{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d24d903c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import random\n",
    "from matplotlib import pyplot as plt\n",
    "import cv2\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from keras.utils import image_dataset_from_directory\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Rescaling, Conv2D, MaxPooling2D, Flatten, Dense, BatchNormalization, Activation, Dropout\n",
    "from keras.metrics import Precision, Recall\n",
    "\n",
    "import keras_tuner as kt\n",
    "from keras.optimizers import Adam\n",
    "from keras.callbacks import EarlyStopping\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bb7a4955",
   "metadata": {},
   "outputs": [],
   "source": [
    "top_dir = 'C:/Users/coco/Untitled Folder'\n",
    "train_dir = top_dir + '/train'\n",
    "test_dir = top_dir + '/test'\n",
    "\n",
    "train_real = os.listdir(train_dir + '/REAL')\n",
    "train_fake = os.listdir(train_dir + '/FAKE')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "36f4f987",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot k-number of images from the dataset\n",
    "\n",
    "def plot_im(directory, k):\n",
    "\n",
    "    files = os.listdir(directory)\n",
    "    im = random.choices(files, k = k)\n",
    "\n",
    "    fig = plt.figure()\n",
    "\n",
    "    for i in range(k):\n",
    "        im_i = cv2.imread(directory + im[i])\n",
    "        fig.add_subplot(int(np.sqrt(k)), int(np.sqrt(k)), i + 1)\n",
    "        plt.imshow(im_i)\n",
    "        plt.axis('off')\n",
    "    \n",
    "    return plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6b1efe2c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "real_im = plot_im(directory = train_dir + '/REAL/', k = 9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7ff6d9bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fake_im = plot_im(directory = train_dir + '/FAKE/', k = 9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0d882fc1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 100000 files belonging to 2 classes.\n",
      "Found 20000 files belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "# Generate tf datasets\n",
    "\n",
    "train = image_dataset_from_directory(\n",
    "    train_dir,\n",
    "    label_mode = 'binary',\n",
    "    image_size = (32, 32)\n",
    ")\n",
    "\n",
    "test = image_dataset_from_directory(\n",
    "    test_dir,\n",
    "    label_mode = 'binary',\n",
    "    image_size = (32, 32)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e1636263",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m75s\u001b[0m 23ms/step - accuracy: 0.8521 - loss: 0.3378 - precision: 0.8496 - recall: 0.8529 - val_accuracy: 0.7082 - val_loss: 0.6587 - val_precision: 0.9957 - val_recall: 0.4183\n",
      "Epoch 2/5\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m85s\u001b[0m 27ms/step - accuracy: 0.9123 - loss: 0.2164 - precision: 0.9103 - recall: 0.9138 - val_accuracy: 0.8600 - val_loss: 0.3534 - val_precision: 0.7894 - val_recall: 0.9820\n",
      "Epoch 3/5\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m71s\u001b[0m 23ms/step - accuracy: 0.9238 - loss: 0.1882 - precision: 0.9217 - recall: 0.9258 - val_accuracy: 0.9063 - val_loss: 0.2333 - val_precision: 0.9523 - val_recall: 0.8554\n",
      "Epoch 4/5\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m72s\u001b[0m 23ms/step - accuracy: 0.9333 - loss: 0.1702 - precision: 0.9316 - recall: 0.9346 - val_accuracy: 0.9310 - val_loss: 0.1755 - val_precision: 0.9118 - val_recall: 0.9543\n",
      "Epoch 5/5\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m58s\u001b[0m 18ms/step - accuracy: 0.9402 - loss: 0.1554 - precision: 0.9385 - recall: 0.9418 - val_accuracy: 0.9275 - val_loss: 0.1865 - val_precision: 0.9609 - val_recall: 0.8912\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.history.History at 0x1e971eafc10>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "simple_model = Sequential([\n",
    "    Rescaling(scale = 1./255),\n",
    "    Conv2D(filters = 64, kernel_size = 3, activation = 'relu'),\n",
    "    MaxPooling2D(),\n",
    "    Conv2D(filters = 64, kernel_size = 3, activation = 'relu'),\n",
    "    MaxPooling2D(),\n",
    "    Flatten(),\n",
    "    Dense(units = 64),\n",
    "    BatchNormalization(),\n",
    "    Activation('relu'),\n",
    "    Dense(units = 1, activation = 'sigmoid')\n",
    "])\n",
    "\n",
    "simple_model.compile(\n",
    "    optimizer = 'adam',\n",
    "    loss = 'binary_crossentropy',\n",
    "    metrics = ['accuracy', Precision(), Recall()]\n",
    ")\n",
    "\n",
    "simple_model.fit(train, validation_data = test, epochs = 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "dd7677ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hyperparameter tuning with keras_tuner\n",
    "\n",
    "def model_builder(hp):\n",
    "    \n",
    "    # Hyperparameters\n",
    "    dense_units = hp.Int('dense_units', min_value = 16, max_value = 128, step = 16)\n",
