https://www.nitingupta.com/tags/data-cleaning/ Data Cleaning Source Themes Academic (https://sourcethemes.com/academic/)en-us© Nitin Gupta. All Rights Reserved.Sun, 25 Dec 2016 00:00:00 +0000 https://www.nitingupta.com/img/icon-192.png https://www.nitingupta.com/tags/data-cleaning/ https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/ Sun, 25 Dec 2016 00:00:00 +0000 https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/ <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/htmlwidgets/htmlwidgets.js"></script> <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/jquery/jquery-3.6.0.min.js"></script> <link href="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/leaflet/leaflet.css" rel="stylesheet" /> <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/leaflet/leaflet.js"></script> <link href="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/leafletfix/leafletfix.css" rel="stylesheet" /> <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/proj4/proj4.min.js"></script> <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/Proj4Leaflet/proj4leaflet.js"></script> <link href="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/rstudio_leaflet/rstudio_leaflet.css" rel="stylesheet" /> <script src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/leaflet-binding/leaflet.js"></script> <div id="other-posts-in-this-series" class="section level4"> <h4>Other posts in this series:</h4> <ul> <li><a href="https://www.nitingupta.com/casestudies/ames-housing-part2-eda/">Ames Housing - Part 2 - Walkthrough - An Exploratory Data Analysis</a></li> <li><a href="https://www.nitingupta.com/casestudies/ames-housing-part3-models/">Ames Housing - Part 3 - Closing the deal - Building Models</a></li> </ul> <p>In this case study, we use the <a href="http://www.amstat.org/publications/jse/v19n3/decock.pdf">Ames Housing dataset</a> to predict the sale prices of houses. The case study has three posts:</p> <ol style="list-style-type: decimal"> <li>In this post, we build a data pipeline: a series of steps that fixes errors in the data, fills in missing values, creates new features and drops features that we do not need. At the end, we save the finalized dataset.</li> <li>In <a href="https://www.nitingupta.com/casestudies/ames-housing-part2-eda/">Part 2</a>, we explore the finalized dataset to find which features affect the sale price.</li> <li>In <a href="https://www.nitingupta.com/casestudies/ames-housing-part3-models/">Part 3</a>, we build models that predict the prices. A model is a formula that a computer learns from examples.</li> </ol> </div> <div id="neighborhoods-of-ames" class="section level2"> <h2>Neighborhoods of Ames</h2> <p>Location is very important for the price of a house. The version of the dataset that we use (from Kaggle) does not have the location of each property. But the full dataset by Dean De Cock, which is in the R package <code>modeldata</code>, has the latitude and longitude of 2930 properties. We use it to show where the neighborhoods are.</p> <p>On this map, each dot is a property. The color shows the type of its neighborhood. Later in this post, we put the neighborhoods into five types by their median sale price: <strong>Type1</strong> has the most expensive neighborhoods, and <strong>Type5</strong> has the cheapest. The labels show the short names of the neighborhoods that the dataset uses. You can zoom in, move the map, and click a label to see the details of a neighborhood:</p> <pre class="r"><code>library(leaflet) # Neighborhood types, from the most expensive (Type1) to the cheapest (Type5), by median SalePrice. # The same groups are used to create NeighborhoodType later in this post. lNeighborhoodTypes &lt;- list(Type1 = c(&quot;StoneBr&quot;, &quot;NridgHt&quot;, &quot;NoRidge&quot;), Type2 = c(&quot;Veenker&quot;, &quot;Timber&quot;, &quot;Somerst&quot;), Type3 = c(&quot;Crawfor&quot;, &quot;CollgCr&quot;, &quot;ClearCr&quot;, &quot;Blmngtn&quot;, &quot;Gilbert&quot;, &quot;NWAmes&quot;, &quot;SawyerW&quot;), Type4 = c(&quot;Mitchel&quot;, &quot;NPkVill&quot;, &quot;NAmes&quot;, &quot;SWISU&quot;, &quot;Sawyer&quot;, &quot;Blueste&quot;, &quot;BrkSide&quot;, &quot;Edwards&quot;, &quot;OldTown&quot;), Type5 = c(&quot;IDOTRR&quot;, &quot;BrDale&quot;, &quot;MeadowV&quot;)) # modeldata::ames spells out the neighborhood names; this dataset uses short codes dfNeighborhoodNames &lt;- tribble( ~Name, ~Code, &quot;Bloomington_Heights&quot;, &quot;Blmngtn&quot;, &quot;Blueste&quot;, &quot;Blueste&quot;, &quot;Briardale&quot;, &quot;BrDale&quot;, &quot;Brookside&quot;, &quot;BrkSide&quot;, &quot;Clear_Creek&quot;, &quot;ClearCr&quot;, &quot;College_Creek&quot;, &quot;CollgCr&quot;, &quot;Crawford&quot;, &quot;Crawfor&quot;, &quot;Edwards&quot;, &quot;Edwards&quot;, &quot;Gilbert&quot;, &quot;Gilbert&quot;, &quot;Iowa_DOT_and_Rail_Road&quot;, &quot;IDOTRR&quot;, &quot;Meadow_Village&quot;, &quot;MeadowV&quot;, &quot;Mitchell&quot;, &quot;Mitchel&quot;, &quot;North_Ames&quot;, &quot;NAmes&quot;, &quot;Northridge&quot;, &quot;NoRidge&quot;, &quot;Northpark_Villa&quot;, &quot;NPkVill&quot;, &quot;Northridge_Heights&quot;, &quot;NridgHt&quot;, &quot;Northwest_Ames&quot;, &quot;NWAmes&quot;, &quot;Old_Town&quot;, &quot;OldTown&quot;, &quot;South_and_West_of_Iowa_State_University&quot;, &quot;SWISU&quot;, &quot;Sawyer&quot;, &quot;Sawyer&quot;, &quot;Sawyer_West&quot;, &quot;SawyerW&quot;, &quot;Somerset&quot;, &quot;Somerst&quot;, &quot;Stone_Brook&quot;, &quot;StoneBr&quot;, &quot;Timberland&quot;, &quot;Timber&quot;, &quot;Veenker&quot;, &quot;Veenker&quot;) %&gt;% mutate(Type = map_chr(Code, function(code) names(lNeighborhoodTypes)[map_lgl(lNeighborhoodTypes, ~ code %in% .x)])) # properties with their location; neighborhoods that are not in this dataset are left out dfLocations &lt;- modeldata::ames %&gt;% transmute(Name = as.character(Neighborhood), Longitude, Latitude, SalePrice = Sale_Price) %&gt;% inner_join(dfNeighborhoodNames, by = &quot;Name&quot;) %&gt;% mutate(Type = factor(Type, levels = names(lNeighborhoodTypes))) nLeftOut &lt;- nrow(modeldata::ames) - nrow(dfLocations) # one label per neighborhood, at the median location of its properties dfCenters &lt;- dfLocations %&gt;% group_by(Name, Code, Type) %&gt;% summarise(Longitude = median(Longitude), Latitude = median(Latitude), Properties = n(), MedianPrice = median(SalePrice), .groups = &quot;drop&quot;) %&gt;% mutate(Popup = glue::glue(&quot;&lt;b&gt;{Code}&lt;/b&gt;: {str_replace_all(Name, &#39;_&#39;, &#39; &#39;)}&lt;br&gt;&quot;, &quot;Neighborhood type: {Type}&lt;br&gt;&quot;, &quot;Properties: {Properties}&lt;br&gt;&quot;, &quot;Median sale price: {scales::dollar(MedianPrice)}&quot;)) palType &lt;- colorFactor(quant_pal(&quot;metro&quot;)(5), domain = levels(dfLocations$Type)) leaflet(dfLocations, width = &quot;100%&quot;, height = 600) %&gt;% addTiles() %&gt;% addCircleMarkers(~Longitude, ~Latitude, color = ~palType(Type), radius = 3, stroke = FALSE, fillOpacity = 0.6) %&gt;% addCircleMarkers(data = dfCenters, ~Longitude, ~Latitude, popup = ~Popup, radius = 6, color = &quot;white&quot;, weight = 1, fillColor = ~palType(Type), fillOpacity = 1, label = ~Code, labelOptions = labelOptions(noHide = TRUE, direction = &quot;right&quot;, textOnly = TRUE, style = list(&quot;font-weight&quot; = &quot;bold&quot;, &quot;font-size&quot; = &quot;12px&quot;, &quot;color&quot; = &quot;#222&quot;, &quot;text-shadow&quot; = &quot;0 0 3px #fff, 0 0 3px #fff&quot;))) %&gt;% addLegend(&quot;bottomright&quot;, pal = palType, values = ~Type, title = &quot;Neighborhood type&quot;, opacity = 1)</code></pre> <div id="htmlwidget-1" style="width:100%;height:600px;" class="leaflet html-widget"></div> <script type="application/json" data-for="htmlwidget-1">{"x":{"options":{"crs":{"crsClass":"L.CRS.EPSG3857","code":null,"proj4def":null,"projectedBounds":null,"options":{}}},"calls":[{"method":"addTiles","args":["https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png",null,null,{"minZoom":0,"maxZoom":18,"tileSize":256,"subdomains":"abc","errorTileUrl":"","tms":false,"noWrap":false,"zoomOffset":0,"zoomReverse":false,"opacity":1,"zIndex":1,"detectRetina":false,"attribution":"&copy; <a href=\"https://openstreetmap.org/copyright/\">OpenStreetMap<\/a>, <a href=\"https://opendatacommons.org/licenses/odbl/\">ODbL<\/a>"}]},{"method":"addCircleMarkers","args":[[42.054035,42.053014,42.052659,42.051245,42.060899,42.060779,42.062978,42.060728,42.06112,42.059193,42.05848,42.057419,42.058151,42.057268,42.059169,42.061239,42.06035139999999,42.05885,42.056673,42.054453,42.050346,42.05027,42.049297,42.055147,42.054592,42.055227,42.053395,42.056375,42.055318,42.051835,42.051684,42.051685,42.049672,42.050683,42.050681,42.049811,42.062352,42.062109,42.0611331,42.0612885,42.060872,42.0608079,42.060879,42.060298,42.059617,42.058388,42.058599,42.057102,42.057164,42.058252,42.057181,42.061928,42.06257,42.062303,42.061443,42.060977,42.061193,42.060936,42.057103,42.054386,42.053243,42.051721,42.051674,42.052457,42.052437,42.050909,42.05306,42.051827,42.051653,42.050084,42.052347,42.051429,42.036255,42.036113,42.036026,42.03548,42.036659,42.034589,42.0347,42.036633,42.034555,42.034678,42.03607540000001,42.036535,42.034793,42.035685,42.034856,42.03457,42.035333,42.034603,42.048253,42.046825,42.045773,42.047452,42.048086,42.048566,42.047899,42.047876,42.047782,42.0471748,42.046145,42.046145,42.046424,42.046114,42.0470316,42.04638,42.034561,42.048606,42.046322,42.043038,42.043306,42.043306,42.043329,42.043822,42.043821,42.046574,42.042205,42.043363,42.039085,42.038971,42.040326,42.041146,42.040391,42.0418387,42.039211,42.039211,42.038538,42.035256,42.035482,42.035943,42.036095,42.037424,42.048494,42.0459375,42.049514,42.047447,42.044895,42.043984,42.042908,42.0430839,42.044847,42.043969,42.044052,42.044176,42.044883,42.0435919,42.042308,42.040396,42.04137,42.038515,42.039841,42.038443,42.038521,42.038249,42.0381282,42.035869,42.0372097,42.040085,42.041119,42.035867,42.03594,42.0347571,42.034848,42.034748,42.034851,42.036873,42.037093,42.033313,42.032134,42.031363,42.031381,42.0303783,42.033279,42.033644,42.032391,42.03128,42.0303113,42.030383,42.030467,42.030388,42.030331,42.028099,42.029146,42.028145,42.028083,42.03005,42.028164,42.02916,42.028165,42.033642,42.030187,42.033621,42.033563,42.0336,42.033665,42.033631,42.031556,42.031337,42.032302,42.027986,42.028888,42.028875,42.026914,42.020693,42.024141,42.026741,42.033454,42.032432,42.028256,42.026752,42.02616,42.025255,42.022825,42.034451,42.033386,42.033387,42.033008,42.03237,42.029867,42.03069,42.029924,42.029731,42.030519,42.031187,42.03136,42.03128,42.030082,42.03009,42.03387,42.032164,42.026547,42.024673,42.023079,42.028226,42.026533,42.024208,42.025173,42.02526,42.0247497,42.024126,42.025019,42.024162,42.03374,42.032076,42.033783,42.03068,42.0311198,42.025747,42.024614,42.024604,42.024604,42.030601,42.0298532,42.024318,42.021321,42.018411,42.01867,42.022186,42.022045,42.021395,42.021133,42.018662,42.01906899999999,42.016121,42.018721,42.017749,42.016275,42.016253,42.01404,42.014092,42.014023,42.0203585,42.021245,42.021041,42.020635,42.021163,42.020306,42.019849,42.01962899999999,42.020196,42.021793,42.021583,42.017726,42.022674,42.022702,42.022089,42.017737,42.0177987,42.0163944,42.019091,42.019093,42.016226,42.014937,42.014682,42.011005,42.010606,42.009345,42.009337,42.010797,42.022772,42.021414,42.021474,42.021456,42.022653,42.021327,42.02152,42.00868699999999,41.99706,41.997052,41.996065,41.994784,41.994013,41.997741,41.995346,41.995278,41.994284,41.99353,41.995489,41.99435,41.992989,41.99321,41.990934,41.993253,41.991004,41.991861,41.991758,41.991876,41.992212,41.992119,41.9917,41.991824,41.991558,41.988729,41.9889358,41.98766,41.986647,41.9913,41.989848,41.989185,42.0530363,42.05139,42.049322,42.059956,42.063294,42.060896,42.062956,42.060598,42.062903,42.061981,42.061633,42.061178,42.059283,42.059129,42.059161,42.058274,42.058059,42.059645,42.058194,42.05793,42.058173,42.057854,42.058201,42.058747,42.059242,42.060265,42.059817,42.058439,42.056796,42.053532,42.055408,42.055081,42.055572,42.05184,42.052507,42.052102,42.052719,42.05287999999999,42.052915,42.050345,42.0496539,42.049394,42.049106,42.051619,42.0503771,42.049406,42.053166,42.052355,42.052448,42.05267000000001,42.04891,42.056398,42.054626,42.055228,42.055184,42.055154,42.05337,42.053306,42.053307,42.0564419,42.056461,42.052758,42.052338,42.051683,42.051661,42.051823,42.051822,42.050615,42.050705,42.050705,42.050682,42.050533,42.050213,42.05009,42.050134,42.050222,42.050397,42.0488934,42.048486,42.0623099,42.063304,42.063025,42.061902,42.06128,42.062991,42.062992,42.062994,42.062942,42.062258,42.062257,42.062776,42.062789,42.0623985,42.062108,42.060934,42.060353,42.061382,42.061437,42.060576,42.060575,42.061678,42.060426,42.060178,42.0599123,42.0598989,42.061404,42.059241,42.058782,42.058886,42.0592,42.059177,42.059179,42.059183,42.058339,42.058336,42.057177,42.05851,42.057164,42.057089,42.057019,42.056983,42.057277,42.057467,42.06232,42.063183,42.062058,42.062404,42.062562,42.062557,42.06230499999999,42.062266,42.062081,42.062086,42.061562,42.059377,42.061739,42.061149,42.061114,42.061656,42.061365,42.063283,42.058368,42.05859,42.059481,42.059068,42.059605,42.057839,42.057834,42.057833,42.057098,42.057093,42.056955,42.058668,42.053749,42.053911,42.055626,42.053441,42.053041,42.052668,42.050411,42.049701,42.05161,42.050215,42.049921,42.050971,42.054282,42.055036,42.054302,42.054152,42.054151,42.053212,42.053258,42.053108,42.051804,42.051797,42.051793,42.051645,42.051648,42.051652,42.050406,42.051717,42.051795,42.052337,42.052346,42.052127,42.05142,42.051378,42.052122,42.051414,42.051247,42.051251,42.050895,42.037557,42.037611,42.037661,42.036001,42.036011,42.035282,42.037923,42.038127,42.037756,42.037434,42.036852,42.03607,42.036905,42.036576,42.034533,42.035518,42.037607,42.035968,42.035756,42.0365385,42.035287,42.035806,42.034747,42.034615,42.035289,42.03466,42.035331,42.034595,42.034602,42.034581,42.037236,42.047669,42.047627,42.048416,42.048488,42.047338,42.047253,42.047101,42.046144,42.046418,42.047171,42.043782,42.043553,42.043019,42.034509,42.047708,42.048009,42.04896,42.048761,42.04876,42.047652,42.046476,42.046476,42.045806,42.045799,42.043236,42.043842,42.043673,42.044693,42.048796,42.0467129,42.046635,42.044821,42.043935,42.043145,42.043135,42.042241,42.044843,42.044774,42.044926,42.04488,42.040645,42.038492,42.0407264,42.040011,42.039902,42.040888,42.041104,42.039463,42.038596,42.035489,42.035443,42.03464,42.036182,42.034741,42.037452,42.035951,42.036178,42.049306,42.048294,42.046827,42.048608,42.048724,42.04806,42.04579,42.044886,42.043966,42.044125,42.043991,42.044083,42.045473,42.044914,42.04297,42.042233,42.043238,42.040931,42.040857,42.0404571,42.039674,42.0385265,42.038356,42.038398,42.037122,42.0372232,42.034918,42.034686,42.034754,42.036223,42.035829,42.034746,42.0380659,42.037679,42.037144,42.037328,42.036015,42.036002,42.034668,42.034574,42.0347572,42.0347398,42.035257,42.03528,42.0417879,42.04015,42.040295,42.04124,42.038509,42.038349,42.039047,42.040766,42.038811,42.037129,42.035712,42.035787,42.035841,42.037233,42.03602,42.035968,42.03572,42.035706,42.037214,42.03475419999999,42.034858,42.037049,42.038948,42.038257,42.035287,42.0342224,42.033273,42.0320748,42.0323259,42.032396,42.032202,42.032108,42.0319053,42.031305,42.0304742,42.030327,42.0303538,42.030282,42.0310603,42.033381,42.033579,42.032048,42.032301,42.032227,42.030523,42.030196,42.0296722,42.029007,42.0290352,42.028957,42.029305,42.029001,42.0281106,42.028104,42.0276972,42.028556,42.0282044,42.027062,42.0272,42.027227,42.023218,42.023222,42.03399599999999,42.033558,42.0333765,42.031564,42.031542,42.033406,42.033616,42.03329,42.033347,42.033496,42.033474,42.031395,42.03252,42.03246,42.032359,42.031292,42.032375,42.03235,42.0310861,42.030062,42.026759,42.028948,42.027854,42.027998,42.027039,42.026956,42.025306,42.023085,42.025796,42.0260296,42.0242487,42.024143,42.022828,42.032681,42.028101,42.028098,42.028049,42.0269586,42.022825,42.022978,42.033376,42.033633,42.032367,42.032388,42.029863,42.031342,42.030628,42.031138,42.033427,42.033277,42.032302,42.032652,42.032564,42.032503,42.026548,42.023229,42.0240278,42.028468,42.027656,42.027506,42.024483,42.023982,42.023916,42.026133,42.025996,42.026307,42.024993,42.0246851,42.02472119999999,42.023548,42.022754,42.034183,42.033714,42.0318449,42.03254,42.030625,42.030623,42.030678,42.03252,42.031622,42.031834,42.031623,42.031545,42.0314669,42.0314606,42.0312822,42.03057,42.030531,42.0310946,42.0309301,42.026626,42.026482,42.026414,42.026372,42.025707,42.025293,42.025577,42.024628,42.024596,42.024613,42.024765,42.024327,42.024399,42.02322,42.029772,42.02792,42.0255464,42.021171,42.021176,42.021207,42.020817,42.019784,42.018996,42.0197228,42.018597,42.017569,42.016903,42.018667,42.016862,42.016399,42.01571,42.016551,42.01613,42.015713,42.021384,42.020087,42.019184,42.019697,42.019548,42.019691,42.020666,42.020991,42.021788,42.020843,42.017109,42.018722,42.018847,42.01877,42.018703,42.016973,42.016978,42.0181479,42.018875,42.018866,42.018006,42.017856,42.017856,42.016819,42.016797,42.016274,42.016271,42.016468,42.013957,42.014054,42.013512,42.013392,42.020123,42.021246,42.021219,42.018988,42.0186549,42.0196717,42.018928,42.017026,42.01896,42.022478,42.019423,42.020321,42.0217997,42.021496,42.019004,42.01822,42.019381,42.018592,42.022092,42.022088,42.022091,42.020123,42.019713,42.017564,42.016921,42.016182,42.019094,42.018931,42.017889,42.017835,42.0189256,42.016943,42.016835,42.016957,42.016067,42.0191252,42.019181,42.018707,42.017583,42.017574,42.016684,42.0158534,42.016152,42.014688,42.013169,42.014943,42.013753,42.0125,42.009489,42.009489,42.009489,42.009489,42.010636,42.01141,42.009134,42.022721,42.021425,42.021596,42.020184,42.020002,42.008683,42.0183,41.996888,41.996672,41.997119,41.99539,41.997067,41.997069,41.996919,41.995905,41.996066,42.001246,42.00138,41.995027,41.995058,41.993759,41.994727,41.994695,41.997684,41.998952,41.9981656,41.996327,41.996093,41.995359,41.992964,41.9935397,41.992193,41.993128,41.992122,41.991786,41.992919,41.992185,41.9928219,41.99217,41.992539,41.991786,41.991365,41.987535,41.988776,41.988222,41.987659,41.986502,41.986509,41.991493,41.99095,42.0525525,42.051446,42.049587,42.05996,42.061231,42.060771,42.060822,42.060369,42.060639,42.0623,42.062976,42.061374,42.060865,42.06076,42.059408,42.059364,42.059411,42.058445,42.05701,42.059518,42.059636,42.061394,42.06056,42.060241,42.059792,42.058896,42.0586718,42.056015,42.054514,42.054562,42.054439,42.055358,42.054503,42.054455,42.054211,42.055432,42.052544,42.050326,42.051473,42.050679,42.049449,42.04986,42.049609,42.051074,42.056306,42.05539,42.055304,42.054328,42.0564348,42.054693,42.052794,42.052755,42.051479,42.052332,42.051678,42.051687,42.051659,42.050001,42.05068,42.050257,42.04845,42.062046,42.0632692,42.063129,42.062925,42.062259,42.0624143,42.062432,42.061133,42.06027,42.060427,42.060766,42.060425,42.0599011,42.058484,42.058683,42.058573,42.058649,42.057028,42.058694,42.057164,42.0579589,42.0582,42.056766,42.0570189,42.057446,42.057179,42.0571198,42.057468,42.057094,42.056616,42.063336,42.062299,42.062285,42.062004,42.063301,42.061316,42.059635,42.061544,42.0609759,42.06132,42.061312,42.058377,42.057986,42.058595,42.0600066,42.059507,42.05695,42.058581,42.053879,42.053666,42.055748,42.055777,42.0522319,42.052272,42.050383,42.05162,42.049501,42.049696,42.050692,42.049484,42.054236,42.055157,42.055102,42.054415,42.054366,42.054306,42.054319,42.055517,42.054227,42.053183,42.051646,42.05165,42.050307,42.052278,42.052456,42.052275,42.0520487,42.05138,42.052173,42.051224,42.050899,42.037836,42.035271,42.035132,42.037918,42.037806,42.03805,42.036525,42.037258,42.035875,42.035878,42.035878,42.034988,42.036089,42.034535,42.035666,42.036839,42.036501,42.036466,42.036039,42.035179,42.034614,42.03569,42.035703,42.03459,42.0375,42.034532,42.047666,42.048607,42.047467,42.048089,42.048191,42.047