{ "cells": [ { "cell_type": "markdown", "id": "8d90fd8d-5228-4349-87ff-7acb66178eec", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "source": [ "### Fitting latent variables onto data - Exercises & Answers" ] }, { "cell_type": "markdown", "id": "6519e989-c91c-4a1a-b086-bf3dae0c25b1", "metadata": {}, "source": [ "### 1. Grit revisited - are we *sure* its there?\n", "Armed with new knowledge of CFA, we can now examine directly whether the two-factor structure claimed in the research literature exists in the data we have to hand. While our use of EFA indirectly explored the latent variables in that dataset, with CFA we explicitly test the presence of two factors and set certain questions to load onto these variables. \n", "\n", "First, lets re-obtain the data that has the grit questionnaire in it.\n", "\n", "Download the data from this link: https://openpsychometrics.org/_rawdata/duckworth-grit-scale-data.zip\n", "\n", "You will need to unzip it and grab the `data.csv` file.\n", "\n", "Import everything we need first, adding `semopy` to the list of packages we will use.\n", "\n", "### Important ###\n", "After importing everything, run this command just underneath your imports. It'll ensure the results will match when running this code elsehwere.\n", "`np.random.seed(36)`" ] }, { "cell_type": "code", "execution_count": 1, "id": "0ff2cc71-3c36-44c8-a7f1-56bb7f7cd3a3", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "# Your answer here\n", "# Import what we need\n", "import pandas as pd # dataframes\n", "import seaborn as sns # plots\n", "import statsmodels.formula.api as smf # Models\n", "import marginaleffects as me # marginal effects\n", "import numpy as np # numpy for some functions\n", "import pingouin as pg\n", "from factor_analyzer import FactorAnalyzer # Note we write from factor_analyzer\n", "from horns import parallel_analysis\n", "import semopy as sem # semopy imported here\n", "\n", "np.random.seed(36)" ] }, { "cell_type": "markdown", "id": "fa9d704e-487a-4b6f-8e97-1c862af625b9", "metadata": {}, "source": [ "Read in the data into a dataframe called grit, specifying the separator as '\\t' (sep='\\t'), as before." ] }, { "cell_type": "code", "execution_count": 2, "id": "219ffa16-6fa0-4dff-9a44-a1afd22c829c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
countrysurveyelapseGS1GS2GS3GS4GS5GS6GS7GS8...O7O8O9O10operatingsystembrowserscreenwscreenhintroelapsetestelapse
0RO17411333231...5454WindowsChrome136676869590307
1US12022332133...4345MacintoshChrome128080033657134
2US9933334344...5544WindowsFirefox1920108095550138
3KE509813424154...4254WindowsChrome160090044440
4JP34012332224...4132WindowsFirefox192010803337
5AU51512513145...5255WindowsChrome192010802090554
6US12621341111...5555WindowsChrome136676836212
7RO20831141344...5343WindowsChrome13667686207
8EU13013314154...5145WindowsMicrosoft Internet Explorer1600100014183
9NZ12923224244...4344MacintoshChrome144090068143
\n", "

10 rows × 98 columns

\n", "
" ], "text/plain": [ " country surveyelapse GS1 GS2 GS3 GS4 GS5 GS6 GS7 GS8 ... O7 O8 \\\n", "0 RO 174 1 1 3 3 3 2 3 1 ... 5 4 \n", "1 US 120 2 2 3 3 2 1 3 3 ... 4 3 \n", "2 US 99 3 3 3 3 4 3 4 4 ... 5 5 \n", "3 KE 5098 1 3 4 2 4 1 5 4 ... 4 2 \n", "4 JP 340 1 2 3 3 2 2 2 4 ... 4 1 \n", "5 AU 515 1 2 5 1 3 1 4 5 ... 5 2 \n", "6 US 126 2 1 3 4 1 1 1 1 ... 5 5 \n", "7 RO 208 3 1 1 4 1 3 4 4 ... 5 3 \n", "8 EU 130 1 3 3 1 4 1 5 4 ... 5 1 \n", "9 NZ 129 2 3 2 2 4 2 4 4 ... 4 3 \n", "\n", " O9 O10 operatingsystem browser screenw screenh \\\n", "0 5 4 Windows Chrome 1366 768 \n", "1 4 5 Macintosh Chrome 1280 800 \n", "2 4 4 Windows Firefox 1920 1080 \n", "3 5 4 Windows Chrome 1600 900 \n", "4 3 2 Windows Firefox 1920 1080 \n", "5 5 5 Windows Chrome 1920 1080 \n", "6 5 5 Windows Chrome 1366 768 \n", "7 4 3 Windows Chrome 1366 768 \n", "8 4 5 Windows Microsoft Internet Explorer 1600 1000 \n", "9 4 4 Macintosh Chrome 1440 900 \n", "\n", " introelapse testelapse \n", "0 69590 307 \n", "1 33657 134 \n", "2 95550 138 \n", "3 4 4440 \n", "4 3 337 \n", "5 2090 554 \n", "6 36 212 \n", "7 6 207 \n", "8 14 183 \n", "9 68 143 \n", "\n", "[10 rows x 98 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Read in\n", "grit = pd.read_csv('data.csv', sep='\\t')\n", "grit.head(10)" ] }, { "cell_type": "markdown", "id": "0a4685da-ef39-4d1d-9e06-74ba2c037e2e", "metadata": {}, "source": [ "Like before, get the grit-related columns by keeping only the columns with 'GS' in them. Store in a dataframe called `grit2`." ] }, { "cell_type": "code", "execution_count": 3, "id": "3abe182d-0109-445c-ab6e-c0b25154612f", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
GS1GS2GS3GS4GS5GS6GS7GS8GS9GS10GS11GS12
0113332313233
1223321332132
2333343443333
3134241541131
4123322243344
\n", "
" ], "text/plain": [ " GS1 GS2 GS3 GS4 GS5 GS6 GS7 GS8 GS9 GS10 GS11 GS12\n", "0 1 1 3 3 3 2 3 1 3 2 3 3\n", "1 2 2 3 3 2 1 3 3 2 1 3 2\n", "2 3 3 3 3 4 3 4 4 3 3 3 3\n", "3 1 3 4 2 4 1 5 4 1 1 3 1\n", "4 1 2 3 3 2 2 2 4 3 3 4 4" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "grit2 = grit.filter(regex='GS\\d+')\n", "grit2.head()" ] }, { "cell_type": "markdown", "id": "5ef22dae-8ca2-430e-8500-a5a3947d3b9b", "metadata": {}, "source": [ "### Building a latent factor model of grit\n", "We take the model directly stated from the original Duckworth et al (2007) paper that described the grit scale. The questions loading onto the two latent grit factors are as follows:\n", "- Consistency in interest\n", " - GS2\n", " - GS3\n", " - GS5\n", " - GS7\n", " - GS8\n", " - GS11\n", "- Perseverance in effort\n", " - GS1\n", " - GS4\n", " - GS6\n", " - GS9\n", " - GS10\n", " - GS12\n", "\n", "With that in mind, create a CFA model that tests whether the two latent variables (consistency and perseverance) are measured by the above stated questionnaire variables. Create it and fit it in `semopy`." ] }, { "cell_type": "code", "execution_count": 4, "id": "3970aad3-24e6-4a3e-96c8-698bc20dbb3d", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "SolverResult(fun=0.3782276478265812, success=True, n_it=24, x=array([ 1.122537 , 1.36719906, 1.39375519, 1.44706539, 0.80596871,\n", " 0.85883226, 1.32252661, 1.43824065, 1.57281968, 1.28650847,\n", " 0.68358768, 1.09217894, 0.98222584, 0.57969277, 0.8487556 ,\n", " 0.96281907, 1.19213773, 0.82104992, 0.57220757, 0.61039529,\n", " 0.78652695, 0.66075116, 0.45345085, -0.2184887 , 0.29510074]), message='Optimization terminated successfully', name_method='SLSQP', name_obj='MLW')" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Model string\n", "mdspec = \"\"\"\n", "consistency =~ GS2 + GS3 + GS5 + GS7 + GS8 + GS11\n", "perseverance =~ GS1 + GS4 + GS6 + GS9 + GS10 + GS12\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(grit2)" ] }, { "cell_type": "markdown", "id": "8ab0363d-affb-4c97-a8e4-b0afa1bca164", "metadata": {}, "source": [ "If you have fitted the model successfully, inspect the standardised loadings. Do they appear significant and sensible?" ] }, { "cell_type": "code", "execution_count": 5, "id": "b19561e1-da35-45d0-96c7-b4427c64cfbb", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
lvaloprvalEstimateEst. StdStd. Errz-valuep-value
0GS2~consistency1.0000000.590099---
1GS3~consistency1.1225370.6102720.03590431.2648170.0
2GS5~consistency1.3671990.7127110.03937534.722140.0
3GS7~consistency1.3937550.7685580.03840136.2951860.0
4GS8~consistency1.4470650.7395570.04074835.5126110.0
5GS11~consistency0.8059690.4803140.03097626.0194010.0
6GS1~perseverance1.0000000.549115---
7GS4~perseverance0.8588320.3929290.04118120.854890.0
8GS6~perseverance1.3225270.6886570.04298730.7658740.0
9GS9~perseverance1.4382410.6929690.04659430.8674410.0
10GS10~perseverance1.5728200.6329470.05362429.3307660.0
11GS12~perseverance1.2865080.6762210.04222930.4648430.0
12consistency~~consistency0.4534511.0000000.02311619.6163050.0
13consistency~~perseverance-0.218489-0.5972810.010538-20.7324450.0
14perseverance~~perseverance0.2951011.0000000.01676217.6058340.0
15GS1~~GS10.6835880.6984730.01643441.5971620.0
16GS10~~GS101.0921790.5993780.02798539.0276270.0
17GS11~~GS110.9822260.7692990.0224943.6729730.0
18GS12~~GS120.5796930.5427250.01561437.1266570.0
19GS2~~GS20.8487560.6517830.0203741.6674570.0
20GS3~~GS30.9628190.6275680.02339441.1572120.0
21GS4~~GS41.1921380.8456070.02692944.2694720.0
22GS5~~GS50.8210500.4920430.02199137.3360670.0
23GS6~~GS60.5722080.5257510.01568636.4783810.0
24GS7~~GS70.6103950.4093180.01807233.775120.0
25GS8~~GS80.7865270.4530550.02196635.8064840.0
26GS9~~GS90.6607510.5197940.01823236.2413370.0
\n", "
" ], "text/plain": [ " lval op rval Estimate Est. Std Std. Err z-value \\\n", "0 GS2 ~ consistency 1.000000 0.590099 - - \n", "1 GS3 ~ consistency 1.122537 0.610272 0.035904 31.264817 \n", "2 GS5 ~ consistency 1.367199 0.712711 0.039375 34.72214 \n", "3 GS7 ~ consistency 1.393755 0.768558 0.038401 36.295186 \n", "4 GS8 ~ consistency 1.447065 0.739557 0.040748 35.512611 \n", "5 GS11 ~ consistency 0.805969 0.480314 0.030976 26.019401 \n", "6 GS1 ~ perseverance 1.000000 0.549115 - - \n", "7 GS4 ~ perseverance 0.858832 0.392929 0.041181 20.85489 \n", "8 GS6 ~ perseverance 1.322527 0.688657 0.042987 30.765874 \n", "9 GS9 ~ perseverance 1.438241 0.692969 0.046594 30.867441 \n", "10 GS10 ~ perseverance 1.572820 0.632947 0.053624 29.330766 \n", "11 GS12 ~ perseverance 1.286508 0.676221 0.042229 30.464843 \n", "12 consistency ~~ consistency 0.453451 1.000000 0.023116 19.616305 \n", "13 consistency ~~ perseverance -0.218489 -0.597281 0.010538 -20.732445 \n", "14 perseverance ~~ perseverance 0.295101 1.000000 0.016762 17.605834 \n", "15 GS1 ~~ GS1 0.683588 0.698473 0.016434 41.597162 \n", "16 GS10 ~~ GS10 1.092179 0.599378 0.027985 39.027627 \n", "17 GS11 ~~ GS11 0.982226 0.769299 0.02249 43.672973 \n", "18 GS12 ~~ GS12 0.579693 0.542725 0.015614 37.126657 \n", "19 GS2 ~~ GS2 0.848756 0.651783 0.02037 41.667457 \n", "20 GS3 ~~ GS3 0.962819 0.627568 0.023394 41.157212 \n", "21 GS4 ~~ GS4 1.192138 0.845607 0.026929 44.269472 \n", "22 GS5 ~~ GS5 0.821050 0.492043 0.021991 37.336067 \n", "23 GS6 ~~ GS6 0.572208 0.525751 0.015686 36.478381 \n", "24 GS7 ~~ GS7 0.610395 0.409318 0.018072 33.77512 \n", "25 GS8 ~~ GS8 0.786527 0.453055 0.021966 35.806484 \n", "26 GS9 ~~ GS9 0.660751 0.519794 0.018232 36.241337 \n", "\n", " p-value \n", "0 - \n", "1 0.0 \n", "2 0.0 \n", "3 0.0 \n", "4 0.0 \n", "5 0.0 \n", "6 - \n", "7 0.0 \n", "8 0.0 \n", "9 0.0 \n", "10 0.0 \n", "11 0.0 \n", "12 0.0 \n", "13 0.0 \n", "14 0.0 \n", "15 0.0 \n", "16 0.0 \n", "17 0.0 \n", "18 0.0 \n", "19 0.0 \n", "20 0.0 \n", "21 0.0 \n", "22 0.0 \n", "23 0.0 \n", "24 0.0 \n", "25 0.0 \n", "26 0.0 " ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "model.inspect(std_est=True)" ] }, { "cell_type": "markdown", "id": "904f44bb-5129-4e72-9145-dbcbf8804af2", "metadata": {}, "source": [ "Finally, check the fit statistics. Is this model any good? Does it describe the data well?" ] }, { "cell_type": "code", "execution_count": 6, "id": "eae80e83-7860-4abe-b131-11f77caa2725", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
DoFDoF Baselinechi2chi2 p-valuechi2 BaselineCFIGFIAGFINFITLIRMSEAAICBICLogLik
Value53661615.0320560.015958.9625240.9017150.8988010.8739790.8988010.8776080.08308949.243545208.2277720.378228
\n", "
" ], "text/plain": [ " DoF DoF Baseline chi2 chi2 p-value chi2 Baseline CFI \\\n", "Value 53 66 1615.032056 0.0 15958.962524 0.901715 \n", "\n", " GFI AGFI NFI TLI RMSEA AIC \\\n", "Value 0.898801 0.873979 0.898801 0.877608 0.083089 49.243545 \n", "\n", " BIC LogLik \n", "Value 208.227772 0.378228 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "sem.calc_stats(model)" ] }, { "cell_type": "markdown", "id": "20c9153a-fd0b-4096-b55c-4a5aa6fae1d7", "metadata": {}, "source": [ "What do the guidelines suggest we should do with this model? Do the statistics suggest it fits the model well?" ] }, { "cell_type": "code", "execution_count": 7, "id": "ca0c784c-40ce-4298-9f1d-0e4f1e7a64cb", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "# Your answer here\n", "# Not fully. its close in many regards but only just squeaks across the line." ] }, { "cell_type": "markdown", "id": "b377e9d2-88d0-4c24-8cd1-d4032efd84fa", "metadata": {}, "source": [ "### 2. Grit and conscientiousness\n", "Let us now expand the use of CFA and dip into the idea of SEM a little. Here we'll expand our previous exercise where we conducted an EFA on scores from the Big 5 trait Conscientiousness at the same time as the grit scale. In the following we'll see what happens if we specify our grit model like we just did, but also include the data for Conscientiousness and its associated questions. As such, we'll find three latent variables (the two grit related ones from the grit questionnaire, and one for Conscientiousness). \n", "\n", "First, get the right questions out of the main `grit` dataframe and store it in a dataframe called `grit_consc`. It should include all questions with `GS` in them and `C`." ] }, { "cell_type": "code", "execution_count": 8, "id": "c075049a-d575-4404-ac70-cf3e57129cd3", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
GS1GS2GS3GS4GS5GS6GS7GS8GS9GS10...C1C2C3C4C5C6C7C8C9C10
01133323132...2443243224
12233213321...4343135253
23333434433...2242345334
31342415411...4151414143
41233222433...3131423234
\n", "

5 rows × 22 columns

\n", "
" ], "text/plain": [ " GS1 GS2 GS3 GS4 GS5 GS6 GS7 GS8 GS9 GS10 ... C1 C2 C3 C4 C5 \\\n", "0 1 1 3 3 3 2 3 1 3 2 ... 2 4 4 3 2 \n", "1 2 2 3 3 2 1 3 3 2 1 ... 4 3 4 3 1 \n", "2 3 3 3 3 4 3 4 4 3 3 ... 2 2 4 2 3 \n", "3 1 3 4 2 4 1 5 4 1 1 ... 4 1 5 1 4 \n", "4 1 2 3 3 2 2 2 4 3 3 ... 3 1 3 1 4 \n", "\n", " C6 C7 C8 C9 C10 \n", "0 4 3 2 2 4 \n", "1 3 5 2 5 3 \n", "2 4 5 3 3 4 \n", "3 1 4 1 4 3 \n", "4 2 3 2 3 4 \n", "\n", "[5 rows x 22 columns]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Get grit and conscientiousness\n", "grit_consc_names = ['GS1', 'GS2', 'GS3', 'GS4', 'GS5', 'GS6', 'GS7', 'GS8', 'GS9', 'GS10', 'GS11', 'GS12',\n", " 'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10']\n", "\n", "# Extract them\n", "grit_consc = grit[grit_consc_names]\n", "\n", "# Show \n", "grit_consc.head()" ] }, { "cell_type": "markdown", "id": "93e185be-f405-4872-8703-158dffc4b13d", "metadata": {}, "source": [ "Set up a CFA model in which the two grit latent variables are captured exactly like in the last exercise, and the conscientiousness variable is captured by its own questions." ] }, { "cell_type": "code", "execution_count": 9, "id": "fd06cef2-8e94-4c8e-a203-df98e9129393", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "SolverResult(fun=0.9710707937399814, success=True, n_it=52, x=array([ 1.09309815, 1.33427212, 1.36288773, 1.42898751, 0.78738655,\n", " 0.85031695, 1.39139981, 1.49488101, 1.55740394, 1.34510828,\n", " -0.94987594, 0.48332316, -1.02903567, 1.07197578, -1.08683903,\n", " 0.76819026, -0.81086835, 1.03272723, 0.65672684, 0.75008356,\n", " 0.79145626, 1.3814536 , 0.83952471, 1.02987412, 0.9654017 ,\n", " 1.30351542, 0.96074342, 0.90503006, 0.95851657, 0.70041189,\n", " 1.14501598, 0.98512611, 0.56303226, 0.83185583, 0.97223287,\n", " 1.20778227, 0.83141406, 0.54799373, 0.61779007, 0.77584295,\n", " 0.6474266 , 0.56482701, -0.29713507, 0.47013957, 0.32249194,\n", " -0.21642159, 0.27898365]), message='Optimization terminated successfully', name_method='SLSQP', name_obj='MLW')" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Model string\n", "mdspec = \"\"\"\n", "consistency =~ GS2 + GS3 + GS5 + GS7 + GS8 + GS11\n", "perseverance =~ GS1 + GS4 + GS6 + GS9 + GS10 + GS12\n", "conscientious =~ C1 + C2 + C3 + C4 + C5 + C6 + C7 + C8 + C9 + C10\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(grit_consc)" ] }, { "cell_type": "markdown", "id": "57cc7567-0248-4262-a465-c72e43d1a8bb", "metadata": {}, "source": [ "Inspect the standardised loadings once the model has been estimated." ] }, { "cell_type": "code", "execution_count": 10, "id": "b9957844-a36d-41b5-8564-9c0662f5b9fa", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
lvaloprvalEstimateEst. StdStd. Errz-valuep-value
0GS2~consistency1.0000000.600909---
1GS3~consistency1.0930980.6051480.03455331.6352240.0
2GS5~consistency1.3342720.7082860.03774435.3505610.0
3GS7~consistency1.3628880.7652910.03673637.0993890.0
4GS8~consistency1.4289880.7436750.03918436.4683260.0
5GS11~consistency0.7873870.4778300.02999326.2519820.0
6GS1~perseverance1.0000000.533716---
7GS4~perseverance0.8503170.3783010.04187420.3067530.0
8GS6~perseverance1.3914000.7045410.04486831.0110950.0
9GS9~perseverance1.4948810.7004010.04835130.9169780.0
10GS10~perseverance1.5574040.6094710.05449728.5777710.0
11GS12~perseverance1.3451080.6875450.04393330.61750.0
12C1~conscientious1.0000000.655405---
13C2~conscientious-0.949876-0.5191230.032091-29.5992850.0
14C3~conscientious0.4833230.3685370.02236621.6093930.0
15C4~conscientious-1.029036-0.6061270.030383-33.8689950.0
16C5~conscientious1.0719760.6340580.03047435.1762780.0
17C6~conscientious-1.086839-0.5818520.03323-32.7066050.0
18C7~conscientious0.7681900.5075160.02648129.0089590.0
19C8~conscientious-0.810868-0.5394030.026482-30.6195150.0
20C9~conscientious1.0327270.6212310.02986534.5800680.0
21C10~conscientious0.6567270.4851320.02357427.8576260.0
22conscientious~~conscientious0.5648271.0000000.02511522.4897180.0
23conscientious~~perseverance-0.297135-0.7485260.012762-23.2836670.0
24consistency~~consistency0.4701401.0000000.02336920.1184690.0
25consistency~~conscientious0.3224920.6258180.01395623.1077270.0
26consistency~~perseverance-0.216422-0.5975820.010431-20.747020.0
27perseverance~~perseverance0.2789841.0000000.01615217.2723950.0
28C1~~C10.7500840.5704450.01865440.2100650.0
29C10~~C100.7914560.7646470.01808343.7671340.0
30C2~~C21.3814540.7305120.03191843.2807760.0
31C3~~C30.8395250.8641810.01867144.9629120.0
32C4~~C41.0298740.6326100.02476341.5896920.0
33C5~~C50.9654020.5979710.02362940.8567070.0
34C6~~C61.3035150.6614480.03093342.1406380.0
35C7~~C70.9607430.7424270.02210943.455710.0
36C8~~C80.9050300.7090440.02107142.9505410.0
37C9~~C90.9585170.6140720.02326141.2078290.0
38GS1~~GS10.7004120.7151470.01646742.5342520.0
39GS10~~GS101.1450160.6285450.02810940.7341820.0
40GS11~~GS110.9851260.7716780.02249343.7965320.0
41GS12~~GS120.5630320.5272820.01487437.853950.0
42GS2~~GS20.8318560.6389080.02000841.5763130.0
43GS3~~GS30.9722330.6337950.02344341.4715650.0
44GS4~~GS41.2077820.8568880.02703544.6745340.0
45GS5~~GS50.8314140.4983320.02194637.8849970.0
46GS6~~GS60.5479940.5036220.01480637.0120810.0
47GS7~~GS70.6177900.4143300.01791434.4857050.0
48GS8~~GS80.7758430.4469470.02158335.947570.0
49GS9~~GS90.6474270.5094390.01739237.2261830.0
\n", "
" ], "text/plain": [ " lval op rval Estimate Est. Std Std. Err z-value \\\n", "0 GS2 ~ consistency 1.000000 0.600909 - - \n", "1 GS3 ~ consistency 1.093098 0.605148 0.034553 31.635224 \n", "2 GS5 ~ consistency 1.334272 0.708286 0.037744 35.350561 \n", "3 GS7 ~ consistency 1.362888 0.765291 0.036736 37.099389 \n", "4 GS8 ~ consistency 1.428988 0.743675 0.039184 36.468326 \n", "5 GS11 ~ consistency 0.787387 0.477830 0.029993 26.251982 \n", "6 GS1 ~ perseverance 1.000000 0.533716 - - \n", "7 GS4 ~ perseverance 0.850317 0.378301 0.041874 20.306753 \n", "8 GS6 ~ perseverance 1.391400 0.704541 0.044868 31.011095 \n", "9 GS9 ~ perseverance 1.494881 0.700401 0.048351 30.916978 \n", "10 GS10 ~ perseverance 1.557404 0.609471 0.054497 28.577771 \n", "11 GS12 ~ perseverance 1.345108 0.687545 0.043933 30.6175 \n", "12 C1 ~ conscientious 1.000000 0.655405 - - \n", "13 C2 ~ conscientious -0.949876 -0.519123 0.032091 -29.599285 \n", "14 C3 ~ conscientious 0.483323 0.368537 0.022366 21.609393 \n", "15 C4 ~ conscientious -1.029036 -0.606127 0.030383 -33.868995 \n", "16 C5 ~ conscientious 1.071976 0.634058 0.030474 35.176278 \n", "17 C6 ~ conscientious -1.086839 -0.581852 0.03323 -32.706605 \n", "18 C7 ~ conscientious 0.768190 0.507516 0.026481 29.008959 \n", "19 C8 ~ conscientious -0.810868 -0.539403 0.026482 -30.619515 \n", "20 C9 ~ conscientious 1.032727 0.621231 0.029865 34.580068 \n", "21 C10 ~ conscientious 0.656727 0.485132 0.023574 27.857626 \n", "22 conscientious ~~ conscientious 0.564827 1.000000 0.025115 22.489718 \n", "23 conscientious ~~ perseverance -0.297135 -0.748526 0.012762 -23.283667 \n", "24 consistency ~~ consistency 0.470140 1.000000 0.023369 20.118469 \n", "25 consistency ~~ conscientious 0.322492 0.625818 0.013956 23.107727 \n", "26 consistency ~~ perseverance -0.216422 -0.597582 0.010431 -20.74702 \n", "27 perseverance ~~ perseverance 0.278984 1.000000 0.016152 17.272395 \n", "28 C1 ~~ C1 0.750084 0.570445 0.018654 40.210065 \n", "29 C10 ~~ C10 0.791456 0.764647 0.018083 43.767134 \n", "30 C2 ~~ C2 1.381454 0.730512 0.031918 43.280776 \n", "31 C3 ~~ C3 0.839525 0.864181 0.018671 44.962912 \n", "32 C4 ~~ C4 1.029874 0.632610 0.024763 41.589692 \n", "33 C5 ~~ C5 0.965402 0.597971 0.023629 40.856707 \n", "34 C6 ~~ C6 1.303515 0.661448 0.030933 42.140638 \n", "35 C7 ~~ C7 0.960743 0.742427 0.022109 43.45571 \n", "36 C8 ~~ C8 0.905030 0.709044 0.021071 42.950541 \n", "37 C9 ~~ C9 0.958517 0.614072 0.023261 41.207829 \n", "38 GS1 ~~ GS1 0.700412 0.715147 0.016467 42.534252 \n", "39 GS10 ~~ GS10 1.145016 0.628545 0.028109 40.734182 \n", "40 GS11 ~~ GS11 0.985126 0.771678 0.022493 43.796532 \n", "41 GS12 ~~ GS12 0.563032 0.527282 0.014874 37.85395 \n", "42 GS2 ~~ GS2 0.831856 0.638908 0.020008 41.576313 \n", "43 GS3 ~~ GS3 0.972233 0.633795 0.023443 41.471565 \n", "44 GS4 ~~ GS4 1.207782 0.856888 0.027035 44.674534 \n", "45 GS5 ~~ GS5 0.831414 0.498332 0.021946 37.884997 \n", "46 GS6 ~~ GS6 0.547994 0.503622 0.014806 37.012081 \n", "47 GS7 ~~ GS7 0.617790 0.414330 0.017914 34.485705 \n", "48 GS8 ~~ GS8 0.775843 0.446947 0.021583 35.94757 \n", "49 GS9 ~~ GS9 0.647427 0.509439 0.017392 37.226183 \n", "\n", " p-value \n", "0 - \n", "1 0.0 \n", "2 0.0 \n", "3 0.0 \n", "4 0.0 \n", "5 0.0 \n", "6 - \n", "7 0.0 \n", "8 0.0 \n", "9 0.0 \n", "10 0.0 \n", "11 0.0 \n", "12 - \n", "13 0.0 \n", "14 0.0 \n", "15 0.0 \n", "16 0.0 \n", "17 0.0 \n", "18 0.0 \n", "19 0.0 \n", "20 0.0 \n", "21 0.0 \n", "22 0.0 \n", "23 0.0 \n", "24 0.0 \n", "25 0.0 \n", "26 0.0 \n", "27 0.0 \n", "28 0.0 \n", "29 0.0 \n", "30 0.0 \n", "31 0.0 \n", "32 0.0 \n", "33 0.0 \n", "34 0.0 \n", "35 0.0 \n", "36 0.0 \n", "37 0.0 \n", "38 0.0 \n", "39 0.0 \n", "40 0.0 \n", "41 0.0 \n", "42 0.0 \n", "43 0.0 \n", "44 0.0 \n", "45 0.0 \n", "46 0.0 \n", "47 0.0 \n", "48 0.0 \n", "49 0.0 " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "model.inspect(std_est=True)" ] }, { "cell_type": "markdown", "id": "7fc695de-1919-4e0d-8001-0c5ab5f38c14", "metadata": {}, "source": [ "Take a look at the model fit statistics now. What does this suggest to us?" ] }, { "cell_type": "code", "execution_count": 11, "id": "76f4ef04-79f0-42a1-822a-e8b0e97e448d", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
DoFDoF Baselinechi2chi2 p-valuechi2 BaselineCFIGFIAGFINFITLIRMSEAAICBICLogLik
Value2062314146.4722890.029702.881490.8662970.8604020.843460.8604020.8500710.06693992.057858390.9482060.971071
\n", "
" ], "text/plain": [ " DoF DoF Baseline chi2 chi2 p-value chi2 Baseline CFI \\\n", "Value 206 231 4146.472289 0.0 29702.88149 0.866297 \n", "\n", " GFI AGFI NFI TLI RMSEA AIC BIC \\\n", "Value 0.860402 0.84346 0.860402 0.850071 0.066939 92.057858 390.948206 \n", "\n", " LogLik \n", "Value 0.971071 " ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "sem.calc_stats(model)" ] }, { "cell_type": "markdown", "id": "561b1c2a-7b46-49d8-abb0-54ca9dad3cdf", "metadata": {}, "source": [ "What should we conclude? Do these three separate latent factors represent the data well?" ] }, { "cell_type": "code", "execution_count": 12, "id": "9d34389d-6e5e-49c4-9475-7d8e626b451a", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "# Your answer here\n", "# Not quite! " ] }, { "cell_type": "markdown", "id": "e39a5b9c-9e4c-4ae2-ac14-627596b5c5fd", "metadata": {}, "source": [ "### 3. Latent regressions\n", "For the final trick, lets expand the last model we made but include an additional part - this time, we'll use the two grit latent variables to predict the latent conscientiousness score. We'd like to see how well these latent variables can predict conscientiousness. If they are measuring the 'same' kind of trait, then we'd expect to see high coefficients for these. \n", "\n", "Rebuild the model from question 2 but include an additional part where the two grit latent variables predict laten conscientiousness." ] }, { "cell_type": "code", "execution_count": 13, "id": "7f716a67-e145-4b2e-a74d-05c4919e6bfe", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "SolverResult(fun=0.9710712122615321, success=True, n_it=44, x=array([ 1.09372032, 1.33481513, 1.36326389, 1.42906892, 0.78786201,\n", " 0.8498987 , 1.3911391 , 1.4946746 , 1.55663121, 1.34467594,\n", " -0.94951483, 0.48329874, -1.02884432, 1.07183657, -1.0865845 ,\n", " 0.768163 , -0.81085453, 1.03271978, 0.65668753, 0.30459112,\n", " -0.82905879, 0.75009116, 0.79133413, 1.38197185, 0.83954118,\n", " 1.03002355, 0.96546841, 1.30376258, 0.96068456, 0.9050144 ,\n", " 0.95839995, 0.70040618, 1.14535444, 0.98499342, 0.56307038,\n", " 0.83191806, 0.9719378 , 1.20764937, 0.83116974, 0.54796473,\n", " 0.61777322, 0.7761055 , 0.64744345, 0.22025482, 0.46986339,\n", " -0.21635048, 0.27907538]), message='Optimization terminated successfully', name_method='SLSQP', name_obj='MLW')" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Model string\n", "mdspec = \"\"\"\n", "consistency =~ GS2 + GS3 + GS5 + GS7 + GS8 + GS11\n", "perseverance =~ GS1 + GS4 + GS6 + GS9 + GS10 + GS12\n", "conscientious =~ C1 + C2 + C3 + C4 + C5 + C6 + C7 + C8 + C9 + C10\n", "\n", "conscientious ~ consistency + perseverance\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(grit_consc)" ] }, { "cell_type": "markdown", "id": "54343eb3-3d06-4838-b92c-9f0d7858fca9", "metadata": {}, "source": [ "Once estimated, inspect the coefficients of that latent regression." ] }, { "cell_type": "code", "execution_count": 14, "id": "0407273b-946c-43f2-8ad0-950b20024f37", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
lvaloprvalEstimateEst. StdStd. Errz-valuep-value
0conscientious~consistency0.3045910.2777820.02277413.374290.0
1conscientious~perseverance-0.829059-0.5827030.038286-21.6543580.0
2GS2~consistency1.0000000.600782---
3GS3~consistency1.0937200.6053120.03457231.6364120.0
4GS5~consistency1.3348150.7083770.03776435.3463610.0
5GS7~consistency1.3632640.7652890.03675537.091020.0
6GS8~consistency1.4290690.7435400.03919936.4563360.0
7GS11~consistency0.7878620.4779690.03000826.2553310.0
8GS1~perseverance1.0000000.533780---
9GS4~perseverance0.8498990.3782130.04185920.3036660.0
10GS6~perseverance1.3911390.7045420.04485531.0141210.0
11GS9~perseverance1.4946750.7004060.0483430.9200590.0
12GS10~perseverance1.5566310.6092870.05447628.5748660.0
13GS12~perseverance1.3446760.6874760.04391730.6187460.0
14C1~conscientious1.0000000.655438---
15C2~conscientious-0.949515-0.5189430.032088-29.5913370.0
16C3~conscientious0.4832990.3685480.02236421.6105120.0
17C4~conscientious-1.028844-0.6060640.030378-33.8676620.0
18C5~conscientious1.0718370.6340310.0304735.1768850.0
19C6~conscientious-1.086584-0.5817620.033225-32.7037370.0
20C7~conscientious0.7681630.5075500.02647829.011770.0
21C8~conscientious-0.810855-0.5394360.026479-30.6224140.0
22C9~conscientious1.0327200.6212870.02986134.5844570.0
23C10~conscientious0.6566880.4851730.0235727.8607510.0
24conscientious~~conscientious0.2202550.3898780.01172218.7903720.0
25consistency~~consistency0.4698631.0000000.02336120.1131870.0
26consistency~~perseverance-0.216350-0.5974630.010429-20.7446720.0
27perseverance~~perseverance0.2790751.0000000.01615517.2746420.0
28C1~~C10.7500910.5704010.01865540.2095970.0
29C10~~C100.7913340.7646070.01808143.7668420.0
30C2~~C21.3819720.7306980.03192843.2838610.0
31C3~~C30.8395410.8641730.01867244.9629580.0
32C4~~C41.0300240.6326860.02476541.5916920.0
33C5~~C50.9654680.5980050.0236340.8580090.0
34C6~~C61.3037630.6615530.03093742.1429770.0
35C7~~C70.9606850.7423930.02210743.4555030.0
36C8~~C80.9050140.7090090.02107142.9503170.0
37C9~~C90.9584000.6140020.02325841.2068530.0
38GS1~~GS10.7004060.7150780.01646842.5325510.0
39GS10~~GS101.1453540.6287690.02811540.7388190.0
40GS11~~GS110.9849930.7715450.02249143.794620.0
41GS12~~GS120.5630700.5273770.01487437.8561340.0
42GS2~~GS20.8319180.6390610.02000841.5792650.0
43GS3~~GS30.9719380.6335970.02343941.4673140.0
44GS4~~GS41.2076490.8569550.02703244.6751860.0
45GS5~~GS50.8311700.4982020.02194237.8803450.0
46GS6~~GS60.5479650.5036210.01480637.0108570.0
47GS7~~GS70.6177730.4143320.01791434.4855150.0
48GS8~~GS80.7761050.4471480.02158535.9556930.0
49GS9~~GS90.6474430.5094320.01739337.2248080.0
\n", "
