new

frontend/src/components/DataSpace.vue

@@ -3,7 +3,7 @@
   	<div id="BeeSwarm"></div>
     <div id="Sliders"></div>
     <div id="TextLabels"></div>
-    <div id="NoAction" style="width: 2525px !important; height: 225px !important;"></div>
+    <div id="NoAction" style="width: 2525px !important; height: 250px !important;"></div>
 </div> 
 </template>
 
@@ -32,7 +32,7 @@ export default {
       var tooltip = d3.select("#tooltip")
 
       var width = 2525
-      var height = 225
+      var height = 250
       var rectWidth = 20
       var rectHeight = 10
       
@@ -226,7 +226,7 @@ export default {
       svg.selectAll("*").remove();
 
       var width = 2525
-      var height = 225
+      var height = 250
 
       var xScaleOp = d3.scale.linear()
         .domain([0, width-50])
@@ -388,7 +388,7 @@ export default {
       svg.selectAll("*").remove();
 
       var width = 2525
-      var height = 225
+      var height = 250
 
       var svgLines = d3.select('#TextLabels').append('svg');
       svgLines.attr('width', width);

frontend/src/components/FeatureSpaceDetail.vue

@@ -833,7 +833,7 @@ export default {
         var MIVar = JSON.parse(this.dataFS[28+this.quadrantNumber])
         MIVar = MIVar.concat(this.MIRemaining)
         //var colorCateg = d3.scaleOrdinal(d3.schemeDark2)
-        var colorCateg = d3.scaleOrdinal().domain([0, 1, 2, 4]).range(['#808000','#7570b3','#1b9e77','#d95f02'])
+        var colorCateg = d3.scaleOrdinal().domain([0, 1, 2, 4]).range(['#808000','#7570b3','#d95f02','#1b9e77'])
 
         var corrTargetFormatted = []
         for (let i = 0; i < Object.keys(corrTarget).length; i++) {

frontend/src/components/FeatureSpaceOverview.vue

@@ -1186,7 +1186,7 @@ export default {
         var legendRectSize = 14;                                
         var legendSpacing = 3;                                    
         var labelsData = JSON.parse(this.overallData[1])
-        var color = d3v5.scaleOrdinal().domain(labelsData).range(['#808000','#7570b3','#1b9e77','#d95f02'])
+        var color = d3v5.scaleOrdinal().domain(labelsData).range(['#808000','#7570b3','#d95f02','#1b9e77'])
 
         var svgLegend = d3v5.select('#legendTarget').append('svg')                    
             .attr('width', 130)

frontend/src/components/Main.vue

@@ -8,7 +8,7 @@
           <mdb-card>
             <mdb-card-header color="primary-color" tag="h5" class="text-center"><DataSetSlider/></mdb-card-header>
             <mdb-card-body>
-              <mdb-card-text class="text-left" style="font-size: 18.5px; min-height: 220px">
+              <mdb-card-text class="text-left" style="font-size: 18.5px; min-height: 248px">
                 <DataSpace/>
               </mdb-card-text>
             </mdb-card-body>
@@ -149,7 +149,7 @@ export default Vue.extend({
       DataResults: '',
       keyNow: 1,
       instancesImportance: '',
-      RetrieveValueFile: 'IrisC', // this is for the default data set
+      RetrieveValueFile: 'HeartC', // this is for the default data set
       SelectedFeaturesPerClassifier: '',
       FinalResults: 0,
       selectedAlgorithm: '',
@@ -862,7 +862,7 @@ body {
   top: 0px;
   bottom: 0px;
   margin-top: -4px !important;
-  //overflow: hidden !important; # remove scrolling
+  overflow: hidden !important; // remove scrolling
 }
 
 .modal-backdrop {

run.py

@@ -502,11 +502,11 @@ def dataSetSelection():
 
 def create_global_function():
     global estimator
-    def estimator(n_estimators, eta, max_depth):
+    def estimator(n_estimators, eta, max_depth, subsample, colsample_bytree):
         # initialize model
         n_estimators = int(n_estimators)
         max_depth = int(max_depth)
-        model = XGBClassifier(n_estimators=n_estimators, eta=eta, max_depth=max_depth, n_jobs=-1, random_state=RANDOM_SEED, silent=True, verbosity = 0, use_label_encoder=False)
+        model = XGBClassifier(n_estimators=n_estimators, eta=eta, max_depth=max_depth, subsample=subsample, colsample_bytree=colsample_bytree, n_jobs=-1, random_state=RANDOM_SEED, silent=True, verbosity = 0, use_label_encoder=False)
         # set in cross-validation
         result = cross_validate(model, XData, yData, cv=crossValidation, scoring='accuracy')
         # result is mean of test_score
@@ -568,11 +568,11 @@ def executeModel(exeCall, flagEx, nodeTransfName):
     # Bayesian Optimization CHANGE INIT_POINTS!
     if (keyFirstTime):
         create_global_function()
-        params = {"n_estimators": (5, 200), "eta": (0.05, 0.3), "max_depth": (6,12)}
+        params = {"n_estimators": (5, 200), "eta": (0.05, 0.3), "max_depth": (6,12), "subsample": (0.8,1), "colsample_bytree": (0.8,1)}
         bayesopt = BayesianOptimization(estimator, params, random_state=RANDOM_SEED)
-        bayesopt.maximize(init_points=10, n_iter=5, acq='ucb') # 35 and 15
+        bayesopt.maximize(init_points=35, n_iter=15, acq='ucb') # 35 and 15
         bestParams = bayesopt.max['params']
-        estimator = XGBClassifier(n_estimators=int(bestParams.get('n_estimators')), eta=bestParams.get('eta'), max_depth=int(bestParams.get('max_depth')), probability=True, random_state=RANDOM_SEED, silent=True, verbosity = 0, use_label_encoder=False)
+        estimator = XGBClassifier(n_estimators=int(bestParams.get('n_estimators')), eta=bestParams.get('eta'), max_depth=int(bestParams.get('max_depth')), subsample=bestParams.get('subsample'), colsample_bytree=bestParams.get('colsample_bytree'), probability=True, random_state=RANDOM_SEED, silent=True, verbosity = 0, use_label_encoder=False)
         columnsNewGen = OrignList
 
     if (len(exeCall) != 0):