final kfold added

main.py

@@ -30,10 +30,21 @@ max_params = combinations[0]
 max_avg_auc = 0
 max_avg_auc = 0
 
 
 for params in combinations:
 for params in combinations:
-    avg_acc, avg_auc = cnn.run_nn(input_, output_, n_experiences, params)
+    avg_acc, avg_auc, X_train_kfold, X_test_kfold, y_train_kfold, y_test_kfold = cnn.run_nn(input_, output_, n_experiences, params)
     if avg_auc > max_avg_auc:
     if avg_auc > max_avg_auc:
         max_avg_auc = avg_auc
         max_avg_auc = avg_auc
         max_params = params
         max_params = params
+        X_train_kfold_opt = X_train_kfold
+        X_test_kfold_opt = X_test_kfold
+        y_train_kfold_opt = y_train_kfold
+        y_test_kfold_opt = y_test_kfold
 
 
+# once we have chosen the optimal parameters we can do the normal kfold
 
 
-## to do: final kfold with optimal parameters, code to execute in paralel, make sure it's the same random shuffle ...
+#note: the test data remains unbalanced
+acc, auc = cnn.run_kfold(X_train_opt, X_test_kfold_opt, y_train_kfold_opt, y_test_kfold_opt, max_params)
+
+
+
+## TO DO: create confusion matrix, etc...
+## TO DO: code that allows to execute in parallel, make sure it's the same random shuffle ...