" ], "text/plain": [ " lval op rval Estimate Est. Std Std. Err z-value \\\n", "0 conscientious ~ consistency 0.304591 0.277782 0.022774 13.37429 \n", "1 conscientious ~ perseverance -0.829059 -0.582703 0.038286 -21.654358 \n", "2 GS2 ~ consistency 1.000000 0.600782 - - \n", "3 GS3 ~ consistency 1.093720 0.605312 0.034572 31.636412 \n", "4 GS5 ~ consistency 1.334815 0.708377 0.037764 35.346361 \n", "5 GS7 ~ consistency 1.363264 0.765289 0.036755 37.09102 \n", "6 GS8 ~ consistency 1.429069 0.743540 0.039199 36.456336 \n", "7 GS11 ~ consistency 0.787862 0.477969 0.030008 26.255331 \n", "8 GS1 ~ perseverance 1.000000 0.533780 - - \n", "9 GS4 ~ perseverance 0.849899 0.378213 0.041859 20.303666 \n", "10 GS6 ~ perseverance 1.391139 0.704542 0.044855 31.014121 \n", "11 GS9 ~ perseverance 1.494675 0.700406 0.04834 30.920059 \n", "12 GS10 ~ perseverance 1.556631 0.609287 0.054476 28.574866 \n", "13 GS12 ~ perseverance 1.344676 0.687476 0.043917 30.618746 \n", "14 C1 ~ conscientious 1.000000 0.655438 - - \n", "15 C2 ~ conscientious -0.949515 -0.518943 0.032088 -29.591337 \n", "16 C3 ~ conscientious 0.483299 0.368548 0.022364 21.610512 \n", "17 C4 ~ conscientious -1.028844 -0.606064 0.030378 -33.867662 \n", "18 C5 ~ conscientious 1.071837 0.634031 0.03047 35.176885 \n", "19 C6 ~ conscientious -1.086584 -0.581762 0.033225 -32.703737 \n", "20 C7 ~ conscientious 0.768163 0.507550 0.026478 29.01177 \n", "21 C8 ~ conscientious -0.810855 -0.539436 0.026479 -30.622414 \n", "22 C9 ~ conscientious 1.032720 0.621287 0.029861 34.584457 \n", "23 C10 ~ conscientious 0.656688 0.485173 0.02357 27.860751 \n", "24 conscientious ~~ conscientious 0.220255 0.389878 0.011722 18.790372 \n", "25 consistency ~~ consistency 0.469863 1.000000 0.023361 20.113187 \n", "26 consistency ~~ perseverance -0.216350 -0.597463 0.010429 -20.744672 \n", "27 perseverance ~~ perseverance 0.279075 1.000000 0.016155 17.274642 \n", "28 C1 ~~ C1 0.750091 0.570401 0.018655 40.209597 \n", "29 C10 ~~ C10 0.791334 0.764607 0.018081 43.766842 \n", "30 C2 ~~ C2 1.381972 0.730698 0.031928 43.283861 \n", "31 C3 ~~ C3 0.839541 0.864173 0.018672 44.962958 \n", "32 C4 ~~ C4 1.030024 0.632686 0.024765 41.591692 \n", "33 C5 ~~ C5 0.965468 0.598005 0.02363 40.858009 \n", "34 C6 ~~ C6 1.303763 0.661553 0.030937 42.142977 \n", "35 C7 ~~ C7 0.960685 0.742393 0.022107 43.455503 \n", "36 C8 ~~ C8 0.905014 0.709009 0.021071 42.950317 \n", "37 C9 ~~ C9 0.958400 0.614002 0.023258 41.206853 \n", "38 GS1 ~~ GS1 0.700406 0.715078 0.016468 42.532551 \n", "39 GS10 ~~ GS10 1.145354 0.628769 0.028115 40.738819 \n", "40 GS11 ~~ GS11 0.984993 0.771545 0.022491 43.79462 \n", "41 GS12 ~~ GS12 0.563070 0.527377 0.014874 37.856134 \n", "42 GS2 ~~ GS2 0.831918 0.639061 0.020008 41.579265 \n", "43 GS3 ~~ GS3 0.971938 0.633597 0.023439 41.467314 \n", "44 GS4 ~~ GS4 1.207649 0.856955 0.027032 44.675186 \n", "45 GS5 ~~ GS5 0.831170 0.498202 0.021942 37.880345 \n", "46 GS6 ~~ GS6 0.547965 0.503621 0.014806 37.010857 \n", "47 GS7 ~~ GS7 0.617773 0.414332 0.017914 34.485515 \n", "48 GS8 ~~ GS8 0.776105 0.447148 0.021585 35.955693 \n", "49 GS9 ~~ GS9 0.647443 0.509432 0.017393 37.224808 \n", "\n", " p-value \n", "0 0.0 \n", "1 0.0 \n", "2 - \n", "3 0.0 \n", "4 0.0 \n", "5 0.0 \n", "6 0.0 \n", "7 0.0 \n", "8 - \n", "9 0.0 \n", "10 0.0 \n", "11 0.0 \n", "12 0.0 \n", "13 0.0 \n", "14 - \n", "15 0.0 \n", "16 0.0 \n", "17 0.0 \n", "18 0.0 \n", "19 0.0 \n", "20 0.0 \n", "21 0.0 \n", "22 0.0 \n", "23 0.0 \n", "24 0.0 \n", "25 0.0 \n", "26 0.0 \n", "27 0.0 \n", "28 0.0 \n", "29 0.0 \n", "30 0.0 \n", "31 0.0 \n", "32 0.0 \n", "33 0.0 \n", "34 0.0 \n", "35 0.0 \n", "36 0.0 \n", "37 0.0 \n", "38 0.0 \n", "39 0.0 \n", "40 0.0 \n", "41 0.0 \n", "42 0.0 \n", "43 0.0 \n", "44 0.0 \n", "45 0.0 \n", "46 0.0 \n", "47 0.0 \n", "48 0.0 \n", "49 0.0 " ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "model.inspect(std_est=True)" ] }, { "cell_type": "markdown", "id": "d554cd7a-d822-4a58-9839-940effa07093", "metadata": {}, "source": [ "By how much does an increase in latent grit alter conscientiousness?" ] }, { "cell_type": "markdown", "id": "118627aa-48cf-40c5-bb71-31b3ea67ebae", "metadata": {}, "source": [ "### 4. Testing the presence of the Big Five - in Big Data\n", "We'll now test the presence of the Big Five in a massive dataset of over 1 million respondents! You can find the dataset here: https://openpsychometrics.org/_rawdata/IPIP-FFM-data-8Nov2018.zip\n", "\n", "Download it, unzip it, and read in the `data-final.csv` file, which is an enormous dataset containing responses to a Big 5 Questionnaire (the IPIP). These are the questions, the short-hand prefixes showing what trait the question is measuring:\n", "\n", "\n", "- EXT1\t- I am the life of the party.\n", "- EXT2\t- I don't talk a lot.\n", "- EXT3\t- I feel comfortable around people.\n", "- EXT4\t- I keep in the background.\n", "- EXT5\t- I start conversations.\n", "- EXT6\t- I have little to say.\n", "- EXT7\t- I talk to a lot of different people at parties.\n", "- EXT8\t- I don't like to draw attention to myself.\n", "- EXT9\t- I don't mind being the center of attention.\n", "- EXT10\t- I am quiet around strangers.\n", "- EST1\t- I get stressed out easily.\n", "- EST2\t- I am relaxed most of the time.\n", "- EST3\t- I worry about things.\n", "- EST4\t- I seldom feel blue.\n", "- EST5\t- I am easily disturbed.\n", "- EST6\t- I get upset easily.\n", "- EST7\t- I change my mood a lot.\n", "- EST8\t- I have frequent mood swings.\n", "- EST9\t- I get irritated easily.\n", "- EST10\t- I often feel blue.\n", "- AGR1\t- I feel little concern for others.\n", "- AGR2\t- I am interested in people.\n", "- AGR3\t- I insult people.\n", "- AGR4\t- I sympathize with others' feelings.\n", "- AGR5\t- I am not interested in other people's problems.\n", "- AGR6\t- I have a soft heart.\n", "- AGR7\t- I am not really interested in others.\n", "- AGR8\t- I take time out for others.\n", "- AGR9\t- I feel others' emotions.\n", "- AGR10\t- I make people feel at ease.\n", "- CSN1\t- I am always prepared.\n", "- CSN2\t- I leave my belongings around.\n", "- CSN3\t- I pay attention to details.\n", "- CSN4\t- I make a mess of things.\n", "- CSN5\t- I get chores done right away.\n", "- CSN6\t- I often forget to put things back in their proper place.\n", "- CSN7\t- I like order.\n", "- CSN8\t- I shirk my duties.\n", "- CSN9\t- I follow a schedule.\n", "- CSN10\t- I am exacting in my work.\n", "- OPN1\t- I have a rich vocabulary.\n", "- OPN2\t- I have difficulty understanding abstract ideas.\n", "- OPN3\t- I have a vivid imagination.\n", "- OPN4\t- I am not interested in abstract ideas.\n", "- OPN5\t- I have excellent ideas.\n", "- OPN6\t- I do not have a good imagination.\n", "- OPN7\t- I am quick to understand things.\n", "- OPN8\t- I use difficult words.\n", "- OPN9\t- I spend time reflecting on things.\n", "- OPN10\t- I am full of ideas.\n", "\n" ] }, { "cell_type": "code", "execution_count": 15, "id": "8eafff30-6ade-4ec9-be48-a46d3c5a411d", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
EXT1EXT2EXT3EXT4EXT5EXT6EXT7EXT8EXT9EXT10...dateloadscreenwscreenhintroelapsetestelapseendelapseIPCcountrylat_appx_lots_of_errlong_appx_lots_of_err
04.01.05.02.05.01.05.02.04.01.0...2016-03-03 02:01:01768.01024.09.0234.061GB51.54480.1991
13.05.03.04.03.03.02.05.01.05.0...2016-03-03 02:01:201360.0768.012.0179.0111MY3.1698101.706
22.03.04.04.03.02.01.03.02.05.0...2016-03-03 02:01:561366.0768.03.0186.071GB54.9119-1.3833
32.02.02.03.04.02.02.04.01.04.0...2016-03-03 02:02:021920.01200.0186.0219.071GB51.75-1.25
43.03.03.03.05.03.03.05.03.04.0...2016-03-03 02:02:571366.0768.08.0315.0172KE1.038.0
\n", "

5 rows × 110 columns

\n", "
" ], "text/plain": [ " EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10 ... \\\n", "0 4.0 1.0 5.0 2.0 5.0 1.0 5.0 2.0 4.0 1.0 ... \n", "1 3.0 5.0 3.0 4.0 3.0 3.0 2.0 5.0 1.0 5.0 ... \n", "2 2.0 3.0 4.0 4.0 3.0 2.0 1.0 3.0 2.0 5.0 ... \n", "3 2.0 2.0 2.0 3.0 4.0 2.0 2.0 4.0 1.0 4.0 ... \n", "4 3.0 3.0 3.0 3.0 5.0 3.0 3.0 5.0 3.0 4.0 ... \n", "\n", " dateload screenw screenh introelapse testelapse endelapse \\\n", "0 2016-03-03 02:01:01 768.0 1024.0 9.0 234.0 6 \n", "1 2016-03-03 02:01:20 1360.0 768.0 12.0 179.0 11 \n", "2 2016-03-03 02:01:56 1366.0 768.0 3.0 186.0 7 \n", "3 2016-03-03 02:02:02 1920.0 1200.0 186.0 219.0 7 \n", "4 2016-03-03 02:02:57 1366.0 768.0 8.0 315.0 17 \n", "\n", " IPC country lat_appx_lots_of_err long_appx_lots_of_err \n", "0 1 GB 51.5448 0.1991 \n", "1 1 MY 3.1698 101.706 \n", "2 1 GB 54.9119 -1.3833 \n", "3 1 GB 51.75 -1.25 \n", "4 2 KE 1.0 38.0 \n", "\n", "[5 rows x 110 columns]" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "big5 = pd.read_csv('data-final.csv', sep='\\t')\n", "big5.head()" ] }, { "cell_type": "markdown", "id": "e3b1860b-c614-4b06-b964-fc34db78d94a", "metadata": {}, "source": [ "There are some other columns which can be gotten rid of by running the following code:" ] }, { "cell_type": "code", "execution_count": 16, "id": "0c8c9253-d3a2-4d93-9ba1-6b3a2e254871", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
EXT1EXT2EXT3EXT4EXT5EXT6EXT7EXT8EXT9EXT10...OPN1OPN2OPN3OPN4OPN5OPN6OPN7OPN8OPN9OPN10
04.01.05.02.05.01.05.02.04.01.0...5.01.04.01.04.01.05.03.04.05.0
13.05.03.04.03.03.02.05.01.05.0...1.02.04.02.03.01.04.02.05.03.0
22.03.04.04.03.02.01.03.02.05.0...5.01.02.01.04.02.05.03.04.04.0
32.02.02.03.04.02.02.04.01.04.0...4.02.05.02.03.01.04.04.03.03.0
43.03.03.03.05.03.03.05.03.04.0...5.01.05.01.05.01.05.03.05.05.0
\n", "

5 rows × 50 columns

\n", "
" ], "text/plain": [ " EXT1 EXT2 EXT3 EXT4 EXT5 EXT6 EXT7 EXT8 EXT9 EXT10 ... OPN1 \\\n", "0 4.0 1.0 5.0 2.0 5.0 1.0 5.0 2.0 4.0 1.0 ... 5.0 \n", "1 3.0 5.0 3.0 4.0 3.0 3.0 2.0 5.0 1.0 5.0 ... 1.0 \n", "2 2.0 3.0 4.0 4.0 3.0 2.0 1.0 3.0 2.0 5.0 ... 5.0 \n", "3 2.0 2.0 2.0 3.0 4.0 2.0 2.0 4.0 1.0 4.0 ... 4.0 \n", "4 3.0 3.0 3.0 3.0 5.0 3.0 3.0 5.0 3.0 4.0 ... 5.0 \n", "\n", " OPN2 OPN3 OPN4 OPN5 OPN6 OPN7 OPN8 OPN9 OPN10 \n", "0 1.0 4.0 1.0 4.0 1.0 5.0 3.0 4.0 5.0 \n", "1 2.0 4.0 2.0 3.0 1.0 4.0 2.0 5.0 3.0 \n", "2 1.0 2.0 1.0 4.0 2.0 5.0 3.0 4.0 4.0 \n", "3 2.0 5.0 2.0 3.0 1.0 4.0 4.0 3.0 3.0 \n", "4 1.0 5.0 1.0 5.0 1.0 5.0 3.0 5.0 5.0 \n", "\n", "[5 rows x 50 columns]" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Run this to keep only needed columns\n", "big5 = big5.filter(regex='[A-Z]\\d+').loc[:, lambda x: ~x.columns.str.contains('_E')]\n", "big5.head()" ] }, { "cell_type": "markdown", "id": "33070757-1cdc-4300-8e43-99a49f84353c", "metadata": {}, "source": [ "With the dataset ready, prepare a CFA model that tests whether each of the ten questions load onto their respective latent factor (e.g. EXT1, EXT2 on Extraversion, OPN1, OPN10 onto Openness, and so on)." ] }, { "cell_type": "code", "execution_count": 17, "id": "c1131f49-9a00-4c26-8a4c-9b22a1f54229", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "SolverResult(fun=5.362960399983457, success=True, n_it=57, x=array([-1.06200145, 0.95417843, -1.030203 , 1.08329996, -0.81910411,\n", " 1.18386593, -0.80722522, 0.94436833, -1.02606913, -0.86188908,\n", " 0.96142689, -0.74789625, 1.01326436, -0.82631873, 0.78080518,\n", " 1.01344284, 0.62361959, 1.18156628, -1.09149167, 0.5441523 ,\n", " -1.07460931, 1.11229093, -1.2430879 , 0.77959885, -0.80312943,\n", " 1.03367876, 0.60953939, -1.12362769, 0.68604513, -1.43895163,\n", " 1.20265913, -1.16395247, 1.14358636, -1.07432794, -1.40957848,\n", " -0.85553933, -0.65361763, 0.74696009, -0.480897 , 0.72871028,\n", " 1.06146634, 1.0741923 , 1.14944716, 0.97562448, 0.94016725,\n", " 1.43674748, 0.90548163, 0.84803146, 1.4625594 , 0.52875896,\n", " 0.85281192, 1.00831611, 0.78594071, 0.78610704, 0.64851546,\n", " 0.87507333, 0.91986045, 1.32519691, 0.94056273, 0.9785403 ,\n", " 0.99411219, 1.21806196, 0.96280453, 0.97306684, 1.08424766,\n", " 0.9874805 , 1.02212687, 1.1566158 , 0.89358005, 1.37933939,\n", " 1.18518101, 0.83121266, 0.72093211, 0.71466859, 0.89804327,\n", " 0.86073212, 0.92737012, 0.92128132, 0.8048323 , 0.74866263,\n", " 0.76710141, 1.00770347, 0.92762256, 1.13732951, 1.1543183 ,\n", " 0.93733664, 0.50929538, 0.93945862, 0.8324595 , 0.96393237,\n", " 0.57244287, 0.94478514, 0.77220248, 1.16343713, 0.92790156,\n", " 0.35866303, -0.17700163, -0.06727421, 0.50955039, -0.06273015,\n", " 0.05657929, -0.18473819, 0.04077642, 0.73729111, 0.11323818,\n", " 0.82428062, -0.01288336, -0.1776681 , -0.03575431, 0.40169197]), message='Optimization terminated successfully', name_method='SLSQP', name_obj='MLW')" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Model string\n", "mdspec = \"\"\"\n", "extra =~ EXT1 + EXT2 + EXT3 + EXT4 + EXT5 + EXT6 + EXT7 + EXT8 + EXT9 + EXT10\n", "open =~ OPN1 + OPN2 + OPN3 + OPN4 + OPN5 + OPN6 + OPN7 + OPN8 + OPN9 + OPN10\n", "consc =~ CSN1 + CSN2 + CSN3 + CSN4 + CSN5 + CSN6 + CSN7 + CSN8 + CSN9 + CSN10\n", "agree =~ AGR1 + AGR2 + AGR3 + AGR4 + AGR5 + AGR6 + AGR7 + AGR8 + AGR9 + AGR10\n", "neuro =~ EST1 + EST2 + EST3 + EST4 + EST5 + EST6 + EST7 + EST8 + EST9 + EST10\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(big5) # EDIT THIS" ] }, { "cell_type": "markdown", "id": "cd23a2ac-6b7e-4be3-9bb0-fab2eec864c7", "metadata": {}, "source": [ "Check the model fit statistics:" ] }, { "cell_type": "code", "execution_count": 18, "id": "01b64a87-43a3-4bef-a669-2499dc93e317", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
DoFDoF Baselinechi2chi2 p-valuechi2 BaselineCFIGFIAGFINFITLIRMSEAAICBICLogLik
Value116512255.445234e+060.01.902557e+070.7138370.7137940.6990540.7137940.6990990.067841209.2740791510.6549375.36296
\n", "
" ], "text/plain": [ " DoF DoF Baseline chi2 chi2 p-value chi2 Baseline \\\n", "Value 1165 1225 5.445234e+06 0.0 1.902557e+07 \n", "\n", " CFI GFI AGFI NFI TLI RMSEA AIC \\\n", "Value 0.713837 0.713794 0.699054 0.713794 0.699099 0.067841 209.274079 \n", "\n", " BIC LogLik \n", "Value 1510.654937 5.36296 " ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "sem.calc_stats(model)" ] }, { "cell_type": "markdown", "id": "16cfff63-d58e-4151-9951-5746275960ef", "metadata": {}, "source": [ "Based on this truly massive dataset, what do we make of the Big 5 model? " ] }, { "cell_type": "markdown", "id": "25b2266d-5bc4-451e-ae71-9a50ffa62845", "metadata": {}, "source": [ "### 5. Exploring and confirming\n", "For the final exercise, we'll see how EFA and CFA work together. \n", "\n", "The last dataset we'll see contains a series of scales that researchers thought might measure the 'DISC' personality model (see [here](https://en.wikipedia.org/wiki/DISC_assessment)), which has four central traits:\n", "\n", "- Dominance: active use of force to overcome resistance in the environment\n", "- Inducement: use of charm in order to deal with obstacles\n", "- Submission: warm and voluntary acceptance of the need to fulfill a request\n", "- Compliance: fearful adjustment to a superior force.\n", "\n", "This model of personality is used in business, but has no real empirical underpinning. To try to address this, researchers took four sub-scale measures from the IPIP, which measure similar-sorts of things - namely, **assertiveness, social confidence, adventurousness, and dominance**.\n", "\n", "Our goal now will be to *explore* a set of latent factors in *half* of this data, and then *confirm* it in the other half. Thus, we will use both EFA and CFA.\n", "\n", "First, lets read in the data, which is accessible from here: http://openpsychometrics.org/_rawdata/AS+SC+AD+DO.zip \n", "\n", "Extract the `data.csv` file and read it in, and show the head. I've renamed it `data-disc.csv` my end and read it in with that title." ] }, { "cell_type": "code", "execution_count": 19, "id": "0b51415a-04bd-440c-95d1-35058fd6bd80", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
AS1AS2AS3AS4AS5AS6AS7AS8AS9AS10...DO3DO4DO5DO6DO7DO8DO9DO10agegender
04433541311...31325421292
14344323343...32323322492
25445332211...33344523521
34332334341...33444531342
44444432120...43435544522
\n", "

5 rows × 42 columns

\n", "
" ], "text/plain": [ " AS1 AS2 AS3 AS4 AS5 AS6 AS7 AS8 AS9 AS10 ... DO3 DO4 DO5 DO6 \\\n", "0 4 4 3 3 5 4 1 3 1 1 ... 3 1 3 2 \n", "1 4 3 4 4 3 2 3 3 4 3 ... 3 2 3 2 \n", "2 5 4 4 5 3 3 2 2 1 1 ... 3 3 3 4 \n", "3 4 3 3 2 3 3 4 3 4 1 ... 3 3 4 4 \n", "4 4 4 4 4 4 3 2 1 2 0 ... 4 3 4 3 \n", "\n", " DO7 DO8 DO9 DO10 age gender \n", "0 5 4 2 1 29 2 \n", "1 3 3 2 2 49 2 \n", "2 4 5 2 3 52 1 \n", "3 4 5 3 1 34 2 \n", "4 5 5 4 4 52 2 \n", "\n", "[5 rows x 42 columns]" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "disc = pd.read_csv('data-disc.csv')\n", "disc.head()" ] }, { "cell_type": "markdown", "id": "4bf0efb3-f4f3-4217-b2fc-25626bfc6c9f", "metadata": {}, "source": [ "The column names here represent the four sub-scales, like so:\n", "\n", "- **Assertiveness**\n", " - AS1 Express myself easily.\n", " - AS2 Try to lead others.\n", " - AS3 Automatically take charge.\n", " - AS4 Know how to convince others.\n", " - AS5 Am the first to act.\n", " - AS6 Take control of things.\n", " - AS7 Wait for others to lead the way.\n", " - AS8 Let others make the decisions.\n", " - AS9 Am not highly motivated to succeed.\n", " - AS10 Can't come up with new ideas.\n", "- **Social Confidence**\n", " - SC1 Feel comfortable around people.\n", " - SC2 Don't mind being the center of attention.\n", " - SC3 Am good at making impromptu speeches.\n", " - SC4 Express myself easily.\n", " - SC5 Have a natural talent for influencing people.\n", " - SC6 Hate being the center of attention.\n", " - SC7 Lack the talent for influencing people.\n", " - SC8 Often feel uncomfortable around others.\n", " - SC9 Don't like to draw attention to myself.\n", " - SC10 Have little to say.\n", "- **Adventurousness**\n", " - AD1 Prefer variety to routine.\n", " - AD2 Like to visit new places.\n", " - AD3 Interested in many things.\n", " - AD4 Like to begin new things.\n", " - AD5 Prefer to stick with things that I know.\n", " - AD6 Dislike changes.\n", " - AD7 Don't like the idea of change.\n", " - AD8 Am a creature of habit.\n", " - AD9 Dislike new foods.\n", " - AD10 Am attached to conventional ways.\n", "- **Dominance**\n", " - DO1 Try to surpass others' accomplishments.\n", " - DO2 Try to outdo others.\n", " - DO3 Am quick to correct others.\n", " - DO4 Impose my will on others.\n", " - DO5 Demand explanations from others.\n", " - DO6 Want to control the conversation.\n", " - DO7 Am not afraid of providing criticism.\n", " - DO8 Challenge others' points of view.\n", " - DO9 Lay down the law to others.\n", " - DO10 Put people under pressure." ] }, { "cell_type": "markdown", "id": "d7691071-bffc-452d-a670-1a79f6378216", "metadata": {}, "source": [ "The first task is to prepare the data. Drop the `age` and `gender` columns, and the use the `.sample` dataframe method to extract 50% of the data for exploration and the other half for confirmation. This can be a tricky step, so take care." ] }, { "cell_type": "code", "execution_count": 20, "id": "8f5acfa4-4368-429b-8049-2ff4c8cc9289", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "# Your answer here\n", "# Drop columns\n", "disc = disc.drop(columns=['age', 'gender'])\n", "\n", "# Sample half\n", "explore = disc.sample(frac=.50, random_state=42)\n", "\n", "# Get the other half not in the first half\n", "confirm = disc[~disc.index.isin(explore.index)]" ] }, { "cell_type": "markdown", "id": "afa8af02-3cf2-4dd7-aae7-34add704d08f", "metadata": {}, "source": [ "Fit an EFA to the first half of the data. Lets start with four factors, which would represent the four subscales. What does that solution look like? Fit it and examine a plot of the loadings.\n", "\n", "If you want the figure to be larger, you can run the following lines of code before you make a heatmap:\n", "`import matplotlib.pyplot as plt\n", "plt.figure(figsize=(10, 10))`" ] }, { "cell_type": "code", "execution_count": 21, "id": "74d8d9a9-83bf-4fc0-a9d9-13c01319cfdc", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAykAAAMtCAYAAACb+z58AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdeVzVVeL4/xf7vu8uLIIgm4orKu6AS+6amqKTaaMtVpM55ThT48ykZdknq6mpBMvURNPU0BREVFARNTdAEVFB9n1H4ML9/cGP99frBTP1InDP8/G4j0f3vM8573Pszb337BpyuVyOIAiCIAiCIAhCO6H5tAsgCIIgCIIgCIJwL9FIEQRBEARBEAShXRGNFEEQBEEQBEEQ2hXRSBEEQRAEQRAEoV0RjRRBEARBEARBENoV0UgRBEEQBEEQBKFdEY0UQRAEQRAEQRDaFdFIEQRBEARBEAShXdF+2gVoS6ampk+7CIKa+PLLL592EQQ1kZ6e/rSLIKiJLl26PO0iCGpi0aJFT7sIrdLQ0HjaRWhRZzybXYykCIIgCIIgCILQrohGiiAIgiAIgiAI7YpaTfcSBEEQBEEQhEfVXqd7dUZiJEUQBEEQBEEQ1MyXX36Ji4sL+vr69O/fn9jY2AfG37ZtG3369MHQ0BAHBwcWLVpEUVGRysrXJiMpp06dYvjw4QQFBXHo0CGFa7t372b9+vVcu3aNxsZGHB0dGT9+PBs2bAAgJyeHFStWcP78eVJTU3nttdf49NNP26LYndKSJUt47bXXsLe35+rVq7zzzjucPn26xbhfffUV8+fPVwq/evUqgwcPBmDevHn873//U4pjY2NDbW3tky280OH07t0bNzc3dHV1KSoqIiEhgbKysgem6d69O3369MHExISKigouXbrEnTt3pOve3t44OjpiampKQ0MDBQUFXLhwgfLyclVXR2in+vfvj7+/P8bGxhQUFBAVFaXwzNzL2NiYsWPH4uDggKWlJWfPniUqKqrVvL28vJg+fTopKSn89NNPqqqC0EH06tULHx8fDAwMKC0tJSEhgby8vFbj29nZMWjQIMzNzampqeHKlSukpKQoxHFycqJfv37SZ9758+fJyMhQdVUENRceHs4bb7zBl19+ybBhw/j666+ZMGECycnJODo6KsWPi4tj4cKF/N///R+TJ08mKyuLZcuWsWTJEn7++WeVlLFNRlLCwsJYvnw5cXFxCn94R44cYe7cucyaNYuEhATOnz/P+++/T11dnRSntrYWGxsbVq9eTZ8+fdqiuJ3WjBkz+OCDD/j4448JCAjg9OnT7N69m27durUY/+2338bNzU169erVi+LiYvbu3asQr6ysTCGem5ubaKAIeHl50atXL86ePcuvv/5KTU0NY8eORVu79b4Ra2trhg8fzq1btzhw4AC3bt1i+PDhWFlZSXHs7OxISUnh0KFDHDlyBA0NDcaMGYOWllZbVEtoZzw9PQkKCuLkyZNs2rSJO3fuMHfu3FZ3c9TS0qK6upqTJ08+8MclNO0IOXbsWPGDUQDAxcWFQYMGcenSJfbv309eXh5BQUEYGRm1GN/Y2JigoCDy8vLYv38/ly5dYvDgwTg5OUlxbGxsGDVqFDdu3GDfvn3cuHGD0aNHY21t3VbVEv4gDQ2Ndvn6oz755BMWL17MkiVL8PT05NNPP6V79+589dVXLcaPj4/H2dmZ1157DRcXFwICAli6dCnnzp173H/SVqm8kVJVVcXOnTt56aWXmDRpEt999510LSIigoCAAFauXImHhwfu7u5MmzaNzz//XIrj7OzMxo0bWbhwIWZmZqoubqf26quvsmXLFrZs2cL169d55513yMrKYvHixS3GLy8vJz8/X3r5+flhbm7O1q1bFeLJ5XKFePn5+W1RHaGd8/T0JDExkTt37lBWVsapU6fQ1tbGxcWl1TS9evUiJyeHpKQkysvLSUpKIjc3F09PTynO0aNHuXnzJmVlZZSWlnL69GmMjY0VGjKC+hg8eDAXL17k4sWLFBUVERUVRXl5Of369WsxfllZGVFRUVy5cuWBnSkaGhpMmzaNEydOUFJSoqriCx2It7c3qamppKamUlZWRkJCAlVVVfTq1avF+L169aKqqkoaQW5O6+Pjo5BndnY2V65coaysjCtXrpCdnY23t3dbVUvoJGpraykvL1d4tfYZV1dXx/nz5wkODlYIDw4O5tSpUy2mGTp0KJmZmRw8eBC5XE5eXh4//fQTzzzzzBOvSzOVN1LCw8Px8PDAw8ODkJAQNm/eLO3lbG9vT1JSEomJiaouhtrT0dGhb9++HD16VCH86NGj0tSt37Nw4UKOHTumNI3C2NiYxMRErl69ys6dO+ndu/cTK7fQMRkbG2NgYEBOTo4U1tjYSF5e3gN7CG1sbBTSAGRnZz8wjY6ODoAYvVNDmpqaODg4cOvWLYXwmzdvtjpC/LCGDx9OdXU1ly5deqx8hM5BU1MTKysrsrKyFMKzs7OxtbVtMY2NjQ3Z2dkKYVlZWVhbW0s93zY2Nkp5ZmVltZqnILRm3bp1mJmZKbzWrVvXYtzCwkIaGhqws7NTCLezsyM3N7fFNEOHDmXbtm3MmTMHXV1d7O3tMTc3VxhYeNJU3kgJDQ0lJCQEgPHjx1NZWUl0dDQAy5cvZ+DAgfj6+uLs7MzcuXMJCwt7Ij82WmpRdsaDbh6WlZUV2traSqMc+fn5Sg9pS+zs7AgKCuL7779XCE9NTeWll15i7ty5vPDCC9y9e5fIyEhcXV2faPmFjkVfXx+Au3fvKoTfvXsXAwODB6b7o2kGDBhAfn7+7651ETofQ0NDNDU1qaysVAivqqrC2Nj4kfPt1q0bffr04cCBA49bRKGT0NPTQ1NTU+nzqaamptXPJwMDA2pqahTC7t69i6ampvQZaWBg8Ic/84Sn62lP62rttWrVKsrKyhReq1at+t263Esul7c6dSw5OZnXXnuNd999l/Pnz3Po0CFu3brFsmXLnti/7f1UunA+JSWFhIQE9uzZ03QzbW3mzJlDWFgYgYGBGBkZceDAAdLS0oiJiSE+Pp4VK1awceNGTp8+jaGh4SPfe926daxZs0YhTFdXFz09vceqU2ejoaHxUI23+fPnU1ZWRkREhEL42bNnOXv2rPQ+Pj6e2NhYli5dyl//+tcnXl6hfXJ2dlYYkYuJiWmT+w4cOBBzc3MiIyPb5H5Cx/Cwn2st0dXVZerUqRw8eFDpB6YgtPRcPehZe5jn8I/mKQgt0dPTe+jfuNbW1mhpaSmNmjyo43rdunUMGzaMlStXAk0b4xgZGTF8+HD+85//4ODg8HgVaIFKGymhoaHIZDK6du0qhcnlcnR0dCgpKcHCwgIAV1dXXF1dWbJkCatXr8bd3Z3w8HAWLVr0yPdetWoVb775pkLYveVQN0VFRchkMqUhZBsbm4daQ7JgwQJ27NhBfX39A+PJ5XJ+++03MZKiZjIzMyksLJTeNy9i19fXV/ihd//7+929e1fqYfy9NAMGDKBbt25ERkZSXV39uFUQOqDq6moaGxuVRk0MDQ2pqqp6pDwtLCwwNzdn9uzZUlhzz+KqVav46quvKC0tfeQyCx1TbW0tjY2NSiMcLY2ENKupqVHqbNXX16exsVFK09JITEsjyoLwJOnq6tK/f3+ioqKYPn26FB4VFcXUqVNbTFNdXa208U3zd72qGtUqa6TIZDK2bNnChg0blBbmzJw5k23btvHqq68qpXN2dn6sL5hmLbUo1fkAnvr6ei5evMiYMWMURkNGjx79u1MaAgICcHV1ZcuWLQ91r969e5OUlPRY5RU6FplMpjTlpqamBgcHB2nRsaamJnZ2dly4cKHVfAoKCnBwcODatWtSmIODg0IDCJpGULp3705UVNRjf1YIHVdjYyM5OTm4uLgobOvq4uLC9evXHynPwsJCvvnmG4WwkSNHoqurKy3KF9RPY2MjRUVFdOnSRWG3t/vf36ugoIDu3bsrhHXt2pXCwkLpR11BQQFdu3YlOTlZIY7YgKb90tTsHEcMvvnmmyxYsIABAwYwZMgQvvnmGzIyMqTpW6tWrSIrK0v67Td58mRefPFFvvrqK8aNG0dOTg5vvPEGgwYNokuXLiopo8oaKREREZSUlLB48WKlXblmzZpFaGgohYWFVFdXM3HiRJycnCgtLeWzzz6jvr6eoKAgKf7FixcBqKyspKCggIsXL6Krq4uXl5eqit8pffHFF3zzzTf89ttvJCQksGjRIrp160ZYWBgA7733Hl26dGHp0qUK6RYuXMjZs2e5evWqUp7vvPMOZ8+eJS0tDRMTE5YtW4avry8rVqxokzoJ7dfVq1fx8fGhoqKC8vJyfHx8kMlkCouchw4dSnV1tfQ3fu3aNYKDg/Hy8iIzM5Nu3brh4ODA4cOHpTQDBw7ExcWFY8eOUV9fL4281NfX09DQ0KZ1FJ6+M2fOMHXqVHJycsjMzMTPzw8zMzN+++03AEaNGoWJiQm//PKLlKZ5OoOuri6GhobY2dnR0NAgLSYtKChQuEdzr/b94YJ6SUpKYvjw4RQVFZGfn4+HhwdGRkZSp0r//v0xNDSUDsS7du0avXr1YuDAgVy/fh1bW1t69uzJ8ePHpTyTk5OZMGECvr6+ZGRk4OjoSJcuXcR6KEHl5syZQ1FREf/617/IycnBx8eHgwcPSltk5+TkKDTAn3/+eSoqKvjiiy9YsWIF5ubmjBkzhg8//FBlZVRZIyU0NJTAwMAWtw2eOXMma9euJSQkhMTERBYuXEheXh4WFhb4+fkRGRmJh4eHFN/Pz0/67/Pnz7N9+3acnJy4ffu2qorfKe3ZswdLS0vefvtt7O3tSU5OZtasWdJuXfb29ko74piamjJlyhTefvvtFvM0MzNj48aN2NnZUV5ezuXLl5kwYQLnz59XeX2E9i05ORltbW0GDRqErq4uhYWFREdHI5PJpDhGRkYKw8SFhYXExcXRp08f+vTpQ2VlJbGxsQon2jZ/Ntw/Qnvq1Clu3ryp4loJ7c3Vq1cxNDQkICBAOsxxx44d0oiHsbGx0vfQkiVLpP92cHDAx8eH0tJS/vvf/7Zp2YWO5datW+jp6UknbpeUlCiM5hoYGCicmVJZWUlUVBSDBw/G09OT6upqzpw5Q3p6uhQnPz+fY8eO0a9fP/z8/KioqODYsWNKo8eCoAovv/wyL7/8covX7j0ypNny5ctZvny5ikv1/2jI1Wh1VmuHewnCk/bll18+7SIIauLeHzyCoEqqmtIhCPd7nDXJqtZeN2DqjNvwd46JdYIgCIIgCIIgdBqikSIIgiAIgiAIQrui0i2IBUEQBEEQBKGzUOedYtuaGEkRBEEQBEEQBKFdUauRlGHDhj3tIghq4tixY0+7CIKaGDNmzNMugqAm7j8MWBAEQZXUqpEiCIIgCIIgCI9KTPdqO2K6lyAIgiAIgiAI7YpopAiCIAiCIAiC0K60yXSvU6dOMXz4cIKCgjh06JDCtd27d7N+/XquXbtGY2Mjjo6OjB8/ng0bNgBNp6R/9dVXXLx4kdraWry9vfnnP//JuHHj2qLonc4zzzzDrFmzsLS0JD09na+//pqkpKTfTefl5cX69eu5ffs2r776qhT+4Ycf0rt3b6X4CQkJvPfee0+07ELHMmrUKMaNG4e5uTnZ2dns2LGD1NTUFuO6ubkxa9Ys7O3t0dXVpaioiBMnThAVFSXF6devHxMnTsTW1hYtLS3y8vKIjIwkPj6+raoktGO+vr64urpKz8+5c+coKyt7YJru3bvTu3dvjI2Nqays5NKlS2RmZrYY18vLi759+3Lt2jV+++03VVRB6ADkcjkHDx7k5MmTVFdX4+zszOzZsx940GV2djYHDhwgIyOD4uJiZs6cqbSW7MCBAxw8eFAhzMTEhA8++EAl9RAenZju1XbapJESFhbG8uXL2bRpExkZGTg6OgJw5MgR5s6dy9q1a5kyZQoaGhokJycTHR0tpT1x4gRBQUGsXbsWc3NzNm/ezOTJkzlz5gx+fn5tUfxOY8SIESxdupT//ve/JCcnM3HiRP7973+zdOlSCgoKWk1naGjIW2+9xcWLFzE3N1e49u9//xsdHR3pvYmJCV9++SWxsbGqqobQAQwcOJC5c+eybds2bty4wYgRI3j99dd59913KS4uVopfW1vL0aNHyczMpLa2lp49e7JgwQJqa2s5ceIEAFVVVRw4cIDc3FxkMhm9e/dm0aJFVFRUPFRDW+i8PD096dWrF6dPn6aiogIfHx9Gjx5NREQEMpmsxTTW1tYMGzaMy5cvk5mZSbdu3QgICCAqKoqioiKFuJaWlri5uVFSUtIW1RHasaioKI4ePcqCBQuwtbXl0KFDfPHFF7z77rvo6+u3mKa+vh4rKyv8/PzYvXt3q3k7ODiwfPly6b2mppjsIqg3lf8FVFVVsXPnTl566SUmTZrEd999J12LiIggICCAlStX4uHhgbu7O9OmTePzzz+X4nz66af89a9/ZeDAgfTs2ZO1a9fSs2dPfvnlF1UXvdOZPn06kZGRHD58mDt37vD1119TUFDAM88888B0r732GjExMVy9elXpWmVlJSUlJdKrX79+1NbWikaKmgsKCiIuLo7Y2FhycnIIDw+npKSEUaNGtRj/zp07JCQkkJ2dTVFREfHx8SQlJdGzZ08pTkpKChcuXCAnJ4eCggKio6PJzMzEzc2tjWoltFe9evUiMTGRzMxMysrKOH36NNra2jg7O7eaxsPDg9zcXJKTkykvLyc5OZnc3Fw8PDwU4mlrazN06FDOnDlDXV2dimsitGdyuZyYmBjGjRtH37596dKlCwsWLKCuro6zZ8+2ms7JyYkZM2YwYMAAtLVb7xvW1NTEzMxMepmYmKiiGoLQYai8kRIeHo6HhwceHh6EhISwefNm5HI5APb29iQlJZGYmPjQ+TU2NlJRUYGlpaWqitwpaWtr07NnT6VpCr/99hteXl6tpgsKCsLBwYFt27Y91H2Cg4M5fvw4tbW1j1VeoePS0tLCyclJaXQjKSkJV1fXh8qje/fuuLq6cv369Vbj9OrVC3t7+1ankAnqwcjICAMDA3Jzc6WwxsZG8vPzsba2bjWdtbU1OTk5CmE5OTnY2NgohA0YMIDs7Gzy8vKebMGFDqeoqIjy8nI8PT2lMB0dHdzc3Lh169Zj519QUMDf/vY33n33XcLCwigsLHzsPIUnT0NDo12+OiOVT/cKDQ0lJCQEgPHjx1NZWUl0dDSBgYEsX76c2NhYfH19cXJywt/fn+DgYObPn4+enl6L+W3YsIGqqipmz579wPvW1tYq/VBubGxU2+FTU1NTtLS0lKYrlJaWYmFh0WKaLl26sGjRIlauXEljY+Pv3sPd3R0XFxc+/fTTJ1FkoYMyNjZGS0uL8vJyhfDy8nLMzMwemHb9+vWYmJigpaXF/v37lUbkDAwM+Oijj9DW1kYul7N161aSk5OfeB2EjsPAwACAu3fvKoTfvXsXIyOjVtPp6+u3mObeKTtOTk5YWloqraUU1FPzZ9r9IxympqYtTmP9I5ydnVm4cCG2trZUVFRw6NAhPv74Y/7+979jbGz8WHkLQkel0kZKSkoKCQkJ7Nmzp+lm2trMmTOHsLAwAgMDMTIy4sCBA6SlpRETE0N8fDwrVqxg48aNnD59GkNDQ4X8fvzxR/75z3+yb9++3z1Uat26daxZs0YhzNXVVWH6iDpqHsVqpqGhoRQGTcPOb7/9Nlu3biUrK+uh8h43bhy3bt16YO+3oD4e9lm71/r169HT06NHjx7MnDmT/Px8EhISpOt3797lX//6F3p6enh6ejJnzhwKCwtJSUlRSR2E9sfZ2ZmBAwdK748fPw4oP2+thT3Ivb2RhoaG9OvXj5iYmIfqpBE6n4SEBH788Ufp/csvvwwoL5z+o89ZS7y9vRXeu7i48N5773HmzBnGjh372PkLQkek0kZKaGgoMpmMrl27SmFyuRwdHR1KSkqkHnxXV1dcXV1ZsmQJq1evxt3dnfDwcBYtWiSlCw8PZ/HixezatYvAwMDfvfeqVat48803FcKeffbZJ1Szjqe8vJyGhgalaXJmZmaUlpYqxTcwMMDd3R1XV1eFD2ZNTU0iIiJYvXo1ly5dkuLr6ekxcuRIfvjhB5XWQ2j/KisraWhoUBo1MTExURpduV/z9IasrCxMTU2ZMmWKQiNFLpeTn58PNK1jcXBwYMKECaKRokYyMzMVpsFoaWkBTZ9Z946MtDRScq/7R02g6XOsOY2lpSUGBgaMHz9euq6pqYmtra30HfUkfpwK7Vfv3r0V1jU1b8Jw/6hwRUUFpqamT/Teenp6dO3aVfq8E9oPdZ2R8zSorJEik8nYsmULGzZsIDg4WOHazJkz2bZtm8JWts2cnZ0xNDSkqqpKCvvxxx954YUX+PHHH393kXczPT09pSlj6vxgyWQyUlNT8fPz49SpU1J4v379OH36tFL86upqli1bphA2adIk+vTpw/vvv68w/xtg+PDh6OjocPToUdVUQOgwGhoaSE9Px8vLiwsXLkjhXl5eXLx48aHz0dDQeOAi02b37i4ndH4ymYzKykqFsJqaGuzt7aXprM2NiQc9b4WFhTg4OCg0cB0cHKSdDnNzczlw4IBCGn9/f2mRvWigdH76+voKDVm5XI6pqSnXrl2je/fuQNPzeOPGDaZOnfpE711fX09ubu5Dr+MThM5IZY2UiIgISkpKWLx4sVKP6qxZswgNDaWwsJDq6momTpyIk5MTpaWlfPbZZ9TX1xMUFAQ0NVAWLlzIxo0b8ff3l34cGxgY/O78dkHRzz//zFtvvUVqaipXr15lwoQJ2NjYSHuzP//881hZWbFhwwbkcjnp6ekK6UtLS6mrq1MKh6apXs3bfwpCVFQUixcv5vbt29y8eZMRI0ZgaWnJsWPHAJgxYwbm5uaEhYUBMHr0aIqLi6WFzD179iQ4OFih0TthwgTS09PJz89HW1sbX19fhgwZ8tCbOgid17Vr1/D29qaiooKKigq8vb2RyWTcvn1bijNkyBCqq6ulEeCUlBQCAwPx9PQkKyuLrl27Ym9vL53NI5PJlM5Zkclk1NbW/u75K0LnpKGhwejRozl8+DA2NjbY2tpy+PBhdHV1FaYgfv/995ibm0sNF5lMJn22NTQ0UFpayp07d9DT05Omru/ZswdfX18sLCykNSl3795l8ODBbV9RQWgnVNZICQ0NJTAwsMWGxMyZM1m7di0hISEkJiaycOFC8vLysLCwwM/Pj8jISGkbyK+//hqZTMYrr7zCK6+8IuXxpz/9SWE7Y+H3nThxAhMTE+bNm4elpSW3b9/m3XfflYaTLS0tf3etT0u6du2Kj48Pf/vb3550kYUO6uzZsxgZGTF58mTMzMzIzs5m48aN0uJSMzMzrKyspPgaGhrMmDEDa2trGhoaKCgoYM+ePdJ6A2gaHZ0/fz4WFhbU19eTk5NDaGjoA7f+FNTD1atX0dbWZuDAgejq6lJYWEhMTIzCGSmGhoYKox+FhYWcPHmS3r1707t3byorK4mLi1M6I0UQ7hUUFER9fT3h4eHSYY6vvvqqwohLSUmJwrqVsrIyhUMZo6OjiY6OpmfPnrzxxhtAUyfg5s2bqaysxNjYGBcXF9566y2Fz0mhfeisO2m1RxpyNRqznjBhwtMugqAm7l2HJQiqdP/J1YKgKo/SiSUIj+Jh1h4/LU96/dGT8ntrPjsi9V2kIQiCIAiCIAhCu6Tyc1IEQRAEQRAEoTMQ073ajhhJEQRBEARBEAShXRGNFEEQBEEQBEEQ2hW1mu7l5OT0tIsgqAlxAJfQVu4/kFAQVOVBh2MKgroQ073ajhhJEQRBEARBEAShXRGNFEEQBEEQBEEQ2hW1mu4lCIIgCIIgCI9KTPdqO23SSDl16hTDhw8nKCiIQ4cOKVzbvXs369ev59q1azQ2NuLo6Mj48ePZsGEDAHFxcbz99ttcu3aN6upqnJycWLp0KX/5y1/aouidzsiRIwkKCpJOAd+1axc3btxoMa6rqyszZszAzs4OXV1diouLiY2NJTo6WiGen58fU6ZMwdramsLCQvbt28fFixfboDZCezZ+/HimTp2KhYUFd+7cISwsjKtXr/5uul69evHvf/+bjIwMVqxYIYUPHjyYmTNn4uDggJaWFjk5Oezfv1/hVHpBPcnlcqKjo0lISKCmpobu3bszdepU7OzsHpguMTGRqKgoioqKsLKyIjg4GG9vb+n6hx9+SGlpqVI6f39/pk6d+qSrIXQAcrmcyMhI4uPjpd8kM2bMwN7evtU0ubm5HDp0iMzMTEpKSpg6dSojRoxQinfy5EmOHTtGeXk59vb2TJ06lR49eqiyOoLQrrVJIyUsLIzly5ezadMmMjIycHR0BODIkSPMnTuXtWvXMmXKFDQ0NEhOTlb4EWxkZMSrr75K7969MTIyIi4ujqVLl2JkZMSf//zntih+p9G/f3+effZZfvzxR9LS0hg+fDivvvoqa9asoaSkRCl+XV0dMTExZGVlUVdXh6urK/Pnz6e2tpa4uDgAXFxcWLJkCfv37+fixYv07duXF198kY8++ojbt2+3cQ2F9mLYsGEsWrSIb7/9lqtXrzJu3Dj+/ve/8/rrr1NYWNhqOkNDQ1577TUuX76Mubm5wrXKykp2795NZmYmMpmMAQMG8Oqrr1JWViYaxWruxIkTxMXFMWvWLKytrYmJiSE0NJQVK1agp6fXYpr09HR+/PFHgoKC8PLyIjk5me3bt7N06VLpO+qVV15BLpdLafLy8ggNDcXX17dN6iW0PzExMRw/fpy5c+diY2PDkSNH+Prrr3n77bdb3cSirq4OKysr+vTpw759+1qMc+HCBfbt28eMGTNwcXHh9OnTfPvtt/z1r3/FwsJClVUShHZL5WtSqqqq2LlzJy+99BKTJk3iu+++k65FREQQEBDAypUr8fDwwN3dnWnTpvH5559Lcfz8/Hjuuefw9vbG2dmZkJAQxo0bR2xsrKqL3ukEBgZy8uRJTp48SW5uLrt27aKkpISRI0e2GP/OnTucO3eOnJwcioqKSEhIIDk5GTc3NynO2LFjuXr1KocPHyYvL4/Dhw9z7do1xo4d21bVEtqhyZMnEx0dzZEjR8jKyiIsLIyioiLGjRv3wHTLli0jNjaW69evK11LSkrizJkzZGVlkZeXx4EDB0hPT8fT01NV1RA6ALlczsmTJxk9ejQ+Pj7Y29vz7LPPUl9f/8DG68mTJ3Fzc2PUqFHY2toyatQoXF1dOXnypBTH2NgYExMT6XX16lUsLS1xcXFpg5oJ7Y1cLufEiRMEBgbSu3dvHBwceO6556irq+PChQutpnN0dGTy5Mn4+fmhrd1y3/CJEycYNGgQ/v7+2NnZMW3aNMzNzTl16pSqqiM8Ig0NjXb56oxU3kgJDw/Hw8MDDw8PQkJC2Lx5s9QzZW9vT1JSEomJiQ+d34ULFzh16lSrP6yFlmlpaeHo6Kg03ebq1asPPZzcvXt3evToQWpqqhTWo0cPpTyTk5PFELUa09bWxtXVlUuXLimEX7x4kV69erWabsyYMdjb2xMeHv5Q9/H19aVLly4kJyc/VnmFjq2kpISKigp69uwphWlra+Pi4kJ6enqr6TIyMhTSALi7u5ORkdFifJlMxsWLFxkwYECn/UEgPFhxcTEVFRW4u7tLYc2fd48zc0Amk5GZmYmHh4dCuIeHh5iRIKg1lU/3Cg0NJSQkBGiao15ZWUl0dDSBgYEsX76c2NhYfH19cXJywt/fn+DgYObPn680RN+tWzcKCgqQyWT885//ZMmSJQ+8b21tLbW1tQphDQ0NaGlpPdkKdhDGxsZoaWlRXl6uEF5eXo6pqekD065bt05KHxERodDTaGpq+kh5Cp2XiYkJWlpaSnP5y8rKlKZwNXNwcCAkJITVq1fT2NjYat6GhoZ8++236Ojo0NjYyDfffKPUGBLUS0VFBdD0GXcvY2PjFteTNKusrGwxTXN+90tOTubu3bv079//8QosdFjN33UmJiYK4SYmJhQXFz9yvlVVVTQ2Nv6h51EQ1IFKGykpKSkkJCSwZ8+epptpazNnzhzCwsIIDAzEyMiIAwcOkJaWRkxMDPHx8axYsYKNGzdy+vRpDA0NpbxiY2OprKwkPj6ed955Bzc3N5577rlW771u3TrWrFmjENa/f38GDBigmsp2EPfOr4aH26Xi448/Rk9Pjx49ejBt2jTy8/M5d+7cY