{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "50cc3f9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import cv2\n",
    "import matplotlib.pyplot as plt\n",
    "from tqdm import tqdm\n",
    "\n",
    "# Directory to Pictures\n",
    "DATA_DIR = \"C:/Users/muhle/Desktop/Code/DataSets/PetAI\"\n",
    "# Kategorien die zu erkennen sind\n",
    "CATEGORIES = [\"DOG\", \"CAT\"]\n",
    "\n",
    "# Die Ordner mit den Kategorien im Namen Suchen, anschließend alle Bilder in Schwarz-Weiß umformen \n",
    "for category in CATEGORIES:\n",
    "    path = os.path.join(DATA_DIR, category)\n",
    "    for img in os.listdir(path):\n",
    "        img_array = cv2.imread(os.path.join(path, img), cv2.IMREAD_GRAYSCALE)\n",
    "        # Anzeigen eines Bildes mit Matplotlib\n",
    "        plt.imshow(img_array, cmap=\"gray\")\n",
    "        plt.show()\n",
    "        break\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7847ac85",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Größe des Bildes definieren\n",
    "IMG_SIZE = 50\n",
    "# Größe des Bild Array's ändern\n",
    "new_array = cv2.resize(img_array, (IMG_SIZE, IMG_SIZE))\n",
    "# Neues verpixeltes Bild anzeigen\n",
    "plt.imshow(new_array, cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "709419fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Liste für Trainingsdaten definieren\n",
    "training_data = []\n",
    "\n",
    "def create_training_data():\n",
    "    for category in CATEGORIES:  # do dogs and cats\n",
    "\n",
    "        path = os.path.join(DATA_DIR,category)  # Pfad zu Hunde und Katzen erstellen\n",
    "        class_num = CATEGORIES.index(category)  # Klassifikation für Hund und Katze bekommen, Hund=0; Katze=1;\n",
    "\n",
    "        for img in tqdm(os.listdir(path)):  # Bilder von Katzen und Hunden wiederholen\n",
    "            try:\n",
    "                img_array = cv2.imread(os.path.join(path,img) ,cv2.IMREAD_GRAYSCALE)  # Bild zu einen Array\n",
    "                new_array = cv2.resize(img_array, (IMG_SIZE, IMG_SIZE))  # Größe zu einer normalen Datengröße machen (50,50)\n",
    "                training_data.append([new_array, class_num])  # Hinzufügen zu unserer Liste trainings_data\n",
    "            except Exception as e:  # Wenn ein Fehler auftritt soll es einfach übersrpingen und weitermachen\n",
    "                pass\n",
    "create_training_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "41c1245c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Länge der training_data\n",
    "print(len(training_data))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cdfe3e93",
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "# Random Werte aus training_data nehmen\n",
    "random.shuffle(training_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f1e9c8e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Array ausgeben\n",
    "for sample in training_data[:10]:\n",
    "    print(sample[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5eaae186",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Input (X) und Output (y) definieren\n",
    "X = []\n",
    "y = []\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "510602f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "for features, label in training_data: \n",
    "    X.append(features)\n",
    "    y.append(label)\n",
    "    \n",
    "# X in einen Numpy Array umwandeln und ihn eine neue Größe geben, 1 Farmkanal (Grau)\n",
    "X = np.array(X).reshape(-1, IMG_SIZE, IMG_SIZE, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b59339a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "# X und y in Pickle dateien schreiben\n",
    "pickle_out = open(\"X.pickle\", \"wb\")\n",
    "pickle.dump(X, pickle_out)\n",
    "pickle_out.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b487159",
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle_out = open(\"y.pickle\", \"wb\")\n",
    "pickle.dump(y, pickle_out)\n",
    "pickle_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6aabb96e",
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle_out = open(\"X.pickle\", \"wb\")\n",
    "pickle.dump(X, pickle_out)\n",
    "pickle_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f0e4ca0",
   "metadata": {},
   "outputs": [],
   "source": [
    "X[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7532702",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3ff3800d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}