353|0.9958959817886353|\n|         song ykur|            1|0.9955539703369141|0.9955539703369141|\n| 28th best selling|            1|0.9953100085258484|0.9953100085258484|\n|             shade|            1|0.9938849806785583|0.9938849806785583|\n|         pop music|            1|0.9925559759140015|0.9925559759140015|\n|      yo everybody|            1|0.9925500154495239|0.9925500154495239|\n|            sketch|            1|0.9912359714508057|0.9912359714508057|\n|             allen|            1|0.9899489879608154|0.9899489879608154|\n|             walls|            1|0.9877820014953613|0.9877820014953613|\n+------------------+-------------+------------------+------------------+\nonly showing top 20 rows\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "print \"Top Keywords from tweets mentioning taylorsiwft\"\nbrand3TweetsKeywordsDF.show()", 
            "execution_count": 39
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "brand1TweetsKeywordsPandas = brand1TweetsKeywordsDF.toPandas()\nbrand2TweetsKeywordsPandas = brand2TweetsKeywordsDF.toPandas()\nbrand3TweetsKeywordsPandas = brand3TweetsKeywordsDF.toPandas()", 
            "execution_count": 40
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from wordcloud import WordCloud\n\n# Process Pandas DataFrame in the right format to leverage wordcloud.py for plotting\n# See documentation: https://github.com/amueller/word_cloud/blob/master/wordcloud/wordcloud.py \ndef prepForWordCloud(pandasDF,n):\n    kwdList = pandasDF['TOPKEYWORDS']\n    sizeList = pandasDF['KEYWORD_SCORE']\n    kwdSize = {}\n    for i in range(n):\n        kwd=kwdList[i]\n        size=sizeList[i]\n        kwdSize[kwd] = size\n    return kwdSize", 
            "execution_count": 41
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/plain": "<matplotlib.figure.Figure at 0x7fb338edf750>", 
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAABRgAAADeCAYAAABWkWwTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsnWd4XMXVgN/ZXrTq3bIkW3LvBjcw\nbmDTYzo4lACBBEJIICSQAiT5IAmBQAiBhNCrKaYYTDHNBdNs3HBvsmVZlmT1ttq+9/txV9Kudlda\nFSOXeZ9nH+/eO+XMjHzPnTNnzghFUZBIJBKJRCKRSCQSiUQikUgkkp6g6W8BJBKJRCKRSCQSiUQi\nkUgkEsnRizQwSiQSiUQikUgkEolEIpFIJJIeIw2MEolEIpFIJBKJRCKRSCQSiaTHSAOjRCKRSCQS\niUQikUgkEolEIukx0sAokUgkEolEIpFIJBKJRCKRSHqMNDBKJBKJRCKRSCQSiUQikUgkkh4jDYyS\n4xYhxHAhhLcH+TRCiJeFEPVCiM/7Sw6JRCKRSLrDEaT3hgoh6ntbToRyjUKIZiFEdgxpTUIIRQiR\nE+X+j4UQS2Ks91UhxJ3dlVcikUgkRw5Hio7sa4QQ3wghrujH+ouEENMC34/ovpL0HmlgPMYIvFi3\nfvxCCEfQ78v7uK6Qh5UQYl7gYXF+X9ZzBHIqMA3IUhRlRn8KIic1EonkeEfqve+FPtV7iqLsUhQl\nsbfldBwPRVFciqLEKYpS1tuyFUV5WlGUc3tbjkQikfQnUkd+L/SpjjzWnE8URSlQFOXrwM+QvhJC\nnCGE2NOP4kn6GF1/CyDpWxRFiWv9LoQoBq5TFOXTw12vEOIc4CXgckVR3j/c9XUXIYROUZS+elDn\nAXsVRXH0sxz9hhBCA6Aoir+zaxKJRHK4kXovMlLvHZ0IIbSKovj6Ww6JRHJsIHVkZKSO7JrDJFuP\n+0pydCA9GI8zhBBmIcRjQohyIUSpEOIBIYQ+cO8MIcQeIcSfhRC1Qoh9QoiLYyjzAuBF4KJgBSKE\nGC2EWCaEqBNCbBdCnBe4fooQ4oAQQgSl/aEQYrUQIk4I4RRCxAeu3yOEcAkhzIHfDwgh7gt8TxZC\nLBRCVAVkvb21TCHEDYG6HxNC1AG/FULohBD/EkLUBFZK5nZox/VCiGIhRJMQYm+ktgshfgY8CswK\nrPz9PnD9poD7d40Q4i0hREbgeusWrBuFEEXAlhj6c0Gg/uEBmd8UQhwKrAAuF0IMC6T7BXAhcFdA\nlkVCiLuEEC93KO8JIcTfA9+/CfTpOiFEQ6DshKC0pwTGoV4IsV4IcXLQvW+EEP8nhFgNtADZEa5d\nLoT4skP9vxdCvNZVuyUSieRwIPXekaf3RAfvDCFEhRBietDv+4QQTwW+W4XqrV8b0E2rhRBJQogH\ngUnAUwG5HhQdtj0H8j0shPgo0MYvhRB5HcQ5L9AHVUKIv3Toz0+DZIo4tkFkCFVHNwkhPhNCDIgl\nb0DGR4QQHwsh7KieHRKJRPK9IHXkkacjgc8BrWj3NJ0Q0JsrAuNQJYR4XghhC5TZ6fyvw3VtYDxL\nhDq/fCaonOFCCG+g3QeAD0R0HXymEOLboHJXCSFWBf3+VghxRuB7hRBieoS++gPwNjA4qK0pEfpD\ncjShKIr8HKMfoBg4rcO1+4FVQCqQAXwL/CFw7wzAC/wNMACnoRqNBkUp/xtgMVALzOhwLx4oBy4H\ntKiTgFqgEBBAETA7KP2HwE2B72uAswPfPw9OG7h3ZuD768AiIC5Q7j7UVTKAGwJtuT5Qvxm4BdgM\nZANpwBeAN5A+CagHCgK/s4ERUdp9A/Bp0O+zgApgLGACngA+DtwzAQrwPpAImCOUNzxIjhuAXUB+\n4LcOuCrQRhPwX+CboLyvAncG/c4HmgBr4LcRqANGBY3Z/kCdccAS4KmgvDWBcdcE2lUFJAXl3QsM\nA/QB2TpeSwAag/9mgO2t4yk/8iM/8nM4P0i9d9TpvcDvCmB60O/7gnTTL4E3Au3RBfrVGjQeVwTl\na607J/D7VaASmBjQUW8Az3VI+1FAzkGo+uyKjm3ubGyD6qlHNQ6agMe7mbcWmIKqe439/f9IfuRH\nfo7ND1JHHpU6MujanMA4ZAb6+r7AvXy6nv+16rafoc7N8gJj8h7wZFAdCvAUYAn0UUQdHMjrCvxr\nCoxteeC7DXACtkC5bTo+Ql+dAezp7/8b8tN3H+nBePxxOfBHRVGqFUU5BNwLXBl03wv8WVEUt6K6\nz38KXNRJeXOAbcDqDtfPB7YoivKyoig+RVG+RTVmXaioT5MXgCsAAis6M4FWL7eVwEwhhBEYgmpU\nmxlYXRkLfBm4dyFwh6IozYqi7AEe7tCWvYqiPBmo3wFcAjyoKEqZoihVqAq1I6OFEKZAmu2dtDuY\ny4EnFEXZpCiKE7gdOE0IkRmU5i+KotQrnbiDCyF+i/rQn6EoSjGAoiheRVFeCLTRCfwZmCyEMEUq\nI5BvLXBB4NK5wD5FUbYGJXtWUZQdiqI0A38EFgSu/wh4S1GUTxVF8SuK8gHq2M4LyvuUoig7FUXx\nKO0u88HXGoA3aR/bE4EU1AmcRCKR9AdS7x2hei9GPKgTv4KATvxWURR7N/K/rijKekVRPMBCYHyH\n+38LyLkP1bNiQVgJnYxtUJrFiqJ8HeiP3wOnCiHSYsz7hqIoqwO619WNtkkkEklvkTryKNCRgbnb\nssA4VATaNjNwr5iu53/Bsj2gKMp+RVEagT+g7kATQWnuVhSlJSBbRB0cyLsJmA5MRTVMfxv4Ph3Y\npChKUyxtkxxbSAPjcUTgwZGJ6sHWyn5gQNDvqsCDMPh+Z6cx3oG6mvGGCLjTB8gDZgRcqeuFelrk\nhUBW4P4LwAUBQ9kC4BNFUaoD91YCs1BX89cCy1AfoCcDmwMPtEzUv9+STtpyoIOs2R2utfWDoih1\nqA/cXwAVQoh3hRCFnbS7Y7nBZdWjevF1JktHNMBtwL8CSgNQY18IddvXXiFEI7ADdZWvM/fx5wko\n6MC/L3a437EPLELdJp0HXNFhzE4kdPwjtaPjtedR+7K1/leUIzS2iEQiObaReu+I1nux8jRq/7wh\n1O17fxVCaLuRvyLoewuqZ0swHfsn0th3NbYh5SiKUgs0B8rqVl6JRCL5vpA68ujRkUKIbKGGwzoY\nmBM+hep12kpX87+IsgW+m4HkwG+/EnpQWmc6uHVcZgS+r0Adl5mB35LjEGlgPI4IrA5VoD7gW8kF\nDgb9Tu3gHZcLdHYaYxOqa3M28ErQA+cAqit4YtAnTlGUWwKy7ENd9TgXdWUp+CH4BTAOOBv14bQR\n1WV7Hu0PqwrAH5AvWluUDrKWAwM7pG9PrCjvK4pyaqAtJairY7FQRlCfCiESUd3FO5OlI37UuB9/\nEWpQ5FauCVyfjbr9eHhrNZ2U+wYwVQgxCrXPFna437EPWgKehwdQvRGDx8yqKMo/u2hHx2srAZMQ\nYirqC0I0BSeRSCSHFan3jmi9F4wddTtWK21eHop6MvTdiqIMR53EXAxc1oM6otGxfyKNfadj27Ec\nIUQyqiGzPMa8fdEOiUQi6RZSRx6xOjLSvQdQdeVoRVHigetonw9C1/O/iLKhttmBul09rO4udHBH\nA+NKum9glPrvGEMaGI8/XgH+KIRIEUKko7pFvxR0X496aIhBCDEH1bj1ZmcFBlZl5gFDgReEeprw\nYmCCEOJSIYQ+UN5UIcTQoKwvAHcBg1Fd5IPL2wrcCKxU1FOJ16I+SFcG0rhQg8L+NRB8tgA1RkRw\nWzryOnCrECJLCJGK6q4OgBBigBDibCGEBTWeRDOqkoqFV4DrhRq42IQaO2pZsCdiLCiKshE4B3ha\nCHF64HJrDIsa1HgX93bIdgi1/4LLaQbeDci1IrDdIZirhRBDhRBxwJ9o337wPHCxEOJUoQYANge+\nZ9INAi8rL6LGG6lRFGVtd/JLJBJJHyP13hGq94LYCCwIeO1PBeYHyXmaEGJkoI8bUbfrtcoZpgN7\nwB1CiAQhRD7wc9p1YjCxjO18IcQUoW7TuxdYrihKZYx5JRKJpL+QOvLI05GVqIe8BBs8bQEZGgPX\nfxWcIYb5X7BsvxZC5Ap1i/m9wMLA/C2MLnTwKlTD72hgQ+AzApiAahSOhUNAemBeKjkGkAbG44+7\nUeNibEV9of+S0HgTxagPjgrgGeAaRVH2dlWooiith4NMQHXZrgdOR/XAK0ddLbkXVUm1sgg1AO/r\nSnjMoZWoqzLrg35bCX1Y/TTw735UV/mngJATtDrwKOqDcCtqXJDXg+5pgd+itrsGNYDtzzspqw1F\nUd5DDX78Lmo7MwmN9xEzAWPcecCLQohTUd3SqwJybSb8Yf0EMCmw1eDVoOvPA2OI7D34IqpyOYiq\nIG4L1L0XdavCn4Fq1H79JT17TrzQSf0SiUTyfSL13hGs9wL8HlVn1AO/Qz34pJUBwDuoXjFbgA9o\nNwL+E7hKqCeSRoqdFQvvA9+hTlYXEWEyGtgq19XYvoQ6iaxGnWD9qBt5JRKJpL+QOvII05EBvXE/\nsC4wxxuPOk7TgQZUQ2okI29n879W/gu8BXyFelhOLR2MlR2IqoMDht9twAZFjWvpB9YB2wP3YuE7\n1H7aH2hrclcZJEc2IoqxWnIcItSj5B9VFCXW+BK9ra81TsZliqLEusohiYHAauBaIEMJCh4shPgG\ndYw7W83ri/ptqCtSwxVFKekqvUQikfQHUu/1D0KIkcBaRVEsXSaWSCQSSb8gdeTRRbT5n0TyfSI9\nGCX9yQKgUSqQviUQ6+RXwEv9qFxuRnXPl8ZFiUQiaUfqPZXRwL7+FkIikUgkRxRSR/aQI2T+J5Gg\n628BJMcnAU+6fOCH/SzKMUXArbwE2Iu6DaE/ZKhAPanzB/1Rv0QikRyJSL2nIoT4Peo2sxv7WxaJ\nRCKRHBlIHdlzjoT5n0TSitwiLZFIJBKJRCKRSCQSiUQikUh6jNwiLZFIJBKJRCKRSCQSiUQikUh6\njDQwSiQSiUQikUgkEolEIpFIJJIe060YjEIIuZ+6m+isNrz2pj4vV2Mw4ne7wq8bTWh0+sNSZ6xo\nLXFozVb8zpaIcugTUvC1NOH3uDstJ1obJRLJ4UNRFNHfMhwtSJ0okUgkxzZSJ8bO0aQTR1x1Jzte\nvg/F5w25nn/2tRgTUtm58P4uyxj384cA8LmdbHni94dFTknvSEjSsmjdEAB+PHcvB4o6n3t2xokz\nrNz56ADWfWHnnp8d7CsRJT3klDNs/ODKJApGGjHHaXA0+6k44OGV/9aw6sP+s4Mc68SiE6UH42Hm\ncBn6ci/+ScTrfpezX42LAL6WZqy5BSSOmxrxfsbMszFlDOiynGhtlEgkXaMxGbCMGUz8rPHEzxyP\nZczg/hbpuCQhXsPcmRYuv9DGeWdau52/4Fdn9LlMw+46j5M+vr3tkzhpUJd5rIUZJE0p6HXd05b+\nptdlSCQSiUTSW4RG26v825+/h/pd6/tIGsmRzsTpVixxGqadFtffohz3nP3DRO76zwDGTbNgsmio\nLvfi8SgUjDLhbPH3t3jHPfIU6Rix5AwiZfIsFK8PfUISzcU7qVq1FIDci67Dvn835pxB6K3x7H/9\nf/jdLpInTkdnS6Ry5Xtt5aROOw1r/lCEELjrqin78DUyZv8Ajd6AeUA+Tbs2Ez9sLBWfLca+fzdZ\np1+MKWMAzft2ULVqKca0LFKnnoo5M4fci64DoOTNp0FRSJ44nYQxk7AX7w6p05pbSNr00xFaHftf\nezyqV2D+5TdT/PK/yTztfIRGS/nHb5D/w5soXvgYqSfNJS5/KE17tlGzZnmn7Y5G2smnEzd4OIrf\n12lfd9bGjnKYs/NImTyb0sXPAZBz3tXUrFmOo2x/t+U7nhl1/6UU/2859qLKTtPp4oxMfuuXfDWv\n65XdVvQJZiYturnLPIdThq4Y8/DltJTUtP22Dctiy20L8Tb3/99Ld9sr9DpSFpxK4rzJCKM+5F7D\nR2uofvkT/I7+b9fxgNEgKNuSj0GvLvbtKvKw+EN72/07f5XEkAIDd/21hpKD3mjF9Dl7HvqQkudW\nkTJ9CLnXzIg5n9DKNUmJRCKRHP1sf+HeXpfhbqrD62rpA2kkRwPrv7Bz1mWJfPmR9I4LJjVTx2U3\nprDwsRpqK7+fd9krbk4F4LPFjfzzd+W4XarztNWmwdly1DhSH7NIA2M3MCSmUvTsPwAYdNUtNG5b\nj6tGNYb4vV5K334uJH3t+i9In3lOyLXE0SdS+u6LOA8dBNHuYWov3oWrthKtyUzFsnew5g/Dvn83\nFZ++heJXGHLjnVSt+ghXVTll7y/EkvMHSt54Kqw+n8uBMTWr/aLQkH32Ava99AjepoZO2+dz2NEa\nzWjNVoRWh8ZowuuwY8kZhGXAIIoX/ofci6+npXQvjrL9Udsdse+S04krHMm+Fx6m8Lo7Ok0brY2R\n5OiKWOU73tl6+2v9LUK/ytBSUkPRQ0vbfuf/ZPZhq8tamIEpK4GaVbv6vGyh1zHgrh9hHpGHt64J\nx9od+OqaQAi0STYSTp+MaUgOpXc/g9/V820ikq7RaODtFzLbjIuR0OkEP7wgjo2bXfzz8frvTTaf\n3YXD7sJRmhpzHvueQ9j3HDqMUkkkRx/GhFRGLlC3Rm5+7i68TnsXOSQdiR84nPy5V9FUuot9Hz/X\n3+JIJKBIA4UknLWf2zlvbN+/ux/tTJ4dxw+uTOK9l+u/NwNjSoZqwnrx4ao24yKAvUl6Lx4JSANj\nN3DVVrUpHVdVBfrE1DYDo6N0X0xlHHjrWVKmzEafkEzNmhU0F20DwNvSjMZsQfF5UbweNDodQqcj\n89TzERoNWqMZoREo/u4pPb0tHp/D3qVxEcBRXoI1fwh+twuh8WDNG4KzvARjSgbOigOAgvNQKaa0\nrDYDY6ztNiSl4qoqB0XBWV3RrTa0EkkOZ1V5SBqhCZ3MxyqfRPJ9kTytEJ/Tc1jKTjr3JMwj8qh9\nYwW1b6xA8YUqWs/BKpIvnk3S/JOpeX35YZFBonLVpTbmzrRw5c8OseprJ8Ub8sLSvPexnT/cmsRZ\np1m+VwNjv3KUTty0WhhcoGPkaD0jRukZOlxHSoqGnFwdVqvA61Vwu6Gh3k9tjZ/SAz727fWyZZOH\nDevc1FT3z0vvlGkGLvmhhRMmGUhL1+ByQUW5j3Vr3LzxWgvfbej8WSQEnHa6iR+cb2b8RD3JKRrs\nzQqVlX52bPPwwRInq1a48Hi+v3HNytYyc46REyYZKByiIzNLizVOoNcJmu1+6mr87C3y8t9/N7Np\no+do/ZM7brDlDEVrMJGQP6q/RZH0EflnX4vP0Uzj/h1kTjkTY0Iq7sYaDn37MXVBW4rNqdlkTj0L\na/ZghEaLo7KEitVLaT5YFFKeKSWbrKlnYsnMQ2sw0bB3C+Vfv4+7sSZqGq+jGXt5cVi6gvNuJC5H\njcn33aO/iii/3+chccgEMiefjiE+JaLssaK3JpB10jnY8oaj1Rtx1JRTsXopTfu393lZeadfhS13\nGDtfeQBPs/pOIbQ6hlx8C0IIdi96GL/XE3V8drx8X7dlzz/zanxuF417N5M59Uw0BhP1u9ZTsWYp\niq99x1qs46O3JjB0wa877atYy5J8P5wwvfvhf/qKyrLvb/ePJHakgbEbGFMzQGgABVNaFtXffNZ2\nTyG2N1h3fTVlH7yK1mSh4Lo72PXoH9tK6Ig1bwhas4XypYuIHzGhvS5FQej06pt/F2/O3uYmtGZr\n0GEzImJdAI6y/SSfOIPGHd8htFoSx0ymZs0KFJ8X27CxgMCUOZCmgFG0O+321NdgTMsCITAmp3eZ\nPlIbnVXlYXL43S50VjUWhtBoMaZlh5YTo3zHAmMfvQp9ogWh01Dz+S72/edTJr16E+uu+h9as4FJ\nr/+c7Xe+Qd23exn3nx+x/e43sRZkMPDKk7EMSmPbHa/RuKUUgLhhWQy+eW5YeQCK10/2xZPJOHMs\nOquRQ0s3U/Ls553Kpnh95CyYStrc0Wj0OkqeW0XVZ1sBSJpS0KcymLKTGHzzXEwDkihbtIaKJRu6\nLK+7dNWvSScOJvuSyWhNevxuH+uvfgJjmo2CW87ANjIbv9dHxpljAdh43dMofgWt2UDej2eQNKUA\nodfis7vY/MuXAQXF62fCM9d12d+2GeNw7i6l5rVlEe/XvL4cy7hC4qaPlQbGw8xVl9gAePXt5qhp\n9u1XX4wG5eqjpgGwFqQz6MZTiRuWhc/ppnrFDvY/tRK/K9Q4JHRa9f/ZaaMwpsXjqbdTvXIHJc99\nEZY2Vgp+dQYZZ6h/q7v//n7b/9uOxFq34vX1uj3F/wv92x1213mAwp4HPiDv+lmkzBiG1mzAWVbP\njj+/jfNgXY/abo0TzJxt5LTTTcycY8Jmi+6NajAIDAaIi9MyIEfLmHHtY6oosHG9m8VvOlj8pgNH\nL7fvjJ+o5/V3Qr1Pp4w7RF1tuxEzLk7wlwcSOfMcU0g6vR4Kh+goHKLj0sstLH3fyV13NNDQEG4A\nzcvX8eC/Exk7PvTvMzFJkJikYegwHT8430zRHi+/+UU9WzZ372/s1ttt3HhzeCyrRa+08IfbwxdF\np55k4Lob4jhlljF4A0gICQkaEhI05A/WMWeuif3FXha+0MLLz7fgdh8/7wNHE02lu0gZMZWGfZu7\nlS9n+gUcWv8pnpbGwySZpDfE54/CkpFH5frP1LBRI6eQO++KNiOdKSWLwgt/gbOmnIMr38Tv9ZA0\n7AQGz7+Rfe89RVPJDkA1kBWcfyPO6jJKly9C8XlJHXcKviBv4UhpjIlp2PKGh6QD2LvkSXTmOEZe\nfXdU2Y0JaWRMnkfl+s/wuZxtsntammgu3R1zH2hNFgov+gWK30f5l0vwOVtIHDqRwedcx94lT7a1\nsbdlfffYbQCULn+doZfeRu7cyyla/B9QFLJPOhdjQiq7Xn8Iv7f9GR1pfOJyhrS1rzuyx+ePID5v\nOIfWfYoxPoX0E07F53ZSuU6dJ8c6Pq11dlZfd8a6Iz5pi+pzdHrBuKmWfqvf65V6/UhEGhi7gc/R\nQs4PrkQfn0jT3m24a6PHihMaDdlnLcCcnYs+PpGqLz/GXVdF3oKbUHxehBDUrf+y0/ocZSWkTptL\nxuxzVe+/VhSFxh0bGXzVrbgbaild/FxbfcaUDDRGY3udtZWUL32dnPOuBhQOvPk0Pqcjcn0VB4gb\nNIyKT95C6HRkzb2Qg+++iN/jpqWkiPwf/ozmvTtwHCyOKrPGYCTr9IswpmYhtFqMKRlUrnwfV80h\n7Pt2MuiKX+Cuj2F1KUIbHWX7I8gh8DTWk3/5zXjtjbh66B15LFD50WYqlmxAo9diSFWNG817DmEZ\nlIYx3Ub9un3YxuRQv2E/WqsRd3Uz7upm6lYXMf7Ja0PKGnDx5IjlAQidBo1ey4Zrn8KQHMeEZ66j\n6pMtOEpro8omdFo8DQ42XPsUlrwUxjxyJY2bD+CqbKRudVHfyXCwjmF3zafooaU0767ghBdvwL67\ngqYd5Z2W11266teGTQeo+Wo33kYH5twUAFxVTWz7wyIKf3MWLcXVlC1aE1Jm/k9no7Ua2XDdM/hd\nHozp8XibnejijAidJqb+1qcn0fDZuk5ldxYdJGHupB63XRIbY0YaaXF0/uJTU+fD7VHISO880PzI\nv1xM1bJtVH68GdvIAWTNn4ghOY6d9yxuTyRg2N3zSZyQR/ni9bSUVGPJSyXrvBOwFmaw9Tev9qgd\nxf9dRtnra0iYEO6BGXPdt7/Wvhim0fS6PcVPrAhbXDOk2Bj+pwvw2p2UPLsKodWQeEI+7sruGx+m\nnmTgymuszJxjxGDo/QGyQsCEEwxMOMHALb+28a9/NPHKSy196lk3YpSer1apsVWtcYLnXkkJMwxG\n4oyzTeTla7nq0toQI+OIUXqeW5hMUnLXcTcLCnUsejeVn1xdy6qVvY/vOmJ0qNwpqRruvichzFga\nC3n5On53dzxXXWvl7t829Il8kr6l8cAONj3T/RN400ZPp3rb19LAeISis9jY8/ZjuOrUuVLDvi2M\nuPIPbfezpp6F3+umaPF/2gxfDUWbKLzwZgbMOJ8dL/0NUD3adCYrdTvW0rBXNUI3dvBoi5SG/dup\n+i58MVbxedu8+6KhNZrZ9dpDbd5wrbJnnDi3WwbGtPGz0FsT2PHy33A3tJa1FVNyBplTz+yWgbGz\nslrxuZ0Uf/Q8Qy78JeknnIqz6iCp406h5JOFbePQSqTxKZh/Y1v7uiO7zhxH0eL/tuW1ZA0iPn9k\nm4Ex1vFprbN2+5qo9XVnrDvicStMmRPHRdclUzjKhE4HB4rcfPh6A++/UkekIwLu+s8ATjkjdL6w\nf7eL60+PbYfcvIsS+NndGVjiIuvSTatb+PWCkoj5zrgkkcHDjRHzRss3aJiRi65LZtw0C8lpOpwO\nP/t2uPj4zQY+WhR5N6MQMOuceE67IIHCUUbik7Q4mv3UVHrZt8PFV5828/Wnzbgc6jvCgHwDl/0s\nhYIRRvKHGtEFQgE9sTT8wMDXHq/h6furondQjMTFa7nzsWxSM/SkZrabrz7eOzwkXbR+GTbOxJwf\nJDBmspkB+QYMJoG90c+L/6rm/YX1YYZKs0XDO1uGto312CkWLvlpMsPGmbHaNNRX+9j4tZ2Fj9ZQ\nui885FR3+/RYQxoYu4G3qYHSd18Iu94xFiKA4vdz8L2Xw64Xv/RI2LVDy98FoCVoO2/r90jpASo+\neSum+gCa9+2ked/OiPeC8bucbP/H7W2/tz/Y/r36m89CPDYhcrv9bhcHl0SWo/LzD7qUIZiObYws\nh8LBJS9FzB9JvmOZ7IsnYx2SQdXHW9q8AJu2lxE3JANzXirlb69jwGVTsA5O6zKW2qGlmxj8i3lh\n5bVS/o66Au2ubcZZVochzdapgRGgern6QtiyvwZ7USW2Edm4Opn890QGxefHOjiN4X++oC2dKSeZ\nph3lXZbXHbrqV/PAZLIvmqRu2Vdg6x2vhW1X7kjS1AK2//6NNi+uSH3TVX/7nW60ts5XErU2qzzk\n5XvAYhZUVHZ+oBWoPuVdrcAeeOlLKt7bCEDlx1tQfH4yz52AtTCj7W8u5eShJE8tZOc9i0Pie7qr\nmxn0s1N73A6fw42jtBZDSlxlT9DqAAAgAElEQVRUo3xXdSdPK6T2K3XiodFrKe5le4LLa8U2MpuD\nr37D/mfaJxmt3suxYDAI5l9o5kfXWhk6/PC9GiUla/jTXxOYM8/EL2+sw97cN1bGESN1bQbGv9yf\nEJNxsS3vKD1//UcCN12venqmpmp48vmkmIyLrWh18M//JHLROTUU7+udm8jQYe39P3K0nv8+nURW\ndu9Oex2Qo+WpF5P536PNPHS/DNAvkRxuvC1NoUYtRaHpQPtcJC5nCPVF34V41YFqZMyePr9ta7K7\nsZbmg0UMmHUR5oxc6rZ/S0tlqAEhljTdwWNvCN1qG5A9cciE6JkiYMsdirOmrM1AFygMe/k+UkZP\nQ2h1KDG61XVWVnA5jspSyr56l+yT5+N3O6ndvoa6nWvDyos0Pub0nB7J7ne7QgyvrvoqrJn5bb9j\nHZ/WOjurrzdjfealCVzxi1RcToXaSi8JyVoKR5u4ebSJCSdbuOemgygdXtW/XNpEQ42PhGQtuYUG\n8oYYY6oL4La/Z3H6xQmsWdFMc6Of4eNMZOcZACja5mTTagdb14UfEtSaz+NW2PCVneZGP3N+EN9l\nvrMWJPKL/8tEowWfT6HmkJeEJC1jJlsYM9nCsnca8UTw5P/1/VnMvTABgJZmP1VlXuKTtOQNMZI3\nxMj0M21cflJRmzEsIVlL1kA9Lc1+dnznZPSJZgB2bXaGneBcfqBvQkIJjRq3vL7GS32Nl7FT1LnO\nptWh/VC0LXx+M3y8mUfeal8kd7b4qa7wkpym46Y/ZXDiTCt3/TjynHBAvoG5Fybw6/uzQIGqCg8a\noR5sc9r5CZw8z8b8MeFxObvbp8ca0sAYIy2l+0IMgBLJ98VJZyVy2yP5ALQ0+bhyQuRtRBuufRLF\n68eSl8qUd25h9fyHKX97LSPuvQjF7WPfY5+SNK2Q/J/ODpmER8Lb6IhYXis+e/sDXFEU1UrSBWmn\njaJiyQYseSlYC9Jp2l7WafqeyOCsaKBlfw3Vy7ZR+uo3WAen4zhQE1N53aGzfjVlJjDszvl8e9lj\n+OwuBl55ckhen92FKSM+rMz6tcUMuGQyex5ait/txZAah7ch3Nu4s/4+eO8LDPzr9Qx+9nfUv/81\nLZv3qoe8ALrkeLJ/+0OEXkfJ7/7Xo3ZLYufVt5vbtklH4/nHMtDrBY882XmM3Ir3N4b8Ln3lazLP\nnUD63NHsCxjkBlw6BWjdLvz90lXdKdOHhhgEe9ueqk+3hhkYfQ53l8+1znC7FQqH6A6rcTGYGbOM\nbNieGXU7cHe54BILzz5pZ+nyNPIHd78Nc89QvQNPnWviv88k9UiG+HgNH3+extCB5V0n7gSDQZCR\nqeWdpakkp/TdyeVCwA03x3HDzXE8/u/YDY1CaEgZOZWU4VMxJaWj+Hy0VJdSuXE5jtqu25o3ewFx\nAwrRW+Lxe9w4asuo3bmWmp1rOglzI0gqnEDysBOxpA5AaHR4Whpx1JTRULyFhuKt+L3tXhM6k5Ux\nV99DS1UpO998CNvAYWSMn4MlNQeh1eKsr6Rm+2qqt0beOWNJzyV5yETisgowJqQitHp8bgfO+krq\n92ygetvXKJHce1B37CQWjCepcAKW1IHoTBb8Pi9eR7Mq7/6t1O78NiTPoHlXkzh4bMg1Z20F21+/\nP2o/GhPSyJhwKpbUAZiSMwEYcclvwtId2rCMstXvRSwjIW8UKSOnYU3PRWs007B/G1WbP6e5rChi\neknPUSL+bYvgBIhYXh5RKHr7sbZf5vQcxv38Qfa8+Sj28n1R0ww65zri80d0SBcrkeTqOixVWA6h\nPr/iBhSGXK/fvZH63RuJFrKqu2V1LMeUlInf7cTT0qT+fxYalA6Ws4jjE3StO7J77B10mKIQGsci\ntvGJVGd4fT0f6+x8A/MGh3uN3nBnOhdcm8xzywr40azQZ8GydxtZ9q664H/mpYnc+rfMqOUHc9Of\nMjj94gSWvdvIfbe0z3cKR5n4z5J8Bg0z8fP5+/H5lJjyvfFkbaf5/r04n2FjTTzx10reeCrcAWH2\nufG8s3koF5+4O+QglEkzrcy9MIFDBz1ceUpsz8Ft6x1tXoKtnn4A//hNOcW7Do8DQ1O9L8QzsdVz\nMZK3Ykd2bHREHHdQ+23K7DguuyGFVx8P32Gp0wvmX5XEGUN2hBmf51+VxE1/ygjL05M+PdaQBkaJ\n5Ahn7MmxbeU98eUbUXx+fC1uiv75EQC+Fje6OBO1X6qrK41bSsk4Ywz23eokvvA3Z2EdlIYpO4nC\nX5/F7vuW0LSjnPQzxpJy8pCw8npKS3E1OpuJic9ej9Bp2PvIJ20eeh1l8NTZ2Xzryz2TQVHY8ce3\nyL9hDie8fCPOsjp23PUmeHydlmfJTaHgV2e0/bYNy6J04VdRq+msX/1uLxVLNjDhyWvxtbip/HhL\nSN6KJRsZ+odzOeHlG/E2OvjuxucA2Pffz8j/yWwmPHMdQqvB2+hgy69f6dYLrWtvGRUPvU7GTeeT\ncukcUi6dEyp3Uwvlf1+IuyR6eAdJ3/DPx+u5/EIbt96QyOuL2+MwajQwbpSR3/w8kYt/EIe9xc9j\nT3fPwCRaX9yD/zY0Ak9DC3sf+aQvxO8eXdTdmacydL89kcpTPF17i3bFc0/ZufIaK7pO3ox8Xija\n42XrZg/bt3nYs8tLQ4Mfo1GQkKAhJ1fLxBMNnHSKgYSEro1jFy+w9ImBcfBgHRdcbAkxLq5a6WLJ\n2w5KSnzExQlOnWvikgUWtFHaN6hAx89vDY2JWF3t59WXWtj8nZu6Wj/5g3RccbW1Wx6SPeGhRxMj\nGhcVBXZs8/D5Chd7i7zU1fqpq/Wj0QhSUjUMG65jxmwjE04wdFr+DTfHsek7D59+5OxcECHIn3cV\niYNUY5jP7cTndhCXOQjb2UMo//bDTrOnjpxG8rBJKH4/npYGdCYrcVkFxGUVkFQwnqKlT0f0YMqb\nfRnJwya11QlgSsrAlJRB4uCxbHnx/0IMjK2YktJJHTmNgTMuwu/14mlpRGeyYknNwXJKDtXbvgrT\nKdb0PIZe8Mu2336vG4+9Ab3FRlzmIOIyBxE/cDhFH4bvCtGZbQw+4xqsGfnt+X1eNHoDRkMqxoRU\nFJ83zMBYv28TXmczOpMVU1ImpqTwSVpYXSYrxvgUfG4n9soS4jIH0VJ1AL8ntB/cTeGTRCE05M6+\njOShJ6oyely4m+tJHDSGxEFjOLRxGWXfRDZKSnqG3hqPMSm9zUtOCA22gUPb7jeV7MCWNxyNzhDy\nt5xQMC7guRg5nJKjshQUsGYNimpMclSWUvzhc4y94e+dpuuu7I6q7u18aSrZScrok3DWlPf6hPlY\ny0ocMp6U0dMo/uBZXPWVDLnkV2RMOYOKb0J3kUVqY3D7+lL2jkQbn9Y6mw/u6XVZkfj6k8iLSp8t\nbuSCa5PJytWTkaPnUGnvve5SA6ccV3Tw4Gv9rdFCUpqW6gpvr/OlZekZNlZdJFzxXuT3rc3ftqDT\nC4aPN7NuVft4tnodZgzQc9vfs3jpkWoOHTw8B1EeiRwsdjNsrAlLfPT3taWv14cZFwF2b1F1s8Eo\nQk6yPt77FKSBUSI54hkXo4Hx20sfi3h940+eaftevWwb1cvaD+nZ80Dkbet7H/mYvY98HHbd2+zi\nq3mhHgabbgoPGxCMp8HRJkPpwq/D7ve1DM7yenb8MXx7fbTyADbfEnlbf2d01q/FTyyn+InIh6g4\nDtTw3Q3PhV332V0U/XNpxDzB7e2qv5vXbMexo4T4WeMxj8hHl2xDURS8NY0c+s/b+O1dTKglfcKW\n7W5+/tsq/vtAGvf/UY3DWThIT0tJAdrAbk+vV+H6W6s40MUpeMZUG66q9hdjQ7rqAeuqbr/mLKvH\nOjidum/24O8DY1t36G7dR2p7yst8vPeOg/MuNIdcr672s+IzJ5995OLLVS6czs6N/s8/bcdsFlx4\nqYVbfhNHfCcvrn2FVgd//Ye6HcfrhTturWfJ4lAP6M+Xu/h8uYtHn0xq+xsM5oGHExk1pt1wuGql\ni5//pC7kUJqN6z28946DR59IYs7c7sdEjJVJU0INhIoC77zl4OEHmig7GP3v4dOP4LF/NTNshJ4l\nH6dGTQdwz30JrF3jpr4u+haltFEnkzhoLIri58DKRdTu/BZF8aPR6UkfN5usSWdGzWsbMISBp1zE\ngc8XUbNjTZsHoC1nKHlzfoht4DAGTD2X0i/fDslnTh2gGiV9Xoo+fIqm0t2Ags4cR0LeSHTmOLyO\nyBNljc5AzikXcvCrd6ja8iWK34cQGpKHTWLgjItJHXlSmBejvXI/tbvW4qgupX7vJtytp89qtKSP\nnUn21HOIzxtJXHZBqKefEAw+41qsGXn4PS7K1nxAfdF3eFoaERoNBlsK8bnDsVeET/jrdq+nbrca\n5iRlxFRyZ14StR/b5DxUzO531fccjd7IuB//jf3LX8UZgxdp5onzSB56Ij63k5IVr9KwbwuK4iep\ncCK5My8mY/wcPM31VG35osuyJLHhczkYfM71HFr3KT6HneSRUzDEJ7fdL1/9IUMuvoWC82+ketMX\n6iEvQydizcyj+MPn2tLZcoeTMnoaTSU7cDfWodHqUFBCjFCR0iQUjgtLhxBoDSa0RvUZb0xKx+dy\n4Hc7Q7Zqe532iLIfWPZaUFEa9LZEdEYLAtFWls/laFs0qNywnMTC8RRccBPV332Ou6kendmKJT0X\nn6uFijWxL953VtbBVeozxJCQQs7sS6jZ8nVbfMKyL94hZ+aF2A/uoelA+1bOSONzaO2nMdXXXdlj\nHZ/WOlNGTY1aX8xjHYGy/ZGNPCVF7QbuAXmGPjEwNtSqz/uk1FBlmxyIua0ohHgS9ibf4BHt27YX\nfhXqcdqRxJTQchtqfTx4Rzm/CmzLnndhAt+tbmHZO42sfL8Rh/3Y2MJrsmiYdU48E0+2kFuoxkS0\nxGna4lt25ku9b0dkr8yWZrVvtNrQA3SPlz7tDGlglEiOYDLzjKTndO6JIZF0xNdop+7dL6l7t/OD\npCSHl6deauTHl8dz4nj15U8TZGdav8nFr/9Yw6pvIh+6FUzyKcMof6s9hlLanJEANGzY33at5vOd\npM4cTub8iZS98W1YGYeTLusOfffqfXs6lNeXPPV4M+ddaMbnhc8+cbLwxRa++dKFv5vvgw6HwkvP\n2Vm5zMn/nkumcEj0162sbC3lZX1nRP3Xg01hxsVWPvvEycIXVE/NjgR7Je7a6eWGa+rweMI72uuF\nu+5oYPrMyIfgWKyCFnvfDVBzs8JPflTL2jXhHnvR2Lndw6MPN/PzW8JPqW4lJVXDzbfGcc/d0T1s\n08bOBKBq8xfU7Fjddt3v9VCx7mPMqQNIHDQmYt7sKWeDEFRvC11YayrdxcGv3iX/tCtIHXUS5d9+\n2OalCKA3q4uKnpYmmkrbDQJeRzM1O0IPB4tE3e71VG5qDxegKH5qdqzGlJxJ+rhZEbdJ71+2MOya\n4vdxaOMyEgePxZKeizUjP8TAmDR4HNaMPFAU9rz/P+wVxUF5/bgaqqja3Pvg/r1FZ7KSPm42AAc+\nX0T93k1t9+r2rEdoteTNXkDWpDOp2fktfo+MUdwXuBqqqNr4OZmT5mFISMHdUEPJx+0x0111lex5\n419kTTuHnJkXgUb1oNu75AmaStpjNbqbatHoDGROPQut0Yzf5aT4vadpOVTSaRpnbXlYuqShE8md\ne3nb7+GX/xZQDYpbn7qr7Xrt1m9w1JSHyR4cZzB17HSyTzkvrCx3Qw3bX/wLAD5nC7sXPUzGlDPI\nmDQPnSUen6sFR3UZVRtWdKs/OysL1NOV88/4ER57A2VftB+YVrPlK2x5w8mdezk7X/0H3pamqOMT\n3L6+lD3W8Wmtc+hlt0WtL9ayIsrhiqyXXA5/265uk6X3B7sBrF7ezFkLEplxVjz//F37AaQXXKMa\n2TetboloaArO99IjNVSVe7rMZ7WpL5eKX/XI6wxnhMMHV77fxKbVDs6/Jol5FyUwfpqF8dMs3HhX\nOu+8WMfCR2vCYiseTYycaOaPjw8gKVWH4oe9O5xs3+igucHHmMkWcgZ1Ps9uNSR2h2O9T7tCGhgl\nkiOYcdN7ftKxRCLpf6adWcrAbB1jRxlIsGlosits3u6muCT2FfKcBVMxZSRgLzqEbUQ2GWePp2bV\nLuxF7Vvda77YSc2qneRfPxvLoDSaNpeCRmDKTiT5pCFsuFbd3qjRa9Enx6G1GjAPVF9YLbmpuGua\n8dndeOrs+N2q94XQaTAkx6G1GjHnpaJPtGAdnI6vxYW7rqXtQKKu6t56+2u4A96JPrur1+0pX7y+\nrby+ZtcOL7ffUs+Xq1xUVfb+5e9AiY+fXl3LWx+kRt0yPftUIwtfDA/Y3hMOlvp45n+db2d7+n92\nrrjaGhoiqwN/+n1DRONiK1VVfj5f4eK0eeFejEOH6di4vm+2A7XYFX58RS0b1sVuXGzlkQeb0Ovh\npzdFNzJedoWFZ56wc7A0soHXGK96H9ftWR/xfn3RxogGRkNcIpb03Kj1Nperhjqh0WLJyAs5/KKl\nuhS/143BlkTurEupWPcx7qa6qGV1pGHflojX63avI33sTAy2ZNxNnR/KFoyroRpLei5aQ+hYJwRi\nKDaWbA8xLh5p2AYOR6PT43M5qC/6Lux+3e71DJx+AVqjmficodTvixznWtI9NFo99bvWU78r8v8d\nAGftIfa9/3Sn5bjqKtn7budxo2NJA1C3cx11O9d1mqb4/fbdKZ3JXvXd5zGdXOx12jm48k0Ornyz\ny7S9KUvxedn12kMR8wW3qZVYxicW2YO9TVsp+SR0V1Cs49Na57bn/i/q/e6U1RGjKbLSM1s0bfrQ\n0UeLY9981sxXnzRz0tw47ngoG3uTj+HjzQwdY6K5wcdjf4p82GZwvic/GsTXnzZjb/Jx1oLEqPla\nDY6KAj+etzfidt6uqKv28swDVTz/UDUTT7Fw5qWJnDzXxmU3pDDhJCu3XrS/ywMJj0SMZg1/+l8O\niSlavvumhb/dUkZtZfuuobv/O6BLA2M3Q6+2caz2aSwcNQZGrc5IZuHJJGeNwmxLR6s34ve6cTub\nsNcfZP+WD3DZY39h6gk6vZlRs27EHJdGyZYPKdvd84DyRyrTLrwfITTUlm1lx1fP9rc4xz1jT5IG\nRknPEDot+qwUNBYTvtpGPFX1/S3SccuBMm+X26A7Y9sdr5N/4xwyzhqHz+mh4t31FD+5MjSRAjvv\nfZes+RNJP2MMqTOHo3j9uCobqfu6fdtQ8slDGfr7c0Oy5v90dtv3Pf/4oC1uaPzYgYy679KQtAOv\nOAmA4idXULZoTUx1e5vavbMaNpZw4MUve9WeAy9Gj4/aFyx+s2uv0u5woMTHIw82c9f/hR/uBDDh\nREOfGRiXvO3o1DAIUHbQx4a1biZOivxSvXWzJyZvwdVfuSMaGPMH952B8W/3NPbIuNjKww80MWmK\ngYknRm6rXi+4/EdW7v9L53FCnbUVEa+76iN76JlTstu+T7gh8qS/TQZzqAHU62imZPmr5M66lJTh\nU0gZNpmSla9TV7QxJu86V4QYhADOQKw1Y0JqmIFRozOQVDgBW85QTEkZ6ExWNHojGp0eoWndUhc6\nOTcnqYcdNEfYAn0kYQ4cCOOsqwg76AJUT01n3SEs6bmYUrJBGhj7hs5WMCT9z3E4PjmDDOzaHB4m\nKLewXT+U7e+5vglGUeDemw7y95cGcsqZNjRaqDnk5YNX63nlsZqoMfmC842ZbGnL21m+fTtVvaDR\nQsFIE3u29DwUks+n8O0KO9+usDNktIn7X85l2FgTM8+x8dnizvXkkcjYyea2beH331YeYlwESE47\n/KawY61PY+GoMDBq9SbGzL4ZS3xG2HWz3oTJmkzRukWHXQ5r0gCsCepLY+rA8cekgVFy5KDRCsZM\ni+55IZFEQmuzkHrFPGzTxyAM6pbHuiVfUv3CRxjzM0k8+yQaPlqNc8/BfpZUEgtFD6lxObf+5tWu\nEysK5YvXUb44uodG9YrtVK/YHlPdDev3h8U77U3dADv+rMaK6qv2AOy8Z3Gn948E3ni1hV/cFhfR\ni3HEyL47MOWrL2Lb3rl+nSeqgfGTpbFNTnZsizxBysyKEOCxB+zd42XRK70zvPp8cO8fG3nr/ejx\nGC+61MxDf2/EG3UNQIl4oAqAL4rBT2MIxPFUFFwNnW8T7nhICUBd0UaayopIHzuD5GGTyZ11KTkn\nn0fVli+oWPdJVHmAiIfGAG15NLrQcbdm5jP49GvQmW2gKDhqyrAf2o/P1YLf6yE+dzjGhLSw8rRG\n1bjsc/WNcfxwodGrISqijVXwPa3BHDWNRCI5uplxlq3tROhg5sxX4xeXlXj69DCOn/w+nTGTLVFP\nMO4q399/VRaTAarigIfdW5wMGW3i0p+m8Jeb++b9fvcWJ1vXtTB5VhxZuZHfF4K3nSek9I3u70tM\nlvbt43VV4bpx2LjDF0s6ErH06bHAUWFgzBl+aptxsbF6HxVFX+J2NKDR6jGYE9DqTfi8hz9mir3u\nIC0N5RityRwq7joOjuToJm2AgSnzEhhxYhw5hUZSMvQYzRq8HoWmeh9l+1zsWNfMqw9H9mzoDUID\nE2bYsMb3/8NaoxWMO9nG+FNsFI61kJlrwBqvRaMTOO1+qsvc7N/pZNNXTaz5pAF74/d7wISkHa3N\nwsC/XI8+KwXF7cF9sBrDgPaJtedQHbaTRqF4vdLAeISg0wl0WnBGiQ0kOTZwOBRWf+Vm3pnhL7NZ\nA/ruOb91c2yTo+1bo6f7dnVsHhzRDltJS+ubQ21efM7e7diXkdiyycOXq1ycfIox4v3EJA3TTjay\namW090iBRqcPOQSiFY0usnHY71GNtAoK2169r0dyex1NlK1+n/JvPyR/7lUk5o8hY8Kp2AYMYdfi\nf7cdGhMuU+RJS6uhLdgLUqMzMPj0a9GZ42gu20Pxpy/haQmd0A6KuzqigdHncaPnyDfKtca31Ooj\nj3/wPb+7b72XJRLJkcNJ82xcfnMqbzxZg8upIDQw74IEfnBVIgBvPtW3OyHPvDQBxa++53VnK2xr\nvooDnpjzPn5vJfe/PJCZZ9vwebN55T817N+tPuvjk7RkDdQzaVYcLz1SHZLvrAWJJKZo+eazZvbt\ndLVtr9bpBNNOi2PsZAsAe7ZGXnj0+RQOFLkZWGDgkp+ksHuzMyRmoU4v8Haxq+Jw0urdKTRw7pWJ\nvP1se7iRCSdZ0RwGT97e9umxwFFhYEwZoMZ58bpb2P7FU9+LMTESXo+DjZ882C91H0u8uWd8yO/m\nBh/XTduCx937B9AvH8xjxvykkGtFm1u4/fxdUXKEM3iUmctuyWLirPiIOwgMWkFKpoaUTD1jpsVx\nwuwEXnmonPUre+7mPGyClfyRZvKHm8kfYSJvmBmjOXySZrFpw/qvM95/ropn7u2ZIUmrFcxbkMIF\nN2aQnBF5EmWN12KNN5M33MyM+Um4/+zn+ulbaa4/+oyMpuQsEgvGEpddgCk5E63RjOL30XRgF/V7\nNlK3ewORTpbIn3cViYXj2fvBU7RU7GfgrEuwDRyK12Hn0IbPqNn6NekT5pA2dgZaowl7eTGlq94M\n21o36MxrSRg0msqNyyn7agn6uETSJ8wmPnck+rgE7OV7qd2+pu3kzUgkXzQLfVYKtW+soHbxKhSX\nhyGL2mPZ+B0uHDsPYB6R30e9Juktf749mdtvTkSfVdR1YslRzbpvIxsYbTaBTkcnHnSxUVXlp6kp\nNj1aUR79GR3NM7EjNTWRrX82W+8NjH4/fLCk716+33jVEdXACHDq6aZODIxgTEzHUR2uSw22pAip\nwVGjnmwsRO/7QvH72ffRc1jScig892dY0nNJKhxP7a7Inr3GxDRaqg6EXTclqQv1rsb2CWZc1mB0\ngS3a+5ctDDMuAujNkbf2O+sOYUpMVw96OYJx1qiHYJiSMhEaDUoHq7XQaNv6xhHDidSSrokU809y\n5HA8jk91hZe3n63l+t+l88ObUqiu8GJL1BAXcORYtbSJ9xaGxrqddloc46dZsdo0WG0acgarizcZ\nA/Tc+0wO9iY/9kY/9iYfrz1eS3MHB4uV7zcx76IEPtg1rO2ay+Hn0EEvaz9v5tX/1lBfE66LW/P9\nc1FezPk2r2nhvlvKuO3vWcyZH8+c+fGqYVJRjXwAy94Jf76nZ+v54U0pXP2rNNwuhdpKLxotJCTr\n2mJWrnyvkdXLmqP27cLHqrnjoWwmzbSyaO0Qqiu86A2ChGQtzz1YxaInD28Iu84o3etm2TuNzJkf\nz413ZXDBNcnUVXvJyNGTlKrjyfsquf636X1aZ1/06dHOEW9gFBotpjg1yHZj9b5+My5K+g6/T0Gj\nbbfcxSVoOWFOAt8s7V2MOKNZw+S5CWHXl78V24NNoxVcflsW869LpztzgsGjzPzh6cGsereOx+88\n0O1ToR77bASZedEnPv1BVr6RWx/Oo2C0pVv5DCYNj3w0god/tZ9NXx6eQxgOBwNnXULKyKlh14VG\nS8Kg0SQMGk3y8Ens/eDpqNvPTEnqCZ1xAwoBMOiNDJx5MV57I9nTzmlLZxs4lIJzfsr2hX+L6IFi\nSsrAmjWIwWddh9bY7hliyxmKLWcoCYPHsv+TFyPmjZs8AufOEmpeWxa1rd7qBkyFA6J3huR7JSG+\nb7y9JEc+nZ0UbTIJmpt7t8h28EDsCzvRDrCpro7dSNliV/B5QdvhTdJi7b1HwNbNHupq++6ExWWf\nOvF4FPT6yLJNnhpl+1dTLQZbMslDTuBgBANj4qCxUfO1VJViScvpudAdaKkqxV6xj/jcERhsKVHT\nJQ0eF3EhKnnIRFyNNSEHxrR6NaIoeFrCdbYxIS3qYTUNxVtIHDSG+LxRWFJzaKku7WaLeo7iU43g\nOnP4aegdaTywA5/bidZoJrFgfFjfJA2ZiEZvxO9xhZzaLZFIjh22rnOw6Mla9u92c/H1yQwZY0Kr\nE+zZ4uTD1xt4/5W6sDQO+zkAACAASURBVMNRJs+K4+wfJoaVZbJomDwrNITVewvrQwyMuYUGNFrw\n+8ARmBcKAUazILfQQG5hMqecaeOnZxXT3BA5nxDQEjjAxWTpPB+ohslt652c96MkTphhJStXj1Yr\nOFTqoeKAh2ceCA/X8fGiekxmwagTLWTn6knL1qEo0Fjr47tvnHz6diMr32vs9KCTzxY34rD7ueDa\nZApHmUgfoKOlyc/eHS5K9vRNTMve8MCvy9m6toUzL00kO99AQrKWshIPbz9bx/ov+j7ER1/06dHO\nEW9g1OraDS8e17Fr6T2e2LiqiYmzQlfEZ56X1GsD45R5CW2xFlrxehRWLen69EWDScPt/8lnwozI\nK/WxcMoPkhg4xMQ91xRRXx27K8qRZlwcNsHK758cTFxiz7btJaTouOuZwTzymxJWvRv7yZf9ib18\nHykjp+Kqr6SheBuOqgN47I1oDEYGzroUvcWGbeAw0sbOoHJDZONd2pjp+L1u9i19Fq3BRM6MC9Ho\nDGROOZOarV+rp40WTiBl5FQM8cnE542IeNqnJT2XQWdcg+LzUr7mQ1oOlaD4feTOWYDBlkRiwVg8\nzedy8MvwuHPaxDjs6zufICk+P0LX/1vvJSrxfeDtJTk6aGiI/japehj07m2zuip2A6PTGbmu8ijb\nnqPh9iiYdaFGO10fvFluXN+3kxJHi8J3GzycODmyIbGgUEdCgoaGhtAZZtXmzxlw0nmkjZmBo7ac\n2l1rQVEQGi2pI6eRNPSEqHUe/OodCs+9kfxTL6diw2dtB8XoTFYM8Skk5I1E8fupWPdxSL7UkdPQ\nmeNoKN6qetQF6kvIH0VcdoHangjGzlYSBo0h84R5VH63XN3WLQQpwyaROmo6pV++HdovrR57QpA6\n6mSqNrfHFrcNGMLAmZcgomwhq9u9nvSxMzGnZFN47g0c/OY96vduwudqQQgNOku86iFpslK1ZVVU\neXtCqxdixvg5OKpK27ZBg7owGLwA53M7qVj7EQNOms/AUy5C8flo2LcZRfGTVDCegdMvAKBi/Sch\n5UgkkqOfhjpfSAzENSuaWbMiNnvCv+6s4F93dj8MVv5QI4+8lYfXq3D39aUh9Qmh3r/n6RzSs/Wc\ndn48i5+ri5hv/Rf2tu3RneULpqrcw5P3VfJkjJE5yko8PH5vZbfb2JGvPlFPv/6+6E5cS59PYcnL\n9Sx5ObKdIVJZjhZ/l3UU73JFTNNXfXo0cxQYGINeBnty7rrkiGP5m7VhBsYTZsVjS9TRVN/zPWKz\nzk8Ou7b2s4Yut+sKAb/+d3TjYnWZm7XLG6k84KahxovFpiU5XceYk2wUjLGEbKPOH2Hm7ucL+cOl\nu3E0H33bhAcOMfGHpwdHjP3osPvZ/HUTuze20Fjrxe9XiE/WUTjWwriTbVhs7Xk0WsEvHsilutzN\n9m/t32cTekTd7vW46iuxH9ofdm/3W/9i+KW/QaM3kjR0YlQDoz4ukd1v/qutDOP/s3fe4VFV+R9+\np9dMei8khECooSsg1QJYsa+9rK66/nZ1d9Xd1e2rW1xXXbfYsHdEFEUBRXoPQmgJSUjvPZlMr78/\nhpRhSibJJASc93l8JPeec++5956595zP+RZNDPEzL0YRnUThR88AoKspIXz0ZMRyFcr4UV4FRrFC\njc2oo/DjZ7HqO7q3F676J2OvexhZeAyxU+bTfGyXR/IAu1aPJM67y14XslHxWBtDGaVHCiELxu8P\nRj/W7cEIA9Tqw2XZG74Expbm/o2zLBYnCsVpAqMPK8H+UHB8kP7iXjh62LfAKBBA1jixR/bspqM7\nUSePJXzUBEYtvomUeVdjM+mRKDUIxWIqt31MyrwVXuMe6upKqPj2XdIvvp3IrBndoldPVmZoK/Z0\nc5aoI0mYfhGJs5bjsNuwGbSIFWHd8R7bSvLoqMj3eZ11uRtInLWMhOkXYdF3IJYpu63hm/PdM7Cb\n2xtpK/6OyKwZpMxbQdyUBViNncjCohArwtA3lNOcv5vk86/wOI/TYad0/WuMXv5DFNFJpC28gbSF\nN+CwWRCKJN2dWlt1wkNgTJm3ApFUgVAqRx7hck+TaqLIvPRe7BYjdosJu9lEQ9632M2+4yJqUrOZ\nfMefsOg7EIpEiOVqavevp/HwFrdyjUe2IQ2LJHbyAjIuuQOH1YzVqEOmie6+Lw2Htng7RYgQIUL0\ni2vujkSuFPLms00eYqbT6YoLuPdbHVfeFkl8smTQ9UKEGGmMSIFx4oL7kMo13QlcuogfPYf40XM8\nyhfsep22OvfBlkwVxYzljwOQ+8XvsZr1SGRqEsfMIyp5MjJFBAgEWIxaOlvKqTu5E327+4qwQCBk\nzrXeM2g2lO6h5OAnAV+T69wXEJk4HpkqCpFIitWsQ9daRWPFAVprj+PLemHudc9Qd3IXZXmfIpVr\nSBq7gMiE8UiVkQgEAizGDtobi6kt2oZJ1+z1GF3EpE4lPuM8VBHJCMVSV936E9QUbsFsaMdht7pZ\njQ4FuadEv94WciKxgAsuj2D9u/7b74vIOInXjMuBuEdffmcsMxZ7iovNtRZe/UM1Bzb7iq1YR8Io\nGf/9drzb1lHj5Nzz+2T+/WhlQG1/YJH3icKkOWoe/Ku7a5JR7+DnlwW+amPop8j58LOjPMRFu93J\nmhcbWPtqI0a998mnXCnkmvvjueb++G73cqFIwMPPpvOLy094mPGPNJwOu1dxEcCibaWz5iTh6RO7\n4zR5w24xuR1DV1dCPBe7n8fpwNRajzopE4nK052/i8a8rW7iIrgyddbt+4r0S24HgYCo7FnU7fvK\nrYz+UDHhS6ajnDoGQ95Jj+OGLchBnpVC66fbPfaFGBgbVyUNqv60KcF938qTUom8YAny5DTE6jAc\nJiPWjjZq3luJXd8zYNVMm0XCipuofPk5ROowohdejCw+CYfVgr6ogOZvvsCmc3eZVI2dQPIt91Dz\nzisYK0uJvvBSwibkIFKpseu0aA9/R/O3X53eJMSaCKIXXYIqazw4HegKjtGy7WvsBu+LD73Ldx27\nrzohCNi1GVzuV9443YKvL7yt+wZDLK0oD77AWFjgP7ZkRqanwOh0Oijb+DoxE+YRnT0bWUQcYrkK\nfUM5DQc30VlTTMzEuShjvLtCt5XkoYxNJSx1nEvMEoqwdrZi0bbSUZlPe8lhjzqtJ/YhFEtQJ2Qg\nC49BoorAZtKhqz1Ja9EB2k7m4c/atbUoF0NTFfE5i1HEpiAQijA0V9NSsA9vPlkVWz5AV19G9Pjz\nkWliECvUmDuaaTyyncYj2/x+9yy6NgrXPE/U2JlEZk5FEZ2ESKbAbjFhM3aib6hwWX2eRuzkBR7b\nhGIpmjT38VRz/m6fAmPphtddFpQxyUjVkTisJowttZjaG7yWr971GR0V+cROugBVfDpSdQQd5cdo\nPr4bbVX/sryGCBEihC9iElziX2e77+9pcrprUaq1sedbN9B6IUKMNEakwBgelxXU40nl4UgVEYyf\ndzdShfukXhEWiyIslqZK74kTDNp6JFIVYqnSbdW5P0QnT2bMrB94CHdSRThRyeFEJU+ivaGQor3v\nYrN6H0gpNXGoo9KYcME9iKXucfHk6hgS1DHEjZrJid1v0N7g3UVy3Pm3EZ2S415XFU1C5jxi02Zw\nYs+b2Mz6IRcYrRYnO79sY9ktMW7bF14dNWCBcf4VkW5xHQHam20c2uY/DmBKppxbHkn02F6Wb+R3\nt5zE0OlfGKuvMPPZK42s+JF7gNhFV0dx4FstewJw+26s9u4Kpm3x/Hg4HU6f5YNB+nj3bJBmo4O/\nP1DG4Z3+76PJ4OD9Z+uoLDLxs+d7ghLHJEq47ZdJvPi4Z8D5swmrzuWGIBCKEIjEXuMwWrTuYrbN\n6F0IcVhccWRFEu+WNACdPiY72vJ8nE4HAoEQVWKGx/7Wj7egnpVN8uO3odtfgLHAJXhKE2OIuHQO\nsXctx9bUTvvnu3yeO0T/WDJ/5GRQVWZkkXz7fTgtFvTFBdgNesQaDfKUdK/CAkDUggtRjZuEobSI\njkP7kScmo5k6E0VaBhUvP4vD5PlNEodHkHLHA4iUKvTFBSAQoEjLQCDxXFGXxsaTeteDrrIlhQhl\nciLOn48yK5uqlS94CIanl7e1tSJNSPJbJ4QLg48FoP5gNIyMoEAN9cFflKruI0ZlWpr3MZ7T4aDp\n2A6fbr6Fq5/1e9yavV/A3i8CayRg1rZQs3ttwOVPRyCSoK0sQFtZEFB5p8NB8/HdNB/f7XW/sbmG\nQy/93Hd9u42Wgr20FOwNuI3+jhcoHeXH6Cj39ALwR2d1USjOYogQIYaUshNmZi5QseKOSA7u1FNT\n3jNvi0kQc+P90cyYr8JidrJ9feeg64UIMdIYkQJj7ro/dv9bIlUy9ZJHAWiqOED50S89ytstvt0n\nANRRKaROXIZEqqK5Kg9tcxk2qwGpXIM6Ko3wmNFom0s96jmdDvK+fqb7b7kqiumnrCIDJTJxPOPO\nv73bWrK+ZCcGbSN2mxm5Opq49FmERY0iIn4c4+bcQf6OV3B6MQlQRaSQPedOxBIFTRUHaK3Lx2rW\nI1dGEps+k/DYMQhFErJm38x3Xz2Fw+65Ut8lLpoN7dQWbcXQUYdQJCEsOp3EMReQPfcuj+x6Q8WW\nT1o9BMasHCVJGTJqy/qfyGfhCk+30O1rW7Hb/U+Wbn0sEYnM3UWxvdnGk3eX9CkudvHO07XEJkuZ\nd5l7IOB7fp9M7rcd2KwjY8LWFzFJnoLXG0/V9Cku9mbnujY3gRFcYusHz9XT3hRYZtIzhkCAOmkM\nmrRxyKOTkKg0LtcyhRphrywGviKl2c3ugYKdXn6DQM/v208mIXOHd6HdYbNg0bYiC4/pdivrja1F\nS/XvXifhp9ehPm8C6vMmAKCaOQ7VzHGYTtZQ//wq7Dr/78wQ/aOswsq7qwcWe+a269WkpwXH1SVi\n9jwEQiFVb/0PU22vhAsCgU+BUZ09mZr3VrqEwlPEXX4dEbPmEjX/Qpq/WedRJ+bCS9GdOEbDutWu\ndL9dp/EimideewsipYqad19Ff9IlnEcvXkb0okuIufhyGtZ+1Gf5vuoMNwqFgPTRYjJGi5g0RUp4\nhAC1WohKLUAu7/lPKnMlb5HJXH93ZQ8cKkxBCB9nNo+M75WvDNWDwV+SHYDo2HMjXMHQ9rIQIYLD\nz9ctICbDd5Iem8XB76ZtDNr5ErM13Ph0DpHJCsoPtvHGvblBO3YIT0bPiiLn8iSSJ4YTkahAphZj\ntzowdljRNpqoOd7BiW1NFO3wTDxytrPq5RbmLVWTminljc2jyT9oRCCA6HgxcUmu8Z7Z6OAfj9bR\nUG31Wa+m3EJHqx2BAMZPU/isFyLESGNECoxWUy9Bo5fY5rBb3fcFSMa0a7CadBze9BwGrWewVoFA\n6FXUO52BZLAeM/MHIBCga6vi2Nb/uQl/HY3FNJTuZcysHxA3aibhcWOISZtGU4VnTB6xVAHIObH7\nLVpre1ZstUBjxQHGzbmD6OTJSGRqopIn0Vx5yK1+VyZus76VI5v/hdXcYwHSVn+CpqpDTFny01Pn\nGXpOHjFQfdJEyhi52/aFK6L44Lm6fh0rbazcw/IOXCKmP+JSpMxc7Omm+tG/6vqVpAXgzb/UMHOJ\nBpmiZ4ISESth3mWRbPsssCzWZ5qlN7tnpSw+bOCbD1v6fRy7zYmoV9B/sUTApbfF8P6z/Xuuw4kq\nIZ3UxTd6uII5HXas+g5EUoVbRmdveMvqPCCcThxW31aqXUKmr/ZYapqo/OWLKCako5iQjjgmHIFA\ngK1V6ze7dIiBk3fcwp+eGdjvPGeiNGgCY5dvqoeW6CdVnbGq3E1cBGjZsoGImXPQ5Mz0KjAiENC0\nca2buAjgPK3fKkZlIktMQVdwxE0sbN25mah5i9FMnk7jutU47Xa/5U+vM9wCY3iEkPkLZcxbICVn\nmpTRmWKEI1CLsgZhMWuY1hj7xKAPvtCp9ZNkByAqagQ+1BAhQgSF+XdlEJfpCqWUNTemj9IhBopU\nKeLGp6cyfrHnIrhILEKqEBGeICd1SgRmg/2cFBg72uz8+Ipyrro9kvMvVJOdo8Bud9LZbufwXgOH\nduvZsKrDw8359HqpmVISU6XY7U6/9UKEGGmMSIEx2AiFYor3v+dVXAQCEhcHikTmWp0rOfCxV6tC\ngIoj64gbNROAuPRZXgVGgObKPDdxsTc1hVuITp4MQFhkmofA2HX8yvyv3cTFLozaBupO7iIle0kA\nVxUctqxp5bbH3OOXLbgqkg+fr+tX6vaFKzyTu5QcM1BZ5N+cY+nNMR5GZG2NVjat6r9Q0Npg5duP\nW7j09li37ZfdGXNWCIxiiYALr3cXGD9/bWAZsA5u0zLrQnfhdtZF4SNWYFTEJJN51Y8RisQ4HXZa\n8vfQXnIEY3Ntt5iXuvhGosef5/9A/em0/hAI/Fqcddmn9HU2Y345xvzy4LQphF9aWgcuLnd0Bu/7\n03nsEOrxk0m5/T7a9myj4+A+7Lo+whvUeoYvsOt1WFqakMbEIVKHeRzDVFOJw9z3gpsy41TG2wp3\nDwGn1YK1ow1pTByS6FgsjfV+y59eZzgQCGDRhTJuuFnJosVyRGfBaCkY4mCwXmODZSjaodP5v0FK\nZUhgPBcRiERII2MRyeRYdVqsHSN/TPZ9YPNLJ4lOU6KMlKKKkKKMlJI4LgxVlO/wMYNihLzbznWu\n/uNkr+Ki3eZEIMAtnNW5KC52YdA5+OB/LXzwv/4Zagy0XojvH+Oun8DE23NYc8UHZ7opHpwFQ+bB\n01C2D21z2RltQ87FgcWb6XJ19iZGlh/1YklyCl1rT0IRiTzMY398xvmYDW00VXgG2+6itmjrsAqM\nn73SyMTz1Exf2JNgJS5Fyn1PpvLSE4HF7Js8N8wj/qHN6uTPd3pOUE/nqns9P4D/e7wKRx9u1b54\n/ckaFlwVhTq8J45T5iQli66OYuunI3tAe+0D8YRH97wO8nN17P5qYJmGv3q72UNgTBsrZ/oiDQe3\n+kqYc+bIuvr/EIrEmDuaKHjvr17LdGWaHC5UcWlek84IJTIUsckAGOrLh7VNIbwjSSwZVP3N241E\nRw4svu/pdB7Lo/NYHkKplLCcmaTc9iNkCcngdFL0h194reMtxiK4REZi4hAr1R4Co7UtsIGvONwV\nukKZkYVY4x5CQl+Uj74oH6fFElD53nWGCplMwO/+rOH6m5R9Fx6BjBTrw5GK3e6y8pT4yHKtVIWc\ni88VRAoVGbf9FGlkLE6bFUtHG7LoOFr2b6Vhy+cACKUy4hdfSd3Gj89wa7+f5K2r9dh2zZ8mM/Na\n7wmTBsuqXx1m1a88kyqFCC45l/bEtX/rgQMUbj93RcQQIUJ453shMNr6iNE4VAj8xFjzh1Ao9iow\n2iwGL6UDQyRVYNF6z6zXhd0ahABO/WTLJ61uAiPA3OURvPanGqzmvmdLi7zEXjywWUtne9/m494y\nXVYWDryvOB1QVWxi/Ez3mDLjpqtGvMA4dpp7m6v6sP70R4eX5DTgSvgyEhGIXe2ym71fs0QZhioh\nfRhbBJpRE7wKjOEZk7rfK7q6UrI++sOAz1F848Drhgge767u5N3VwQ3W7bBY6MjdTUfubpQZWSRc\nfRPK9DEYyj0zi4uUaq/HEKldC1U2/cDbZtO6MqEbSotp29t35vL+lg8mYWECXns3mqnT+/eeKiq0\nUVVpo6nBQWurg06tA73eidHg7P6/4dR/OdMkPPl339njQww9YrFvETEYLuZnCptJH5TEKecKsfMu\nQRoZS9Our2nZtxmH1cKEX7on43FYzChTM89QC0OEOLcxddrOaQvFECFC+OZ7ITCeabt4fXsNpQfX\nBFzeV6zHwcR4EwQQ9tvpdLj8krwpb0NE7qYOdB12N6s/lUbErCUadq/3b0EnlQs5b6mnlcuWTwZm\nVm7UO2iuG1zQXG8CY8aEkZNl1hejJ7q3saFq4JmqfSXHUYePzNeNtbMdqSYKeWQcIpnSI1lL6pIf\nIBhm/8iYKfNpLTqAub1ncCaSKUmcvQxw/VbbCnNRtE9yq+e02hEIBYhjXb8Lp92OQ2/CaXcgUisQ\nSMRYKhsxFlUS4vuBoayY5s3rkaeO8iowypNTPbaJ1GFIo2KwadtdlowDxFh+ErgExeisgATD/pYP\nFmIxvPJWVEDiosXiZPtWM9u3mDmwz8LJ4sBjISUlh1xwzyQSicDv8MYyQhLchBg8YVmTMNSU07Rz\ng99yErXG7/4QIUIMjPY644gJuREiRG/GXT+BcddNQJWkpjGvgdxn9qCt6NEcsm+cyLgbJqCKV6Ov\n03H83SOcXFvYvf/y96+h4L2jxM9MInXhKKp3VLL/H7ux6nrmzpPvmsqYq8ahiFYilLjGfpZOC98+\ntIGWfP/Ce9S4aC59awVf3vYpbcU9BkqRWVFc9s7VfHXnWlpPeCYDXfrK5dgtdrY++g02o43L37+G\nys3lHFl5sLvMlHumk7YknXU3u3QpaZiUG765jTVXfUjOvdNJW5LBiY+Oc/hl7+H6AmVkzvjPEbpi\nOwqFYjpbPa2RhhOb1YhI4l/oEollwyouAlgtTnata2PpaRmlF14d1afAOPvicBQqzwzQh7YNzOKm\nrWHwGblaGz2PERYk98ehQiCAsAj3V8Htv0ri9l8l+agxMEbqfWgvySNu2hKEEhmjL7+X+v0bMLc3\nIRRLUCako0kbj7GlFkV0cO+HL+wWEwKBkLHXPkzz0Z3o68tw2KykXXgT0jBXvNHmIzswd7RQdt8z\nbnXFURpSn7wH/cEiWldtwVRW1+M3KRCQ+IsbkaUn0rpqy7BcS4jhRZGWgbG6wsNXVp6chqnWu6gs\nT05DNXaCm+txzOJlIBCgPTy4AYah7CSmmkrU4yaiyZnhdjyhTIYkMhpzfW1A5X3VCQYP/SKMGbP8\nx/0yGpysfEnHO28aaG8L+SIPNUqVIOiJXjTh/sc3ekNoNnyuIFZp0JUU9FluuBcPQ4T4vmC3hr6T\nIUYe8dMSmP7T2ez49Wbay9pIvzgTY1NPbopx100g597p7H9mDy0FTcRMjGXWI3MRioUUfdLzTZn1\nyFwKPjjKhh9+ziUvXsbku6Zy8N/7AchYNobxN09iwz3r6KzqIGtFNjMeOo+1163C3NF3/PLWwhaa\njzWSdXU2+5/e3b0987KxtJe2uYuLp1R8iUqCpdPC9l9/i93Sf4O0BU8toWxjCSVfFuO0DX4sFPqy\nDgOKsDjEUuWgXJwHi7GzCU10OiKxFLvNu3WaXH1msqptWdPqITBOWxCGJlKMts23dchCL+7ROz5v\nwz7AGIpGw+CzAJv0nscYqZZ7XSjDRB7JboYCiWxkWu80fLeJsNRsFDFJqOJHkXnFfW776/dvoLO6\niKxrfjos7bFbjFRv/Zj0pXcSP/Nij/3tJUeo3eM9HmvMLRfjsNqo/fv7ngHZnE7qn/uYtH8+SMxt\nS6l/YfVQND/EGSTpprsBMFVXYOtoB4EAWVIq8qQUip/8wmsdY2UZyTfdja64AFt7K/KkVOSp6Vha\nmmjd+e2g21T38Tuk3PljEq65hYjZF2BpakAcHokiNZ3O43nUf/qB3/LmhjoEQqHfOoMhJVXE3fep\n/JaprrJz7+2tlJwMZW4cLjQaIQYv39PBEB7h/xvU2hyaEJ8r2A06JBGeCQBPxxJK+BIiRNAQj9Bx\nfogQXYiVEnCCscWArqaTY2/mue2ffPdUjr97hLINLo8fbUUH6iQNk++e5iYwtp5o5vArLsvAsq9L\niJ3Sk9chdnIcLfnN3VaRZRtPMvuxuYRnRNKY5z3h8OkUri5g9mNzOfjvXGxGKwKRkPSlo8l/76hb\nOZvJhkQl4cJ/LePr+7/EYRvYOKb+QB2FHwcvxvnIVj7OFQQCEjLnUl2w6Yw1QdtchiYmg8jEiTRX\nHfJaJiJh3DC3ykXxYQPVJSZSMuXd20RiARdcEclXb3s3I46IEZNzgWcym4G6RwOYjYOfXJj0nsdQ\nnRLwhjBZ+aBQaUamZeFwYbeYKF7zArFTFxKRmYMsPBYEAuxGHfr6cuoPfI3gVIZpgXDo75VYpkJb\neYITHz1N3LQlaFKzkag06OrKaC3YR1vxQZ91ldOy0O0+5jPbg9Nux1RYiWr2+KFqfojTeOwnEYSH\nifo0Dn/8qcFnDGzd8S3qCVOQp6YjGiPHYTFj7WijefN6nFbvC0u6gqO079lO5PwLUWWOw2Exo83L\npembdThMg4/La21rofKlfxI5bzHq8ZNRT5iCrVOL9vABOg7u67O8JmcmTrvNb53BcNtdKp9JPwBa\nWxzceFUzTU2Df4H7O08Id2LjhNTXBVdgTEry//5uDgmM5wy60gIippyHOiMbXdkJr2XCJ86gs/io\n130hzm5u/fd0JiyJ91vm8YnrAzpW8sRwHlw1t/vvjc8Wsu21vhNJdnHxT8ey+L6eWJ/PX7mDxhLf\noUcEQgHjF8eRvTCOtGkRhEXLkCpFGLVW2mqMlOa2kvdFLQ3FwY3d3F+WPDCGyGQFmjg5mjgZmjg5\nivCeMCPJE8P5y/HlPut/+vtj5K7uO6GnKkpKzqWJZM2LJS5TjSpKikAAumYL9cWdFG5rJO/LWiz9\nMBL5+boFxGS4Fhb/unAznc0uqzKhSEDWBTFMvyqF+Cw1EQkKOpvNtFYbOLm7mfzNjbRU6P0dupve\nz3HiRfEDeoa9+15/+x149r2++nwgfW/js4V+j3E63u716fc5PE6OQCgY8L3uD7V7qin5oohlK6+g\nOb+ZTQ9+1W3xJ4+QI49S0JLv7n7ckt/ElHumoYhRYmx2GYu1FPToExatBYmqxwumo6ydtAszUCWo\n0dfrSF0wCrvZjrY88OSpFd+WMuOh2aRfMpqTawtJnpuCLFxO2Qb3xJIOm4NFT1+MKkE9YHERoPlY\n44DreiMkMA4xVrMOiUxN6viLMXTU0Vp73GdZVUQyZkPbkFg6NlUcICV7CWkTl9HeUOhxDqkinKSs\nBUE/b6BsXdPKTSNPzwAAIABJREFUrY+6u6AuXOFbYLzgikhEIvfJWulxIxWFA58QBxKnss9jeFm8\nczqdZzoMqF+kZ9mKozw2mdHXP4ixvoKyNS8H5ZgOm4WGA9/QcOAbr/uddhuHX3rU677yr9+Gr9/2\n2G5qayTvf55B98vWv+63LV0uWxZtK9Xb+mdlKJRJEKrkfssIpGKEspGZcOdcQiIRsPr1BC69KLCM\nxMEQGNt2b6Vt99Z+1RGIxXTmH6Yzv+/smvqifIp+3/9EEnajgeZNX9K86cshKT8YLrvS/+/lD090\nBEVcBIiMOrvetWeS5BQRRw8PPmxJb1JH+RcYK8pDFqrnCk27viYsaxJpN9yLtugohirXxFwaFUvU\nzAWoRmURNmYihf/6zRluaYiRTs3xDupOaEnMdsXrnH51SsBCj0AA067smdtU5rX7FRfTZ0Ry5W8m\nkjDW04BCHS1DHS0jdUoEC+4ezZH1daz90zFMnWfmvbXgh6ORKoZu0V0gFLDoR6NZcPdoZCpPuSIy\nRUFkioLxi+O46Kdj2fhcId+tqe73eTTxcjqbzUSlKLn5uWkkTXCPyxqdpiQ6TUnW3BikSjHf/re4\nz2P6eo79fYa9+15/+h1473sDafPp7Y5IVAy433Xd6x9/ONfjPsPA7nV/cTqc7P/Hbo69fZjMy8Zy\n+fvXsPFH6zC1GhHJT/Wz0+WAU3/bzT3XbDX4vv6