{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "### Screencast\n", "\n", "In the previous video, I brought a few questions we will be exploring throughout this lesson. First, let's take a look at the data, and see how we might answer the first question about how to break into the field of becoming a software developoer according to the survey results.\n", "\n", "To get started, let's read in the necessary libraries we will need to wrangle our data: pandas and numpy. If we decided to build some basic plots, matplotlib might prove useful as well." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
RespondentProfessionalProgramHobbyCountryUniversityEmploymentStatusFormalEducationMajorUndergradHomeRemoteCompanySize...StackOverflowMakeMoneyGenderHighestEducationParentsRaceSurveyLongQuestionsInterestingQuestionsConfusingInterestedAnswersSalaryExpectedSalary
01StudentYes, bothUnited StatesNoNot employed, and not looking for workSecondary schoolNaNNaNNaN...Strongly disagreeMaleHigh schoolWhite or of European descentStrongly disagreeStrongly agreeDisagreeStrongly agreeNaNNaN
12StudentYes, bothUnited KingdomYes, full-timeEmployed part-timeSome college/university study without earning ...Computer science or software engineeringMore than half, but not all, the time20 to 99 employees...Strongly disagreeMaleA master's degreeWhite or of European descentSomewhat agreeSomewhat agreeDisagreeStrongly agreeNaN37500.0
23Professional developerYes, bothUnited KingdomNoEmployed full-timeBachelor's degreeComputer science or software engineeringLess than half the time, but at least one day ...10,000 or more employees...DisagreeMaleA professional degreeWhite or of European descentSomewhat agreeAgreeDisagreeAgree113750.0NaN
34Professional non-developer who sometimes write...Yes, bothUnited StatesNoEmployed full-timeDoctoral degreeA non-computer-focused engineering disciplineLess than half the time, but at least one day ...10,000 or more employees...DisagreeMaleA doctoral degreeWhite or of European descentAgreeAgreeSomewhat agreeStrongly agreeNaNNaN
45Professional developerYes, I program as a hobbySwitzerlandNoEmployed full-timeMaster's degreeComputer science or software engineeringNever10 to 19 employees...NaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
\n", "

5 rows × 154 columns

\n", "
" ], "text/plain": [ " Respondent Professional \\\n", "0 1 Student \n", "1 2 Student \n", "2 3 Professional developer \n", "3 4 Professional non-developer who sometimes write... \n", "4 5 Professional developer \n", "\n", " ProgramHobby Country University \\\n", "0 Yes, both United States No \n", "1 Yes, both United Kingdom Yes, full-time \n", "2 Yes, both United Kingdom No \n", "3 Yes, both United States No \n", "4 Yes, I program as a hobby Switzerland No \n", "\n", " EmploymentStatus \\\n", "0 Not employed, and not looking for work \n", "1 Employed part-time \n", "2 Employed full-time \n", "3 Employed full-time \n", "4 Employed full-time \n", "\n", " FormalEducation \\\n", "0 Secondary school \n", "1 Some college/university study without earning ... \n", "2 Bachelor's degree \n", "3 Doctoral degree \n", "4 Master's degree \n", "\n", " MajorUndergrad \\\n", "0 NaN \n", "1 Computer science or software engineering \n", "2 Computer science or software engineering \n", "3 A non-computer-focused engineering discipline \n", "4 Computer science or software engineering \n", "\n", " HomeRemote \\\n", "0 NaN \n", "1 More than half, but not all, the time \n", "2 Less than half the time, but at least one day ... \n", "3 Less than half the time, but at least one day ... \n", "4 Never \n", "\n", " CompanySize ... StackOverflowMakeMoney Gender \\\n", "0 NaN ... Strongly disagree Male \n", "1 20 to 99 employees ... Strongly disagree Male \n", "2 10,000 or more employees ... Disagree Male \n", "3 10,000 or more employees ... Disagree Male \n", "4 10 to 19 employees ... NaN NaN \n", "\n", " HighestEducationParents Race SurveyLong \\\n", "0 High school White or of European descent Strongly disagree \n", "1 A master's degree White or of European descent Somewhat agree \n", "2 A professional degree White or of European descent Somewhat agree \n", "3 A doctoral degree White or of European descent Agree \n", "4 NaN NaN NaN \n", "\n", " QuestionsInteresting QuestionsConfusing InterestedAnswers Salary \\\n", "0 Strongly agree Disagree Strongly agree NaN \n", "1 Somewhat agree Disagree Strongly agree NaN \n", "2 Agree Disagree Agree 113750.0 \n", "3 Agree Somewhat agree Strongly agree NaN \n", "4 NaN NaN NaN NaN \n", "\n", " ExpectedSalary \n", "0 NaN \n", "1 37500.0 \n", "2 NaN \n", "3 NaN \n", "4 NaN \n", "\n", "[5 rows x 154 columns]" ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "from collections import defaultdict\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "\n", "df = pd.read_csv('./Part I/stackoverflow/survey_results_public.csv')\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now to look at our first question of interest: What do those employed in the industry suggest to help others enter the field? Looking at the `CousinEducation` field, you can see what these individuals would suggest to help others break into their field. Below you can take a look at the full field that survey participants would see." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[\"Let's pretend you have a distant cousin. They are 24 years old, have a college degree in a field not related to computer programming, and have been working a non-coding job for the last two years. They want your advice on how to switch to a career as a software developer. Which of the following options would you most strongly recommend to your cousin?\\nLet's pretend you have a distant cousin named Robert. He is 24 years old, has a college degree in a field not related to computer programming, and has been working a non-coding job for the last two years. He wants your advice on how to switch to a career as a software developer. Which of the following options would you most strongly recommend to Robert?\\nLet's pretend you have a distant cousin named Alice. She is 24 years old, has a college degree in a field not related to computer programming, and has been working a non-coding job for the last two years. She wants your advice on how to switch to a career as a software developer. Which of the following options would you most strongly recommend to Alice?\"]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df2 = pd.read_csv('./Part I/stackoverflow/survey_results_schema.csv')\n", "list(df2[df2.Column == 'CousinEducation']['Question'])" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
indexCousinEducation
0Take online courses; Buy books and work throug...711
1Take online courses551
2None of these523
3Take online courses; Part-time/evening courses...479
4Take online courses; Bootcamp; Part-time/eveni...465
\n", "
" ], "text/plain": [ " index CousinEducation\n", "0 Take online courses; Buy books and work throug... 711\n", "1 Take online courses 551\n", "2 None of these 523\n", "3 Take online courses; Part-time/evening courses... 479\n", "4 Take online courses; Bootcamp; Part-time/eveni... 465" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#Let's have a look at what the participants say\n", "study = df['CousinEducation'].value_counts().reset_index()\n", "study.head()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
methodcount
0Take online courses; Buy books and work throug...711
1Take online courses551
2None of these523
3Take online courses; Part-time/evening courses...479
4Take online courses; Bootcamp; Part-time/eveni...465
\n", "
" ], "text/plain": [ " method count\n", "0 Take online courses; Buy books and work throug... 711\n", "1 Take online courses 551\n", "2 None of these 523\n", "3 Take online courses; Part-time/evening courses... 479\n", "4 Take online courses; Bootcamp; Part-time/eveni... 465" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Oh this isn't what I was expecting, it is grouping items together if a participant provided \n", "# more than just one answer. Let's see if we can clean this up.\n", "# first to change this index column to a more appropriate name\n", "study.rename(columns={'index': 'method', 'CousinEducation': 'count'}, inplace=True)\n", "study.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A quick look through data, allows us to create a list of all of the individual methods marked by a user." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Here is a list of the different answers provided\n", "possible_vals = [\"Take online courses\", \"Buy books and work through the exercises\", \n", " \"None of these\", \"Part-time/evening courses\", \"Return to college\",\n", " \"Contribute to open source\", \"Conferences/meet-ups\", \"Bootcamp\",\n", " \"Get a job as a QA tester\", \"Participate in online coding competitions\",\n", " \"Master's degree\", \"Participate in hackathons\", \"Other\"]" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": true }, "outputs": [], "source": [ "#Now we want to see how often each of these individual values appears - I wrote \n", "# this function to assist with process - it isn't the best solution, but it gets\n", "# the job done and our dataset isn't large enough to computationally hurt us too much.\n", "\n", "def total_count(df, col1, col2, look_for):\n", " '''\n", " INPUT:\n", " df - the pandas dataframe you want to search\n", " col1 - the column name you want to look through\n", " col2 - the column you want to count values from\n", " look_for - a list of strings you want to search for in each row of df[col]\n", " \n", " OUTPUT:\n", " new_df - a dataframe of each look_for with the count of how often it shows up \n", " '''\n", " new_df = defaultdict(int)\n", " for val in look_for:\n", " for idx in range(df.shape[0]):\n", " if val in df[col1][idx]:\n", " new_df[val] += int(df[col2][idx]) \n", " new_df = pd.DataFrame(pd.Series(new_df)).reset_index()\n", " new_df.columns = [col1, col2]\n", " new_df.sort_values('count', ascending=False, inplace=True)\n", " return new_df" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
