refactor: main.py order of magnitude of for loops - threshold<crs<ini…

…t_labelled<data_sets

main.py

@@ -16,29 +16,31 @@
 from src.ssl.self_flexcon import SelfFlexCon
 
 
-crs = [0.05]
-thresholds = [0.95]
+crs = [0.03, 0.05, 0.08, 0.1]
+thresholds = [0.98, 0.95, 0.93, 0.9]
 
 warnings.simplefilter("ignore")
 
 parser = argparse.ArgumentParser(description="Escolha um classificador para criar um cômite")
 parser.add_argument('classifier', metavar='c', type=int, help='Escolha um classificador para criar um cômite. Opções: 1 - Naive Bayes, 2 - Tree Decision, 3 - Knn, 4 - Heterogeneous')
 parent_dir = "path_for_results"
-datasets = ['Car.csv', 'Adult.csv']
-init_labelled = [0.05]
+datasets = ['Car.csv']
+init_labelled = [0.03, 0.05, 0.08, 0.1]
 
 args = parser.parse_args()
 
 # datasets = [f for f in listdir('datasets/') if isfile(join('datasets/', f))]
 # init_labelled = [0.05, 0.10, 0.15, 0.20, 0.25]
 
 for dataset in datasets:
-    for cr in crs:
-        for threshold in thresholds:
-            comite = Ensemble(SelfFlexCon, cr=cr, threshold=threshold)
-            path = os.path.join(parent_dir, dataset)
-            os.makedirs(path, exist_ok=True)
-            for labelled_level in init_labelled:
+    for labelled_level in init_labelled:
+        for cr in crs:
+            for threshold in thresholds:
+                comite = Ensemble(SelfFlexCon, cr=cr, threshold=threshold)
+                path = os.path.join(parent_dir, dataset)
+                os.makedirs(path, exist_ok=True)
+
+
                 fold_result = []
                 df = pd.read_csv('datasets/'+dataset, header=0)
                 seed(214)