iT08QOyWOFZ9cj0VnxdCgY9svN2FqD1yjcFgd\nnFxbxNirszm5tpDRl2ZRt6+mW+DsIjw9gsLcfDSjAksgKFF6l/56X1swCAmMASIQCBFJeiYiApEE\noUjiNdtzb4r2vsuE+fciEIrInnsXHY3FtNWfwGLUIhAKkcjUKDXxaGIzkauiyfv6mSERGI2dLmVa\nro4m58KHqSnahkFbj1AoQh2VRmLmPECA1axHIvPvLjYUbPu0jZt/noiwl2g4ZoqSpNEyaks94xUs\nXOHp+rJ1zeBcXeSqwU/+5ErPj62+0z6iAx1bLZ6T57UrGyk9Ftzs67VlwcxS7sRu6TuOxVnJIHRu\nc3k96tnjkY9JwXTSc6AlH5OCevZ4zOWBmeiHGDj33qbh0ouUHMgzU1xq5aZr1BSXWjl2wkJCrIhZ\n02QIhQIef6qFTduC+1vrF8Mcd3ckMXqMmLh43xOk2ho7G78K3nsrJDAGTuaY4A9Px2b7X1gpKwkJ\njOcK1s52yt//L8lX3IJm7BQ0Y6cAEDZmImFjJgJgrKvEbjpzoYtCDB37V1VRc7wDVYQUZaQUZYSE\n2Aw1kckDS7p44JNqrnhiAgCxGSpScyKoOty3NdLo86KJSOw554E1vi32pl6exLVPTkYkcf9OOJ0u\nK6Xe2wUCyLk0keQJGl69Y1+3Bd5wUpuvRSL3/KYlT3QJHVaT3a+Yqm/1nUxSIhdx0z+nkr0ozmOf\nw+5EIHAJkF2oo6Rc++fJJI7T8OXf8vs159LEyTBqldz/3vmoo2V+yxZu69vKy9tzHMwz7Op7/el3\nMHR9bzD9ruteexMXTyeQez0YDA16jr5+iKwV4xi9fAz57x1F36DD1GYkZkIsdftqusvGTIjF0KDH\n0hl4AtToCbFs+skGGg7WDbiNRWsKmHj7FOJy4km+IJXdf/RMfmhsMXLgub2UbTjJ2GuyKVrTY7Fv\n1VuRhrnHF4/I7Dt0SDAICYx+mHrJI4glCkQSuSsBSi/iRs0kbtRMnE4HdqsJm9WEtukkJw+scivX\n0XSS49tfZux5tyJVaAiPyyI8LsvnOZ1D6EfbWnucqKSJyFRRjJ52tds+m8VIwa7XSBl/IZEJw+8+\n2dpo5fCuTqYtcH/pLLo6ivf/6fnjPD3rsd3mZMfnbYNqg1wZBIHRi0ip7wium1ew0Ws921dZZGLn\nusHdz6HC1FRD/v+eONPNGJG0fLyV5MdvJfWpe9AdKMRUWImt3TXAE2tURN96CQKhgNbVW89sQ78H\n3HCVGoA5y11C77VXqNi42cDPfutyvRgzWsKX7yWyYrmK/6zsOGPt/D4zJsv/EOjr9aagLg6lZ4SG\nXIEyfmLwraynTvN9TLsdigpDAuO5hLmlgdI3n0WZmokqLRNJWAQIBFg7OzBUnkRfefJMNzHEEFG0\no4miHe4eUPNuT+eyXw5sfpO3rpblj2R3xxicsSIlIKFnxoqU7n9bjHaOrPcuNmTMjOK6v0zpNrLo\nqDex/bVSCrc30VZrxOlwIlOLSZsSwbzb0xk7PxaAmHQVt74wnRdv3jOg6xoMr9y+1+v2LrfoxhId\n/71ht9cyfbHidxPdxMXWKgNbXy2haGcznY0mEAgIT5AzbkEsi+7NJDzBZQA099ZRGNosbH4p8N92\nTLqKpT8fhzpahtVk5+BnNRTvaqalyoBEJiQxW8PY+bEkT9RQc9z/WM3Xc9z7YWWfz/CV2/di95Jc\no3ffC7TfwdD2vb7a7Iuue336fbYYbKgipf261wMlbXE6ErWU1sIWrHoLUo2Mziqta6cTjr1xmJz7\nZqCr07mSvEyIJfumSRx4znt/94Y6QU1YssbD2rC/GBr1VG+vZObPzsduslO93XfC4JaCZpa+cgWd\nVVrqcl2JEJuONJB5WRY1u6rQN+hJXZBG7JQ4dHW+hf9gERrt+kGpSeizjEAgRCxVIpYqsRiivZbR\nNpdycMNfiBs1i8ikCajCk7utBK1mHcbORjqaSmmpPtxtaTgUnNj9JrFp04hLn40qIgmhWIrF2EFb\nXQE1hVuwGDswahvPiMAIsOWTVg+Bcf4VkV4FxtM5sFnrNyFMIETFSxAIGNSEMibRcwLT2T6yBUZd\nhx2z0YFM0SOOaiJDr4azEUNeMXXPfEjsPZejnj0e9WmxFu0dOhpXrkN/sOgMtfD7Q3aWBLOl52Vi\nMjlRKnpW3E+WWnngsSY2rkri4fsj+PsLI1PQP5eJjfO/qFRRFlzBac48/5mqQ/Qwc7Z00N/j3oSH\nC/2KloUF1qBnrQ4xMjBUlWCoKum7YIgQPjBqrRz/toGcSxMBmLI8gXV/y8dm9m0UIlWKmHhxTxzI\no+vrvMYJlMhF3PB0TrfAU3WknTd+lOvhgmrW2Sje3Uzx7maWP5LN/LsyAEjNiRj09Y00pl2V3P3v\nkn0tvP3j77Caet07p5P2WiP7Pqwkb10td708i7Sprvtw4YNjKN7dTNWRwIS4i38yFrFMSPXRDt59\n6CDaBnevhepjHeSurkIkFvj9HgXyHP09w+krUrzGpOzd9wLpdzC0fa9LQPbXZl903eu/LdnicZ/b\naowB3+vB4LA6yL5hIurkMIRiIYf+m0tVL+HuxKrjOGwOptw7DVW8K4Zi3v8OUPJF4PMmY6uB9rI2\nrlx1HeByy+6s1pL3vwNUbi3vV3sLV+dz0X+WU/zpiT6zQx9+9SDz/3IhG+75Am1FO0dfP4QiWsEF\nf16MSCqkensl+/+xmwm3TulXGwbCiFcRrGY9u1c/0u96Zn3rgOr1ZrD1e+Ow26gv3UN9af9XmQJt\nR9/lnDRVHqSp0neSiPIjX1B+xHu20aFm/zcd6DrsqMN7XNbiUqSMm66i8KD/QK9bPhl8JkCZQkhs\nspTG6sBNoE8nZYxnPK/ygjPo/hggNaVmN6vQ5Ez/bgIhRi66/QXovytCMXk08tFJiMJdixl2rZ62\ntTtx2ka24H2uEB4mpKa+51536h1ERLi7427eYaShyc6NV6lDAuMZQC737x4ezKQfo8eISewjyUiI\nHqKihUzOkXAkLzhxGC+8RIbQj568b8/Av/shQoQ49znwSVW3wCgPkzDxwngOf+XbAGLyskQk8p53\nfu4n3uMDTr8qmfB419zBbnXw4aN5fca3+/pfReRcmogm3n8M4bMdo9bKh4/kuYuLp2HW2fjg54f4\n2boFSJUiBEIBSx4Yw1sPHAjoHGKZkLYaI2/8KBej1vf3pi9Lvf4+x9Of4QV3ZvgU67r6XiD9Doa2\n72kbTAG12Rtd9/p0cfF0+mMV2V+qd1ZSvbPSb5miNQUUrSnwuX/dzWvc/j6y8iBHVrq0FaFEyLKV\nV1Kzs5Jtj23CbrEjkogYdVEG8/64iJqL3sFuDXweZjW4+mTpV57xKAs/znfL/Hz87cMcf7snnrpV\nb2XXH7Z51Cv9qsfC19Jp4d3zXwu4PYESdIFRIBIScel5hM2bjDQlBoFUgkNvwtrYRvvGXLSbvWcw\nHm5ibr6QqGs9k5o0rvyK9vXBzVIZIjCsFie7vmxj6c0xbtvnXxnpJjCeLuJ1tNg4uFXb7/PZbU5E\nYvcJZnq2YsACo1AkINWLwFiUF/wsWMGm5KjBTWDMnj40cTgFQhHZP/oDIrmCE6/+CZve87kJBEIm\nPfwPTqx8EmunS3QRK1SMv//PbuUMtWWUfPRvv+eLn3sp4VlTkIZH4bBZMdRX0Jy7BV1V14tawMQH\nn8Lc1sjJ95/vrpcw7zJiZ1+IvqaM0lWucyQuuJKYGYsoeuOvmNtdrjfSiGhipi9CnTYWiTocp8OB\nTa/FUFdO+4mD6CrPjKWg027HkFeMIS/4AZJDBIbR5CSmV8y9lhYHmemen9yKKisTs4ffsk17KBft\nodxhP+9IQqfzP4gNZtbnH943/LGNz3auukYRNIHxhpv8J1vatDGYMYJDhAhxrlG6r4W2aiORKa6x\n8vSrU/wKPb1dVJvK9FTmeV9EnHVdave/C7c30Vbdt1GC3eqgcEdTd92IRAXtdSPfmCEQeicZOfBJ\ntd9YjV10NJg49EUN592YBsC4BbH9Ouem/xb7FRcDob/P8fRnGJuh8vkcS/f1JAHsq9/B0Pa9QNvs\ni01DkLhlJBE5JoqIzEg23PM5NmOPWFuzu4qpD8xEGi7rl+t0xiWZdJS103R0aGNSBpugRxxPfPQH\nxN65DHlWMkKFDIFIiEijRD4mOShZeoOFqbQOXW4hxsIqLHWDz94ZIjh4s0ScsyzCLTvz+UvdMyVt\n/7wNu73/qx2lxz1fiDOW9B141hcTZqlQhnlaqBQeGlgMhuFMDJO3s9Pt7+RMOXEpwRc9nA47HScO\nIhAIicie5rWMOm0susqT3eIigN1iomr9e9Rt/YzWI4FZAYtVGuLOuwiBUIi2NB9Tcy3q1Cwyrruf\nqClzulqEsbEaWXSCW7ILZVI6OJ0o4ns+0vLYJOxmE+Z2Vww9WVQ8Wbc8QvSUudhNerSlxzDUlyOU\nSIicOJuw9Oz+3ZwQ5xQl5VZUveK6FhRbmDxeRuJpSUXiY8XYQ0alZ4S2Vv8Win25UAdK2igR11wX\nWDbxED1ce4OSiMjBP4Pps6RMn+X7e9bU6ODggZAF47lE3IJLUaaM7rNcV/KXECH6wumE7z7tsQQb\nc773sFgAUSlKRk2P7P77wCe+rbwSs3sEtQofQpA3eotB8VnqgOuNdDJm9SShKNjSEHC9/G8DL9sb\nh93JsY2DS3woU4sH9BxPF/R8Pcfe88Ex50f7tVwNtO8NdZu9EYx7PdLR1XZiN9vJvmEiimgF8kgF\n8dMSmP3oXFpPNAckLooVYtTJYaQuGkXW1dkcfSNvGFoeXIJuwaieNQ6Ati/20PrpDuxaA0KFDGlK\nDJaa5mCfbsDo9uaj29tjVjr249/j139mmMm46+c0bfsKXemJvgt7Ie2mB2j7biedRUcDKi+SKxj7\ns6co+OvPB3S+YFF82EBNiYnkzJ6XZ0SMmOwZKgpyXZaA5y9zjzcyUPfonevayMpxn/TNviicV35X\nPSDz7DnLPeOgVJeYqCkZmGWE2eg5+ZV6ydgWDPK2a93iMAoEcMlNMbz7j9qgn6stP5foafOJyJ5J\n83eeptsR42fQlu9uWeW022k/8R0A8tjkXgKhb1IuvpHG/Zto2L2+++usiE9h9A0/IWnxNegqirB0\ntGBsqEaVMgZpeDSW9mYEQiGKhFS0pcfRZE7qPp48NglTYzXgOlbU5DkIpTJqN39Cy+FdbueWxyZj\nN7qLtn1Rtv71fpUfMAIBAqEAp33oEkqFgB17TEyd1BNq4NsdRm5coead/8Vz108aqW+yc/+dGkal\nijl09BzNiD7CaWr0/xuYdZ6UN14dnAW6XC7guf9GIhrxAWlGHkqVgEd/HcYTjw082LtIDL/5o/+F\nww/fM4RE/nOMmDkX4bCYMVSX+i0Xt/AytEVHhqlVIc52vvusmgsfHINAKHDLYnw601ckd69Z221O\nDq2t8Vm293GW/yKb5b/o/+K0KvLcie/b24KxvijwcXR9Yf/G3F00luj8umAHQtJ4zZA/R6fD2d3v\npl+ZzNZXvceVDbTvDUebTycY93qkY+4ws/XRb8j50XQm3pGDSCrC0GSgdk81R171HaKuN1Fjo1ny\n/FJsZjvH3j5M+ddnXwzhIVErbM0dNL21EXuHHpxOHAYTpqJqHPqQC0qIvtnyqecqypylLvEuLkVK\nxvgeV96yfCMVJwbmFvDtx62YDO4TTE2UmGW3xPio4ZvoBAlLrvVM/b7+7YGL6t6yT4slAjRRwZ+p\nmgwOtp+WhXvZLdHEJgd/0GJsrMbUXIciLhlZdLzbPqFYimbMJLTFgxvwy6LjCcsYT+PuDW5Lf8aG\natqO70cgFBE5cXb3NgB5zKnYOrHJCMVSWo/2ZAwTqzSIFWoMDT0rgYJTX3CH3TNeiampBquu/277\nw4FqWhZjPvzDmW7GOc8Hn7oPdld9pqOhyc7CuQpKvxuFrnw0z/7J9a557b2R2VfOdY4ftWI0+l5M\nmr9INigrRpEI/v5cBJNzgp8R+fvC9TcpufOegbuXP/a4hkmTfd9/k8nJh++O/DAmIYYGiebcS5AR\nYujoqDdRvNv/uF4ggGlX9CQpObG1EV0Abr6DoXe8vbOdLsHKbnX0GQ+wNwO9x4b2wT+bYAm8/p5j\n7343fUWy1zL96XvD0ebTCca9Phuo21/Dhnu+4KMlb/P+BW8wufMgzkPHMbUFplc0Hm7gw8Vvs3rZ\nexx9LfDQgqtOTh9ok4POkAiMptK64fXvPBf5Ht+/bWtacZzm8jx9kcsCYdZF7u7RW9YMPLmLUWdn\n82pP9/jrfxJPVHz/JoR3PpHsYV1o1NnZ+unA29dc5z0eyKTzhsYVYs2L7u4FCrWIh54ZhUQW/NdE\nl4ViZPZMt+2aMZMRSmQ4bIP7CKlTxwLgdHpaKJmaXbFLFHEu92fjKdGwS2BUJWXgdDrQVxVj1bpE\nV0VskqtsY4/A2F54EJxOkpdcR9KS65DHJA2qzcOFUBlK4DMc5B4yM++ynpVjvcHB9T+sp7HZtXDQ\nZTD//hodr703sJX3EIPDYnFyYL/vd41MJuCJP4T3jp4QMLFxQt7+KJrll5/bQfiHg1//TsOd96j6\n/Rx+9lgYd93rX5x8+3V9n5asIc5dnI7Qsw/RPw70SpiRNjXSY3/GzKjuOI0AB9YEngTDZnZgMdr7\n/Z/ddu70Y6nSZUTRX0s3p8PZZ3Zlb9itg59vy8PcDT+G4jn27ncx6apB973haPPpBONef5/ImKDk\nvGVn5yLYkDjtOEzfD4U6mCiS00lcfj1CqQx9eTHOXgJj5n2/RiCR4LTZad65kY5jrsxY0ectRiiT\nIY2IQZmWCQIBpSv/gd3ovhqvTMskcdn1lLzyN79tcDrsRJ+3mIic8xDKFbQf3kfTtq8AkCekkHDx\nNQhlMhq3fonuZH6fbVAkpZFwybWIVGoEQhHaE4dp+OZTpJExJCy9FmlkLE6Hndbc7bQd7HExbW20\ncmS3jqnze8zkE9NlxKdKmdUrRqLd5mTH2sFlXv3ohXrOXxZBVFyPoBgWIea3b2Ty25uL0bX3/YG7\n7bEk5npxj373H3UeFpL9obPd5uEuDnDlPXHs/brDQ4QdLN6S24yfpeL3b2Xy9ANlaNsCX0kE10ra\npDlhTDpPzQfPuQckbj/xHQnzLyciezr1u76iy+04YvyMAbe/N5Iw1/OY/LNnfZYRyV0fYXN7M3az\nqVtgVCalY25tdCWFaXBlGpN3CYy9LBgNdRWUf7aSxMUriM6ZS3TOXIyNNbQe3kVbQS7OEepzJ1SG\nBI/hYv9Bd6v9PbkmsmZXMP98BWqVkKMFZopKgpPEIsTAWLfWyPyFvkX3S6+QYzSG86ffaP1aO3ah\nVAm48WYl9z2oJip65IRdOZsRCODx32u47EoFT/2hg7yDff9mxo4T88BP/C/G1dfZefk/umA1M8QZ\nRiA+bUojFHpu6yorEqNKGY259ewKnB/izFOwpRF9mwVVpJQZVyd7JNCYfnVP7G5to5ninYF7Mq1+\n4ghH1vtP4HGuY9a75hr9tcoUigSIh8AgIhAsBvfx/lA8x979Dhh03xuONocYHDMvCh+UjnAmGbTA\nmPiL65EmRiOO0iDSuOLZaRZMQbPAPXCy8UQlVU94T4MtkIiJuGQm6rkTkabE4jRZMORX0L5uD6YS\n/zHgTq8rlEuxt+uoffrDPusOBkl8JBGXnY9yUgaS+EgEEhH2DgO2tk5MJ2vQ7z+B4WhZQCKDNDKa\ntJvup/CZX3Vvy7zv193/tuk6XCNsJyRdcTPa/IPdq67SqDhqPnvb63GVqaNJvPQGSl7+a5/iIrgy\n/Lbl7aFl3xbAFQfSUFGM0+Ekcdl13cdIuvIWpBHRtB7Y4bcNQrkSkUKFvqyIjqO5GKpKu6/Ndsp9\nVCASETPXFSunSzgF+Nv9Zby2ZyIqTc8H5g/vjHFLPPLM/5X3W/Q6HV27nXvnHueWRxK55v4ed920\nsXLeOjAZgJZ6Kwe+7aChykJHiw2FSkhknIRJc9SMneppVdFSb+VX1xTR2jh44eCPd5Tw4tYJbtmu\ns3KUfFyYg7bNxt4NHZSfMGLotKMKE6GOEBERIyEyTkJskoTPX2ti57rARdgfnn+MJz/MIjG9Z8I9\nfqaKN3J7YhHWV5ipKTXT2WbDYnYiVwqRq4REx0tIGi1HoXL/wNeWmj0ERpu+k6qv3iHtsjtIvug6\najZ9THTOBYSlZ9O4d2N/b5MHNr0rZtfR5wKJKeqk7dgeonIuQCRXEj52KtXffARAy6EdqNOyCB87\nDV1lMZZ2d4vXzvICOt8o6P5boo4gcdFVJF98A7rKYso+eXHQ1+KPMe//DqfNTsntTwGQ9fGfhvR8\nIQaPwehk45aBJX4KEXw+/dhIp9bJ/1Z6WgN0ce0NSq69oSder90GHR0OjEYnScmigCzrPv3YyC9/\n3s4996t47ImBJxP7vnDr9S289UG0W+zKnGkSVq3tfwgTb/zzb528/N8zIy5GZkWx/J1r+GjBG9gt\nI3Mhqi9G4jXIouJQpY9DneGKBR89Yz4RE70vWjpsVixtLZS99dxwNjHEOYDd6mDlnft4aO18Zl2X\nSsnelm5hZvKyRKZf5XJRdTrhldv39mkM4HT25BicsCT+ey/yFO1oYuJF8YgkQkbPiqI0NzBPsDFz\ng/NtGAgl+1qG/Dn27nfAoPvecLT5+8TP/p1B/l4dG99rQiIVEBUvpaGqJ7663eZkxpJwrvtJIs/8\nuJSWOgsCIfx97Xhe+nUFDVVmnv58PM//tIziwy5DsbgUKZVFnuEF73sqDYVaxIu/qsBsdBCTJKW5\n1mUkdO2DCXzy33oi4yQ8//UEfn1NIbWlwx+icNACoywtHqFShtNux9bWiThKg9Nsxa539zO3d3iP\ncSOJjyT5iVuRJvd6MagVLpFy/mSa3v6ats93B14XEMeEk/b3H/mtOxhU07NIevRGBFJ3N1pxVBji\nqDDkmUmEXzSDsvuexdY2QNe3UxaMIoWKug2rsbQ0IlaqyXrIXUBwWHwnBxCHhWOoLiN24aXUb1gd\n0Gkd5tM7oQDwr577aoO+9AQlr/yNsKyJxC66DJu2nZq17+B02Cn+75/Bi/tqF1azg11ftnPJTT1Z\n2nqLi9pWG99tDV7csveeqUMiFXLF3bEe+6ITJCwNMC5jc62FP95REhRxEVxi5QfP13PrI4ke+zSR\nYrf74w25sn+ree3NNn5z00meWDma0RMVXsskjJKRMGrwbrbakuPYTUbCs3Ko3fwJ4WOnAE7a8g/0\nWbcvdBXFgEs0dzr6nvwYG6oRiiWEZUwAwFBX3r09euo8ZJExtOTt8n2AU1h17VR99Q5hPx6PKjlj\nwO0PlOa3PcVY/YFCjAXlPusoJqSjmjFuCFsVIsTZxaaNJv7zvI7/eziw8BMiMf2yTvxklYHfnEpU\nsndXyMMjEPbvtfCLn7bxzxeCnyDnzZX6MyYuAiTM9h4/62xiJF6DqbEWU2MtLfu3MO7hp2jJ3Ubz\n3m/PdLNCnIM0nNRRdaSd1CkRTL0iqVuYmbK8Z6xedqCV1qq+FxPrC7UkZrsWncYtiEWmFmPWDc54\n4mymZF/PQv74JfEBC4wTlsT3XWiI0LdahuU59u53wKD63nC1+fvC5lUt3PvnNDImKdm2poWCXPcx\nxqQ5YSy8OoqHLs5H1+66x3HJMkaNU/DYy5nd5RIzZN0Coy+mLw7nLz882Z0MtktcBFj/dhMAbY1W\n6ivMRCdIzk6Bsfyh/7j9PfaTP9K5r4D6f33SZ12hQkbyb29DmhiNrV1H89tfYzhahihcRfT1C1Gf\nN57YO5ZibWxDt7cgoLp2nRFpcgyjnrnfZ91BIRAQf/+VCKQSDEfLaH73GyxVTThtdsQxGmRpcajP\nG4/T7ghYXLR2tGE3GtCMn4q2IA95QgrSKJegJZTJuq39IqbP61dTtfmH0JUWkPaD+4mZexHNuzf1\n71pPYaqrQiiVoR6djaWtGVX6WJp3fdNnPXliKuaGWrQnDmNubiD9jocAsLQ2EX3+Ylr2uAZ+8rgk\nzC2NOE9LlrHlk1afAtr2tW0DyvTsjzf/UkP1SRN3/za5O5tyfzi8s5MXHqmgvTm4L+dPX2ogNknC\n0puHZ3WuvcnK4zcUc9ujiSy/LQahaAAByHph9hETxWm30VF0iKgpc9GMmYIyeTT6mlIsHZ5xMfuL\nqaUObckxEhddTf32tThsPYKvQCQiLH08+uoS7GbXQkhXopfwsTnYzSbMLS63KYfNgio1C6FE5uYe\nDaAZMwWcchMCAAAgAElEQVRDXTk2vbvQrYhPRSiRYukYePzNQGnfsM9jmy63AO1m35nKHCZLSGAc\nBpQKAWFqIQ1N7gK3RCLghivVZIwSk3fMwpff6L/PIXdHDC/8s5OICAG33jnwhCKn43DAi//W8cI/\nO7ufcf5xKx0dDsLDQ+7TffHVFyYMhlaeeSECjWbw98vhgOef6eSlf59Zt+jE81P6LjTCGenXYGrw\nnbU3RIhgcOCTalKnRJA1NwZ5mASHzcHYC2J67Q8s9mLu6mqu/I1rcVumFnPRg1l8+fcgzlvPMnoL\nYzOuTmb7a6V0Nvs2pgEIj5cz9fKeOOil+wc/j+gvw/Ucu/odEOp7I4jDO7T87JLjzLwwgpt+kURL\nvZV/PVzWvT8qXsKJ7/Tc9PMkXv2dK/wWArDZnPx4wbF+hT0TCumKLOaBobNnzuF0us5xJhiSGIyB\nEnnlXKSJ0Thtdqp//yaWapfqamvVUvv0h6Q+9UMU2WnE3nYJun0n3BKf+KoLYC6rw3ii0mfdwSBS\nKxBHu9T+1tXbMJ3sGcRYG9qwNrShyy3s1zGdDgc1a98hcfn1xC26HH15EbqTrh+4tb2VMf/3O5cL\n8ZFcLG39e2k6bTaqV7/GqFv/b8ACo8NqoerjlSRccg1CuYL6UxaVfRGRcx5hYyfjdNhxWMzUrV8F\nQNXq14m/aAVjHvwdApEIS0sDVatW4jzN0KwoT+81BiEMLrmLPzatauHI7k5+8HACF1wRiSgAca36\npIlV/65n15ftQ9ImgFd+V03hQQM3/yKRmMShz0hqNTt4/ckavl3dyjX3xzFnWYSbm3Zf2G1O8nP1\nbPmklT0bfN+XtuO5RE2ZS/y85QgEQtqOe7de1IyeiFgZhlAmRxbpsjKVaCKJO+9i7GYTDoupOzt1\nF9Vff8iEB54kPGsyxqYaHGYzEnU4sugERDI5ha892S0wdsVhDBuVja66mN5vbnVaFuCe4AUg/vxL\nkMckYmqpx6ptw241Iw2LRJk4CnDSsPurgO9XMLF3+l+1dBiGfyXr+8gVS1W88UIcyrTS7m0qpZBN\nnyQxc2qPBfCqtTpufaAhJDKOAP70Wy0H9lt56ulwVOrBjcqqKu385pcd7NnpPjFyOCB3r4WLloZi\noQbC1m/NrFjWzFNPhzPngoFbzleU2/jNYx3s2+PbgjT5gjQWPnMJWx7eQNPhenLun0nWNRMwthgo\n33CSwy95fp+UcSom3zOdxDkpyCMUGFsMVG+v4NhrBzF39Dz7URdnMubqbCKzopGGubwxbtx+l9ux\njr+Z5zqHAK7/9g6a8urZ+nOXlfrS164kemIcuU/voniNa3woC5dx7cbbKHj3CIf+s7/fbRpI+VEX\nZzLrsXl9X8MZpi1vj0c88hAhgsmR9XVc/cdJiCRCshfGYrc6uuMGmjptHP+moY8juMhdXcX8uzKI\nTHZ5Dc27PR1Du4Wtr5T0OS4QiQVkL4rj+KbAznW20OW6Kw+T8IN/TuWt+w9gMXr3RpKpxNz07FSk\nyp6QWlteLhmupnYz0OfY9QxrC7S01fSdZfjI+jou+9V4pArRGet7/W3z94HMyUrKC4zsWd9G9Ukj\nT652N+TYta6NvO1afvPmGK5+IIFPX6ynsdpMfbmJq+6N59OX6hk1XkFtiQmrxXXzDZ12N+/NLg5t\n03LVj+J58deVWM0OouIldA4yZFywOaMCY/iSaQB07jzqJhB20b5+P4rsNCQJUchHJ7rFVBxM3cFg\n1xmx64yI1ArCl87EWFyN0zx4d1hjdRmlrz7tdd/J//S4Rbfs3dzz71OxEr1R+UFP7De7yUjpyn/4\nPb/dZKTgr+7x6sre6EmQYWqoofydf3vU89eG+g2rvbpmW9tbqF7tPR7n6fx06YmAygWTxmoLLzxS\nydt/q2XO8ggmnR9G2lg5UfESZHIhVosDbZudunIzRYf0vP/s8MSs2PZZKzu/aGPmhRomzw1jbI6S\nqHgJKo0IoUhAY7UFXYeNmhIzZQVGyvONnDw6uFhvFSeMPPdwBa9GVDNzSTgTZqlIG6sgLkWKMkyE\nWCzAbHJg1NnpaLFRW25my+pWju/XdZtu+8NQX4G5tRFZVBwOm4WO4jyv5ZIuvA6J2j2DuEQdQfzc\n5d1/N3+3lbrtn3f/bTcZqN36KRHjpqNKzEAgEmHTd2KoLUN78ijWU3EaXTgxNVajSh2Dsa7C7TwC\noQi7yeBhkdiU+y0RE2Yhj0lCFhUHCLAb9WhLj9F8cDv66uEf4FT96mUstf4Dijv0JpfKEWJIuWKp\nConEXaR6/OFIZk6V0aF1sDvXxII5cm64Ss3nG/R89NmZs6pa+Mq1RE5wdy9q+q6aXQ997qPGuctX\nXxjZt8fMbXepuPk2JRGR/bOcKzph44N39ax634jVR8bEPbtCAmN/qK6yc8dNrSy+SMbdP1Jz3hzP\nAbcvykpsvPOmgQ/f1WMLcPytilcz5T+XIg2XU7a+mNiceEQyz+GyJj2Ci1+6HKlGRv3+GmpqK4nM\nimLcDRNJmpPC1/d83i3QmTtM1O2tpm5vNeNvnowsQs6Rl7/DYe95FzcfOTUpdEJ7cQvho3vigkaO\njcbpcBI9IbZbYIzIjAKgtahn4bmvNn1x/ccDvoau68h/53Df13CG0RYcOtNNCHGO05WMBFzupb05\n/GVtwFmQ7VYH7z18kPveOb9bJLr4p2PJuSyJ7z6tpjKvHV2LGYfDiUwpJjxBTuxoNaOmRTJ6dhQK\njYTHJ64P3oWNALatLGHRvS630YyZUTz02Xy2vlpC0Y4mtI2uRXJNnJxxC2JZ9KNMIpJ6QjrtfreC\nkr3Db8Ho7zl++vtjfT7DV27bG5BYZ9bbOLaxnukrXGEqhrvvzbw2td9t/j6w5PoYZl8Sjs0GJr2d\nV56o9ChjNTt4+r4S/vjBWFobrGxb08LT95Vyx29SeGXPZGpKTfztnhK6jFy+fr+Zh5/P4MUdk+hs\nt/PYFa5v/5tPVnPbr5P51zeuHA2dbTb+cEvxcF5un5wxgVEcrUEc4xIMjIXeTXmt9T0TemlafLdI\nOJi6g8bppPm9TcTfdwVhcyehnDwa7dY8tNuOYC4LBUc9l2hvtrH+nWbWvxN4Brihxm53su/rDvZ9\n3dF34SCia7ezdU0rW4fAcrTorb4TEJ149Y8DOnbLoR20HNoRUNnS1f/zut1Xopj2wkO0F46sSYyp\npG+3MP2hYopv/MPQN+Z7zpQJ7iJImFrIA3dpcDjgkutrOXjEzDWXqfhoZQK3XBd2RgXGyvWFdJxs\nQRouJzwrBlXS9zsJSUuzg+f/4XKlnTlbyozZUqbPkBCfICI8Qkh4hEt0NBqdFJ2wUllu5/AhC7n7\nLBQX9a1ivfOGnnfeGLh1Vd5BK2NTBzfe0Godgz4GwOwpwycobdlkZssmM6lpIhYsljFjlpTMMWIS\nEkWo1AJEQgEGo5OWZjtlJXb++69OjuT1fwF4yv0zqN5eQe7fd+F0uAb6YrnncHnuHxchC5ez5Wcb\nqNtb3VP/3hlM+uE0pj44m31/cX1/6vfXUL/f9X4efdlYZBFyCt474jNBSltRC7FTEhArJNiMVoQS\nETU7K4ka3zOZDD8lMLYV9kym+2rTYK6h93UEcg0hQngjKkXJhIvikYeJkaslp/4vRv7/7J13eFzV\nmfB/905v0qh3yZJlW5J7bxhjTMchkFCSsEtIW5L92E3ZtE0hdfdL+bLZkAK7KUACBEhCCaYaTHHD\nvclNkm1ZvY2kaZp+7/fHWCONZjQaWd3c3/P4sebec859z5x5Z+59z1ssanLKLZF2Ko3IP/1pDT5X\nEK8rGP7fGcDnDg7roaa3aFh+WwF6iyY8plmN7uJ1MkuMUW3v/Z8VQ8YO4nUF2P34hbhjJ6J0RTpq\n3YAH3YFnmxK0jqXlpIM/3X+Qj/5sKYbUcJRS9mwzN365YtSyXA688ctajFYtq+4oAiCt0MBt3wsX\nm+wPJY2Xvunoy6288tOpC/Edbh3ve3zNuF7nwN8aIwZG5bM3PfjtAw389oH4535wz4Dxz+0I8eWb\nBz6j7Y0+fnJf/O+z5rNevvKB2M9znzPE/3wj1oB5Z3l0eqyv3zr5Tlr9TJ2BMX3gRyTnvg+Qc98H\nErZXmQd2+8fSdzywv36AYLeDzI9dg64kh7QPrCPtA+vwN3Vi33YQ+xsHkbxKIneFicWYXUzRFR9G\nbTAjiCqq//Td8PGsQgrXfwit2UrQ56Fl70s4Gk6i0hqouP3f6KzeQUblWtQ6A7bTe2nZ93JkvML1\nt2HMLCTocdF+5E06q3dO4QwVFGYuuTlqHM4Bz57bbzFjMYs897KbQ8fCHkEvvt6HxyuzZMHYiyaN\nhfPPVUf+rvjkSio+uXIKpZk+eL0yO9/1sfPdxPmfFCaXxoYQTzzWxxOPTUw1dkEUOfyLvRHjIkDQ\nG204zl6aS/q8TBrfro8yzAFhD79/WMis62ez/6e7kAKj9xjvPmMDAVJLrdhOduJ3+ml69wKrvn4F\naoOaoCeItTyNoDeIs9GetEyiRozIM9FzmC6IWh2GvGI0FisyMkGnHU9bY5yihgqTQX5VCjd9ZWTj\nhSDArGVpcc8NZ2C0ZGq5+WuVSckx94rYoo7AqAyMXefdZJaaSMkZeM5sPeOk+cTonQDq9tj45e27\nuPlrlVRtzolU9x2JtppLLCY6jZFCMs9/t5rWUw6u+Zc5mNIGNmzjGRY9jgDbH6pj1x/rJ1HK+FzK\nOrbVOHF0JH+fUX+oZ8o/e6OVWeH9xRSGSA/69MpyEjkShfh/j7rv+OA+UIP7QA3GBaWkbFqCeU0V\n2sIssj5xA2m3XUHbg8/Sd3TyQyQV3j/kLN6E7cw+uk7sQrhYZlPU6Ci74dNceOvPOJvOoEvNZM4t\n/0Lt339F0ONCa0lHVGs59fSP0BhTqLzr69hqDuDr7aBw3a3UPPcL0uYsJ33OcsW4qKAwBiwmgaaW\nAaPER28LVyj+wxMDRYECAZlOW4i8bFVMf4XLD11uPiWf+zK1P/gqcrLxutNo/PcLtpMdBPoSez5m\nLwtX7Ow80hZzLugN4m5zkVJixVKUiv1cz6hl6LkY9pxalobtZCe9dTZ6ztgQRIG0uZl0Hm3DOjud\n3truiCE0GZkGyzPRc5hqBLWa7A03kbZ0HaIm2qNcDgbpPb6X1tdHLkipoDAcB55r4oYvRedaS7bA\nRjx6Wzw88flDZJeZmX9dDmUrM8goMWK0alFrRfx9IRydXjrPuWk42kPdri5az1x+BsZ+9j7dwJGX\nWliyJZ+KK7PJmWvGlKZFlsNVkNtrndTs7OTI1la8zrGnKxsvhq7jytuLxn0Np+qzd/C5psv+c6cw\ndqbMwBiyD4SDNf/oz7gPJF8YZSx9x5u+6vP0VZ9H+N+tpFy5iIzbN6LOTCX/K3dx4Yu/IdA5cYU/\nFN7f2E7vpWjD7RgzC+mu2Y+r9RzGrCJCfi/OprBO+OxduJprSCmqoLsmnHi9szoc7hToc+BzdKE1\npeLr7UBUh2/A5VAQQVQMHjORzI/fgHPHUXznlHQNU42tRyI9LaxHJYVqNq4z0NYR4o13o/PVSCE5\naU8FhZmNcfbEVm+f6PHfL/S1jRy+bsoJbxgs+8Ialn1h+PA3jSn5fJGDsZ/rQQpIkTyMPTU2es91\nIwUkMioz6TzWhnV2GudfHgi9SkamwfJM9BymEkGlpuSuz2IsLCPocuCqO0HA5UAQBNSmFIzFs0lb\nup72t15ECigRR5NJ9ettE5YzsPO8e1LzEb77+3O8+/tzIzccJR3nXHQ87OKth6fOUcUwbx45/3AP\nntoa2v/4WNL9xvv997mC7H2qgb1PxYaEXir/teXdcRsrERO5jjPlszdZ77XC9GLKDIyBjl6C3Q7U\n6SkY5hWNykg4lr4ThewLYN92EPfBWkp/9a+IBh2mlfPofXnvVIumcJniaDzNyad/RGrJfPJW3Uzt\nC78EWR7RXzfkHxQWJEO/h2/znhco3/I5RJWapl3PTZTYChNI2pZ1pG1Zh7+5C+eOozh3HiPQPvM8\nTy4HjlT7uH5TOOfTt76UjijCY085CAYHPO5F8WIotUspIf1+wFQ+sQbAiR7//cLgoiUj0byzAUfD\n8CFpnq5LC+OWghL28z2kFKeiteiwnexCCkj01NpIr8zElGNGbdCEQ6lHIVM8eSZqDlNJxqqrMBaW\n0bnrdbp2v448pLCZIIpkrruWjNWb6Nz52hRJqaAwfTHOmYuo12Osmj/VoigoKMwwprSKtP2NQ2Tc\neRXW61dif/NQVGGWKEQxpurpWPqOCUFIGJId7HYgeXyodBpEw9Tm1VK4vDFmFeGxtdB77ijennCi\n/b7ORkSNjpSiChyNp9GlZmIpnEvboW0jjqdPy6Fu68P0V69SmHm0/+Y5LOsXYlhQRsZHNpPxkc14\naxpx7jiGc3c1IcelF5ZQGB2P/8XJ9ZuM1O0voaRQjcst8cvfRT/EL6jUotcJVJ+auR40uetnMeuD\nVaRV5qCx6Ag4vHRXt3H2mWN0HYktrnbrzn/m1G/3ceaxA8OOWfVPq5l7z3J6Tnbwzj/9Neb8sm9u\nJmtZAbp0IyFfAEedjYZXztDwyumovHnTAcvCpaSuWIs+twBRH650OefbP4lqU/Od6EJSmrQMUpas\nwDSnEk1aBqJOT8jlwH22BttbrxJ0RH+OLAuXkrPl9mHH7373DbrefDlGNn1+EWlXXI2+oBi12YKv\nvQV3zSl69+8i5J66okMzAU9n2OjWtr+FM09Xj9D60ug+YyNzQTbW8nRsJzrCx053kb00j5RZVmAg\nlPpSZJqMOUwVqfOX42m5QOfOV+OelyWJzp2vUf6Zf1cMjDOANQ/dibUqN/LadrCR/V8a3UZ45qoS\nlnz3RmwHGjj8QOz3YSJyNpZTctsiLOVZqI0agn0BPK12dn/mqVGNM5Poq63BsnoN7urjo+qXeduH\n6HnzDUIOx8iNFRQULkum1MDY88IuzGuq0BVnU/zjf6LnhV30HT2L5AugTrOgLczEvLKCYK+Ltgef\nTapvsNeFaNBhXFg6bN+xYF5dSdot63DtOYnnZD3+FhuSx4eo16ItyCTtlnWorOGwk74jdTH9BQRm\nqxeSL5aiFrQ4pG5OBw/gkie3KvBEUaCazXz16qhjfbKLnf6/T5FEly8ZFauxli5ElkKE/D5OPfNj\npKCfs6/+jqL1H6J4452E/F4a3v0r3t4OVFrDCCMKLPiHB0AQ8TttnHvtEYKeycuxMVg3+mSXoheX\ngOOtwzjeOozKYsS8dgGW9QswVJagn1tE5r030v6rZ3HvP43km7kGrZnC08+7uPYqI/fcacHdJ/FP\nX+qkvTO62uqWa00A7NrniTfEtEYQBZZ982qKrg97zQX7Ang6XOjTDeRdWUbelWXUPnGYEw/tielb\nfNM8zvzxQNy9DEEUKLohPOaFrSdjzs+6pYriG+chhyS8XX1oU/VkLMknY0k+BZvLee9rLyMFpk9V\n21Cfm766M/TVnSFt7UZUJjNd219JuPGZtm4j1lVX4Gtrxl13GtnvxzBrNqnLVmOcVU79r38clWMx\n1Oeme+f2Ycf3NJyPuYaxdA4F99yH7Pfjrj1FqM+NuWoR6VdeQ+8+Jf/uSLQfamHBp5aSsyJv1Ma5\nkC+8dqJGTFiBuaemi1nXlZE2J52Ow+G0F7aTnZTfWoG1LA05JNF7dsBDfbQyTcYcpgqtNYOeo++N\n2E6Tmj4J0iiMlZbXTuM824U21YClPPOSxshcUYTapCV7fdmo+hXdsoD5/3Y1AFIghLfThUqnJmVO\n9iXJMVPwnDlD/be/Oep+KevW49izWzEwKii8j5lSA6Pk9dP8gz+S/+W70M8rIvPua+Dua2LaOXfF\n3viMpa9lbRWWDYsQjXpEow6VSR/2dAQyPrKJ1OtWIHl8BDt7afvNC8i+gcSxggCGeUUY5hUlnFvX\nk2/iPRvrvVGoKqdMtSDyOl3MYbFmA7v9LyErnmMKo6Bxx19p3BHr3ePpaqbmhV/GHA/5PRz+n2hP\nmTPP/hcA5rzZpM1ZRvXj3wOgaMPtpM9dQcfRtyZA8vgM1g29YFL0YgyEnH3YX9+H/fV9qNMsmNcv\nxLKmitzP347k8+Pef5q2X8R+dhTGD1mGT32+g2eed3HomI9OW+xDeEdXiP/7ix4efWrmJcuu+ORK\niq6fR9Dt59D/fYvWd88hSzKCSqRg02yWfO0q5ty9FE+7k3PPDvwOh3xBTAWpZC7Oj+vhmLWiEEO2\nmZA3QNMbtdHnlhew+MsbOfKTt2l4+TRS8GI13JVFLPvWZrJXFTH/n9dy/BfTx0DWd7aGvrM1AKQs\nWYnKZKZn11sJi7B0v7ON3r078Xd1RI4JKhWF9/4zhuJSzPPm4zxxNOYayY4PYF21HkEUaXzsN3hb\nwhWEO155Dm16JqE+xdN5JNoPtmI72UnhhhJm3VBO/avRG8oaowZzgYWe2tjoGlezg7S5GWQvzaN5\n5/B5xXrO2FDp1OStKeTMMycA6D7ViSAK5K0pxH6+N8qYnoxMg4vXTMYcpgrJ70NtMCXRTqkkPRNo\neP5Y5O/yT6wmbWH+qMfoOtBI4ZYFtO8YXW658o+HN4dbXj9N9U/fRLpoUFfPwNykCgoKiUmbm8G6\n723CnG+m40g7O76+jaBHKZo3WqbUwAgQ7HbS8M3fY1lTieXKxejL8xH1WoI9LgKtNlz7z+Dadyrp\nviqLETkQxHPywrB9daV5mFdXxh1TZTagMl/09JpXRPv/bo0yMLqPnKXjD69gWlKOtjALtdWEoFEj\n+4MEbHY8pxuwv34Ab21z3PGzxcKYYyYhBZOQiktWCsIoTA1qgwk5FP4CVWn1GNLzcLXEeuBOJEN1\nQ9GL8SHY46R36256t+4m9ZrlpH1wA5YrFikGxknitbeGz1/2u8dn5g6/NlVP+UeXAHDkp+/Q8vbA\nA5sckmh6oxZRI7Lsm5up/MxqGl49Q/CiYaPl7XMUXT+X4psr4xoYi2+qAKDpzbpIn36qPrsWQRSo\n/3u0Z2PH/kaqf7mLFd+9ltJbF3D69/sIuGaul27Q5QRXtNFZDoVwHNmPobgUTeY4eM5crCwUlfFF\nlvHbOsc+9vuEXd/ezuZf38y6717FvDuq6KnrRlSJGHPMZC3K5sIb53nvB+/E9Kt99jRFm0pZ/4Or\nadh+joA7gNaio+mdehrfro+067lYITp3VUHkmP18L0FPkOxledS/FmsoGUmmpzc+OqlzmCr6ms5h\nLp+PNi0Lf0/8z7Q2PQt3w+Te5yhMHV37LvDGTQ+Pqo8hNwVdZthQXffI3ohxESDonrm/MROFoFIK\nRCrMbCrvXkhqaTgFSd7qAgo3zorZfFMYmXE3MNZ8+Duj7yTLOPecxLknNhxqIvp2PfkmXU++Ofpr\nAZLHR+9L79H70sihF/HQCvphjuuU1HcKU4a9vpqUwgoq7/waAD11B+k5e2RSZYinG4pejA+iTotp\n+VwyP34jol6LPIoiBgoKQ8lZXYxKpybg9NG8Pf6NV9O2WhZ98Uo0Fh1ZK4pofTdc7fDCiycpun4u\nBZvKOPbzd6OMiBqzjrwryy62i94cNGSbSasc3rDWdTRsrBQ1ImlVOXTsaxzTHKcjQWfYIC2qNWMe\ny1l9GHPlQgrvuY+ePe9gP7SXkGvmedJOJa5mJ6/c8xyVdy+i6KoSSm+cgxSQ8HS5Of9KHWf/Hr8A\nYdv+ZnZ8/Q3mf3wJRZtKEVUCfe1umt6pj2oX9ARwNTmwFKdGjsmSTPeZLrKX5NJbG1vgZSSZJnsO\nU0Xn7m2YZ1dRes/n6d7/Lu4LNQRcYf3RWFIxlcwhfcVG6p/81RRLqjCd0aYO3Jd6Oqbx96MgMOv7\nP0TQaqn/5r9HebAX3P+v6EpK8Le00PTznw30EUVKf/ifyMEg9d/5NsgyOfd8HNPCRVFD+9vbaPp/\nPx320prMTKxXb0abX4A2N5wns/DfvhLTrvet7XS//NLA5XU6Zv3wPyPj68tmY920CV1RMXIwiLeu\njp43txHoHH7TS5uXR+qVGzGUl6OypCD7/fhbW3Ee2I9z/75h3yvzkiWYl69Al1+AaDIh+3wEHQ78\nrS30nTiB+8QJ5KHV5S+1n8LMYsgzpxxUnpcuhSn3YHy/4Ze9xCvz65VnXpU+hcsHWZJoePeZKZUh\nnm4oenHpCBo1pqVzMK9fiHn5PASdBm9t08WCL6NL2q2gMBhLaThvmaO+e9iiKlJQwlnfQ1pVNqnl\nGREDY9eRFlyNvZiLrBRcXc6FrQOGxMJr56DSqnDW99Bd3RY1XuqgvFu37vznhPLp0kbKNzvNEUUs\n85dgrliALjsXldGEoNUhasZuWOzHWX0EUacn85qbydx8ExmbbqDxdw/ibZ5+4a6TRfPOBp5c87tR\n9fE7fBx9aD9HH9o/qn6Nbyfn6ffinX+JOfbGZ7eOq0xJtY9T4DDZOSRL4S++Q/efnqXvwNh/n7xt\nTTQ9/0cKbv4oWRtuIGvDDTFtQh43vs7WMV9L4fJFZRj4zp3WhgZZxtfYiGHOHLS5ufiawmkvBLUa\nbWE4Okibm4uo0yH5fOHXWdkIGg3e+vMR3XYfP07I5UJlMqPJyUGbkzPipVUmM5qMDGSfF19DA/rS\nUnxNjcj+aENbsDt2QwRAk5mFZcVKsu68K9zO3ovamoZ5+XKMCxbQ8stf4G9vj+mXsmYNmbd9GEQR\nWZIIOeyoTCb0ZWXoy8pwHT4UN1VI1p13YVmxEgDJ6yXU24toNKK9OF/TwkV4/uMHhIYYCi+1n8L4\nM//ji3G1urjw+ujSHSTDqSeOkTYvE1OemeadDdPCI38mohgYJ5l2qZEMMS/qWK/URZ88jXfGFBQm\ngaG6oejF6BFUIoaFs7GsX4h5VQWiMbz77m/pwrnjGN1/fXtqBXwfkpoismqpnuxMFe4+iedfmfn5\n7dTG8EPX0BDmoQQ94RttjTk6V9WFraeY/7m1FN9cEWVg7A+Prn8xNiKhfwxZknE3JS7+NJPz5Qgq\nFUTZ3QAAACAASURBVAX33IdxVjn+znZcZ04Q6LEheT3ocgtI37B53K5lP/gezuOHsCxegXXFWoo/\n83m6d71F17bEBiyF9x95P/g3Wh/4r4TFiaYbztrj1P7veawLVmIsKkNjTkVGJujsxX2hDnv18JXs\nFS4Pln7/JnI2lkcdc523sfPeJ+K215h1LPn+TegyTegzzVF5Fm9451+j2nYfaWbf5/8WM4alLINZ\ndy0jY2kh2nQjIV8Q59kuWl49RfOrp4bdlBsrvsaGsIExvyBiYNQVlyCoVHhqazDMmYuuuARPbTgn\nsLYwnHrB1zDg7e86fAjX4UPheaxeTdbtd454Xe+Feloe+g0w4JXY+fRT+NvaRugZRlCpyLrzLlyH\nD2Pb+ndCTif6WaXk3PNxVBYLadffSPsfH43qYygvJ/NDtyOHQtie+xvO/fuRQ+HwdcPcuWTf9VHS\nb96C7YXno/pp8wuwrFiJHAzS9sjv8dTWRoyrKrMZY2UVKrOZkNM5Lv0UJoY5H66i7rnTEzJ2T203\nL31USSE1VhQD4yTTFKqjKaTE8isoDEXRjbFT9si/49p7CueOo7Q/9PyMehi83BBF2PpkHtduNEaO\n1ZwNRBkYv/vVdL75xTS++j0bP3945uQatR1tpezDC0mfn4OgEuOG3ItqkbTKsPeD7Vj0g0btE4dJ\nX5BL3oZSKj+zmlO/3UvlZ1aTVplN8/Y6zj59NGY82/HwGIIo8MbHnpyAWU08cjBskBVU6mGLsGTd\n8EGMs8ppf+Fp7If2Rp3TF5SMefyhSH4/9v27se/fTcqSleTe9lEkr4fuHZeWRkZhGiIIWD98A6a1\ny5Alib7dh+h97jUAtGXFWG+9Fu2sQlCpCDS20vPkC/gbWhA0anL+/Z/R5GUjaDUU/89/ANBw3zdB\nkij4+bfp+fPf6ds3oK+Fv/oett8/g+fwCbTF+WR+7h/o+O8/kPHJO9DOKiTkcNH+w18huftIu/tW\nTKuXIPn8OF55GzkwsGGRSC7dvDKyv/gpGj87UOE249MfAcD2u6eiph7qc2Hb9xa2fZNXsE5h+nD4\ngZcjfxdumc+CryTeoAm4fOz/0nOR19aqXNY8FDayvbrxwYR9M5YVsuJntyGHJE794h2qf/pmxOsx\nY0Uxi75xLQu+dg0X/naUUw/G5jUdK47du7BevRnLipU494V/O6xXX43k8dD2u98y64f/gXXjVRED\nY8qqNQDYd099UbSu557FsWd35LW3/jztjz1K/v3/gmnhQgT1wG+a2mol777PAdD4o/8kaI/ecPTU\n1ND84H9T/K0H6Dt1Ck/NQJqHkN2O7PcjaLWYlywl0NlJsKcnfM7lGjas+lL7KYw/c++YjzF75AJe\nClOLYmBUUFBQuEw49+mfIPsTe5UpTA733GXh2o1G/vGf29mxx0v94Vjj0NbX3Xzzi2ncdI1xRhkY\nO/Y1EHT70Zh1FFw9m6ZttTFtCq+dg9qoIdgXoPNAbD7ECy+eJG9DKYXXzuHUb/dScPXsi8fjF3Xr\na3XQe6YT67ys8Z3MJBLotqHLLcBQUoa7Jn7e6KAj/LCkScuIOq7LK8C6ct2Yx+/HUFyKp+lC1CaE\nvqA4LINrZhYfUoiPac1SjCsW0f6T/0FlMiLodZFzkrsP996j2B79GwSDWO+4ifR7b6ft+w8iB4K0\nff9BdGXF5Hzz/0QMi6NBlZZK2l1b6H36JQLtXWhLCgjZnaRu2Yx+wVzaf/wwIYeLtI98AJU1JSm5\nfGfOEeoe+L4U1GoMSyrp+vUfI8dyrv4g9hMH8bY3jeGdU1BInrn3rUcQBc48vIfGF6ujztkONHD6\n1ztY/MANFN2ygNrf7xn3IjFBu52Qw4E2Ly9SxEtfMgtfYyOyJOFrakZXUhI5p83Pj/SZavytsakK\nJN9AhXdBpYoYGLV5AxXEi7/1QMJxVWZz1OuQ20XHM0+TdcedWFauwrJiJZ6zZ3EdPoT76JFI+PhQ\nLrWfwviTu2r0FeQVJh/FwKigoKBwmaAYF6cP99xpAeCp51zDtjl/IXzDXFo8frn1JoOAy8/pR/az\n4P71LPnyRqSgROs755AlGUEUyL9qNou+dCUAZx47ELeic/t7DXg6XJjyUyi6bi7mIit9rU464hgj\n+zn+y11c8YtbWP7ANdT86RDO891AOBG/KT+F3PWzkEMSpx+ZnqGPvft3Y65aRN4d9+A6cRTJ50U0\nGGh7dsAj01l9mPQrryVt/SbUKakE7D1oM7IxVy3CefQgKUtXjnp816lqXKeORbXN/+gnAfA2XSBo\n7wVBIHXFWvyd7TirYz1IFWYugu5iegGfH197V9S5YHsXwUHHXO/sJfur98XNuXhJ19aocW7bge9c\nOLen92R4M8K0YQXO13fgv9AMQO/TWzGuWJi0XK4dA/ki9QvmIrk9eE+fixzLWLmRjJUb8dk6sJ88\niOPkIfy98XPAKSiMFX22hdSKsMd+25s1cdv0RAqRqbBW5dK1f/zz3XobGzDNX4A6LQ1Rr0fU6/E1\nXADA13ABfWkp2uwc5GAAUa8Ph/pOAwYbE0dCNFwsvCPLBLqGLwADxOSBBHAfPYL33FlSr9iAZeUq\nDOXlGMrLybjlgzh276LnjW3j2k9h/FAbNOQsUwyMMwHFwKigoKCgoDDOLKzS0edJ/IBu6wnhD8jk\nZKsmSapYiq6fh7UiC61Fh9qkJbU87DlnnZvFmp/cTNDtJ+DyEXD7Ofnwe5F+dU8dxZBjYfYdi1j1\ng+sJ9gXw9XrQWQ2RHI3nnz9B7ZOH415XlmQaXjnNvI+voOq+cKjWhZdOJawabzvSwoHvvcHK719H\n0XVzkYISyOHK0f00vh7/4W460HeuhpanHiX9ys2YqxYhiCIBR7TnaqCnm6bHHiLzmpswVyxAUKnw\ndXbQ/vyfcRw7hKliwajHd52qjmnbveNNzFWL0BfNQlWuR/L76Nr+Cr17dypVMC8z3HsOYVhUQf6P\nv0b7jx7Cf37Aq0+VYibl5qvRV5Uj6nUgCAgq1bgZGAH8jbHeSep0K4HWjsjrYHcvcmAgrH8kudy7\nDqDOSifY2Y1xxSLcuw9Gydvy8lOkVC7FVDKH7A03kr3hRjzN9dhPHsJ++gihvuE3fhQURotlUBGy\nq/72yRHba9OMI7a5FHyNjZjmL0CbnY0q1QqA9/z58P8X6kkFdCUlSB7PxfbTpKjXKL5rIt6Cskzj\nT39ySd9TIaeT7ldepue1VzHMnYtl1RpMCxZg3XQ1hvI5tPz6l5GcjuPRT2H0aIwaspflYS1PJ21O\nOtbyDCxFKQhi2AN30X3LWXTf8oRjPLPpMYKexI4XH3vv0wnPv3Db07hbR86tmV6ZyQ2P3ArAnu+9\nzflX6jDnW1hy/ypyV+QjSzKdx9s59vBBes+GN8cRYM5tlcy9Yz4ak4bmnQ0c+9+D+HqTM7gLokDB\nhmIK1heTtSgHc0EKfqcPd6uT9kOt1L9aR+/ZnqTGGm8UA6OCgoKCgsI4YzQItHWMfKMpAMHgxCR9\nT4aCzeXkrosN39ZYdDHHBxsYAY7/Yidtu+rD+RgX5GLINhNweOk82ET98ydo35v44eXC1lPMu2cF\nhhxz2OD48shJu5u312GtyCJndTHG/BQElUhfq5O+Vgdtuy/Q/Nb0zuPqOnUsxptwKN6mCzQ9+lDc\nc2d/9M24x0czPkDP7rfp2f32iO0UZj6yz0/ng4+iLSkg55v/B/vz23C8HM5JmPl//hGpz0vHf/2e\nUI8dXXkJOf+euEp7IoQ4lc4HGw4HGgoxmwmDH8xHkivkcJF6yzU4Xn4Lw+IK2r4XnR+v9/g+eo/v\nQ2UwkVKxmNTKpRgLyzAUzCJn8624z5/BfvIg9pOHLnmuCgr9aEwDRcj6mkZOdxIawehxqfQbDDVZ\n2WiyskCW8V6oB8BbH/5fV1RMyN4b1X4mEQmnFkV0+QX4mi89DYIsSfSdPk3f6dPoCgvJu+9z6IqK\nMC1eguvQwXHvN5PRCno2pd7Na72/H9M4ay0fZI/zhRHbZS7KYeP/u25M15oqUsvSMGQauebhLVH5\nIgs3lJCzLJ/XP/0C9vO9rPi3dcy9vSpyfs6HKslZnsdrn3iBwAhFFLOX5LLiK+uwzk6POq5PN6BP\nN5AxP5uqf1jMhW1n2feTXXEjiSaSCTUwGgQT2WIRFiENi2hFix61oEFERYggITlIAB9u2Umf7MQh\n2eiWOwjISi6DiaBEVTFFaxF9F6kT9OSIxaSLuZiFVLSCHhVqQgTwy368spseqQOb3Eqv1DXMmDOb\nobphFCyKXkwRI31PeWX3pOgFJNYNh9Rz2evF5URbR4j8nMSeiVXztGg0Aqdro28k1AYT2Wuvx1Ja\nhcaUQl9bIz0n99FTPVD0o2DzHaQvWsvxn38pqq8hp4jyj32R5m3P0F39HvqsAso/9gUCTjuiWk3j\na08iqnUUXHsHks/L0Z89zntffSnS7uyff0Hxlnvjth2MoFKRveoarJUrcLdZsZ9zYj9zlPbdryAF\no29kirfcC8g0vfZn5n3qW6g0enz2Thr+/ijPb/jNqN/bE7/Zw4nf7Bl1PwWF9zP+C83Yfvc0GZ+4\nA8fLbyFo1Ohml0SMeADqnNgcp/LFvIuCKEb+BpC9vrB34UXUmekI6uS8sYO2XjR52XhPhD2OVSnm\nyFjJymVctQR/fRP+hhaCXd1xrxPyuOk5vJuew7tRm1NIrVyKZd5izGUVmGdXYp5dRfOLjycls4LC\ncAQHGQN2fvzxCasUPRK+xkaQZTRZWegKC/G3tiB5w7/dIaeTYE83usICAlotyHKk2vR40V+oSWUy\nj9Dy0gl2d+NrakJXWIh10ybaH//TuIzra2rCW38eY0UlmoyMkTuMsZ9CYgJuP92no591BFEgbW74\nPfZ09eHp6ks4RjJ6eOx/DqKz6tBZ9ehS9aSUpGLKs1y64EBqaRpL71+FMdsUkaHf81Jj0rD0/tXU\n/O1kxLjYn14IIKXEytw7qjjx2PCpamZdX86ab10ZFb0DIAWk6GMClFw3m/SKTLZ9bitem2dM8xoN\nE2JgTBdzKFctwioOn4xdjQa1oEGHAbMQduPm4n3JmeAhWqV6/HKsi6iIyGbdnQgMvIEtofNUBxM/\nbBgEMxu0t8Qcb5MucCywa4S+JjZoPxh17FRwP42h+LkrTEIK67VbEo45GJdsZ7f/paTbj5aR1iPR\nWjjkblpD9cOuRzKEGNi91gg6rtTeGrV+A3JoUQtajIKZdDGH2SykV+qiNnSEHqkjpv1MZCLX4lrd\nR0atFxBfN5LRi3DfWN143Re/wutM0wsIr4dZsE6KXsxRLaZAVTasbqSLOZetXlyOvL3LE8nDOBxf\n+9c0AF5/O/omqfgD96JLy6Hr4NsEXL1YK5ahMaVesiyCqKLzwJtkLttIwdW3E/J5ad/1ClkrriZv\n4wc598wvI+3yrrp12LaDRqR4y72Yi+dgO7wTX3cbuoxcMpdciSGngHN/fSgmbEljTqXklk/Svutl\nBFGFuWQufufUhG4oKLyfMCypQvJ4CTS3o5tdQrAznItQDgQJOVzoK8rw1ZxDU5hH6s2bYvoHO7uR\nQyGMqxbTd/A4si+8geA734h5w0q8J2pBgLSPfiDpIjDunfuxXHcFvtrzhOxOrB++IdI3WblEvRbz\nVWtw70wu52rQ5cC2/x1s+99BbTRjmbOAvBvuVAyMCmPGeS5sCBFEAUt5Fo6aqbk3kzweAjYb2pxc\ntHn5OPZGRx146+sxzV8AMgQ6OyLGx/GifwMiddMmfM1NUeMLKtW4hQ/bXnyBvPs+h2nxErJDEr3b\n38Tf3gaAymhCnZGOsaKSnm2vR/VLWbMGldmC++SJsCfkxfsUQaXCWDUfQ1m42JyvuXlc+ilcOl3H\nO3j13uejjmktWm7fdg8AtX87RfUj8VPwjIahY1R8ZAHLvrBmTGPmripAVIsc/tU+ap45gagRWf2N\nKyneXApA3tpCUmenEXD52fODd2je0cB1v7uFjKrw82jx5rJhDYzZS3NZ+52NEYNkX7ubk48fpWV3\nI65mJxqThswF2cz7yALy1xYBYClO5cofX8sbn90aTi00CYy7gXGBei35qtIxjTFPvYxieS47/H+P\nOSch4ZLtWIS0yDGLaB1xTKuYGfd4qhD/+GAGX6sfhxR/t3Q6ISJSpV49pvVIEdJJUacPux7JEJTD\nO1oWIY2lmo1xDSjDYRUzWSFupiZ4mAuhkcPnpiuTsRaXohcQXzeS0QuIrxvTnemqF3oh+Zw8l4te\nXM78/OFe7v6whS9+1sozzw/k+xJFWDxfx1fut3LHLWbcfRK//r09cl5QqzEVlNGx9w06D2wHoPf0\n2MP4uo/tQRBU5F/9IRpe/iP2M0dQm1LIWnFVVDt7zdFh2/aTUr6QlLL5NGx9FHvtQDhuwGUn/6rb\nSCmbj+NsdN4/Y94sOve/Sffx8AOP7ejIGxhTidU6i6rKO1Gr9Zw99zrNze+N3ElhVPz8J05+/pOR\ncxuNNykpRaxY/jkAgkEv7+74/qTLMJmozCbS7tqCypqCr66erocHNgFtf3iG9I99EMv1Gwk0t2F7\n5C9kf/kzUf0ldx89f3oO64euJ/0fb6Pxc98CoPcvL5Nx7+3k/eBLSF4fjpfeQjSbSAbHK++gzkwn\n52ufRfL5cWzdjjp74L4jGblcOw9guWY9XQ8/kfR7IWq0mMurSJm3GHNpRZRHpoLCpeJpdeA400HK\nvGzKPracI999Zcpk8TU2YF6yFAQhkn+xH299Pealy9AVFuI8GGuYz7jlVkRDuDiMJisbAE1aOrmf\n+jSSx4vk9SB5vfS+tT2SxzEexnkVlHznewTtdgS1GpXJRPerr2B/5+1xmaP33Dk6nnyC7Dvvwrxs\nGeZlyyLGS0EV9gRwHYq9b1Jb07Buvoa0629ADgbDFbRFAZXJHEnv4Dp6hL5TJ8fUL0tTzGz9Et5z\nhp8PKg3rCMhe6ryHSFFlUmlYg140E5D91Hr30xkYvrhdPyvMN2AUU5CRueCrpsF3ihxNCbN0i9jr\nejHSrsqwHp/cx1nvYYxiClXGdZF+Oxx/AWCWbiFqQYNRTCFNnYcA7HI+i1WdEyX3YNkbfCeRkLgi\n5XZEVNR5D9LiD6ekSTQnqzqbKsMVqAUNtmALMnLM+zP0PZrpqLQqGt+u59Tj4fvjkD/Evh/tpGBD\nMSqtCkEUMOWa2fd/d9L0TrgI07H/Pcim/74BgLQ5GYgaFVIg2iCv0qlZ971NEeOi7UQH2z//alT4\nc8AdoHVvM617m1n6L6upvDtcPC1zQTZlN8+l7oXJeWYcVwOjgDhm42I/zaFzw55zSN1YVAOGDZOQ\nioCAnCA7vFWI76VkEExoBX1CL6ShRhQZCac8vT0vBESWaTaRLuaMy3iJ1mMkggQwCSms0FyNRtCN\n3GEIAgLz1MsIEqA5dPaS5ZgqJmstLkUvIL5uJKMXEF83pjOKXihMFtWn/Nz/9U4e+mkWP/lOOKSj\nvFRDX8NsLt7/EgzKfOaLnTS2DHizysEgvu5O0hesxtvZjKOuGlkeH70KuC7mXeoK7/SHPC5ETbiA\nQj/+3s5h2/YXWEiduwgp4MdRF21EdF0IhzuaispjDIwAXQffGZd5TAZFhevR68ObNGWlmxUDo8KM\nxbVzP66d++Oe81bX0PKNn0Yda7wvNs+na8f+qMrNAKEeOx0/j87H5XxzYOPA39BCw6e+Fve6cjCI\n7ZG/YHvkL3H7JiOXoFbRt+8osj9xripBpcZcVklq5VLM5VWImnC+PE/LBRr/NrZ8YgoK/Zz+9Q5W\n/vw2cjfNYVFQ4twTB3CdD3sLa1L0GPNTyV5XihySqHts34TJ4WtswLx0GQDe89H3qL6L+RgRhHA4\n9RBSN2yIOSZotRgrKqOOOfbsGdbA2P7oI6ReeSXa/ALUaWnIXi/+1lYCHe2XMJvhcR89QuOFelLW\nX4FxXgXq9HQElYpgTzeB7m66X3k5po9z/z4EjRZ9aSmajAxUVivIMpLbje9sHa6DB3EdPRITgTHa\nfp2BBuYZVpGiysQR6iJXW8oe5wuoBA3LTNdyvO9dbMFmjGIKq8xb2B96Cbdkj5G3HwGBGs8BHKEu\nNIKOtZZbsQe76Ag0UGFYh1mVhivUg0pQk6stZbfzWQQElpiuprpvZ6RfqioLeyh8j1ekq+SA61WO\n9b2NVtATkH0xcouoIrJD2EFjp+OvmFVWVptvoTvYRkD2DTunPsnBIuMmajz7aAucx6rOZpV5y4jX\nuRxo2B5t3Pc7fXRVd5CzLA8IhzPXvzaQM9x2YpDXswCmPDPOhujPRNnNcyI5HaWAxK5vv5Uwt+LR\nh/dTcl0Zxqxwn4qPLZiZBsbZ6vjVDe2yjV6pE4/sIkgQkFGjRScYsAhWLEIaOsEQaS8j0ywN/8Ds\nkLspYHbktYiISUjBJQ+vnMN5MELYW6tTHj4PhUWMNqK4ZDtSAkNKAB+NoVo0gg6TYEGLDo2gQ2Ty\nKoXOVi+Ia0QZ7VrAyOsxEgICizRXRBlRPLKbLqkFj+zCjxcVavSCkXQxl1Qhfg6LCvVyeqQO+uTJ\n93gYC8OtBQysh1PuZaxrcSl6AYm9exPpBcTXjeEYrBdhndBOum4ksxZDdSNTzJs0vYDhdaNUNT/u\nODNVL94P/O5xB5+6O4UVS8JrLA5y3j50zMeXv2Njx3uxN+kXtj5C0fUfpXjLvQTdDmqf+C+CbseY\n5ek3VEoXqwTL/WE+gwyMA+di2wqCgCzL6KyZiBotC77w/+JeR62P9caV/D6CnplZuVWSlKqQCgrT\nBUGtRlCrMG9cTft//Dp+G1HENGsuKZVLSZmzEFGnB8Df3RGuJH3iIP5e22SKrXCJ5F9XQWpFDmqz\nFo1Jh2V2JmqzjuU/uoWg20fA7Sfo8lH7+/eQQwPPZtnry8hYVojapENt1mIqCt+vGvJSWP7jW/C2\nOwm4/Zx/4gAB19jzancfbeboD15j4devIf/aeeRfOw85KCHLMqJm4B63ZduZMV8rEfYdO7Dv2BH3\nnK+5mXNf+bdh+yY6lyzuE9W4T8RuMA5F8vkSXs/f1jaiPMHeXrpf2kr3S1uTki1gs2F7cfRGrEvp\nd8FXTbGukuq+HTiCXXglF+nqPIKyH1swHErdJznoDraQqSnE7Rv++UkvmllqujbqmEmVij3USbP/\nDIXaeZz2vEeupoyeYBteqQ+DaMGsSo/qV+vdHzEwdgdbcYbC34GDnUkGy52tKYnIrhX0kTauUC/O\nkA2rKhu/7Bl2Tp2BRrSCnrZA2NjWG+ygT3IkvM7lgj1O9WZXkyNiYLSf7yHoHXAu8Dv9hPwhVNrw\nd4XWrI3pP/uD8yJ/t+xuxNWS+LlPCki07G6k/IMVQDi/oynXjLtt4t/ncTMwatAyS1UVdaxX6uJE\n8D3c8sgPRmYhlQwxj1yxGB8efPLwrtfxwpPNgnVY40Z/HrXhsIoZdEoJDIxD+tqlxDclftnHqWDs\nbrEKNSWqeZSrFyfsP1birQXALv/WUa9FqphJp9SUcD1GIlPMj3p9KPA2XVLLMK2PkiZms1C9Fr0Q\nHWoTfv8q4r6305Xh1iJZ3RjNWoxWLyCxboykFzA63RhOL2BydONS16IhFL4ZnGi96JNdnA4eGFY3\nuqTWy0Yv3k+svbGJonw1i+ZrSbWION0yx0/5qW8Y3uvGZ2un7sn/xlRUTvrCNcz7xDdoePmPOM+d\nHLZPP4KYfAqK+CSTnF4k6HHR8ubf4p71O2JvrGQpTiXZaUxD404sKYWoRA11Z6cu3E1BQSEaXcVs\nsu6/h+5H/kKgrTNum7n3fw+VIfxbGXQ76T2+D/uJg3jaRg5FVJhe5F09l6y1s2KODz1W9+he5EF7\nQVlrZlF0S6zji0qvIWvNQN/GF46Pi4ERoO2tWnpPtFHyocVkrirGmJ+KoBLxtDnwtDro2FNP29vx\n8/crXF60+GspsyxBI+ho8vd7jclRESPJIiDwruOpuBFpTf4zrLXcSo1nP4XauZz1Hon0kZGG7def\npimR3AW6OTT5hvN4E8LzGe2cLm5sJ3+dmYmnyx1zbHBl6HhGvpAvGDEwqnTRjjcak4a0OQMOWJ3H\nk/MIdg8xQqaWpc0sA2O2qghxUG69PtnJwcD2qEIGiXDJdlwhOxdCpzEKiRPjO+UeZKSoXH4WMY02\n6ULc9qliBgIDH/7WUD15qlkD5xPkm1OjwSBEV8NyyJeWfzFEkCCJQznGg6FrAeH1SMa4CNFrYRDM\nUe/dWLFLtgTGxTA9Ugd7A9tYp7kxxrsrX1VKbfDIpLyP48Fwa5GsboxmLUarFxCrG1HnRsjDONN0\nYzLXYrTYJRsHA9sTvgeXk16832hsCUaFQSeLu7EOd2Md8z7xDfI3fYgzgwyM/Qa7oYnTNZbkcq+O\nBb+9C31WPo5zJ5BDM8twmCx2+wV27/7xVIuhoKAwBG/1GRo/GxvGPRhBpaa3ej/2EwdxX6iDcUoz\noTD5HPz6peW5PvGz7Zz42fYxXbv3ZBuvbnxwVH28HU7OPLyTMw+P6dIKM5yQHKLFX0uxropz3nDB\nDnuoCxUaMjWFdAWaMIopZGjyOes7knAsj+SkVL8oMo5FlY47ZEcihFdy0xvsoFhXhVY00BVsivTp\nCzmi+omokEgckTFYbrOYTkec/JBmlRWLKp3eUCcB2TvsnDySi4DsI1dTSlvgPCmqTIyq1KSvM5Px\nO2NDlwdXtfY7Yzc1Bv9MCUOMtmlzMyO5FwGW3r+KpfevGrVc+jT9yI3GgbG6OkQY6snUHDqbtHFx\nKCOF+vUXekl0/cEMzTHXLJ2NyhWXIqYPaywYGgIK07/AS7z34lJztHlk17iFXvpkDwcDbyXZto8z\nodjqUCrUpCYId59uDLcWl6IbI63FaPUCYnUjWb2Amacbk7kWo6FfL5IxDl4ueqGQiFidc7ecRzOo\nyAqA3xHWNUNOUdRx67xlEyfaRew1RxFEkYwlVwzTYvyM7woKCgqj5cwvH6DlpT/jrq9RjIsKLRs0\npgAAIABJREFUCgpTQoP/JCW6+ZFnq5Ac5LD7dWbrlrAx5SMsNV3Dib5duEO9CceRkUlVZbMx5SNc\nlfIxKgxrogxQjf5TlOuX0eQ7E9XnkHtbVL+hRquR5G7x10Y9F7pCPWxIuYNlpus55dmNV3IlnJOM\nxNG+t5itX8qVKXdSpKuga5AhcbjrzHRkSY4yJsZDCoxuvuNlGFTpxr2+c1zG7SraIfnJvHLfeA0d\nl6EFLRKHQA88eMvI2CVbVMVdNRpMQiouOVbBhxolwkacxF8EU83QtYCJX49kqA0dIcjwyUiH0ha6\nQIV6GWqi8xBYhQxstI63eBPCZK/FaPQCYnUjWb2Amacbil4ozASMucUUXHsnznMn8dm7kEMhrJXL\nsQ+pJG2vOUbOupsovvEf6Tz4NnIoSMrs+Wit8QuajSf22mPYa4+Sd+UH0Gfm4W4+hyAIaK2ZpMxe\nyPm/PhQpEqOgoKAw2chBxZtfQUFhagnJQcQhphZHyMZeV3I5Iwdz2L1t2HNdgSbesD8Wc9wjOeP2\nq/cdT3itfrmb/TWRY37Zyy7ns3HbJ5pTb7B92H7xrnM5MDgX7PCNRjemZkhOxpA/hBwa5SCAFJwc\nQ+64GRglOdrl1jAkT9h4M7SghV4wokYb86AuIER59vTJTkIEYwwxqWIGrjg7CEOr5Lrk3oQFXqYD\nQ9cCJn49RsIne2kN1Y+qj0SIXslGppgXddwopjCCh/e0YbLXIlm9gPi6kaxewMzTDUUvFCaTB76c\nTtksNffe3zFy40H4nd34ejqwVq1AbbQgh4K073qFrkPRFZgDzh7qn/tfcq+4mdz1NyFLEo5z1TS9\n/hQVn35gPKcSB5mGl/5IxpIrSJ+/GuvcpchSEL+zB+e5akLeqTfcKygoKCgoKChMFbN0C7EFE+ey\nn470y91fkGWmX+dyINgXvWn23vff4cIb54ZpPfWMm4FxaLhgoWoODaGaCcsJFq+YhEW00iNFP8yZ\nhFTUaCKvnXI4+bxdtkUZYqxCJs3EhhEPDQOdziGg/cQL3SxUzeFc6MQUSBOmXWqIm2R2JNyynUyi\nDSkaYisrAajTLGT94zWYFpWhSjUhB4IEOntpe+hFPKenJrfDcGsxUbqRrF5AfN1IVi/C484s3Zjs\ntUiGydALhanhw1tMVM3TjtrAGHQ7adgauxMdD3fTWc4+FZsfqvrBr0b+9nY2c/znXwLAee5k5G+A\n7mO76T62O6bdcG2jkGVsh3dgOxy/WuVgGrY+mtR8EpGXu4zKytsBeOvtbyPH2TAYqZ/f72LX7h/H\n9E1JKWLF8s8NO4bNdoajx5Jbk8Fj1da9RGPjLgAyMuaSl7eCFEshWq2ZUMhHX5+Nzq4TNDSM/B6O\nN/HkTCRjc/NeQqHkPK01GiOZGRWkpZdjMeeh06WiUmmRpAD+QB9udzvdthra2o8QDHpHHnAQgiCQ\nkTGPjPR5pKYWo9Va0GgMyLJMMOjB4+nB6Wqht/c8HR2JvTSSoaT4SmbPvuHiK5ma2q00Ne0Ztn28\nuQuCMOZ5z53zAazWWej1VlQq/Yihbk5nM/sPxK+wHEYgK7OSrKz5F99HM4Ig4ve7cTib6Ow8QUfH\nsUil+WTQ6VLIzV1KWtpsTMZsNBoDgqAiFPLj8zno6+uk116PzXaGvr6upMdVUFBQmEnkaWdTYViN\nR3JxxP3mVIuTNJMl90x9f6YSb2/0PYMh0zhFkiTHuBkYbXIr5SyKvNYLRlZqr+F4YHfCKraXSr+3\n1OCiDRbBSg/RD3PWIXnJ+o0gQ40hqWIGQxEQMAupUcfscuIK0tOBoWsB4fUwC6kTshbJ0C0lV+1o\nKAE59oFmaGhoP/lf+jDG+bPwnm3Bue80AqDJTSfQFlvRFKDi2e9y5q4fIgcmrkjBcGsxUbqRrF5A\nfN1IRi9gZurGZK9FMkyGXihMDcWFk5Pn5P1CR+dx5sz5AGq1joyMeXR1jVxRGyA3byAnZVvb4aQN\nk+OByZiNKKqprPgwOTmLo86JoprUVBOpqcXY7Rew2xsmTa6h9Mu5eNG9UccHy1hUuJ7qE0+OKKda\nbWD9un9HFFUx51QqHQaVDoM+jcyMCkpLN3Pi5NN0d9clJWda2mzmzf0gRmNsvllBAK3WglZrITW1\nmMKCNWwfBwNjv3FRliVOn36W1rZDcdup1QbmztlCdvaiuHO/1HkLgsCc8pspLFw7tokMwmjMYn7V\nnVgsBTHn9Horer2V7KwFlJRspLr6yaSMgcVFV1BWdh2iGPu9p1brUav1mEzZZGXNZ075zWx/6xvj\nMhcFBQWF6Uar/yyt/kurfTCVTJbcM/X9mUp667rDYdUX9xYz5k98OqSxMG5PQHbJxh7/K6zV3hg5\nZhHSWKe9OfK6W2qjMVRLh9Q85mSeEhItoXMUqsojx/LEWTSEouP4Z6mqol43S2F3UofcjUPuJkVI\nB8K56tLFnKgH/lyxBJGBG0WJEC2h6euO2k//WqzRXh9VUXii1mIkZGQ6pUtzEY9X7UoUYmsT6Yqz\nMc6fhftIHY3ff/ySrjURDLcWQ3XjaGDHpOoFxNeNgOwbUS9gZupGsmsxWboxGXqhMHXUnA2wbJFu\n5IYKSREKBThy9A+sWP45Fi74KDt3/YhAwJ2wT+msq0mzlgFgtzdQd/aVuO0cjsYYg8esWZsoK712\nTDLn568kP38lDkcTe/f9N2539EZPZmYFVZV3snzZZ3G5Wtl/4NfIU1CQol/ORDLqdClJyRn2IrTR\n2nqAzq6TeDzRm1aCoKKoaB3ls29AozGxZPEnaW7ey5maF4aVLyN9LosWfTzitRcIeGhoeIe29iP4\nfINDqwRMpiwy0uei0Y4t/UVZ2bXMKtlEKBTgePXjdHfXJmwfDHpwOBqxWArizv1S5m00ZrJ61RcQ\nBJHqE3+O65FZVnots2ZtAsIeurv3/BhJGt6Ivm7tV9Hrw/mTHY4mTp9+Fpe7LaqNxZLP/Kq7MJty\nWbP6S/T1dbJv/y+RpNjNWEEQWb/u62i1ZgAuXHib8/XbY9oKgoDZnE9WZhVmS17MOAoKCgoKCokI\nuAPIkowgCqSUpI7c4TLC1+vlwM92s+LL6wAouXY2ni4Ph37x3hRLFp9xdbFwyj2ECKIaZth0MZd0\nMRe/7KNVOk9TqA63fOlx9w45+sbVLEYXndAKeoyCOfLaK7sJyANlwe2SjRRVeuR1qpBJNwOGlKEh\noE6595LCGacCp9zD0cBOFmrWxV2P8V6LRIQITsj7Zl4xl6yPXY06PQVVSthV2LSknIpnvxtp4zpY\nQ9N/PBl5nXLFAqzXrUBXmgvAvKe/FTWm7a/v0vnk9nGVc6S1AFis2TBpegGJdWMkvYCZqxvJrMVQ\n3TgTjO+xMlYmSi8UpgePPe1UDIzjjMPRiNvdjsmUQ07OYpqadidsn5s74L3Y2npwosWLi9/v5MjR\nP8QNie3qOk31iSdZsviTmM15ZGZU0JmkZ+Z44/c7Y4yLEC0jkJSce/f997DnZDlEQ8MOjMZM8vNW\nApCeXj5se0EQmTv3AxHjotfby6HDv8XrjReZION2d8Sdx/DyxH4Hz559PSXFGwE4evQReu31SY3V\n3LKXpub4N/ujnTdAVuZ8hIsbR8OFe9df2E5+/kq0WjNarRmrtSyhMbTfuNjX18nhI7+LG/budLZw\n+MgfWLP6C6hUOozGLPLyltPcvDemrU5riRgXQyE/Z89tI172elmWcTqbcTqbE85ZQUFBQUEhHrIk\n42p2YClKJX9dETqrHl/v6NKNzGTqXjhN5d0LMeVZAKj46AJ8di8nHjuSsGiMqBYpuKKY7jM23K2x\n6cImgnF3eXnP/yq9UmfCNlpBR4mqgvXaLazQbCZLjA3TSIah4Zwq1FFGE6swJARU7hnyOrq/dUg4\n6NAqudM9x9xQOqSmEddjvNYiEUF5YvLbBW0OHDur6f77brr/Hs6J5G+x0fn4G5F/9u1HovqEnB7c\nR+rofm4nAJ1/3h7V3n1kYly2J3MtRtILSKwbI+kFzGzdSGYtYGA9ZppeKEwPHn7UzouvuVlQqYSu\njyf9hsK8QaHP8bBaZ2EwhDdKQiE/7R3HJly2eDQ3702Yb29wmGxu7tLJECku8YxH/XR319Hbez7y\nejzkbGsb+G3W6WI3wfrJzKzEYBj4DTp56i/DGBcvjVDIF/W6vPymiHExEHAnbVyE+MbKoSQ7bwCj\naeQQKEkK4XYPbACajMP3UakGci6fPbctYU5Nn89OW9vhyOv8vBXxrz/Ik1Wl0mIwpMVtp6CgoKCg\nMFaad4ZrKmhTdFz94I3krsxHY7z42yaA1qIlZZaVnGWXn6e8FJDY8fU3CXoHIgQWf3YFNz95O5V3\nL8RSmII534K1PJ38dUVUfmwhG350DR965W42/OgajNmTV9h03JNEuWUH+wLbyBWLma1ehElISdg+\nXcwhXczBKfdSFzxCp9SS9LXi5ZszC1b6ZBcAVjH6RssudQ15HZ0zLnWI0WVolVzHNM8xF4/B67FI\nc0XCtmNZi0RMVGVh7/k2vOfDoT3agkzSb1lLoKMH27M7h+3jPnoW99GwETHrH66h+/ndE5qDMera\no9CNidQLSKwbI+kFzHzdmKy1SMR0rritMHbSrSp++F897HutkK3b+jh8zIe7b/g1f/C3U5Mfd6bR\n2naY8vKbsJjzMZtzcbna4rYb7L3Y0VkdY0iaLLp7kssvCJCaWjyBkiRmJDk7u05itZYC4yOnd1D4\nsCiqEEV13BDc9PQ5kb+dzuYoQ+d4MNj4O3fOFgoLw+FHPp+DI0f+MK7XguTnDSAKsbkc4xGSBjar\nhASpMlJTSgCQpAA22+kRx+3pPU9BwRoAzOZ81GodwWC0HoU9X8NexQCLF3+CM2eep6dHya+loKBw\naazeaOSnj8YaiL73r+28+aIrTo/pw/L1Bj7z5XRK52pxOaT/z955h0d21Xf/c6f3qt7Lquxqe3f3\n2tgGbNwwNpiOQ0kCmARIeJ28lACBBJIXCAFCiB1jY7DBGBuM+7p7i9fbq1arVa+j0Yw0vd33j7HK\naKqkUdnd+3mefVZzy5nfOWfOzL3f+yvs2unj+/+Q2aFBIndO/Oowddc3oDKpsTbaueo/3g1ANBhB\nppQjyOLRDkF3kEeveyBtOyqjirobGtEVG1DqVagMSpQGFcbKxNDry75zNUFXgLA3RMgTIuwNE/aE\nCHtCnHpk8YvnOk85eOXLz3Hpt69CZYpHSZlrLWz43DY2fG7botuTjgXLQj8Q62Iw1E2RrIIqeRNW\nWVHG442ChQ3KKxmO9XAg/EpO7xEjhkd0TeaLAzAIZoaI5zWb6aXlEhMFRq/oJkJ4spKuStCgFfT4\nRe/k6+m4zyEvrZkMxLoQw6/Oei6ORfYSEi8c9+PFYDZrYy5zkW1dQOa1kcu6OF/WxkLPhcSFS/+x\nmsm/b3m3nlvenfnJoSQw5sb0vIslJRtpa/tz0jEymZKiwjWTr/v79i2Kbanw+XK/sZioiBwO+xfQ\notRks3O6kJsPO2M55po0mSom/56NWJsrkWj8e7yu7ppJcdHvd3Lw4P/gz6On5AS59hvAH8jtd1Wv\nm/rd8gdcaY8zmSoB8PqG04qa0wkFp0KpBEFAo7biiSQL+qdPP8m6dR9FEOTotHY2rL8Lj6efvv59\nDA4ezporVUJCQuJ8oKRcwb/8TykqdVzk0upk3HinSRIY84jf4eOlv32Ga39xY8J2uXp2kpbGpmPj\n3duzHmdrTnaymWApBEaAgTd7eeojj7HxC9upvKJmsvBLJlxtTvyOxfstXtAylyIig7FuBmPdGAQL\nFfIVlMprUGaodlooq8AgWPCI6S+SpjMWcybki9PLzBCNV7k1yaa2i4hJYZzxbSPYZCWT20yCfVJI\nmU6MKN4lqsCcL+YyFxcpbbwVfinn+ZDIjZlrY4vqHXmdi3TrArKvjWzrYmZ49Lm+Nmb7PSWtC4lc\n2LtfEqAXmpLi9Zw583RSwZGiwhYUiviTXZ/PMasw1/wiZgyPToVCoVsCgTG7naFQotdINjt1ukLs\n9kYM+hK0WhtKpQ6FQotMppz03MsFlXIqvcfMgjH5IBoJUlG+nZrqHZPbHCMn5yUuzuy7Vmufdb8B\nhoeOTdplMlUwNpZcFKysdHNCKoDRDCKsSh3P22Q0lHHVjn+eTZcAUCi1Kbc7R9s4cPBemptumazw\nbTCU0tjwHhpWXM+Is5X+/rdwOI7nFEYuISEhcS5y6TX6SXFRYuFwHB3irX/fRfmlVVjqbahM8fu1\n0HiIgNOP68wojqO552I+F/EOeHj1K89jqrFQtaOWoo2lFKwuQq6SEfFH8Dt8uDtcOI4OMbCnh9HT\ni+sItKAC43Q8oouTkX20Rg5QLK+kWt6c4GE1nY3KK9kV+jNh0ueHmSCpoIUQd23VC+aEKrcTBWhm\n4hJHsDElpJhlNgZjXcnvExs9bwoyzGYu1IJuVvMhMXs8oouXg4/ldS7SrQvIbW1kWheGpPyLF97a\nkNaFRDYuuX7hihnYTfXUlVyGSV+KXJYshHcM7qK151kASm1rWVN7C8++9Y2k465c+yUUCg3P7/8W\nADKZgivWfhFEkZcP/xsxMbEabbl9PS01NyW0P4FJV0ZtySWY9OWolQYikQD+kAuH+zTdw/sIRfL3\n5DQYHEOtNqFSGbDbG3E4EsM9l0NxFwBRzC0v33Tk8sXP2ZmLnbFYYs7YTHZu3vRXCZ6H80GhmPKW\nj0byH+ZuMJRiszUmbKusuBjPeB/9A7Mr8GW11FFf/8689X3c08fg0GGKi9ayccOn6OndjdvdSTjk\nRa02Ybc3UVKyfvL4M+3PZhSKp4/lXMgUsu1ynWXP3h9QXLSWyspLMBrjeYsFQUaBvZkCezN+v5P2\ns88xOHhoXnZISCwUq6/5PAZ7YgoI9+BpTrz4X0tkUXpm2joXOy2lTTRc/GHcA620vv7LfJt4wWEv\nSi2r6A0yvB4pLVI+OfXIsXl5EI51unho+y/yZo/zhCNrewd+tIcDP0qf7zpTSHc6xjpcHL3vANx3\nIPvBi8iiCYwTxIjSH+2gP9qBVVZEg3xdUj44jaCjUbGBY5H0kzDBTK9EnRB/QmuWJYoC6Qo6uGMO\npmktacWdmYLN+UAucwGzmw+JuZHvuUi3LiC3tZFpXczMV3i+r42tymukdSGxbCg0N7JhxQfwB0dp\n63uRWCxKecEGTLpS3N5e2gdeZcw7tzyhsViE/pHDVBVtpdDSxOBoYqXgEttqAPpHEgUCm7GWTQ0f\nIhoL43CfJhTxoVYZsegrqC29jK7hN+fW2TQMDOynuvpKAEpLNiUIjGq1Gau1HgBRjM1aJMongiAg\nk8mJxaLZD36bpcgVmYudM4XsdHaWlm6aFNhEMcbYWDdjYz34A07CYR+RSIBYLIJCoWHN6g9mtS0a\nDU2KmQshvqrV8d8zp7ONUNhDSXFcsGtuvgV/YDTnnI+lpZtobrplMgfi9L6PjffMut8TnDjxKHK5\nigJ7M1WVl0Jlci5tUYzS3v581qrqEwJtMOjG5e7M2YYJgqHM1SdFMcbA4EEGBg9i0JdQUrqRkuJ1\nqFTx6w+t1kbLqjsosDdz7PjDs35/CYmFZrhjHz5XHwq1Hr2lDLUhucjhcmG6rbaKNdlPSIG5uBG5\nUoO1vCXP1l2YOAZTp56QxEWJC41FFxinMxobYm/sOcrldaxUbEnwqiqT19IaPUhYzHyxPbOghRwF\nWsGQJBS6ZhR4mcA9Iy+jSZZOYDy3iljMlkxzAbnPh8T8ycdcpFsXftGT09rItC6me0PC+b829oal\ndSGxfFhRtgNRjLHv9C/xB+Mh+n0jh7h8zd3oNXYc7tNJIcOzodexn6qirZTbNyQJjDZjLeP+Qcb9\ngwnbKwu3IAgy9rXez5ivP2GfTmMnHPHN2Z5U9PW/RXX1FYCA3d6MQqElEomH65YUr0MQ4iFKIyOt\nhLKIIguNUqknGBzL+filyL8I2e2cCK+dIJWdGo2FpsabABh1tXPixKNpKz5rNLlVGw6HvahUhsn2\n840oRjlx4lEGBg8iCDJUSj02WwOCIGfN6g/x6mvfzNrGRL8nxMVMfc+13xPEYmGOHHmAHVd+G1GM\nEYtFkcnkRKNB/H4no6Pt9Pbtxe/P/jscejsXotc3zLFjv5mVHbPF4x2gre3PnDnzNAUFK6mp2YHR\nUAZAcfE6SWCUWJYMnp4S6StWX0vF6muX0JrMTLd1+/u/P6c23IOtFNVvw9lzNF9mpUWlNVO+6ip6\njr9A2J/7b+K5xGvPefnMV+yoNVKYtMSFzZIKjBP0RtuJiBHWTatyLCDDLhQzICaHK08nVUELvWBK\nFlHE1AJjSAziFz1ohfgFrAIVOsGQUHEXkj3CzldSzQXkPh/nGoJSvmhVpGfLfOYi3bpIKTCmWBuZ\n1sV0b0i4MNbGhbYuJJYnMkGOUVeCN+CYFBcBorEQLm8PheZGNCpTwr7ZMu4fZMzXh91Uj1ppJBie\nXuhBluS9GN+e/mLaF8j/Awi/fwSXqwOLpRaZTE5x0Rp6+/YCUFS0dvK4/v6lK+4ygV5fnLPAGAyO\nTQqli002Ow2GqZQZ6ewsLlqLTKYgEglw5MgDSRWHp6NU6nKya3y8b7JCscVSl9M5syEaDTMweBCI\ne+AdPfYQGzd+GoO+BKVSm1Mxm4l+A1n7nmu/pxBoWfV+wmEfBw/dx/j43FMvjI11A2A0lBLPCr/w\nqU1EMcbw8DEcjhO0rLqDoqK5eVpJSEjkH1f/Kd589B8X5b0sZSspbriEwbZd563AONgX4e/v6ufT\nf2enpkGJezTGU79N39drv7KWvQ+ewdXj5eZ/2cKKy0r48z8d4PjTyfl2JVJjXllC6dVNWNeVoy03\nI1criHhDeDudDOxspfuPRxAjiQ/dL3v4EwQGxjhwzxNc8egn8fW6OP79F3CfGKD+Y9upunUdCAJ9\nTx2j9WevIcbiv5VKs5Ydj3+KsdYhdn/q19i3VFH7gc2YV5Xi7XLS++dj9DxxZPL4CxnZUhswwWCs\nC/cMTyjdjFDMdMwUOLSCHoNs6kl3UPQRSFG4ZYKZ72sUbJMVdAGiRPCK5+eXYSpSzQXkPh/nErpV\n1UttQkbmMxep1oWAkPPaSLcupleQvpDWxoW0LiTyS3Wlgqsv13LjO/VcdZmW6oo5Ptt7W8ibmRtx\n+ja5TJm0b7b0OPYjCAJl9nUJ20VRpN95JOn4AWfc+2FTw4epLbkMlTJzxex80DdNPCwujtup1dox\nGuNeUqGQB8fIyZTnLib2Gfn9MuGeQ9hqvshmZ2HBqsm/09k54Z03Pt6bUVwEsFpqc7LL6Tw9+bfF\nUjM5vwtFJBLk8OFfTnq+rl79QYQMuQch0SsxW99z7fcExUVrKCpaw+m2J+clLgKT4d5KpR67PffP\nZT4QxRgdHS9Ovs70UEJCQuL8w1KyuN85S8X+N/x8+uYerms5y+2XdnLfD9MXDGu+ppzxIT+VG+0Y\ni7Xc/+GXufTTzYto7bmNeWUJ2356B1W3rcfYUIggEwg4vMi1Sixrymi++0rWf+uGlOcaau00370D\nEDHWF7Dun66n6pZ11H9sG2I0htKopvr2jZS9c1XSufoqKxU3rmHTv96CpaUUMRrD1FjEyi/sYO3X\n340gk37flo3ACDAeS1yECiG3m7CZOeC0ggH5NOfMdN6Lk/tnhIgaBHOCl9b4eVTEIldmzgXkPh/n\nEmV/exuln72Z4rveRendt2DcvnKpTUpirnORal3MZm1kWxcTtl1Ia+NCWRcS+UEmg9bdVbTtrebp\nh8t49L4SnnmkjLY3qzm1q4rPfMyMbBa/wrFYBG9gBL3ajkoxJeLJBDkWfQWRaAhfYP4exQPOI0Rj\nYcrtUwUkNCoTzvF2gmFP8vGjxzje+UdApKH8Kq5Y87esq7sds7583rakY3j46KSQYzbXoFIZKCpa\nPWXTwIF5hYrni7KyLajV5rT7pwt7S5kvsqxsS9p9dlsjZvPUw7h0dsZi8WiAmeHUM1GpjFRVXZ6T\nXUPDRxM8K1tW3TEZMr1QBAIuDh3+JdFoGKuljuammzMeP9FvyNz32fR7AputAQCvdzDLkdmJRqcK\nkjU13jiZGzE7QlqRVSbLLL5OR6GcejgpVZOWkLhwEGRyTEUrltqMZUcsHCMairH2xmr2PXSG4TNj\nqPXSPUWuuE8M0PfMCU79+BVeuf1eXnjnT3j1jnvZ+e6f0vpfrwFQuL0W67rka1GlSYO22Mgrt99L\n0OlDU2ig8a8uY//f/YGXbvlvTv/8dQCKLq9POleuUbLyCzs49Z+vsPOGn/Hie37Gse89jxiNUXz5\nCipulDz1l9WnWCEkel6EcsxrlspTazrp8i9O4J4ppMjMCV5ZqbyWzndmzgXkPh/nCr3/+jD2916G\n8aKVCHIZYccYnj1L7/Eyk7nORap1YZxRATrT2si2LuDCWxsXwrqQyA9yOfz65yXUVk99ZkRx0gmR\nuhol//GdAnZcquXOTw8QzbEOSHv/y6ypvZVNDR+ic2g3MTFKuX0DaqWR1t7nU3o3pmMiZ9xMItEQ\ng6PHKLOvx2KoxOXppsTaQl+K8OgJehz76Xceocy+joqCTRRbV1JsXcnZgdc53ft8zjblSjQaZmjo\nEGVlWxEEAbutMcHLbimrR09HLlexYf1dHD/x28kQ1QkK7M2sWnU7AGNjPTidrUthIhC302SqzGgj\nZLZzbDwe1qXXFVFasjGlEGm11NHcfGvOImEsFqHtzFO0rLoDAJ2ukC2bP8vZjhcYGjqaFKqtUhkw\nm6spsDdz4uSjOb1HKsbHezl+/GHWrPkQpaWb8PkcdHa9nPLYiX5D+r7Ptt8TTIiXdXXXcubMM/h8\njqSK3rPB5x9Bp7Wj0VjZuuVztJ99jqGhI0nVp+VyNUZjGTbbCoqL4jkTZ342AMrLtlFUtJaBwYM4\nna34/akfcJhMlTQ33ZrRNrlcSWPDjZjN1fj9I5xqfSIhj2Vd7TUUFrUQCftpO/MUbnd45ZvbAAAg\nAElEQVQ8NUlJ8Xqqqi4DoKvr1cmQd4kLD4Vaz+ZbvoHX2cORZ3+AuaSRspU70FsrkMkV+MeGGGrf\nw1Db7rw+gDLYqyio3oCxsA6NsQCZXEk0FMA/NshI10EGz+xGnFFEa6IK9O7ffCltu5Vr30X5qqvx\njHRx9Lkf5cXWxks+gq1ybcI2v3uAQ09ly+MoUFC9noKaTeit5SjUeqLhIOHAGD5XP86eo4z2HiMW\njX8/aYwFlK+6Gp2lDJ25BOHthxFr35Xc374TL9J16Mm89O9cwusMsumOOqq3FvL0Px9EppAhUywr\n369lz9HvPJu0LRaO0vHrtyi+ogFzczGWllJGDyVHAHT+7gDhsQDDb7RTccNq3McHcOyNR2gMvd5O\nw6cuQVeeOvdz//On6PzdVOXm3iePYaixU/2+DdTcvpHuPxzOUw/PTZaNwCggYBOKE7b5xWRPiVSM\nzyhooZkpMIqpK0hPnT+acL5eMKMVpi6SzscquZlINReQ+3wsBaFeBydv/fqszhnffYLx3ScWxqA8\nMZ+5SLUupodHQ+a1kW1dwIW1Ns7FdSGxdHzyw2ZuebeeH/7cxUOPejh5OoTPL2LQy2huUHLnrUb+\n8uMmbr1ezyc/bOZn/+vOqd1+5xF0Gjv1pVewqvo9AHj9wxzreJzekVQ31qk9hQRBhkKuTuuB3OPY\nT5l9PWX29bg83RRbW3iz9X8z2haNheke3kf38D5sxlpW19xMbckljIy14RzvyKl/s6Gv/y3KyrYC\nUFS8drJ6sdvdhdc3lFMbKpUBjcaKQqGJ/5NrUCg1WC1TT621ugKqq64gEg0QicT/RSNBwmFfxvfx\n+YYJBNzYbCvYvOkv8flH8PmGQRTR64vRauP5cCORACdOPrpkXl0TdmayMRc7h4eP4Q+MotVYWbny\nNsrLtzPu6SMSCaBS6jGbq9DpCgHo7HqFAnvTZH7FTAwOHkKnK6S25iogXvm5uekWmptuJhBwE4n4\nEWRyVEojSqV28rz5CIwAw46pIkf19dfi8zsYHj6Wsd9AUt8LC1bOqd8Q9xYtK9uC3daYMoxdFGNE\nIn483kEcw8fp69+X4Kk4k8OH72f9+rvQqM2oVIbkcRTkKJXa3L0bBQGzuQqzuQqIF+XxeocJR3yI\nYgyFQoNeV5jgxTtTzJygrGwrpaWbANDpCqivv26yGI3VWkdNzY7JY1c238buPf+OUqmjufm9k56U\nzc3vZcR5asmKJUksD7SmIopXXETt5luJRSOE/WMIaj16azm1m27FXNxA62v35+W9DPYqVl/z+cnX\nsUiIkM+NUmvEWFiLsbAWS2kzJ1/5n4TzBk6/zgp7FYJMniQ+AiAIFNZsBmCofU9ebAVw9hwhHPSi\nVOvRmovRmnL7LqrfdgeFtXF7ouEAIZ8LhUqH1hRvw1axhv1PfGtSYFSq9aj1NqLhAJ6RLoyF8fQQ\nXmc30Ujid1TAc2E5K0zwzD8fYttHG3j6WweJhmKUtlg5tbNvqc06b/D1uDA3F6MwqNPuBwg64vdx\n4+1TjjXhsfjvlEKrSnnu8Gtnkrb1P3eS6vdtQFtmRltiwj9wYaQQS0VeBEaNoKNCvoLeaPucb7Yb\nFOsScrvFiDESG8jpXHFGQQstU0+IY0QZSxHWOJ0YMcZjTsyyAgB0gjEhr9q5VMRiIeYCZjcfEnGW\nei5SrQvDNA/GbGsj27qAc2dtLPVcSFx4fOLO+A36l76WeOHs8cbYdzDIvoNBDh0P8ov/V8Qn7jTm\nLDAatEXUFF/Mmf6XOdP3UtbjI7HUYoNJV4YgyBDTeDy6PN14Aw6Kras4O/AaZn15QihoNpzjZ2nr\n28nqmpsx6ysWRGAcG+vG6x1Ery9ODDWeRXGXyspLqK66IuMxOq2d+vrrkrYHg2O8/sZ3054nkyk4\ncvRBWla9n4KCZnRaOzqtPem4g4fuzUsI7FyZsPOKy7+e1sZA0M3Ro7/KaGcsFuHw4V+ybevdAJhM\nFZOi7wSiKHK24wU6OnaiVGhyFtrOnn0er3eQhhXXo1ZP/A4Jb1eWzn916Ql6e/dQXr4NEFi18nb2\nB36elAtxot/r131sUkib2fe59ntsrJuOjp3U1r4j5X5BkKFU6rFa6rBa6qiouJiDh+5N60no8zl4\n880f09x0E4WFLcSLvWQex3DYl1YUnLldqdRjsaTPwer1DnI0TQXrCRF2Av201zP3ad/+jGo01oQw\nbZlMjkZjIxyeX75KiXMbmUJFzaZb6DzwBAOn30CMRREEGYW1m6ndchu2ivyFMHpGuhju2IdvtJeR\n7iOEfHHRQpDJKW26nKp112MpW4mpqJ6xoSlRYqTrINXrb8BWsZqRruToAEtJEyqdmVgkxEhn/rxy\nHZ0HcHTGPa+K6rdRt+V9Wc/RW8sorN1MLBbh1Cv34h44zcTDS6XGgKV0JUqNgXBgqijcuKOT4zt/\nCoBcoWbLbd8G4Myeh/G5petmgL6jozz25b2Tr/uPjdJ/LLNmIZGIXKOk5OpG7Juq0FfbUJk1yLUq\nZCo5MmXmFB4Rb/z6OBaKXwdHPFMRaROFWgR56nyKvr7ka3Zv19Tvrq7cIgmM80WOgjr5aurkqzkV\n2c9IrB+PmNvNklYw0KBYR4kssdjGQLSTCLmHgbhjI5jkcSFFJUwp1WMxJyLZ3eBdogMzcSFFhoxi\neSUAEcJ5KWIhICBHkVA8ZgIZMlSCmogYJpaDrZmYPhfO2CDDsd6c5yPdXMDs50NieczFzHVhF6Yq\ngeayNtKtC1j4tSGtC4lzmaYVKsbGM39uH3hknO9/o4DmhtRPSFNhM9YglykJBF1vC4SZ32Pc1w+A\nUqEjHPEBoJCraSxPLVhMp9exn8aKa6krzZw3zmKoxO3tSfJsm8jBmCpvY77o699Hw4rrJ19HoyEG\nh5ZHaIpSqSMaDXH4yC8pKFhJaclGTKYKlEoD0WgQn8/BsOMYY2NLWzFyys4H0trY27sno2fcBF7v\nIGfan5302pPLVUQiAYLBMUZHz9Df/xYeb/zm0u3umvRAzYWhoSM4HMcpKlyDzdaAyVSJSqVHLlcj\nilHCYR9+vxP3WDcOR36iE1pP/xGNxord3ohcrmTd2o/w5r6fEAwm/nZ4vYPs2ftDysu3J/V9YODA\nnPptNlfT1HQTBn0JsViYQMBNNDotJYcgIJer0ahNyN4u7qTV2liz5kO8+eaP0343hMNejhx9CL2+\nmJLidVgsdWi1VhQKHSASiQTw+52Mj/fiHG3D6Tyd9uFCf/9bjI/3UliwCpOpCq3WhkplQC5XAgLR\naIhAwMW4p4/h4eOMjJxI6wEbCLhmvB5N+TcwOf7B4BiiKE4WjRHFWNLcSFyYjHQeoP/Uq5OvRTHG\nUPtetOYSSptmlws1G2d2J4vmYixK34kXsVWuxWCrxGCvThAYxViUwbZdFDdcklJgLKyL58V1dB0k\nmqVo1kKj1MQf6oT947gHElNkhAMehs++uRRmnZcUNZoZapW+w3LBsrqU9d+8AZVVhxgT8bQ7cB0f\nIDIeIBqIULC1Gl2lNe35YiTxAbsYzT2KJBZOfjgfDUTiursAcs2yCRJeEvLe+ybFRiCel8wjuvCI\nLgKijwhhomIEmSBDgQqdYMAsK5j0rppOmBCt0QNJ2zORLlQzW3j0BO7YCEwTuifEjlRFHdJhFCxU\nyBtQCioUKFGgRCPoUQjKlMLiBDrByJWq9wJxj6iIGCZCiAhhImKYgOgjTJBTkdklgLfJirHJ4k/J\nQ2KQgVjHrOcC5jYfEonMnIuZa0MktiBzMXNdTP8c5rI20q0LyH1tpFoX8TWhyrg2sq2Lif9nuzay\nzcXMtVEpT135TloXEpkIhUVGRzPnQ4zFYHQ0it2We6GEwdETVBVto6XmJlpqbprcHgx7cHt7ON37\nAt7AVJhHIDTGgPMoF6/6SxxjbcgEOTZjDZFoEG/AgVad/uKrd+QQK8qvptS2hkAo/QXvhvoPACIu\nby+BkBtBEDDpyjDpSvEGHAyOJoeV5ovu7tfp7n59zuefOfMMZ848k0eLppBNq+jtcJzIm+iVbybs\nzJeNnZ0v0dn5Utbj+gf2z7qwTSwWZWDw4Lzy7I2NdbPzxXtyOlYUYxw6/L85HRuJBHLqey79tlrq\nWLfu48hkckKhcV57/btkSndgtzfTsuoO5HIlBn0JVmt9QgXuVEyIwfPF4xnA45m/R1Jv7x6slrrJ\nHIzt7VO5W53O0/QP7KeosIVw2E9r6xMAhELjnGl/etILubPrZUIhKW2JBDh7jqbc7ujcT2nT5aj1\nNoLehY/CCYw7MNgqUag0SfsG23ZRsfpadOaSBK8+hUqHtbwFgOE8hkfPFa+zh1gkhFpvpW7r7fQe\ne46gV/K0Wwiuu2cdD3zslaU2Y9kj1ygmxUXnwR6OfPNpgiPehGPW/dP1GQXG+b1/8r2rXKuMBwUA\nEf+F7XyyYPKqSlBjE4qxkVsYyARecYzXQ3+a9fv1Rc+yUrFlMl/cBB3R3Ip2DMQ6aRI3oZ4RAtkd\nzXyBNh2dYKJS3pDz8amY8NpS8bYX5tsf1BixWQuM01EJaqrkTbM6xyuOcST8xgWVZ28xmMvamOtc\npFsXkNvaSLcuIPe1sZDrAua3NhZzLiQuLO781CBP/ro04zH/8DdWaquV3Pih/pzaXFt3G8WWlbxx\n/KcJIuIE9WVXcknLX3Og7dcMu6e8DA6fnVsuunDEx5DrJCXWFt46/UDa41489K9zal9CQiIZtdrE\nhg1/AUA0GuS117+T8XhRjOFwHKer6+XJUGqLpSarwLjciET8HDx0b8p9oihy4sTvOHHid0n7urpe\npavr1RRnSVzI+MdS58f1OnsAEZ2lNC8Co7GghsZLP4pSYwRRxOvqJ+hxEAn5iUXDGGwTqRKSQy0n\nQopbrvkc+x//JtFwPOXAyh2fQSZT0HngCcYdnfO2cb6Egx72/u4elBojJY2Xsfqaz8f7C0QjQQZP\nv07PseeJRbJ7uS8FRaUKfvjrMsqrk0WhzrYQf/OhfhyD2VPAfPwLNj5+9+wFq8trpzxXv/LWTYwP\np045AWAsTL7fyoXmtWo+dreNrZdrUShSh/WmIxIROX4wyIt/8vD8Ex7cWR6OLwd7CrbVTnouvvWl\nxxAjiR77ugoLRZfUzasPmaj78BYO/mOiXtVw18VAPHzauT+5KNqFRF4ERp84zoHwy1TJG7HLMt9Q\npcMveuiOnqYzempO54vEGBdHMQtTuYN8ooeQmH4Rz8QtOigSEvMFnWsiwvS5sMlKEFL8oGVj+lzk\nEl4ukZrlMBep1kXcttzXRqp1AefW2lgOcyFxYfHcyz7++u+Hue8/injiKS/tnRGCIRG9TqCxXsl7\n32Pgpnfq+cb3nBw/FaKxXjlZYXo6p9qmnoLaDDX4Q66U4iIwGT6pkM/tAnUmKoWOIkszLk8X3sCF\nmYRdQmKxUSp1k38Hg7mnIdHqCib/lrz4JC50ZIrUqUfkCjUg5EUMk8mVNF76MZQaA2NDZzi961eE\n/YlrtvHSj6IxFqZpIZ7D0WCvorB2MwOtr6G3lqG3liHGogx3vDVvG/NJODBO9+E/03PkacwljRTV\nb8NWvpqylVdhKm7g2PM/Tl2wZgkpKY+Li6WVyeLi6WNBvviRflzOxbN5qHWMez/wYtr9dz28I+2+\nVKg1Al/8ViHX3WpMeQ2ZCwqFwNrNGtZu1vDX/2jnHz49wK6dvjm1tVj2eN4uyCLIBCpvXkvX76ai\nGmwbK1n1pauZswE5UHRpPXUf2UrHw/sRw1HKrltJ5S3x6uydj8zdIex8IS8Co4jIcKyX4VgvBbIy\nLLICdIIRLXrUgg45cmSCHBlyYkSJiBHCBPGKbsZjozjFQVyx1DdMs2FPaH5hTgfD83NJHox18Wzw\noXm1MV+mz4UCJRZZ4eR8WITCBZ+L3ugZeqPJlZXmSkf0BB3R5RlWlo1MczFzbQDSulhAZjMXM9fG\nQLRDWhcSs2bPMxU0rVCi18n40G3pK7J+7cs2vvbl1CkRAJSlU5+bcDSAVmVGozIRCCXexKiVBioK\n4hVY3d785PRrKH8HMkHO2cE38tKehIREdkKhqTAvjcaKUqnNWhVZry+mqLBl8vXoaPuC2SchcS6g\nNRbgdSZ7EWlNRQAEPPO/vjYV1aPUxAuLtu3+dZK4CEx6+qVjoPU1Vlx0J0X12xlofY2CmvjvuLPn\nKJGgN+O5S4UoxnD1n8TVfxK9rYJVOz6DwVaJvWo9jmUkipZWKvnhr8soKU+WO47tD/Dlj/fjGVtc\nh4Fd97Vm3O/syn3OTRYZ/3JvKS0b8vNQGSAShsNv5u6ctVT2eLtH6X/uJKXXNNP82Suovm0DoVEf\n2hITKqsO17F+ep44QuNnLs2bLdM5c99uVnziIuo+vJVYMDJZqXrwlTZ6njiyIO95LpH3EGlHrA9H\nrI+qe+4gtq0JT0zk2C3fzFv7cr2G2m9/FFWZjcEHX2Tkid1Zz1EVWWj878+n3Df4yxcYfjR7/qb5\ntlF1zx2YtjUh5nk8MhEhPDkf5zItj/1fBJnA2J5TdP3zw+ekHefLXJxrZFu3bTms/Xyw+vGvLura\nl1h6Nq5VZz9olnQMvkFL9XvYvvLTDI4eJxgeRyFXo1PbKDCtQCZTcKb/ZXzBuXsXr6q+gVgsiklX\nisVQyaDrBMOuuUUWSEhIzJ5QaJxxTx9GQxkymYKWVe/n4KH7Uh4rkykoKd5Aff07J3NpOkfblrQq\nuYTEcsBWuW6yWvJ0CmritQLykUNw0ktSFBMqKE+gMRZgsFdlbGOk+xDV629AZy5Bby3DXrUBgKFl\nkHsxF7zOHsaHz2IpW4lGn/phaSw6FYmhUBsWxa6yqri4WFyWLHXsf8PPVz45QMA3O3HRNRJleCCC\n2SpHpZ6bd9yJZzNXuJ9eVTobf/8vRRnFvHBIpL8ngtsZJRgUKSyWU1yuRKNNb/vOP3nwZilQOF97\nhvojaLUCBpNsXvYc/e5zuI72U359C7pyCyqLFl+vm87fHaTzkf3oq9M/vJ8vfU8fx31qkJo7NmFZ\nVcJY6xC9fz5GzxNHJitQX8iccyVuNPWlaGrj+dIsl7fkJDDGwhGC3cPITTrkRh2CbPZfCvloQ0JC\nYnGR1q3EUlHa0pH3Nnsd+wlFvFQVbqXE2oJCoUGMRQmExxlynaTHsR/n+Nl5vYfNUINGZSYU8dI5\nuIu2vvShPBISEgtDW9ufWb/uEwiCDJutgfXrPo7P5yAaDSLIFCgVWrRaOyZTBTLZ1KW83+/k5Im5\n5VyVkDifsFWsprzlGvpPvhQXuASBwprNFDdckrf38Lvezp8sCBSvuJiB1qlcoObiBmq33JY1JY8Y\nizJ4ZjcVq6+lbOVVqLQmgl4n7oHlk0O1qH47So2B0d7j+Nz98HYleEEmx1regqmoHgCvK7V4Joox\n/GNDaE1FlK28Eu9oz2S+yYl28h1a/R8Pl1FYkixzvLHTx1f/aoBQcPYi0GMPuHnsgXjBO61Ohtkm\nw2KVY7bJMVvllFQo+Isvzk/UyrWK9KXX6LnsWn3S9lgMnn50nD8/MsaxA0GiMyojy2Rx8bVpjZrN\nl+rYcpmWotKpcXr8V3OrYL0U9ojRGN2PH6b78cMp94+3DfPslT9M2v7qHYm5fs8+tI+zD+1L2BZ2\n+1OeO2m3SoFjdweO3R1pj7mQOecExsCZfgIdg6iKrYw+l1sF18ioh9Of/Wn8hQD6lhpqv/2RWb1v\nPtqQkJBYXKR1K7FUOBYop8+w69SCehS+duzHC9a2hIREboyOtnPkyIM0N9+KSmXAZmvAZktfLE0U\nYwwMHqSt7SnC4eUZVikhsZj0HHmGyjXXUd5yNWGfG7lKh0KlBcDZnSxIFNRswmCrRK7SoFBq0Vni\nNQX01gqaL7+LSDhANOwnGg7QffjpuGg2Poyjcz8F1Rup2XgTpU2XEQ54UOutKDVGxh2dDLXtomr9\nDRltHWzbRfmqq7FXrQNgqP1N0lWNT2VrNjsnsJa3YC5egVypQa7UojXFc0Oq9TaaL7+LoM9FNByg\n78ROIqGptAxqvYXyVe+gcs07iUUjhANjIMhQqvXI5HHP6ZGug4z2pk/d03v8BVZs/wCW0mY23fJ1\nwj43gkyBQqOn+/DT9J98KeMYzYbKWmVKcfHFJz188wtDRCLz9zDz+2L4fTEGeqaKw9gK5fMWGHOt\nIn3D+01J26JRkf/zyQF2v5g+f2IsBj0dYXo6wrzwx3iu3pXr1Fxzs5HqeiUnDgXnZPdys2fBkfxV\nMnLOCYxRb4C2u/9r7g2IEAvMM7FvPtqQkJBYXKR1KyEhISFxjuAYOcmu3f9GcdFaqqovR60yIJOp\nEMUI0WiIYHAMr28Yt6uDYccJgsG5eZ5ISJyPDHfswzPaQ1nzFeitFQgyOd7RXoba9zDUlhz9VlC1\nHkvZyqTtCpU2aXvP0WcRo3Hh7szu3zA+3EFR/TY0BjtKtYGAZ4SB1lfpP/kKWnNRVlvDgXFGug5S\nULMJUYwxfDZzmGwqW7PZCWApbaZ4xUVJ7ckUqoRzB9t2JQiMw+1vIpOrMBbWojHYUeksIIpEgl7G\nRtsY7niLka5DZBJFHR1vEQ0HKG26HL21HJXeSjQcwOfqxz+Wv5QOVfVKfvhQedL2P/92nH/9yhCx\nJa7RWNxkzrhfb8+eXkcQYPMl2qTtf/rNeEYxLx0nDgXnJeQtN3sklp5zTmCUkJCQkJCQkFiOjI11\ns/PFe5bajKycK3YuBSXF61nV/D7Odu7kbMcLsz6/omw7K+rfRevpP9I3sC/lMQX2Ztau/nDCtqPH\nf83Q8NGEbdFokL7+N+nrfzOp/Z7uXWnbX2xy6bOExGIikylw9Z3A1ZdbQbyTr/zPnN5HFGMMtr3B\nYFvqYmje0T52/+ZLWduJvR0i7O4/RciX+WHBXG09u+9Rzu6bfQqFgGeEzgNPzOk9pzPae4zR3mNz\nPl/M4nhY06Dihw+VYS2QJ2x/9H43P/qGI+v5i8HHH7qS8eH0RVSMhdkLpKTLAbn/jczFwBaK5WaP\nxNKz8ALjcljNywlpPCQkLkyktS8hISFxQRATo8TESNr97rEujhz7FUqlnpKidVgstXltfylYjjZJ\nXMAI504Mo0yunBYefW4Ud1kKMhVlqW1U8YOHyrDaE8XFB38yys+/N/fCd/lmqHWMez+QPrf1XQ/v\nyNqGVi9LvWOJPvLLzZ5zAVN5E3VXfYSx3lbad96/1ObknYUXGKVKOolI4zF7RJFl8S21XOyQODeR\n1r6EhITEeU9P3256+jIXIAyHfQw7jgOg0xXMSmDMpf3FZjnaJCFxrlDccDFyhZqg18lo7/GlNmfZ\n4vWkvo6ub46Li2Zrorj48+85efAn868Wnk923deacb+zK3sOXXeaHN9bL9Py4pOeOdk1H5abPecC\npvJG5EoNlqqWpTZlQVhwgVGMxD90qhIrBTdfhGHjCpQ2I7FACH97P6PPH8T9ytG05wtyGS2//8eU\n+5zPvEXfT55cELsXivmOxwTa+lJs79yEbnU1Sls8sWp4ZAzvoXZG/riXYN9IxvM11UWYLl6JfnU1\n6spC5AYtYiRK1O3Df3YA92vHcL96NG06DVWRhcb//jwAJz78faJjPhRmPbbrt2Da3oyq0AwygbBz\nHN+JHpxP7sV/pj9lW+ZLW7BeuwFtXSkyjYrwyBie/W0M//4NwsNuxHAEQa7KOibzZb52zGdMK7/8\nXsyXtjDy5Jv0//wpFFYDBTddhHFLA8qC+FhGHGN4Dp/F8fguQn3Zn8apKwqwXrMBfUs1yiILcoOG\nmC9IxO0l2DuC52A743tOER4ZS3m+wqzHfsNWDJtWoCqxIlMribh9+E/34nrxMGN7TmZKt5JX5tuX\n+bD68a8CcOzWbyXkskl3nPu1Y3R/LzkEJV9rX0JCIjNynZ7Cy67DsGIVCoOJaCBAyDmM+/BeXIfi\n3iEKgwn7RVcz+NxjSefXfPTzKPRG2n7yz4BIybveh6aojL4//ZqSa25GW1mHGA7j7+2g+7fJoWq6\nqnrsF12FtqwamVpN1O8j0N/NwDO/J+xePp4UEhISEhJLi6W0mco17wSg+8gzCUVZJBLxeZLHpqFF\nzf97oBTTDHHxh1938Oj9yy8v7YlnU1fanuCxL2fOvwng88boOhOmql6ZsP1dt5l4+RnvnPIezofl\nZs9Cka2y9GwY622loGk7ro4jeWlvubHgAmMsEMKwrpaq/3MHMu2UOCNXajGsq8Owrg7zRSvp/v6j\nqW/eRZFA5xAKkw65SYcgT+OGe44w3/EQZAIln7gW+w3bkpzp1OV21OV2rNdtYuhXLzL86OspbSj/\n6xuwXrsxuW25DFmRGWWRGdO2JqxXraPz279BDGeuhqq0G1EVmqn6h/ejtBsTbSqzoy6z4345eQEJ\nchkVX7wV8yWrErarSqzY3r0Fy5Vr6fzOI0TGfKg0Cycw5sOOfI2puqIAbUM5NV+7E7kxMWGuqsyG\nrcyGZcdaur7zCJ4DZ1L3RyGn9K7rsL1rc9JnRG7UIjdqUVcUYNrWROeQK6UoZ7poJRV335TwGYX4\nXCvtzZi2N+M5cIbu7z1K1Js+l8h8yUdflgvz/i6UkJDIiYpbP4bKXoxzz0uEx10oDCb01StQGKeS\nq0c8Y5hXb2Jo5x8Ro1NhnUqLHW15DcOvPM30JyjqwhKq7/xLvJ1tDD7zexQmC/btOxBkcsTY1Pe5\nefVmyt7zAUIuJ869LxF2u1BabOiq6ol4lt/NzvlOaclGVja9lzf3/wSVykBt1Q4MhhKi0TAjzlO0\ntT9DKDSe8lwxFmXj+k9hNJSCIODx9NPR+RIjzuQq7kZjOVs2/lXCtlOnH6e3L/vNYi7Mtv10/R4a\nPpqyz3MZp9naNJHbct+Bn7Gi7l05jSuAWm2mtvoq7LZGlCo9odA4DscJWtv+lHasqisvx2SsQKUy\nEIkECARdjIycordvDyGpyrbEMmDljk+jMdhR6+NVh0e6DuLoeGuJrVreeGcIjP5UJT0AACAASURB\nVE1r1Pz7A2UYzYnaQCzKshQX0yGTJ97gxKLZvTdeesrDRz5rndEOfOe/S/jNf7v51U9H8Ywt3r3E\ncrNnuTPWe4qDD/zDUpuxYCy4wChTK6n8u/cRC0VwPPYGvtZexHAUTXUhBTdfHBdeLl5Jyceuof9/\nnkk6X4yJtH3+Z5OvVcUWGn/++YU2e8GY73iUfubd2K7bBECwx4HzqX0EuoYRBAF1dRH267egKrFS\n/JGrQSZj+LevJrXhPdGN9dqNBHtHGH+zFX9bHxGnB5lOjaamCPv1W1FYDRg21FNw4/a0QuUE2hVl\nFH9wB3KTDverx/Ae7yLm8aOwGdE2lKNvqcJ7rDPpvOKPXD0p6oWH3Tj+sItAxxAytQLdykps12+l\n+p47FlxsyYcd+RpTbX0p1ffcjtygiXsJ7m0lOuZFWWTBetU69GtqkKmVVPzNLbR+8ofEguHEBgSo\n+vvbMG5tAkCMxhjbdRLfiS4iYz7kBi2qEivGDfXI9Go8+9uSbDBubqDq728DQSDiHGfkyTcJ9jiI\n+UOoSixYrl6PrqkCw4Z6qr7yPjq+9iDiQoT/5qEvy4n5rn0JCYnsCAoFuqo6HK8/z8junZPbnXte\nSjpWrtVhbFrD2PEDk9vMqzeBKOI+nCiSyFRqXIf2MPjcHya3xUIBtOXV+LrbJ48pue5WQs5hzt77\n78TCUtX65UJN1RUU2FfiHG2jf2A/BkMpJcUbMJuqeXP/fxKJJD8os1jqAJH+gf1otVbstkbWrfkw\nBw7fx+ho4gM+n8/B4WMPolLoKShYSYG9Oa/2z7X9mf0uL9uWsc+zGae52lRVcSm5jqteV8jG9Z9E\nodQx6mzDPzKKwVBCRflFnO3cSTic6BljtdSxfu3HiUZDjDhbCYd9qNVGTKZKqquvlMK4JZYNkaAX\nubWcwLiD4bNv0ncifV4+iTje8an7sNpGVUpxMRIR+eYXhhbbtDmx+oYqrv3KWjTGKc+/0W4vP73h\n2aznPvq/bm79iBmDKbH/crnABz9j4eYPmnji12M89ks3A70Lnxt3udkjsbQsvMCoUxNx+zjzxf8m\nPDz1NMF7tAPXy0ep//5dqEpt2N+zlZGn3swa+hn1n9sX7PMZD8OG+klx0XPgDJ3f+s1k2CWA51A7\no8+8RfXX7kTfUk3xB6/Es78tKTTZ/fIRQr0j+E71JNk3/mYr7leOsuKHn0GmVWG+fHVWgbHs0+8m\n4vbS9oX/Itg1nLRfkAlJApSqxIr9xu0AhIZcnPniL4iOTV0ojr/VhuulI9R//y+QG7JX1Jor+bIj\nX2MqN2hA1ND13UcY231y2p5OXDsPUfWV92G6aCUKsw7TtiZcM0Jqbe/cPCnIBXscdH7z14QGkvOP\nDAAyrSppXmQaFRV33wSCgL+tj7P33J8oYh4C5zP7qbj7JixXrUO/thbzFWtwvXg47djMlfn2ZbmR\n7+9CCQmJZMRIhNDIMJZ12wgM9jJ+6iikCTkLu5xY1m9PEhi9Ha2Ex1xJx7sO7Ep47e/rRmWxw9sC\no76mEZlaw8jOP0ri4jKjsGAVh478khHnVP6rpoYbKS/bRnXVFZxpT36oY7PWs/PlKQ+D8rJtNDXc\nSFXFpUlCWDQaxOGIV6tVKLV5Fxjn2n6qfmfq82zGaa42FRWuznlcV628HaVSx8Ej9+N0np7cXltz\nNfV113HyVGKKg4qy7QiCjAOH72V8fHooooBOa08SJCXOLyJBb04Vm5cDp994cKlNOOeYCJEuKlXw\n/ftLk8TFUFDkq381wBs7z411ftlnmvnlR17mnf+wnse+vJeNt9cSDmaOGpxgdCTK9+4Z5uv/UZyy\nnpHeKOMDn7Jwx19Y2Puyjz89MsYrTy+c9/Zs7PnqXw/wxvM+IpHlfd8mMXcWJd545I+7E26oJ4h6\n/Aw++PYTG0HAetW6xTBnyZnreBTcfBEQ9+Tq/Y8nEsTFCWLBML0/eiJekEQQKLzt0qRjxGgspRA2\nQWjQhffIWSAespsNQSmn+99+n1JcBFIKP5ar1iHI4t9AQw+9lCDqTRDscTDyZH7Ci9KRLzvyOaau\nV4/OEBenGH70jcm/tQ3liTsFKLz1krg9kSid3/5NSkFuglgKsd76jvXITToAen/8x2QPybcZ+N/n\np865en3a95gzeejLckT6LpSQWHh6Hr2P8LiLils/RsPnvkrRldejMJiSjnMd3I2+ZgVKSzxETVtW\nhcpWOJmncSZhd+J3kBiNICimntMqrXYAQo7BfHVFIk+43F0JohlAe8cLiIiUFKf+DfP5E3NZ9/fv\nQ0TEaCxPefxyZGa/s/V5LuM0W3IdV4u5BqOhjGHH8QRxEaCr6xVKitYhExLzrk3e2YozrztFfH5H\nPsyXkJBYAsJhkXBYxGiW8f37SyksSfaR+vrnBs8ZcRFAoZLhaB8HAbzOIK/+7CRrb6zO+fwXn/Tw\n3b8bIhxKL9TJZLB9h45v/bSEu79ewIqVC5dyLFd7vvXTEv6wt3rB7Vlu1F31UTZ94t8S/q265cs5\nnWtv2ELT9Z9l/Ye/ndTGxL/lxMJXkQY8+1PnioO4d5cYExFkArpVVYthzpIzl/GQqZXo19QA4D3S\nQXgkdc4ggNDAKL4T3ehWVWHYuAJBLpt1mHHYEc9lJyjkCEp5xjyMgbMD+I53zap9fcvbfRNFxvak\nzrsD8fFIJZLmi8W0I9cxTZWvcoJg79QF8oQQOIGmqghlUTzH2NjuU3PygDNuaQQgMuohcDb9TXLE\n7SUy6kFhNaBtzP/NVj76shyRvgslLkwENMVl6CrqMTWvR2kwIdfqEORKBJmMWCSMGAkTDfiJeNyE\nx10EHYMEHf34B7qJeGaXWzXoGKTjvh+gq16Bdf12bNuuxLrlMnof+yWetqkKna5Deyi4/Dos67Yx\n/PJTmFo2EQ34416PKcjmlSi8nShWStK//Bj3JCfWD4e9+H0OdLpCVCoDoVBitUuXqz3hdUyMEgn7\nUMjVC2prPpnZ72x9nss4zZZcx9VqqYsf7+5IaiMaCyOTKdHq7Hi9U+GQQ0OHKSxYxfq1H6e753X6\nBvbN214JCYmlx+eJoVILfOcXpdQ0pBalbrzTxBs7vcRycwJccsKBKFqLClEEe60Rz3AAtX520sxT\nvxvnxKEgX/p2IWu3ZI74e+9Hzbz3o2aO7Q/wyL1uXn7ak/exytUek1W+KPYsJ1ydR4gEPCg0BrSW\nYjSW4pzOq770DgoatyJGI4z1nyYa9KMvrEJtijss+Zy9ePrbs7SyuCyKwBjsTy8OxIJhwoOjqEpt\nqMuze8udD8xlPNSVhZMFbgLtA1nfw392EN2qKmQaJerKQgIdM8QiQUC/pgbjhno0NUUobEbkRi0y\ntRKZSomgUiQcm/G9TvdltWcm6opCAMLDY8R8wbTHBXsyV8OeL3m1I09jmknYm26joEh0QNbUlUz+\n7T3Wkd3eFGhXlAKgsBomqyJnQ6ZWIlMr03o7zoV89GU5In0XSlxIyLV6bJsuw7J6C0qTNe1xMqUK\nlCrkWj0qa/JnPzTqYGTvTkYPzS5/mq+zDV9nG8qXnqTqzr+k5LpbaZsmMEa843haj2FesxnHq89g\nWrWesWNvJRR9mQ0hd/x3Qm0rwt/TMac2pqMuKKX+E7k93c5Ez+P3M3bq0LzbmQ3LzfZI2J9yeyjs\nRUchSqU+SYhKVQxETPKMW96k6nemPs9lnGZLruOqVscfMjbUX09D/fUp21LIE29gB4ePIG/VUF97\nLXW111BbczUOxwk6u19hbDx9lImEhMTyJhgQ+dqPilm7Ob1otf1KHZ/7vwX88Ovnhrfyrvta0ZiU\n7Ln/NHc9vAOAtx6evVDUcTrEZ2/v5bJr9Xz8C7asXoEtGzV8Y6OGng4b9/7AyQtPeJKdvufBcrNn\nueA8sx/nmf0AFDRuo/rS27OeYyxroKBxK7FohJNP/AD/aDztnSCTU3vlh7DWrCXsG6N7zx+ytLS4\nLIrAGPOnF24gHh4IINcvXK695cRcxmN6ReFIijDepDbcUxdwMz3ddM0VlH/2PagrCxO2i5EoUW+Q\nsMuDXK/JeT7mUkV4ou3IeOa+xHwLV6E4n3bkc0yjWWxJh2LaPEecc7sBkOu12Q9KgaCQQx4Fxnz0\nZTkifRdKXCjYNl5K0eXXI1PN39tLZS1AjOXqFSgwvfozxEOb/T1nMa3akHT06IE3qGr+DLatV6DQ\nG3EdmntaDl/HaWKhILatl+M+vh8xIiUyXzZkeVCaivPCE3W2/Z7DOM2WnMf1bVMcIyfThjcHU1QA\n7+t/k8GhQ5QUr6e8dCuFhS0UFq6is/vVlHknJSQklj9FpQqKSrPLFu/9qJmuM2Eee2D5V5Fuf32I\n8SE/o11efnzt06j0Cly9cw/xfvVZL68+62Xr5Tpu+7iZbVfoMn6lV9Qo+eoPirnlQ2b+5e+H6GrP\n333ccrTnXMRStRoAV8fhSXERQIxFGTj0PNaatZjKmxBksllcJy88iyIwyvUaop70Ao3cGBcTJm6u\nz3fmMh7TcwMqzPqs76GwTB0zXWys/r8fwLi5AYCuf344bVhwxRduxrJjbdb3gdQ5FrMR9QVRmHXI\nDZkFLUGlzLh/vuTDjnyP6VyrZk8XehVmXYYjM7ThDyLXqQn2ODj91z+ZUxv5IB99WUxytVH6LpQ4\n3yl5x63YNuY3rUVgsBfXkdyEP4XeQPF1t6Irr0GuNyJGI4RdTtxH3+LU9+9JOt57tpWhF/9E0VXv\nYfC5xwgMzN3LKRrwc+r7/wd9bROV7/sLtGVVCAolUb8Xf08H/U/9lqh/4ZKsS6THbmvkbMcLCdtU\nKgMWczXBoBuv9/zMmzmz39n6vJzGaXDwEGUlmwGRtjNPzercaDREb99eevvi3xulxRtZ2fxeIpEA\nnV0vL4C1EhKZkcuUXLzucxw+/Vvcnu6lNuecJRaDxx9081//6sTvi/FP/1nCFe9KvC/+m38q4MBu\nPx2nl3d+9vEhP3KljKJGM2Wrray8rpzytTb+ZfPj82p37ys+9r4ypRvI5LD1ch3f/UUpshTVN9Zs\n1vDgC1VEIiJ3f6CPI/vy69wz057tO3S85/0mLnmHfknsOZeQyeNSXSyaLLbG3o62EQQBmVxJNJbZ\niWUxWRSBUV1ekLYAhkyjQllkAeLFNC4E5jIegc4hYqEIMpUiIXw0HZq6eKhrLBRJaEffEk8eG+pz\nZsw5qCxMToifT0K9DhTmKlRFZmQaJbFA6qcUqjLbsrdjuYxpoGMqD5FuZRXOZ/bPvo32fvSra1CX\n25EbtUTHl0boykdf8o1Mq04r/E2E2mdD+i6UOJ8x1K3Mu7gI4Gk/kfOxEe84vb+/f1btCzIZYjSK\n+2jq75mBp37LwFO/Tdoe6O8m0J98o+g9ewrv2fS/BRKLj8lYQYG9GcfIVAG12pqrAYGBwYNLZ9gC\nM7Pf2fq8nMZp1NXO2HgPBfaVlBSvT3p/g6EUj6c/YZvZXM3YWHeSl6TJVAFAKIXHo4TEYhCNhXn1\nwL8vtRnnNO2nQvzrV4Y5fnBKcPrW3w5SXF5O89rEaInv/qKET9/ci3t0+Sb1q7+0mMqNBVSst6HU\nyOk/7uLA7zry/j6xKOx+0ccHruzifZ8wc9OdJpSqZFdChULgaz8q5gNXdmUs1DJfdr/oY/eLPkor\nlcvCnuWMu/s4BU3bsdaso//As4S8rsl9xS2XAzDe3040vHzERVgkgdG4uSHtTbVpW9NkFV/vLAuF\nnKvMZTzESBTP/jZM25vRr65GWWCaLBoyE1WZDV1T/GLKe/BMgkecoI574kUzhPwqrAZ0TZWz69Qs\n8R7viheyEASMW5twv5I6qb5xQ/2yt2O5jGmgvZ+Iy4vCosd08UqUD+wkPDK74gjje1vRr64BQcD2\nrs0MP/LqwhibhXz0Jd+oK+z4TqZZt9ubc2pD+i6UOG8RBIp33Jj1sPG2Y/j7OgiODBIN+ImFgwgy\nBTKlErnOgFJvQmUrRG0vRlNcgUylxtN+PGu7c0WmUmPdfBnuo/uWnXdh0NHPyR/cg0KnR66d+hd/\nbZj2tx5dRd1Sm5vAbGxX6E0pc2/mE5e7kzUtH2TEeYpAwIXRWI7ZVIXPP0Jn9yvzbt9ua0Sh0KBQ\naLCYawCwWVYAEAi6GRlpZSJ0XxDk2G0NyOVqFAoNJkO8UFqBvRmFQks0GiQc9uEcbcup/UgkQCQa\nTHiPdP0uL92asc+zGae52jQbjh1/mA3r7mJV8/uoKNuOxzuAIMjQqC0EQm5OnHw04fi1LR8CYGys\nm0DQjSAIGA1lGI3leH3DDA2lL6InISGxvHngP0cTxEWI52b8yl/0819/qKC4bErWKKtS8u2fl/A3\nd/YRDi9PceqOH19M+65Bdt3bSseeYaLhhQ1x7e8O86NvOPjNz1188VuFXHRVcvRVUamC62838ocH\nF/6ea7nZsxxxdR3H1XUUS9VqVt36Zdxdx4iGAugLq9AVVBIN+une/fulNjOJRREYbTdsxfXiYYJ9\niYUy5AYtRXdeCcTDbF07FzcJ+VIx1/EY/t3r/5+9846Oozr78DOzvWi1q94tWe4V44Jt3DBgE5pj\nOhhIIEBIIEAIISSENJKQQgghJCR8EAihJvTebYPBBWzcu2zJkqzeVqvtO/P9sVjSaqvKqph5zuFg\n3Zm5993ZuzN3fvOWoAihEsm/aQUVdz8dVolY1GnI/965nWmoGl78NGS7r6ENbbYVXUEGKrMhzCNL\n0KjI/965CBrVwHzYKLSu3k7m+QtAgOxVp+DYUhZmiyY9hfQVc4e9HcPlnMqSTNOrG8i+8lREnYai\nOy+m4tfP4m+O/MZe1GtAEJBcXSEEze9sIeP8k1Gnmsi6eBHu8nraN0X3xDGU5uKtax3wkN6B+CwD\nTcbK+Ry5579h7YaxeaSdMTOhPpRr4VeHD17M46HH7Dz/WvwcorZUkVUXpjCqQE1NXYAXXndQUTmy\n8veZSyagS49dEa9t92aqX38q8U4FAX1mLu6G+IXNeouo1WGZMB3rjLmIag2N694d8DEGAsnrxut1\nQ2vsQmOTbh9+njGJ2j5QBWFi0di4m6rqTxlVuIg021gCAS81dVsoO/QOfn//w6+mTbkCQQiN9Qrm\n/psMwOqP7uosZKJW65g25YqwPnKyZ5CTHcwT6ve7+eiTuxPuv+cYx+j5ueN95t6cp77a1Btc7mY+\n2/wgRYULycyYRE72DCQ5gNdjp+xw+G+24shaMjMnY0ktIk2lIxDw4va0cqj8faqq1xOQlJxeQ838\n6TdRfvRj0q1jybSNJxDwUlW3ibKqNXQXo4ty5lGUcxJ6XSouTyvlR9dRXb+5V33EYvq4SwgEvGSl\nT+LgkfdRq3QU5y2koXU/Ow4E13rzpt9IfdMuyqpWdx5XWnAKWemTWb/tQQAKc+ZSmD0Hg85KQPLS\n3lHD3vI36OiWN3TJrB+jUesBga37nqKhJXRdbTJkMKbwNGyWElQqLV5vO0dqN1BRE/oMpxCZ5oYA\nP7q6hr+/kI/R1HVNmjZLz+2/z+Q3t9bHOHrouP+UNymamU7pghxOvnYCPref6u3NfPS3xKM2+kJ9\njZ8ffauGH/w6kxWrwqPrFp9hHlRBb7jZM7yQOfThE4w743rMOaOxFk9HEAR8TjuN+zZQs+19vI6W\noTYyjKQLjAGHC0GtZvS936L5jc/o2FOJ7PWhK8oiY+U8tF+GBDa/vhFvbewTJIgCKmOXC7SoUfe6\neq2gEhENXdWMBF2wuq/sTfxhrj999Od8uA5UU//sR2Rduhjz9BLG/OV6mt/8DPeRBgD0xdmknzUb\nbU6wWmfjy+tx7gkN37J/spuM8+Yj6rWM+tml1D+zFm91E4JWjXFCARkr5qErClad1hcnVj69L3iq\nGml+63PSzpyFNsdG6Z+vpfGl9XiO1COoVRjG5pN+9hwEQSBgd4YVqhlOdgyXcwrB79x84hhMU0Zh\nKM1l3D9upG3dblz7qwh0uFGlGNCkpWAYm49p8igq7n4Gx7auimWS20vlH1+g+JeXI6hVjLrzYhzb\nDuHYfBBfiwNBJaJONaErzMQ0ZRTaHBsHb/pHUnIG9vez9KSvv9tj37tl7gRK7r6Clg+34atvQ5Vi\nwDSthLTTZxDocCP41DGLswzktVBh+LNonoFtO728/FYHOZkq2tol2h3hb6ezMlRseKeAwm5v3n/5\nozSu+l59QuLkcCF10okxt7dsW0/NO+FhxjGRZdz1R/thVXREnZ6cM87HXX+UI88+jK9N+c0dz4ii\nmvqGndQ3RI5S6E5t3dao4cDr1t8TsX31R3clbIvP5+TDtXcmvH9v++9Obz53b/fvrU19Oa8APr+L\nssPvRhQUe3Kkah1Hqtb1yi6FwWd88ZkcqHiHQ1UfkmouZGLJObg8rRxt+AKAwuyTKC08hb2H38De\nUY3FXMCE4rMQBRWVdZsS6iMeGdZx7D78Kg5XA2OKTqOmYRtb9j7BrElXU27Kob0j/ostm6WYcUXL\n2H7gORzOerRaM2mW0WHFh9Z8fg8qUcvSOT8N60OnTWH25Gto76hl2/5n8XjtmI1ZeH19L/bxVeTQ\nPi8/v6GO3z+ai9jNn2P5yhQqD/l44sHhd4/3Ov3Ya100ltkRBMgal8r4pXlJFxiP8cCvGpl7ijHE\n8xNgwvT+F+c7HuwZLhTMOQdzzmgOr32a5rLNQ21OQiRdYLRv3Efbut0U/ehCMi9aSKRMZfZP91D7\n+PvhxqWaKPn1lYgmHSqjPkQcALAunY516XTkgITk9BBweujYUU71X1/t3MdQmkvBrStRGXWIRn3Q\n06kbWRcvIuviRZ3VfiWXh9a1O6h/es2A9jEQ5wOg/tm1yIEAWZcuQZefTu61Z4TvJMs0vPgpdf/5\nIPz4/32MeUYp+pJsjOMLKP7Fqh7HQv3Ta3BsO8To318d0YaBouZf76DJSCFlzni0WVbyvv21kO0B\nh5uKu58h86IFpMwcO2ztGE7nVA5IVNz9DPk3nkPqwsmIOg22U6djO3V6wn107Cin/K7/UHjbeajT\nUjBPH415evTwu74WpYnHQHyWsX/7br9/t5V/epGin1yMqNNgmlaCaVpJyPZAu4uK3zxHzpVLw7Z1\np7+/fYWRx6L5ehzlo1GpQJbho/Uubr6zkV17uzxtb/+elcI8NUeq/bzxbgdjRms4fbGRR/+SxSeb\nXNTUDd/8QV0ImIrHR93q72inbvWrUbcPBf72Nvb+4UdDbYbCYDEI1ZGHJcOwirSCQnNbGVX1nwPQ\n4WrEZimmMGdOpzg4umAx5UfXUdO4rXMfo87G6IIlnQJjvD7iIcl+ahq2otemMrbodKrqP6O9oxa3\n145BZ0tIYFSJwedSj9eBy9OCy9NCW3vkAi4ykdfKBVmzkWWZrfufJhAIrg2c7the3wqR2bjWyV9+\n2cj3fxWacuNbt6ZxpMzHmreG10vbG95cRu2eVqq3t7B/TQ1rHtiFp2Pwold8XplPP+hg5RWpIe0m\ns4hWJ+D1DG5o+XCzZ7iQMe4kkGW8jmYEUYUsDf/ngqQLjM7dR3BsOcjBmx4i47yTMc8oRZOWguTx\n4iqrpeX9L6LmvRPUKnRF8YsnCCoRVYoBVYoBX4MtZJto0KEriJ/bR1CrgpVgU41oM0Mn9kD0cYz+\nnI9jNPxvHfb1e0k/ew7m6aNRZ1hAkvE12+nYXh70aqyI7A4uOT0c+tG/SF8xl9STJ6HNS0cA/HYn\nzr2VNL2+CeeeSgSNCtkfQFAnL6xX9gWo+O1zWBdNxXbaCehH5yDqtPia7LR/foDGFz/B19SOp7Ix\nqQJjf+0YTucUvvRCvPcFGl/ZgO3U6RgnFaHJTEXUaQg4XATanLjK62jftA/n3sgLoY5dFez79l+x\nLZ1OypxxGEpyOr03/W1OPFWNOHdV0PbJ7qQWJOnvZxmI361j6yEO3vJPMr4+D/PUEjQZFmRZxtdo\np/2z/TS+vB5/iwNPVWNMgXEgfvsKI4vpk7veugoCLJ5vYPXL+cw+rZKKquAicvkpwd/Viitq2Lkn\n+HDxl99k8N2rU7nqUgu/vX/4vXXvidaWjtpojrq9bc8WJO/wSkCtoKCgoDA0OFwNIX+3O2vJSpsI\ngFZjQqsxY3dUh+zT5qhmdMEp6LQpcftIBJ8/GHlzLGz+mMegJPkRhcQej5taD1DdsIU5U67F7qim\numELNY3bkKTERaIUUy72jupOcVGhf7z0nzYKSzRccFXXel4Q4M77sqit8rF3x/BZi/zl1LeG2gTc\nrnDRLhCQ8Q9R3srhZs9AYS2aTEruWFRaPSqtHn1qFgC6lDTGLLsGr6OFgNdN7fYPCXhDowJbDm8j\nfexsxp91Y0i75PfidbTQVr2P2m0f4HcPHwE9aQLjkd8+F/K3t66Vow+90as+fE12dq74Vb/s6NhZ\nPiz6GIjz0R1PVSNH//Fmn46VPD4a/vtxzAIesi/ArvN/E3W7t7613+ckOBC0rt1B69roibdrH3uP\n2sfe6/9YSbSjv+e08o8vUPnHFyJu60mi5911oBrXger4O0ZB9vppfnszzW8PvTt2Xz/LgMxRghXC\nj/499u/16D/f4ug/Iy8WutvR39++wshi9ulVlB/xkZOt5ubrUrnmcgs/vsXG9bcFH46KCjT4/XKI\nV+Mf/trKd69OZelCw4gQGHUZOTG3O8qSV6RFQUFBQWFkIfbI3SkgdKZOFEV1Z2skjgmCsfpIiB65\nQXtWHo+GStUVTScjs/fw65RXf0xe5gxKC06hOG8hn+16BK8v8Yf9/uQpVQjnwV83hgiMADq9wD2P\n5HLdiioaakdWjutkUjJWG9bW3BBASm69magMN3sGCkvBRDInzAtrF9VaUgu6Xow07F0fIjDqrdkg\niMiyhIDQVS1aAFGlRW/NRm/NxlY8jR3P3d2z+yEjqR6MVjGTItU4tvs+ibg9UyzALFg4HOh6+Bil\nmoAKNYcCO0Pa8lQl1AUqO9vj9T1QTNOczJHAflqlhojbU8UMilUT2eZLZUGSMgAAIABJREFUvNpu\nmphNs1Q3UCYOGbm/vJHW59/FtWN/Z1v2j66h5Zk38B6pSbifvhwTCeOcqaSetRhNYQ61v/hbv/tL\ndKwjV4fnVFFQ6C/Zt15Hy/9ex1uZnBx0Csmn7LCPrTuDi4FWu5fv/LCBGVN1nLa4K5er0SBgb5dC\nnnWqa/0cqfYzrjR8oTUc0VhsMbd7GpJ3LVZQUFBQGFmYDKHRaSnGnM6wYLfHjtfXQao5n6a2rkrq\nqeYC3N62zmJDsfoYKAJ+D2p1aG5tszE8l7rb28ah6jVU1HzK/BNuIjdjOhU1iT2fdrjqycucgShq\nkAa5CNGCkmuobd/LwcaBy1tamj6fvNQpfHzo4QHrs7dIEhzY7WHspNDcfelZKn73aA43XHgUt3Pk\nKlb5ozTUVPr6LbqNGqNl9iJDWPu+Xnp5Djd7hiNHPn2eI58+36tjDLYcJpxzM7IUoOy9f2E/ur9H\neLSAwZbDmNO/hdYcex0+2Ijxd0keDVJViLgIUBHYG7ZfRWAvVYGDYe0jlTGqaUNtQtKo+/0jvRb2\noh2jHZWHcdaUhPtxbtpBzc8fJNASudLwQDKYYykMLtqifIwnTk24faDHOUbdfQ8r4uIIZ9MX4Yui\nTVvc5GWHpkmI5LxQU+snzTakt+iEUZtSom6TfF78zuETtqHw1aKmdgsfrr2T8orV8Xc+jujt5/6q\nnieFoSEtdTSF2SdhMmSQlzmD7PQpnbkVQeZw9VpG5S0gN2M6JkMGuRnTKcqdx6Fu1Zxj9zEwtDqO\nkJsxnfTUUkyGDIrzFmI1F3Vuz0qbRH7WiaSYcjDobGTYxqJRGyIKnaIQvO/3rLxeWbcJUVQzfdzF\nWFOKMOrTSE8t7VW4t0I4d3yrlsa6cE/FsZN0/Oz+LMSRsbyKyLW3pfHfdaO45gdpFJZo4h8QgVFj\ntPzukRxUqnBP4Q9f792abbjZc7yQNXkRolpL3c61tFXtiZB7UcbVUkNb5fCLEkp6DkY9RqZpFmDA\nRJNcw0H/dgAKVePIV42mSarlgD9yVbm+9j1aNYV0MYcGqZryQLAS0yT1HIxCCipBTZPUte8MzRKa\npVqsYiY6DGzxrcaPj1LVVMxiKlqiV4U9hhYd0zUL0GHAjbPTq7KnHWbBSol6EhYxjRmaJQBs9a1F\n7pVP/zAiyS79xhMnIblH/lsLhZGFcfokJE94Lpxo7QM9jsLxgb1dwpoavoK1WESO1UXSaKIXVJAk\nUEdYaA1HRE10T0sl96KCgoKCQneO1KzHZhnF2FHLkCQf5TXrONrQ9Sx4pHYDkixRWnAKel0qLk8b\nB4+8T3X9loT7GAgOVa1BqzEzdeyFiIKahpa97Cl/neK8BQDIcoDCnLkYdDZEQYXL08rBI+/T0NLl\nLDOm8FRG5c5HFIPCy/RxlyLLARyuBjZs/ztuTxuf7XqUMYWnMWPCFYiiGo/XTvnR5EboHe801Pq5\n45paHnwuD70xdC224HQT19+Rzt9/23uPV51ewJQiYk4RMZqD/zdZgv/PzIksq6y8IhVHewBnu4yj\nXcLpkHC0S3Q4JJztEn5/75+ns3LVXHmjjStvtHH0iI+Na51s3eCmbJ+HqnIfkeqAaHUCU2fpOfUc\nM8tXpqDRhq8xj5T5WPt2x4i353hAa7IC4PfEriivs8SvNzDYJF1gVAtadniDVVFP0i6nRiinQ7ZT\nGdiPHy9mwTqgfWsEHVYxk89873Oi5hRa5HrapCb2+j9HIhi/vlC7goNs7+xHIhAS4mwSLGSq8tno\nfYf52rPi2qEXTGz2foiExGzNaZiFVNSCNqIdO33rsWoz+cK3ps+feyjJ+83NCHod7l1lITlDDCdM\nwPr1U9EU5lD3+0fx7C/v3KYbXYht1dkgCGhH5dHyzBu0v78+6jHqtFTSrj4P3ZgiZH8A8+LZABz9\nyf0gSZgXz8L69dOQ/X5aX/6Ajk8SqxgXCd3oQtKuPBfRYkZQqXB+toPmJ18DwLx4FpYzFyNqNQmP\npc5OJ+3KFWiyM0AKYH/nE9o/2BB3LIXBw7xgDpblixG12uD3+tp7dGzYgjrNStoV56MbPSo47xbM\nAaD+/v+L2H70F38CSUKdlU7aZSvRZGWAJGF//2Pa16xHNBrIvesW2j/8BMvyJTjWbaL15bejjnOs\nP8O0iVjPOR1Nfg519z2M52A5ANpRBaRdsgJ1mhXJ6aLlxbdw7dgTMo554RxEo7FzLIWhY8ceL0sX\nhoZ6FOapOWeZCb1OYNX5KRw+EgyHSjGLaDQCvm5JrDPTVbR3jIwQHkGMvpSQfIqIrqCgoKDQRUDy\ns/3Af2PuU1W3iaoYHomJ9BGLdVvvB8Dnd/Lehp91tn+67YHOf/sDHnYeDM+PXvOlkNnQso+Gln0x\nxzlY+QEHKz+IuY/DWcfWfU8lbLtCYuzf6eFXt9Tz63/khHksXnKtlYoyH288Z4/bz31P5jF2khZT\nioha3fsXvz2rWvfE45bpaJd47Vk7j97X3Ov+84o0rLwitbP6ss8rU1/jx2GX6GgPriN1BoGJ0/Ux\nPTcDAZl7bq/H5+2fA1E8e3QGgfQsNVm56kGxZ6Tiaj6KJX88WZMWYK/eh8ceWlRVY0wlZ9pSLPnj\nkQLDK69o8qtIy/ZODz2H1IpRMNMhx/8x97VvnWDELgffSNjlZlIEG+20MkE9E5WgRpIDaAQtAkLn\nsT3zKxoFM+1SKzIyDrktrh3tcgsSwR+wFw8qQYNZSA2zo42BzQ0y6Igira+sxrlpO7oxReT89PrO\nTa6te3Ft3Uveb28JO8y26mzaXluNa+teTPNOoGP91pjH+JvbqL/3MdKvvRBfVS32t0LzW7r3Hqbq\n+79Dk5dFzl3f6ZfAaPnaQhwfb6b9gw0IajUqmyVkHOfm3UgOZ2JjCQKZN66i6dEX8JZXI5oM5P7q\nJrzl1XjKKmOOpTB4uPeX4fxiJ1KHE01uFjk/uoGODVvwN7dS/5dHSb/qYnzVtdjfXdt5TLR2BIHM\nb19B0xPP462o6hT7vBXV+OoaUKfbELQaau6+n7y7b6Njw2Z8tQ3R+wNc2/fg2r6HvF/8IKQ968Zv\n0vTYf3Ht3o86K52c279L3R//QaDd0TnO0Z/diyrVEjKWwtDwr6ftnDwniyf+lk15pY/sDBXnn2PG\nkiLy8QYXjz+Y1ZmvRpbhtEUG3vog+JZyzGgNY0Zr2LjZPYSfoBcII8PTUkFBQUFh6BmIO4Zy14nP\nKNtMimwzMWhS8Ute2t117K57jw5v1/OoSlAzLe8cssxjqXccYHftu/ilYOSBUWtjTMYC0oxFaFVG\nPP52Klu3cahpfcg4YzIWUGQ7EZWgoaLl8zA7ci2TKE2fj1Frw+2zU9m6lfLmTYMSwbfuvQ4euqeJ\nG+5MD9v2g19ncLTCxxcbXBGO7CIrV02qTRVzn/6g0wvo9Cps6QMzhkYrkD+qd6HKXo/Mz2+oY9eW\ngV93Djd7Rgq1O1ZjHTUVfWoWUy74MR57Iz63AwEBjdHSmXdR8nsp//jZIbY2lKQLjCYhNVjZCzCL\n1rCciwPdtxcv2WIhAKlCGo1SNWliNhpByzbfOjRoyVGNCumn5+XNKTtIEa0ICJiE+CJQpAtku9wa\nZscxVKhCBM6RgjrDhnNT0PPTc/AIvrrEBFNBq0H2BZV12d9/hV2Tm0n6t84HGUSjHlQinTGHvcTx\n0eekfePraEfl4/hkM5595SHjWL62MPjwnMBY6gwb2sIcsm65MrQ9JwNPWWXMsRQGD01OFpbTF4F4\n7Hs1gCjSl+zE6gwb2oJcsm74Rmh7dga+uqC41/7hJ0guN/76JlTW1D6LfpLLjWt3sKCSv74J956D\n6CePp2PD5s5xAAJt9n6PpdB/nniunYUnGfjmpaH5CV99u4NLrqvjDz9L5+pVFp5/1YGogif+ns2f\nH2qluVXixm8F3/q+/l7ssAgFBQUFBQUFhZ6kGYsYn7WUrdUv4fA0olWbSDeOwuMPzR1flDaL8qaN\nbKh4gjlFqyjNmM+++mCuy4DkxeNrZ9vRV/H4HdgMBUzJPZN2dx0NHYcAyE+dyuj0eeyufYcWVzXz\nS67C7etyJMowlTA55wx21b5Nm7sGkzadqTlfQxDEMKEyWTz3SCsFxRpWrAp9plerBe5+KIfrV1ZR\nVT64BXb6g9czsPrBwT1e7v1JA7u39k3MG272HC/43R3seeXPZE46GWvhZPTWLLQpaSBJ+D1O2mvK\naK85QOP+jficA+O8N1AkXWB0yG1M05yMHhON0lE6ZDsCAlPU8zCJFtRoMGhMlPl34JadTNLMwSKk\nYRItHPRvwyt7mKSZg1lIRUTV2R6tb2RoFuqZrTmNRqmGVqkRjaBjNFM4UbMEj+zCIbXGtLlDttMk\n1TBVMx+X3LciHm1SY5gdEBQja6UjnKRdjkvu6FX16WQiiCLZJy3HNmEWKp0RZ30lR9e+iLupNvpB\nCeZgbHnmDTK+cwm+qjoEjRrnZzvjHxQFMcVE5o2rOHLNXagsZgr+emef+wJw7djP0Tvuw3DiRGwX\nLMff3EbjQ892jlNz1wP4ahoSG0sQkAMBqm79fUSxKtpYw5E+zYchIm3yXApOvSikzdvWxN5//yZs\nX9FsIvPbV1Bz9/34autRpZgp+NPPwvZLnC+/8zvuCfvORWMwPFZyfXmDlOV+enrFPrZznAEZS2Eg\nuPbWepwuibxcNY1NAd5838nr73Ygy/D9uxr5/l3B+8KEsVpWnmnilz9K6zy27LCPv/8rvge9goKC\ngoKCgkJ3VKIGkPH4HTh9rTh9rbS6qsP2a3PVcKAx+CxaY9+FzVDQuc3j72Bfw5rOv53eFkbZZmHR\n53QKjEW2E6mx76aqLeiA0txRgVHbVdF2TMYCjrRspsa+u7OPI61bKbLOGDSBEeD+nzeSV6Rm9kJj\nSLvFKvL7f+Vy/coq2ttGRlqa3/+ogfWrnXx9lYWps/V9CtsG2LvDw2tP23njf/aIORJHqj3HEwGf\nm9ptH1C7LXaaheFGUgXGVqkhLPwYgiLbDv+nEY/Z4Qtvj9Tmkjsi9g1wOLCLw4FdnX/7ZA+bfO9G\n3DdaLsQDX4qY8WiTGtkmdYmE3QXDnnYcY68/3H18qEmbPJes2ad3/m0uGMOoM7/J/if/gCwHL7j+\nphaMc6bh3LQdbXE+muxwd/NIaPKycG7cTvNTr/eqMIzscqPOCC27Luq1nX2Yl56UcF/R0JYU4DtS\ng3PTDnzV9eT87Dsh4wRa2xMey9/QjL+2idQzF9H2+ppg/0W5+I42IPv9UccajiQyH0Yiol4HyATa\ngm96zEvmhe0ju1yo020Jtfsbm/HXNZK6fDFtbwXf+GoLchPyHIw2TnTbtRgmj8e1ax/qrHT0E8fS\n9sbIuuF8Fbn5zsa4++w94OXks6q5+dpU8vPU7Njt5d6/tWJvH96/NVGrQ21ORW0yD7UpfULU6jEW\njsZYMBpdRg5aazpqoxlRo0VQqZB8XgJuF97WJrxN9TirD9FRcQB/R99ePCocH/R23jR8+o4yZxQU\nvqR7jsOh7ON4p9FxiKrW7cwddSVt7hqq2rZztG0XkhwaSWZ313T+2xdwoxZ1nX+rRC0labPJNI/F\noLEgIKJW6WjoKOvcx6RN6xQPARyehhCB0azLxGrIZ3R6+HpbEMSIzxQb1zpZVFIW1t4fAgGZH1xZ\nE3/HCFx+6pEBtaW/BAIyq99wsPoNB3qjyNSZeiadoCOvSENuoZrMHDVGk4jOIKDTC8gyOB0ylYe9\nlO31UrbHy2cfOwfMa7O39nS0yzgdEk0N/qTYozD0JN2DUWFkYCmdGtams2WhS8vG3fTlBTkgYf36\nUmwXnYF710Fc27oSG6dfeyHaghzU2elkXHsBgTYHLc+8gaesEgQB88JZGGdNAVGg5q4HCLQ5Yh8D\ntH+wgYzvXkr+n+9AcjipuesB/A0ttH+wgbx7vo9j3Wb89V1h2pk3XIY6Ox2VLYXMGy8j0Oag+YlX\n8FZG97