    "    conv2d_filters_0 = hp.Int('conv_filters_0', min_value = 16, max_value = 128, step = 16)\n",
    "    conv2d_filters_1 = hp.Int('conv_filters_1', min_value = 16, max_value = 128, step = 16)\n",
    "    dropout_rate = hp.Int('dropout_rate', min_value = 0, max_value = 5, step = 1)\n",
    "    learning_rate = hp.Choice('learning_rate', values = [1e-2, 1e-3, 1e-4])\n",
    "    \n",
    "    model = Sequential([\n",
    "        Rescaling(scale = 1./255),\n",
    "        Conv2D(filters = conv2d_filters_0, kernel_size = 3, activation = 'relu'),\n",
    "        MaxPooling2D(),\n",
    "        Conv2D(filters = conv2d_filters_1, kernel_size = 3, activation = 'relu'),\n",
    "        MaxPooling2D(),\n",
    "        Dropout(rate = dropout_rate/10),\n",
    "        Flatten(),\n",
    "        Dense(units = dense_units),\n",
    "        BatchNormalization(),\n",
    "        Activation('relu'),\n",
    "        Dense(units = 1, activation = 'sigmoid')\n",
    "    ])\n",
    "    \n",
    "    model.compile(\n",
    "        optimizer = Adam(learning_rate = learning_rate),\n",
    "        loss = 'binary_crossentropy',\n",
    "        metrics = ['accuracy', Precision(), Recall()]\n",
    "    )\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e326f9ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 30 Complete [00h 10m 21s]\n",
      "val_accuracy: 0.9369000196456909\n",
      "\n",
      "Best val_accuracy So Far: 0.9488000273704529\n",
      "Total elapsed time: 02h 40m 21s\n"
     ]
    }
   ],
   "source": [
    "# Optimize model for validation accuracy\n",
    "\n",
    "stop_early = EarlyStopping(monitor = 'val_accuracy', patience = 3)\n",
    "\n",
    "tuner = kt.Hyperband(\n",
    "    hypermodel = model_builder,\n",
    "    objective = 'val_accuracy',\n",
    "    max_epochs = 10,\n",
    "    factor = 3,\n",
    "    directory = 'tuning',\n",
    "    overwrite = True\n",
    ")\n",
    "\n",
    "tuner.search(\n",
    "    train,\n",
    "    validation_data = test,\n",
    "    callbacks = [stop_early]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7729e909",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m89s\u001b[0m 27ms/step - accuracy: 0.7996 - loss: 0.4318 - precision_3: 0.7911 - recall_3: 0.8103 - val_accuracy: 0.8865 - val_loss: 0.2705 - val_precision_3: 0.8360 - val_recall_3: 0.9617\n",
      "Epoch 2/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m84s\u001b[0m 27ms/step - accuracy: 0.9037 - loss: 0.2442 - precision_3: 0.9009 - recall_3: 0.9062 - val_accuracy: 0.9000 - val_loss: 0.2349 - val_precision_3: 0.8495 - val_recall_3: 0.9723\n",
      "Epoch 3/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m84s\u001b[0m 27ms/step - accuracy: 0.9209 - loss: 0.2030 - precision_3: 0.9194 - recall_3: 0.9219 - val_accuracy: 0.9233 - val_loss: 0.1907 - val_precision_3: 0.8899 - val_recall_3: 0.9662\n",
      "Epoch 4/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m85s\u001b[0m 27ms/step - accuracy: 0.9304 - loss: 0.1793 - precision_3: 0.9279 - recall_3: 0.9328 - val_accuracy: 0.9277 - val_loss: 0.1793 - val_precision_3: 0.8959 - val_recall_3: 0.9680\n",
      "Epoch 5/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m113s\u001b[0m 36ms/step - accuracy: 0.9363 - loss: 0.1651 - precision_3: 0.9341 - recall_3: 0.9383 - val_accuracy: 0.8899 - val_loss: 0.2645 - val_precision_3: 0.8264 - val_recall_3: 0.9871\n",
      "Epoch 6/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m115s\u001b[0m 37ms/step - accuracy: 0.9391 - loss: 0.1547 - precision_3: 0.9374 - recall_3: 0.9405 - val_accuracy: 0.9143 - val_loss: 0.2128 - val_precision_3: 0.9804 - val_recall_3: 0.8454\n",
      "Epoch 7/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m124s\u001b[0m 40ms/step - accuracy: 0.9449 - loss: 0.1436 - precision_3: 0.9417 - recall_3: 0.9479 - val_accuracy: 0.9445 - val_loss: 0.1432 - val_precision_3: 0.9413 - val_recall_3: 0.9481\n",
      "Epoch 8/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m127s\u001b[0m 41ms/step - accuracy: 0.9471 - loss: 0.1357 - precision_3: 0.9451 - recall_3: 0.9490 - val_accuracy: 0.9406 - val_loss: 0.1469 - val_precision_3: 0.9314 - val_recall_3: 0.9513\n",
      "Epoch 9/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m139s\u001b[0m 40ms/step - accuracy: 0.9496 - loss: 0.1287 - precision_3: 0.9477 - recall_3: 0.9513 - val_accuracy: 0.9448 - val_loss: 0.1411 - val_precision_3: 0.9600 - val_recall_3: 0.9282\n",
      "Epoch 10/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m133s\u001b[0m 37ms/step - accuracy: 0.9534 - loss: 0.1233 - precision_3: 0.9520 - recall_3: 0.9546 - val_accuracy: 0.9346 - val_loss: 0.1659 - val_precision_3: 0.9694 - val_recall_3: 0.8975\n",
      "Epoch 11/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m115s\u001b[0m 37ms/step - accuracy: 0.9559 - loss: 0.1158 - precision_3: 0.9538 - recall_3: 0.9579 - val_accuracy: 0.9334 - val_loss: 0.1703 - val_precision_3: 0.9004 - val_recall_3: 0.9746\n",