+UpdH73PxethWlqavKXv/yFHTt2kJOT88A8a2pqWLFiBfr6+vTu3ZtFixaRl5dHUlLSEyu30L5duHCBvXv3Su//9Kc/PbG85XJ5q59f586dw93dXXTAqJHz58/z008/Se+bO0fvf0Ye9Nz8EeK7s2PQ1BSnd7QVlTZSQkNDkclkdO3aVQqTy+Xo6OhQUlIiLQZzdXXF1dWVJUuWsHr1atzd3QkPD2fRokVSuuY5wL6+vuTl5fHPf/7zgY2UVatW8eabbyqE3btTkLqprKykoaEBMzMzhXATExOlkZD7FRUVAZCdnY2JiQmTJk2SGinl5eWPlKfQeVVUVNDQ0KC02NPMzIyysjKl+Pr6+ri5ueHi4sKLL74INH1Za2pqsmvXLtasWSNNCZXL5eTm5gJw+/ZtunXrxowZM0QjRY14eXnRvXt36X1DQwPQ9Bl3bwOipZGSexkbG1NZWakQVlVV1WKakpISbty4Ic0KENSDt7c3Tk5O0nuZTAYozxaorKxUGl35I4yMjNDU1FQaNXncfAWho1NZI0Umk7FlyxY2bNhAcHCwwrWZM2eybds2Xn31VaV0zs7OGBoaUlVV1WrecrlcaSrX/fT09JSmjKnrVC9o+iLPyMjA09NTYTGpp6fnH5ouo6GhgY6OjvT+5s2beHp6KuzI5unpyc2bN59IuYWORyaTkZaWRp8+fThz5owU3qdPHxISEpTi19TU8MYbbyiEjR8/Hh8fHz7++GPy8vIeeL97n0eh87v/s10ul2NiYkJqaipdunQBmp7BW7duMX78+FbzcXR0JDU1lYCAACksNTVV2tnrXufPn8fY2FhpzYDQuenr6yvs2NX8rF2/fp1u3boB/+/zbtKkSY98H21tbbp168b169cVdo67fv26wpbYgqBuVNZIiYiIoKSkhMWLFyv1tM+aNYvQ0FAKCwuprq5m4sSJODk5UVpaymeffUZ9fT1BQUEA/Pe//8XR0VFacBsXF8fHH3/M8uXLVVX0TuvIkSMsWrSI9PR0bt68yfDhw7GwsODEiRMA0m4izTuwjRw5kuLiYulHoqurK0FBQcTExEh5Hj16lBUrVhAcHMylS5fo06cPnp6efPTRR21eP6H9+OWXX3jttde4ceMGKSkpBAcHY21tTWRkJADz58/HysqKzz77DLlcrrRYuaysjPr6eoXwGTNmkJaWRm5uLtra2vTr149Ro0bxzTfftGndhPZFQ0ODYcOGcezYMaytrbGysuLYsWPo6OjQt29fKd7OnTsxNTWVGi7Dhg3jm2++4fjx43h6enL16lVu3LjB0qVLFfJvbGzk/Pnz9OvXT607uoSmZ23EiBFER0djY2ODtbU10dHR6Orq4ufnJ8Xbvn07ZmZmPPPMM0BTQ6b5e7ShoYGysjKysrLQ09PD2toagBEjRvDjjz/SrVs3nJ2diY+Pp6SkhCFDhrR9RYUHEtPy2o7KGimhoaEEBgYqNVCgaSRl7dq1hISEkJiYyMKFC8nLy8PCwgI/Pz8iIyOlHqvGxkZWrVrFrVu3pF00PvjgA6UvEuH3NfcGPvPMM5iampKdnc0XX3whLfgzMzPD0tJSiq+hocG0adOwtramsbGRgoICfv75Z4Xtn2/evEloaChTpkxhypQpFBQU8O2334odSdTcyZMnMTExYfbs2VhYWJCRkcH7779PQUEBABYWFtKX88PS09PjxRdfxMrKirq6OrKysti4caPCRg6CehoxYgT19fXs27dPOszxhRdeUBhxKS0tVfhx4eTkxNy5c4mKiiIqKgpLS0uee+45pZGUGzduUFpaKhbMCwCMHj2a+vp6du/eTU1NDY6Ojvz5z39WGHG5/1krLy/nk08+kd4fO3aMY8eO4erqyssvvww0HbdQXV1NVFQU5eXlODg4sGTJEoXvZEFQNxryllaydlLLli172kUQ1ER+fv7TLoKgJsQ6CaGt6OrqPu0iCGricabPqZqNjc3TLkKLmjsBO5M2OXFeEARBEARBEDo6Md2r7Yh91ARBEARBEARBaFdEI0UQBEEQBEEQhHZFTPcSBEEQBEEQhIcgpnu1HbVqpNx7ZoMgqNK8efOedhEENZGdnf20iyCoCTs7u6ddBEEQ1IiY7iUIgiAIgiAIQruiViMpgiAIgiAIgvCoxHSvtiNGUgRBEARBEARBaFfaZCTl1KlTDB8+nKCgIA4dOqRwbffu3axfv55r167R2NiIo6Mj48ePZ8OGDUr5nDx5kpEjR+Lj48PFixfbouidzrPPPsuf/vQnrK2tSUtL4+OPP+bChQstxu3fvz+bNm1SCp8+fbp0ovy3337LgAEDlOLExsby2muvPdGyCx1L3759GThwIMbGxhQWFnL06FGysrJajGtkZMSoUaOwt7fHwsKC8+fPExMToxDHysqKgIAA7OzsMDMz4+jRo5w/f74tqiK0cz4+PvTr1w9DQ0OKi4uJjY0lJyen1fhdunQhICAAS0tLqqqq+O2330hKSlKIo6uri7+/P66urujp6VFeXs7JkydJT09XdXWEds7Ly4sePXqgq6tLUVERFy5coLy8/IFpunbtio+PD0ZGRlRVVXHlyhWl9WT6+vr07t0be3t7tLS0qKys5OzZs5SWlqqwNoLQfrVJIyUsLIzly5ezadMmMjIycHR0BODIkSPMnTuXtWvXMmXKFDQ0NEhOTiY6Olopj7KyMhYuXMjYsWPJy8tri2J3OsHBwaxcuZJ169Zx8eJFZs6cyRdffMHMmTPJzc1tNd3UqVOpqqqS3peUlEj/vWLFCnR0dKT3ZmZmhIeHExUVpZpKCB2Ch4cHY8aMISoqiqysLPr06cOsWbMICwujoqJCKb6WlhY1NTXEx8fTv3//FvPU0dGhtLSUlJQURo8ereoqCB2Em5sbw4cP5/jx4+Tk5ODt7c3kyZPZvn07lZWVSvFNTEyYPHkySUlJREVF4eDgwMiRI7l79y5paWkAaGpqMnXqVGpqavj111+pqqrC2NiYurq6tq6e0M54eHjg7u7O2bNnqaiowNPTkxEjRnDo0CFkMlmLaSwtLfH39ycpKYmsrCy6du3KkCFDiImJobi4GGj6fBszZgz5+fnExsZSW1uLsbEx9fX1bVk94SGI6V5tR+XTvaqqqti5cycvvfQSkyZN4rvvvpOuRUREEBAQwMqVK6U//GnTpvH5558r5bN06VLmzZvHkCFDVF3kTiskJIS9e/fy888/c+vWLT7++GNyc3N59tlnH5iuuLiYoqIi6dXY2ChdKy8vV7jm7+/P3bt3RSNFzQ0YMIArV65w5coViouLiYmJoaKigr59+7YYv7y8nKNHj5KUlERtbW2LcXJzczl+/DjXrl2joaFBhaUXOpK+ffuSnJxMcnIyJSUlxMXFUVlZia+vb4vxfXx8qKioIC4ujpKSEpKTk7l69Sp+fn5SHE9PT/T19Tl48CC5ublUVFSQk5NDUVFRW1VLaKd69uzJ1atXycrKory8nLNnz6KlpSV1vrbE3d2dvLw8rl27RkVFBdeuXSM/P5+ePXtKcXr16kV1dTXnzp2jpKSE6upq8vPzFToIBUHdqLyREh4ejoeHBx4eHoSEhLB582bkcjkA9vb2JCUlkZiY+MA8Nm/eTFpaGu+9956qi9tpaWtr4+npyenTpxXC4+Pj6dOnzwPT7tixg8jISP73v/+1OLXrXtOmTePw4cPcvXv3scssdEyamprY29tLUwKb3b59m65duz6dQgmdkqamJra2tty5c0ch/M6dO9jb27eYxt7eXil+RkYGNjY2aGo2fSW6uLiQm5vLyJEjeeGFF3juuefo37+/6EFVc0ZGRhgYGCjM5mhsbKSgoAArK6tW01lZWSnNAMnNzVVI06VLF0pKSvD392fy5MkEBgbi4uLy5CshCB2IyhspoaGhhISEADB+/HgqKyul6VzLly9n4MCB+Pr64uzszNy5cwkLC1PoSU1NTeWdd95h27ZtaGuLzcgelYWFBdra2tLQcrOioqJWP1wLCwv517/+xVtvvcVbb71Feno6X3/9Nf369Wsxvre3Nz179uTnn39+4uUXOg4DAwM0NTWVegCrqqowMjJ6SqUSOqPmZ626ulohvLq6GkNDwxbTGBkZtRhfS0sLfX19oGnaqqurKxoaGvzyyy+cO3cOPz+/3+2kETq35ufj/k642tpa6Vpr6e5Pc/fuXYU0RkZGuLq6UllZSWxsLDdv3sTPzw8nJ6cnWAPhSdDU1GyXr85Ipb/6U1JSSEhIYM+ePU0309Zmzpw5hIWFERgYiJGREQcOHCAtLY2YmBji4+NZsWIFGzdu5PTp0+jp6TFv3jzWrFmDu7v7H7p3bW2t0rSRxsbGTvs/8mE1j2I109DQUAprlp6errBI9PLly9jZ2bFw4UJ+++03pfjTpk0jNTVVaQGqIMCDnzVBaE80NDSoqakhJiYGuVxOQUEBRkZG+Pn5cfbs2addPKGNODo6KqyRi42NfWJ53z8qp6GhQXFxsTSzpLS0FFNTU1xdXcVmDYLaUmkjJTQ0FJlMpjDFQy6Xo6OjQ0lJCRYWFgC4urri6urKkiVLWL16Ne7u7oSHhzN9+nTOnTvHhQsXePXVV4GmhoZcLkdbW5vIyEjGjBnT4r3XrVvHmjVrFMLs7OxwcHBQUW3bt5KSEmQymdKoiaWlpdLoyoNcuXKFiRMnKoXr6+szbtw4vvrqq8cuq9Cx1dTU0NjYqDRqYmhoqNSDLQiPo/lZu3/U5EHPWlVVVYvxGxoapN7uqqoq6bumWXFxMUZGRmhqaiqsyxM6r+zsbIV1SFpaWoDyyIient4DpzjfP2rSUpqamhqlHcLKy8vp1q3bY9VBEDoylQ0ryGQytmzZwoYNG7h48aL0unTpEk5OTmzbtq3FdM7OzhgaGlJVVYWpqSlXrlxRSL9s2TI8PDy4ePEigwcPbvX+q1atoqysTOFlZ2enquq2ezKZjKtXr+Lv768Q7u/vz6VLlx46n169elFYWKgUHhQUhK6uLgcPHnzssgodW2NjI7m5uUrTFJycnFrdglgQHkVjYyP5+fl0795dIbx79+6t7liYm5vbYvyCggKp8ZGTk4OZmZlCHHNzc6nxIqgHmUxGVVWV9CovL6empkbht4SGhgY2NjYP3FShqKhI6feHnZ2dQpqioiJMTEwU4piYmIiF8+2QhoZGu3x1RiobSYmIiKCkpITFixcrfdjPmjWL0NBQCgsLqa6uZuLEiTg5OVFaWspnn31GfX09QUFBaGpq4uPjo5DW1tYWfX19pfD76enpoaenpxCm7lO9tm7dyn/+8x+Sk5O5fPkyM2bMwN7enp9++gloWiNka2vLP/7xDwDmzZtHdnY2N2/eRFtbm2eeeYbAwEBWrFihlPe0adM4duwYZWVlbVonoX06d+4czzzzDLm5uWRnZ9OnTx9MTU2lBvHw4cMxMTFRaNTa2toCTedTGBoaYmtrS0NDg/RFrqmpibW1NdDUo2lsbIytrS11dXXiHAE1dvHiRYKCgsjPzyc3Nxdvb2+MjY2laTNDhgzByMiII0eOAJCYmEjv3r0JCAggKSkJe3t7vLy8iIyMlPJsjjNixAguX76MmZkZAwYM+EMdOkLnlJqaSq9evaioqKCyshJPT08aGhrIyMiQ4gwcOJCamhrpGUxNTWXUqFF4eHiQnZ1Nly5dsLOzUzgL6vr164wZM4ZevXpx584dLC0t6dGjhzgLSlBrKmukhIaGEhgYqNRAAZg5cyZr164lJCSExMREFi5cSF5eHhYWFvj5+REZGYmHh4eqiqa2IiMjMTMz489//jPW1tbcuHGD5cuXS4eeWVtbK+yIo6Ojw1/+8hdsbW2pra0lLS2N5cuXExcXp5Cvo6Mj/fr1Y9myZW1aH6H9SklJwcDAgKFDh2JkZERhYSG7d++WpjMYGxsr9Rr+6U9/kv67+YdjWVkZ33zzjZTm3jiDBg1i0KBBZGRkEB4e3ga1EtqjGzduoK+vz8CBAzEyMqKoqIiIiAjpPB5DQ0OFZ62iooJffvmFgIAAfH19qaqq4sSJE9IZKQCVlZXs37+fgIAA5s6dS1VVFZcuXWpxLZ6gXlJSUtDS0qJfv37o6upSXFzMiRMnFM5IuX86YVFREfHx8fj4+ODj40NlZSXx8fEKU61LSko4deoUvr6+eHl5UVVVxcWLFxUaP4KgbjTkarSS9d598AVBlebNm/e0iyCoCQMDg6ddBEFNqPOUaaFt/d75bU/T/dNF24v7t1bvDNR7/pMgCIIgCIIgCO2OaKQIgiAIgiAIgtCuiNMRBUEQBEEQBOEhdNadtNojMZIiCIIgCIIgCEK7olYjKaNHj37aRRDURGpq6tMugqAmZs+e/bSLIKiJGzduPO0iCIKgRtSqkSIIgiAIgiAIj0pM92o7YrqXIAiCIAiCIAjtimikCIIgCIIgCILQrojpXoIgCIIgCILwEMR0r7bTJo2UU6dOMXz4cIKCgjh06JDCtd27d7N+/XquXbtGY2Mjjo6OjB8/ng0bNgBw7NixFhe8X716lV69erVF8TuVoUOHMnr0aExNTcnNzWXv3r3cunWrxbi+vr4MHTqUrl27oq2tTW5uLocPHyYlJaXF+H379mXhwoVcuXKFzZs3q7IaQgcwcuRIxo0bh5mZGdnZ2YSHh7e68NbNzY0ZM2Zgb2+Prq4uxcXFnDhxgiNHjkhxHBwcmDp1Ko6OjlhbWxMeHk50dHRbVUdox+RyOQcPHuTkyZNUV1fj7OzM7Nmz6dKlS6tpsrOzOXDgABkZGRQXFzNz5kzGjBmjEOfAgQMcPHhQIczExIQPPvhAJfUQ2j9vb2/69OmDoaEhJSUlnDx5ktzc3FbjOzg4MHToUCwsLKiurubixYskJye3GNfV1ZWgoCBu3brF4cOHVVUFQegw2qSREhYWxvLly9m0aRMZGRk4OjoCcOTIEebOncvatWuZMmUKGhoaJCcnt/jDIyUlBVNTU+m9jY1NWxS9U+nbty/Tpk1j9+7d3Lp1i6FDh/LnP/+ZDz/8kNLSUqX4rq6uXL9+nYMHD1JTU8OgQYNYvHgxGzduJCsrSyGuhYUFU6ZMIS0trY1qI7RnAwYMYM6cOWzfvp0bN24wYsQIXnvtNf75z39SXFysFL+2tpaYmBgyMzOpq6vDzc2NkJAQamtriY2NBUBXV5eCggLOnz8vdrQSFERFRXH06FEWLFiAra0thw4d4osvvuDdd99FX1+/xTT19fVYWVnh5+fH7t27W83bwcGB5cuXS+81NcUsaXXl6urK0KFDiY2NJTc3Fy8vL5555hnCw8OprKxUim9iYsLEiRO5evUq0dHR2NvbM3z4cGpqapQ6B42NjRkyZAjZ2dltVR1BaPdU3kipqqpi586dnD17ltzcXL777jveffddACIiIggICGDlypVSfHd3d6ZNm6aUj62tLebm5qoubqc2cuRIzpw5w5kzZwDYu3cvHh4eDBs2jAMHDijF37t3r8L7gwcP4uPjg7e3t0IjRUNDg/nz53P48GFcXFwwMDBQaT2E9i8oKIi4uDji4uIA2LlzJ97e3owcOZKff/5ZKf6dO3e4c+eO9L6oqAg/Pz969uwpNVLS09NJT08HYPr06W1QC6EjkMvlxMTEMG7cOPr27QvAggULWLVqFWfPnmX48OEtpnNycsLJyQmAffv2tZq/pqYmZmZmT7zcQsfTu3dvrl27xrVr14CmWSLdu3fHy8uLhIQEpfheXl5UVlZy6tQpAEpLS7GxsaFPnz4KjRQNDQ3Gjh3LuXPncHBwQFdXt20qJDwS0VHRdlT+Lx0eHo6HhwceHh6EhISwefNm5HI5APb29iQlJZGYmPi7+fj5+eHg4MDYsWOJiYlRdbE7HS0tLbp168b169cVwlNSUnB2dn6oPDQ0NNDT06O6ulohPDg4mKqqKqnxI6g3LS0tHB0dlaY0JCcn4+rq+lB5dO/eXRrJE4QHKSoqory8HE9PTylMR0cHNze3Vqey/hEFBQX87W9/49133yUsLIzCwsLHzlPoeDQ1NbGxsVHoTAHIzMzE3t6+xTR2dnZkZmYqhN25cwcbGxuFH7r9+/fn7t27UuNHEIQmKh9JCQ0NJSQkBIDx48dTWVlJdHQ0gYGBLF++nNjYWHx9fXFycsLf35/g4GDmz5+Pnp4e0DTU/s0339C/f39qa2v54YcfGDt2LMeOHWPEiBGt3re2tpba2lqFMJlMhra2eu4VYGRkhJaWFhUVFQrhFRUVmJiYPFQeo0aNQldXl4sXL0phzs7ODB48WFpDJAjGxsZoaWlRXl6uEF5eXq4wZbMlH374oZT+l19+kUZiBKE1zc/Z/Z9jpqamLU4t/COcnZ1ZuHAhtra2VFRUcOjQIT7++GP+/ve/Y2xs/Fh5Cx2Lvr4+mpqa1NTUKIRXV1fTvXv3FtMYGhoqNWpqamrQ0tJCX1+f6upq7O3t6dWrFz/99JPKyi4IHZVKf7GnpKSQkJDAnj17mm6mrc2cOXMICwsjMDAQIyMjDhw4QFpaGjExMcTHx7NixQo2btzI6dOnMTQ0lEZhmg0ZMoQ7d+7w8ccfP7CRsm7dOtasWaMQ5u/vz5AhQ1RT2Q6ieRSrmYaGhlJYS/z8/AgODiYsLEyae6unp8f8+fPZuXMnVVVVKimv0Hk8zI4o69evR19fHxcXF2bMmEF+fj5nz55tg9IJHUVCQgI//vij9P7ll18GlJ+vh/lc+z3e3t4K711cXHjvvfc4c+YMY8eOfez8hY7vYb9D740PTc+njo4OY8aM4fjx49y9e1dVRRSEDkuljZTQ0FBkMhldu3aVwpr/MEtKSrCwsACaFqO5urqyZMkSVq9ejbu7O+Hh4SxatKjFfP39/dm6desD771q1SrefPNNhbC///3vj1mjjquqqoqGhgalnmxjY+MWF/zdq2/fvsyZM4fvv/+e1NRUKdzKygorKysWL14shTV/AH/00Ud88MEHFBUVPcFaCB1BZWVli8+aiYmJ0ujK/Zqfl6ysLExNTZk8ebJopAgKevfurTBFVSaTAU0jKveuHamoqPjdkbs/Sk9Pj65du5Kfn/9E8xXav7t379LY2Ki05tLAwEBpdKVZdXU1hoaGCmH6+vo0NDRQW1uLhYUFpqamTJgwQbre/B365z//mR07dvzuZ6bQ9sQWxG1HZY0UmUzGli1b2LBhA8HBwQrXZs6cybZt23j11VeV0jk7O2NoaPjAnvkLFy7g4ODwwPvr6elJU8aaqetUL4CGhgYyMzNxd3fnypUrUri7uztJSUmtpvPz82Pu3Ln88MMPXL16VeFafn4+69evVwibMGECenp67N27t8Udw4TOr6GhgYyMDLy8vBSmBnp6enLp0qWHzkdDQ0Ot/2aFlunr6yvs2CWXyzE1NeXatWvStBuZTMaNGzeYOnXqE713fX09ubm5D722Sug8GhsbKSgooHv37ty+fVsK79q1q8L7e+Xl5UmbMzTr3r07BQUFNDY2UlpaSnh4uML1QYMGoaOjw8mTJ3+3A1EQOjuV/QKIiIigpKSExYsXK+2MMmvWLEJDQyksLKS6upqJEyfi5OREaWkpn332GfX19QQFBQHw6aef4uzsjLe3N3V1dWzdupXdu3c/cMtIoWXHjx9n3rx53Llzh9u3bzNkyBAsLCyknUeeeeYZTE1NpakUfn5+zJs3j59//pn09HRpznd9fT13795FJpMp7Q/f3KP0oH3jhc4vKiqKF154gfT0dNLS0hgxYgSWlpYcP34caNqdy9zcXDpPZ9SoURQXF0vPjZubG8HBwRw9elTKU0tLS+qc0NbWxtzcnG7dulFbW0tBQUEb11BoLzQ0NBg9ejSHDx/GxsYGW1tbDh8+jK6uLgMHDpTiff/995ibm0sNF5lMRk5ODtDUsC4tLeXOnTvo6elha2sLwJ49e/D19cXCwkJak3L37l0GDx7c9hUVnrrLly8zZswY8vPzycvLw8vLCxMTE2mTkEGDBmFkZCRt7pOcnIyPjw9Dhgzh6tWr2NnZ0atXL+n8p4aGBkpKShTuUVdXB6AULgjqSGWNlNDQUAIDA1vcunHmzJmsXbuWkJAQEhMTWbhwIXl5eVhYWODn50dkZKS0DqWuro633nqLrKwsDAwM8Pb25sCBA0ycOFFVRe+0Ll68iKGhIcHBwZiampKTk8O3334rfRiamJhIU/Cgaf2PlpYWs2bNYtasWVJ4QkICO3bsaPPyCx3HuXPnMDIy4plnnpEOc/z888+lhcxmZmZYWlpK8TU0NJg+fTrW1tZSj+WePXs4ceKEFMfc3Fzavhxg3LhxjBs3jpSUFLFxg5oLCgqivr6e8PBw6TDHV199VWHEpaSkRGGaRllZmcKhjNHR0URHR9OzZ0/eeOMNoGnL2M2bN1NZWYmxsTEuLi689dZbWFlZtVndhPYjLS0NfX19BgwYgKGhIcXFxRw8eFAa8TAyMlLYwKGiooKDBw8ydOhQfHx8qKqq4uTJk09k1znh6RHTvdqOhvxJrC7sIO5foyIIqiKG6YW2Ig62FNrKjRs3nnYRBDWxbNmyp12EVrm5uT3tIrSoM/59ihNpBEEQBEEQBEFoV8SqVEEQBEEQBEF4CGK6V9sRIymCIAiCIAiCILQropEiCIIgCIIgCEK7olbTvcQQndBWamtrn3YRBDUhNmkQ2oquru7TLoIgPHXit2TbESMpgiAIgiAIgiC0K6KRIgiCIAiCIAhCu6JW070EQRAEQRAE4VFpaor+/bbSJo2UU6dOMXz4cIKCgjh06JDCtd27d7N+/XquXbtGY2Mjjo6OjB8/XuEE6draWv71r3+xdetWcnNz6datG6tXr+aFF15oi+J3KkOHDmXUqFGYmpqSm5vLvn37Wj391tfXl6FDh9KlSxe0tbXJzc0lMjKSlJSUFuP37duXBQsWkJiYyObNm1VZDaEDGDNmDBMnTpROnN+2bRvXr19vMW7Pnj2ZM2cODg4O6OrqUlhYyLFjxzh8+LBCPENDQ2bOnCmd+FxYWMiPP/7I5cuX26JKQjsll8s5cuQIZ86coaamBkdHR6ZOnYq9vf0D0125coXIyEiKioqwsrJi3Lhx+Pj4SNcbGho4cuQIFy5coKKiAlNTU/r378+YMWPEDxU11atXL3x9fTEwMKC0tJQzZ86Ql5fXanx7e3sGDRqEubk5NTU1XL58Wek71MnJiX79+mFqakp5eTm//fYb6enpqq6KILR7bdJICQsLY/ny5WzatImMjAwcHR0BOHLkCHPnzmXt2rVMmTIFDQ0NkpOTiY6OVkg/e/Zs8vLyCA0Nxc3Njfz8fGQyWVsUvVPp27cvU6dOZc+ePdy6dYshQ4bw4osvsn79ekpLS5Xi9+jRg+vXr3Pw4EFqamoYOHAgL7zwAp999hlZWVkKcS0sLJg8eTJpaWltVBuhPRs0aBDz589ny5YtXL9+ndGjR7NixQpWrVpFcXGxUvza2lqOHDnCnTt3qK2txd3dneeff57a2lqOHTsGgJaWFitXrqS8vJwvvviC4uJiLC0tuXv3bhvXTmhvjh8/TmxsLLNnz8ba2pro6Gg2bdrEypUr0dPTazFNeno627dvJzg4GG9vb5KSkti2bRsvvfSS9B11/Phx4uPjmT17NnZ2dmRmZrJr1y709fUJCAhoyyoK7YCLiwuDBw/m9OnT5OXl0atXL4KDg9mzZw9VVVVK8Y2NjQkKCuL69escP34cOzs7hgwZwt27d6VGiI2NDaNHj5YaJk5OTowePZoDBw5QUFDQ1lUUhHZF5Y2Uqqoqdu7cydmzZ8nNzeW7777j3XffBSAiIoKAgABWrlwpxXd3d2fatGnS+0OHDnH8+HFu3ryJpaUlAM7Ozqoudqc0YsQIEhISOHPmDAD79u3Dw8ODoUOHcvDgQaX4+/btU3j/66+/4uPjg5eXl0IjRUNDg/nz53P48GF69OiBgYGBaisitHvjx4/nxIkTHD9+HIDt27fj6+vL2LFj2bVrl1L8jIwMMjIypPeFhYX0798fd3d3qZEyYsQIjI2N+c9//kNDQwMARUVFqq+M0K7J5XLi4uIYM2aMNAoyZ84c/v3vf3PhwgX8/f1bTBcXF4ebmxujR48GwNbWlps3bxIXF8e8efOApoaMl5cXnp6eAFhaWnLp0iUyMzPboGZCe+Pj48P169elEeEzZ87QtWtXevXqxfnz55Xi9+rVi6qqKuk7t6ysDGtra3x9faVGire3N9nZ2dJo8OXLl7G3t8fb21v67BPaF7G7V9tR+Xh1eHg4Hh4eeHh4EBISwubNm5HL5UDTMGhSUhKJiYmtpt+/fz8DBgxg/fr1dO3aFXd3d9566y1qampUXfRORUtLi27duikNM6ekpDx0o09DQwM9PT2qq6sVwoODg6msrCQhIeFJFVfowLS0tHB2dlb6u05MTMTNze2h8nB0dMTNzU3hefXz8+PGjRssXLiQzz77jPfff59JkyaJLww1V1xcTEVFBT179pTCtLW16dGjxwOnzKSnp+Pu7q4Q5u7urpDG2dmZtLQ0qUc7Ozub27dv06tXrydcC6G909TUxMrKiuzsbIXwrKwsbG1tW0xja2urNOsgKysLa2tr6XOrtTit5SkIT9KXX36Ji4sL+vr69O/fn9jY2AfGr62tZfXq1Tg5OaGnp4erqythYWEqK5/KR1JCQ0MJCQkBmnpXKysriY6OJjAwkOXLlxMbG4uvry9OTk74+/sTHBzM/PnzpSH65p4tfX19fv75ZwoLC3n55ZcpLi5+4D9MbW2t0lkVMpkMbW313CvAyMgILS0tpTMVKisrMTExeag8Ro4cia6uLpcuXZLCnJ2dGTRoEJ988skTLa/QcZmYmKClpUVZWZlCeFlZGWZmZg9M+3//939S+p9//lkaiYGmaRGenp6cPn2aTz75BDs7OxYuXIiWlpbSqJ+gPioqKgCUPseMjY0pKSlpNV1lZSXGxsZKaZrzAxg1ahR3795lw4YNaGhoIJfLGTduHH379n1yFRA6BD09PTQ1NZU6SGtqajA0NGwxjYGBQYvxNTU10dfXp6amptU4YkaCoGrh4eG88cYbfPnllwwbNoyvv/6aCRMmkJycLE15vV9bL79Q6S/2lJQUEhIS2LNnT9PNtLWZM2cOYWFhBAYGYmRkxIEDB0hLSyMmJob4+HhWrFjBxo0bOX36NIaGhjQ2NqKhocG2bdukHziffPIJs2bN4r///W+rf8jr1q1jzZo1CmH+/v4MHTpUlVVu95pHsf4oPz8/goOD2bx5s9TQ0dPTY968eezatavF+biCerv/WWv+kfcg77//Pvr6+ri6ujJ79mzy8/OJj48HmnoyKyoqpNHY27dvY25uzsSJE0UjRY1cuHBB+k4BWLRoUYvx5HL5746ytXT93rBLly5x4cIF5s6di52dHTk5Ofzyyy/SAnpB/TzK59qTuI/QfrTX0fuWOuf19PRaXZf3ySefsHjxYpYsWQLAp59+yuHDh/nqq69Yt26dUvynsfxCpY2U0NBQZDIZXbt2lcLkcjk6OjqUlJRgYWEBgKurK66urixZsoTVq1fj7u5OeHg4ixYtwsHBga5duyr0wHp6eiKXy8nMzFQY4r/XqlWrePPNNxXC/vGPf6iglh1DVVUVDQ0NLfY23ttz2JK+ffsye/ZstmzZQmpqqhRuZWWFlZWVwi5rzX+869ev58MPPxRrBtRQRUUFDQ0NmJubK4Q371zzIIWFhQBkZmZiZmbGtGnTpEZKaWkpDQ0NCl/eOTk5mJubo6WlJa1TETo3Ly8vunfvLr1v7sVr3n2rWVVVldJIyb1a+uy7f3Tl4MGDjBo1Sho5cXBwoKSkhJiYGNFIUTO1tbU0NjYqjZo0j4i0pKUREQMDAxobG6UNP1qLIzYEEf6oljrn33vvPf75z38qxa2rq+P8+fO88847CuHBwcGcOnWqxfzvXX7xww8/YGRkxJQpU/j3v/+tspE/lTVSZDIZW7ZsYcOGDQQHBytcmzlzJtu2bePVV19VSufs7IyhoaHUMz9s2DB27dql8OVx/fp1NDU16datW6v3b6n1qK5TvaBpK83MzEzc3d0V1gq4u7uTlJTUajo/Pz/mzJnD1q1buXr1qsK1/Px8PvroI4WwCRMmoKenx969e1vcMUzo/BoaGrh9+zbe3t4Ki0m9vb25cOHCH8rr3r/Z1NRU/P39FXou7ezsKCkpEQ0UNXL/Z7tcLsfExITU1FSpQ0wmk3Hz5k0mTJjQaj5OTk6kpqYyfPhwKez69es4OTlJ7+vr65V6TTU1NUUvtxpqbGykqKiILl26KKxb6tKli8KmH/fKz89XaFA3xy8sLJSeofz8fLp06aLwPdy1a1fy8/NVUAuhM2upc761UZTCwkIaGhqws7NTCLezsyM3N7fFNI+6/OJxqOxXe0REBCUlJSxevFhpHvqsWbMIDQ2lsLCQ6upqJk6ciJOTE6WlpXz22WfU19cTFBQEwLx58/j3v//NokWLWLNmDYWFhaxcuZIXXnhBzNn8g06cOMFzzz1HZmYmt2/fxt/fHwsLC06fPg0gnWnx448/Ak0NlOeee469e/eSnp4ujcLU19dz9+5dZDKZ0sPc3KPU2kMuqIdDhw6xdOlSbt26xY0bNxg9ejRWVlYcPXoUgGeffRYLCwu++eYbAMaOHUtRURE5OTlAU+N5woQJHDlyRMrz6NGjBAYGMn/+fKKiorC3t2fy5MlERUW1fQWFdkNDQ4OAgABiYmKwtrbG2tqamJgYdHR08PPzk+KFh4djamoqNVya52AfO3YMLy8vkpOTuXHjBi+99JKUxtPTk6NHj2Jubo6dnR3Z2dnExsYyYMCANq+n8PQlJiYyYsQICgsLyc/Px8PDA2NjY65duwZA//79MTIy4sSJEwBcu3YNT09PBg0aREpKCra2tgo7FgIkJyczceJEfH19pSMaunTpwoEDB55GFYWH0F7PSHrQ1K7W3N8J86Bpso+6/OJxqKyREhoaSmBgYIsLZWfOnMnatWsJCQkhMTGRhQsXkpeXh4WFBX5+fkRGRuLh4QE0DclHRUWxfPlyBgwYgJWVFbNnz+Y///mPqoreaV28eBFDQ0OCgoIwNTUlJyeHTZs2SYtLTU1NFabo+Pv7o6WlxcyZM5k5c6YUfvbsWXbs2NHWxRc6kISEBIyNjZk6dSrm5uZkZWXxySefSNP/zMzMpDmt0PRB+eyzz2JjY0NDQwP5+fns2rWLmJgYKU5xcTEfffQR8+bN4z//+Q+lpaVERkaKL3OBkSNHUl9fz969e6mpqaF79+4sWbJE4Qu7tLRU4cvX2dmZ5557jsjISCIjI7G0tGT+/PkKC0anTp3K4cOH2bt3L5WVlZiamjJ48GDGjh3bpvUT2odbt26hp6dH3759MTQ0pKSkhMjISGnmh6GhIUZGRlL8yspKoqKiGDRoEJ6enlRXVxMfH68wEpOfn8+xY8fo168f/fr1o6KigpiYGHFGiqBS1tbWaGlpKXUo5+fnK42uNHvU5RePQ0OuRuPWK1aseNpFENRE89oKQVC16dOnP+0iCGqipYNYBUEV7l3r2t40n8fU3jzoOI+WDB48mP79+/Pll19KYV5eXkydOrXFhfPffPMNb7zxBvn5+dLyi3379jFjxgwqKytVMpLSPsesBEEQBEEQBKGd0dDQaJevP+rNN99k06ZNhIWFcfXqVf7yl7+QkZHBsmXLgKY1LgsXLpTiz5s3DysrKxYtWkRycjInTpxQ+fIL9V1JLgiCIAiCIAhqaM6cORQVFfGvf/2LnJwcfHx8OHjwoLR5SE5OjsKmEE9j+YVopAiCIAiCIAiCmnn55Zd5+eWXW7z23XffKYX16tWrTTerEY0UQRAEQRAEQXgI7XV3r85IrRopCxYseNpFENSEOIhLaCv/+9//nnYRBDVhb2//tIsgCIIaEc1BQRAEQRAEQRDaFbUaSREEQRAEQRCER/UoO2kJj0aMpAiCIAiCIAiC0K60yUjKqVOnGD58OEFBQRw6dEjh2u7du1m/fj3Xrl2jsbERR0dHxo8fz4YNGwB4/vnn+f7775Xy9PLyIikpqS2K36nI5XJ++uknoqOjqayspGfPnrzwwgt079691TTR0dGcOHGCO3fuAODi4sJzzz2Hm5ubFKehoYFdu3YRFxdHaWkpFhYWjBw5khkzZohFZmpKLpezd+9ejh07RlVVFa6urixYsIBu3bq1mubYsWOcPHmSzMxMoOlU8FmzZuHq6irF+eWXXzh//jw5OTno6OjQs2dPZs+ejYODg8rrJLRPY8aMYcKECZibm5OVlcX27du5fv16i3HvfV50dXUpKioiJiaGyMhIKU5AQABLlixRSvviiy9SX1+vsnoI7Z+/vz8jRozAxMSEvLw8IiIiuH37dotxvb298ff3x8HBAW1tbfLy8jhy5AipqalSHFtbW4KDg+natSsWFhb88ssvnDx5so1qIwjtW5s0UsLCwli+fDmbNm0iIyMDR0dHAI4cOcLcuXNZu3YtU6ZMQUNDg+TkZKKjo6W0Gzdu5IMPPpDey2Qy+vTpw7PPPtsWRe909u/fz4EDB3jppZdwcHBgz549vP/++/zf//1fq4fxJCUlMXToUDw8PNDR0WH//v28//77bNiwAUtLS6Dp1NEjR47w8ssv061bN27evMlXX32FoaEhEydObMsqCu3EwYMHOXToEC+++CL29vbs37+fjz76iA8++KDVZ+3atWv4+/vj5uaGjo4OBw8e5OOPP+b999+XnrWUlBTGjh2Li4sLjY2N/PTTT3z00UesW7cOPT29tqyi0A4MGjSIefPmsWXLFlJTUxk9ejRvvvkmf/vb31o8Ib22tpYjR45w584d6urq6NmzJ88//zy1tbUcP35cilddXc2qVasU0ooGinrr3bs3kyZNYt++fdy+fZvBgwezaNEiPvnkE8rKypTiu7i4kJqayuHDh6mpqWHAgAH86U9/4ssvvyQ7OxtAaihfvnyZSZMmtXWVhEcgpnu1HZV3cVdVVbFz505eeuklJk2apLDvckREBAEBAaxcuRIPDw/c3d2ZNm0an3/+uRTHzMwMe3t76XXu3DlKSkpYtGiRqove6cjlcg4ePMj06dMZPHgwjo6OvPLKK9TW1hIXF9dqutdee41x48bh7OxM165dWbp0KXK5nCtXrkhxUlNTGTBgAP369cPW1hZ/f3969+7NzZs326JqQjsjl8s5fPgwU6ZMYcCAAXTr1o0XX3yRuro64uPjW023bNkyxo4di5OTE126dOGFF16gsbGR5ORkKc5bb73F8OHD6datG46OjixZsoSioiJu3brVFlUT2plx48Zx4sQJTpw4QU5ODtu3b6e4uJgxY8a0GD8jI4MzZ86QnZ1NYWEhp0+f5sqVK3h4eCjFLSsrU3gJ6i0gIIBz585x9uxZCgoKiIiIoKysDH9//xbjR0REcOLECTIzMykqKuLw4cMUFRXh6ekpxcnMzOTXX3/l8uXLNDQ0tFVVBKFDUHkjJTw8HA8PDzw8PAgJCWHz5s3I5XKgaTvDpKQkEhMTHzq/0NBQAgMDpRMxhYeXn59PaWkpvXv3lsJ0dHTw8vJqdWpES2pra5HJZBgbG0thHh4eJCYmSr1Dt2/fJiUlBT8/vydXAaHDKCgooKysDB8fHylMR0cHDw8PhakOv6e2tpaGhgaFZ+1+NTU1AA+MI3ROWlpaODs7K32HJCYmKkxHfRBHR0d69uzJtWvXFML19PT4+OOP+eSTT3jjjTekGQCCetLS0qJr165Kn1+pqakP/XtEQ0MDPT09qqurVVFEQeh0VD7dKzQ0lJCQEADGjx9PZWUl0dHRBAYGsnz5cmJjY/H19cXJyQl/f3+Cg4OZP39+i9M2cnJy+PXXX9m+ffvv3re2tpba2lqFsLq6OnR1dZ9MxTqg0tJSoGl06l5mZmYUFBQ8dD7bt2/H0tISX19fKWzq1KlUV1fz5ptvoqmpSWNjI3PmzGHYsGFPpOxCx9Lc62xqaqoQbmpqSlFR0UPns2vXLiwsLPDy8mrxulwuZ/v27bi7uz9wrYvQOZmYmKClpUV5eblCeHl5udLn3P0++eQTKf3evXs5ceKEdC0nJ4dNmzaRmZmJgYEBQUFBrF69mnfffZe8vDyV1EVo3wwNDdHS0qKiokIhvKKiAnd394fKY/jw4ejo6HD58mVVFFFoI2KdbdtRaSMlJSWFhIQE9uzZ03QzbW3mzJlDWFgYgYGBGBkZceDAAdLS0oiJiSE+Pp4VK1awceNGTp8+jaGhoUJ+3333Hebm5kybNu13771u3TrWrFmjELZ06VKWLVv2xOrX3sXGxvLtt99K79955x1AeT6lXC5/6DmW+/bt4+TJk7z33nsKDb5Tp04RFxfH8uXL6d69O7dv3+b777/H0tKSkSNHPoHaCO3ZqVOnFKZyvvnmm0DLc3cf9lk7cOAA8fHxvPPOO612Lvzwww9kZmayevXqP15oodNoHp1v9jDP2Nq1a9HX18fV1ZVnn32WvLw8zpw5A0BaWhppaWlS3NTUVNasWUNgYCDbtm17soUXOjQNDQ2l568lffr0ITAwkC1btlBVVdUGJROEjk+ljZTQ0FBkMhldu3aVwuRyOTo6OpSUlGBhYQGAq6srrq6uLFmyhNWrV+Pu7k54eLjCuhO5XE5YWBgLFix4qNGQVatWST+Umt0/nN/ZDRgwgJ49e0rvmxd9Nu++1exheh2haVelvXv38ve//11peHvbtm1MnTpVGjlxdHSkoKCAvXv3ikaKGvDz81PYgav5WSsrK8Pc3FwKLy8vVxpdacnBgweJiIjgr3/9a6vTbH744QcuXLjA3/72N2lRvaBeKioqaGhoUPr8MjEx+d01JIWFhUDTmgBTU1OmTZsmNVLuJ5fLuXXrFnZ2dk+m4EKHU11dTUNDAyYmJgrhxsbGVFZWPjBt7969mTlzJtu2bePGjRuqLKYgdCoqG7OSyWRs2bKFDRs2cPHiRel16dIlnJycWu2NcnZ2xtDQUKmn4fjx49y4cYPFixc/1P319PQwNTVVeKnbVC8DAwOFTQe6deuGubm5wlCzTCYjOTn5d4er9+/fz+7du1m1apXCj9FmtbW1Sr2XmpqaD9XDJHR8BgYG2NnZSa+uXbtiZmamsFZAJpORkpKi0HBuycGDB9m/fz8rVqzAxcVF6bpcLmfLli2cO3eOt99+GxsbmydeH6FjaGho4Pbt23h7eyuEe3t7/6EfgxoaGujo6DwwTvfu3aUps4L6aWhoICsrS2mtk5ubG+np6a2ma96NdMeOHaSkpKi6mEIb0NDQaJevzkhlIykRERGUlJSwePFipV6uWbNmERoaSmFhIdXV1UycOBEnJydKS0v57LPPqK+vJygoSCFNaGgogwcPVliIK/wxGhoaTJw4kb179+Lg4IC9vT179+5FT0+PgIAAKd4XX3yBpaUl8+bNA5qmeO3cuZPXXnsNW1tb6YtaX18ffX19APr378/PP/+MtbU13bp14/bt2xw4cIDRo0e3eT2Fp09DQ4Nx48YRERGBnZ0d9vb2/PLLL+jq6irshPP1119jYWHB7NmzgaYpXnv27GHZsmVYW1u3+Kxt2bKF+Ph4Xn/9dfT19aU4hoaGatcRIcDhw4f585//zO3bt7lx4wajRo3CysqKmJgYoOn7xsLCQpr6OnbsWIqKisjJyQGazk0ZP348R44ckfKcOnUqaWlp5OXlYWBgQGBgII6Ojvzwww9tX0Gh3YiLi2P27NlkZWWRnp7O4MGDMTc3l0bgxo0bh5mZGTt37gSaGiizZ8/ml19+ISMjQ9rco76+Xlozq6Wlha2trfTfpqamODg4UFdX94fW7wlCZ6SyRkrzLlwtTSOaOXMma9euJSQkhMTERBYuXEheXh4WFhb4+fkRGRmpsB1kWVkZu3fvZuPGjaoqrtqYMmUKdXV1hIaGUlVVhZubG3/7298Uzq0oKipSWBgWFRWFTCbjk08+Uchr1qxZ0nk1ixYtIjw8nNDQUMrKyrC0tCQwMJBZs2a1TcWEdmfixInU1dWxZcsWqqur6dGjBytXrlR41oqLixWetaNHjyKTyfjiiy8U8po2bRrTp0+X4kDTurN7LVmyhOHDh6uqOkI7lZCQgLGxMVOnTsXMzIysrCw++eQT6Qeeubk5VlZWUnwNDQ1mzZqFjY0NDQ0N5Ofns2vXLo4dOybFMTQ05Pnnn8fMzIyamhrS09NZt26d2OZazV2+fBlDQ0PGjh2LiYkJubm5fPfdd1JHiampqcL01sGDB6OlpcW0adMU1tKeP3+eXbt2SWlef/116drIkSMZOXIkN2/e5JtvvmmLaglCu6UhV6P5OBcvXnzaRRDUxN27d592EQQ18b///e9pF0FQE/b29k+7CIKauPcQ7/Zm0KBBT7sILUpISHjaRXjixD5qgiAIgiAIgiC0K6KRIgiCIAiCIAhCu6LywxwFQRAEQRAEoTPorDtptUdiJEUQBEEQBEEQhHZFrUZSfu/AJUF4Umpqap52EQQ1oaWl9bSLIKgJQ0PDp10EQRDUiFo1UgRBEARBEAThUd27bb6gWuJfWhAEQRAEQRCEdkU0UgRBEARBEARBaFfaZLrXqVOnGD58OEFBQRw6dEjh2u7du1m/fj3Xrl2jsbERR0dHxo8fz4YNG6Q427ZtY/369aSmpmJmZsb48eP5+OOPFU4RFh6OXC5n//79HD9+XDoFfP78+XTt2rXVNFlZWezdu5f09HSKioqYO3cuQUFBCnFqamrYu3cvv/32GxUVFTg6OvLcc8/h4uKi6ioJ7ZRcLufgwYOcPHmSmpoanJycmDNnDg4ODq2mycnJISIigjt37lBcXMzMmTMZPXp0q/EPHz7ML7/8wqhRo5g1a5YqqiF0AKNHj2bcuHGYm5uTlZXFjh07SE1NbTGum5sbs2bNwsHBAV1dXYqKijh+/DhRUVFSnH79+vHMM89ga2uLlpYWeXl5REZGcvr06baqktBODRgwgCFDhmBiYkJ+fj6RkZFkZGS0GNfY2JigoCAcHBywsrLizJkzREZGKsXT09NjzJgx9OrVCwMDA0pKSoiKiuLGjRuqro7wCMTuXm2nTUZSwsLCWL58OXFxcQp/zEeOHGHu3LnMmjWLhIQEzp8/z/vvv09dXZ0UJy4ujoULF7J48WKSkpLYtWsXZ8+eZcmSJW1R9E7n119/JTIykvnz5/P3v/8dU1NTNmzY8MCF3nV1ddjY2DBz5kzMzMxajPP999+TnJzMkiVLWLNmDd7e3mzYsIGSkhJVVUVo544cOUJMTAyzZ89m5cqVmJqa8vnnn3P37t1W09TV1WFtbc2UKVMwNTV9YP7p6emcOnXqgQ1sofMbOHAgc+fO5cCBA6xZs4bU1FTeeOMNLC0tW4xfV1fH0aNH+fDDD/n73/9OREQE06dPZ8SIEVKcqqoqIiIiWLt2Le+99x4nT55k0aJFeHt7t1W1hHbIy8uLcePGERcXxzfffENGRgbz5s1r9bNKS0uL6upq4uLiyM3NbTGOpqYmISEhmJub89NPP/Hf//6XiIgIKioqVFkVQegQVN5IqaqqYufOnbz00ktMmjSJ7777TroWERFBQEAAK1euxMPDA3d3d6ZNm8bnn38uxYmPj8fZ2ZnXXnsNFxcXAgICWLp0KefOnVN10TsduVzOkSNHeOaZZ+jfvz/dunVj8eLF1NXVcebMmVbTubi4MHv2bAYPHoy2tvLgW11dHefPn2fWrFl4eHhgZ2fH1KlTsba2JiYmRpVVEtopuVxOTEwM48aNo2/fvnTp0oUFCxZQX1//wL9dJycnpk+fzoABA1p81prV1tby3Xff8dxzz2FgYKCKKggdRHBwMLGxscTGxpKTk8OOHTsoLi5m1KhRLcbPyMggISGB7OxsioqKiI+PJzExEXd3dylOSkoKFy5cICcnh4KCAo4cOUJmZiY9e/Zso1oJ7dGQIUO4cOECFy5coLCwkMjISMrKyhgwYECL8cvKyjh8+DCXL1+mtra2xTh+fn4YGBgQHh7OnTt3KCsr486dO+Tl5amyKoLQIai8kRIeHo6HhwceHh6EhISwefNm5HI5APb29iQlJZGYmNhq+qFDh5KZmcnBgweRy+Xk5eXx008/8cwzz6i66J1OYWEhZWVlCr2BOjo6eHh4kJaW9sj5NjQ00NjYiI6OjkK4jo6OGK5WU0VFRZSXl9OrVy8pTEdHBzc3N27evPnY+YeHh+Pj46OQv6B+tLS0cHJyIikpSSE8OTkZNze3h8rD0dERNzc3UlJSWo3j6emJvb09169ff6zyCh2XpqYmDg4OSt+VN2/epHv37o+cr7u7O5mZmUyYMIE333yTZcuWERAQIKYUtWMaGhrt8tUZqXxNSmhoKCEhIQCMHz+eyspKoqOjCQwMZPny5cTGxuLr64uTkxP+/v4EBwczf/589PT0gKZGyrZt25gzZw53795FJpMxZcoUhdEW4eGUlZUBKA1Nm5qaUlRU9Mj5GhgY4OrqSkREBA4ODpiZmXHmzBlu3bqFra3tY5VZ6JjKy8sBMDExUQg3MTGhuLj4sfI+d+4cd+7c4a9//etj5SN0fCYmJmhpaUnPW7OysjJ8fHwemPajjz6S0u/bt4/Y2FiF6wYGBnz88cdoa2sjl8vZunUrycnJT7wOQsdgaGiIpqYmVVVVCuFVVVUYGRk9cr4WFha4uLhw5coVfvzxRywtLZkwYQKampqcOHHicYstCB2aShspKSkpJCQksGfPnqabaWszZ84cwsLCCAwMxMjIiAMHDpCWlkZMTAzx8fGsWLGCjRs3cvr0aQwNDUlOTua1117j3XffZdy4ceTk5LBy5UqWLVtGaGhoq/eura1VGl6tq6tDV1dXlVVuV+Lj49myZYv0/vXXXweUF33J5fLHboUvWbKEzZs389Zbb6GpqYmTkxODBw8mPT39sfIVOoazZ8/y448/Su9feukloOUFho/zrJWUlLB7925eeeUVpZE7QWimoaEhjdi35sMPP0RPTw9XV1dmzpxJfn4+CQkJ0vW7d++yZs0a9PT08PT0ZM6cORQUFDxwxEUQ/igNDQ1pDZRcLicnJwcTExOGDBkiGimC2lNpIyU0NBSZTKawsFUul6Ojo0NJSQkWFhYAuLq64urqypIlS1i9ejXu7u6Eh4ezaNEi1q1bx7Bhw1i5ciUAvXv3xsjIiOHDh/Of//yn1Z2C1q1bx5o1axTCFi1axAsvvKCi2rY/ffr04b333pPey2QyoKmX0dzcXAqvqKj43UXKv8fW1pa3336b2tpaampqMDc353//+x/W1taPla/QMfj6+uLs7Cy9b37WysvLFTZbqKioUBpd+SMyMjKoqKhg/fr1UlhjYyNpaWmcOHGCTz/9VBy0pUYqKipoaGhocXT4/tGV+xUWFgJNuxeampoydepUhUaKXC4nPz8fgDt37uDg4MDEiRNFI0VNVVdX09jYqDRqYmRkpDS68kdUVlbS0NCg0KguLCzExMQETU1NGhsbHzlvQTXEd0zbUVkjRSaTsWXLFjZs2EBwcLDCtZkzZ7Jt2zZeffVVpXTOzs4YGhpKf/TV1dVKC2i1tLQAHthTtmrVKt58802FMHVbbG9gYKCwqFgul2NmZkZycjJOTk5A0/+nlJSUJ7Z9q56eHnp6elRVVZGYmMizzz77RPIV2jd9fX309fWl93K5HFNTU65duybN15bJZNy4cYOpU6c+8n08PDz429/+phC2detW7OzsCAoKEl8eaqahoYH09HS8vb25cOGCFO7l5aXw/mE8aKMGaOrx/r04QufV2NhITk4OPXr0UGio3v/+j7pz547S1ERLS0sqKipEA0VQeyr7xI2IiKCkpITFixcrbVs7a9YsQkNDKSwspLq6mokTJ+Lk5ERpaSmfffYZ9fX10jkckydP5sUXX+Srr76Spnu98cYbDBo0iC5durR6/+Yfy/dSp6leLdHQ0CAwMJADBw5gZ2eHra0tBw8eRFdXl8GDB0vxNm3ahIWFBTNnzgSaflxmZ2dL/11SUkJGRgZ6enrY2dkBkJiYiFwux97envz8fHbt2oW9vT3Dhg1r+4oKT52GhgajR48mMjISW1tbbGxsOHz4MDo6Ogo74WzZsgUzMzOp4SKTyaStOmUyGaWlpWRmZqKnp4eNjQ36+vpKf/e6uroYGRk98PNA6LwiIyNZsmQJt2/fJi0tjREjRmBpacnx48cBmDFjBhYWFtL04NGjR1NcXExOTg4APXv2ZNy4cRw9elTKc+LEidy+fZv8/Hy0tbXp3bs3Q4YMYevWrW1fQaHdOH36NNOnTycnJ4fMzEz69euHmZkZ58+fB2DMmDGYmJiwb98+KU3zd2Tz55SdnR0NDQ3SSN65c+cYOHAg48ePJyEhASsrKwICAhRG9QRBXamskRIaGkpgYGCL52rMnDmTtWvXEhISQmJiIgsXLiQvLw8LCwv8/PyIjIzEw8MDgOeff56Kigq++OILVqxYgbm5OWPGjOHDDz9UVdE7tQkTJlBfX8/WrVupqqqiR48evPnmmwojLsXFxQrrBkpLSxWmzh0+fJjDhw/j4eEhLV6uqalh9+7dlJSUYGRkRP/+/Zk+fbroeVRjgYGB1NXVER4eTnV1Nc7Ozrz66qsKIy73P2tlZWV88MEH0vvo6Giio6Nxc3PjjTfeaMviCx3E2bNnMTY2ZvLkyZiZmZGVlcXGjRulzUDMzc0VzkzR0NBg5syZWFtb09DQQEFBAbt375YaNdDUyRUSEoKFhQX19fXk5OSwadMmzp492+b1E9qP5ORkDA0NGTFiBMbGxuTn57N9+3ZpUxpjY2Ol3zxLly6V/rtLly74+vpKHbLQNCV227ZtBAcHs2zZMsrLy0lISODkyZNtVzHhD+msO2m1Rxry31td2InExcU97SIIauJBh2MKwpO0Y8eOp10EQU00TxMWBFV79913n3YRWnXvwa/tSWfcaEFM4BYEQRAEQRAEoV0Rc3EEQRAEQRAE4SGIDVrajviXFgRBEARBEAShXRGNFEEQBEEQBEEQ2hW1mu4lhuiEtlJXV/e0iyCoCSsrq6ddBEFNiMN5BUHs7tWWxK92QRAEQRAEQRDaFdFIEQRBEARBEAShXVGr6V6CIAiCIAiC8KjEdK+2I0ZSBEEQBEEQBEFoV9pkJOXUqVMMHz6coKAgDh06pHBt9+7drF+/nmvXrtHY2IijoyPjx49nw4YNUpz//ve/fPHFF9y+fRtHR0dWr17NwoUL26LonY5cLmffvn0cP36cqqoqevTowYIFC+jatWuraY4fP87JkyfJysoCwNnZmZkzZ9KjRw8pTkpKCr/++ivp6emUlpayfPly+vXrp/L6CO2XXC7n8OHDxMfHU11djZOTEzNnzsTe3r7VNLm5ufz6669kZmZSUlLC1KlTGTlypEKctLQ0YmJiyMzMpLy8nEWLFuHr66vq6gjt2JAhQxg5ciQmJibk5eWxf/9+bt++3WJcHx8f/P396dKlC9ra2uTl5REVFcX169cV4owZMwYrKyu0tLQoLCzkxIkT/Pbbb21UI6G98vb2xs/PD0NDQ4qLizl58iQ5OTmtxu/SpQtDhw7F0tKSqqoqLl68SFJSknTdw8ODsWPHKqX7+uuvaWhoUEkdBKGjaJNGSlhYGMuXL2fTpk1kZGTg6OgIwJEjR5g7dy5r165lypQpaGhokJycTHR0tJT2q6++YtWqVXz77bcMHDiQhIQEXnzxRSwsLJg8eXJbFL9TOXjwIIcPH2bx4sXY29vzyy+/8PHHH7N27VoMDAxaTHPt2jX8/f1xc3NDR0eHgwcP8vHHH/P+++9jYWEBQG1tLd27dycgIID//ve/bVkloZ06evQox48f57nnnsPGxoaoqCj+97//8c4776Cvr99imrq6OqysrOjbty979+5tNU6XLl0YNGgQ3333neoqIHQIffr0YfLkyezdu5fbt28zePBgFi9ezIYNGygtLVWK7+LiQmpqKocOHeLu3bsMGDCA559/ni+++ILs7GwAqquriY6OpqCgAJlMhqenJ88++yyVlZUKjRlBvbi5uREQEMCJEyfIzc3Fy8uLSZMm8eOPP1JZWakU38TEhGeeeUb6XWNvb8+IESOoqanh5s2bUrza2lq2b9+ukFY0UNovsVNs21F5I6WqqoqdO3dy9uxZcnNz+e6773j33XcBiIiIICAggJUrV0rx3d3dmTZtmvT+hx9+YOnSpcyZMweAHj16EB8fz4cffigaKX+QXC4nKiqKSZMmMWDAAACWLFnC66+/Tnx8PKNHj24x3dKlSxXeL1q0iHPnzpGcnMywYcMA6N27N71791ZtBYQOQy6Xc+LECQIDA6XnYt68ebz77rv89ttvDB06tMV0jo6OUidGREREi3E8PT3x9PRUTcGFDmf48OGcPXuWhIQEAH755Rfc3d3x9/dXGrlvvn6vQ4cO4eXlhZeXl9RIufcHJMDJkycZMGAAzs7OopGixvr06cPVq1e5evUq0PRcODo64uPjQ3x8vFJ8b29vKisrOXnyJAAlJSXY2trSt29fpWespqZG9RUQhA5G5c3B8PBwPDw88PDwICQkhM2bNyOXywGwt7cnKSmJxMTEVtPX1tYq9boaGBiQkJBAfX29Ssve2RQUFFBWVoaPj48UpqOjg4eHBzdu3HjofGpra2loaMDIyEgVxRQ6geLiYioqKvDw8JDCtLW1cXV1bXUajiD8UVpaWnTt2lWp4ZCamoqzs/ND5aGhoYGenh7V1dWtxnFzc8PGxoZbt249TnGFDkxTUxMbGxvu3LmjEH7nzh3s7OxaTGNvb68UPyMjAxsbG4XeeB0dHRYsWMDChQuZOHGiOI9GEP5/Kh9JCQ0NJSQkBIDx48dTWVlJdHQ0gYGBLF++nNjYWHx9fXFycsLf35/g4GDmz5+Pnp4eAOPGjWPTpk1MmzaNfv36cf78ecLCwqivr6ewsBAHB4cW71tbW0ttba1CWF1dHbq6uqqtcDtWVlYGgKmpqUK4mZkZhYWFD53PTz/9hIWFBd7e3k+0fELnUV5eDjRNd7iXiYkJJSUlT6NIQidkZGSElpaW0lSbiooKpWevNSNGjEBXV5dLly4phOvr67N69Wq0tbVpbGzk559/JjU19YmVXehY9PX10dTUVBrxqK6upnv37i2mMTQ0VGr81tTUoKWlhb6+PtXV1ZSWlnL06FGKiorQ1dWld+/eTJ8+nZ07d0rf2UL7Inb3ajsqbaSkpKSQkJDAnj17mm6mrc2cOXMICwsjMDAQIyMjDhw4IC2EjY+PZ8WKFWzcuJHTp09jaGjIP/7xD3Jzc/H390cul2NnZ8fzzz/P+vXr0dLSavXe69atY82aNQphL7zwAosXL1ZllduV06dP8/3330vv33jjDUD5D0wulz/0H93Bgwc5c+YMb7/9Njo6Ok+srELHdv78eXbt2iW9X7JkCdDysyYIT9r9z5WGhsZDPWt9+/YlKCiI7777jqqqKoVrtbW1fPrpp+jq6tKzZ08mT55McXGx0jQdQb209Kw9Tj55eXnk5eVJ4Tk5OcyePRtfX1/i4uIevaCC0AmotJESGhqKTCZT2DlKLpejo6NDSUmJtOja1dUVV1dXlixZwurVq3F3dyc8PJxFixZhYGBAWFgYX3/9NXl5eTg4OPDNN99gYmLywCHRVatW8eabbyqEqdvOLH379lXYgUsmkwFNIyrm5uZSeHl5udLoSkt+/fVXIiIiWLlyZas9R4J68vb2ltaSwP9b9Hn/s1VZWfnQPdyC8HuqqqpoaGhQeqaMjY1bXMh8rz59+jBr1iy2bt3a4nRXuVxOUVER0PTD0dbWltGjR4tGipq6e/cujY2NGBoaKoQbGBi0OlWwurq6xfgNDQ1KMz3ulZ+fj5mZ2eMXWhA6OJWtSZHJZGzZsoUNGzZw8eJF6XXp0iWcnJzYtm1bi+mcnZ0xNDRU6tXS0dGhW7duaGlpsWPHDiZNmvTAHRb09PQwNTVVeKnbVC8DAwPs7OykV5cuXTAzM1PY/lAmk5GSkoKbm9sD8/r111/55ZdfWLFiBS4uLqouutDB6OvrY2NjI73s7OwwMTFRWCsgk8lIS0t76LUCgvB7GhoayMrKomfPngrhPXv2fODap759+zJ79mx+/PFHrl279tD309YW5x+rq8bGRgoKCpQ66Lp166YwEnKv3NxcunXrphDWvXt3CgoKaGxsbPVe1tbWD1wjJTxdmpqa7fLVGansEzciIoKSkhIWL16s1CMwa9YsQkNDKSwspLq6mokTJ+Lk5ERpaSmfffYZ9fX1BAUFAXD9+nUSEhIYPHgwJSUlfPLJJyQmJipMYxIejoaGBkFBQUREREgNl4iICPT09PD395fiffvtt5ibm/Pss88CTVO8fv75Z5YuXYq1tbU0T1ZPT0/a1ODu3bvk5+dLeRQUFJCRkYGRkRFWVlZtWEuhPdDQ0GDEiBEcOXIEa2trbGxsOHLkCLq6ugrn52zfvh1TU1MmTZoENDVkmr/wGxoaKCsrIysrC11dXWxsbICmaTj3rqEqLi4mKysLQ0NDaXRWUB+xsbHMmTOHzMxMMjIyGDx4MObm5tJuS+PHj8fMzIzw8HCgqYEyZ84c9u/fT3p6OsbGxkDTs3f37l0ARo8eTWZmJkVFRWhpadGrVy/69+/Pzz///HQqKbQLly5dYuzYseTn55OXl4eXlxcmJibS5j/+/v4YGRlJxygkJSXh6+vL0KFDuXr1KnZ2dnh6ehIVFSXlOWDAAPLy8igrK0NHR4fevXtjZWXFiRMnnkodBaE9UVkjJTQ0lMDAwBaHLGfOnMnatWsJCQkhMTGRhQsXkpeXh4WFBX5+fkRGRkq7AjU0NLBhwwZSUlLQ0dFh9OjRnDp1SvTGPqKJEydSX1/PDz/8QFVVFa6urqxYsULhjJSioiKFebZHjx5FJpMpnX8ydepUabvo27dv8+GHH0rXduzYAcCwYcOk9QmCehkzZgz19fXs3r2bmpoaHB0dWbp0qcJufSUlJQrPWnl5ucJBrseOHePYsWO4urryyiuvAE276Xz55ZdSnH379gEwcOBAnnvuOVVXS2hnLl26hKGhIYGBgZiampKbm0tYWJh0RoqpqanC9NbBgwejpaXF9OnTmT59uhR+7tw5du7cCYCuri7Tp0/HzMyM+vp68vPz2bFjh9LiekG93LhxAz09PQYMGICRkRFFRUVERERIUwsNDQ2lRi80beBw4MABhg0bhq+vL1VVVcTFxSlMGdTT02PUqFEYGhpKHTB79+5V6PQTBHWlIVejlaynTp162kUQ1ITYwUpoK8ePH3/aRRDUhOgcFNrKyy+//LSL0Krx48c/7SK0qKVzoTq6zjmJTRAEQRAEQRCEDks0UgRBEARBEARBaFfEViWCIAiCIAiC8BDEYY5tR4ykCIIgCIIgCILQrqjVSMrvHe4lCE/KvVvkCoIq3X+mlCCoSvMWzYIgCG1BrRopgiAIgiAIgvCoOuvBie2R+JcWBEEQBEEQBKFdEY0UQRAEQRAEQRDalcee7nXq1CmGDx9OUFCQ0kEyu3fvZv369Vy7do3GxkYcHR0ZP368dKJ0Tk4OK1as4Pz586SmpvLaa6/x6aefKt1j9+7d/OMf/yAtLQ1XV1fef/99hZOChYcnl8v59ddfOXnyJDU1NTg5OTF79mwcHBxaTZOTk8OBAwe4c+cOxcXFzJgxg9GjRyvEiY2NJS4ujuLiYgDs7e0ZP3483t7eKq2P0L716dMHd3d3dHV1KSws5MyZM9JJ4K1xdHTEz88PExMTKioquHDhAhkZGdJ1Ozs7vL29sbKywtDQkKNHj3Lnzh0V10Roz4YPH05gYCBmZmbk5OTw008/kZaW1mLcPn36MHz4cLp164a2tjY5OTkcPHiQq1evSnE0NTUZN24cgwcPxtzcnLy8PPbt20dycnJbVUlop/r06aNw4vyxY8fIyspqNX63bt0YOXIkVlZWVFZWcu7cOS5fvixd19TUZNCgQXh5eWFsbExJSQmxsbHcvn27DWojPAqxu1fbeeyRlLCwMJYvX05cXJzCD4kjR44wd+5cZs2aRUJCAufPn+f999+nrq5OilNbW4uNjQ2rV6+mT58+LeZ/+vRp5syZw4IFC7h06RILFixg9uzZnDlz5nGLrpaOHDlCTEwMzz77LG+99RampqZ88cUXD1wQWVdXh7W1NVOmTMHU1LTFOObm5kyZMoWVK1eycuVK3N3d+fbbb8nJyVFVVYR2zsfHBy8vL86cOcOBAweoqakhKCgIbe3W+0ZsbGwYOXIkaWlp7N+/n7S0NEaOHIm1tbUUR1tbm5KSEvEZIADQr18/Zs2axeHDh1m3bh03btzglVdewcLCosX4bm5uXLt2jS+//JIPP/yQ1NRUli1bRrdu3aQ4kydPJiAggF27dvHvf/+buLg4XnzxRYU4