ozL5qFceZkCASQ3B6aH3sJoGuc396C5PbEHCvnzm93jlN//xOkrTqb/D/fgaAS8dU0UH/f\nv8EffazhSELzYQTib2ymfc168n7xg+D3+unn+BtCQ/3b12wg47pV5P/uJ0gdTmruvj96uyxT//fH\nSbvoXPJ/9xMEtQpfTT31Dz4W15Zo46RfdTHa/BzUWelkXHUxgbZ2Wv73OvUPPk7aJStIv/ICJJeb\n5idfwFdb3+kpqTCy2bXXy3U/GEYh7YKI2pSCJiUVtTkVTUoqhtyioKBoTkWTYkHU6uN2o7WmM+n2\n+/pkgqexlrJ//aFPx8bDXDKBwvOuRlBFXwaJWj2iVo/GYsNUNAbbjPkgy3RUHKBl23rs+7f36qWZ\nwsjGXDIB6/S5pJRO6t28OWFev+eMymBi/PfujritecvH1L7f+/VEyRW3YMgtirq9+vWnaNu9uVd9\nCioVE77/OwQxPJfY7j/c2msbFRQU+oaMzO66dznUvIH81KmMyVjA6LS5bDjyJF5/V1VevxS9ENu0\n3LMw6zLYWfMWdnctAdnP/OKrwsfqdk07VpMAQEBAJWo40PAR1W07Ihw3vF+ijgTcTonPPnby2cfD\nI6XOcLNHYWgQ5F4sdARBUFbSxyklK64jZdSEsPZ9T9yDp7XvD73GWVNIWTafut8/AgEJ0aBHP3Vc\nZy5HheFJsuZDMuhNiLRCfGRZVmKrE+R4vSfmLrsQ2wnh3gaDzUAKjIJaTc7Sr3/5uQZuirfu/Iza\n914Y8mrZsUTcqlf+jX1fYlEZQ4EuI5fSq38YcdtQ25677IIBnzPQt3lTsOIbWMZPD2sPuF3se6B3\n6WKMhaUUX3pDzH3ctVUceqJ3LwesU+eQ97VLIm4bqQKjck9MnOP1njgSyU4Zj0bUYffU4Q94sOhz\nmJp7FtuOvkq94wAAC0quobZ9Lwcb1wHBcOaclAmsO/xI53aXr5WdtW+jFrWMzVxEpqmU8pbPONDw\nEQAF1ulMyl7Grtq3aXVVM6/4G3j8HXx86GEAssxjmVFwHgcb11HXvg8BAbMuE0EQOdrW95RZCgoK\nQ0Mi90TFg1EBAHvZjjBByVlTjqc1fnhfLJxf7MYwdRz599yKHAiADEd/8ud+9amQfJI1HxQUvir8\n9s50DlX4eOTJ6ImXtRqBG69JZe8BL2++r7ztTRZqs4XClVfH9NbqK9Yps9Fn5VP54qP47C0D3r/C\n0KE2W7CdMD8pffdl3rhqKyMKjCq9AU2KFV977Pzi3THml8TdR5eVi6DWIPsTD1vTZ+VHbPe2KGsH\nBYXBRJYDjEqbhUFjRRREXL429jes7RQXE2FHzRtMzlnO4tLv4PV3cLh5Ex6/I2Sf6tbtGDVWxmct\nRSWo2V33bkg4dL3jAFuqnqc0/WRK0+chyRId3uZBzb+ooKAwuCgCowIAzbs2okmxYpswG1Gnx1G5\nn6NrXiS8xnYvCUg0PfbigNioMHgkbT4oKHxF+OGNVj5a74opMHp9Mr+6I40duxWBMVmozRaKL70R\nrS0jaWPos/IovuxGyp/5G7625qSNozB4HJs3yaS388ZdWxl1my4rr3cCY0F8gVEQVRhyCnBWHU64\nX31WXsT2WLYrKCgMPPWOg9Q7Dsbc55in4jEONq7r9GYEaHPX8Gn54zH7kJHZ37CW/Q1rO9t6hkM3\nOMpocAxsTkUFBYXhiyIwKgDBPBi169+idv1bQ22KwjBAmQ8KCoNDQ1OA0cWaoTYjBL+zHW9rU/wd\nv0RtNEXNxyhLAXz2xIWP7vTXI1DU6hl18Xfjiot+hx3H4b24ao7ga20i4HEDMqJGi8ZiQ5eVh7l4\nHLqM3Kh9aCw2ii/5Lof/cz9+pyPqfgrDn+E6b1y1lQRf8oVHJ+kz83CU7Q5rj4yAIa84oT0NecW9\nFBgjezC6aodXkQQFhd7w6upsiksjPzJ7vTKzxhwdZIsUFBRGMhMma7jngTQKClVs3uTl+suPLy9/\nRWBUUFBQUFAYInxeGUvm8Erx1bDubRrWvZ3w/rFyNvrsrRx8eCjyoQrkn70KXXpW1D08jTXUf/w2\n7Qd3QZxk83WAPruAzPnLSBk7JeI+mtQ0Cr7+TSqefQhZCvTHeIUhY+DnTeHKq6POGUh83kgeN96W\nRrS2zLBt0TwHI6HLyEalT6w4mCFvVML9aq3piLrILxpctVUJ96OgoKCgoHA8843rzJSODcpw8xfp\n4uw98lAERgUFBQUFhSHAkiKSl6vG6VJSDww0thPmkjJmcsx9Dj1+X6+EQHddFZUv/QvL+Onkfe0S\nRG34otBYMJrMk5dR/7Hi/T0SiTdvmjatpv6jN3s1b+LNGUh83rhqKyMKjLrM6F6SYWNFyL/obW7A\nZ2/BVDwudN8EPR0huvcisoy7ThEYFUYu/3ygnaJiFVabCluaiDVNZPxEDbZ0cahNU1BQGIH0osby\niEQRGBUUFBQUFPpJbraKNJsqpM1kFJk8QRtx/8x0FbfdYEWnFdi42T0YJn5lUBvNZC0+O+Y+dWte\n67OXoX3fNnz2VkZdfH1EwSj9pKW07d2Kp6GmT/0rDA3x5k3dmtdo2rS6T33HmzOQ2Lxx11aSOvHE\nsHZdWiaCWo3s98e1JVL+RXddFd7WpjCBUW22JFxARhfFi9LTXI/k9cQ9XkFhuPLGS+E5kn/xBxvn\nXWIc0HGuuSGFWXO1vPK8k7decQ1o3woKCsOHfz/sYOJULfkFKta8f/w9AygCo4KCgoKCQj/58c02\nvnNVakjbzOk6tq4ujHvsXx5uS5ZZX0nS556KShc9BLR1x8Y+C0XHcNVUUP3G0xSuvCpsmyCqyF5y\nDkf+93C/xlAYXGLNm2TPGUhs3kQNNRZEdOk5CXkKGiJ4MLrqqvC2NETZvxjf3q1x+42af7FGyb+o\noJAIF19pIjtXxZbPvENtioKCQhLZt9vHylPrhtqMpDFoAqO5YAyZJ56CMa8EUaXG295Ce8VeGjZ/\niM8R+nAlanVkzlhC6tgT0KWmI/m8eO1NtJXtoGnbOgLeka/0ai1pWEZPwZBZgD4jF7UxBZVWh6DW\nIPt8BHweAu4OPK0NtO7dgqO6jIC7Y6jNThqiRktK0XiMeaMxZOahTUlDbTAjqDUgS0g+LwGvG6+9\nGZ+9mcbtn+BurEaWYuc/UhhZiFod1nEzyDpxKWqTBUGtRvJ68NqbcdVXYj+0i/byPcjd8l4lI9eZ\nIIiYi8aRMmoC1nEzgg+dgojk8+Brb8HVWIOjcj/2QzsJeJLzllmbYsNSOhVT/mj0tmy0tkzkQKDz\nutBx9DD2Qztx1SuhZ8OB3z/QyvbdXk5bbGDJyQbSe3gzRmLfQR/33N/Cq28fv9f2wUZlMGGbHjkf\nJAQLc9R+8PKAjNV+YEfUbeaSCRjyinEdLR+QsRSSS6x5M9Bzpm3PloheiBB/3rjrqoKxVUKEQi9Z\neXEFRrUpBa01PUK/1XiaaiMeY8gbhT0RgTE7ssCoVJBWUIhPSama7Nz46wYFBQWF4c6ACozWcTMo\nOuMKAKpXP0/Tjk8ByJy5lNyTz6J75TudNROdNRPbhFmUv/oIHTVdVerGXXob2tSuBZBKpcagN2LI\nKiRtyjzKX30Ed1Ps0KNpN90X8re3rYm9/+5bovnME08hd8E5IW2u+ioOPHtflCOiYy4YQ/a8r2HK\nDX+DfAxBq0PU6tCYLOjTc0ktnQbIuOqrad23mZZ9W/A723s9ds9zEg93Uy37n/pDr8fpDTpbFpkz\nT8E67kREdbRKqiIqlRqV3ojWkgaAbdIcAh4X9sO7ady6dliLLJHmT8DjYvcjP0cOxA9nikfhslXY\nJszs/NtVX8mBZ/8c9zidLYvxV9yR8DjJnA+CqCLzxCVkzTotLHxMpTNgyMzHkJlP2uS5eO3N1Kx7\nlbaD2wGQBvCFgyCKpE2ZT9asU9GYU8O2q3QGVDoD+ow8bBNmIvl9NO/aQN3Gdwi4w0No+oIhq4Ds\nuWdgKZ5Iz2qhglpENFvRmK2YC8aSPWcZrvoqate/RXvFngEZX6FvVNf6eeRJO488aUerEWgpK+Gz\nLzycfVnke5XbI+P3H+dJWIYA6+RZiJrIYekADevfG9BwTVmSEMTIebjSZy6kShEYRwSx5s1Az5n6\nj97EMv6EPs0byevB01yPLj07bJs+M36hl0j5FyEoXAbcTvwOO2qzJfSYBAq9qPRGNCnWiNtcNYrA\nqKAQj3mLIhdIUlBQUBhpJC07rSGrAIDU0mnknnw2PR+Uj6HSGSg+51uojSnBv7X6EHGxJ9oUG8Xn\nfCtm+NNwRFCpKVx2GaPP+25McTFGDxiyCshduIIxF9404PYNNoJKTe7CFYxbdTtpk06KIS5GR6Uz\nYJswE1EzvKsvtez9PMTjDoK2W0om9btvUaMltTS0OmXzns/63e9gojaYKD3/BnLmnxU1N1V3tJY0\nRp35TQpOvQhBFPG7Bsb7S2fNZMxFt5C/5LyI4mIkRLWGjOkLGX/FHZgLx8U/IAaCIJJ78tmMvfj7\nWIonEe2a2RNDVgElK66laPnlw/638FXB65PZvc9LICDj6JAi/qeIi8khdcrsqNskj5vWHZsGdLz2\ngzujbksZNxWVfmBzdCkkh2jzJhlzxtfW3K95E80jMJFCL5HCo332ls4XZK4IfeuzCxBUsT2ronkv\nypKEu6E6rl0KCl915i5Q1m8KCgrHB0kUGAsRNVrylpwXd1+V3kj23DMAyOnh6RUJrSWNjOkL+23j\nYCGIKkpWXIttwqwB6a9lhAlIPdGYLIy56GYyZyyO+gY/UTytDXRUlw2QZcnB72zHUbEvrH0g5kNq\n6dQQUUkOBGjdt6Xf/Q4WKp2B0efdgDG3uNfHpk2eS9HyK/C7HP22w5hbzJiLbu58MdJb1AYzJV+/\nDuv4yGFv8RDVGorPvYbMmUsjhr4lgnX8iYy58HudL2sUhpZtO5UcSoONJsWKPkqhCQD7/m3Ift+A\njtm2M/r9WFCpMZf2/0WSQnKJNW+SMWegf/MmmkdgrLl/DGNBcViuxmXUAAAgAElEQVRb97DqSOKl\noFKjj3NvjJZ/0dNQk1DhGQWFrzJGk8CseYrAqKCgcHyQtByM+rQc0qctQGOyIPl9tOz9HFddJSq9\nkYxpC8JCKWwTZtG841PSJ89F8ntp2bs5uL9OT/rUk8O8GtOmzqdu07vJMn9AyZpzOuaCsRG2yDjr\nKnHWlONta/oyt6SMSmfsDI/WZ+ajMXWFq8iSRPOuDX2yo2n7J6gMJtQGE2q9CZU++G9BNbi1fkov\n+F5ML1VkGVfjUXwdbQRcHchSAJXWgDY1HV1aFqK6K4ypedfGQbC4/zTv2URK8cSQtpTiiaj0pn7l\n1rT2ECnth3clHKobcDvD5oTOljWo86Fw+Sr06Tlh7V57M21lO/DZmwl43agNZnS2LCwlk0IEtNSx\n08O8Q3uLPj2HknOvjegVLXk91H/+Pn5XB7IsoTaYMWYXYi4aj0obGs4iCCKFyy7D195Kx9FDvbBA\noOjMb5AyakLErb72FuyHd+Osq0ClNaA2WTAXjsWYXUhPL0d9Rh6jV15P2f/+elzkqh3J/O6vLWSk\nKfmUBhNTyfiY2x1lA59GoKPiALLfj6COfN00l0ygbdfnAz6uwsARa94kY85A/+ZNJC9DCOaRVJst\n+B32qOPqs8OFQnddl4dhtIIshrxRuGoqovcbrcBLrVLgRSGcvHwVy881MHuujtHjNFitInq9gMsl\nY2+TqKzws3+Pj42feNj4iQe3K7rH/9vrc8jLD95rV7/r5uZrmhKyYfuR4Jw9Wh3gjHmR848mC5NZ\nYNZJOsZO1HDu+UaKStR097e48TYLN95mid4BMHfiUZwd0c/Lq6uzKS4NXl+WzqylsSGYr1xUwYLF\nes690MiY8RqyclSIAjQ2BHj+qQ4+fNfNkcOhLwXGTdTw/DtZnX8/eK+dhx9IPFXXTbdbuObGrrX7\nylPrKDswOC8eVm/JJT0jeHK7n7MT52hZcaGJGbO0ZOWokGVoqAuweaOH119ysXlj79Ni2NJFzjzX\nyPzFOkrHabCliQgCbFjn4aMP3Lz5ijPmdxbP7mTYPBBMnqblmdczAbjzlhZee9FJfqGKW+9MZc58\nHVIAtm3x8tc/2jmw14cgwIWXm7j0m2YKilS0Nkus/cDNg/faaW2O/0yXniFy609TmTApOH/NZhGv\nV8bhkDhy2M/uHT7eed3F9i29f9E/YbKGFRcaOWGWlvwCNSkWkUQei6cVRffU/8sj6ZyyLHYKhFjH\nJ8qc+TrmL9YxY7aOnFwVVpuIVivgckk01Aevqzu3etm4zsO2LV6SWcYiaUqCoFKRe/LZSD4Ph158\nCGdd1yKjZc9njL30ByHCmajWUHz21SAIHPzvA7gbj3Zua969iXGrbg/ZPyjA5eCOkpR6uKDSGcia\neWpYu7OmnMoPnsPTHL+CUMaMxVjHnoAxZxT2w7vwdURfPMaies0LEdtFjZaMExaRM+/MPvXbGwRR\njCouuptqqP/8A9or9kYVyQRBxJBVQP7SC9Gn544Yb077oaDw1z3sSRBVWMfNoGn7uj71qflSaOpO\ny57EQ7n8LkfEOTFY8yF1zLQvQ4FDkXwe9j7+GyDCTVgQSJs4h9wF53SeS+u4GX22QRBVFC2/PExc\nlCWJ+s/fp2HL6oi5t0SNjqxZp5I169QQj0NBEClafjn7n/5jwsVfMmYsingefO0tVK95Afvh3RGP\n01ozKFhyAeai0NBsfXoueYvPo/K9pxMaXyE5HDzk4+Chgfd8UoiOMa845nZnVW+E/8SQfF5cdVUY\n8yOPbSwcPeBjKgwsseZNMuYM9G/euOuro+b+1Gfm4YghMApi+EsPVzcPxmjipTFvFM2bo3aLPjuy\n96RS4EWhJz/8WSqXftOEWh0erWEyC5jMKnLzVcyZr+Pyb5npcMjc95s2/vfU8VMM7YRZOv76WAxH\niwEmK0eksSFAQZGaP/0jjYlTwtNSFZnU3HpnKgajyEN/Dr2G7N/jY/NGDzNPCnpZXnCZiUf+1k4i\nNRZFEc4+r+vZZ8cX3kETF3tSUqrh0EEfd/zSysqLw9NQmMxqikvVnH+ZiXded/HD7zYn1K8owjU3\npnDV9SmYzOHzesnpepacrufGH1r4y+/aeOm53uVsLynVcNEVpoRs/vkPWxIWMZNB6Xg1mdkqHn8+\nM6Ro0ZLT9cyap+OKFfVcfKWZS75h6tyWnaviostNzJ6n47Jz6ulwRLZ/znwdV15nZsESPT1vfwa1\ngMGoIjNLxcyTdFxxjZmPPnRz1w9aaGmKr6QZTQJ3/sbKOeeNzLQ28xfr+cd/Il9TzCki5hSRklI1\ni5bq+e6t8Pg/HNz327aI+w8ESXdVqv/s/RBxEYIho/WfvU9+j/BpTYoNIERchC89rbZ9TM78s0La\njbklw15gTB0zLSx3jbe1kUMv/wPJl5iy3vjFWhq/WIs2NR1BGPio9mMVmgeDzJlLw9pkWaLmo1do\n3L4uWB0xBrIs4aw7woFn/oTGZOlTsZuhQA74ad3/BenTTg5pt02Y2WeB0Tr+xJD54He2016xt192\nwuDMB0GlJnfBuWHtcsBP2Qt/J6K4CCDLNO/eSEfNYcZc+D1UelPk/RIk+6Tl6DNCH4wkv5eK1x+j\n/Uh4WHvnPj4PtevfxN1UQ9EZl9Pdk1CTYiV3wTlUffDfuOPr0rLJ7XFdA3A1VHPohb/F/B6OXUdy\nTz477HdlmzgL++FdtB3cFtcGBYXjBV2MEFFfeyt+Z//TKUTCXVsZVSjSpFhRGUwEBihXrMLAE2ve\nJGvOQN/njez34WmqjVjURZeVh+Nw79YB3UOkA64OfG3NaFLTQvYxxCj0IqjUaNPCi86AUuBFIRRB\ngCuuMYe1B/wQkGS02sii45Hy4yvM3tEusXtH1wtIUQx6Th2joS5AQ31sUSSQgLh3jKwcFe12mSde\nyuz0iovGxx9GXnc+/VhHp8CYk6di8al6Vr8b/1lh7gIdOXldz8G9FdcGktFj1Vx/SwqLT+vyJpNl\nCATkMMF7+dkGfnabgMsZ+7lUpxe49+9pIX0eQwoAAp1iWFq6yC//aGPcJA1/+EVbvEfeELu7i4ux\nbM7KUXH95Y1x7U4WpeM0fP/HFrJzVZ0ecsc+v9ks8P07U1m0NHiupEDQo/YYJaVqLv2GmUf+FvnZ\nfsnp+s5ju+PzyahUQpjouGipnoeeyODyFQ0x856r1QIPPJrOnPldaQqamyQ+3+ChpUkixSJwwmxd\np6f0MT7f4GH3Dh+fro39O/jfUx3s2u7FliZitQX/KxmjIa9gYKKcbviBhetu6l16rDdfSe7vMLkC\noyxHDWFtO7A1TGAEonquOSoPhLXpYoXZDhN6ihcAzbs3Jiwudsfblpjb/3BFbTCTNeu0kDZZkqh4\n83Hsh6InPI9GXz05h4qWPZ+FCYzGnFHorJl4Wht63V/PHI4tezcjJ9PfeQCxjpvRWRG8O0c/ehlX\nffwHEk9LPUfeeYqSFdf12QZNii2i4H30o1diiovdad3/Bab8UtKnzg9pt02YTe36t+IK4Lnzzw4L\nSfc72zn8ysMJi7w1n7yOJiUN67gTQtrzl5yH/fBO5N6sQhUURjCRKusew9vc+2tsonhbYvetz8yl\n48jBpI2v0D9izZtk0p95466tjCgw6hMo9NIdf0d7WEi1q7YyTGDUpKahNqXg7wi/p+kzcyN6U8p+\nP57Gml7Zo3B8s3R5V7TIpk89PP5PBzu3emltCa5ddXqBklI1k6drmb9Ix4IlehrqA2z6dGhCP5PF\nts1eLjmrvvPvFIvIJzu7frvP/aejVyHI8SgereaWH6eSniHidsm8+ryTT9a6qawI4OyQSEtXMW6S\nhm/flMKu7ZGfTz9810VdTaDTK+3iK80JCYxfv6jLEcDtknn71aETGG/7aSq2dBFHu8Rj/3Dw/psu\njlT4CfiDYbdLlxu44TYLaenB69lNt1v4/S9ie3n97B5riLhYWeHn0b+1s26Nh4a6AIIAF6wycc0N\nKZ1C66qrzLQ2S/zzL4l9x7f9NDVhm2fM0iZkd7KYu0CHRiNw32/beOaxDjQa+MUfbSw7K/jbX7BE\nj6Nd4qe3trDmfTdpaSoe+FcaU6YH058tO9sQVWD8zyMOLv2GGUGEe+9uY9sWL2X7fTg7ZAQB8gpU\nLF1u4LqbUki1Bs/HpKkaVl5sjOkBfdlVphBx8Yn/c/DA7+14vV2ipCjCZVeZuf3nXUVAKw75uffu\n+Od53Wo361aH/lYu/1ZoX33lm9eb+fbNXeLikcN+Xn3BybbNXhrrA7jdMqnWoKA58yQti5bqaW6S\n2LsruRFWSSvyAuBurotagMHvckQUzNorIue7cUcIJdakhAsUw41IBRe87S1DYMnQk3HCQkSNNqSt\nYfOHfRIXRyLOuiMR57FtYu+LvejTc8PE65ESLg5gGz8zrM1rb+5VftH2ir29zHUYSvrU+WEPRc7a\nCpp3ru9VP3Ub3kLuESciqFRxC1FpLWkRK4nXbXi71565NeteQfKHLgrVxhSsY/sePq7QP4wGgezM\n8LeTGo3AqvNT+OmtNs5eZuprTR+FHqh0hrD7S3d8bYmFOvUFb5y+NRZb0sZW6B/x5k0y6c+8iRbK\nHEl07CL8YtPdezFe34YooeS6KPkXj4VyKygc45gQU17m59uXN7JutbtTXATwuGX27vLxwtMd/OD6\nZpacWMPN32pK2NNLITI33GZh9Bg1O7d6OXtxHb++s5XV77o5uM/H0aoAO7d5efGZDs5aWBf1XAf8\n8N8nu0SaeQt1FBXH9lNKsYicsrxLfHvvLReOKOGvg4EtXcThkLliZSP/99d2DpcFhTqApkaJ/z3V\nwSVn1VNXE1zTX/pNc2cey0icdqaBc87v8izc9KmH85fV8+KzTuprA8gySBL89z8dnHdaHds2d63T\nv/N9C1NnJHbvsaWLCduciN3JRKcT+PAdF4//w4HHI+NwyPzqjlY8nuD3Lorwp9/Y+fAdN1IgmPvz\nwXu7nnnGT9JE9GQGqKkO8OObmznz5FqefNTBji+8neHgsgzVlQH+84iDqy5sDBEHl58TnmO/O6uu\n7vKq/nyDh3vvbgs5HoLf45OPOnj1+S6B/OzzjZgjhMQPFuMnabjlji6R8r7ftHHuKXU8/EA7Gz/x\nUHbAT3VlgN07fLzxkpNf3dHK6SfV8t0rk++wllSBMZpYeAxnbXjC6NZ9X0TcV/J5wjzWNOb+K7/J\nxhMhhDt3/tlhBSKOdwxZhWTNPj2krXb9m9Suf3OILBoaDr/0UFhRkp7nJS6CQMmKa8Oa3U0jw1Mg\ndez0sLyBAOWv/l+vH0Yq3nicqOHUMUgZNSGYP7EbciDAoRcf6nVfflcHtZ+Gz+Os2adGLBxzjJIV\n14VVjG7a8SlNvRQ4AXyONg6//HBYe+Gyy2IXVFJIGg/fl0XV9uKQtt/8JB3nkdH85bcZzDlRz5MP\nZeE9WsrFXw8PGesrBedOZdmam1m25mZOe+9GFj57FUteupaMOdFDHI8HtOlZMbfH8xbrD97m+pjb\ndRnhhawUhgfx5k0y6c+8ad2+MaJ3ui4zN2o1aeu0kyL205OWLz6NmK4mfdaiyP1OnR2xvfmLTyK2\nK3x1ORa2mVugYuLk+OKKyykPWb6+4wmdTuDO77dw2bkN1NdGj2qJFUYK8H9/bee1F4ICiyDAU69m\notNFF1j++WR65/bKCj8//f7QOtjIMpy9sJay/dG9t2qPBlg+L1gURxThwSi5MsdN1HDfP7ocnR7/\np4NrLmmMWpAoKGw2dIq0oghPvZIZEj4ey+5EbT7WdzS7B4OH/hzqJGFvk3jjpa689C88HepN+Ola\nd6cAKQjBcxuNt19zcbQ6dmTWwX0+Hrqvy4Y583URc75C0OMyt1vo85/viR0d+edueQt1OoFl5wxN\nzkZzishzb2R1hoWXHfDz+D8dcQu3yDKd8ySZJFVgjJe7RgqE/1h8zuhfbE8vIVEdfQIOFyKFWmpS\nrIw+/4aI1XOPV3oWI4GR5XE3UPg67DgSDL+NhrlgDBpzaBX2kRQGa8gI93bwddgjenfGw+9y4Glt\n7PVxxpxwscVZWx7mBZgoroZI1b8EjLnFEfdX6QzobJlh7fbDu/o0PkBHzeGIhWVMeSV97lOh70yb\nFPrwlGIW+c5VFiQJll14lHMvr+Hqm4ICw6oLepc7JRalVwZFhJr39vLhWQ/x8SWPsWbl/9G0OXrq\nAV2GmYk3L0GX3r+cpkNJvPVAMq+R8foWRsBa5avKUK4j+zNv5EAgavixLooXo7Eg/F7gPBr+ol/y\nuvE0hd+P9TmFEJYHXIjqNakUeFHoybG8XzqdwKPPZfC9H1rIzBqYPGQK0Qn44b03Eys8GI+nH+sS\nh1KtYlTvsNKxaqac0LUOevm/ziH3RK2s8NOcQMEPSYKN64Jh+YWjgpWEezJrri7k7zXvJXZ+P3wn\nNFR2do9+IlFZEV9k724zRLd7MGioC7+3OTtin3ePu2ty6PT99wo8JlgeQxvlNPf0lvR5Y09Sf4+v\nQj1El6+ZJ2lD8le+/NzwyvOd1JkX8Mb+sUXyVooWUg3QU5aNlPNluOGsrQgrcgNgyMxn7KW3UXj6\npeijJMc+njBkFoT87bU343MMTX6IoaZld7iw2jMPXyx65l4EsJf3XZgabCLlJfW09F5cPIZ3gATG\nvgicx4h23dL2EIK7xi8mYrhaf7xQZTli0StjriIwDgU52Wrs7V33rAvONZNiFnnl7Q62bA8uAl97\n14nLLXPClPgLzEQw5FjQZQRFwrLHNiB5uxZ5ciD64i5zXjGFK6ejsYxcz/pI1XG7k8xQzZ4vP3vS\ns9CbwvAh3rxJJv2dN9EKqETLw9izoIzP3oI/yjrMFUF4FDVa9FmhfWutaYgRntwkrwdPHA9Nha8e\n3UNEDUaBa7+Xwrsbcnjg0XSWn20YEGFBIZxDB31RPet6y67tXnZ80fU9XnJl5BeTK7rlXpQkQkJL\nh4qqBIS6Yxw6GNxXEGBUSfgz2rgJoW379yTW9/7doc5Vk6bFf8mVqN3HbIbodg8G9rbw9Va8d7zd\nl2iJyDvnXmDktrtS+etj6fznpUxeW5PNextz+GhbLpv25fHDnyUW5drTGzKeR2luj0IvDfVD4+Az\nflLovPni8745yCSLpCp0kq/3CSTlntLwcUD1h/+LuJAURBHbxNmMu/xHjF75nWDF6SFc7CYTfVpo\nGFIihTyOV9oO7QzzNLOMnpzQsaJaQ2rptLD2SKLlcCWS515fRMJjeNp6f6whqyCsrT9FlKQInoMA\nKn1k13l9RvgDoOT14Gtv7bMNAO7mcIHRkBk5P5ZCckkxCTS3dMuJszIYBv2vp7q89H0+mYamABlp\nA3Mr1qR2eRK46hPP45k+q2hAxh9KpEDstUMyX0jGFTfj2KYwdMSbN8mkv/MmWq7ESEVrVEYz2h73\nXmfV4ah9R/JshPBq0vpo+RfrqiKGWSso3HlLS0iInkodrA77x7+nsXpLLj//nZWJUxSv74GkpXlg\nX7A9/XiXt9SUE7Rh35eogrPP61qPrP/IHZIjcKiw2xO/JrV1yw1qTgkXvq1pXddvn0/G0Z7YOW5u\nCj0PVlv8tUmidne3GSLbnWykQJg/2IAya66OR5/L4Nf32bjyWjOLT9UzfaaWUaPVZOeqsNpE9AYh\n4fzmFYf8tHTzaj3ta7HzNXbfLknwxWdDI+zZejw31FQN/e+rO0mVtvuyqD4eF+Kuhmoq3vw3RWdc\njqiOnHPEXDgWc+FY/C4HLXs307xrA55+eFQNN3rmy/QkMR/WcEcO+Gnd/0VI5WHbhFm0HdgW91jL\n6Klh3gJ+Z3vcfKfDiUh5CftTETzg6q1buIBaH/7GNXfBOeQuOKfPdkRCFWEcAHUE4dHX0X+P3kje\nKJE+q0LyaWqRSLMFF6CjCtQsnm+gtj7A+x+FitFSQB6wQi8qfdctXfYntsITNSrSZhQOjAFDSLyX\nk4I6ecudeH3LfXjZqjA4DOVL7f7Om2ghyJFyNxpywn/jrqPlUfuOts2QOyqYo/HYWFG8JaOJnwoK\nr73o5N03XXz9IiOXXWWmpFsxCrNZ4PzLTJx/mYmNn3j406/bkl7t9KuA3zewYv+7r7u47a5gVWoI\nVpT+xe1d+RUXLtGT0a3I3UvPDb33IgTXW4ni7hayazCGL9KMpq623niHSlIwfPdYbkpzAmHMidrd\n3WaIbHey8ffiHPeWm263cM2NoSmFqisDlJf5aG6SaGuVcLtkXC6ZE2ZqWbg0flSOzyfz3yc7Oisx\nn3O+kU2feiJ63M5bqOOb13flTH/jZWdIkarBxJwSOm864oSgDzbJ9Z2Ve/9h5eP0jaf90E4OPHMf\nBaddjClGyKLaYCZzxmIyZyzGUXWQxi/W9isv23ChpydXwOuOsudXg5bdm0IExpRRE1AbTPjjiGW2\nieHVl1v2bRlRlRojFTiSfH1/AyT5e7f4VOn0YcVVkoUY5QFSjCCy9uccdPXhCWuL5kWpkFy27vSw\n/JTguf/prWmIIvz7WXtIEnVR/DKUuo9VFdVmHdN/eSb6DDO6DDNqU9cLrGVrbg7Z990lf+n8t7HA\nSsmq2VjGZGIqTkPUBB8E5j92edgYh5/+nAMPD/9iDQF37AeYZP4O4vUdKTeqwvAg3rxJJv2dN57G\nmv9n77zD46jOvn3P9qbd1apXW5Zky7130zGYZlMcQkjgIxASWkJCSCXJm4SEEEIKgYTQCWCaaQ7F\nYBuMweDe5SZbvdeVtL3O98daZbWr1apaMntfFxfemTNnnpk9mj3zO09B9HpDhEq5IR6JXBH0mxJO\nYIzkwehqqsfvciJRBv9eq9OCvZ17KygznAKjUqHnrPk/ZdMXvxq2c8QYXlxOkddesPHaCzZmzlWw\ncrWGiy9Xozd0vTQvXKrk1feSeej3bbz8XOSc/jFGFo9HZO1LNm77YUCUuXSVmr/+oQ1Le+BdZNW1\nXc+2VrOfzRtGxztff0Lwu4tzHZWKu9PdY1Gljr5fiZSgwjjW9r7f36K1u6egGM7uscrXvqkNEhf/\n9kAb77/tCJvvEeCWO+KiEhgBnvm3hbMvUDF5mhxBgD/8LZ7rbtSyc5sLc7MfXZyE2fMVLFjS5eBT\nVeHlr/efvlRvHQWzOlCrhVH1fQ9riPTouczRgcvcQPHax6hY/wIuc9+5aXSZeYy/4hb046eMgHXD\nRyC/YPBDz/8VFxjt9RVBHqqCRIpx4pyIx8g0ceiyJoVsNx/ZOeT2DReCRBI23+RAi6sM5NhIlZ1H\nCqkinMAYKg72F184gVExcoJqjC5eWhsIUT65axw3fSMOq83Po08HT0amTVagUgqUlA3MQ0OQCEhk\nUtytDiwnG7Gc6PIMNx+oDvqvOwqDGk26Aa/dTfuxrudQ+/H6kOMctWMjV25vueQ6kGqGrlJ3T2R9\n9O21DtxDO8bw0te4GU4GO25Evx9nLwXGFD3S0qh6CIx+twtnY6ScvyKOutD84coeVbfDhWMDOGtD\nj40RIxwH9ri5/xetnDenjp99v4XCA11zOokUfv47A2dHKRTEGDnWvmTrXDBVqQVWXhMQFY3xEs65\nsOv7ev9tO54h9qAcKN0F7L7oHoJqtYTa3z3sXC4Xoi6okpAYnBojGg+4aO3uGTYbzu6xiFQGd96r\n7/xst4k8/x9rr+IiQH/qtzkdInfd1My2z7veoabNUnDz7XH8+FcGvnd3XJC4uHu7i29/rSmqgkHD\nRc9xk5w6ulLsnZ7sn19pRFpP7Kf15AEME6aTOPtstOkTIh4xfuV3aC85TNUnr+O1R59Xa7QghBE3\nxAF4t55ptBzdRdrSyzs/Gwvm0nTg817bGyfODskj5mioGlxhkJGmN6FrEJ7L/a0OOzoquoYZ/0Mg\nAgphCseIiLHVntPAa+9YWX6uhhuvjcNm9/Pdexqpbwweq5cvD4Svf7FzYB5unnYnu+5+o/OzYUoq\nC//9dYCg7T1pPVzbuV+qlnPB+jsAKHxwI9bSgeciPZ34PW58DhtSdfiUAAqDadjOLe+j78HmVo0x\nfPQ1boaToRg3jtpK1GmhRcuUiSmBPIin6OnB6Kgt7zPKyF5dhnbcxB5bu35jBIkUuTEx5Dif04G7\ndWw+R2KcPjwekfXrHKxf5+Dab2m574/GzmnRrd+P47NPwjsmdA8f7Ue9xBiDpLHBx6b1TlacqiJ9\n9Tc0rHnOyiWr1MjlXc+J0RIeDTA+N/oBMiE/8K4gilBeEroIfORQ8LaCqXJ2bevbUaBncY5oUgBE\na3eHzdC73WORqTMUmBK63n/ffKXv1Fg9hdy+aGzwcf8vWnn+jUSSU6V4PCJuV8Ar1G4XaajzcXCv\nm/Xr7EFC5Omi5ETwdztjjoKjhaPn+449iqOgPxV+o0YUaSs+SFvxQVQJaZimLSK+YF6v3lX6CVPJ\nT76H0nVPji1BCfCHyXEULkz2q0brsd2kLrkUQQg8NDUp2Sjjk3v1bg1XPdp8dOwUd4GAGCj6fCHV\nMSWDEP3CCdiRbQifc6tx72YcDVVh9w0UV2v4XKM+Z6igJJEPvpJw2GqeLgcxhXHkEUW45e4GXn/H\nyt6DLhqbQ4XwhiYff3rEzPOvjr2Fo9GIs7EWbXZe2H09PbqGkp7FM3riagotvhRj9BBp3AwnQzFu\nes3DmNCVh1GmjUOm0wftt1eX9dl3uErS3VHEJ4YtntSbTTFiRMvrL9mYu1DJJasC70TTZymQSMIX\nj+jupZWUHJ2oMGX6aFho7kL0B8/RZLKxEXXyynPWToExv0DO5GlyLruyKzz6yCEPRUdHj+iRlCwl\nO0dGRWnk3LsSKSw85bFWXuLFGiaNzc4vgkWmc5erohIYz7so+P139/a+j4lmXHe3GXq3eyzS0zvv\nWBRC2tyF4Wte9MakKXKefT2ROL2EynIv3/92c1BV7tFGz7G26msaXnuhvzUJho+YwBgF0jAv7UOJ\ns7mWmi1vU7v1XYz5M0mcdTbq5NB8OXKdgZyVt1L0ysOnNfmVU70AACAASURBVG9Q/xHxu11B4ock\nJjDisbVjrSgiblxB57b4gnnUbfsgpK3SlBJS+Vj0+2gt2jvsdg41PrcTWQ9vEaGX4kfREC6fYcTz\n9/K342yuo7Vo34Dt6JcNYXJrSeQDvwddfYQ+q8KJmTFGjo829/6sfvqlWOjsUOKsr+5VKAqILIZh\nCYlVp/VeJMdrs+C1xQTk0UzEcTNMYwaGZtw46sIvinUv9BK2wEuE/IudbfoSGHsR7cOFVvfF5NxV\nZKTOD8mrqNdlsGDm7Rw9+Q7V9bt7PX5cxjLyx6/gZPlGyqq29Pv8MUYfJ4s8QLf5XS+aW12Nj4Kp\nAcEwv0CG0SShtY+qydf9v+FLmTEQ7HYRvz+QlxkgM3t0hTv2xr7dbo4d9nTe/0tWaZg+u2su+/Zr\no0fw6ODq6zT840+R514XXaom/pTH3Ccbws+hK8u97PjCxcKlgXn3lddqeO5xa1CF9J6kpEmDBNid\nX7qorhya6r/dbY5k91ikZ5EiQx+Vt5eeqyK/IPpFhI68ix1h7j++rWVUi4sAzU1+tn3uYvFZgfE3\nbebg3yGHkmHNwXimMBQv/tEg+ryYj+3hxKt/p/jNf2GrCZ0AyuOMQWG1YwWPPfhhrtAPX7jaWMJ8\nNDh/onFS+DyM8WG2t5ce6bMozGgkXP7NvnJRRaK/ORV9LkfYvI09Rc/hxOsITVgeqLQ+uFVruc4Y\nei7XWFqMiBFj4NgrT0bcr83KHfJzShRKVD0Wf7pjr+5byIlxeok0boZjzMDQjRtXc33YAmHdvSN7\n5l9EFLH3IR5CYDHO3RLeCx9AaQrvgemoHVkPxqy0xeSPX0FxxaaYuDiKkcoC4k40KBQCyy/tmtuV\nlXrx96LD7N3Z5ckjkwn84Kf68A1PsWKlmlVfG13F7/x+qCrvEjTOOl8VJBaNZl55vms+e/YFqs6w\ndpdL5IN3Rp/I9a1bdMxZ0Pt7fWq6lJ/8xgCA3wdvvdL7HPo//7B0ZniK00v4y7/je63crNUJ/OXf\npqDq0089Gv3iY7Q2R2P3WKOsJFjsi1S8JTdfxh//Ht+v/tPSpZ2h6z4vnDg+erxuI/HEI8HaSrTP\nNZ1O6KwAP1yMjafXaUYeN/JimK26mOI3HqVy06sh++Inzx9RMWQo6Bkqqk7qfWL9VaKtuDDIm02h\nN6FJGx/SzpA7I2RbT3FyrOBqawrZpjRGDhWLhFwbeTIZ1oaW0DB0ZfzwhVD2JFwotkSmQKHv349i\nT1Sm0IT74YsAxBhNKORjIxxqtGMrP9FrCgQAXe7kIT+nbvykkJQPQTaVHhvyc8YYWiKNm+EYMzCE\n40b042wIfcYrDPF0LFipkjOC9jmbaqMutOeoDRUiO6JRFL0IjCMZIp2ZuoBJEy6jpOJjSis/HbHz\nxug/MqnAbx+K5x9PJXDtDVqmz1agi+t6DZVIISNLysrVGl5+N6nTKw7gnQiecO++acft7vJwWn29\nlocfNzFtpqIzJ6NaIzB/sZI//iOeB/9pGpV177Z83PU3aTBKeOKlRBYuVaLVBYwVhICINSFPxvzF\nwxtZ1x8+eMfRWXBiQl5XYOTH6x2dVaVHC263iEIh8MSaRH70CwMTJ8s7x0h8goRrrtfyynvJJKUE\nns0vPmOloqz3OcWeHS6e+XeXSDh3oZK3NqZwzfVaUtKkCELge1t9vZY3N6Ywa26XSLjmOSs7vogu\nl5/bLUZtczR2jzXKir0Un+i6nsVnKTH28GJMSZPyvbvjePndZEwJkn4VYPF1ayqVwa8fiCd3orxf\n1cFPB3t3unn+P10C//1/jefV95O58VYds+cpGDdBRkaWlImT5Zy7XMV3fxDHf15K5NP9aSw+e3gj\nSb8yIdKRJnJ9Ee6lfaQwH9lJ1oXXBW0TJFJ0mfm0nth/mqzqP87GmqBq2EpTMlKFCt9XvJq06PPS\nWrSPhOlLOrfFT5qDvbas87PKlIKyxxj0OqxYyo6OkJVDi7Oxmrjs4GrYPa+vP6i65ZqKFkdDZUjI\nuSYtZ8A29Bd7fQWi34cgCX4uqRLTcbe3DKhPQZAE5d3qwFbXt6dKjKHn2lU6Xl8X6qnak3iDhLXP\npnLhNTUjYNWZjd/jxlpyjLj8aWH3x+VPR6JQ4ncPXYJuw9TQ3LidiH4sxUeG7FwxhodI42Y4xgwM\n7bhx1lWiyQj+/RJkcmRaHV6bBWVSWtA+RxT5Fzuw15SF2KpKSsNeXRY2h6TXZhmxokZpybMpyL2C\nksrNlFRuHpFzxhg851+s4vyLu15u/T7weAPCTzjhb8cXLl58pvff0uYmP4/9pZ177uvy4LroMjUX\nXaZGFMHrFYOKjgCsX+fozO8YicxsGResUKGLk6DTC8TFSQL/jhPInxQQQOVygf++mYTV4sdqEbFa\n/VjbA/+vrfbx3lvReZI9/4SVlas1GIwB4aRgqpynXgkUUXI5ReQKoTOEutXs5+yZoyMfv8sl8tar\nNm6+PS5o+2gq7tLB049ZWHKOillzFXz7dh3fvj0QPeX1imHzXj72cN9pbB57uB1jvITV1wccfzKy\npPzfg4Foog6v2x5Tfdavc/DX+6NPkfP0YxbuuEcflc17d7qjsnus8dc/tPHYcwmdfwOf7kujoc5H\ne5ufeJMkSGA9cczDj77bwrtbUqJaTKiv9bF3p7vTS/Tq6zS9elvbbSLNTT6OFnr48F0Hn3zoCJsb\nFgILAldeq0EXJyFOL6DVSYiLE9DpJWTnBEtwj7+QgNUqYm33B/5vCTxH1jwX+T3iHw+2odMLneNv\nynQ5U6YbIh4zEnxlBMb+hlF2HqfSnvZwXnt9BZqU7KBtikF4fJ0OrFUnSJ5/YednQZBgnDSH5kNf\nnkarRgfmo7uCBEZD3kyqt7zdWVlZH8Z7sfXYHsTenmijHEdTqJCiiItHYUjA3da/ypOCRDogcdJS\ncRzTtMVB21SmZBR604AFvv4g+rw4GqrQpAZX/9TnTKG9pHBAfWozJoQtntRdrI4xcjz/aDKtbX42\nfNr7JDtnnJz31qQxMXd0JZwfy7Qe2tGrwCiRK4ifuZjmXZ8O2fl0uVN63WctPY7XeuZN9M9Eehs3\nwzFmFPGJQzpuegtJlhsS8HvcKIzBc9j+hO07qkMXqJQRBEbHCHovTsm7ErujhdKYuDgm8PmgpspH\nemaPIn9SUEpDVQCPR+Tl52z888/tRHBMBwLinEwucNeP9UFijiAQIi6+9oKNh+9vi0pgnDxNzo9/\nFfllXRBg9vzw4atlxd6oBcbGeh933tTM358wBYklAErV6Pakeu0FGzd9N67z3tdU+dj55emvtNsT\ntVrCnf+vmd8+ZAwKwQ8n1H34rgOXs+8iKX4f/P7nrRwr9HDXvXqMpmCv3O60t/l54hELLz7d9+Jz\nT7s3fuCIyub/+4k5KrvHGls3O/nNvWZ+/YARpSogtqemS0lND77JWz52ct8PzbS3+ako9TJuQnRS\n149va+afzyQE5RANh0YroNHKyBon46LL1BzY4+aGq8KnEklMkgSFrkdi6bnhPQr7Ehj9/sD427XN\nxZ8fjV6vctiHd4x8ZQRGiVyJPM7Y75VVQ+40TrcvvaOxOkRgHO7CM0ONraYUn8sRJPSapi6i+dA2\nvuoVbu09PMxkmji0aTnYakoAMOSFCowtY6x6dHds1SUEvvPgv6u48ZNpPrC1X33psvKRDKBAjKX8\nOH6Pu0d+VYGE6Uuo/eK9fvc3EFqL9oUKjBOmIWx+E7G3ZEMRMOTNDNnmaqnvtSp5jOFFKhVY+2wq\nhgklYfcvnq/iredTSTRJ2Xvw9E3E/e6usaYwDmwhbjRhKT6Ku7UZhTEh7P7EhefTemjnkBVKC1dF\ntwPz/tgC2lgh0rgZ6jGTfNalQzpuehP1FAYT4X5r++PB6GysDfmtVCamIZErkGnjQtsPscAoCL3f\np9rGA6QnzyZ//AqKSkOL48UYXXi9Ipcuq+MHP9NTMFVBZpaU+AQJKrWAVCJgd4i0NPkoLvKye7uL\nj95z0FAX/Vzo6ccsfPg/B9dcr2HeQiXjcmTo9AJ2m0hjvY+dX7p59007hQdCc5aOFg7udbPq/AZW\nrdZw9oUq8gtk6A2BvwFLu0hzo4+Txz0c3De6csTVVvvYvNHBBSsCc4h1a22duQlHEwajgKXdz49v\na2HBEiUrV2uYNU9BcooUvz8g8u7e7uLdt+zs3dm/cfL6SzY+eMfOpVdpOPsCFfmT5JgSJPj9gWIu\nWz918sE7AwsbNxiFYbF5rPG/N+zs3ubi2hu1XPstLWq1BEu7n+YmH3t2uNnwviNI2D6w1x2VwHjh\nJWpuvkM3oEIpM+f2XuF+JFm/zoHT0czis1TMnKsgMVmCwShBIgmIiQ11PkqLvezf7WbzBieV5cMb\nQv+VERgBdBm5mI/t6dcxpikLh8ma6AknJoYrEjGaCRcKrE7OJGHGEpoPfnEaLRudGPJmYKspQaE3\noU4Kzp/kaKzGGcYLcKzgsbZiqy5BmxGcPD9xxrLAWOjHrKS3ojh94fe4aD2+F9O0RUHbTdMD49Ft\nMQ+o3/5gPrKT1MWXBFV+lql1JExfQtOBz/vVl1xnJH7KgpDtTQf7J9jGGDpu+0kjTzycxNQCBYeP\nBU/6rl2l45lHklEpBd7faOebt9WfJitB9PmxVZjRZscz/hvzaC9qwGvrslcil+L3DE2VwxFB9NP0\n5QbSL/1G2N1SjY60i1ZT9b8XBn0q/cTQxZ8OnA3VWE4eHvQ5YowQEcbNUI8ZfcGsXvcPZNy4Wxrx\nu51Ieniwyw2mznyJHXhtFtyt/YgUEP2BEOxuxW6UianIDeEFfEdt/ytIA/jFwDNGIkg7/w2gUvbu\n/XHkxFs4XW1MyDoPi62W2oZ9Azp3jJHD76fPCr6DoarCyyMPRtf/jOy+81Nv/MARVbuhxGrxs+Y5\na5+eS32x8ryRnVd0ZJEQRVj3+ugLjwZQKLsWW3Z+6RpyL0urVeT1F228/uLQFuDssHs4bB4KDh90\n9/l38tc/tPHXP7T1uj/akP+aah//+FN7VM+RX91j5lf3RH6f+9UDRq79ViC82Grx8+LTVrZ/7qKh\n3oe3hwanUAgY4iXkT5Jx7Q1aps4ICJLnXaTm4w9DCxqVFntH9PmxeYOTzRtGR+q5M7bISziBwNRN\n3IoG48Q5YQtujDS6zPyQbSMRxjnUNO77FFEMlvjTll4e4p35VaTnfYkbH0gsr58QGrJlHsPeix20\nHt8bsk0Zn4xpcqhI1huqhNQBC4wADbs3IfqChROpQkXWxd9EkA7/2ovP7aTlSGihnpSFF52qKB09\n6WetQiILDrP1uZ2Yj+4elI0xBs5zL7dz931NrH81nfHZXd/Nz++O56XHU1ApBZ74bzvXfLsWm/30\nLn2WvBgYh4kLxnHuO9/lrJdv4uy1t3DhxrvIvjrUM3a003p4T8RQTX3BLJKWXjyoc6jTsnsVMQEa\nPh0ZT+gYQ0ekcTMSYwYGOm5EHHWhhcPkBlNI/sWBVDW39/B4VCakhIRddzBQD0aHMzBnj9MFL6im\nJPYu4gOUVHxCY/NRJueuQq+LFQ8c6+gWTSPzT7cz4eXfohyf1uu+GKMLjVbgvIsCCxw7vnBRUz06\nFyVHY3GfaBirdo8FOsRFu03k+pWNPP53C/t2u6mu9FFfG/xfZbmXwv1u3n7Nzre/1kRzU2Duvvjs\nsRVVOhKcsR6M5e89S/43fhy0TZuWw4zv/5W67etp3Ptpr1UDJXIFSbPPJWXRikHbMeMHfwNErFXF\nWEoPY6k4jrO5LqpjFXoTqUsuRaYJDkMxH9tN28mDg7YNAEFAKleGzd0mSKXI1Dp8bmfEypzR4m5t\n4vB/7mPa7X/q3CaRK8n7+g8BsFYW0bhnM9bq4ojnk6o0JM+7EEPudBSGBA49du+YzUfYQeVHa8he\ncUPnZ6UxiYQZS0lbekVQu/biQzTt/2z4DBmh8dBcuA2/103WRd8M2p554ddJO3sVh/9zH72FzseN\nKyD9nKsGVXkaAiJ94X9+Tu41dwWFKmvTJzD9zofwuZ007f2U+p0bIvYj1xnQpuegy5xI3LgC5HFG\nXOYGjr/4YJ821Gx5G0vpEXKu/C4dYWxSlZbJN/8fHmsbNVveoq24kHD3QmFIIOPca4gbVxCyz1J+\njNJ1T/Z5/hjDy+PPtRFvkHBiRzYut4hSIeB0iXzztnrW/m/ovdDbjtSx4dxH+n1c7cZj1G48g6od\ni35KX/g7prlnkXrBVWGbJC29GMPk2TRs/RBL0cGofkNUSekkLr0ooueiu7WZ0hf+hs8ZupodY5TT\nx7hJWnoxSUsvxt3S0K9xk3nlTRHHDAx+3Jj3f4k2Oy9om75gZkghsead/c9X2LJ7C4mLLuj8LNPG\nkXbR6pB27UUH8doH9lyrqPkCi62GiTkr0GlS8It+mlqOcax4HYnxEyMcKXLg2BrUKhMLZt6O3dHE\nnsJn8PvPnAqqXyWs2wuxbi9k3GM/7te+GKeXfz6T0Flx966b+pdLPUaM0cBDv2ulrDj63w2nQ+TI\nQTdnna9Co4kpwD05YwVGR2M17aWH0edMDd4hCKQuvpSEGcuwlB3B2VSLz2VHIlciU+tQp2ajy8jr\nlm9GpG7bh6QuvmQQ1gjoMvPQZeaRRiC82dlci7OpFo+1DZ/bid/jQpDKkCrVKA0JaFLHo07OoGfu\nHJ/TTu3Wd/t1dlViOgnTl6CIi0eiVCFVqJAq1Kf+rQw5RwdKYxJTbv09EAhx9rmdeCytAXtdDnxu\nFz6njZrP10Vti9/jwuuwIVNrQ/bpsiaiy5qI3+vB0ViF127B57CBIEGiUCLXxKGMTw4RXM8E2ooP\nheSoTJpzXkj186HIvdgxHqTK7uMg8G95nIH+jAe/y4nP7QwaE5UbX47KDvOxPcRPWRDioStVqCi4\n6T7aiw/ham/G73YiVWpQGBKIG1cQIizW79xAyoKL+n8jANHno+y9Z8ldfVdIv1KFipRFK4gvmIez\ntQGfw4bf50UqVwRyuuoMKI3JISFo/cVScZyG3Z+QPO+CoO1ynYFxl30bj7WV9tIj2OvKkMhVyLV6\ndJl5aNLGEe678lhbqdr06qBsijF0PPAPMyqVwC/ujgfg2lvqWP/x6AwfOtNo2fM5qqR0jDPCpzpR\nmJLJXHkjXrsVW+kx7DXleFqb8TkdiKIv8HeuN6JKSkc7fiKq5Iyw/XTgd7uofPvZwYmLggSZRotE\noUKqVAWez0o1EoUSiVJ9apsaqTLwObA/fHLwDlIvuJKEBefhd7vwu534XKee3S5HYNupf3strfhc\np9o47QOrnCxIUBgT+mW7VK0btO2BfzsGbz9jc9yE8xwMV9zQWR/q6dgXXrsVT1sLckOX16IsjJf9\nYPMvmttK2XUwdGHsk22/Dfrscrez6YtfBW1zOFvYsuOPgzp/jJFBf/5cjCuXISgUiF4v5jc2Y/ls\n/4D7k6eaSLzlCuSpCeDz0/bBl7Rt2IlEqyLrz3fS9uF29OfPRaJT0/7JHlpe3TSEVxNj2iwFC5Z0\nzYPd7lGYfDFGjD5Qa/oX1KvTCcxeEBj3w53PcCxyxgqMAFWbXiX/+p8g1+pD9sm1ekxTF4U5Kpj6\nnRtp2L2J5LnnD1pI6ECm1qHLzA8b+hwJv9tF2fvP4bVb+nWc0pgUlPtwIAhSGTK1DlmPFwHR5+2X\nwAhQ8uZj5Kz6HvI4Y9j9EpkcbVrOgG0di4g+L60n9pPQrbJxz+rlXocVS9nRQZ9rKMcDYV4MoxUY\nASrWv0je136AwpgYtF2hN5E4+5w+j28+sJX67R9imrKw32HFHXjtForfeIyclbeiTg4NsVIYE0Ps\nG2rqvnwfQSojKcw1y3VGEqYvieo781jMlLz9Hzy2WOXa0cRvHmyh3eLngfsSuGy5NiYwjiA1H61F\nkMsxTO49nYJMo8MwdR6GqfMGfB6/20XF2idwNUaXR6g3dOMnkv217w6qj57IdIawglAk7NWllK15\ntN/nGmr7B2I7DNz+DsbauHGfEjilqsiFmnqmBYkWe00ZBkPkCpW9VbOOEaM7jiOl2HYdxWexo8hI\nIuP+7w5cYBQEUn50HY1PvIOrpAaJVk3Wn+/AVVKDu7YJWZIRQSGn4p5/Io2PI/vvd2P5bD+emqah\nvaivKCq1wP892PUuZ7XGxMUYY5NVX9Pw2gs2vN6+x7BGK/DgYyZ0uoCTx4b3R0few9HEGS0weh02\nSt76N+OvuGVA4ZT1OzdQv/1DAOwNlegy8/o4YvhwmRuo+GgNjoaxP4FzttRT9MpfyTx/ddjKt19V\nzEd2BgmMPWk9vndA1YVHM16HleI3H2P8Fd8JK+5FonHvp53evI76CuS66QO3w27h5Np/krb0chJm\nLotYuTIa/N7+V/ir/XwdrpZ60s++skd16+iwVByncsPL/V6AiDE4DnyaFXG/3SHSbvHTZvFTU+/l\n1hv0nLUo1ONs5rlj/9k+KhH9VL/3Mj6HDdOcs4blFF5rO5VvPzvgAhcxRiGnxk0kgXGwDPW4cdZX\noh0XKZx44Dhqyvu4F+KQV5COcWYiT0/CeMXSQGI5ESRaFYJUgujrf6ojebIR5bhUUn8SnG5Hnp6I\nuzYgIrZ9uB0An9mCp64ZmUkfExj7iSBASpoUc4sfl1MkTi9h9jwFd/1Ez6QpXTmmH/9bbHE7xtik\nYKqcZ15P5F8Pt7Nnp4uemcAkUsibKOes81R8/UYtqeldEYbFRaOrqvto4IwWGCEgzJ187R+kLFqB\naeqikEIIveFsqukUFwEcAxQYy959moQZy9BlTxywaFG79V2a9n92RolLPqeN8g/+izYjl+S55wfy\nyPUzi62jsRrzkZ1Dkn9RY8qg4OI7sDaVU7Tx9OSus9eV4zI3oIxPDrs/XEGQMwGPrZ2Ta/9J8rwL\nSJpzXp/imqu1kdrP1tFedqRzm72hEn3uwAVGOOWN+9k7tBzZSfK8CzDkzQjJXxXxeL8PW00J5iO7\naDt5YEA2tBzejrWyiJRFKzBOnIMg6fuZ4Wypp2HnR7QWDTzEKMbAmTKp/2LwQI6JMQhEP3Wb3sZZ\nV0XqhVeFVNsdDLbyIqrffxmvNfZidcYh+qn54JUhHzMwPOPGUTu8AmMk3OZmfK5Y3tEYfZN6z3VU\n/uzfeKobkRq0jH/y54PoTUD0+ii/86+BEtXdkGgDf7N+ezfvIpFYxYwBoNYIbNieCoDHIyKXh97D\nrZudrHl26HNLx4gxnOze7mLeokCE6ux5Cp5+NRG3W6Su2ofdLiKTgUYnISVFSrj6n1s3x7wXwzGk\nAmNr0T5ai/ZF3b76k7VUf7I26vbH/juw/Co+l4OaLW/TsHMj+tzp6DImoErMQKbWIlVqAm3cDtyt\nTdjrymkrKcRWXRzUR+3Wd/ud+xCgvfQI7aVHkCiUaNMnoE0bj8KQgEKfgFxrQCKXI8jkSKRy/D4P\nfrcLr9OOq6UOR2M11qqT2GvLBnTdHbSdPMDBf94zqD6GC1t1MaXVxcg0cehzpqJNG48yITWQL1Kh\nQiKV4fd5Eb1uvE477vYW3K1NNO75JGyl8A4UWiNelx2/190Pa0T8noHlbBoquhcGUSj15Ey+hPjE\nfORKHXPn3YHLYabo4Bu0myNP9iMxGseD6PNSv+MjmvZ/hnHibJLmnIdMq0eQSvF73HjaW7DXV9Be\nXEh7+VEQg13YG3ZtomHX0OTVcTbVUPHhi0hVGvQ5U0mYsRSF3oRUoUKQSPF73YH8ow4rLnMjTnM9\n9tpybDXF+D39GW/hcbe3ULnhZWq3voshfya6zDxUCWkoDYn4fV58DhuutkbsteW0lx7GXjfwsRBj\n8Nzyw4bTbUKMKGkt3IWt/AQp569CP2kGveWbjRZPu5ny1/4zNMbFGJUMx5hp+Hw9bYd3D42B3RhO\nD0JnfTWi14sgC//a4KiLee/GiBJRxGcORFroly8YVFeeBjOeumbiVy7D/E6gAKJiXGrMQ3EYCScu\nfvaJk5/e2dJT440RY9Rz+w3N/OQ3BlZfr6XDp0ShEMjOiSyRNTb4eOpRC6+/ZBsBK8ceI+LBeI6w\nEjkBj41qSjgq7h2J0zJDWIwLB8fFgGeP12GlpXAbLYXbRuT83fG7XVjKjmIpO3rKrkKOitF5HF0o\nrOaIuJsayobXyNOI126h5fB2Wg5vH3RfgkTK9Ct/RsnnazBXFEZ1jL2lmr2v/KrvhiPI5DnXY0iY\ngLWtiqa6wyCAWpOA095yWuzR6tOYe/aP2PrBL4etQqPP5aD50Jc0H/pyWPrvly1OO+ajuzAPQWGd\ngeC1W2g+sJXmA1tPy/ljRMcLr8VC0scSHksrVev+izIpjZxvfn8Anmki9qpSWg/tpO3wnmGxMcbo\novuYSZh7FvqCWf0cN8FjZriiURx9CIzulsYB9y36fTjqK9FkhM+PHQuPjhEtbRt2kvXw9/E7XVg+\n3YenvmtOm/KjryNPTUBqiiPlh1/H12al8Zn3cFfUBe3L+N13OrfX/WUNCTdeyrh/34sgleKuaaTu\nwZdO4xWeefi8sGeHi+wcGXqDBKkUzC1+Cvd7ePdNOx9/6Oi57h8jxpjA5RL5w32tPPu4hYuvUDN3\noZIJeTKMJikatYDHK+Kwi7SZ/ZSVeiku8vDlFlfYMOoYXQhiP54IgiAM+PEhR8EC4XxaaBhRgdGJ\njSLx4IicL1r6a9dXQWAcSvSpeUy6+HZObn4uaoFxtCGVqVi64vc47M3s3vwXRPH0Lwtm5p7DhMmX\nDavAGOP0I4piLH4oSgbzmxjj9CPI5GizctFk5aJKTkcRn4hUozuVpkFA9LjxOe24W5txtzRiry7F\nVnESr7XtdJse4zTS33HTuG1jbMyMYWK/idET+038ajHjsntpqThI1aENEdtJpHJmXP5Tire9gqWh\nZISs62Lm5T+jqWwP1YWnt4J4tPfrdDPrqZtp/uw4lS9+MeA+Zj/7HRo3Habq5ZF36hpppBoF89bc\njlQbCLUuvOdl2gur+t0GQKKUMfvp73DiofdpPzQ6BcoBlwAAIABJREFUF+6i+U0csRyMHtyIjOzv\nzkFxdA7q0WrXmYI+Y9LpNmHQqNSBqmxOW/OoEBcB4pOGJ7dTjBgxRg86QwZTF93Cif1raakfeNV6\npdrAguX3UXZ0PZUnNg+hhUOH6PVgLT2GtfTYaTm/QiXhW7/M5txrk3BYfWxZ28jLfw6dUHa0m3th\nPHEmWa9tx0/V8tOnJ/LUL0vZt7l1SGwcyj7vfiyPBRd3VSJ+6JbjHPhsaIS3cPfoe/OGZzF7JMbN\nQO77uMkaHvjfNJ79TRkfvxJL3RAjRoxRwik54kyqJXA6kOmUZH5rKYlnT0IWp8JW2kTZE59gOVwd\n1C5p+VQyv74IZYoBV30bVS9vo/GTIz36UiF6fficY79Ayem+Fp/dzY6rHkGZomfui7cNuE2AwB+L\n6B3bfytnfJGXGF8NDBkFpEw+C40pHblaD0Deed8Oabf7hZ90CnYypZbZ1/0+aL+1oZSj6x8LOU4Q\nJMy78S+U73gbn9tB9oJV+H1eKna+gzFrKqZx03G0NVC69VUcrXVBx0rlKlKnnYdp3AyUOhN+nwdr\nUwV1hZ/QXnuys50pZTI5k1agUOmRK7RAQNQ7+/KHOtu0NBylcOdzIfYpVQZmLbsLhUKH22Whqa6Q\nihOb8LjtIW1VGhMpmXMxJReg0piQyVS4XRbMjUUUHXwjqG1S+izSxi1Ep09HJlcDsOzSB4LaVJz4\nmLLjH4WcJ0aMGAHGZcnIy5Gj1Uiw2vwUl3oorxpZL2CFMg63K/pw7sFWUh9OMvPOpaFyT7+uZzRy\n+a1pXPCNZB6/t5g4k5zqk+GLZHS02/hSPSf3WyO2BZDKht7haij6fOoXpaz9WxXzLzZx7T2ZQ2BV\nF+Hu0ZnAcHyXMWLEiDGS+L0e9q8bWB2FGF0kXzIT3cRUTjz8Ae4GC+mr5zP5/tXsu/kpPK2B9z39\ntEzy772M0n9vwry7lPh5OeT/7DKc9W1BQmTOXRfS/NlxWr48cbouZ8gYLdci+vp2pOurjd/lYc8N\nYz+396gQGLXoWSxcxH7xC5qo7dw+S1iGDDm7xS7vh3OFVRwUt6FCQ45QgBINTuzsEz/HQVeizQuF\n1QDUU8khcUfY80qRkiNMIYVMlKjx4cVGO4XiTpyECjPh0GFgkbCcfeJWmgkWlmYJy9gvduVPWyqs\nQI2uT7uSySBPmIYKLTbCVxjUYSBfmIGRRET8NFHLcfEAHoKLlJhIIVeYShwGaijjhHgIH2deeKvo\n92M312A312DMnILamIq5/BBOS3DOoe5etD6Pk5LP1yBTalEbU0iauLjP8xizpqA2pGBtLMeQUcCE\nZd/A47BgqS/FkFFA7jk3ULjuL53t5Wo9BRffjsqQjMvSRGvVEWQqHfrUPAzpEynb9iaNRQGPVrej\njYaaQF5OuUJL5oSzcdgaqavsSgbvsIUmrtbokpm55HasbdW02I+i1aeRkbMMU3IB+794LERkzJxw\nNunjl2Btr8HcWITP58ZomkBq9gJOFr4TFP7s9dgwNxZhbiwic8LZyBVayo5/FORV2d5S1ud9ixHj\nq4hEAse+zCZnXKjYUVLm4e9PtPHkC23DnhhdKlMhV2ijEuSsbdXs+Oj3fbY7nYyfvAJz/bExLzBO\nW6KnpsTJ1nXNUbV7/neRCzqVHbZxx+LoC+1Fw1D2abf4sFt81JYOfdXFaO/RWGE4vssYMWLE+KrS\nn5Rwo5WaN3ZS++YuRH/gWkr/tYnkFTPQT8+i+fPjAMRNy8RR3ULtuoAHf221mZTLZ6OfltklMAoC\nxtnjaP7s+PAaPBL3fKSuJUa/GBUCo412WmkmQ8ihSQwIjHIUJJDMsTCFULKEPKRIOS4ewIsHI4kh\nguAW8X/MFs7q9ZwCArOEs4jDQIl4hHbMyJBjFBJx0btnQE+stGGljVQhi2axS2DssL87X4gfIkcR\n0S4dBqYLi6inkqPiXlRomCLMC2qjRss84VzMNHFA/AIpMvKF6cwUFrNb/LSznYlkZgvLqKGMk+Ih\nJgtzmS0Y2CNuGfFw9eGmvbaI9toiABQaI2pjKs0luyPmYBT9PppLAg9gjSkjKoHRkD6RwnV/wdFa\nT/aCK0mZfBaH3/s7PreDSRd9D33aROTqODyOwItvzpJrURmSqT30MVX71nc+bLUJmRRcchfjFl5F\ne20RLksz1vYarO01AXt0yWROOBun3UzlycjhhQWzv4FcoeHQjqc7t42beBHjJl5ITsGlIV6JFSc+\nprrsCxzWLvFVEKTMWPw9ElKm0FjblRvU3HgCc2NgRSg1cx5yhZaq4i2xHIwxYvSBVAqvPJkaJC6K\nIginHJImjJfz6J8SOW+Zmuu/V4dvGKMh4pPycFjPnKqao9m7sj/oE+S0NvYd0hNtu68ysXsUI0aM\nkSBj2oUk5S5EodYjSKT43A6ObX4KUfQxbcWPOLT+b9jNNZ3tNfHpTL/kHgo//AdKnYmEcbPxe12Y\nsmZQeeADMmdeQmv1UU5+0b/CNCp9ElkzL0GfkofP46L++FZqj20BQKpQM2/1/ex75w9kzrgYU9Z0\nQKDu+OdUHfwQCJ8PcMZl93Lw/Yc7P6dOXEbKxKUodPH4PS5s5hrKd7+No70rBYNEKid3yfXEZ0xF\n9PvY/78H8Hm6FpHmrr4fmUIFCBzf8iyt1cGhujMuu5fao1vQp+QRnzktbB897znQed+tzdFVsJfK\nFOQvuwFj+hR8PhcNJ7ZRdXADdHsfThg3m4xpF6CMS8Rta+Xgew91OlSEs7O19hhlu94KsrX79yKV\nKXE72oO8N3ver3B99IoYLJR2/rubo7ujshm5UYNEKcfv8iBVK1Ak6LCXBeaA0/52PdrcZKRqBQW/\nvarzuKqXt1Hx/Odddipl5P/8ckyL8xA9Psy7Syl5dCM+W7AjUyQkKgWTfrWK+IUT8Dk91L27n8oX\nt3be8sTzJpN5/WJU6fG4G9qp/+AANd0EVIC0K+eQunIOyhQDPrsbe3EDJf/ahKOyOepr6YtI5+gg\nGlsHy4K3foBMqwIBjv76Tcw7ioP2z3rqZmrW7sQwaxympflhv5fM6xeTcslMFAlaBFngb8VrdXHk\nF69jPV4bcs7hYlQIjADVYglThHkoUOHGSQpZiAQ8/XqiJY5t4oZOkayV0BcnD2789O4Wkkga8SSy\nX9xKUzfPww6Bsz/UihVMECZzlL34CbwlJpMZVsLry65sIR8XDg6Luzqvbyrzg9qMFwrw4eWQuK2z\nL4/oYp5wHiaSaSHw4J8gTKWNZo6KgUqXB8VtLBKWk0QGDYQmFo3RNy6rGUdrPQCWumJSJp+Fzx0Q\npO0tNejTJqLQGPA4LKiNKRgyJ+Nsb6J634dBKzm25iqaTu4kedJSEvPmB/YPAINpAjpDBk11wUJq\nZfFmsnLPITljNicK3w7Ke+J2WaCH548o+mio2oNalzQgO2LEiBHMrTcYuOpSLY882crLb1o5dsKN\n3SGi00ooyJdz/dVx3P5tPVdfpuXWGwz85/mhLwKRM+VSElKnotYmdimbp9j67s+DPJHzZ60mNXtB\n5+fje1+loSp8DjtDwgQy885FH5+NVKbE47Fjba2i+NA7OO3miDadtfKhkLyMgkRKdv75JGfNRaky\n4HZZaao5QNmxj/D7ukSjjusBmHPePRGvR67Qkj1pOaaUyShUcfg8Thy2JuordlNXsTOijQNBZ5Sx\n+oeZzFtuJC5eTlO1i82vN/LBs3X4u4XEXHVnOuesTsKYrECuEEjLUbHmROC+/+yyQ1QVOfps17Mt\nELTv8XuLI3pFTl4Qx+W3ppE3W4daK8Vi9lJ62MbDtxZ1trn1gRzO/VrX70FvfUqlAmddnciSyxPI\nyFcTFy/j41caeO3hKpz24c0h1Nc9+mb+zl5tbG30sGeTOcTOm+8fz7gCDU/8rIQbfjWOnGlaTuyz\nsuZPFdSV9d/z0pSq4JKbUlnzYOhL8G9fn4IxWc6PzjvQOT2I9r7Hpyi48dfZzFhmQBRh/6etveZd\nlMkFVt2ezrIrE5HJBbZ/0MLaf1ThdgTPRaMdwzFifFVJHD+X1IJzOLLhURyWRuZf+wD73/0TXlfA\nycXaVE5y3iLKdr3VeUxSznwcbXXYWqpQ6kwYMyZTuuMNHG31ZM28lGOfPMnk5XegOZIeJExGQqHW\nM3X5XdjNNZz4/L9IFWq8TltIu/xlN9BUtpfaI5uRqXT0J1xCn5xL1pzLOfn5C9jb6pGrdBhS8nE7\ngucqKZOWUXv0Uw5v+CcKjZGMaRdSse+9zv173vg1EpmC+dc+0PMUnYyfdxW1x7aE7aPnPU/JW0z2\nnCuC7ns0pExaRuW+96k6+BG6xHGMX3ANLmsLjSW7ADCkTSJnwWpKd72BrbkSlT6ZtCnnUXP4417t\nzFtyfZCtPb8Xt70NtTE14v3q2Ud/MC3KA7+f9gNdvy8tX56k7fwKJv/hGir/u5WsG5fSuKmwU6w6\n+su1CAoZC974PkUPvNu53e8J/r1OWzWXmjd2cugHL6JI0pP/s8vJun4xZU99GrV9aVfOofzpLVT8\ndytxU9KZ8IOLcNW10rChEOO8HHJ/eDHFj2zAerwWdaaJvHtWIMgkVL2yHQD9jCzG3XoeRfevw17e\nhDxeg2HWONxNln5dSyT6OgcQla1Dwc6r/4lUJWfh/37Ua5ucu5ZT++ausN9L0gVTSV89n0N3r8FR\n1ULq5bMY/93z2HvTk3jbo3eeGwpGjcBYTxUTmUka2ZRTRKqQRRM1eAldkW6mftAeePFCEl48QeLi\nQKmjgjymkUQa9aeEu1QhKyRkOhriMGKmKeL1mUiihcYgobId86nj42mhAQkSDJgoFg93trHShgsH\nJiGZBjEmMA4Ej6MrZL1DWOzA73UDIEgDHkv6tHwALHUnwxZqcZgD40NrGnguKmPiBADamkuDbfF5\ncDrMaHTJqLWJ2C31ffblcrWjPFVcJkaMGIPj5uvjALj3/4LFAavNz+79Lnbvd3HgiIun/57MzdfH\nDYvA2Fh9gJaG4ySmTcfaVhUk/vUMFyopfJeqk1swJuaRN+Oqnl11kpw5h0mzv47T3kJVyee47GZU\nGhOGxAm4nOFTenSQPmEZ5cc+6lH0RWDKvBswJuVTU/oFNks92rgU0icsQ2fI4OC2JzsXZzquZ8aS\n71G0//WI1zN5/g1odMlUFX+G29mGQhmHITEXhUrf123rNyqNlN+8OpmEVAXrn6+nsdJF3iwt1/0k\ni6xJGh6/t2sVesf6Fo7ttgJw+18mYGvz8sIfAi8HjZWuqNr1bAtw70UHmbZEz02/HR/R1mVXJnLb\nnyfQUOVi/bN1NNW4ScpQMHlh8H158Y8VvPdUbZ99+nwiF1yXTGO1i3efrMXW5uW2hyYgSASe/21Z\nRFsGS1/3KJKNkxfGcdENKWHtzJyo5pcvFHB4u4UNL9Rz2XdS+clTE/npJYfwefs392ypc7P4ChMv\nP1RB92lAUqaS/Nk63vl3TVAUWTT3Xa6UcN+LBSSmK3j/mToaq1zMPNvAHX/NDWkrCHD3Y/lMW6zn\noxfrEQS4+MYUcqZq+OONxzpt6s8YjhHjq4ouaRy25spODz6JVI7akIKlITAHry/6kvHzr6Zy3/v4\nvC4EQULC+NnUHv20sw/R56WpdDcKjZGsWZdhaSrDbW9FqTVFLTAm5y9GRKTos+fxeXv3KGuvP0l9\n0alKwO39K/wkkSlABLfDgsvajMvajLUpNAWFrbmSqoOBPOyOtnoypi8P7ayPYpW2lqpe++h5z5vK\n9jJ+/tVB9z0a2utOUH/iy8A52huIS55AysQlnQJjxvTl1Bd9QXNZIDWF09JEzvyrgwTGnnY2le1D\nlzS+c3+478VpCXaC6nm/evYRLTK9mpw7LqD6jV142rq9j4oijR8fJvcHF5F29TzUmQlUvtBVDdrn\n9CA55XXnd3vxOdxh+7cW1VLx30CaN3t5M02bjxA3NaNfNrbtK6fu3cD9dFQ2o5+eRerKOTRsKCTr\nW0uoXbeXplPFZ5zVZureP0DK5bM7RTupWgGiiLvFirO2FWdtK5YjXX8j0V5LJPo6BxCVrUNFXx6R\nthN1vX4vcVPSsR6v6/S8bPz4MBO+vxxNdkLYitXDyagRGP34qKOCdCGHerEKI4kcEMOXR3eL0bvn\n9oYcJS6GJg+QCwdmGkkVsqkXq1Cixkgih8T+DzoFSrxE/gORoySNbNKE7JB9SkEFIkiRIyDg6dGX\nGxcKFP22K0YAv7crNLgvkVuhCYh1SRMXkTRxUa/tpEr1gO1RqgLnyJ16BblTrwjbRiZTBX0WBAlJ\n6TNISJmKNi4VmUKLVKZAKpFTcfKTAdsSI0aMLiblKWi3RJ5Uv/i6hYd/l0hB/vA8k61tgXw7On06\nVnMVNkvvi14+rwuHtTGiACeVKcmdfiUOWxP7tjyCz9ft96Uo/DEdiytp4xeRO20ln//vp0H7E9Om\nYkqdwtFdL9JUe6hzu8vZTu60lSSkTKG57nDQ9QARr0cikWFIyKGi6BOqTn7aub2q+LNer20wXPad\nVDJy1dx//VGO7Qqsem95s5HGKhdfvzeLre80cWhrQECuKXFSUxKYe7idfmztPo7uCBVmo23XQW2p\nE2Ny5HGk0kr5f78ZR125k/tWFeLq5sH29r+CJ9ROm4/aUl+ffQL8+prDQZ/PuiqRecuNPP/bPg8d\nFP25Rz1t/PztJhLTlWHtVGmkfLq2kRdPiZV2q48b7ssmb5aO47v7n/czPlnBlIV6Dm/rsm3pygQA\ntr4T/PIZzX1fdmUCaTkqnvxFKVveCKQ6+XRtI3f9PZfFlycEtZ23PJ455xt55K6T7PyoBQiInjf+\nehxzzo9nz6aASN+fMRwjxlcVR1s9puwZKLRG3LbWwGJ+W1e6oeaK/WTPuYKE8bNoOLkDY8ZkZEoN\nTWVd0QAdXnfdvfNFnw+JNPpXck18BrbmyojiIhBWEIyW1tpjNJbsZOrF38fWXElj8U6ayvbg9wWn\nSLI2B0cayuTB7xzREKmPnvc8PnNqyH2PBkdbsMBqb605FToeQGNIIy5xPOlTzw9qJ0iknVFgPe30\nue1BtkbzvfTVRzQIMikFv7kSZ00rlS9sDdpnWpJP7o9WcOjuF3HWtGKYmU3B767m+P3raNsf/Xiw\nHg+eX3ktTqRaZb/stFcEL7DbShpIWBpwvtHkJBE3JYPM60LfjwWZBNHrx7yzhIaPDjH9kW9hPV5H\nw0cHafz4CH730KXpiuYc0dg6UkT6XuzlzSScPQllsh5XQzumJfn4XV4clS0jZl8Ho0ZgBKgWS8kS\n8kghCzeuIfEu7A0vHhT07w8lEnViBQXCHGTISSETH96ggjXR4saFjMjVDz24aaaeMvFY2OMhcH0i\nYoiYqEBJGyM/0M4c+uG9cCok0d5SHXFV0mWJnOA/Gprrj4QtAAPgdnW91AgSKdMXfgdjQi52Sz3N\nDUdx2prxep3o9OmDtiNGjBgBPB6RFnPkMA2/H8xmHwkm6QhZNTiMiXnI5CpKj7wfLC5GwOd1kZI1\nl7wZV1FxfFPI/sT0Gfh87k4RsYPWxoBiaUjMDdnXF36/F7u1kdTs+djaamiuOxzWi3yomLc8nupi\nR6cw08Gmlxv4+r1ZLLzENCrEmWlL9GjipLzyUGWQuDjUVB63M3WxHolUGNWhtZHs/OTVrpfXkkMB\nL8mkTOWABEaXw8/SlQlBAuOSlQmcPGAdUMGbaYsNiH7Y9l7w3GHH+pYQgXHBJSZcDj+7N3V5+xZ+\nEbBjyqK4ToFxrIzhGDFOJw0nthGXNJ5ZK3+Jz+3kxGf/xeOydu4X/T4ai3eQnLeYhpM7SMyZS1tt\nUVD001DkwBcgqgIa3UXMfiOKlO16i5rDn5A0YT4Z0y8ibcp5HNn4LzzOrueEvw+RMxoi9dHznrvt\nrSH3PSp6pIkBoeseCgJSmYKqgx92ejR20D3FVF/XGs33Muj7JUDej1cgN2o49MM1IeJW+ur5NG4s\nxFnTCkDbgQrqPzxI5jcX90tgHIg3YIipPe65wKnbIwhIVXIqnv+chg2htRI6r0kUKXl0I9Wvbifp\noulk3bCMjK8v5NCPXsZjDk0JMCD6Oke0to4Qkb6X+vf3o5+awZz/fhefzYWr0cKx37+Npy36VAJD\nxagSGK200UYLKUIG9VQOayGSVrGRLCE3KGfhYKinigJmk0Q6SUIG9VRFzLXYG1baiCcJAaHX6zfT\ngB4TVtp6bSPip5UmEoQ0Sk8JkToMKFFjFvu36hNjYLjtgcm4tbGc8u1vDss5OkISW5tOUl26tY/W\nkJo1H2NCLubGExTufCbopVsikQVytcWIEWPQ7C90sWBO5FVptUogPU3Gtl1DX1V3OFBrA+KFwxr9\nb2acMYvkzDk4bM1UFIUKjCptAlKpgmVXPBj2eLlCMyBbj+56kYmzr2Xy/BtwO9upr9xDTekXuPsI\n4x4IydlKju4MFZ7sFh/WNi8p2UO3mDkYkk/ZUX1yaHPxTJiu5cLrk8mdocOQJEdnDEwte77OnU56\n2qhUS5ArA8WCwtnZWN31Euh1B+ZZcsXArmjXBjPzLzbx3G/L8bj8jJ+iISNXPeAQclOagvYWD25n\n8ByzpS70xSM1W4VSLeHFY/ND9ukMXa8AY2UMx4hxutGasjj2yRO014dPG1B/4kvSppxHXFIOxowp\nlGx7dchtsLfVkTRhPhKpfMAios/jQqroO4LKbW+lunAjtce2MPPyn5GYMzco5Hsk6OueR4NaH1x4\nVRuf3hW+LIrYWmtQG1I7390GwlB8L30x7pZzMMzM5tDdL+G1hs4dJXIpojd0cVsi71rIFv1+EEGQ\nDm/BPHV28IKXNjcFZ3VL4H4XN6AZnxiU67A3XI0WqtZ8Se2bu5j97HdIunAqNWsDOZaH6lp6PUc/\nbT3d6CalcuTnr9N2ILriR8PFsAuMucI0cijo/KwhjgwhkDfukLi9M2dhB3vFLSwQLuB4mOrR0ZJN\nPhOFmac+JZAiZAFwVNxDNafyZFBFs7iOXGEqU5iHEjVe3LTTSqG4IyS0uC98eDkobmOWsAwndvZ0\nq+YcalPvdh0Wd5FCJouFi1GhwUY7doIH9BFxDxp0TBMWYiIZKTJcOGihgSJxP75ThWb2iFuIJ4n5\nwnnEEU8t5WwR/9fvaxtrOMwBz9H47BkRq0gPN/VHPkOfmkvypCUAVO76X9APjiCRYsiYjKW+OCSf\nY7SUF23AlDyJ3KkrQwRGqUyJSmPC1t7lSSuTByYTltaKIHFRZ8ggf/rVVJX0HkJos9Sh1iVhTMqn\npf5on7ZJpQpmLLkNrS6F8hOb+qyGHSPGWGRSXqjHudcH3/5+Azd/U88L/0rh3Y9sFJd5cDhFNGqB\niblyrrlCx6oVWn72+2b+9njrMFs5NIt1HTkPVdqEkLyvvR/Twtb3fsG4SRex7IoHQ0KknfYWdIYM\nvnz/V0Nand5uqWf/Z492flaq45mx9HYEQcKuTX8asvMA1Je7SMkOFZM1eik6g4z6isF7eAwFTVWB\n3/7UcaoBeeKF4+fPT2LS3DhuX7QPpy0w97jz77ks6eFJdzoJZyNEtrNnAZTB8J+fFvPo57O482+5\n/OPOE9zz+ER2bzSzcc3AFrdb6tzkzdQiVwh43F1/26bU0LDqhioX46ZquGXG7qC2PRkrYzhGjNOJ\nIJUh+r1MvuB2AJyWRir3f0BLZVd6D7e9jWOfPMmU5XfScHIbzeUDf5/tjcr9H9Bcto/8ZTcQlzwB\nr8tO7dFPO3MMRsOxzU8xfv5VzL3m9whSKa1VwdWd4zOmkDljBUpdAoJEisvWQu3RT6k7Hn113syZ\nK0grOAfJqdz0k865GdHvw9FWz6H1f4uqj573XBT9uKzNIfe9L5pKd5O37Abi06fg97mpP/Elxdte\n69xfuP7vGNMnM/Wi76OJz0D0+yjZ/hotlQejPkfP70UileO2t7H/f70XuOkPs5/9DupMEwBz19ze\nud1e3sT+W58F4NAP15D5zSXMeeF7KExa3I0W6t7bz6EfrelsL3r9HLpnDRPuWk7+zy7D2+6k4rnP\naNg4tO/MjRsLmXjfSkyLcvE5vdS/t48TD78PwIE7/kv8wlymP/IttLnJiD4/jsoWql/bTvPngQgW\n06I8sm5ahirNiCCV4Kpvo3rtTmrf3jNk1xLNOfqyNefOC0m5ZAYSRUBWm/a36xG9fixHqim89xWA\nPttk33QW6dfMR6IM7J98/zWIXj/2iiYO3PZ8VNcikUvxe3xM/ct1gXvjF3HWmKl49jOat/aSx2iY\nGHaBsVgspJjoB6yISK3Yu+r6qbiuzz4qOEGFeKLPdl48HBf3c5yhefg3Uccm8Y1B2QQB8bO+jyIs\ndqxR5Xg008gu8asl7DQV7yZt+gUk5M5FZUzB2daARCZHKldzfMPjne2MWVORq3RIFWpU+kDVRoXO\nRPqM5fg8DnxuF7aWqk7BciCUfvEaE5cbSJ60hPjs6djNNfjcThQaA2pjKlKFioNv/nHAAiPA0b1r\nmLH4e8xedhfW9loEQYpKbUQfP47G2oMc39/1A9pYc4Ds/AvIyjsXpdqIy2FGrU0iMW16rxVjO6gt\n305i2nQmz/kmTTUH8XqdyOQamusKQ6pYA2jiUokzBArYJGfMjgmMMc5ICj8PzYXbk29cret13w++\na2DerP/f3r0HR1mdcRz/7mZzz+ZK7nckIAKmEVREq5VSRKTW8VJbiraiTseOHTujo2OntmOndOpM\nZxxtnanUOoiOzuhMcSzIRa0CIlcJSSQhQghJNjHX3dzYzd77RyRhzSZZNsSE9Pf5L+973rPPe7LJ\nSZ593nOiWfvL8TdiCpfbdZaoGPOYazCGorvzFF6Pi9zZ36XDciykhOC5DzIaaj8gPikbc0o+fbbh\nNYg6WypJzyklp3jZKGskGgiWIB37fkZe43TY6LWeIT23NPglE3DgfSv3Pp7H/GsTA9YAXLE2E4DD\nu6bHsiRf7O9hwO5l1S8y+WxrF27nxJNoJpOrJtmsAAAIMElEQVQBe583IHG3cFnShPu9mILFmJBs\n+tbi9Ptg33tdrFyXyeVXm0nLiWLzhvDXRqve38vS1alctyaNPf8eXhrlmlWpI9oe3G5l6epUfnBf\nJu//K/DnxXDeE4KXyntYZKoYjBEsWPlrbM3VfLlnEz6vm1nFi7ls2Vq633kmYD48t+ljx+kjAX1Y\nGyuxNg4mrDzOsxx88wkAKrY+d8Hx2Lu/onb3q0HPeV2Oob5H43UPUPfZW6OetzVXY2uuHvU8QOW2\nv455zFKxA0vFjrD7CDbmRqOJ1MLSoOM+mnPjO15Csrulhu6W4AUUweK0VO3CUrUr4NhY35dQ+xhN\n+fpXxm3j9/po2vzpiLUZv6nveDMVj2wKeu5csvJ8Ta/vo+n14HtjBHMu1rESW7aDdUM7PwdjPXAK\n64FT477WWPcynlBfY6xY61/6kPqXRj6hcyFtGjftpXHT2Mn78b4vi15Yh23/KU78YQs+lwdjVASz\nbryckqfWYDv4wgXtrj1R0+oRaYAM8mjhzFSHIZcwt6OXEztfIrdsNfFpecSlZONx2rFbmwPaFS69\ni6i4wH8wouKSyC1bNfR16/FPaDryn7Bj8Tjt1Gz/OxlzryO1uIyE9CKMxgjcjj762+uxNVZNqBwf\nBiuAju55ge9c/ysy8xbj93lwDvTS1nyU1sZDI9pWHdhI0bxVpGVegdFowt7fRu2xt2lvLqdo3spR\nX8fWeZLqI5vJL1nOrOxFGAxGnAPddAVJLgLY+1rp67EQn5BJe3P5hO5RZKbKzTJxz+0Jk5pg7Omq\nZ+HSB7HU7cHv82CKjKWlfrjSwWCMICrajCkyhnjzYEIhNn4W8YnZeD1OXM4+fF43HreD08ffo6T0\nLspueoy2ps9xDfQSFZNIcvoc6iq34Dg72pqyfmqPvsWS5U9SvufFoUeVO1u+oLOliuIFtxGXmE1v\nVz0YDMTGp5GWtYCq/RtxOgJ/R/q8bmYvvH3U+zGn5FNSejfWtsE1Zn0+DwnJeWTkldHRfPGrSXZs\nauXaW1N54uUStm9qo71xgDllCSy/N4OD261U7J7ctesiTAZyLoslr2SwQj2zKIaCy+Nw9Hvp6XIP\nVePZe728saGRB/9UzIZ3F7B3Sye2NhfJGVEsXJbIXx6oHeovOSOSOLNp3D4rP+1l/rWJ3P9MIRW7\nu0nLiaanw0ViauCfl5FRBpLTo4g1R5Aze7BSLndOLLZ2N44+L92d7ouS8AwmWIyr12cFjXOy7N3S\nyZqHsrn3iXz6ezwc+3hk1fJY497aMDA05nvf7eS2h7J44Nki0vOj6bS4uGKpmdmL4kf0eXinlUM7\nrKx9qoCCuXHUHu0jqzCGxStS+PP9J4Yeq57q97DIdBeXnENccjbHd/1tKIHY3XKC/NLVmKLjcJ23\nzmJaYRmOnrYJbbIiwcccRh93kf9nccXpVD32Bt6B4aclbYdOU7D+RkyJsbi6LnDd0gmYFglGE5HE\nYSaGWEoMV7LXv3WqQ5JLnN3awsmPxv6kp+KdP4bcn9/v4/Brjwcc62utCzjWfGwnzcd2jrzW66Gt\nZi9tNaE/VmDvb2fP1ifHb/g1j9vOkU9GfjIWTK+tkcoDG4OeO1M7Mv7zdY5SrRiM1+uifO+LIbUV\nuVRlLzgz1SGMy+noxufzMGfRHfj9fux9bQEJuaS02Sy67uGAawrmraBg3goA6o9vw1K3G4DWhkM4\n7d3kzbmJgrnLMRojcbvO0mtrwOMeuxLb63FhNJq44ur7qdz3j68rD/zUfP4GOdbrySq4mvTcUvw+\nDwOObrraqnG7Ri5OXX34dYrm3zLq/TjtNhz9HWTmX0VktBmfz4PTbqPhxE6a60L/PRwq14CPDetq\nuPs3edz843TMKSY6m52887yFrRvDr4AP1fxrzDz92vBSNHc+msudj+YC8OZzTWx7ZTiGj9/uoLPF\nxZqHs/nRIzlERhvps7k5Wd4/an9j9bntla+IT4rg+h+m8f2fZtBhcfLP39bz+7fmB1y/ZGUqjz5/\nWcCxnz09XP378lOnA6rxLqZgMW5/tRXLKceIOCeL5UsHDTV25l6VwIdvtuNxj6zKHWvczx9z14CP\nDfed4L7fFbJ6fRZ+H5R/0s2zP6nh+f9eGXC93w8vPnaKlesy+d496SxZmUJXi4ujH9no7xmu/Jnq\n97DIdOc824XP6yZr3g101B0G/BQtuZOzVgsuRy9GUxSRMWbiUrLJKFnK6QNvT3XIl7xgYx6bmEle\n6a1D4y4ig3xOD9l3LKZtVxX4/MQWpFH4wI30n2z9VpOLAAZ/CLtQDTU2GCZl1xUzySw23ARAOxaq\n/Z+Pc4WIiEwGv98/nfZmmNYma04UEZHpQXNi6Gb6nJiUNZe8K28hNikLY4SJjrpDWKp24h7ox5xe\nzLybH8LnddNWu4/mLz6Y6nBnhG+OucvRQ09L7dC4i8ig5KuKyP/5DcQVzcIYacLV1YftcD1Nmz/F\n3X3xdpIOZU6cFglGERGZHvTPVOg0J4qIzGyaE0OnOVFEZGYLZU6c3P3JRUREREREREREZEZTglFE\nRERERERERETCpgSjiIiIiIiIiIiIhO1Cd5HuBBomIxAREZlyhVMdwCVGc6KIyMylOfECaL1KERG5\noE1eRERERERERERERM6nR6RFREREREREREQkbEowioiIiIiIiIiISNiUYBQREREREREREZGwKcEo\nIiIiIiIiIiIiYVOCUURERERERERERMKmBKOIiIiIiIiIiIiETQlGERERERERERERCZsSjCIiIiIi\nIiIiIhI2JRhFREREREREREQkbP8DQqXbobkKmZ8AAAAASUVORK5CYII=\n"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "%matplotlib