Ändern der Reihenfolge recall_0 und recall_1 zu recall_not_oos und recall_oos...

Ändern der Reihenfolge recall_0 und recall_1 zu recall_not_oos und recall_oos / Logik für augmentierte Daten angepasst

main.py

@@ -90,8 +90,8 @@ def run_model(config, config_knn, worksheet, r):
     training_duration_model_str = format(training_duration_model)
     print('Die Dauer für das Erstellen des Modells beträgt: ' + training_duration_model_str)
     # learning curves
-    print("Erstelle den Plot Graphen")
-    plot_values(history, config)
+    # print("Erstelle den Plot Graphen")
+    # plot_values(history, config)
     print("Inhalte in die Excel-Tabelle schreiben")
     worksheet, r = create_excel_result(worksheet, r, callback, config, config_knn, evaluate_metrics,
                                        training_duration_model)

ressources/config/config.yaml

 bilder:
   # original_path: "assets/Bilder/Datengrundlage-Reduziert-Test/"
-  original_path: "assets/Bilder/Datengrundlage/"
+  # original_path: "assets/Bilder/Datengrundlage/"
+  original_path: "assets/Bilder/Datengrundlage-Augmentiert"
   knn_path: "assets/Bilder/AktuelleTrainingsUndTestdaten/"
 knn:
-  epochs: 10
+  epochs: 3
   exception_activation_funktion_1:
     ['ReLU']
 result:

src/data/copyData.py

@@ -44,8 +44,8 @@ def copy_images_with_exclusion(src_dir, dest_dir, exclude_dir=None):
             os.remove(file_or_dir_path)
 
     # Erstelle die gewünschte Ordnerstruktur im Zielverzeichnis
-    os.makedirs(os.path.join(dest_dir, '0_OOS'))
-    os.makedirs(os.path.join(dest_dir, '1_!OOS'))
+    os.makedirs(os.path.join(dest_dir, '!OOS'))
+    os.makedirs(os.path.join(dest_dir, 'OOS'))
 
 
     # Durchlaufe die Unterordner im Quellverzeichnis
@@ -58,7 +58,7 @@ def copy_images_with_exclusion(src_dir, dest_dir, exclude_dir=None):
         for folder in os.listdir(os.path.join(src_dir, subdir)):
             if folder == 'OOS':
                 for img in os.listdir(os.path.join(src_dir, subdir, folder)):
-                    shutil.copy(os.path.join(src_dir, subdir, folder, img), os.path.join(dest_dir, '0_OOS'))
+                    shutil.copy(os.path.join(src_dir, subdir, folder, img), os.path.join(dest_dir, 'OOS'))
             elif folder == '!OOS':
                 for img in os.listdir(os.path.join(src_dir, subdir, folder)):
-                    shutil.copy(os.path.join(src_dir, subdir, folder, img), os.path.join(dest_dir, '1_!OOS'))
+                    shutil.copy(os.path.join(src_dir, subdir, folder, img), os.path.join(dest_dir, '!OOS'))

src/data/pictureHandling.py

 import os
 import shutil
 import tensorflow as tf
+from src.enum.marktEnum import Markt
 
 
 def createPathWithPictures():
@@ -26,27 +27,28 @@ def createPathWithPictures():
 
 
 def generateAugmentedNotOOSPictures():
-    # Pfad zum Ordner "!OOS"
-    folder_path = "/assets/Bilder/Modelldaten/only!OOS"
+    for markt in Markt:
+        # Pfad zum Ordner "!OOS"
+        folder_path = "assets/Bilder/Datengrundlage_Test/" + markt.name + "/"
 
-    # Erstellen Sie eine Instanz des ImageDataGenerator
-    datagen = get_images()
+        # Erstellen Sie eine Instanz des ImageDataGenerator
+        datagen = get_images()
 
-    # Pfad für den neuen Ordner, in dem die augmentierten Bilder gespeichert werden sollen
-    save_to_dir = "/assets/Bilder/Modelldaten/alleMaerkteAugmented/!OOS"
+        # Pfad für den neuen Ordner, in dem die augmentierten Bilder gespeichert werden sollen
+        save_to_dir = "assets/Bilder/Datengrundlage-Augmentiert/" + markt.name + "/!OOS"
 
-    for i in range(9):
-        # Trainingsdaten konfigurieren
-        train_it = datagen.flow_from_directory(
-            folder_path,
-            class_mode='binary',
-            batch_size=1425,
-            target_size=(224, 224),
-            save_to_dir=save_to_dir,
-            save_prefix="augmented_" + str(i) + "_",
-            save_format='jpg',
-        )
-        train_it.next()
+        for i in range(9):
+            # Trainingsdaten konfigurieren
+            train_it = datagen.flow_from_directory(
+                folder_path,
+                class_mode='binary',
+                batch_size=1425,
+                target_size=(224, 224),
+                save_to_dir=save_to_dir,
+                save_prefix="augmented_" + str(i) + "_",
+                save_format='jpg',
+            )
+            train_it.next()
 
 
 def get_images():
@@ -65,4 +67,4 @@ def get_images():
     return augmentedImageDefinition
 
