diff --git a/backup.ipynb b/backup.ipynb deleted file mode 100644 index a9a52049143171217d17418469293b143065a0d7..0000000000000000000000000000000000000000 --- a/backup.ipynb +++ /dev/null @@ -1,93 +0,0 @@ -{ - "cells": [ - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "###########################################################################\n", - "# ImageGenerator with dataframes\n", - "###########################################################################\n", - "\n", - "#Size of our input images\n", - "SIZE = 128\n", - "# Size of ba\n", - "batch_size = 64\n", - "\n", - "# Pfad zum Ordner, der nur Bilder der Klasse A enthält\n", - "src_path = \"data/cell_images\"\n", - "\n", - "\n", - "# Klasse \"uninfected_train\"\n", - "# Pfad zum Ordner mit den Bildern\n", - "src_path_train = \"data/cell_images/uninfected_train\"\n", - "# Liste der Dateinamen im Ordner\n", - "file_list_train = os.listdir(src_path_train)\n", - "# Liste der Labels (Klassen) für die Bilder\n", - "labels_train = ['uninfected_train'] * len(file_list_train) # Bilder im Ordner werden Klasse \"uninfected_train\" zugeordnet\n", - "# Erstellen eines DataFrames mit Dateinamen und den entsprechenden Labels\n", - "df_data_train = pd.DataFrame({'filename': file_list_train, 'label': labels_train})\n", - "\n", - "# Klasse \"uninfected_test\"\n", - "src_path_test = \"data/cell_images/uninfected_test\"\n", - "file_list_test = os.listdir(src_path_test)\n", - "labels_test = ['uninfected_test'] * len(file_list_test)\n", - "df_data_test = pd.DataFrame({'filename': file_list_test, 'label': labels_test})\n", - "\n", - "# Klasse \"parasitized\"\n", - "src_path_parasitized = \"data/cell_images/parasitized\"\n", - "file_list_parasitized = os.listdir(src_path_parasitized)\n", - "labels_parasitized = ['parasitized'] * len(file_list_parasitized)\n", - "df_data_parasitized = pd.DataFrame({'filename': file_list_parasitized, 'label': labels_parasitized})\n", - "\n", - "\n", - "#Define generators for training, validation and also anomaly data\n", - "# Konfigurieren des ImageDataGenerator für das Rescaling der Pixelwerte\n", - "datagen = ImageDataGenerator(rescale=1./255)\n", - "\n", - "# Erstellen eines ImageDataGenerator-Objekts, um Bilder und Labels zu laden, Klasse \"df_data_train\"\n", - "train_generator = datagen.flow_from_dataframe(\n", - " df_data_train,\n", - " src_path_train, # Verzeichnis, das die Bilder enthält\n", - " x_col='filename', # Name der Spalte im DataFrame, die die Dateinamen enthält\n", - " y_col='label', # Name der Spalte im DataFrame, die die Labels enthält\n", - " target_size=(SIZE, SIZE), # Größe der Eingabebilder\n", - " batch_size=batch_size, # Anzahl der Bilder pro Batch\n", - " class_mode='categorical', # 'categorical' für Klassifikation, 'binary' für binäre Klassifikation\n", - " shuffle=True\n", - ")\n", - "\n", - "validation_generator = datagen.flow_from_dataframe(\n", - " df_data_test,\n", - " src_path_test,\n", - " x_col='filename',\n", - " y_col='label',\n", - " target_size=(SIZE, SIZE),\n", - " batch_size=batch_size,\n", - " class_mode='categorical',\n", - " shuffle=True\n", - ")\n", - "\n", - "anomaly_generator = datagen.flow_from_dataframe(\n", - " df_data_parasitized,\n", - " src_path_parasitized,\n", - " x_col='filename',\n", - " y_col='label',\n", - " target_size=(SIZE, SIZE),\n", - " batch_size=batch_size,\n", - " class_mode='categorical',\n", - " shuffle=True\n", - ")" - ] - } - ], - "metadata": { - "language_info": { - "name": "python" - }, - "orig_nbformat": 4 - }, - "nbformat": 4, - "nbformat_minor": 2 -}