methodcount
12Take online courses15246
1Buy books and work through the exercises11750
8Part-time/evening courses7517
3Contribute to open source7423
0Bootcamp5276
2Conferences/meet-ups5244
11Return to college5017
10Participate in online coding competitions3610
4Get a job as a QA tester3376
9Participate in hackathons2747
5Master's degree2639
7Other2348
6None of these604
\n", "
" ], "text/plain": [ " method count\n", "12 Take online courses 15246\n", "1 Buy books and work through the exercises 11750\n", "8 Part-time/evening courses 7517\n", "3 Contribute to open source 7423\n", "0 Bootcamp 5276\n", "2 Conferences/meet-ups 5244\n", "11 Return to college 5017\n", "10 Participate in online coding competitions 3610\n", "4 Get a job as a QA tester 3376\n", "9 Participate in hackathons 2747\n", "5 Master's degree 2639\n", "7 Other 2348\n", "6 None of these 604" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Now we can use our function and take a look at the results\n", "# Looks like good news for Udacity - most individuals think that you \n", "# should take online courses\n", "\n", "\n", "study_df = total_count(study, 'method', 'count', possible_vals)\n", "study_df" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
methodcountperc
12Take online courses152460.209432
1Buy books and work through the exercises117500.161408
8Part-time/evening courses75170.103260
3Contribute to open source74230.101968
0Bootcamp52760.072476
2Conferences/meet-ups52440.072036
11Return to college50170.068918
10Participate in online coding competitions36100.049590
4Get a job as a QA tester33760.046376
9Participate in hackathons27470.037735
5Master's degree26390.036251
7Other23480.032254
6None of these6040.008297
\n", "
" ], "text/plain": [ " method count perc\n", "12 Take online courses 15246 0.209432\n", "1 Buy books and work through the exercises 11750 0.161408\n", "8 Part-time/evening courses 7517 0.103260\n", "3 Contribute to open source 7423 0.101968\n", "0 Bootcamp 5276 0.072476\n", "2 Conferences/meet-ups 5244 0.072036\n", "11 Return to college 5017 0.068918\n", "10 Participate in online coding competitions 3610 0.049590\n", "4 Get a job as a QA tester 3376 0.046376\n", "9 Participate in hackathons 2747 0.037735\n", "5 Master's degree 2639 0.036251\n", "7 Other 2348 0.032254\n", "6 None of these 604 0.008297" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# We might also look at the percent\n", "\n", "study_df['perc'] = study_df['count']/np.sum(study_df['count'])\n", "study_df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We might want to take this one step further and say we care more about the methods that are suggested by those who earn more, or those who have higher job satisfaction. Let's take a stab at incorporating that into this analysis." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# let's rewrite part of this function to get the mean salary for each method\n", "\n", "def mean_amt(df, col_name, col_mean, look_for):\n", " '''\n", " INPUT:\n", " df - the pandas dataframe you want to search\n", " col_name - the column name you want to look through\n", " col_count - the column you want to count values from\n", " col_mean - the column you want the mean amount for\n", " look_for - a list of strings you want to search for in each row of df[col]\n", " \n", " OUTPUT:\n", " df_all - holds sum, square, total, mean, variance, and standard deviation for the col_mean\n", " '''\n", " new_df = defaultdict(int)\n", " squares_df = defaultdict(int)\n", " denoms = dict()\n", " for val in look_for:\n", " denoms[val] = 0\n", " for idx in range(df.shape[0]):\n", " if df[col_name].isnull()[idx] == False:\n", " if val in df[col_name][idx] and df[col_mean][idx] > 0:\n", " new_df[val] += df[col_mean][idx]\n", " squares_df[val] += df[col_mean][idx]**2 #Needed to understand the spread\n", " denoms[val] += 1 \n", " \n", " # Turn into dataframes\n", " new_df = pd.DataFrame(pd.Series(new_df)).reset_index()\n", " squares_df = pd.DataFrame(pd.Series(squares_df)).reset_index()\n", " denoms = pd.DataFrame(pd.Series(denoms)).reset_index()\n", " \n", " # Change the column names\n", " new_df.columns = [col_name, 'col_sum']\n", " squares_df.columns = [col_name, 'col_squares']\n", " denoms.columns = [col_name, 'col_total']\n", " \n", " # Merge dataframes\n", " df_means = pd.merge(new_df, denoms)\n", " df_all = pd.merge(df_means, squares_df)\n", " \n", " # Additional columns needed for analysis\n", " df_all['mean_col'] = df_means['col_sum']/df_means['col_total']\n", " df_all['var_col'] = df_all['col_squares']/df_all['col_total'] - df_all['mean_col']**2\n", " df_all['std_col'] = np.sqrt(df_all['var_col'])\n", " df_all['lower_95'] = df_all['mean_col'] - 1.96*df_all['std_col']/np.sqrt(df_all['col_total'])\n", " df_all['upper_95'] = df_all['mean_col'] + 1.96*df_all['std_col']/np.sqrt(df_all['col_total'])\n", " return df_all" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
CousinEducationcol_sumcol_totalcol_squaresmean_colvar_colstd_collower_95upper_95
3Contribute to open source1.392267e+0822531.239062e+1361796.1454951.680845e+0940998.10367160103.21454563489.076445
7Other4.491415e+077383.851360e+1260859.2816941.514792e+0938920.33107358051.23441163667.328977
5Master's degree4.284612e+077213.771773e+1259425.9692771.699862e+0941229.38760956416.46251762435.476037
11Return to college8.733691e+0714747.243713e+1259251.6361451.403567e+0937464.20878157339.03727661164.235014
0Bootcamp9.583229e+0716228.502989e+1259082.7949981.751510e+0941851.04391757046.04867161119.541325
9Participate in hackathons4.641498e+077964.044473e+1258310.2770601.680908e+0940998.87926455462.07023661158.483884
2Conferences/meet-ups9.699603e+0716778.366275e+1257839.0153071.643482e+0940539.88644655898.70108059779.329533
4Get a job as a QA tester5.852363e+0710325.017250e+1256708.9407711.645773e+0940568.12277154233.79298759184.088555
1Buy books and work through the exercises1.909928e+0833931.624985e+1356290.2322111.620639e+0940257.16056054935.64403357644.820390
12Take online courses2.415638e+0844932.011544e+1353764.4880121.586442e+0939830.16788352599.82576254929.150262
8Part-time/evening courses1.124542e+0821179.290785e+1253119.6131181.566963e+0939584.87914051433.35145854805.874778
6None of these5.377087e+061124.273189e+1148009.7063571.510415e+0938864.06365340811.98104155207.431672
10Participate in online coding competitions4.332660e+079313.326921e+1246537.7031611.407734e+0937519.78518144127.56702648947.839295
\n", "
" ], "text/plain": [ " CousinEducation col_sum col_total \\\n", "3 Contribute to open source 1.392267e+08 2253 \n", "7 Other 4.491415e+07 738 \n", "5 Master's degree 4.284612e+07 721 \n", "11 Return to college 8.733691e+07 1474 \n", "0 Bootcamp 9.583229e+07 1622 \n", "9 Participate in hackathons 4.641498e+07 796 \n", "2 Conferences/meet-ups 9.699603e+07 1677 \n", "4 Get a job as a QA tester 5.852363e+07 1032 \n", "1 Buy books and work through the exercises 1.909928e+08 3393 \n", "12 Take online courses 2.415638e+08 4493 \n", "8 Part-time/evening courses 1.124542e+08 2117 \n", "6 None of these 5.377087e+06 112 \n", "10 Participate in online coding competitions 4.332660e+07 931 \n", "\n", " col_squares mean_col var_col std_col lower_95 \\\n", "3 1.239062e+13 61796.145495 1.680845e+09 40998.103671 60103.214545 \n", "7 3.851360e+12 60859.281694 1.514792e+09 38920.331073 58051.234411 \n", "5 3.771773e+12 59425.969277 1.699862e+09 41229.387609 56416.462517 \n", "11 7.243713e+12 59251.636145 1.403567e+09 37464.208781 57339.037276 \n", "0 8.502989e+12 59082.794998 1.751510e+09 41851.043917 57046.048671 \n", "9 4.044473e+12 58310.277060 1.680908e+09 40998.879264 55462.070236 \n", "2 8.366275e+12 57839.015307 1.643482e+09 40539.886446 55898.701080 \n", "4 5.017250e+12 56708.940771 1.645773e+09 40568.122771 54233.792987 \n", "1 1.624985e+13 56290.232211 1.620639e+09 40257.160560 54935.644033 \n", "12 2.011544e+13 53764.488012 1.586442e+09 39830.167883 52599.825762 \n", "8 9.290785e+12 53119.613118 1.566963e+09 39584.879140 51433.351458 \n", "6 4.273189e+11 48009.706357 1.510415e+09 38864.063653 40811.981041 \n", "10 3.326921e+12 46537.703161 1.407734e+09 37519.785181 44127.567026 \n", "\n", " upper_95 \n", "3 63489.076445 \n", "7 63667.328977 \n", "5 62435.476037 \n", "11 61164.235014 \n", "0 61119.541325 \n", "9 61158.483884 \n", "2 59779.329533 \n", "4 59184.088555 \n", "1 57644.820390 \n", "12 54929.150262 \n", "8 54805.874778 \n", "6 55207.431672 \n", "10 48947.839295 " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_all = mean_amt(df, 'CousinEducation', 'Salary', possible_vals)\n", "\n", "# To get a simple answer to our questions - see these two tables.\n", "\n", "df_all.sort_values('mean_col', ascending=False)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
methodcountperc
12Take online courses152460.209432
1Buy books and work through the exercises117500.161408
8Part-time/evening courses75170.103260
3Contribute to open source74230.101968
0Bootcamp52760.072476
2Conferences/meet-ups52440.072036
11Return to college50170.068918
10Participate in online coding competitions36100.049590
4Get a job as a QA tester33760.046376
9Participate in hackathons27470.037735
5Master's degree26390.036251
7Other23480.032254
6None of these6040.008297
\n", "