      "Epoch 12/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m117s\u001b[0m 37ms/step - accuracy: 0.9568 - loss: 0.1130 - precision_3: 0.9559 - recall_3: 0.9574 - val_accuracy: 0.9470 - val_loss: 0.1376 - val_precision_3: 0.9324 - val_recall_3: 0.9638\n",
      "Epoch 13/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m145s\u001b[0m 38ms/step - accuracy: 0.9596 - loss: 0.1059 - precision_3: 0.9584 - recall_3: 0.9607 - val_accuracy: 0.9481 - val_loss: 0.1328 - val_precision_3: 0.9464 - val_recall_3: 0.9501\n",
      "Epoch 14/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m116s\u001b[0m 37ms/step - accuracy: 0.9628 - loss: 0.0989 - precision_3: 0.9616 - recall_3: 0.9638 - val_accuracy: 0.9507 - val_loss: 0.1315 - val_precision_3: 0.9618 - val_recall_3: 0.9388\n",
      "Epoch 15/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m116s\u001b[0m 37ms/step - accuracy: 0.9615 - loss: 0.0980 - precision_3: 0.9592 - recall_3: 0.9636 - val_accuracy: 0.9076 - val_loss: 0.2485 - val_precision_3: 0.8504 - val_recall_3: 0.9892\n",
      "Epoch 16/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m118s\u001b[0m 38ms/step - accuracy: 0.9661 - loss: 0.0922 - precision_3: 0.9653 - recall_3: 0.9667 - val_accuracy: 0.9469 - val_loss: 0.1394 - val_precision_3: 0.9368 - val_recall_3: 0.9584\n",
      "Epoch 17/20\n",
      "\u001b[1m3125/3125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m118s\u001b[0m 38ms/step - accuracy: 0.9666 - loss: 0.0883 - precision_3: 0.9662 - recall_3: 0.9668 - val_accuracy: 0.9483 - val_loss: 0.1359 - val_precision_3: 0.9377 - val_recall_3: 0.9604\n"
     ]
    }
   ],
   "source": [
    "optimal_hps = tuner.get_best_hyperparameters()[0]\n",
    "model = tuner.hypermodel.build(optimal_hps)\n",
    "\n",
    "history = model.fit(\n",
    "    train,\n",
    "    validation_data = test,\n",
    "    epochs = 20,\n",
    "    callbacks = [stop_early]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "046ad861",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_1\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential_1\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ rescaling_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">30</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)     │         <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">15</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">15</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">83,040</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3456</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)             │        <span style=\"color: #00af00; text-decoration-color: #00af00\">55,312</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_1           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)             │            <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)             │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)              │            <span style=\"color: #00af00; text-decoration-color: #00af00\">17</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ rescaling_1 (\u001b[38;5;33mRescaling\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m3\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m30\u001b[0m, \u001b[38;5;34m96\u001b[0m)     │         \u001b[38;5;34m2,688\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m15\u001b[0m, \u001b[38;5;34m15\u001b[0m, \u001b[38;5;34m96\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m96\u001b[0m)     │        \u001b[38;5;34m83,040\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m96\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m6\u001b[0m, \u001b[38;5;34m96\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3456\u001b[0m)           │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)             │        \u001b[38;5;34m55,312\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ batch_normalization_1           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)             │            \u001b[38;5;34m64\u001b[0m │\n",
       "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_1 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m)              │            \u001b[38;5;34m17\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">423,301</span> (1.61 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m423,301\u001b[0m (1.61 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">141,089</span> (551.13 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m141,089\u001b[0m (551.13 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">32</span> (128.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m32\u001b[0m (128.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">282,180</span> (1.08 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m282,180\u001b[0m (1.08 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.summary()\n",
    "model.save('AIdetect.keras')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c38bc7bf",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}