806,42.047571,42.047548,42.047102,42.047103,42.046381,42.046377,42.046372,42.046088,42.046085,42.046961,42.046962,42.046647,42.047171,42.047171,42.047171,42.045012,42.044419,42.044642,42.044693,42.036729,42.037578,42.04872599999999,42.047928,42.0492538,42.049141,42.048762,42.046399,42.042943,42.04378,42.0458538,42.0471033,42.046584,42.047541,42.048185,42.04535509999999,42.044262,42.044058,42.044851,42.043142,42.043845,42.044101,42.042398,42.044773,42.044773,42.041229,42.04066,42.039235,42.040112,42.0417754,42.0417764,42.041811,42.038323,42.04102,42.04186989999999,42.041172,42.040296,42.035582,42.035571,42.034563,42.035741,42.0380577,42.036178,42.036102,42.03713,42.034679,42.048952,42.047782,42.0470858,42.0470799,42.046935,42.046572,42.042373,42.0423032,42.044736,42.043988,42.044791,42.044853,42.044304,42.042187,42.0438199,42.042023,42.042304,42.040479,42.040301,42.040391,42.0404436,42.039531,42.038516,42.039445,42.0372039,42.03471,42.037302,42.037459,42.035893,42.035723,42.035938,42.035222,42.040021,42.040142,42.041173,42.038686,42.039987,42.039568,42.03807,42.037125,42.037104,42.035813,42.037162,42.037006,42.037291,42.034811,42.034564,42.034789,42.034686,42.0347301,42.034636,42.034801,42.036973,42.037025,42.035724,42.0342174,42.034128,42.033379,42.032611,42.03305659999999,42.032392,42.032419,42.03136,42.030427,42.0304142,42.030418,42.030482,42.032161,42.032272,42.030406,42.0303597,42.033435,42.033571,42.033249,42.032008,42.032094,42.032381,42.0311,42.0304823,42.029028,42.029058,42.02928,42.029253,42.029281,42.028287,42.02793,42.027073,42.027032,42.029084,42.02813,42.02882839999999,42.0287814,42.028397,42.029209,42.0273991,42.0272517,42.02774,42.026691,42.02726639999999,42.03361,42.0343116,42.03338,42.03047,42.033456,42.033568,42.033699,42.03337,42.033511,42.033321,42.033473,42.033366,42.033493,42.031354,42.031293,42.031493,42.032476,42.031461,42.031496,42.031467,42.031432,42.031049,42.030388,42.031057,42.029399,42.028426,42.028274,42.027065,42.029556,42.02903,42.029038,42.0286305,42.0255246,42.024111,42.024265,42.0249938,42.0242588,42.022574,42.02295,42.026478,42.0262558,42.024885,42.0242331,42.023716,42.034391,42.034409,42.033423,42.031281,42.02393,42.033604,42.033225,42.032365,42.031571,42.031248,42.031123,42.031348,42.03156,42.030552,42.034428,42.034428,42.033454,42.033834,42.034414,42.032495,42.032566,42.032155,42.03392,42.033692,42.032653,42.033559,42.029453,42.024662,42.0243033,42.02521,42.028568,42.028595,42.027332,42.027507,42.027555,42.028381,42.028233,42.028188,42.0280297,42.024025,42.0257161,42.025036,42.0236702,42.0244599,42.022902,42.022769,42.02283,42.031766,42.033912,42.03085,42.031077,42.030621,42.025411,42.025275,42.027368,42.027193,42.029437,42.030276,42.0249247,42.02578099999999,42.021173,42.01915,42.019153,42.018501,42.018786,42.017423,42.017517,42.017914,42.017766,42.016276,42.015974,42.01626,42.015949,42.022136,42.021234,42.018839,42.019769,42.020694,42.02107,42.021275,42.020796,42.020991,42.018735,42.018412,42.018354,42.018709,42.017598,42.018006,42.01727,42.017484,42.018186,42.016975,42.016207,42.016341,42.014435,42.016709,42.017866,42.016277,42.016254,42.017854,42.016262,42.014549,42.013952,42.013912,42.013641,42.013622,42.013603,42.015975,42.015368,42.0138316,42.015927,42.015927,42.016121,42.016122,42.01974999999999,42.019937,42.020905,42.021244,42.019997,42.019996,42.02099,42.019861,42.018991,42.018852,42.018912,42.01883,42.01738599999999,42.016804,42.018898,42.020483,42.0202945,42.020394,42.019544,42.020112,42.021582,42.01797,42.017962,42.0198028,42.019959,42.018541,42.017874,42.02224,42.022238,42.022089,42.0214496,42.019986,42.018715,42.017413,42.017083,42.019272,42.019092,42.019005,42.01791,42.017834,42.0176853,42.01789,42.016887,42.0168678,42.016072,42.019833,42.019617,42.018443,42.018888,42.018766,42.017528,42.016769,42.01601,42.015667,42.014567,42.012638,42.012488,42.009489,42.009489,42.022607,42.021806,42.022797,42.022516,42.0213831,42.021276,42.0214037,42.020379,42.021389,42.018806,42.021162,42.021506,42.021594,42.022458,42.022626,42.009326,42.008803,42.008806,42.0091899,41.99507,41.997124,41.997062,41.997069,41.996919,41.996903,41.995991,41.998543,42.00514,41.998999,41.995141,41.994841,41.99373,41.995142,41.993751,41.995175,41.999812,41.999575,41.9976055,41.998554,41.995512,41.99639,41.99423,41.99312,41.993611,41.99338,41.993335,41.990054,41.993059,41.993185,41.99261,41.991372,41.990899,41.991883,41.991779,41.993249,41.992133,41.991933,41.988675,41.987109,41.988934,41.98811,41.988076,41.987289,41.98658,41.986498,41.991885,42.022745,42.051297,42.051357,42.049671,42.050443,42.06127499999999,42.060466,42.061664,42.061326,42.061242,42.063067,42.059592,42.058953,42.059562,42.059517,42.059242,42.059293,42.057006,42.05817,42.057533,42.057022,42.059375,42.059239,42.059309,42.059304,42.060431,42.060144,42.061202,42.060096,42.059743,42.058893,42.058998,42.058996,42.054368,42.053916,42.055545,42.0545,42.054395,42.054941,42.055217,42.054427,42.05622,42.056211,42.052353,42.053271,42.05026,42.049832,42.050075,42.050372,42.04995599999999,42.051039,42.048765,42.055392,42.055264,42.054603,42.053407,42.054612,42.053547,42.05346,42.053503,42.0562952,42.05639499999999,42.05467,42.054789,42.055319,42.05266900000001,42.052698,42.052696,42.052294,42.051833,42.050397,42.05057,42.049996,42.0488205,42.062574,42.063296,42.063298,42.0632689,42.0632722,42.063141,42.0619544,42.061976,42.0619709,42.061284,42.062259,42.0628153,42.0624082,42.062434,42.061504,42.061131,42.06027,42.060269,42.061133,42.060576,42.060427,42.059561,42.06113,42.05976099999999,42.059677,42.058634,42.058619,42.059211,42.059202,42.059197,42.059194,42.059175,42.059175,42.059176,42.059176,42.05826,42.058276,42.058284,42.056827,42.05712399999999,42.05737999999999,42.0568529,42.057209,42.063388,42.063381,42.062483,42.062279,42.0626729,42.06208,42.062067,42.062533,42.062887,42.062931,42.062974,42.063019,42.063106,42.063342,42.062584,42.062581,42.062946,42.063301,42.061713,42.061429,42.061483,42.061541,42.061531,42.061318,42.061392,42.060021,42.058858,42.059181,42.057981,42.058401,42.059526,42.057836,42.058897,42.0533209,42.053364,42.055751,42.055627,42.05359,42.053436,42.053467,42.05198,42.052361,42.052247,42.051813,42.05048,42.049518,42.049509,42.052416,42.054408,42.055757,42.054318,42.054177,42.054417,42.054354,42.054309,42.054514,42.054154,42.054153,42.054153,42.054157,42.053185,42.053186,42.053227,42.053108,42.053032,42.053067,42.051802,42.051795,42.051655,42.051657,42.050206,42.050301,42.051681,42.052269,42.052119,42.051377,42.052267,42.052146,42.051054,42.037586,42.03763600000001,42.037643,42.03722,42.036028,42.035259,42.036038,42.035071,42.036464,42.034569,42.035678,42.0362103,42.034678,42.035946,42.035871,42.035857,42.034532,42.034724,42.035675,42.034652,42.034819,42.037652,42.037502,42.03466,42.034644,42.034673,42.047384,42.045776,42.047618,42.0474249,42.047406,42.047711,42.047705,42.047699,42.047687,42.047667,42.047104,42.047104,42.046144,42.046145,42.046145,42.046405,42.047171,42.046318,42.046137,42.046135,42.046411,42.044383,42.044762,42.044989,42.042994,42.043048,42.042961,42.037103,42.0349702,42.048488,42.046549,42.047758,42.048067,42.04849,42.047615,42.043363,42.043005,42.045483,42.04259,42.04389,42.0421698,42.047811,42.048292,42.046504,42.044758,42.043984,42.043929,42.043121,42.042124,42.04227,42.044036,42.042577,42.042485,42.042479,42.044921,42.043609,42.043627,42.039141,42.040143,42.0417369,42.039069,42.0413937,42.0417604,42.04074,42.041245,42.041332,42.039212,42.0392,42.036575,42.035684,42.036023,42.036223,42.034615,42.0362,42.03465,42.034906,42.034755,42.049216,42.048934,42.046714,42.045809,42.04593250000001,42.046882,42.044053,42.043143,42.044735,42.043932,42.043946,42.04230949999999,42.044218,42.045669,42.043952,42.043233,42.0426969,42.042728,42.04302,42.0418697,42.041243,42.040773,42.040451,42.03981,42.039369,42.039482,42.03846,42.03837,42.0384781,42.038545,42.034789,42.0347206,42.034812,42.035932,42.036077,42.036027,42.03725,42.0380982,42.0357,42.035779,42.035854,42.034878,42.039338,42.039983,42.040146,42.038888,42.035733,42.035921,42.037329,42.037203,42.035954,42.035876,42.037106,42.035862,42.036002,42.037084,42.034685,42.034614,42.034564,42.036949,42.035423,42.0346,42.035034,42.034403,42.032254,42.03147329999999,42.03131,42.031369,42.0311108,42.033405,42.032343,42.033385,42.033489,42.033313,42.032276,42.032357,42.030579,42.0318018,42.030311,42.03013,42.0299846,42.029218,42.029198,42.029223,42.0280396,42.028076,42.028847,42.02819,42.027154,42.0282,42.0277574,42.026089,42.0268089,42.0272605,42.0272703,42.033311,42.034339,42.031846,42.030823,42.033482,42.033612,42.033336,42.032399,42.031433,42.031438,42.03142,42.032505,42.032386,42.031321,42.031483,42.03148729999999,42.0304477,42.029451,42.026912,42.026998,42.027088,42.0299664,42.02911,42.02752,42.0275775,42.025229,42.02543,42.021576,42.0249982,42.0242321,42.0227,42.0238856,42.022963,42.0238391,42.023518,42.026723,42.025893,42.0261614,42.022951,42.033524,42.033285,42.028048,42.028004,42.033272,42.032426,42.034453,42.034429,42.033193,42.032385,42.032098,42.031533,42.031203,42.029562,42.031215,42.029679,42.031139,42.029786,42.030591,42.034422,42.032528,42.034421,42.033557,42.032638,42.033726,42.034018,42.032699,42.02868,42.02384,42.024087,42.026185,42.023229,42.023021,42.028424,42.028236,42.024949,42.025818,42.025001,42.0205201,42.020186,42.0246868,42.023874,42.023945,42.023451,42.022824,42.034453,42.033662,42.033576,42.031546,42.032467,42.031566,42.030801,42.02667,42.026642,42.025666,42.025707,42.02559400000001,42.025264,42.025281,42.025558,42.024598,42.024594,42.024633,42.024614,42.025726,42.025156,42.024635,42.024485,42.023226,42.029694,42.029803,42.027956,42.027969,42.024734,42.024305,42.022228,42.021324,42.02121,42.020966,42.020912,42.019032,42.019083,42.019028,42.018378,42.017907,42.017826,42.016896,42.016829,42.016297,42.016214,42.021992,42.021386,42.0218643,42.022239,42.01972,42.021163,42.021056,42.021025,42.020861,42.021737,42.018518,42.018424,42.018849,42.016973,42.016961,42.016212,42.0172,42.018564,42.018864,42.0185298,42.018017,42.017856,42.016277,42.017753,42.015974,42.014038,42.014039,42.014823,42.013576,42.013544,42.01343,42.013467,42.01469,42.0210331,42.020703,42.021221,42.021195,42.02103,42.020893,42.019926,42.01969589999999,42.019902,42.01917,42.018283,42.018873,42.018922,42.0164532,42.016642,42.018973,42.018965,42.020062,42.020783,42.020504,42.022655,42.021436,42.017962,42.01965,42.018943,42.01873,42.019937,42.01902,42.018761,42.01804,42.018039,42.017888,42.017889,42.01892100000001,42.017103,42.017106,42.0162813,42.016216,42.016217,42.016217,42.016853,42.016067,42.01995369999999,42.019952,42.01879599999999,42.017572,42.020037,42.018504,42.018941,42.0168895,42.017573,42.016961,42.01629,42.013003,42.0145503,42.01265,42.012488,42.011506,42.009489,42.011836,42.009321,42.011343,42.010333,42.010076,42.022836,42.022474,42.022658,42.022658,42.022566,42.022685,42.0228,42.021547,42.021278,42.021497,42.020161,42.020077,42.0214218,42.021555,42.009401,42.009401,41.994357,41.994207,41.994206,41.995218,41.995973,41.995931,41.995512,41.994188,42.005692,41.99854,42.0004101,41.994241,41.994137,41.9941,41.994824,41.995088,41.998023,41.997992,41.997364,41.997126,41.996089,41.995508,41.997299,41.995341,41.993986,41.997352,41.993288,41.992975,41.993261,41.990054,41.992318,41.993356,41.993107,41.9927922,41.991211,41.992917,41.99284,41.992145,41.992516,41.992479,41.991709,41.991641,41.991868,41.991612,41.991098,41.989581,41.990264,41.989488,41.991144,41.991371,41.989905,41.987606,41.989768,41.98822,41.9881729,41.98651,41.986588,42.053406,42.053327,42.0531078,42.052519,42.050507,42.059909,42.061407,42.060425,42.06048,42.060736,42.061879,42.062828,42.06281,42.062802,42.062798,42.062891,42.062956,42.060625,42.0591,42.058421,42.058201,42.059599,42.059332,42.058136,42.060459,42.060527,42.060516,42.061478,42.061008,42.060111,42.059957,42.061096,42.0593867,42.056702,42.05856300000001,42.059947,42.058816,42.055607,42.054367,42.0543376,42.054956,42.0562619,42.05327,42.0524029,42.053418,42.050514,42.050176,42.049771,42.051077,42.052362,42.052522,42.052359,42.04977,42.049608,42.055131,42.053459,42.05446,42.054385,42.055319,42.052655,42.0524739,42.0527019,42.051798,42.051839,42.051691,42.051672,42.051673,42.051664,42.050595,42.048947,42.048946,42.050705,42.050441,42.06147199999999,42.06330699999999,42.061901,42.061284,42.0628074,42.062109,42.061134,42.060342,42.060272,42.060271,42.060425,42.060835,42.060577,42.060866,42.060424,42.060357,42.060377,42.061544,42.061377,42.059636,42.059825,42.058544,42.05869,42.059216,42.059017,42.059184,42.057164,42.058231,42.058238,42.057164,42.057164,42.05758,42.063304,42.062442,42.062361,42.061915,42.061928,42.063031,42.062576,42.061984,42.061987,42.062574,42.063298,42.063329,42.063185,42.062374,42.0623,42.063301,42.063301,42.063301,42.062831,42.062873,42.060239,42.060293,42.059803,42.061727,42.060836,42.060074,42.05865,42.05853,42.057905,42.057838,42.05696,42.058461,42.05443,42.05384,42.053801,42.053314,42.055601,42.053154,42.053466,42.05178,42.049423,42.051073,42.049687,42.049528,42.055056,42.054318,42.054176,42.054175,42.053183,42.053033,42.0518,42.052265,42.052272,42.052346,42.051355,42.051426,42.051382,42.051245,42.051225,42.051224,42.037843,42.036149,42.037915,42.037911,42.037827,42.036605,42.03718,42.036849,42.034595,42.03674,42.035555,42.035629,42.03748,42.035795,42.035235,42.036051,42.036087,42.036206,42.035302,42.035094,42.034692,42.035696,42.035841,42.035325,42.035536,42.03457,42.03458,42.034663,42.037065,42.045763,42.048608,42.048371,42.047997,42.048058,42.047592,42.04768,42.047665,42.047642,42.047618,42.047595,42.046145,42.046431,42.046127,42.046107,42.046429,42.046416,42.044609,42.043779,42.037409,42.045632,42.048199,42.048997,42.048602,42.046472,42.04806,42.043548,42.044669,42.046496,42.045706,42.045728,42.045728,42.045885,42.04587799999999,42.04795,42.04657,42.045685,42.044308,42.043214,42.044978,42.042114,42.040072,42.038434,42.04102,42.0414084,42.041314,42.040273,42.040143,42.039195,42.038351,42.039202,42.035836,42.03537,42.035379,42.035666,42.0380632,42.034612,42.036024,42.037463,42.034831,42.035987,42.037247,42.0493,42.049031,42.04665,42.048528,42.0422973,42.042156,42.043129,42.044735,42.043897,42.04498,42.044975,42.044254,42.042105,42.043626,42.0427,42.042905,42.041821,42.040745,42.040252,42.039447,42.038357,42.039387,42.038602,42.0383475,42.038836,42.040276,42.039351,42.034934,42.034617,42.0347269,42.0378602,42.03717959999999,42.037001,42.037097,42.037041,42.0358,42.0367836,42.035857,42.034574,42.035277,42.034761,42.040944,42.040811,42.040105,42.040081,42.040297,42.040148,42.0385,42.038497,42.038297,42.041874,42.0416529,42.040768,42.03902,42.036098,42.035782,42.035996,42.036099,42.036053,42.034805,42.03471,42.034662,42.034829,42.034757,42.034708,42.036792,42.03548,42.034618,42.03475299999999,42.0345614,42.033267,42.0334994,42.033385,42.032144,42.032346,42.03139,42.031506,42.031347,42.03121,42.030396,42.030463,42.031196,42.033623,42.0330459,42.032305,42.030267,42.032064,42.032267,42.031964,42.030368,42.03146,42.031138,42.029142,42.027993,42.029136,42.028951,42.029281,42.027953,42.0281148,42.027953,42.028963,42.029025,42.027992,42.0272703,42.02297,42.03405,42.03179,42.030786,42.033306,42.033369,42.032299,42.031499,42.032396,42.032478,42.031415,42.031436,42.032418,42.032417,42.0310818,42.030389,42.029526,42.02686,42.028544,42.026886,42.02689,42.025258,42.025459,42.0249549,42.0242386,42.0242578,42.023117,42.022825,42.0238089,42.034411,42.034576,42.033296,42.0331453,42.0277962,42.028002,42.027997,42.033521,42.033388,42.0333,42.03224,42.032304,42.032383,42.031356,42.030611,42.032216,42.030702,42.031362,42.031347,42.030231,42.030232,42.033285,42.03247,42.032291,42.032307,42.033686,42.032478,42.031635,42.032157,42.032575,42.029453,42.028836,42.023651,42.024705,42.02653,42.027664,42.026667,42.026614,42.02821,42.028191,42.026458,42.025846,42.025524,42.025102,42.025197,42.026308,42.0247523,42.022824,42.032082,42.032493,42.030933,42.0315156,42.024601,42.024635,42.024555,42.024615,42.025687,42.025218,42.025194,42.02437,42.024481,42.023306,42.024376,42.023308,42.023138,42.029884,42.028168,42.025573,42.02438,42.024592,42.022227,42.022077,42.021209,42.0209,42.019047,42.018043,42.018123,42.017631,42.0168729,42.017929,42.01686,42.016217,42.016913,42.021388,42.020145,42.02007,42.019697,42.021811,42.021889,42.020998,42.01834,42.018664,42.017854,42.017055,42.018188,42.014494,42.016979,42.016904,42.01869,42.017856,42.017867,42.017858,42.017186,42.016275,42.016252,42.016618,42.015914,42.016092,42.014636,42.014109,42.015793,42.016004,42.01397,42.013991,42.01576499999999,42.013519,42.014593,42.0195855,42.020995,42.020342,42.020592,42.020547,42.021113,42.01899,42.021168,42.018648,42.018961,42.01896,42.018959,42.018813,42.018812,42.01881,42.01881,42.018809,42.02011,42.022676,42.022659,42.022641,42.021432,42.018848,42.017727,42.022089,42.022088,42.02197,42.021559,42.020871,42.020286,42.018837,42.018282,42.01924,42.019093,42.01895,42.018808,42.018857,42.016941,42.0165794,42.016086,42.016072,42.017609,42.01874,42.019059,42.019759,42.017131,42.016827,42.016385,42.017589,42.0157866,42.016124,42.013296,42.013495,42.014936,42.01359,42.011793,42.022786,42.022709,42.020441,42.020855,42.020207,42.02153,42.019025,42.019349,42.019099,42.019099,42.022603,42.009398,42.008375,41.996849,41.994357,41.995987,41.996907,41.995702,41.996063,42.004509,41.99984,41.997906,41.999553,41.998023,41.999342,41.996331,41.994347,41.994229,41.993258,41.993047,41.99222,41.99329,41.993173,41.99315,41.990054,41.990993,41.990523,41.991731,41.993062,41.991574,41.992303,41.992159,41.991708,41.992533,41.992522,41.992497,41.99171,41.991532,41.9900428,41.990134,41.9887374,41.987686,41.988964,41.988314,41.98651,41.990921,41.989265],[-93.619754,-93.619756,-93.6193873,-93.61732000000001,-93.63893299999999,-93.638925,-93.633792,-93.633826,-93.632852,-93.63906799999999,-93.63694700000001,-93.63595100000001,-93.63864700000001,-93.634626,-93.632913,-93.62654999999999,-93.6235954,-93.628804,-93.62297100000001,-93.636655,-93.633962,-93.63637199999999,-93.639366,-93.626231,-93.626537,-93.628806,-93.627112,-93.62276900000001,-93.62437300000001,-93.62727099999999,-93.62753600000001,-93.62728,-93.62723200000001,-93.625924,-93.62596600000001,-93.625848,-93.653201,-93.654645,-93.65373289999999,-93.6528306,-93.65271300000001,-93.65331999999999,-93.65233600000001,-93.65230699999999,-93.655051,-93.650436,-93.65606699999999,-93.654853,-93.654121,-93.649447,-93.65101300000001,-93.644356,-93.64163499999999,-93.642252,-93.644076,-93.641487,-93.642447,-93.64132499999999,-93.64133699999999,-93.651381,-93.652334,-93.654206,-93.653616,-93.650356,-93.650578,-93.65297200000001,-93.642368,-93.649976,-93.650318,-93.647105,-93.64277300000001,-93.642224,-93.69230899999999,-93.692412,-93.68980999999999,-93.68641700000001,-93.68769399999999,-93.686263,-93.68626399999999,-93.68628,-93.685028,-93.685029,-93.67906939999999,-93.677205,-93.673669,-93.670956,-93.672096,-93.672067,-93.66968799999999,-93.66067200000001,-93.646749,-93.65177300000001,-93.652773,-93.647988,-93.64614,-93.645599,-93.64488900000001,-93.64488900000001,-93.64489,-93.6468879,-93.645606,-93.645585,-93.645482,-93.645478,-93.64125299999999,-93.643556,-93.65921,-93.639516,-93.633351,-93.63687899999999,-93.637463,-93.637531,-93.637788,-93.631435,-93.632554,-93.626137,-93.62592600000001,-93.623383,-93.62693899999999,-93.62753600000001,-93.626785,-93.627809,-93.625728,-93.62290399999999,-93.62286400000001,-93.620651,-93.620504,-93.62928700000001,-93.629313,-93.62475000000001,-93.624601,-93.621431,-93.61822100000001,-93.6184183,-93.614307,-93.614244,-93.617693,-93.617553,-93.61711699999999,-93.6177041,-93.61862600000001,-93.61658199999999,-93.61488199999999,-93.61330100000001,-93.61512999999999,-93.61424099999999,-93.613624,-93.615622,-93.617092,-93.61722,-93.61842799999999,-93.615475,-93.61164199999999,-93.618788,-93.61920139999999,-93.620341,-93.61167620000001,-93.607275,-93.606905,-93.61060999999999,-93.605