inline\nmaxWords = len(brand1TweetsKeywordsPandas)\nnWords = 20\n\n#Generating wordcloud. Relative scaling value is to adjust the importance of a frequency word.\n#See documentation: https://github.com/amueller/word_cloud/blob/master/wordcloud/wordcloud.py\nbrand1KwdFreq = prepForWordCloud(brand1TweetsKeywordsPandas,nWords)\nbrand1WordCloud = WordCloud(max_words=maxWords,relative_scaling=0,normalize_plurals=False).generate_from_frequencies(brand1KwdFreq)\n\nbrand2KwdFreq = prepForWordCloud(brand2TweetsKeywordsPandas,nWords)\nbrand2WordCloud = WordCloud(max_words=nWords,relative_scaling=0,normalize_plurals=False).generate_from_frequencies(brand2KwdFreq)\n\nbrand3KwdFreq = prepForWordCloud(brand3TweetsKeywordsPandas,nWords)\nbrand3WordCloud = WordCloud(max_words=nWords,relative_scaling=0,normalize_plurals=False).generate_from_frequencies(brand3KwdFreq)\n\n\nfig, ax = plt.subplots(nrows = 1, ncols = 3, figsize = (23, 10))\n\n# Set titles for images\nax[0].set_title('Top Keywords for katyperry')\nax[1].set_title('Top Keywords for justinbieber')\nax[2].set_title('Top Keywords for taylorswift')\n\n                \n# Plot word clouds\nax[0].imshow(brand1WordCloud)\nax[1].imshow(brand2WordCloud)\nax[2].imshow(brand3WordCloud)\n\n# turn off axis and ticks\nplt.axis(\"off\")\nax[0].tick_params(axis='both',which='both',bottom='off',top='off',left='off',right='off',\n                 labelbottom='off',labeltop='off',labelleft='off',labelright='off') \n\nax[1].tick_params(axis='both',which='both',bottom='off',top='off',left='off',right='off',\n                 labelbottom='off',labeltop='off',labelleft='off',labelright='off') \n\nax[2].tick_params(axis='both',which='both',bottom='off',top='off',left='off',right='off',\n                 labelbottom='off',labeltop='off',labelleft='off',labelright='off') \n\nplt.show()", 
            "execution_count": 42
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "_____________\n\n<a id=\"enrichpi\"> </a>\n## Step 8: Enrich the data with <a href=\"https://www.ibm.com/watson/developercloud/personality-insights.html\" target=\"_blank\" rel=\"noopener no referrer\">Watson Personality Insights (PI)</a>\n\nIn this tutorial, we will create personality profiles for a sample of 100 users as that is the limit of what we can run with free plan for Watson <a href=\"https://www.ibm.com/watson/developercloud/personality-insights.html\" target=\"_blank\" rel=\"noopener no referrer\">Personality Insights (PI)</a>. \n\nIn practice, you can run personality profiles for all users, or you can choose to run personality profiles for only a selected subset of users; for example, for users with most negative sentiment tweets or users with largest number of followers or posts."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Number of unique users tweeting about justinbieber:  533\nNumber of negative tweeting folks:  163\nNumber of positive tweeting folks:  201\nSample size:  95  Fraction:  0.169946332737  Seed:  91\nRecords in brandUserSampleDF:  107\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }
            ], 
            "source": "import random\n## Aggregate users by sentiment\nbrandUserSentimentDF = brand2TweetsDF\\\n                        .groupBy('USER_SCREEN_NAME', 'SENTIMENT_LABEL','SENTIMENT',\\\n                                 'USER_FAVOURITES_COUNT','USER_STATUSES_COUNT',\n                                 'USER_FOLLOWERS_COUNT','USER_FRIENDS_COUNT')\\\n                        .count()\\\n                        .orderBy('SENTIMENT', ascending = True)\n\n## Get negative and positive tweeting users\nnegativeTweetersDF = brandUserSentimentDF.where(col('SENTIMENT_LABEL') == \"negative\")\npositiveTweetersDF = brandUserSentimentDF.where(col('SENTIMENT_LABEL') == \"positive\")  \n\n# Take a random sample of 100 users\nnum_users = brandUserSentimentDF.count()\nsample_num_users = 95\nusrfraction = float(sample_num_users)/float(num_users)\nusrseed = random.randint(1, 100)\n\n## Start off by finding the number of unique users tweeting about Coke\nprint 'Number of unique users tweeting about justinbieber: ', len(brand2TweetsDF.select('USER_SCREEN_NAME').distinct().collect())\nprint 'Number of negative tweeting folks: ', negativeTweetersDF.count()\nprint 'Number of positive tweeting folks: ', positiveTweetersDF.count()\nprint 'Sample size: ', sample_num_users, ' Fraction: ', usrfraction, ' Seed: ', usrseed\n\nbrandUserSampleDF = brandUserSentimentDF.sample(False, usrfraction, usrseed)\n\nprint 'Records in brandUserSampleDF: ', brandUserSampleDF.count()", 
            "execution_count": 43
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "In this step, we collect enough tweets for each unique user and run those tweets through Personality Insights to extract the Big Five personality traits, also known as OCEAN (openness, conscientiousness, extraversion, agreeableness, and neuroticism)."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Libraries and Credentials for Twitter and Personality Insights\nimport urllib2, requests, json\nimport tweepy\nfrom tweepy import OAuthHandler\nfrom watson_developer_cloud import PersonalityInsightsV3\n\n## Credentials for Twitter Developer account\nconsumer_key = credentials_json['twitter_consumer_key']\nconsumer_secret = credentials_json['twitter_consumer_secret']\naccess_token = credentials_json['twitter_access_token']\naccess_token_secret = credentials_json['twitter_access_token_secret'] \n\nauth = OAuthHandler(consumer_key, consumer_secret)\nauth.set_access_token(access_token, access_token_secret)\n \napi = tweepy.API(auth)\n\n## Get PI object using PI credentials\npersonality_insights = PersonalityInsightsV3(\n  version=credentials_json['pi_version'],\n  username=credentials_json['pi_username'],\n  password=credentials_json['pi_password'])", 
            "execution_count": 44
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Now we'll define several functions to programmatically paint a portrait of each user and their personality.\n\n- `getTweets()` - to collect tweets from a given user ID\n- `getPersonality()` - to call Personality Insights on the users' tweets\n- `extractOCEANtraits()` - to gather the OCEAN scores for each users PI data\n- `getPItraits()` - to put it all together and return a list of traits for each user"
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "## Collect tweets for a given user\ndef getTweets(username):\n    twitter_id = username    \n    try:\n        tweet_collection = api.user_timeline(screen_name = twitter_id, count = 100, include_rts = True)\n        i = 0\n        tweets = []\n        for status in tweet_collection:\n            i = i+1\n            tweets.append(status.text)\n    except Exception:\n        tweets = None\n    return tweets\n\n## Call Personality Insights on the tweets for the user\ndef getPersonality(tweets):\n    # get tweets by user\n    if tweets == None:\n        profile = None\n    else:\n        tweets_content = ' '.join(tweets)\n        # UTF-8 encoding\n        twt = tweets_content.encode('utf-8')\n        # call PI to get personality profile\n        try:\n            profile = personality_insights.profile(twt, content_type = 'text/plain', content_language = None,\n                                           accept ='application/json', accept_language = None, raw_scores = False,\n                                           consumption_preferences = False, csv_headers = False)\n        except Exception:\n            profile = None\n    \n    return profile\n\n\n## Extract OCEAN percentiles from PI data\ndef extractOCEANtraits(profile):\n    if profile == None:\n        openness = None\n        conscientiousness = None\n        extraversion = None\n        agreeableness = None\n        neuroticism = None\n    else:\n        personality = profile['personality']\n        openness = personality[0]['percentile']\n        conscientiousness = personality[1]['percentile']\n        extraversion = personality[2]['percentile']\n        agreeableness = personality[3]['percentile']\n        neuroticism = personality[4]['percentile']\n    \n    return openness, conscientiousness, extraversion, agreeableness, neuroticism\n\n## Combine function calls for a user\ndef getPItraits(user, verbose = F):\n    # get tweets by user\n    if verbose == F:\n        try:\n            tweets = getTweets(user)\n            # run PI profile on extracted tweets\n            profile = getPersonality(tweets)\n            # extract OCEAN traits\n            openness, conscientiousness, extraversion, agreeableness, neuroticism = extractOCEANtraits(profile)\n        except Exception:\n            return None\n        return openness, conscientiousness, extraversion, agreeableness, neuroticism\n    else:\n        print 'Getting tweets for user: ', user\n        try:\n            tweets = getTweets(user)\n            # run PI profile on extracted tweets\n            profile = getPersonality(tweets)\n            # extract OCEAN traits\n            openness, conscientiousness, extraversion, agreeableness, neuroticism = extractOCEANtraits(profile)\n        except Exception:\n            return None\n        return openness, conscientiousness, extraversion, agreeableness, neuroticism", 
            "execution_count": 45
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Now we'll extract Personality Insights traits for the users sampled in step 7, then push the data back to Spark for machine learning.  \n\n**This cell will also send data to the API, so be mindful of the number of calls you are making and be patient for the results.**"
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "text": "Number of records in userPersonalityInsightsDF:  83\nroot\n |-- USER_SCREEN_NAME: string (nullable = true)\n |-- SENTIMENT_LABEL: string (nullable = true)\n |-- SENTIMENT: double (nullable = true)\n |-- USER_FAVOURITES_COUNT: long (nullable = true)\n |-- USER_STATUSES_COUNT: long (nullable = true)\n |-- USER_FOLLOWERS_COUNT: long (nullable = true)\n |-- USER_FRIENDS_COUNT: long (nullable = true)\n |-- count: long (nullable = true)\n |-- OPENNESS: double (nullable = true)\n |-- CONSCIENTIOUSNESS: double (nullable = true)\n |-- EXTRAVERSION: double (nullable = true)\n |-- AGREEABLENESS: double (nullable = true)\n |-- NEUROTICISM: double (nullable = true)\n\n", 
                    "output_type": "stream", 
                    "name": "stdout"
                }, 
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  USER_SCREEN_NAME SENTIMENT_LABEL  SENTIMENT  USER_FAVOURITES_COUNT  \\\n0   KidrauhlSwag17        negative  -0.893496                    285   \n1     flatlinesIut        negative  -0.838696                  19964   \n\n   USER_STATUSES_COUNT  USER_FOLLOWERS_COUNT  USER_FRIENDS_COUNT  count  \\\n0                  171                    52                  92      1   \n1                20223                  2548                 118      1   \n\n   OPENNESS  CONSCIENTIOUSNESS  EXTRAVERSION  AGREEABLENESS  NEUROTICISM  \n0  0.226331           0.332067      0.484590       0.650300     0.325146  \n1  0.479193           0.012110      0.115281       0.217535     0.047131  ", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>USER_SCREEN_NAME</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>SENTIMENT</th>\n      <th>USER_FAVOURITES_COUNT</th>\n      <th>USER_STATUSES_COUNT</th>\n      <th>USER_FOLLOWERS_COUNT</th>\n      <th>USER_FRIENDS_COUNT</th>\n      <th>count</th>\n      <th>OPENNESS</th>\n      <th>CONSCIENTIOUSNESS</th>\n      <th>EXTRAVERSION</th>\n      <th>AGREEABLENESS</th>\n      <th>NEUROTICISM</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>KidrauhlSwag17</td>\n      <td>negative</td>\n      <td>-0.893496</td>\n      <td>285</td>\n      <td>171</td>\n      <td>52</td>\n      <td>92</td>\n      <td>1</td>\n      <td>0.226331</td>\n      <td>0.332067</td>\n      <td>0.484590</td>\n      <td>0.650300</td>\n      <td>0.325146</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>flatlinesIut</td>\n      <td>negative</td>\n      <td>-0.838696</td>\n      <td>19964</td>\n      <td>20223</td>\n      <td>2548</td>\n      <td>118</td>\n      <td>1</td>\n      <td>0.479193</td>\n      <td>0.012110</td>\n      <td>0.115281</td>\n      <td>0.217535</td>\n      <td>0.047131</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 46
                }
            ], 
            "source": "## Convert to Pandas to use the Watson PI API\nbrandUserSampleDF = brandUserSampleDF.toPandas()\nbrandUserSampleDF['OPENNESS'], brandUserSampleDF['CONSCIENTIOUSNESS'],\\\nbrandUserSampleDF['EXTRAVERSION'], brandUserSampleDF['AGREEABLENESS'],\\\nbrandUserSampleDF['NEUROTICISM'] = zip(*brandUserSampleDF['USER_SCREEN_NAME'].map(getPItraits))\n\n# Conver back to Spark Dataframe from Pandas dataframe\nuserPersonalityDF = spark.createDataFrame(brandUserSampleDF)\n\n## Drop rows without any PI enrichment\nuserPersonalityDF = userPersonalityDF.na.drop()\n\n## Check row count and schema\nprint 'Number of records in userPersonalityInsightsDF: ', userPersonalityDF.count()\nuserPersonalityDF.printSchema()\n\nuserPersonalityDF.limit(2).toPandas()", 
            "execution_count": 46
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "________________\n\n<a id=\"sparkml\"> </a>\n## Step 9: Spark machine learning for user segmentation \n<a href=\"https://spark.apache.org/docs/latest/ml-guide.html\" target=\"_blank\" rel=\"noopener no referrer\">Spark MLlib</a>  includes a rich set of machine learning algorithms that are very powerful in extracting insights from data. Typically, you will need to convert the data into the right format before you can apply these machine learning algorithms. In this step, we apply several steps to process the data so it can run the <a href=\"https://spark.apache.org/docs/latest/mllib-clustering.html#k-means\" target=\"_blank\" rel=\"noopener no referrer\">Kmeans</a> clustering algorithm including normalizing **USER_FOLLOWERS_COUNT** and **USER_STATUSES_COUNT** fields as well as extracting the relevant feature set into a Vector to be used for clustering.\n\nWe run Kmeans clustering using two different feature sets.\n* **FeatureSet 1** (no Personality Traits): (SENTIMENT, USER_FOLLOWERS_COUNT, USER_STATUSES_COUNT)\n* **FeatureSet 2** (with Personality Traits): (SENTIMENT, USER_FOLLOWERS_COUNT, USER_STATUSES_COUNT, OPENNESS, CONSCIENTIOUSNESS, EXTRAVERSION, AGREEABLENESS, NEUROTICISM)\n\nFirst, we'll convert the follower and statuses counts to a vector as per the Spark documentation. Then we can run <a href=\"https://spark.apache.org/docs/2.1.0/ml-features.html#minmaxscaler\" target=\"_blank\" rel=\"noopener no referrer\">MinMaxScaler</a> and normalize the counts to a range between 0 and 1. These are necessary preprocessing steps for the clustering algorithm to work properly."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  USER_SCREEN_NAME SENTIMENT_LABEL  SENTIMENT  SCALED_FOLLOWERS  \\\n0   KidrauhlSwag17        negative  -0.893496          0.000886   \n1     flatlinesIut        negative  -0.838696          0.043420   \n2  JohnellaOrlando        negative  -0.718087          0.003289   \n3      realwonders        negative  -0.696614          0.468525   \n4          anotbok        negative  -0.696614          0.933421   \n\n   SCALED_STATUSES  OPENNESS  CONSCIENTIOUSNESS  EXTRAVERSION  AGREEABLENESS  \\\n0         0.000841  0.226331           0.332067      0.484590       0.650300   \n1         0.105617  0.479193           0.012110      0.115281       0.217535   \n2         0.012384  0.005824           0.739157      0.101717       0.929923   \n3         0.001583  0.999433           0.615764      0.709900       0.532056   \n4         1.000000  0.545226           0.326542      0.429713       0.587216   \n\n   NEUROTICISM                               PI_ENRICHED_FEATURES  \\\n0     0.325146  [-0.893495976925, 0.000886132033673, 0.0008412...   \n1     0.047131  [-0.83869600296, 0.04342046965, 0.105616545007...   \n2     0.295270  [-0.718087017536, 0.00328891312498, 0.01238367...   \n3     0.288864  [-0.696614027023, 0.468525271804, 0.0015832292...   \n4     0.237858  [-0.696614027023, 0.93342081047, 1.0, 0.545225...   \n\n                                       BASE_FEATURES  \n0  [-0.893495976925, 0.000886132033673, 0.0008412...  \n1    [-0.83869600296, 0.04342046965, 0.105616545007]  \n2  [-0.718087017536, 0.00328891312498, 0.01238367...  \n3  [-0.696614027023, 0.468525271804, 0.0015832292...  \n4              [-0.696614027023, 0.93342081047, 1.0]  ", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>USER_SCREEN_NAME</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>SENTIMENT</th>\n      <th>SCALED_FOLLOWERS</th>\n      <th>SCALED_STATUSES</th>\n      <th>OPENNESS</th>\n      <th>CONSCIENTIOUSNESS</th>\n      <th>EXTRAVERSION</th>\n      <th>AGREEABLENESS</th>\n      <th>NEUROTICISM</th>\n      <th>PI_ENRICHED_FEATURES</th>\n      <th>BASE_FEATURES</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>KidrauhlSwag17</td>\n      <td>negative</td>\n      <td>-0.893496</td>\n      <td>0.000886</td>\n      <td>0.000841</td>\n      <td>0.226331</td>\n      <td>0.332067</td>\n      <td>0.484590</td>\n      <td>0.650300</td>\n      <td>0.325146</td>\n      <td>[-0.893495976925, 0.000886132033673, 0.0008412...</td>\n      <td>[-0.893495976925, 0.000886132033673, 0.0008412...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>flatlinesIut</td>\n      <td>negative</td>\n      <td>-0.838696</td>\n      <td>0.043420</td>\n      <td>0.105617</td>\n      <td>0.479193</td>\n      <td>0.012110</td>\n      <td>0.115281</td>\n      <td>0.217535</td>\n      <td>0.047131</td>\n      <td>[-0.83869600296, 0.04342046965, 0.105616545007...</td>\n      <td>[-0.83869600296, 0.04342046965, 0.105616545007]</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>JohnellaOrlando</td>\n      <td>negative</td>\n      <td>-0.718087</td>\n      <td>0.003289</td>\n      <td>0.012384</td>\n      <td>0.005824</td>\n      <td>0.739157</td>\n      <td>0.101717</td>\n      <td>0.929923</td>\n      <td>0.295270</td>\n      <td>[-0.718087017536, 0.00328891312498, 0.01238367...</td>\n      <td>[-0.718087017536, 0.00328891312498, 0.01238367...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>realwonders</td>\n      <td>negative</td>\n      <td>-0.696614</td>\n      <td>0.468525</td>\n      <td>0.001583</td>\n      <td>0.999433</td>\n      <td>0.615764</td>\n      <td>0.709900</td>\n      <td>0.532056</td>\n      <td>0.288864</td>\n      <td>[-0.696614027023, 0.468525271804, 0.0015832292...</td>\n      <td>[-0.696614027023, 0.468525271804, 0.0015832292...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>anotbok</td>\n      <td>negative</td>\n      <td>-0.696614</td>\n      <td>0.933421</td>\n      <td>1.000000</td>\n      <td>0.545226</td>\n      <td>0.326542</td>\n      <td>0.429713</td>\n      <td>0.587216</td>\n      <td>0.237858</td>\n      <td>[-0.696614027023, 0.93342081047, 1.0, 0.545225...</td>\n      <td>[-0.696614027023, 0.93342081047, 1.0]</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 47
                }
            ], 
            "source": "from pyspark.ml.feature import VectorAssembler\nfrom pyspark.ml.feature import MinMaxScaler\nfrom pyspark.ml.linalg import Vectors\n\n## Define columns to be converted to vectors\nfollowersVector = VectorAssembler(\n  inputCols=[\"USER_FOLLOWERS_COUNT\"], outputCol=\"USER_FOLLOWERS_COUNT_VECTOR\")\n\nstatusesVector = VectorAssembler(\n  inputCols=[\"USER_STATUSES_COUNT\"], outputCol=\"USER_STATUSES_COUNT_VECTOR\")\n\n## Define our input and output columns for MinMaxScaler\nfollowersScaler = MinMaxScaler(inputCol=\"USER_FOLLOWERS_COUNT_VECTOR\", outputCol=\"USER_FOLLOWERS_COUNT_SCALED\")\nstatusesScaler = MinMaxScaler(inputCol=\"USER_STATUSES_COUNT_VECTOR\", outputCol=\"USER_STATUSES_COUNT_SCALED\")\n\n## Invoke the VectorAssembler transformations and select desired columns\nuserPersonalityDF = followersVector.transform(userPersonalityDF)\nuserPersonalityDF = statusesVector.transform(userPersonalityDF)\n\n## Fit MinMaxScalerModel on our vectors and rescale\nfollowersScalerModel = followersScaler.fit(userPersonalityDF)\nstatusesScalerModel = statusesScaler.fit(userPersonalityDF)\n\nuserPersonalityDF = followersScalerModel.transform(userPersonalityDF)\nuserPersonalityDF = statusesScalerModel.transform(userPersonalityDF)\n\n## User-defined function to convert from vector back to float\nfrom pyspark.sql.types import DoubleType\nfrom pyspark.sql.functions import lit, udf\n\ndef ith_(v, i):\n    try:\n        return float(v[i])\n    except ValueError:\n        return None\n\nudfVecToFloat = udf(ith_, DoubleType())\n\n## Add column for floating point scaled followers\nuserPersonalityDF = userPersonalityDF.withColumn(\"SCALED_FOLLOWERS\", udfVecToFloat('USER_FOLLOWERS_COUNT_SCALED', lit(0)))\\\n                                     .withColumn(\"SCALED_STATUSES\", udfVecToFloat('USER_STATUSES_COUNT_SCALED', lit(0)))\\\n                                     .select(\"USER_SCREEN_NAME\", \"SENTIMENT_LABEL\", \"SENTIMENT\", \"SCALED_FOLLOWERS\", \\\n                                            \"SCALED_STATUSES\", \"OPENNESS\", \"CONSCIENTIOUSNESS\", \"EXTRAVERSION\", \\\n                                            \"AGREEABLENESS\", \"NEUROTICISM\")\n        \n## Lastly, we'll have to add columns for features both with and without Personality Insights\nassemblerWithPI = VectorAssembler(\n    inputCols = ['SENTIMENT','SCALED_FOLLOWERS','SCALED_STATUSES','OPENNESS','CONSCIENTIOUSNESS','EXTRAVERSION',\\\n               'AGREEABLENESS','NEUROTICISM'],\n    outputCol = \"PI_ENRICHED_FEATURES\")\n\nassemblerWithoutPI = VectorAssembler(\n    inputCols = ['SENTIMENT', 'SCALED_FOLLOWERS', 'SCALED_STATUSES'],\n    outputCol = \"BASE_FEATURES\")\n\n\nuserPersonalityDF = assemblerWithPI.transform(userPersonalityDF)\nuserPersonalityDF = assemblerWithoutPI.transform(userPersonalityDF)\n\n## View transformed DF\nuserPersonalityDF.limit(5).toPandas()", 
            "execution_count": 47
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "### Kmeans clustering \n\nNow that we have prepared the data, we can run Kmeans to assign cluster labels to each user."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  USER_SCREEN_NAME SENTIMENT_LABEL  SENTIMENT  SCALED_FOLLOWERS  \\\n0   KidrauhlSwag17        negative  -0.893496          0.000886   \n1     flatlinesIut        negative  -0.838696          0.043420   \n2  JohnellaOrlando        negative  -0.718087          0.003289   \n3      realwonders        negative  -0.696614          0.468525   \n4          anotbok        negative  -0.696614          0.933421   \n\n   SCALED_STATUSES  OPENNESS  CONSCIENTIOUSNESS  EXTRAVERSION  AGREEABLENESS  \\\n0         0.000841  0.226331           0.332067      0.484590       0.650300   \n1         0.105617  0.479193           0.012110      0.115281       0.217535   \n2         0.012384  0.005824           0.739157      0.101717       0.929923   \n3         0.001583  0.999433           0.615764      0.709900       0.532056   \n4         1.000000  0.545226           0.326542      0.429713       0.587216   \n\n   NEUROTICISM                               PI_ENRICHED_FEATURES  \\\n0     0.325146  [-0.893495976925, 0.000886132033673, 0.0008412...   \n1     0.047131  [-0.83869600296, 0.04342046965, 0.105616545007...   \n2     0.295270  [-0.718087017536, 0.00328891312498, 0.01238367...   \n3     0.288864  [-0.696614027023, 0.468525271804, 0.0015832292...   \n4     0.237858  [-0.696614027023, 0.93342081047, 1.0, 0.545225...   \n\n                                       BASE_FEATURES  BASE_PREDICTIONS  \\\n0  [-0.893495976925, 0.000886132033673, 0.0008412...                 0   \n1    [-0.83869600296, 0.04342046965, 0.105616545007]                 0   \n2  [-0.718087017536, 0.00328891312498, 0.01238367...                 0   \n3  [-0.696614027023, 0.468525271804, 0.0015832292...                 0   \n4              [-0.696614027023, 0.93342081047, 1.0]                 4   \n\n   PI_PREDICTIONS  \n0               3  \n1               3  \n2               3  \n3               3  \n4               2  ", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>USER_SCREEN_NAME</th>\n      <th>SENTIMENT_LABEL</th>\n      <th>SENTIMENT</th>\n      <th>SCALED_FOLLOWERS</th>\n      <th>SCALED_STATUSES</th>\n      <th>OPENNESS</th>\n      <th>CONSCIENTIOUSNESS</th>\n      <th>EXTRAVERSION</th>\n      <th>AGREEABLENESS</th>\n      <th>NEUROTICISM</th>\n      <th>PI_ENRICHED_FEATURES</th>\n      <th>BASE_FEATURES</th>\n      <th>BASE_PREDICTIONS</th>\n      <th>PI_PREDICTIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>KidrauhlSwag17</td>\n      <td>negative</td>\n      <td>-0.893496</td>\n      <td>0.000886</td>\n      <td>0.000841</td>\n      <td>0.226331</td>\n      <td>0.332067</td>\n      <td>0.484590</td>\n      <td>0.650300</td>\n      <td>0.325146</td>\n      <td>[-0.893495976925, 0.000886132033673, 0.0008412...</td>\n      <td>[-0.893495976925, 0.000886132033673, 0.0008412...</td>\n      <td>0</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>flatlinesIut</td>\n      <td>negative</td>\n      <td>-0.838696</td>\n      <td>0.043420</td>\n      <td>0.105617</td>\n      <td>0.479193</td>\n      <td>0.012110</td>\n      <td>0.115281</td>\n      <td>0.217535</td>\n      <td>0.047131</td>\n      <td>[-0.83869600296, 0.04342046965, 0.105616545007...</td>\n      <td>[-0.83869600296, 0.04342046965, 0.105616545007]</td>\n      <td>0</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>JohnellaOrlando</td>\n      <td>negative</td>\n      <td>-0.718087</td>\n      <td>0.003289</td>\n      <td>0.012384</td>\n      <td>0.005824</td>\n      <td>0.739157</td>\n      <td>0.101717</td>\n      <td>0.929923</td>\n      <td>0.295270</td>\n      <td>[-0.718087017536, 0.00328891312498, 0.01238367...</td>\n      <td>[-0.718087017536, 0.00328891312498, 0.01238367...</td>\n      <td>0</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>realwonders</td>\n      <td>negative</td>\n      <td>-0.696614</td>\n      <td>0.468525</td>\n      <td>0.001583</td>\n      <td>0.999433</td>\n      <td>0.615764</td>\n      <td>0.709900</td>\n      <td>0.532056</td>\n      <td>0.288864</td>\n      <td>[-0.696614027023, 0.468525271804, 0.0015832292...</td>\n      <td>[-0.696614027023, 0.468525271804, 0.0015832292...</td>\n      <td>0</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>anotbok</td>\n      <td>negative</td>\n      <td>-0.696614</td>\n      <td>0.933421</td>\n      <td>1.000000</td>\n      <td>0.545226</td>\n      <td>0.326542</td>\n      <td>0.429713</td>\n      <td>0.587216</td>\n      <td>0.237858</td>\n      <td>[-0.696614027023, 0.93342081047, 1.0, 0.545225...</td>\n      <td>[-0.696614027023, 0.93342081047, 1.0]</td>\n      <td>4</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 48
                }
            ], 
            "source": "from pyspark.ml.clustering import KMeans\n\n## Define model parameters and set the seed\nbaseKMeans = KMeans(featuresCol = \"BASE_FEATURES\", predictionCol = \"BASE_PREDICTIONS\").setK(5).setSeed(206)\npiKMeans = KMeans(featuresCol = \"PI_ENRICHED_FEATURES\", predictionCol = \"PI_PREDICTIONS\").setK(5).setSeed(206)\n\n## Fit model on our feature vectors\nbaseClustersFit = baseKMeans.fit(userPersonalityDF.select(\"BASE_FEATURES\"))\nenrichedClustersFit = piKMeans.fit(userPersonalityDF.select(\"PI_ENRICHED_FEATURES\"))\n\n## Get the cluster IDs for each user\nuserPersonalityDF = baseClustersFit.transform(userPersonalityDF)\nuserPersonalityDF = enrichedClustersFit.transform(userPersonalityDF)\n\n## Check our work\nuserPersonalityDF.limit(5).toPandas()", 
            "execution_count": 48
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "____________\n\n<a id=\"clusters\"> </a>\n## Step 10: Visualize user clusters\n\nOK, great - we have successfully clustered our users both with and without enrichment data from Watson Personality Insights. Now we can visualize users through their cluster identity.\n \nIn this step, we visualize the different user clusters obtained with and without using Personality traits. This visualization illustrates the differences between two clustering approachs and shows how Watson <a href=\"https://www.ibm.com/watson/developercloud/personality-insights.html\" target=\"_blank\" rel=\"noopener no referrer\">Personality Insights</a> can help provide finer user segmentation.\n\nBrand managers and marketing teams can use this segmentation to craft different messaging to target the various user segments to improve brand adoption in the marketplace."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "scrolled": true
            }, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/plain": "<matplotlib.figure.Figure at 0x7fb348f2dc90>", 