gvpxd3dn1KhRnDlzhq1bt5KVlcX06dMxMTFpMb6pqSnTp08nKyuLrVu3kpCQwOjRo+nZs6cUZ9iwYfTu3ZuYmBi+//57Ll26xJQpU7CxsWmraglCu/VYjZSqqip27tzJSy+9xKRJk/juu++kaxEREQQEBLBy5Uo8PDxwd3dn2rRpfP7551IcZ2dnNm7cyMKFCzEzM2vxHp9++ilBQUGsWrWKXr16sWrVKsaOHdviiIvwYHK5nGPHjhEcHEzfvn3p0qULISEh1NfXc+7cuVbTOTk5MW3aNPr379/qD0xfX1+8vb2xtbXF1taWyZMno6enJ3qD1JinpydXrlwhIyOD0tJS4uLi0NbWpkePHg9Mk52dTWJiIuXl5SQmJpKTk4OXl5cUJysrS2l0RVBfY8eO5fTp05w6dYq8vDx2795NSUkJw4cPbzH+7t27OXLkCBkZGRQUFLB//34KCgrw9fWV4gwaNIjDhw+TlJREUVERsbGxXL16lbFjx7ZVtYR2qH///iQmJpKYmEhxcTHHjh2joqKi1U7WPn36UF5ezrFjxyguLpbS9u/fX4rj6enJmTNnuHXrFmVlZVy+fJnbt28zYMCAtqqWILRbj9VICQ8Px8PDAw8PD0JCQti8eTNyuRxomu6TlJREYmLiYxXw9OnTBAcHK4SNGzeOU6dOPVa+6qioqIjy8nJ69eolheno6ODm5satW7ee2H0aGxs5f/48dXV1ODs7P7F8hY7D2NgYQ0NDsrOzpbDGxkZyc3Mf2ENoY2OjNPqWnZ0tehWFFmlpadG9e3eFqVoAV69efWBj+F4aGhro6elRXV0thWlra1NfX68Qr76+HldX18cvtNAhaWpqYmdnR3p6ukJ4eno6Xbp0aTGNg4NDi/Ht7OykHaK0tLRoaGhQiCOTyVrNU3j6NDU12+WrM3qsNSmhoaGEhIQAMH78eCorK4mOjiYwMJDly5cTGxuLr68vTk5O+Pv7ExwczPz589HT03voe+Tm5mJnZ6cQZmdnR25u7gPT1dbWUltbqxBWV1eHrq7uQ9+7sykvLwdQmrJlYmIirSV5HNnZ2WzYsAGZTIaenh5Llix54FoXofMyMDAAoKamRiH87t27GBkZPTDd/Wlqamqk/AThXsbGxmhpaUmfbc0qKipanZp6v7Fjx6Krq8v58+elsOZRkxs3blBYWIiHhwe9e/cWc9HVmIGBAZqamkrnElVXV2NoaNhiGiMjI4XGLzTNQNHS0sLAwICqqirS09Pp168fmZmZlJaW4ujoiKurq3jWBIHHaKSkpKSQkJDAnj17mjLS1mbOnDmEhYURGBiIkZERBw4cIC0tjZiYGOLj41mxYgUbN27k9OnTrf5Rt+T+P1a5XP67f8Dr1q1jzZo1CmEhISEsWLDgoe/b0Z09e5YdO3ZI75ctW9ZivIf593wYtra2vPPOO9TU1HDx4kW2bt3Ka6+9JhoqasDFxYUhQ4ZI76Ojo59iaQQBaVT/Qfr378/EiRP5+uuvFQ77/emnn5g3bx7vvvsucrmcwsJCTp8+rfCMC8LDuP85bP6ubQ6PiYkhKCiI559/HoDS0lKSkpLEpjOCwGM0UkJDQ5HJZHTt2lUKk8vl6OjoUFJSIi1adHV1xdXVlSVLlrB69Wrc3d0JDw9n0aJFD3Ufe3t7pVGT/Px8pdGV+61atYo333xTIezEiRMPdc/OwtfXV2G6lUwmA5pGVO5dA1RZWdnqwr8/QltbW5qW4+joSHp6OsePH2fu3LmPnbfQvt25c4fCwkLpvZaWFqA8MqKvr680UnKvlkZNWhpdEQRo+uxqaGhocXS4oqLigWn79etHSEgImzZtIiUlRSnfb775Bm1tbYyMjCgrK2Pq1KkUFRU98ToIHUNNTQ2NjY1KI8GGhoZKoyXNqqqqWozf0NAgbVZTU1PD/v37pdGVyspKhg8fTllZmWoqIjw2McrVdh5pEptMJmPLli1s2LCBixcvSq9Lly7h5OTEtm3bWkzn7OyMoaGh0nDpgwwZMoSoqCiFsMjISIYOHfrAdHp6epiamiq81G2ql76+PjY2NtLL3t4eU1NThS9kmUzGjRs3cHFxUUkZ7p/XLXROMpmMiooK6VVaWkp1dbXCKJqmpib29vYUFBS0mk9BQYHSyJuDg8MD0wjqq6GhgTt37iisswPo1asXN2/ebDVd//79WbBgAZs3byYpKanVeDKZjLKyMjQ1NfHz81PYOlZQL42NjeTl5eHo6KgQ7uTkpLD27l45OTk4OTkpxc/Ly6OxsVEhvKGhgcrKSjQ1NenZs2erW2gLgjp5pJGUiIgISkpKWLx4sdKuXLNmzSI0NJTCwkKqq6uZOHEiTk5OlJaW8tlnn1FfX09QUJAU/+LFi0BTz1VBQQEXL15EV1dX2s3n9ddfZ8SIEXz44YdMnTqVffv2ceTIEeLi4h6xyupLQ0ODUaNGERkZKTVcIiMj0dHRUdhJZMuWLdKWwtD0Rd08mtX8pZ2ZmYmenp40crJ//368vLywsLCgtrZWOvvm5ZdfbvuKCu3C1atX6d27NxUVFZSXl+Pr64tMJlP48RgQEEB1dTW//fablGb8+PH4+PiQkZGBo6MjXbp04ddff5XSaGtrK4z8mZiYYGFhQV1d3R/qABE6h+joaP70pz+RkZHBzZs3CQgIwNLSUvqOmDJlCubm5mzZsgVoaqD86U9/YteuXdy+fVsahamrq5N6t52dnTEzMyMzMxNzc3OeeeYZNDQ0lDrMBPVy/vx5JkyYQF5eHjk5Ofj6+mJiYsKlS5eAps8zY2Nj6cy4S5cu0bdvX0aOHMmVK1dwcHDAx8eHgwcPSnna29tjbGxMQUEBxsbG0pTCB+24KQjq4pEaKaGhodLBWfebOXMma9euJSQkhMTERBYuXEheXh4WFhb4+fkRGRmJh4eHFN/Pz0/67/Pnz7N9+3acnJykrWuHDh3Kjh07+Pvf/84//vEPXF1dCQ8PZ/DgwY9SdLUXGBhIfX09O3fupLq6GmdnZ1555RX09fWlOCUlJQrDmWVlZXz44YfS++joaKKjo3Fzc+P1118Hmhaq/vDDD5SXl6Ovr0+XLl14+eWXlXo4BfWRmJiIlpYWgwcPRk9Pj4KCAqKioqRph9C0sPTeOdsFBQWcOHECPz8/+vbtS0VFBcePH1eYSmZlZcX/x96dh0Vd9Y//fzIgy8ywyqos4oIKCIIKgiKKoqYmlWllZWVattzVXWlpt0tl2m136ae7u6ystDJTK01RNPcNExfcxQUxF3aURTbZfn/44/1tnMFMHbZ5Pa5rros5c86Zc+DNe+bsgwcPVp736NEDgDNnzrBr1656qJloTA4cOIBGo+Gee+7Bzs6OjIwMPv30U2UzEHt7e50zU3r37o25uTkPP/ywzlTU33//ne+++w643hC+9957cXZ2pry8nGPHjrFo0SKZdmjiTp06hY2NDT179lQOc1yxYoUytVCj0eh0oBQWFrJixQqio6MJDg6muLiYLVu2cPr0aSWOhYUFvXr1wt7enoqKCtLS0khISNDb+Ec0Hs11J63GyKzmVlYXNhO//fZbQxdBmAg5xFLUl6SkpIYugjARsgWzqC83riluTO6///6GLoJBK1asaOgi3HXSHBRCCCGEEEI0Knd0TooQQgghhBCmQnb3qj8ykiKEEEIIIYRoVKSRIoQQQgghhGhUTGq614gRIxq6CMJEbN68uaGLIEzEpUuXGroIwkScP3++oYsgRIOT6V71R0ZShBBCCCGEEI2KNFKEEEIIIYQQjYpJTfcSQgghhBDidsl0r/pj9EZKdnY2U6dOJSEhQTl5Pjg4mBkzZhAREQFAcnIys2bNYvv27RQUFODt7U10dDQTJ07Ez88PgJdffpmdO3dy9OhROnfuzMGDB41d9GZp3LhxvPzyy7i7u3PixAnefPNNEhMTDcadP38+jz76qF74iRMnCAsLA+DRRx9l/vz5enFqT2oWpqumpoZffvmFLVu2UFxcTLt27XjyySfx9PSsM82WLVvYsWMHFy9eBMDX15dRo0bVeYjcqlWrWLZsGYMGDeLxxx83Sj1E4xcaGkrPnj3RarXk5OSwceNGLly4YDCuRqNhwIABuLu74+TkxN69e9m4caNOnI4dOxIZGYmjoyMqlYorV66wZ88ejh49Wh/VEY1Yr169iImJwc7OjszMTFasWMHZs2cNxg0KCqJXr160bt0aCwsLMjMzWbduHSkpKTrxbGxsGDJkCEFBQajVai5fvszKlSs5ceJEfVRJiEbL6I2UESNGUFFRwaJFi2jbti1ZWVls2rSJy5cvAxAfH8+IESMYNGgQixcvpl27dmRnZ7N8+XKmTp3K0qVLgetfeMaOHcuePXs4fPiwsYvdLD3wwAP8+9//5tVXX+X333/nqaee4ueff6ZHjx7Kl8I/mzRpEtOmTVOeW1hYsHv3br1TTQsKCggNDdUJkwaKiI+PJyEhgWeffRZ3d3d+/fVX3n//fT744ANsbGwMpjlx4gQRERH4+fnRokUL4uPj+fe//83777+Pk5OTTtzU1FS2bNmCt7d3fVRHNFKdO3cmNjaWdevWcfHiRUJCQnjooYf44osvKCws1ItvYWFBSUkJu3btUjpbblRaWsquXbvIy8ujqqqKDh06MGzYMIqLi0lLSzN2lUQjFRISwv33389PP/1EWloakZGRPPvss8yePZv8/Hy9+O3atePkyZPEx8dTWlpKeHg448aNY+7cucqGF+bm5jz33HMUFRWxcOFC8vPzcXBwkM9QITByIyU/P5+dO3eydetWoqOjAfDx8VE+GEpKSnjqqacYMmSIzhdfX19fwsPDdf7pP/74YwBycnKkkXKbXnzxRb799lsWLVoEwJtvvsmAAQMYN24cM2bM0ItfWFio8yE/bNgwHBwc+P7773Xi1dTUkJ2dbdSyi6alpqaGdevWERcXR48ePQB49tlneeGFF0hMTKR///4G0z3//PM6z8eNG0dSUhLHjh0jKipKCS8rK+Ozzz7j6aefZuXKlUarh2j8wsLCOHToEIcOHQJg48aNtG3bltDQULZu3aoXv6CggA0bNgAQHBxsMM8bd7Hau3cvXbp0wcvLSxopJqxv377s2bOH33//HYAVK1bQqVMnevfuTXx8vF78Gzv01qxZQ2BgIIGBgUojJTw8HLVazbx586iurgbgypUrRq6JuBMy3av+GHXhvFarRavVsnLlSoO9AuvXryc3N5dJkyYZTO/g4GDM4pmUFi1aEBISorc17qZNmwgPD7+lPMaMGcOWLVv0plFotVqOHTtGSkoKy5cvJygo6K6VWzRNOTk5FBQU0KVLFyWsRYsWdOrUidOnT99yPuXl5VRVVaHVanXCFy5cSNeuXQkMDLxrZRZNj0qlwsPDQ2+6TVpa2k2nFf5dbdq0wcnJSbbgNWHm5uZ4enrqTdVKSUmhTZs2t5SHmZkZ1tbWFBcXK2GBgYGcO3eOBx98kHfffZc33niDAQMGyBdhITDySIqFhQULFy5k/PjxzJ8/n9DQUKKjo3n44YcJCgpSvqx06tTprr93eXm5XsOopqbGZP/xW7ZsiYWFhd6IR05ODm5ubn+Z3s3NjdjYWMaOHasTfurUKSZMmMCxY8ews7PjueeeY8OGDURGRpKamnpX6yCajtpRUHt7e51we3t7cnNzbzmfpUuX4ujoSEBAgBK2e/duzp07xzvvvHNXyiqaLrVajUql0vnSB1BcXIxGo7mjvK2srPjHP/6Bubm5MjJ47ty5O8pTNF0ajQZzc3OKiop0wouKirCzs7ulPPr27YulpaXOmtqWLVvSoUMH9u/fz+eff46LiwsPPvgg5ubmrF+//m5WQYgmp17WpAwdOpQdO3awe/du1q1bx5w5c1iwYAE1NTVGe9/Zs2fz9ttv64S1aNECKysro71nU2Dod34rf4fHHnuMgoICvSHtvXv3snfvXuX57t272blzJ88++2ydI2Si+dm1axdff/218vz11183GO/v/M/Hx8eze/du3nrrLSwtLQHIy8vju+++44033lDChDCG8vJyvvrqK1q0aEGbNm0YMGAA+fn5MpoidJiZmd3SfS00NJTBgwfz1VdfcfXqVZ30V69eZenSpdTU1HDx4kXs7e3p16+fNFIaKVPt7G4I9bIFsbW1NbGxscTGxjJt2jTGjRvH9OnTmTdvHnB9uLR2p6+7ZfLkybz66qs6Ya1atbqr79GU5OXlUVlZqTdq4uLickvrSR577DGWLFlCRUXFTePV1NRw4MCBOndjEs1TaGiozt+8srISuD7/39HRUQkvLCzUG10xZM2aNaxatYo333xTZ2F8WloahYWFTJ06VQmrrq7m5MmTbNiwgYULF6JSyfFPpqKkpITq6mq9URONRqM3unI7atcGZGdn4+zsTGRkpDRSTFRxcTFVVVXY2trqhGu1Wr3RlRuFhITw8MMPs3DhQk6dOqXzWmFhIVVVVToNnaysLOzt7TE3N6eqquruVUKIG3z66ad88MEHZGRkEBAQwLx583TWf9Zl165dREdHExgYaNTddhvk09zf35/i4mIGDhyIs7Mzc+bMMRjP0G4Zt8rKygo7Ozudhym3fisqKkhOTqZfv3464TExMezZs+emaXv37k379u359ttvb+m9unTpQlZW1m2XVTQ9NjY2uLu7K4/WrVtjb2+vs2VrZWUlKSkpdOjQ4aZ5xcfHs3LlSiZNmkTbtm11XgsICGD27Nm89957ysPX15fIyEjee+89aaCYmOrqajIyMvD19dUJ9/X1Nbhj4Z0yNze/63mKpqGqqoqLFy/SsWNHnfCOHTvedBpgaGgojzzyCN999x3Hjx/Xez0tLQ0XFxed7ycuLi4UFBRIA0UY1dKlS3nllVd46623SE5OJioqinvuuecvO2IKCgoYM2ZMnRvg3E1GHUnJy8tj5MiRjB07lqCgIGxtbdm3bx9z5swhLi4OjUbDggULGDlyJMOHD+ell16iffv25ObmsmzZMs6fP8+PP/4IwJkzZ7h69SqZmZmUlpYqLTd/f3+Z9nGLPvnkE7788kuSk5NJSkpSzqz46quvAJgxYwYeHh48++yzOunGjBnD3r17De7Z/uabb7J3715SU1OxtbXlueeeIygoiNdee61e6iQaJzMzMwYPHsyqVatwc3PD3d2dVatWYWlpSWRkpBJv/vz5ODo68tBDDwHXGyg//fQTzz//PM7OzkpHhbW1NdbW1tjY2ODl5aXzXlZWVmi1Wr1wYRqSkpIYPnw4GRkZXLp0iZCQEOzs7Dhw4ABwfR2Ara0tq1evVtK4uroCYGlpiVqtxtXVlerqamW9VEREBBkZGeTn52Nubk67du3o0qUL69atq/8KikZj69atPProo1y4cIFz584RERGBo6Mju3btAq7vgGlvb8/ixYuB6w2URx99lF9++YVz584pozAVFRWUlZUB13uko6KiuP/++9mxYwcuLi7Exsayffv2hqmk+EvNpcP7o48+4umnn2bcuHEAzJs3j/Xr1/PZZ58xe/bsOtM9++yzjB49GnNzc6PvrmnURopWqyU8PJy5c+eSmppKRUUFXl5ejB8/nilTpgAQFxdHYmIis2fPZvTo0RQWFuLl5UVMTAwzZ85U8ho3bhzbtm1TnoeEhADXeyFudWcNU/fLL7/g5OTEG2+8gbu7O8ePH+fBBx9Udutyd3fX+6JnZ2dHXFwcb7zxhsE87e3t+fjjj3Fzc6OwsJBDhw4xePBg9u/fb/T6iMZt2LBhXLt2jYULF1JSUkK7du144403dM5Iyc3N1bnhb9y4kcrKSmXL8Vr3338/I0aMqLeyi6bjxIkT2NjY0Lt3b+Uwx6VLlyrbp2u1Wr2FzbUfygAeHh4EBgaSn5/Pp59+ClxvvAwePBhbW1sqKyvJy8tj1apVcrieiUtOTkatVjNo0CDs7OzIyMjg888/V6YF2tnZ6UxvjYyMxNzcnJEjRzJy5EglPCkpiR9++AG4PmNk/vz53HfffUyaNImCggK2bdvGpk2b6rdyoskztGGUlZWVwbXY165dY//+/bz55ps64QMHDqzzgG+Ab775htTUVL7//nud7+jGYlZjzNXrjcyNc0mFMJYbt3oWwlhqz/wQwtjkPCxRX2rXLDdGtSP/jU3nzp31NoyaPn26wXPw0tPTad26Nbt27dKZ3TBr1iwWLVrEyZMn9dKcPn2a3r17s2PHDvz8/JgxYwYrV6406pqUelk4L4QQQgghRFPXWKd7Gdow6q92tL2xLnUd1VFVVcXo0aN5++238fPzu/PC3iJppAghhBBCCNGE1TW1yxBnZ2fMzc3JzMzUCc/OzjZ4dl5RURH79u0jOTmZF198Ebi+cUlNTQ0WFhb89ttvxMTE3HklbiBb4QghhBBCCGEiLC0t6datm9504drDuG9kZ2fHkSNHOHjwoPKYMGECHTt25ODBg4SHhxulnDKSIoQQQgghxC1orNO9/q5XX32Vxx9/nO7duxMREcEXX3zB+fPnmTBhAnB9+tilS5f49ttvUalUBAYG6qR3dXXF2tpaL/xuMqlGyp9PeRXCmFJSUhq6CMJEyH1N1BcHB4eGLoIQ4i556KGHyMvL45133iEjI4PAwEDWrl2Lj48PABkZGQ1+eK1J7e7VXFq/ovG71YMvhbhTsi2uqC9yJpmoL4Z2pGosHnnkkYYugkFLlixp6CLcdSY1kiKEEEIIIcTtUqlkOXd9kd+0EEIIIYQQolEx+khKdnY2U6dOJSEhgaysLBwdHQkODmbGjBlEREQA109xnTVrFtu3b6egoABvb2+io6OZOHEifn5+HDp0iPfff5+dO3eSm5tLmzZtmDBhAi+//LKxi9/sPPfcc0ycOBEPDw+OHTvGK6+8ws6dO+uMP3r0aCZNmkSHDh0oKChg3bp1vP7661y+fBm4fnLzmDFjlIVT+/fvZ8qUKezdu7de6iMat6CgIDp06IClpSW5ubkkJSVRUFBw0zTe3t4EBwdja2tLUVERBw8e5MKFC8rrgYGBeHl5YW9vT1VVFTk5ORw4cEA5YVyYnvDwcKKiorC1tSU7O5s1a9Zw7tw5g3EDAgIICwujVatWmJubk52dzaZNmzh9+rRevNjYWJycnLh8+TK//fYbx48fr4faiMasR48eREZGKtfaunXr6py3r9VqGTRoEB4eHrRs2ZI9e/awbt06vXjW1tbExMTQuXNnbGxsuHLlCr/99pveNSmEqTH6SMqIESM4dOgQixYt4tSpU6xatYq+ffsqX3Lj4+Pp2bMn5eXlLF68mBMnTvDdd99hb2/P1KlTgetffF1cXPj+++85duwYb731FpMnT+aTTz4xdvGblVGjRjFv3jzee+89QkJC2LFjBwkJCXh5eRmM36tXL7799lu++uorAgICGDlyJD169GDBggVKnL59+7JkyRL69etHREQE58+f57fffqNVq1b1VS3RSAUEBNC5c2eSkpJISEigrKyMAQMGYGFRd9+Is7MzUVFRpKWlER8fT1paGn369MHZ2VmJ4+rqysmTJ0lISGDjxo2YmZnRv3//m+Yrmq8uXbowdOhQtm7dyieffMK5c+d44oknsLe3Nxi/TZs2nDlzhoULF/K///2Ps2fP8vjjj+Ph4aHE8fLy4uGHHyY5OZn//ve/JCcn88gjj+Dp6Vlf1RKNUEBAAIMHD2bHjh3Mnz+f8+fP89hjj9V5rVlYWFBcXMyOHTv0zqOoZW5uzuOPP46DgwPLli3jv//9L6tXr5ZOl0bMzMysUT6aI6MunM/Pz8fR0ZGtW7cSHR2t93pJSQk+Pj707t2bFStWGExf124iL7zwAidOnGDz5s23XJ7m+ke8Vb///jsHDhzg+eefV8KOHz/OypUrmTJlil781157jeeee4727dsrYS+++CKTJk3C29vb4HuoVCquXLnCiy++yHfffXf3K9FEyML56x0UKSkpHDt2DLh+bYwcOZIDBw7U2UMYFRVFixYtdP6vY2JiuHbtWp0jflZWVowaNYr169eTnZ199yvSyJn6wvnnnnuO9PR0fv31VyXslVde4fjx4/z222+3lMfLL7/MkSNHlOvu4YcfxsrKikWLFilxnnzySUpLS1m6dOndrUATYuoL58eNG0dGRgZr1qxRwl544QVSUlLYtGnTTdM++eSTZGZm6o2kdO/encjISD755BOqq6uNUu6mqDEvnH/ssccauggGff/99w1dhLvOqCMpWq0WrVbLypUrKS8v13t9/fr15ObmMmnSJIPpb7bdYUFBAU5OTnerqM1eixYt6Natm96H9m+//Wbw4B6AxMREPD09ueeee4DrPdgPPvigzg36Rmq1mhYtWigjZcI0abVa1Go16enpSlh1dTVZWVm4uLjUmc7FxYWMjAydsIyMjJumqf3idO3atTsstWhqzM3NadWqlV6j98yZM8o2mn/FzMwMKysrSkpKlDBvb2/OnDmjE+/06dN1ds6I5q/2WktNTdUJT01NrXM2wq3o2LEjFy9eZOjQobz++us8//zzREVFmXynqhBg5EaKhYUFCxcuZNGiRTg4ONCrVy+mTJnC4cOHAZQPlk6dOv2tfHfv3s2yZct49tln73qZmytnZ2csLCzIysrSCc/KysLd3d1gmt27d/Poo4+ydOlSrl27RlZWFvn5+fzjH/+o833ef/99Ll26xMaNG+9q+UXTYmNjA0BZWZlOeFlZmfKaIdbW1pSWluqElZaW3jRNt27dlGtTmBa1Wo25ubneWTFFRUVotdpbyqN3795YWlpy5MgRJUyr1erlefXqVWxtbe+80KJJUqvVqFQqiouLdcKLi4tv+VozxNHREX9/f8zMzFi8eDHbt28nIiKCPn363GmRhZE09LQuU5ruVS9rUtLT01m1ahWDBg1i69athIaGsnDhQm5nptmxY8eIi4tj2rRpxMbG1hmvvLycwsJCnYdA73duZmZW59+hc+fOfPzxx7zzzjt069aNQYMG4evry/z58w3GnzhxIo888ggPPPCAwZEz0Xz5+vry8MMPK4+b3TD/7v/9zfIKCwvD0dHxpps/iObP0H3tVgQFBdG/f3+WLFmi9+XT0HVqQseKiTrc7WvAzMyM4uJiVq9eTUZGBkePHmXHjh107979rr6PEE1Rvaw0tba2JjY2ltjYWKZNm8a4ceOYPn068+bNA66fzl2709fNHD9+nJiYGMaPH8+//vWvm8adPXs2b7/99t0ofrOQm5tLZWWl3qiJq6ur3uhKrcmTJ7Nr1y7+85//AHDkyBGKi4vZuXMn//rXv3QWAr722mtMmTKFAQMG6PRICtNw4cIFcnNzlee1+8jfODJibW2tN7ryZ4ZGWgyNrsD1XXY8PT357bffdKbqCNNRUlJCVVWV3giHoZGQG3Xp0oUHHniAJUuW6E3hMTRqcit5iuarpKSE6upqvVETjUZzR9dFUVER1dXVOo2fnJwcbG1tMTc3p6qq6rbzFqKpa5BzUvz9/SkuLmbgwIE4OzszZ84cg/H+PH3j2LFj9OvXjyeeeIL33nvvL99j8uTJFBQU6DxMWUVFBfv379cbfYqNjSUxMdFgGrVarbeQr/aG+eeeytdff52pU6cyePBg9u/ff5dLLpqCyspKioqKlEdBQQElJSU6OyapVCrc3NzIycmpM5+cnBydNAAeHh56aXr06IG3tzcbNmyQL44mrKqqivT0dJ3NPQDat2/PH3/8UWe6oKAgHnzwQZYuXcrJkyf1Xj9//rzBPOvaalY0f7XXWrt27XTC27Vrp7NF+t914cIFnJycdD5TW7ZsSVFRkTRQGqmGntYl073ukry8PGJiYvj+++85fPgwaWlpLF++nDlz5hAXF4dGo2HBggWsWbOG4cOHs3HjRs6dO8e+ffuYNGkSEyZMAP5fAyU2NpZXX32VzMxMMjMzb/plx8rKCjs7O52Hqfvoo48YN24cTz31FJ06deKjjz7C29tbmb41a9Ysnd1sVq9ezQMPPMCECRPw9fUlMjKSjz/+mD179iiLmydOnMjMmTMZO3Ys586dw83NDTc3NzQaTYPUUTQeKSkpdOnSBS8vLxwcHIiMjKSyspK0tDQlTmRkJCEhITppPDw8CAgIwM7OjoCAADw8PEhJSVHihIWF0bZtW3bs2EFFRQXW1tZYW1tjbm5er/UTjcPOnTvp3r073bp1w8XFhSFDhmBvb09SUhIAAwcO5MEHH1TiBwUFMXLkSNauXcuFCxeUDV6srKyUOImJibRv354+ffrg4uJCnz59aN++Pbt27ar3+onGY/fu3YSGhhISEoKzszODBg3C3t6effv2AdC/f3/uv/9+nTTu7u64u7tjaWmJWq3G3d1dZyOQvXv3YmNjw+DBg2nZsiUdOnQgKipKuX6FMGVGne6l1WoJDw9n7ty5pKamUlFRgZeXF+PHj1e2vI2LiyMxMZHZs2czevRoCgsL8fLyIiYmhpkzZwKwfPlycnJyWLx4MYsXL1by9/HxqfPALqFv2bJltGzZkmnTpuHh4cHRo0cZMmSI0jvo4eGhs3vNokWLsLW15cUXX+TDDz8kPz+fzZs388Ybbyhxnn/+eaysrPj555913mvGjBky3c7EHTt2DHNzc8LCwrCysiI3N5dNmzZRWVmpxLmxMZuTk8OOHTvo2rUrwcHBXL16le3bt+tMJevYsSMAgwYN0km7a9cuzp49a8QaicboyJEjqNVqYmJisLW1JSsri0WLFikj8ba2tjo7RYaFhWFubk5cXBxxcXFK+P79+5X72Pnz51m6dCmxsbEMGDCAy5cv8+OPP3Lx4sX6rJpoZI4dO4ZarSY6OhqtVkt2djaLFy9WZmrY2trqnZlS29kK0KpVK4KCgsjPz1emuxcWFvLdd98xePBgnnvuOQoLC9mzZ4+ssxMCI5+T0tg01+Ew0fjIOSmivpj6OSmi/pj6OSmi/jTmc1KeeOKJhi6CQX+eCdNcNMiaFCGEEEIIIYSoizRShBBCCCGEEI1KvWxBLIQQQgghRFMnSwfqj4ykCCGEEEIIIRoVkxpJ8fX1begiCBORl5fX0EUQJuLGQweFMJY/78wnhBDGZlKNFCGEEEIIIW6XTPeqPzLdSwghhBBCCNGoSCNFCCGEEEII0ajIdC8hhBBCCCFugUz3qj9Gb6RkZ2czdepUEhISyMrKwtHRkeDgYGbMmEFERAQAycnJzJo1i+3bt1NQUIC3tzfR0dFMnDgRPz8/8vLyePTRRzl8+DB5eXm4uroSFxfHrFmzsLOzM3YVmpXHHnuM8ePH4+rqyqlTp5g5cyZ79+6tM76lpSX/+Mc/uO+++3B2diYzM5NPP/2U5cuX68UdNmwYH3/8Mb/99hsTJkwwZjVEExAUFES3bt3QaDTk5eWxbds20tPT64zfunVr+vTpQ8uWLSkuLmbfvn0cOXJEJ05ISAhdunTBzs6O0tJSTp8+za5du6iqqjJ2dUQjFhoaSnh4OFqtlpycHDZu3MjFixcNxtVoNPTv3x93d3ecnJzYt28fGzdu1Inj5+dHZGQkjo6OqFQqrly5QlJSEkePHq2P6ohGrFu3bkRERGBra0tOTg7r16/nwoULBuNqtVpiY2Px8PDAycmJpKQkfvvtN504QUFBxMXF6aWdNWuW3NeEyTN6I2XEiBFUVFSwaNEi2rZtS1ZWFps2beLy5csAxMfHM2LECAYNGsTixYtp164d2dnZLF++nKlTp7J06VJUKhVxcXHMnDkTFxcXzpw5wwsvvMDly5f54YcfjF2FZmPo0KH861//Ytq0aezfv5/Ro0fz9ddfM2jQoDq/PP73v//F2dmZN998k3PnztGyZUssLPQvm1atWjF58mSSkpKMXQ3RBPj5+REdHc3mzZtJT08nKCiI++67j++++46ioiK9+HZ2dtx3330cPXqUdevW0apVK2JiYigtLeXMmTMAdOzYkV69erFhwwYyMjJwcHBg4MCBAGzfvr1e6ycaj86dOzNgwADWr1/PxYsXCQkJ4aGHHuLLL7+ksLBQL76FhQUlJSUkJibSo0cPg3mWlZWRmJhIXl4eVVVVtG/fnqFDh1JcXExaWpqxqyQaKX9/fwYNGsTatWu5ePEioaGhjB49ms8++8zgtWZubk5xcTE7d+4kPDy8znzLysr49NNPdcKkgSKEkRsp+fn57Ny5k61btxIdHQ2Aj48PYWFhAJSUlPDUU08xZMgQVqxYoaTz9fUlPDyc/Px8ABwdHXnuueeU1318fHj++ef54IMPjFn8Zufpp59m+fLlLFu2DIB3332XqKgoHn30UYO/yz59+hAeHk50dDQFBQUAXLp0SS+eSqVi3rx5/N///R/du3eX0S1BaGgox44d49ixYwBs27YNHx8fgoKC2LVrl178oKAgioqK2LZtGwBXrlzBzc2Nbt26KY0UDw8P0tPTOXnyJACFhYWcPHkSd3f3eqqVaIzCwsI4dOgQhw4dAmDjxo34+voSEhKiXE9/VlBQoIycBAUFGczz/PnzOs/37dtHly5d8PLykkaKCevZsyfJyckcPHgQgN9++4127drRvXt3Nm/erBe/oKBAGTkJDg6+ad7FxcV3vbzCOGS6V/0x6sJ5rVaLVqtl5cqVlJeX672+fv16cnNzmTRpksH0Dg4OBsPT09P55ZdflIaP+GstWrQgMDCQHTt26ITv2LGD0NBQg2kGDBjAkSNHePbZZ0lMTGTTpk1MnjwZKysrnXgvvfQSeXl5SuNHmDaVSoWrqyt//PGHTvgff/yBh4eHwTTu7u4G47u6uqJSXb9Npaen4+bmhpubG3B99MXX11e+NJowlUqFu7u73jWQlpaGp6fnXXsfHx8fnJyc9BovwnSoVCo8PDw4e/asTnhqauodX2u106pffvllHnroIel4EeL/Z9SRFAsLCxYuXMj48eOZP38+oaGhREdH8/DDDxMUFMTp06cB6NSp0y3l98gjj/Drr79SWlrKvffey4IFC+qMW15ertcwqqmpMdkWsKOjIxYWFuTm5uqE5+Xl4eLiYjCNl5cX3bt3p7y8nAkTJuDk5MQ777yDg4MDb7zxBnB9fu7IkSMZNmyY0esgmgYbGxtUKhUlJSU64SUlJajVaoNpNBqNXiOlpKQEc3NzrK2tKSkp4dSpU9jY2DBq1Cjg+lSKQ4cOsW/fPuNURDR6arUalUql1wtdXFyMRqO5o7ytrKx48cUXMTc3p6amhvXr13Pu3Lk7ylM0XTe71rRa7W3nm5eXx6pVq8jOzsbS0pLw8HCefPJJvvjiC2VavBCmyuhbEI8YMYL09HRWrVrFoEGD2Lp1K6GhoSxcuJCampq/ldfcuXM5cOAAK1euJDU1lVdffbXOuLNnz8be3l7nUTt9zJQZ+p3X9XdQqVTU1NTwz3/+k8OHD7N161bee+89RowYgZWVFRqNho8++ogpU6Zw5coVYxddmDhPT0/CwsLYvHkzP/zwA6tXr8bX11eZPipErbvRGVVeXs7XX3/NwoUL2bZtG/3798fb2/sulE40ZTd+XpqZmf3t7zJ/dunSJY4cOUJWVhYXLlzgp59+Ii8vr871UqLhmZmZNcpHc1QvWxBbW1sTGxtLbGws06ZNY9y4cUyfPp158+YBkJKSouz0dTPu7u64u7vTqVMnWrZsSVRUFFOnTjU4hWTy5Ml6jZi/mhPanF25coXKykq9UZOWLVvqja7Uys7OJjMzU2eh85kzZ5RhbxsbG7y8vPjyyy+V12un5pw6dYoBAwbI9AgTVFpaSnV1td6oiVqt1htdqVVcXGwwflVVFWVlZQBERERw4sQJZZ1LXl4eLVq0oH///rJhg4kqKSmhurpab9RErVbflTn+tZ0v2dnZtGzZkoiICLmnmajaa+3GUZO7da39WXp6Ok5OTnc1TyGaogY5zNHf35/i4mIGDhyIs7Mzc+bMMRjvZiMftT0Xhta6wPWhejs7O51Hc21p3oqKigqOHj1K7969dcJ79+7NgQMHDKbZv38/bm5uOl8efX19qaqqIiMjg9TUVAYPHsywYcOUx8aNG/n9998ZNmwYGRkZRq2TaJyqq6vJzs7W63X29vau85rIzMzUi+/j40N2djbV1dUABneVq53Cacr/26asurqazMxMfH19dcJ9fX3r3IL4dpmZmWFubn5X8xRNR3V1NRkZGbRt21YnvG3btnf9WnN3d+fq1at3NU8hmiKjjqTk5eUxcuRIxo4dS1BQELa2tuzbt485c+YQFxeHRqNhwYIFjBw5kuHDh/PSSy/Rvn17cnNzWbZsGefPn+fHH39k7dq1ZGVl0aNHD7RaLcePH2fSpEn06tWLNm3aGLMKzcpXX33Fhx9+yJEjRzhw4ACPPPIIrVq1YvHixQBMnDgRNzc3Xn/9dQBWrVrFiy++yJw5c5g3bx6Ojo5MnjyZ5cuXK43DU6dO6bxH7TaMN4YL03LgwAEGDRpEVlYWGRkZdOnSBVtbWw4fPgxAr1690Gg0ys43hw8fJjg4mD59+nDkyBE8PDwICAggISFByTMtLY2QkBBlhM/BwYGIiAjOnj17R9MtRNOWlJTEvffeS0ZGBpcuXaJr167Y2dmRnJwMQHR0NLa2tsTHxytpXF1dgesLltVqNa6urlRVVZGXlwdcH7XLyMggPz8flUpFu3btCAwMZP369fVfQdFo/P7779x3332kp6dz6dIlQkJCsLe3Z//+/QDExMRga2vLr7/+qqSp3eij9lpzc3OjqqpKmcHQp08fLl68yOXLl7GysiIsLAw3Nzede59oXKRTrP4YtZGi1WoJDw9n7ty5pKamUlFRgZeXF+PHj2fKlCkAxMXFkZiYyOzZsxk9ejSFhYV4eXkRExPDzJkzgesLcb/88kv++c9/Ul5ejpeXFw888ABvvvmmMYvf7KxZswZHR0f+8Y9/4OLiwqlTpxg7dqxyRoqLiwutWrVS4peUlDBmzBhmzJjBr7/+ypUrV1i7di0ffvhhQ1VBNBGnTp3C2tqanj17olarycvL49dff1WmDmo0Gp2tqgsLC1m5ciXR0dEEBQVRXFzM1q1ble2HAfbs2UNNTQ2RkZFotVpKSkpIS0sjMTGx3usnGo8TJ05gY2NDr169lMMcly1bpnSYaLVavW3Rn376aeXn2gZxfn4+n332GXB9N8RBgwZha2tLZWUleXl5rF69mhMnTtRfxUSjc/z4cWxsbOjTp49yrS1ZskTZot/QtfbMM88oP7dq1YouXbqQn5/Pf//7X+D6rI+hQ4ei1WopLy8nMzOTRYsW3fTgWyFMhVmNCXVB3jhMK4SxvPTSSw1dBGEiSktLG7oIwkRUVlY2dBGEiZg6dWpDF6FO48ePb+giGPTn9cHNRb0snBdCCCGEEKKpk+le9adBFs4LIYQQQgghRF2kkSKEEEIIIYRoVGS6lxBCCCGEELdApnvVH5NqpGRmZjZ0EYSJaNmyZUMXQZiI06dPN3QRhImQAwaFEPVJpnsJIYQQQgghGhWTGkkRQgghhBDidsl0r/ojIylCCCGEEEKIRsXoIynZ2dlMnTqVhIQEsrKycHR0JDg4mBkzZhAREQFAcnIys2bNYvv27RQUFODt7U10dDQTJ07Ez89PJ7+8vDyCg4O5dOkSV65cwcHBwdhVaFaeeeYZXnnlFdzd3Tlx4gQTJ06s88Tuzz//nMcff1wv/Pjx43Tv3l15bm9vz4wZM4iLi8PBwYFz584xefJk1q9fb7R6iKYhKCiIDh06YGlpSW5uLklJScrpzHXx9vYmODgYW1tbioqKOHjwIBcuXFBeDwwMxMvLC3t7e6qqqsjJyeHAgQPKCePC9PTo0YPevXsrp4AnJCTwxx9/GIyr1WoZPHgwrVq1wsnJiT179pCQkKAXLyIigh49emBvb09JSQnHjh1j48aNcqChiQsKCqJbt25oNBry8vLYtm3bTU+Hb926NX369KFly5YUFxezb98+jhw5ohMnJCSELl26YGdnR2lpKadPn2bXrl1UVVUZuzpCNGpGH0kZMWIEhw4dYtGiRZw6dYpVq1bRt29fLl++DEB8fDw9e/akvLycxYsXc+LECb777jvs7e0Nnjj69NNPExQUZOxiN0sjRoxgzpw5zJkzh4iICHbt2sXKlSvx9PQ0GH/ixIn4+voqjw4dOpCXl8eKFSuUOC1atCA+Ph4fHx9Gjx5NcHAwL7zwwk1v2sI0BAQE0LlzZ5KSkkhISKCsrIwBAwZgYVF334izszNRUVGkpaURHx9PWloaffr0wdnZWYnj6urKyZMnSUhIYOPGjZiZmdG/f/+b5iuar8DAQO655x62bdvGZ599xh9//MFjjz2Gvb29wfgWFhYUFxezbds2srKyDMYJCgpiwIABbNmyhf/+97+sXLmSwMBABgwYYMyqiEbOz8+P6OhokpKSWLx4Menp6dx3333Y2toajG9nZ8d9991Heno6ixcvJikpib59+9K+fXslTseOHenVqxd79uzh22+/ZcOGDfj5+dGrV6/6qpb4m8zMzBrlozky6qd6fn4+O3fuZOvWrURHRwPg4+NDWFgYACUlJTz11FMMGTJE54uvr68v4eHh5Ofn6+T32WefkZ+fz7Rp0wz2fImbe+mll1i0aBELFy4EYNKkSQwYMIDx48czffp0vfiFhYU6vdP33nsvjo6OfPvtt0rYE088gaOjI/369VN6GP/c6y1MV6dOnTh69KhyPezatYuRI0fi6+tb545UnTt3JiMjg6NHjwJw9OhRXF1d6dSpEzt37gRg8+bNOmkSExMZNWoUTk5OZGdnG7FGojGKjIzkwIEDHDhwAICEhATat29Pjx492Lhxo178/Px85fMjNDTUYJ5eXl5cuHBB6fHOz8/nyJEjdXboCNMQGhrKsWPHOHbsGADbtm3Dx8eHoKAgdu3apRc/KCiIoqIitm3bBsCVK1dwc3OjW7dunDlzBgAPDw/S09M5efIkcP1z9+TJk7i7u9dTrYRovIw6kqLVatFqtaxcuZLy8nK919evX09ubi6TJk0ymP7PU7mOHz/OO++8w7fffotKJUtp/q4WLVoQEhLCpk2bdMI3bdpEz549bymPJ554gs2bN+s0QoYOHcqePXuYN28eaWlp7N27l4kTJ8rfyMRptVrUarXOiFp1dTVZWVm4uLjUmc7FxYWMjAydsIyMjJumsbS0BODatWt3WGrR1Jibm+Ph4UFqaqpO+JkzZ/D29r7tfP/44w88PDxo3bo1AI6Ojvj5+XHq1Kk7Kq9oulQqFa6urnrTCGuvFUPc3d0Nxnd1dVU+I9PT03Fzc8PNzQ24Pvri6+tLWlqaEWohRNNi1JEUCwsLFi5cyPjx45k/fz6hoaFER0fz8MMPExQUpPSmdurU6ab5lJeX88gjj/DBBx/g7e3N2bNn//K9y8vL9RpGNTU1zXZI7K84OztjYWGhN70hOztbuTnejLu7OwMHDuTJJ5/UCW/Tpg3R0dEsXbqUBx54gHbt2jF37lwsLCyYPXv23ayCaEJsbGwAKCsr0wkvKytDo9HUmc7a2prS0lKdsNLSUiU/Q7p160ZWVpbeyKto/tRqNebm5ly9elUnvLi4GK1We9v5Hj16FI1Gw9NPP42ZmRnm5uYkJSWxY8eOOy2yaKJsbGxQqVSUlJTohJeUlKBWqw2m0Wg0eo2UkpISzM3Nsba2pqSkhFOnTmFjY8OoUaOA6w3vQ4cOsW/fPuNURNwxU/0e2RCMPol7xIgRDB06lB07drB7927WrVvHnDlzWLBgATU1NbeUx+TJk+ncuTOPPfbYLb/v7Nmzefvtt3XCLCwsaNGixd8qf3Nz4+/czMzslv4Ojz32GPn5+axevVonXKVSkZOTwwsvvEB1dTXJycl4eHjwz3/+UxopJqR2imatG6dk/dmt/t/XutkHQlhYGI6OjrJJg9Dzd6+zP2vTpg19+vQhPj6eixcv0rJlS+655x6io6OVqTtC3A2enp6EhYWxefNmMjMzcXBwIDo6muLiYpKSkhq6eEI0qHqZk2NtbU1sbCzTpk0jMTGRJ598kunTpys7d6WkpNw0/ebNm1m+fDkWFhZYWFjQv39/4ProgKG1FHC9YVNQUKDzMOWFtbm5uVRWVurNc3VxcbmlefxjxoxhyZIlVFRU6IRnZmZy+vRpqqurlbDa+bSm3iA0JRcuXGDNmjXKo3YU09raWieetbW13ujKn5WVlemNmhgaXYHrOzp5enqyYcMGvd5NYRpKSkqoqqrSGzXRaDQUFxffdr79+/fn0KFDHDhwgOzsbE6cOMHGjRuJioqSXlQTVVpaSnV1td6oiVqtrvP+U1xcbDB+VVWVch+MiIjgxIkTHDt2jLy8PFJTU0lMTKRHjx7GqYgQTUiDLBzw9/enuLiYgQMH4uzszJw5cwzGq52+8fPPP3Po0CEOHjzIwYMHWbBgAQA7duzghRdeMJjWysoKOzs7nYcpf7hUVFSQnJxMTEyMTnhMTAy///77TdNGRUXRvn17Fi1apPfa77//Trt27XR+t+3btycjI0OvQSOar8rKSoqKipRHQUEBJSUlOnO1VSoVbm5u5OTk1JlPTk6O3vxuDw8PvTQ9evTA29ubDRs26E31EaajqqqKjIwM2rVrpxPerl07zp8/f9v5tmjRQm8kxpSnC4vra+qys7P11jp5e3vrraOrlZmZqRffx8eH7OxspWPPUOdp7bUm11vj1NC7eJnS7l5GbaTk5eURExPD999/z+HDh0lLS2P58uXMmTOHuLg4NBoNCxYsYM2aNQwfPpyNGzdy7tw59u3bx6RJk5gwYQJw/QMnMDBQefj6+gLXdwJydXU1ZhWalY8//pgnn3ySMWPG0LFjR/7973/j5eWlNPrefvttvvzyS710TzzxBElJSRw/flzvtS+++AInJyf+85//0L59ewYPHszEiRP5/PPPjV4f0bilpKTQpUsXvLy8cHBwIDIyksrKSp0FoZGRkYSEhOik8fDwICAgADs7OwICAvDw8NAZbQ0LC6Nt27bs2LGDiooKrK2tsba2xtzcvF7rJxqHxMREQkNDCQkJwdnZmcGDB2Nvb8/evXsBGDBgAA888IBOGnd3d9zd3bG0tEStVuPu7q6zOcPJkyfp0aMHgYGBODg40K5dO2JiYkhJSbmjaWSiaTtw4ACBgYH4+/vj6OhInz59sLW15fDhwwD06tWLgQMHKvEPHz6MnZ0dffr0wdHREX9/fwICAti/f78SJy0tjS5duuDn54ednR3e3t5ERERw9uxZudaEyTPq/CetVkt4eDhz584lNTWViooKvLy8GD9+PFOmTAEgLi6OxMREZs+ezejRoyksLMTLy4uYmBhmzpxpzOKZnJ9//pmWLVsyefJk3N3dOX78OPfff7+yW5e7uzteXl46aWr3eZ84caLBPC9dusS9997LnDlzSEpKIj09nU8//ZQPP/zQ6PURjduxY8cwNzcnLCwMKysrcnNz2bRpk85heDcuos/JyWHHjh107dqV4OBgrl69yvbt28nNzVXidOzYEYBBgwbppN21a9ctbaohmpejR49iY2ND3759sbW1JTs7m++//145NNTW1lbvzJTnn39e+bl169YEBwdz5coV5s6dC1zfWrampob+/ftjZ2dHcXExJ0+e1NsdUZiWU6dOYW1tTc+ePVGr1eTl5fHrr79SVFQEXL+f2dnZKfELCwtZuXIl0dHRBAUFUVxczNatW5XthwH27NlDTU0NkZGRaLVaSkpKSEtLq/OQZSFMiVmNCTXV69qBQ4i7TUaSRH2p68wZIe42Jyenhi6CMBGvvPJKQxehTi+++GJDF8GgTz75pKGLcNfJYRZCCCGEEEKIRkUaKUIIIYQQQohGxXT35BVCCCGEEOJvaK47aTVGMpIihBBCCCGEaFRMaiTF0dGxoYsgTMSVK1caugjCRMiGIKK+qFTSrymEqD8m1UgRQgghhBDidsl0r/oj3SJCCCGEEEKIRkUaKUIIIYQQQohGxejTvbKzs5k6dSoJCQlkZWXh6OhIcHAwM2bMICIiAoDk5GRmzZrF9u3bKSgowNvbm+joaCZOnIifnx9geHjts88+Y8KECcauQrPyxBNPMGHCBFxdXTl16hTTp08nKSmpzviWlpb885//5IEHHsDFxYWMjAw+/vhjli5dqsQZN24cY8aMoVWrVly5coU1a9Ywe/ZsysvL66NKopHq0qULoaGhaDQaLl++zPbt20lPT68zfuvWrYmKisLJyYni4mL279/P0aNHldcfeOABPD099dKlpaWxevVqo9RBNA0hISGEh4ej1WrJzc1l48aNXLx40WBcjUZDTEwM7u7uODk5sW/fPr2T5IODgwkMDMTFxQWAzMxMtm3bRkZGhtHrIho3ua8Jme5Vf4zeSBkxYgQVFRUsWrSItm3bkpWVxaZNm7h8+TIA8fHxjBgxgkGDBrF48WLatWtHdnY2y5cvZ+rUqTpfhr/55hsGDx6sPLe3tzd28ZuV4cOHM2PGDKZMmcLevXt5/PHH+f777+nbt2+dN9n58+fj4uLC66+/TlpaGs7OzlhY/L/L5v7772fy5Mm89tpr7Nu3j7Zt2zJ37lwAZsyYUR/VEo1Qhw4d6NOnD1u3biU9PZ3AwECGDx/O999/z9WrV/Xi29nZMXz4cI4ePcr69etp1aoVffv2pbS0lNTUVADWrFmDubm5ksba2prRo0dz5syZequXaHw6derEgAEDWL9+PZcuXaJr166MGjWKBQsWUFhYqBffwsKC0tJSdu/eTY8ePQzm6e3tzfHjx7l06RKVlZX07NmThx56iAULFhi8foVpkPuaEPXLqI2U/Px8du7cydatW4mOjgbAx8eHsLAwAEpKSnjqqacYMmQIK1asUNL5+voSHh5Ofn6+Tn4ODg64u7sbs8jN2vjx4/nxxx9ZsmQJANOnTyc6OpoxY8bw/vvv68Xv27cvPXv2JDIyUvlb3Ng72a1bN/bt28fKlSuV13/99Ve6du1qzKqIRi4kJIRjx45x7NgxAHbs2IGPjw9BQUEkJibqxQ8MDKSoqIgdO3YA13dHc3V1JTQ0VPkwv3Fkzs/Pj8rKSk6fPm3k2ojGLCwsjEOHDnH48GEANm3ahK+vLyEhIWzbtk0vfkFBARs3bgQgKCjIYJ439mAnJCTQsWNH2rRpo9MLLkyL3NeEqF9GXZOi1WrRarWsXLnS4NSf9evXk5uby6RJkwymd3Bw0Hn+4osv4uzsTI8ePZg/fz7V1dXGKHaz1KJFC4KCgvQ+tLdt20b37t0Nphk4cCCHDx/mueeeY9++fezYsYOpU6dibW2txElKSqJLly5Ko8Tb25uYmBi96RPCdKhUKlxdXTl//rxO+Pnz5/Hw8DCYxsPDQy/+H3/8gaura53bnvr7+3Pq1CkqKyvvTsFFk6NSqXB3d+fcuXM64efOnaN169Z37X1atGiBSqWitLT0ruUpmha5r4laZmZmjfLRHBl1JMXCwoKFCxcyfvx45s+fT2hoKNHR0Tz88MMEBQUpPQWdOnX6y7zeffdd+vfvj42NDZs2beK1114jNzeXf/3rXwbjl5eX6zWMampqmu0f8q84OTlhYWFBbm6uTnhubi6urq4G03h7e9OjRw/Ky8sZN24cTk5OzJo1CwcHB1577TUAVq1aRcuWLVmxYgVmZma0aNGCRYsW8b///c/odRKNk42NDSqVipKSEp3wkpKSOs/0UKvVBuObm5tjbW2t95qbmxvOzs7SGDZxarUalUpFcXGxTnhxcTEajeauvU90dDRXr17VawwJ0yH3NSHqn9F39xoxYgTp6emsWrWKQYMGsXXrVkJDQ1m4cCE1NTW3nM+//vUvIiIi6Nq1K6+99hrvvPMOH3zwQZ3xZ8+ejb29vc6jqKjoblSpSbvxd25mZlbn30GlUlFTU8OLL77IwYMH2bx5M2+//TajRo1SRlMiIiJ46aWXmDJlCoMHD+bpp59mwIABvPLKK8auimiC/s7//M06FPz9/cnNzSUrK+tuFEs0cX/nuvq7wsPD8ff355dffqGqqspo7yOaLrmvCWEc9bIFsbW1NbGxsUybNo3ExESefPJJpk+fruzclZKS8rfz7NmzJ4WFhXX+M0+ePJmCggKdh62t7R3Voym7fPkylZWVym41tVq2bElOTo7BNNnZ2WRmZuo07k6fPo1KpVKGtydOnMjPP//MkiVLSElJYd26dbz//vu8+OKLJjtqZepKS0uprq7W611Uq9V1Tpcx1BtpY2NDVVUVZWVlOuEWFhb4+fkp88KF6SopKaG6uhqtVqsTrtFo9EZXbkdYWBgREREsXbq0zvukMA1yXxO1GnpalylN92qQc1L8/f0pLi5m4MCBODs7M2fOHIPxblw4/2fJyclYW1vrrVupZWVlhZ2dnc6juf4Rb0VFRQWHDx+mT58+OuF9+vRh3759BtPs3bsXd3d3nZts27ZtqaqqUrbitLGx0VsbVNvbaMq/b1NWXV1NdnY23t7eOuHe3t51buGakZFhMH52drbe9dWhQwfMzc05efLk3S24aHKqq6vJzMykTZs2OuFt2rTh0qVLd5R3WFgYkZGRLFu2jMzMzDvKSzR9cl8Tov4ZtZGSl5dHTEwM33//PYcPHyYtLY3ly5czZ84c4uLi0Gg0LFiwgDVr1jB8+HA2btzIuXPn2LdvH5MmTVLOQFm9ejVffvklR48eJTU1lQULFvDWW2/xzDPPYGVlZcwqNCtffvkljzzyCA899BDt27dnxowZtG7dmu+++w6AN998k//7v/9T4q9YsYIrV64wd+5cOnToQHh4OFOnTuXHH39UeoE2bNjAmDFjGD58OF5eXkRFRTFx4kQ2bNggGxuYsOTkZAICAvD398fR0ZGoqCi0Wi1HjhwBIDIyktjYWCX+0aNHsbW1JSoqCkdHR/z9/QkICODAgQN6efv7+3P27Fm9nkhhmpKSkggODiYoKIiWLVvSv39/7OzsSE5OBq6vJxk2bJhOGldXV1xdXWnRogVqtRpXV1datmypvB4eHk6fPn1ISEigoKAAjUaDRqOhRYsW9Vo30bjIfU2I+mXUhfNarZbw8HDmzp1LamoqFRUVeHl5MX78eKZMmQJAXFwciYmJzJ49m9GjR1NYWIiXlxcxMTHMnDkTuL6zyqeffsqrr75KdXU1bdu25Z133uGFF14wZvGbnVWrVuHo6Mg///lPXF1dOXnyJI8//rjS4+jm5karVq2U+CUlJTz88MPMnDmThIQErly5wurVq3VGvv7v//6PmpoaJk2ahLu7O5cvX2bDhg38+9//rvf6icbj9OnTWFtbExYWhkajIS8vj1WrVilTB9Vqtc70y8LCQlatWkVUVBRBQUFcvXqVbdu2Kdt01nJwcKB169Y6W5YL05aSkoKNjQ29evVCo9GQm5vL8uXLlTNStFotdnZ2OmnGjh2r/Ozh4UFAQAAFBQV89tlnAISGhmJhYcH999+vk27nzp3s3LnTyDUSjZXc1wTILJH6ZFZjzBWHjczd3JJSiJt54403GroIwkTcuEOQEMZS1y5WQtxtL730UkMXoU6vvvpqQxfBoI8++qihi3DXNciaFCGEEEIIIYSoi1GnewkhhBBCCNFcyHSv+iMjKUIIIYQQQohGRRopQgghhBBCiEbFpKZ73em++ULcqg8++KChiyBMRF1nRQlxt8k0FyHk/6A+yUiKEEIIIYQQolGRRooQQgghhBCiUTGp6V5CCCGEEELcLpVK+vfri9EbKdnZ2UydOpWEhASysrJwdHQkODiYGTNmEBERAUBycjKzZs1i+/btFBQU4O3tTXR0NBMnTsTPz0/Ja+HChXz00UecOnUKBwcHHnzwQT755BNjV6FZqamp4ZNPPmHp0qUUFhYSHBzMtGnT6NChQ51pHn/8cZKSkvTCo6Oj+eKLL/TCP//8cz766CPGjBnDW2+9dVfLL5qOrl270qNHD7RaLbm5uWzevLnOdWEajYa+ffvi7u6Oo6Mj+/fvZ8uWLTpxWrZsSe/evXFzc8Pe3p7Nmzezf//++qiKaOQCAgIIDg5GrVZz5coVdu3aRWZmZp3xPTw8iIyMxNHRkZKSEg4ePMjx48cNxm3Xrh2xsbGkpaWxfv16Y1VBNBH+/v4611piYuJfXmsRERE619qJEyeU1319fQkJCcHOzg6VSkVBQQGHDx/m9OnT9VEdIRo1ozdSRowYQUVFBYsWLaJt27ZkZWWxadMmLl++DEB8fDwjRoxg0KBBLF68mHbt2pGdnc3y5cuZOnUqS5cuBa6fpPnhhx/ywQcfEB4eTllZGWfPnjV28ZudL7/8km+++Yb333+fNm3a8Nlnn/HUU0+xbt06tFqtwTT//e9/qaioUJ7n5+cTFxfH4MGD9eIePnyYpUuX0rFjR6PVQTR+HTt2JCYmhg0bNnDp0iWCg4N58MEH+frrrykqKtKLb25uTmlpKb///jvdunUzmGeLFi3Iz8/n5MmT9OvXz9hVEE1Eu3btiIyMZMeOHWRmZuLv78/QoUNZunQpV69e1Ytva2vLkCFDOHHiBJs2bcLd3Z2oqChKS0tJS0vTiavVaomIiCA9Pb2+qiMasdprbefOncq1NmTIEJYtW1bntXbPPfeQkpLC5s2bcXd3p3fv3pSVlSnXWllZGQcOHCA/P5/q6mq8vb3p27cvpaWlXLx4sb6rKESjYtRGSn5+Pjt37mTr1q1ER0cD4OPjQ1hYGAAlJSU89dRTDBkyhBUrVijpfH19CQ8PJz8/H4ArV67wr3/9i9WrV9O/f38lXkBAgDGL3+zU1NTw7bffMmHCBAYOHAjAv//9byIjI4mPj+fhhx82mO7G3YPWrFmDtbW1XiOluLiYiRMnMnPmTD777DOj1EE0Dd27d+fIkSMcOXIEgC1btuDr60vXrl3ZsWOHXvzCwkI2b94MQGBgoME8MzMzlR7LPn36GKnkoqkJCgoiJSWFlJQUABITE/Hy8sLf39/gCLC/vz9Xr14lMTERuP455eLiQnBwsE4jxczMjP79+7Nv3z48PDywtLSsnwqJRqtLly5615qnp+cdXWsZGRk6aY4ePYqfnx/u7u7SSGmkZHev+mPUiXVarRatVsvKlSspLy/Xe339+vXk5uYyadIkg+lrvxxv2LCB6upqLl26ROfOnfH09GTUqFFcuHDBmMVvdi5evEhOTg69e/dWwiwtLenRowfJycm3nM/PP//M0KFDUavVOuHvvPMO0dHRREZG3rUyi6ZHpVLh7u7OuXPndMLPnTtH69atG6ZQollSqVS4uLjofRZcvHgRd3d3g2nc3Nz0vvxduHABFxcXnbnm3bp1o6ysTPlCKkxb7bV247Vz8eJF3NzcDKap61pzdnauc11D69atcXBw0Gu8CGGKjNpIsbCwYOHChSxatAgHBwd69erFlClTOHz4MIAy57JTp043zefs2bNUV1cza9Ys5s2bx08//cTly5eJjY3l2rVrxqxCs5KTkwNcn9v/Z87OzuTm5t5SHocPH+bUqVOMHDlSJ3zNmjUcP36c11577e4UVjRZNjY2qFQqiouLdcKLi4vRaDQNVCrRHFlbW6NSqSgtLdUJLykp0etEqaVWqykpKdEJKy0txdzcHGtrawDc3d3p1KkT27ZtM07BRZNT17VWWlpa57VmY2NjMP6frzW43lk4duxYxo0bx+DBg9m1a5ec6yYE9bQmZejQoezYsYPdu3ezbt065syZw4IFC6ipqbmlPKqrq6moqODjjz9WpiktWbIEd3d3tmzZwqBBg/TSlJeX643eWFlZYWVldeeVaiJWrVrF9OnTleeff/45oD9Ueat/B4CffvoJPz8/goKClLCMjAzee+89vv76a5P6/Yq/x8zM7G9da0Lcrr97rdXeE2tqamjRogUxMTFs27aNsrIyYxVRNCM3u9bqeu3P4deuXeOnn36iRYsWtG7dmoiICAoLC2U0pZGS6V71p162ILa2tiY2NpbY2FimTZvGuHHjmD59OvPmzQMgJSVF2enLEA8PD+D6/M5aLi4uODs7c/78eYNpZs+ezdtvv60TNn36dGbMmHFnlWlCYmJiCA4OVp7Xjjrl5ubi6uqqhOfl5eHs7PyX+ZWWlrJmzRpeeuklnfBjx46Rl5fHAw88oIRVVVWxd+9eFi9ezJEjRzA3N7/T6ogmorS0lOrqar1RE0M92ELcibKyMqqrq7GxsdEJN9SDXcvQKIu1tTVVVVWUl5fj6OiInZ0d99xzj/J67ZeSZ555hh9//JHCwsK7XBPR2N3OtWZolMXGxka51v6s9prKy8vDwcGBkJAQaaQIk9cg56T4+/uzcuVKBg4ciLOzM3PmzNFZOF8rPz9fmSYGcPLkSTw9PQG4fPkyubm5+Pj4GHyPyZMn8+qrr+qEmVovf+2aoFo1NTW4uLiwa9cupcF37do19u7dy+uvv/6X+SUkJHDt2jWGDx+uE96zZ09Wr16tEzZ58mTatm3L+PHjpYFiYqqrq8nMzMTHx0dnG00fHx/OnDnTgCUTzU11dTU5OTl4eXnprIFq3bq13pqoWllZWXqfG15eXuTk5FBdXU1+fr6yq2StsLAwWrRowa5duwzu4iSav9przdPTU+fauvH5nxm61jw9PcnNzaW6urrO9zIzM5PPTSEwciMlLy+PkSNHMnbsWIKCgrC1tWXfvn3MmTOHuLg4NBoNCxYsYOTIkQwfPpyXXnqJ9u3bk5uby7Jlyzh//jw//vgjfn5+xMXF8fLLL/PFF19gZ2fH5MmT6dSpU51bkZra1K5bYWZmxpgxY/j8889p06YNPj4+fP7551hbWzNs2DAl3qRJk3Bzc9NbX/LTTz8xYMAAHB0ddcK1Wq3OeTZwvdfcwcFBL1yYhn379jF06FAyMzNJT08nODgYOzs7Dh06BEBUVBS2trasXbtWSVM7umdpaYlarcbV1ZWqqiry8vKA6wtXa0f8zM3N0Wq1uLq6cu3aNWUnQGF6Dh8+TExMDNnZ2WRlZeHv74+tra1y7klYWBgajUY5d+f48eMEBgYSERHBiRMncHNzo1OnTmzcuBG4Pgp85coVnfeoHYW+MVyYliNHjtCvXz9ycnLIysqic+fOaLXam15rAQEBetfapk2blDy7du1KTk4OhYWFmJub4+XlRYcOHdi5c2eD1FH8NZnuVX+M2kjRarWEh4czd+5cUlNTqaiowMvLi/HjxzNlyhQA4uLiSExMZPbs2YwePZrCwkK8vLyIiYlh5syZSl7ffvst//znPxk6dCgqlYro6GjWrVtHixYtjFmFZmf8+PGUl5fz9ttvU1BQQHBwMF9//bXOiEtGRobeziNpaWns37+fr7/+ur6LLJqgkydPYmNjQ2RkJBqNhtzcXH7++WdlSoNWq8XW1lYnzRNPPKH87O7ujr+/PwUFBcqBoVqtVidOWFgYYWFhnD9/Xq/nW5iO1NRUrK2t6d69O2q1msuXL7N27VplxEOj0ehca0VFRaxdu5bIyEgCAwMpLi5m165demekCHGj1NRUrKys6Natm3KtJSQkKNeaWq3W+SwtKioiISGBiIgIAgICDF5rLVq0ICoqCo1GQ2VlJfn5+WzZsoXU1NR6r58wPZ9++ikffPABGRkZBAQEMG/ePKKiogzG/eWXX/jss884ePAg5eXlBAQEMGPGDIPrwu8WsxpZySrEXffBBx80dBGEibixsSeEsUgPsqgvzz77bEMXoU6TJ09u6CIYNHv27L8Vf+nSpTz++ON8+umn9OrVi88//5wFCxZw/PhxvL299eK/8sortGrVin79+uHg4MA333zDf/7zH/bs2UNISMjdqoaOBlmTIoQQQgghRFPTXBrrH330EU8//TTjxo0DYN68eaxfv57PPvvMYIOndrOrWrNmzeLXX39l9erVRmukGPWcFCGEEEIIIYRxlZeXU1hYqPMwdJA6XF9nt3//fuVYj1oDBw4kMTHxlt6vurqaoqIinJyc7rjsdZFGihBCCCGEEE3Y7Nmzsbe313nUNQUsNzeXqqoq3NzcdMLd3NzIzMy8pff78MMPKS4uZtSoUXdc9rrIdC8hhBBCCCFuQWOd7nU7R28YOtz7Vuq3ZMkSZsyYwa+//qpz7t7dZlKNFNnfXtSXG3dHE8JYardoFsLYXFxcGroIQog6/J2jN5ydnTE3N9cbNcnOztYbXbnR0qVLefrpp1m+fDkDBgy47fLeCvkmJYQQQgghhImwtLSkW7dubNiwQSd8w4YNREZG1pluyZIlPPnkk/zwww8MHTrU2MU0rZEUIYQQQgghbldjne71d7366qs8/vjjdO/enYiICL744gvOnz/PhAkTgOvTxy5dusS3334LXG+gjBkzhv/7v/+jZ8+eyiiMjY0N9vb2RimjNFKEEEIIIYQwIQ899BB5eXm88847ZGRkEBgYyNq1a/Hx8QGuH+x9/vx5Jf7nn39OZWUlL7zwAi+88IIS/sQTT7Bw4UKjlFEaKUIIIYQQQpiY559/nueff97