" ], "text/plain": [ " method count perc\n", "12 Take online courses 15246 0.209432\n", "1 Buy books and work through the exercises 11750 0.161408\n", "8 Part-time/evening courses 7517 0.103260\n", "3 Contribute to open source 7423 0.101968\n", "0 Bootcamp 5276 0.072476\n", "2 Conferences/meet-ups 5244 0.072036\n", "11 Return to college 5017 0.068918\n", "10 Participate in online coding competitions 3610 0.049590\n", "4 Get a job as a QA tester 3376 0.046376\n", "9 Participate in hackathons 2747 0.037735\n", "5 Master's degree 2639 0.036251\n", "7 Other 2348 0.032254\n", "6 None of these 604 0.008297" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "study_df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Although we can see the mean salary is highest for the individuals who say that you should contribute to open source, you might be asking - is that really a significant difference? The salary differences don't see that large...\n", "\n", "By the Central Limit Theorem, we know that the mean of any set of data will follow a normal distribution with a standard deviation equal to the standard deviation of the original data divided by the square root of the sample size, as long as we collect a large enough sample size. With that in mind, we can consider two salaries significantly different if a second salary is two standard deviations or more away from the other.\n", "\n", "Using the lower and upper bound components, we can get an idea of the salaries that are significantly different from one another. " ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Quiz - perform a similar analysis looking at career and job satisfaction for this individuals\n", "# to determine which you want to be like" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "df_jobsat = mean_amt(df, 'CousinEducation', 'JobSatisfaction', possible_vals)" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
CousinEducationcol_sumcol_totalcol_squaresmean_colvar_colstd_collower_95upper_95
9Participate in hackathons14884.02073115166.07.1799324.0038042.0009517.0937957.266070
7Other13797.01937106521.07.1228704.2574892.0633687.0309807.214760
11Return to college27767.03904212971.07.1124493.9650701.9912487.0499857.174912
4Get a job as a QA tester21294.03000162716.07.0980003.8570631.9639417.0277217.168279
3Contribute to open source42374.05999324340.07.0635114.1724962.0426697.0118207.115202
2Conferences/meet-ups30868.04371236106.07.0620004.1446352.0358387.0016457.122354
0Bootcamp30404.04307231670.07.0592063.9567921.9891696.9997997.118613
1Buy books and work through the exercises66788.09492508944.07.0362414.1095172.0271946.9954597.077023
8Part-time/evening courses42797.06100324829.07.0159024.0277802.0069336.9655377.066266
12Take online courses85746.012222651056.07.0157094.0490092.0122156.9800357.051384
6None of these3000.043322898.06.9284064.8794012.2089376.7203437.136470
5Master's degree14459.02091108711.06.9148734.1744852.0431566.8272987.002448
10Participate in online coding competitions18184.02675135660.06.7977574.5045182.1223856.7173276.878187
\n", "
" ], "text/plain": [ " CousinEducation col_sum col_total \\\n", "9 Participate in hackathons 14884.0 2073 \n", "7 Other 13797.0 1937 \n", "11 Return to college 27767.0 3904 \n", "4 Get a job as a QA tester 21294.0 3000 \n", "3 Contribute to open source 42374.0 5999 \n", "2 Conferences/meet-ups 30868.0 4371 \n", "0 Bootcamp 30404.0 4307 \n", "1 Buy books and work through the exercises 66788.0 9492 \n", "8 Part-time/evening courses 42797.0 6100 \n", "12 Take online courses 85746.0 12222 \n", "6 None of these 3000.0 433 \n", "5 Master's degree 14459.0 2091 \n", "10 Participate in online coding competitions 18184.0 2675 \n", "\n", " col_squares mean_col var_col std_col lower_95 upper_95 \n", "9 115166.0 7.179932 4.003804 2.000951 7.093795 7.266070 \n", "7 106521.0 7.122870 4.257489 2.063368 7.030980 7.214760 \n", "11 212971.0 7.112449 3.965070 1.991248 7.049985 7.174912 \n", "4 162716.0 7.098000 3.857063 1.963941 7.027721 7.168279 \n", "3 324340.0 7.063511 4.172496 2.042669 7.011820 7.115202 \n", "2 236106.0 7.062000 4.144635 2.035838 7.001645 7.122354 \n", "0 231670.0 7.059206 3.956792 1.989169 6.999799 7.118613 \n", "1 508944.0 7.036241 4.109517 2.027194 6.995459 7.077023 \n", "8 324829.0 7.015902 4.027780 2.006933 6.965537 7.066266 \n", "12 651056.0 7.015709 4.049009 2.012215 6.980035 7.051384 \n", "6 22898.0 6.928406 4.879401 2.208937 6.720343 7.136470 \n", "5 108711.0 6.914873 4.174485 2.043156 6.827298 7.002448 \n", "10 135660.0 6.797757 4.504518 2.122385 6.717327 6.878187 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_jobsat.sort_values('mean_col', ascending=False)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
01234567891011121314151617
0Bootcamp3040443072316707.059213.956791.989176.99987.11861Bootcamp9.58323e+0716228.50299e+1259082.81.75151e+09418515704661119.5
1Buy books and work through the exercises6678894925089447.036244.109522.027196.995467.07702Buy books and work through the exercises1.90993e+0833931.62499e+1356290.21.62064e+0940257.254935.657644.8
2Conferences/meet-ups3086843712361067.0624.144632.035847.001657.12235Conferences/meet-ups9.6996e+0716778.36627e+12578391.64348e+0940539.955898.759779.3
3Contribute to open source4237459993243407.063514.17252.042677.011827.1152Contribute to open source1.39227e+0822531.23906e+1361796.11.68084e+0940998.160103.263489.1
4Get a job as a QA tester2129430001627167.0983.857061.963947.027727.16828Get a job as a QA tester5.85236e+0710325.01725e+1256708.91.64577e+0940568.154233.859184.1
5Master's degree1445920911087116.914874.174482.043166.82737.00245Master's degree4.28461e+077213.77177e+12594261.69986e+0941229.456416.562435.5
6None of these3000433228986.928414.87942.208946.720347.13647None of these5.37709e+061124.27319e+1148009.71.51042e+0938864.14081255207.4
7Other1379719371065217.122874.257492.063377.030987.21476Other4.49141e+077383.85136e+1260859.31.51479e+0938920.358051.263667.3
8Part-time/evening courses4279761003248297.01594.027782.006936.965547.06627Part-time/evening courses1.12454e+0821179.29078e+1253119.61.56696e+0939584.951433.454805.9
9Participate in hackathons1488420731151667.179934.00382.000957.093797.26607Participate in hackathons4.6415e+077964.04447e+1258310.31.68091e+0940998.955462.161158.5
10Participate in online coding competitions1818426751356606.797764.504522.122396.717336.87819Participate in online coding competitions4.33266e+079313.32692e+1246537.71.40773e+0937519.844127.648947.8
11Return to college2776739042129717.112453.965071.991257.049997.17491Return to college8.73369e+0714747.24371e+1259251.61.40357e+0937464.25733961164.2
12Take online courses85746122226510567.015714.049012.012216.980037.05138Take online courses2.41564e+0844932.01154e+1353764.51.58644e+0939830.252599.854929.2
\n", "
" ], "text/plain": [ " 0 1 2 3 4 \\\n", "0 Bootcamp 30404 4307 231670 7.05921 \n", "1 Buy books and work through the exercises 66788 9492 508944 7.03624 \n", "2 Conferences/meet-ups 30868 4371 236106 7.062 \n", "3 Contribute to open source 42374 5999 324340 7.06351 \n", "4 Get a job as a QA tester 21294 3000 162716 7.098 \n", "5 Master's degree 14459 2091 108711 6.91487 \n", "6 None of these 3000 433 22898 6.92841 \n", "7 Other 13797 1937 106521 7.12287 \n", "8 Part-time/evening courses 42797 6100 324829 7.0159 \n", "9 Participate in hackathons 14884 2073 115166 7.17993 \n", "10 Participate in online coding competitions 18184 2675 135660 6.79776 \n", "11 Return to college 27767 3904 212971 7.11245 \n", "12 Take online courses 85746 12222 651056 7.01571 \n", "\n", " 5 6 7 8 \\\n", "0 3.95679 1.98917 6.9998 7.11861 \n", "1 4.10952 2.02719 6.99546 7.07702 \n", "2 4.14463 2.03584 7.00165 7.12235 \n", "3 4.1725 2.04267 7.01182 7.1152 \n", "4 3.85706 1.96394 7.02772 7.16828 \n", "5 4.17448 2.04316 6.8273 7.00245 \n", "6 4.8794 2.20894 6.72034 7.13647 \n", "7 4.25749 2.06337 7.03098 7.21476 \n", "8 4.02778 2.00693 6.96554 7.06627 \n", "9 4.0038 2.00095 7.09379 7.26607 \n", "10 4.50452 2.12239 6.71733 6.87819 \n", "11 3.96507 1.99125 7.04999 7.17491 \n", "12 4.04901 2.01221 6.98003 7.05138 \n", "\n", " 9 10 11 12 \\\n", "0 Bootcamp 9.58323e+07 1622 8.50299e+12 \n", "1 Buy books and work through the exercises 1.90993e+08 3393 1.62499e+13 \n", "2 Conferences/meet-ups 9.6996e+07 1677 8.36627e+12 \n", "3 Contribute to open source 1.39227e+08 2253 1.23906e+13 \n", "4 Get a job as a QA tester 5.85236e+07 1032 5.01725e+12 \n", "5 Master's degree 4.28461e+07 721 3.77177e+12 \n", "6 None of these 5.37709e+06 112 4.27319e+11 \n", "7 Other 4.49141e+07 738 3.85136e+12 \n", "8 Part-time/evening courses 1.12454e+08 2117 9.29078e+12 \n", "9 Participate in hackathons 4.6415e+07 796 4.04447e+12 \n", "10 Participate in online coding competitions 4.33266e+07 931 3.32692e+12 \n", "11 Return to college 8.73369e+07 1474 7.24371e+12 \n", "12 Take online courses 2.41564e+08 4493 2.01154e+13 \n", "\n", " 13 14 15 16 17 \n", "0 59082.8 1.75151e+09 41851 57046 61119.5 \n", "1 56290.2 1.62064e+09 40257.2 54935.6 57644.8 \n", "2 57839 1.64348e+09 40539.9 55898.7 59779.3 \n", "3 61796.1 1.68084e+09 40998.1 60103.2 63489.1 \n", "4 56708.9 1.64577e+09 40568.1 54233.8 59184.1 \n", "5 59426 1.69986e+09 41229.4 56416.5 62435.5 \n", "6 48009.7 1.51042e+09 38864.1 40812 55207.4 \n", "7 60859.3 1.51479e+09 38920.3 58051.2 63667.3 \n", "8 53119.6 1.56696e+09 39584.9 51433.4 54805.9 \n", "9 58310.3 1.68091e+09 40998.9 55462.1 61158.5 \n", "10 46537.7 1.40773e+09 37519.8 44127.6 48947.8 \n", "11 59251.6 1.40357e+09 37464.2 57339 61164.2 \n", "12 53764.5 1.58644e+09 39830.2 52599.8 54929.2 " ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pd.DataFrame(np.hstack([df_jobsat, df_all]))" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "collapsed": true }, "outputs": [], "source": [ "pd.DataFrame?" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 4307\n", "1 9492\n", "2 4371\n", "3 5999\n", "4 3000\n", "5 2091\n", "6 433\n", "7 1937\n", "8 6100\n", "9 2073\n", "10 2675\n", "11 3904\n", "12 12222\n", "Name: col_total, dtype: int64" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_jobsat.col_total" ] }, { "cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [], "source": [ "df_dotplot = pd.DataFrame(np.array(['Method', \"At least Master's\", \"Less Than Master's\", \n", "\"Master's Degree\",0.0589517,0.0293459,\n", "\"Bootcamp\",0.0746172,0.071824,\n", "\"Become QA Tester\",0.0484688,0.0457388,\n", "\"Buy Books\",0.162073,0.161205,\n", "\"None of these\",0.00836278,0.00827705,\n", "\"Part Time Courses\",0.103298,0.103248,\n", "\"Return to College\",0.0687279,0.0689754,\n", "\"Online Courses\",0.207892,0.2099,\n", "\"Contribute to Opensource\",0.097821,0.10323,\n", "\"Coding Competitions\",0.0453475,0.0508806,\n", "\"Other\",0.0269729,0.0338607,\n", "\"Hackathons\", 0.0316254,0.0395937,\n", "\"Conferences\", 0.0658422, 0.0739201]).reshape((14, 3)))" ] }, { "cell_type": "code", "execution_count": 71, "metadata": {}, "outputs": [], "source": [ "df_dotplot.columns = df_dotplot.iloc[0]" ] }, { "cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [], "source": [ "df_dotplot.drop(0, inplace=True)" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "collapsed": true }, "outputs": [], "source": [ "import seaborn as sns" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "df_dotplot.prop = df_dotplot.prop.astype(float)" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [], "source": [ "df_dotplot = df_dotplot.melt(id_vars='Method', value_name='prop', var_name='status')" ] }, { "cell_type": "code", "execution_count": 69, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFXCAYAAAC7nNf0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xd4VEXbwOHflvROSEghjRJKqAF9UREFFMVXUV5pKoKA\n9FCU3nsTEZRIlS4ifIACCoqAVEFpoQgCAUJLQkklbfv3R8JKSGETUiA896UXm3PmzJmzhH12zpmZ\nR2EymUwIIYQQ4omnLO0GCCGEEKJoSFAXQgghyggJ6kIIIUQZIUFdCCGEKCMkqAshhBBlhAR1IYQQ\nooxQl3YDHtXt23dLuwlCCCFEifHwcMpzn/TUhRBCiDJCgroQQghRRkhQF0IIIcqIJ/6ZuhBFzWg0\nYDQaS7sZooxQKpUolarSboZ4SkhPXYj7ZGSkodfrSrsZogzR63VkZKSVdjPEU0J66kJkMRoNKJVK\nrK1tS7spokyxQqvNyPr9kh67KF7SUxcii9FolA9dUSyUSpU80hElQoK6EEIIUUZIUBfiKXbs2BEa\nN27IiRMR5m2nTp2gceOGxMRE5yiv0Wj49detAEyZMp7Tp08V+Jzvv/9u4RsshMiXBHUhnnIVK/qz\nd+/v5p93796Fl5d3rmXj4+P49ddtJdU0IUQByUA5IZ5y9euHcvLkcfPP165dwd8/kKSkRObMmUla\nWhr29vaMGjWe775bxdmzf/PDD+sBWLNmJcnJySiVKqZO/Qy9Xs/EiWPRaDJQqdQMHz4ab28fvvpq\nFqdPnyIoqFJpXaYQTwUJ6kI85ZRKJZUrV+XChXPo9Xpq1Ajh1KmTrFq1jBYtWtK8eQt27drB6tUr\nef/9D7lx4zqtW7fhzJnThIY+w7vvtuPzz6dx7NgRIiKO8dprb9CixescPvwnCxbMpWPHLsTERLNo\n0XIiIy9w6tSJ0r7kp5LBZOJwUiK74uO4o9XioFLxnKsbL5dzx14lA0TLCgnqQghefrk5e/fuRqvV\n0rLlm5w6dRKNRsO6dWv44Yf1GAwGKlb0y3FctWrVAXBzK4dGo+HKlSjatOkAQJ06dfn66y+5ejWK\nqlWrAVClSlVsbGxK7sIEAOkGA9MuX+R0SvYEWH+npvDjrVjGVq5KRVu7UmqdKEoS1EWpiNdpuas3\n4GZlhbNafg1LW8OGz7Jy5VJcXFwJDAwCwMrKiq5de/Cf/zzHuXP/cP36VRQKBSbT/VOzFNnq8fML\n4PTpk3h7+3DyZAQ+Pr74+vrxyy8/AxAVdRmtVltSlyWyfH3tSo6Afs8dnY5JFyP5qkYINkoZZvWk\nk09TUaIi7ibzf7ExnE1NATJDQgNnFzp4+VDJ3r50G/cUU6vVeHv7ZOuNf/hhF5YvX8KqVcvQ6/UM\nGzYaNzc3EhISWLPm21zr6dSpC1OnTuDHHzegUCgYPnwMFSv6UaVKMN27dyYgIBA7O/l7toTJZMII\nGO//M+u1Ids+E0YT//6Ztc2Qte2OVssfiQn5nuu2TsuBhHiauZcvgSsTxUlhMplMpd2IRyH51J8c\nv8fHEX41itx+4awVCsZUrkqIY955govbveVh1WqrUmvDkyrXAMQDQacAgSgzgPFAuZx1Gwp7PkvO\ncd82w4PnyKVuwwPXkv36jFk/P/ya7r0u6Q/mZ51dGF6pSgmfVRRGfvnUpacuSkSiTseCa1fy/KDS\nmkzMvnKZBTVro1Yo8ihVdHL7ENUa9BhNoMwRbHILOrkFhvs+/C340LYkyD0YdAwWBruHB0DLzpdX\nkMstgIsnW7qseFcmSFAXJWJn/B10D7kpFK/TMeL8P7io1TkDqQW9LUMBApEQZZUSUCoU5j+NJtCa\nHv5b7yUDGMsECeqiRESmWZal6mK6ZLN6mjwYgP79U4FSkce2bH/mvk11//4H687vfFl/qgp4vvzO\noUKByWhAqQArlTr3a3lgW67tt/B8D9IYjXT/+yQpBkO+fxfNy8nz9LJAgrooEWVlTK2SrA98CwLD\nwwKRqgCBIa9gp8qx7SFBJ4/zFcU1FTgQAYoSeNTyOCjN8Ro2SiUf+VQk/NqVPMs0LedOsINDCbZK\nFBcJ6qJE1HR04mBS4kPLNS/njp+tXa7BQFXYHp2FvSCjQY8SBdZqK4t7QUI8CZq5l0elULAi+jqJ\ner15u7VCwRsennzg7VuKrRNFSUa/ixKRqtfT88xp0ox53wL0tbHly+o1Sy14Fqg3pdFife4iVhej\nUKSlY7KzRVfJH131qphs5dmkyO5xmVmhN5mISE7itlaLg1pNAydnHGSdiCdOfqPfy8pdUfGYc1Cr\nGRQYhHUeAdtZpWZwYKUnojesSE7BcdOv2B45gSohCaVGiyoxGdtjp3H48ReUiUmFrvte1rQdO37N\ntr1z5w5MmTK+wPVt2LDW4rJTpoxn69YtDy23ZMlCOnT4H2FhPejT52M++aQv58//U+C2iZKnViho\n6OJKSw9PmriVk4BeBklQFyWmvrML04Or85JbOayygre9UknL8h7MrFaDALsnYJlKkwn7nftRpqTm\n3AUo09Kx27EPHmF6UEBAIDt3bjf/fPFiJOnp6YWqa8WKpRaXLVfOnfLlPSwq26HD+4SHL2LevG8Y\nOHAI48ePQqPRFKqNQoiiI1/TRIkKtLNnQEAQYf6BaIxGbJXKJ6J3fo8q+iaqhNzHBty7ClVyCuqr\nN9AH5lwr3RJVqlTl6tUrpKSk4OjoyK+/bqVFi5bcvBkLZPa+9+z5nfT0dFxdXZk69XNiYqKZNm0C\nKpUao9HIuHGT+eWXn0lOTuLzz6czcOBgZs6cyvXr1zAajXTv3pvQ0IZ8+GE7/PwCsLJSM3jwSGxt\nbTl5MoLw8Dmo1WpsbW2ZPHkG9vZ5D6IKCAgkOLg6J09GUKNGCNOnTyQpKfNuxcCBQ6hcuQo//fQj\nGzasw9nZBbXaiubNXwXg5583YzQa6datJ8nJyaxduxqlUkmdOvXo3bsfKSkpudYnhMidBHVRKlQK\nxROZGUp9PcbicoUN6gAvvdSMPXt28cYbb3H27N988EFnbt6MxWg0kpSUxJw581AqlXz6aRhnz/7N\nhQvnqVEjhD59BnDixHFSU1Po3LkbGzasY/Dg4fzww3pcXFwZMWIsSUmJ9O3bg2+/XUd6ejoffdSN\n4ODq5nPv27eHZs1eoV2799m/fy/JyXfzDeoA5cqVIykpkZUrl9KgwbO0bt2Ga9euMnXqBKZNm8W3\n365k+fLvsLKyon//XubjnJycmD79C5KTk+jT52O++WYVtra2TJo0hsOHD3H48F856ps/f0mh31ch\nyrpiC+pGo5Hx48dz7tw5rK2tmTx5MgEBAeb9P/30EytWrEClUhEcHMz48eMB8j1GiNKmeMhcXzPD\noy1x8+qrrzNr1nR8fHypW7e+ebtSqcTKyorx40dhZ2fHrVu30Ov1vPnm26xevYJBg/rh4OBIz559\ns9V38WIkJ08e58yZ05nNM+hJTMy84+DvH5it7IcfdmHlyqUMGNAbDw9Patas9dD2xsbG8tJLzfnl\nl585duyI+fHB3bvJXL9+jaCgIGxtbQGoVauO+Th//8x/39evXyMxMYHBg/sDkJaWxo0b17l0KTJH\nfUKIvBVbUN+xYwdarZa1a9cSERHB9OnTmT9/PgAZGRnMmTOHLVu2YGdnx6effsrvv/+OwWDI8xgh\nHgcGVxeLyhndLCuXF1/fiqSnp7N+/ff07BlGdPQNACIjL7B3724WL15BRkYG3bp1BGD//j3UrVuf\nrl178Ntvv7B69QpGjhzHvcktAQGBeHp60qlTVzSaDFasWIqzszOQc6749u1beeONNwkLG8iqVcvY\nvHkjXbv2yLOtly5dJCrqEiEhtQgICKRFi5q0aPE6CQnxbNnyIxUr+nHlShQaTQZWVtacPfs3AQGB\nWefOHNbj7e2Lp2cF5syZh1qtZuvWLVStGsy1a1dz1CeEyFuxBfWjR4/y4osvAlCvXj1Onz5t3mdt\nbc3333+PXdbAKL1ej42NDfv27cvzGCEeB7rKAdgeiUChz7vHblIq0VUJfORzNW/+Kr/+uhV//wBz\nUK9Y0Q87Ozt69+4KgLt7ee7cuU1ISG0mTx7HihVLMBqN9Ov3KQCBgUFMnDiG4cPHMGPGZMLCepCa\nmkLr1m1R5pFms0aNWkyfPhk7OzsUCgVDh47KUeb7779jx47tqFQqVCo1kyfPQK1W06lTV6ZPn8Tm\nzRtJS0ula9ceuLq68sEHnenTpzvOzs5oNBrUajX6++ZLu7m50b79B4SF9cBgMODt7UOzZq/mWp8Q\nIm/FNk991KhRtGjRgpdeegmAl19+mR07dqB+YArFqlWr2LNnD4sXL2b06NEWHXM/vd6AWv3kPZsV\njx+tVktSUtpD5xKr/rmIzd4/867n+Qboa1Ur6uY9sfR6PatWLadLl48xmUz06tWNXr36Ur9+g9Ju\nWonR63W4uNhjbW1d2k0RZVyx9dQdHR1JTf132o/RaMwWnI1GIzNnzuTy5cvMnTsXhULx0GNyk5Ag\na4WLonFvgRAekvJFXyUIg0qFzdFTqJL/XfzI6OiAJrQWuipBoJe0Mf9SkpaWTqdO76FWW1GzZi1q\n1aqH/il6j/R6I3FxKaW++IwoG0ol9WpoaCi///47b7zxBhEREQQHB2fbP3bsWKytrZk3b575NuDD\njhHicaEP8kcf6IfqdjyK9HRMtrYYPMpBHre0n3Y9e/bNMXhPCFH0iu32+73R7+fPn8dkMjF16lTO\nnDlDWloatWrV4t1336Vhw4bmQTqdOnWiefPmOY6pXLlyvueRZWJFUXlclvIUZY/8bomilF9PXdZ+\nFyKLfPCK4iK/W6IoydrvQgghxFNAVpQTojA0KVhH7sLq8n4U6QmYbF3QBTyPLvgVTLbOpd06IcRT\nSm6/C5HF0lukirs3cdgxBWXq7Rz7jHZupL0yCqNL4fJTHzt2hLFjRxAYGARkTrMbPHh4tmVcHyY5\nOYlDhw7SosXrhWqDKHpy+10UJbn9LkRRMRmx3/NFrgHdBCjTE7Db/Tnkkzf+YRo0aEh4+CLCwxfx\n8ce9+OabBQU6PjLyAgcO7Cn0+YUQTy65/S5EAahiT6NKvJrrPnOWtruxqK8fRe//7COf7+7dZFxd\n3Th//h9mz56JSqXC2tqaoUNH4+XlxZo137JzZ+bKbnXr1qdPn/6sXLmUyMgLbNq0kdDQhsyYMRmd\nToetrS3jx08lISGOuXNnYzQaSUxMZPDg4dSuXZf27d+hVq06XLt2lQYNniE1NYWzZ//G3z+AMWMm\nMWXKeEwmE7du3SQ9PY3Royeal3sVQjweJKgLUQDqGycsKxd9otBB/ejRI4SF9UCn0xEZeZ5p02Yx\nY8YUhg8fTdWq1di3bzfh4V/QpUsPdu36jQULlqJSqRg1aigHDuyjU6eubNq0gbff/h/Dh39Kx44f\n0ajR8+zfv4cLF85x924yYWGfULlyFbZv/4WtW7dQu3ZdYmNj+PLLBZQvX56WLZuxaNFyPvlkKO3a\nvc3du5mPuXx9KzJ69AQOHtzPvHlfMmPG7EJdoxCieEhQF6IAFEatZQUNuoeXyUODBg2ZMGEaAFev\nRtGzZ1dMJhNVq2YuPVu3bigLFoRz5UoUISG1zasu1q1bj8uXL2bLqnb16hVzVrTGjTOXXz5xIoLl\ny7/BxsaGtLQ0HBwy06o6O7vg5eUFgJ2dHUFBlQBwcHBEq9UAEBr6DAC1atXlq6++KPQ1CiGKhzxT\nF6IADC6W5Ug3ulYskvO5ubkD4OPjS2TkBQAiIo7h5+dPQEAgZ86cRq/XYzKZiIg4jp9fAEqlEqPx\nXna2IM6e/RuA7du3sX7993z55Uy6devJ6NETqFy5ijmT24PZ2nJz7txZAE6dOkFQUP4LQwkhSp70\n1IUoAF3QC9ge+w6FQZNnGZNSha5Sk0Kf497td5VKRVpaKv36fUKVKlWZPfszTCYTKpWK4cPH4Otb\nkWbNXqF3726YTCbq1KlLkyYvc+fObS5dimTduu/o23cAM2dOZcWKJdja2jJ27CT0ej1jxgzDyckZ\nDw9PkpISLW7boUN/sH//HoxGIyNHjiv0NQohiodMaRMii6XTjqwif8fu0KI892c07Iy2etmbTjZl\nyniaN29Bo0bPl3ZTnjgypU0UpVJJ6CJEWaWr0hSTlR02EWtR3Y01bzc6eqKp0wZdpRdLsXVCiKeZ\n9NSFyFLg3pTJiCruEor0REy2zhjcq0iWNpEr6amLoiQ9dSGKg0KJoXyV0m6FEEKYSbdCCCGEKCMk\nqAshhBBlhNx+F6IQjLq7pF37mfTonRg08aisXbH1boqD/1sorV1Ku3lCiKeUDJQTIoulg5n0aTHE\nHx6KIT02xz6ljTvlnpmBlWNAsbRRPJlkoJwoSvkNlJOgLkQWSz54TSYjd/7ojf7upTzLqOx98Gi8\nFIVSVeA2FEXq1eKUlpbGwoXhnDlzGhsbW+ztHQgLG4i//79fYlavXsG6dd+xbt1mbGxssh1/+PAh\nVqxYCsDp0yfNS9iGhX1C9eo1LG7Hhg3rePfddkVwRSVDgrooShLUhbCAJR+8mjtHiT8y/KF1udYb\ni51XweerHzt2hE2bNpjXfv/rr0OsX/89n302p8B1FYdRo4YQGtqQd99tD8CFC+eZMGEUCxYsw9HR\nEYDOnTvQoMEzVKkSzBtvvJVnXa1avcbmzb8WuA16vZ62bVvxww9bC3cRpUCCuihKMqVNiCKiuX3Y\nsnJ3jhQqqD/oXupVgIsXI5kzZyYmkwkXFxdGjBiHg4MDs2d/xtmzf6PT6enWrQcvvvgyc+fO5uTJ\nCABeffV12rV7jylTxqNWq4mNjUGn09G8eQsOHNjLzZuxTJ/+Bb6+FVmwIJwTJ45jNBpp3/4DmjV7\nxdyWO3duc/36NaZMmWneVrVqMC+80IQ9e3bx3/+24tixI/j4VOSdd95l4sSx+Qb1Bx09ephvvpmP\nSqWmYkU/Bg8ewY0b15k2bSJqtRqTycSECVPZvPkHEhMTmD37M8LCPmHmzKlER9/AYDDQq1cYdevW\np2PHdvj5+WNra8u4cZMf+e9BiCeFBHUhCsBkaZY2S8vlIrfUqwAzZkxmxIixBAVV4qeffmT16hVU\nr16TpKREFi9eSXJyMmvXrkapVBETE82iRcsxGAz07t2NBg0ys6t5eXkzbNhoZs6cSkzMDT7//CuW\nLFnIgQN78fMLICbmBvPnL0Gj0dCzZxeeeeY/ODll9gpiY2Pw9vbJ0V5vb29iY2MA+OmnTbz11jv4\n+wdiZWXF33+fJiSkVo5jcrxdRiMzZ05lwYJluLq6smBBOL/+upXU1FRq1apDr15hREQcIzk5mU6d\nurJly4988slQ1q//Hnf38owcOY7ExET69evBqlXrSE1N4eOPe1G5sqwjIJ4uEtSFKAC1hQPgLC2X\nm9xSr/7441auXLnMrFnTATAY9FSs6M/Vq1cICcl8Lu3s7Ez37r357ruV1K1bD4VCgVqtJiSkNlFR\nmWMA7j2bd3R0IiAgEAAnJyc0Gi2XLkVy7tw/hIX1ADJvc8fGRuPklJnytUIFL2JionO099q1q/j7\nB5KcnMzBgwdISIhn/fq1pKamsHHjWouCenx8HPHx8YwePRQAjSYDKysrOnbszLffrmDQoH44OjrS\ns2dYtuMuXrzImTOnOHUqM8+9Tqc3536//zm/EE8LCepCFICdT3Punv8GkyEj70IKNXa+LYrkfPdS\nr0JmkBo9eiJeXl6cPBlBXNwd1Go1v/++E4CUlBTGjh3Ou++2Z+vWzbRv/wF6vZ7Tp0/SsuWbwB/5\nplcNCAikfv2GDBs2CqPRyPLl3+Dr+28KWQ8PT3x9/cyD1ObPn4vRaGTfvj18883HbN++lTfffJu+\nfQcAkJGRQdu2rUhISMDNze0h11kODw8PPvtsNvb2DuzduxsnJyf27NlNaGhDunXryS+//MyaNasY\nMmQkRqMxq80BVKxYkQ8+6ExGRgYrVy41P9u3JJWsEGWNBHUhCkBp5YhzjT4knf4izzLO1XqgsilX\n6HPklnrVxsaWQYNGMHnyWAwGAwqFguHDx+Dn58+RI3/Ru3c3DAYDXbp057nnXuD48aP07NkFnU5H\ns2avUK3aw0fPv/BCE44fP0qfPh+Tnp5GkyZNsbd3yFZm9OgJLFwYTvfunVEqldja2uLpWYFLlyLZ\nsmUTY8ZMNJe1tbXlpZeasWXLD3Tq1DXfc6tUKsLCPmHQoP6YTCYcHBwZM2Yi7u7lmTZtIlZWVhiN\nRgYMGIRSqaRiRT+mTBnP4MHDmTFjCmFhPUhNTeHdd9tLMBdPNRn9LkSWgoxQTo/dw93zyzCk3TBv\nU9l54VilE/a+rxZbGx9HKSkp3Lp1k0qVKpd2Ux5bMvpdFCWZ0iaEBQr6wWsyGdElncOoSUBp7YKV\na3UUioLPTRdlnwR1UZRkSpsQxUChUGLtavmCKUIIUdwkoYsQQghRRkhQF0IIIcoIuf0uRCHo9XeJ\njf2ZW7d3oNXGY2XliodHM7y93sLKSrK0CSFKhwyUEyKLpYOZMjKiOXV6KBpNzixt1tbu1Ar5DHv7\nwi18UtQJXYqqviVLFuLu7s4777QpVDuedjJQThSl/AbKye13IQrAZDJy5uz4XAM6gFYbx5mzYzGZ\nDIU+R4MGDQkPX0R4+CI+/rgX33yzoNB1FUd9QojHl9x+F6IAEpOOk5aWd9pVgIyMG8TF/0F596JN\n6BIW1oMhQ0YSEBDIjz+uJy4ujvLlPbh+/Rp9+w7IWnzmfRYvXpkj5Wlu9Z0//w+zZ89EpVJhbW3N\n0KGj8fLyYs2ab9m5czsqlYq6devTp09/8/HXr19jwoRRDBs2hrS0VMLD56BWq7G1tWXy5Bk5FqsR\nQpQsCepCFEBCwl8Wljtc6KCeV0KX3Lz66mt07dqRXr3C+PPPg4SGNswR0PNOEDOF4cNHU7VqNfbt\n2014+Bd06dKDXbt+Y8GCpahUKkaNGsqBA/sAuHr1Cj//vJmxYyfj5+fP119/SbNmr9Cu3fvs37+X\n5OS7EtSFKGUS1IUoAKOF2dcsLZebvBK63O/eSBh7ewfq1Qvlr78OsnXrZj76qLvF9d25c5uqVTOT\ntdStG8qCBeFcuRJFSEht1Gp11vZ6XL58EYBDh/5ApVKhVGY+tfvwwy6sXLmUAQN64+HhSc2aD0/c\nIoQoXvJMXYgCsLcPtKicg4XlHub+hC7W1jbExd0BMm+d3/PWW63ZsmUTCQkJVKlS1eL6ypf3IDLy\nAgAREcfw8/MnICCQM2dOo9frMZlMREQcx88vc9Bfu3bv0b//p0yZMh6DwcD27Vt54403mTt3IUFB\nldi8eWORXLMQovCkpy5EAXh6NCcqajFGY95Z2hQKNZ6ehc/SlldCl7Zt2zNr1nQqVPCifHkPc/mQ\nkFrcuHGN1q3bFqi+YcNGMXv2Z5hMJlQqFcOHj8HXtyLNmr1C797dMJlM1KlTlyZNXiYy8jwAzzzT\niN9/38nq1Sto0OBZpk+fjJ2dHQqFgqFDRxX6moUQRUOmtAmRxdJpR7E3txEZmfdz7kpBffHxaV2k\nbcuP0Wikd+9ufPHFXBwcHEvsvMJyMqVNFCWZ0iZEEfKq0JLq1cZga+ubbbuNjRfBVYeVaECPjr5B\n164dad68hQR0IYT01IW4pzBZ2lJSzplXlHNykixtInfSUxdFSbK0icdGRsZNomM2cufObnS6ZGxs\nyuPp0QJv71ZP3PKqCoUSJyfJ0iaEeHxIT12UmOS7Z/j77xEYDKk59tnYVKB2rc+xtfUuhZZlkt6U\nKC6Pw++WyWQgLm4/N2/9ikZzC7XaEXf3JlTwbIFaLY9uniT59dQlqIsSYTCkc+Toh+h0iXmWcXCo\nSr2681AoFCXYsn89Dh+8omwq7d8tvT6VM2fHkJx8Msc+a+vyhIRML7JpmKL4ye13Uepu3/4934AO\nkJp6gctRC7Gzq0hmWFdCVoBXPPgaRdZ/iqzXimyvM/fkvu/+us2vUWA0KbC3CwJ099X9oMxtGYYU\n/o7/hXOJv5OqT8BO7UKwy0vULvcGdmqXzPPnODyvLyu5n8ey47NvK60vROLxdiFyVq4BHUCrvcOZ\nMyMJrb8Ulcq2hFsmipoEdVEiEhOPWlQuOnp9MbckbyqVGzVrTEanz79csu4O26K/IkUfZ96WYUjm\nz1vfcipuCy19+uNq7VWoNpw8cYYJE75gwcIZeHhkLhSzdOn3+FX05tUWL1lYiyKXVzl/ymvb5s2/\nsmXLr3Ts2JaXX34eAK1Wx86d+2jZsjmrVq3Dzc2VN998Lfe6FPdvKdgXmdxfWtLuXMoo8jrawjbl\nenwBryfzPwwGA7fv7EWp0P9bVqHI+oKa2+usL6Eosr6o5fUFNbdymeM97r3WauOJi9ubR7szaTS3\nuH3nd7wqtMy3nHj8SVAXJcJoekikfEKYTEZ2xi7KFtDvl2ZI4rfYBbzrNwZlIUfCW1lZ8cWshUyd\nNqKQPW9TLq9y/pSXAwf+ZMTIfgQF+Zv/3uLi7rDtlx289noTTCYjJgwYTXkshftEP9ArHkYD3Lix\nFoMhobSbkqf4+D8kqJcBEtRFiXBwqER8/IGHlrOz88fKypXMyGAic8iH5a8hM/BmMuX5GpPxXmlz\nHUqlMyhUefTFMkvfSD9HvPZGvteQrLvN1dRTBDrWe+j15qZuvZqYjCa2bPmNVq2yr0y3YcPP7Nl9\nCJVKSa3a1enW7T2+XbWB2Ju3SUpM4uatO/Ts8SENGtbh5MmzrFi+DqVKibd3Bfr372pe0x3gZuxt\nZs9ehMFgAIWC3r078c8/F4mMjGLO7MWMGNkPLy9PAL7//keuXr3B6tWZS8EeOniUffv+4m7yXT7s\n1JZGjULZt/dPNm7cilKpJKRWNbp27cDff59j8eLVqFVqbGysGTV6ANbWVsz9ainR0bEYjSY6d25L\nnbo1C/VeiaJjMOS9SqJ4ckhQFyWiQoXXuXZtNWDMs4xa7UL9egtQKq1LrmH3sWQw082kHRbVdVN3\ng1oO7e7bklv39YF+tAlsbRNRqxz5ZPAQevbsyouN38RK7YqNjSexMUYO7I9g4cIVqFQqRo8ewYmI\naKysXLGoqDULAAAgAElEQVS30zN2zAwOH/6Tdeu+p3Hjt5j71XDCw+fj5ubKkiWL2bP7b956q5X5\n3MuWLaJdu440bvwiFy5c4LOZM1i0cDF79xzl008HERAQYC7buXMvrly5xcfd+rNs2TI8PSsydOhQ\njh8/zvfff0+D0JdYvXoTCxYuwNbGlqlTp3L6VDRHjvxDs6YtaNPmXf744w90Wgf27T1EuXJejBgx\nlqSkJAYM+ITly5cCpsx3JNvb8uD7lv/7aMr24uHveZ7HY8qlqGX1mb8s3rfLgBFX14aYTKlgMmVd\nadb/9/9sfk3WF897lRiz9t2r+74vpQ/Ul+1LrsmE3pCCVpv7naX72T2wmJJ4MklQFyXC1qYCQUE9\nuXx5fh4llFStMqjUArqlDHndcs6lXPZb5w+/ja5QgEKhQqFQ4ObmTv/+g5g2bRK1a9dFoVBx7doN\nQkLqYG1tB0C9eqFERV1BoVARHFwTlcoab28/tFodd++mERcXx/jxYwHQaDQ888x/sg2EunLlGvXr\nN0KlsqN69TrcvnUbtdoehUKJWm2HWm1vLvvvdgeUSitq1KiDWu2Ip6cfGo2O2Nh4kpKSGDE8c/33\ntLQ0bt6Mo3PnXqxcuZTBg4fh4eFJ7drPEhV1g5Mnj/PPP5nJZIxGE6mpJlxdXS16b59Eer2OqlU+\nLZXR7waDhsNHOqDX5z9TqEKF10uoRaI4yTKxosT4+rxLteCR2Nn5Zdvu5FidWiEzcHd/vpRaZjl3\n20CLypWzsFx+Gjdugp9fAFu3/gSQbwa1Bx+9u7i44unpyfTpXxAevojOnbvSoMEz2coEBgZy8uRx\nAC5cOEe5cu7kRaFQ3vf4Iuf5vL198fSswJw58wgPX0SbNu0JCamdaya3gIBAXnnlNcLDFzFr1lc0\nbfoKzs7OhX2bxEOoVDZUCuqTb5kKni1xcqpeQi0SxUl66qJEeXg0o3z5pqSmXkSvT8baujz29v6l\n3SyLVXNrzr6YxejzydKmVKip6Vb4LG33GzBgEEePHgagcuUq+WZQy9YGpZIBAwYzZMgATCYT9vYO\njBkzIVuZvn0HMmPGZNas+Ra9Xs+IEWPybIebmxs6nZ55877CxsYm1/3t239AWFgPDAYD3t4+NGv2\nKlqtLkcmt/LlPZgxYzJhYT1ITU2hdeu25hztonh4er6KQqHictQitNo75u1KpS0+Pq0J8P+o9Bon\nipQsPiNEFksXCDkdt40d1/PO0vaST1/qe5RcUhfx+CvtxWfuMZkMJCYeJUNzC7XKATe3Z1GrHUq1\nTaLgZPEZIYpQLfeWWKvs+SNmKYn3jYR3tvbiuQqdqVHu1VJsnRB5UyhUuLk9W9rNEMVIeupCZClM\nlrbYtHOk6eOxU7viZV+90HPTRdn2uPTURdkgPXUhioFCocTbQbK0CSEeHzI6RQghhCgjJKgLIYQQ\nZYTcfheiEFIM6fyaEMHupL+J16fgqrLnRZeavOEWivN9i7YIIURJKraBckajkfHjx3Pu3Dmsra2Z\nPHnyfctOZkpPT6dLly5MmTKFypUrA9C6dWscHR0BqFixItOmTcv3PDJQThQVSwczxWoTGH1lDbd0\nSTn2lVM7MingPfxsyheqDceOHWHs2BEEBgahUCjQaDS0aPE6bdp0sOj4ixcjuXs3mXr1Qs3bfvnl\nZ376aRNarZaoqMsEB1cDYNy4ycyePZOpU2cWqq0PSk5O5uuv53DjxnX0ej0VKlRgyJBR5n/PTzMZ\nKCeKUqkMlNuxYwdarZa1a9cSERHB9OnTmT//3yVCT506xbhx47h586Z5m0ajwWQysWrVquJqlhCP\nxGgyMfXaxlwDOkC8PoXJ19Yzr3IPVIrCPd1q0KAhEyZkfpnVarW8//67vPbaf3Fyyvsf8j27d+/E\n3d09W1B//fX/8vrr/yUmJppx40YSHr7IvK+oAjrA+PGjePvt//HSS00BWLt2NTNnTjFfixCi+BVb\nUD969CgvvvgiAPXq1eP06dPZ9mu1Wr7++muGDh1q3vbPP/+Qnp5O165d0ev1fPrpp9SrV7hMV0IU\nhxOpUURpbuVbJkabwJ93L/C8c7VHPl9aWhpKpRKVSsXx40dZtmwxRqOR9PR0xo2bjJWVFcOGfYKz\nswuhoQ3Ztu0n1GorgoOrU7NmrYfW36rVa2ze/CthYT2oUiWYy5cvYmdnR5069fnrr4OkpKTwxRfh\n2NvbM3PmVK5fv4bRaKR7996EhjY01xMbG0N8fJw5oAO0adOB9PR0ALZv38a6dWuwsrLCz8+foUNH\nsX37Nq5ciaJ3735oNBo++KAN69dvISysB25u5UhOTmbQoKFMmzYRlUqN0Whk3LjJVKjgxYIF4Zw4\ncRyj0Uj79h/QrNkrbNz4f2zb9hNKpZIaNWoycOCQR37/hXjSFFtQT0lJyXbbTaVSodfrzakfGzRo\nkOMYW1tbunXrRtu2bYmKiqJ79+788ssv2dJFPsjNzR61WuYGi0en1WpJSkpDrc67h3087ZJFdUWk\nXaJJuYJPd1OplBw7doR+/XqiVCpQq9UMHjwMZ2dHrly5zIQJU/Dw8GD58iXs2bOT1157g/j4OFas\n+A4rKysUCnB3L0+dOnVyrVuhIMf1qdVKFAoFtWrVYvDgoQwc2Bd7ezvCwxcwceJYTp06zp07d3Bz\nc2PMmPEkJSXSq9fHrFmz3lxHQsIdfH19s9WtViuxsbEiKSmRpUsXsmLFGhwcHJgz53O2bNmInZ19\n1jUqMRiU2dry2muv8/LLzVi/fi0hIbUICxtARMRxMjLS+Ouvg8TGRrN48TI0Gg0ff9yZ5557jm3b\ntjBkyAhq1gxhw4b/A4z5fnaULCUuLvZYWz/eCYvEk6/YfuMdHR1JTU01/2w0PvwfWFBQEAEBASgU\nCoKCgnB1deX27dt4e3vneUxCQlqRtVk83e4998wvPazGoMtzX/ZyevT6vOvJi8FgJDS0YY5b1nq9\nEXf38syaNQM7O3tu375F7dp1MRiMeHv7oFCo0OuNGI0mjEZjruc2GIyYTOTYp9cbMZlMVKlSDb3e\niIODI/7+gVmvnUhPz+DChQucPHmc06dPZR2j586deHNmtfLlK3Dz5s1sdev1enbt+g0/P38CAyth\nY2OHXm+kdu36HD58iJo1a2E0mtDrjej1hmxt8fX1R6830rJlK1avXsGAAWE4ODjSs2dfLlw4zz//\nnKVXr48B0On0XL9+nREjxrJmzbfExEQTElI7qy0F/zsoDnq9kbi4FHmmLopEfs/Ui21KW2hoKHv3\n7gUgIiKC4ODghx6zfv16pk+fDsDNmzdJSUnBw8OjuJooRIH521j2++hfyIFy+ZkxYwojR45j1Kjx\nlC//bzsU9z27VyqVGI2FG/uqeDD12n0ellnNw8MTFxdX9u3bbd62bt0a9u3bg7e3L1FRl8234iMi\njuHn54+1tTVxcZnJRc6d+yfb+e4leNm/fw9169bnyy/n07Rpc1avXkFAQCD16zckPHwRX321gGbN\nXsHXtyKbN//I4MEjCA9fxIUL5zh16kSh3gchnmTF1lN/9dVXOXDgAB06dMBkMjF16lS2bNlCWloa\n7du3z/WYNm3aMGLECN577z0UCgVTp059jG6fCQEvuYSw/ObvaEx599jVKGnmWrvIz/3aay3p06c7\ndna2uLm5c+fO7RxlqlWrwbx5XxIYGJTtmfejevvt/z00s9qYMRP54osZrFnzLTqdDl/figwbNhpH\nR0e6du1J//49USiUVKzoR69eYWi1Wn78cQO9e3ejWrUaODjkTCxSvXpNJk8ex4oVSzAajfTr9ynB\nwdU4fvwoffp8THp6Gk2aNMXe3oHKlavQt2937O3t8fDwsGhMgRBljaz9LkQWS6cdbU84QXjM1jz3\nd6/wCm+5P5PnfvH0kSltoijJ2u9CFKEWbnWxV1nz7a09RGsTzNsrWLnyvkdjmhZDL10IISwhPXUh\nshS0N2U0mbiQHk2CIRUXlT3Bdj6FnpsuyjbpqYuiJD11IYqBUqGgmr1vaTdDCCHMpFshhBBClBES\n1IUQQogyQm6/C1EIKXot2+OvsDvxOgk6DS5qa1509aWleyDOapvSbp4Q4iklA+WEyGJxljZNKmMv\nHeSWLudqhuXUtkyo9Bx+tg9PvpKbB7O0paam4uPja17nPTe5ZWZ7FMnJSRw6dJAWLV63qPzNm7GE\nh88hISEejUZDtWo1GDBgUJ7tXbJkIe7u7vj7B7Jp04anIuGLDJQTRalUVpQToiwymkxMv3I414AO\nEK/PYGrUXxhMhV+etEGDzNXS5s5dyNKl36JWq9m/f0+e5Xfv3klUlGVr0lsiMvICBw7kfb77GQwG\nRowYRIcOHQkPX8TixStQq9V8882CImuPEMJycvtdiAI4mXKbqIzkfMvEaFP5KzmW51x8Hvl8Op2O\nuLg7ODllLsn6YHay2rXrZMvMNnbsCFavXo+NjQ3z588lICAQLy9v5s+fi5WVFa1atWbNmlXUqxfK\nxYuRAEyf/kW25EsrVy4lMvICmzZt5NlnGzFt2kQMBgMKhYIBAwZTteq/Sz6fPBmBp2cFQkL+Xb2t\nd+9+3LsBuGbNt+zcuR2VSkXduvXp06d/rte5a9cO1q5djVKppE6devTu3Y/ExEQmTBiFTqfDzy+A\nY8cOs3btjxw/fpRFi+ahUqnw8fFl6NBRsvKkEFmkpy5EARy7m3/a1X/L5VzC1VJHjx4hLKwHHTu2\npWvXD2jSpCkNGz7LwYMHiIm5wfz5S/jqqwWsXLkUW1s7WrZ8kw4d3s93WVStVsu8ed/w+uv/JTU1\n1byOu4eHJ4cOHchWtlOnrjRo0JC33/4fX389h7ZtO/D114sZMGAQ06dPylb2zp3b+Phkn9ZnY2OD\nra0tFy9GsmvXbyxYsJQFC5Zy/fo1DhzYl6NtyclJLF26kC+/nM/8+Uu4c+cWhw8fYuXKJbz44suE\nhy+iWbPmGAwGTCYTM2ZMYerUmeb2b926pdDvtRBljXy9FaIAtBbeVtcZDYU+R4MGmVnakpIS+eST\nvnh7Z/b4L12K5Ny5fwgL6wFkZkGLjY3Os577h8v4+wdk2xccnJnr3dOzAlqtNs86oqKiqFs381l9\n1arVuHXrZrb9Xl7e7NmzK9u2pKRETp06iVarJSSktrkXXbduPS5fvpjjHNevXyMxMYHBgzN78Wlp\nady4cZ2oqChatnwTgDp16gOQmJhAXNwdxowZDoBGo+GZZ/6TZ/uFeNpIUBeiAPxtLBsA51/IgXL3\nc3FxZcyYSfTv34vq1b8zZycbNmwURqOR5cu/wde3YrbMbPcyn3l7+xAZeZ7AwCAAlMoHM7DlnZHt\n/voCAwM5efI4jRu/xIUL5yhXzj1b2ZCQ2kRHR3PmzGlq1qyFyWRi6dJF2NjY8Npr/+X7779Fr9ej\nUqmIiDjO66//l8jI89nq8Pb2xdOzAnPmzEOtVrN16xaqVg3m+vXrnD59iqpVq/H336fM74mnp6f5\nkcH+/Xuws7N/lLdZiDJFgroQBdDErSIrYs6gMeXdE1crFDR18yuS8wUFVaJNm/bMmTOTSZOm55qd\n7P7MbO+/34khQwbg5eWDk1Phvlj4+lbk0qVI1q37jr59BzJjxmTWrMkMziNGjMlWVqlUMmnSdGbP\n/oyMjAzS09MJCalF9+59sLKyolmzV+jduxsmk4k6derSpMnLOYK6m5sb7dt/QFhYDwwGA97ePjRr\n9iodO37EpElj2bXrN8qX90CtVqNUKhkwYDBDhgzAZDJhb+/AmDETCv3+ClHWyJQ2IbJYOu3ot/gr\nfH0971zdH/vU4s3ylYq0bU+jgwf34+rqRo0aIRw+/CerVi3jq6+ezFH1MqVNFCVZ+12IIvRquQDs\nlVasjj1LtDbVvL2CtT0dKlQrsl76087b25dp0yaiUqkwGo0MHDi4tJskxGNPeupCZClMlrbI9EQS\ndBm4qG2oau+GSpH3s2rx9JKeuihK0lMXohgoFQqC7d1KuxlCCGEm89SFEEKIMkKCuhBCCFFGyO13\nIQohRa9ne9wd9ibEkaDT4aK2orFbOV4v74GzLFkqhCglMlBOiCyWZ2nTMP7ieW7lshJbOSsrxleu\nSkVbu0K3Izr6Bl9/PYekpCQMBj2VKwfTp08/7O0d8jymVavX2Lz5V778chbt23+Al5dXoc+/d+9u\n/u//1mAymdBoNLz//oc0bfpKoesTMlBOFC0ZKCdEETGaTMy4fDHXgA4Qr9Mx7dJFvqoRUqiR8BpN\nBsOHf8qwYWPMSVK2bfuJ8eNH8dlncx56/IABgwp8zvudOnWCdeu+47PP5mBvb09SUiI9e3YhMLAS\nQUEy916Ix51FQf3atWvs3r2bK1euoFAoCAgIoGnTpvj6+j78YCHKkJMpd7mSkZ5vmRithsNJiTRy\nLfjI+D/+2E+9eqHZsp61bPkmP/ywnujoGyxbthgrKytiY2OIi7vDyJHjqVaturlsWFgPhgwZyY4d\nvxITE01CQgI3b8bQr9+n/Oc/zz00w9mWLT/Stu172NtnLr3q4uLKokUrcHJy4u7du0yaNIbU1FQM\nBgPdu/emQYNnaNPmrYdmhrtyJYrjx49iMOh56aVmdOz4ERcvRjJnzkxMJhMuLi6MGDEOnU7HuHEj\nMBqNaLVahgwZQdWq1Qr8PgrxtMo3qN+6dYupU6cSHR1NaGgo/v7+qNVqrl+/zsCBA/H19WX48OGP\ndKtPiCfJ8eQky8rdTS5UUI+OvoGvb8Uc2729fYiNjQEyk6gMHTqKzZt/YPPmjQwZMjLXuqysrJk1\n6ysOHz7EmjWrefbZRsyYMYX587/Bza0cixfPZ+vWLbRq1dp8TG5Z15ydM9O+rlixhIYN/0O7du9x\n+/Yt+vT5mHXrNuV5LVqtlsWLVwDQps1bzJ27EHf38uasajNmTGbEiLEEBVXip59+ZPXqFdSuXRdn\nZxfGjJnA5cuXSU/P/wuUECK7fIP6rFmzCAsLo0qVKrnu/+eff5g1axYzZ84slsYJ8bjRGi0bgqI1\nWpbN7UEeHp6cOfN3ju03blynQoXML8/3eq6enhU4dSrv5Wr/zcTmhVarsSjDWYUK3ty6dTNHzvRy\n5dy5cuUyLVq8bm6nvb0DCQnx2Y7PKzPc2LGTWLBgLnFxcTRq9DwAV65cZtas6QAYDHoqVvSnUaPn\nuX79KsOHD0KtVtO5c7f83i4hxAPyDeozZszI9+Dq1atLQBdPFX9bW4vK+RVyoFzjxi+xcuVSc9Yz\nyLwl7uLiau7BKyx8Vv9gMUsynP33v2+xYEE4oaENsbOzIyEhnqlTJzJ58gwCAoI4cSKC4ODq3L59\ni7t3k3F2dnloZjitVsvvv+9k/PipAHTs2JZXXnkNf/8ARo+eiJeXFydPRhAXd4fjx4/i7l6e2bO/\n5vTpkyxc+DVz5y4s1HspxNPIomfq0dHRTJo0iUOHDmFlZcWLL77IqFGjKFeuXHG3T4jHShO3cqyM\nuYEmn564WqGg6QMpSi1lb2/PjBmz+eqrWSQnJ6HXG6hSpSrjx08pbJPNLMlwVqtWHVq1as0nn/RF\nrVaj0WTQq1dfqlSpSqdOXZg2bSK7d+9Eo9GYn8c/LDOctbU1zs7O9OjxETY2NjzzTCMqVPBi0KAR\nTJ48FoPBgEKhYPjwMbi4uDBu3Eh++GE9BoOBLl26P/J1C/E0sWhK2/vvv88bb7zBO++8g8lkYsOG\nDRw4cIDFixeXRBvzJVPaRFGxdNrRjrg7zLt2Jc/93Xz9+K+HZ5G2TTzZZEqbKEr5TWmzaEW5lJQU\nOnbsiKOjI05OTnz00UfcvHmzyBooxJPkFffyDA6shLeNTbbtFayt6e8fKAFdCFFqLLr9HhISwqZN\nm3j77bcB2L17NzVr1izWhgnxOHve1Y1GLq5EpqWRqNfhrFZT1d5BsrQJIUqVRbffn3vuORISErCx\nsUGpVGabZqJQKDh79myxNjI/cvtdFBW5RSqKi/xuiaKU3+13WSZWiCzywSuKi/xuiaL0yMvEpqen\nEx4ezsGDBzEYDDRq1IgBAwaYV50SQgghROmzqKc+YsQI7OzsaNeuHQDr1q3j7t27j8Ucdempi6JS\nkN5Uit7Ib7cz2BefQYLOiItayQvlbHjNww5nK8loLLKTnrooSo98+71Vq1Zs3rw527Y33niDrVu3\nPnrrHpEEdVFULP3gvakxMOFcIre0Oeeql7NSMjbYhYp2hc+VdOnSRebP/4qMjAzS09N57rkX6Nq1\nh8WLztyzZ8/vhITUonx5j2zb72Vy+/nnTbi7u/POO20eWpdGo2H79m289dY7Fp17w4a1vPtue4vK\n6vV6Vq1axuHDf6JUKlGr1XTv3ifb+vdPOgnqoig98pQ2k8lEcnKy+efk5GRUKtWjt0yIJ4zRZOKz\nyKRcAzpAvM7I9MhkDIUcqnL37l3Gjx9J//6DmDt3IQsXLuPixUg2bdpQ4Lr+7//WkJqammP7gAGD\nCpyvIT4+ji1bfrS4/IoVSy0u+803C0hPTyM8fBHh4YsYPnwM06dPJDr6RoHaKISw8Jn6Rx99RNu2\nbWnatCkAu3btokePHsXaMCEeR6eSdVxJN+RbJlZj4HCilkZuNvmWy83+/XsIDX0GPz9/AFQqFaNH\nT8DKKrOHN3fubE6ejADg1Vdfp12795gyZXyOzG1xcXeIjDzP5MljGTNmEqNHD8XZ2YXnnnuBgwcP\nmJPA7N27m127dpCRkcHAgYOpWbOWOTc7wLhxI3j77Xf57bdfiIq6zLJli2nb9j2mT59IUlJmcpuB\nA4dQufK/+SFWrFhCcnISn38+nYEDBzN16gSio29gMBjo0OEDmjdvke2at2/fxrp1m1AqM/sYXl7e\n/O9/7di27Sfq12/AypVLUSqVxMXF0apVa959t12uGd7On/+H1atXYmWlJjr6Bs2bt6Bz527s2bOL\nb79dgVqtpnx5DyZMmEpqamqhM845OTmzbNliTCYTwcHVGTJkBCdOHM83+50QJcWi37qmTZtSu3Zt\nDh8+jNFoZO7cuVSrJukQxdPneHLuedQfFJFUuKCeW5a0ewNSDxzYR0xMNIsWLcdgMNC7dzcaNHgG\nyD1zW5UqwQwZMhIrKyvi4+NYsuRbrKysOHjwgLlub28fhgwZyaVLF5k8eSxLl67OtV2dOnXl4sVI\nunTpzrx5X9GgwbO0bt2Ga9euMnXqBObPX2Iu27lzNzZsWMfgwcPZsGEtrq6ujB07ibS0VLp27UiD\nBs/i6uoKQEJCPE5OzjkCoI+PL2fOnDa/J0uXrsZkMtKpUweaNXsl1wxvzzzzH27ejGH58jXodDre\need1Onfuxm+//cr7739I06avsG3bT6SmphY645xer6dDh9YsXrwCN7dyrF69glu3bj40+50QJcWi\noP7BBx+wbds2goODH15YiDJMZ2mWtkLefq9QwZvz5//Jti06+ga3bt3kypXL1K1bD4VCgVqtJiSk\nNlFRl4CHZ27z9vYx9/bvV7duKACVKlUmLi4ux/7cLuPSpUiOHTvCzp3bAbh7NzlnoSxRUVE0bPgs\nAPb2DgQGBnHjxnVzUHd0dOLu3WT0en22wH79+lVzVrpatepgbW1tbueNG9dzzfCWub8KarUatVqN\njU1m8p1+/T5h1arlbNiwjoCAQJo0ebnQGeeSkhJxcnLCzS0z78UHH3QmISH+odnvhCgpFj1Tr169\nOj/++COXLl0iOjra/L8QTxtLB8D52Rbu1usLLzTmzz//4MaN60DmILK5c2dz6dJFAgKCzLfe9Xo9\np0+fNAez3AbRKZVKjFmJZxSK3P+pnz2bmeb14sVIcxDV6/WkpaWh0+m4fPmi+XiTKbOugIBA2rV7\nn/DwRUyaNJ0WLVrmqPdeQAwMDOTkyeMApKWlcvHiRXx8fMzlrKysaNr0FRYtmmdu640b1/nhh/W0\nbPkmABcunMdgMJCRkcHly5eoWNHfnOEtPHwRvXv35/nnG2e1M+c1bt78A9269SA8fBEmk4m9e3eb\nM84BuWacM5lMREaev++9zKzYza0cKSkpJCdnPnqYM2cmMTHR5ux34eGL6Ny5q/kOihAlzaJPnhMn\nTnDiRPZv/wqFgp07dxZLo4R4XDUpZ8O311PQ5JMuXa2Al8tblqL1QQ4OjowaNYEZMyZjNBpJS0vj\nhRdepHXrNigUCo4fP0rPnl3Q6XQ0a/YK1apVz7OuWrXqMHnyOIYOHZVnmZiYG/Tv3wudTmt+zt6u\n3Xv07PkRPj6+eHl5A+Dm5oZOp2fevK/o1Kkr06dPYvPmjVm31HOOrwkMDGLixDGMGDGWGTMm07t3\nNzQaDV27djf3cu/p3bsfS5cuomfPj1CrrbC2tmbYsNH4+lbk5s1Y9Ho9gwf3Jykpic6du+Hq6ppr\nhrc7d27neo01aoQwdOhA7O0dsLOz4/nnG/PCCy8WKuOcUqnk00+HMWTIQJRKJcHB1ahRI+Sh2e+E\nKCmyopwQWSyddrTzdjrzr6Tkub+LnwP/rSALMxWFY8eOsGnTBiZMmFbaTXkkMqVNFKVHntIWHR1N\nnz59CA0N5dlnn2Xw4MHEx8c//EAhyqDmHnZ8WskZb5vs0zo9rZX0C3KSgC6EKDUFzqduNBrZuHGj\n5FMXZU5Be1NGk4nIVD2JOiPOVkqqOqglS5vIlfTURVF65LXf7+VTv+ejjz5i48aNj94yIZ5gSoWC\nYEf5kBZCPD4suv1+L5/6PZJPXQghhHj8FCifuq2tLQqFQvKpizJJbpGK4iK/W6IoST51ISxQkA/e\nDB38HaPg3E0laRqws4ZgTyO1fEzYWRd3S8WTRoK6KEqFHv0+a9asbIlcHpSYmPhYpF8VoiQlpcPa\noyoOXlIRn6ogQ68gIU3Bn1Eqvj+iIj5nDpUCuXTpIkOGDKBfv558/HEnlixZiKXfvQ8d+oMpU8YD\nMHLkkEdrCLBp00b69u1OWFgPevfuyrFjRx65TkvFxsayf/9eIDOzXGxsLMnJSWzf/gsAq1YtNy8l\nKwQ5JUAAACAASURBVITIlO9AuZYtW9K3b188PT1p2LAhXl5eqFQqoqOjOXToELdu3WLkyJEl1VYh\nSp3JBFtPq7ibkdsodxOpWgU/n1bxwTMGlIVIq34vS9uUKTPx8/PHYDAwZsxwNm3aYFGK1PtNnfpo\nX7h37PiVw4f/5Msv56NWZyZJCQvrwdKlq83LvBanY8cOc+VKFI0bN2HAgEFZ245w4MAeWrR4nQ8/\n/KjY2yDEkybfoF6zZk1WrVrFoUOH2LVrF7t370ahUODv70/79u157rnnSqqdQjwWriUoiEvNa9pa\n5vakdAWX4xRU9ij4k63CZGmLirrMtGkTsbW1w87OFicnZwBztrWwsB5UrVqNS5cukpaWwqRJM/Dy\n8mb58m/Yu/d3XF3dyMjI4OOPexEa2tDclk2bNtKv3yfmNdl9fHxZtmw1Li6uxMREM23aRPOKbgMG\nDKZq1WDat3+HWrXqcO3aVRo0eIbU1BTOnv0bf/8AxoyZxJQp4zGZTNy6dZP09DRGj55IQEAg69d/\nz2+//YpCoaB58xb8739t+fbb5WRkZFC7dh2+/341Q4aMZOXKpURGXmDTpo2cPn2S5s1b0LDhs7lm\ngsvtut3cyjF27HBSU1PJyMigR48+PPtsowL/PQnxuLJoSlujRo1o1Eh+8cWj05v+v737jpOivB84\n/nlmdvd6oxy9HkUUECFWJIAao2jMT4lCMCRWLAG7IiqKBgQ1RgWRiCUq0dhNxKhBkaCxoKCogFKl\nd6633Z2Z5/fH7u3t3u3dLQe3V/i+X6977cwzz8w8u3d733nKzGPzZdF6Pin8gSKrjNbuNE7LGMDA\nlG5Rn1/e1GzJja2MW3LrF9TrM0vbk08+zhVXXMXxx5/E3//+HFu2bK523MCjTG/mySfn8sEH/+Gk\nk07hiy8+46mnXsCy/Pz+92NrKEvniLSMjEANfe7cR7nwwrEMGzaC9evXMmvWn3jmmQXs3r2Lxx77\nK23atOHss09j/vznuPHG27jool9TVBQY/9KpU2fuuutePv/8fzzxxGNcffUkFi/+gCeeeBqAG2/8\nIyeeeBK/+90lwZr6cF5+OTB73O9/fxn/+tcb/PrXF7Bq1XcA/Otfb0SdCS7a+x46dBgFBQU8/PBs\n8vLy2LZty0H/jpqzLwv28vrufeT7IdHQjGidxq/bdsGsT7OSaJJiCuqffPIJjz76KAUFBRF9e/Ls\nd3Ewcv3F3LvtVX4q3xORvqRgFUNSezK58/kkGk17lJldyzPf65OvqvrM0rZ161b69esPwIABg6IG\n9T59+gaP344DBw6wZctP9Ot3DKZpYpomRx3Vr9o+7dt3YO/e3aSmVs6VvmzZ5+Tk9Gbz5s2hGd56\n9+7L3r2B32l6egbt2wcmhklKSqJHj55A4Jn2Pp8XgMGDA5Od9O9/LLNn/4VNmzayZ89urr/+GiDQ\nBbFt27aYP7OaZoKL9r579szh17++gGnT7sSyLH7zm+oXMy2R37G5+ccf2VmaDWSH0l8qhTd2rWfm\nUR3pmljz4CvRfMR0eTZ9+nQmTpzIc889xwsvvBD6ESJWtnaYvu21agG9woriTczZ+W6cS3XwWqXE\nVvuONV9V9ZmlrUePHqFa648/ro563KqtID165PDjj6txHAefz8e6dWur7XPOOefx3HPPYFkWAFu3\nbuGBB6ZjmkbE7Gvr16+lVavWUc8Tzdq1gVtgv//+W3r0yKFr1250796TOXOe5PHH5zNq1Lnk5PRG\nKRWaGa5CYOa5yM+2tpngqpZn48YNlJaW8NBDj3Hnnffy6KNHxkDfKesqAnp1XqsVt/+wC59jx7lU\noiHEVFPPyspi5MiRDV0W0YJ9XbyRDeW7a83zSeEPDC3sRzt3RoOVo7agYweDl+mq/rWo2CshA0rd\nJnYtMVsBSZk2P5WH7xkjEy674yb+9NQDOI5DeXk5xx03mOPO/jlKKVLXdeCyO67CsixOOGsYCd1a\ncf41lzB//hM8/5+XSUtLw53mYUv5PmiXwpbyffhaudnp5EP5PvI8XgqTLFydMug38kQuv+MqUlPT\nUO1TOeAqY2t55UxnfU4dzMbinVx517W43S4cx+Gqu2+jKMnm/Kv/wDPPPMkL77+CZdtcdvv1bPXu\nh+zkwCtELNutE9hh5VGapvj4xy/56O5P0Voz4ZprScxuRe+hg7jyzqvx+y1yevbixHSD1J7ZLH33\nJdr895/4s9zsdgpJapfK+oJtPPXP5ylLVew3Shl81nC+fuqvXHHH1fh8Ps6fMJ6SZI0/M7CP6T1A\ngdtHUaKNyk7h8w3f8J87l+I4Dhdc9Qe2e6vPI1+P31zYftX3tOzA35bLruVfbj1PGO184XaUl/JT\ncetac/nsLF7ZvYXxHXvWrxCiyYjpPvWHHnoIy7IYNmwYCQkJofTjj2/8OYPlPvXm4dEd7/BRwfeN\nXQwhGo+GQOOoGfjRgVeFK7Rcka5wReQBE6VdlXlC21yoKvnQwbRQujt4vNq1StzH/P7ypNDm4JCf\n/f7dd4GmvTVr1oTSlFLSBC9iVmyXN3YRhAgG1srgFwqYdQbGQF5VJU/1fV1hwTrK/vWu/ze8clsG\ny7UEMQX1BQsWNHQ5RAuX7UmPKZ9LmbhqGepRn55qXa+9Yjx2+KFV/c5Tr73q8SDIJvPZaQVVaqIa\nIxgwq9dQq9VSawqgVfZXNQVvEVWyWc/RnaJJiSmoL1++nGeeeYbS0lK01jiOw86dO/noo49q3Mdx\nHKZNm8batWvxeDxMnz6dbt26ReQpKyvj0ksvZcaMGeTk5MS0j2ieTs8YyDu5KwIrThqG3QOlE9Gq\nBMe1CVQZbmXyt94TSXc1znzkR8qjPLXW+DX4HY0v+Op3NF6t8TsV6eHL4AvmiS09sOxzNH6t8Tng\n18F8DkjoiORW4DYUbgWe4KvbUKFlj6FwG+BWKpAeSgvPG9hekdejKvcptLw8vKmkzib4M9qkxukd\ni4YUU1C/6667uPLKK3nrrbcYP348H3/8cZ2ztH344Yf4fD5eeeUVVq5cyaxZs5g3b15o+/fff889\n99zDnj17Yt5HNF85Se0Znj6A/+3LwrT7RGzT/hNwXN9zYcfkRgvo8WZXBEGt8TmVgS8QJOtO9zsE\ng2fd+4YH70Cgbex337QoqCEoVgbLiuWo6eFBODxPRbA2FKZj4TEUiW53RLpbBabwbVgJvH9gJ2sK\nWteYI9mdy/9l927gcoh4iCmoJyYmMnr0aHbs2EF6ejrTp0/nggsuqHWfFStWMGzYMAAGDRrEqlWR\nz2j2+XzMnTuX2267LeZ9RPOltcYpPxXT9lfbpjAxrUHYvqS4lseqqGkGA16Z38avwVZU1jCj1Dwr\nlmuqnUZLrxpoaxs9fyRyqYqgGgyQFTXNsOVAzTVKnmAwrjW9SjAOTzdVbLfiHYrgjRW4XI3T/H9P\nTh+mbljL2sJWqCrdW+meAzx4VHdc8gCaFiGmoJ6QkEB+fj49evTg22+/5eSTT6a0tLTWfYqLi0lN\nrWzOMU0Ty7JCj5wcMmTIQe8TTVZWcqN9UUTs1uSXsyy/ekAP9/buMkZ2yyTJNPA6Gp9dEQid0HJk\nevCnYpsdzFslvXJbZLrE1UiRNdGqNdbqQdMTSg9vJq4aOKsG5MB6QkVepXAZYDaDpwkeGoOMjGQ8\nnsZ7uNJT7U5gfWEBT2/8iT1lFqkug990a8+Ido1/F5M4fGIK6pdccgk33ngjc+bM4Te/+Q0LFy6k\nf//+te6TmppKSUnldFWO49QanOu7T15e7RcXomn455a6bz20geuW7Wr4wjRRZrC2auLgLS5CWz6w\n/CS6XHRu3w5l2/hKi2nXpg3ffPUFg/r3p1V6ekTwrQiuVYNyXemuONRWo9IabYPVwi+xLMvhwIHi\nRh+vkYnBLZ1zItLktuDm55BvaTv77LM566yzUErx5ptvsnnzZo466qha9xk8eDBLlixh1KhRrFy5\nkj59+tSav777iObhgL95DI8KDTyKGIxUWUt1K/DvN1B+hUuH/xBaTnRDp6M1HjNsXxWt1hqZbiqF\n11vOlVf+gamTp3LMMYEL5/fee4clz81nxIjT2bJlM9dcM4mJj7zOhSf3oVtXGdwkhKgUU1AvKCjg\noYceYuvWrTz22GMsWLCA22+/nYyMmp/89Ytf/IJPP/2UsWPHorXm/vvvZ+HChZSWljJmzJiY9xEt\nQ5oZW3+dAhJCAbTmgUrRBjNVq4UGl2tKDzUrBwM5th9DqVprU0V7FJu/r7u7p6vfJiP74Gufn332\nPwYNGhwK6ABnn30ub7zxKg89dD9ZWa0YMGAgAM8+O5+8vFzKysqYNm0GnTp15q9/fZxvv/0Gx3EY\nM+ZiTjvtDCZOnEBWVisKCwv5y1/mYJrSXSVESxVTUJ86dSpDhw7lu+++IyUlhezsbG699Vbmz59f\n4z6GYXDfffdFpOXk5FTLF34PfLR9RMswtFUCSw7U/QCa2cdk0SEppj/Lw85y6m5+LtodWxN10W5F\nRqeDD+o7d+6gU6fO1dI7derMBRdcGDFr2SmnnMovfzmKZ555kv/+dzE9e/Zi164dzJv3DF6vl6uu\nupTjjz8RgDPO+CXDh8ujnoVo6WKqPm3fvp0xY8ZgGAYej4cbb7yR3btrf463EOEGprvpm1J7sB7R\nOqHRAnqsdIy9CLHmq6pt22x27ao+rmDHju04TuRB+/YNzKzWunVrysvL2bRpA2vX/sjEiRO4+eZJ\nWJbF7t07AejaVZ73IMSRIKagbpomRUVFoYE0mzdvxpDbH8RBMJRicq+MGgP7SVkeJnRr+lM/JqbH\nVvuONV9Vp546nOXLl7FmTeXtnAsX/pOMjExM04yYtazqwLZu3bpz3HE/4/HH5zN79l857bQzQrV+\n+b4KcWSIqVo0adIkxo8fz65du7j22mtZuXKl9HeLg5buNvjTUZl8V+jn01wvRbZDa7fBiNaJ9Epx\nNc7o64OU2UWz63uNtmsuq1KazG71C+rJyck88MAjzJ79MIWFBViWTa9evZk2bQa7du3ghReepU+f\n6INUhw79Od98s4Jrr72CsrJSfv7zkSQnp9SrHEKI5immWdpyc3N56aWXWLJkCY7jMHDgQNq2bcvE\niRPjUcZaye0Y4nCJ9TGxuT8pdnxd82CzDsfatOnVsm/REgfnSHkEsYiPQ76l7corr6Rv374yp7oQ\nQKseGtNts3u1ga+4ssbuTta0O9ohq561dCGEOFQxj0qS5nYhKmV01qR3sinLA6tcYSZokltBM+hB\nEEK0YDEF9TPOOIPXXnuNk046KeIe144dOzZYwYRo6pSC5FZQz8lThRDisIspqBcVFTF//nyysrJC\naUopFi9e3GAFE0IIIcTBiSmoL1q0iM8//5zExMSGLo8QQggh6immoN6lSxcKCgokqAtRoRzc3ytc\nPxioEtDJYPV18B+rIX4zyAohRISYnkihlOKcc87ht7/9Lb///e9DP0IciVQ+JL1o4vmfiXFAocoV\nRq7C87lJ0gITdaD+x/766+Xcc8+UiLR58+bw7rsLYz7GjBnT+OKLz2LKu3TpEvbv38euXTuZMOGS\ngymqEKIJiqmmfvXVVzd0OYRoHjQkLDQxCqsPc9dojBJF4tsmZX+wY7xkblyvvfYPune/o1Hn+RZC\nHD4xBfUTTjihocshRLNgbFWY+6Pft6YIpBv5CnOjwu59+EbFO47NrFl/Yu/ePRw4sJ+hQ3/OhAnX\nsm3bVh54YDp+v5/ExESmTau89XT16lU8+uhD/OlPD1BaWsycOY/gOA75+fnccsvtFBUVsWHDOqZP\nv5upU/9Efn4eU6bczP79++nVqzeTJ9/Frl07mTnzPmzbRinF9dffQu/efRg79nwGDDiWrVu30KpV\nK6ZPf5AdO7Yzc+a9mKYLx3G4557ptGvX/rB9BkKIujXt2TOEaGJcm2O7Ed3cXP+gvmLFciZOnBBa\n37lzB1dccTXHHDOA22+fitfr5YILRjFhwrXMnfsov/vdJZx00in8739LWb9+LQCrVn3HihVf8uCD\nj5CV1YrFixcxceKN5OT0YtGi93n33YVMnnwXvXr14dZb78DtdlNaWsKUKfeQmprKmDHnk5eXy9y5\nj3LhhWMZNmwE69evZdasP/HMMwvYuXMHjz02j3bt2nPNNZfxww9rWLv2B/r1O4Zrr72eb7/9hpKS\n4nq9fyFE/UlQF+JgWDHms+t/iiFDfsa9984Mrc+bN4eSkhJ++mkjX3+9nJSUFHy+wGNHt27dQv/+\ngfnVTz11OAAffPA+X375BaWlpZhm4Cvepk02zz33NAkJCZSWlpKSUv2Z8B06dCI9PR2ArKwsysvL\n2bx5M8ceOxiA3r37snfvHgAyMjJDtfDs7Hb4fF7OPffXvPji89x88yRSUlK56qo/1v9DEELUSzPo\n9ROi6XBax1b71jHmOxipqWncc890xo79HV5vOVprunXrwQ8/rAZg0aL3eP31lwG47LIJjBkzjocf\nngXAY489xOWXX8Vdd91LTk4vKqZ8MAwjNKVrtAl1unfvznfffQPA+vVradWqdY15//e/pRx77HE8\n9tg8Ro48nRdffP4wfwJCiLpITV2Ig2AdpfF8olFWzc3w2tBYRx/eoG4YBsuWfc7q1d/jdrvp3LkL\n+/fv449/vJ6HHrqf559/hsTERO6++0+sXfsjAL/61f+xZMmHLFr0PmeeeTZTp04mLS2dtm2zKSjI\nB6B//4FMn34Pt912Z9Tz/vGPN/DAA9P5xz/+jmVZTJkytcYyHnXU0Uyffg/PP/8MjuMwadJNh/Uz\nEELULaZZ2poymaVNHC6xzqTlWqVI+KDmWdq8I2ys45r110ocZjJLmzicDnmWNiFEJau/RntsPJ8a\nGPmVNXYnXeM/2TnstXQhhIiVBHUh6sHuoynrbWPsBlWq0Ekapz0ySkUI0agkqAtRXwqcDiCztAkh\nmgqpVwghhBAthAR10TgcB/xloJ3GLokQQrQY0vwu4srI24pnzULcW5ehbD/anYy/56l4jz4PndK6\nsYsXO6+NZ00urvV5qFILneTC6pWB/+jW6CT5WokmzO+gyiy0x4BE+VttaeSWNhE35s7vSF76Z5Tt\nr7bNSUin9Bd34WR2aYSSBcR625Eq9JG8cBNGUZT3keyi7Fc9cLLqP03xpk0bmTdvNuXl5ZSVlXHy\nyUO57LIJUR/4UpM33niFN998jcsum8Dpp59Z77KIw6Mp3NKm8r0krNiLa2MBygn827c6puAbnI3d\nObXRyiUOXm23tEnzu4gPXynJnzwWNaBrwPAWkvTxo02/OV5rkv6zJWpA14BRapH03hZw6netXFRU\nxLRpd3DddTczZ86TPPnk39i4cQP/+tcbB3WcpUuXcN99sySgCwCM/WWkvLkB9/r8UEAHcO0sIemd\nn3D9mNuIpROHk7S9iLhwb/oY5S+Nuq2i/mkW7iThy+dwsrqAMsAwQZno4CuGEXUdw0RXWzfAcIWO\noyuOV7GvMuAgar4VzO3FmAfKa30fRqEP1+ZCrJ4ZB338//1vKYMHH0+XLl0D5zNN7rrrXtxuN3Pm\nPMJ3360E4Be/OIuLLvotM2ZMw+12s3v3Lg4c2M8dd0zjxx/XsG7dj8yadR/33juTzz77hA8++A9K\nKU4//UwuvHAsM2ZMo6CggMLCAh588FFeeukFvv32GxzHYcyYiznttDOYOHECvXv3ZdOmjZSWFvOn\nPz1A+/YdeO65p/nkk6XYts3//d9o/u//RvP66y9XO8fSpR/x978/j8vlok2bttx77/0YhtQj4k5r\nEhdvQ/miXzArIPHjnZR0TkWnyhS8zZ0EdREXrr0/xpQvYf0HDVySSlpVXgRgmPgSMigbOgnDVIAK\nC/oqGLEV7g0lMR3btWkfdrYveIzgcaIcr+q2/bu20ym7NXiLQ+nJLvj0s6Xs2rGN+XPnYdsO10y6\nhiGDjgXHpn12J267+VbeXvg2b//rDW69eTIffPA+t948GW95GYsXf8ATc58C4MabJnLiiScBgYlj\nxoy5mM8//5Rdu3Ywb94zeL1errrqUo4//kQA+vU7huuvv5knn5zLBx/8hxNPPIllyz5j/vzncByH\nv/71cTZt2hg4xxNPB85x4x858cST+OCD/zBu3HhGjjyD9957h5KSEtLSam42FIfA1mA5KL+N8jvg\nc1CWA34Hc08pZp631t2Vo3H/kIfv+HZxKrBoKBLURXw0wWZ1pZ1AuZxAU7pSLtA2OLXU4K0Yp1/z\n+1Hegx/v0SEzmbWbNmEU7w2l7dyzl3WrVjCoT3fMot2YQP9e3dmy5muUv5S+HVth5G+nfYrBquJc\njPytKKscVbiTzZu3sGfndm6YeBkARcUl7FizHOUtoltWEkbuFn5avZy1a1Yx6eo/AGD7ytiz/huU\n7eWoDpkYBTtpn57IgbwDbFv7Lf1yeuAq3QcorrtkHIs/+YQ9u3Zww8QrQEFRUTHb16/mussvYcEr\nr/DGqy/SrWtXhh8/COWyibyoCbzq8PWaLoSqrodfGFXbP9p6E1ERgH3BAOyvDMDKHwjMlcuV6fjD\ngnbV7fXs7gln7onekiaaFwnqIi7s1j1xb/uq7nxZXdEJaeDY4NiBwOvYwWBrg3ZQoXUHtIVynLD1\n4D4NxEmPrflYp9cvkAw9fjDPv/4W55+1m84d2mNZFrOfeZ7BA45h5eo1jP31uViWxfc/rmPUacP5\n4uvoM6ZV6NqpIz26duYv99yBUoqX//UOOd27suSzzzEUoG26dezA4AFHc/sfr8JxHP726ht0atsa\ntA5c8FjlYPvB8dOtfVveeucddHkxjuNw830zmXjpeHp06RRxjl6d2vL2wre4/MJf0yozgweemM/H\nSxYx6rQRUcvZ8GE3estLYLWWFhQHsHRgyl1bo/ygLA2WDrz6Qdka/IF15Q/fVvHqBLZ5/WArlG5i\nFxkV/GWNXQJxGEhQF3HhzxlBwndvohw/NorVyW0odHlo4y+jb1kuCnCSW1Fy9oxAX/ihCNXAndDF\ngAq7KMCxKy8MKta1g2VbOGnZOEbFZC06+LC4QC1IofEfZeP5fnvgn3VNpzfA37cVOjF4ARC6wSTy\neIF0HQwfgW0p6QlMvekGZj3xFNpxKC0rY+gJP+PC837Fnv25XHnbXViWxWmnnkzfnJw6P4rePbrz\ns4EDuPr2u/H7/fTr04u2rVpF5Dn1hCF8vWo110y5m9KycoafdAIpyUlRj9enZ3dOGjyIq26finY0\n5599Zo3nOLp3L26dPovkpCSSEhMZ+rMhdZb3kDiB4BsKtlbYepXX2rZFHOOw3hsUr/EEDoHC+0AF\nB3Q6rWrdA0CpLcDRDVoy0fDkljYRN64N/2XxusX8Pfto9nlSQundygu4cs8qjj3+D9jtG++fSqy3\nHbl/yCVx6Y4at5cP7YB/QJvDWrY6RVw41HYRUbFey7YqFzJ1XZTUea4q+yt0WB9w2GtFzdbSgWZl\nC5TlBOKSpSNryOEB2B98bZb/yRygIvj6AoG44jVaWihY+6psC89rRfQ4aA267HKUU3N/ucbC6b6Y\n0rNuabB3Kg4fmaVNNAn/SOvIi52Pr5a+JTGDqd2GcnNiBqc2QrkOlr9fK7THJOHL3RgFvlC6k+bG\ne3w7rD5Z8S9UtSbkw0NrHagBR+njDe/7rb7dDgZrJzDq2jr8fcDxphXgUmi3ApeBdilwK7RLRSzj\nUmhXZV7tUjiGxvXTYlx2USAAKy9Ke0H7gEDXUWVrUi3r9aAUvNZpLb/Y1YZMq/qUwTaaBV238Nv6\nP1pBNCES1EVc7PaW8NKemkfAa+CvO77j+PR2JBxq83scWDkZWD3TMfaWYZRaOEkmTnYygY7qRnII\nAThiQFZEALZRTW+MY520AtwG2m2g3WZg2RVYr0yv/kMwjxOWz6nYz6y8YAq/JIm8PNHVt2uwbD/k\njMNluoPbdZU9qizrKNu1DlxkaSts/IiNdgLdTVpXdCs5aKcij0NR4R6eLsnlzbY7uXhHBqfvSyXJ\nMXDQLMsq4x+d8lmdbtDDM5DB9fy8RdMhze8iLv6++wde37segOxyF8MPJJNumezzWPy3TQmF7kDk\naO9JJt2VAARriVVE/gOM9o+v+vaa/sCrbdeBFFXZy13tPNHKUtNXKNr+4eV0aUiyFB5bkWgrEm2D\nBBuSQuuKhGB6YlhatR/LINEJLLua6iCsWthoykxNuelQZmrKTIcyI/hakW4ElktNh3Ij+BpcLwtu\nD60bGq+h4zH6rllyOZBumaHPsMKp6e25pfsJjVgyEStpfheNbmt5EW4HJm1qw1l7UzHD/uNes7kV\nL3UuYEHnfHb7Stnta2K31mhwa0WSrUiyDZJsRbJjBIOvEUhzAtuSg3kSncq8SVXyJNmKJMdoxgHY\noTQs+FYE44oAWx4WkMuMyrzlYWmV+TV+JQE4niwDcj3Vm/ILHKsRSiMONwnqIi4SlMltG9py2v7q\nz5j2aINLtmVhaHi+a/6hnSgsACeHBdDwAJwclhYegCvyJ0YJ0mYzjDo2lYEzsvZbmVYRgEur5AkF\naiNsfwnALVpWsIVMNG8S1EVcjPRnMXx/7U9j++32TFp3ak2rxCRclsb0g8vSgWVL47Iq110WgTR/\n5HbT0hjNsEPJUWC7VODHrbBdYesuI7DuVmFplXkdt4HtUjgR+yu0AShF5RC6ymicGPyByjF2qoZo\nXeVRLtXWwkdaq1qXol8PhN9nH3V7lGOpyAy15q3rWEQtf/TPI+r2GvcPMAwz9Hjcg/qMo3wuNf4O\natm/yPIxae1HWDV2RAUMz+xc63bRPEhQF3Fxwk4PUHtQd6M4Z5lTZ77Gpg3AbVYbXFVtYJY7Mp3g\ntqqDtXAbgQF2Vf6Ju5AvqDh0qaabC7J78+redTXmOSalNcelZcexVKKhyP8MEReu0sbpr4sagD3h\nwfjQA7AQTd3Ydn3xa4d/7dtA1ZsZBqdlc1PXIRjyN90iSFAX8ZEQ25+aNhU6yVU9mFYEYE/1KpKW\n7gAAIABJREFUW5MignbVAGzKrGBCGErxhw5HM6p1D5bmb2efr5QU080pGR3plZzZ2MUTh5Hc0ibi\nwtxaRPK7m+vMV/zbPugMGbAjhBA1qe2WNqnGiLiwu6RiZ0d/nngFf+9MCehCCHEIJKiL+FCKsrO7\nY7eNHtj93dMoH94pzoUSQoiWRfrURdzoBAO7UxFm/i6w00CboCzwFOF06hl4DKcQQoh6k6Au4ibx\ns+V41m0K3q9VWLnBgcSvvgUNvoH9Gq18QgjR3Enzu4gL40BeIKDXIuGbVeD1xqlEQgjR8khQF3Hh\nXv9TnXmUbeOuI/ALIYSomTS/i7gwimN7SlzSV9+SsHodTlYGdlYGTlZm4DUzHVzy5yqEELWR/5Ii\nLrTbHXNeo7QMo7QM147dlfsrhZOWitMqAyczA7tVJk5WBk5aKhjS4CSEECBBXcSJ1b0zno2b672/\n0hqzsAizsAjYHkrXpomTmR5Zq8/KQCcnyaNchRBHHAnqIi6sLh2xszIx82qeWtXftRPlJxyHmZeP\nkVeAmVeAkZePUVCEquHBh8q2MQ/kYR7Ii0jXHk8owNutMgO1+6wMSPAc1vclhBBNiTwmVsSNKikl\n+T9LMfMLqm2zOmRTesYwiNZMb9sYBUUYefnBQF+AmZuPUVJ60GVwUpJDwT7Qb58Z6K83zfq8JSGE\niLvaHhMrQV3El23j2rwd909bUeVedEoS/l49sDp3OPjmcp8PM68wLNgHaviG13dQh9FK4aSnRQ7O\na5WBTk2R/nohRJMjQV0cObRGlZVH1uqDr8q2D+5QphkW6IO1+qwMdFKi9NeLZsd2YP1exQ+7FcXl\nCo8LemU7HN1BkxT7OFbRBEhQF8JxMIpKAjX5sIBvFNbcX1/joRI8OMEAb2dl4LTKxM7MAI/8ZxRN\nk9cPb39nsqeo+sVosltz3rE2bVIboWCiXiSoC1ETy8YoKAwNzgvU7PMxSsoO+lBOanKoNl/RjO9k\npEl/vWh0//7e4KcD0bqSNKBI8Wh+d6KNW/5Um4XagnqDjX53HIdp06axdu1aPB4P06dPp1u3bqHt\nH330EXPnzsXlcjF69GguuugiAM4//3xSUwOXjJ07d2bmzJkNVUTRCLR28O1fQdnupTj+QsyENiR1\n+gXujKNQjdGk7TJxWmfhtM6KTPf6Qv30ZliwVz5/jYcyiksxikth285QmlYKJyMt4nY7OysTnZYi\nTfjisNMaHB1oareCP3nF1BDQITARA5T4FOv2KI7p2KzreIIGDOoffvghPp+PV155hZUrVzJr1izm\nzZsHgN/vZ+bMmbz++uskJSXx29/+ltNOO420tDS01ixYsKChiiUakeMrIPfru/Hnr4lIL922kMR2\nw8gceDvKbCK3nCV4sNu3xW7fllAY1xpVWhZxu52ZV4CRX4CynaiHUVpj5hdi5hfiDntSrna5AvfX\nt4qs2eukxAZ/a6LhOcGAatf4qiLX7fDtKiIo29X2V1h2ZHp4Pk39LhZ/OiBBvSVosKC+YsUKhg0b\nBsCgQYNYtWpVaNvGjRvp2rUrGRkZAAwZMoSvvvqKjh07UlZWxmWXXYZlWdx0000MGjSooYoo4khr\nJ2pAr1C+5xMKVieSOfC2OJfsICiFTknGTknG7tyhMt1xMIqKK2+1q+ivLyqu+f56y8Lcn4u5Pzci\n3UlMiBiUF6jhp0e/1U/USuvqQS8iSNoq9sBrVz1W9cAbfmxHN79WGOvgxpGKJqrBgnpxcXGoGR3A\nNE0sy8LlclFcXExaWmWfQEpKCsXFxSQmJnL55Zdz4YUXsnnzZq688kref/99XLU88zsrKxmXSzqC\nmrqinV/UGNArlO38gE6DxpGQ1hVlmCjDBcpAqWZwW1m7DKBTRJL2W+gDeeh9eeh9uTj7A68U13x/\nvVHuxdi1F9euvZEbMtIw2rZCtc1CtQm+tspEmU37s9FahwKe3wbL1lh2IIBYTmDZH7YeWq7IFwyo\n/oj1ymP4bR3KE9oveK4aGk9EDTq0dtO2bRNpKRP11mBBPTU1lZKSykk8HMcJBeeq20pKSkhLS6NH\njx5069YNpRQ9evQgMzOTffv20aFDh2rHr5CXd/APIBHxl//je6HlAtWT7eYIvKSTpPfT1fmAFL0H\ngA3vXhplbwOUiTJMUMFlZQZfXWCYYWlGaFsozXBVSQs/XthxQvkjj1NxjMjzVimLUXmMiHyGCYkG\ndHWhurUH1Rll2ZiFpRgFJZiFJRj5xbjyi1F+G4VCaQMwUOHNqAVFOAVFsGFLKEkbRrC/PiNigJ5O\nrd5fX7U5uHrNVFVpAq6lqdipDLZ2lO1Vm4ypZ3NwS2MojcsAlwFm8MdlBl8NHZkeli/wqkP5I9Mr\nt0dLd5mB3/3zy0zK/bX/Hnpketm3T6Y+bg4aZaDc4MGDWbJkCaNGjWLlypX06dMntC0nJ4ctW7aQ\nn59PcnIyy5cv5/LLL+f1119n3bp1TJs2jT179lBcXEzbtm0bqogijhx/IRaJfOWewk7z5xHb1uhL\nybHf4ljrCRTRqlcOaAdtVw5Sa449fxqFjQebBBw82MqDjQfH7cFu68FWCYH1YJ7AckIwfwKO8mCT\nGJHHUR5sy4O9z4W934dDAbYqC5xHebCVGxs3GmnNAlDosEBac3Csb+Ct7diNNS7SNGBo590s/qnm\nylG/jK1kp3WMY6lEQ2mwW9oqRr+vW7cOrTX3338/a9asobS0lDFjxoRGv2utGT16NBdffDE+n48p\nU6awc+dOlFLccsstDB48uNbzyC1tzUP+qtl8uOdkdpsn15inl/Uax1pPNGg5NODgrgyuwcAaGWiD\ny+GBtyLQqipBN7R/MD0YmKtus0lAK+kXr2Dqcgx8mHgxtQ8Df2AZH4b2YeLDVH5MrMAyFqaygmk2\nhrJwYWEqO5juYCobl7IxlYNhOLgMG1NpXMoJ1HSVg2kYwVYVs3qrSkWaESUtrGWnzrRqLTxGHeeo\naOExGuQOEK01+z+9ii2lnfjOdQ2lRmXwdutieluvcpR+hXY/fx4zsc1hP784/OQ+ddHotm7fwtsb\ncmrPpG3O6rUHjycR29HBPtfAq+2AHewvtR0dGOSkw5qOgz+WNgLL2sB2jMC6NrC1GXyVOYwqVATP\nQHD1YWpv9GV8wcDrDVuv2B4IxqFATDBdhy2H1gPnMvBLg3xNqnQvVe9Cqtr9FH6REHkBUXHh4PiK\n8OV+AwRai/YbAymlPW6KyXZW4KIcgNRevyet1/jGfPciRo3S/C5EuI0l3evOpEze33hkNQEaSmOq\nimZfHXiNqFk6uJSDoZxgLTRQE3UpCyO4Hli2MLFw2eW4vCWBn/IizPJi3OXFmHZ5qBbsCgZXlAM4\naDS6Ylnp0KvjMdHuih8Dx1RggNY2aDv0irbRTthy+KsTfMUGbTXuh90c6GBXEw3T1aTQtHW+Bb6t\nts2Xt/ownkk0FgnqIi5KfU25bhY5yKhqf6qrYpBSjf2pYdurDWSKPoCp4jX6fDEViYepH1xrVHFJ\nlWfhB6e0dQ5uiLj2uLEzw56F36piStuEGIviRAZ/x65+keBUuWDQViDQOVEuJkIXFVYwzanleFbw\n1Yl+3rDjVCtnqLxWWJoTtcyB8toRZSbqWBEhDj8J6iIukmK8U8Y0NAmuaAH28Afeim2GauEPd1MK\nnZaKlZYKXcNuu7NtjMLiwEN0cgMP0TFz8zGKS2o+lM+Pa+9+2Ls/It1JToqc5S4rIzClbZXbUQN3\nHhiA+4hqgtfBGnj4xUHlxURNFyJWlDSnyoWGBdUudiKPZ5Vsp3znh3WW0ZPZLw6fhGhoEtRFXPTJ\n1qzZVXsehebiE2zS5aFq8RGchc7JysDqGZbu94emtA09Cz+vAKO85tudjNIyjNIyXDt2h9K0Ujhp\nqZFz17fKwElLPeKmtK28mHHF/WJGa83+wvVYxcHbITWg3aD8lXcbKjfJXUbFuWSiIchAOREXWsO/\nvjXYnl/zP/NjOjiM7CvNlE1VtSltc/Mx8gtR1sH1lWvTDDwit+qUtslJLbzJpPH4CzeS//l0MvaM\nIi3/l7jsVjiqnOK0j8lv/SpJx19AcqczG7uYIkYy+l00CV4/vL/GYFte9cB+VHuHkX0cmvgD0kRV\nWqOKSjDzCzByw6a0LSg86ClttccTFugrm/FJkKecHSqVD4mvglFSvXFWGw7eX2nsns06FBxRJKiL\nJkNr2F0I6/calPshNQH6tndondLYJROHlW1jFBRVm+XOqOURuTVxUpLCnoUfrNlnpAcGU4i6aUh8\n2cTcrdDoyCcVVmRxa0ovsyG5EconDpoEdSFE0+DzV6/V5+VjeH0HdRitFE56WuTc9VkZOGkpR1x/\nfQQH8AW6y/GD8oGxW5GwpO4LIN+pNv7jm3U4OGJIUBdCNF1aB/vrKwflVQR8ZR/c1GE6OPivajO+\nTkpsWv31GrAJBV78wUDsUxEBOZAeTKsSrPFHyWvX/z1a3R2858uYluZAHj4jhGi6lEInJ2EnJ2F3\nal+Z7jiB++tzCzDyg7fd5RVgFBbVPKWtbUef0jbBEzaVbVh/vSeGR/dqKoNulWAbU6Ctmq8iQDex\n6VmVxPMWQYK6EKJpMgx0ehpWehrQuTLdsjEKCisfolNRsy8pRWsFuEC7ArdtBZeV341R5MLY6uDS\nRaDLgf1oVxLanYw2E8H0oPGAY6KsykCs6pjdrCnTbh2Y5NBb93uws+NQINHgJKgLIeIvvPk5omm5\n9qbmwLKB8rcGf5tQbdjxg+OrR/OzN3CrdmOHba00eILXIRWvbl1lHbQnLD20HgzewXwV++Ai8MYc\nSHrOxCio+V1qpbEGSFW9JZCgLoSonQYswmquFcuqep+wX1XpI64lWDuNHUrrRysbXDoQUBMUOsGo\nFmhDQTkUdKk1WGPScFcWBpT/0ibxDROjhose7zAHndlA5xdxJUFdiJbEIUpABeVTMdaKo+Uj6m1Q\nzYGOWtvVwVqsH2WXo/xl4C/FKC9BeUsAP0pZwSsYC4hcViqsP98Gx0nASc6octtdBribzlS7+0oV\nBW0VXfIhs7zy+qHEDTsywPIpumrdpMYSivqRoC7iTuWCa71ClSmcVI19lEanNnap4uyQmp9rqRUf\nwujnxtSgzc81MoGU4E+QZWHkF0beW58XeARuTYxyL8auvbh27Y1Id1JTIp+F3yojcH99nG+5s/2w\nZ42B44G12eC2wWOBZYC34jPaaVC63yGlbVyLJhqA3NIm4seChA8MXD9G/lPTSuMfovGf6jR+52Y0\n0vwcQZu6SkClxqZm7Y6S11M9KDdo8/Ph4PVGznKXm4+ZX4Dy+eveN4w2DJyMtGpPzdOpKdVuudNO\nYFI4xw7+VMzdYqnqaTY4FelV9vGVKHzFdX+4Wd0cOv9M+tWbA7mlTTQ+DQnvGbg2VK+lKK3wLA/8\n0/EPO8R/KuHNz+EB90hvfq5aq40h0EYN1hUB+EiTkIDdPhu7fTZ+Ak9GdCyNLixD5RVBXgm6oASK\nyqHYi6MNHOXCVi4czMBrxXK5C2eXC3u3C0eZOPixVTG26cEx3NiYONqAON/y5jv4h/2JJkiCuogL\nYxdRA3o49wqF004FgkZtD+Woek9wePC1mmnwDW9+jqjtxlorjpaPpl37bSBaV6nlWqGZSytrs4Hp\n1cNqtCpKXnBsFSUv6FArixtIjyzAoTxqtRErymbTGQIgDoEEdREXrh/q7kdUWpH476ZfDTwim58P\noxqblW0VEYirNSvbVQO1ipI3sNwSP0ylbUxtYWBhKI1hapRbYXgMVIILlWhiuBSGCYaLwHYz8Hnv\nWW1Q12eS0blZ98SKIAnqIi5UceOcV5qfD47W0WqxwRprtVpsZS23WnpwH10tL3FvVo4PHQykgYCq\nwgJrZZANTw9uc4HhWJjlJbhKi3GVFOEqLsBVWIDL7w0EcG1jYqGoPegG+uvTsVsFn4WfEei31ynJ\nWFvzOVDUusZ9E1Ux6R0TaIkXQ0caCeoiLnRSjPk8Gp0izc/RxLdZuWVRVYKrYYKqNehWyV8RqE0d\ncYyKdGUcyqPlTQJN+JXN+I7W+EtKA3PW5xWgK56eV1CEcqK30SvHwczLx8zLB7aE0rXbRW+/A8mn\ncsDdtdp+SXY+R5cswd59fORjekWzJKPfRVwYWxRJb9ZexdVoyi6z0RlxKtRhpp1otdgqzcrhATbG\nUcwtvlnZ0FVqsJGBVUVJV1UCbkXNt3reYNBtKR+b4wSntA2f/CYfo6ikzl01UGi2Za8nB6+Rgkv7\naO3fSmv/NgwcfDndKB9+csO/B3HIZPS7aHROV01puia5sOb/riVdNKqBAnp9mpWj3TokzcoVy7pa\nbbVqs3L1QK2jNEEHarkiRoYResCNRVit2+/HzCvEyM/HyC2ofC5+uTeURQEZ9j4yyvZFP3Qt9+KL\n5kOCuoiLskLYnKbo44V0b/Xt+5Nho6HoVRD4J18ZRKVZuTaH2qwckX7Ym5VF3Ljd2NmtsbMj+81V\nWTmeb1eTsGZ9nYfQiQkNVToRRxLURVzkbTawTFiTHQjqrUvBZYPPhH2pUOoBtGL9hy3nT7K2ZuVo\n6dUC9JHUrCwahE5KxDfwaDw/bqyxL76Cv0e3OJVKNKSW8x9UNGm+ii4/BYWJgZ/GpaPUVmtoVq5S\nm1VVgqs0K4umTCcn4TuqFwlr1tWYx26dhdW1YxxLJRqKBHURFwf7YAtpVhbi8PGeMAhlWXjWbaq2\nzWrbmrIzhhHvZ9KLhiGj30VcFOxQbP2i7hu8c06zSMqUgCtEQzDy8nGv/wmjuBTtcePv0RW7Yzv5\nwjUzMvpdNLr0DpqEdI23ltHvaR0ckrPiWCghjjBOVibeE45r7GKIBiTtLSIulAHdT7FJSI3eMJTS\nRtPleJkhSgghDoU0v4u4cmwo2K7I36awfQp3kiarmyatg5YWQCGEiIE0v4smwzAhq1sgkAshhDi8\npPldCCGEaCEkqAshhBAthAR1IYQQooWQoC6EEEK0EBLUhRBCiBZCgroQQgjRQkhQF0IIIVoICepC\nCCFECyFBXQghhGghJKgLIYQQLYQEdSGEEKKFkKAuhBBCtBAS1IUQQogWQmZpE3GV793JN/veYF3+\nfymzC0lzt6Ff1i8Y1OZ8kt1ZjV08IVosy/GxNm8xa/IWUejbQ4KZRu/MYQxofS7JrszGLp44TGQ+\ndRE3O4q/518/3YnPKa22LdXdlt/k/JnMhE6NUDIhWrZyq5C3Nt3OnrJ11bYluTI5v8dMspN7N0LJ\nRH3UNp+6NL+LuPDZZbyzeVrUgA5Q7N/HvzffRzO/xhSiSfrPtgejBnSAMiuff/50Jz67LM6lEg1B\nmt9FXPyYv5gyu6DWPPvKN/LRjsfI9FTW1pVSh3DW2vdVtW6va9+6Tl1zjtrPG4tDeF91fJ51l+xQ\n3tchnLvOv4ND+V3XdeQ69q3zM22sv+GAYv9+fir8otY8pVYuP+YvZmDrcw+qdKLpkaAu4mJb0dcx\n5fv+wDsNXBIhRDSbC5dJUG8BpPldxIWj7cYughCiFn6nvLGLIA4DqamLuGib1IuNhZ/WnS+xF6me\nNoGVWvrXNYfW9177/rUf+9D6/es4dkO+rzrKfWjnPsRj1/q7buBz17K9znPX+bfQSO8rrFxeu5hC\n/546z5aV0KXOPKLpk6Au4uKYVmfx5Z6/41BzjT3ZlcmY3rNxGZ44lkyIls1yfDzzwzjKrPxa8/Vv\nPSpOJRINSZrfRVykedoyvNO1NW43MPlFl1sloAtxmLkMD6d1uo7aBt0NanM+2Um94lco0WDkPnUR\nV+vzP+aLPS9woHxzKK1jSn9OaX8ZnVMHNl7BhGjhNhZ8xic7nyTftyOUlmimMbjthRyfPRalpI7X\nXNR2n7oEdRF3WmtyvVsot4pIdbcmI6FjYxdJiCOC1g47S1ZT5N9LgplKl9RBuIyExi6WOEgS1IUQ\nQogWQp4oJ4QQQhwBGiyoO47D3XffzZgxYxg/fjxbtmyJ2P7RRx8xevRoxowZw6uvvhrTPkIIIYSo\nWYMF9Q8//BCfz8crr7zCzTffzKxZs0Lb/H4/M2fO5Nlnn2XBggW88sor7N+/v9Z9hBBCCFG7BrtP\nfcWKFQwbNgyAQYMGsWrVqtC2jRs30rVrVzIyMgAYMmQIX331FStXrqxxHyGEEELUrsGCenFxMamp\nqaF10zSxLAuXy0VxcTFpaZUd/SkpKRQXF9e6T02yspJxucyGeRNCCCFEM9JgQT01NZWSkpLQuuM4\noeBcdVtJSQlpaWm17lOTvLzoU3kKIYQQLVGjjH4fPHgwH3/8MQArV66kT58+oW05OTls2bKF/Px8\nfD4fy5cv57jjjqt1HyGEEELUrsHuU3cch2nTprFu3Tq01tx///2sWbOG0tJSxowZw0cffcTcuXPR\nWjN69GguvvjiqPvk5OTUeh65T10IIcSRRB4+I4QQQrQQ8vAZIYQQ4gjQ7GvqQgghhAiQmroQQgjR\nQkhQF0IIIVoICepCCCFECyFBXQghhGghJKgLIYQQLYQEdSGEEKKFkKB+BHrqqac49dRT8Xq9AKxd\nu5avvvoqIs/27du56KKLDsv5du7cyUcffVQt/bTTTuPyyy+PSPvb3/5G3759D/ocX331FT/++GNM\neZctW8b48eMP+hxC1GbZsmXceOONDXb8WbNmMX78eM466yxGjBjB+PHjue666w77eW+//XZ+9rOf\n4fP5QmmrV6+mb9++LFu27KCOVdN3vyb1+e6LSBLUj0Bvv/02o0aN4t///jcAixYtYsOGDQ12vi++\n+IKvv/466ra9e/eSm5sbWl+6dGloSt6D8cYbb7B3796Y8rZp04bs7OyDPocQjen2229nwYIFTJgw\ngXPPPZcFCxYwe/bsBjlX27ZtQ/NwACxcuJAuXboc9HFq++5Hc/TRRx/0OUSkBpulTTRNy5Yto2vX\nrowdO5Zbb72VoUOH8tZbb+F2uznmmGMYOHBgtX2+/PJLHnnkEUzTpEuXLtx33314vV7uvPNOioqK\n2Lt3L+PGjWPcuHG8+OKL/POf/8QwDAYMGMCUKVOYP38+5eXlHHfccZx++ukRx/7lL3/J+++/z7hx\n49i4cSNdu3Zl/fr1AKxbt45Zs2Zh2zZ5eXlMmzaNwYMHM2XKFLZs2UJ5eTm///3v6dWrF5988gmr\nV6+mV69efPvttzz33HMYhsGQIUO45ZZbmDNnDt988w2lpaXMmDGDO++8E6/Xy/XXX09xcTFlZWXc\neOONnHrqqXH5PYgjR7Tvz/bt25kyZQoulwvHcXj44YdJSEjghhtuQGuN1+vl3nvvpV+/fjGdY8uW\nLVxxxRXk5uYycuRIJk2axJdffsnjjz+O1pqSkhIefvhh3G43N998M+3bt2fbtm0MGDCAe++9t9rx\nzjnnHN555x3OOOMMHMdh9erVDBgwAAhMq12f737nzp2ZPn06AJmZmaH5QP785z/jdru56KKLeOqp\npwB45JFHWLZsGZZlceaZZzJhwoTD9Nto+SSoH2Fee+01LrzwQnr27InH42H37t2cf/75tGnTJmpA\n11ozdepUXnrpJVq3bs2jjz7KW2+9xTHHHMM555zDmWeeyZ49exg/fjzjxo3jzTff5J577mHgwIG8\n9NJLaK2ZMGECmzZtqhbQAc4991ymTp3KuHHjePvtt/nVr37F4sWLAdiwYQOTJ0+mb9++LFy4kDff\nfJM+ffrw1Vdf8eqrrwLw6aef0r9/f4YNG8aoUaNITk5mzpw5vPHGGyQlJXHrrbfy6aefAtCzZ0/u\nuuuu0LnXr19Pfn4+Tz/9NAcOHGDz5s0N8ImLI1lN3x+/38/AgQO59dZbWb58OUVFRaxdu5bMzEwe\nfPBBNmzYQGlp7NNKe71ennjiCWzbZsSIEUyaNIn169fz0EMP0a5dO/7617/y/vvv86tf/YrNmzfz\nzDPPkJSUxBlnnMG+ffto27ZtxPEGDhzIokWLKC0tZeXKlZx44ols3LgRCFxA1Oe7f9FFF3H//ffT\nq1cvXnvtNZ5++mlOOeUUvF4vr732WsT5Fy5cyAsvvEB2djZvvvnmof8ijiAS1I8gBQUFfPzxx+Tm\n5rJgwQKKi4v5+9//TteuXWvcJzc3l71793LDDTcAUF5ezimnnMLw4cN5/vnnWbRoEampqViWBcDM\nmTN59tlnefDBBxk0aBB1PYW4Q4cOAOzatYuvv/46dB6A7OxsnnjiCRITEykpKSE1NZXU1FTuuOMO\npk6dSnFxMeedd17E8bZu3Upubm7oyr6kpIStW7cC0KNHj4i8vXv3ZsyYMdx0001YliX97OKwq+n7\nc+211/LUU09xxRVXkJaWxo033sjPf/5zNm/ezLXXXovL5eKaa66J+Ty9e/fG4/EA4HIF/q23a9eO\nGTNmkJyczJ49exg8eDAAXbt2JTU1FQg0s1eMranq9NNPZ/HixXz22Wdce+21/OUvfwEC3Vf1+e5v\n3Lgx1Crg9/vp3r07UP17CfDQQw/x8MMPs3//foYNGxbz5yAkqB9R3n77bUaPHs3kyZMBKCsr4/TT\nT6dbt244jhN1n6ysLNq3b88TTzxBWloaixcvJjk5mWeffZZBgwYxbtw4vvjiC5YuXQrAq6++yr33\n3ktCQgKXX34533zzDYZh1Hh8gFGjRjFr1iyOO+44lFKh9BkzZvDnP/+ZnJwcZs+ezY4dO9i7dy+r\nV69m7ty5eL1ehg8fzq9//WuUUmit6dy5Mx06dODZZ5/F7Xbz5ptv0q9fPz788EMMI3IIydq1aykp\nKWH+/Pns3buXsWPHMnLkyEP9mIUIqen7s3jxYoYMGcLEiRN55513ePrppznvvPPIzs7m2Wef5Ztv\nvuEvf/kLCxYsiOk84d+bClOnTuWDDz4gNTWVyZMnh4JstLzRnHvuudx///0opSL60+v73e/RowcP\nPPAAHTt2ZMWKFezbtw+g2vfS5/Px/vvvhy4iRo0axTnnnEOnTp1iKveRToL6EeS1116L3nIWAAAE\nOUlEQVTjwQcfDK0nJSVx5plnYpomL774Ijk5OZx00kkR+xiGwZ133smECRPQWpOSksKDDz6IUorp\n06fz7rvvkpaWhmma+Hw++vbty7hx40hJSaFdu3Yce+yxpKamMm/evFCTfVVnnXUWM2bM4J///GdE\n+nnnncf1119Peno67du3Jy8vj7Zt27Jv3z7Gjh2LYRhcdtlluFwujj32WP785z/z6KOPcskllzB+\n/Hhs26ZTp06cffbZUT+P7t27M3fuXN577z0cx+G66647DJ+yOJJ9+umnXHDBBaH1hx9+OOr3p6Sk\nhMmTJzNv3jwcx2HKlCl07NiRm266iX/84x9YlsUf//jHQyrLeeedx8UXX0xSUhJt2rSJeSBphZyc\nHPLy8hg9enRE+siRI+v13Z82bRqTJ0/GsiyUUsyYMSNqmTweDxkZGVx00UUkJiYydOhQOnbseEif\nxZFEZmkTQgghWgi5pU0IIYRoISSoCyGEEC2EBHUhhBCihZCgLoQQQrQQEtSFEEKIFkKCuhCimtmz\nZ7N8+fJa83z00Uf87W9/i1OJhBCxkKAuhKjmq6++wrbtWvOsXr2a4uLiOJVICBELefiMEEe43bt3\nc8stt1BaWophGIwYMYJVq1Zx11138fjjj1NQUMAjjzxCeXk5BQUF3HrrrfTu3ZuXX34ZgI4dO7Jz\n504AJk2aBASm1X3hhRcoLi7m7rvvxrIsEhISmDlzZujxoEKIw0+CuhBHuNdff50RI0ZwxRVXsGzZ\nMlatWkX//v2ZOHEiffv25brrrmP69Onk5OTw+eefc//997Nw4ULGjh0LwOjRo5kzZ07UYz///PNc\neumlnH322bz77rusXLlSgroQDUiCuhBHuJNPPplJkybxww8/MHz4cH73u9/x3//+N7T9oYceYsmS\nJbz//vt8++23lJSUxHzs4cOHc9999/HJJ58wcuRIfvnLXzbAOxBCVJA+dSGOcEOGDOHf//43p556\nKu+++y5XX311xPZx48bx3Xff0b9//2rbKlRMqFPB7/cDgef6v/XWWwwcOJDnn3+ee+65p+HeiBBC\naupCHOkefPBBsrOzueSSSzjxxBM5//zz6devH7Ztk5+fz+bNm3nppZdISEhgzpw5oQF0pmmGpu3M\nyspi2bJlAHz33XehGbhuuOEGzjnnHMaOHUtOTg4zZ85snDcpxBFCJnQR4gi3a9cubr75ZkpKSjBN\nkyuuuIJdu3bx8ssv88ADD7Bo0SI+/PBDUlNTGTRoEO+99x5Llixh9erVTJ48mUsvvZRzzz2X66+/\nnv3793PMMcewceNGZs+eTXFxMXfeeSeO42CaJrfddhsnnHBCY79lIVosCepCCCFECyF96kIIIUQL\nIUFdCCGEaCEkqAshhBAthAR1IYQQooWQoC6EEEK0EBLUhRBCiBZCgroQQgjRQkhQF0IIIVqI/wfm\n97sud/y2JQAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.pointplot(data=df_dotplot, x='status', y='prop', hue='Method');\n" ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnYAAAFJCAYAAAAWkGAcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIBAIKglC2AkEAoFAIBBUEv4fsCwcmM5O6lcAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "for idx, row in df_dotplot.iterrows():\n", " plt.plot([0,1],[row[\"At least Master's\"], row[\"Less Than Master's\"]]);\n", " plt.text(1.05, row[\"Less Than Master's\"], row['Method']);\n", "plt.xticks([0,1], [\"At least Master's\", \"Less Than Master's\"]);" ] }, { "cell_type": "code", "execution_count": 75, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Method\n", "At least Master's\n", "Less Than Master's\n" ] } ], "source": [ "for row in df_dotplot:\n", " print(row)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.1" } }, "nbformat": 4, "nbformat_minor": 2 }