969,-93.609218,-93.60780099999999,-93.605943,-93.606742,-93.60483499999999,-93.60370500000001,-93.617014,-93.617135,-93.62024,-93.62024099999999,-93.61791119999999,-93.614002,-93.615431,-93.61544600000001,-93.608954,-93.6090761,-93.610466,-93.610468,-93.60678900000001,-93.60678900000001,-93.61688700000001,-93.61058800000001,-93.61221,-93.608878,-93.60824599999999,-93.607238,-93.608752,-93.607088,-93.625737,-93.62582500000001,-93.625586,-93.624566,-93.622597,-93.62352300000001,-93.62244800000001,-93.62348299999999,-93.621482,-93.620409,-93.62338,-93.621674,-93.621835,-93.62163200000001,-93.62949500000001,-93.626876,-93.620386,-93.65623100000001,-93.65553800000001,-93.658265,-93.656924,-93.658081,-93.657573,-93.655849,-93.67203600000001,-93.671021,-93.671076,-93.673401,-93.672381,-93.678141,-93.676659,-93.677992,-93.676654,-93.67667400000001,-93.671511,-93.673565,-93.672523,-93.670503,-93.670114,-93.663619,-93.668378,-93.66790399999999,-93.675038,-93.671178,-93.662762,-93.66357000000001,-93.667708,-93.66525999999999,-93.66609699999999,-93.66588759999999,-93.66579900000001,-93.66336699999999,-93.659808,-93.685001,-93.68598799999999,-93.683167,-93.68202599999999,-93.68015679999999,-93.69019,-93.69126900000001,-93.69244999999999,-93.692521,-93.68329,-93.68018049999999,-93.68213799999999,-93.69188,-93.69054199999999,-93.688258,-93.68773,-93.68780700000001,-93.68774500000001,-93.683727,-93.687859,-93.68552299999999,-93.687764,-93.680864,-93.682686,-93.68122700000001,-93.6837,-93.687922,-93.684015,-93.68596100000001,-93.6783313,-93.677482,-93.677553,-93.67754499999999,-93.676372,-93.675268,-93.675175,-93.6663375,-93.66750999999999,-93.664841,-93.660168,-93.66194400000001,-93.655762,-93.655653,-93.656718,-93.65032100000001,-93.6514413,-93.6507568,-93.64678000000001,-93.646683,-93.640041,-93.639781,-93.642793,-93.639678,-93.645014,-93.645701,-93.645729,-93.641677,-93.625291,-93.629496,-93.626689,-93.615272,-93.606593,-93.608271,-93.60775,-93.61599299999999,-93.604947,-93.602823,-93.603972,-93.600797,-93.602002,-93.65789400000001,-93.649203,-93.64920100000001,-93.649096,-93.64947099999999,-93.647645,-93.64681400000001,-93.652495,-93.60742999999999,-93.608372,-93.60762200000001,-93.608223,-93.60359,-93.60431800000001,-93.60442500000001,-93.603435,-93.60359,-93.601806,-93.603398,-93.601449,-93.60685599999999,-93.6035354,-93.6030339,-93.600582,-93.599791,-93.60000599999999,-93.60014700000001,-93.61889549999999,-93.619562,-93.61552399999999,-93.639388,-93.639915,-93.63639999999999,-93.63583800000001,-93.63744199999999,-93.634265,-93.63373900000001,-93.63354699999999,-93.63302400000001,-93.63921000000001,-93.63906900000001,-93.63906900000001,-93.636955,-93.637046,-93.637671,-93.635824,-93.63575299999999,-93.636822,-93.636047,-93.639325,-93.63464399999999,-93.633245,-93.627751,-93.62863900000001,-93.629357,-93.623975,-93.637343,-93.636512,-93.63505600000001,-93.63241699999999,-93.63777399999999,-93.63745,-93.638366,-93.63421199999999,-93.63704299999999,-93.633796,-93.637289,-93.634694,-93.635919,-93.639516,-93.639511,-93.63923079999999,-93.639366,-93.632915,-93.632178,-93.630253,-93.63143199999999,-93.63179100000001,-93.625856,-93.626554,-93.62870700000001,-93.626603,-93.62654999999999,-93.627037,-93.625657,-93.625542,-93.62349859999999,-93.6223289,-93.62739599999999,-93.628119,-93.627565,-93.629456,-93.629851,-93.629881,-93.627244,-93.624729,-93.624748,-93.625945,-93.625918,-93.625846,-93.625688,-93.62569000000001,-93.625696,-93.62570700000001,-93.62560359999999,-93.626521,-93.65887699999999,-93.65627490000001,-93.652716,-93.65782799999999,-93.657483,-93.654241,-93.654144,-93.653955,-93.652863,-93.654439,-93.654567,-93.65822399999999,-93.65783499999999,-93.657049,-93.654437,-93.65885400000001,-93.65673700000001,-93.65315200000001,-93.652884,-93.654724,-93.65476,-93.652084,-93.654652,-93.65272399999999,-93.65412669999999,-93.65477539999999,-93.643771,-93.65603999999999,-93.65486,-93.6545189,-93.65048,-93.649779,-93.649743,-93.64972299999999,-93.65025799999999,-93.650244,-93.655483,-93.6552449,-93.65420399999999,-93.65421499999999,-93.654405,-93.6514849,-93.651014,-93.650238,-93.645752,-93.64466,-93.644457,-93.645627,-93.641561,-93.641515,-93.642094,-93.64259699999999,-93.642911,-93.642914,-93.64274899999999,-93.64331199999999,-93.64410599999999,-93.64177199999999,-93.641488,-93.64262600000001,-93.640051,-93.64006500000001,-93.649649,-93.642813,-93.642932,-93.641919,-93.64425100000001,-93.64137700000001,-93.641313,-93.64130299999999,-93.641367,-93.641397,-93.641312,-93.640879,-93.65429,-93.657163,-93.65095700000001,-93.652608,-93.65237999999999,-93.65555500000001,-93.65598799999999,-93.654048,-93.65150300000001,-93.65211699999999,-93.65052900000001,-93.65082200000001,-93.644121,-93.64247,-93.64255300000001,-93.643944,-93.642579,-93.64236,-93.641251,-93.641296,-93.65032100000001,-93.6504,-93.65044,-93.650417,-93.65037700000001,-93.650338,-93.648275,-93.648809,-93.64877799999999,-93.643001,-93.64287,-93.64384200000001,-93.64247400000001,-93.64375200000001,-93.64084699999999,-93.641746,-93.643131,-93.64150100000001,-93.639832,-93.69153,-93.69201699999999,-93.69247200000001,-93.69143,-93.691506,-93.69170699999999,-93.691868,-93.691068,-93.69015899999999,-93.69040200000001,-93.689136,-93.69139,-93.6888,-93.687843,-93.686263,-93.686267,-93.687804,-93.68503699999999,-93.68395700000001,-93.6790752,-93.680198,-93.67636399999999,-93.677008,-93.67405100000001,-93.67571599999999,-93.66955400000001,-93.669538,-93.673518,-93.669707,-93.669845,-93.660327,-93.650333,-93.64776500000001,-93.646316,-93.646338,-93.64703899999999,-93.64691599999999,-93.644205,-93.64566000000001,-93.645482,-93.64179900000001,-93.645923,-93.64805699999999,-93.645546,-93.65921,-93.63534199999999,-93.63463400000001,-93.63480300000001,-93.63197099999999,-93.63188100000001,-93.631519,-93.633163,-93.631851,-93.632537,-93.632823,-93.637964,-93.637905,-93.63264700000001,-93.631899,-93.62890299999999,-93.62704699999999,-93.624612,-93.630714,-93.627416,-93.625775,-93.622991,-93.62246500000001,-93.623485,-93.622635,-93.62247600000001,-93.623411,-93.62941499999999,-93.628658,-93.62731269999999,-93.627836,-93.627985,-93.62697900000001,-93.62458100000001,-93.622685,-93.62169899999999,-93.629499,-93.62918999999999,-93.62563299999999,-93.62568,-93.623705,-93.624612,-93.62264399999999,-93.622494,-93.61967300000001,-93.617963,-93.615686,-93.615602,-93.61555799999999,-93.614619,-93.614216,-93.619029,-93.61878400000001,-93.61864,-93.61488,-93.616434,-93.614751,-93.614059,-93.614768,-93.61588,-93.614543,-93.62001600000001,-93.618478,-93.61654729999999,-93.618549,-93.6190376,-93.61721900000001,-93.61239999999999,-93.61720699999999,-93.61679119999999,-93.618605,-93.618607,-93.62008,-93.62033599999999,-93.61718999999999,-93.617182,-93.61462060000001,-93.61442090000001,-93.61239,-93.612241,-93.61404899999999,-93.615465,-93.61404899999999,-93.613899,-93.6117397,-93.61202999999999,-93.612253,-93.61225399999999,-93.610355,-93.60765499999999,-93.61058800000001,-93.610649,-93.60819599999999,-93.60892699999999,-93.60922600000001,-93.606573,-93.60552,-93.609059,-93.608977,-93.61060999999999,-93.608903,-93.60890999999999,-93.606891,-93.606889,-93.607814,-93.605968,-93.606786,-93.609945,-93.61060500000001,-93.60498699999999,-93.60404699999999,-93.60387900000001,-93.603701,-93.6179738,-93.615583,-93.619013,-93.6195025,-93.62026,-93.61559800000001,-93.615599,-93.61953609999999,-93.618486,-93.6200683,-93.620216,-93.61763669999999,-93.615381,-93.6129105,-93.60886600000001,-93.607776,-93.608829,-93.606842,-93.606841,-93.61046899999999,-93.60678799999999,-93.61864800000001,-93.617054,-93.61786789999999,-93.61388700000001,-93.612245,-93.61058800000001,-93.6130761,-93.61058199999999,-93.61486119999999,-93.610044,-93.6076506,-93.608704,-93.608707,-93.60705799999999,-93.60520699999999,-93.60542100000001,-93.6289369,-93.628848,-93.62798149999999,-93.625968,-93.626116,-93.62456400000001,-93.624567,-93.622513,-93.623515,-93.62167699999999,-93.622443,-93.624689,-93.623645,-93.623644,-93.623643,-93.62362899999999,-93.62256499999999,-93.621351,-93.6213592,-93.62725999999999,-93.625362,-93.62152500000001,-93.621504,-93.621506,-93.621635,-93.621633,-93.628299,-93.626716,-93.62186199999999,-93.6217097,-93.6250433,-93.62515399999999,-93.599575,-93.659205,-93.658207,-93.65817300000001,-93.657107,-93.6601886,-93.657241,-93.65838100000001,-93.669839,-93.67540099999999,-93.67228799999999,-93.673382,-93.67799100000001,-93.672651,-93.67184899999999,-93.671369,-93.666432,-93.66636,-93.666813,-93.662126,-93.659279,-93.660877,-93.66798,-93.671177,-93.67215729999999,-93.666242,-93.66599600000001,-93.665994,-93.667776,-93.668212,-93.667209,-93.66527099999999,-93.66526999999999,-93.66014699999999,-93.663366,-93.6641939,-93.66104,-93.661005,-93.65985499999999,-93.684708,-93.685733,-93.685986,-93.68340999999999,-93.68106299999999,-93.682025,-93.682176,-93.679789,-93.679784,-93.6791403,-93.679109,-93.679784,-93.67952729999999,-93.67909689999999,-93.6791456,-93.67978100000001,-93.679856,-93.67912509999999,-93.679181,-93.687067,-93.68885899999999,-93.688993,-93.689091,-93.69026599999999,-93.690971,-93.690364,-93.689072,-93.689071,-93.691433,-93.69157,-93.690111,-93.689066,-93.689081,-93.685125,-93.67929700000001,-93.68256459999999,-93.691883,-93.69213499999999,-93.688372,-93.69040099999999,-93.691355,-93.691585,-93.69286679999999,-93.69054800000001,-93.690324,-93.691667,-93.688383,-93.689431,-93.691175,-93.689438,-93.689942,-93.691672,-93.68862799999999,-93.686916,-93.687212,-93.687219,-93.685068,-93.684887,-93.684954,-93.68315699999999,-93.683159,-93.68317,-93.68351,-93.687819,-93.686218,-93.68268999999999,-93.68262300000001,-93.68298299999999,-93.686744,-93.68567299999999,-93.67914999999999,-93.680706,-93.68110900000001,-93.681343,-93.681411,-93.68078199999999,-93.68218,-93.68167,-93.681297,-93.681439,-93.68411500000001,-93.685333,-93.68605100000001,-93.68453599999999,-93.68429999999999,-93.67830500000001,-93.677407,-93.676372,-93.672234,-93.6755239,-93.6739723,-93.675248,-93.676608,-93.66995199999999,-93.664779,-93.666315,-93.66468500000001,-93.6644556,-93.65919100000001,-93.664812,-93.6644024,-93.66433599999999,-93.663302,-93.657963,-93.65666400000001,-93.655674,-93.651825,-93.652027,-93.651786,-93.64857000000001,-93.651669,-93.64655500000001,-93.648414,-93.646655,-93.64842899999999,-93.6459433,-93.64688099999999,-93.648419,-93.644704,-93.64463600000001,-93.64280549999999,-93.6397,-93.642371,-93.64012,-93.640592,-93.644295,-93.6439975,-93.639431,-93.642945,-93.64288500000001,-93.640762,-93.641356,-93.63940700000001,-93.645743,-93.645743,-93.645743,-93.645743,-93.644864,-93.643539,-93.643947,-93.62529000000001,-93.626839,-93.625299,-93.616895,-93.61539500000001,-93.6165709,-93.588227,-93.606944,-93.607252,-93.607055,-93.603717,-93.604716,-93.604491,-93.604519,-93.602356,-93.602424,-93.64978000000001,-93.652119,-93.65315,-93.651832,-93.651867,-93.65303900000001,-93.650453,-93.646455,-93.644884,-93.6446004,-93.646793,-93.646638,-93.646816,-93.651703,-93.6042638,-93.60835899999999,-93.60819600000001,-93.60821,-93.604252,-93.604218,-93.60347899999999,-93.60063599999999,-93.603111,-93.601975,-93.60176,-93.60112700000001,-93.60668800000001,-93.604928,-93.60386699999999,-93.60277699999999,-93.609453,-93.606925,-93.59986599999999,-93.600191,-93.6188286,-93.61632890000001,-93.617424,-93.639062,-93.63779599999999,-93.638775,-93.638778,-93.63748200000001,-93.636718,-93.63694799999999,-93.633876,-93.633781,-93.633905,-93.633797,-93.639151,-93.63900700000001,-93.63776300000001,-93.636798,-93.63727900000001,-93.632013,-93.631581,-93.62883100000001,-93.628901,-93.627911,-93.622299,-93.629808,-93.62256669999999,-93.63914939999999,-93.639715,-93.63782500000001,-93.63780800000001,-93.636511,-93.636505,-93.63511099999999,-93.63029899999999,-93.633168,-93.635153,-93.63710500000001,-93.63719,-93.633905,-93.635216,-93.63417690000001,-93.631129,-93.63023099999999,-93.626544,-93.627875,-93.62651099999999,-93.630208,-93.62426499999999,-93.62487299999999,-93.62897099999999,-93.627629,-93.62711400000001,-93.62946199999999,-93.62765,-93.627309,-93.629425,-93.627234,-93.625986,-93.625848,-93.62649,-93.658873,-93.6573126,-93.652941,-93.652744,-93.654642,-93.65665509999999,-93.657832,-93.657889,-93.657188,-93.656801,-93.65369,-93.654734,-93.654443,-93.656958,-93.65572,-93.65254899999999,-93.65121600000001,-93.655587,-93.65491400000001,-93.65418699999999,-93.6513669,-93.649744,-93.653868,-93.65276899999999,-93.65184499999999,-93.651163,-93.6514243,-93.649997,-93.65060699999999,-93.65089,-93.640176,-93.640958,-93.6424399,-93.643039,-93.642326,-93.642641,-93.643522,-93.643964,-93.64177099999999,-93.64141600000001,-93.640051,-93.649587,-93.64133099999999,-93.642397,-93.64426979999999,-93.643632,-93.64134199999999,-93.640959,-93.654594,-93.65412000000001,-93.653324,-93.650932,-93.651562,-93.65527899999999,-93.656036,-93.65392900000001,-93.65579,-93.653908,-93.65180099999999,-93.65099600000001,-93.64410700000001,-93.64414499999999,-93.64233400000001,-93.64210799999999,-93.64209200000001,-93.642376,-93.64397599999999,-93.64403,-93.644024,-93.643788,-93.650397,-93.650358,-93.64825500000001,-93.64383599999999,-93.64118999999999,-93.640924,-93.64304229999999,-93.643692,-93.641651,-93.644018,-93.639662,-93.692497,-93.692431,-93.691695,-93.691836,-93.69076099999999,-93.690364,-93.69124600000001,-93.69123999999999,-93.689769,-93.689024,-93.688981,-93.689801,-93.68880299999999,-93.687819,-93.687831,-93.687847,-93.68642800000001,-93.686278,-93.674093,-93.675743,-93.673965,-93.670997,-93.67250300000001,-93.670551,-93.66203,-93.660522,-93.653482,-93.653232,-93.647488,-93.646063,-93.646113,-93.64489,-93.644891,-93.644891,-93.644172,-93.64411800000001,-93.646984,-93.646986,-93.646989,-93.647047,-93.64704399999999,-93.64374599999999,-93.64372899999999,-93.643378,-93.641785,-93.64176999999999,-93.64174199999999,-93.647655,-93.64870500000001,-93.648954,-93.645713,-93.655582,-93.658315,-93.63922509999999,-93.63509500000001,-93.63432720000002,-93.63497599999999,-93.63152599999999,-93.63177899999999,-93.6324549,-93.63124500000001,-93.62643559999999,-93.62614309999999,-93.624762,-93.62461999999999,-93.62463200000001,-93.6304229,-93.629079,-93.630152,-93.63073199999999,-93.625925,-93.624499,-93.625799,-93.621336,-93.62267300000001,-93.62266200000001,-93.629259,-93.629555,-93.62774089999999,-93.629884,-93.62770979999999,-93.6267969,-93.626171,-93.626762,-93.623267,-93.62192259999999,-93.621566,-93.62207600000001,-93.62820499999999,-93.62936000000001,-93.627578,-93.628236,-93.6285065,-93.62264399999999,-93.62264399999999,-93.62355599999999,-93.620487,-93.61669999999999,-93.616552,-93.6178602,-93.61699159999999,-93.615685,-93.61561399999999,-93.61717899999999,-93.6193303,-93.619022,-93.618634,-93.617544,-93.617543,-93.61322199999999,-93.614856,-93.61307699999999,-93.61239399999999,-93.613383,-93.617238,-93.61708400000001,-93.617086,-93.61940430000001,-93.615621,-93.61707,-93.612264,-93.6165173,-93.61845700000001,-93.615484,-93.615484,-93.61076,-93.612269,-93.61226499999999,-93.612253,-93.607422,-93.610663,-93.605189,-93.605659,-93.607174,-93.604106,-93.60821899999999,-93.607957,-93.608909,-93.60688,-93.605971,-93.60597,-93.60678799999999,-93.610604,-93.60794799999999,-93.60876499999999,-93.608903,-93.6071073,-93.607799,-93.60780099999999,-93.604986,-93.604837,-93.603702,-93.6195726,-93.61941499999999,-93.615582,-93.62005000000001,-93.61648000000001,-93.615596,-93.616991,-93.61710600000001,-93.6189373,-93.6195822,-93.6198478,-93.62021900000001,-93.613969,-93.615447,-93.615031,-93.61128869999999,-93.608867,-93.60887,-93.607771,-93.606764,-93.607713,-93.607724,-93.606802,-93.6077128,-93.61844000000001,-93.62017899999999,-93.620183,-93.618296,-93.618297,-93.617386,-93.616885,-93.616998,-93.616996,-93.612239,-93.613845,-93.6096888,-93.6096482,-93.61000799999999,-93.608752,-93.60755209999999,-93.6059697,-93.606196,-93.60448599999999,-93.6043548,-93.628806,-93.62671159999999,-93.625728,-93.62592600000001,-93.62471499999999,-93.625584,-93.62558900000001,-93.623666,-93.624566,-93.623514,-93.621532,-93.621532,-93.622102,-93.62468800000001,-93.62468699999999,-93.623632,-93.62256600000001,-93.622552,-93.62148500000001,-93.620392,-93.620391,-93.624591,-93.624471,-93.6218858,-93.627,-93.625556,-93.625539,-93.625434,-93.625623,-93.62167599999999,-93.621065,-93.62082939999999,-93.6259506,-93.62844699999999,-93.627729,-93.6261621,-93.6265884,-93.62949999999999,-93.625289,-93.62373799999999,-93.621921,-93.624819,-93.6246096,-93.62457999999999,-93.658159,-93.65697900000001,-93.65795799999999,-93.657031,-93.658395,-93.675551,-93.66980599999999,-93.672241,-93.676873,-93.677717,-93.67794499999999,-93.672718,-93.674179,-93.671905,-93.666954,-93.666912,-93.665521,-93.669465,-93.66372699999999,-93.66535500000001,-93.665356,-93.666979,-93.66213399999999,-93.660673,-93.66081200000001,-93.660521,-93.67812499999999,-93.67621899999999,-93.6739889,-93.67358000000002,-93.666066,-93.66621600000001,-93.666147,-93.66515,-93.665571,-93.663054,-93.663123,-93.660595,-93.66086249999999,-93.667534,-93.66506590000002,-93.66525999999999,-93.6644616,-93.66187789999999,-93.662924,-93.658528,-93.659949,-93.683223,-93.683302,-93.682029,-93.68203200000001,-93.682175,-93.691284,-93.691095,-93.68698000000001,-93.686789,-93.68119900000001,-93.681136,-93.68265129999999,-93.679198,-93.692009,-93.691213,-93.69161200000001,-93.690719,-93.689772,-93.690276,-93.690113,-93.68899500000001,-93.689019,-93.691706,-93.68910099999999,-93.688958,-93.68895000000001,-93.685742,-93.686888,-93.687518,-93.685158,-93.68586500000001,-93.6844,-93.68428900000001,-93.68315800000001,-93.68330899999999,-93.68786,-93.687704,-93.68618900000001,-93.68422700000001,-93.68439600000001,-93.68271799999999,-93.68557199999999,-93.685569,-93.68522,-93.686644,-93.686982,-93.688193,-93.687674,-93.68385600000001,-93.67994,-93.681121,-93.68369,-93.67950500000001,-93.68031000000001,-93.687825,-93.686931,-93.68520700000001,-93.684451,-93.684397,-93.684343,-93.68343499999999,-93.68343499999999,-93.68384469999999,-93.683127,-93.682991,-93.68143600000001,-93.68140099999999,-93.67830300000001,-93.678304,-93.67755099999999,-93.677407,-93.673447,-93.673571,-93.673137,-93.671443,-93.672408,-93.67529399999999,-93.675173,-93.67325099999999,-93.67558199999999,-93.674898,-93.670627,-93.665043,-93.66499229999999,-93.6653819,-93.66556199999999,-93.661895,-93.660115,-93.664925,-93.665829,-93.66450599999999,-93.663479,-93.66345,-93.66176299999999,-93.6558,-93.65668100000001,-93.656735,-93.655851,-93.64856,-93.65181,-93.651785,-93.651635,-93.647024,-93.646748,-93.648411,-93.646438,-93.646438,-93.6447083,-93.646288,-93.64842,-93.6447134,-93.64673500000001,-93.64288599999999,-93.642821,-93.64140500000001,-93.63961999999999,-93.63949700000001,-93.639467,-93.64108400000001,-93.64259699999999,-93.643378,-93.64293000000001,-93.640311,-93.640293,-93.645743,-93.645743,-93.628409,-93.627459,-93.625291,-93.625288,-93.6291062,-93.62838499999999,-93.62584289999999,-93.626835,-93.62514299999999,-93.62521700000001,-93.615426,-93.616848,-93.616849,-93.604195,-93.604193,-93.616242,-93.61658399999999,-93.61598499999999,-93.615921,-93.607342,-93.60712100000001,-93.60489,-93.60454900000001,-93.60454799999999,-93.602912,-93.602429,-93.65887499999999,-93.639793,-93.65022900000001,-93.653122,-93.653025,-93.651718,-93.649198,-93.649475,-93.650431,-93.644679,-93.64605400000001,-93.6470597,-93.64427000000001,-93.646726,-93.64664500000001,-93.647834,-93.653218,-93.64721399999999,-93.608267,-93.608343,-93.610145,-93.608197,-93.606374,-93.605351,-93.60464899999999,-93.604184,-93.603601,-93.604269,-93.601444,-93.603178,-93.603458,-93.61010899999999,-93.60510719999999,-93.604761,-93.603917,-93.603916,-93.603886,-93.60829099999999,-93.603722,-93.600066,-93.577427,-93.618504,-93.618503,-93.617273,-93.61727999999999,-93.63988000000001,-93.63736,-93.636146,-93.636972,-93.63718899999999,-93.63434100000001,-93.63634500000001,-93.640613,-93.63786500000001,-93.637883,-93.