gazc2PLZu3Wr8At3A6I2U7Oxspk6dSkJCAllZWTg6OhIcHMyMGTOIiIgAIDk5mVmzZrF9+3YKCgrw9vYmOjqaiRMn4ufnx8KFC3nqqacM5p+VlWXUnQWam5qaGr744gt++eUXioqKCAwM5I033qBdu3Z1plm1ahVvv/22XnhiYqKySGvYsGFkZGToxRk5ciRvvvnm3auAaDKCg4Pp0aMHGo2GvLw8tmzZwqVLlwzG1Wg0REdH4+bmhqOjIwcOHDB4Q+zQoQO9evXC3t6egoICdu7cyZkzZ4xcE9HYdevWjZ49e6LVasnJyWHDhg1cuHDBYFytVkv//v3x8PDAycmJvXv36s3L/jN/f3/uv/9+Tp48yU8//WSsKogmwt/fn6CgINRqNVeuXGH37t033bLVw8ODnj174ujoSElJCYcOHeLEiRPK623atCEkJAQ7OztUKhUFBQUcOXKE06dP10d1xG1oLtO9mgKjN1JGjBhBRUUFixYtom3btmRlZbFp0yYuX74MQHx8PCNGjGDQoEEsXryYdu3akZ2dzfLly5k6dSpLly7loYceYvDgwTr5Pvnkk5SVlUkD5W9atGgRixcvZsaMGXh7e/PVV1/x/PPP88svv6DRaOpMp9Fo+OWXX3TC/ryLxHfffUdVVZXyPDU1leeff97oOz+Ixqljx47069ePTZs2cenSJYKCgnjggQdYuHAhRUVFevHNzc0pLS1lz549dOvWzWCeHh4eDBs2jF27dnHmzBnat2/PsGHD+PHHH295X3fR/HTu3JnY2FjWrVvHhQsXCA0N5eGHH+bzzz+nsLBQL765uTklJSXs2rWLsLCwm+ZtZ2dH//79daY8CNPVtm1bIiIi2LlzJ1lZWXTu3Jl77rmHZcuWUVxcrBff1taWwYMHk5KSwpYtW3Bzc6N3796UlZWRlpYGXD+ALzk5mfz8fKqqqvDx8SE6OprS0lIuXrxY31UUolExaiMlPz+fnTt3snXrVqKjowHw8fFRPhhKSkp46qmnGDJkCCtWrFDS+fr6Eh4eTn5+PnB9UY6NjY3yek5ODps3b+arr74yZvGbnZqaGn744QfGjh1LTEwMAG+//bbyAT9ixIg605qZmeHs7Fzn646OjjrPFy5ciKenZ51fOEXz1q1bN44cOcKRI0eA68PEbdq0ITg4mJ07d+rFLywsZMuWLQAEBgbWmecff/xBUlISAElJSco1tmbNGiPVRDR24eHhHDx4kIMHDwLXd6dp27YtoaGhBkfjCgoKlJGT4ODgOvM1MzPjvvvuY/v27Xh5eWFtbW2M4osmJCgoiJMnT3Ly5EkAdu/ejaenJ/7+/uzdu1cvfufOnbl69Sq7d+8Grn8ncnFxISgoSGmk3DgD4ejRo/j5+eHu7i6NFGHyjLoFsVarRavVsnLlSsrLy/VeX79+Pbm5uUyaNMlgegcHB4Ph3377LWq1mgcffPBuFrfZu3TpEnl5efTs2VMJq92G7tChQzdNW1paytChQ7nnnnt4+eWXSUlJqTNuRUUFa9euJS4uToZFTZBKpcLNzY0//vhDJ/yPP/6gVatWt52vh4cH586d0wk7d+7cHeUpmjaVSoWHh4fyha/W2bNn8fT0vKO8o6KilOk5QqhUKpydnfUaDhcvXqzzXAk3NzeD8V1cXOr8bGzVqhX29vYGp0+LxsHMzKxRPpojo46kWFhYsHDhQsaPH8/8+fMJDQ0lOjqahx9+mKCgIGXOZadOnf5Wvl9//TWjR4/WGV25UXl5uV7DqKKi4pYPummOag99a9mypU64k5PTTW+Ivr6+zJgxg/bt21NcXMySJUsYO3YsP/74I97e3nrxt2zZwtWrV7n33nvvbgVEk2BjY4NKpaKkpEQnvLi4mDZt2tx2vhqNRi/PkpIS1Gr1becpmja1Wo1KpdI7qLe4uBitVnvb+Xp6ehIcHMyCBQvutIiimbC2tkalUlFaWqoTXlpaWuc9yMbGxmB8lUqFtbW18lqLFi147LHHMDc3p7q6ml27dtW5fk8IU2L0wxxHjBhBeno6q1atYtCgQWzdupXQ0FAWLlxITU3N385v9+7dHD9+nKeffvqm8WbPno29vb3O48MPP7zdajRJa9eupXfv3sqjsrLSYLyampqbtsK7dOnCkCFD8PPzIyQkhPfffx8fHx9+/PFHg/F//fVXIiMj5XRiE3fj/3dz7ekRjY+Zmdltfb7A9dHluLg41q5dq/cFUwhj3NcqKir4+eefWbFiBXv37qVnz554eHjccb5CNHX1sgWxtbU1sbGxxMbGMm3aNMaNG8f06dOZN28eACkpKcpOX39lwYIFdO3a9S/XOkyePJlXX31VJ6yiouK2yt9URUdH06VLF+X5tWvXgOsjKn9uQFy5cgUnJ6dbzlelUuHv729w95yMjAySkpL44IMP7qDkoikrLS2lurpabyMGtVptcHHprSouLtbrsVSr1XqjK8J0lJSUUF1drTdqcifXmqOjIw4ODowaNUoJq/0iOnnyZD777DNlvaQwHWVlZVRXV+vdg6ytreu8B5WWlurN+LCxsaG6upqysjKd8NpNHvLy8nB0dKRr164y5auRkg63+mP0kRRD/P39KS4uZuDAgTg7OzNnzhyD8W78ILh69SrLli37y1EUuL7zlJ2dnc7D1KZ6aTQavLy8lEfbtm1p2bIle/bsUeJUVFSwf//+my4gvVFNTQ2nTp0yuJB+1apVODo60rt377tSB9H0VFdXk5WVpRwIVcvHx4f09PTbzjcjI0MvzzZt2txRnqJpq66uJiMjA19fX51wX1/f2150nJubyxdffMGCBQuUx6lTpzh37hwLFiwwuGOYaP6qq6vJzc2ldevWOuGenp5kZWUZTJOVlaW3NsrT05OcnJy/HOkzNze/swIL0QwYdSQlLy+PkSNHMnbsWIKCgrC1tWXfvn3MmTOHuLg4NBoNCxYsYOTIkQwfPpyXXnqJ9u3bk5uby7Jlyzh//rzOlKKlS5dSWVnJo48+asxiN1tmZmaMHj2ar7/+Gi8vL7y9vfn666+xtrbW2eJ52rRpuLi48I9//AOAL774gsDAQLy9vSkuLubHH3/k5MmTvPHGGzr5V1dXs2rVKoYNG4aFhZwTasr279/PPffcQ1ZWFunp6cr/f+0i5N69e6PValm3bp2SpnZ0r0WLFqjValxcXKiqqlK2Kz9w4AAPPfQQPXr0IDU1lXbt2uHt7V3ntENhGvbs2UNcXBwZGRlcvHiRkJAQ7O3tOXDgAAB9+/bF1taW1atXK2lqFzpbWlqiVqtxc3OjqqqK3NxcqqqqyMnJ0XmP2l7vG8OFaTl8+DD9+vUjNzdX2YJYq9Uq557UngtVu6vciRMnCAgIoGfPnqSkpODm5kbHjh3ZvHmzkmfXrl3JycmhsLAQlUqFt7c3fn5+7NixoyGqKESjYtRvklqtlvDwcObOnUtqaioVFRV4eXkxfvx4pkyZAkBcXByJiYnMnj2b0aNHU1hYiJeXFzExMcycOVMnv6+++ooHHnhAb7tbceueeOIJysvLef/995XDHP/3v//pTM3JzMzUGc4sKirivffeIy8vD61WS8eOHVmwYIHeVrF79uwhMzOTuLi4equPaJxOnjyJtbU1PXv2VA5zrD1AFK6P8tnZ2emkGTNmjPKzu7s7nTt3pqCgQFm8nJ6eTnx8PL1796ZXr17k5+cTHx8vZ6SYuBMnTqBWq5WGb05ODj/++KMy4qHVarG3t9dJM27cOOVnDw8PAgMDyc/P53//+1+9ll00LWfPnsXa2prQ0FDUajWXL18mISFB2bhBrVbrTD0sKipi3bp1REREEBAQQHFxMYmJiTq70VlYWNC7d280Gg2VlZXk5+ezefNmzp49W+/1E7dGpnvVH7Oa211d2ATduAOMEMby+eefN3QRhIm4cW67EMYim6GI+vLMM880dBHqNGPGjIYugkGNtVx3okHWpAghhBBCCCFEXWThgBBCCCGEELdApZL+/foiv2khhBBCCCFEoyKNFCGEEEIIIUSjYlLTvW48XE4IYzl16lRDF0GYCH9//4YugjARsrOmELK7V32SkRQhhBBCCCFEoyKNFCGEEEIIIUSjYlLTvYQQQgghhLhdMt2r/hi9kZKdnc3UqVNJSEggKysLR0dHgoODmTFjBhEREQAkJycza9Ystm/fTkFBAd7e3kRHRzNx4kT8/PwA2Lt3L2+++Sb79+/HzMyMHj16MGfOHLp27WrsKjQrNTU1fPLJJyxbtozCwkKCgoKYNm0aHTp0qDPN448/zt69e/XCo6OjlUMLP//8czZs2KCcyBsSEsJrr71G27ZtjVYX0bhFR0czcOBA7O3tSU9PZ9myZZw5c8Zg3Hbt2vHAAw/g7u6OpaUlly9fZvv27WzatEmJ4+HhwfDhw/H29sbZ2Zlly5bpvC5MV1BQEKGhoWg0GvLy8ti+fTvp6el1xm/dujVRUVG0bNmS4uJi9u/fz5EjR5TXR4wYgaenp166tLQ0Vq1aZZQ6iKbD39+ftm3bYmlpSV5eHsnJyRQWFt40TevWrQkMDESj0VBcXMyRI0f0rlFra2uCgoJwd3fH3Nycq1evsnfvXvLz841YGyEaL6M3UkaMGEFFRQWLFi2ibdu2ZGVlsWnTJi5fvgxAfHw8I0aMYNCgQSxevJh27dqRnZ3N8uXLmTp1KkuXLqWoqIhBgwYRFxfHp59+SmVlJdOnT2fQoEFcvHiRFi1aGLsazcaCBQtYuHAhs2fPpk2bNsyfP5+xY8eSkJCAVqs1mOa///0vFRUVyvP8/Hzuu+8+Bg0apITt3buX0aNH06VLF6qqqpg7dy7jxo0jPj4etVpt9HqJxqV79+6MGjWKH374gdTUVPr06cM//vEPZsyYwZUrV/TiX7t2ja1bt3Lx4kWuXbtG+/btefTRR7l27Ro7duwAwNLSktzcXPbv38+oUaPqu0qikerQoQN9+vRhy5YtpKen06VLF+Li4vj+++8pKirSi29nZ0dcXBxHjx5l/fr1tGrVin79+lFaWqo0ouPj4zE3N1fSWFtb8+ijj3L69Ol6q5donDp27Iifnx979+6lqKiIzp0706dPH9atW0dlZaXBNE5OTvTs2ZNjx45x6dIlWrduTUREBFu2bFG+C7Vo0YKYmBiys7PZsWMH5eXlaLVanc9eIUyNURsp+fn57Ny5k61btxIdHQ2Aj48PYWFhAJSUlPDUU08xZMgQVqxYoaTz9fUlPDxc6T04efIkV65c4Z133sHLywuA6dOnExQUxPnz52nXrp0xq9Fs1NTU8O233zJhwgQGDhwIwPvvv0+vXr2Ij4/n4YcfNpjOwcFB5/natWuxtrZm8ODBStiCBQt04syePZvIyEiOHTtGjx497m5FRKM3YMAAdu3axa5duwBYtmwZ/v7+REdHs3LlSr34Fy5c4MKFC8rzvLw8QkJCaN++vdJI+eOPP/jjjz8AuP/++41fCdEkhIaGcuzYMY4dOwbA9u3b8fHxoUuXLiQmJurF79KlC0VFRWzfvh2AK1eu4ObmRmhoqNJIKS8v10nj5+dHRUWFNFIEHTp04MSJE1y6dAm43kF377334u3tzdmzZw2m8fPzIysri5SUFABSUlJwcXGhQ4cO7NmzB4BOnTpRUlLCvn37lHQlJSVGro24HTLdq/4YdeG8VqtFq9WycuVKvZs+wPr168nNzWXSpEkG09d+Oe7YsSPOzs589dVXXLt2jdLSUr766isCAgLw8fExZhWalYsXL5KTk0OvXr2UMEtLS3r06EFycvIt5/PTTz8xZMiQm46Q1PZg2tvb336BRZNkbm6Ot7c3x48f1wk/fvz4LXcoeHl50bZtW9nKWdyUSqXC1dWV8+fP64T/8ccfeHh4GEzj4eGhNHb/HN/V1bXOk6QDAgI4depUnT3lwjRoNBpsbGzIyspSwqqrq8nJyaFly5Z1pmvZsqVOGoDMzEydNK1ateLKlSv07NmTe++9lwEDBuDr63v3KyFEE2LUkRQLCwsWLlzI+PHjmT9/PqGhoURHR/Pwww8TFBSk9Ep16tTppvnY2tqydetW4uLiePfdd4HrPRPr16/HwsJwFcrLy/UaRpaWllhZWd2FmjVNOTk5AHo305YtW950/vafHT58mNOnT/Pee+/VGaempob333+fbt26KWuKhOnQarWYm5vrzdEuKirCzs7upmnff/99Jf3q1auVkRghDLGxsUGlUun1OJeWltZ5LpZaraa0tFQnrKSkBHNzc6ytrfXycnNzw9nZmY0bN97dwosmx9raGoCysjKd8PLy8pt22llbW+ulKSsrU/KD6w2gdu3acerUKVJSUnByciIkJITq6mq9RrUQpsLoWxCPGDGC9PR0Vq1axaBBg9i6dSuhoaEsXLiQmpqaW8qjtLSUsWPH0qtXL37//Xd27dpFQEAAQ4YM0fuwqTV79mzs7e11HrNnz76bVWv0Vq9eTWhoqPKo7QU0NFR5q8OXP/30Ex06dCAoKKjOOO+++y4nT57kww8/vL2CC5P1wQcfMGvWLBYvXkz//v1lqqC4JYY+S272+XKrnz1wfRQlNzdXrydcNH/e3t7cf//9yuNuTvO5MS8zMzOuXLnC0aNHyc/P5+zZs5w9e1amszdCZmZmjfLRHNXLFsTW1tbExsYSGxvLtGnTGDduHNOnT2fevHnA9fmZtTt9GfLDDz9w7tw5du/erQzH//DDDzg6OvLrr78aXEsxefJkXn31VZ0wS0vLu1epJqBfv346jYlr164BkJubi6urqxKel5d306HqWqWlpaxdu5aXXnqpzjjvvvsumzdv5vvvv8fd3f0OSi+aqqtXr1JVVaU3amJra/uXO+Dk5eUBkJ6ejp2dHcOGDTO4s5wQcP2eVF1drTdqYmNjU+d8/pKSEr1eb7VaTVVVlV5vt4WFBX5+fvz+++93t+CiSUhPT1fuSYCymcKNIyNWVlZ6186f3ThqYihNaWmp3v2xsLDQ4C5zQpiKBjnM0d/fn+LiYgYOHIizszNz5swxGK924XxJSQkqlUqnpVj7vLq62mBaKysr7OzsdB6mNtVLq9Xi4+OjPNq3b4+Li4vOYtJr166xd+9eQkJC/jK/hIQErl27xr333qv3Wk1NDe+88w4bNmxg4cKFcmM1YVVVVZw/f57OnTvrhHfu3JnU1NS/lVdd0zmFgOvrAbKzs/H29tYJ9/b2JiMjw2CajIwMg/Gzs7P1Pk86dOiAubm5suBZmJbKykqKi4uVR2FhIaWlpbi5uSlxzMzMcHFx0WnM3CgvL08nDVyfRvjnNHl5edja2urEsbW1pbi4+C7VRoimx6iNlLy8PGJiYvj+++85fPgwaWlpLF++nDlz5hAXF4dGo2HBggWsWbOG4cOHs3HjRs6dO8e+ffuYNGkSEyZMACA2NpYrV67wwgsvcOLECY4dO8ZTTz2FhYUF/fr1M2YVmhUzMzPGjBmjnGly6tQpJk+ejLW1NcOGDVPivfHGGwanav38888MGDAAR0dHvdfeeecdVq9ezX/+8x80Gg05OTnk5OTctHdJNF8bN26kd+/eREZG4u7uzsiRI3FyclJ2VLrvvvt48sknlfh9+/YlKCgIV1dXXF1diYyMZODAgSQlJSlxzM3N8fT0xNPTEwsLCxwcHPD09MTFxaW+qycakQMHDhAQEIC/vz+Ojo706dMHW1tb5dyT2mup1pEjR7CzsyMqKgpHR0f8/f0JCAjgwIEDenkHBASQmpoq9zGhOH36NJ06daJVq1bY2dkRFhamdMzU6tGjB4GBgTpp3Nzc6NixI7a2tnTs2BE3Nzed3eJOnTpFy5Yt6dSpExqNRtk85O927Ajja+hpXTLd6y7RarWEh4czd+5cUlNTqaiowMvLi/HjxzNlyhQA4uLiSExMZPbs2YwePZrCwkK8vLyIiYlh5syZwPWF9atXr+btt98mIiIClUpFSEgI69atq3MHF2HYuHHjKCsr45133qGgoICgoCC++uornTNS0tPT9S74tLQ09u/fz1dffWUw3yVLlgAwZswYnfBZs2bxwAMP3OVaiMZu3759aDQahg4dqhzm+MknnyhnAtjb2+Pk5KTENzMz47777sPZ2VnZLeeXX35Rth+G67v9TZ06VXk+cOBABg4cyMmTJ/noo4/qr3KiUTl9+jQ2NjaEh4ejVqvJy8vj119/VXYY1Gg0Oj3UhYWF/Prrr/Tp04egoCCKi4vZtm2b3kGjDg4OtG7dWmd7fCFOnjyJubk5oaGhOgfP/nnntxunE+bl5fH7778TGBhIYGAgV69e5ffff1fuh3B9K+zExES6dOmizDY5ePCg3s51QpgSs5q/s4KwiTOhqooGVjsKKISx+fv7N3QRhIlo1apVQxdBmIiRI0c2dBHq1Fg3YZo8eXJDF+GukwnfQgghhBBC3ILmOrWqMWqQhfNCCCGEEEIIURdppAghhBBCCCEaFZnuJYQQQgghxC2Q6V71x6QaKVVVVQ1dBGEiAgICGroIwkSY2vlPouGcO3euoYsghDAhMt1LCCGEEEII0aiY1EiKEEIIIYQQt0ume9UfGUkRQgghhBBCNCpGH0nJzs5m6tSpJCQkkJWVhaOjI8HBwcyYMYOIiAgAkpOTmTVrFtu3b6egoABvb2+io6OZOHEifn5+AGzatImpU6dy5MgRtFotY8aM4b333sPCQgaD/o6amho+/fRTli9fTmFhIUFBQfzrX/+iffv2N01XWFjI//3f/7Fx40YKCwvx9PRk4sSJ9OnTB4DKykr+97//sWbNGnJzc3FxcSEuLo4JEyagUklb2BR16dKF0NBQNBqNcipzenp6nfFbt25NVFQUTk5OFBcXs3//fo4ePaq8/sADD+Dp6amXLi0tjdWrVxulDqJp8Pf3Jzg4GLVarZzcnZmZWWd8Dw8PIiIicHR0pKSkhIMHD3LixAnldV9fX0JCQrCzs0OlUlFQUMDhw4c5ffp0fVRHNGJdu3alR48eaLVacnNz2bx5M5cuXTIYV6PR0LdvX9zd3XF0dGT//v1s2bJFL56fnx+9evXCwcGB/Px8du7cKdeaENRDI2XEiBFUVFSwaNEi2rZtS1ZWFps2beLy5csAxMfHM2LECAYNGsTixYtp164d2dnZLF++nKlTp7J06VIOHz7MkCFDeOutt/j222+5dOkSEyZMoKqqiv/85z/GrkKz8tVXX7Fo0SLee+892rRpw+eff864ceNYs2YNGo3GYJpr164xbtw4WrZsydy5c3F3dycjI0Mn/ldffcWyZcuYNWsW7du35+jRo/zrX//C1taWxx9/vL6qJxqJDh060KdPH7Zu3Up6ejqBgYEMHz6c77//nqtXr+rFt7OzY/jw4Rw9epT169fTqlUr+vbtS2lpKampqQCsWbMGc3NzJY21tTWjR4/mzJkz9VYv0fi0a9eOyMhIdu7cSWZmJv7+/gwZMoRly5YZvNZsbW255557SElJYfPmzbi7u9O7d2/KyspIS0sDoKysjAMHDpCfn091dTXe3t7K9Xjx4sX6rqJoJDp27EhMTAwbNmzg0qVLBAcH8+CDD/L1119TVFSkF9/c3JzS0lJ+//13unXrZjDPVq1ace+99yoNkw4dOnDvvfeyZMkSMjIyjF0lcRtkulf9MWojpbZHYOvWrURHRwPg4+NDWFgYACUlJTz11FMMGTKEFStWKOl8fX0JDw8nPz8fgB9//JGgoCCmTZsGQPv27Zk9ezaPPPII06dPx9bW1pjVaDZqamr47rvveOaZZ4iNjQVg1qxZ9OnThzVr1jBq1CiD6VasWEFhYSGLFy+mRYsWwPUb658dOnSImJgY5e/cunVr1q5dy7Fjx4xYI9FYhYSEcOzYMeXvv2PHDnx8fAgKCiIxMVEvfmBgIEVFRezYsQOAK1eu4OrqSmhoqNJIKS8v10nj5+dHZWWl9DiauC5dupCSkkJKSgoAiYmJeHp64u/vT1JSkl58f39/rl69qlyH+fn5uLi4EBwcrDRSbvxyePToUfz8/HB3d5dGignr3r07R44c4ciRIwBs2bIFX19funbtqty7/qywsJDNmzcD1+9xhnTr1o1z586xZ88eAPbs2YOXlxfdunUjPj7eSDURomkw6jwcrVaLVqtl5cqVel8wANavX09ubi6TJk0ymN7BwQG4/uXE2tpa5zUbGxvKysrYv3//XS93c3Xx4kVyc3Pp1auXEmZpaUn37t1JTk6uM92WLVsIDg5m5syZ9OnTh7i4OL744gudLZ1DQkL4/ffflS0qU1JSSE5OJioqymj1EY2TSqXC1dWV8+fP64SfP38eDw8Pg2k8PDz04v/xxx+4urrWOV3Q39+fU6dOUVlZeXcKLpoclUqFi4uLXsPh4sWLuLm5GUzj5uamF//ChQs4OzvXea21bt0aBwcH6dk2YSqVCnd3d71tmM+dO0fr1q1vO99WrVrp5ZmWlqbXESiEKTLqSIqFhQULFy5k/PjxzJ8/n9DQUKKjo3n44YcJCgpSekA7dep003wGDRrEvHnzWLJkCaNGjSIzM5OZM2cC+j1etcrLy/UaRubm5iZ9pkBubi4ALVu21Alv2bLlTdcKXLx4kT179jBs2DA+++wz/vjjD2bOnEllZSXPP/88AOPGjePq1asMGzYMc3NzqqqqePnllxk6dKjxKiQaJRsbG1QqFSUlJTrhJSUlqNVqg2nUarXB+Obm5lhbW+u95ubmhrOzM5s2bbq7hRdNirW1NSqVitLSUp3w0tLSOq81Gxsbg/FvvNYsLS157LHHUKlU1NTUsHPnzjrXHojmr/a+VlxcrBNeXFxc51TpW6HRaAze++4kT2Fcss62/hj9Nz1ixAjS09NZtWoVgwYNYuvWrYSGhrJw4UJqampuKY+BAwfywQcfMGHCBKysrPDz81O+/P55jvqfzZ49G3t7e53Hv//977tWr6YgPj6e7t27K4/aHucb51PW1NTcdI5ldXU1Tk5OzJgxg4CAAIYMGcIzzzzD0qVLlTgJCQnEx8czZ84cli9fzqxZs/jmm29YuXKlUeommqZb/Z+Hm8/79ff3Jzc3l6ysrLtRLNEM3exaq+u1P4dfu3aNn376iRUrVrB3714iIiLqHAkUpsvMzOxv3dcMudP0QjRX9bI1lrW1NbGxscTGxjJt2jTGjRvH9OnTmTdvHnB9alDtTl91efXVV/nnP/9JRkYGjo6OnDt3jsmTJ+Pr62sw/uTJk3n11Vd1wupq0DRX/fr1o0uXLsrziooKAGX3rVqXL1/WG135MxcXFywsLHR+f+3atSM3N5dr165haWnJhx9+yNNPP82QIUOA6+sF0tPTWbBgAffdd99drplozEpLS6murtbryVar1Xo92LUMjbLY2NhQVVVFWVmZTriFhQV+fn78/vvvd7fgoskpKyujuroaGxsbnXBDoyW1DI2y1F5rN46+FxYWApCXl4eDgwMhISEy5ctE1d7XbhzhMDQK/HcYGolRq9V6IzZCmKIGGbPy9/enuLiYgQMH4uzszJw5cwzGq104X8vMzIxWrVphY2PDkiVL8PLyIjQ01GBaKysr7OzsdB6mNtVLo9Hg4+OjPNq1a4ezs7POwuVr166xb98+QkJC6swnJCSE8+fPU11drYSdO3cOFxcXLC0tges38BuHQM3NzXXSCNNQXV1NdnY23t7eOuHe3t51fsHLyMgwGD87O1vvGurQoQPm5uacPHny7hZcNDnV1dXk5OTobU3t6elZ5yhbVlaWwfi5ubk3vV+ZmZmZXEeX+H+qq6vJzMzEx8dHJ9zHx+eOpgGmp6fTpk0bnbA2bdrcdAq2aFhmZmaN8tEcGbWRkpeXR0xMDN9//z2HDx8mLS2N5cuXM2fOHOLi4tBoNCxYsIA1a9YwfPhwNm7cyLlz59i3bx+TJk1iwoQJSl4ffPABR44c4dixY7z77ru8//77fPzxx/Kh8TeYmZnx+OOP8+WXX7Jx40ZOnz7NW2+9hbW1tc7akcmTJzN37lzl+UMPPUR+fj6zZ8/m3LlzbNu2jS+//JJHHnlEidO3b1+++OILtm3bxqVLl9i4cSOLFi2if//+9VpH0TgkJycTEBCAv78/jo6OREVFodVqlV1xIiMjlR3m4PruSba2tkRFReHo6Ii/vz8BAQEcOHBAL29/f3/Onj2rN8IiTNORI0fo1KkTHTt2xMHBgYiICLRaLcePHwcgLCyMfv36KfGPHz+OVqslIiICBwcHOnbsSKdOnTh06JASp2vXrrRu3RpbW1scHBzo0qULHTp0kJ3kTNy+ffsICgoiMDAQJycn+vXrh52dnXLtREVFKbMJarm6uuLq6oqlpSVqtRpXV1edmQv79++nTZs2hIWF4eTkRFhYGD4+PrIpkBAYebqXVqslPDycuXPnkpqaSkVFBV5eXowfP54pU6YAEBcXR2JiIrNnz2b06NEUFhbi5eVFTEyMsjgerq95eO+99ygvLyc4OJhff/2Ve+65x5jFb5aefvppysvLeffdd5XDHL/88kud4eaMjAydVrmHhwdffvkl//73v7n//vtxc3Pjscce4+mnn1bivPXWW3z88ce8++67XL58GVdXV0aOHMlzzz1Xr/UTjcPp06extrYmLCwMjUZDXl4eq1atUs4SUKvVOluHFxYWsmrVKqKioggKCuLq1ats27ZN2X64loODA61bt9bZslyYttTUVKysrOjWrRtqtZrLly+TkJCgnJGiVqvRarVK/KKiIhISEoiIiCAgIIDi4mJ27dqlbD8M0KJFC6KiotBoNFRWVpKfn8+WLVv0rkdhWk6ePImNjQ2RkZFoNBpyc3P5+eeflWmBWq1W70iEJ554QvnZ3d0df39/CgoK+OKLL4DrIymrV6+md+/e9O7dm/z8fFavXi3TCoUAzGpMaMWWbFUq6sunn37a0EUQJsLUprGKhlP7ZVwIY5s4cWJDF6FOf55p0pj885//bOgi3HWyj5oQQgghhBCiUZFGihBCCCGEEKJRqZctiIUQQgghhGjqmutOWo2RjKQIIYQQQgghGhWTGknJzc1t6CIIE+Ho6NjQRRAm4k7OaBDi73BycmroIgghTIhJNVKEEEIIIYS4XTLdq/7IdC8hhBBCCCFEoyKNFCGEEEIIIUSjckfTvbKzs5k6dSoJCQlkZWXh6OhIcHAwM2bMICIiAoDk5GRmzZrF9u3bKSgowNvbm+joaCZOnIifnx8AL7/8Mjt37uTo0aN07tyZgwcP6r3XkSNHePHFF0lKSsLJyYlnn32WqVOnyrDb31RTU8PChQtZvXo1RUVF+Pv788orr+Dr63tL6Tdt2sQ777xD7969ee+995Twhx56iMzMTL349913X7M8YEjcmqCgIDp06IClpSW5ubkkJSVRUFBw0zTe3t4EBwdja2tLUVERBw8e5MKFC8rrgYGBeHl5YW9vT1VVFTk5ORw4cEAOmjNhISEhhIeHo9Vqyc3NZePGjVy8eNFgXI1GQ0xMDO7u7jg5ObFv3z42bdqkEyc4OJjAwEBcXFwAyMzMZNu2bXIKuMDf35+goCDUajVXrlxh9+7dBj/7anl4eNCzZ08cHR0pKSnh0KFDnDhxQnm9U6dOdOjQQVnvk5OTw969e8nJyTF6XcTtke+d9eeORlJGjBjBoUOHWLRoEadOnWLVqlX07duXy5cvAxAfH0/Pnj0pLy9n8eLFnDhxgu+++w57e3umTp2q5FNTU8PYsWN56KGHDL5PYWEhsbGxtGrVir179/Lf//6X//znP3z00Ud3UnyTtGTJEpYtW8Yrr7zC559/jpOTE6+99holJSV/mTYzM5PPPvuMoKAgvdc+//xzfvnlF+Xx4YcfAtC3b9+7XQXRRAQEBNC5c2eSkpJISEigrKyMAQMGYGFRd9+Is7MzUVFRpKWlER8fT1paGn369MHZ2VmJ4+rqysmTJ0lISGDjxo2YmZnRv3//m+Yrmq9OnToxYMAAEhMT+eabb7hw4QKjRo3Czs7OYHwLCwtKS0vZvXs32dnZBuN4e3tz/PhxfvjhB7799lsKCwt56KGH0Gq1xqyKaOTatm1LREQEycnJ/PLLL2RmZnLPPfeg0WgMxre1tWXw4MFkZmbyyy+/kJycTGRkpE6noIeHB6mpqcTHx7Ny5UqKi4sZMmQIarW6vqolRKN125/q+fn57Ny5k61btxIdHQ2Aj48PYWFhAJSUlPDUU08xZMgQVqxYoaTz9fUlPDyc/Px8Jezjjz8GrvcgHD58WO+9Fi9eTFlZGQsXLsTKyorAwEBOnTrFRx99xKuvviqt2ltUU1PD8uXLefzxx+nTpw8AkydP5v7772fjxo0MHz68zrRVVVXMnDmTp556isOHD3P16lWd1x0cHHSe//DDD7Ru3ZquXbve7WqIJqJTp04cPXpUGQXZtWsXI0eOxNfXl9OnTxtM07lzZzIyMjh69CgAR48exdXVlU6dOrFz504ANm/erJMmMTGRUaNG4eTkVOeXTtF8hYWFcejQIeWzY9OmTfj6+hISEsK2bdv04hcUFLBx40YAgx0uZ5gDUQABAABJREFUAKtXr9Z5npCQQMeOHWnTpo1ybQrTExQUxMmTJzl58iQAu3fvxtPTE39/f/bu3asXv3Pnzly9epXdu3cD1783ubi4EBQURFpaGgBbtmzRSbN9+3Z8fX1p3bp1nfdJIUzFbY+kaLVatFotK1eupLy8XO/19evXk5uby6RJkwymv/FL7c3s3r2b6OhorKyslLBBgwaRnp7OuXPn/m7RTVZGRgaXL1+me/fuSpilpSXBwcF/+cG7aNEi7O3tGTp06F++T0VFBRs2bOCee+6RBqSJ0mq1qNVq0tPTlbDq6mqysrKUKTSGuLi46E2pycjIuGkaS0tLAK5du3aHpRZNjUqlwt3dXe9z4Ny5c7Ru3fquvU+LFi1QqVSUlpbetTxF06JSqXB2dtabRnjx4kXc3NwMpnFzczMY38XFpc7PRgsLC1QqlcHvVaJxMDMza5SP5ui2GykWFhYsXLiQRYsW4eDgQK9evZgyZYrSm1XbA9CpU6c7LmRmZqbeTaD2+c3mggpdtdPwbtzr3tHRUXnNkCNHjrB27VomTpx4S++zY8cOrl69yj333HP7hRVNmo2NDQBlZWU64WVlZcprhlhbW+t9ESwtLb1pmm7dupGVlaUzOitMg1qtRqVSUVxcrBNeXFxc5xSc2xEdHc3Vq1elU8yEWVtbG2yolpaW1jk1y8bGxmB8lUqFtbW1wTRhYWEUFxfL+UdCcIcL50eMGMHQoUPZsWMHu3fvZt26dcyZM4cFCxZQU1Nzt8oI6C9Uqs2/rtZjeXm5Xk9EeXm5zmhMc7dhwwZlbQjA+++/Dxj+Xdb1eywpKWHmzJm8/vrrtzz6tXbtWsLCwnTWEYjmrXYaZ60bp2T92d+9N9yshygsLAxHR0fWr1//t/IUzcvd/rz5s/DwcPz9/fnhhx+oqqoy2vuIpuHGa+1u9mAHBwfTrl074uPj5VoTgrtwmKO1tTWxsbHExsYybdo0xo0bx/Tp05k3bx4AKSkpyk5ft8vd3V1vxKR27nldw6yzZ8/m7bff1gl77bXXeP311++oLE1Jr1696Ny5s/K8oqICgLy8PFq2bKmE5+fn13lC+qVLl8jMzGTKlClKWHV1NQAxMTF89913OtMqMjMz2b9/P+++++5drYto3C5cuEBubq7yXKW6Pkh748iItbW13ujKnxkaaTE0ugLQo0cPPD09+e23325p4wfR/JSUlFBdXa23oF2j0eiNrtyOsLAwIiIi+PHHH2W3JRNXVlZGdXW13qiJtbV1nfcfQ6PANjY2VFdX690Hg4KC6Nq1K2vWrLnpzAbR8Jrr1KrG6K6fk+Lv709xcTEDBw7E2dmZOXPmGIz3d6ZmREREsH37dp0557/99hutWrWiTZs2BtNMnjyZgoICncc//vGPv1OVJk+tVuPp6ak82rRpo2y5WauiooJDhw4RGBhoMA9vb2+++eYbFixYoDx69epFSEgICxYswNXVVSd+QkICDg4O9OzZ06h1E41LZWUlRUVFyqOgoICSkhI8PDyUOCqVCjc3t5t+2cvJydFJA9d3v7kxTY8ePfD29mbDhg16mzgI01FdXU1mZqbe50CbNm3ueLpMWFgYkZGRLFu2TKYVC6qrq8nNzdVb6+Tp6UlWVpbBNFlZWXh6eurFz8nJ0RmRCQoKIjQ0lISEBJ3OHiFM3W2PpOTl5TFy5EjGjh1LUFAQtra27Nu3jzlz5hAXF4dGo2HBggWMHDmS4cOH89JLL9G+fXtyc3NZtmwZ58+f58cffwTgzJkzXL16lczMTEpLS5VzUvz9/bG0tGT06NG8/fbbPPnkk0yZMoXTp08za9Yspk2bVmeL1srKSm9ql6n3tpqZmTFy5EgWL16sNFy+//57rKysGDBggBLvvffew8XFhWeeeQYrKyvatm2rk09tr+WN4dXV1SQkJDB48GDZDlaQkpJCly5dlIZLYGAglZWVyq42AJGRkZSWlpKcnKykGThwIAEBAVy4cAEvLy88PDx0pnOFhYXh6+vLli1bqKioUOZ2V1RUyBQJE5SUlMS9995LZmYmly5domvXrtjZ2SnXVHR0NLa2tsTHxytpajtXWrRogVqtxtXVlaqqKvLy8oDrU7yioqJYvXo1BQUFyvqWa9euKSPSwvQcPnyYfv36kZubS1ZWFp07d0ar1SrnnvTo0QONRsPWrVsBOHHiBAEBAfTs2ZOUlBTc3Nzo2LGjznTY4OBgunfvzubNmykqKlJGXioqKqisrKz3OgrRmNz2N0mtVkt4eDhz584lNTWViooKvLy8GD9+vDI1KC4ujsTERGbPns3o0aMpLCzEy8uLmJgYZs6cqeQ1btw4na0iQ0JCAEhLS6NNmzbY29uzYcMGXnjhBbp3746joyOvvvoqr7766u0W32Q98sgjlJeXM3fuXK5evUrnzp35z3/+ozOEnZ2drUzX+Tv2799PVlYWQ4YMuZtFFk3UsWPHMDc3JywsDCsrK3Jzc9m0aZPOB++Ni5tzcnLYsWMHXbt2JTg4mKtXr7J9+3ad3sWOHTsC13f4+7Ndu3Zx9uxZI9ZINEYpKSnY2NjQq1cvNBoNubm5LF++XDncU6vV6p2ZMnbsWOVnDw8PAgICKCgo4LPPPgMgNDQUCwsL7r//fp10O3fuVLbCFqbn7NmzWFtbExoailqt5vLlyyQkJCijuWq1WmfqYVFREevWrSMiIoKAgACKi4tJTEzU6ajx9/fH3Nyc2NhYnffav38/+/fvr5+Kib9FpnvVH7MaY644bGRkyF7Ulw0bNjR0EYSJkF2ARH25cWdIIYzlmWeeaegi1Ol///tfQxfBoBdeeKGhi3DX3fU1KUIIIYQQQghxJ2ThgBBCCCGEELdApnvVHxlJEUIIIYQQQjQq0kgRQgghhBBCNComNd1r5cqVDV0EYSJkK1xRX+65556GLoIwEX/elVMIY2rMC+dvZ/dTcXvkNy2EEEIIIYRoVKSRIoQQQgghhGhUTGq6lxBCCCGEELdLdveqPzKSIoQQQgghhGhUjDqSkpiYSFRUFLGxsaxbt04JP3fuHL6+vspzrVaLt7c3ffv25ZVXXqFDhw7KaxkZGbz22mvs37+f06dP89JLLzFv3jxjFrtZCwgIIDg4GLVazZUrV9i1axeZmZl1xvfw8CAyMhJHR0dKSko4ePAgx48fNxi3Xbt2xMbGkpaWxvr1641VBdFEBAYGEhISglqt5vLly+zcuZOMjIw647dq1YpevXrh5OREcXExycnJHDt2THm9U6dO9O/fXy/d/PnzZaMCE1dTU8Py5cvZtGkTV69epUOHDjz99NN4eXnVmWbjxo1s376dCxcuANC2bVseeeQR2rdvr8QpLS1l6dKlJCUlUVBQgK+vL08++aROHGFaBg4cSFxcHA4ODly8eJFvvvmGlJSUv0zXsWNH3n77bS5cuMDEiROVcE9PTx566CHatm2Lq6sr33zzDWvXrjVmFYRoMow6kvL111/zj3/8g507d3L+/Hm91zdu3EhGRgaHDh1i1qxZnDhxguDgYDZt2qTEKS8vx8XFhbfeeovg4GBjFrfZa9euHZGRkRw4cICffvqJjIwMhg4dilarNRjf1taWIUOGkJGRwU8//cSBAwfo1auXTgOzllarJSIigvT0dGNXQzQB7du3p3fv3uzbt49ly5aRkZHBvffee9NrbdiwYWRkZLBs2TL2799PVFQUbdu21YlXXl7ON998o/OQBor49ddfWbNmDWPHjmX27Nk4ODgwc+ZMSktL60xz/PhxevXqxfTp05k5cyYtW7Zk5syZXL58WYkzf/58Dh8+zIsvvsiHH35IUFAQ7777rk4cYToiIyN56qmn+Pnnn5k0aRInTpzgrbfewtnZ+abp1Go1L774IkeOHNF7zcrKiuzsbBYvXsyVK1f+P/buOyyqK3/8+JuhD70PFoqNDmIBUbGBJRE1iSYaY4wJJtHdGE3b6Ca7GzeJbnZj1rj5pphgXLMmsSb2BlZERVFQFLGigvQOA+gw/P7g4f4cZ0BjGETmvJ7nPo9z5nPunKPXO3PuafoqutCKjIyM2uXREemtkVJdXc3atWuZPXs2MTExrFy5UivGyckJhUJBt27dmDBhAvHx8YSHhxMbGyv98PDy8uLzzz9n+vTp2NnZ6au4BiE4OJjz589z/vx5ysrKSEpKoqqqCn9/f53x/v7+VFVVkZSURFlZmZT37saikZERUVFRnDhxgsrKyraoitDO9e7dm4yMDDIyMigtLSUxMZHKykoCAwN1xgcGBlJZWUliYiKlpaVS3tDQUK1YpVKpcQiGraGhge3bt/Pkk08SHh6Oh4cHf/zjH6mrqyMxMbHZfK+//jqjR4/Gy8uLzp07M2vWLBoaGqQfkrdu3eLYsWNMmzYNf39/FAoFzzzzDK6uruzevbutqie0IzExMezdu5e9e/eSk5PDypUrKSoqYtSoUS3me+WVV0hMTOTChQta712+fJkffviBpKQkbt++ra+iC8IjSW+NlDVr1uDj44OPjw/Tpk3j+++/p6GhoeXCyGTMnTuXa9eukZKSoq+iGSSZTIaLi4s0tKFJdnY2CoVCZx43Nzeys7M10m7cuIGLi4vGOuF9+/altrb2vrq8hY6v6Vq7u/f0xo0bzV5rCoVC69q8fv261rVmamrK9OnTeeGFFxg7duw9n2AKHV9BQQFlZWUaD09MTU3x9/cnMzPzvs9TV1eHSqWSevvq6+tRq9WYmppqxJmZmYl7nQEyMTGhW7dupKWlaaSfPn0aHx+fZvMNGzYMNzc31q1bp+8iCkKHo7dGSlxcHNOmTQNgzJgxVFVVaQzjao6vry/QOG/l96irq6OiokLjMOSnFBYWFshkMq3hD0qlErlcrjOPXC7XelJdU1ODsbExFhYWQOOPS19fXw4cOKCfgguPnOautZqamhavNV3xd15rpaWlJCQksG3bNnbv3o1KpeKpp54SPawGrqysDEDrOrCzs6O8vPy+z7N69WocHR0JCgoCwNLSkl69erFhwwZKSkpQq9UcPHiQS5cuiWE5BsjGxgZjY2PpemtSVlaGvb29zjwKhYLnnnuOZcuWoVar9V9IoU087GFdYrjX75SZmUlycjJTpkwBGp9ATJ48mRUrVtwzb1Nvy+/9C1+8eDF2dnYah5jMrc3IyOiePVx3x0Pjv5OpqSkjRozgwIED1NbW6quIwiPqt1xX9xOfn5/PhQsXKC4uJjc3l127dlFeXk5wcPDvKabwiDl06BDPP/+8dDQNDb77O+O3XH+bNm3i8OHDvP3225iZmUnpr732Gg0NDcyaNYupU6eyY8cOBg0aJHacFiTNfYc2jQxpmpMnCMJvp5fVveLi4lCpVHTu3FlKa/pRe68nUBkZGQA6J2f/FgsWLODNN9/USPvvf//7u875KKutrUWtVmNpaamRbmlp2ezkUl29LBYWFtTX11NXV4eDgwO2trY89thj0vtNPxReeeUVfv75ZyoqKlq5JkJ713St3X3tWFpaNjuHRNe1ZmlpSX19fYsN4Pz8fNGTYmD69eunsQJkUw95WVkZDg4OUnpFRcV9XRubN2/ml19+4S9/+Quenp4a7ykUChYuXEhtbS01NTU4ODjw73//G1dX11aqjfCoqKyspL6+XqvXpLkeOwsLC3r06IG3tzexsbFA4/ejTCbj559/5qOPPiI9Pb0tii4Ij6xWb6SoVCpWrVrFkiVLtCaTTZw4kdWrVxMTE6Mzr1qtZtmyZXh7e+ucMPtbmJubY25urpF299hiQ6JWqyksLKRr164aQ+k6d+7c7NC6/Px8rS/trl27UlhYiFqtpqysjDVr1mi8HxYWhqmpKYcPH6aqqqq1qyE8Au681q5evSql3/36Tnl5eXh5eWmkeXh4SNdac5ydncVKSwbG0tJS42FLQ0MD9vb2nD59Wnq4pVKpOHfuHM8991yL59q8eTMbNmzgvffeo3v37s3GWVhYYGFhQVVVFWlpadJQZsFwqFQqrly5QnBwMMnJyVJ6cHAwx48f14qvqanRelA6evRoAgMDWbJkCQUFBXovs6Afoie17bR6I2Xr1q2UlpYSGxur9RRr0qRJxMXFSY2U4uJi8vLyUCqVpKens3TpUpKTk9m2bRvGxsZSvtTUVACqqqooLCwkNTUVMzOzZlelEnQ7ffo0I0aMoKCggPz8fPz9/bGxsZH2PQkLC8PKyop9+/YBjUt0BgYGEhERQUZGBm5ubvj6+hIfHw80Tiy9u2fs1q1bAGLMtoFLTU0lOjqawsJC8vLypGutad+TAQMGYGVlJc1TS09PJygoiEGDBnHu3DkUCgV+fn4aqyj179+fvLw8ysvLMTMzIzg4GGdnZw4ePPhQ6ii0D0ZGRjz++OP88ssvuLu7o1Ao+OWXXzA3N2fw4MFS3BdffIGjoyNTp04FGod4rVmzhtdffx1XV1dprkFTgwT+/3dPp06dyMvL44cffqBTp04MGzasLasotBNbt25lzpw5XL58mQsXLhAdHY2zs7N0n5o6dSqOjo588cUXNDQ0aC0GUl5ezu3btzXSTUxM6NKli/RnJycnvLy8qK2tbXEPM0EwBK3eSImLiyM6OlpnN/vEiRNZtGiR9OQzOjoaaJw06+npyfDhw1m+fLnWRll39qqkpKTw448/4unp+bsn1xuay5cvY2FhQb9+/aQN9rZv3y71eFhZWWFjYyPFV1ZWsn37dgYOHEhgYCDV1dUcPny42afhgtDk0qVL0rVmZWVFcXExW7ZskZaolsvlWtfa1q1bGTx4MEFBQVRXV3Po0CGuXLkixZiZmTF8+HDkcjl1dXUUFRXxyy+/iCeSAhMmTODWrVt89913VFdX06NHD9577z2NHpeioiKNeStNiy989tlnGueaNGkSzzzzDNA4DPGnn36iuLgYa2trwsPDefbZZzEx0es+yEI7lZSUhLW1NZMmTcLBwYEbN26waNEiioqKAHBwcPjNKw46ODjwr3/9S3o9fvx4xo8fz9mzZ/nggw9as/iC8Mgxavits1sfYV9//fXDLoJgIMQGg0JbubO3QBD06aOPPnrYRRAMRHtesvl+FoF6GF566aWHXYRWJwbWCYIgCIIgCIKB+fLLL/H29sbCwoK+ffty6NChFuMPHDhA3759sbCwoFu3bnp/+C8aKYIgCIIgCIJgQNasWcO8efN47733OHXqFJGRkTz22GNaGzE3uXr1Ko8//jiRkZGcOnWKP//5z7z++uts2LBBb2UUjRRBEARBEARBuA8Pe9PG1trM8bPPPiM2NpaZM2fi5+fH0qVL6dq1K1999ZXO+K+//hoPDw+WLl2Kn58fM2fO5KWXXuLTTz/9vX+lzRKNFEEQBEEQBEF4hNXV1VFRUaFx1NXV6Yy9desWKSkpWluFjBo1iqSkJJ15jhw5ohU/evRoTpw4Ie1X1doMaokSsSyu0FaalpQUBH3Lzs5+2EUQDITYfFAQ2q/FixezcOFCjbS//e1vOleJKyoqor6+Hjc3N410Nze3Zpe+zsvL0xmvUqkoKirC3d3991VAB4NqpAiCIAiCIAjCg3qQoVVtYcGCBVobiN69qfnd7q5LQ0NDi/XTFa8rvbWIRoogCIIgCIIgPMLMzc3v2Shp4uzsjLGxsVavSUFBgVZvSROFQqEzvmkTUn0Qc1IEQRAEQRAEwUCYmZnRt29f9uzZo5G+Z88eBg4cqDNPRESEVvzu3bvp168fpqameimnXntSkpKSiIyMZOTIkezcuVNKz8rKwtvbW3ptbW2Nh4cHw4YNY968efTs2VN6b+PGjXz11VekpqZSV1dHQEAAH3zwAaNHj9Zn0TusPn36EB4ejrW1NYWFhcTHxzc7pt3KyoqoqCgUCgWOjo6cOHGC+Ph4jZhevXoxcOBAHBwckMlklJaWkpycLMYuCwAEBwfTs2dPzMzMKCoqIjk5mfLy8hbzeHh4EBISgo2NDZWVlaSmpnLjxg3pfVdXVwICAnB0dEQul7N//36N9wXD09DQwK5duzh69ChKpRJPT08mTpyIQqFoNk9eXh47duwgOzub0tJSJkyYwNChQzViLl++zL59+8jOzqaiooIXX3yRoKAgfVdHaMeeffZZXnrpJVxcXLh06RKLFy8mJSWl2XhTU1P+8Ic/MH78eJydncnLy+Obb75h48aNAPTo0YM5c+YQEBBA586dWbx4MatWrWqr6ggPoL0O9/qt3nzzTZ5//nn69etHREQEy5cv5/r168yaNQtoHD6Wk5MjXY+zZs3iiy++4M033+Tll1/myJEjxMXF8dNPP+mtjHrtSVmxYgVz5swhMTFR57rL8fHx5ObmkpaWxqJFi8jIyCAkJISEhAQp5uDBg4wcOZLt27eTkpLC8OHDGTduHKdOndJn0TskPz8/oqOjSUpKYsWKFWRnZzN58mRsbW11xpuYmKBUKklKSiI/P19nTG1tLUlJSaxatYq4uDhOnz7N2LFjNRqhgmEKCAjAz8+P5ORkduzYQW1tLdHR0ZiYNP9sxNnZmcjISK5evcrWrVu5evUqQ4YMwdnZWYoxMTGRGsOCALB3714OHDjAU089xRtvvIGNjQ1ff/01tbW1zea5desWTk5OxMTEYGNj02xMp06deOqpp/RVdOER8thjjzF//ny++eYbnnrqKVJSUvjmm29anDD873//m4iICN5//30ee+wx3n77ba5cuSK9b2FhwY0bN/jss88oLCxsi2oIAgCTJ09m6dKl/P3vf6d3794cPHiQ7du34+npCUBubq7Gb3dvb2+2b9/O/v376d27Nx9++CHLli1j4sSJeiuj3npSqqurWbt2LcePHycvL4+VK1fy17/+VSPGyclJetLVrVs3xo0bR1RUFLGxsVy+fBljY2OWLl2qkWfRokVs2rSJLVu2EBoaqq/id0hhYWGkpaWRlpYGNDYSvb29CQ0N5cCBA1rx5eXlUs9JcHCwznPe3fg8ceIEQUFBdO3alatXr7ZyDYRHia+vL+np6VIvx+HDh3n66afx9vbm4sWLOvP4+fmRm5sr9cSlp6fj6uqKr68viYmJANy8eZObN2+2TSWEdq+hoYGDBw8SHR0t3aemTp3KX//6V06ePNns0AUPDw88PDwA2Lp1q84YPz8//Pz89FNw4ZHzwgsvsHHjRtavXw80rqY0aNAgpkyZwr///W+t+MGDB9O/f39GjRol9SDffe9KT0+X7nd3T3oWBH37wx/+wB/+8Aed761cuVIrbejQoZw8eVLPpfr/9NaTsmbNGnx8fPDx8WHatGl8//330ioAzRZGJmPu3Llcu3at2e5TtVpNZWUljo6O+ih2hyWTyVAoFFoNh6tXr7bqcrmenp44Ojo2u2OpYBisra2Ry+UaX8hqtZr8/HxcXFyazefi4kJubq5GWm5ubot5BMNWUlJCZWUlPj4+UpqJiQndu3cnKyvr4RVM6FBMTU0JCAjg8OHDGumHDx9u9oHpiBEjOHv2LLGxsezfv58dO3bwzjvv3PfkZqF9kslk7fLoiPTWkxIXF8e0adMAGDNmDFVVVSQkJBAdHd1iPl9fX6Bx3kpYWJjW+0uWLKG6uppnnnmmxfPU1dVpbWKjUqlaHGrSkcnlcmQyGdXV1Rrp1dXVWFlZ/a5zm5ub89prr2FsbCyNDRc/DgybpaUlgNZwm9ra2havNwsLC2pqajTSampqpPMJwt0qKioAtIZs2djYiL2xhFZjb2+PiYkJRUVFGunFxcUaw1Hv1KVLF/r06UNdXR1z5szBwcGBv/71r9jZ2fH++++3RbEF4ZGml1/smZmZJCcnSxPDTExMmDx5MitWrLhnI6WlNZd/+uknPvjgAzZt2oSrq2uL59G1qc2IESPu+fmGpjUmgNXV1bFixQpMTU3x8vIiKiqKsrIy0ZtiQLy9vQkPD5de7927t9nYe/Wo3q2jTFIUWkdKSgrr1q2TXs+cORNofv1+QdAnIyOjZq81mUxGQ0MD77zzDlVVVQB88sknLF26lA8//LDZ3cAFQWikl0ZKXFwcKpWKzp07S2kNDQ2Ympre88lWRkYGgNbE6zVr1hAbG8u6devuq6Gha1Obzz///H6r0OEolUrUarXWU2y5XK7Vu/Igmv5dCwoKcHJyIiIiQjRSDMiNGzc0njA2dT3f3TNiYWHR4mTm2tparV4TXb0rguEKCAiQ5pIA1NfXA409KncuAlJVVdXshHhB+K3KyspQqVRavSaOjo4UFxfrzFNYWEh+fr7UQIHGFeOahl9fu3ZNr2UW9EM8OGs7rT6ITaVSsWrVKpYsWUJqaqp0pKWl4enpyerVq5vNq1arWbZsmTSZu8lPP/3EjBkz+PHHHxk7dux9lcPc3BxbW1uNw1CHekHj321eXp5W48/b27vZJYgflJGREcbGxq16TqF9U6lUVFZWSkd5eTlKpVJj1RuZTIabm1uLK9gUFhZqrZTj7u4uVr0RJBYWFri4uEiHm5sbNjY2XLhwQYpRqVRcvnwZLy+vh1dQoUO5ffs2Z8+e1VqIYeDAgc2uNnry5ElcXV2Ry+VSmpeXF/X19Vqb4gmCoK3Vf7Vv3bqV0tJSYmNjsbOz03hv0qRJxMXFERMTAzSO5czLy0OpVJKens7SpUtJTk5m27Zt0o/cn376ienTp/P5558zYMAA6T+2paWl1vmFliUnJzNu3Dhyc3PJycmhd+/e2NraSjfYoUOHYmNjo7HSTdOwOjMzM+RyOa6urtTX10tPjiIiIsjNzaWsrAyZTEb37t0JDAxk165dbV9BoV05f/48QUFBUsMlMDAQlUqlsXjDwIEDqampka7B8+fPM2rUKAICArhx4wZdu3bF3d1d43oyMTHReEJubW2Ng4MDdXV1KJXKtqug0C4YGRkxZMgQ4uPjcXZ2xsXFhfj4eMzMzOjTp48U9+OPP2Jrayt9/6hUKmlp9fr6esrLy8nJycHMzExaqKGurk6jh7CkpIScnBzkcjkODg5tWEuhPfjvf//LP/7xD9LT00lNTeWZZ57B3d2dNWvWAPDGG2/g5ubG/PnzAdi2bRuzZ8/m448/5osvvsDBwYF33nmHjRs3SkO9TE1N6d69u/TnptUMlUqlGI0gGLxWb6TExcURHR2tswExceJEFi1aRElJCYA0bEsul+Pp6cnw4cNZvnw5PXr0kPJ88803qFQq/vjHP/LHP/5RSn/hhRd0Lo8mNC8jIwNLS0sGDRokbea4du1aaeKptbW11p4psbGx0p/d3d0JCAigrKyMr776Cmi8qY4ePRobGxtUKhXFxcVs2bJFGrYnGK6zZ89ibGxMWFgY5ubmFBUVkZCQgEqlkmLuHn5YWFjIoUOH6N27NyEhIVRVVXHw4EGNH4pOTk6MGjVKet2vXz+gcRhFUlKSnmsltEcjRozg9u3bbNiwgZqaGjw8PHj11VexsLCQYkpLSzWGaVRUVLBkyRLp9f79+9m/fz/du3eXvmtu3LjBl19+KcVs2rQJgP79+/Pss8/qu1pCO7Njxw7s7e35wx/+gIuLCxcvXmTWrFnSKoYuLi4aPcFKpZLY2Fjef/991q1bR1lZGTt37tQYeu7i4sIvv/wivY6NjSU2Npbk5GReeOGFtquccN/EcK+2Y9RgQLMLFy9e/LCLIBiI1lzWWRBaIpZjF9rK22+//bCLIBiI9vygs6VpCw/Tc88997CL0Oo65sLKgiAIgiAIgiA8sgx3JrkgCIIgCIIg/AZiuFfbET0pgiAIgiAIgiC0K6KRIgiCIAiCIAhCu2JQw72cnJwedhEEA3HnMruCoE+3b99+2EUQDIQh7zUmCE2aNisW9E/8TQuCIAiCIAiC0K6IRoogCIIgCIIgCO2K6LsVBEEQBEEQhPsgVvdqO3ptpCQlJREZGcnIkSPZuXOnlJ6VlYW3t7f02traGg8PD4YNG8a8efPo2bOn9F5iYiLvvvsu58+fR6lU4unpyauvvsobb7yhz6J3WP7+/gQHByOXyyktLeXIkSPk5eU1G+/u7s6AAQNwcHBAqVSSlpamscmSr68vPXv2lDaUKyws5Pjx4xQWFuq9LkL71q9fPyIiIrCxsaGgoIDdu3dz/fp1nbHW1taMHDkSd3d3nJycOHbsGLt379aKMzc3Z8SIEfj6+mJpaUlpaSl79uzh0qVL+q6O0I75+voSFBSEpaUlZWVlHDt2jPz8/GbjFQoFYWFh2NvbU1NTw+nTp8nMzJTet7e3p0+fPjg5OWFjY8PRo0c5d+5cW1RFaOcmT57MjBkzcHFx4fLly3zyySecPHlSZ2y/fv34/vvvtdLHjx8vzVs0MTFh5syZjB8/HldXV7Kysvj3v//N4cOH9VoPQXgU6LWRsmLFCubMmcN3333H9evX8fDw0Hg/Pj6egIAAlEolZ86c4fPPPyckJIQtW7YQFRUFgJWVFa+99hrBwcFYWVmRmJjIq6++ipWVFa+88oo+i9/hdOvWjYiICBITE8nPz8fPz4/HHnuMtWvXUl1drRVvY2PDmDFjOH/+PPv27cPNzY3BgwdTW1sr3WDd3d25fPkySUlJqFQqevfuzeOPP866detQKpVtXUWhnfD392f06NFs376dGzdu0KdPH6ZOncqXX35JRUWFVryxsTFKpZLExETCw8N1nlMmkzFt2jSUSiXr16+noqICW1tbbt26pe/qCO2Yt7c34eHhHDlyhPz8fHx9fRk1ahQbN27UeV9rahBfuHCBAwcO4ObmRkREBLW1tVy7dg1o/OFYWVnJ1atXm70eBcMzevRo3n33XT766CNOnTrF008/zVdffcWECRNafNgXExNDVVWV9Lq0tFT685w5cxg7diwLFy7k6tWrDBw4kKVLl/L8889z/vx5vdZHENo7vc1Jqa6uZu3atcyePZuYmBhWrlypFePk5IRCoaBbt25MmDCB+Ph4wsPDiY2Npb6+HoDQ0FCeffZZAgIC8PLyYtq0aYwePZpDhw7pq+gdVnBwMJmZmWRmZlJWVsaRI0eoqqrC399fZ7yfnx9VVVUcOXKEsrIyKW9wcLAUs2/fPs6dO0dxcTHl5eUcPHgQIyMjOnfu3FbVEtqhiIgITp06xalTpygqKmL37t2Ul5fTr18/nfHl5eXs2rWL06dPU1dXpzMmNDQUS0tL1qxZw40bNygvL+fGjRstPjEXOr7AwEAuXLjAhQsXKC8v59ixY1RXV+Pr66sz3tfXl+rqao4dO0Z5eTkXLlzg4sWLBAUFSTFFRUUcP36cq1evSt9FgjB9+nQ2btzIxo0buXr1Kv/85z/Jy8tj8uTJLeYrKSmhuLhYOtRqtfReTEwM3333HYcOHSI7O5u1a9eSlJTECy+8oO/qCA/IyMioXR4dkd4aKWvWrMHHxwcfHx+mTZvG999/T0NDQ8uFkcmYO3cu165dIyUlRWfMqVOnSEpKYujQofoodoclk8lwdnYmOztbIz07Oxs3Nzededzc3HTGu7i4NPsfwsTEBJlM1uwPTaHjk8lkUg/bna5cuULXrl0f+Ly9evUiOzubxx57jDfffJNZs2YxePDgDntzFu5NJpPh5OTEzZs3NdJzcnJwdXXVmcfV1ZWcnByteGdnZ3EtCc0yMTHB39+fpKQkjfSkpCR69+7dYt61a9eyd+9evv32W/r376/xnpmZmdb3ZW1tLaGhoa1SbkF4lOmtkRIXF8e0adMAGDNmDFVVVSQkJNwzX9PTr6ysLI30Ll26YG5uTr9+/fjjH//IzJkzWzxPXV0dFRUVGoch7ydgYWGBTCajpqZGI72mpga5XK4zj6Wlpc54mUyGhYWFzjxhYWFUV1dr/QgQDIdcLkcmk2kNtamursbKyuqBz+vg4IC/vz8ymYyffvqJQ4cOMWDAACIjI39vkYVHlLm5eZvc1wTBwcEBExMTiouLNdKLi4ub3YOtqKiIDz74gDfffJM33niDrKwsvvvuO/r27SvFJCUlMX36dDw8PDAyMiIiIoLhw4fj4uKi1/oIwqNAL3NSMjMzSU5OZuPGjY0fYmLC5MmTWbFiBdHR0S3mbeptufuJ1qFDh6iqquLo0aPMnz+fHj168OyzzzZ7nsWLF7Nw4UKNtJiYGMaNG/cgVeow7u7Nas0nhyEhIXTv3p2tW7eKIRJCqzMyMqK6upqtW7fS0NBAbm4uNjY2REREcPDgwYddPOEh0nVfu1fPvSC0hpa+Q7OysjQeuKalpaFQKHjhhRek0SL/+Mc/+OCDD9i8eTMNDQ3cuHGDTZs2MWHCBH0XXXhAose17eilkRIXF4dKpdKYl9DQ0ICpqanGhDFdmlaOunP1rztfBwUFkZ+fzwcffNBiI2XBggW8+eabGmk//PDDb6pHR1JbW4tardZ6umhhYdHsBPeamhosLS010iwtLVGr1dTW1mqkBwcH07t3b7Zt20ZJSUnrFl54pCiVStRqtVaviZWVlc6JzPerqqqK+vp6jR+fRUVF2NjYIJPJNMZ5C4ahrq6u2fva3b0lTX7LfU0QmpSWlqJSqbR6TRwdHbV6V1py+vRpYmJiNM47d+5czMzMsLe3p6CggDfeeEOMRhAE9DDcS6VSsWrVKpYsWUJqaqp0pKWl4enpyerVq5vNq1arWbZsGd7e3i2Ox2xoaLjnnAdzc3NsbW01DlNT0weu16NOrVZTVFSkNaG9S5cuzU48zs/Pp0uXLlrxhYWFGj8Ug4OD6dOnDzt27KCoqKj1Cy88UtRqNbm5uXTr1k0jvVu3bty4ceOBz3vjxg1pqesmjo6OVFZWigaKgVKr1RQXF9OpUyeN9E6dOlFQUKAzT0FBgc74oqIi0fsiNEulUnHu3DkiIiI00iMiIkhNTb3v8/j6+upcov/WrVsUFBRgYmJCdHQ0+/bt+71FFoRHXqv3pGzdupXS0lJiY2Oxs7PTeG/SpEnExcVJTxGKi4vJy8tDqVSSnp7O0qVLSU5OZtu2bRgbGwPwf//3f3h4eEhzVRITE/n000+ZM2dOaxe9wzt9+jTDhw+nqKhIWoLY2tpa6r3q378/VlZW7N+/H2js1QoICGDAgAGcP38eNzc3fHx82Lt3r3TOkJAQ+vXrx969e6msrJSeUN6+fRuVStXmdRTahyNHjvDkk0+Sm5tLdnY2ffr0wc7OThriMGLECGxsbNi0aZOUp2kBBzMzM6ysrHBzc6O+vl5q+J44cYL+/fszZswYkpOTcXJyYvDgwSQnJ7d9BYV2Iz09nSFDhlBUVERBQQE+Pj5YW1tLy7f27dsXKysraUjg+fPn8fPzIywsjMzMTFxdXenVq5d034PGCfn29vZA4/LYVlZWODo6cvv2bSorK9u6ikI7sWrVKhYvXszZs2dJS0vj6aefxt3dnbVr1wIwd+5cXF1dee+99wCYNm0aN2/e5NKlS5iamhITE8OoUaOYN2+edM6goCBcXV2la3H27NnIZDKd+6sI7YMY7tV2Wr2REhcXR3R0tFYDBWDixIksWrRIGg7UND9FLpfj6enJ8OHDWb58OT169JDyqNVqFixYwNWrVzExMaF79+784x//4NVXX23tond4V65cwcLCgj59+iCXyykpKWHHjh3S+u1yuRxra2spvrKykp07dxIREUFAQADV1dUkJSVJe6RA434YxsbGjBw5UuOzUlJSml2hTej4zp07h1wuZ8iQIVhbW1NQUMCPP/5IeXk50LhXxd33iDv/T3fq1ImgoCDKyspYtmwZABUVFaxevZpRo0Yxa9YsKioqSE5OFpueGbirV69ibm5O7969pU1qd+/eLQ0tlMvlGkMPq6qq2LNnD2FhYfj5+aFUKjl69Ki0R0pTnieeeEJ6HRQURFBQELm5uezYsaPN6ia0L7t27cLe3p5Zs2bh4uLCpUuX+MMf/kBubi4ALi4uuLu7S/Gmpqa89dZbuLq6UldXJ8XfuYWCubk5c+bMoUuXLiiVSg4dOsSf//xn0RgWBMCowYD6t5