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAABPEAAAJBCAYAAAAnRyDXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3Xe4XFW9//H396RRAqEkBKT3TtQo\nHQPoVVQQBVQQG6iIYqF5vdc2Lq9e2+WHCpYroNguFkQUG1ioShM0iIhIkw6hhSRA2lm/P/YOHMLJ\nyUkyc9aU9+t55jnJnJm9P7NnkvM9373W2pFzRpIkSZIkSVL76isdQJIkSZIkSdLQbOJJkiRJkiRJ\nbc4mniRJkiRJktTmbOJJkiRJkiRJbc4mniRJkiRJktTmbOJJkiRJkiRJbc4mnoqIiI9HxHdL5xiO\niLg9Il5S//lDEXH6CO33rRFx2TI+Z3ZEbNaqTCu6/4HHUstm4L+ZiNioPtajSudaJCJ+FRFvKZ1D\nkqROY108rP1aF3ewiNgkInJEjK7/3lZ1Y0QcHhEXlM4hDYdNPLVMRLwhIv5U/wC7t/7Pes8mbv8Z\nPwxGQs75v3POb1/a4yLioohY6uNWxGD7yDmPzznf2sr9DmXg/iPizIj45EjsdziFXX28nqw/jw9G\nxDkRsd5I5Gu2nPMd9bFeCCv+eav/HW2xgplennP+Vr29ZS60JUnqZtbF1sUrUhfXtdXC+vPzWET8\nJSL2b17akdXMurEZv3PknL+Xc37pgG2ucG0stYpNPLVERBwPfAH4b2AysBHwFeDAkrkGGskip532\n3ePek3MeD2wFrAGcvKwb6MX3rhdfsyRJzWJd3L777jCX13XsGsAZwA8jYs1l2UAvHutefM3qbjbx\n1HQRMQH4BHBMzvmcnPOcnPP8nPN5OecPDPL4vSPirsXuGzhUf+f6zOVjEXF/RPy/+mGX1F8frc9K\n7VY//siI+HtEPBIR50fExgO2myPimIj4J/DPJeR/U0T8KyIeiogPL/a9gVMaV4qI79aPezQiro6I\nyRHxKWAv4NQ616mD7XuwM6aDnEWMiDg1ImZGxI0R8eL6zqH2scWi9yEivh0RM+rX85GI6Ku/99aI\nuCwi/qc+TrdFxMuXcDyOiIjzBvz9nxHxowF/vzMinjtw/xFxFHA48O91vvMGbPK5EXFd/Zp+EBEr\nDdjWOyLi5oh4OCJ+FhHPqe9f4rGKiG2BrwG71ft6dLDXMVDO+WHgx8AO9bbG1cfijvoz9rWIWLn+\n3t4RcVdEfDAi7gO+GRETI+Ln9fv+cERcOuDYbltnezQi/hYRrxqQ+cyI+HJE/CIiZkXElRGx+YDv\nf7E+no9FxDURsdcS3pOnjsdgn4V6Hyct9pyfRcRxg2xr0b+j6fXzX7+E17xm/Zpn1J+Zn0fEBsN9\nPyLiFRFxQ/26746IE5f2PkmS1OnCuti6eDnr4iXJOfcD3wBWBjav97V/VKPzHo2IP0bETgMy3V7X\ndNcBc+r68YN1PTYrIv4x4FiOi4gvRMQ99e0LETGu/t6i+vCEiHggqhGlRwzYzysj4s/1Z/POiPj4\nkl7DUHVjRLyw/myPGvD4gyJi+iDbGfTYLuE1/0dE3FK/5hsi4jUDtvPUaMAYvDZeYu0vjTQ/eGqF\n3YCVgJ80aXtfBL6Yc16d6gfVD+v7X1R/XaMern55RBwIfAg4CJgEXAqctdj2Xg3sAmy3+I4iYjvg\nq8CbgOcAawMbLP642luACcCG9eOOBp7IOX+43u976lzvGc6+l2AX4BZgItAAzomItZayj0VOqfNt\nBkwD3gwcMeD7uwD/qLf9OeCMiIhBtnMxsFdE9EXVVBtL9R4T1Tof44HrBj4h5/x14HvA5+p8Bwz4\n9uuA/YBNgZ2At9bb2hf4dP399YB/Ad9f2gHKOf+d6thfXu9rjaU9JyImAgcDf67v+gzV6LznAlsA\n6wMfG/CUdYG1gI2Bo4ATgLuoPmOTqT5zOSLGAOcBFwDrAO8FvhcRWw/Y1qFAAtYEbgY+NeB7V9cZ\n1gL+D/jR0oq5JXwWvgUcNqA4nQi8pN7m4s9f9O9oSv38HyzhNfcB36z/vhHwBHDqINtb0vtxBvDO\nnPNqVM3T3w/1uiRJ6hLWxdbFy1wXDyWqRufbgdlUDdDnUTX13kl17P8X+Nmi5lvtMOCVVKP4Ngfe\nA7ywrsteBtxeP+7DwK5U9egUYGfgIwO2sy7VcVwfeBvw5Xh6NOAcquO6Rr2vd0XEq4d6LYPVjTnn\nq4GHgJcOeOibgG8P8vyhju1TrznnvIDqs7NXnT8B341BltZZQm08aO0/1GuTWsUmnlphbeDB+j/L\nZpgPbBERE3POs3POVwzx2KOBT+ec/17v/7+pznBtPOAxn845P5xzfmKQ5x8C/DznfEnOeS7wUaB/\niFxrA1vknBfmnK/JOT+2lNcy1L4H8wDwhfqM7Q+oiotXLu1J9ZmrQ4H/zDnPyjnfDpxE9QNwkX/l\nnE+r11X7FlXjbPLi26rX8phF9cP8RcD5wD0RsQ1VEXRpfUZwuL6Uc76nHg13Xr1dqM6ifSPnfG19\n7P+T6qzcJsuw7aXuO6qRYdOBe4Hj6wLtKOC4+r2ZRfW5OXTA8/qBRs55bv3ezac6XhvX782lOedM\nVfSMBz6Tc56Xc/498HOqImKRn+Scr6o/n98b8PrJOX835/xQznlBzvkkYBwwsAE4LDnnq4CZwIvr\nuw4FLso5378Mm3nGa65z/Tjn/Hh9jD5F9f4P13xgu4hYPef8SM752mV4riRJncq6eMmsi5dcFw9m\n17qOvY+qtnxNznkmVR37vznnK+tj/y1gLlVdOnA/d9bHeiFVjbldRIzJOd+ec76lftzhwCdyzg/k\nnGdQNbsGHqf59ffn55x/SdVI3Lo+NhflnP+ac+7POV9H1TBellpxoG8BbwSIiLWoGo3POhm9FANf\nMznnH9XHur/+/PyTqkk5HEuq/aURZxNPrfAQMDGat/7A26hGSd0Y1dD8oRZx3Rj4Yj3U+VHgYSCo\nzhYtcucQz3/OwO/nnOdQvZ7BfIfqB/f36+Hmn6tHYg1lqH0P5u7FfkD8q864NBOBMfXjBz534HG4\nb9Efcs6P138cv4TtXQzsTVWsXAxcRPVDeVr992Vx34A/Pz5gn88ZmDfnPJvq2A/MvKLeV5/hWz/n\nfHhdnEwCVgGuGfC5+XV9/yIzcs5PDvj756lG0V0QEbdGxH8MeA13Lla8LfG488zXT0ScGNWUl5l1\njglU7+XyeKr4qb9+Zxmf/4zXHBGrRMT/RjUF5TGqaTtrxPCvkHsw8ArgXxFxcdTTfCRJ6nLWxUtm\nXTxEXTiIK+o6dmLOedec82/r+zcGTlj0Ptfv9YY889gMfB9vBo4FPg48EBHfr0cVwmL1OM8+xg8t\n1pB+KnNE7BIRF0Y1ZXkmVRN5eevY7wIHRMSqVKMVL80537uM23jG5ysi3hxPTzl+lGpmyHDzLan2\nl0acTTy1wuVUZ3+GHD49wByqJgrw1NmypxooOed/5pwPo5qe+Fng7Po/9MHOftxJNWVvjQG3lXPO\nfxzwmKHOmtxL9UNvUZZVqM4qPkt9FiblnLcDdgf2pxpCPtQ+Bt4/p/66yoD71l3s8esvNpR/I+Ce\npewD4EGqM0YDz7RuBNw9xHOGsqhY2av+88UsvVhZ1rNT9zAgb/0er02VeWnHakXOhD1INTV0+wGf\nmQm5Wjh40O3XZ3FPyDlvBryKakTfi+vXsOFia2QM67hHtf7dv1MVKmvmahrqTKpie2kGe/3fBQ6M\niCnAtsC5w9jOUNs8gepM6y65msKzaKrBYPmelSfnfHXO+UCqf8fn8vT0H0mSupl1sXXx0vKtqDuB\nTy32Pq+Scx44dXrxWvb/cs57Uh2TTPVZgsXqcZ55jJfm/4CfARvmnCdQrXW3XHVszvluqn87B1GN\nBBzqZPRSP1/16NPTqKYRr13X2dcPM99Qtb804mziqenqYd0fo1oj4dX1CJ4xEfHyiPjcIE+5CVgp\nqsVQx1Ctu/DUGg4R8caImFSPblp00YJ+YEb9dbMB2/oa8J8RsX393AkR8dpliH82sH9E7BkRY6kW\nIh7030lE7BMRO9bF1WNUxcGiEVj3L5brWepRYHcDb4yIURFxJPXitAOsA7yvPn6vpWrG/HJp+6in\nAvwQ+FRErFb/4DqeqrGzPC4G9gFWzjnfRbXuyH5Uhdyfl/CcpR6DxZwFHBERz63X8Phv4Mp6iP/S\njtX9wAb1e7ZM6s/VacDJEbEOQESsHxEvW9Jzolo8eIu6kJxJNS2hH7iS6ozkv9fv2d7AAQxjbT9g\nNWAB1ed6dER8DFh9mC/jWce6fp+upip6fryUqSrDea9Wo2p2PlpPa2gsZXtPvR8RMTYiDo+ICTnn\n+VT/XpZlqokkSR3JuhiwLh4yXxOcBhxdj4SLiFi1/vysNtiDI2LriNi3rrefpKrvFr1XZwEfiYhJ\nUa2p/DGGf5xWAx7OOT8ZETsDbxjm85ZUx3+b6gT3jsA5S3n+0o7tokb3DKguUEJ9gbvhbHOI2l8a\ncTbx1BK5Ws/reKrCYwbVGaL3MMhooLq4eTdwOk+Puhp4Va79gL9FxGyqxXwPzdUaXY9Trcv1h3pY\n9K45559QnUn6flRT/q4HBr261BJy/w04hupM0r3AI4tlGWhdquLmMeDvVD/QF50l+iJwSFRXuPrS\nELt8B/ABqqkJ2wN/XOz7VwJbUp1B/BRwSM550TSGpe3jvVTH8lbgsvo1fWOILEuUc76Jas2LS+u/\nP1Zv9w91YTSYM6jW2ng0IpY6CqyeEvBRqqvG3ktVuA1cl26oY/V74G/AfRHx4DK8tEU+SDVE/or6\nc/Nbhl6Lbsv6MbOpzhJ+Jed8Yc55HlXT7uVU79lXgDfnnG8cRobzqabx3kQ1deFJhj/NZEmfhW9R\nFT5Lm0r7ceBb9Xv1uiU85gtUV0F7ELiizrokg70fbwJur4/v0VRrrkiS1PWsi62LWca6eBnz/Inq\n2J1K9R7dzNAXyBhHdVG3B6mm865DtRY1wCeBP1FdnOOvwLX1fcPxbuATETGLqvk33FkXS6rjf0I1\nKvAnA6Y4D2apxzbnfAPVOoiXUzXodgT+MMQ2P84za+NBa//hvDip2cL1GCWpO0XEi6jOnm7s4ruS\nJEnqJBFxC9WU8N8u9cFSj3AkniR1oXoKzvuB023gSZIkqZNExMFUU2B/XzqL1E6adZUkSVKbiIht\nqaZCTAeOKBxHkiRJGraIuAjYDnhTvf6jpJrTaSVJkiRJkqQ253RaSZIkSZIkqc3ZxJMkSZIkSZLa\nnE08SZIkSZIkqc3ZxJMkSZIkSZLanE08SZIkSZIkqc3ZxJMkSZIkSZLanE08SZIkSZIkqc2NLh1A\nkiS11jXXXLPO6NGjTwd2oD1P4PUD1y9YsODtU6dOfaB0GEmSJHWWXql3beJJktTlRo8effq66667\n7aRJkx7p6+vLpfMsrr+/P2bMmLHdfffddzrwqtJ5JEmS1Fl6pd5tx+6kJElqrh0mTZr0WDsWNAB9\nfX150qRJM6nOnEqSJEnLqifqXZt4kiR1v752LWgWqfNZl0iSJGl59ES9a7EsSZJa7rWvfe0ma621\n1pQtt9xy+9JZJEmSpGYbiXrXNfEkSeoxkWJqM7eXG/mapT3myCOPfPD973//A0ccccSmzdy3JEmS\ntLhurXcdiSdJklru5S9/+exJkyYtKJ1DkiRJaoWRqHdt4kmSJEmSJEltziaeJEmSJEmS1OZs4kmS\nJEmSJEltziaeJEmSJEmS1OZs4kmSpJY74IADNt1zzz23ue2228ZNnjx5p5NPPnli6UySJElSs4xE\nvTu62RuUJEntLTfyNSO9z/POO++2kd6nJEmSelO31ruOxJMkSZIkSZLanE08SZIkSZIkqc3ZxJMk\nSZIkSZLanE08SZIkSZIkqc3ZxJMkSZIkSZLanE08SZIkSZIkqc2NLh1AkiR1v8cffzx22WWXbebN\nmxcLFy6MAw444JGTTz75ntK5JEmSpGYYiXrXJp4kST0mpTS1mdtrNBrXLO0xK620Ur7sssv+MWHC\nhP65c+fGC1/4wq1/97vfzXzxi188p5lZJEmSpG6td51OK0mSWq6vr48JEyb0A8ybNy8WLFgQEVE6\nliRJktQUI1Hv2sSTJEkjYsGCBWyzzTbbTZ48ecq0adMe23fffR2FJ0mSpK7R6nrXJp4kSRoRo0eP\n5sYbb7zhjjvuuO7aa69d9eqrr16pdCZJkiSpWVpd79rEkyRJI2rixIkL99prr1nnnXfehNJZJEmS\npGZrVb1rE0+SJLXcPffcM/rBBx8cBTB79uy48MILV992222fLJ1LkiRJaoaRqHe9Oq0kSWq5O++8\nc8xb3/rWTRcuXEjOOQ488MCHDzvssJmlc0mSJEnNMBL1buScm7k9SZLUZqZPn377lClTHiydY2mm\nT58+ccqUKZuUziFJkqTO0iv1rtNpJUmSJEmSpDZnE0+SJEmSJElqczbxJEmSJEmSpDZnE0+SJEmS\nJElqczbxJEmSJEmSpDZnE0+SJEmSJElqczbxJEnSiFiwYAHbbrvtdvvss88WpbNIkiRJzdbqend0\nKzYqSVoxEfENYH/ggZzzDqXzqMtETG3q9nK+ZjgP++QnPzl5iy22eGL27Nmjmrp/SZIkaaAurXcd\niSdJ7elMYL/SIaRmueWWW8acf/75E97xjnc8WDqLJEmS1GwjUe/axJOkNpRzvgR4uHQOqVmOOeaY\nDT/3uc/d1ddn6SFJkqTuMxL1rpW0JElqqbPOOmvCxIkTF+y1116Pl84iSZI6R0R8IyIeiIjrS2eR\nhjJS9a5r4kmSpJa67LLLxv/mN79ZY/31158wd+7cvjlz5vQdeOCBm/70pz+9rXQ2SZLU1s4ETgW+\nXTiHNKSRqncdiSdJklrqy1/+8t3333//dXffffdfzzzzzFt33XXXWTbwJEnS0rjEjDrFSNW7NvEk\nSZIkSZKkNud0WklqQxFxFrA3MDEi7gIaOeczyqZS18j5mlK73n///Wftv//+s0rtX5IkST2gS+td\nm3iS1IZyzoeVziBJkiRJah9Op5UkSZIkSZLanE08SZIkSZLUduolZi4Hto6IuyLibaUzSSU5nVaS\nJEmSJLUdl5iRnsmReJIkSZIkSVKbs4knSZIkSZIktTmn00oaVKRYBVgTWGsZvq4M9AML668D/zzU\nfQO/Nwt4oL7dP8ifH8yN3N/K1y6p+W6++eYxhx9++KYPPvjgmIjgLW95y4yPfvSjD5TOJUnqXZFi\nFDAZWHfAbb3661rAWGAM1e/NYwbchvP30cDjwGPAzGF8fdZ9uZFntfDlS2qykah3beJJPShSrAZs\nVd+2rL9uDKxN1ZBbExhXLODQFkaKh1hyk+9+4Bbgn7mRFxRLKbWxCKY2c3s5c83SHjNmzBhOOumk\nu/bcc8/HH3nkkb7nPe95273iFa94bOrUqU82M4skSZFiZaradrDm3MDbRFo7O231ej/LJVI8DtwG\n3DrYLTeyP0OlJejWetcmntSlIsU4YAue2ahb9OflLibawChgnfq2wxCPmxcp/gn8rb7dUH+1uScV\nsPHGG8/feOON5wOsueaa/ZtvvvkTd9xxx1ibeJKk5RUpxgBbU9WEA2+b0h1LR60CbF/fFpcjxX0M\n3uC7JTfyvSOWUhIwMvWuTTypw0WKNYCdgW14ZqNuI7qjeFleYxm86LG5JxX2j3/8Y+wNN9ywyrRp\n02aXziJJan+Rog/YnGc367akmr7ai4JqdOF6wB7P+maKJ4CbgD8BV9W36613pZHRqnrXJp7UYSLF\nVsDu9W0PYFuqH+IanqGaezdRNfX+CvwBuNxpClJzzZw5s++ggw7a/DOf+cyda621lutbSpKeoV72\nZVfguTzdrNuWau1lDd/KwJT69rb6vscjxZ95uql3VW7kWwvlk7pWK+tdm3hSG4sUKwEv4OmG3W7A\npKKhutdYni4UX1ffNzdSXAlcDFxE1dR7okw8qfPNnTs3XvnKV27+2te+9uG3vOUtj5bOI0kqL1Ks\nA+w14DaFavkUNd8qVL9TPDVyL1I8CFxN1dS7GrgyN/KDZeJJna/V9a5NPKmNRIp1ebphtzvwfKrm\nksoYB7yovn2UarTeVTzd1PtjbuTHy8WTOkd/fz+HHnroxltttdWTH//4x+8vnUeSVEakWB94MVV9\ntRfVUjAqZyLw8voGQKS4naqp90fgV7mRbyoTTeosI1Hv2sSTCqqbdq8A9qFq3G1aNpGWYiywZ337\nMDA/UlzN0029P+RGnlMuntS+fvOb34w/99xz195yyy2f2GabbbYDSCnd/frXv35m6WySpNap12/e\nG3hJfdu6aCANxyb17XXAFyLFzcAvgV8AF+dGnlsumtS+RqLejZxzs7YlaRgixfOB/evbC3A9u26y\ngGrx4IuB84FLciMvLBtJgunTp98+ZcqUtp8aM3369IlTpkzZpHQOSdLyqy9CsSfwMqqm3VScHttN\n5gC/o2rq/TI38p2F80hA79S7jsSTWixSrEw1ZeAA4JXA+mUTqYVGUy3EvCvwQeChSPEz4CfABZ61\nlCRJ3ahu3L0IeC1wMDC5bCK10KrAq+obkeKvVCP0fkm11IwnsKUWsokntUCkGE810u5gqvUlVi2b\nSIWsDRxR32ZFil9RNfR+kRt5VtFkkiRJK2Cxxt1BwLplE6mQHevbfwCPRIoLqJp6v/ICGVLz2cST\nmiRSTKA6I3Uw1fSBlcomUptZjWpdkddRXfX2AuAs4GeuoydJkjpB3bjbi6qesXGnxa0JvL6+9UeK\nPwLfBX6YG/mRosmkLmETT1oBkWIV4BCqH1QvwSvJanjGUU2vPgB4vJ5yexbw69zI84omkyRJGmBA\n427RVFkbdxqORWsj7gl8KVL8gqqh93PrXWn52cSTlkOk2BF4J/BGYELhOOpsqwCH1rdHIsU5VA29\nC3Mj9xdNJkmSetKAi1O8Dht3WnFjgdfUt0cixY+A7+RGvqxsLKnz2MSThqm+QMXrgaOA3QrHUXda\nE3hbfbszUnwNOC038oyysSRJUi+IFGsBRwLvAjYrHEfdaU2q36eOihQ3A2cAZ+ZGvq9sLKkz9JUO\nILW7SLF9pPgScA/wTWzgaWRsCHyKqpn37UjxwtKBpBV19tlnr77JJpvssNFGG+3woQ99yFEdktQm\nIsXzIsUZwF3A57GBp5GxBfBpqnr33Eixf6QYVTqUtCJaXe86Ek8aRD3q7nVUZ4l2LxxHvW0c8Cbg\nTZHiKuBU4AeuJaIVE1Obu718zdIesWDBAo477riNzj///Js222yz+VOmTNn24IMPfnTq1KlPNjeL\nJGk4IsVYqrWd34MnqVXWaODA+nZ3pDgTOCM38m1FU6nDdWe960g8aYBIsV2k+CJwN3AmNvDUXnYG\nvk11tvKTkWKD0oGk4broootW3Xjjjedut91281ZaaaV80EEHPXz22WevUTqXJPWaSLFBpPgv4A7g\ne9jAU3tZH/gwcHOk+HGk2Ll0IGm4RqLedSSeel6kWInqalvvBPYoHEcajnWoipsPRoqfAqfmRr6o\nbCRpaHfeeefY9ddf/6kRpBtssMG8K6+8cnzJTJLUSyLFPsAxVKOd/D1Q7a4POAg4KFJcBHw2N/Kv\ny0aShjYS9a7/eatnRYpVqBbtPRGvuKXONJrqinEHR4q/Al8GvpsbeU7ZWJIkqR1EivFUy3IcA2xf\nOI60vPYG9o4UfwE+B/wwN/LCspGkMpxOq54TKVaNFB8AbgP+Bxt46g47Al8D7ooUJ0eKDUsHkgba\ncMMN5919991jF/39rrvuesaZSklS80SKSZHi81RLxHwFG3jqDs8F/g/4Z6Q4pl7HXGobI1Hv2sRT\nz4gU4yPFfwC3U53BWadsIqkl1gCOpSpuvhgpJpcOJAFMmzZtzu23377SjTfeOPbJJ5+Mc845Z62D\nDz740dK5JKmbRIo1I8V/A7dSzTZZvXAkqRU2pbrY278ixUcjxVqlA0kwMvWu02nV9SLF6sB7geOA\ntQvHkUbKOOB9wNsjxSnA53IjP1w4k3rYmDFjOOmkk+7Yb7/9tlq4cCFveMMbHnzBC17glWklqQnq\nevd4qnrXxp16xSTgE8C/R4rTgZNyI99VOJN62EjUu5Fzbub2pLYRKSZQNTGOA9YsHEcqbSbw/4CT\ncyPPKh1GI2v69Om3T5ky5cHSOZZm+vTpE6dMmbJJ6RyS1CkixapU9e6JgKOR1OvmU023/Vxu5BtK\nh9HI6pV615F46jqRYtF0wvdTTS2UBBOABLw3UnwW+HJu5CcKZ5IkScuhXgvs3cAHqUYjSYIxwFuA\nN0WKbwMfzo18T+FMUlPZxFPXqNdCOI7qbKTTCKTBTQQ+DxwfKT4FnJYb2YsLSJLUASLFOOAo4D+B\n9QrHkdpVH/BW4LWR4n+oRuY9XjaS1Bxe2EIdL1KsEikS1QUrPoINPGk41qNaEPimSHFkpBhVOpAk\nSRpcpBgTKY4C/gl8CRt40nCsCjSoLvj21khh/0Mdzw+xOlqkOAT4O/AxYLXCcaROtDFwBnBDpDg0\nUkTpQGqJ/v7+/rZ+b+t8/aVzSFI7iRR9keItwD+A/wU2LBxJ6kTPAb4J/ClS7FM6jFqmJ+pdm3jq\nSJFim0hxAfAjYKPSeaQusBVwFvCXSPHi0mHUdNfPmDFjQrsWNv39/TFjxowJwPWls0hSu4gUU4Er\ngDOBTcumkbrC84DfR4qfRoqtSodR0/VEvevVadVRIsV4qlF3x1ItXCqpNb4LnJAb+YHSQbTirrnm\nmnVGjx59OrAD7XkCrx+4fsGCBW+fOnWqnzlJPS1STAA+BbyL9vw/W+oG84GvASk38kOlw2jF9Uq9\naxNPHSNSHEa1IP/6pbNIPeIRqqvenZ4b/rCQJKnVIsUbgJOAdUtnkXrEo8AngVO82Js6gU08tb1I\nsT3VAvx7F44i9ao/AEfnRnaqoyRJLRAptga+AuxbOovUo24B/iM38tmlg0hDsYmnthUpVgcS8B5g\ndOE4Uq+bTzUy4BO5kZ8oHUaSpG4QKVYGPgx8ABhbOI4k+DXwjtzId5UOIg3GJp7aTn11zDcBnwMm\nF44j6ZluA47Jjfyr0kEkSeoa4uZMAAAgAElEQVRkkeKVwCl40Qqp3TwGHJ8b+YzSQaTF2cRTW4kU\nz6WaOrtH6SyShvQj4P25ke8tHUSSpE4SKTYEvgS8unQWSUM6n2pU3p2lg0iL2MRTW4gUY6imzv47\nMKpwHEnD8xjVFKCv5EbuLx1GkqR2FilGA8cBDWDVwnEkDc9jwIm5kU8rHUQCm3hqA/VCvt8DppbO\nImm5XA0clRv5L6WDSJLUjurZJt8BdiidRdJyuYBqVN4dpYOot9nEU1GR4miqxfJXKZ1F0gpZSPVv\n+SO5keeXDiNJUjuo13o+FvgMXrhC6nSOylNxNvFURKSYBJwBHFA6i6Smuho4NDfyraWDSJJUUqSY\nDJwJ7Fc4iqTm+g3wdkflqYS+0gHUeyLFy4G/YgNP6kYvBP4cKV5fOogkSaXU9e512MCTutG/AddH\nineWDqLe40g8jZhIsTLweeCY0lkkjYgzgPflRn68dBBJkkZCpBgHfBZ4HxCF40hqvd9Sjcr7V+kg\n6g028TQi6sV8/w/YtnQWSSPqBuD1uZGvLx1EkqRWihTbAmcBU0pnkTSiHgOOyI18Tukg6n428dRS\nkaIPOBH4L1zMV+pVTwDH5Ub+39JBJElqhXpa3f/Di7VJvezzwH/mRl5YOoi6l008tUyk2BD4NrB3\n4SiS2sOPgHfkRp5ZOogkSc0QKdYCTgdeUzqLpLZwIdVF3h4oHUTdySaeWiJSvA74X2CN0lkktZXb\nqQqbK0sHkSRpRUSKvYHvABsUjiKpvdwNHJIb+YrSQdR9bOKpqerps58Hji+dRVLbmg98BPh8bvhD\nSJLUWSLFaCAB/wH0FY4jqT3NA47Pjfzl0kHUXWziqWkixepUi/m+onQWSR3hfODNTjeQJHWKSLEm\ncDawb+kskjrCd4F35kZ+vHQQdQebeGqKSLEZcB6wXekskjrKfcDBuZH/WDqIJElDiRRbAj8Htiqd\nRVJHuY6q3r25dBB1Pod/a4VFimnAVdjAk7Ts1gV+V6+jKUlSW4oU+wBXYANP0rLbCfhTpHhV6SDq\nfI7E0wqJFO8AvgyMKZ1FUkfLwIdzI3+6dBBJkgaKFG8HvoL1rqQVk4FPAx/NjdxfOow6k008LZdI\nMQo4GXhv6SySusrpwLtyIy8oHUSS1Nu8YJukFvkNcFhu5IdKB1HnsYmnZRYp1gB+ALy0dBZJXem3\nwCG5kWeWDiJJ6k2RYjzVBdv2L51FUle6HdgvN/I/SgdRZ7GJp2VSL+h7HrB16SySutr1wCtzI99R\nOogkqbdEio2o6t2dSmeR1NUeBF6RG/nq0kHUObywhYYtUrwYuBIbeJJabwfgikgxtXQQSVLviBS7\nUF2wzQaepFabCPw+UjjDTcNmE0/DEineDfwaWLN0Fkk9Yz3gEq/kJUkaCZHiUOAiYHLhKJJ6x3jg\n55HisNJB1BmcTqsh1Qv6ngK8u3QWST2rHzg+N/IXSweRJHWnSPFxoFE6h6SelYFjcyN/qXQQtTeb\neFqiSDEa+DbgWQFJ7eAU4LjcyAtLB5EkdYdIMQb4Fta7ktrDp3Mjf6h0CLUvm3gaVF3QnAUcXDqL\nJA3wc+DQ3MhzSgeRJHW2SDEW+BHgsg2S2skZwDs9ca3B2MTTs0SKccDZwP6ls0jSIK4FXpob+aHS\nQSRJnamud38MvLJ0FkkaxM+oTlw/UTqI2otNPD1DpFgZ+AnwstJZJGkI04F9cyM/XDqIJKmzRIqV\ngHOx3pXU3i4DDsiN/GjpIGofNvH0lEixKlXHf9/SWSRpGP4MvMRGniRpuCLFKlT17otLZ5GkYfgr\nsF9u5HtKB1F7sIknACLFasAvgT1LZ5GkZXAtVSPvkdJBJEntrT5h/QtgWukskrQM/kW1lMxNpYOo\nPJt4IlKsAfwa2KV0FklaDtdQNfKcaiBJGpQnrCV1uBnAtNzIfy8dRGXZxOtxkWIt4DfA80tnkaQV\ncDXwb7mRZ5YOIklqL5FidaoT1ruVziJJK+BuYK/cyLeVDqJybOL1sEgxCfgtsFPpLJLUBFdRNfIe\nKx1EktQe6hknFwAvLJ1FkprgNmBP18jrXX2lA6iMSLEecDE28CR1j52B8+sRF5KkHlfPOPkdNvAk\ndY9Ngd9Giomlg6gMm3g9KFJsQNXA27Z0Fklqsl2BX9drH0mSelT9C+7vcckYSd1nW6oT1xNKB9HI\ns4nXYyLF+sAlwJals0hSi+wG/CpSjC8dRJI08iLFOsCFwJTSWSSpRZ4P/CJSrFI6iEaWTbweUk8x\n+yXVEFxJ6mZ7YCNPknpOPTLlt8AOpbNIUovtAZwbKcaVDqKRYxOvR0SKMcCPcQ08Sb1jT6ozlKuW\nDiJJar36F9mfAjuWziJJI+TfgO9HitGlg2hk2MTrHacDLykdQpJG2IuAn3uGUpK6W6ToA74LTCud\nRZJG2KuBb0aKKB1ErWcTrwdEik8Cby6dQ5IK2Rs4rXQISVJLfRE4pHQISSrkjcCXS4dQ69nE63KR\n4h3Ah0vnkKTC3hQpPlI6hCSp+SLFh4D3lM4hSYW9K1J8tnQItVbknEtnUItEildSrQsyqnQWSWoD\nGTg0N/IPSweRJDVHpHgr8M3SOSSpjXw0N/InS4dQazgSr0tFihcAP8AGniQtEsC3IsUupYNIklZc\nSuml27P9u0rnkKQ281+R4sjSIdQajsTrQpFiU+ByYHLpLJLUhu4HdsmN/K/SQSRJyyeltBNwWSav\ndCEXXnkJl+xZOpMktZF5wN65kS8vHUTNZROvy0SKtYA/AluXziJJbeyvwB65kWeVDiJJWjYppfWA\nK4ENF913PddfdDZn710slCS1n3uBF+RGvqd0EDWPTbwuEilWAn4L7FE6iyR1gF8BB+RGXlg6iCRp\neFJKqwKXAM9f/Hv3cu9lp3Harv30jx75ZJLUlq4CXpQbeW7pIGoO18TrEpGiD/gONvAkabheDpxc\nOoQkaXhSSn3A/zFIAw9gPdbb8ziOmz6WsY6ylqTKzsDXS4dQ89jE6x4nAYeUDiFJHea9keKY0iEk\nScPSAF411ANWY7WpJ3LiPauz+r0jlEmS2t2bI8VxpUOoOZxO2wUixdHAV0vnkKQOtRDYPzfyr0sH\nkSQNLqW0P/AzqiuNL1U//feezumz7+GeLVubTJI6wkJgv9zIvy0dRCvGJl6HixQ7A5cCY0tnkaQO\n9hiwe27kv5UOIkl6ppTS5sCfgDWW5XmZ/NiP+NHNN3DDoNNvJanHPAy8MDfyraWDaPnZxOtgkWJt\n4Fpgo9JZJKkL3A7skhv5gdJBJEmVlNIqwOXATsvz/EyefyEXXnkJl+zZ3GSS1JGuB3bLjTy7dBAt\nH9fE61D1hSy+hw08SWqWTYBzI8WY0kEkSU/5OsvZwAMIYsy+7LvnwRx8cRMzSVKn2gH4dqQY1tIE\naj828TrXR4GXlQ4hSV1mN+AzpUNIkiCl9F7g8GZsa0d2nPZO3nlpH30LmrE9SepgrwE+VjqElo/T\naTtQpHgZ8EtswkpSK2TggNzIvygdRJJ6VUppD+BCoKmjo2cx65pTOGWrecxbrZnblaQOk4GDciOf\nWzqIlo1NvA4TKTaiWgdv7dJZJKmLPQRMyY18d+kgktRrUkrrUtW767Vi+/OY949TOXXCYzy2biu2\nL0kdYjbV+njXlw6i4bOJ10FSSqPnMe+nP+bHz/kH/3hu6TyS1OUuBfbJjbywdBBJ6hUppdHA74G9\nWrmffvrvOY3T5tzLvVu2cj+S1OZuBJ6fG/mJ0kE0PE7H7CyfGsvYVxzKoTsezMEXBdFfOpAkdbG9\ngFQ6hCT1mE/Q4gYeQB99zzmKo9bZhm3+3Op9SVIb2wb4f6VDaPgcidchUkr7Ua2D99RVZGYz+9qv\n8tUN5zBnUrlkUpuZD3wTWAj0A9sB+wC3AhfU9z8HeBUwapDnXwD8k2qViM2Al9fPOQt4DHghsHP9\n2J8BL6i3p27VD7w0N/LvSgeRpG6XUtoTuJgRHGiQyfN/z++vvJRL9xypfUpSG3pVbuTzSofQ0tnE\n6wAppfWA6cCzmnX99N//A35wr9NrpVoG5gHjqJpv36C6jvPZwJuBiVSTdNYAnr/Yc+8AfgMcUf/9\nG8CLgbnA/VTjAr4BvB24D7gSOLB1L0Vt4z5gp9zIM0oHkaRulVJaDbgO2KTE/v/KXy/+MT+eVmLf\nktQGZlDVu/eVDqKhOZ22M5zBIA08gD76Jju9VhogqBp4UDXxFlL9TzeKqoEHsDlwwxKeu6B+zqKv\n4+vnzqcak7XovMfvqUb4qResC5xWOoQkdbkvUaiBB7AjO047iqMu7aNvQakM0nKbD3wd+CrwZarr\nOkM1E+Vr9f1nUF22azD3AafXz/1Kvb0FwHfq+64a8NifAfc0N77awiTgzEgRS32kirKJ1+ZSSkdR\nTehboiBG7ciOe5/ACX9ZlVUdKSL1UxUrn6dq2K1f37foOqM3UE2NXdyGVL8+/A9wErAF1Y+zzYBH\nqYqbXaiWf10PWL1VL0Bt6MBI8fbSISSpG6WUDgLeWjrHc3jOXsdx3F/GMnZW6SzSMhkNvAV4F3A0\ncDNwJ/Bz4OD6/h2BSwZ57kLgHGB/4Biqf4mj6m1sVD/3uvqx91Gd0HYpmW71MuD9pUNoaDbx2lhK\naTOqVsKwjGf880/ghP6t2fovLYwltb8+qoLjeKrG3QPAIcD5VGcpxzJgdckBHgIerJ93PHAb8C+q\nQuYQqqJoe+AKYHfg18APqJp66gVfiBRblA4hSd2kXjbm66VzLLIaq73gRE68Z3VWd0qZOsdgM1Gi\nvs2t758LrDbIc28BJlPNOwBYhadnsTgTpRd9JlLsWDqElswmXptKKfUB36KazDdsTq+VBliZamTd\nzVSj7I4EjgI2BtYe5PE3AhtQFUHjqEbi3bnYY64GpgB3ASsBrwUub350taVVge9EisEuiSJJWj7f\nYPCfysWMZezWx3Js/3qs98/SWaRhW3wmygZUF3L7HtWwkOnAYJdveYiq2fcdqqm3l9X3OxOlV40D\nvhUpxpQOosHZxGtfxzP4f7NL5fRa9bQ5wBP1n+dTrQUyEZhd37cA+APVVWUXNwG4nafPYP6LZ65G\n+QRwE1UTbz5Pj+ab37T0an+7Ah8pHUKSukFK6Rhgv9I5BtNH33OO4qh1tmGbP5fOIg3L4jNR7qc6\n0Xw4cALwPKpZKYvrp7q420FUJ7xvpKqfnYnSy54HfLh0CA3Oq9O2oZTS9sA1PD0oerl59Vr1nPuA\nc3l66P/2wN7ABVQNuEzVwNutfvzdwJ+orjLbD/yCqnkH1Ui8gb9a/BrYGtiUqnF3FjCr3t4uLXo9\nakcLgD1yI1+11EdKkgaVUtoa+DPVuPm2lcnzfstvr/4Df9ijdBZp2C4CxlDVuItWOHsU+C7wnsUe\n+1eqWSuvqf9+MdUaewM/8VdQ/Wa6OtUslRdRzRk7ovnR1TYWADvnRvZERpuxiddmUkpjgCuput9N\nkckLr+f6S8/hnBdlsqMvJWnF3QTslBt57lIfKUl6hpTSaKoxQoONi29L13HdxedwzrTSOaRBzaEa\nibcy1Ynm71A14c4F3kY1K+Va4J/A6xd77hNUDbkjqUbffZfqZPdWA77/I+CNVNXP/cBeVNNsj2rV\nC1KbuB6Ymht5Xukgetro0gH0LB+miQ08eHp67aZseu1X+eqGc5gzaenPkiQNYSvgg8AnSgeRpA50\nHB3UwAPYiZ2mrc3al57BGbv10+/vUGovs3j2TJStqdbE+yHVEjArUc08gWoq7D3AvlSNv92A0+rv\nbcnTDTyoRubtRdUk3By4iqq101H/grWcdgA+DnyocA4N4Ei8NpJS2oZqydGxrdqH02slqWmeBLbP\njXxr6SCS1ClSShsDN1BdA7PjzGLWn07hlG3mMW+ZLj4nSR1qIbC7y8i0D6dWtpev0cIGHjzj6rUX\ne/VaSVohKwFfKh1CkjrMqXRoAw9gNVZ7wYmceNdqrHZ/6SySNAJGAWdGipb2KTR8NvHaRErpCGBE\n1tmop9dO8+q1krTCXhkpXrP0h0mSUkoHAfuXzrGixjJ2m2M5dsF6rHdz6SySNAK2pVoGQW3A6bRt\nIKU0kWplgrVHet9Or5WkFXYHsG1u5MdLB5GkdpVSWg34O7B+6SzNkskzf8APbr2RG5u6nrUktaHZ\nwFa5ke8tHaTXORKvPZxEgQYeOL1WkppgI+BjpUNIUpv7L7qogQcQxITX8/rt92CPP5TOIkktNh74\nbOkQciRecSmlfYDfl84BMJvZXr1WkpbPfGBKbuS/lw4iSe0mpfR8qmtajiqdpRUyOV/HdZf8hJ+M\nyNI4klRIBvbIjXx56SC9zCZeQSmlccB1PPMi3kU5vVaSltuFuZH3LR1CktpJSqkPuBJ4QeksrXY3\nd196Bmfs1k//6NJZJKlF/gTskhvZWXyFOJ22rA/QRg08cHqtJK2AfSLF4aVDSFKbOYYeaOABrM/6\nex3LsX8Zy9jZpbNIUou8ADiydIhe5ki8QlJKzwFuAlYtnWVJnF4rScvsPmCb3MgzSweRpNLqevfv\nwOqls4ykecy78RROWXMWsyaXziJJLTAD2NJ6twxH4pXzKdq4gQcwnvHPP4ET+rdm67+UziJJHWJd\nqsXbJUlVvdtTDTyAsYzd5liOnb8u695SOosktcAkIJUO0asciVdAvbjvn4AonWU4Mnnh9Vx/2Tmc\ns1cm2/iVpKEtBF6YG/nPpYNIUikppR2Bv9DDgwYyeeb3+f5trjUtqQstoLqo2w2lg/Sanv2hWtjJ\ndEgDDyCIUTuy47QTOOEvq7LqjNJ5JKnNjQK+Eik65v95SWqBz9Hjv2sEMeFQDt1ud3b/Q+ksktRk\no4Evlg7Ri3r6B2sJKaWDgBeVzrE8nF4rScO2K/C20iEkqYSU0kuA/UrnaAdBjP03/m33V/Pqi0pn\nkaQme0mkeE3pEL3G6bQjKKU0FrgB2Lx0lhXh9FpJGpb7gM1yIz9ROogkjZSUUgDXAM8rnaXd3MVd\nl57BGbtn8qjSWSSpSW4DtsuN/GTpIL3CBszIej8d3sADp9dK0jCtC7yrdAhJGmGHYwNvUBuwwV7H\ncdy1Yxgzp3QWSWqSTYEPlA7RSxyJN0JSSmsBtwITSmdppn767/8BP7jXBXslaVAPAJvmRn68dBBJ\narWU0jjgJmCj0lna2Vzm/v1UTl17FrPWKZ1FkprgcWCb3Mh3lg7SCxyJN3JOpMsaeAB99E0+lEN3\nPJiDLw6iv3QeSWoz6wDHlA4hSSPkfdjAW6pxjNv2WI6dN5nJt5TOIklNsArwodIheoUj8UZASmkS\n1VzxVUtnaaXZzL72q3x1wznMmVQ6iyS1kRlUo/GcPiWpa9WzTm4B1iidpVNk8syzOOu2m7jJGS2S\nOt1cYPPcyHeXDtLtHIk3Mj5IlzfwwKvXStISTALeUzqEJLXYR7CBt0yCmHAYh223G7v9sXQWSVpB\n44B/Lx2iFzgSr8VSSutRnZVcuXSWkeLVayXpWR6iGo03q3QQSWq2lNL6VPXuuNJZOlEm5+lMv/hc\nzt27dBZJWgFPUNW795cO0s1ssLTeh+ihBh549VpJGsTawHtLh5CkFjkRG3jLLYh4Ls/d+2287ZIg\nFpbOI0nLaWWqnwdqIUfitVBKaUPgn/RwUePVayXpKQ9TnZ18rHQQSWqWlNJE4F9UC5trBc1k5tWn\ncup285nf9UvxSOpKc4BNciM/WDpIt3IkXmt9hB5u4IFXr5WkAdYC3l86hCQ12bHYwGuaCUx44Ymc\neMd4xjubRVInWhU4rnSIbuZIvBZJKW0C3ASMKZukfXj1WkniUaqzkzNLB5GkFZVSmkA1Cm9C6Szd\nZiEL7/o6X597P/dvXjqLJC2jx6jq3UdKB+lGjsRrneOxgfcMXr1WkliDatSKJHWDY7CB1xKjGLXB\n0Ry99lZsNb10FklaRqsD7ysdols5Eq8FUkprA3fg1IJBefVaST1uJtXZyUdLB5Gk5ZVSWoVqFN7E\n0lm6WSbPvYALrrmcy3cvnUWSlsEjwMa5kWeVDtJtbKC0xruxgbdEXr1WUo+bQDVaW5I62VHYwGu5\nIMa9lJfudiAHXlw6iyQtgzWpRmuryRyJ12QppZWoRuG57tswePVaST3KtUIkdayU0ljgVmD90ll6\nyZ3ceck3+MYemTyqdBZJGoYZVPXu46WDdBNH4jXfW7GBN2xevVZSj1odeFfpEJK0nN6CDbwRtyEb\nvuhYjr1mDGPmlM4iScMwCTi6dIhu40i8Jkop9QH/ALYonaUTefVaST3mX8BmuZE9gSGpY6SURlHV\nu141tZC5zL3hFE6ZNJvZ1syS2t19wKa5kZ8sHaRbOBKvuV6DDbzl5tVrJfWYjYFXlA4hScvoAGzg\nFTWOcdsdx3FzJzP51tJZJGkp1gWOLB2im9jEa64PlA7Q6ZxeK6nHOKVWUqd5d+kAglGM2uBojl5r\nS7acXjqLJC2FPzeayOm0TZJS2gW4onSObuL0Wkk9oB/YIjfybaWDSNLSpJS2pJpKG6WzqJLJc8/n\n/Guu4IrdS2eRpCHskRv5j6VDdANH4jWPoymazOm1knpAH/DO0iEkaZjehQ28thLEuJfxst1exasu\nLp1FkoZwVOkA3cKReE2QUloTuAdYqXSWbpTJC6/n+svO4Zy9MtnGs6RuMwPYIDfyvNJBJGlJUkor\nA3cDa5bOosHdwR2XfJNv7pHJo0pnkaTFPAE8Jzfyo6WDdDobIs3xVmzgtUwQo3Zkx2kncMJfVmXV\nGaXzSFKTTQIOKR1CkpbiMGzgtbWN2OhFx3LsNWMY83jpLJK0mJWBw0uH6AY28VZQSimAo0vn6AWL\nptduwzZ/Lp1FkprMBX8ltTv/n+oAE5iw84mcePt4xnviW1K7cUptE9jEW3H7AluVDtEr+uib/Hpe\nP+UQDvHqtZK6yR6RYsfSISRpMPUF3KaWzqHhGce47Y7juCfXYR0vmiSpnewUKXYpHaLT2cRbcY7C\nG2FB9O3ADk6vldRtvECSpHblKLwOM4pRG76Ld62xBVtcVzqLJA3wjtIBOp0XtlgBKaX1gDuA0aWz\n9Kp++u//IT+850ZufF7pLJK0gmZRLfg7u3QQSVokpbQ2cBeu/9yRMnnur/n1tVdy5W6ls0gSMAdY\nLzfyrNJBOpUj8VbMkdjAK8rptZK6yGrAm0qHkKTFHIENvI4VxLj92G/XAzjg4tJZJAlYFS9wsUJs\n4q0Yf9lqA06vldRFnFIrqd28pXQArZggYipTpx3JkZd40ltSG/ACFyvA6bTLKaX0QuCq0jn0TE6v\nldQF9syN/IfSISQppbQj4JpqXeRRHr3qy3x5h/nMX6V0Fkk97QW5ka8pHaITdcVIvIhYKSKuiojp\nEfG3iEgjsFtH4bUhp9dK6gJeMElSu3hD6QBqrjVYY+cTOfE2Z69IKszReMupK0biRUQAq+acZ0fE\nGOAy4P055ytasb+U0mjgbmCdVmxfzTGb2dd+la9uOIc5k0pnkaRlMBuYlBv5ydJBJPWulFIAtwKb\nFI6iFljIwju+xtcWzmDGpqWzSOpJXtBtOXXFSLxcWfTmj6lvrexOvhQbeG1vPOOffwIn9G/DNn8u\nnUWSlsF4qp8zklTS7tjA61qjGLXRu3n3GpuzudOlJZWwGnBo6RCdqCuaeAARMSoi/gI8APwm53xl\nC3f3xhZuW03k9FpJHerg0gEk9Tyn0na5INZ8I2/camd2vrx0Fkk9ySbecuiK6bQDRcQawE+A9+ac\nr2/29lNK44H7AReD7TCzmX3t1/jahrOZ7fRaSe3uEWBybuT5pYNI6j310jH3ANZMPSCT87Vce8l5\nnDetdBZJPWUB1RIyj5YO0km6ZiTeIjnnR4ELgf1atIuDsIHXkcYz/vnHc3x2eq2kDrAmsE/pEJJ6\n1r9hA69nBBFTmTrtCI5w5oqkkTQa2L90iE7TFU28iJhUj8AjIlamKjxubNHuDmvRdjUC+uhbx+m1\nkjqEU2ollXJ46QAaeRuz8bT38/6rRzP6idJZJPWMV5cO0Gm6YjptROwEfAv+P3v3HSZlfaj///3s\nbO9sYxtdVIp0FJAiUUGNqFHRGGvq135ywJhf4kn2jDGJSSSaeBI1VVGjYsWODRdREFCKgCC9LHVh\nK7s7Mzvz/P6YWensAjPzmXK/rmsvZmdn5rmHC3aeuedTcOAvJmfYtn1vsI/jdDpzgN34N86QKKfp\ntSIS4Xbi37VLHziISNg4nc50/L9/Mk1nETNaaFnxMA933se+AtNZRCTmNeKfUttiOki0iImReLZt\nL7Nte7Bt2wNs2+4figIv4GJU4MUMTa8VkQjXGRhtOoSIxJ1JqMCLa6mk9pvClKZCCjeYziIiMS8T\nOM90iGgSEyVeGF1uOoAEl6bXikiE05RaEQk3/d4RHDi63sqtOb3otcx0FhGJeZpSexxiYjptODid\nzjSgGm1qEbM0vVZEItAWoJtdoRdrEQk9p9OZhP98N9t0FokMNnbLW7y1eAELRprOIiIxaxdQoiVk\nOkYj8TrufFTgxTRNrxWRCNQFGG46hIjEjXGowJMDWFipF3LhiG/yzUrTWUQkZhUBo0yHiBYq8Tru\nEtMBJPQ0vVZEIpCmtolIuEwyHUAij4VlDWf4uJu4SefHIhIqmlLbQSrxOsDpdFr4N7WQOGBhJfSn\n/7ipTF2SSeZu03lEJO5pPVYRCReVeHJU3ek+7k7uXJhIYrPpLCIScy41HSBaqMTrmLPw7xIocUTT\na0UkQpxiOa0BpkOISGxzOp39gB6mc0hk60Sns+7irvUZZFSbziIiMeUUy2n1Nx0iGqjE65gLTQcQ\nMzS9VkQihKbUikioXWQ6gESHVFL7TWHKvkIKN5rOIiIxRVNqO0AlXsecbzqAmKPptSISATSlVkRC\n7QLTASR6OHB0u5Vbs3vR6wvTWUQkZqjE6wDLtm3TGSKa0+nMBvYAiaaziHk+fLtmMKNqFasGm84i\nInGn2K6wd5oOISKxx8sds14AACAASURBVOl0ZgB7gWTTWSS62Ngtb/LmkoUsHGE6i4jEhC52hb3V\ndIhIppF47TsHFXgSoOm1ImLQaNMBRCRmfQMVeHICLKzUi7jozG/yzUrTWUQkJmiDpXaoxGvfeaYD\nSGTR9FoRMWSM6QAiErO0/rOcMAsrYTjDx93IjfqQW0RO1ljTASKdSrz2aT08OSLtXisiYaaTGhEJ\nFZ3vyknrQY9xd3LngkQSm01nEZGodbbpAJFOa+Idg9PpLAM0H1uOycb2rWDFRy/y4hgbW8W4iISK\nD+hkV9j1poOISOxwOp2dgR2mc0jsaKFl+Z/5c0kTTfmms4hIVOpmV9ibTYeIVCocjk1TaaVdml4r\nImGSAIwyHUJEYo7W25SgSiW1/1SmNhZSuNF0FhGJShqNdwwq8Y5NJZ50mKbXikgYaF08EQk2lXgS\ndA4c3W7l1qye9PzCdBYRiToq8Y5BJd6xaf0hOS7avVZEQkwlnogEm94sSUhYWPnXc/0pwxg233QW\nEYkqel06Bq2JdxROp7MUqDKdQ6JXI42fP8qjXRppLDSdRURihgvIsStsl+kgIhL9nE5nOlAHJJrO\nIrHLxvYtZOHcN3lTAyREpCO8QJ7WgT4yjcQ7upGmA0h00/RaEQmBFGC46RAiEjNGoAJPQszCSjiT\nM8feyI1zNFNFRDrAgf/1SY5AJd7RqcSTk6bptSISAhrJICLBovXwJGx60GPsHdyxIJHEFtNZRCTi\naUrtUajEOzo1vxIU2r1WRIJM6+KJSLDoTZKEVR55I+7irjXppO8xnUVEIppen45Ca+IdgdPpTALq\ngVTTWSS2+PDtmsGMqlWsGmw6i4hErXqgk11ha3SviJwwp9PpAGqALNNZJP548W56hEeoprqb6Swi\nEpEagVy7wvaaDhJpNBLvyAahAk9CQNNrRSQIsoGBpkOISNQbgAo8McSBo9tt3JbZgx7LTWcRkYiU\nib+XkUOoxDsyrYcnIaPptSISBJpSKyInS1OVxCgLK/8Gbug1lKHzTWcRkYik16kj0G5UR6YSrwNe\neeUVvvrqKzIyMrjtttsAaGpq4oUXXqC2tpbc3FwmT55MWlraYfd95513WLNmDbZt07NnTy688EK8\nXi/PPPMM9fX1DB8+nDPPPBOAV199lWHDhlFaWhrW5xdqgd1rd81gxmJNrxWR4zQG+LPpECIS1Yaa\nDiBiYaVdzMVnFlFU+RZvjTOdR0QiytnofPcwGol3ZDqp6YBBgwZx3XXXHXTd3Llz6dGjB3feeSc9\nevRg7ty5h91v8+bNbNmyhVtuuYVbb72Vbdu2sXHjRtauXUvXrl255ZZbWLZsGQA7duzAtu2YK/Da\naHqtiJygs0wHEJGod4bpACLgn6VyFmeNu4EbdD4sIgfSSLwjUIl3CKfTmQ70Mp0jGnTv3v2wUXar\nV69m0CD/1PVBgwaxatWqw+5nWRatra14vd6v/8zMzMThcODxePD5fLRtuPLBBx8wfvz40D8ZgzS9\nVkROQBfLaWWaDiEi0cnpdCYAfU3nEDlQT3qOu4M7FiSS2GI6i4hEhDLLaXUxHSLSqMQ7XD/093LC\nGhsbycryr5GcmZlJY2PjYbfp0qUL3bt354EHHmDatGmccsopFBYW0rNnT2pra/nHP/7BWWedxapV\nqygpKSE7OzvcT8OIwPRa+3ROX2w6i4hEhdNNBxCRqHUKcPh6JyKG5ZE3YipT16STvtd0FhGJCPrA\n6RAqqw6nqQVBYlkWlmUddv2ePXuorq5mypQpTJkyhQ0bNrBp0yYcDgdXXnklN998M/369WP+/PmM\nGjWKt99+m+eee+6Io/pijabXishxUIknIidqgOkAIkeTRtoZU5hSn0/+JtNZRMS400wHiDQq8Q6n\nk5qTkJmZSUNDAwANDQ1kZGQcdptVq1ZRXl5OSkoKKSkpnHLKKWzZsuWg2yxcuJCBAweydetWUlNT\nmTx5MvPmzQvLczBN02tFpIP6mA4gIlFLH1pLREsksfvt3J7Rgx4rTGcREaNU4h1CJd7hdFJzEk47\n7TSWLFkCwJIlSzjttMP/z+Xk5LBx40a8Xi9er5dNmzZRWFj49c+bm5v56quvGDhwIB6P5+vRfB6P\nJzxPIkJoeq2ItEMj8UTkROl8VyKehVVwAzf0HMrQT01nERFjdL57iETTASKQTmo66IUXXmDjxo00\nNTUxbdo0xo8fz+jRo3n++edZvHgxOTk5TJ48GYCqqioWLVrEpZdeSt++fdmwYQOPPPIIAKeccspB\nZV9lZSVjxowhISGBXr16sWDBApYvX86wYcOMPE+TAtNrC1awovJFXhxjY6t4F5E2GoknIidK57sS\nFSystIu5eHghhZVv8/Y403lEJOw0Eu8QVtsuoAJOp7MzsMN0DpEjaaTx80d5tEsjjYXt31pE4oAb\nyLAr7FbTQUQkejidzgygHs3IkSizjnWVT/LkWODwRbdFJJZl2RX24Ttmxim9eB9Mn0pKxNL0WhE5\nRDLQ03QIEYk6/dB7AIlCveg17k7unJ9IYovpLCISVqeaDhBJ9AJ+MG1fLBFNu9eKyCE0pVZEjpc+\ntJaolUfeyKlM/Sqd9L2ms4hI2GhK7QFU4h2sl+kAIu3R7rUicgAt9isix6uf6QAiJyONtAFTmFKX\nT/5m01lEJCx0vnsAlXgH07QkiRqaXisiaCSeiBy/HqYDiJysRBJ73M7t6d3pvsJ0FhEJOY3EO4BK\nvIOpxJOooum1InFPn0yKyPHqZjqASDBYWAU3cmOPIQz51HQWEQkplXgHUIl3MH0yKVFH02tF4ppK\nPBE5XirxJGZYWOmTmDR8IhMrTWcRkZA51XJa2pU6QCVegNPpLAHSTOcQOVGaXisSl3Isp1ViOoSI\nRAen05kJ5JnOIRJMFlbCSEaOu57rKwHbdB4RCbp0oNx0iEihEm8/bWohUU/Ta0XikkbjiUhHaRSe\nxKxe9Bp3B3fMd+Bwmc4iIkGnKbUBKvH203p4EhM0vVYk7mhzCxHpqO6mA4iEUj75I+/irtVppNWY\nziIiQaUPrQNU4u2nEk9iiqbXisSNrqYDiEjU0Eg8iXlppA2YytTafPI3m84iIkGjkXgBKvH200mN\nxJy26bWTmazptSKxq8B0ABGJGjrflbiQSGKP27gtrTvdV5jOIiJBoTWgA1Ti7VdsOoBIKFhYCf3o\nN+4u7lqq6bUiMUklnoh0lEo8iRsJJBTeyI09BjN4geksInLS8k0HiBQq8fbrbDqASChlkDFY02tF\nYpJKPBHpKJV4ElcsrPRLuGTYBCbMMZ1FRE6KzncDVOLtp5F4EvM0vVYkJumkRkQ6SiWexB0LK2EU\no8Zex3WVgG06j4icEI3EC1CJBzidTgsoNJ1DJBw0vVYk5uj1S0Ta5XQ6E9CH1hLHTuGUcXdwx3wH\nDpfpLCJy3FTiBajE88sHEk2HEAknTa8ViRm5ltNymA4hIhEvF7BMhxAxKZ/8kXdx1+o00mpMZxGR\n45JsOa0s0yEigUo8P62HJ3FJ02tFYkICkGc6hIhEvE6mA4hEgjTSBkxlam0eeVtMZxGR46IlZFCJ\n10YlnsQtTa8ViQk6qRGR9qjEEwlIJLHH7dye2o1uK01nEZEO05RaVOK1UYkncU/Ta0Wimko8EWmP\nSjyRAySQUHgTN3UfxKAFprOISIfofBeVeG2KTAcQiQSaXisStXRSIyLtUYkncggLK/1SLh06gQlz\nTGcRkXZpJB4q8drkmA4gEik0vVYkKmmHWhFpj0o8kSOwsByjGDX2Wq6tBGzTeUTkqPShNSrx2mSa\nDiASaTS9ViSq6KRGRNqjEk/kGHrTe9zt3D7PgcNlOouIHJFG4qESr41KPJEj0PRakaihEk9E2qMS\nT6QdBRSMmsrUVWmk1ZrOIiKH0fkuKvHaqMQTOQpNrxWJCjqpEZH2qMQT6YB00gdOZerePPK2mM4i\nIgfRSDxU4rVRiSfSDk2vFYloKvFEpD25pgOIRItEEnvezu2p3ei20nQWEfmazndRiddGJZ5IB2h6\nrUjE0kmNiLRHI/FEjkMCCYU3cVO3QQxaYDqLiAAaiQeoxGujEk+kgzS9ViQiZZgOICIRL910AJFo\nY2FlXMqlQ8/n/Dmms4gIOaYDRAKVeH4q8USOk6bXikQUh+kAIhLxEk0HEIlGFpbjbM4eey3XVgK2\n6TwicUznu6jEa6MST+QEaHqtSMTQm3MRaY/e/IichN70Hnc7t89z4HCZziISp3S+i0q8NkmmA4hE\nK02vFYkIenMuIu3R7wmRk1RAwaipTF2VRlqt6SwicUivY6jEa2OZDiAS7TS9VsQofTIpIu3Rmx+R\nIEgnfeBUpu7NI2+r6SwicUavY6jEa6MSTyQINL1WxBid1IhIe/R7QiRIEknseTu3p3Sl65ems4jE\nEb2OoRKvjUo8kSDR9FoRI3RSIyLt0e8JkSBKIKHwu3y360AGLjSdRSROaOYJKvHa6O9BJMg0vVYk\nrHRSIyLtUYknEmQWVsZlXDbkPM6bYzqLSBzQ6xg66W+jkXgiHWT5fL4kj6c5yeNpTna7W5LdbneK\ny+VOdrncKW63J8Xl8qa0tLSmuFze1b7T3EPc97RO+u6jM4eP+TjXdHaRWNXqo9l0BhGJeHrzIxIC\nFpZjNKPHnrdo78wX7cFJvuJkR3mXzXZp2TZKi3c5igv3JGRntjgsvecUOSk+G4/pDJFAJZ6ffqFK\n1HO0trqTPJ6WZLe7Ocnjafm6WHO5PAeUa94Ul8ub4nbbKS4XyS6Xnex2W8lud0KSx2MltbYmJno8\nDofXmxT4Sk7w+VIs2062bDsNSLMgGcgIfAHgJcH3BWdseJfzd7zHea1LGJS1m8LuNgl5ANVv/tfS\n/HwGGvqrEYkHNaYDiEjEU4knEiK5NTVVd77++fl3sST1Oa5e/D/82TOTHsPASgRISWlxFxburiku\n3lFfVla1r7x8a3OXLltay8qqvCUl262iol2JnTrVJGdlNaSnpTVnJia2dkpIsLNNPy/x+9734PXX\noagIli8//OdPPw2/+x3YNmRlwSOPwMCBsHs3fOtbUFsL990Hl13mv/2ll/pvU1oa3ucRA1ymA0QC\nlXh+KvEkNGzbDoxaawmMWnO1jVxLcbk8KS5X69flmtvtS3G5/OWa220nud0JbeVaYmtrYuArKcHn\nS3J4vcmWbaceoVxLBkL6gt9CiutTzlr3DhP2zGa8vZK+eXXk9ASrF9Dr0Nunpja35OXtPT2UmUQE\nr+kAIhLxVOKJhMiNjz++1YIyBz6+wzNDv8MzVFG66+f85stnuKaHy5XadevWLp23bu3SedGi4R16\nzMRET2t+/p6akpLt9aWl2xrLy7e2dOmyxV1WVuUtLd1mFRXtcuTl7U3Ozq5PS09vykxK8uRYlp1r\nWVoqKthuugluvx1uuOHIP+/RAyoroVMneOst+NGP4NNP4Zln4Oab4fLL4aKL/CXea6/B4MEq8E6Q\nNk5EJV4blXhxJsHrbU3yeJoCpVpLktvtSvGXa54DyjVfW7mW7HIRKNdIdrutJI/HkeTxJLSVaw6v\nNynB50tK8PlSEny+VMu2U/AXa6lAeuAr6tSRXf8RYza8zQV1cxibsJZTOjeT1gOsvh19jIkTZ62y\nLAaFMqeI6KRGRNqlEk8kBAYtXrwgt67urEOvL2Nb0RPcVPRvvmu/wJWf38OvXWs5ZRhYSR153NbW\npMSdO4sLd+4sLlyyZHCHsiQkeH15eXtrOnfeWVtWVrWvrKyqqbx8q7u8fKu3tHSb3bnzTkd+/p6k\nnJy6tPT0pszkZHd2QoKvk2WpFziWsWNh48aj/3zUqP2XR4yArVv9l5OSoKkJXC5wOKC1FR56yF/k\nyQnR+S4q8dpoBEOESPR4WgJrrbnaRq4dUqz5yzWXy5victnJLhcpbjdJ/nItIcnjcSS1tiYktrYm\nOlpbkwLlWtuU0LZyLd3y/9vXEPUDVFG6azbjN89iYuM8RqZspmuph6SuYJ3UNNjrrnuqNlgZReSo\ndFIjIu1xmw4gEmuS3O6mi1977ZhjqhKwrat4fshVPM8OOu++h1+vfJLru3tI7hbsPD6fI6G6urBT\ndXVhpxUr+nf4fjk5tXXFxTvqSkq2N5SVVTV36bLFVV6+tbWsrIrOnXdaBQXVSbm5tWkZGfvSk5Pd\n2Q6Ht5NlkRLs/LHgn/+ECy/0X/7Od/xff/ubf7rtX/8K118P6VE5vCMi6HwXlXhtNLf6GCyfz3vA\nRgbuFJerJdntdie7XJ7Ug8s1O8Xl8iW7XL4Ut9tK9o9cs5I8noQkj8eR2NrqaBu1Flhr7etyDf+I\ntbaRa6mGn3JM82HZX3Hq5vc4b9u7nO9axLDMHRR39eEoAoqCfbzx42fnBfsxReQw+jBKRNqzz3QA\nkVhzxQsvLHT4fOM6evtidhb+kx+M+zs/tF/hsiU/47fNX3HqULCSQ5mzPXV1uTl1dbk5q1d3fAWc\nzMyGfUVFu2pKSrY3lJdvbS4v39oSKP58xcU7rMLC3Um5ubUpmZmNGSkprqzExNZOlhWds5M6avZs\nf4k3d67/+5wceOMN/+WaGrj/fnj5ZfjhD/3fT50KI0eayxuFQtrbWJZ1AfAn/CPX/2Hb9v2hPN6J\nUonnF5UlnqO11Z3kdjcnB8q1ZLfblbx/rTVPqsvlTQ6Ua6n+cs1OdrutwMi1timhjsTW1iNtZJAS\nWGstNbDWWmbgS6KIh8TWxQxe/w4Tdr3Pud5lDMjdS14PsLoBQf/071BpaU3NeXl7Twv1cUREJZ6I\ntEslnkgQFe3Ysf7Ur746oQomAdu6nJcHXc7L7KJwzy/41fLHuamrm5Qewc4ZKo2NWRmNjVkZ69cf\ntiT2UaWmNrcUFu6uKS3d1lBaum1fly5bWsrLt3rKyqrskpLtFBbuTuzUqSYlK6shPTW1Jcu/zl90\nzJ5atgx+8AP/mnj5+Yf//Fe/gnvu8a+TN3o0XHmlf628WbPCnzWKtYTqgS3LcgB/Ac4HtgILLct6\n1bbtlaE65olSiecXvH8MgY0M2nYITXa53Cn7NzNoTWlpaU11uVqTA9NBU1wuO8XttpNdLiswaq2t\nXEt0eL2OxNbWJIfPl5zg38igbTODQzcyyAlafola+0hv+oRR62YxseZDzrFWc1pBI5k9wToVONVE\npgsueHuVZdGxRTxE5GQ0mg4gIhFPJZ5IEN0wfXqdBT1P9nGK2J3/GDePe4ybeZVJS3/GbxtX0nco\nWDE3O6mlJS11y5auJVu2dC3p6H2SktyewsLdtZ0776wrL9+6LzDd19O2s2/nzjsT8/L2JrXt7HvA\nBh9hW/d+82Z/Iffkk3DqEd51rVnjXyfvnHNg6VJITQXLgubmcCWMGSEr8YAzgbW2ba8HsCzrWeBS\nQCVeJOq7YsXK9H37GlPcbm9KS0vbDqF2stttJ7vdVtsOoYFRawduZJAc2MjgwB1CU4jijQwkOuym\nYO+HnLNxFhPrP+bs5A30KHaR0h2sM0xnO9B11z1VbzqDSJzQ/zURaY9KPJEgOXvu3E8ymppGtX/L\n43MJrw28hNeoJr/mf/nfBf/k+2UtpHV8qFsM8niSk7ZtKyvctq2scPHiIR26j8PR6s3L21tbXLyj\nvrR0W0NgxJ+rbWffzp13JuTn70nJzq5PTU9vykhK8uQENvg44gZA11wDH34I1dVQXg5OJ3g8/p/d\nfDPcey/s2QO33uq/LjERFi3af/977oFf/3r/Y112mX9q7b33nujfStwKZYlXBmw54PutwGEb1kQC\ny7Zt0xnMs6zZwDmmY4gcyQa6b/uAb2ydxcSmTzkrvYqyci+JUbEp+Z49ecvy8moGmM4hEgdmgX2B\n6RAiErmcTucMYLLpHCLRLrW5ue4nv/+9O8G2C8NxvLe44Iu7+X39cvoPASstHMeMR5blszt1qqkv\nKtpVW1q6rbG8fGtTeflWV9vOvsXFOxwFBdVJOTl1qenpTRkpKa62nX2NrmcYZz4Fe0QoHtiyrCuB\nC2zb/kHg++uBs2zbvj0UxzsZGonn12A6gIiXBN9y+m98l/N3vMd57sUMzt5NYXebhFIgKkq7A6Wl\nNTV36lTT8dVxReRk1JkOICIRTyPxRILgmv/8Z2mCbY8N1/Eu5O0zLuRtasitc1Kx8G/8qKSZ9N7h\nOn68sO0Ea+/e/Jy9e/NzVq3q0+H7ZWXVNwR29q0vLd3W1KXLFndgnT+Ki3dYBQXVibm5tWmZmY3p\nKSmubIfDm2tZqIw9MaF8HasCuhzwfXnguoijEs9PJd5x2ALcAOwELOBHwH8d5bYLgZHAs8CVwGrg\nO4AHeCzws1bgAuBV4mcOcgsproUMXz+LidWzGW+voF+nOnJ6gtWTIKytEQkuuujNLy2Ljo15F5GT\npem0ItIelXgiJ6nrxo0ru2zZMtrEsTtRm/MQ/z32If6bdzlv+U/4Q+1SBg4BK17eQkWkhobsrIaG\n7Kw1azq+BHl6+r7moqJdNcXFOxoCI/6aD9zZt6hoV1KnTjXJmZmNGampLVmJia25lqVNJgnt69hC\noLdlWT3wl3ffxl9dRByVeH4q8Y5DIjANGIL/L24o/i1c+h5yOy/wU2DCAdc9hn/P5u74i78XgUeA\n64jdAq+O7PqPGLNhFhNr5zDWsYbeRc2k9QCr4x/xRKHrrntK/69Ewkcj8USkPSrxRE6C5fP5vvOf\n/2BBguks5/Ne/yUMppacuvv4nzmPcEvnJjJOM51LOqapKSNt48YeaRs3dnwz4uRkl7uwcHdNYMRf\n286+7rKyKl9JyXarqGiXIy9vb0pgg4+sxMTWHMuys8O5wUcYhOx1zLbtVsuybgdmAQ7gX7ZtrwjV\n8U6GSjw/lQ3HoSTwBZAF9MFfVR9a4j0MXIG/0m6TBDQFvpKAWuA14O0Q5g2nKkp3zWb85llMbJzH\nyJTNdC31kNQVrIGms4Xb2LFz8kxnEIkjKvFEpD0q8UROwnnvvjs3xe0O2zTajsilLucBfjL2AX7C\nbM5Z+RP+sPczhg4GK8N0NgkutzsluaqqvHNVVXnnjt7H4Wj1FhRU1xQX76grK6vaV1ZW1VxevtVd\nXr7VV1q6zQ7s7JucnV2flp7elJmU5Mm2LLuTZZkvqo8ipOe7tm2/CbwZymMEg0o8P5V4J2gjsJjD\nt22pAl4GZnNwiXcb/qm4Lvyj8n4F/JwI+DjrOPmw7DX03vIe5217hwmuRQzL2EFxVx+OIqDIdD7T\n0tP3NXXqVKNPA0XCR9NpRaQ9KvFETlBmQ8PukfPmRfSH8uP5sO8ihlNPVsNv+PlH/8ftBfvIjOmZ\nP3JsXm+iY+fO4oKdO4sLli4d1KH7JCR4fZ061dQE1vlrCEz19QSm+xLY2Te5bYOP5GR3286+4eiW\n9obhGBFPJZ6fSrwT0Ih/pN1DQPYhP/sx8DsOL+e6Ah8GLq/Fv29zH+B6wI2/1Ov4agLh4SGxdQmD\nNrzDhJ3vc653GQNy9pDfA6yu+J+SHOKb33xjldbDEwkrjcQTkfY0mg4gEq2unz79KwvONp2jI7Jp\nyLqfn425n5/xEaNX3cUDuxdw5iCwskxnk8jn8zkS9uwp6LRnT0GnFSv6d/h+OTm19Z0776wtKdl+\n0M6+gXX+HAUF1Yk5OXWpGRn7MlJSXFkOh7eTZZF6nPFU4qESr41KvOPkwV/gXQtcfoSfL8K/EiRA\nNf4xqYnAZQfc5h7gPuDPwA/wr5P3c+DpkCTumH2kN81j5PpZTNz7IedYqzg9v5HMnmD1BrQLVAdd\ne+3T+j8lEl4q8USkPbtMB4h0dXV1vPzyyzQ2NmJZFkOHDmXEiBE8//zzVFdXA9DS0kJqaiq33HLL\nQfetrq7m+eef//r7mpoaxo8fz8iRI3n33XdZs2YNxcXFXH65/8x56dKlNDU1MXLkyPA9QTkhp61a\ntaRo9+6oKPAONYa5p3/KiNMbyGz8HT/96M/cmd9A9qGrIImctLq63Oy6utzsr77q+GSsjIzGfUVF\nu2pLS7c1BNb5c/lH/VVRXLydgoLqpE6dalIzMxvTUlJc2W538u4MTRRXiRdQbTpANLGB7+MfQTfl\nKLfZcMDlm4CLObjAqwRK8bdiTfhH7CUELodLNfk1H3LOhllMbPiYsxPX07PYRUp3sDr+kYMckdbD\nEwk7TacVkfbsMB0g0iUkJDBhwgRKS0txuVw89thj9OzZk8mTJ399m1mzZpGSknLYfQsKCr4u9nw+\nH9OmTaNPnz60tLSwfft2br31VmbOnMnOnTvJy8tjyZIlXHfddWF7bnJiErxez5XPPx/1I9iyaMy8\nj1+MuY9fMI8Rq6cybec8Rg4C69AJVSJhs29fZsaGDZkZGzb07Ohdamw7lImig0o8v22mA0STj4En\ngTOAtpn1vwE2By7f3M79bfwj8J4LfP8j/CP6WvHvVBsKG+i+7QO+sXUWE5sWcGZaFWXlrSSVAZ1C\ndMi4lZHRuC83t/Z00zlE4oxG4olIe7abDhDpsrKyyMry9zUpKSkUFhbS0NBAUZF/uWPbtlmxYgU3\n3njjMR9n/fr15OXlkZubi8vlwuv1Yts2Ho+HhIQEPvnkE84880wcDkfIn5OcnEteffXjRK/3HNM5\ngmkk80/7hLNP20d60wPcNfdB/ju3jlwNYpBosNN0gEigEs9PJd5xGI2/iOuoxw/53gLePeD7PsDn\nJ5mpjZcE30r6bnqHCdvf5Xz3YgZn76awm01CKf7BfxJiF1/8+mqthycSdrtNBxCRiKcS7zjU1NSw\nfft2ysrKvr5u06ZNZGRkkJ+ff8z7Ll++nP79/Z1ISkoKvXv35tFHH6Vnz56kpqaydetWxo0bF9L8\ncvI67dmzdcDSpYfu3xczMmhKr+De0RXcy0KGrZnKtO1zGT3QJiHHdDaRo1CJB1i2xiOCZSXi31fB\nMh1FOs5Fsnshw9e9w4TqD/iGvYJ+nWrJ7akt1c2aOfOSyksueU1npiLh4wNSwG41HUREIpvT6dyL\nZiG0y+Vy8fjjctqndQAAIABJREFUjzNmzBj69t2/fNjrr79OXl4eo0aNOup9W1tbmTZtGrfddhuZ\nmZmH/XzmzJkMHz6c7du3s27dOjp37qxCL0L9+I9/XJBTX3+m6Rzh1ERa84P892fTmJpTQ94ZpvOI\nHCLDtsO6AldEOnTz0Phk261oFENEqyer4Q0uWnonf6ocxOK5GTSuTqXFGsPcPr/il2M+ZvTYWjqd\noQLPPK2HJxJ221XgiUgHaTReO7xeLzNmzOCMM844qMDzer18+eWX9OvX75j3X7t2LSUlJUcs8LZv\n9//1FxQUsHLlSq666ipqamrYs2dPcJ+EnLTBn30WdwUeQDrNaffwm9F7yT/jcwavO4fZcyx8NaZz\niQCNKvD8NJ12v21AkekQAtso2T2b8ZveYULjJ4xK2US3Ug9JXcEaaDqbHFtGRuO+nJy6PqZziMSZ\nLaYDiEjU2A5oZ8qjsG2bmTNnUlBQcNhou/Xr11NQUEBOzrFnGn7xxRecccaRBzB98MEHTJo0Ca/X\ni8/nA8CyLDweT3CegARFktvd9M033oj7ZXgGs6TXbL7Rq5nUloe545Pf8dOMveTr/ZiYoqm0AVFR\n4lmW5QAWAVW2bV8cosNsY/8+DRImqzl183uct+1dzm9ZxLCM7ZR09eHoDBSazibHb9Kk11ZZFkNN\n5xCJMyrxRKSjNBLvGDZv3syyZcsoKirikUf8262de+65nHrqqQetc9emvr6eV1999etdZt1uN+vX\nr2fSpEmHPfaXX35JaWkp2dn+zUCLi4v561//SufOnSkuLg7xM5PjceXzzy90+Hya4xyQRkvq3fxh\n1N38gS/ov2EKf9z8PueeYZOg2TcSTirxAqJiTTzLsqYAw4DskJV4lvV34AcheWzBQ2LrEgZteIcJ\nO9/nXO8yBuTsIb8HWFo4NYa89trFlRdf/IZOekTC649gTzUdQkQin9Pp/ANwl+kcIpGqePv2dT96\n7LGuFiSZzhLJWkhx/ZVbP/stP0uvpmAgWFpbXkLtFdvmW6ZDRIKIH4lnWVY58E3g18CUEB5KIxmC\npIm05nmMXP8OE/bMZrz1JX3yG8nsCVZvoLfpfBI6o0fPPfZ2bSISCptNBxCRqKGReCLHcP306Q0q\n8NqXiitlCg+OmsKDrKDvxrt4YNMsJvazSSgwnS34vPjHE5UBrx/ys83AjUBt4Hb3AxcBHwO3AMnA\nM/jfAtcCVwFvo60JToj6moCIL/GAh4C7gawQH2djiB8/Ju0hr7aScRtmMbF+LqMT19OzuIXU7mAd\ne9VfiTmZmQ2NOTl1p5vOIRKHdFIjIh2lEk/kKEZ/9NHH6c3NZ5vOEW36sbL7W1zU3UWy+zH+37zf\n8POUnXQeHDuj8/4E9AHqj/Cz+/AXc7cAK/EXeBuBacCbgcuPBr6/D/g5KvBO2GrTASJFRJd4lmVd\nDOyybfszy7LOCfHhNob48aPeRrpt/4BvbHmHCU2fclbaVsrLWkkqBwabzibmTZr02mqthydihEo8\nEemoKtMBRCJRanNz3fgPPjjVdI5oloI7+U4eHnknD7OaUzdPZdqGt7iwrw9HFK91vhV4A7gH+OMR\nfm6xv9yrA9r2Q0kCmgJfScA6/Kdr54Qwa8xTiRcQ0SUecDZwiWVZFwGpQLZlWU/Ztn1dCI61IQSP\nGZW8JPhW0nfTu5y//V3Ody9mcNYuirrbJJQAJabzSWS69tqnG0xnEIlTKvFEpKPWmQ4gEom+8/TT\nSxNse6zpHLHiNL7q+jqTurpJ8vyT78//Fb9I2k7JYLCibBjaj4HfA0d7m/O/wATgYWAf8F7g+p8B\nNwBpwJP4lyK9L5RB44FKvICILvFs2/4Z/v8BBEbi3RWiAg/8n0x6iLM1EFwkuxcyfN07TKj+gG/Y\nK+jXqZbcHmD1AHqYzifRY/TouVH8KZtI1HKj3bpEpOO24R8akm46iEik6LZx48ryrVtHm84Ri5Lx\nJN3CoyNu4VHW0mvrXTyw7nUuPs1LYhRsyfw6UAQMBT48ym2eAW4CpgLzgOuB5cAgYH7gNnPwj4Ox\ngavx1w3TgM6hiR2bmvAPixSiZHdaOKjEC83utP6DrAV6hezxDasnq+EjxmyYxcSaOYx1rKF3YRPp\nPcGKq+JSgi8rq76hri4nzbIi+4MBkRi0HuyYfd0SkeBzOp1LgQGmc4hEAsvn8/30/vtXp7jdfUxn\niRceElsf56bPnFQkVFE2NHJH5/0M/yi6RKAF/7TZy4GnDrhNP/wbVXQJfN8Tf3lXFPjeBiYCzwJ3\nAL/Bv4rXO/j37ZQOWmLbWsKrTdS84bZt+0OOXoEHy0ZipMTbQefdsxm/aRYTGz9hVMomupW4Se4G\nlk7aJOguueTV1ZbFMNM5ROLQRtMBRCTqrEUlnggAE95556MUt3uc6RzxJInWxB/yj7N+yD9YT4+q\nu/n92le47FQviRG2bNNvA1/gryEe4OACD6Ar8D7+0Xhf4i/7DpycNB3/Zhd5+AeTJQS+mkKUOWZp\nKu0BoqbEC5OoXBfvK3pveZ9zq95hQssihmVsp6RLYIiypjdKWFx77dONpjOIxKnlpgOISHBYlpWK\nf95VCv5z9Bds264IwaHWhOAxRaJOVn39rrPmzx9kOkc868mGsheYXNaKwzudGxb+L/9rb6HLULAc\nprMd3S+BYcAl+KfF/hB4EP8mF48H/gR/Ufc4/lF3AFPwF3rJwH/CljZGqMQ7gEq8g601HeBYWnF4\nlzBo/bucv+s9zmtdysCcveT1sEnowv4xvCJhd/bZHxe1fysRCQGVeCKxwwV8w7btRsuykoC5lmW9\nZdv2/PbueJy+CvLjiUSl66dPX2P5N1IUwxLxOr7Hv4d/j3+zmS7b7+b3X73IFae0klRmOpvfOezf\nWfbeA67vC3x8lPukA7MP+H4M8EWwg8WLVaYDRBKVeAeLmDdDTaQ1z2Pk+neYsGc2461VnJ7fQFZP\nsHoDvU3nE2mTlVXfkJXVcJrpHCJxKmJet0Tk5Nj+harbRrYnBb5CsXj1lyF4TJGocvqXXy4urK5W\ngReBurKl5FmuKfGS4PsP31n0S+71bqT7ULDUXcQvjcQ7gP4jHGyZiYPuIa+2knEbZjGxfi6jE9fT\ns7iF1O5g9TORR+R4XHbZK6ssi+Gmc4jEKZV4IjHEsiwH8BlwCvAX27Y/DcFhVOJJXHO0trqveOGF\nXNM55Ngc+BKu56lh1/MUWynb+f9x/6rnuLpXK0nlprNJ2GkE+QGiZnfasLGsGiBkv9Q302X7+5y7\n5R0mNH3KWalb6FKuX0QSzd5+e2LlxInvaEFgkfDbDHY30yFEJPgsy8oFXgbusG076GW90+ncBkTY\nIvIi4fGtl16qHLBsmc5do5APy36Oqz//H+7zrKfnULCSTGeSkKuybdSXHEAj8Q73Bf4J6yfFh2Wv\npO/Gdzl/+3uc5/6cIVk76dzNJqEEnTRJDBk5cp42UBExQ6PwRGKUbdu1lmXNBi4gNP/Xv0TnoxKH\n8vbs2XLGsmVnms4hJyYB27qGZ4dew7Nso2T3z/nNyv/wne4ekvWhZuzSVNpDqMQ73DKOs8Rzkez+\njKHrZzFx9wd8w7ec/p1qye0JVg+gR2hiipiXnV1Xr/XwRIxRiScSQyzLKgQ8gQIvDTgf+F2IDrcS\n+EaIHlskYt34xBM7LG0IGBNK2V74ON8d9y++Z7/IFYvv4dcta+g9FKxk09kkqFTiHUIl3uGOuS5e\nA5mNHzFm/Swm1sxhrOMrTi1sIr0nWKcDp4cpo0hEuOyyV1ZrPTwRY1TiicSWEuCJwLp4CcAM27Zf\nD9GxjKwDLWLSkEWLPs2urz/LdA4JrgRsazIvDJ7MC+ykqPoefr3iSa7v5ialu+lsEhQq8Q6hEu9w\nX+/7vIPOu2czftMsJjZ+wqiUTXQrcZPcDawBJgOKRIprr326yXQGkTimEk8khti2vQwYHKbDLQzT\ncUQiQpLbve+iN9/UuloxrjO7Cv7BD8f9gx/yCpcu+Rm/bVrF6UPBSjGdTU6YSrxDqMQ7xFIGLJ3E\na59uo7Sbl8RiQOt9iRzFyJHzikxnEIlTXrTDpIicuOVAM5BmOohIOEyeMWOhw+c7x3QOCZ/LmDno\nMmaym4K9v+Te+f/mu11cpPY0nUuO20rTASKNdqc9AstiJdDHdA6RSJaTU1tXU9Mp07JwmM4iEoe+\nAlvrUYrICXM6nR8Do0znEAm14m3b1v7ob3/rZoF2Mo1zr/PNZT/ld40r6TsErFTTeaRdO2xbmzAd\nKsF0gAi1yHQAkUh3+eUvfaUCT8SYpaYDiEjU05RaiQs3PPnkPhV4AnAxbwxYQf9R1RS03M7Dc1Jp\nXms6kxzTp6YDRCKVeEemEk+kHddc84zWwxMx5xPTAUQk6i0wHUAk1MZUVs5Na24eaDqHRJZ89uY+\nzJ1jm0k/5S0u+GIAS+eC3Ww6lxxmvukAkUgl3pGpxBNpx4gR8zubziASxz42HUBEop5KPIlpaU1N\nteNnzz7ddA6JbBcw64ylDBq9lzz3j3lwThpNa0xnkq9pJN4RaE28I7As0oAG0FRBkSPJza2p27s3\nL8uy9EGAiAFNQA7YraaDiEh0czqde4FOpnOIhML3//73OeVVVWNN55Do8x7nrvgJf6hZwqAhYKWb\nzhOnvECubdNoOkik0RvwI7BtmoEVpnOIRKrLL39ptQo8EWMWqMATkSDRungSk7qvX7+irKpqtOkc\nEp3O4/1+ixkyupbc1p/w+4/S2bfadKY4tEIF3pHpTfjRaUqtyFFcc80zWjNCxBxNpRWRYFGJJzHH\n8vm81zzzjMPSe105STnUZ/+en47ZR+Zpszln5TAWfgS2iqXw0FTao9AvtqP7yHQAkUil9fBEjFKJ\nJyLBonXxJOZMnDXr42SPR2vhSVCdQ2XfhZw5pp5sfsZvPsqk4UvTmWLcPNMBIpVKvKN733QAkUiU\nl7enNiNj36mmc4jEKR86qRGR4FGJJzElq75+55mffjrIdA6JXVk0Zv6Ge8Y0kN3nI0Z/eRbzPwK7\nwXSuGDQnFA9qWVYXy7JmW5a10rKsFZZl/VcojhNKKvGOwrbZAmjuu8ghLr/8pa+0Hp6IMSvArjUd\nQkRiQ0VFxQ50visx5Ibp09dZkG06h8SH0XzcZz4jxzSQlfAL7p2bTZ3W1Q+OKttmXYgeuxWYatt2\nX2AEcJtlWX1DdKyQ0BvxY3vPdACRSKP18ESM0lRaEQk2ne9KTOi7YsXnBdXVo0znkPiTyb6Me6kY\nXUduv3mMWH02c+eAXWc6VxQLySg8ANu2t9u2/XngcgPwJVAWquOFgkq8Y9NJjcghzjrr02LTGUTi\nmEo8EQk2ne9K1HO0trq/9dJLnUznEBnBp6fNZczYRjKT7uUXH+dQu9x0pihUGY6DWJbVHRhMlG2i\noRLv2GYDXtMhRCJFfn51bXp6k9bDEzFHJZ6IBJvOdyXqXTpz5ieJXm8P0zlE2mTQlP4L7ju7lk79\nFzJszVgq51j4tCRKx4S8xLMsKxN4Efixbdv1oT5eMKnEOwbbpg5YaDqHSKQIrIdnmc4hEqe2gr3B\ndAgRiS0VFRV1wCLTOUROVH519eb+X3wxwnQOkaMZxme9Kzln7D4yUn/Dzz7uxN5lpjNFsF22zapQ\nHsCyrCT8Bd7Ttm2/FMpjhYJKvPZpioFIgNbDEzHqLdMBRCRmvWs6gMiJuuGJJ3ZZkGo6h0h70mhJ\n/Rn3n72X/AGLGbRuPB9UWvhqTOeKMCFbDw/AsiwL+CfwpW3bfwzlsUJFJV77VOKJBJx55oIS0xlE\n4tibpgOISMzS+a5EpWELF87PbmgYZjqHyPEaxNJeH3DuuCbS03/PTz7Op3qp6UwRYnaIH/9s4Hrg\nG5ZlLQl8XRTiYwaVZdu26QwRzbJIBvYCGaaziJiUn19ds3t3Ya6m04oY4QbywW40HUREYo/T6dT5\nrkSdZJer8af339+QYNv6kFliwhf03zCVaZvf47z+Ngn5pvMY0sW22Wo6RCTTSLx22DZuQjykUyQa\nXHHFi1oPT8ScShV4IhIqFRUVbuAj0zlEjsfkGTM+U4EnseQMlvd4h4njmknLepAfzytk12KIq1FX\nn6nAa59KvI7RFAOJe9dc80yL6QwicewN0wFEJObpfFeiRsm2bWt6rVt3tukcIqGQgjv5x/xp5C46\nD15J380X8mZlAt7dpnOFwSumA0QDlXgdo5MaiXtnnrmg1HQGkTim9fBEJNS0uYVEB9u2r58+vdmC\nRNNRREKtD6u6vck3xzWTlvswt8/rzI7PY3h0nkq8DlCJ1zFfADtNhxAxpaBg9960tOZTTOcQiVNr\nwF5jOoSIxLwvQNOYJPKNraz8OK2lZYDpHCLhlIwn6Xb+MnIHJUNWcfqWSbxamYB3l+lcQbTWtllu\nOkQ0UInXAbaNDbxvOoeIKVde+YLWwxMxR1NpRSTkKioqbOBl0zlEjiWtqanmnA8/7GM6h4hJp/FV\n11e5dFwLqXmPcPP8UqoWge0zneskzTQdIFqoxOs4/aOSuHXNNc+4TWcQiWOaSisi4fKi6QAix3Lt\nU08ttyBed+0UOUgSrYk389iIKsqHreWUbd/ipUoHrTtM5zpBmkrbQSrxOu51QDsDSlwaNmyRdv4S\nMaMRqDQdQkTixkdoCRmJUD3Wr19eum3baNM5RCJRL9aXv8QV41pILfwH319QzpaFUTQ6bxfwiekQ\n0UIlXgfZNk3Aa6ZziIRbYeGuPVoPT8SY98DWSFgRCYuKigofGg0hEcjy+bzffuaZJAst7yJyLIl4\nHd/nX2duoevw9fTcMZnnKx20bjOdqx2v2TbRUjgapxLv+DxrOoBIuE2e/PwarYcnYoyWchCRcNOU\nWok4F7z99txkj+c00zlEokkPNpbO4KpxLaR2fpwbF3Zl0wKwvaZzHYE+PDoOKvGOz9tArekQIuH0\n7W8/q1FAIma0AC+ZDiEicWc2sNd0CJE22XV1O4YvWDDEdA6RaJWI13Ej04dvovuZm+i26xr+U5mI\nJ1J2I28E3jMdIpqoxDsOto0b7dolcWbYsEWlpjOIxKk3wK43HUJE4ktFRUUrGgUsEeSG6dM3WJBl\nOodILOjKlpL/cO24FlJLn+S6RT1YPx/sVoORZtk2LQaPH3VU4h0/TamVuFFUtLM6NbWll+kcInHq\nadMBRCRuaUqtRIS+y5d/lr9nz0jTOURijQNfwnU8PWw9vUZspXzv9UyvTMSzxUAUfWh0nFTiHb/3\n8e+eIhLzrrpqxlqthydiRC3wpukQIhK33gXqTIeQ+OZobXV96+WX803nEIl1ZWwrms6N41yklD/D\ntz/vxdp5YHvCcOhW4PUwHCemqMQ7TraNF3jBdA6RcNB6eCLGvAi2y3QIEYlPFRUVbvTGSgy77OWX\n5yV6vd1N5xCJFwnY1rd5bshaeo/cTkntd/lXZRLuTSE85BzbpiaEjx+TVOKdGE2plbgwZMjnZaYz\niMQpTaUVEdNmmA4g8Su/unpTvxUrRpjOIRKvitlZ+C++P66F1K4vcMXiU1n9CdjBHuDxXJAfLy6o\nxDsxc4FI2c1FJCQ6d96xOy1N6+GJGFAFVJoOISJx701gh+kQEp9ufPzx3Rakms4hEu8SsK0reGnw\nak4ftZPODT/kb5XJuDYE4aGb0OCoE6IS7wTYNjb6dFJi3NVXP7fWdAaROPUs2D7TIUQkvgV2qZ1u\nOofEn+ELFszPamwcZjqHiBysiN35f+P/jXOR2uMVLl3ah5Ufn8TyLy/aNvVBDRgnVOKdOLXGEtOu\nvvq5cCxmKiKH01RaEYkU/zIdQOJLssvVeMFbb3UznUNEju1SXh24kn5n76Jo3y38tTKFlnXH+RB6\nfTlBlm3bpjNELctiLaDphhKTmptT16WmuvTvWyS8vgS7r+kQIiJtnE7nR8Bo0zkkPlw3fXplr/Xr\nx5nOISLH7w0uWvb/cX/DcvoPBetY0+HXAb0DMxzlOGkk3sn5j+kAIqFQUrJttwo8ESP0uiIikeaf\npgNIfCitqlrTc/36s03nEJET803eHPAFA86upqDlDv48J42mNUe56eMq8E6cSryT8zeg1XQIkWC7\n6qoZWg9PJPxagX+bDiEicojngQbTISTG2bZ93ZNPtliQaDqKiJycfPbm/pn/GttERu+3mbh8AEs/\nBrs58GMf8LjBeFFPJd5JsG22Aq+YziESbFdf/ZzKaZHwmwl2lekQIiIHqqio2IfWgpYQO+fDDz9O\na2k5w3QOEQmuibzTfymDzq6hk3sK0+aUUvVUoEeRE6QS7+Q9bDqASLANHry4zHQGkTj0F9MBRESO\nQlNqJWTS9+3bO7ayUuvBisSwXOpypnHX2CrKZ5rOEu1U4p0k22YOsMx0DpFgKS2t2pWa6uppOodI\nnFkB9mzTIUREjqSiouJTYIXpHBKbrn3qqZUW5JnOISIhtw141XSIaKcSLzg0Gk9ixtVXP3e824OL\nyMn7q+kAIiLt0Gg8Cbpea9d+UbJ9uzazEIkP/8C2tWzTSVKJFxxPA3tNhxAJhquumuExnUEkztQD\n002HEBFpx5NAi+kQEjssn8979bPPplhgmc4iIiHnBf5uOkQsUIkXBLZNM/p0UmLE4MGLu5jOIBJn\npoPdaDqEiMixVFRUVOMv8kSC4sI335yb1Np6qukcIhIWr2Pb2tAiCFTiBc9f8W+XLCH3PaAI6H/A\ndUuAEcAgYBiw4Cj3vQDIBS4+5PprgQHAzw+47j7ibfPhsrKtO1NS3D1M5xCJM9rQQkSixTTANh1C\nol92Xd32YYsWDTGdQ0TC5v9MB4gVKvGCxLbZCLxmOkd8uAl4+5Dr7gYq8Jd59wa+P5KfcPiHyMuA\ntMCfC4E6YDvwKXBZUBJHC62HJxJ274O9ynQIEZGOqKioWA28YTqHRL8bnnhiowVZpnOISFgsxrbf\nMx0iVqjECy5tcBEWYzl8AysL/7JS4C/hSo9y33M5/HwhCWjGP5DSAziAXwLOYISNKldf/ZzXdAaR\nOKNReCISbaaZDiDRrf8XXyzK37t3pOkcIhI2fzAdIJaoxAsi2+Z9YKXpHPHpIfyj7LoAdwG/PY77\n9gEKgSHAJGAt/kIv/kb4Dxy4VOvhiYTPFuBV0yFERI5HRUXFh8BnpnNIdEr0eFoue/nlQtM5RCRs\nNgAzTIeIJSrxgk9zvY14BHgQ/3viB4HvH+f9H8I/FXcq8AvgV8CvgauIl010ysu37EhJcXc3nUMk\njvwf2Br9KiLRSKPx5IRc9vLLnzp8vm6mc4hI2PwRW+e7waQSL/im45/PKWH1BHB54PJkjr6xRXtm\nAkOBRmAd/g8NXgCaTjZgxPv2t5/Vengi4VOL/9MHEZFo9Dyw2XQIiS4Fu3dv6rty5QjTOUQkbKqB\nf5kOEWtU4gWZbbMP/UM1oBSoDFz+AOh9Ao/hwT8i7278a+RZgeu9gPtkA0a8q66aod2VRcLn/8Bu\nMB1CROREVFRUtAJ/Mp1DossNTzxRbUGK6RwiEjb/h23H/miYMFOJFxoPAC2mQ8Sua4CRwGqgHPgn\n/imvU4GBwM+BvwVuuwj4wQH3HYN/pN77gfvOOuBnfwFuBNKBAfhH352Bf2RebmieSgQZMGBZV9MZ\nROLEPvyfGIiIRLO/o9kn0kFnzp8/L6uxcajpHCISNk1oqbGQsGzbNp0hJlkWfwT+23QOkY7o0mXz\n9s2bu5WYziESJx4CW68PIhL1nE7nH/DvKCZyVMkuV8NP779/X4JtF5vOIiJh8zC2fafpELFII/FC\n5378oy1EIt63v/3setMZROKEGy0ILyKx40E0+0Ta8e1nn/1cBZ5IXGlF57shoxIvRGybXcDDpnOI\ndMRVV83QkFyR8Pg32FtNhxARCYaKioptaJMeOYayrVtXd9+wYbTpHCISVs9j25tMh4hVKvFC6w9A\nvekQIu3RengiYeEGfmM6hIhIkP0WzT6RI7Ft+7onn3Rb4DAdRUTCxgf82nSIWKYSL4Rsm71o8XKJ\ncN26bdyenOxRiScSev8Ce7PpECIiwVRRUbEb7VQrRzB+9uy5qS7XGaZziEhYPYVtrzAdIpapxAu9\nPwI1pkOIHM011zyj9fDCZMsWGD8e+vaFfv3gT4G3PEuWwIgRMGgQDBsGCxYc+f4//Sn07+//eu65\n/ddfey0MGAA///n+6+67D155JXTPRY6bRuGJSCz7A1BrOoREjvR9+/aMmTOnv+kcIhJWbuCXpkPE\nOpV4IWbb1KFFHSWCTZ78vNbDC5PERJg2DVauhPnz4S9/8V+++26oqPCXeffe6//+UG+8AZ9/7r/N\np5/CAw9AfT0sWwZpaf4/Fy6EujrYvt1/m8suC/9zlKP6F9hbTIcQEQmFioqKWnS+Kwe47sknV1rQ\nyXQOEQmrR7UWXuipxAuPPwHVpkOIHEn//ss1lTZMSkpgyBD/5aws6NMHqqrAsvyFHPhLuNLSw++7\nciWMHesvAjMy/CPv3n4bkpKguRl8PvB4wOGAX/4SnM7wPS9pVwsahScise8hdL4rQK+1a5cV79ih\nzSxE4ksjcJ/pEPFAJV4Y2DaNwO9N5xA5VPfuG7ZpPTwzNm6ExYvhrLPgoYfgJz+BLl3grrvgt789\n/PYDB/pLu6YmqK6G2bP903P79IHCQn85OGkSrF3rL/TaykKJCA9qFJ6IxLqKiopG4H7TOcSsBK+3\n9epnn02zwDKdRUTC6kFse7fpEPEg0XSAOPIXYCrQ2XQQkTbXXPPMBuAI474klBob4Yor/OVddjb8\nz//Agw/6r5sxA77/fXjvvYPvM2GCf7rsqFH+0m7kSP+oO/A/TptJk+Cxx+DXv4alS+H88+GHPwzf\nc5PD7MK/c6OISDz4KzAFnVvErYvefPPjpNbWcaZziEhYVQMPmA4RLzQSL0xsmyb0Rk4ijNbDCz+P\nx1/WXXstXH65/7onnth/efLko29scc89/jXx3n0XbBtOPfXgn8+cCUOH+kvCdev8heALL/hH74kx\nvwS7wXROdLN4AAAgAElEQVQIEZFwqKioaEbLB8StnNra7UM++2yo6RwiEna/xbbrTYeIFyrxwutR\noMp0CJE2/fsv72Y6Qzyxbf8ouz59YMqU/deXlkJlpf/yBx9A796H39frhT17/JeXLfN/TZiw/+ce\nj39E3t13+9fIs6z993O7Q/N8pF0rgH+YDiEiEmb/f3t3Hm9VXe9//LUO4Gyaeh1xSEQKM0HLDOch\nM+dKr8pgk9X1ltfu1bLy1vL8rqU+sqzM363sVyFg5pBm5qw5oeIA4cQgKiIgSIrMcA7nfH5/rEOC\niUx77+/ae7+ej8d+iICcV4+SNu+91nddCUxOHaHaO33o0Jcz2CR1h6SaeoXirkPViCNeDUWwBPC4\neZXCrru+MLVHj6U7pu5oJiNHwrBhxVDXr1/xuvVWuPJKOOec4ty773wHfvWr4uc/8QSccUbx7fZ2\nOPBA6NsXvvxlGD68eMjFMldcAZ/9LGy0UfHQi4ULYc89iyvzNt+89v9ZBcC5EB2pIySplvI8bwPO\nTd2h2tpz7Ngntpg9e7/UHZJq7gIilqSOaCZZhHfT1VKW0QI8CnwkdYua2/nnXzjywgu/u3/qDqlB\n3QFxVOoISUqltbX1buDw1B2qvu7t7Yu/ddFFM7t1dnqHh9RcngM+RPihdS15JV6NRdAJnAl0pm5R\nczvppOtd8KXq6MCrUCTpP4ClqSNUfZ/64x9HOeBJTeksB7zac8RLIIIngV+m7lBz22OPZ3dJ3SA1\nqN9APJM6QpJSyvP8OeDy1B2qrn957bXJHxg3zttopeZzDRH3po5oRo546ZwPzEodoebUq9ekaT16\nLO2ZukNqQPOA76aOkKSSuACYmTpC1XP60KGvZ7B+6g5JNTUPOCd1RLNyxEskgtnAN1N3qDmddtrv\nJ6dukBrUJRD+gVWSgDzP5wLfTt2h6vjoI488vMmCBfuk7pBUc61ETE8d0awc8dIaCoxMHaHm43l4\nUlW8APw4dYQklczvgFGpI1RZ6y9ePPfIO+/slbpDUs09C/w0dUQzc8RLKIIA/p3iEHSpZvr2fW6X\n1A1SA/oKxKLUEZJUJnmeB3AW4AeIDeSUa64Z0xKxTeoOSTX3VSJ8aFFCjniJRfAU8PPUHWoevXtP\nnOp5eFLF/Q7intQRklRGeZ4/Dvw2dYcqo+crr0zYZfLkA1J3SKq5q4m4P3VEs3PEK4fvAa+mjlBz\n8Dw8qeJew8N9JWlVvg28mTpC6ygiBg0f3p5Bt9QpkmpqLnBu6gg54pVChP9CqHY+85kbUidIjeZs\niDdSR0hSmeV5/ho+1K3uHXbPPQ9tsGTJB1N3SKq5nAgvPCqBLMLjKcoiy7gXODR1hxpbW1uPaT16\nLN0hdYfUIG6FOCZ1hCTVi9bW1nuAw1J3aM1tPH/+38+59NJuGbw3dYukmnoK2JsIz/IvAa/EK5ev\nAu2pI9S4dt99wisOeFLFzAfOTB0hSXXmS8DC1BFac4OHDRvvgCc1nXbgcw545eGIVyIRjAN+lLpD\njeu0037/cuoGqYGcDzEldYQk1ZM8z18E/jt1h9ZM74kTx24zc+b+qTsk1dyFRIxJHaG3OOKVTw78\nLXWEGtNnPnNDlrpBahCj8MnikrS2fgo8mjpCq6elo2Ppyddeu1EGvo8siVcozmDqC+xB8S8UwClA\nv67XLl1/fbsJy/2cfsB7gJ90/dh5wIeA05f7+cOX+3E1nSeAH6SO0Io8E6+Esoy+wJPABqlb1Fja\n27tP7969Y/vUHVKdawf2gXg6dYgk1avW1tY+wBhgw9QtenfH3Xzz/XuPHn1w6g695dWu197APGAf\n4CaKUW+Zc4DNgO+9y6/TAexA8cnk5sBJwF3AGcDZwG7AscDtQI+K/idQHVhCcQ7ec6lDtCKvxCuh\nCJ7Dp3epwvr0GT/FAU+qiEsc8CRp3eR5PgE4P3WH3t3ms2dP7z969IdTd2hF21EMeACbAh8Api33\n4wFcC5y2il/nHqAXsDPFMNDe9c8upBjtLgXOwgGvSX3XAa+cHPFKKoLLKT70kCpi4MCrPQ9PWneP\nAa2pIySpQfwUeCB1hFbu9KFDX8lg49QdWrnJFJe0fnS573sQ2AbovYp/9hreGvo2BY4G+lOMhJtR\nXKF3YgVbVTdG4ln9peXttCWWZWwLPA1slbpF9e+ZZ/Z4eI89nhuQukOqY/OA/hAvpA6RpEbR2tq6\nKzAW2CR1i1b0obFjH//UjTd+JHWHVm4+cDDFJa2fXu77z6S4Ffacd/ln24DtgWcpBr+3OwP4d2A0\ncCfFWXk+kaYpLAT2ImJS6hC9M6/EK7EIZlD8/imtsz59JrwvdYNU577qgCdJldX1tNpzU3doRd3b\n2xcd/6c/bZu6QyvXDnwGGMSKA95S4I8UD7l4N7dR3JL7TgPeGIrbavsA11HcmvsC8Py6Jas+nOeA\nV26OeCUXwZ+AK1N3qL69//3jpnTv3rFd6g6pjo2AGJY6QpIaUZ7nv6TYHVQSn77hhlHdOjt3TN2h\ndxbAFynOwvuvt/3Y3cD7gZ6r+DV+z8rPzPsu8D8UQ2FH1/e1UFyipYZ2D3BF6gi9O0e8+vCf+MGH\n1sGgQSM8D09aey9R3FEiSaqeL1L8fqvEtp4586X3jx/vESwlNhIYBtwL9Ot63dr1Y8ufc7fMdIrz\n7pZZQPEU2k/zz24CPkxxq+3mXb/2nsBiYK/K5KucZgNfwPPWSs8z8epElrEvxe/X3VO3qP48+2zf\nh/v2HeebMWnNLQUOhHg0dYgkNbrW1tZ9gYfwYZhJnfvDH47eeMGCvVf9MyU1iABOIOLPqUO0al6J\nVycifCKi1t7uu0/0PDxp7VzggCdJtZHn+WPAeak7mtnHHn74YQc8qen80AGvfjji1ZeLKK7Gk1Zb\n377PTvY8PGmt3Efx+64kqUbyPL8MuDl1RzNaf/HiuUfcddduqTsk1dSDFA84Vp1wxKsjEXQAQ4C5\nqVtUPwYNGvFK6gapDr0BDIHoTB0iSU3oc8CU1BHN5rTf/35MS8TWqTsk1cxrwKlELE0dotXniFdn\nIniJ4uBfabV86lM3dkvdINWhMyCmpo6QpGaU5/ls4BSKh2OqBnacMmXcTi+/fEDqDkk10wkMJGJ6\n6hCtGUe8OhTB9cCFqTtUH3r3fn7X1A1SnfkBxI2pIySpmeV5/ijwndQdzSDr7OwcNHx4ZwZ+8Cs1\nj/9DxD2pI7TmHPHq1/congAurdQeezzzUvfuHdum7pDqyC3Af6eOkCQB8CPgL6kjGt1h99wzcv22\ntj1Sd0iqmbuA/0kdobXjiFenIgiK8/GeSd2i8ho8eLi3A0qrbzwwCCJSh0iSIM/zAE4HXkzd0qg2\nnj9/1v4jR+6ZukNSzUwDBhGe+1yvHPHqWATzgeOB11O3qJxOPPEmb4uQVs8c4AQIHxwkSSWS5/kb\nwHH4YLeqGHLVVRMz2Dx1h6SaWAqcQsSs1CFae454da7rQRcnU/wLKa2gd+/nd0vdINWBTuA0iImp\nQyRJ/yzP8+eAU4GO1C2NpPeECWO3ee21/VN3SKqZM4kYmTpC68YRrwFE8Ffg66k7VC577vnUS926\ndW6dukOqA9+BuC11hCRp5fI8vw34RuqORtHS0dF+8nXXbZy6Q1LNXErEr1NHaN054jWICK4AfpW6\nQ+UxaNAIz8OTVu0aiEtSR0iSVi3P88uAK1N3NIJjb7llZI+lS71jQ2oONwHnpY5QZTjiNZavAQ+m\njlA5nHjiTd1TN0glNwb4QuoISdIa+SpwX+qIerb57NnT+o0Z85HUHZJqYjQ+yKKhOOI1kAjagc8A\nU1K3KL3ddpvUK3WDVGKzgBMhFqUOkSStvjzPl73ffSF1S7367O9+Ny0Db6WVGt9U4DgiFqYOUeU4\n4jWYCGZRPLF2QeoWpbPXXn970fPwpJVaRDHg+YGHJNWh5Z5YOyd1S73Za8yYxzafM2ff1B2Sqm4+\nxYA3PXWIKssRrwFFMBb4HBCJU5TIwIFXT0vdIJXUUuBfIR5OHSJJWnt5no8DTsEn1q627u3ti477\n85+3T90hqeo6gdOI+FvqEFWeI16DiuB64NupO5SG5+FJK/UliFtSR0iS1l2e53cAZ6XuqBcnXX/9\nqG6dnT1Td0iqunMJ3+82Kke8BhbBJcClqTtUe716veDTxqR/9i2I36WOkCRVTp7n/wt8N3VH2W09\nY8aLu0+YMCB1h6Sq+18iLksdoepxxGtwEXwD+G3qDtVOv35jXujWrfNfUndIJXMZxCWpIyRJlZfn\n+YXAj1N3lNnpV101J4P1UndIqqprga+ljlB1OeI1hy8Bf0ododoYNGiE5+FJKxoOnJM6QpJUVefi\nB9fvaMBDDz288cKF/VN3SKqqvwCDiehMHaLqcsRrAhF0AKcC9yVOUQ2ccMKfeqRukErkduALED7o\nR5IaWJ7nQfHB9R9Tt5TJ+osXzzn8nnt6p+6QVFV/BU4ioj11iKrPEa9JRLAYOAEYnbpF1dWr1wu+\nUZMKo4CTwDc0ktQM8jzvAAYCd6duKYuBV189tiXCY1akxvUocDwRi1OHqDYc8ZpIBHOBTwDPpG5R\ndfTvP/qFlpbYKnWHVALjgWMgFqQOkSTVTp7nS4ATKf5g29R2evnlcTtOmXJA6g5JVTMW+CQR81OH\nqHYc8ZpMBH8HjgAmpG5R5Q0ePNzz8CR4ETgS4vXUIZKk2svzfAFwNE38wXXW2dk5cMSIyPzzntSo\nJgBHEvFm6hDVlr+pN6EIZgKHU/xBVw3k+ONv9qljanbPAwdDvJI6RJKUTp7ns4EjgRdSt6Rw+N13\nP7R+W1vf1B2SqmIycAQRr6UOUe054jWpCKYBhwFTUreoMrKsM3bd9cXdUndICU0ADoGYmjpEkpRe\nnuevUrzfnZS6pZY2mTdv1oCHH94rdYekqpgOHE74frdZOeI1sQheprgib3rqFq27vff2PDw1teco\nBjx/P5Mk/UOe51OAg4FxqVtqZchVV03MYLPUHZIqbgbFFXjeUdfEHPGaXASTKIY8z1Krc4MGjfC/\nQzWrp4FDIWakDpEklU+e59MphryxqVuqrc/48X/betas/VN3SKq4qcDBRDTNBxJ6Z454IoLxwP7A\nxNQtWnvHH3/z+qkbpATGAoeBZ4JIklYuz/NZwKHA46lbqqWlo6P9pOuu2zR1h6SKewk4kAj/vC5H\nPBW6bq09AHgydYvWXJZ1xvve91Lv1B1SjY2mGPD+njpEklR+XQ+7OAIYmbqlGo67+eaR3Ts6eqXu\nkFRREygGvMmpQ1QOjnj6hwiWfUJ5T+oWrZl99nlyUktLbJm6Q6qhx4DDId5IHSJJqh95ns8FPgHc\nm7qlkt77xhtT9xo7dt/UHZIq6mmKW2g9Nkn/4IinFUQwDzgGuCF1i1bfkCHDXk3dINXQI8DHId5M\nHSJJqj95ni+geL97W+qWSjl96NDpGWyUukNSxTxCMeDNTB2icnHEK7Esy3bMsuyvWZY9l2XZs1mW\nnV2LrxvBEuBfgV/V4utp3R177C2eh6dm8ReKAW9u6hBJUv3K83wxcCJwY+qWddV/9OjHNp8zx6vw\npMZxB8VTaGenDlH5OOKV21LgnIjoC+wHfDXLsr61+MIRdEbwFeD7tfh6WntZ1hm77DLZ8/DUDH4B\nnACxIHWIJKn+5XneRvHB9f9L3bK2erS1LTzmllu2T90hqWKuBY4nYmHqEJWTI16JRcSrETG669vz\ngHHADrVt4L+BrwNRy6+r1ffhDz/xfEtLbJG6Q6qiAL4FcSZER+oYSVLjyPN8aZ7nZwDnU4fvd0+6\n/vrHu3V29kzdIakifgGcRkRb6hCVlyNenciybBegPzCq1l87gp8CQ4D2Wn9trdqQIcNmpG6QqqgN\nGARxSeoQSVLjyvP8B8BAYEnqltW1zYwZL/SeOHFA6g5J66wTOIeIM4noTB2jcnPEqwNZlm1C8aCJ\nr0ekOQcqghHACYCX9ZaM5+Gpgb0JHAnx+9QhkqTGl+f5NcDhwOupW1bH6VddNS+DHqk7JK2T+cCJ\nRPw4dYjqgyNeyWVZ1oNiwBsREX9M2RLBbcDHgTdSdugtLS0dsfPOL++eukOqgpeB/SHuTx0iSWoe\neZ6PBD4GTErd8m4OePDBkRstXNgvdYekdTIVOJCIP6cOUf1wxCuxLMsyioN2x0VJlvkIHgb2BZ5O\n3SLYd9/Hnm9pifem7pAqbDSwH8RzqUMkSc0nz/PnKR4q91DqlneywaJFcw69914/xJXq2xPAvkT8\nLXWI6osjXrntT3EW3WFZlv2t63V06qgIXqD4hPIPqVua3aBBIzwPT43mVuAgCP+3LUlKJs/z14Ej\ngNId6TBwxIixLRH/krpD0lq7ATiIiFdTh6j+OOKVWEQ8FBFZRHwoIvp1vW5N3QUQwYIITgW+Afi0\nyESOPfaWDVI3SBV0GXA8xILUIZIk5Xm+BBgEfD91yzI7T578XM+pUw9I3SFprV0EnEzEotQhqk9Z\nRN09SV0lk2UcTnFV3papW5pJS0tHZ3t7j7ktLbF56hZpHS0EzvABFpKksmptbT0Z+A2wSaqGrLOz\n87yLL56wflvbB1I1SFprbcCXiRiaOkT1zSvxtM4iuAfYh+IcK9XIRz866nkHPDWAl4ABDniSpDLL\n8/w64CNAsvNaP37XXQ854El1aRpwmAOeKsERTxURsexJklyVuqVZDB483DPDVO/uAj4MMTZ1iCRJ\nq5Ln+XiKB7zV/IOnTebNm7XfI4/sVeuvK2md3Qn0J2Jk6hA1Bm+nVcVlGWcBPwa6p25pZJMn7/zY\nzjtP2Td1h7QWArgQuACiM3GLJElrrLW19SzgR0CPWny9M6+4YuTWs2btX4uvJakiOoELgO8Tvt9V\n5TjiqSqyjIOAa4FtUrc0oq7z8Oa1tMRmqVukNfQ6MBji9tQhkiSti9bW1v2A64Ce1fw6fcaNG3Pq\nH/7Qv5pfQ1JFzQQGEnFv6hA1Hm+nVVVE8ADFOXmjUrc0oo997JGJDniqQ6OA/g54kqRGkOf5o8De\nwN3V+hrdli5tO+n6633PJ9WP+4B+DniqFkc8VU0E04CDgSsobp9ThQwaNGJm6gZpDf0MOBDildQh\nkiRVSp7ns4BPAN+nCu93j7v55ke6d3TsWulfV1LFBfAD4AgiPLtcVePttKqJLONI4DfADqlbGsHL\nL+/02E47veJ5eKoH04EvQNyROkSSpGpqbW2t6Pvd977++tSzLr98yww2rMSvJ6lqiuNiwrtNVH1e\niaeaiOBO4IPA8NQt9a6lpaNzxx1f6ZO6Q1oN1wJ7OuBJkppBnud3AntSoafXfnbo0OkOeFLp3Uvx\n9FkHPNWEI55qJoI3IxgCnAT8PXVPvRow4OGJWYZno6jM3gQGQZwC8UbqGEmSaiXP89l5ng8ETgXW\n+v8D937yycc2mzvXuy6k8loAfI3i9lmPi1HNOOKp5iK4geKqvJtTt9SjwYOHex6eyuxuiqvvrk4d\nIklSKnme/4Hi/e4aX53To61twdF/+YtH0Ejl9RCwFxFX4PlkqjHPxFNSWcbngZ8A70ndUi+mTNnx\n8R13nPqR1B3S2ywCzgN+Dv4fiyRJy7S2tv4bcCmw8er8/IHDh9/fe9Kkg6tbJWktLAbOB35CRGfq\nGDUnr8RTUhH8FvgQxaO4tQrdui3t6Nlz6u6pO6S3eQLYG+JyBzxJklaU5/kvgH7AI6v6udtOnz5p\nt0mTBlS/StIaGgX0I+LHDnhKyRFPyUXwMnAY8HWKq3m0Ep6Hp5JpB1qBj0GMTx0jSVJZ5Xk+CTgQ\n+A7QtrKfN2TYsAUZ9KhZmKRVaQO+DexPxITUMZIjnkohgojgp8DewOOpe8pqyJBhr6VukLrcD/SD\nuABiaeoYSZLKLs/zjjzPLwL2Ah54+48f8MADD220aNFetS+TtBKjgX2IuJiIjtQxEjjiqWQiGA8M\nAP4LmJM4p3Q++cnbNkrdoKb3GnA6xCEQz6WOkSSp3uR5Ph44BPgi8DrAhgsXvnnoX//aJ2WXpH+Y\nC/wn8FEinkkdIy3PB1uotLKMrYGLgM8DWeKc5Lp1W9rR3t5jQZb5EBAl0Qn8EvgOxJupY7IsOwr4\nKdAN+HVEXJw4SZKkNdba2roV8KMvXnnlLj2nTTsodY8khgHfJGJG6hDpnTjiqfSyjI8APwP2S92S\n0sEH3/fcffcd2jd1h5rSk8CZEKW41T3Lsm7ARODjwFSKW/BPi/DKQElSncqyg4ArgA+mTpGa1FPA\n14h4MHWI9G68nValF8HjFLfYfg5o2k9EhgwZNit1g5rOHOAsYN+yDHhd9gUmRcSLEdEGXAOckLhJ\nkqS1F/EA0J/iSJm5iWukZjIHOBvY2wFP9cART3Wh68EXQ4HdgR/yLk/1alRHHXX7xqkb1FSuBt4P\n8XOIztQxb7MD8Mpyfz+16/skSapfEUuJuAzoA4xInSM1uACuAvoQ8TMfXKF64YinuhLBvAi+CewJ\n3Ja6p1a6dVvasf320z3sWLUwCjgYYhB4FogkSTUXMYOIwcDHAK8MkirvKeAgIj5LxMzUMdKacMRT\nXYpgYgRHA8cBk1L3VNuBBz44PsvYNHWHGtp44NMQ+0E8kDpmFaYBOy739z27vk+SpMYR8SgRB1G8\n3/UJmdK6m0LxVOi9iXgodYy0NhzxVNciuAXYA/g2MD9xTtUMGTLs76kb1LCmAV8CPghxY+qY1fQ4\n0DvLsvdlWbYecCpwc+ImSZKqI+IWYC/g86x4nISk1TML+DqwOxG/8dZZ1TOfTquGkWVsBZwDfA3Y\nJHFORU2dusMTO+ww/cOpO9RQZgMXA5dDLEods6ayLDsa+AnQDfhNRHw/cZIkSdWXZRtQvNf9NrBF\n4hqp7OYClwKXEdGwF3youTjiqeFkGVtSPNnra8B7Euess+7d25e2ta23OMsaa5hUMouAy4GLIWan\njpEkSWshyzYHvgX8B7Bh4hqpbBYDPwcuJuL11DFSJXk7rRpOBK9HcD6wC/A/FI8Nr1sHH3z/BAc8\nVUAH8GugN8R5DniSJNWxiDeJ+BbQG7gSaE9cJJXBUuBXwG5EfMMBT43IEU8NK4LZEXyPYsy7AHgz\nadBaGjx4uOfhaV0sBn4B9IH4EoQPgJAkqVFETCPiy8CuFMdMLEhcJKWwFBgO9CXiK4Tvd9W4vJ1W\nTSPLeA/FLQf/SR2dITJt2vZPbr/9q/uk7lDdeRP4v8DPIGamjpEkSTWQZVsCZ3W96ub9rrSW5lPc\naXIZEVNSx0i14IinppNlbEpxXt45wJaJc96V5+FpLUwDLgN+BTEvdYwkSUogyzYGvkxxTnTPxDVS\npc0Efgb8L+ERMWou3k6rphPBvAguorjN9lvAjLRFK3fIIfeNd8DTahoPfAHYFeJHDniSJDWxiAVE\nXAb0Ar4ITEhcJFXCRIpxemcifuCAp2bkiKemFcH8CC4BdgL+Fbg3cdI/GTx4uIexalUeAU4E+kL8\nFqItdZAkSSqJiDYifgP0BU4CHk1cJK2NR4BPAR8g4koilqQOklLxdlppOVlGH+DfgM8C702cw/Tp\n2z253XYzPA9Pb7cYuA74JcTI1DGSJKmOZFk/ive7g8A7PlRa7cBNwM+IeCh1jFQWjnjSO8gyNgRO\nAc4E9k3R0KNHW/uSJeu3ZRkbp/j6KqVngCuBYeDtA5IkaR1k2aYUQ95XgH6Ja6RlJlK83x1KxKzU\nMVLZOOJJq5Bl7E3xaeVAqN2gduSRdzxzxx1HfbBWX0+ltQi4luJBFQ+njpEkSQ0oyz5K8X73FGDD\nxDVqPouBG4Aribg/dYxUZp6JJ61CBKMj+DKwPXAW8Gwtvu6gQSM8D6+5PUXxFOXtID7ngCdJkqom\nYhQRn6d4v/t1YFziIjWHZ4Czge2JGOyAJ62aV+JJayHLOBA4AzgB2KwaX+PVV7d9ctttZ3oeXnOZ\nC1xPcdXdqNQxkiSpiWXZAOA04GRgm8Q1ahwLKO4yuZKIR1LHSPXGEU9aB1nGesDHKZ72dQIVehhG\n13l47VnGRpX49VRqc4GbKR5UcQf4tC1JklQiWdYNOAQ4Ffg0sEXSHtWjxcCtwB+AW4hYmLhHqluO\neFKFZBk9gMMoBr0Tga3W9tf6xCduf/r22z+5Z6XaVDoOd5Ikqf5kWQ/gSIpB7wRg07RBKrE24G7g\nGuAmIuYl7pEagiOeVAVZxrJPLE8GPgVsvSb//LBhg+8bPHjEIZUvU0IOd5IkqXFk2QbAMRSD3jH4\nQAwVD2S7neIhFbcQMSdxj9RwHPGkKssyWoCDKK7Q+zSw3ar+mRkzthmzzTav9a92m6ruNeAOinPu\nHO4kSVJjyrKNgEOBTwJHAb3SBqmGZgF3ATcCtxGxIHGP1NAc8aQa6hr0BgDHAkcA/XnbU6LXW29J\n2+LFGyz1PLy6tBh4CLiT4s3MWPA3WUmS1GSyrDfFoPdJirtTNkjao0pqA0ZSvN+9ExiDo4JUM454\nUkJZxhYUn1oeTjHq9T7qqNuevu22oz0Prz4E8BTFYHcn8CDE4rRJkiRJJZJlGwIH89ao1zttkNbC\nON4a7e73ajspne6pA6RmFsEbFGdG3ACQZex46KF/HUBxG8KBeCtCGU2nOKT3zuKvMTNxjyRJUnlF\nLDsn7XbgbLKsF8VRMwO6Xh8AsnSBegfTKO4uKT6ojnglcY+kLl6JJ5Vatj3Fm5yDgAOAvkC3pEnN\npR0YAzwKPFK84uW0SZIkSQ0ky94L7Mdbo96+wCZJm5rLQuAJYBTFe95RRExLmyRpZRzxpLqSbQjs\nSXGW3rLXnvg0sEroBCZQvIl5EngcGO3tsZIkSTWUZd2AD7HiqLcrbztHWmslKG6NHcVbo90zRHQk\nrZK02hzxpLqXdQPez4rDXn9g85RVJTeDYrCbQPFG5klgDMT8pFWSJEn6Z8W5en2BPbpeH+z66054\nK+7KTKd4n7v8azQRc5JWSVonjnhSw8p2BnanOFdv17e9NksYViuLgOd5a6xb7hVzU4ZJkiSpArJs\nU51YMysAAAPKSURBVN4a95YNe7sDPWmO8987gBeB8aw41o13rJMakyOe1JSyLfjnYa8XsB2wJbAF\n5X7jswh4leKKumWvZX8/hWKsmwL+BidJktR0iltydwB2AXbu+mvPru9b9tqKcl/F105xNd10igdN\nTHvbt6cBU4hYkqxQUs054kl6B1lGcbXeVhSj3pYr+fZmQI+VvLqv5PsWUxygu6jrr6v69uusONLN\n8Eo6SZIkrZMsWw/Ynrfe067O6z0U72ezrlfLct9+p+8LYAEwr+s1fyXfXvb3s3lrqJuFf1iX9DaO\neJIkSZIkSVLJ+YQfSZIkSZIkqeQc8SRJkiRJkqSSc8STJEmSJEmSSs4RT5IkSZIkSSo5RzxJkiRJ\nkpRElmXdsiwbk2XZLalbpLJzxJMkSZIkSamcDYxLHSHVA0c8SZIkSZJUc1mW9QSOAX6dukWqB454\nkiRJkiQphZ8A3wQ6U4dI9cART5IkSZIk1VSWZccCr0XEk6lbpHrhiCdJkiRJkmptf+D4LMsmA9cA\nh2VZNjxtklRuWUSkbpAkSZIkSU0qy7JDgHMj4tjULVKZeSWeJEmSJEmSVHJeiSdJkiRJkiSVnFfi\nSZIkSZIkSSXniCdJkiRJkiSVnCOeJEmSJEmSVHKOeJIkSZIkSVLJOeJJkiRJkiRJJeeIJ0mSJEmS\nJJWcI54kSZIkSZJUco54kiRJkiRJUsk54kmSJEmSJEkl54gnSZIkSZIklZwjniRJkiRJklRyjniS\nJEmSJElSyTniSZIkSZIkSSXniCdJkiRJkiSVnCOeJEmSJEmSVHKOeJIkSZIkSVLJOeJJkiRJkiRJ\nJeeIJ0mSJEmSJJWcI54kSZIkSZJUco54kiRJkiRJUsk54kmSJEmSJEkl54gnSZIkSZIklZwjniRJ\nkiRJklRyjniSJEmSJElSyTniSZIkSZIkSSXniCdJkiRJkiSVnCOeJEmSJEmSVHKOeJIkSZIkSVLJ\nOeJJkiRJkiRJJeeIJ0mSJEmSJJWcI54kSZIkSZJUco54kiRJkiRJUsk54kmSJEmSJEkl54gnSZIk\nSZIklZwjniRJkiRJklRyjniSJEmSJElSyTniSZIkSZIkSSXniCdJkiRJkiSVnCOeJEmSJEmSVHKO\neJIkSZIkSVLJOeJJkiRJkiRJJeeIJ0mSJEmSJJWcI54kSZIkSZJUco54kiRJkiRJUsk54kmSJEmS\nJEkl54gnSZIkSZIklZwjniRJkiRJklRyjniSJEmSJElSyTniSZIkSZIkSSXniCdJkiRJkiSVnCOe\nJEmSJEmSVHL/H+sh6g+e7sIuAAAAAElFTkSuQmCC\n"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "## Aggregate users in each cluster and convert to Pandas for plotting\nbaseClusterAggDF = userPersonalityDF.groupBy('BASE_PREDICTIONS').agg(F.count('USER_SCREEN_NAME').alias('NUM_USERS')).toPandas()\nenrichedClusterAggDF = userPersonalityDF.groupBy('PI_PREDICTIONS').agg(F.count('USER_SCREEN_NAME').alias('NUM_USERS')).toPandas()\n\n%matplotlib inline\n# Code courtesy of \n#pandas_softdrink_tweets_grouped_by_sentiment=df_softdrink_tweets_grouped_by_sentiment.toPandas()\n#pandas_softdrink_tweets_grouped_by_sentiment.count()\nplot1_labels = baseClusterAggDF['BASE_PREDICTIONS']\nplot1_values = baseClusterAggDF['NUM_USERS']\nplot1_colors = ['green', 'gray', 'red', 'blue', 'yellow']\nplot2_labels = enrichedClusterAggDF['PI_PREDICTIONS']\nplot2_values = enrichedClusterAggDF['NUM_USERS']\nplot2_colors = ['green', 'gray', 'red', 'blue', 'yellow']\n\nfig, axes = plt.subplots(nrows=1, ncols=2, figsize=(23, 10))\naxes[0].pie(plot1_values,  labels=plot1_labels, colors=plot1_colors, autopct='%1.1f%%')\naxes[0].set_title('Cluster distrubtion without Personality traits')\naxes[0].set_aspect('equal')\naxes[0].legend(loc=\"upper right\", labels=plot1_labels)\naxes[1].pie(plot2_values,  labels=plot2_labels, colors=plot2_colors, autopct='%1.1f%%')\naxes[1].set_title('Cluster distrubtion with Personality traits')\naxes[1].set_aspect('equal')\naxes[1].legend(loc=\"upper right\", labels=plot2_labels)\nfig.subplots_adjust(hspace=1)\nplt.show()", 
            "execution_count": 49
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "### Scatter plots of clusters using Principal Components Analysis (PCA)\n\nTypically you would visualize clusters by plotting some aggregate measure of the data, then filling in the data points with different colors based on cluster ID. In the absence of aggregate metrics, we can use Principal Components Analysis (PCA) to compress our data set down to two dimensions. After we've performed PCA we can then plot the values of the two components on the X and Y axis to form a scatterplot.  \n\nLet's try that now."
        }, 
        {
            "cell_type": "code", 
            "metadata": {
                "collapsed": true
            }, 
            "outputs": [], 
            "source": "from pyspark.ml.feature import PCA\n\n## Get the first two principal components for features with and without enrichment from Personality Insights\npcaBase = PCA(k = 2, inputCol = \"BASE_FEATURES\", outputCol = \"pcaFeaturesBase\")\npcaEnriched = PCA(k = 2, inputCol = \"PI_ENRICHED_FEATURES\", outputCol = \"pcaFeaturesEnriched\")\n\n## Fit the model to our data\npcaBaseModel = pcaBase.fit(userPersonalityDF)\npcaEnrichedModel = pcaEnriched.fit(userPersonalityDF)\n\n## Transform the data get our principal components\nuserPersonalityDF = pcaBaseModel.transform(userPersonalityDF)\nuserPersonalityDF = pcaEnrichedModel.transform(userPersonalityDF)", 
            "execution_count": 50
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "We have the principal components, but before we can plot them we have to convert them from a feature vector back to individual columns.  We'll accomplish this using a lambda function to build a RowRDD, then convert back to a DataFrame."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "execute_result", 
                    "data": {
                        "text/plain": "  USER_SCREEN_NAME  PC1_BASE  PC2_BASE  BASE_PREDICTIONS  PC1_ENRICHED  \\\n0   KidrauhlSwag17  0.892480 -0.008917                 0      1.009923   \n1     flatlinesIut  0.837680  0.101299                 0      0.904827   \n2  JohnellaOrlando  0.717167  0.003790                 0      0.855621   \n3      realwonders  0.716606  0.271009                 0      0.943213   \n4          anotbok  0.718642  1.350662                 4      0.875348   \n\n   PC2_ENRICHED  PI_PREDICTIONS  \n0     -0.694524               3  \n1     -0.218704               3  \n2     -0.961229               3  \n3     -1.217703               3  \n4     -1.015899               2  ", 
                        "text/html": "<div>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>USER_SCREEN_NAME</th>\n      <th>PC1_BASE</th>\n      <th>PC2_BASE</th>\n      <th>BASE_PREDICTIONS</th>\n      <th>PC1_ENRICHED</th>\n      <th>PC2_ENRICHED</th>\n      <th>PI_PREDICTIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>KidrauhlSwag17</td>\n      <td>0.892480</td>\n      <td>-0.008917</td>\n      <td>0</td>\n      <td>1.009923</td>\n      <td>-0.694524</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>flatlinesIut</td>\n      <td>0.837680</td>\n      <td>0.101299</td>\n      <td>0</td>\n      <td>0.904827</td>\n      <td>-0.218704</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>JohnellaOrlando</td>\n      <td>0.717167</td>\n      <td>0.003790</td>\n      <td>0</td>\n      <td>0.855621</td>\n      <td>-0.961229</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>realwonders</td>\n      <td>0.716606</td>\n      <td>0.271009</td>\n      <td>0</td>\n      <td>0.943213</td>\n      <td>-1.217703</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>anotbok</td>\n      <td>0.718642</td>\n      <td>1.350662</td>\n      <td>4</td>\n      <td>0.875348</td>\n      <td>-1.015899</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
                    }, 
                    "metadata": {}, 
                    "execution_count": 51
                }
            ], 
            "source": "from pyspark.sql import Row\n\n## Split the features vector into columns with the rdd-based API, then convert to DF and reorder columns\npcaDF = userPersonalityDF.select(\"pcaFeaturesBase\", \"pcaFeaturesEnriched\", \"USER_SCREEN_NAME\", \"BASE_PREDICTIONS\", \"PI_PREDICTIONS\")\\\n                         .rdd.map(lambda x: Row(**{'PC1_BASE': float(x[0][0]), \n                                                   'PC2_BASE': float(x[0][1]),\n                                                   'PC1_ENRICHED': float(x[1][0]),\n                                                   'PC2_ENRICHED': float(x[1][1]),\n                                                   'USER_SCREEN_NAME': str(x[2]),\n                                                   'BASE_PREDICTIONS': int(x[3]),\n                                                   'PI_PREDICTIONS': int(x[4])}))\\\n                         .toDF()\\\n                         .select(\"USER_SCREEN_NAME\", \"PC1_BASE\", \"PC2_BASE\", \"BASE_PREDICTIONS\",\\\n                                 \"PC1_ENRICHED\", \"PC2_ENRICHED\", \"PI_PREDICTIONS\")\npcaDF.limit(5).toPandas()", 
            "execution_count": 51
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Now that the dimensionality of our dataset has been reduced to 2, we can view the components on a typical scatter plot.  Let's see how the clusters look from this perspective."
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/vnd.plotly.v1+html": "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>", 
                        "text/html": "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>"
                    }, 
                    "output_type": "display_data"
                }, 
                {
                    "metadata": {}, 
                    "data": {
                        "text/vnd.plotly.v1+html": "<div id=\"cb18c1e8-d868-4f63-8e54-a0a5ce2f4097\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"cb18c1e8-d868-4f63-8e54-a0a5ce2f4097\", [{\"name\": \"BASE_PREDICTIONS\", \"text\": [0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], \"marker\": {\"color\": [0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], \"size\": \"12\"}, \"mode\": \"markers\", \"y\": [0.8924802978482049, 0.8376797562613812, 0.717166796884082, 0.7166056580408464, 0.7186418594380158, 0.6852398929217907, 0.6608889633076349, 0.6616177490738482, 0.6366062770842857, 0.6362056013866877, 0.6359023063262487, 0.6177485313910458, 0.5925474840303098, 0.5869854706145748, 0.5749564337928471, 0.5600681410038006, 0.5411692985424806, 0.5198152464442554, 0.5080116317602078, 0.46692620559057535, 0.45345732569203195, 0.4530856476493637, 0.42116520204685826, 0.29825763548538853, 0.2982753606473062, -2.5650821051136286e-06, -0.0002774863474669601, 0.0007544654179948537, 5.760607533752699e-05, -0.00015067940850315524, -0.016039880659827564, 0.04345823675220718, -0.0006066590998510797, -2.2914445635504048e-05, 2.2100426247592813e-05, -2.011111919843442e-05, -0.00010528229709318268, 0.0002576460230158269, 0.0010558565424087238, -0.00011206170501954236, -0.006301481495769368, -0.0002442355478737391, -0.0020833601018750785, -0.0021055136103810624, -0.0026730954241142918, 5.514431688527466e-06, -0.0005576479228905408, 0.0015586347001302619, 0.0002527894944168214, -0.0005255309731090898, -0.000341016139806688, 4.2588738396880464e-05, 0.0004423056271415827, -0.00317011622342031, 2.0854897969478174e-05, -0.02596769064294192, -0.029826230012264365, -0.07826976471609678, -0.23969429066875278, -0.44264189275945315, -0.49731709901282634, -0.5094416786427068, -0.5524596960440226, -0.5642346141299643, -0.568394108442482, -0.591012040216381, -0.6553617093686552, -0.6559067967503555, -0.6939359951815318, -0.7173953522284979, -0.7354271553334483, -0.736372083930102, -0.7890300755524398, -0.7991058845224106, -0.7951974655299332, -0.8045060110276984, -0.808317801679788, -0.8372385082166419, -0.8342120650477965, -0.846482698966001, -0.8554982592761436, -0.880263540173367, -0.8816623705089347], \"x\": [-0.008917387062338975, 0.10129878302038337, 0.0037900294142462817, 0.27100865087265313, 1.3506619689228945, 0.050747873123150086, 0.003453387912846738, 0.07813648252300272, -0.0036212722841407062, 0.023256989603594035, 0.03617503501098932, 0.0016781276913304158, 0.013330587204817447, 0.03526230766229999, 0.006972651652509394, -0.005595212649334634, -0.00564759386036948, 0.00475984961307487, 0.22114081119007067, -0.002047123597962948, 0.002887968463491701, 0.15630450432019843, 0.009086555725614731, -0.000380031978078281, -0.0033276099560379533, 0.0010950183110153813, 0.019293598769904247, 0.011520033425725542, 0.0012799936931326681, 0.015618472263071585, 0.7391341026473907, 0.6358630429358322, 0.039148754708945945, 0.007108801380236422, 0.0009720562372245558, 0.002023434047310698, 0.025257110315410454, 0.05050767979556683, 0.03497552087826855, 0.013395503959282505, 0.34735879705027084, 0.043108490599350534, 0.10446184796267088, 0.15185203851126416, 0.1435221736386101, 0.0015166682821729199, 0.028388714198332842, 0.11710136408750385, 0.010959396546541166, 0.03325092830640606, 0.023789991820992954, 0.003597153673625718, 0.02944266533255544, 0.2677261865253255, 0.0004266593014739903, 0.3068143814270467, 0.05587263705244179, 0.07577557987003046, 0.08477528033316467, 0.07217095941454983, 0.5374285231973983, 0.040009906368505224, 0.2850657129660654, 0.021033784951724, 0.031710879345033964, 0.020209080723566603, 0.11826666686531058, 0.03123643468871769, 0.03404865757063656, 0.13506339943940138, 0.20186667100351727, 0.24752730065077627, 0.008947245133019305, 0.24979701840225893, 0.01517806128611034, 0.026333761397580355, 0.14205931683334327, 0.23006554070383728, 0.03258761772664834, 0.10712280865805235, 0.040735959464641924, 0.03438178041212821, 0.045741159853044853], \"type\": \"scatter\"}], {}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>", 
                        "application/vnd.plotly.v1+json": {
                            "data": [
                                {
                                    "mode": "markers", 
                                    "type": "scatter", 
                                    "text": [
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        4, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        0, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        4, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2, 
                                        2
                                    ], 
                                    "marker": {
                                        "color": [
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            4, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            0, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            4, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2, 
                                            2
                                        ], 
                                        "size": "12"
                                    }, 
                                    "y": [
                                        0.8924802978482049, 
                                        0.8376797562613812, 
                                        0.717166796884082, 
                                        0.7166056580408464, 
                                        0.7186418594380158, 
                                        0.6852398929217907, 
                                        0.6608889633076349, 
                                        0.6616177490738482, 
                                        0.6366062770842857, 
                                        0.6362056013866877, 
                                        0.6359023063262487, 
                                        0.6177485313910458, 
                                        0.5925474840303098, 
                                        0.5869854706145748, 
                                        0.5749564337928471, 
                                        0.5600681410038006, 
                                        0.5411692985424806, 
                                        0.5198152464442554, 
                                        0.5080116317602078, 
                                        0.46692620559057535, 
                                        0.45345732569203195, 
                                        0.4530856476493637, 
                                        0.42116520204685826, 
                                        0.29825763548538853, 
                                        0.2982753606473062, 
                                        -2.5650821051136286e-06, 
                                        -0.0002774863474669601, 
                                        0.0007544654179948537, 
                                        5.760607533752699e-05, 
                                        -0.00015067940850315524, 
                                        -0.016039880659827564, 
                                        0.04345823675220718, 
                                        -0.0006066590998510797, 
                                        -2.2914445635504048e-05, 
                                        2.2100426247592813e-05, 
                                        -2.011111919843442e-05, 
                                        -0.00010528229709318268, 
                                        0.0002576460230158269, 
                                        0.0010558565424087238, 
                                        -0.00011206170501954236, 
                                        -0.006301481495769368, 
                                        -0.0002442355478737391, 
                                        -0.0020833601018750785, 
                                        -0.0021055136103810624, 
                                        -0.0026730954241142918, 
                                        5.514431688527466e-06, 
                                        -0.0005576479228905408, 
                                        0.0015586347001302619, 
                                        0.0002527894944168214, 
                                        -0.0005255309731090898, 
                                        -0.000341016139806688, 
                                        4.2588738396880464e-05, 
                                        0.0004423056271415827, 
                                        -0.00317011622342031, 
                                        2.0854897969478174e-05, 
                                        -0.02596769064294192, 
                                        -0.029826230012264365, 
                                        -0.07826976471609678, 
                                        -0.23969429066875278, 
                                        -0.44264189275945315, 
                                        -0.49731709901282634, 
                                        -0.5094416786427068, 
                                        -0.5524596960440226, 
                                        -0.5642346141299643, 
                                        -0.568394108442482, 
                                        -0.591012040216381, 
                                        -0.6553617093686552, 
                                        -0.6559067967503555, 
                                        -0.6939359951815318, 
                                        -0.7173953522284979, 
                                        -0.7354271553334483, 
                                        -0.736372083930102, 
                                        -0.7890300755524398, 