63906799999999,-93.63784800000001,-93.637299,-93.637917,-93.63583300000001,-93.63332899999999,-93.634513,-93.63291700000001,-93.63162199999999,-93.63163299999999,-93.62901599999999,-93.629075,-93.626366,-93.622874,-93.629538,-93.63007399999999,-93.628754,-93.62872299999999,-93.63787499999999,-93.638853,-93.63787600000001,-93.637816,-93.637158,-93.634665,-93.633973,-93.631528,-93.63266899999999,-93.631795,-93.638554,-93.635411,-93.63724000000001,-93.634714,-93.634581,-93.63649700000001,-93.633898,-93.633757,-93.63219100000001,-93.627736,-93.62633099999999,-93.626565,-93.62911800000001,-93.626722,-93.627112,-93.628047,-93.62805,-93.6247461,-93.621646,-93.624798,-93.624798,-93.624431,-93.62974699999999,-93.62865599999999,-93.627855,-93.62977699999999,-93.627242,-93.62723800000001,-93.62457499999999,-93.62567199999999,-93.6257971,-93.6588809,-93.65801499999999,-93.65767900000002,-93.6569765,-93.65664029999999,-93.65424299999999,-93.6570383,-93.65665679999999,-93.65627019999999,-93.657967,-93.654774,-93.65705419999999,-93.6562861,-93.657444,-93.65331,-93.657524,-93.657211,-93.657235,-93.65542479999999,-93.654689,-93.654617,-93.65091700000001,-93.657492,-93.656249,-93.6565389,-93.65698999999999,-93.6545189,-93.650597,-93.650499,-93.65045000000001,-93.650406,-93.649838,-93.649828,-93.64980799999999,-93.649798,-93.64939699999999,-93.64931900000001,-93.64929100000001,-93.653414,-93.652565,-93.65205399999999,-93.6514105,-93.65060799999999,-93.64693700000001,-93.646429,-93.645875,-93.64573,-93.64478799999999,-93.644415,-93.6443,-93.645732,-93.64437700000001,-93.64437599999999,-93.644375,-93.644374,-93.644447,-93.64436600000001,-93.641795,-93.641716,-93.640513,-93.642177,-93.644256,-93.644064,-93.643062,-93.643961,-93.64395399999999,-93.64178099999999,-93.639977,-93.64021,-93.64228799999999,-93.64333999999999,-93.641279,-93.642836,-93.643641,-93.641323,-93.640658,-93.6575919,-93.655711,-93.652894,-93.651096,-93.652641,-93.652777,-93.649901,-93.65727099999999,-93.655159,-93.655905,-93.65370799999999,-93.653712,-93.65334,-93.652946,-93.653864,-93.64411800000001,-93.64376900000001,-93.64206900000001,-93.64176,-93.642261,-93.64224299999999,-93.642234,-93.64396000000001,-93.644086,-93.64383599999999,-93.64248000000001,-93.64234399999999,-93.64354400000001,-93.643449,-93.642197,-93.641251,-93.643748,-93.642275,-93.650341,-93.65042,-93.650299,-93.650279,-93.64758399999999,-93.64793299999999,-93.648742,-93.643974,-93.64397,-93.643792,-93.64204599999999,-93.641233,-93.639574,-93.69179,-93.692245,-93.68912,-93.691244,-93.689026,-93.68976699999999,-93.688804,-93.689449,-93.68642800000001,-93.686413,-93.685185,-93.6791142,-93.677007,-93.674475,-93.673987,-93.673889,-93.66955900000001,-93.669552,-93.672318,-93.673518,-93.671941,-93.661846,-93.66184800000001,-93.660522,-93.66198300000001,-93.66213399999999,-93.653082,-93.652658,-93.647974,-93.64733000000001,-93.647166,-93.646653,-93.646637,-93.646607,-93.646576,-93.646478,-93.64410700000001,-93.644096,-93.64567099999999,-93.645639,-93.645595,-93.645482,-93.641813,-93.64218200000001,-93.639931,-93.639878,-93.643456,-93.64820399999999,-93.645798,-93.645589,-93.64577,-93.64554699999999,-93.645544,-93.658232,-93.6574973,-93.639515,-93.634545,-93.635094,-93.634528,-93.63589899999999,-93.63155399999999,-93.638051,-93.63687,-93.633748,-93.63167300000001,-93.63130200000001,-93.6299222,-93.62918999999999,-93.626518,-93.62589800000001,-93.624758,-93.62459200000001,-93.62454099999999,-93.624745,-93.623636,-93.624606,-93.623104,-93.621352,-93.62134399999999,-93.62232,-93.622834,-93.62145200000001,-93.62144000000001,-93.62694500000001,-93.63002299999999,-93.6279875,-93.626864,-93.62763750000001,-93.6249615,-93.624484,-93.62326,-93.621368,-93.622804,-93.621503,-93.629661,-93.62954499999999,-93.628654,-93.62858199999999,-93.625632,-93.623704,-93.62148500000001,-93.620497,-93.620491,-93.61938300000001,-93.61706,-93.61568699999999,-93.61609900000001,-93.61814299999999,-93.61553499999999,-93.618786,-93.618404,-93.61935200000001,-93.617521,-93.615281,-93.61686580000001,-93.613046,-93.61169700000001,-93.612347,-93.613457,-93.61398299999999,-93.61087499999999,-93.612301,-93.61926769999999,-93.617242,-93.61848000000001,-93.615622,-93.61842799999999,-93.617075,-93.612414,-93.61240100000001,-93.61245,-93.61207499999999,-93.612252,-93.618606,-93.61975179999999,-93.62034300000001,-93.61719100000001,-93.61844499999999,-93.615616,-93.61548500000001,-93.6133009,-93.61553600000001,-93.61241800000001,-93.6139,-93.613899,-93.607117,-93.609148,-93.610561,-93.605388,-93.610609,-93.61061100000001,-93.610623,-93.607957,-93.607963,-93.607963,-93.606932,-93.605968,-93.60674,-93.606782,-93.605942,-93.606746,-93.606746,-93.60483600000001,-93.603701,-93.603703,-93.604562,-93.61896900000001,-93.61698800000001,-93.61905449999999,-93.62024,-93.617135,-93.6176729,-93.615432,-93.61544600000001,-93.60861,-93.608868,-93.60685700000001,-93.606842,-93.60772299999999,-93.61047000000001,-93.6082631,-93.60678900000001,-93.618662,-93.6190848,-93.61706100000001,-93.61829400000001,-93.616911,-93.61885029999999,-93.617037,-93.615578,-93.612061,-93.612191,-93.60888199999999,-93.60945319999999,-93.608678,-93.6047622,-93.60417649999999,-93.6051157,-93.62818729999999,-93.62627160000001,-93.627605,-93.627487,-93.62471499999999,-93.623521,-93.62150099999999,-93.623643,-93.623631,-93.624539,-93.622551,-93.621504,-93.62150200000001,-93.6224,-93.62240199999999,-93.62121189999999,-93.62091319999999,-93.62576,-93.625435,-93.62528500000001,-93.62528399999999,-93.6241649,-93.62033,-93.622147,-93.62200319999999,-93.62844699999999,-93.62830099999999,-93.629496,-93.6277573,-93.62751369999999,-93.629501,-93.6274918,-93.628263,-93.62637649999999,-93.625297,-93.622793,-93.62202499999999,-93.62187999999999,-93.623814,-93.655699,-93.65612400000001,-93.657089,-93.655924,-93.675549,-93.67555900000001,-93.672207,-93.669741,-93.67441700000001,-93.67368999999999,-93.677504,-93.676953,-93.677778,-93.678149,-93.675714,-93.677997,-93.677601,-93.676681,-93.672119,-93.668544,-93.66403699999999,-93.662239,-93.66067099999999,-93.660691,-93.66198300000001,-93.661985,-93.65935500000001,-93.676272,-93.673958,-93.674582,-93.669606,-93.671221,-93.671511,-93.66504,-93.663415,-93.66712,-93.66061000000001,-93.663517,-93.6644275,-93.66468,-93.6643138,-93.658545,-93.66057000000001,-93.65965899999999,-93.658537,-93.693153,-93.684102,-93.684102,-93.68347,-93.68330899999999,-93.683226,-93.681061,-93.68682800000001,-93.686941,-93.690342,-93.69067,-93.691248,-93.691213,-93.69101499999999,-93.690234,-93.689071,-93.689071,-93.69002999999999,-93.69135,-93.688941,-93.688923,-93.68988400000001,-93.688917,-93.688931,-93.685142,-93.684944,-93.682439,-93.679216,-93.6825099,-93.683612,-93.691971,-93.692069,-93.688862,-93.69040200000001,-93.690545,-93.692286,-93.692696,-93.692415,-93.69063300000001,-93.688952,-93.68892,-93.68934400000001,-93.68951800000001,-93.691469,-93.692052,-93.684962,-93.686678,-93.6874333,-93.68773400000001,-93.684837,-93.68499300000001,-93.683235,-93.684354,-93.68315800000001,-93.68316900000001,-93.687856,-93.686575,-93.682805,-93.686795,-93.687786,-93.68675,-93.683813,-93.67958299999999,-93.681196,-93.6800822,-93.679805,-93.681274,-93.681085,-93.67950500000001,-93.687684,-93.686903,-93.684118,-93.68413699999999,-93.684714,-93.684625,-93.684285,-93.684271,-93.68373200000001,-93.6783521,-93.67754499999999,-93.676222,-93.67145499999999,-93.671454,-93.671453,-93.673197,-93.67470480000001,-93.67525000000001,-93.67432700000001,-93.674307,-93.673253,-93.67318,-93.6762195,-93.676241,-93.670587,-93.66573699999999,-93.664768,-93.663509,-93.664686,-93.661542,-93.660303,-93.665875,-93.664821,-93.66346,-93.663455,-93.663329,-93.65016199999999,-93.65181,-93.646624,-93.646688,-93.646817,-93.64672,-93.6461654,-93.644544,-93.64466899999999,-93.6447083,-93.64451099999999,-93.64469800000001,-93.64476000000001,-93.646289,-93.646416,-93.6435257,-93.643215,-93.64281,-93.641948,-93.644426,-93.641322,-93.642359,-93.6428531,-93.640497,-93.641156,-93.642663,-93.642883,-93.64400619999999,-93.639393,-93.640326,-93.64184899999999,-93.645743,-93.64492300000001,-93.645785,-93.643548,-93.643429,-93.640182,-93.628412,-93.62949999999999,-93.596569,-93.597005,-93.626875,-93.62826,-93.62531300000001,-93.629496,-93.628236,-93.626689,-93.62822199999999,-93.616969,-93.61506369999999,-93.61388599999999,-93.615697,-93.615618,-93.606577,-93.606672,-93.607597,-93.607551,-93.606933,-93.60250499999999,-93.60104,-93.60177899999999,-93.63966000000001,-93.64975699999999,-93.64944620000001,-93.650588,-93.650569,-93.650561,-93.651732,-93.650417,-93.64661700000001,-93.64436600000001,-93.646782,-93.64679099999999,-93.646788,-93.646894,-93.64663400000001,-93.64758500000001,-93.647199,-93.662162,-93.653109,-93.652429,-93.647569,-93.610102,-93.610118,-93.608192,-93.605507,-93.6072813,-93.607395,-93.605355,-93.60535400000001,-93.60354599999999,-93.601862,-93.601827,-93.601507,-93.60209399999999,-93.60342199999999,-93.604923,-93.60464399999999,-93.60452100000001,-93.60678,-93.604524,-93.602559,-93.6015199,-93.607114,-93.60485799999999,-93.605119,-93.604146,-93.6030815,-93.606641,-93.60485799999999,-93.61846199999999,-93.618182,-93.6192621,-93.616439,-93.61743199999999,-93.639067,-93.637871,-93.63748200000001,-93.63756600000001,-93.63637799999999,-93.63637300000001,-93.63605200000001,-93.63596699999999,-93.635935,-93.635837,-93.63579799999999,-93.635932,-93.631604,-93.635716,-93.636949,-93.639293,-93.634231,-93.63157200000001,-93.632571,-93.62960099999999,-93.62933700000001,-93.62917400000001,-93.628607,-93.62877,-93.62776700000001,-93.62633700000001,-93.62647699999999,-93.6220148,-93.625169,-93.63006999999999,-93.624062,-93.628269,-93.638026,-93.637874,-93.63266449999999,-93.63151499999999,-93.63300199999999,-93.635486,-93.6353259,-93.636044,-93.637387,-93.637191,-93.635367,-93.639516,-93.632369,-93.630348,-93.632355,-93.630289,-93.631162,-93.62626,-93.625373,-93.62798100000001,-93.627849,-93.62441099999999,-93.62990599999999,-93.62923189999999,-93.627932,-93.62710300000001,-93.62732800000001,-93.62736599999999,-93.62987699999999,-93.629846,-93.629516,-93.62716899999999,-93.628857,-93.62895,-93.62476700000001,-93.625709,-93.65886399999999,-93.658413,-93.658226,-93.657991,-93.65665369999999,-93.654774,-93.657984,-93.658788,-93.657118,-93.657149,-93.656864,-93.65301599999999,-93.654642,-93.652334,-93.654831,-93.653452,-93.6531488,-93.643412,-93.643862,-93.65499699999999,-93.655461,-93.657883,-93.65308400000001,-93.650626,-93.6534099,-93.649714,-93.654138,-93.649553,-93.649525,-93.654124,-93.65414,-93.650617,-93.64645899999999,-93.645841,-93.645774,-93.64441100000001,-93.64436600000001,-93.644627,-93.64575499999999,-93.64421299999999,-93.644306,-93.641662,-93.641946,-93.640287,-93.640227,-93.641321,-93.64189999999999,-93.642261,-93.642241,-93.642116,-93.641518,-93.64152199999999,-93.64458,-93.642588,-93.6436049,-93.644094,-93.64146700000001,-93.64021099999999,-93.642713,-93.64291299999999,-93.64128100000001,-93.64136000000001,-93.641283,-93.641053,-93.655739,-93.65451899999999,-93.654444,-93.65785099999999,-93.6532,-93.65227,-93.650059,-93.653915,-93.655997,-93.655704,-93.65360800000001,-93.653479,-93.64231700000001,-93.642082,-93.64178,-93.64179300000001,-93.643635,-93.643811,-93.650361,-93.64407,-93.64389799999999,-93.64286199999999,-93.642309,-93.642374,-93.64357,-93.643173,-93.64385900000001,-93.64397700000001,-93.69257500000001,-93.69140299999999,-93.69178700000001,-93.691754,-93.69095900000001,-93.69124600000001,-93.691247,-93.689021,-93.68782,-93.68769500000001,-93.686267,-93.686267,-93.68665,-93.681595,-93.68140699999999,-93.674167,-93.675963,-93.67712299999999,-93.675723,-93.677012,-93.66955299999999,-93.671037,-93.672411,-93.670919,-93.670998,-93.669708,-93.669757,-93.671926,-93.66064299999999,-93.653876,-93.65351200000001,-93.652348,-93.650474,-93.650057,-93.64783199999999,-93.64652599999999,-93.644891,-93.644891,-93.644891,-93.644891,-93.645617,-93.645482,-93.64547899999999,-93.645478,-93.643373,-93.64113999999999,-93.648931,-93.64606499999999,-93.658237,-93.633752,-93.63460000000001,-93.634738,-93.635873,-93.633466,-93.630258,-93.63247200000001,-93.633098,-93.62793499999999,-93.62597599999999,-93.626097,-93.626161,-93.627517,-93.628069,-93.62689999999999,-93.626345,-93.624612,-93.629155,-93.623414,-93.623564,-93.62231199999999,-93.630156,-93.62676500000001,-93.627065,-93.6282949,-93.621621,-93.621362,-93.622017,-93.62357,-93.622649,-93.621399,-93.628568,-93.62918500000001,-93.629256,-93.628176,-93.62702609999999,-93.62473799999999,-93.623704,-93.62264500000001,-93.62049399999999,-93.621426,-93.62143,-93.61967300000001,-93.61685199999999,-93.616012,-93.615629,-93.6181153,-93.620355,-93.61838899999999,-93.61933399999999,-93.617499,-93.61862499999999,-93.61288599999999,-93.612566,-93.615882,-93.61329599999999,-93.614251,-93.61246199999999,-93.618319,-93.618481,-93.618593,-93.617225,-93.617069,-93.612464,-93.61349100000001,-93.6136767,-93.61547400000001,-93.61547,-93.612263,-93.62035400000001,-93.617182,-93.6168701,-93.6143498,-93.61337259999999,-93.61396499999999,-93.610772,-93.612239,-93.6139,-93.61164269999999,-93.61226600000001,-93.61404899999999,-93.612404,-93.613899,-93.608909,-93.60898400000001,-93.60758199999999,-93.60942,-93.61027799999999,-93.610501,-93.607257,-93.60842199999999,-93.609073,-93.60650800000001,-93.60538400000002,-93.60674,-93.605727,-93.609054,-93.61050299999999,-93.608903,-93.608904,-93.60674299999999,-93.609053,-93.609053,-93.609053,-93.606892,-93.606893,-93.60674400000001,-93.60370399999999,-93.605839,-93.604687,-93.604568,-93.6050674,-93.618539,-93.61955189999999,-93.617164,-93.61852,-93.61713899999999,-93.618489,-93.619513,-93.615681,-93.618799,-93.618335,-93.618334,-93.615638,-93.614011,-93.6145858,-93.61461500000001,-93.613905,-93.607862,-93.608833,-93.610354,-93.60893799999999,-93.607744,-93.606803,-93.61829299999999,-93.61688599999999,-93.613889,-93.61536099999999,-93.615365,-93.615337,-93.61279580000001,-93.61376300000001,-93.608901,-93.610438,-93.60966500000001,-93.606195,-93.606869,-93.628478,-93.62602200000001,-93.627385,-93.625576,-93.62043300000001,-93.623642,-93.62454,-93.62455300000001,-93.62455300000001,-93.623481,-93.621484,-93.622415,-93.620411,-93.62437779999999,-93.62238600000001,-93.62577,-93.622773,-93.62543599999999,-93.621562,-93.622624,-93.626858,-93.62663099999999,-93.6275675,-93.6273219,-93.6255067,-93.628416,-93.629501,-93.62658879999999,-93.657017,-93.65564999999999,-93.65815000000001,-93.6578344,-93.66003959999999,-93.655885,-93.65578600000001,-93.67555,-93.67121299999999,-93.675399,-93.67445600000001,-93.67451699999999,-93.675408,-93.675606,-93.676455,-93.67227800000001,-93.674385,-93.67360100000001,-93.672645,-93.670204,-93.670514,-93.665569,-93.665505,-93.66562999999999,-93.667081,-93.669544,-93.66412,-93.668392,-93.66708199999999,-93.66203899999999,-93.67816500000001,-93.67635199999999,-93.674402,-93.676681,-93.66603499999999,-93.66515299999999,-93.66276499999999,-93.66347500000001,-93.661562,-93.660664,-93.66022700000001,-93.66625500000001,-93.667188,-93.666095,-93.666096,-93.66018699999999,-93.6627359,-93.658506,-93.685537,-93.683235,-93.68218,-93.68015149999999,-93.689071,-93.68981599999999,-93.690366,-93.69123399999999,-93.688937,-93.688925,-93.688924,-93.689485,-93.68906699999999,-93.68908399999999,-93.68878599999999,-93.688934,-93.688929,-93.68216,-93.681349,-93.68209499999999,-93.683877,-93.680401,-93.691896,-93.691873,-93.68830199999999,-93.690363,-93.69200499999999,-93.692606,-93.692221,-93.69021499999999,-93.691931,-93.689083,-93.68922999999999,-93.691243,-93.688577,-93.686629,-93.686825,-93.687158,-93.68498200000001,-93.683468,-93.683246,-93.682979,-93.687703,-93.686288,-93.684421,-93.68550399999999,-93.685697,-93.687675,-93.685571,-93.683718,-93.679147,-93.681479,-93.67980300000001,-93.682686,-93.681895,-93.681191,-93.683728,-93.68377700000001,-93.687684,-93.687685,-93.687826,-93.684089,-93.68411599999999,-93.68411599999999,-93.68536899999999,-93.685872,-93.68343499999999,-93.684332,-93.683758,-93.67826699999999,-93.67636899999999,-93.676361,-93.675297,-93.67554800000001,-93.671454,-93.67237900000001,-93.67130400000001,-93.671612,-93.670025,-93.669977,-93.669927,-93.670063,-93.670039,-93.669985,-93.66996,-93.66992399999999,-93.661951,-93.659182,-93.659222,-93.659222,-93.66011899999999,-93.66345800000001,-93.661995,-93.65675299999999,-93.6567,-93.65564000000001,-93.659043,-93.655631,-93.648562,-93.651811,-93.651285,-93.646942,-93.646619,-93.646438,-93.64841699999999,-93.648416,-93.64644,-93.6447118,-93.648415,-93.64702699999999,-93.644307,-93.64425799999999,-93.642431,-93.64048,-93.643443,-93.641155,-93.642319,-93.641002,-93.6416718,-93.640612,-93.642884,-93.644203,-93.639622,-93.639428,-93.643489,-93.628467,-93.62841,-93.628298,-93.629496,-93.626683,-93.625147,-93.615307,-93.61385,-93.61501199999999,-93.61502400000001,-93.604344,-93.61602600000001,-93.616242,-93.599969,-93.606736,-93.60367100000001,-93.603173,-93.601191,-93.603854,-93.644527,-93.651077,-93.649773,-93.64609900000001,-93.646575,-93.64617200000001,-93.646643,-93.64675800000001,-93.64789,-93.64897999999999,-93.647474,-93.610268,-93.608343,-93.60819499999999,-93.60819499999999,-93.607142,-93.60439100000001,-93.603922,-93.60351799999999,-93.60427900000001,-93.601089,-93.60318100000001,-93.60352399999999,-93.60043899999999,-93.60204299999999,-93.6019,-93.601844,-93.601615,-93.602345,-93.60447459999999,-93.603534,-93.60868840000001,-93.606842,-93.604776,-93.60268000000001,-93.606847,-93.60019,-93.599996],3,null,null,{"interactive":true,"className":"","stroke":false,"color":["#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#FFC425","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#F37735","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#F37735","#F37735","#F37735","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#FFC425","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735"],"weight":5,"opacity":0.5,"fill":true,"fillColor":["#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#FFC425","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#F37735","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#F37735","#F37735","#F37735","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#FFC425","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#F37735","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#D11141","#D11141","#D11141","#D11141","#D11141","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#F37735","#F37735","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#00AEDB","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#00AEDB","#00AEDB","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#00B159","#F37735","#F37735","#F37735","#F37735","#F37735","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#FFC425","#FFC425","#F37735","#FFC425","#FFC425","#FFC425","#FFC425","#FFC425","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735","#F37735"],"fillOpacity":0.6},null,null,null,null,null,{"interactive":false,"permanent":false,"direction":"auto","opacity":1,"offset":[0,0],"textsize":"10px","textOnly":false,"className":"","sticky":true},null]},{"method":"addCircleMarkers","args":[[42.0625775,42.009489,42.051828,42.032468,42.0280685,42.018662,42.016226,42.0212445,42.060293,42.022658,41.991883,41.9920035,42.041119,42.050441,42.05178,42.06027,42.04977,42.030482,42.033521,42.035282,42.05123500000001,42.0190125,42.060241,41.995424,42.0430335],[-93.6418475,-93.645743,-93.62802550000001,-93.6246885,-93.67204150000001,-93.687158,-93.64281,-93.66527049999999,-93.640959,-93.6259506,-93.603458,-93.60468,-93.618182,-93.625848,-93.653324,-93.65420949999999,-93.63417690000001,-93.615365,-93.671941,-93.686267,-93.64397199999999,-93.6505389,-93.63006999999999,-93.64843500000001,-93.6484545],6,null,null,{"interactive":true,"className":"","stroke":true,"color":"white","weight":1,"opacity":0.5,"fill":true,"fillColor":["#00AEDB","#F37735","#FFC425","#F37735","#00AEDB","#00AEDB","#00AEDB","#F37735","#00AEDB","#FFC425","#FFC425","#F37735","#F37735","#F37735","#D11141","#D11141","#00AEDB","#F37735","#F37735","#00AEDB","#00B159","#F37735","#D11141","#00B159","#00B159"],"fillOpacity":1},null,null,["<b>Blmngtn<\/b>: Bloomington Heights<br>Neighborhood type: Type3<br>Properties: 28<br>Median sale price: $191,500","<b>Blueste<\/b>: Blueste<br>Neighborhood type: Type4<br>Properties: 10<br>Median sale price: $130,500","<b>BrDale<\/b>: Briardale<br>Neighborhood type: Type5<br>Properties: 30<br>Median sale price: $106,000","<b>BrkSide<\/b>: Brookside<br>Neighborhood type: Type4<br>Properties: 108<br>Median sale price: $126,750","<b>ClearCr<\/b>: Clear Creek<br>Neighborhood type: Type3<br>Properties: 44<br>Median sale price: $197,500","<b>CollgCr<\/b>: College Creek<br>Neighborhood type: Type3<br>Properties: 267<br>Median sale price: $200,000","<b>Crawfor<\/b>: Crawford<br>Neighborhood type: Type3<br>Properties: 103<br>Median sale price: $200,624","<b>Edwards<\/b>: Edwards<br>Neighborhood type: Type4<br>Properties: 194<br>Median sale price: $125,000","<b>Gilbert<\/b>: Gilbert<br>Neighborhood type: Type3<br>Properties: 165<br>Median sale price: $183,000","<b>IDOTRR<\/b>: Iowa DOT and Rail Road<br>Neighborhood type: Type5<br>Properties: 93<br>Median sale price: $106,500","<b>MeadowV<\/b>: Meadow Village<br>Neighborhood type: Type5<br>Properties: 37<br>Median sale price: $88,250","<b>Mitchel<\/b>: Mitchell<br>Neighborhood type: Type4<br>Properties: 114<br>Median sale price: $153,500","<b>NAmes<\/b>: North Ames<br>Neighborhood type: Type4<br>Properties: 443<br>Median sale price: $140,000","<b>NPkVill<\/b>: Northpark Villa<br>Neighborhood type: Type4<br>Properties: 23<br>Median sale price: $143,750","<b>NoRidge<\/b>: Northridge<br>Neighborhood type: Type1<br>Properties: 71<br>Median sale price: $302,000","<b>NridgHt<\/b>: Northridge Heights<br>Neighborhood type: Type1<br>Properties: 166<br>Median sale price: $317,750","<b>NWAmes<\/b>: Northwest Ames<br>Neighborhood type: Type3<br>Properties: 131<br>Median sale price: $181,000","<b>OldTown<\/b>: Old Town<br>Neighborhood type: Type4<br>Properties: 239<br>Median sale price: $119,900","<b>Sawyer<\/b>: Sawyer<br>Neighborhood type: Type4<br>Properties: 151<br>Median sale price: $135,000","<b>SawyerW<\/b>: Sawyer West<br>Neighborhood type: Type3<br>Properties: 125<br>Median sale price: $180,000","<b>Somerst<\/b>: Somerset<br>Neighborhood type: Type2<br>Properties: 182<br>Median sale price: $225,500","<b>SWISU<\/b>: South and West of Iowa State University<br>Neighborhood type: Type4<br>Properties: 48<br>Median sale price: $136,200","<b>StoneBr<\/b>: Stone Brook<br>Neighborhood type: Type1<br>Properties: 51<br>Median sale price: $319,000","<b>Timber<\/b>: Timberland<br>Neighborhood type: Type2<br>Properties: 72<br>Median sale price: $232,106","<b>Veenker<\/b>: Veenker<br>Neighborhood type: Type2<br>Properties: 24<br>Median sale price: $250,250"],null,["Blmngtn","Blueste","BrDale","BrkSide","ClearCr","CollgCr","Crawfor","Edwards","Gilbert","IDOTRR","MeadowV","Mitchel","NAmes","NPkVill","NoRidge","NridgHt","NWAmes","OldTown","Sawyer","SawyerW","Somerst","SWISU","StoneBr","Timber","Veenker"],{"interactive":false,"permanent":true,"direction":"right","opacity":1,"offset":[0,0],"textsize":"10px","textOnly":true,"style":{"font-weight":"bold","font-size":"12px","color":"#222","text-shadow":"0 0 3px #fff, 0 0 3px #fff"},"className":"","sticky":true},null]},{"method":"addLegend","args":[{"colors":["#D11141","#00B159","#00AEDB","#F37735","#FFC425"],"labels":["Type1","Type2","Type3","Type4","Type5"],"na_color":null,"na_label":"NA","opacity":1,"position":"bottomright","type":"factor","title":"Neighborhood type","extra":null,"layerId":null,"className":"info legend","group":null}]}],"limits":{"lat":[41.986498,42.063388],"lng":[-93.693153,-93.577427]}},"evals":[],"jsHooks":[]}</script> <ul> <li>All three of the most expensive neighborhoods (Type1) are in the north of the city.</li> <li>The other types are spread across the city. For example, the cheapest type (Type5) has BrDale in the north, IDOTRR near the center and MeadowV in the south. And Timber, a Type2 neighborhood, is in the far south of the city, like Mitchel (Type4) and MeadowV (Type5). So the part of the city alone does not decide the price: neighborhoods that are close to each other can have very different prices.</li> <li>11 properties of the full dataset are in neighborhoods that our dataset does not have, so the map does not show them.</li> </ul> </div> <div id="data-summaries" class="section level2"> <h2>Data Summaries</h2> <p>The Ames Housing dataset has the sale prices of properties in Ames, Iowa, and 80 other attributes of each property. We call each attribute a feature. Each property has an <strong>Id</strong> number. The data comes in two parts: a training set and a testing set. Here are the number of rows (properties) and columns (features) in each set:</p> <pre class="r"><code>print(&quot;Dimensions of the training set&quot;)</code></pre> <pre><code>[1] &quot;Dimensions of the training set&quot;</code></pre> <pre class="r"><code>dim(dfTrain)</code></pre> <pre><code>[1] 1460 81</code></pre> <pre class="r"><code>vTrainIDs &lt;- dfTrain$Id print(&quot;Dimensions of the testing set&quot;)</code></pre> <pre><code>[1] &quot;Dimensions of the testing set&quot;</code></pre> <pre class="r"><code>dim(dfTest)</code></pre> <pre><code>[1] 1459 81</code></pre> <pre class="r"><code>vTestIDs &lt;- dfTest$Id dfClean &lt;- dfTrain %&gt;% bind_rows(dfTest)</code></pre> <p>The data description explains each feature, and the meaning of each category. It is long, so it is folded:</p> <pre class="r"><code># data_description.txt has one line &quot;Feature: description&quot; for each feature, # followed by one line &quot;code&lt;TAB&gt;meaning&quot; for each of its categories, if it has any vLines &lt;- readLines(&quot;data_description.txt&quot;) vStarts &lt;- grep(&quot;^[[:alnum:]]+:&quot;, vLines) vEnds &lt;- c(vStarts[-1] - 1, length(vLines)) dfDescription &lt;- map2_dfr(vStarts, vEnds, function(iStart, iEnd) { vParts &lt;- str_split_fixed(vLines[iStart], &quot;:&quot;, 2) vCategories &lt;- trimws(vLines[seq_len(iEnd - iStart) + iStart]) vCategories &lt;- vCategories[vCategories != &quot;&quot;] vCodes &lt;- str_split_fixed(vCategories, &quot; &quot;, 2) tibble(Feature = vParts[1], Description = trimws(vParts[2]), Categories = if (length(vCategories) &gt; 0) paste0(&quot;&lt;b&gt;&quot;, trimws(vCodes[, 1]), &quot;&lt;/b&gt;: &quot;, trimws(vCodes[, 2]), collapse = &quot;&lt;br&gt;&quot;) else &quot;&quot;) }) cat(&#39;&lt;details class=&quot;code-fold&quot;&gt;&lt;summary&gt;Show the data description&lt;/summary&gt; &#39;)</code></pre> <details class="code-fold"> <summary> Show the data description </summary> <pre class="r"><code>cat(knitr::kable(dfDescription, format = &quot;html&quot;, escape = FALSE), sep = &quot; &quot;)</code></pre> <table> <thead> <tr> <th style="text-align:left;"> Feature </th> <th style="text-align:left;"> Description </th> <th style="text-align:left;"> Categories </th> </tr> </thead> <tbody> <tr> <td style="text-align:left;"> MSSubClass </td> <td style="text-align:left;"> Identifies the type of dwelling involved in the sale. </td> <td style="text-align:left;"> <b>20</b>: 1-STORY 1946 &amp; NEWER ALL STYLES<br><b>30</b>: 1-STORY 1945 &amp; OLDER<br><b>40</b>: 1-STORY W/FINISHED ATTIC ALL AGES<br><b>45</b>: 1-1/2 STORY - UNFINISHED ALL AGES<br><b>50</b>: 1-1/2 STORY FINISHED ALL AGES<br><b>60</b>: 2-STORY 1946 &amp; NEWER<br><b>70</b>: 2-STORY 1945 &amp; OLDER<br><b>75</b>: 2-1/2 STORY ALL AGES<br><b>80</b>: SPLIT OR MULTI-LEVEL<br><b>85</b>: SPLIT FOYER<br><b>90</b>: DUPLEX - ALL STYLES AND AGES<br><b>120</b>: 1-STORY PUD (Planned Unit Development) - 1946 &amp; NEWER<br><b>150</b>: 1-1/2 STORY PUD - ALL AGES<br><b>160</b>: 2-STORY PUD - 1946 &amp; NEWER<br><b>180</b>: PUD - MULTILEVEL - INCL SPLIT LEV/FOYER<br><b>190</b>: 2 FAMILY CONVERSION - ALL STYLES AND AGES </td> </tr> <tr> <td style="text-align:left;"> MSZoning </td> <td style="text-align:left;"> Identifies the general zoning classification of the sale. </td> <td style="text-align:left;"> <b>A</b>: Agriculture<br><b>C</b>: Commercial<br><b>FV</b>: Floating Village Residential<br><b>I</b>: Industrial<br><b>RH</b>: Residential High Density<br><b>RL</b>: Residential Low Density<br><b>RP</b>: Residential Low Density Park<br><b>RM</b>: Residential Medium Density </td> </tr> <tr> <td style="text-align:left;"> LotFrontage </td> <td style="text-align:left;"> Linear feet of street connected to property </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> LotArea </td> <td style="text-align:left;"> Lot size in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> Street </td> <td style="text-align:left;"> Type of road access to property </td> <td style="text-align:left;"> <b>Grvl</b>: Gravel<br><b>Pave</b>: Paved </td> </tr> <tr> <td style="text-align:left;"> Alley </td> <td style="text-align:left;"> Type of alley access to property </td> <td style="text-align:left;"> <b>Grvl</b>: Gravel<br><b>Pave</b>: Paved<br><b>NA</b>: No alley access </td> </tr> <tr> <td style="text-align:left;"> LotShape </td> <td style="text-align:left;"> General shape of property </td> <td style="text-align:left;"> <b>Reg</b>: Regular<br><b>IR1</b>: Slightly irregular<br><b>IR2</b>: Moderately Irregular<br><b>IR3</b>: Irregular </td> </tr> <tr> <td style="text-align:left;"> LandContour </td> <td style="text-align:left;"> Flatness of the property </td> <td style="text-align:left;"> <b>Lvl</b>: Near Flat/Level<br><b>Bnk</b>: Banked - Quick and significant rise from street grade to building<br><b>HLS</b>: Hillside - Significant slope from side to side<br><b>Low</b>: Depression </td> </tr> <tr> <td style="text-align:left;"> Utilities </td> <td style="text-align:left;"> Type of utilities available </td> <td style="text-align:left;"> <b>AllPub</b>: All public Utilities (E,G,W,&amp; S)<br><b>NoSewr</b>: Electricity, Gas, and Water (Septic Tank)<br><b>NoSeWa</b>: Electricity and Gas Only<br><b>ELO</b>: Electricity only </td> </tr> <tr> <td style="text-align:left;"> LotConfig </td> <td style="text-align:left;"> Lot configuration </td> <td style="text-align:left;"> <b>Inside</b>: Inside lot<br><b>Corner</b>: Corner lot<br><b>CulDSac</b>: Cul-de-sac<br><b>FR2</b>: Frontage on 2 sides of property<br><b>FR3</b>: Frontage on 3 sides of property </td> </tr> <tr> <td style="text-align:left;"> LandSlope </td> <td style="text-align:left;"> Slope of property </td> <td style="text-align:left;"> <b>Gtl</b>: Gentle slope<br><b>Mod</b>: Moderate Slope<br><b>Sev</b>: Severe Slope </td> </tr> <tr> <td style="text-align:left;"> Neighborhood </td> <td style="text-align:left;"> Physical locations within Ames city limits </td> <td style="text-align:left;"> <b>Blmngtn</b>: Bloomington Heights<br><b>Blueste</b>: Bluestem<br><b>BrDale</b>: Briardale<br><b>BrkSide</b>: Brookside<br><b>ClearCr</b>: Clear Creek<br><b>CollgCr</b>: College Creek<br><b>Crawfor</b>: Crawford<br><b>Edwards</b>: Edwards<br><b>Gilbert</b>: Gilbert<br><b>IDOTRR</b>: Iowa DOT and Rail Road<br><b>MeadowV</b>: Meadow Village<br><b>Mitchel</b>: Mitchell<br><b>Names</b>: North Ames<br><b>NoRidge</b>: Northridge<br><b>NPkVill</b>: Northpark Villa<br><b>NridgHt</b>: Northridge Heights<br><b>NWAmes</b>: Northwest Ames<br><b>OldTown</b>: Old Town<br><b>SWISU</b>: South &amp; West of Iowa State University<br><b>Sawyer</b>: Sawyer<br><b>SawyerW</b>: Sawyer West<br><b>Somerst</b>: Somerset<br><b>StoneBr</b>: Stone Brook<br><b>Timber</b>: Timberland<br><b>Veenker</b>: Veenker </td> </tr> <tr> <td style="text-align:left;"> Condition1 </td> <td style="text-align:left;"> Proximity to various conditions </td> <td style="text-align:left;"> <b>Artery</b>: Adjacent to arterial street<br><b>Feedr</b>: Adjacent to feeder street<br><b>Norm</b>: Normal<br><b>RRNn</b>: Within 200’ of North-South Railroad<br><b>RRAn</b>: Adjacent to North-South Railroad<br><b>PosN</b>: Near positive off-site feature–park, greenbelt, etc.<br><b>PosA</b>: Adjacent to postive off-site feature<br><b>RRNe</b>: Within 200’ of East-West Railroad<br><b>RRAe</b>: Adjacent to East-West Railroad </td> </tr> <tr> <td style="text-align:left;"> Condition2 </td> <td style="text-align:left;"> Proximity to various conditions (if more than one is present) </td> <td style="text-align:left;"> <b>Artery</b>: Adjacent to arterial street<br><b>Feedr</b>: Adjacent to feeder street<br><b>Norm</b>: Normal<br><b>RRNn</b>: Within 200’ of North-South Railroad<br><b>RRAn</b>: Adjacent to North-South Railroad<br><b>PosN</b>: Near positive off-site feature–park, greenbelt, etc.<br><b>PosA</b>: Adjacent to postive off-site feature<br><b>RRNe</b>: Within 200’ of East-West Railroad<br><b>RRAe</b>: Adjacent to East-West Railroad </td> </tr> <tr> <td style="text-align:left;"> BldgType </td> <td style="text-align:left;"> Type of dwelling </td> <td style="text-align:left;"> <b>1Fam</b>: Single-family Detached<br><b>2FmCon</b>: Two-family Conversion; originally built as one-family dwelling<br><b>Duplx</b>: Duplex<br><b>TwnhsE</b>: Townhouse End Unit<br><b>TwnhsI</b>: Townhouse Inside Unit </td> </tr> <tr> <td style="text-align:left;"> HouseStyle </td> <td style="text-align:left;"> Style of dwelling </td> <td style="text-align:left;"> <b>1Story</b>: One story<br><b>1.5Fin</b>: One and one-half story - 2nd level finished<br><b>1.5Unf</b>: One and one-half story - 2nd level unfinished<br><b>2Story</b>: Two story<br><b>2.5Fin</b>: Two and one-half story - 2nd level finished<br><b>2.5Unf</b>: Two and one-half story - 2nd level unfinished<br><b>SFoyer</b>: Split Foyer<br><b>SLvl</b>: Split Level </td> </tr> <tr> <td style="text-align:left;"> OverallQual </td> <td style="text-align:left;"> Rates the overall material and finish of the house </td> <td style="text-align:left;"> <b>10</b>: Very Excellent<br><b>9</b>: Excellent<br><b>8</b>: Very Good<br><b>7</b>: Good<br><b>6</b>: Above Average<br><b>5</b>: Average<br><b>4</b>: Below Average<br><b>3</b>: Fair<br><b>2</b>: Poor<br><b>1</b>: Very Poor </td> </tr> <tr> <td style="text-align:left;"> OverallCond </td> <td style="text-align:left;"> Rates the overall condition of the house </td> <td style="text-align:left;"> <b>10</b>: Very Excellent<br><b>9</b>: Excellent<br><b>8</b>: Very Good<br><b>7</b>: Good<br><b>6</b>: Above Average<br><b>5</b>: Average<br><b>4</b>: Below Average<br><b>3</b>: Fair<br><b>2</b>: Poor<br><b>1</b>: Very Poor </td> </tr> <tr> <td style="text-align:left;"> YearBuilt </td> <td style="text-align:left;"> Original construction date </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> YearRemodAdd </td> <td style="text-align:left;"> Remodel date (same as construction date if no remodeling or additions) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> RoofStyle </td> <td style="text-align:left;"> Type of roof </td> <td style="text-align:left;"> <b>Flat</b>: Flat<br><b>Gable</b>: Gable<br><b>Gambrel</b>: Gabrel (Barn)<br><b>Hip</b>: Hip<br><b>Mansard</b>: Mansard<br><b>Shed</b>: Shed </td> </tr> <tr> <td style="text-align:left;"> RoofMatl </td> <td style="text-align:left;"> Roof material </td> <td style="text-align:left;"> <b>ClyTile</b>: Clay or Tile<br><b>CompShg</b>: Standard (Composite) Shingle<br><b>Membran</b>: Membrane<br><b>Metal</b>: Metal<br><b>Roll</b>: Roll<br><b>Tar&amp;Grv</b>: Gravel &amp; Tar<br><b>WdShake</b>: Wood Shakes<br><b>WdShngl</b>: Wood Shingles </td> </tr> <tr> <td style="text-align:left;"> Exterior1st </td> <td style="text-align:left;"> Exterior covering on house </td> <td style="text-align:left;"> <b>AsbShng</b>: Asbestos Shingles<br><b>AsphShn</b>: Asphalt Shingles<br><b>BrkComm</b>: Brick Common<br><b>BrkFace</b>: Brick Face<br><b>CBlock</b>: Cinder Block<br><b>CemntBd</b>: Cement Board<br><b>HdBoard</b>: Hard Board<br><b>ImStucc</b>: Imitation Stucco<br><b>MetalSd</b>: Metal Siding<br><b>Other</b>: Other<br><b>Plywood</b>: Plywood<br><b>PreCast</b>: PreCast<br><b>Stone</b>: Stone<br><b>Stucco</b>: Stucco<br><b>VinylSd</b>: Vinyl Siding<br><b>Wd Sdng</b>: Wood Siding<br><b>WdShing</b>: Wood Shingles </td> </tr> <tr> <td style="text-align:left;"> Exterior2nd </td> <td style="text-align:left;"> Exterior covering on house (if more than one material) </td> <td style="text-align:left;"> <b>AsbShng</b>: Asbestos Shingles<br><b>AsphShn</b>: Asphalt Shingles<br><b>BrkComm</b>: Brick Common<br><b>BrkFace</b>: Brick Face<br><b>CBlock</b>: Cinder Block<br><b>CemntBd</b>: Cement Board<br><b>HdBoard</b>: Hard Board<br><b>ImStucc</b>: Imitation Stucco<br><b>MetalSd</b>: Metal Siding<br><b>Other</b>: Other<br><b>Plywood</b>: Plywood<br><b>PreCast</b>: PreCast<br><b>Stone</b>: Stone<br><b>Stucco</b>: Stucco<br><b>VinylSd</b>: Vinyl Siding<br><b>Wd Sdng</b>: Wood Siding<br><b>WdShing</b>: Wood Shingles </td> </tr> <tr> <td style="text-align:left;"> MasVnrType </td> <td style="text-align:left;"> Masonry veneer type </td> <td style="text-align:left;"> <b>BrkCmn</b>: Brick Common<br><b>BrkFace</b>: Brick Face<br><b>CBlock</b>: Cinder Block<br><b>None</b>: None<br><b>Stone</b>: Stone </td> </tr> <tr> <td style="text-align:left;"> MasVnrArea </td> <td style="text-align:left;"> Masonry veneer area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> ExterQual </td> <td style="text-align:left;"> Evaluates the quality of the material on the exterior </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Average/Typical<br><b>Fa</b>: Fair<br><b>Po</b>: Poor </td> </tr> <tr> <td style="text-align:left;"> ExterCond </td> <td style="text-align:left;"> Evaluates the present condition of the material on the exterior </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Average/Typical<br><b>Fa</b>: Fair<br><b>Po</b>: Poor </td> </tr> <tr> <td style="text-align:left;"> Foundation </td> <td style="text-align:left;"> Type of foundation </td> <td style="text-align:left;"> <b>BrkTil</b>: Brick &amp; Tile<br><b>CBlock</b>: Cinder Block<br><b>PConc</b>: Poured Contrete<br><b>Slab</b>: Slab<br><b>Stone</b>: Stone<br><b>Wood</b>: Wood </td> </tr> <tr> <td style="text-align:left;"> BsmtQual </td> <td style="text-align:left;"> Evaluates the height of the basement </td> <td style="text-align:left;"> <b>Ex</b>: Excellent (100+ inches)<br><b>Gd</b>: Good (90-99 inches)<br><b>TA</b>: Typical (80-89 inches)<br><b>Fa</b>: Fair (70-79 inches)<br><b>Po</b>: Poor (&lt;70 inches<br><b>NA</b>: No Basement </td> </tr> <tr> <td style="text-align:left;"> BsmtCond </td> <td style="text-align:left;"> Evaluates the general condition of the basement </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Typical - slight dampness allowed<br><b>Fa</b>: Fair - dampness or some cracking or settling<br><b>Po</b>: Poor - Severe cracking, settling, or wetness<br><b>NA</b>: No Basement </td> </tr> <tr> <td style="text-align:left;"> BsmtExposure </td> <td style="text-align:left;"> Refers to walkout or garden level walls </td> <td style="text-align:left;"> <b>Gd</b>: Good Exposure<br><b>Av</b>: Average Exposure (split levels or foyers typically score average or above)<br><b>Mn</b>: Mimimum Exposure<br><b>No</b>: No Exposure<br><b>NA</b>: No Basement </td> </tr> <tr> <td style="text-align:left;"> BsmtFinType1 </td> <td style="text-align:left;"> Rating of basement finished area </td> <td style="text-align:left;"> <b>GLQ</b>: Good Living Quarters<br><b>ALQ</b>: Average Living Quarters<br><b>BLQ</b>: Below Average Living Quarters<br><b>Rec</b>: Average Rec Room<br><b>LwQ</b>: Low Quality<br><b>Unf</b>: Unfinshed<br><b>NA</b>: No Basement </td> </tr> <tr> <td style="text-align:left;"> BsmtFinSF1 </td> <td style="text-align:left;"> Type 1 finished square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> BsmtFinType2 </td> <td style="text-align:left;"> Rating of basement finished area (if multiple types) </td> <td style="text-align:left;"> <b>GLQ</b>: Good Living Quarters<br><b>ALQ</b>: Average Living Quarters<br><b>BLQ</b>: Below Average Living