cvX/6wiyAYiJY29hKE1nT3kExB0Jd///vfD7sIgoE4c+bMwy5Cs9avX/+wi6DTpEmTHnYRWp1ed5wXBEEQBEEQhI5CJtPb7h3CXcTftCAIgiAIgiAI7YpopAiCIAiCIAiC0K6I4V6CIAiCIAiCcB/E6l5tx6AaKYa8T4rQtiwsLB52EQQD0bNnz4ddBMFAGNA6O4IgtANiuJcgCIIgCIIgCO2KQfWkCIIgCIIgCMKDEsO92o7oSREEQRAEQRAEoV3Ra09KUlISkZGRjBw5kp07d0rpWVlZeHt7S6+tra3x8PBg2LBhzJs3r9kx1ocPH2bo0KEEBgaSmpqqz6J3aL1798bHxwczMzMKCws5evQoZWVlLebx9PSkT58+2NjYUFlZSUpKCtevX9eI8fX1JTAwEEtLS8rKykhOTiY/P1+PNRHas969exMWFoa1tTVFRUXs3buX7OxsnbFWVlYMHz4chUKBg4MDKSkp7N27VyPGycmJwYMHo1AosLOzIyEhgZSUlLaoitDONTQ0sHnzZg4ePIhSqcTb25vnnnuOzp07N5snJyeHTZs2ce3aNYqLi5k8eTIjR47UiKmtreXXX3/l5MmTVFZW4uHhwZQpUzS+vwTDMnnyZF588UVpx/lPPvmEkydPNhtvamrK7NmziYmJwdnZmfz8fJYvX84vv/wixURHRzNnzhy6du3KjRs3WLZsGQkJCW1RHUFo1/Tak7JixQrmzJlDYmKi1g9agPj4eHJzc0lLS2PRokVkZGQQEhKi8z9neXk506dPJyoqSp9F7vCCgoIICAjg6NGjbNmyhZqaGkaPHo2JSfPtVRcXF4YNG8alS5fYtGkTly5dYvjw4Tg7O0sx3t7ehIWFkZaWxubNm8nPz2fkyJFYWVm1RbWEdsbX15eoqCiOHj3KypUryc7OZtKkSdjY2OiMNzY2RqlUcuTIEQoKCnTGmJqaUl5ezoEDB6iqqtJn8YVHzM6dO9mzZw9Tp07l/fffx87Ojs8++4za2tpm89y6dQsXFxcmTpyInZ2dzpiVK1dy7tw5Zs6cyQcffIC/vz+fffYZpaWl+qqK0I6NGTOG+fPn8+233/L0009z8uRJvv76axQKRbN5lixZQnh4OH/961+JiYnhnXfe4erVq9L7ISEhfPrpp2zZsoWJEyeyZcsWPv30U4KCgtqiSsIDMDIyapdHR6S3Rkp1dTVr166VniCsXLlSK8bJyQmFQkG3bt2YMGEC8fHxhIeHExsbS319vUbsq6++ytSpU4mIiNBXkQ2Cv78/p0+f5tq1a5SVlXHo0CGMjY3p3r17s3kCAgK4efMmZ86coby8nDNnznDz5k0CAgI0Yi5evMjFixcpLy8nOTmZ6upqfH1926JaQjvTr18/Tp8+zenTpykpKWHv3r1UVlYSGhqqM76iooK9e/dy9uxZ6urqdMbk5eWxf/9+zp8/r3V/EAxXQ0MD8fHxjB07lr59+9K5c2deeuklbt26xbFjx5rN5+3tzdNPP01YWJjOhzS3bt3i5MmTTJo0iV69euHm5saECRNwdnZm//79eqyR0F5Nnz6djRs3smHDBq5cucInn3xCXl4eU6ZM0Rk/aNAg+vXrx+zZszl69Cg3b94kPT1dYyTI888/z5EjR/juu++4evUq3333HceOHeP5559vo1oJQvult0bKmjVr8PHxwcfHh2nTpvH999/fc/lCmUzG3LlzuXbtmsYwju+//57Lly/zt7/9TV/FNQjW1tbI5XJycnKkNLVaTX5+Pq6urs3mc3Fx0cgDjUMlmvLIZDKcnJy0Ym7evNnieYWOSSaToVAoyMrK0ki/evVqi8NvBOFBFBUVUV5ervHQxNTUFB8fHy5duvTA51Wr1ajVaq2l601NTbl48eIDn1d4NJmYmODv709SUpJGelJSEiEhITrzDB8+nLNnz/LSSy+RkJDA1q1befvttzE3N5diQkJCtM55+PBhevfu3ep1EIRHjd7mpMTFxTFt2jSgsYu0qqqKhIQEoqOjW8zX9OQ9KyuLsLAwLl68yPz58zl06FCLQ5KEe5PL5QDU1NRopNfU1GBtbd1sPktLS61hE7W1tVhaWgJgbm6OTCbTiqmpqZFiBMMhl8uRyWRUV1drpCuVSjH8T2h15eXlANja2mqk29raUlxc/MDntbCwoHv37mzZsgV3d3dsbW05duwYV69eFQ9fDJCDgwMmJiZa11RxcbHG0Oc7denShT59+nDr1i3mzp2Lg4ODNBzxL3/5CwDOzs6/6ZzCwyeTiTWn2opefvVnZmaSnJzMxo0bGz/ExITJkyezYsWKezZSmnpbjIyMqK+vZ+rUqSxcuJBevXr9pjLU1dVpDRu5ffu2QW3o2K1bNwYOHCi93rNnT7Ox9+rl0vX+3Wn3EyMYNnE9CL/X0aNH+eGHH6TXr7/+us641rjWYmNjWblyJW+//TYymQwPDw/CwsJ0zrEUDMPd15WRkVGz15pMJqOhoYF3331Xmkf3r3/9i88++4yPPvpI+o3yW84pCIZEL42UuLg4VCqVxtCOhoYGTE1N7znhMCMjA2gcL1xZWcmJEyc4deoUr732GtDYBd/Q0ICJiQm7d+9mxIgROs+zePFiFi5cqJE2fvx4nnjiid9Rs0fL9evXKSwslF4bGxsDjT0jd/am6OopuZOuHhELCwspT11dHWq1WivmXucVOialUolardbqNZHL5SiVyodUKqGj6N27t8bqWiqVCmic12Rvby+lV1ZWavWu/Faurq786U9/oq6ujpqaGuzt7fn666/FU24DVFpaikql0vq3d3R0bLbHrrCwkIKCAo2FPq5cuYJMJsPNzY3r169TVFT0m84pCIak1fusVCoVq1atYsmSJaSmpkpHWloanp6erF69utm8arWaZcuW4e3tTWhoKLa2tpw5c0bjPLNmzcLHx4fU1FTCw8ObPdeCBQsoLy/XOMaOHdva1W3XVCoVlZWV0lFWVoZSqaRTp05STNPNsrkVlaDxRnv3XILOnTtLedRqNcXFxRrnBejUqVOL5xU6JrVaTV5eHl5eXhrpXl5eWvOWBOG3srCwwM3NTTo6deqEnZ0dZ8+elWJUKhWZmZn06NGjVT7T3Nwce3t7qqurOXv2rJgvYIBUKhXnzp3TWrwnIiKCtLQ0nXlOnTqFi4uLxgM8T09P6uvrpeX509LStM45cOBAsc1CO/awV/EypNW9Wr0nZevWrZSWlhIbG6u1rOOkSZOIi4sjJiYGaBx3mZeXh1KpJD09naVLl5KcnMy2bdukp/6BgYEa53B1dcXCwkIr/W7m5uYak9MAgxrq1Zxz584RHBxMRUUFFRUVBAcHU19fz+XLl6WYyMhIlEqltHjBuXPneOyxxwgKCuL69et4eHjQqVMntm3bJuU5e/YskZGRFBcXU1BQgI+PD1ZWVpw/f77N6yg8fCdOnGDs2LHk5eWRk5ND7969sbW1lb54hwwZgrW1Ndu3b5fyNI3zNzMzQy6X4+rqSn19vfREUSaTSU8cjY2NsbGxwdXVlVu3bt1znx+h4zIyMiI6Oprt27dLDZdt27ZhZmam8SArLi4Oe3t7Jk6cCDT+6Lx586b057KyMq5fv465uTlubm4ApKenA0gPctavX49CoWDQoEFtXEuhPVi1ahWLFy/m7NmzpKWlMWnSJNzd3VmzZg0A8+bNw9XVlT//+c8AbNu2jVmzZvHRRx/xf//3fzg4OPDWW2/xyy+/SEO9/ve//7Fy5Upeeukl9u3bx/DhwxkwYADTp09/aPUUhPai1RspcXFxREdH61x3fuLEiSxatIiSkhIAaX6KXC7H09OT4cOHs3z58lZ7+iVoO3PmDMbGxkRERGBmZkZRURG7du2ShkxA48Z6d46HLSgoYP/+/fTp04fQ0FAqKyvZv38/RUVFUszVq1cxNzcnJCQEuVxOaWkpe/bs0Zo8LRiG8+fPY2FhwcCBA7GysqKoqIj169dTUVEBNF5jdw/FmTFjhvRnhUKBv78/5eXlfPPNN0Dj6nR3xoSFhUnzA37++We910lov8aMGcOtW7dYvXo11dXVdOvWjTfffBMLCwsppri4WONpY1lZGX//+9+l17t27WLXrl306tWLP/3pT0DjUNeNGzdSWlqKlZUVffr04cknnxSLuBionTt3Ymdnx6xZs3BxceHixYvMnj2b3NxcoHESvLu7uxRfU1PDyy+/zJ///GfWrFlDeXk5O3fu5D//+Y8Uk5qayjvvvMOcOXOYM2cON27c4J133uHMmTNtXj9BaG+MGgxodtb333//sIsgGIg75wIJgj6JvaOEtjJ79uyHXQTBQDT1YrZHd44iaU864pQGsY6aIAiCIAiCIAjtimikCIIgCIIgCILQroiBtYIgCIIgCIJwHzrqSlrtkehJEQRBEARBEAShXTGonpQ7N1QSBH26eyNDQdCXqVOnPuwiCAbizlWpBEEQ9M2gGimCIAiCIAiC8KDEcK+2I4Z7CYIgCIIgCILQrohGiiAIgiAIgiAI7YoY7iUIgiAIgiAI90EmE8/324peGylJSUlERkYycuRIdu7cKaVnZWXh7e0tvba2tsbDw4Nhw4Yxb948evbsKb23f/9+hg8frnXujIwMfH199Vn8DikoKIjQ0FCsrKwoKSnh0KFD3Lx5s9n4Tp06ERkZiaOjI9XV1Zw8eVJrJ1gzMzMiIiLo3r075ubmVFRUkJiYyLVr1/RdHaEdCwwMJDQ0FLlcTklJCYmJieTm5jYb36lTJwYNGiRda6dOneLs2bPS+76+vkRFRWnl+/rrr6mvr9dLHYRHw/Tp03n11VdxdXXlwoULLFy4kOTk5GbjzczMmDdvHk8++SQuLi7k5eXxn//8hzVr1gCwdu1aIiIitPIlJCQwY8YMfVVDeAQ0NDQQHx/PsWPHqKmpwcPDgwkTJqBQKFrMd+bMGXbv3k1xcTFOTk6MHj2awMBA6f36+nri4+M5deoUlZWV2Nra0rdvX0aMGCF+FAsGS6+NlBUrVjBnzhy+++47rl+/joeHh8b78fHxBAQEoFQqOXPmDJ9//jkhISFs2bJF68dIZmYmtra20msXFxd9Fr1D6tmzJ5GRkezfv5/c3FwCAwMZN24cq1ev1rnyma2tLePHj+fs2bPs3r0bd3d3hg0bRk1NDZcvXwYanyg88cQT1NTUsGPHDqqqqrC2tub27dttXT2hHenRoweDBw/mwIED5OXlERAQwLhx4/jxxx91Xms2NjbExMRw7tw54uPjUSgUDB06lJqaGq5cuSLF1dXV8eOPP2rkFQ0UwzZu3Dj+9re/8d5773HixAmee+45Vq1axYgRI5p9APPVV1/h7OzMO++8Q1ZWFs7OzhgbG0vvv/LKK5iamkqvHRwc2LVrF9u2bdN7fYT27cCBAxw6dIhnnnkGZ2dnEhIS+O6773jnnXcwNzfXmefatWv8+OOPjBo1ioCAAM6ePcvq1auZPXu29LvowIEDHD16lGeeeQY3Nzeys7NZt24dFhYWDB48uC2rKAjtht4aKdXV1axdu5bjx4+Tl5fHypUr+etf/6oR4+TkJD196NatG+PGjSMqKorY2FguX76s8aXh6uqKvb29voprEHr37s25c+c4d+4cAIcOHcLDw4OgoCCOHDmiFR8YGEhlZSWHDh0CoLS0FFdXV0JDQ6VGir+/PxYWFqxfvx61Wg1AZWVlG9VIaK969+5NRkYGGRkZACQmJtK1a1cCAwM5evSoVnzTtZaYmAhoXmt3NlIAlEql/isgPDJefvll1qxZw88//wzAwoULGTp0KM8//zyffPKJVvywYcMIDw9n8ODBlJWVAZCdna0R05TeZPz48dTU1LB161a91EF4NDQ0NJCYmMiIESOkXpDJkyfz4YcfcurUKQYMGKAzX2JiIj169JBGhbi6unLlyhUSExOlJcSvXbuGv78/fn5+ADg6OpKWlqZ1bQoPn1jdq+3orQ9xzZo1+Pj44OPjw7Rp0/j+++9paGhouTAyGXPnzuXatWukpKRovBcaGoq7uztRUVHs27dPX8XusGQyGa6urly/fl0j/fr167i7u+vMo1AodMa7urpK3c/e3t7k5uYydOhQYmNjmTp1Kv369RP/iQ2YTCbDxcVF69q5ceNGs0MiFAoFN27c0Ei7fv06Li4uGkMdTE1NmT59Oi+88AJjx47F2dm59SsgPDJMTU0JCgri4MGDGukHDx6kX79+OvOMHDmS06dPM2vWLI4fP86BAwd4//33sbCwaPZzpkyZwubNm6mpqWnV8guPlpKSEiorKzWGpJuYmNCtW7cWhzdfu3aNXr16aaT16tVLI4+XlxeXL1+msLAQgJs3b5KVlSWGtQsGTW89KXFxcUybNg2AMWPGUFVVRUJCAtHR0S3ma/oPmZWVRVhYGO7u7ixfvpy+fftSV1fHDz/8QFRUFPv372fIkCHNnqeuro66ujqNtNu3b2t04RsSS0tLZDKZ1lPompoa5HK5zjxyuVzrS1mpVGJsbIyFhQVKpRI7Ozu6dOlCZmYmmzdvxt7enqFDh2JkZMTx48f1Vh+h/bKwsEAmk2ldO7/1WqupqdG41kpLS0lISKC4uBgzMzOCg4N56qmnWLNmDeXl5Xqrj9B+OTo6YmJiIv2wa1JUVNTskGAPDw/69+9PXV0dL7/8Mg4ODnz88cfY29vz9ttva8X37t0bX19f3nnnHb3UQXh0NI0SsLGx0Ui3tramtLS02XxNw6DvznPnqINhw4ZRW1vLkiVLMDIyoqGhgdGjR9O7d+/Wq4AgPGL00kjJzMwkOTmZjRs3Nn6IiQmTJ09mxYoV92ykNPW2ND2Jb+qNaRIREcGNGzf49NNPW2ykLF68mIULF2qkjRkzhscff/yB6mSo7u790tVDUlNTw759+2hoaKCwsBArKyv69OkjGikG7l49p781Pj8/n/z8fOl1bm4ukydPJjg4WBqSKBgmXfep5q6npp65119/XfqR+Pe//51vvvmG999/n9raWo34yZMnc/78eVJTU1u/4EK7durUKel3DMCLL76oM66hoeGeowd0vX9nWlpaGqdOnWLKlCm4ubmRm5vLli1bpAn0gmCI9NJIiYuLQ6VS0blzZymtoaEBU1PTFp82ANIY9jtX/7rbgAED+N///tfieRYsWMCbb76pkfbdd9/dq+gdVk1NDWq1WutJtqWlZbNj/JVKpc74+vp66YtcqVSiVqs1fhCUlpZiZWWFTCaT5qkIhqO2tlYv15ou+fn52NnZ/f5CC4+kkpISVCoVrq6uGulOTk4UFRXpzJOfn09eXp7GU+xLly4hk8lQKBRkZWVJ6RYWFowfP54lS5bopfxC++bv70/Xrl2l1yqVCkBafatJdXW1Vk/Jne7uNQHt3pXt27czbNgwqefE3d2d0tJS9u3bJxop7YwYzt52Wn1OikqlYtWqVSxZsoTU1FTpSEtLw9PTk9WrVzebV61Ws2zZMry9vQkNDW027tSpU83Oo2hibm6Ora2txmGoQ72g8e+2oKBA44YLjUMfmlsWNi8vT2tFNg8PDwoKCqTGR25urtaPRHt7e6qqqkQDxUCp1WoKCwu1rrWuXbuSl5enM09eXp7Oa7OwsLDF68jZ2VlMpDdgt2/f5syZM0RGRmqkR0ZGcuLECZ15Tpw4gZubm0ajuFu3btTX12tdn+PGjcPMzEzjabpgOMzNzXF2dpYONzc3bGxsuHjxohSjUqm4cuUKnp6ezZ7H09NTIw/AhQsXNPLcvn1b68evTCb7zT3SgtCRtHojZevWrZSWlhIbG0tgYKDGMWnSJOLi4qTY4uJi8vLyuHLlCps3byY6Oprk5GTi4uKklb2WLl3Kr7/+ysWLFzl79iwLFixgw4YNvPbaa61d9A4vNTWVgIAA/Pz8cHBwYPDgwVhbW0v7nkRERDBy5EgpPj09HRsbGwYPHoyDgwN+fn74+/tz6tQpKebMmTNYWFgwZMgQ7O3t8fLyol+/fpw5c6bN6ye0H6mpqdJKNQ4ODgwaNAgbGxtp35MBAwZoLDPedK0NGjRIutb8/Pw0rrX+/fvTtWtXbG1tcXZ2ZsSIETg7O2vt2yMYlm+//ZYpU6YwefJkevTowd/+9jc6d+4s9ba/++67/Pvf/5bif/31V0pLS1myZAk9e/YkPDyc9957jzVr1mj12k2ZMoXdu3drrfYlGCYjIyMGDx7Mvn37SE9PJy8vj3Xr1mFqaqrxYHXNmjXs2LFDej1o0CAuXrzI/v37KSgoYP/+/Vy6dEljaWE/Pz/27t1LRkYGJSUlpKenc+jQIQICAtq0joLQnrT6cK+4uDiio6N1DsGYOHEiixYtoqSkBECanyKXy/H09GT48OEsX76cHj16SHlu3brF22+/TU5ODpaWlgQEBLBt2zYxt+QBXLx4EQsLC8LCwrCysqK4uJgtW7ZI3dBWVlYa3c8VFRVs3ryZyMhIgoODqaqq4uDBg9Lyw9DYZb1p0yYiIyN59tlnqa6uJi0tTWt1NsGwXLp0CQsLC/r166fzWpPL5RqTTysrK9m6dSuDBw8mKCiI6upqDh06pLH8sJmZGcOHD0cul1NXV0dRURG//PILBQUFbV4/of3YsmULDg4OzJ07F1dXVzIzM3nhhRfIyckBwM3NTWPosVKpZOrUqfz9739n27ZtlJaWsnXrVv71r39pnNfb25uwsDBpiVhBABg6dCi3b9/m119/paamhq5duzJz5kyNPVLKyso0ekW8vLx49tln2b17N7t378bR0ZHnnntOY6TChAkT2LVrF7/++itVVVXY2toSHh6ucwNb4eESw73ajlGDAfUl/uc//3nYRRAMhNghWGgr//jHPx52EQQDIb5DhbbyxBNPPOwiNCs+Pv5hF0Gney1M9SgSv6QEQRAEQRAEQWhX9LZPiiAIgiAIgiB0JGKkRNsRf9OCIAiCIAiCILQropEiCIIgCIIgCEK7YlDDvcrLyx92EQQD4ejo+LCLIBiIwsLCh10EwUAoFIqHXQRBeOjE6l5tR/SkCIIgCIIgCILQrohGiiAIgiAIgiAI7YpopAiCIAiCIAiC0K7odU5KUlISkZGRjBw5kp07d0rpWVlZeHt7S6+tra3x8PBg2LBhzJs3j549e2qcp66ujr///e/873//Iy8vjy5duvDee+/x0ksv6bP4HVLfvn2JiIjA2tqawsJCdu/ezY0bN3TGWltbEx0djbu7O46OjiQnJ7Nnzx6NmODgYMaPH6+Vd/HixdTX1+ulDsKjISAggNDQUORyOSUlJRw+fJjc3Nxm4zt16sTAgQNxdHSkurqa1NRUzp49qzO2R48ejBo1iitXrmjcWwTD9Morr/Dmm2+iUCg4d+4c77zzDocPH9YZ++233/L8889rpZ87d44+ffoAsHv3boYMGaIVs2PHDp588snWLbzwSGloaODXX39l//79VFdX0717d55//nm6dOnSbJ79+/dz+PBhsrOzgcYd6CdNmkT37t2lmC1btpCSkkJubi6mpqb07NmTZ555Bnd3d73XSRDaK702UlasWMGcOXP47rvvuH79Oh4eHhrvx8fHExAQgFKp5MyZM3z++eeEhISwZcsWoqKipLhnnnmG/Px84uLi6NGjBwUFBahUKn0WvUPy9/dn1KhR7Nixgxs3btCnTx+effZZvv76ayoqKrTijY2NUSqVJCYmEh4e3ux5a2tr+eqrrzTSRAPFsPXo0YPBgwdz8OBB8vLy8Pf3JyYmhp9++omqqiqteBsbG8aOHcu5c+dISEhAoVAwZMgQampquHLlikastbU1AwcO5ObNm21VHaEdmzRpEp9++ilz584lKSmJmTNnsmnTJkJDQ3U+gHnrrbd4//33pdcmJiYkJyezceNGKW3y5MmYmZlJrx0dHTl+/LhGjGCYtm/fzs6dO3n55ZdRKBRs3ryZf/3rX/zjH//A0tJSZ57z588zYMAAevTogampKdu3b+fTTz/l448/lhZZyczMJCoqCm9vb9RqNevXr+df//oXixcvxtzcvC2rKAjtht6Ge1VXV7N27Vpmz55NTEwMK1eu1IpxcnJCoVDQrVs3JkyYQHx8POHh4cTGxko/cnfu3MmBAwfYvn070dHReHl5ERYWxsCBA/VV9A4rPDyc1NRUUlNTKS4uZs+ePVRUVNC3b1+d8eXl5ezevZszZ85QV1fX4rmrq6s1DsGwhYSEkJGRQUZGBqWlpRw+fJiqqioCAwN1xgcEBFBVVcXhw4cpLS0lIyOD8+fP07t3b404IyMjRo4cyfHjx3U2rAXD8/rrr7Ny5Uq+//57MjMzeeedd8jOzuaVV17RGV9RUUF+fr509OnTBwcHB1atWiXFlJaWasRERUWhVCrZsGFDW1VLaIcaGhrYtWsX48ePp1+/fnTp0oWXX36ZW7ducfTo0WbzzZo1i6ioKDw9PenUqRMvvfQSarWac+fOSTFvv/02kZGRdOnSBQ8PD2bOnElxcTFXr15ti6oJv4GRkVG7PDoivTVS1qxZg4+PDz4+PkybNo3vv/+ehoaGlgsjkzF37lyuXbtGSkoKAJs3b6Zfv37885//pHPnzvTq1Yu3336bmpoafRW9Q5LJZLi7u2s9lb5y5UqL3dT3w8zMjDlz5vD6668zefJk3Nzcftf5hEebTCbDxcVF6yn2jRs3mr02FAqFVvz169dxcXHR2N23X79+1NTUkJGR0foFFx45pqam9OnTh/j4eI30+Ph4BgwYcF/nmDFjBnv37uX69estxqxbtw6lUvm7yis82goLCykvL9d42GJqaoqPjw8XL1687/PU1dVRX1+PtbV1szFNv3FaihGEjk5vjZS4uDimTZsGwJgxY6iqqiIhIeGe+Xx9fYHGeSvQ+CM6MTGR9PR0fvnlF5YuXcr69ev54x//2OJ56urqqKio0DgMeYiYXC5HJpNp9XJUV1f/rptgcXExmzdvZs2aNfzyyy+oVCpmzJiBg4PD7y2y8IiysLBAJpNpPUhQKpXI5XKdeeRyudYPwJqaGoyNjbGwsAAaGzJ+fn7s379fL+UWHj3Ozs6YmJhQUFCgkV5QUHBfD0sUCgWjR4/m+++/bzamX79+BAYGthgjGIamvdZsbW010m1tbX/TPmzr1q3DwcEBf39/ne83NDTw448/0qtXr9/9EFEQHmV6aaRkZmaSnJzMlClTgMYxv5MnT2bFihX3zNvU29LUdaVWqzEyMmL16tWEhYXx+OOP89lnn7Fy5coWe1MWL16MnZ2dxnHw4MFWqN2jTVdv1r16uFqSk5NDeno6BQUF3Lhxgw0bNlBcXEz//v1/TzGFDuDu6+pBu6MbGhowNTUlOjqa/fv3U1tb2xrFEzoQXdfa/dzXnn/+ecrKyti8eXOzMTNmzCA9PZ0TJ0787nIKj5akpCReeeUV6Wgahq7rXna/97dt27Zx9OhR5syZozHv6U4//PAD2dnZzJ49+8ELL+jNwx7WZUjDvfQycT4uLg6VSkXnzp2ltKYfGqWlpS3mbRrG0bT6l7u7O507d8bOzk6K8fPzo6GhgezsbK2VwJosWLCAN998UyPts88+e6D6dARKpRK1Wq3Va2JlZdXqc0hyc3PFjusGrLa2FrVardVrYmlp2exwGV29LJaWltTX11NXV4eDgwO2trY8/vjj0vtNN+VZs2bx448/ijkqBqioqAiVSqXVa+Li4qLVu6LLCy+8wI8//sjt27d1vm9pacnTTz/N3//+91Ypr/BoCQ0N1ViBq+k6KS8vx97eXkqvqKjQ6l3RZfv27WzdupU//elPWgsJNfnhhx84deoUf/7zn8X3qGDwWr2RolKpWLVqFUuWLGHUqFEa702cOJHVq1cTExOjM69arWbZsmV4e3sTGhoKwKBBg1i3bh1VVVXSD+wLFy4gk8la7AY1NzfXWhHDxESvi5m1a2q1mtzcXLy9vcnMzJTSvb29uXDhQqt+lpub2339QBA6JrVaTWFhIV27dtWY9NmlSxdpGOfd8vLy8PLy0kjr2rUrhYWFqNVqysrK+PnnnzXeDwsLw8zMjMTERJ0rhgkd3+3btzl58iRRUVEavSFRUVFs3bq1xbxDhgyhR48eOhd1aTJp0iTMzc356aefWqvIwiPE0tJSY8WuhoYG7OzsSE9Px9PTE2j8zZOZmckzzzzT4rm2b9/O5s2befvttzW2YLjz3D/88AMpKSksWLAAFxeX1q2MIDyCWv1X+9atWyktLSU2Nlaj9wMab/hxcXFSI6W4uJi8vDyUSiXp6eksXbqU5ORktm3bhrGxMQBTp07lww8/5MUXX2ThwoUUFRXxzjvv8NJLLzW73J+g27Fjx5gwYQK5ublkZ2fTp08f7OzsOHnyJADDhw/HxsZG48u+6QmlqakpVlZWuLm5UV9fT1FREQCRkZHk5ORQUlKCubk5/fv3x83NTexdYeDS0tKIioqioKCA/Px8/P39sbGxIT09HYABAwZgZWUlzVM7e/YsQUFBDBw4kIyMDNzc3PDz85P25amvr6ekpETjM27dugWglS4YlmXLlrFixQpOnjzJ0aNHiY2NpWvXrnz77bcAfPjhh3Tq1InY2FiNfDNmzODYsWMaKyzdbcaMGWzevFlcYwLQ2Hs7evRotm7dipubGwqFgi1btmBmZqaxUMM333yDg4OD1HDZtm0bGzduZNasWTg7O1NWVgY0zt9rmnO3atUqjh49yty5c7GwsJBi5HJ5s8PChIejow6tao9avZESFxdHdHS0VgMFGntSFi1aJN3wo6Ojgcb/hJ6engwfPpzly5fTo0cPKY+1tTV79uxhzpw59OvXDycnJ5555hk++uij1i56h3fu3DksLS2JjIyUNnP8+eefpQl/1tbWWv9uL7/8svTnTp06ERgYSFlZGV988QXQeJMdO3YsVlZW1NXVkZeXx6pVq8QeFgbu0qVLmJub069fP6ysrCguLmbr1q1Sj4dcLtcYelhZWcm2bdsYNGgQQUFBVFdXk5iYqLUanSDcbf369Tg6OvLnP/8ZhULB2bNneeKJJ6TVuhQKBV27dtXIY2tryxNPPMHbb7/d7Hl79OjBoEGDGDt2rF7LLzxaHn/8cW7dusWqVatQKpV069aNd955R+OhaUlJicaqhHv37kWlUknfm02eeOIJaXPQvXv3Ao3zae80c+ZMIiMj9VUdQWjXjBp+z6zpR4xo2AhtRYwlFtrK3XPvBEFfxMp6Qlu53yXEH4b2ugjTkCFDHnYRWp3hTtIQBEEQBEEQhN9ADPdqO3rbJ0UQBEEQBEEQBOFBiEaKIAiCIAiCIAjtimikCIIgCIIgCMJ9eNibNj6MzRxLS0t5/vnnpc3RmzbCbc7t27d59913CQoKwsrKik6dOjF9+vTfvKiSQc1JMTU1fdhFEAzE3RsTCoK+iOVJhbZSU1PzsIsgCMJDMHXqVLKzs6XtJV555RWef/55tmzZojNeqVRy8uRJ/vKXvxASEkJpaSnz5s1j/PjxnDhx4r4/16AaKYIgCIIgCIIg3J+MjAx27tzJ0aNHCQ8PB+Dbb78lIiKCzMxMfHx8tPLY2dlJ+5w1+c9//kNYWBjXr1/Hw8Pjvj5bNFIEQRAEQRAE4T6019W96urqqKur00gzNzfH3Nz8d533yJEj2NnZSQ0UaFwi2s7OjqSkJJ2NFF3Ky8sxMjLC3t7+vj9bzEkRBEEQBEEQhEfY4sWLpTkjTcfdm4M+iLy8PFxdXbXSXV1dycvLu69z1NbWMn/+fKZOnYqtre19f7Zee1KSkpKIjIxk5MiR0jg2gKysLLy9vaXX1tbWeHh4MGzYMObNm0fPnj2l92bMmMF///tfrXP7+/tz9uxZfRa/QwoNDSUsLAxra2uKiopISEggOztbZ6yVlRUjRozAzc0NR0dHUlJSSEhI0IhxdnZm8ODBKBQK7OzsSEhI+E3jDYWOrXfv3vTq1QszMzOKioo4evRoi5PtADw9PQkNDcXGxobKykpOnjwp7R7exMfHh8DAQORyOaWlpSQnJ1NQUKDHmgjt2cyZM3n99ddRKBRkZGQwf/58jhw5ojP2q6++4rnnntNKz8jIkJ4UTp06la+//lorxsXFRetJpWBYGhoa2Lp1K4mJiSiVSry8vHj22Wfp1KlTs3lu3rzJli1buHbtGiUlJTz99NNERUVpxZWWlvLLL79w9uxZbt26hZubG88//zyenp76rJLQQSxYsEBrc9+WelE++OADFi5c2OI5jx8/DujuPWpoaLivXqXbt28zZcoU1Go1X3755T3j76TXnpQVK1YwZ84cEhMTtX5kAMTHx5Obm0taWhqLFi0iIyODkJAQjR/Cn3/+Obm5udJx48YNHB0defrpp/VZ9A7J19eXqKgojhw5wsqVK8nOzubpp5/GxsZGZ7yxsTFKpZIjR440+wPQxMSEsrIyDhw4QFVVlT6LLzxiAgMD8ff35+jRo2zdupWamhpGjRqFiUnzz0ZcXFwYOnQoly9fZvPmzVy+fJlhw4bh7OwsxXh5eREWFsbp06fZvHkzBQUFjBw5Eisrq7aoltDOPPXUU/zjH//g008/ZfDgwRw5coQNGzbQpUsXnfHvvvsuPXr0kA5fX19KSkr49ddfNeLKy8s14nr06CEaKAK7d+8mISGBKVOmMH/+fOzs7Pj888+pra1tNs+tW7dwdnbmySefbPYpcnV1Nf/6178wNjbmtdde429/+xuTJk0Si7C0Qw97Fa/mDnNzc2xtbTWOlhopr732GhkZGS0egYGBKBQK8vPztfIXFhbi5ubW4t/V7du3eeaZZ7h69Sp79uz5Tb0ooMdGSnV1NWvXrmX27NnExMSwcuVKrRgnJycUCgXdunVjwoQJxMfHEx4eTmxsLPX19UDj5BuFQiEdJ06coLS0lBdffFFfRe+w+vfvz+nTpzl9+jTFxcUkJCRQWVlJaGiozviKigoSEhI4e/Zss1/OeXl57N+/n4yMDOnfTBCgsbfz9OnTXL9+nbKyMg4dOoSJiQndunVrMc/Nmzc5c+YM5eXlnDlzhtzcXPz9/aWYgIAALl68yMWLFykvLyc5OZnq6ur7HhcrdCyvvfYaq1atYtWqVVy4cIH58+eTk5NDbGyszviKigoKCgqkIzQ0FHt7e/73v/9pxDU0NGjEiZ46oaGhgYSEBB577DFCQ0Pp3LkzL7zwArdu3SI5ObnZfF5eXkycOJH+/fs3+5Bm9+7dODo68sILL+Dt7Y2zszO+vr64uLjoqzqCgWu6xlo6LCwsiIiIkL5rmxw7dozy8nIGDhzY7PmbGigXL14kPj4eJyen31xGvTVS1qxZg4+PDz4+PkybNo3vv/+ehoaGlgsjkzF37lyuXbtGSkqKzpi4uDiio6NF9+dvJJPJUCgUXL16VSP96tWrdO7c+SGVSuiorK2tkcvlGmuiq9XqZse2NnFxcdFaRz0nJ0fKI5PJcHJy0oq5efNmi+cVOiZTU1N69+7N3r17NdL37t2rMcmzJdOnT2f//v3cuHFDI93a2pr09HQyMjJYu3YtwcHBrVZu4dFUVFRERUUFfn5+UpqpqSk9e/bkypUrv+vcaWlpeHh4sHz5ct555x0+/vhjDh069HuLLAi/m5+fH2PGjOHll1/m6NGjHD16lJdffpmYmBiNh4O+vr788ssvAKhUKiZNmsSJEydYvXo19fX15OXlkZeXx61bt+77s/XWSImLi2PatGkAjBkzhqqqKq35DLr4+voCjfNW7pabm8uOHTuYOXPmPc9TV1dHRUWFxqFSqX5bJToQuVyOTCZDqVRqpFdXV4thMkKrs7S0BLT3VaipqZHeay5fS3nMzc2RyWS/+bxCx+Tk5ISJiYlWL0dBQcE9hyEAuLm5MXLkSK15jxcvXmT27NlMmTKFl156idraWnbv3k337t1btfzCo6WiogJAa8iKra2t9N6DKioq4uDBg7i6ujJnzhwiIyNZu3YtR48e/V3nFYTWsHr1aoKCghg1ahSjRo0iODiYH374QSMmMzOT8vJyALKzs9m8eTPZ2dn07t0bd3d36UhKSrrvz9XLxPnMzEySk5PZuHFj44eYmDB58mRWrFhBdHR0i3mbelt0TcZZuXIl9vb2PPHEE/csw+LFi7UmBEVFRTFy5Mj7rEXHdHdvVntdSk94tHTr1o2IiAjpdXx8vM64B7ne7iePuI6FOxkZGd2z5x7gueeeo7y8nK1bt2qkHz9+XJowCnD06FEOHTrEq6++yp/+9KdWL6/QPh07dowff/xRev3HP/4R0L7f3M+1di8NDQ14enpKv288PDzIzc3lwIEDDBgw4HefXxB+D0dHR60hsXe78/+Bl5dXq/y/0EsjJS4uDpVKpTGMqKGhAVNTU0pLS1vMm5GRAaCx+ldT/hUrVvD888/f1w7LulY5+M9//nO/VehwlEolarVaq9dELpdTXV39kEoldBTXr1+nsLBQem1sbAxo94xYWFi0uGu1rh6RO/PU1dWhVqtbjBEMR3FxMSqVSmuon4uLy33NIXn++ef5+eefuX37dotxDQ0NnDx5UvSkGJiQkBCN3yJNozHKy8uxs7OT0isrK3/zhOC72dnZ4e7urpGmUCg4efLk7zqvIDzKWn24l0qlYtWqVSxZsoTU1FTpSEtLw9PTk9WrVzebV61Ws2zZMry9vbUmcx84cIBLly41OxnybrpWOWhpVaGOrmk+gJeXl0a6l5cXOTk5D6dQQoehUqmorKyUjrKyMpRKpcaynE3zolr68VhYWKi1lGenTp2kPGq1muLi4hZjBMNx+/ZtUlNTGTFihEb68OHDOXbsWIt5Bw8eTPfu3Vm1atV9fVZwcPB97wkgdAwWFha4urpKh7u7O7a2ttLDVGi89128eLHFBUHuR/fu3bVWUMrPz3+gycaCfj3sVbyaOzqiVv/VvnXrVkpLS4mNjdV40gAwadIk4uLiiImJARqfguXl5aFUKklPT2fp0qUkJyezbds26Ulsk7i4OMLDwwkMDGztIhuM48ePExMTQ15eHjdv3iQkJARbW1tSU1MBGDJkCDY2Nmzbtk3K0/SE0tTUFEtLS1xdXamvr6e4uBho/OHZtDysTCbD2toaV1dXbt26dc/9MISO7dy5cwQHB0tzwoKDg1GpVBoTTAcPHoxSqZSeFp47d47HHnuMwMBAbty4QdeuXenUqRPbt2+X8pw9e5bIyEiKioooLCykV69eWFlZkZmZ2eZ1FB6+L774guXLl3Py5EmSk5N58cUX6dKlCytWrADgb3/7G506deLVV1/VyDd9+nSOHz+u8YOzyfz58zl+/DiXL1/GxsaGWbNmERQUxFtvvdUmdRLaJyMjI6Kioti5c6fUcNm5cydmZmaEhYVJcd9//z329vY8+eSTQGNDJjc3F4D6+nrKysq4ceMG5ubm0ndsVFQU//znP9mxYwd9+/YlKyuLxMREnXv6CIKhaPVGStPqW3c3UAAmTpzIokWLKCkpAZDmp8jlcjw9PRk+fDjLly+nR48eGvnKy8vZsGEDn3/+eWsX16CcP38eS0tLBg0ahJWVFUVFRaxbt06a8Gdtba3VZX3nUs/u7u4EBARQXl4ubXRmbW2tERMeHk54eDjXr1/np59+aoNaCe1Veno6JiYmDBgwAHNzcwoLC9m9e7fGAhbW1tYaeQoLCzlw4AB9+vQhNDSUyspK9u/fT1FRkRSTlZWFubk5vXv3xtLSktLSUuLj48WwRQO1ceNGHB0deffdd1EoFJw7d45JkyZJq3UpFAqtPVNsbW0ZP3487777rs5zNu194ebmRkVFBadPn+axxx5rdtVJwXCMGjWKW7du8dNPP6FUKvH29ub111/HwsJCiikpKdF4sl1WVsbHH38svd6zZw979uyhZ8+eUsPXy8uLWbNm8euvv7Jt2zacnZ15+umn73uVOkHoiIwaWmNmyyPik08+edhFEAzE/awsJAit4fXXX3/YRRAMxKZNmx52EQQDMXz48IddhGa11xXXOuICC3rdcV4QBEEQBEEQBOG3Eo0UQRAEQRAEQRDaFcNd7koQBEEQBEEQfoOOupJWeyR6UgRBEARBEARBaFcMqifFkPdJEdrW3evdC4K+3L2RoSDoi9iLSBCEtiR+tQuCIAiCIAjCfRDDvdqOGO4lCIIgCIIgCEK7IhopgiAIgiAIgiC0K3od7pWUlERkZCQjR45k586dUnpWVhbe3t7Sa2trazw8PBg2bBjz5s2jZ8+eGudZvXo1//znP7l48SJ2dnaMGTOGTz/9FCcnJ30Wv0MKCQmhf//+WFlZUVxczL59+8jJydEZa2VlxdChQ3Fzc8PBwYGTJ0+yf/9+rbiePXsyaNAg7OzsKC8vJzExkUuXLum5JkJ7FxoaSlhYGNbW1hQVFZGQkEB2drbOWCsrK0aMGIGbmxuOjo6kpKSQkJCgERMSEkJAQAAuLi4A5OXlcfDgQXJzc/VeF6F9e+6555g5cyaurq5cvHiRjz76iBMnTjQbb2ZmxmuvvcaECRNwcXEhLy+PL7/8kvXr1wONu4rPnj0bT09PTExMyMrKYsWKFfz6669tVCOhPQsICKB79+6YmppSUlJCSkoKFRUVLebp0qULgYGBWFtbU1VVxZkzZzS+ewMCAggMDNTIU1NTw+bNm/VSB+HBieFebUevPSkrVqxgzpw5JCYmcv36da334+Pjyc3NJS0tjUWLFpGRkUFISIjGj5PExESmT59ObGwsZ8+eZd26dRw/fpyZM2fqs+gdko+PD8OHD+fYsWP88MMPZGdn89RTT2FjY6Mz3tjYmJqaGo4dO0ZhYaHOGHd3d2JiYjh37hw//PAD586dIyYmBoVCoc+qCO2cr68vUVFRHDlyhJUrV5Kdnc3TTz/d4rWmVCo5cuRIs5Nzu3btSkZGBj/99BM//PADFRUVPPPMM1hbW+uzKkI79/jjj/Pee+/x1VdfMX78eI4fP05cXBzu7u7N5lm2bBkDBw5kwYIFjBw5knnz5nH58mXp/fLycr788kuefvppYmJi2LBhA//4xz+IjIxsiyoJ7Zivry8+Pj6kpKQQHx9PbW0tw4YNa3FhHicnJyIiIrh27Rq7du3i2rVrDBw4EEdHR4248vJyNm3aJB27du3Sd3UEoV3TWyOlurqatWvXMnv2bGJiYli5cqVWjJOTEwqFgm7dujFhwgTi4+MJDw8nNjaW+vp6AI4ePYqXlxevv/463t7eDB48mFdffbXFp2SCbn379uXMmTOcOXOGkpIS9u/fT2VlJSEhITrjKyoq2LdvH+fOnaOurq7Zc167do3k5GRKSkpITk7m+vXr9O3bV59VEdq5/v37c/r0aU6fPk1xcTEJCQlUVlYSGhqqM76iooKEhATOnj3b7LW2detWTp06RUFBASUlJezcuRMjIyM8PT31WRWhnXvppZdYt24da9eu5fLly3z88cfk5uby3HPP6YwfMmQIYWFhxMbGkpSURE5ODqdPn+bUqVNSzLFjx9izZw+XL1/m+vXr/Pe//yUzM1Pc1wR69erFuXPnyMnJoby8nGPHjmFsbNzifahXr17k5+eTkZFBZWUlGRkZ5Ofn06tXL404tVpNbW2tdDR3LxQEQ6G3RsqaNWvw8fHBx8eHadOm8f3339PQ0NByYWQy5s6dy7Vr10hJSQFg4MCBZGdns337dhoaGsjPz2f9+vWMHTtWX0XvkGQyGW5ubly7dk0j/dq1a3Tq1OmBz+vu7k5WVpZGWlZW1u86p/Bok8lkKBQKrl69qpF+9epVOnfu3GqfY2pqikwmo7a2ttXOKTxaTE1NCQwMJDExUSM9MTGRPn366MwTFRXFmTNneOWVV0hMTGTPnj3Mnz8fc3PzZj8nIiICb29vjh8/3qrlFx4tVlZWWFpakpeXJ6Wp1WoKCwtbHH7u5OSkkQcah6s6OztrpNnY2DB+/HjGjh1LREQEVlZWrVsBoVUYGRm1y6Mj0lsjJS4ujmnTpgEwZswYqqqqtMaY6+Lr6wsg/fAdOHAgq1evZvLkyZiZmaFQKLC3t+c///mPvoreIVlaWiKTyVAqlRrp1dXVv+tGaGVlpXVOpVKJXC5/4HMKjza5XK6Xa+1uQ4cOpaqqSquRLBgOBwcHTExMKCoq0kgvLi7W+gHYpGvXrvTr14+ePXvyhz/8gY8++ogxY8bwwQcfaMRZW1uTlpZGRkYG3333HX//+985fPiwvqoiPAIsLCwAtB6M1NbWSu81l+9eeYqLizl27BgHDhzgxIkTWFhYEBUVhZmZWSvWQBAeLXpppGRmZpKcnMyUKVOAxk0UJ0+ezIoVK+6Zt6m3palVeO7cOV5//XX++te/kpKSws6dO7l69SqzZs1q8Tx1dXVUVFRoHCqV6nfW7NF3d29WR219Cw+fPq+1sLAw/Pz8+OWXX6ShoYJwp+Z67mUyGQ0NDbz55pucPn2aAwcOsGjRIiZOnKjRm1JdXc348eN56qmn+Oyzz/jzn/9MeHh4WxVfaAc8PT156qmnpEMma72fTHffD/Py8sjOzqa8vJz8/HwOHjwIgJeXV6t9piA8avSyuldcXBwqlUpjaEdDQwOmpqaUlpa2mDcjIwNAWv1r8eLFDBo0iHfeeQeA4OBgrKysiIyM5KOPPmp2cuTixYtZuHChRtrIkSMZPXr0A9frUVZTU4NardZ6ki2Xy6murn7g81ZXV2v1msjlcq2n6ILhUCqVernWmoSFhREREcGaNWuaXdBBMAylpaWoVCqtXhMnJyeKi4t15ikoKCA/P5+qqiop7fLly9IwxaYhsQ0NDdKfMzIy6N69O7NmzeLYsWN6qo3Q3uTk5GhcR02NlLt7RnT1lNxJV0+Lubl5i3nq6+spLy9vdrER4eERD3fbTqv3pKhUKlatWsWSJUtITU2VjrS0NDw9PVm9enWzedVqNcuWLcPb21uaYKtUKrWeXhgbGwPNPykDWLBgAeXl5RpHVFRUK9Tw0aRWq8nPz9ea3Ofp6cnNmzcf+Ly5ubla5/Ty8vpd5xQebWq1mry8PK0ngF5eXs0ud32/wsLCGDhwIOvWrdMa4y0Yntu3b5Oens7gwYM10gcPHszJkyd15klJScHV1VXj4Yq3tzf19fUtXlNGRkZi6I2BUalUVFVVSUdFRQU1NTUaq1fKZDJcXFyabRRD41AuNzc3jTSFQqE1TPFOMpkMW1tbampqfn9FBOER1eqNlK1bt1JaWkpsbCyBgYEax6RJk4iLi5Nii4uLycvL48qVK2zevJno6GiSk5OJi4uTGiLjxo1j48aNfPXVV1y5coXDhw/z+uuvExYW1uLkbHNzc2xtbTWOlpYINAQpKSkEBQURGBiIo6Mjw4YNw8bGhrS0NKDxi33MmDEaeVxcXHBxccHU1BS5XI6Li4vGsoknT57Ey8uL/v374+joSP/+/fHw8JAWPhAM0/HjxwkJCSEoKAgnJydGjBiBra0tqampQOMKS3cvfuHq6oqrqyumpqZYWlri6uqqMRk1LCyMyMhItm/fTnl5OVZWVlhZWWFqatqWVRPamRUrVvD0008zadIkunfvznvvvYe7uzs//vgjAG+//Tb/+te/pPgtW7ZQVlbGJ598Qo8ePejfvz/vvvsu69evl1ZTmjVrFoMGDaJr165069aNl156iSeeeIJNmzY9lDoK7ceFCxfw8/Ojc+fO2NnZERYWRn19vcaiNOHh4QQFBWnkUSgU+Pr6YmNjg6+vL25ubly4cEGKCQkJwcXFBSsrKxwdHRk4cCCmpqZizp1g0Fr9V3tcXBzR0dHY2dlpvTdx4kQWLVpESUkJANHR0UDjMBBPT0+GDx/O8uXL6dGjh5RnxowZVFZW8sUXX/DWW29hb2/PiBEj+OSTT1q76B1eZmYmFhYWDBgwQNrMcePGjVRWVgKNk+BtbW018kyfPl36s0KhwM/Pj/Lycr777jsAbt68ydatWxk8eDCDBg2irKyMrVu3iqfcBu78+fNYWloyaNAgrKysKCoqYt26ddKGZ9bW1lrX2osvvij92d3dnYCAAMrLy/n6668B6NOnDyYmJjz55JMa+RITE8WEZgO2fft2HBwceO2113B1deXChQvMnDlT6s11cXHReKClVCp54YUX+Otf/8ovv/xCWVkZ27dv57PPPpNiLC0tWbhwIQqFgtraWq5cucJbb73F9u3b27x+Qvty/vx5jI2N6du3L2ZmZhQXF3PgwAGNOa9yuVxjpEdxcTFHjhyRHhJWV1eTlJQk/RZqyhMREYGZmRl1dXUUFxcTHx8vhk4LBs2o4V7rAncgS5YsedhFEAyEWKRBaCvffvvtwy6CYCA+/vjjh10EwUBMnjz5YRehWe11pEhH3MdJrzvOC4IgCIIgCIIg/FaGPUlDEARBEARBEO6TWN2r7YieFEEQBEEQBEEQ2hXRSBEEQRAEQRAEoV0xqOFed69TLgj6cvXq1YddBMFADBgw4GEXQTAQ1tbWD7sIgvDQieFebUf0pAiCIAiCIAiC0K6IRoogCIIgCIIgCO2KQQ33EgRBEARBEIQHJYZ7tR3RkyIIgiAIgiAIQrui156UpKQkIiMjGTlyJDt37pTSs7Ky8Pb2ll5bW1vj4eHBsGHDmDdvHj179tQ4z//93//xxRdfkJWVhYeHB++99x7Tp0/XZ9E7tODgYHr06IGZmRnFxcUkJydTXl7eYp6uXbsSEhKCjY0NlZWVpKWlcePGDen9nj170qtXL6ysrAAoLy/nzJkz3Lx5U691Edqvvn37EhERgY2NDYWFhezatUvjmrmTtbU1I0eOxN3dHUdHR5KTk9m9e7dGTHBwMBMmTNDKu2jRIurr6/VSB+HREB0dzeOPP469vT05OTn873//IzMzU2dsr169mDJlCu7u7pibm1NUVMTevXs1vqMARo8eTXR0NE5OTlRWVpKcnMzatWu5fft2W1RJaKcaGhrYtWsXR48eRalU4unpycSJE1EoFM3mycvLY8eOHWRnZ1NaWsqECRMYOnSoRszly5fZt28f2dnZVFRU8OKLLxIUFKTv6ghCu6bXRsqKFSuYM2cO3333HdevX8fDw0Pj/fj4eAICAlAqlZw5c4bPP/+ckJAQtmzZQlRUFABfffUVCxYs4Ntvv6V///4kJyfz8ssv4+DgwLhx4/RZ/A7J398fX19fjhw5QkVFBUFBQURFRbF582ZUKpXOPM7OzkRGRkoNk65duxIZGcmuXbsoLi4GQKlUcurUKSorKwHo1q0bQ4cOZfv27fdsAAkdj7+/P6NHj2b79u1kZ2fTp08fpk6dyldffUVFRYVWvLGxMdXV1SQmJhIeHt7seWtra/nyyy810kQDxbCFh4czbdo0Vq5cyYULFxgxYgTvvPMO7777rnR/ulNdXR179uzh+vXr1NXV4ePjw4svvkhdXR379u0DYODAgUyePJlvv/2WixcvolAoePXVVwFYvXp1m9ZPaF/27t3LgQMHePbZZ3FxcWHPnj18/fXXzJ8/HwsLC515bt26hZOTE7179+bXX39tNqZTp06EhYWxcuVK/VVA+N3EcK+2o7fhXtXV1axdu5bZs2cTExOj8z+dk5MTCoWCbt26MWHCBOLj4wkPDyc2Nlb64fHDDz/w6quvMnnyZLp168aUKVOIjY3lk08+0VfROzQ/Pz/S09O5ceMG5eXlJCUlYWJiotGzdTdfX19yc3M5e/YsFRUVnD17lry8PPz8/KSYnJwcbt68SWVlpdTTolKpcHZ2botqCe3MgAEDOHXqFKmpqRQVFbF7924qKiro16+fzvjy8nJ2797N6dOnqa2tbfHc1dXVGodg2B577DH279/P/v37uXnzJv/73/8oLi6WHnTd7dq1axw5coScnByKioo4fPgwZ86cwcfHR4rp0aMHFy9e5MiRIxQVFZGens6RI0fo1q1bW1VLaIcaGho4ePAg0dHRBAcH4+7uztSpU7l16xYnT55sNp+Hhwfjx48nNDQUExPdz4b9/Px4/PHHCQ4O1lfxBeGRo7dGypo1a/Dx8cHHx4dp06bx/fff09DQ0HJhZDLmzp3LtWvXSElJARqfet39dMLS0pLk5GTR7f4bWVtbY2lpSW5urpSmVqvJz89vsTHh4uKikQfg5s2bzeYxMjLC09MTExMTioqKWqfwwiNDJpPh7u7OlStXNNIvX75Mly5dfte5zczMmDNnDnPnzmXy5MktDrEQOj5jY2O8vb1JT0/XSE9PT9caNtwcT09Pevbsyfnz56W0Cxcu4OXlJTVKXFxcCAkJITU1tdXKLjx6SkpKqKys1GjQmpiY0L17d7Kysh5ewQShg9LbcK+4uDimTZsGwJgxY6iqqiIhIYHo6OgW8/n6+gKN81bCwsIYPXo03333HU888QR9+vQhJSWFFStWcPv2bYqKinB3d9d5nrq6Ourq6jTSbt++jampaSvU7tHU1Ni7+0l1bW2tNJekuXy68lhaWmqk2dvbM3r0aIyNjVGpVBw4cEAM9TJAcrkcmUym1ctRXV39uzaDKy4uZvPmzRQUFGBmZkZ4eDgzZsxg+fLllJSU/N5iC48gGxsbjI2Nte4z5eXl2Nvbt5h32bJlUv6NGzeyf/9+6b2jR49iY2PDX//6V6Dxh2h8fDxbtmxp7SoIj5Cmoao2NjYa6TY2NpSWlj6MIgkPgRju1Xb00kjJzMwkOTmZjRs3Nn6IiQmTJ09mxYoV92ykNPW2NF0Ef/nLX8jLy2PAgAE0NDTg5ubGjBkz+Oc//4mxsXGz51m8eDELFy7USHvyySd56qmnfk/VHileXl4a4/ubxlvrS0VFBdu2bcPMzAwPDw8GDhzInj17REPFQN3dc2pkZHTP3tSW5OTkkJOTI72+ceMGL7/8Mv3792fXrl0PfF7h0afrurrXtfbhhx9ibm5Ojx49mDx5Mvn5+Rw5cgRoHHozYcIEVq5cyaVLl1AoFEybNo0nnnii2TkFQseTkpLCunXrpNczZ84EtH+k/p77miAIzdNLIyUuLg6VSkXnzp2ltIaGBkxNTe/5tCEjIwNAmiNhaWnJihUr+Oabb8jPz8fd3Z3ly5djY2PT4hClBQsW8Oabb2qkbdiw4UGr9EjKzs7WGG7V1KizsLCgpqZGSr/79d1qa2u1htzpyqNWq6mqqgIau8WdnJzw9fXl2LFjv7suwqNDqVSiVqu1ek3kcnmrzyG5efMmjo6OrXpO4dFRWVlJfX29Vq+JnZ3dPR+OFBYWAo33STs7O5566impkTJp0iQOHz4s9a5kZ2djbm7OSy+9xKZNm8SPUgMREBCgseBP01zZiooKbG1tpfSqqiqt3hVBEH6/Vp+TolKpWLVqFUuWLCE1NVU60tLS8PT0bHFlFLVazbJly/D29iY0NFTjPVNTU7p06YKxsTE///wzMTExyGTNF9/c3BxbW1uNw9CGeqlUKqqqqqSjvLycmpoajSFyMpkMNze3FueOFBYWag2rc3d3v6/5Ji39Gwkdk1qtJjc3V2uScbdu3cjOzm7Vz1IoFFLDWDA89fX1XL16lcDAQI30wMBALl68eN/nMTIy0pjQbGZmhlqt1ohRq9VimIeBsbCwwMXFRTrc3NywsbHhwoULUoxKpeLy5ct4eXk9vIIKbcrIyKhdHh1Rq/ekbN26ldLSUmJjY7Gzs9N4b9KkScTFxRETEwM0jjHPy8tDqVSSnp7O0qVLSU5OZtu2bdJT/wsXLpCcnEx4eDilpaV89tlnpKen89///re1i24QMjIyCAwMpLKykoqKCgIDA1GpVFy9elWKGThwIEqlUpokev78eUaNGoW/vz/Z2dl06dIFd3d3jSE2vXv3JicnB6VSiampKZ6enri5ubF37962rqLQDhw9epQnnniCmzdvkpOTQ2hoKHZ2dtKCGCNGjMDGxoZNmzZJedzc3IDGH4hyuRw3Nzfq6+ulxvCQIUPIzs6mpKQEc3NzwsLCcHNzY8eOHW1fQaHd2LFjB7Nnz+bKlStcunSJ4cOH4+TkREJCAgDPPPMMDg4OfPPNN0DjnirFxcXSHk4+Pj48/vjjGvvynDp1iscee4xr165x+fJl3NzcmDRpEidPnhS9KAbMyMiIIUOGEB8fj7OzMy4uLsTHx2NmZkafPn2kuB9//BFbW1vpt45KpSI/Px9obFiXl5eTk5ODmZkZLi4uQOM82jsf/JWUlJCTk4NcLsfBwaENaykI7UerN1Li4uKIjo7WaqAATJw4kUWLFkmTXJvmp8jlcjw9PRk+fDjLly+nR48eUp76+nqWLFlCZmYmpqamDB8+nKSkJPHU4gGdO3cOExMTwsLCMDMzo6ioiISEBI09UqysrDS+iIuKikhMTCQkJISQkBCqqqo4dOiQxh4EFhYWDBo0CEtLS27fvk1paSl79+4lLy+vTesntA/nzp3D0tKSIUOGYG1tTWFhIT/99JM0BMfa2lpjuATAK6+8Iv25U6dOBAUFUVZWxn/+8x+gsXd07NixWFtbU1dXR15eHv/973/FhqEG7tixY9jY2PDkk09ib29PdnY2//rXv6T7k729vcbQYCMjI5555hlcXFxQq9UUFBSwZs0ajQcqv/76Kw0NDTz99NM4ODhQUVHBqVOnNOYnCIZpxIgR3L59mw0bNlBTU4OHhwevvvqqxpDo0tJSjSfbFRUVLFmyRHrdtGR29+7d+eMf/wg0zrG7cw+opgc4/fv359lnn9V3tQShXTJqMKDHQv/73/8edhEEA3Fnz5Qg6FNzO6sLQmsTP5aFtjJ27NiHXYRm3b3keXtx97DXjkBMGBAEQRAEQRAEoV0RjRRBEARBEARBENoVvW3mKAiCIAiCIAgdSUddSas9Ej0pgiAIgiAIgiC0KwbVkyJWARLaSqdOnR52EQQDsXLlyoddBMFAREVFPewiCIJgQERPiiAIgiAIgiAI7YpopAiCIAiCIAiC0K6IRoogCIIgCIIgCO3KA89JSUpKIjIykpEjR7Jz504pPSsrC29vb+m1tbU1Hh4eDBs2jHnz5tGzZ0/pvdzcXN566y1SUlK4ePEir7/+OkuXLtX6rA0bNvCXv/yFy5cv0717dz7++GOefPLJBy26QevduzdhYWFYW1tTVFTE3r17yc7O1hlrZWXF8OHDUSgUODg4kJKSorErM4CTkxODBw9GoVBgZ2dHQkICKSkpbVEVoZ3z8/MjKCgIS0tLysrKOHr0KPn5+c3GKxQKwsPDsbe3R6lUcubMGc6fPy+9b29vT58+fXB2dsbGxoajR49y9uzZtqiK0M4999xzvPzyy7i6unLx4kU+/PBDTpw40Wy8mZkZc+bMYcKECTg7O5OXl8eXX37J+vXrtWJjYmL4/PPP2bNnD7NmzdJnNYRHgK+vL4GBgdJ9LTk5ucX7mpubG2FhYdjb21NTU8OZM2e0NmD19PSkT58+2NjYUFlZSUpKCtevX9d3VYQHJFb3ajsP3JOyYsUK5syZQ2Jios7/TPHx8eTm5pKWlsaiRYvIyMggJCSEhIQEKaaurg4XFxfee+89QkJCdH7OkSNHmDx5Ms8//zxpaWk8//zzPPPMMxw7duxBi26wfH19iYqK4ujRo6xcuZLs7GwmTZqEjY2NznhjY2OUSiVHjhyhoKBAZ4ypqSnl5eUcOHCAqqoqfRZfeIR4e3sTHh5Oamoqv/76K3l5eYwePRorKyud8dbW1owaNYq8vDx+/fVX0tLSGDBgAF5eXlKMiYkJlZWVnDhxAqVS2UY1Edq7sWPH8v777/Pll18ybtw4jh8/zooVK3B3d282z7Jly4iIiGD+/PmMHDmSefPmceXKFa24Tp06MX/+fJKTk/VZBeER4e3tTVhYGGlpaWzevJn8/HxGjhzZ4n1t5MiR5Ofns3nzZtLS0ggPD8fT01OKcXFxYdiwYVy6dIlNmzZx6dIlhg8fjrOzc1tVSxDarQdqpFRXV7N27Vpmz55NTEyMztVlnJycUCgUdOvWjQkTJhAfH094eDixsbHU19cD4OXlxeeff8706dOxs7PT+VlLly5l5MiRLFiwAF9fXxYsWEBUVJTOHhehZf369eP06dOcPn2akpIS9u7dS2VlJaGhoTrjKyoq2Lt3L2fPnqWurk5nTF5eHvv37+f8+fPSv6sgBAYGcuHCBS5cuEB5eTnHjh2juroaPz8/nfF+fn5UV1dz7NgxysvLpbxBQUFSTFFREcePH+fKlSviWhMkL730EuvWrWPt2rVcvnyZjz76iNzcXJ577jmd8UOGDJG+i5KSksjJyeH06dOcPHlSI04mk/Hvf/+bzz//nBs3brRFVYR2LiAggIsXL3Lx4kXKy8tJTk6muroaX19fnfG+vr5UV1eTnJxMeXm5lDcwMFDjnDdv3uTMmTOUl5dz5swZbt68SUBAQFtVSxDarQdqpKxZswYfHx98fHyYNm0a33//PQ0NDS1/kEzG3LlzuXbt2m8aDnTkyBFGjRqlkTZ69GiSkpIepOgGSyaToVAoyMrK0ki/evUqnTt3fjiFEjokmUyGs7MzOTk5Guk5OTm4urrqzOPq6qoz3tnZWXStC80yNTUlMDCQxMREjfTExET69OmjM09UVBRnzpzhlVde4fDhw8THx7NgwQLMzc014ubMmUNJSQnr1q3TW/mFR4dMJsPJyUnrPnXz5s1m72suLi5aWx/cfV9zcXH5TfdK4eEzMjJql0dH9EBzUuLi4pg2bRoAY8aMoaqqioSEBKKjo1vM1/S0ISsri7CwsPv6rLy8PNzc3DTS3NzcyMvLazFfXV2d1tN/lUqFiYlBbQ0jkcvlyGQyqqurNdKVSmWzXdWC8CAsLCyQyWTU1NRopNfU1GBpaakzj6Wlpc54mUyGhYWF1nuCAODg4ICJiQlFRUUa6UVFRbi4uOjM4+HhQb9+/airq2P27Nk4OjqycOFC7OzsmD9/PgB9+/bl6aefZty4cXqvg/BoMDc3RyaTUVtbq5H+W+9rtbW1Gvc1S0tLrXPW1tY2e05BMCS/uSclMzOT5ORkpkyZAjSOE588eTIrVqy4Z96m3pbf2uK7O76hoeGe51i8eDF2dnYax759+37T5xqKe/WCCUJbENeh8KDuvnaMjIyavZ6a3nvjjTc4ffo0+/fv5+OPP2bixImYm5tjZWXFkiVLeO+99ygtLW2L4guPEF3XVUv3rvu5r/3WcwqCofjN3QpxcXGoVCqNIUINDQ2Ympre84aekZEBoLH6170oFAqtXpOCggKt3pW7LViwgDfffFMj7Ysvvrjvz+1olEolarVaq9dELpeLSchCq6qtrUWtVms9CdT1VLFJTU0NcrlcK16tVms9ZRSEJqWlpahUKq1eEycnJ63elSaFhYXk5+drLPRx+fJlZDIZ7u7uWFpa0rVrV5YvXy69L5M1Ps/LzMxk5MiRYuUlA1RXV9fsfa25e5Su+5qFhYXGfU1XT4yFhYW477VjHXVoVXv0m3pSVCoVq1atYsmSJaSmpkpHWloanp6erF69utm8arWaZcuW4e3t3exEbV0iIiLYs2ePRtru3bsZOHBgi/nMzc2xtbXVOAx1qBc0/v3n5eVprJYEjYsX3D0eVhB+D7VaTVFRkdZcp06dOjW7SlxBQQGdOnXSSOvcuTNFRUXiiaLQrNu3b5Oens6gQYM00gcNGqQ1Eb5JSkoKrq6uGj8evb29qa+vJzc3l8uXL/PYY48xbtw46UhISODo0aOMGzeO3NxcvdZJaJ/UajXFxcVa96mW7muFhYX3vK8VFhZq3Ss7d+7c7DkFwZD8pl/tW7dupbS0lNjYWK3VuCZNmkRcXBwxMTEAFBcXk5eXh1KpJD09naVLl5KcnMy2bdswNjaW8qWmpgJQVVVFYWEhqampmJmZ4e/vD8DcuXMZMmQIn3zyCRMmTGDTpk3Ex8drTZQU7u3EiROMHTuWvLw8cnJy6N27N7a2ttK/wZAhQ7C2tmb79u1SnqbJe2ZmZsjlclxdXamvr6e4uBj4/5OkoXHJYhsbG1xdXbl16xZlZWVtWj+h/UhPT2fo0KEUFhZSUFCAr68v1tbW0r4n/fr1Qy6Xc/DgQaCxl9XPz4/w8HDOnz+Pq6srvXr1Yv/+/dI5ZTIZ9vb20p/lcjmOjo7cvn2bysrKtq6i0E6sWLGCTz/9lDNnznDq1CmmTJlCp06d+PHHHwF4++23USgUvP322wBs3ryZ1157jU8++YTPP/8cBwcH5s+fz/r166V5jBcuXND4jIqKCp3pgmE5e/YskZGRFBcXU1BQgI+PD1ZWVtJ9rW/fvsjlcg4dOgTA+fPn8fX1pX///ly4cAFXV1d69uzJgQMHpHOeO3eOxx57jKCgIK5fv46HhwedOnVi27ZtD6WOgtCe/KZGSlxcHNHR0TqXC544cSKLFi2ipKQEQJpEL5fL8fT0ZPjw4SxfvpwePXpo5LuzVyUlJYUff/wRT09PaRWqgQMH8vPPP/P+++/zl7/8he7du7NmzRrCw8N/U0WFxhumhYUFAwcOxMrKiqKiItavXy99AVtZWWFra6uRZ8aMGdKfFQoF/