 
-generateAugmentedNotOOSPictures()
+# generateAugmentedNotOOSPictures()

src/knn/createKNN.py

@@ -19,7 +19,7 @@ def evaluate_model(model, test_it, datagen, config, config_knn):
     if config_knn.excluded_folder != Markt.Kein_Markt:
         test_it = get_markt_data(datagen, config["bilder"]["original_path"] + config_knn.excluded_folder.name)
     all_metrics = model.evaluate(test_it, steps=len(test_it), verbose=1)
-    loss, accuracy, recall_0, recall_1 = all_metrics
+    loss, accuracy, recall_not_oos, recall_oos = all_metrics
     print('loss: %.3f' % loss + ' - accuracy: %.3f' % accuracy +
-          ' - recall_0: %.3f' % recall_0 + ' - recall_1: %.3f' % recall_1)
+          ' - recall_not_oos: %.3f' % recall_not_oos + ' - recall_oos: %.3f' % recall_oos)
     return all_metrics

src/knn/defineKNN.py

@@ -23,11 +23,11 @@ def define_model(config_knn):
     opt = get_optimization_method(config_knn.optimization_method)
     # Die Klassifizierungsmetriken und die Loss-Funktion werden definiert
     loss_fn = tf.keras.losses.BinaryCrossentropy()
-    recall_0 = tf.keras.metrics.Recall(name='recall_0', class_id=0)
-    recall_1 = tf.keras.metrics.Recall(name='recall_1', class_id=1)
+    recall_not_oos = tf.keras.metrics.Recall(name='recall_not_oos', class_id=0)
+    recall_oos = tf.keras.metrics.Recall(name='recall_oos', class_id=1)
 
     # Füge die Metriken dem Modell hinzu
-    model.compile(optimizer=opt, loss=loss_fn, metrics=['accuracy', recall_0, recall_1])
+    model.compile(optimizer=opt, loss=loss_fn, metrics=['accuracy', recall_not_oos, recall_oos])
     return model
 
 

src/result/customCallback.py

@@ -35,10 +35,10 @@ class CustomCallback(tf.keras.callbacks.Callback):
             logs = {}
         self.epoch_training_losses.append(logs.get('loss'))
         self.epoch_training_accuracies.append(logs.get('accuracy'))
-        self.epoch_training_recall_oos.append(logs.get('recall_0'))
-        self.epoch_training_recall_not_oos.append(logs.get('recall_1'))
+        self.epoch_training_recall_oos.append(logs.get('recall_oos'))
+        self.epoch_training_recall_not_oos.append(logs.get('recall_not_oos'))
         self.epoch_test_losses.append(logs.get('val_loss'))
         self.epoch_test_accuracies.append(logs.get('val_accuracy'))
-        self.epoch_test_recall_oos.append(logs.get('val_recall_0'))
-        self.epoch_test_recall_not_oos.append(logs.get('val_recall_1'))
+        self.epoch_test_recall_oos.append(logs.get('val_recall_oos'))
+        self.epoch_test_recall_not_oos.append(logs.get('val_recall_not_oos'))
         self.epoch_durations.append(time.time() - self.epoch_start_time)

src/result/plotResult.py

@@ -15,11 +15,11 @@ def plot_values(history, config):
     pyplot.subplot(212)
     pyplot.title('Classification Accuracy')
     pyplot.plot(history.history['accuracy'], color='blue', label='Accuracy')
-    pyplot.plot(history.history['recall_0'], color='blue', label='Recall OOS')
-    pyplot.plot(history.history['recall_1'], color='yellow', label='Recall !OOS')
+    pyplot.plot(history.history['recall_oos'], color='blue', label='Recall OOS')
+    pyplot.plot(history.history['recall_not_oos'], color='yellow', label='Recall !OOS')
     pyplot.plot(history.history['val_accuracy'], color='orange', label='Test Accuracy')
-    pyplot.plot(history.history['val_recall_0'], color='red', label='Test Recall OOS')
-    pyplot.plot(history.history['val_recall_1'], color='green', label='Test Recall !OOS')
+    pyplot.plot(history.history['val_recall_oos'], color='red', label='Test Recall OOS')
+    pyplot.plot(history.history['val_recall_not_oos'], color='green', label='Test Recall !OOS')
     # save plot to file
     filename = sys.argv[0].split('/')[-1]
     pyplot.savefig(config["result"]["plot_path"] + filename + '_plot.png')