Quarters<br><b>Rec</b>: Average Rec Room<br><b>LwQ</b>: Low Quality<br><b>Unf</b>: Unfinshed<br><b>NA</b>: No Basement </td> </tr> <tr> <td style="text-align:left;"> BsmtFinSF2 </td> <td style="text-align:left;"> Type 2 finished square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> BsmtUnfSF </td> <td style="text-align:left;"> Unfinished square feet of basement area </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> TotalBsmtSF </td> <td style="text-align:left;"> Total square feet of basement area </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> Heating </td> <td style="text-align:left;"> Type of heating </td> <td style="text-align:left;"> <b>Floor</b>: Floor Furnace<br><b>GasA</b>: Gas forced warm air furnace<br><b>GasW</b>: Gas hot water or steam heat<br><b>Grav</b>: Gravity furnace<br><b>OthW</b>: Hot water or steam heat other than gas<br><b>Wall</b>: Wall furnace </td> </tr> <tr> <td style="text-align:left;"> HeatingQC </td> <td style="text-align:left;"> Heating quality and condition </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Average/Typical<br><b>Fa</b>: Fair<br><b>Po</b>: Poor </td> </tr> <tr> <td style="text-align:left;"> CentralAir </td> <td style="text-align:left;"> Central air conditioning </td> <td style="text-align:left;"> <b>N</b>: No<br><b>Y</b>: Yes </td> </tr> <tr> <td style="text-align:left;"> Electrical </td> <td style="text-align:left;"> Electrical system </td> <td style="text-align:left;"> <b>SBrkr</b>: Standard Circuit Breakers &amp; Romex<br><b>FuseA</b>: Fuse Box over 60 AMP and all Romex wiring (Average)<br><b>FuseF</b>: 60 AMP Fuse Box and mostly Romex wiring (Fair)<br><b>FuseP</b>: 60 AMP Fuse Box and mostly knob &amp; tube wiring (poor)<br><b>Mix</b>: Mixed </td> </tr> <tr> <td style="text-align:left;"> 1stFlrSF </td> <td style="text-align:left;"> First Floor square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> 2ndFlrSF </td> <td style="text-align:left;"> Second floor square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> LowQualFinSF </td> <td style="text-align:left;"> Low quality finished square feet (all floors) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> GrLivArea </td> <td style="text-align:left;"> Above grade (ground) living area square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> BsmtFullBath </td> <td style="text-align:left;"> Basement full bathrooms </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> BsmtHalfBath </td> <td style="text-align:left;"> Basement half bathrooms </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> FullBath </td> <td style="text-align:left;"> Full bathrooms above grade </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> HalfBath </td> <td style="text-align:left;"> Half baths above grade </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> Bedroom </td> <td style="text-align:left;"> Bedrooms above grade (does NOT include basement bedrooms) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> KitchenAbvGr </td> <td style="text-align:left;"> Kitchens above grade </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> KitchenQual </td> <td style="text-align:left;"> Kitchen quality </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Typical/Average<br><b>Fa</b>: Fair<br><b>Po</b>: Poor </td> </tr> <tr> <td style="text-align:left;"> TotRmsAbvGrd </td> <td style="text-align:left;"> Total rooms above grade (does not include bathrooms) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> Functional </td> <td style="text-align:left;"> Home functionality (Assume typical unless deductions are warranted) </td> <td style="text-align:left;"> <b>Typ</b>: Typical Functionality<br><b>Min1</b>: Minor Deductions 1<br><b>Min2</b>: Minor Deductions 2<br><b>Mod</b>: Moderate Deductions<br><b>Maj1</b>: Major Deductions 1<br><b>Maj2</b>: Major Deductions 2<br><b>Sev</b>: Severely Damaged<br><b>Sal</b>: Salvage only </td> </tr> <tr> <td style="text-align:left;"> Fireplaces </td> <td style="text-align:left;"> Number of fireplaces </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> FireplaceQu </td> <td style="text-align:left;"> Fireplace quality </td> <td style="text-align:left;"> <b>Ex</b>: Excellent - Exceptional Masonry Fireplace<br><b>Gd</b>: Good - Masonry Fireplace in main level<br><b>TA</b>: Average - Prefabricated Fireplace in main living area or Masonry Fireplace in basement<br><b>Fa</b>: Fair - Prefabricated Fireplace in basement<br><b>Po</b>: Poor - Ben Franklin Stove<br><b>NA</b>: No Fireplace </td> </tr> <tr> <td style="text-align:left;"> GarageType </td> <td style="text-align:left;"> Garage location </td> <td style="text-align:left;"> <b>2Types</b>: More than one type of garage<br><b>Attchd</b>: Attached to home<br><b>Basment</b>: Basement Garage<br><b>BuiltIn</b>: Built-In (Garage part of house - typically has room above garage)<br><b>CarPort</b>: Car Port<br><b>Detchd</b>: Detached from home<br><b>NA</b>: No Garage </td> </tr> <tr> <td style="text-align:left;"> GarageYrBlt </td> <td style="text-align:left;"> Year garage was built </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> GarageFinish </td> <td style="text-align:left;"> Interior finish of the garage </td> <td style="text-align:left;"> <b>Fin</b>: Finished<br><b>RFn</b>: Rough Finished<br><b>Unf</b>: Unfinished<br><b>NA</b>: No Garage </td> </tr> <tr> <td style="text-align:left;"> GarageCars </td> <td style="text-align:left;"> Size of garage in car capacity </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> GarageArea </td> <td style="text-align:left;"> Size of garage in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> GarageQual </td> <td style="text-align:left;"> Garage quality </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Typical/Average<br><b>Fa</b>: Fair<br><b>Po</b>: Poor<br><b>NA</b>: No Garage </td> </tr> <tr> <td style="text-align:left;"> GarageCond </td> <td style="text-align:left;"> Garage condition </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Typical/Average<br><b>Fa</b>: Fair<br><b>Po</b>: Poor<br><b>NA</b>: No Garage </td> </tr> <tr> <td style="text-align:left;"> PavedDrive </td> <td style="text-align:left;"> Paved driveway </td> <td style="text-align:left;"> <b>Y</b>: Paved<br><b>P</b>: Partial Pavement<br><b>N</b>: Dirt/Gravel </td> </tr> <tr> <td style="text-align:left;"> WoodDeckSF </td> <td style="text-align:left;"> Wood deck area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> OpenPorchSF </td> <td style="text-align:left;"> Open porch area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> EnclosedPorch </td> <td style="text-align:left;"> Enclosed porch area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> 3SsnPorch </td> <td style="text-align:left;"> Three season porch area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> ScreenPorch </td> <td style="text-align:left;"> Screen porch area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> PoolArea </td> <td style="text-align:left;"> Pool area in square feet </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> PoolQC </td> <td style="text-align:left;"> Pool quality </td> <td style="text-align:left;"> <b>Ex</b>: Excellent<br><b>Gd</b>: Good<br><b>TA</b>: Average/Typical<br><b>Fa</b>: Fair<br><b>NA</b>: No Pool </td> </tr> <tr> <td style="text-align:left;"> Fence </td> <td style="text-align:left;"> Fence quality </td> <td style="text-align:left;"> <b>GdPrv</b>: Good Privacy<br><b>MnPrv</b>: Minimum Privacy<br><b>GdWo</b>: Good Wood<br><b>MnWw</b>: Minimum Wood/Wire<br><b>NA</b>: No Fence </td> </tr> <tr> <td style="text-align:left;"> MiscFeature </td> <td style="text-align:left;"> Miscellaneous feature not covered in other categories </td> <td style="text-align:left;"> <b>Elev</b>: Elevator<br><b>Gar2</b>: 2nd Garage (if not described in garage section)<br><b>Othr</b>: Other<br><b>Shed</b>: Shed (over 100 SF)<br><b>TenC</b>: Tennis Court<br><b>NA</b>: None </td> </tr> <tr> <td style="text-align:left;"> MiscVal </td> <td style="text-align:left;"> $Value of miscellaneous feature </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> MoSold </td> <td style="text-align:left;"> Month Sold (MM) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> YrSold </td> <td style="text-align:left;"> Year Sold (YYYY) </td> <td style="text-align:left;"> </td> </tr> <tr> <td style="text-align:left;"> SaleType </td> <td style="text-align:left;"> Type of sale </td> <td style="text-align:left;"> <b>WD</b>: Warranty Deed - Conventional<br><b>CWD</b>: Warranty Deed - Cash<br><b>VWD</b>: Warranty Deed - VA Loan<br><b>New</b>: Home just constructed and sold<br><b>COD</b>: Court Officer Deed/Estate<br><b>Con</b>: Contract 15% Down payment regular terms<br><b>ConLw</b>: Contract Low Down payment and low interest<br><b>ConLI</b>: Contract Low Interest<br><b>ConLD</b>: Contract Low Down<br><b>Oth</b>: Other </td> </tr> <tr> <td style="text-align:left;"> SaleCondition </td> <td style="text-align:left;"> Condition of sale </td> <td style="text-align:left;"> <b>Normal</b>: Normal Sale<br><b>Abnorml</b>: Abnormal Sale - trade, foreclosure, short sale<br><b>AdjLand</b>: Adjoining Land Purchase<br><b>Alloca</b>: Allocation - two linked properties with separate deeds, typically condo with a garage unit<br><b>Family</b>: Sale between family members<br><b>Partial</b>: Home was not completed when last assessed (associated with New Homes) </td> </tr> </tbody> </table> <pre class="r"><code>cat(&quot; &lt;/details&gt; &quot;)</code></pre> </details> <p>Now, let’s put the training set and the testing set together in one dataset, and count the missing values in each feature:</p> <pre class="r"><code>eda_plot_missing_counts(dfClean)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/missing_values-1.png" alt="" width="672" /></p> <div id="whats-notable" class="section level4"> <h4>What’s notable?</h4> <ul> <li>The combined dataset has 2919 properties.</li> <li>Very few properties have a pool, a fence or access to an alley.</li> <li>Very few properties have a special feature that other features do not describe (<code>MiscFeature</code>).</li> <li>More than 12 features have at least 1 missing value. The dataset is small, so we do not want to remove properties. Instead, we fill in each missing value with a likely value. This is called imputation.</li> </ul> </div> </div> <div id="data-cleaning-transformation" class="section level2"> <h2>Data Cleaning &amp; Transformation</h2> <p>In this section, we look at the features of the complete dataset, and we make a set of steps to clean the data.</p> <div id="fixing-data-errors" class="section level3"> <h3>Fixing Data Errors</h3> <p>First, we check that the data makes sense:</p> <ul> <li><code>YearRemodAdd</code> (the year of the last remodel) must not be earlier than <code>YearBuilt</code>: 1 property to fix</li> <li><code>YrSold</code> (the year of the sale) must not be earlier than <code>YearRemodAdd</code>: 3 properties to fix</li> </ul> <pre class="r"><code>dfClean %&gt;% select(Id, YearBuilt, YearRemodAdd, YrSold) %&gt;% dplyr::filter(YearRemodAdd &lt; YearBuilt)</code></pre> <pre><code># A tibble: 1 × 4 Id YearBuilt YearRemodAdd YrSold &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 1877 2002 2001 2009</code></pre> <pre class="r"><code>dfClean %&gt;% select(Id, YearBuilt, YearRemodAdd, YrSold) %&gt;% dplyr::filter(YrSold &lt; YearRemodAdd)</code></pre> <pre><code># A tibble: 3 × 4 Id YearBuilt YearRemodAdd YrSold &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 524 2007 2008 2007 2 2296 2007 2008 2007 3 2550 2008 2009 2007</code></pre> <pre class="r"><code># fix errors dfClean &lt;- dfClean %&gt;% mutate(YearRemodAdd = ifelse(YearRemodAdd &lt; YearBuilt, YearBuilt, YearRemodAdd)) %&gt;% mutate(YrSold = ifelse(YrSold &lt; YearRemodAdd, YearRemodAdd, YrSold))</code></pre> <ul> <li><code>GarageYrBlt</code> (the year the garage was built) must not be earlier than <code>YearBuilt</code>: 18 properties to fix</li> <li><code>GarageYrBlt</code> must not be later than <code>YrSold</code>: 1 property to fix</li> </ul> <pre class="r"><code>dfClean %&gt;% select(Id, YearBuilt, GarageYrBlt, YrSold) %&gt;% dplyr::filter(GarageYrBlt &lt; YearBuilt)</code></pre> <pre><code># A tibble: 18 × 4 Id YearBuilt GarageYrBlt YrSold &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 30 1927 1920 2008 2 94 1910 1900 2007 3 325 1967 1961 2010 4 601 2005 2003 2006 5 737 1950 1949 2006 6 1104 1959 1954 2006 7 1377 1930 1925 2008 8 1415 1923 1922 2008 9 1419 1963 1962 2008 10 1522 1959 1956 2010 11 1577 2010 2009 2010 12 1806 1935 1920 2009 13 1841 1978 1960 2009 14 1896 1941 1940 2009 15 1898 1935 1926 2009 16 2123 1945 1925 2008 17 2264 2006 2005 2007 18 2510 2006 2005 2007</code></pre> <pre class="r"><code>dfClean %&gt;% select(Id, YearBuilt, GarageYrBlt, YrSold) %&gt;% dplyr::filter(GarageYrBlt &gt; YrSold)</code></pre> <pre><code># A tibble: 1 × 4 Id YearBuilt GarageYrBlt YrSold &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 2593 2006 2207 2007</code></pre> <pre class="r"><code># fix errors dfClean &lt;- dfClean %&gt;% mutate(GarageYrBlt = ifelse(GarageYrBlt &lt; YearBuilt, YearBuilt, GarageYrBlt)) %&gt;% mutate(GarageYrBlt = ifelse(GarageYrBlt &gt; YrSold, YrSold, GarageYrBlt))</code></pre> </div> <div id="imputing-missing-values-new-features" class="section level3 tabset tabset-fade tabset-pills"> <h3>Imputing Missing Values &amp; New Features</h3> <!-- #### <span style="color:red">Basement Features</span> --> <div id="basement-features" class="section level4"> <h4>Basement Features</h4> <ol style="list-style-type: decimal"> <li><p>For one property (<code>Id</code> = 2121), all the basement features are missing (NA). We set <code>TotalBsmtSF</code> (the total basement area) to 0.</p></li> <li><p>Now 79 properties have no basement (<code>TotalBsmtSF</code> = 0). For these properties, we change all the missing basement values to <strong>None</strong>.</p></li> <li><p>The quality of basements is different in different neighborhoods. So we fill in each remaining missing value with the most common value in the same <code>Neighborhood</code>.</p></li> </ol> <pre class="r"><code># 1. Change TotalBsmtSF to 0 when all basement features are NA ##################################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Bsmt&quot;)) %&gt;% dplyr::filter(is.na(TotalBsmtSF))</code></pre> <pre><code># A tibble: 1 × 13 Id Neighborhood BsmtQual BsmtCond BsmtExposure BsmtFinType1 BsmtFinSF1 BsmtFinType2 BsmtFinSF2 BsmtUnfSF TotalBsmtSF BsmtFullBath BsmtHalfBath &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 2121 BrkSide &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; NA &lt;NA&gt; NA NA NA NA NA</code></pre> <pre class="r"><code>dfClean &lt;- dfClean %&gt;% mutate(TotalBsmtSF = if_else(Id == 2121, 0, TotalBsmtSF)) # 2. Change NA values to None or 0 if there is no basement ##################################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Bsmt&quot;)) %&gt;% dplyr::filter(TotalBsmtSF == 0)</code></pre> <pre><code># A tibble: 79 × 13 Id Neighborhood BsmtQual BsmtCond BsmtExposure BsmtFinType1 BsmtFinSF1 BsmtFinType2 BsmtFinSF2 BsmtUnfSF TotalBsmtSF BsmtFullBath BsmtHalfBath &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 18 Sawyer &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 2 40 Edwards &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 3 91 NAmes &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 4 103 SawyerW &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 5 157 NAmes &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 6 183 Edwards &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 7 260 OldTown &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 8 343 NAmes &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 9 363 Edwards &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 10 372 ClearCr &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; &lt;NA&gt; 0 &lt;NA&gt; 0 0 0 0 0 # ℹ 69 more rows</code></pre> <pre class="r"><code>vFeaturesChr_bsmt &lt;- dfClean %&gt;% select(contains(&quot;Bsmt&quot;)) %&gt;% select_if(is.character) %&gt;% colnames() vFeaturesNum_bsmt &lt;- dfClean %&gt;% select(contains(&quot;Bsmt&quot;)) %&gt;% select_if(is.numeric) %&gt;% colnames() dfClean &lt;- dfClean %&gt;% mutate_at(all_of(vFeaturesChr_bsmt), ~ case_when(TotalBsmtSF == 0 ~ &quot;None&quot;, TRUE ~ .)) %&gt;% mutate_at(all_of(vFeaturesNum_bsmt), ~ case_when(TotalBsmtSF == 0 ~ 0, TRUE ~ .)) # 3. Change NA values in remaining features to the most common value in the neighborhood where TotalBsmtSF &gt; 0 ##################################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Bsmt&quot;)) %&gt;% dplyr::filter_all(any_vars(is.na(.)))</code></pre> <pre><code># A tibble: 9 × 13 Id Neighborhood BsmtQual BsmtCond BsmtExposure BsmtFinType1 BsmtFinSF1 BsmtFinType2 BsmtFinSF2 BsmtUnfSF TotalBsmtSF BsmtFullBath BsmtHalfBath &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt; 1 333 NridgHt Gd TA No GLQ 1124 &lt;NA&gt; 479 1603 3206 1 0 2 949 CollgCr Gd TA &lt;NA&gt; Unf 0 Unf 0 936 936 0 0 3 1488 Somerst Gd TA &lt;NA&gt; Unf 0 Unf 0 1595 1595 0 0 4 2041 Veenker Gd &lt;NA&gt; Mn GLQ 1044 Rec 382 0 1426 1 0 5 2186 Edwards TA &lt;NA&gt; No BLQ 1033 Unf 0 94 1127 0 1 6 2218 IDOTRR &lt;NA&gt; Fa No Unf 0 Unf 0 173 173 0 0 7 2219 IDOTRR &lt;NA&gt; TA No Unf 0 Unf 0 356 356 0 0 8 2349 Somerst Gd TA &lt;NA&gt; Unf 0 Unf 0 725 725 0 0 9 2525 CollgCr TA &lt;NA&gt; Av ALQ 755 Unf 0 240 995 0 0</code></pre> <pre class="r"><code># determine most common values by Neighborhood dfNeighborhood_mcv &lt;- most_common_value_by_grouping(dfClean %&gt;% dplyr::filter(TotalBsmtSF &gt; 0), &quot;Neighborhood&quot;, c(&quot;BsmtQual&quot;,&quot;BsmtCond&quot;,&quot;BsmtExposure&quot;)) bsmt_fintype2_mcv &lt;- dfClean %&gt;% dplyr::filter(BsmtFinSF2 &gt; 0) %&gt;% pull(BsmtFinType2) %&gt;% find_most_common_value() dfClean &lt;- dfClean %&gt;% mutate(BsmtQual = if_else(is.na(BsmtQual), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, BsmtQual[Neighborhood == .x])), BsmtQual)) %&gt;% mutate(BsmtCond = if_else(is.na(BsmtCond), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, BsmtCond[Neighborhood == .x])), BsmtCond)) %&gt;% mutate(BsmtExposure = if_else(is.na(BsmtExposure), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, BsmtExposure[Neighborhood == .x])), BsmtExposure)) %&gt;% mutate(BsmtFinType2 = if_else(is.na(BsmtFinType2), bsmt_fintype2_mcv, BsmtFinType2)) # cleanup rm(vFeaturesChr_bsmt, vFeaturesNum_bsmt) rm(dfNeighborhood_mcv, bsmt_fintype2_mcv)</code></pre> <p>Below are histograms of the numerical basement features, and their correlations with <code>SalePrice</code>. A histogram shows how many properties have values in each range. A correlation is a number from -1 to 1. It tells how closely two features move together in a straight line: near 1, they go up together; near -1, one goes down when the other goes up; near 0, they have no straight-line relationship.</p> <p>We can check that: <code>TotalBsmtSF</code> = <code>BsmtFinSF1</code> + <code>BsmtFinSF2</code> + <code>BsmtUnfSF</code></p> <p>We also create new features:</p> <ul> <li><code>BsmtBath</code> = <code>BsmtFullBath</code> + 0.5 * <code>BsmtHalfBath</code></li> <li><code>HasBsmt</code> = <code>TotalBsmtSF</code> &gt; 0</li> </ul> <pre class="r"><code># New features: BsmtBath, HasBsmt dfClean &lt;- dfClean %&gt;% mutate(BsmtBath = BsmtFullBath + 0.5*BsmtHalfBath) %&gt;% mutate(HasBsmt = (TotalBsmtSF &gt; 0), HasBsmt = as.character(HasBsmt)) dfBsmt &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Bsmt&quot;)) vFeaturesNum &lt;- dfBsmt %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfBsmt, vFeaturesNum, nCols = 2, plotTitle = &quot;Basement Features&quot;) p2 &lt;- eda_plot_correlation_matrix(dfBsmt[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) cowplot::plot_grid(plotlist = list(p1, p2), ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_num_bsmt-1.png" alt="" width="960" /></p> <ul> <li>Most properties have a basement.