v7+lJeX88033wCN68DfGRMWFkZYWBjXr1/n559/1nudhPbp6tWrWFhYEBoailwup7S0lN27d0tDbCwtLbG2tpbiq6qq2L17N+Hh4fj5+aFUKjl69KjGanRyuVxjE9fg4GCCg4PJzc3VaFgLhmXbtm3Y29szZ84cXFxcuHjxIrGxsdKqSq6urhp7piiVSqZPn87f/vY3fv31V8rKyti2bRufffbZw6qC8Ii4evUq5ubmhISESPe1PXv2SAvSWFpaagyprqqqYs+ePRr3tWPHjnHt2jUppqCggP3799OnTx9CQ0OprKxk//79zQ5XFB4+Mdyr7Rg1GNBYin/+858PuwiCgXBycnrYRRAMxKJFix52EQQD8f777z/sIggG4sUXX3zYRWhWZmbmwy6CTj4+Pg+7CK3ugXecFwRBEARBEARB0AfDnUkuCIIgCIIgCL+BGO7VdkRPiiAIgiAIgiAI7YpopAiCIAiCIAiC0K4Y1HCvuzdMEgR9EctHCm3lT3/608MugmAg2uuEYUFoS2K4V9sRPSmCIAiCIAiCILQropEiCIIgCIIgCEK7IhopgiAIgiAIgiC0K60+J2XGjBn897//bTy5iQmOjo4EBwfz7LPPMmPGDGSy/98uSkpK4qOPPuLIkSPU1NTQs2dPZsyYwbx58zA2NpbiPv74Y7Zt20ZqaipmZmaUlZW1drENRlBQEKGhoVhZWVFSUsKhQ4eknZl16dSpE5GRkTg6OlJdXc3JkydJT0/XiDEzMyMiIoLu3btjbm5ORUUFiYmJGrvqCoZnzJgxTJgwAQcHB27cuMGKFSvIyMi4Zz5fX18+/PBDrl+/zltvvSWlh4eHM3HiRNzd3TE2NiY3N5fNmzdz4MABfVZDeAT4+/tr7AKelJREXl5es/Hu7u5ERETg4OCAUqkkNTVV49r09vYmNDQUW1tbZDIZ5eXlnD59mosXL7ZFdYR2bMCAAQwZMgQbGxvy8/PZunUrWVlZOmMDAgIYMGAA7u7umJiYkJ+fT3x8vMZ15OrqyqhRo+jcuTMODg5s2bKFw4cPt1FtBKF900tPypgxY8jNzSUrK4sdO3YwfPhw5s6dS0xMDCqVCoBffvmFoUOH0qVLF/bt28f58+eZO3cuH3/8MVOmTKGhoUE6361bt3j66aeZPXu2PoprMHr27ElkZCQnTpzg559/5ubNm4wbNw5ra2ud8ba2towfP56bN2/y888/c+LECYYMGUL37t2lGJlMxhNPPIGtrS07duzgf//7H3v37qW6urqtqiW0Q4MGDeLFF19kw4YNvPXWW2RkZPD+++/j7OzcYj65XM7rr7/O6dOntd6rqqpiw4YNzJ8/nzfeeIO9e/fy2muv0bt3bz3VQngUdO/enYEDB3Lq1Ck2bNhAXl4ejz/+eLP3NRsbGx577DHy8vLYsGEDp06dYtCgQXh7e0sxtbW1nDx5kl9//ZX169eTmZnJsGHD6NKlS1tVS2iHgoODiYmJYd++fSxbtoysrCxefPFF7OzsdMZ7e3tz8eJFVq5cyX/+8x+uXLnCCy+8QKdOnaQYMzMziouL2bFjBxUVFW1VFUF4JOhldS9zc3MUCgUAnTt3pk+fPgwYMICoqChWrlzJs88+y8svv8z48eNZvny5lG/mzJm4ubkxfvx41q5dy+TJkwFYuHAhACtXrtRHcQ1G7969OXfuHOfOnQPg0KFDeHh4EBQUxJEjR7TiAwMDqays5NChQwCUlpbi6upKaGgoly9fBhqfYFpYWLB+/XrUajUAlZWVbVQjob0aN24cCQkJxMfHA7BixQp69+7N6NGjWb16dbP5Zs2axaFDh1Cr1YSFhWm8d/bsWY3X27ZtY/jw4fj5+ZGamtrqdRAeDUFBQZw/f57z588DjT30Xbp0wd/fn+TkZK14f39/qqqqSEpKAqCsrAwXFxdCQkK4evUqALm5uRp50tPT6dWrFwqFguzsbD3XSGivBg8ezIkTJzh+/DgAW7dupVevXgwYMIBdu3ZpxW/dulXj9a5du/D398fPz08awZCdnS1dU4899pieayC0BrG6V9tpszkpI0aMICQkhI0bN7J7926Ki4t5++23teLGjRtHr169+Omnn9qqaAZBJpPh6urK9evXNdKvX7+Ou7u7zjwKhUJnvKurqzRsz9vbm9zcXIYOHUpsbCxTp06lX79+4j+xATMxMaF79+6kpaVppKempuLr69tsvhEjRqBQKFizZs19fU5QUBCdOnWSGt2C4ZHJZLi4uGg1HLKzs3Fzc9OZx83NTSv+xo0bODs7awxHvlPnzp2xt7fXarwIhsPY2JjOnTtrDfm7ePEinp6e93UOIyMjzM3NUSqV+iiiIHQ4bbpPiq+vL6dPn+bChQsA+Pn5NRvXFPOg6urqqKur00i7ffs2pqamv+u8jypLS0tkMpnWzbGmpga5XK4zj1wup6amRiNNqVRibGyMhYUFSqUSOzs7unTpQmZmJps3b8be3p6hQ4diZGQkPW0SDIuNjQ3GxsZac8fKy8uxt7fXmcfd3Z1p06bx3nvvST1yusjlcr799ltMTU1Rq9UsX75cqzEkGA4LCwtkMpnWfaql+5qlpaXO+Dvva9A4DGfatGnIZDIaGhpITEwkJydHPxUR2j25XI6xsbHWSIHKykp69ep1X+eIjIzE1NRU53BWQRC0tWkjpaGhQeMJ+53zTlqKexCLFy+Whok1GTNmDI8//vjvOq+hufvfSNe/S01NDfv27aOhoYHCwkKsrKzo06ePaKQYOF3/v3WlyWQy3njjDX7++ed7PqmuqanhrbfewsLCguDgYF588UXy8/O1hoIJQnPfLy29d/dcyPXr12Nqakrnzp2JiIigoqJC9KYIGoyMjFq81pqEhIQQHR3NqlWrxJzNR5wYKdJ22rSRkpGRgbe3t/TUISMjg4EDB2rFnT9/Hn9//9/1WQsWLODNN9/USPvuu+9+1zkfZTU1NajVaq2ni5aWls12PSuVSp3x9fX11NbWSjFqtVrjJl1aWoqVlRUymazFp+JCx1RZWUl9fT0ODg4a6XZ2dpSXl2vFW1hY0KNHD7y9vXn55ZeBxi8BmUzGunXrWLhwobSiXENDg7RqU1ZWFl26dOGpp54SjRQDVVtbi1qtxtLSUiNdV29JE129LE33tbt735smMhcXF2Nvb09oaKhopBgopVJJfX09NjY2GunW1tZUVVW1mDc4OJiJEyeyevVqLl26pM9iCkKH0maNlL1793LmzBneeOMNRo0ahaOjI0uWLNFqpGzevJmLFy/y4Ycf/q7PMzc3x9zcXCPNUId6AajVagoKCujatStXrlyR0j08PDRe3ykvL09jxZum+IKCAqnxkZubq9XVbW9vT1VVlWigGCiVSsXly5cJCQnh2LFjUnpISIjOicw1NTXMmzdPI23MmDEEBgby6aefkp+f3+LnGfL/a0OnVqspLCykS5cuGsvA3v36Tvn5+VpzCLp06UJRUVGL9ywjIyONpfEFw1JfX09OTg49evTQeCjSo0ePFufFhYSEMGnSJH766ScyMzPboqiC0GHoZeJ8XV0deXl55OTkcPLkSRYtWsSECROIiYlh+vTpWFlZ8c0337Bp0yZeeeUVTp8+TVZWFnFxccyYMYNJkybxzDPPSOe7fv06qampXL9+nfr6elJTU0lNTb3n0wtBU2pqKgEBAfj5+eHg4MDgwYOxtraWnlJHREQwcuRIKT49PR0bGxsGDx6Mg4MDfn5++Pv7c+rUKSnmzJkzWFhYMGTIEOzt7fHy8qJfv36cOXOmzesntB9btmwhKiqKESNG0LlzZ1588UWcnZ3/X3v3HR5llT7+/53ee0/opJNCQg3Su9JUEEEFooIVBVxRgd21USzr2tZ1AfcjrLoCIiwoJrSEGkIPCZCQQgmkk57MJCSZ/P7gl+fLMAmikjr367rmupwz93nmHJw8M6eza9cuAB5//HFefvll4OboSGZmptajtLSUmpoaMjMzld7thx9+mNDQUNzc3PDy8mLSpEkMHz6cAwcOtFo9RetLSkrC398fPz8/7O3tiYiIwNraWvnh2L9/f0aMGKHEnz9/HmtrayIiIrC3t8fPzw9/f3+ttU29e/fGy8sLGxsb7O3tCQ4OxsfHR85J0XOHDh2iX79+9O3bFxcXFyZOnIi9vb3SGTNu3Dit3y6hoaFMnz6dHTt2kJmZibW1NdbW1lodqEZGRnh4eCjnP9na2uLh4YGTk1OL10+ItqZZRlKio6OVw4scHBwIDQ3ls88+Y86cOcruKdOmTSM2NpaVK1cydOhQ1Go13t7eLFu2jIULF2rN+fvrX/+qHBAJEBYWBkBsbCzDhw9vjip0SGlpaZibm9O/f3+srKwoLCzkp59+UhYCWllZaZ0tUFZWxvbt2xkyZAghISFUVFRw4MABZfthuHl2xbZt2xgyZAgzZ86ksrKSM2fOcPLkyRavn2g7Dh8+jI2NDdOnT8fBwYHMzExWrFhBQUEBAA4ODr96ZsrtzMzMmDdvHk5OTty4cYOsrCw+/fRTOfhMz2VkZGBmZkafPn2wtLSkqKiIqKgopRPL0tJS675WXl5OVFQUERER9OrVi8rKSg4fPqxsPww3R+eGDBmClZUVtbW1lJSUEBsbq3XvE/onMTERS0tLRo0ahY2NDbm5uaxbt07ZJMTW1lZrc5ABAwZgZGTEgw8+yIMPPqiknzx5kh9++EHJs2DBAuW1YcOGMWzYMC5evKh1RIMQ+sig/m5WfHUQn3/+eWsXQeiJ2NjY1i6C0BPjxo1r7SIIPXFrQ06I5vTee++1dhGa1FY7K249aLujaNGF80IIIYQQQrRXsrtXy2mxwxyFEEIIIYQQ4m5II0UIIYQQQgjRpsh0LyGEEEIIIe6CTPdqOXrVSLl1hxchmtO1a9dauwhCT9TV1bV2EYSeaNidUwghWoLccYQQQgghhBBtil6NpAghhBBCCPF7yXSvliMjKUIIIYQQQohGFRcXM2vWLOzs7LCzs2PWrFnKIaZ349lnn8XAwIBPPvnkN73vPR9JiYyMVE6HNzY2xtHRkZCQEGbOnElkZKTWnNa4uDiWL1/OkSNHUKvV+Pj4EBkZycKFCzEyMgLg8uXLvPvuu8TExJCbm4unpydPPPEEy5Ytw9TU9F4Xv8Pz9/cnKCgICwsLSkpKOHbsGHl5eU3Gu7m50b9/f+zt7VGr1SQlJXHhwgWtmK5duxIeHo6NjQ3l5eWcPHmSzMzM5q6KaOOmTp3KrFmzcHJy4uLFi3z88cckJCQ0GhseHs6//vUvnfRHHnmEK1euADBhwgTefPNNnZjBgwdz48aNe1p20b706tWLsLAw5cT5w4cPk5OT02S8p6cngwYNwtHRkcrKShISEjh37lyjsd7e3owdO5aLFy8SHR3dXFUQ7cSAAQMYMmQINjY25Ofns2PHDi5fvtxobK9evejfvz+enp4YGRmRn5/P3r17SUtL04kbM2YMjo6OFBUVsWvXLs6fP98CtRHi7jz22GNcu3ZNuQc+88wzzJo1i59++ulX8/7vf//j6NGjeHp6/ub3bZbpXuPHj+frr7+mrq6OvLw8oqOjWbBgAZs3b2b79u0YGxuzdetWpk+fzpNPPklsbCz29vbs2bOH1157jfj4eDZt2oSBgQEpKSloNBpWr16Nt7c3Z8+eZd68eVRWVvK3v/2tOYrfYXXv3p3+/ftz5MgR8vPz8fPzY8yYMWzdupXKykqdeGtra8aMGUNqaioHDhzA1dWViIgIqqqqlB+OLi4uDB8+nFOnTpGZmUmXLl0YMWIEO3bs4Pr16y1dRdFGjB49mldeeYUPPviAM2fO8NBDD/HJJ5/w6KOP3rFRPG3aNK3PYnFxsdbrFRUVPPLII1pp0kDRb97e3gwePJgDBw6Qm5tLYGAgEydO5Pvvv6eiokIn3sbGhgkTJnD+/Hn27t2Lu7s7Q4cORa1Wc/HiRa1Ya2trBg0aRHZ2dktVR7RhwcHBTJgwge3bt3PlyhX69+/PnDlz+OSTTygtLdWJ79atG+np6ezatYuqqir69OnDrFmz+PLLL5VGdOfOnZkxYwZ79uzh/PnzBAYGMnPmTFavXi0bsIjfpLq6murqaq00MzMzzMzM/tB1k5OTiY6OJj4+ngEDBgCwdu1aIiIiuHDhAn5+fk3mzcrKYv78+ezcuZMJEyb85vdululeZmZmuLu74+XlRXh4OEuXLmXbtm1ERUWxbt06KisrmTdvHpMnT2bNmjX07t2bbt26MXfuXNavX8/mzZvZtGkT8P8aPGPHjqVHjx5MnjyZV199lS1btjRH0Tu0Xr16kZaWRlpaGqWlpRw7dozKykr8/f0bjff396eyspJjx45RWlqq5A0KCtK6ZnZ2NklJSZSWlpKUlER2dja9evVqqWqJNuixxx5j+/btbNu2jcuXL/Pxxx+Tl5fH1KlT75ivqKiIwsJC5aHRaLRer6+v13q9sLCwOash2oHQ0FCSk5NJTk6muLiYw4cPU1FRoXWfulWvXr2oqKjg8OHDFBcXk5ycTEpKCr1799aKMzAwYMyYMRw/fpyysrIWqIlo6wYPHszJkyc5ceIEBQUF7Nixg9LSUuWH2+127NjBwYMHycrKorCwkF27dlFYWEhAQIASc99995Gens7+/fspKChg//79ZGRkcN9997VUtUQHsWrVKmU6VsNj1apVf/i6R44cwc7OTutzPnDgQOzs7IiLi2syn0ajYdasWSxevPh3/yZssTUpI0eOJDQ0lC1btih/qK+++qpO3KRJk/D19eX7779v8lqlpaU4Ojo2Z3E7HENDQ5ycnMjKytJKz87OxtXVtdE8Li4uOj2IWVlZODs7KwvHXFxcdK6ZlZXV5DVFx2dsbIy/vz9Hjx7VSj969CghISF3zPvtt9/yyy+/8MUXX9CnTx+d1y0sLNi2bRs//fQTf//73/H19b2nZRfti6GhIS4uLly9elUr/erVq7i5uTWax93dXSc+MzMTFxcXrenIffv2Ra1Wk5ycfO8LLtodIyMjPD09daZqpaen07Vr17u6hoGBAWZmZqhUKiWtS5cupKena8WlpaXRpUuXP15ooVeWLFlCaWmp1mPJkiV/+Lq5ubmN/qZzdXUlNze3yXzvv/8+xsbGvPzyy7/7vVt04by/vz+XL18mNTUVQKs34fa4hpjbZWRk8Pnnn/Pcc881Wzk7IjMzMwwNDamqqtJKV6vVWFhYNJrHwsICtVqtlVZVVYWhoSHm5uZKzO3XrKqqavKaouOzt7fH2NhYZ5SjqKgIJyenRvNcv36dFStW8Prrr/P6669z5coVvvjiC8LCwpSYK1eu8M477/Dqq6/yl7/8herqar766is6d+7crPURbZe5uTmGhoY69ymVSoWlpWWjeSwtLbV+JMLN+6CRkZFyX3N3dycgIIB9+/Y1S7lF+2NpaYmRkZHOFMLy8vK7PoNt8ODBmJqakpSUpKRZW1vrXLOiogIbG5s/XmjRLAwMDNrkw8zMDFtbW63HnaZ6vfXWW796zRMnTih1vl19fX2TO52dPHmSTz/9lHXr1v2h3dBadAvi2ytUX19/V3ENsrOzGT9+PI888ghz586943s1NjevpqYGExOT31HyjqOxf/Om/j/82mu/95pCPxkYGDT5ucjMzNTabCEpKQk3Nzcef/xxTp8+DcDZs2c5e/asEnPmzBm++eYbpk+fzkcffdS8hRdt2u2fq9/7pVhfX4+JiQmjR49m3759Oh0wQvzez1pISAijRo3im2++0VkDKt+hojXMnz+fGTNm3DGmW7duJCYmNrqWtKCgoMkR64MHD5Kfn681IlhXV8ef/vQnPvnkkyY3m7hdizZSkpOT6d69uzJFIzk5mUGDBunEpaSkEBgYqJWWnZ3NiBEjiIiIYM2aNb/6XqtWreLtt9/WSps8eTIPPvjg769AO1ZdXY1Go9EZ4WhsJKSBWq3W6Y00NzdHo9EoeRobiTE3N5cvdz1WUlJCbW2tzqiJg4MDRUVFd32ds2fPMn78+CZfr6+v5/z58zKSoseqqqrQaDQ69ykLCwud0ZIGjY2yWFhYUFdXR3V1NQ4ODtja2vLAAw8orzf8EH3uuef473//K2tU9JBKpaKurk5nhKOxkZDbBQcH8/DDD/P999+TkZGh9VpjoyZ3c00h/ihnZ2ecnZ1/NS4iIkJZx9y/f3/g5vTt0tLSRn/DA8yaNYvRo0drpY0bN45Zs2bx5JNP3nUZW2y6V0xMDElJSUydOpWxY8fi6OjYaO/n9u3bSUtLY+bMmUpaVlYWw4cPJzw8nK+//lpr3nBTGpub93t2FugoNBoNhYWFOlvAeXp6kp+f32iegoICnXgvLy+uX7+u9PIUFBTg5eWlE9PUNUXHV1tbS0pKinIza9C/f38SExPv+jq+vr6/ujDe19dXdpHTYxqNhoKCAp2GaqdOnZrcRS43N5dOnTpppXXu3JmCggI0Gg0lJSVs2LCBTZs2KY9Lly6RlZXFpk2b5MejnqqrqyM7Oxtvb2+tdG9vb2W3y8aEhIQwbdo0Nm7cqLN9P9wcRW7smrKNf9vV2tO6mno0l4CAAMaPH8+8efOIj48nPj6eefPmMXHiRK2dvfz9/dm6dSsATk5OBAUFaT1MTExwd3e/425gt2uWRkp1dTW5ublkZWVx6tQpVq5cyZQpU5g4cSKzZ8/GysqK1atXs23bNp555hkSExO5fPky//73v4mMjGTatGlMnz4duDmCMnz4cDp37szf/vY3CgoKyM3NveNiHaDRuXn6PtXr3Llz+Pr64uPjg52dHf3798fKyoqUlBQA+vTpw5AhQ5T4lJQUrKys6NevH3Z2dvj4+ODj46M15eb8+fN4enoSHByMnZ0dwcHBeHp6NnnmgNAP//3vf5kyZQqTJk2iW7duLFq0CHd3d2VXvhdeeIG33npLiZ8xYwbDhg2jc+fO9OjRgxdeeIFRo0Ypu/wBzJ07l4EDB+Lp6YmPjw9//vOf8fX1lZ3+9NyZM2cICAjA398fBwcH7rvvPmxsbJT71MCBAxk1apQSf+7cOWxsbBg0aBAODg74+/sTEBCgnOFTV1dHUVGR1uPGjRvcuHGDoqIinR3nhP44dOgQffv2pU+fPri4uPDAAw9gZ2fHsWPHABg7dizTpk1T4kNCQnjkkUf45ZdfuHr1KtbW1lhbW2utE4iLi8Pb25uhQ4fi4uLC0KFD8fb25vDhwy1ePyGa8t133xEcHMzYsWMZO3YsISEhfPPNN1oxFy5caHQr7j+iWaZ7RUdH4+HhgbGxMQ4ODoSGhvLZZ58xZ84cZRRk2rRpxMbGsnLlSmWPem9vb5YtW8bChQuVVuGuXbtIT08nPT1dp/dL5mz+NpcuXcLMzIzQ0FAsLS0pLi5m9+7dyvxYCwsLrKyslPiKigp2797NgAEDCAgIQKVScfToUa1eo/z8fPbt20d4eDhhYWGUl5ezb98+6d3Wc3v27MHOzo6nn34aZ2dnMjIyWLRokdK54OzsrDWX1cTEhJdffhkXFxeqq6u5dOkSCxcu1Nre0MbGhiVLluDk5ERFRQWpqak8++yzcuiZnktPT8fMzIy+fftiZWVFYWEhP//8szLiYWlpqbWwuby8nB07dnDfffcRHBxMZWUlhw4d0jkjRYjbJSUlYWlpyciRI7GxsSEvL4/169crJ2/b2Nhgb2+vxPfv3x8jIyOmTJnClClTlPSTJ0/y448/AjdHUjZu3MiYMWMYPXo0RUVFbNiwQc5IEW2Ko6Mj33777R1jfu03+d2uQ7mVQb0e/dL/+uuvW7sIQk98+eWXrV0EoSciIyNbuwhCT8gPZ9FSVq5c2dpFaFJbnYrXEbetbtEtiIUQQgghhBDi10gjRQghhBBCCNGmtOgWxEIIIYQQQrRXzbmTltAmIylCCCGEEEKINkWvRlJu3x1MiOYiu5uJliLndoiWIt+hQoiWpFeNFCGEEEIIIX4vme7VcmS6lxBCCCGEEKJNkUaKEEIIIYQQok2R6V5CCCGEEELcBZnu1XLueSMlMjKS9evX37y4sTGOjo6EhIQwc+ZMIiMjMTT8f4M3cXFxLF++nCNHjqBWq/Hx8SEyMpKFCxdiZGSkxE2ePJmEhATy8/NxcHBg9OjRvP/++3h6et7r4nd49fX1/PLLLxw+fBi1Wk3Xrl159NFH8fDwaDJPTk4OP//8M1evXqWoqIipU6cyYsSIJuN37tzJTz/9xPDhw5k2bVpzVEO0A0888QTz5s3D1dWV1NRUli9fzvHjx5uMNzU15aWXXuLBBx/E2dmZ3Nxc/vnPf/LDDz/oxE6cOJHPPvuMXbt28dxzzzVnNUQ70Lt3b/r374+1tTXXr18nJiamydPRraysGDFiBO7u7jg4OHDy5EliYmK0YpycnBg8eDDu7u7Y2dmxd+9eTp482RJVEW1cr169CAsLw9LSkqKiIg4fPkxOTk6T8Z6engwaNAhHR0cqKytJSEjg3LlzjcZ6e3szduxYLl68SHR0dHNVQYh2o1mme40fP56cnBwuX75MVFQUI0aMYMGCBUycOJHa2loAtm7dyrBhw+jUqROxsbGkpKSwYMECVqxYwYwZM6ivr1euN2LECDZt2sSFCxf48ccfycjIkB+/v9OePXuIjY1l+vTpLF68GFtbWz7//HOqqqqazHPjxg2cnZ2ZPHkytra2d7z+lStXiIuLw8vL614XXbQjEyZM4M9//jNffPEFEydO5MSJE/zf//3fHTsWPv/8cwYNGsQbb7zB6NGjWbBgARkZGTpxnp6eLFmyhGPHjjVnFUQ74e/vz6hRo4iPj2fdunVcu3aNadOmYWNj02i8kZERKpWKI0eOkJ+f32iMiYkJpaWl7N+/X3ZPEwpvb28GDx7MyZMn+eGHH8jJyWHixIlYW1s3Gm9jY8OECRPIycnhhx9+4NSpUwwePJgePXroxFpbWzNo0CCys7ObuxpCtBvN0kgxMzPD3d0dLy8vwsPDWbp0Kdu2bSMqKop169ZRWVnJvHnzmDx5MmvWrKF3795069aNuXPnsn79ejZv3symTZuU6y1atIiBAwfStWtX5UdMfHw8NTU1zVH8Dqu+vp7Y2FjGjRtH79698fT0ZNasWdTU1HDixIkm83Xt2pWHHnqIvn37Ymzc9OBbdXU169atY+bMmVhYWDRHFUQ78fTTT/PDDz+wadMmMjIyePfdd8nJyeHxxx9vNH7o0KEMGDCAp556isOHD5OVlUViYiKnTp3SijM0NOSTTz7h008/JTMzsyWqItq4vn37kpiYSGJiIkVFRcTExFBeXk5YWFij8WVlZcTExHDu3Dmqq6sbjcnNzWXfvn2kpKRQV1fXnMUX7UhoaCjJyckkJydTXFzM4cOHqaioICgoqNH4Xr16UVFRweHDhykuLiY5OZmUlBR69+6tFWdgYMCYMWM4fvw4ZWVlLVATIdqHFls4P3LkSEJDQ9myZQu7du2isLCQV199VSdu0qRJ+Pr68v333zd6naKiIr777jsGDRqEiYlJcxe7QyksLKSsrAx/f38lzcTEBG9vby5evPiHr79x40aCgoK0ri/0j4mJCUFBQRw8eFAr/eDBg4SHhzeaZ/To0SQlJfHss88SFxfH3r17WbJkCWZmZlpxL7/8MoWFhVqdGEJ/GRoa4u7uzuXLl7XSL126JKO54p4yNDTExcWFq1evaqVfvXoVNze3RvO4u7vrxGdmZuLi4qI19b1v376o1WqSk5PvfcGFaMdadHcvf39/Ll++TGpqKgABAQFNxjXENHj99dexsrLCycmJzMxMtm3bdsf3qq6upqysTOtx48aNe1ORdqqhh+b2aRA2NjZ/uPfmxIkTXL16lcmTJ/+h64j2z8HBAWNjY50DLQsLC3FxcWk0T+fOnenbty++vr4899xzvPvuu9x///288847SkyfPn145JFHWLp0abOWX7QflpaWGBoaUllZqZWuUqmwsrJqpVKJjsjc3BxDQ0PUarVWukqlwtLSstE8lpaWqFQqrTS1Wo2RkRHm5ubAzYZMQEAA+/bta5ZyC9GetWgjpb6+XmtXhFvXndwpDmDx4sWcPn2aXbt2YWRkxOzZs5vMD7Bq1Srs7Oy0Hhs2bLg3FWknjh8/ziuvvKI8GqYtNLYzxR/ZraK4uJgff/yROXPmyOiWUDT299nU36yhoSH19fUsWrSIxMRE9u3bx4oVK5g6dSpmZmZYWVnx97//naVLl1JcXNzcRRcdwJ2+H4T4vW7/XP3e7876+npMTEwYPXo0+/btu+O6UNG2GBgYtMlHR9SiWxAnJyfTvXt3fH19leeDBg3SiUtJSSEwMFArzdnZGWdnZ3x9fQkICKBz587Ex8cTERHR6HstWbKEV155RSvt9uknHV1wcDDdunVTnjdsWlBWVoadnZ2SXl5e3uQi07uRmZlJeXk5H3zwgZKm0WjIyMjgwIEDfPLJJ1pD26JjKy4upra2VmfUxMnJSWd0pUF+fj65ubmUl5craenp6RgaGuLh4YGFhQWdO3dm7dq1yusNn6nU1FRGjx4ta1T0kEqlQqPR6IyaNNaDLcQfUVVVhUaj0Rk1sbCwaPKz1tgoi4WFBXV1dVRXV+Pg4ICtrS0PPPCA8nrDj83nnnuO//73v7JGRei1FmukxMTEkJSUxKJFixg7diyOjo589NFHOo2U7du3k5aWxrvvvtvktRp6Mppa9Ag3F+/fPp/d1NT0D9Sg/TE3N1eGlOHmv5utrS0pKSl07twZuNlwSU9PZ8qUKb/7ffz8/HSm4Hz77be4ubkxZswYaaDomZqaGs6ePcvgwYPZtWuXkj548GD27NnTaJ6TJ0/ywAMPaP247N69O3V1deTk5FBfX8/48eO18rzyyitYW1vzzjvv3HELUNFxaTQacnNz6datG2lpaUp6t27dSE9Pb8WSiY5Go9FQUFBA586duXTpkpLeqVMnnTVRDRo+m7fq3LkzBQUFaDQaSkpKdGZ49O/fH1NTUw4dOiQ7ywm91yyNlOrqanJzc6mrqyMvL4/o6GhWrVrFxIkTmT17NkZGRqxevZoZM2bwzDPPMH/+fGxtbdm7dy+LFy9m2rRpTJ8+HYBjx45x7NgxBg8ejIODAxcvXuSvf/0rPXv2bHIURTTOwMCAESNGsGvXLlxdXXFxcWHnzp2YmJjQt29fJe4///kPdnZ2SsOltraW3Nxc5b9LSkq4du0aZmZmuLi4YG5urrO1rKmpKVZWVnKWjZ7697//zUcffURSUhKnTp1i5syZeHp68t133wE3p2+6ubkpm2ds376d+fPn88EHH/DJJ5/g4ODAkiVL+OGHH5TOiNvXqTX0MN6eLvTLiRMnmDBhArm5uWRlZdG7d29sbW1JSEgAbu4cZ21tzS+//KLkcXV1BW7epywtLXF1daWuro7CwkLg5iids7MzcHPLYhsbG1xdXblx4wYlJSUtWj/Rdpw5c4ZRo0aRn59PXl4egYGB2NjYcPbsWQAGDhyIlZUVe/fuBeDcuXMEBwczaNAgkpOTcXNzIyAggN27dwNQV1dHUVGR1ns0rJ29PV20HR11alVb1CyNlOjoaDw8PDA2NsbBwYHQ0FA+++wz5syZo/SqT5s2jdjYWFauXMnQoUNRq9V4e3uzbNkyFi5cqHwILCws2LJlC2+++SaVlZV4eHgwfvx4NmzYoDNSIn7d6NGjuXHjBhs3bkSlUtGtWzfmz5+vNeJSVFSk9UdYWlrKe++9pzzfu3cve/fuxdvbm4ULF7Zk8UU7sWPHDhwcHHjppZdwcXEhNTWVp556SjkDwMXFRasBq1KpmD17Nm+99Rbbtm2juLiYX375hY8++qi1qiDaiZSUFMzNzRk0aBBWVlZcv36dzZs3K41YKysrnfOdIiMjlf92d3cnMDCQ0tJSVq9eDdw8s+LWmP79+9O/f38yMzP1bm2j+H/S09MxMzOjb9++WFlZUVhYyM8//6yMeFhaWmqdmVJeXs6OHTu47777CA4OprKykkOHDt2T3TSF0AcG9Xq0urCh90KI5vbss8+2dhGEnnjuuedauwhCTzR1aKEQ99oLL7zQ2kVoUludXuzh4dHaRbjnWnThvBBCCCGEEO2VTPdqObKiWQghhBBCCNGmSCNFCCGEEEII0aZII0UIIYQQQgjRpujVmhTZOlK0FNk+UrSU2w+LE6K5GBkZtXYRhBB6REZShBBCCCGEEG2KXo2kCCGEEEII8XvJ7l4tR0ZShBBCCCGEEG3KPR9JiYyMZP369TcvbmyMo6MjISEhzJw5k8jISOXEeYC4uDiWL1/OkSNHUKvV+Pj4EBkZycKFCxud+1pdXc2AAQM4c+YMp0+fpnfv3ve6+HohMDCQHj16YGpqSmFhIadPn1ZOZ26Kl5cXQUFBWFlZUVlZSVJSknJ6eANzc3NCQkJwd3fHyMiIiooKjh8/LmuB9NTTTz/Nyy+/jJubGykpKSxZsoQjR440GvvPf/6Txx57TCc9OTmZiIgInfSHH36Y//u//2PHjh08/vjj97zson0JCgoiPDwcS0tLioqKOHjw4B0PXPP09GTw4ME4OjpSWVnJqVOnOHfunFaMqakpAwcOpGfPnpiZmVFWVsbhw4e5cuVKc1dHtGGBgYGEhoZiaWlJcXExcXFx5ObmNhnv4eFBREQEDg4OqFQqEhISSE5OVl7v3r07YWFh2NraYmhoSGlpKYmJiaSlpbVEdYRo05plJGX8+PHk5ORw+fJloqKiGDFiBAsWLGDixInU1tYCsHXrVoYNG0anTp2IjY0lJSWFBQsWsGLFCmbMmEF9fb3OdV977TU8PT2bo8h6w8/PD19fX06fPs2ePXuoqqpi6NChGBs33V51dHRk4MCBXLlyhd27d3PlyhUiIiJwdHRUYkxMTBg5ciQajYaDBw+yc+dOzpw5Q01NTUtUS7QxDz30EKtWreJvf/sbQ4cO5ciRI/zwww906tSp0fg33ngDX19f5REYGEhRURHbtm3Tie3cuTPvvvsucXFxzV0N0Q54e3szZMgQTpw4wcaNG8nOzmbSpElNno5uY2PDpEmTyM7OZuPGjZw8eZKhQ4fSs2dPJcbQ0JApU6Zga2tLVFQU3333HbGxsVRUVLRUtUQb1LNnTwYNGsTp06f58ccfyc3N5YEHHrjjZ+3+++8nNzeXH3/8kdOnT3PffffRvXt3JaaqqopTp07xv//9j82bN3PhwgWGDx/e5L1StD4DA4M2+eiImqWRYmZmhru7O15eXoSHh7N06VK2bdtGVFQU69ato7Kyknnz5jF58mTWrFlD79696datG3PnzmX9+vVs3ryZTZs2aV0zKiqKXbt28be//a05iqw3fHx8SE5OJisri7KyMo4fP46RkRFdunRpMo+vry95eXmkpKRQXl5OSkoK+fn5+Pj4KDH+/v6oVCpOnDhBcXExKpWK/Px8KisrW6Jaoo158cUX+eabb/jmm29ITU1lyZIlZGVl8dRTTzUaX1ZWRn5+vvIICwvD3t6e7777TivO0NCQNWvW8N5773H58uUWqIlo63r37s358+c5f/48xcXFHDp0iIqKCoKDgxuNDwoKory8nEOHDlFcXMz58+dJTk4mLCxMiQkICMDc3JxffvmF3NxcysvLycnJobCwsKWqJdqg4OBgUlJSSElJoaSkhLi4OCoqKggMDGw0PjAwkIqKCuLi4igpKSElJYULFy4QGhqqxDR06JaUlFBWVsbZs2cpLCzE3d29paolRJvVYmtSRo4cSWhoKFu2bGHXrl0UFhby6quv6sRNmjQJX19fvv/+eyUtLy+PefPm8c0338h2m3+AlZUVFhYW5OXlKWkajYaCggKcnJyazOfk5KSVByA3N1crj6enJ8XFxQwcOJBJkyYxevRord4ioT9MTEzo3bs3sbGxWumxsbEMGDDgrq4xa9Ys9u3bx9WrV7XSX3/9dQoLC/nmm2/uWXlF+2VoaIirq6vO5+Tq1atN/shzd3fXic/MzMTFxUWZjty9e3dyc3MZNmwYTz31FDNnzqRPnz4dtrdS/DpDQ0NcXFy4du2aVvq1a9dwc3NrNI+bm5tO/NWrV3F2dtaa+n4rLy8v7O3t7zhdUQh90aK7e/n7+5OYmEhqaipws7eqqbiGmPr6eiIjI3nuuefo27fvXfeeVldXU11drZVWU1ODiYnJ769AO2dubg7cHF6+VXV19R0bf+bm5jp5qqqqlOvBzQZQz549SU1NJSUlBUdHR8LCwtBoNDKHW884OTlhbGxMfn6+Vnp+fj6urq6/mt/NzY3Ro0czd+5crfQBAwbwxBNPMGTIkHtaXtF+WVhYYGhoiEql0kpXqVRN3tOsrKzIzMzUiTcyMsLc3ByVSoWdnR02Njakpqby008/YW9vz7BhwzA0NOT48ePNVh/Rdpmbm2NoaIhardZKV6vVTX7WLCwsGo2/9bMGN9c/PfHEExgaGlJfX8+hQ4fIyspqnooI0Y60aCOlvr5eqyeqsXUnt8d9/vnnlJWVsWTJkt/0XqtWreLtt9/WSps2bRrTp0//jaVuv7p06UKfPn2U5wcPHrxn1769R9HAwICioiLOnj0L3Dw409bWlp49e0ojRU/d/vdtYGDQ5N/8rR577DFKS0vZsWOHkmZtbc2aNWtYsGCBHJQpmp2BgQFqtZrY2Fjq6+spKCjAysqKsLAwaaQIHXe6r93pd06DGzdusHnzZkxMTPDy8iIiIoKysjIZTRF6r0UbKcnJyXTv3h1fX1/l+aBBg3TiUlJSlDmeMTExxMfHY2ZmphXTt29fHn/8cWUnsdstWbKEV155RSvt559/vhfVaDeys7O15lA37Jh2+8iImZmZzkjJrW4fNWksj1qt1tkhrKysTBb/6aHCwkJqa2t1pkC4uLhQUFDwq/mfeOIJNm7cqLXpQvfu3enatSsbNmxQ0hqmS1y/fv03jbKKjkOtVqPRaHR6si0tLXVGVxpUVlY2Gl9XV6fc0yorK9FoNFo/JIuKirCyssLQ0BCNRnOPayLauqqqKjQaDRYWFlrpjY2WNGhslMXCwoK6ujqdmR4N35+FhYXY29sTFhYmjRSh91psTUpMTAxJSUlMnTqVsWPH4ujoyEcffaQTt337dtLS0pg5cyYAn332GWfOnCEhIYGEhAR++eUXADZu3MiKFSuafD8zMzNsbW21Hvo21au2tpbKykrlUVZWhlqt1vrxaGBggIuLyx0XhBYWFur84HRzc9PKU1hYiI2NjVaMjY2NLJzXQzU1NSQkJDB8+HCt9OHDh3P06NE75h08eDA9e/bUWXOSmppKREQEQ4YMUR5RUVEcPHiQIUOGyNQIPaXRaMjPz6dz585a6Z07d25yW9jc3NxG4wsKCpTGR05ODnZ2dlox9vb2SuNF6J+G9Zu3d7x16tRJZ81mg7y8vEbjr1+/fsfPkYGBQaPHMIi2obV38dKn3b2aZSSlurqa3Nxc6urqyMvLIzo6mlWrVjFx4kRmz56NkZERq1evZsaMGTzzzDPMnz8fW1tb9u7dy+LFi7WmZd2+61TDVn89e/aUXvrfIS0tDX9/f8rLy6moqCAgIIC6ujqtOdr9+vVDrVYrU7fS0tIYPnw4fn5+ZGdn4+npiZubm9bC6NTUVEaOHIm/vz9Xr17F0dGRHj16cPLkyRavo2h9X3zxBatXryYhIYFjx44RGRlJp06d+PrrrwH461//iqenJ88995xWvlmzZnH8+HGtcwTg5j3l9rTS0lIAnXShXxISEhgzZgz5+fnk5ubSq1cvrK2tlftXREQEVlZW7NmzB4CzZ88SEhLC4MGDOXfuHO7u7gQGBrJr1y7lmg0xQ4cOJTExETs7O/r27cuZM2dapY6ibUhKSmLEiBEUFBSQl5dHQEAA1tbWnD9/HoD+/ftjZWWlfDeeP3+eXr16ERERQXJyMm5ubvj7+7N3717lmr1796agoICysjKMjIzo3LkzPj4+HDp0qFXqKERb0iyNlOjoaDw8PDA2NsbBwYHQ0FA+++wz5syZo0zRmDZtGrGxsaxcuZKhQ4eiVqvx9vZm2bJlLFy4sMO2ClvbhQsXMDIyIjw8HFNTU4qKijhw4IByfg2gMzxdWFhIfHw8QUFBBAUFUVFRQXx8vNbagIZDrYKDgwkMDKSyspKEhASdBapCP2zduhVHR0dee+013NzcSE5OZvr06cquSu7u7jqdDLa2tkyaNIk33nijNYos2qn09HTMzc3p168fVlZWFBYW8vPPP1NeXg7cvJ/dOspbXl7OTz/9xODBgwkODqayspIDBw6QkZGhxFRUVLB9+3YGDx7MjBkzqKys5MyZM5w6darF6yfajoyMDMzMzOjTp49ycGhUVJRyfo6lpaXWmSnl5eVERUURERFBr169qKys5PDhw1y6dEmJMTExYciQIVhZWVFbW0tJSQmxsbFan0ch9JVB/d2sZO0gfvjhh9YugtAT8+bNa+0iCD2xfPny1i6C0BP6NmVatJ5nn322tYvQpOvXr7d2ERrl7Ozc2kW451psTYoQQgghhBBC3A1ppAghhBBCCCHalBbdglgIIYQQQoj2StZMtxwZSRFCCCGEEEK0KXo1kmJvb9/aRRB6wsrKqrWLIIQQ91TD7pxCCNES5I4jhBBCCCGEaFOkkSKEEEIIIYRoU6SRIoQQQgghhGhT7vmalMjISNavX3/z4sbGODo6EhISwsyZM4mMjNSa0xoXF8fy5cs5cuQIarUaHx8fIiMjWbhwIUZGRkpct27duHLlitb7vP7667z33nv3uvgdXn19Pb/88guHDx9GrVbTtWtXHn30UTw8PJrMk5OTw88//8zVq1cpKipi6tSpjBgxosn4nTt38tNPPzF8+HCmTZvWHNUQ7cCcOXN47rnncHV1JTU1lTfffJNjx441GW9qasqiRYt4+OGHcXFxIScnh88++4yNGzcqMXPnzmX27Nl4enpSXFzMjh07WLVqFdXV1S1RJdFGBQUFER4erpwCfvDgQXJycpqM9/T0ZPDgwTg6OlJZWcmpU6c4d+6cVoypqSkDBw6kZ8+emJmZUVZWxuHDh3W+i4R+CQgIIDQ0FAsLC4qLi4mPjyc3N7fJeHd3dwYOHIiDgwMqlYrExESSk5OV1/38/PD19cXBwQG4eVDg8ePHKSgoaPa6iN9HdvdqOc2ycH78+PF8/fXX1NXVkZeXR3R0NAsWLGDz5s1s374dY2Njtm7dyvTp03nyySeJjY3F3t6ePXv28NprrxEfH8+mTZu0PgjvvPOO1ine1tbWzVH0Dm/Pnj3ExsbyxBNP4OrqSnR0NJ9//jl//etfMTc3bzTPjRs3cHZ2JiwsjC1bttzx+leuXCEuLg4vL6/mKL5oJyZPnsxbb73F0qVLOX78OLNmzeLbb79l+PDhZGdnN5rnX//6Fy4uLrz66qtcunQJZ2dnjI3/3y3qoYceYsmSJfzpT3/ixIkT9OjRg48//hiAt956qyWqJdogb29vhgwZwv79+8nJyaFXr15MmjSJ//73v1RUVOjE29jYMGnSJM6dO8fu3bvx8PBg2LBhVFVVkZGRAdxcID5lyhTUajVRUVFUVlZibW3NjRs3Wrp6og3p0aMHERERHD58mLy8PPz9/Rk/fjw//PADlZWVOvE2NjaMHz+elJQU9u3bh5ubG/fddx9qtZrLly8DNxvM6enp5OXlUVdXR2hoKPfffz+bN29GpVK1cA2FaFuapZFiZmaGu7s7AF5eXoSHhzNw4EBGjRrFunXrmDlzJvPmzWPy5MmsWbNGyTd37lzc3NyYPHkymzZt4tFHH1Ves7GxUa4pfp/6+npiY2MZN24cvXv3BmDWrFksXbqUEydOMHjw4Ebzde3ala5duwKwffv2Jq9fXV2t/P+Njo6+5+UX7ce8efPYsGED33//PQBvvvkmw4YNY/bs2Y2OgA4fPpyBAwcyaNAgSkpKALh27ZpWTJ8+fThx4gT/+9//lNe3bdumfJaFfurduzfnz5/n/PnzABw6dIguXboQHBzMkSNHdOKDgoIoLy/n0KFDABQXF+Pq6kpYWJjSSAkICMDc3Jwff/wRjUYDQHl5eQvVSLRVwcHBXLhwgQsXLgAQHx9Pp06dCAwM5Pjx4zrxAQEBVFRUEB8fD0BJSQkuLi6EhIQojZTY2FitPAcPHqR79+54eXmRlpbWvBUSoo1rsTUpI0eOJDQ0lC1btrBr1y4KCwt59dVXdeImTZqEr6+v8uOmwfvvv4+TkxO9e/dmxYoV0qP1OxQWFlJWVoa/v7+SZmJigre3NxcvXvzD19+4cSNBQUFa1xf6x8TEhJCQEPbv36+Vvn//fvr27dtonrFjx5KYmMjzzz/PiRMnOHjwIH/5y1+0RveOHTtGcHCw0ijp0qULI0eOZO/evc1WF9G2GRoa4urqytWrV7XSr1692mSnlru7u058ZmYmLi4uynTk7t27k5uby7Bhw3jqqaeYOXMmffr0kWkeeszQ0BBnZ2eysrK00rOysnBzc2s0j6urq078tWvXcHFxafKzZGxsjKGhoUxhbcMMDAza5KMjatFzUvz9/UlMTCQ1NRW42cvQVFxDDMCCBQsIDw/HwcGBY8eOsWTJEi5dusRXX33V5HtVV1fr/JHfuHEDU1PTe1CT9qmsrAy4OSp1KxsbG4qKiv7QtU+cOMHVq1d57bXX/tB1RPvn6OiIsbEx169f10q/fv06rq6ujebp0qUL/fr1o7q6mrlz5+Lo6MjKlSuxt7fnT3/6E3BzFM/JyYmtW7diYGCAiYkJ69ev54svvmj2Oom2ycLCAkNDQ51pMSqVCktLy0bzWFlZkZmZqRNvZGSEubk5KpUKOzs7bGxsSE1N5aeffsLe3p5hw4ZhaGjYaI+56PjMzc0b/ayp1WosLCwazWNpaakzIqxSqTA0NMTc3By1Wq2Tp1+/flRWVuo0boTQRy3aSKmvr9dq7dXX199V3KJFi5T/DgkJwcHBgWnTpimjK41ZtWoVb7/9tlbaE088wezZs/9IFdqV48ePa41IPf/880Dji77+SCu8uLiYH3/8kRdffBETE5PffR3Rsdz+921gYNDk37yhoSH19fXMnz9fmVbz9ttvs2bNGpYtW0ZVVRURERG8/PLLLF26lNOnT9OtWzfeeecd8vPz+eSTT5q7OkKPGBgYoFariY2Npb6+noKCAqysrAgLC5NGivhNGrsPNiUkJISePXuyY8cO6urqmrtoQrR5LdpISU5Opnv37vj6+irPBw0apBOXkpJCYGBgk9cZOHAgAOnp6U02UpYsWcIrr7yilXbw4MHfW/R2KTg4mG7duinPa2trgZsjKnZ2dkp6eXm5zujKb5GZmUl5eTkffPCBkqbRaMjIyODAgQN88sknclKxHikqKqK2thYXFxetdCcnpyZ3rMnPzyc3N1dr3n9aWhqGhoZ4eHhw6dIlFi9ezI8//qg0vFNSUrC0tOSDDz7g008/bbIBJDoutVqNRqPRGTWxtLRsctFxZWVlo/F1dXVUVVUpMRqNRuszVVRUhJWVFYaGhso6FaE/qqqqGv2sWVhYNDoiAo2P6FlYWKDRaJTPWoOGqay//PLLH57ZIERH0WKNlJiYGJKSkli0aBFjx47F0dGRjz76SKeRsn37dtLS0nj33XebvNbp06cB7rhtrpmZGWZmZlpp+jbVy9zcXGtOf319Pba2tqSkpNC5c2fgZsMlPT2dKVOm/O738fPzY+nSpVpp3377LW5ubowZM0YaKHqmpqaGxMREhg4dqrWBwtChQ9m5c2ejeY4fP87EiRO1flz26NGDuro6ZSvZhi/3WzX0Nt5plEZ0XBqNhvz8fDp37qy1rq5z585cunSp0Ty5ubl0795dK61z584UFBQon6+cnBylM62Bvb290ngR+kej0XD9+nW8vLyURe9wc3Ogpralzs/Pp0uXLlppXl5eFBQUaN2vQkJCCAsLIyoqSmearBD6rFl+PVZXV5Obm0tWVhanTp1i5cqVTJkyhYkTJzJ79mysrKxYvXo127Zt45lnniExMZHLly/z73//m8jISKZNm8b06dMBOHLkCB9//DEJCQlcunSJTZs28eyzzzJ58mSdP35xZwYGBowYMYJdu3Zx5swZsrOz+eabbzAxMdFa0Pyf//yHbdu2Kc9ra2u5du0a165do7a2lpKSEq5du6b0ipubm+Pp6an1MDU1xcrKCk9Pzxavp2h9a9euZebMmTz66KN4e3vz1ltv4eXlxTfffAPAG2+8waeffqrEb926leLiYj7++GN8fHwYMGAAf/nLX9iwYYPS47h7925mz57N5MmT6dy5M0OGDGHx4sXs3r1bfjjqsYSEBAIDAwkICMDBwYHBgwdjbW3N2bNnAYiIiGD06NFK/NmzZ7GxsWHw4ME4ODgQEBBAYGCg0vnVEGNubs7QoUOxt7ena9eu9O3bl8TExBavn2g7kpKSlHNN7O3tGThwINbW1sq5J/369WP48OFKfHJyMtbW1gwcOBB7e3t8fX3x8/PT+hyFhITQt29f9u/fT3l5ORYWFlhYWGhtvy6EvmqWv4Lo6Gg8PDwwNjbGwcGB0NBQPvvsM+bMmaP0qk+bNo3Y2FhWrlzJ0KFDUavVeHt7s2zZMhYuXKjM2zQzM2Pjxo28/fbbVFdX07VrV+bNmycLtH+n0aNHc+PGDTZu3IhKpaJbt27Mnz9fa8SlqKhIa95saWmp1raxe/fuZe/evXh7e7Nw4cKWLL5oJ7Zv346DgwOLFi3C1dWVCxcuMGvWLGUxqJubm1YDVqVSMWPGDJYvX05UVBTFxcX89NNPWlMIG6Z0vfbaa7i7u1NUVMTu3bt5//33W7x+ou1IT0/H3Nycfv36YWVlRWFhIT///LMyddDS0lJrOmt5eTk//fQTgwcPJjg4mMrKSg4cOKBsPwxQUVHB9u3bGTx4MDNmzKCyspIzZ85w6tSpFq+faDsuXryImZmZ1sGh0dHRynk8lpaWWFlZKfHl5eVER0cTERFBYGAgKpWKI0eOaI3EBAYGYmRkxJgxY7Te6+TJk/J5a6M66k5abZFBvR7Nkdi9e3drF0HoicjIyNYugtATS5Ysae0iCD1x+xRqIZrLrYd3tzUNZ3m1Nfb29q1dhHtOFgsIIYQQQggh2hSZ9CiEEEIIIcRdkOleLUdGUoQQQgghhBBtijRShBBCCCGEEG2KXk33Onr0aGsXQegJR0fH1i6C0BNNHZApxL3WqVOn1i6CEEKPyEiKEEIIIYQQok2RRooQQgghhBCiTdGr6V5CCCGEEEL8XrK7V8u5542UyMhI1q9ff/PixsY4OjoSEhLCzJkziYyMVE6cB4iLi2P58uUcOXIEtVqNj48PkZGRLFy4ECMjI63r7tixg3feeYfExESsrKwYOnQoW7ZsudfF7/D69OlDREQE1tbWFBQUsGvXLq5evdporLW1NaNHj8bDwwNHR0eOHTumcyBmSEgIkydP1sm7atUq6urqmqUOon149NFHiYyMxMXFhYyMDN5///0mT1Du27cvX3/9tU765MmTuXTpEnDzfjJ37lwmT56Mq6srly9f5uOPP+bw4cPNWg/R9vXr14/77rtPua9FRUWRmZnZaKy1tTXjxo3D09MTR0dHjh49SnR0tFZMZGQk3bt318mbmprKd9991yx1EO1DQEAAoaGhWFhYUFxcTHx8PLm5uU3Gu7u7M3DgQBwcHFCpVCQmJpKcnKy87ufnh6+vLw4ODgBcv36d48ePy1ozIWimkZTx48fz9ddfU1dXR15eHtHR0SxYsIDNmzezfft2jI2N2bp1K9OnT+fJJ58kNjYWe3t79uzZw2uvvUZ8fDybNm1SWqs//vgj8+bNY+XKlYwcOZL6+nqSkpKao+gdWmBgIGPHjiUqKoqrV68SHh7OzJkz+de//kVZWZlOvJGRESqVikOHDjFgwIAmr1tVVcWXX36plSYNFP02btw4Xn/9dZYvX87p06d55JFH+PLLL5kyZcodv9AnTpxIRUWF8ry4uFj575deeokJEybw9ttvc+nSJQYNGsQnn3zCrFmzSElJadb6iLarV69ejB8/nh07dpCZmUnfvn154okn+OKLLygtLdWJNzY2RqVSceDAASIiIhq95saNG7U6yiwsLHj++ec5d+5cs9VDtH09evQgIiKCw4cPk5eXh7+/P+PHj+eHH36gsrJSJ97Gxobx48eTkpLCvn37cHNz47777kOtVnP58mUAPD09SU9PJy8vj7q6OkJDQ7n//vvZvHkzKpWqhWsoRNvSLI0UMzMz3N3dAfDy8iI8PJyBAwcyatQo1q1bx8yZM5k3bx6TJ09mzZo1Sr65c+fi5ubG5MmT2bRpE48++ii1tbUsWLCADz/8kKefflqJ9fPza46id2gDBgwgISGBhIQEAHbv3k3Pnj3p06cPsbGxOvGlpaXs2rULgN69e9/x2o3doIX+mj17Nlu2bFFGOz/44APuu+8+Hn30UT799NMm8xUVFVFeXt7oaxMnTmTt2rUcPHgQgE2bNnHfffcxZ84clixZcu8rIdqFQYMGcfr0aWWULjo6Gm9vb/r168eePXt04ktKSoiKigIgLCys0Wuq1Wqt50FBQdTU1EgjRc8FBwdz4cIFLly4AEB8fDydOnUiMDCQ48eP68QHBARQUVFBfHw8cPOz5+LiQkhIiNJIuf279+DBg3Tv3h0vLy/S0tKat0JCtHEttnB+5MiRhIaGsmXLFnbt2kVhYSGvvvqqTtykSZPw9fXl+++/B+DUqVNkZWVhaGhIWFgYHh4e3H///fJl8RsZGhri4eHBxYsXtdIvXrz4h7eVNDU15aWXXuLll1/m0Ucfxc3N7Q9dT7RvxsbGBAYGEhcXp5UeFxf3q43dTZs2ERMTw9q1a+nXr5/Wa6amplRXV2ulVVVVNflDU3R8RkZGeHh4kJ6erpWekZFB586d79n7hIeHc/bsWWpqau7ZNUX7YmhoiLOzM1lZWVrpWVlZTX7nubq66sRfu3YNFxeXJtc1GBsbY2hoqHOvE0IftejuXv7+/ly+fJnU1FTgZi9DU3ENMQ0/qt966y3+/Oc/8/PPP+Pg4MCwYcMoKipqmYJ3AJaWlhgaGuqMeFRWVmJtbf27r1tYWMj27dvZuHEjW7dupba2lsjISGV+rdA/Dg4OGBsbU1hYqJVeWFiIk5NTo3muX7/OW2+9xSuvvMKiRYu4fPkyX331FX369FFi4uLimD17Nl26dMHAwICIiAhGjBiBi4tLs9ZHtF2WlpYYGRnp3NcqKir+0H3tVl5eXri5uTW5nkroB3NzcwwNDXWmYKnVaiwsLBrNY2lpqTMqp1KpMDQ0xNzcvNE8/fr1o7KyUqdxI4Q+atHdverr67V6D+rr6381TqPRALBs2TKmTp0KwNdff02nTp344YcfePbZZxu9RnV1tU5PRG1tLcbG+r2hWWP/5k39f7gbWVlZWjfTq1evMnfuXPr166dMFRMC7rwjyuXLl5XpDwBnzpzB3d2dOXPmcPLkSQDee+893nrrLbZv3059fT1Xr15l27ZtTJkypbmLLtq42+9hBgYGf+i+dqvw8HDy8vLkR6P4XRr7bDYlJCSEnj17smPHDlnX2YbJ7l4tp0VHUpKTk+nevTu+vr7K88akpKTg4+MDgIeHB3Bz0XcDMzMzevTo0eTuLXBzdyk7Ozutx4EDB+5VVdodlUqFRqPR6V20srK65+tJcnJy5MR1PVZcXExtba3OqImjo6PO6MqdJCYm0rVrV63rLliwgP79+zNu3DgmT56MSqWSH496TKVSUVdX12z3NRMTE4KCgmQURVBVVYVGo8HS0lIr3cLCQme0pIFKpWo0XqPRUFVVpZUeHBxM7969iYqKklkiQvz/WqyREhMTQ1JSElOnTmXs2LE4Ojry0Ucf6cRt376dtLQ0Zs6cCdzcMtfMzExZqAZQU1PD5cuXtX7A3G7JkiWUlpZqPYYOHXrvK9ZOaDQacnJydLbV7N69O9euXbun7+Xm5qa1Q5PQL7W1tZw/f15n56SIiAhl04a74e/v3+g2nDdu3CA/Px9jY2NGjx7d6KYPQj/U1dWRk5NDz549tdJ79OjR5Nbqv0WvXr0wMjLizJkzf/haon3TaDRcv34dLy8vrXQvLy/y8vIazZOfn99ofEFBgdYIS0hICOHh4URHR3P9+vV7X3gh2qlmmftUXV1Nbm6u1hbEq1atYuLEicyePRsjIyNWr17NjBkzeOaZZ5g/fz62trbs3buXxYsXM23aNKZPnw6Ara0tzz33HG+++SadO3ema9eufPjhhwA88sgjTZbBzMwMMzMz7crq+VSvo0ePMmXKFHJycrh27Rrh4eHY2dkpvYQjRozAxsaG7du3K3kaFgSamJhgZWWFm5sbdXV1yo10yJAhZGVlUVRUhJmZGf369cPNzU3n3AGhX/7zn/+watUqzp07x5kzZ3jkkUfw8PBg06ZNACxYsABXV1eWLVsGwBNPPEF2djbp6emYmJgwceJExo4dy8KFC5VrBgcH4+rqyoULF3B1deX555/H0NCw0fNVhP6Ii4vj4YcfJjs7m6tXr9K3b1/s7OyU3ZZGjx6NjY0NW7duVfI07D5pamqKlZUV7u7u1NXV6TSKw8PDSUlJabKnXOiXpKQkhg8fTkFBAfn5+fj7+2Ntba3MCunXrx9WVlbs27cPuDlbJDAwkIEDB5KSkoKrqyt+fn7ExMQo1wwJCaFv377ExMRQXl6urG+pqamhtra2xesoRFvSLL/ao6Oj8fDwwNjYGAcHB0JDQ/nss8+YM2eOcpjjtGnTiI2NZeXKlQwdOhS1Wo23tzfLli1j4cKFWnP+PvzwQ4yNjZk1axZqtZoBAwYQExMji7N/o/Pnz2NhYcGQIUOUQ882bNignCVgbW2NnZ2dVp558+Yp/+3p6UlQUBAlJSX84x//AG4uJpwwYQJWVlZK4/Q///kP2dnZLVcx0ebs3LkTe3t7nnvuOVxcXEhPT+eFF14gJycHABcXF2UqJ9xsBP/pT3/C1dWV6upqJb5hu2G42fHw0ksv0alTJ1QqFQcPHmTp0qVNblks9MO5c+ewtLRk2LBh2NjYkJ+fz3fffXfH+9rzzz+v/LeXlxchISEUFxfzySefKOlOTk507dpVOZxYiIsXL2JmZkZ4eDiWlpYUFRURHR2tzBywtLTEyspKiS8vLyc6OpqIiAgCAwNRqVQcOXJEa/1dYGAgRkZGjBkzRuu9Tp48KdMMhd4zqL9XqwvbgeXLl7d2EYSe2LhxY2sXQeiJhx9+uLWLIPTEH92uXoi7dWsHaVvTVjvGbGxsWrsI95x+z38SQgghhBDiLsnuXi2nRXf3EkIIIYQQQohfI40UIYQQQgghRJsijRQhhBBCCCFEm6JXa1Ju369ciOZiZGTU2kUQeiI4OLi1iyD0RHFxcWsXQQihR2QkRQghhBBCCNGm6NVIihBCCCGEEL+X7O7VcmQkRQghhBBCCNGmSCNFCCGEEEII0abc8+lekZGRrF+//ubFjY1xdHQkJCSEmTNnEhkZiaHh/2sXxcXFsXz5co4cOYJarcbHx4fIyEgWLlyoLDzet28fI0aMaPS9jh07Rr9+/e51FTq83r174+fnh6mpKQUFBcTHx1NSUnLHPF27diU8PBwbGxvKy8s5efIkmZmZWjH+/v4EBQVhYWFBSUkJx44dIy8vrxlrItqy6dOnM2fOHJydncnIyODDDz/k9OnTjcb27duXr776Sif9wQcf5PLly8rzxx9/nEceeQR3d3dKSkrYs2cPn332GTdu3Giuaoh2oL6+npiYGI4fP45araZz585MmjQJNze3O+Y7e/Yse/bsoaioCEdHR8aMGUOvXr2U1z/88MNG740DBgxg8uTJ97oaoh0ICAggNDQUCwsLiouLiY+PJzc3t8l4d3d3Bg4ciIODAyqVisTERJKTk5XX/fz88PX1xcHBAYDr169z/PhxCgoKmr0uQrR1zTKSMn78eHJycrh8+TJRUVGMGDGCBQsWMHHiRGprawHYunUrw4YNo1OnTsTGxpKSksKCBQtYsWIFM2bMoL6+HoBBgwaRk5Oj9Zg7dy7dunWjb9++zVH8Di04OJhevXoRHx/PTz/9hFqtZty4cRgbN91edXFxYfjw4aSnp7Nt2zbS09MZMWIEzs7OSkz37t3p378/Z86cYfv27eTl5TFmzBisrKxaolqijRk7diyLFy/mq6++YsaMGZw+fZovvvgCd3f3O+abPHkyo0aNUh63NoQfeOABXn75ZVavXs3DDz/M22+/zdixY3n55ZebuzqijTt48CCHDx9m0qRJvPDCC1hbW/P1119TXV3dZJ7MzEw2btxIWFgYL730EmFhYWzYsIGrV68qMS+88AJvvPGG8njyyScBCAoKavY6ibanR48eREREcPr0abZu3Upubi7jx49v8nvOxsaG8ePHk5uby9atW0lISCAiIoJu3bopMZ6enqSnp/Pzzz+zbds2KioquP/++7G0tGyhWgnx64qLi5k1axZ2dnbY2dkxa9asX+3cBkhOTmby5MnY2dlhY2PDwIEDdTq476RZGilmZma4u7vj5eVFeHg4S5cuZdu2bURFRbFu3ToqKyuZN28ekydPZs2aNfTu3Ztu3boxd+5c1q9fz+bNm9m0aRMApqamuLu7Kw8nJye2b9/OU089JYuXfofAwEASExO5cuUKJSUlHDx4ECMjI3r27Nlknl69epGdnU1SUhKlpaUkJSWRnZ2t1ePYq1cv0tLSSEtLo7S0lGPHjlFZWYm/v39LVEu0MbNmzWLr1q1s3bqVS5cu8eGHH5Kbm8sjjzxyx3zFxcUUFhYqD41Go7wWEhJCQkICUVFRZGdnc+TIEaKjowkMDGzu6og2rL6+nsOHDzN8+HB69eqFm5sb06ZNo6amhjNnzjSZLy4ujp49ezJs2DBcXFwYNmwYPXv2JC4uTomxsrLCxsZGeVy4cAFHR0e6d+/eElUTbUxwcDAXLlzgwoULlJSUEB8fT0VFRZP3oICAACoqKpTZChcuXCA1NZWQkBAlJjY2luTkZIqKiigtLeXgwYMYGBjIkQmiTXnsscdISEggOjqa6OhoEhISmDVr1h3zZGRkMHjwYPz9/dm3bx9nzpzhL3/5C+bm5nf9vi22JmXkyJGEhoayZcsWdu3aRWFhIa+++qpO3KRJk/D19eX7779v9Drbt2/n+vXrREZGNnOJOx5ra2ssLS3JyspS0jQaDXl5ebi6ujaZz8XFRSsPQFZWlpLH0NAQJycnnZjs7Ow7Xld0TMbGxgQEBHDkyBGt9Pj4eEJDQ++Yd8OGDezevZvVq1frjJSePn2awMBApRfby8uLwYMHc/DgwXtbAdGuFBcXU1FRgbe3t5JmbGxMt27d7thjl5mZiY+Pj1aaj49Pk3lqa2tJSEigT58+0kGmhwwNDXF2dm70u7CpaYWurq468deuXcPFxaXJz5CxsTGGhoZ3HAUUrcvAwKBNPppLcnIy0dHRfPXVV0RERBAREcHatWv5+eefuXDhQpP5li1bxgMPPMAHH3xAWFgYPXr0YMKECb/pd2GLbkHs7+9PYmIiqampwM1ehqbiGmJu9+9//5tx48bRuXPnO75XdXW1zh95TU0NJiYmv6PkHUPD8LFardZKV6vVWFtbN5nPwsKCqqoqrbSqqiosLCyAmyNnhoaGOjFqtVqJEfrDwcEBY2NjioqKtNILCwu1pgjeqqCggHfeeYfz589jamrKxIkTWbNmDXPnzuXUqVMA7Ny5EwcHB77++msATExM2LRpk/Jc6Kfy8nIAnXuYtbX1HacjVFRUNJqn4Xq3S05OpqqqivDw8D9WYNEumZubY2hoiEql0kq/0/ecpaUl165d00pTqVQYGhpibm6u810M0K9fPyorK3UaN0L8msZ+95qZmWFmZvaHrnvkyBHs7OwYMGCAkjZw4EDs7OyIi4vDz89PJ49Go2HHjh289tprjBs3jtOnT9O9e3eWLFnCgw8+eNfv3aKNlPr6eq3WXsO6k1+La3Dt2jV27typTAW7k1WrVvH2229rpU2ePPk3/eO0dz169GDQoEHK8927dzcZ29T/izu9fnva3cQI/XH7/3sDA4MmPw9XrlzhypUryvPExETc3NyYPXu20kjp27cvc+fOZeXKlSQlJdG5c2dee+015s2bx9q1a5uvIqJNSUhIYNu2bcrz2bNnNxn7W3sXm/ruAThx4gQ+Pj7Y2tr+pmsK/dbYfbApISEh9OzZkx07dlBXV9fcRRMdTGO/e998803eeuutP3Td3NzcRkc/XF1dm9w0Ij8/n4qKCt577z2WL1/O+++/T3R0NA8//DCxsbEMGzbsrt67RRspycnJdO/eHV9fX+X5rT+iG6SkpDQ6x/Prr7/GycnprnZVWbJkCa+88opW2oYNG35nydunzMxMrR1CGnZMs7Cw0OrBaWyk5FaN9RSZm5sreaqrq9FoNDoxv3Zd0TEVFxdTW1uLk5OTVrqjoyOFhYV3fZ2kpCQeeOAB5fkLL7zAjh072Lp1KwDp6elYWFjwl7/8ha+++koaxHoiICBAayS9YTOWiooKrQZEYyMlt7K2tqaiokIrrbKystE8xcXFZGRk8Nhjj/3R4ot2qqqqCo1Go7Og/fbv01upVKpG4zUajc53Y3BwML179+aXX37RGYUW4m409rv3TqMob731lk6j5nbHjx8HGm9c36lTp2E96ZQpU1i0aBFwc2fZuLg4/vWvf911I6XF1qTExMSQlJTE1KlTGTt2LI6Ojnz00Uc6cdu3byctLY2ZM2dqpdfX1/P1118ze/bsu5qyZWZmhq2trdZD36Z61dbWUl5erjxKSkpQqVR4enoqMYaGhri5uZGfn9/kdQoKCnQW8Xl5eSl5NBoNhYWFWteFm7uW3Om6omOqra0lOTmZiIgIrfQBAwbccSHz7fz8/Lh+/bry3NzcXGshPdz87DX3fFzRtpiZmeHk5KQ8XF1dsba2Jj09XYmpra3l8uXLdOnSpcnrdOnSRSsPQFpaWqN5Tp06hZWVVaPTGoR+0Gg0XL9+vdHvwqa22s/Pz280vqCgQKtTJSQkhPDwcKKjo7XueUL8Fo397r1TI2X+/PkkJyff8REUFIS7u3ujn/GCgoIm12M5OztjbGysM+AQEBDwm3b3apaRlOrqanJzc6mrqyMvL4/o6GhWrVrFxIkTmT17NkZGRqxevZoZM2bwzDPPMH/+fGxtbdm7dy+LFy9m2rRpTJ8+XeuaMTExXLp0iaeffro5iqw3zp8/T0hICGVlZZSVlRESEkJdXR0ZGRlKzJAhQ1CpVJw8eVLJc//99xMcHExmZiZdunTB09OTHTt2KHnOnTvHkCFDKCwsJD8/Hz8/P6ysrEhJSWnxOorW980337BixQrOnTtHYmIiU6dOxcPDg82bNwPw0ksv4erqyl/+8hfg5vkn2dnZZGRkYGJiwgMPPMCYMWO0eoUOHDjAE088QUpKCklJSXTp0oUXXniB/fv36zRehP4wMDDgvvvuY//+/Tg5OeHs7My+ffswMTHR2qjhhx9+wNbWlnHjxgEQERHBV199xYEDBwgICCA5OZmMjAyeeeYZretrNBpOnTpFeHi4Mhot9FNSUhLDhw+noKCA/Px8/P39sba2Vs496devH1ZWVuzbtw+4OVskMDCQgQMHkpKSgqurK35+fsTExCjXDAkJoW/fvsTExFBeXq7MSKipqVFGCYVoDs7Ozk2uE71VRESEsmtr//79ATh69CilpaWNzoaCmzvz9uvXT2dhfWpqKl27dr3rMjZLIyU6OhoPDw+MjY1xcHAgNDSUzz77jDlz5iiHOU6bNo3Y2FhWrlzJ0KFDUavVeHt7s2zZMhYuXKjTM/rvf/+bQYMGNbnYXtydpKQkjIyMiIiIwNTUlOvXr7Nz506tm6GVlZVWL09+fj779u0jPDycsLAwysvL2bdvn1aPz6VLlzAzMyM0NBRLS0uKi4vZvXs3lZWVLVo/0Tbs2rULe3t7nn32WZydnUlPT2f+/Pnk5OQAN3eM8/DwUOJNTExYtGgRrq6uVFdXk5GRwfz58zl06JASs3btWurr63nxxRdxdXWluLiYAwcO8I9//KPF6yfaliFDhlBTU8P27dupqqqiU6dOPPnkk1q9iKWlpVrfK127duXRRx9l9+7d7NmzB0dHR2bMmKGzKUtGRgYlJSX06dOnxeoj2qaLFy9iZmZGeHg4lpaWFBUVER0drUwbtLS01Dozpby8nOjoaCIiIggMDESlUnHkyBGtA2oDAwMxMjJizJgxWu918uRJZT2eaFv0beQ+ICCA8ePHM2/ePFavXg3AM888w8SJE7VGl/39/Vm1ahUPPfQQAIsXL+bRRx9l6NChjBgxgujoaH766SelEX83DOr1aCK37AIkWsqnn37a2kUQeuLPf/5zaxdB6Ini4uLWLoLQE/PmzWvtIjSpqTVIra05d1MtKiri5ZdfZvv27cDNjaj+8Y9/YG9vr8QYGBjw9ddfax0R8n//93+sWrWKa9eu4efnx9tvv82UKVPu+n1bdOG8EEIIIYQQov1wdHTk22+/vWNMY2MeTz31FE899dTvft8WWzgvhBBCCCGEEHdDGilCCCGEEEKINkUaKUIIIYQQQog2Ra/WpNx+qJIQQrR3erT3iWhlcu6VEPq3u1drkpEUIYQQQgghRJsijRQhhBBCCCFEmyKNFCGEEEIIIUSbcs/XpERGRrJ+/fqbFzc2xtHRkZCQEGbOnElkZKRy4jxAXFwcy5cv58iRI6jVanx8fIiMjGThwoUYGRkpcampqSxevJjDhw9z48YNgoODWb58OSNGjLjXxdcLvXr1omfPnpiYmFBUVMTJkycpKyu7Y55OnToRFBSEtbU1FRUVJCUlkZWVpXXNoKAgrTxqtVo5+Efon+nTpzNnzhycnZ3JyMjgww8/5PTp043G9u3bl6+++kon/cEHH9Q6nfnxxx/nkUcewd3dnZKSEvbs2cNnn33GjRs3mqsaop0IDAykR48emJqaUlhYyOnTp3/1vubl5UVQUBBWVlZUVlaSlJREdna2Voy5uTkhISG4u7tjZGRERUUFx48fp6SkpBlrI9qqPn36MHDgQKytrSkoKGD37t1cvXq10Vhra2tGjRqFh4cHjo6OHD9+nN27dzd57cDAQB566CEuXLjA5s2bm6sKQrQbzTKSMn78eHJycrh8+TJRUVGMGDGCBQsWMHHiRGprawHYunUrw4YNo1OnTsTGxpKSksKCBQtYsWIFM2bM0FoMOmHCBGpra4mJieHkyZP07t2biRMnkpub2xzF79D8/f3x8/Pj5MmT7Nmzh6qqKoYPH46xcdPtVScnJyIiIrhy5Qo7d+7kypUrDBo0CEdHR6240tJStm3bpjx27tzZ3NURbdTYsWNZvHgxX331FTNmzOD06dN88cUXuLu73zHf5MmTGTVqlPLIzMxUXnvggQd4+eWXWb16NQ8//DBvv/02Y8eO5eWXX27u6og2zs/PD19fX06fPq3c14YOHXrH+5qjoyMDBw7kypUr7N69mytXrhAREaF1XzMxMWHkyJFoNBoOHjzIzp07OXPmDDU1NS1RLdHGBAQEMGbMGA4fPsxXX33F1atXmTFjBra2to3GGxkZoVKpOHz4MHl5eXe8tq2trc49Twh91yyNFDMzM9zd3fHy8iI8PJylS5eybds2oqKiWLduHZWVlcybN4/JkyezZs0aevfuTbdu3Zg7dy7r169n8+bNbNq0CYDr16+Tnp7OG2+8QUhICD4+Prz33nuoVCrOnTvXHMXv0Hx9fTl//jxZWVmUlpZy9OhRjIyM6Nq16x3z5OXlkZycTHl5OcnJyeTl5eHr66sVp9FoqKqqUh7V1dXNXR3RRs2aNYutW7eydetWLl26xIcffkhubi6PPPLIHfMVFxdTWFioPDQajfJaSEgICQkJREVFkZ2dzZEjR4iOjiYwMLC5qyPaOB8fH5KTk8nKyqKsrIzjx49jZGREly5dmszTcF9LSUmhvLyclJQU8vPz8fHxUWL8/f1RqVScOHGC4uJiVCoV+fn5VFZWtkS1RBszYMAAEhISSEhIoLCwkN27d1NWVkZ4eHij8aWlpezevZukpKQ7fh8aGBjw4IMPcuDAAYqLi5ur+EK0Oy22JmXkyJGEhoayZcsWdu3aRWFhIa+++qpO3KRJk/D19eX7778HbvbiBwQE8J///IfKykpqa2tZvXo1bm5u9OnTp6WK3yFYWVlhYWGhNQKl0WgoKCjAycmpyXxOTk46o1a5ubk4OztrpdnY2DB58mQmTJhAREQEVlZW97YCol0wNjYmICCAI0eOaKXHx8cTGhp6x7wbNmxg9+7drF69mr59+2q9dvr0aQIDA5VphV5eXgwePJiDBw/e2wqIdqXhvnZrT/Xd3tdu793Ozc3VyuPp6UlxcTEDBw5k0qRJjB49mu7du9/7Sog2z9DQEA8PDy5duqSVfvHiRTp16vSHrj1kyBBUKhVnzpz5Q9cRoqNp0XNS/P39SUxMJDU1Fbg5dNpUXEOMgYEBu3fvZsqUKdjY2GBoaIibmxvR0dHY29s3+V7V1dU6PRc1NTWYmJjcm8q0Q+bm5gBUVVVppVdVVd3xDBlzc/NG8zRcD6CwsJCjR49SXl6Oubk5gYGBjBo1iujoaFkvoGccHBwwNjamqKhIK72wsFCnYdugoKCAd955h/Pnz2NqasrEiRNZs2YNc+fO5dSpUwDs3LkTBwcHvv76a+DmVJxNmzYpz4V+auq+Vl1d/Yfva1ZWVvTs2ZPU1FRSUlJwdHQkLCwMjUbDlStX7mEtRFtnaWmJoaEhFRUVWumVlZVYW1v/7ut26tSJ0NDQRtfkCaHvWrSRUl9fr3UITlOHkN0aV19fzwsvvICrqysHDx7EwsKCr776iokTJ3L8+HE8PDwavcaqVat4++23tdKmTp36q9NNOpKuXbtqjTbdyx7n2w8zunWkpbS0lOvXrzNhwgS6deumNDiFfrn979vAwKDJv/krV65o/ehLTEzEzc2N2bNnK42Uvn37MnfuXFauXElSUhKdO3fmtddeY968eaxdu7b5KiLalC5durTYfc3AwICioiLOnj0LQElJCba2tvTs2VMaKQK4833t15iamjJlyhR++eUX1Gr1PS6ZaC5ymGPLadFGSnJyMt27d1fWMiQnJzNo0CCduJSUFGWeeUxMDD///DPFxcXK4rR//vOf7N69m/Xr1/PGG280+l5LlizhlVde0UrTt52msrKyKCwsVJ437Kx2ew9iYz2Kt7q9dxFurju6U566ujpKS0uxsbH5vcUX7VRxcTG1tbU6U20cHR21Po+/JikpiQceeEB5/sILL7Bjxw62bt0KQHp6OhYWFvzlL3/hq6++kpPX9UR2drbW56hhJ8jb72O/do+6m/uaWq3W2SGsrKzsD0/vEe2PSqVCo9HojJpYWlr+7jVKDg4O2NvbM336dCWt4QfwkiVL+PLLL2UXOaHXWqyREhMTQ1JSEosWLWLs2LE4Ojry0Ucf6TRStm/fTlpaGu+++y5w88YAaG1d3PD81kW1tzMzM8PMzEwrTd+metXW1uoMTavVamX7Vrj57+ji4kJiYmKT1yksLMTNzU1rRMTd3Z3r1683mcfQ0BBbW1sKCgr+WCVEu1NbW0tycjIRERHExsYq6QMGDGDfvn13fR0/Pz+tz5i5ubnO37xGo8HAwOAP9WaK9qW2tlbZJbKBWq3Gzc1Nua8ZGBjg4uJCUlJSk9dpuK+lpaUpaW5ubloNoMLCQp2OFhsbG1k4r4c0Gg05OTl0796dCxcuKOndu3f/3bMFrl+/zpo1a7TShg0bhqmpqbIoXwh91iyNlOrqanJzc6mrqyMvL4/o6GhWrVrFxIkTmT17NkZGRqxevZoZM2bwzDPPMH/+fGxtbdm7dy+LFy9m2rRpSs9CREQEDg4OzJkzh7/+9a9YWFiwdu1aLl26xIQJE5qj+B1aamoqAQEBlJeXU1FRQUBAAHV1dVpTFwYMGIBKpVK+4FNTUxk5ciT+/v5kZWXh5eWFm5sbe/fuVfKEhoaSnZ2NSqXCzMyMwMBATExMtM64EPrjm2++YcWKFZw7d47ExESmTp2Kh4eHsvf/Sy+9hKurK3/5y1+Am+efZGdnk5GRgYmJCQ888ABjxozRGg09cOAATzzxBCkpKSQlJdGlSxdeeOEF9u/ff8cOC9HxpaWl4e/vr3Nfu3U71379+qFWq5WpW2lpaQwfPhw/Pz+ys7Px9PTEzc1Nq2F9673v6tWrODo60qNHD06ePNnidRSt7+jRo0yZMoWcnByuXbtGWFgYdnZ2ypTU4cOHY2Njw08//aTkcXNzA25O7bK0tMTNzY26ujquX79OXV2dTkdew0iedPAJ0UyNlOjoaDw8PDA2NsbBwYHQ0FA+++wz5syZo4yITJs2jdjYWFauXMnQoUNRq9V4e3uzbNkyFi5cqAx5Ojs7Ex0dzbJlyxg5ciQ1NTX06tWLbdu2/epOQUJXSkoKRkZG9OnTRzn0bP/+/Vo9k5aWllq90oWFhRw5coTg4GCCgoKorKwkLi5Oa2G0paUlERERmJqaUl1dTWFhIXv27FFGwoR+2bVrF/b29jz77LM4OzuTnp7O/PnzycnJAcDFxUVrPZmJiQmLFi3C1dWV6upqMjIymD9/PocOHVJi1q5dS319PS+++CKurq4UFxdz4MAB/vGPf7R4/UTbcuHCBYyMjAgPD8fU1JSioiIOHDigc1+7VWFhIfHx8QQFBREUFERFRQXx8fFa97Xi4mLi4uIIDg4mMDCQyspKEhIS5CwLPZWcnIylpSWDBw9WDnPcsGGDMuJhbW2NnZ2dVp65c+cq/+3h4UFQUBAlJSV88cUXLVp2Idojg3o9miOxcePG1i6C0BOrVq1q7SIIPbFs2bLWLoLQE7IJimgpbfm+1lbPgLt9iUNH0KIL54UQQgghhGivZHevltNihzkKIYQQQgghxN2QRooQQgghhBCiTZFGihBCCCGEEKJN0as1Kbfv7iJEc7n17AUhmtOlS5dauwhCTzQcxCyEEC1BRlKEEEIIIYQQbYpejaQIIYQQQgjxe8nuXi1HRlKEEEIIIYQQbco9H0mJjIxk/fr1Ny9ubIyjoyMhISHMnDmTyMhI5cR5gLi4OJYvX86RI0dQq9X4+PgQGRnJwoULMTIyUuJOnTrF66+/zvHjxzEyMmLq1Kn8/e9/x9ra+l4Xv8Orr69n165dHD16FJVKRZcuXXj44Ydxd3e/Y77ExESio6MpLCzEycmJ+++/n+DgYOX1qqoqdu7cSVJSEhUVFXh5eTFlyhS6dOnS3FUSbdS8efNYuHAh7u7uJCcn89prrxEXF9do7OrVq3niiSd00s+fP0+/fv2U53Z2drz55ptMmTIFe3t7Ll++zNKlS9m5c2ez1UO0fb1796Z///5YW1tz/fp1YmJiuHbtWqOxVlZWjBgxAnd3dxwcHDh58iQxMTFaMU5OTgwePBh3d3fs7OzYu3cvJ0+ebImqiHYgMDCQHj16YGpqSmFhIadPn1ZOnW+Kl5cXQUFBWFlZUVlZSVJSEtnZ2Vox5ubmhISE4O7ujpGRERUVFRw/fpySkpJmrI0QbVezjKSMHz+enJwcLl++TFRUFCNGjGDBggVMnDiR2tpaALZu3cqwYcPo1KkTsbGxpKSksGDBAlasWMGMGTOor68HIDs7m9GjR+Pt7c3Ro0eJjo7m3LlzREZGNkfRO7zY2FgOHDjAQw89xIIFC7C1tWXNmjVUVVU1mefy5ct8++239OnThz/96U/06dOHb775hitXrigxP/zwA6mpqcycOZNXX30VX19f1qxZQ2lpaUtUS7QxU6dO5YMPPuCDDz5g0KBBxMXFsXXrVjp16tRo/OLFi+nRo4fy8PHxobCwkK1btyoxJiYm/PTTT3Tt2pXHH3+c3r17M3/+fJ0veqFf/P39GTVqFPHx8axbt45r164xbdo0bGxsGo03MjJCpVJx5MgR8vPzG40xMTGhtLSU/fv3U1FR0ZzFF+2Mn58fvr6+nD59mj179lBVVcXQoUMxNm66z9fR0ZGBAwdy5coVdu/ezZUrV4iIiMDR0VGJMTExYeTIkWg0Gg4ePMjOnTs5c+YMNTU1LVEtIdqkZmmkmJmZ4e7ujpeXF+Hh4SxdupRt27YRFRXFunXrqKysZN68eUyePJk1a9bQu3dvunXrxty5c1m/fj2bN29m06ZNAPz888+YmJjwxRdf4OfnR79+/fjiiy/48ccfSU9Pb47id1j19fUcPHiQUaNGERwcjIeHBzNmzODGjRucPn26yXwHDx7Ex8eHUaNG4erqyqhRo/Dx8eHgwYMA1NTUkJSUxIQJE+jZsyfOzs6MGzcOR0fHJnvORcf20ksvsX79etavX8+FCxd47bXXuHbtGvPmzWs0vqysjLy8POURHh6Og4MD33zzjRIze/ZsHBwcePTRR4mPj+fq1ascOXKEpKSklqqWaIP69u1LYmIiiYmJFBUVERMTQ3l5OWFhYY3Gl5WVERMTw7lz56iurm40Jjc3l3379pGSkkJdXV1zFl+0Mz4+PiQnJ5OVlUVZWZkyw+NOswZ8fX3Jy8sjJSWF8vJyUlJSyM/Px8fHR4nx9/dHpVJx4sQJiouLUalU5OfnU1lZ2RLVEqJNarE1KSNHjiQ0NJQtW7awa9cuCgsLefXVV3XiJk2ahK+vL99//z0A1dXVmJqaak0Ts7CwAODQoUMtU/gOoqioiPLycvz8/JQ0Y2NjevbsyeXLl5vMd+XKFa08cLM3qSFPXV0dGo0GExMTrRgTExPZHlUPmZiYEBYWxt69e7XSY2JiGDBgwF1dY86cOcTGxnL16lUlbcKECRw7doyPP/6YS5cucfz4cV599VWte4PQL4aGhri7u+vcvy5duoSXl1frFEp0WFZWVlhYWJCXl6ekaTQaCgoKcHJyajKfk5OTVh642RC+NY+npyfFxcUMHDiQSZMmMXr0aLp3737vKyFEO9Ki3+7+/v5cvnyZ1NRUAAICApqMa4gZOXIkubm5fPjhh9y4cYPi4mKWLl0KQE5OTpPvVV1dTVlZmdZD34dNy8vLAXTW8lhbWyuvNZXvTnnMzc3p2rUru3fvprS0FI1Gw8mTJ8nMzLzjdUXH5OTkhLGxsc5Umry8PNzc3H41v7u7O2PHjmXdunVa6d26dePBBx/EyMiIhx56iPfff5+XX36Z11577V4WX7QjlpaWGBoa6vQ2q1QqrKysWqlUoqMyNzcH0JkeXV1drbzWVL7b81RVVWnlsbKyomfPnlRUVHDw4EEuXrxIWFgYXbt2vYc1EPeCgYFBm3x0RC26BXF9fb3WP2TDupM7xfXq1Yv169fzyiuvsGTJEoyMjHj55Zdxc3PTWlx/u1WrVvH2229rpc2YMYPHHnvsHtSkfTh16hSbN29Wnj/99NNA49vn/doH/NfyzJw5k02bNvHuu+9iaGiIl5cXYWFhTS5eFR3f7X/fBgYGTf7N3+qJJ56gpKSEn376SSvd0NCQgoIC5s+fj0ajISEhAQ8PDxYuXMh77713T8su2r+7+awJcSddunShT58+yvOGKc73wu3fqQYGBhQVFXH27FkASkpKsLW1pWfPnlrrP4XQJy3aSElOTqZ79+7KqbXJyckMGjRIJy4lJYXAwEDl+WOPPcZjjz1GXl4eVlZWGBgY8Pe///2OQ6FLlizhlVde0Urbs2fPPapJ+xAYGKj1b9CwaUF5eTm2trZKekVFxR13SrOxsdEZEbk9j7OzMy+88ALV1dVUV1dja2vLN998o7UwUOiHwsJCamtrdUZNXF1dm1yofKtZs2axYcMGnZHP3Nxcamtr0Wg0StqFCxdwd3fHxMRE70dK9ZFKpUKj0eiMmlhaWqJSqVqpVKKjyM7OprCwUHne0DF6+8iImZnZHTefuX3UpLE8arVaZ4ewsrKyJjcbEUIftNh0r5iYGJKSkpg6dSpjx47F0dGRjz76SCdu+/btpKWlMXPmTJ3X3NzcsLa2ZuPGjZibmzNmzJgm38/MzAxbW1utx+1rJjo6c3NznJ2dlYebmxs2NjbKVDq42XDJyMigW7duTV6na9euWnng5o/DxvI0/LurVCouXLhAUFDQvaqOaCdqamo4ffo0I0eO1EofMWIER48evWPeIUOG4O3trWxjfqv4+Hh69Oih1QPp7e1NTk6ONFD0lEajITc3V+de1K1bN7KyslqnUKLDqK2tpbKyUnmUlZWhVqu1OmAMDAxwcXHRaszcrrCwUKfTxs3NTStPYWGhzo50NjY2snBe6LVmaaRUV1eTm5tLVlYWp06dYuXKlUyZMoWJEycye/ZsrKysWL16Ndu2beOZZ54hMTGRy5cv8+9//5vIyEimTZvG9OnTlev94x//4NSpU6SmpvLFF18wf/58Vq1ahb29fXMUv8MyMDBgyJAh7N27l6SkJHJycti4cSOmpqZaO+F8//33/PLLL8rzIUOGkJqaSkxMDPn5+cTExJCWlsaQIUOUmAsXLpCSkkJhYSGpqan861//wsXFReuMC6E/Pv/8cyIjI5k9ezZ+fn68//77dO7cma+++gqAt99+m7Vr1+rkmzNnDseOHeP8+fM6r61duxZHR0c+/PBDvL29GTduHIsXL2b16tXNXh/Rdp04cYKQkBCCg4NxdHRk5MiR2NrakpCQAMDQoUN54IEHtPK4urri6uqKqakplpaWuLq6ai1iNjQ0VGKMjIywsbHB1dVVvnMEaWlp+Pv74+npia2tLf3796euro7MzEwlpl+/floddGlpabi5ueHn54eNjQ1+fn64ubmRlpamxKSmpuLk5IS/vz9WVlZ07tyZHj16kJGR0aL1E6ItaZbpXtHR0Xh4eGBsbIyDgwOhoaF89tlnzJkzR9mJZ9q0acTGxrJy5UqGDh2KWq3G29ubZcuWsXDhQq3e0mPHjvHmm29SUVGBv78/q1evZtasWc1R9A5vxIgR1NTUsGXLFtRqNV26dGHevHlaQ9HFxcVa//7dunXj8ccfJzo6mp07d+Lk5MSsWbO0FvSp1WqioqIoKSnB0tKS4OBg7r///juuGxId148//oijoyNvvPEG7u7unD9/nocffljZrcvd3V1nGoOtrS1Tpkxh8eLFjV4zKyuLyZMn816MEFAAABIMSURBVP7773P06FGys7P54osv+Pvf/97s9RFtV0pKCubm5gwaNAgrKyuuX7/O5s2blakzVlZWWtNbAa1zttzd3QkMDKS0tFRp8FpbW2vF9O/fn/79+5OZmcmGDRuavU6i7bpw4QJGRkaEh4djampKUVERBw4cUKZTw83phrcqLCwkPj6eoKAggoKCqKioID4+nqKiIiWmuLiYuLg4goODCQwMpLKykoSEBK3GjxD6xqBej1YX3r4QV4jmMmPGjNYugtATb775ZmsXQegJ2RJXtJRHHnmktYvQpLZ6dlJH7BSWAwaEEEIIIYQQbYo0UoQQQgghhBBtijRShBBCCCGEEG2KNFKEEEIIIYQQbUqLHubY2n7tVHUh7hULC4vWLoLQE7efvyBEczE1NW3tIggh9IheNVKEEEIIIYT4vaTDu+XIdC8hhBBCCCFEmyKNFCGEEEIIIUSb8pume0VGRrJ+/fqbGY2NcXR0JCQkhJkzZxIZGamcJg8QFxfH8uXLOXLkCGq1Gh8fHyIjI1m4cKHWgTMrVqxgx44dJCQkYGpqSklJic77ZmZm8uKLLxITE4OFhQWPPfYYf/vb32R+7O9QX1/Prl27iI+PR6VS0bVrVx5++GHc3d2bzJObm0t0dDTXrl2juLiYKVOmMHToUK2YjIwM9u3bx7Vr1ygrKyMyMpLg4ODmro5ow5588knmz5+Pm5sbFy5cYNmyZcTHxzca+/nnnzNz5kyd9JSUFAYPHgyAn58fb7zxBqGhoXTp0oVly5YpJ4QLERoaiq+vL6amply/fp2jR482+n1yqy5duhAWFoaNjQ3l5eWcPn1a64RvNzc3evXqhZOTE5aWlsTExHD16tVmroloy+rr69m9ezdHjx5FrVbTpUsXHnzwwTt+hwIkJSWxc+dOCgsLcXJyYvz48QQFBSmvV1VVsWvXLs6ePUtFRQVeXl5MnjyZzp07N3eVhGizfvNIyvjx48nJyeHy5ctERUUxYsQIFixYwMSJE6mtrQVg69atDBs2jE6dOhEbG0tKSgoLFixgxYoVzJgxg1sPub9x4waPPPIIzz//fKPvV1dXx4QJE6isrOTQoUNs2LCBH3/8kT/96U+/s8r6LTY2lv379/PQQw+xcOFCbGxsWL16NVVVVU3muXHjBk5OTkyYMAEbG5smYzw9PXnooYeaq+iiHXnwwQdZsWIFH3/8MSNGjODIkSNs2LABLy+vRuOXLl1KYGCg8ggODqaoqIjt27crMZaWlly5coV3332XvLy8lqqKaAeCgoIIDAzk6NGj7NixA7VazZgxYzA2brofzsXFhWHDhpGRkcH27dvJyMhg2LBhODs7KzHGxsYUFxdz9OjRlqiGaAf27dvHwYMHefDBB3n55ZexsbFh7dq1d/wOvXLlCt999x3h4eEsWrSI8PBwvv32W60G8ebNm0lLS2PGjBm88sor+Pj4sHbtWkpLS1uiWkK0Sb+5kWJmZoa7uzteXl6Eh4ezdOlStm3bRlRUFOvWraOyspJ58+YxefJk1qxZQ+/evenWrRtz585l/fr1bN68mU2bNinXe/vtt1m0aFGTve67du3i/PnzfPvtt4SFhTF69Gg++ugj1q5dS1lZ2e+vuR6qr6/nwIEDjB49mpCQEDw8PJg5cyY3btzg9OnTTebr0qULkyZNIiwsrMkv/YCAAO6//35CQkKaq/iiHXn++ef57rvv+Pbbb0lLS+PPf/4z2dnZPPnkk43Gl5eXk5+frzx69+6Nvb09//3vf5WY06dP89Zbb7F161aqq6tbqiqiHQgICCApKYnMzExKSko4dOgQxsbG9OjR4455srOzOXv2LGVlZZw9e5acnBwCAwOVmKysLJ3RFaG/6uvrOXToECNHjiQ4OBh3d3ceffRRampqSEhIaDLfwYMH8fHxYeTIkbi6ujJy5Ei8vb05ePAgADU1NZw9e5YHHniAHj164OzszNixY3FwcODIkSMtVDsh2p57siZl5MiRhIaGsmXLFnbt2kVhYSGvvvqqTtykSZPw9fXl+++/v+trHzlyhKCgIDw9PZW0cePGUV1dzcmTJ+9F8fVGUVER5eXl+Pr6KmnGxsb07NmTy5cvt17BRIdiYmJCaGgosbGxWumxsbH079//rq7xxBNPsH//fq5du9YcRRQdiLW1NZaWlmRnZytpGo2G3NxcXFxcmszn4uJCTk6OVlp2dvYd8wj91tR3aI8ePbhy5UqT+TIzM/Hx8dFK8/X1VfLU1dWh0Wh0OgFNTEzku7kNMjAwaJOPjuieLZz39/fn8uXLpKamAjd7qZqKa4i5G7m5uTrnADg4OGBqakpubu7vL7Aeahh5un3Klo2NjYxKiXvGyckJY2NjCgoKtNILCgpwdXX91fxubm6MGjWKb7/9trmKKDqQhjOJ1Gq1VnpVVdUdzyuysLDQyaNWq+WMI9Gk8vJy4GbD+FbW1tbKa03la+x7tyGPubk5Xbt2Ze/evZSWlqLRaDh16hRXr16V72ah1+7ZOSn19fVaLblb153cKe5uNBb/a9eprq7WmRJSU1ODiYnJb3rv9uzkyZNs3rxZeT537lxA99/z9/w/EeLX3H4PMDAwaPK+cKsZM2ZQWlrKL7/80lxFE+1Y9+7diYiIUJ7v3bu3FUsjOrJTp06xZcsW5XnDdNV78X15+/fujBkz2LRpEytWrMDQ0BAvLy969+5NVlbWH34vIdqre9ZISU5Opnv37sowaHJyMoMGDdKJS0lJ0Zrz+2vc3d11Fi0WFxdTU1Nzx5OWV61axdtvv62VNnPmTB577LG7fu/2rlevXnTt2lV53rCxQVlZGba2tkp6RUVFkwvihfitCgsLqa2t1Rk1cXZ21hldaczjjz/ODz/8QE1NTXMVUbRjV69e5fr168rzht0ibx8ZMTc31xkpuVVjoyaNja4I/RUYGEiXLl2U5w3foeXl5b/pO/TWUZNb89w6IuPk5MTzzz/PjRs3qKqqwtbWlm+//RZHR8d7VR0h2p17Mt0rJiaGpKQkpk6dytixY3F0dOSjjz7Sidu+fTtpaWmNbjXalIiICGVBY4Ndu3ZhZmZGnz59msy3ZMkSSktLtR6PPPLIb6tYO2dubo6zs7PycHNzw8bGRmu6XW1tLRkZGXTr1q31Cio6lJqaGs6cOcPw4cO10ocPH86xY8fumPe+++6jR48eMtVLNKm2tpby8nLlUVJSgkqlwsPDQ4kxNDTE3d39jo3igoICrTwAHh4ed9WQFvqhqe/QtLQ0Jaa2tpaLFy9qdQjerkuXLlp5ANLS0hrNY2pqiq2tLSqVitTU1N/UqStER/ObR1Kqq6vJzc2lrq6OvLw8oqOjWbVqFRMnTmT27NkYGRmxevVqZsyYwTPPPMP8+fOxtbVl7969LF68mGnTpjF9+nTlepmZmRQVFZGZmUldXZ2yQ4a3tzfW1taMHTuWwMBAZs2axYcffkhRURGvvvoq8+bN0+rJuJ2ZmRlmZmZaafo01asxBgYGDB06lL179+Li4oKzszN79+7F1NSUsLAwJe6///0vdnZ2TJgwAbh5E27Y8rWuro7S0lKysrIwMzNTtuusrq7W6t0sKioiKysLS0tLHBwcWrCWoi348ssv+ec//0lCQgLHjx9nzpw5eHl5sW7dOgD+/Oc/4+HhwYsvvqiV7/HHH+fEiROkpKToXNPExAQ/Pz/g5he5h4cHQUFBVFZWcunSpWavk2i7kpOTCQkJoby8nLKyMoKDg5Ufjw0GDx6MSqXi1KlTSp6GsyoyMzPp0qULnp6eREVFKXmMjY21eshtbGxwcHDgxo0bVFZWtlwFRZtgYGDA4MGDiYmJURouMTExmJiY0Lt3byVuw4YN2NnZcf/99wM3P3v/+te/iI2NpVevXpw7d460tDReeOEFJc+FCxeAmxs6XL9+nR07duDi4kK/fv1atI5CtCW/uZESHR2Nh4cHxsbGODg4EBoaymeffcacOXOUwxynTZtGbGwsK1euZOjQoajVary9vVm2bBkLFy7Umof517/+VTkgElB+LMfGxjJ8+HCMjIzYsWMHL7zwAvfdd5/WYY7itxsxYgQ1NTX8+OOPykFUzzzzDObm5kpMSUmJ1v+jsrIy/v73vyvP9+3bx759++jZs6dyk7169SpffvmlEtNwvkXfvn1/08iZ6Bj+97//4eDgwKuvvoqbmxspKSnMnDlT2a3Lzc2NTp06aeWxsbFh4sSJLFu2rNFruru7s2/fPuX5/PnzmT9/PocPH2bKlCnNVhfR9p09exYjIyMGDBiAmZkZBQUF7N69W5meA2BlZaW1JqqgoIADBw4QFhZG7969KS8vZ//+/VqdLQ2H7jVo+MGYnp7O4cOHW6Bmoq0ZPnw4NTU1bN26FbVaTefOnZk3b94dv0O7devGY489xs6dO9m1axdOTk48/vjjWlPJqqqqiIqKorS0FEtLS4KDgxk3bpzW4deibZA1vC3HoP5uVrJ2ED///HNrF0HoicjIyNYugtATjU2tFaI52Nvbt3YRhJ6QjicB93ALYiGEEEIIIYS4F6SRIoQQQgghhGhTpJEihBBCCCGEaFOkkSKEEEIIIYRoU/Rq4bz47aqrq1m1ahVLlizR2dJZiHtJPmuipchnTbQU+awJ8ftJI0XcUVlZGXZ2dpSWlt7xXBoh/ij5rImWIp810VLksybE7yfTvYQQQgghhBBtijRShBBCCCGEEG2KNFKEEEIIIYQQbYo0UsQdmZmZ8eabb8qCP9Hs5LMmWop81kRLkc+aEL+fLJwXQgghhBBCtCkykiKEEEIIIYRoU6SRIoQQQgghhGhTpJEihBBCCCGEaFOkkSKEEEIIIYRoU6SRIoQQQgghhGhTpJEimvTPf/6T7t27Y25uTp8+fTh48GBrF0l0QAcOHGDSpEl4enpiYGDA//73v9YukuiAVq1aRb9+/bCxscHV1ZUHH3yQCxcutHaxRAf05ZdfEhISgq2tLba2tkRERBAVFdXaxRKi3ZFGimjUxo0bWbhwIcuWLeP06dMMGTKE+++/n8zMzNYumuhgKisrCQ0N5R//+EdrF0V0YPv37+fFF18kPj6e3bt3U1tby9ixY6msrGztookOplOnTrz33nucOHGCEydOMHLkSKZMmcK5c+dau2hCtCtyTopo1IABAwgPD+fLL79U0gICAnjwwQdZtWpVK5ZMdGQGBgZs3bqVBx98sLWLIjq4goICXF1d2b9/P0OHDm3t4ogOztHRkQ8//JCnn366tYsiRLshIylCx40bNzh58iRjx47VSh87dixxcXGtVCohhLh3SktLgZs/HoVoLnV1dWzYsIHKykoiIiJauzhCtCvGrV0A0fZcv36duro63NzctNLd3NzIzc1tpVIJIcS9UV9fzyuvvMLgwYMJCgpq7eKIDigpKYmIiAiqqqqwtrZm69atBAYGtnaxhGhXpJEimmRgYKD1vL6+XidNCCHam/nz55OYmMihQ4dauyiig/Lz8yMhIYGSkhJ+/PFH5syZw/79+6WhIsRvII0UocPZ2RkjIyOdUZP8/Hyd0RUhhGhPXnrpJbZv386BAwfo1KlTaxdHdFCmpqZ4e3sD0LdvX44fP86nn37K6tWrW7lkQrQfsiZF6DA1NaVPnz7s3r1bK3337t0MGjSolUolhBC/X319PfPnz2fLli3ExMTQvXv31i6S0CP19fVUV1e3djGEaFdkJEU06pVXXmHWrFn07duXiIgI1qxZQ2ZmJs8991xrF010MBUVFaSnpyvPL126REJCAo6OjnTp0qUVSyY6khdffJH//ve/bNu2DRsbG2Wk2M7ODgsLi1YunehIli5dyv3330/nzp0pLy9nw4YN7Nu3j+jo6NYumhDtimxBLJr0z3/+kw8++ICcnByCgoL4+OOPZatOcc/t27ePESNG6KTPmTOHdevWtXyBRIfU1Hq6r7/+msjIyJYtjOjQnn76afbu3UtOTg52dnaEhITw+uuvM2bMmNYumhDtijRShBBCCCGEEG2KrEkRQgghhBBCtCnSSBFCCCGEEEK0KdJIEUIIIYQQQrQp0kgRQgghhBBCtCnSSBFCCCGEEEK0KdJIEUIIIYQQQrQp0kgRQgghhBBCtCnSSBFCCCGEEEK0KdJIEUIIIYQQQrQp0kgRQgghhBBCtCnSSBFCCCGEEEK0Kf8fs2oUhlC2LXkAAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Your answer here\n", "# EFA\n", "efa = FactorAnalyzer(n_factors=4).fit(explore)\n", "\n", "# Get loadings\n", "loadings = pd.DataFrame(efa.loadings_, index=explore.columns)\n", "\n", "# Plot\n", "import matplotlib.pyplot as plt\n", "plt.figure(figsize=(10, 10))\n", "sns.heatmap(loadings, annot=True, cmap='Grays', fmt='.2f')" ] }, { "cell_type": "markdown", "id": "6dab11b4-c4d3-4949-953e-8733d7cf4867", "metadata": {}, "source": [ "We can see immediately that aspects of the Assertiveness and Social Confidence scales align with the first factor, dominance with the second, adventurousness with the third, while some questions in adventurousness (e.g. 1-4) seem to align with the fourth factor. Examine the variance explained next." ] }, { "cell_type": "code", "execution_count": 22, "id": "4d3020d5-1627-4570-8204-0047283047d7", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "(array([6.4104039 , 4.89054046, 3.33641926, 1.72314879]),\n", " array([0.1602601 , 0.12226351, 0.08341048, 0.04307872]),\n", " array([0.1602601 , 0.28252361, 0.36593409, 0.40901281]))" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "efa.get_factor_variance()" ] }, { "cell_type": "markdown", "id": "4736efd0-8f9b-490c-9c3a-58f69d02905b", "metadata": {}, "source": [ "This explains around 40%, with the most common from the first two factors. Perhaps we could do away with a 4 factor solution and retain a 3 factor one? Fit that below." ] }, { "cell_type": "code", "execution_count": 23, "id": "0353f475-f766-461d-a849-ae48683e980b", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Your answer here\n", "# EFA\n", "efa = FactorAnalyzer(n_factors=3).fit(explore)\n", "\n", "# Get loadings\n", "loadings = pd.DataFrame(efa.loadings_, index=explore.columns)\n", "\n", "# Plot\n", "import matplotlib.pyplot as plt\n", "plt.figure(figsize=(10, 10))\n", "sns.heatmap(loadings, annot=True, cmap='Grays', fmt='.2f')" ] }, { "cell_type": "markdown", "id": "127a044f-b7a0-4939-b87a-941a0b1544c1", "metadata": {}, "source": [ "Examine the variance of this fit." ] }, { "cell_type": "code", "execution_count": 24, "id": "ddcb22b6-6cce-4925-9185-a64354120a58", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "(array([6.54110144, 5.0344487 , 3.69496587]),\n", " array([0.16352754, 0.12586122, 0.09237415]),\n", " array([0.16352754, 0.28938875, 0.3817629 ]))" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "efa.get_factor_variance()" ] }, { "cell_type": "markdown", "id": "71702f1a-a50c-484d-9428-b5e1f7ec900e", "metadata": {}, "source": [ "This loses only around 2% variance and has some more informative factors. We'll then say there are three factors underpinning this four-questionnaire test. The first captures Assertiveness and Social Confidence, the second dominance, and the third adventurousness.\n", "\n", "Now, how might we translate this setup into a CFA model that we can test on the other half of the data?\n", "\n", "One straightforward way would be to use a coarse measure as just described above - that is, *all* questions for assertiveness and social confidence go on one factor, all dominance questions on another, and all adventurousness questions on the final one. \n", "\n", "Build that below, but using the other slice of the data to confirm whether this structure 'holds'. Check the fit statistics." ] }, { "cell_type": "code", "execution_count": 25, "id": "6169c9b7-aa08-4b95-ad58-ef34d88e0c57", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
DoFDoF Baselinechi2chi2 p-valuechi2 BaselineCFIGFIAGFINFITLIRMSEAAICBICLogLik
Value7377804041.3731730.09776.7343350.6327140.5866340.5625160.5866340.6112850.094506149.930922500.2399068.034539
\n", "
" ], "text/plain": [ " DoF DoF Baseline chi2 chi2 p-value chi2 Baseline CFI \\\n", "Value 737 780 4041.373173 0.0 9776.734335 0.632714 \n", "\n", " GFI AGFI NFI TLI RMSEA AIC \\\n", "Value 0.586634 0.562516 0.586634 0.611285 0.094506 149.930922 \n", "\n", " BIC LogLik \n", "Value 500.239906 8.034539 " ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "# Model string\n", "mdspec = \"\"\"\n", "latent1 =~ AS1 + AS2 + AS3 + AS4 + AS5 + AS6 + AS7 + AS8 + AS9 + AS10 + SC1 + SC2 + SC3 + SC4 + SC5 + SC6 + SC7 + SC8 + SC9 + SC10\n", "latent2 =~ DO1 + DO2 + DO3 + DO4 + DO5 + DO6 + DO7 + DO8 + DO9 + DO10\n", "latent3 =~ AD1 + AD2 + AD3 + AD4 + AD5 + AD6 + AD7 + AD8 + AD9 + AD10\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(confirm) \n", "\n", "# Fit statistics\n", "sem.calc_stats(model)" ] }, { "cell_type": "markdown", "id": "99cc7fee-7f59-4824-9109-7cac51fafb65", "metadata": {}, "source": [ "That does not look great! Inspect whether the individual coefficients are associated in the way we'd expect:" ] }, { "cell_type": "code", "execution_count": 26, "id": "df1bd5df-3f58-46ca-bdb0-35fedd6ec22b", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
lvaloprvalEstimateEst. StdStd. Errz-valuep-value
0AS1~latent11.0000000.704852---
1AS2~latent10.8183800.6091850.06332412.9236570.0
2AS3~latent10.8262510.5966970.06524412.6640050.0
3AS4~latent10.8035850.6458720.05872413.684130.0
4AS5~latent10.6688980.4960570.06334710.5592470.0
5AS6~latent10.6210500.4714900.06184210.0425510.0
6AS7~latent1-0.679325-0.5188000.061552-11.0366180.0
7AS8~latent1-0.452216-0.3448300.061406-7.3643760.0
8AS9~latent1-0.400243-0.2693130.069506-5.7583910.0
9AS10~latent1-0.311342-0.2297520.063346-4.9149590.000001
10SC1~latent10.8948730.5898800.07146312.5221080.0
11SC2~latent11.0867370.6574650.0780513.9236680.0
12SC3~latent11.0292090.6006780.08074212.7468090.0
13SC4~latent10.9574030.6513460.06939113.7972920.0
14SC5~latent10.8903990.6318870.06647413.3946720.0
15SC6~latent1-0.957331-0.6149290.073398-13.0429580.0
16SC7~latent1-0.773334-0.5997380.060762-12.7272660.0
17SC8~latent1-0.874095-0.5796280.071015-12.308520.0
18SC9~latent1-0.910533-0.6012520.071365-12.7587530.0
19SC10~latent1-0.769383-0.5332470.067851-11.339380.0
20DO1~latent21.0000000.538283---
21DO2~latent21.1428250.5927630.11391510.0322410.0
22DO3~latent21.1446160.6660740.10615210.7828120.0
23DO4~latent21.2857730.7440640.11210611.469270.0
24DO5~latent21.2824940.6886270.11666410.993040.0
25DO6~latent21.0356380.6104910.10130110.2233550.0
26DO7~latent20.9962470.5713670.1017299.7931660.0
27DO8~latent20.9570890.5981860.09484210.091370.0
28DO9~latent21.0406930.6045910.10242610.1604430.0
29DO10~latent21.1776560.6702890.10881210.8228280.0
30AD1~latent31.0000000.478837---
31AD2~latent30.5961620.3421520.0940786.3368980.0
32AD3~latent30.4370530.2686540.0840175.2019670.0
33AD4~latent30.7694670.4270060.1031457.4600150.0
34AD5~latent3-1.349486-0.6987100.13627-9.9029980.0
35AD6~latent3-1.606016-0.8194330.152228-10.5500870.0
36AD7~latent3-1.390536-0.7783190.134248-10.3579740.0
37AD8~latent3-1.274979-0.6433480.133857-9.5249670.0
38AD9~latent3-0.792595-0.4070090.109885-7.2129670.0
39AD10~latent3-1.226555-0.6344940.129663-9.4595470.0
\n", "
" ], "text/plain": [ " lval op rval Estimate Est. Std Std. Err z-value p-value\n", "0 AS1 ~ latent1 1.000000 0.704852 - - -\n", "1 AS2 ~ latent1 0.818380 0.609185 0.063324 12.923657 0.0\n", "2 AS3 ~ latent1 0.826251 0.596697 0.065244 12.664005 0.0\n", "3 AS4 ~ latent1 0.803585 0.645872 0.058724 13.68413 0.0\n", "4 AS5 ~ latent1 0.668898 0.496057 0.063347 10.559247 0.0\n", "5 AS6 ~ latent1 0.621050 0.471490 0.061842 10.042551 0.0\n", "6 AS7 ~ latent1 -0.679325 -0.518800 0.061552 -11.036618 0.0\n", "7 AS8 ~ latent1 -0.452216 -0.344830 0.061406 -7.364376 0.0\n", "8 AS9 ~ latent1 -0.400243 -0.269313 0.069506 -5.758391 0.0\n", "9 AS10 ~ latent1 -0.311342 -0.229752 0.063346 -4.914959 0.000001\n", "10 SC1 ~ latent1 0.894873 0.589880 0.071463 12.522108 0.0\n", "11 SC2 ~ latent1 1.086737 0.657465 0.07805 13.923668 0.0\n", "12 SC3 ~ latent1 1.029209 0.600678 0.080742 12.746809 0.0\n", "13 SC4 ~ latent1 0.957403 0.651346 0.069391 13.797292 0.0\n", "14 SC5 ~ latent1 0.890399 0.631887 0.066474 13.394672 0.0\n", "15 SC6 ~ latent1 -0.957331 -0.614929 0.073398 -13.042958 0.0\n", "16 SC7 ~ latent1 -0.773334 -0.599738 0.060762 -12.727266 0.0\n", "17 SC8 ~ latent1 -0.874095 -0.579628 0.071015 -12.30852 0.0\n", "18 SC9 ~ latent1 -0.910533 -0.601252 0.071365 -12.758753 0.0\n", "19 SC10 ~ latent1 -0.769383 -0.533247 0.067851 -11.33938 0.0\n", "20 DO1 ~ latent2 1.000000 0.538283 - - -\n", "21 DO2 ~ latent2 1.142825 0.592763 0.113915 10.032241 0.0\n", "22 DO3 ~ latent2 1.144616 0.666074 0.106152 10.782812 0.0\n", "23 DO4 ~ latent2 1.285773 0.744064 0.112106 11.46927 0.0\n", "24 DO5 ~ latent2 1.282494 0.688627 0.116664 10.99304 0.0\n", "25 DO6 ~ latent2 1.035638 0.610491 0.101301 10.223355 0.0\n", "26 DO7 ~ latent2 0.996247 0.571367 0.101729 9.793166 0.0\n", "27 DO8 ~ latent2 0.957089 0.598186 0.094842 10.09137 0.0\n", "28 DO9 ~ latent2 1.040693 0.604591 0.102426 10.160443 0.0\n", "29 DO10 ~ latent2 1.177656 0.670289 0.108812 10.822828 0.0\n", "30 AD1 ~ latent3 1.000000 0.478837 - - -\n", "31 AD2 ~ latent3 0.596162 0.342152 0.094078 6.336898 0.0\n", "32 AD3 ~ latent3 0.437053 0.268654 0.084017 5.201967 0.0\n", "33 AD4 ~ latent3 0.769467 0.427006 0.103145 7.460015 0.0\n", "34 AD5 ~ latent3 -1.349486 -0.698710 0.13627 -9.902998 0.0\n", "35 AD6 ~ latent3 -1.606016 -0.819433 0.152228 -10.550087 0.0\n", "36 AD7 ~ latent3 -1.390536 -0.778319 0.134248 -10.357974 0.0\n", "37 AD8 ~ latent3 -1.274979 -0.643348 0.133857 -9.524967 0.0\n", "38 AD9 ~ latent3 -0.792595 -0.407009 0.109885 -7.212967 0.0\n", "39 AD10 ~ latent3 -1.226555 -0.634494 0.129663 -9.459547 0.0" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "model.inspect(std_est=True).query('op == \"~\"')" ] }, { "cell_type": "markdown", "id": "62361912-5761-46dc-bc46-bdb0b8467b51", "metadata": {}, "source": [ "These seem to match the pattern seen in the loadings, broadly, but our model fit suggests this is not good. As a final push, we could consider trying to reflect the loadings we saw from EFA more closely. From the loadings matrix we can actually discern which factor each question had the highest affinity with by using the `.idxmax(axis='columns')` command, like so:\n", "\n", "`loadings.idxmax(axis='columns')`\n", "\n", "Run that below and examine the output." ] }, { "cell_type": "code", "execution_count": 27, "id": "3894d07a-5eec-40b4-afca-318e9ac87d98", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "outputs": [ { "data": { "text/plain": [ "AS1 0\n", "AS2 0\n", "AS3 1\n", "AS4 0\n", "AS5 1\n", "AS6 1\n", "AS7 2\n", "AS8 2\n", "AS9 2\n", "AS10 2\n", "SC1 0\n", "SC2 0\n", "SC3 0\n", "SC4 0\n", "SC5 0\n", "SC6 1\n", "SC7 2\n", "SC8 1\n", "SC9 1\n", "SC10 2\n", "AD1 1\n", "AD2 0\n", "AD3 1\n", "AD4 1\n", "AD5 2\n", "AD6 2\n", "AD7 2\n", "AD8 2\n", "AD9 2\n", "AD10 2\n", "DO1 1\n", "DO2 1\n", "DO3 1\n", "DO4 1\n", "DO5 1\n", "DO6 1\n", "DO7 1\n", "DO8 1\n", "DO9 1\n", "DO10 1\n", "dtype: int64" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# You answer here\n", "loadings.idxmax(axis='columns')" ] }, { "cell_type": "markdown", "id": "bd97e26e-aa98-464f-9e78-ccd90fce31d2", "metadata": {}, "source": [ "What is interesting here is that reveals a rather disparate pattern for the assertiveness scale. Some questions loading on factor 0 (the first) while others are more associated with the second factor, which was related to dominance. Consider AS3 - \"automatically take charge\". Should we be surprised this is more closely associated with a different factor? We can try to recreate this by tying each question as shown above to a specific factor. This is a tricky process. See if you can build this model below and examnine its fit statistic. It will more closely resemble the EFA, and may emerge as a better model." ] }, { "cell_type": "code", "execution_count": 28, "id": "7ec08844-4b6b-4526-a388-270d4c4cf9b4", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "hide-input" ] }, "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", "
DoFDoF Baselinechi2chi2 p-valuechi2 BaselineCFIGFIAGFINFITLIRMSEAAICBICLogLik
Value7377804906.107060.09776.7343350.5365980.4981850.4689070.4981850.5095610.106154146.492616496.80169.753692
\n", "
" ], "text/plain": [ " DoF DoF Baseline chi2 chi2 p-value chi2 Baseline CFI \\\n", "Value 737 780 4906.10706 0.0 9776.734335 0.536598 \n", "\n", " GFI AGFI NFI TLI RMSEA AIC BIC \\\n", "Value 0.498185 0.468907 0.498185 0.509561 0.106154 146.492616 496.8016 \n", "\n", " LogLik \n", "Value 9.753692 " ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Your answer here\n", "mdspec = \"\"\"\n", "latent1 =~ AS1 + AS2 + AS4 + SC1 + SC2 + SC3 + SC4 + SC5 + AD2\n", "latent2 =~ DO1 + DO2 + DO3 + DO4 + DO5 + DO6 + DO7 + DO8 + DO9 + DO10 + AS3 + AS5 + AS6 + SC6 + SC8 + SC9 + AD1 + AD3 + AD4\n", "latent3 =~ AS7 + AS8 + AS9 + AS10 + SC7 + SC10 + AD5 + AD6 + AD7 + AD8 + AD9 + AD10\n", "\"\"\"\n", "\n", "# Create model\n", "model = sem.Model(mdspec)\n", "\n", "# Fit it\n", "model.fit(confirm) \n", "\n", "# Fit statistics\n", "sem.calc_stats(model)" ] }, { "cell_type": "markdown", "id": "fe2dfc2b-2dc1-41c9-b7ee-b8333d446ac8", "metadata": {}, "source": [ "Amusingly, this is *even worse*. Despite our best efforts, we're unable to create a solid, stable set of latent variables that underpin this collection of data. This is a common experience, and the field of psychology has numerous measurement issues to which poor latent factor models contribute. Nonetheless, EFA and CFA are incredibly powerful approaches but they need to be treated with a lot of circumspection." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.2" } }, "nbformat": 4, "nbformat_minor": 5 }