                                        -0.7991058845224106, 
                                        -0.7951974655299332, 
                                        -0.8045060110276984, 
                                        -0.808317801679788, 
                                        -0.8372385082166419, 
                                        -0.8342120650477965, 
                                        -0.846482698966001, 
                                        -0.8554982592761436, 
                                        -0.880263540173367, 
                                        -0.8816623705089347
                                    ], 
                                    "x": [
                                        -0.008917387062338975, 
                                        0.10129878302038337, 
                                        0.0037900294142462817, 
                                        0.27100865087265313, 
                                        1.3506619689228945, 
                                        0.050747873123150086, 
                                        0.003453387912846738, 
                                        0.07813648252300272, 
                                        -0.0036212722841407062, 
                                        0.023256989603594035, 
                                        0.03617503501098932, 
                                        0.0016781276913304158, 
                                        0.013330587204817447, 
                                        0.03526230766229999, 
                                        0.006972651652509394, 
                                        -0.005595212649334634, 
                                        -0.00564759386036948, 
                                        0.00475984961307487, 
                                        0.22114081119007067, 
                                        -0.002047123597962948, 
                                        0.002887968463491701, 
                                        0.15630450432019843, 
                                        0.009086555725614731, 
                                        -0.000380031978078281, 
                                        -0.0033276099560379533, 
                                        0.0010950183110153813, 
                                        0.019293598769904247, 
                                        0.011520033425725542, 
                                        0.0012799936931326681, 
                                        0.015618472263071585, 
                                        0.7391341026473907, 
                                        0.6358630429358322, 
                                        0.039148754708945945, 
                                        0.007108801380236422, 
                                        0.0009720562372245558, 
                                        0.002023434047310698, 
                                        0.025257110315410454, 
                                        0.05050767979556683, 
                                        0.03497552087826855, 
                                        0.013395503959282505, 
                                        0.34735879705027084, 
                                        0.043108490599350534, 
                                        0.10446184796267088, 
                                        0.15185203851126416, 
                                        0.1435221736386101, 
                                        0.0015166682821729199, 
                                        0.028388714198332842, 
                                        0.11710136408750385, 
                                        0.010959396546541166, 
                                        0.03325092830640606, 
                                        0.023789991820992954, 
                                        0.003597153673625718, 
                                        0.02944266533255544, 
                                        0.2677261865253255, 
                                        0.0004266593014739903, 
                                        0.3068143814270467, 
                                        0.05587263705244179, 
                                        0.07577557987003046, 
                                        0.08477528033316467, 
                                        0.07217095941454983, 
                                        0.5374285231973983, 
                                        0.040009906368505224, 
                                        0.2850657129660654, 
                                        0.021033784951724, 
                                        0.031710879345033964, 
                                        0.020209080723566603, 
                                        0.11826666686531058, 
                                        0.03123643468871769, 
                                        0.03404865757063656, 
                                        0.13506339943940138, 
                                        0.20186667100351727, 
                                        0.24752730065077627, 
                                        0.008947245133019305, 
                                        0.24979701840225893, 
                                        0.01517806128611034, 
                                        0.026333761397580355, 
                                        0.14205931683334327, 
                                        0.23006554070383728, 
                                        0.03258761772664834, 
                                        0.10712280865805235, 
                                        0.040735959464641924, 
                                        0.03438178041212821, 
                                        0.045741159853044853
                                    ], 
                                    "name": "BASE_PREDICTIONS"
                                }
                            ], 
                            "layout": {}
                        }, 
                        "text/html": "<div id=\"cb18c1e8-d868-4f63-8e54-a0a5ce2f4097\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"cb18c1e8-d868-4f63-8e54-a0a5ce2f4097\", [{\"name\": \"BASE_PREDICTIONS\", \"text\": [0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], \"marker\": {\"color\": [0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], \"size\": \"12\"}, \"mode\": \"markers\", \"y\": [0.8924802978482049, 0.8376797562613812, 0.717166796884082, 0.7166056580408464, 0.7186418594380158, 0.6852398929217907, 0.6608889633076349, 0.6616177490738482, 0.6366062770842857, 0.6362056013866877, 0.6359023063262487, 0.6177485313910458, 0.5925474840303098, 0.5869854706145748, 0.5749564337928471, 0.5600681410038006, 0.5411692985424806, 0.5198152464442554, 0.5080116317602078, 0.46692620559057535, 0.45345732569203195, 0.4530856476493637, 0.42116520204685826, 0.29825763548538853, 0.2982753606473062, -2.5650821051136286e-06, -0.0002774863474669601, 0.0007544654179948537, 5.760607533752699e-05, -0.00015067940850315524, -0.016039880659827564, 0.04345823675220718, -0.0006066590998510797, -2.2914445635504048e-05, 2.2100426247592813e-05, -2.011111919843442e-05, -0.00010528229709318268, 0.0002576460230158269, 0.0010558565424087238, -0.00011206170501954236, -0.006301481495769368, -0.0002442355478737391, -0.0020833601018750785, -0.0021055136103810624, -0.0026730954241142918, 5.514431688527466e-06, -0.0005576479228905408, 0.0015586347001302619, 0.0002527894944168214, -0.0005255309731090898, -0.000341016139806688, 4.2588738396880464e-05, 0.0004423056271415827, -0.00317011622342031, 2.0854897969478174e-05, -0.02596769064294192, -0.029826230012264365, -0.07826976471609678, -0.23969429066875278, -0.44264189275945315, -0.49731709901282634, -0.5094416786427068, -0.5524596960440226, -0.5642346141299643, -0.568394108442482, -0.591012040216381, -0.6553617093686552, -0.6559067967503555, -0.6939359951815318, -0.7173953522284979, -0.7354271553334483, -0.736372083930102, -0.7890300755524398, -0.7991058845224106, -0.7951974655299332, -0.8045060110276984, -0.808317801679788, -0.8372385082166419, -0.8342120650477965, -0.846482698966001, -0.8554982592761436, -0.880263540173367, -0.8816623705089347], \"x\": [-0.008917387062338975, 0.10129878302038337, 0.0037900294142462817, 0.27100865087265313, 1.3506619689228945, 0.050747873123150086, 0.003453387912846738, 0.07813648252300272, -0.0036212722841407062, 0.023256989603594035, 0.03617503501098932, 0.0016781276913304158, 0.013330587204817447, 0.03526230766229999, 0.006972651652509394, -0.005595212649334634, -0.00564759386036948, 0.00475984961307487, 0.22114081119007067, -0.002047123597962948, 0.002887968463491701, 0.15630450432019843, 0.009086555725614731, -0.000380031978078281, -0.0033276099560379533, 0.0010950183110153813, 0.019293598769904247, 0.011520033425725542, 0.0012799936931326681, 0.015618472263071585, 0.7391341026473907, 0.6358630429358322, 0.039148754708945945, 0.007108801380236422, 0.0009720562372245558, 0.002023434047310698, 0.025257110315410454, 0.05050767979556683, 0.03497552087826855, 0.013395503959282505, 0.34735879705027084, 0.043108490599350534, 0.10446184796267088, 0.15185203851126416, 0.1435221736386101, 0.0015166682821729199, 0.028388714198332842, 0.11710136408750385, 0.010959396546541166, 0.03325092830640606, 0.023789991820992954, 0.003597153673625718, 0.02944266533255544, 0.2677261865253255, 0.0004266593014739903, 0.3068143814270467, 0.05587263705244179, 0.07577557987003046, 0.08477528033316467, 0.07217095941454983, 0.5374285231973983, 0.040009906368505224, 0.2850657129660654, 0.021033784951724, 0.031710879345033964, 0.020209080723566603, 0.11826666686531058, 0.03123643468871769, 0.03404865757063656, 0.13506339943940138, 0.20186667100351727, 0.24752730065077627, 0.008947245133019305, 0.24979701840225893, 0.01517806128611034, 0.026333761397580355, 0.14205931683334327, 0.23006554070383728, 0.03258761772664834, 0.10712280865805235, 0.040735959464641924, 0.03438178041212821, 0.045741159853044853], \"type\": \"scatter\"}], {}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "import plotly \nfrom plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\nimport plotly.graph_objs as go\ninit_notebook_mode(connected=True)\n\npcaDF = pcaDF.toPandas()\n\n## For Base Features\ndata = [go.Scatter(x = pcaDF.PC2_BASE, \n                   y = pcaDF.PC1_BASE, \n                   mode = 'markers',\n                   name = 'BASE_PREDICTIONS',\n                   marker = dict(color = pcaDF.BASE_PREDICTIONS, size = '12'),\n                   text = pcaDF.BASE_PREDICTIONS\n                                 )\n       ]\n\nplotly.offline.iplot(data)", 
            "execution_count": 52
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "What is interesting to note about the clustering without Personality Insights enrichment is that it appears to have a very clear stratification from -1 to 1.  These clusters appear to be somewhat intuitively grouped together.  What about when we run the same algorithm on the PI-enriched data?"
        }, 
        {
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "metadata": {}, 
                    "data": {
                        "text/vnd.plotly.v1+html": "<div id=\"4b2c2aa9-49a6-4bda-83cb-6fa2965a0c38\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"4b2c2aa9-49a6-4bda-83cb-6fa2965a0c38\", [{\"name\": \"Clusters with PI\", \"text\": [3, 3, 3, 3, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 3, 1, 1, 1, 1, 1, 2, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4], \"marker\": {\"color\": [3, 3, 3, 3, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 3, 1, 1, 1, 1, 1, 2, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4], \"size\": \"12\"}, \"mode\": \"markers\", \"y\": [1.009923311803248, 0.9048269203013821, 0.8556210493601677, 0.9432134781467301, 0.8753482625169718, 0.8544545678342522, 0.710251271024436, 0.7234032268508362, 0.6847660115385386, 0.6925720297756528, 0.7936505899412516, 0.775061465720449, 0.67068181523185, 0.7557763597051093, 0.6517765414695528, 0.6041789635795567, 0.570992957734065, 0.647957234824169, 0.6587623824014626, 0.546901156879275, 0.6484812696698836, 0.5762339929635214, 0.5884657211247108, 0.37574036949196743, 0.40797512138489733, 0.07269511911045146, 0.07260247399848173, 0.16297329676829347, 0.10658443526666136, 0.04199288349644573, 0.046306360504653275, 0.18784134777838182, 0.10260867250287549, 0.046176432576408855, 0.07735465679926617, 0.11679242378872458, 0.12204405331398985, 0.036935754536308936, 0.0740488185290661, 0.10132108224606243, 0.06369930903985357, 0.07690391112769027, 0.06199412402820889, 0.05025094246688025, 0.1820102286524334, 0.15449946642663887, 0.10284935895712702, 0.0859985459163844, 0.05147079329578247, 0.06628264753211402, 0.07254430285128532, 0.10430929657711402, 0.16884143703269, 0.137265433692317, 0.061414578282531655, 0.22867324506841818, 0.06495730950216559, 0.012719773886536525, -0.12901943137276697, -0.3701941024009937, -0.34901493768592234, -0.3791057821401316, -0.49147575576591546, -0.4728521483555232, -0.5023894312196765, -0.5047997930938318, -0.5268626941812119, -0.5733219558812409, -0.5057545905981684, -0.5401315605521926, -0.6625040661967789, -0.6492812529409252, -0.6626154254901774, -0.6261595018889144, -0.7012409341611875, -0.725357523845665, -0.6783413494008895, -0.6981295732743854, -0.7510637792202864, -0.7628604025563008, -0.8228531821564232, -0.8114490651866719, -0.7798036694036138], \"x\": [-0.6945239289928102, -0.21870414847610822, -0.9612292989122622, -1.2177027044233877, -1.0158994762583236, -0.7456449979803448, -0.332498657418115, -0.38534537996287327, -0.2957781878478083, -0.36506355273053626, -0.9189652615758528, -0.9766329254090614, -0.5117454410123465, -1.1419988424999175, -0.4450015563600886, -0.2771099684461742, -0.07608279928711745, -0.8369347327021317, -0.8636216281157335, -0.5390751049674114, -1.1122180498588348, -0.7729813144969008, -1.1006450287410103, -0.5350734456600351, -0.7780963502930831, -0.526005273457274, -0.48259699880500584, -0.8124645275269012, -0.7690130497500625, -0.24451465097927008, -0.5483645639591972, -0.8951701369516797, -0.582294897877866, -0.33193925881205694, -0.49123111053484747, -0.832639652669443, -0.8524204914549983, -0.2809789352139572, -0.43551280664953584, -0.5120755484624278, -0.5221544153415156, -0.5072100246395643, -0.38892164838103715, -0.4097248854800819, -1.0815893509407422, -1.05767483400609, -0.6707665430838269, -0.5606673667609451, -0.3427825646656021, -0.47121265323215084, -0.5070376818034162, -0.7266616176120605, -1.161925971145593, -1.003734131233414, -0.43858398998886255, -1.5390019640783712, -0.5658624124860434, -0.49884137444393784, -0.7346243919622337, -0.5556136410127439, -1.130632697917763, -0.9429206033840992, -0.5051411913501477, -0.6709270459525463, -0.4007249899020467, -0.5856987298657882, -0.932327690996167, -0.6098365391324154, -1.2228477030742313, -1.2143203496232635, -0.577440826763362, -0.696491320473642, -0.9075206832026554, -1.2001576406452397, -0.6693023479324943, -0.5064545840339492, -0.8684817775788476, -0.9870938798256813, -0.6008046044624968, -0.6200962319024507, -0.2829976950452319, -0.5671113266459343, -0.6165428036560149], \"type\": \"scatter\"}], {}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>", 
                        "application/vnd.plotly.v1+json": {
                            "data": [
                                {
                                    "mode": "markers", 
                                    "type": "scatter", 
                                    "text": [
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        2, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        3, 
                                        1, 
                                        3, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        2, 
                                        0, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        1, 
                                        2, 
                                        1, 
                                        1, 
                                        1, 
                                        4, 
                                        2, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4, 
                                        4
                                    ], 
                                    "marker": {
                                        "color": [
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            2, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            3, 
                                            1, 
                                            3, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            2, 
                                            0, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            1, 
                                            2, 
                                            1, 
                                            1, 
                                            1, 
                                            4, 
                                            2, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4, 
                                            4
                                        ], 
                                        "size": "12"
                                    }, 
                                    "y": [
                                        1.009923311803248, 
                                        0.9048269203013821, 
                                        0.8556210493601677, 
                                        0.9432134781467301, 
                                        0.8753482625169718, 
                                        0.8544545678342522, 
                                        0.710251271024436, 
                                        0.7234032268508362, 
                                        0.6847660115385386, 
                                        0.6925720297756528, 
                                        0.7936505899412516, 
                                        0.775061465720449, 
                                        0.67068181523185, 
                                        0.7557763597051093, 
                                        0.6517765414695528, 
                                        0.6041789635795567, 
                                        0.570992957734065, 
                                        0.647957234824169, 
                                        0.6587623824014626, 
                                        0.546901156879275, 
                                        0.6484812696698836, 
                                        0.5762339929635214, 
                                        0.5884657211247108, 
                                        0.37574036949196743, 
                                        0.40797512138489733, 
                                        0.07269511911045146, 
                                        0.07260247399848173, 
                                        0.16297329676829347, 
                                        0.10658443526666136, 
                                        0.04199288349644573, 
                                        0.046306360504653275, 
                                        0.18784134777838182, 
                                        0.10260867250287549, 
                                        0.046176432576408855, 
                                        0.07735465679926617, 
                                        0.11679242378872458, 
                                        0.12204405331398985, 
                                        0.036935754536308936, 
                                        0.0740488185290661, 
                                        0.10132108224606243, 
                                        0.06369930903985357, 
                                        0.07690391112769027, 
                                        0.06199412402820889, 
                                        0.05025094246688025, 
                                        0.1820102286524334, 
                                        0.15449946642663887, 
                                        0.10284935895712702, 
                                        0.0859985459163844, 
                                        0.05147079329578247, 
                                        0.06628264753211402, 
                                        0.07254430285128532, 
                                        0.10430929657711402, 
                                        0.16884143703269, 
                                        0.137265433692317, 
                                        0.061414578282531655, 
                                        0.22867324506841818, 
                                        0.06495730950216559, 
                                        0.012719773886536525, 
                                        -0.12901943137276697, 
                                        -0.3701941024009937, 
                                        -0.34901493768592234, 
                                        -0.3791057821401316, 
                                        -0.49147575576591546, 
                                        -0.4728521483555232, 
                                        -0.5023894312196765, 
                                        -0.5047997930938318, 
                                        -0.5268626941812119, 
                                        -0.5733219558812409, 
                                        -0.5057545905981684, 
                                        -0.5401315605521926, 
                                        -0.6625040661967789, 
                                        -0.6492812529409252, 
                                        -0.6626154254901774, 
                                        -0.6261595018889144, 
                                        -0.7012409341611875, 
                                        -0.725357523845665, 
                                        -0.6783413494008895, 
                                        -0.6981295732743854, 
                                        -0.7510637792202864, 
                                        -0.7628604025563008, 
                                        -0.8228531821564232, 
                                        -0.8114490651866719, 
                                        -0.7798036694036138
                                    ], 
                                    "x": [
                                        -0.6945239289928102, 
                                        -0.21870414847610822, 
                                        -0.9612292989122622, 
                                        -1.2177027044233877, 
                                        -1.0158994762583236, 
                                        -0.7456449979803448, 
                                        -0.332498657418115, 
                                        -0.38534537996287327, 
                                        -0.2957781878478083, 
                                        -0.36506355273053626, 
                                        -0.9189652615758528, 
                                        -0.9766329254090614, 
                                        -0.5117454410123465, 
                                        -1.1419988424999175, 
                                        -0.4450015563600886, 
                                        -0.2771099684461742, 
                                        -0.07608279928711745, 
                                        -0.8369347327021317, 
                                        -0.8636216281157335, 
                                        -0.5390751049674114, 
                                        -1.1122180498588348, 
                                        -0.7729813144969008, 
                                        -1.1006450287410103, 
                                        -0.5350734456600351, 
                                        -0.7780963502930831, 
                                        -0.526005273457274, 
                                        -0.48259699880500584, 
                                        -0.8124645275269012, 
                                        -0.7690130497500625, 
                                        -0.24451465097927008, 
                                        -0.5483645639591972, 
                                        -0.8951701369516797, 
                                        -0.582294897877866, 
                                        -0.33193925881205694, 
                                        -0.49123111053484747, 
                                        -0.832639652669443, 
                                        -0.8524204914549983, 
                                        -0.2809789352139572, 
                                        -0.43551280664953584, 
                                        -0.5120755484624278, 
                                        -0.5221544153415156, 
                                        -0.5072100246395643, 
                                        -0.38892164838103715, 
                                        -0.4097248854800819, 
                                        -1.0815893509407422, 
                                        -1.05767483400609, 
                                        -0.6707665430838269, 
                                        -0.5606673667609451, 
                                        -0.3427825646656021, 
                                        -0.47121265323215084, 
                                        -0.5070376818034162, 
                                        -0.7266616176120605, 
                                        -1.161925971145593, 
                                        -1.003734131233414, 
                                        -0.43858398998886255, 
                                        -1.5390019640783712, 
                                        -0.5658624124860434, 
                                        -0.49884137444393784, 
                                        -0.7346243919622337, 
                                        -0.5556136410127439, 
                                        -1.130632697917763, 
                                        -0.9429206033840992, 
                                        -0.5051411913501477, 
                                        -0.6709270459525463, 
                                        -0.4007249899020467, 
                                        -0.5856987298657882, 
                                        -0.932327690996167, 
                                        -0.6098365391324154, 
                                        -1.2228477030742313, 
                                        -1.2143203496232635, 
                                        -0.577440826763362, 
                                        -0.696491320473642, 
                                        -0.9075206832026554, 
                                        -1.2001576406452397, 
                                        -0.6693023479324943, 
                                        -0.5064545840339492, 
                                        -0.8684817775788476, 
                                        -0.9870938798256813, 
                                        -0.6008046044624968, 
                                        -0.6200962319024507, 
                                        -0.2829976950452319, 
                                        -0.5671113266459343, 
                                        -0.6165428036560149
                                    ], 
                                    "name": "Clusters with PI"
                                }
                            ], 
                            "layout": {}
                        }, 
                        "text/html": "<div id=\"4b2c2aa9-49a6-4bda-83cb-6fa2965a0c38\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"4b2c2aa9-49a6-4bda-83cb-6fa2965a0c38\", [{\"name\": \"Clusters with PI\", \"text\": [3, 3, 3, 3, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 3, 1, 1, 1, 1, 1, 2, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4], \"marker\": {\"color\": [3, 3, 3, 3, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 3, 1, 1, 1, 1, 1, 2, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4], \"size\": \"12\"}, \"mode\": \"markers\", \"y\": [1.009923311803248, 0.9048269203013821, 0.8556210493601677, 0.9432134781467301, 0.8753482625169718, 0.8544545678342522, 0.710251271024436, 0.7234032268508362, 0.6847660115385386, 0.6925720297756528, 0.7936505899412516, 0.775061465720449, 0.67068181523185, 0.7557763597051093, 0.6517765414695528, 0.6041789635795567, 0.570992957734065, 0.647957234824169, 0.6587623824014626, 0.546901156879275, 0.6484812696698836, 0.5762339929635214, 0.5884657211247108, 0.37574036949196743, 0.40797512138489733, 0.07269511911045146, 0.07260247399848173, 0.16297329676829347, 0.10658443526666136, 0.04199288349644573, 0.046306360504653275, 0.18784134777838182, 0.10260867250287549, 0.046176432576408855, 0.07735465679926617, 0.11679242378872458, 0.12204405331398985, 0.036935754536308936, 0.0740488185290661, 0.10132108224606243, 0.06369930903985357, 0.07690391112769027, 0.06199412402820889, 0.05025094246688025, 0.1820102286524334, 0.15449946642663887, 0.10284935895712702, 0.0859985459163844, 0.05147079329578247, 0.06628264753211402, 0.07254430285128532, 0.10430929657711402, 0.16884143703269, 0.137265433692317, 0.061414578282531655, 0.22867324506841818, 0.06495730950216559, 0.012719773886536525, -0.12901943137276697, -0.3701941024009937, -0.34901493768592234, -0.3791057821401316, -0.49147575576591546, -0.4728521483555232, -0.5023894312196765, -0.5047997930938318, -0.5268626941812119, -0.5733219558812409, -0.5057545905981684, -0.5401315605521926, -0.6625040661967789, -0.6492812529409252, -0.6626154254901774, -0.6261595018889144, -0.7012409341611875, -0.725357523845665, -0.6783413494008895, -0.6981295732743854, -0.7510637792202864, -0.7628604025563008, -0.8228531821564232, -0.8114490651866719, -0.7798036694036138], \"x\": [-0.6945239289928102, -0.21870414847610822, -0.9612292989122622, -1.2177027044233877, -1.0158994762583236, -0.7456449979803448, -0.332498657418115, -0.38534537996287327, -0.2957781878478083, -0.36506355273053626, -0.9189652615758528, -0.9766329254090614, -0.5117454410123465, -1.1419988424999175, -0.4450015563600886, -0.2771099684461742, -0.07608279928711745, -0.8369347327021317, -0.8636216281157335, -0.5390751049674114, -1.1122180498588348, -0.7729813144969008, -1.1006450287410103, -0.5350734456600351, -0.7780963502930831, -0.526005273457274, -0.48259699880500584, -0.8124645275269012, -0.7690130497500625, -0.24451465097927008, -0.5483645639591972, -0.8951701369516797, -0.582294897877866, -0.33193925881205694, -0.49123111053484747, -0.832639652669443, -0.8524204914549983, -0.2809789352139572, -0.43551280664953584, -0.5120755484624278, -0.5221544153415156, -0.5072100246395643, -0.38892164838103715, -0.4097248854800819, -1.0815893509407422, -1.05767483400609, -0.6707665430838269, -0.5606673667609451, -0.3427825646656021, -0.47121265323215084, -0.5070376818034162, -0.7266616176120605, -1.161925971145593, -1.003734131233414, -0.43858398998886255, -1.5390019640783712, -0.5658624124860434, -0.49884137444393784, -0.7346243919622337, -0.5556136410127439, -1.130632697917763, -0.9429206033840992, -0.5051411913501477, -0.6709270459525463, -0.4007249899020467, -0.5856987298657882, -0.932327690996167, -0.6098365391324154, -1.2228477030742313, -1.2143203496232635, -0.577440826763362, -0.696491320473642, -0.9075206832026554, -1.2001576406452397, -0.6693023479324943, -0.5064545840339492, -0.8684817775788476, -0.9870938798256813, -0.6008046044624968, -0.6200962319024507, -0.2829976950452319, -0.5671113266459343, -0.6165428036560149], \"type\": \"scatter\"}], {}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
                    }, 
                    "output_type": "display_data"
                }
            ], 
            "source": "## For Enriched Features\ndata = [go.Scatter(x = pcaDF.PC2_ENRICHED, \n                   y = pcaDF.PC1_ENRICHED, \n                   mode = 'markers',\n                   name = 'Clusters with PI',\n                   marker = dict(color = pcaDF.PI_PREDICTIONS, size = '12'),\n                   text = pcaDF.PI_PREDICTIONS\n                                 )\n       ]\n\nplotly.offline.iplot(data)", 
            "execution_count": 53
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Not quite the same intuitive results.  Perhaps we should iterate over this and try a different clustering algorithm some time.\n\n**At this point you may be wondering what the utility of all this work might be.**  Well, now that we've enriched the user data with Watson APIs and grouped them into clusters, we can track these clusters over time to see how they respond to various metrics.  This could be purchase history, click patterns, or the response to different advertisement campaigns.  As we build up this data set we'll be able to classify new users and what group they fall into, resulting in a stronger user segmentation model over time."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "_______\n\n## Summary\nWe've accomplished quite a lot in this notebook, so let's take a moment to review.\n\nFirst we loaded our tweet data from Db2 Warehouse on Cloud.  Then we enriched it with several Watson APIs - Natural Language Understanding and Personality Insights.  Using Spark and its elegant API we shaped our data, explored it visually, and prepared it for machine learning.  Our approach was to discover user segmentation by assigning cluster IDs to each data point.  Finally, we plotted the results of our clustering with the aid of principal components analysis.  \n\nWhile more work remains to be done - no one said data science was easy - we have seen how we can seamlessly move between Watson APIs, Spark, and Python using the Data Science Experience. \n_______"
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {
                "collapsed": true
            }, 
            "source": "### Authors\n**Rafi Kurlansik** is an Open Source Solutions Engineer specializing in big data technologies, such as Hadoop and Spark. He's responsible for developing and delivering demonstrations of IBM tech to both enterprise clients and the larger analytics community. Kurlansik has hands-on experience with machine learning, natural language processing, data visualization, and dashboard development. If you're wondering where he comes down on the biggest data science debate of our day, Rafi is, in his own words, \"an avid R fan, especially RStudio!\"\n\n**Joseph Kozhaya** is an IBM Master Inventor and a Watson Solution Architect working closely with partners, clients, and universities to build cloud and cognitive solutions."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "Copyright \u00a9 IBM Corp. 2017. This notebook and its source code are released under the terms of the MIT License."
        }, 
        {
            "cell_type": "markdown", 
            "metadata": {}, 
            "source": "<div><br><img src=\"https://upload.wikimedia.org/wikipedia/commons/thumb/5/51/IBM_logo.svg/640px-IBM_logo.svg.png\" width = 200 height = 200>\n</div><br>"
        }
    ], 
    "nbformat": 4, 
    "metadata": {
        "kernelspec": {
            "display_name": "Python 2 with Spark 2.1", 
            "name": "python2-spark21", 
            "language": "python"
        }, 
        "language_info": {
            "version": "2.7.11", 
            "codemirror_mode": {
                "version": 2, 
                "name": "ipython"
            }, 
            "mimetype": "text/x-python", 
            "nbconvert_exporter": "python", 
            "file_extension": ".py", 
            "name": "python", 
            "pygments_lexer": "ipython2"
        }
    }, 
    "nbformat_minor": 1
}