</li> <li>The column plots show that one category is much more common than the others in <code>BsmtFinType2</code> and in <code>BsmtCond</code>.</li> </ul> <pre class="r"><code>vFeaturesChr &lt;- dfBsmt %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfBsmt, .x)))</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_chr_bsmt-1.png" alt="" width="672" /></p> <pre class="r"><code># cleanup rm(dfBsmt, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="bathroom-features" class="section level4"> <h4>Bathroom Features</h4> <ul> <li>We create a new feature for the total number of bathrooms: <code>TotalBath</code> = <code>FullBath</code> + 0.5 * <code>HalfBath</code> + <code>BsmtBath</code></li> </ul> <pre class="r"><code># New features: HasFireplace dfClean &lt;- dfClean %&gt;% mutate(TotalBath = FullBath + 0.5*HalfBath + BsmtBath) dfBath &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Bath&quot;)) vFeaturesNum &lt;- dfBath %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfBath, vFeaturesNum, plotTitle = &quot;Bathroom Features&quot;) p2 &lt;- eda_plot_correlation_matrix(dfBath[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) cowplot::plot_grid(plotlist = list(p1, p2), ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_bath-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfBath, vFeaturesNum)</code></pre> </div> <div id="fireplace-features" class="section level4"> <h4>Fireplace Features</h4> <ol style="list-style-type: decimal"> <li>1420 properties have no fireplace. For these properties, we change <code>FireplaceQu</code> (the fireplace quality) to <strong>None</strong>.</li> </ol> <pre class="r"><code># 1. Change FireplaceQu to None if there is no fireplace ############################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Fire&quot;)) %&gt;% dplyr::filter(Fireplaces == 0)</code></pre> <pre><code># A tibble: 1,420 × 4 Id Neighborhood Fireplaces FireplaceQu &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; 1 1 CollgCr 0 &lt;NA&gt; 2 6 Mitchel 0 &lt;NA&gt; 3 11 Sawyer 0 &lt;NA&gt; 4 13 Sawyer 0 &lt;NA&gt; 5 16 BrkSide 0 &lt;NA&gt; 6 18 Sawyer 0 &lt;NA&gt; 7 19 SawyerW 0 &lt;NA&gt; 8 20 NAmes 0 &lt;NA&gt; 9 27 NAmes 0 &lt;NA&gt; 10 30 BrkSide 0 &lt;NA&gt; # ℹ 1,410 more rows</code></pre> <pre class="r"><code>dfClean &lt;- dfClean %&gt;% mutate_at(vars(&quot;FireplaceQu&quot;), ~ case_when(Fireplaces == 0 ~ &quot;None&quot;, TRUE ~ .))</code></pre> <ul> <li>We create a new feature: <code>HasFireplace</code> = <code>Fireplaces</code> &gt; 0</li> <li>Many properties have a fireplace.</li> </ul> <pre class="r"><code># New features: HasFireplace dfClean &lt;- dfClean %&gt;% mutate(HasFireplace = (Fireplaces &gt; 0), HasFireplace = as.character(HasFireplace)) dfFireplace &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Fireplace&quot;)) vFeaturesNum &lt;- dfFireplace %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() vFeaturesChr &lt;- dfFireplace %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfFireplace, vFeaturesNum, plotTitle = &quot;Fireplace Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfFireplace, .x))) p3 &lt;- eda_plot_correlation_matrix(dfFireplace[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_fireplace-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfFireplace, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="garage-features" class="section level4"> <h4>Garage Features</h4> <ol style="list-style-type: decimal"> <li><p>If <code>GarageYrBlt</code> is missing, we set it to <code>YearBuilt</code>.</p></li> <li><p>157 properties have no garage. For these properties, <code>GarageType</code>, <code>GarageFinish</code>, <code>GarageQual</code> and <code>GarageCond</code> are <strong>None</strong>.</p></li> <li><p>Garages are different in different neighborhoods. So we fill in each remaining missing value with the most common value (or the median value, for numbers) of the same <code>Neighborhood</code> and <code>GarageType</code>.</p></li> </ol> <pre class="r"><code># 1. Change GarageYrBlt to YearBuilt if it is NA ############################################################################################################ dfClean &lt;- dfClean %&gt;% mutate(GarageYrBlt = if_else(is.na(GarageYrBlt), YearBuilt, GarageYrBlt)) # 2. Change NA values to None if there is no garage ############################################################################################################ dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Garage&quot;)) %&gt;% dplyr::filter(GarageArea == 0)</code></pre> <pre><code># A tibble: 157 × 9 Id Neighborhood GarageType GarageYrBlt GarageFinish GarageCars GarageArea GarageQual GarageCond &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; 1 40 Edwards &lt;NA&gt; 1955 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 2 49 OldTown &lt;NA&gt; 1920 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 3 79 Sawyer &lt;NA&gt; 1968 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 4 89 IDOTRR &lt;NA&gt; 1915 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 5 90 CollgCr &lt;NA&gt; 1994 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 6 100 NAmes &lt;NA&gt; 1959 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 7 109 IDOTRR &lt;NA&gt; 1919 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 8 126 IDOTRR &lt;NA&gt; 1935 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 9 128 OldTown &lt;NA&gt; 1930 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; 10 141 NAmes &lt;NA&gt; 1971 &lt;NA&gt; 0 0 &lt;NA&gt; &lt;NA&gt; # ℹ 147 more rows</code></pre> <pre class="r"><code>vFeaturesChr_garage &lt;- dfClean %&gt;% select(contains(&quot;Garage&quot;)) %&gt;% select_if(is.character) %&gt;% colnames() dfClean &lt;- dfClean %&gt;% mutate_at(all_of(vFeaturesChr_garage), ~ case_when(GarageArea == 0 ~ &quot;None&quot;, TRUE ~ .)) # 3. Change NA values in remaining features to the most common or median value in the Neighborhood for that GarageType ###################################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Garage&quot;)) %&gt;% dplyr::filter_all(any_vars(is.na(.)))</code></pre> <pre><code># A tibble: 2 × 9 Id Neighborhood GarageType GarageYrBlt GarageFinish GarageCars GarageArea GarageQual GarageCond &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; 1 2127 OldTown Detchd 1910 &lt;NA&gt; 1 360 &lt;NA&gt; &lt;NA&gt; 2 2577 IDOTRR Detchd 1923 &lt;NA&gt; NA NA &lt;NA&gt; &lt;NA&gt; </code></pre> <pre class="r"><code>dfNeighborhood_mcv &lt;- most_common_value_by_grouping(dfClean, c(&quot;Neighborhood&quot;,&quot;GarageType&quot;), c(&quot;GarageQual&quot;,&quot;GarageFinish&quot;,&quot;GarageCond&quot;)) dfNeighborhood_med &lt;- median_by_grouping(dfClean, c(&quot;Neighborhood&quot;,&quot;GarageType&quot;), c(&quot;GarageCars&quot;,&quot;GarageArea&quot;)) %&gt;% ungroup() %&gt;% mutate_if(is.numeric, as.integer) dfClean &lt;- dfClean %&gt;% mutate(GarageFinish = if_else(is.na(GarageFinish), map2_chr(Neighborhood, GarageType, ~ with(dfNeighborhood_mcv, GarageFinish[Neighborhood==.x &amp; GarageType==.y])), GarageFinish)) %&gt;% mutate(GarageQual = if_else(is.na(GarageQual), map2_chr(Neighborhood, GarageType, ~ with(dfNeighborhood_mcv, GarageQual[Neighborhood==.x &amp; GarageType==.y])), GarageQual)) %&gt;% mutate(GarageCond = if_else(is.na(GarageCond), map2_chr(Neighborhood, GarageType, ~ with(dfNeighborhood_mcv, GarageCond[Neighborhood==.x &amp; GarageType==.y])), GarageCond)) %&gt;% mutate(GarageCars = if_else(is.na(GarageCars), map2_dbl(Neighborhood, GarageType, ~ with(dfNeighborhood_med, GarageCars[Neighborhood==.x &amp; GarageType==.y])), GarageCars)) %&gt;% mutate(GarageArea = if_else(is.na(GarageArea), map2_dbl(Neighborhood, GarageType, ~ with(dfNeighborhood_med, GarageArea[Neighborhood==.x &amp; GarageType==.y])), GarageArea)) # cleanup rm(vFeaturesChr_garage, dfNeighborhood_mcv, dfNeighborhood_med)</code></pre> <ul> <li><code>GarageArea</code> and <code>GarageCars</code> have almost the same correlation with <code>SalePrice</code>.</li> <li>We create a new feature: <code>HasGarage</code> = <code>GarageArea</code> &gt; 0</li> <li>Most properties have a garage.</li> <li>The column plots show that one category is much more common than the others in <code>GarageQual</code> and in <code>GarageCond</code>.</li> </ul> <pre class="r"><code># New features: HasGarage dfClean &lt;- dfClean %&gt;% mutate(HasGarage = (GarageArea &gt; 0), HasGarage = as.character(HasGarage)) dfGarage &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Garage&quot;)) vFeaturesNum &lt;- dfGarage %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() vFeaturesChr &lt;- dfGarage %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfGarage, vFeaturesNum, nCols = 2, plotTitle = &quot;Garage Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfGarage, .x))) p3 &lt;- eda_plot_correlation_matrix(dfGarage[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_garage-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfGarage, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="masonry-features" class="section level4"> <h4>Masonry Features</h4> <ol style="list-style-type: decimal"> <li><p>For one property (<code>Id</code> = 2611), <code>MasVnrArea</code> (the area of the masonry veneer) is 198, but <code>MasVnrType</code> (the type of masonry) is missing. We set <code>MasVnrType</code> to the most common type in the same neighborhood, among the properties that have masonry.</p></li> <li><p>We fill in the missing values of <code>MasVnrType</code> with the most common values for the same <code>Neighborhood</code> and <code>YearRemodAdd</code>.</p></li> <li><p>We fill in the missing values of <code>MasVnrArea</code> with the median values for the same <code>Neighborhood</code> and <code>MasVnrType</code>. The median is the middle value: half the values are below it and half are above it.</p></li> </ol> <pre class="r"><code># 1. Change MasVnrType in Id 2611 as the most common value in the Neighborhood ############################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Mas&quot;)) %&gt;% dplyr::filter(Id == 2611)</code></pre> <pre><code># A tibble: 1 × 4 Id Neighborhood MasVnrType MasVnrArea &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; 1 2611 Mitchel &lt;NA&gt; 198</code></pre> <pre class="r"><code>masvnrtype_mcv &lt;- dfClean %&gt;% dplyr::filter(MasVnrArea &gt; 0, Neighborhood == &quot;Mitchel&quot;) %&gt;% most_common_value_by_grouping(c(&quot;Neighborhood&quot;), &quot;MasVnrType&quot;) %&gt;% pull(MasVnrType) dfClean &lt;- dfClean %&gt;% mutate(MasVnrType = if_else(Id == 2611, masvnrtype_mcv, MasVnrType)) # 2. Change remaining NA values in MasVnrType to most common values by Neighborhood and YearRemodAdd ############################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Mas&quot;)) %&gt;% dplyr::filter_all(any_vars(is.na(.)))</code></pre> <pre><code># A tibble: 23 × 4 Id Neighborhood MasVnrType MasVnrArea &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; 1 235 Gilbert &lt;NA&gt; NA 2 530 Crawfor &lt;NA&gt; NA 3 651 Somerst &lt;NA&gt; NA 4 937 SawyerW &lt;NA&gt; NA 5 974 Somerst &lt;NA&gt; NA 6 978 Somerst &lt;NA&gt; NA 7 1244 NridgHt &lt;NA&gt; NA 8 1279 CollgCr &lt;NA&gt; NA 9 1692 Gilbert &lt;NA&gt; NA 10 1707 Somerst &lt;NA&gt; NA # ℹ 13 more rows</code></pre> <pre class="r"><code>dfNeighborhood_mcv &lt;- most_common_value_by_grouping(dfClean, c(&quot;Neighborhood&quot;,&quot;YearRemodAdd&quot;), &quot;MasVnrType&quot;) %&gt;% mutate(MasVnrType = if_else(is.na(MasVnrType), &quot;None&quot;, MasVnrType)) dfClean &lt;- dfClean %&gt;% mutate(MasVnrType = if_else(is.na(MasVnrType), map2_chr(Neighborhood, YearRemodAdd, ~ with(dfNeighborhood_mcv, MasVnrType[Neighborhood == .x &amp; YearRemodAdd == .y])), MasVnrType)) # 3. Change NA values in MasVnrArea to median values by Neighborhood and MasVnrType ############################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Mas&quot;)) %&gt;% dplyr::filter_all(any_vars(is.na(.)))</code></pre> <pre><code># A tibble: 23 × 4 Id Neighborhood MasVnrType MasVnrArea &lt;dbl&gt; &lt;chr&gt; &lt;chr&gt; &lt;dbl&gt; 1 235 Gilbert None NA 2 530 Crawfor None NA 3 651 Somerst None NA 4 937 SawyerW None NA 5 974 Somerst Stone NA 6 978 Somerst None NA 7 1244 NridgHt Stone NA 8 1279 CollgCr BrkFace NA 9 1692 Gilbert None NA 10 1707 Somerst Stone NA # ℹ 13 more rows</code></pre> <pre class="r"><code>dfNeighborhood_med &lt;- median_by_grouping(dfClean, c(&quot;Neighborhood&quot;,&quot;MasVnrType&quot;), &quot;MasVnrArea&quot;) dfClean &lt;- dfClean %&gt;% mutate(MasVnrArea = if_else(is.na(MasVnrArea), map2_dbl(Neighborhood, MasVnrType, ~ with(dfNeighborhood_med, MasVnrArea[Neighborhood == .x &amp; MasVnrType == .y])), MasVnrArea)) # cleanup rm(masvnrtype_mcv, dfNeighborhood_mcv, dfNeighborhood_med)</code></pre> <ul> <li>We create a new feature: <code>HasMasVnr</code> = <code>MasVnrArea</code> &gt; 0</li> <li>Many properties have masonry.</li> </ul> <pre class="r"><code># New features: HasMasVnr dfClean &lt;- dfClean %&gt;% mutate(HasMasVnr = (MasVnrArea &gt; 0), HasMasVnr = as.character(HasMasVnr)) dfMasVnr &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;MasVnr&quot;)) vFeaturesNum &lt;- dfMasVnr %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() vFeaturesChr &lt;- dfMasVnr %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfMasVnr, vFeaturesNum, plotTitle = &quot;Masonry Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfMasVnr, .x))) p3 &lt;- eda_plot_correlation_matrix(dfMasVnr[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_masvnr-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfMasVnr, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="pool-features" class="section level4"> <h4>Pool Features</h4> <ol style="list-style-type: decimal"> <li><p>If the property has no pool, we change <code>PoolQC</code> (the pool quality) to <strong>None</strong>.</p></li> <li><p>We fill in the remaining missing values of <code>PoolQC</code> with the most common value in the same neighborhood, among the properties that have a pool.</p></li> </ol> <pre class="r"><code># 1. Change NA values in PoolQC to None if there is no pool ############################################################################################################### dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Pool&quot;)) %&gt;% dplyr::filter(PoolArea == 0)</code></pre> <pre><code># A tibble: 2,906 × 4 Id Neighborhood PoolArea PoolQC &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; 1 1 CollgCr 0 &lt;NA&gt; 2 2 Veenker 0 &lt;NA&gt; 3 3 CollgCr 0 &lt;NA&gt; 4 4 Crawfor 0 &lt;NA&gt; 5 5 NoRidge 0 &lt;NA&gt; 6 6 Mitchel 0 &lt;NA&gt; 7 7 Somerst 0 &lt;NA&gt; 8 8 NWAmes 0 &lt;NA&gt; 9 9 OldTown 0 &lt;NA&gt; 10 10 BrkSide 0 &lt;NA&gt; # ℹ 2,896 more rows</code></pre> <pre class="r"><code>dfClean &lt;- dfClean %&gt;% mutate(PoolQC = if_else(PoolArea == 0, &quot;None&quot;, PoolQC)) # 2. Change NA values in remaining PoolQC to the most common value in the Neighborhood where PoolArea &gt; 0 ############################################################################################################### (dfMissing &lt;- dfClean %&gt;% select(Id, Neighborhood, contains(&quot;Pool&quot;)) %&gt;% dplyr::filter_all(any_vars(is.na(.))))</code></pre> <pre><code># A tibble: 3 × 4 Id Neighborhood PoolArea PoolQC &lt;dbl&gt; &lt;chr&gt; &lt;dbl&gt; &lt;chr&gt; 1 2421 NAmes 368 &lt;NA&gt; 2 2504 SawyerW 444 &lt;NA&gt; 3 2600 Mitchel 561 &lt;NA&gt; </code></pre> <pre class="r"><code>dfNeighborhood_mcv &lt;- most_common_value_by_grouping(dfClean %&gt;% dplyr::filter(PoolArea &gt; 0), &quot;Neighborhood&quot;, &quot;PoolQC&quot;) %&gt;% mutate(PoolQC = if_else(is.na(PoolQC), &quot;None&quot;, PoolQC)) dfMissing &lt;- dfMissing %&gt;% mutate(PoolQC = map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, PoolQC[Neighborhood == .x]))) dfClean$PoolQC[dfClean$Id %in% dfMissing$Id] &lt;- dfMissing$PoolQC # cleanup rm(dfNeighborhood_mcv, dfMissing)</code></pre> <ul> <li>We create a new feature: <code>HasPool</code> = <code>PoolArea</code> &gt; 0</li> <li>Most properties do not have a pool.</li> </ul> <pre class="r"><code># New features: HasGarage dfClean &lt;- dfClean %&gt;% mutate(HasPool = (PoolArea &gt; 0), HasPool = as.character(HasPool)) dfPool &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Pool&quot;)) vFeaturesNum &lt;- dfPool %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() vFeaturesChr &lt;- dfPool %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfPool, vFeaturesNum, plotTitle = &quot;Pool Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfPool, .x))) p3 &lt;- eda_plot_correlation_matrix(dfPool[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_pool-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfPool, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="porch-features" class="section level4"> <h4>Porch Features</h4> <ul> <li>We create new features for: <ul> <li>The total porch area: <code>PorchSF</code> = <code>OpenPorchSF</code> + <code>EnclosedPorch</code> + <code>3SsnPorch</code> + <code>ScreenPorch</code></li> <li>Whether the property has a porch: <code>HasPorch</code> = <code>PorchSF</code> &gt; 0</li> </ul></li> </ul> <pre class="r"><code># New features: PorchSF, HasPorch dfClean &lt;- dfClean %&gt;% mutate(PorchSF = OpenPorchSF + EnclosedPorch + `3SsnPorch` + ScreenPorch) %&gt;% mutate(HasPorch = PorchSF &gt; 0, HasPorch = as.character(HasPorch)) dfPorch &lt;- dfClean %&gt;% select(SalePrice, contains(&quot;Porch&quot;)) vFeaturesNum &lt;- dfPorch %&gt;% select(-SalePrice) %&gt;% select_if(is.numeric) %&gt;% colnames() vFeaturesChr &lt;- dfPorch %&gt;% select(-SalePrice) %&gt;% select_if(is.character) %&gt;% colnames() p1 &lt;- eda_plot_histograms(dfPorch, vFeaturesNum, plotTitle = &quot;Porch Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfPorch, .x))) p3 &lt;- eda_plot_correlation_matrix(dfPorch[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_porch-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(dfPorch, vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="built-area-features" class="section level4"> <h4>Built Area Features</h4> <ul> <li>We create a new feature for the total built area in square feet: <code>TotalSF</code> = <code>GrLivArea</code> + <code>TotalBsmtSF</code></li> </ul> <pre class="r"><code>dfClean &lt;- dfClean %&gt;% mutate(TotalSF = GrLivArea + TotalBsmtSF) vFeaturesNum &lt;- c(&quot;TotalSF&quot;,&quot;GrLivArea&quot;,&quot;TotalBsmtSF&quot;) p1 &lt;- eda_plot_histograms(dfClean, vFeaturesNum, plotTitle = &quot;Built Area Features&quot;) p2 &lt;- eda_plot_correlation_matrix(dfClean[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) cowplot::plot_grid(plotlist = list(p1, p2), ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_area-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(vFeaturesNum)</code></pre> </div> <div id="construction-year-features" class="section level4"> <h4>Construction Year Features</h4> <ul> <li>We create new features for: <ul> <li>The period when the property was built (<code>Vintage</code>): <strong>1945 or earlier, 1946-1999, 2000 or later</strong></li> <li>The age of the property when it was sold, counted from its last remodel (or from its construction, if it was never remodeled): <code>PropertyAge</code> = <code>YrSold</code> - <code>YearRemodAdd</code></li> <li>Whether the property was new or newly remodeled when it was sold: <code>IsNew</code> = <code>YearRemodAdd</code> == <code>YrSold</code></li> <li>Whether the property was ever remodeled: <code>IsRemod</code> = <code>YearRemodAdd</code> &gt; <code>YearBuilt</code></li> </ul></li> </ul> <pre class="r"><code># New feature: Vintage dfClean &lt;- dfClean %&gt;% mutate(Vintage = cut(YearBuilt, breaks = c(1800, 1945, 1999, 2100), labels = c(&quot;1945 or ealier&quot;,&quot;1946-1999&quot;,&quot;2000 or later&quot;), ordered_result = T)) %&gt;% mutate(PropertyAge = YrSold - YearRemodAdd) %&gt;% mutate(IsNew = (YearBuilt == YrSold), IsNew = as.character(IsNew)) %&gt;% mutate(IsRemod = (YearRemodAdd &gt; YearBuilt), IsRemod = as.character(IsRemod)) vFeaturesNum &lt;- dfClean %&gt;% select(contains(&quot;Year&quot;),contains(&quot;Yr&quot;),&quot;PropertyAge&quot;) %&gt;% colnames() %&gt;% sort() vFeaturesChr &lt;- c(&quot;IsNew&quot;,&quot;IsRemod&quot;,&quot;Vintage&quot;) p1 &lt;- eda_plot_histograms(dfClean, vFeaturesNum, nCols = 2, plotTitle = &quot;Construction Year Features&quot;) p2 &lt;- cowplot::plot_grid(plotlist = map(vFeaturesChr, ~ eda_plot_column(dfClean, .x))) p3 &lt;- eda_plot_correlation_matrix(dfClean[c(&quot;SalePrice&quot;,vFeaturesNum)], outlineVar = &quot;SalePrice&quot;, plotTitle = &quot;Correlation with SalePrice&quot;) grid.arrange(arrangeGrob(p1, p2, nrow = 2), p3, ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_construction_years-1.png" alt="" width="960" /></p> <pre class="r"><code># cleanup rm(vFeaturesNum, vFeaturesChr)</code></pre> </div> <div id="neighborhood-features" class="section level4"> <h4>Neighborhood Features</h4> <p>Type of Neighborhood: The dataset has 25 neighborhoods. In real estate, location is very important. Some neighborhoods clearly have higher prices than others.</p> <p>We can put the neighborhoods into fewer groups, by their median <code>SalePrice</code>:</p> <ul> <li><strong>Type1</strong>: StoneBr, NridgHt, NoRidge</li> <li><strong>Type2</strong>: Veenker, Timber, Somerst</li> <li><strong>Type3</strong>: Crawfor, CollgCr, ClearCr, Blmngtn, Gilbert, NWAmes, SawyerW</li> <li><strong>Type4</strong>: Mitchel, NPkVill, NAmes, SWISU, Sawyer, Blueste, BrkSide, Edwards, OldTown</li> <li><strong>Type5</strong>: IDOTRR, BrDale, MeadowV</li> </ul> <pre class="r"><code>dfClean &lt;- dfClean %&gt;% # lNeighborhoodTypes is defined with the neighborhood map at the start of this post mutate(NeighborhoodType = fct_collapse(Neighborhood, !!!lNeighborhoodTypes)) %&gt;% mutate(NeighborhoodType = factor(NeighborhoodType, levels = c(&quot;Type5&quot;,&quot;Type4&quot;,&quot;Type3&quot;,&quot;Type2&quot;,&quot;Type1&quot;), ordered = TRUE)) p1 &lt;- eda_plot_box(dfClean, &quot;Neighborhood&quot;, &quot;SalePrice&quot;, yLabelsScale = scales::label_dollar(accuracy = 1)) p2 &lt;- eda_plot_box(dfClean, &quot;NeighborhoodType&quot;, &quot;SalePrice&quot;, yLabelsScale = scales::label_dollar(accuracy = 1)) cowplot::plot_grid(plotlist = list(p1, p2), ncol = 2)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_neighborhood-1.png" alt="" width="960" /></p> </div> <div id="other-missing-features" class="section level4"> <h4>Other Missing Features</h4> <ol style="list-style-type: decimal"> <li><p>In <code>MiscFeature</code>, <code>Alley</code> and <code>Fence</code>, we change the missing values to <strong>None</strong>.</p></li> <li><p>In <code>Utilities</code>, <code>Functional</code> and <code>SaleType</code>, we fill in the missing values with the most common value of each feature.</p></li> <li><p>In <code>LotFrontage</code>, we fill in the missing values with the median value of the same <code>Neighborhood</code>.</p></li> <li><p>In <code>MSZoning</code>, <code>KitchenQual</code>, <code>Exterior1st</code>, <code>Exterior2nd</code> and <code>Electrical</code>, we fill in the missing values with the most common value of the same <code>Neighborhood</code>.</p></li> </ol> <pre class="r"><code># 1. &amp; 2. change NA values to None or most common value of that feature #################################################################################################################### dfClean &lt;- dfClean %&gt;% mutate_at(c(&quot;MiscFeature&quot;, &quot;Alley&quot;, &quot;Fence&quot;), change_na_to, value = &quot;None&quot;) %&gt;% mutate_at(c(&quot;Utilities&quot;), change_na_to_most_common_value) %&gt;% mutate_at(c(&quot;Functional&quot;), change_na_to_most_common_value) %&gt;% mutate_at(c(&quot;SaleType&quot;), change_na_to_most_common_value) # 3. In LotFrontage, change NA values to the median value by neighborhood #################################################################################################################### dfNeighborhood_med &lt;- median_by_grouping(dfClean, &quot;Neighborhood&quot;, &quot;LotFrontage&quot;) dfClean &lt;- dfClean %&gt;% mutate(LotFrontage = if_else(is.na(LotFrontage), map_dbl(Neighborhood, ~ with(dfNeighborhood_med, LotFrontage[Neighborhood == .x])), LotFrontage)) # 4. In MSZoning,KitchenQual,Exterior1st,Exterior2nd,Electrical change NA values to the median value by neighborhood dfNeighborhood_mcv &lt;- most_common_value_by_grouping(dfClean, &quot;Neighborhood&quot;, c(&quot;MSZoning&quot;,&quot;KitchenQual&quot;,&quot;Exterior1st&quot;,&quot;Exterior2nd&quot;,&quot;Electrical&quot;)) dfClean &lt;- dfClean %&gt;% mutate(MSZoning = if_else(is.na(MSZoning), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, MSZoning[Neighborhood == .x])), MSZoning)) %&gt;% mutate(KitchenQual = if_else(is.na(KitchenQual), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, KitchenQual[Neighborhood == .x])), KitchenQual)) %&gt;% mutate(Exterior1st = if_else(is.na(Exterior1st), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, Exterior1st[Neighborhood == .x])), Exterior1st)) %&gt;% mutate(Exterior2nd = if_else(is.na(Exterior2nd), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, Exterior2nd[Neighborhood == .x])), Exterior2nd)) %&gt;% mutate(Electrical = if_else(is.na(Electrical), map_chr(Neighborhood, ~ with(dfNeighborhood_mcv, Electrical[Neighborhood == .x])), Electrical)) # cleanup rm(dfNeighborhood_med, dfNeighborhood_mcv)</code></pre> </div> </div> <div id="label-encoding" class="section level3"> <h3>Label Encoding</h3> <p>The data description shows that the categories of many features have a natural order. These features are called ordinal features:</p> <ul> <li><code>LotShape</code>: Reg, IR1, IR2, IR3</li> <li><code>LandSlope</code>: Gtl, Mod, Sev</li> <li><code>ExterQual</code>: Ex, Gd, TA, Fa, Po</li> <li><code>ExterCond</code>: Ex, Gd, TA, Fa, Po</li> <li><code>BsmtQual</code>: Ex, Gd, TA, Fa, Po, None</li> <li><code>BsmtCond</code>: Ex, Gd, TA, Fa, Po, None</li> <li><code>BsmtExposure</code>: Gd, Av, Mn, No, None</li> <li><code>BsmtFinType1</code>: GLQ, ALQ, BLQ, Rec, LwQ, Unf, None</li> <li><code>BsmtFinType2</code>: GLQ, ALQ, BLQ, Rec, LwQ, Unf, None</li> <li><code>HeatingQC</code>: Ex, Gd, TA, Fa, Po</li> <li><code>CentralAir</code>: Y, N</li> <li><code>KitchenQual</code>: Ex, Gd, TA, Fa, Po</li> <li><code>Functional</code>: Typ, Min1, Min2, Mod, Maj1, Maj2, Sev, Sal</li> <li><code>FireplaceQu</code>: Ex, Gd, TA, Fa, Po, None</li> <li><code>GarageFinish</code>: Fin, RFn, Unf, None</li> <li><code>GarageQual</code>: Ex, Gd, TA, Fa, Po, None</li> <li><code>GarageCond</code>: Ex, Gd, TA, Fa, Po, None</li> <li><code>Street</code>: Grvl, Pave</li> <li><code>PavedDrive</code>: Y, P, N</li> </ul> <p>Most of these features use the same order: <strong>Ex</strong> (excellent), <strong>Gd</strong> (good), <strong>TA</strong> (typical/average), <strong>Fa</strong> (fair), <strong>Po</strong> (poor). Some of them also have the category <strong>None</strong>. So we can give these features one common set of ordered categories: <strong>Ex, Gd, TA, Fa, Po, None</strong>.</p> <p>Two features already use numbers for their order:</p> <ul> <li><code>OverallQual</code>: 10 to 1</li> <li><code>OverallCond</code>: 10 to 1</li> </ul> <p><code>MoSold</code> (the month of the sale) repeats in a cycle every year, so we treat it as a category, not as a number.</p> <p><code>YrSold</code> has only 5 values, from 2006 to 2010, so we also treat it as a category.</p> <p>In categorical features where several categories have fewer than 10 properties, we put these rare categories together in one category named <strong>Other</strong>.</p> <pre class="r"><code>is.nominal &lt;- function(Z) { return(is.factor(Z) &amp; !is.ordered(Z)) } dfClean &lt;- dfClean %&gt;% mutate(MoSold = as.factor(MoSold)) %&gt;% mutate(YrSold = as.factor(YrSold)) %&gt;% mutate(MSSubClass = as.character(MSSubClass)) %&gt;% mutate_at(c(&quot;ExterCond&quot;,&quot;ExterQual&quot;,&quot;BsmtQual&quot;,&quot;BsmtCond&quot;,&quot;HeatingQC&quot;,&quot;KitchenQual&quot;,&quot;FireplaceQu&quot;,&quot;GarageQual&quot;,&quot;GarageCond&quot;), factor, levels = c(&quot;Ex&quot;,&quot;Gd&quot;,&quot;TA&quot;,&quot;Fa&quot;,&quot;Po&quot;,&quot;None&quot;), ordered = TRUE) %&gt;% mutate_at(c(&quot;BsmtFinType1&quot;,&quot;BsmtFinType2&quot;), factor, levels = c(&quot;GLQ&quot;, &quot;ALQ&quot;, &quot;BLQ&quot;, &quot;Rec&quot;, &quot;LwQ&quot;, &quot;Unf&quot;, &quot;None&quot;), ordered = TRUE) %&gt;% mutate(LotShape = factor(LotShape, levels = c(&quot;Reg&quot;,&quot;IR1&quot;,&quot;IR2&quot;,&quot;IR3&quot;), ordered = TRUE)) %&gt;% mutate(LandSlope = factor(LandSlope, levels = c(&quot;Gtl&quot;, &quot;Mod&quot;, &quot;Sev&quot;), ordered = TRUE)) %&gt;% mutate(BsmtExposure = factor(BsmtExposure, levels = c(&quot;Gd&quot;, &quot;Av&quot;, &quot;Mn&quot;, &quot;No&quot;, &quot;None&quot;), ordered = TRUE)) %&gt;% mutate(Functional = factor(Functional, levels = c(&quot;Typ&quot;, &quot;Min1&quot;, &quot;Min2&quot;, &quot;Mod&quot;, &quot;Maj1&quot;, &quot;Maj2&quot;, &quot;Sev&quot;, &quot;Sal&quot;), ordered = TRUE)) %&gt;% mutate(GarageFinish = factor(GarageFinish, levels = c(&quot;Fin&quot;, &quot;RFn&quot;, &quot;Unf&quot;, &quot;None&quot;), ordered = TRUE)) %&gt;% mutate(Street = factor(Street, levels = c(&quot;Pave&quot;,&quot;Grvl&quot;), ordered = TRUE)) %&gt;% mutate(PavedDrive = factor(PavedDrive, levels = c(&quot;Y&quot;,&quot;P&quot;,&quot;N&quot;), ordered = TRUE)) %&gt;% mutate(OverallCond = ordered(OverallCond, levels = 1:10), OverallQual = ordered(OverallQual, levels = 1:10)) %&gt;% mutate(CentralAir = ordered(CentralAir, levels = c(&quot;Y&quot;,&quot;N&quot;)), HasGarage = ordered(HasGarage, levels = c(&quot;TRUE&quot;, &quot;FALSE&quot;)), HasFireplace = ordered(HasFireplace, levels = c(&quot;TRUE&quot;, &quot;FALSE&quot;)), HasMasVnr = ordered(HasMasVnr, levels = c(&quot;TRUE&quot;, &quot;FALSE&quot;)), HasPorch = ordered(HasPorch, levels = c(&quot;TRUE&quot;, &quot;FALSE&quot;)), IsNew = ordered(IsNew, levels = c(&quot;TRUE&quot;,&quot;FALSE&quot;)), IsRemod = ordered(IsRemod, levels = c(&quot;TRUE&quot;,&quot;FALSE&quot;))) %&gt;% # mutate_if(is.ordered, fct_rev) %&gt;% # reverse the factor levels before assigning an integer order # mutate_if(is.ordered, function(Z) as.integer(Z) - 1) %&gt;% mutate_if(is.character, as.factor) %&gt;% mutate_if(is.nominal, ~ fct_lump_min(.x, min = 10)) %&gt;% mutate_if(is.logical, as.factor)</code></pre> </div> <div id="features-to-drop" class="section level3"> <h3>Features to Drop</h3> <div id="highly-correlated-features" class="section level4"> <h4>Highly Correlated Features</h4> <p>We drop some features from the rest of the analysis, because they are strongly correlated with other features, or because a similar new feature replaces them.</p> <pre class="r"><code>(vFeatures_to_exclude &lt;- c(&quot;BsmtFullBath&quot;,&quot;GarageCars&quot;,&quot;GarageYrBlt&quot;, &quot;GrLivArea&quot;,&quot;PoolArea&quot;, &quot;YearBuilt&quot;,&quot;YearRemodAdd&quot;,&quot;Neighborhood&quot;, &quot;OpenPorchSF&quot;,&quot;EnclosedPorch&quot;,&quot;3SsnPorch&quot;,&quot;ScreenPorch&quot;))</code></pre> <pre><code> [1] &quot;BsmtFullBath&quot; &quot;GarageCars&quot; &quot;GarageYrBlt&quot; &quot;GrLivArea&quot; &quot;PoolArea&quot; &quot;YearBuilt&quot; &quot;YearRemodAdd&quot; &quot;Neighborhood&quot; &quot;OpenPorchSF&quot; [10] &quot;EnclosedPorch&quot; &quot;3SsnPorch&quot; &quot;ScreenPorch&quot; </code></pre> <p>Let’s check these choices. For every pair of features, we measure their association, which tells how strongly the two features are related. We use a measure that suits each type of pair: the Pearson correlation for two numerical features, Cramer’s V for two categorical features, and the correlation ratio for one numerical and one categorical feature. All these measures are near 0 when there is no relationship. Features with many categories can seem related to everything by chance, so Cramer’s V and the correlation ratio include a correction (bias correction) for this.</p> <pre class="r"><code>mAssoc &lt;- association_matrix(dfClean %&gt;% select(-Id)) # one row per pair of features, strongest association first mUpper &lt;- which(upper.tri(mAssoc), arr.ind = T) dfAssocPairs &lt;- tibble(feature1 = rownames(mAssoc)[mUpper[, 1]], feature2 = colnames(mAssoc)[mUpper[, 2]], association = mAssoc[mUpper], measure = attr(mAssoc, &quot;measure&quot;)[mUpper]) %&gt;% drop_na(association) %&gt;% arrange(desc(abs(association))) assoc_of &lt;- function(f1, f2) { dfAssocPairs %&gt;% filter((feature1 == f1 &amp; feature2 == f2) | (feature1 == f2 &amp; feature2 == f1)) %&gt;% pull(association) %&gt;% round(2) } vFeaturesAssoc &lt;- dfAssocPairs %&gt;% filter(abs(association) &gt;= 0.85) %&gt;% select(feature1, feature2) %&gt;% unlist() %&gt;% unique() vPorchAssoc &lt;- map_dbl(c(&quot;OpenPorchSF&quot;,&quot;EnclosedPorch&quot;,&quot;3SsnPorch&quot;,&quot;ScreenPorch&quot;), ~ assoc_of(.x, &quot;PorchSF&quot;))</code></pre> <p>Here are the 36 features that have an association of 0.85 or more with at least one other feature. Related features are next to each other:</p> <pre class="r"><code>eda_plot_association_matrix(dfClean[vFeaturesAssoc], plotTitle = &quot;Features with an association of 0.85 or more with another feature&quot;)</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_association_matrix-1.png" alt="" width="960" /></p> <ul> <li>Most of the features that we drop have a strong association with another feature: <code>GarageCars</code> with <code>GarageArea</code> (0.89), <code>GrLivArea</code> with <code>TotalSF</code> (0.87), <code>BsmtFullBath</code> with <code>BsmtBath</code> (0.97), <code>PoolArea</code> with <code>HasPool</code> (0.94), <code>YearBuilt</code> with <code>Vintage</code> (0.9), <code>GarageYrBlt</code> with <code>YearBuilt</code> (0.86), <code>YearRemodAdd</code> with <code>PropertyAge</code> (-1), and <code>Neighborhood</code> with <code>NeighborhoodType</code> (1).</li> <li>The four porch features have only a medium association with <code>PorchSF</code> (between 0.19 and 0.61). So they do not repeat its information. We drop them because <code>PorchSF</code>, their total area, replaces them.</li> <li>The matrix also finds features that repeat information, which we have not dropped yet. <code>HasGarage</code>, <code>HasFireplace</code> and <code>HasMasVnr</code> give the same information as the <em>None</em> category of the garage, fireplace and masonry features (Cramer’s V of about 1). Also, <code>MSSubClass</code> has a strong association with <code>BldgType</code> (0.89) and with <code>HouseStyle</code> (0.83). Models that use decision trees are not affected by such repeated information. But in linear models, features that repeat each other (collinear features) make the results unreliable. So we will remove them before we build the models in <a href="https://www.nitingupta.com/casestudies/ames-housing-part3-models/">Part 3</a>.</li> </ul> </div> <div id="skewed-categorical-features" class="section level4"> <h4>Skewed Categorical Features</h4> <p>If more than 95% of the properties have the same category in a feature, the feature probably does not help to predict the price. An extreme example is <code>Utilities</code>, which has only 2 categories in the dataset: <strong>AllPub</strong> and <strong>NoSeWa</strong>. Only 1 property has <strong>NoSeWa</strong>, and all the others have <strong>AllPub</strong>. So features like this one do not help to predict the price.</p> <p>For each categorical feature, let’s look at the share of its most common category. Let’s also count its rare categories, which have fewer than 10 properties:</p> <pre class="r"><code>p_dominance &lt;- eda_plot_category_dominance(dfClean, threshold = 0.95, minCount = 10) p_dominance</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_category_dominance-1.png" alt="" width="960" /></p> <pre class="r"><code>dfDominance &lt;- p_dominance$data dfRare &lt;- dfDominance %&gt;% filter(share &lt; 0.95, n_rare &gt; 0) %&gt;% arrange(desc(n_rare), desc(share))</code></pre> <ul> <li>11 features are above the 95% line, so we drop them. The plots below show them.</li> <li>Of the other features, 13 have at least one rare category with fewer than 10 properties, such as Functional, GarageQual, Electrical. For a model, each category becomes a separate feature with the values 0 and 1 (a dummy variable). A dummy variable for a rare category has only a few properties, so the model cannot measure its effect reliably. One way to solve this problem is to put the rare categories together in one category.</li> </ul> <pre class="r"><code>vFeatures_skewed &lt;- dfClean %&gt;% select_if(is.factor) %&gt;% gather(feature, value) %&gt;% group_by(feature, value) %&gt;% count() %&gt;% mutate(perc = n/nrow(dfClean)) %&gt;% dplyr::filter(perc &gt; 0.95) %&gt;% pull(feature) cowplot::plot_grid(plotlist = map(vFeatures_skewed, ~ eda_plot_column(dfClean, .x)))</code></pre> <p><img src="https://www.nitingupta.com/casestudies/ames-housing-part1-data-pipeline/index_files/figure-html/plot_skewed_features-1.png" alt="" width="960" /></p> <pre class="r"><code># Features to exclude for modelling vFeatures_to_exclude &lt;- c(vFeatures_to_exclude, vFeatures_skewed) %&gt;% unique() # cleanup rm(vFeatures_skewed)</code></pre> </div> </div> </div> <div id="finalized-data" class="section level2"> <h2>Finalized Data</h2> <pre class="r"><code># print(&quot;Following features are dropped&quot;) # print(vFeatures_to_exclude) # Finalized dataset dfClean &lt;- dfClean %&gt;% select(-one_of(vFeatures_to_exclude)) # determine the numerical features vFeaturesNum &lt;- dfClean %&gt;% select(-Id) %&gt;% select_if(is.numeric) %&gt;% colnames() %&gt;% sort() # determine the categorical features vFeaturesCat &lt;- dfClean %&gt;% select(-Id) %&gt;% select_if(is.factor) %&gt;% colnames() %&gt;% sort() # vFeaturesCat1 &lt;- vFeaturesCat[map_int(vFeaturesCat, ~ length(levels(dfClean[[.x]]))) &gt; 5] vFeaturesCat1 &lt;- dfClean %&gt;% select_if(is.ordered) %&gt;% colnames() %&gt;% sort() vFeaturesCat2 &lt;- setdiff(vFeaturesCat, vFeaturesCat1) print(&quot;Dimensions of the finalized dataset&quot;)</code></pre> <pre><code>[1] &quot;Dimensions of the finalized dataset&quot;</code></pre> <pre class="r"><code>dim(dfClean)</code></pre> <pre><code>[1] 2919 73</code></pre> <pre class="r"><code>print(glue::glue(&quot;Excluding Id, there are {(ncol(dfClean)-1)} features in the finalized dataset.&quot;))</code></pre> <pre><code>Excluding Id, there are 72 features in the finalized dataset.</code></pre> <pre class="r"><code>print(glue::glue(&quot;There are {length(vFeaturesNum)} numerical, {length(vFeaturesCat1)} ordinal and {length(vFeaturesCat2)} nominal features.&quot;))</code></pre> <pre><code>There are 26 numerical, 26 ordinal and 20 nominal features.</code></pre> <p>The finalized dataset has 2919 properties and 73 columns: the <code>Id</code> and 72 features, including <code>SalePrice</code>. We save it, so that the next posts can use it:</p> <pre class="r"><code># cache = FALSE: always save the current data. Part 3 checks the contents of this file (not its date), so saving # the same data again does not make Part 3 tune its models again. save(dfClean, vTestIDs, vFeaturesNum, vFeaturesCat, vFeaturesCat1, vFeaturesCat2, vFeatures_to_exclude, file = &quot;Ames_Housing_Clean.Rda&quot;)</code></pre> </div> <div id="next-steps" class="section level2"> <h2>Next Steps</h2> <p>The data is now clean and ready. In <a href="https://www.nitingupta.com/casestudies/ames-housing-part2-eda/">Part 2</a>, we explore this dataset, one feature at a time, then two and three features at a time, to find which features affect the sale price.</p> </div>