training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0ecbb814",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:27:49.706576Z",
     "iopub.status.busy": "2024-05-03T04:27:49.706211Z",
     "iopub.status.idle": "2024-05-03T04:28:03.000595Z",
     "shell.execute_reply": "2024-05-03T04:28:02.999612Z"
    },
    "papermill": {
     "duration": 13.304788,
     "end_time": "2024-05-03T04:28:03.003092",
     "exception": false,
     "start_time": "2024-05-03T04:27:49.698304",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-05-03 04:27:52.361759: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2024-05-03 04:27:52.361856: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2024-05-03 04:27:52.486507: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
     ]
    }
   ],
   "source": [
    "%%capture\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import cv2\n",
    "from tqdm.notebook import tqdm_notebook\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator, img_to_array, load_img\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "44227977",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:28:03.017487Z",
     "iopub.status.busy": "2024-05-03T04:28:03.016908Z",
     "iopub.status.idle": "2024-05-03T04:28:03.021159Z",
     "shell.execute_reply": "2024-05-03T04:28:03.020317Z"
    },
    "papermill": {
     "duration": 0.013221,
     "end_time": "2024-05-03T04:28:03.022964",
     "exception": false,
     "start_time": "2024-05-03T04:28:03.009743",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "input_dir = '/kaggle/input'\n",
    "output_dir = '/kaggle/working'\n",
    "working_dir = '/kaggle/working'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ab16b0e5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:28:03.035882Z",
     "iopub.status.busy": "2024-05-03T04:28:03.035607Z",
     "iopub.status.idle": "2024-05-03T04:28:03.977460Z",
     "shell.execute_reply": "2024-05-03T04:28:03.976443Z"
    },
    "papermill": {
     "duration": 0.951118,
     "end_time": "2024-05-03T04:28:03.979969",
     "exception": false,
     "start_time": "2024-05-03T04:28:03.028851",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ffhq-face-data-set\n",
      "person-face-dataset-thispersondoesnotexist\n",
      "\n",
      "Real image count: 70000\n",
      "Fake image count: 10000\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "ls /kaggle/input\n",
    "echo\n",
    "real_img_count=`ls /kaggle/input/ffhq-face-data-set/thumbnails128x128 | wc -l`\n",
    "echo \"Real image count: $real_img_count\" \n",
    "fake_img_count=`ls /kaggle/input/person-face-dataset-thispersondoesnotexist/thispersondoesnotexist.10k | wc -l`\n",
    "echo \"Fake image count: $fake_img_count\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21b2c0b8",
   "metadata": {
    "papermill": {
     "duration": 0.006649,
     "end_time": "2024-05-03T04:28:03.993849",
     "exception": false,
     "start_time": "2024-05-03T04:28:03.987200",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ebdacdc2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:28:04.008924Z",
     "iopub.status.busy": "2024-05-03T04:28:04.008377Z",
     "iopub.status.idle": "2024-05-03T04:28:04.013072Z",
     "shell.execute_reply": "2024-05-03T04:28:04.012165Z"
    },
    "papermill": {
     "duration": 0.014385,
     "end_time": "2024-05-03T04:28:04.014979",
     "exception": false,
     "start_time": "2024-05-03T04:28:04.000594",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "real_img_dir = f\"{input_dir}/ffhq-face-data-set/thumbnails128x128\"\n",
    "fake_img_dir = f\"{input_dir}/person-face-dataset-thispersondoesnotexist/thispersondoesnotexist.10k\"\n",
    "\n",
    "input_shape = (128,128,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "202f45af",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:28:04.029562Z",
     "iopub.status.busy": "2024-05-03T04:28:04.028997Z",
     "iopub.status.idle": "2024-05-03T04:30:29.777439Z",
     "shell.execute_reply": "2024-05-03T04:30:29.776360Z"
    },
    "papermill": {
     "duration": 145.758161,
     "end_time": "2024-05-03T04:30:29.779737",
     "exception": false,
     "start_time": "2024-05-03T04:28:04.021576",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "32ee2fcd79b64f9b88caa9d32ffb7165",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5ee82844b614cf6aba62ca60616a3e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/3000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# processing data\n",
    "X = []\n",
    "Y = []\n",
    "\n",
    "for img in tqdm_notebook(os.listdir(real_img_dir)[:3000]):\n",
    "    X.append(img_to_array(load_img(f\"{real_img_dir}/{img}\", target_size=input_shape)) / 255.0)\n",
    "    Y.append(0)\n",
    "\n",
    "for img in tqdm_notebook(os.listdir(fake_img_dir)[:3000]):\n",
    "    X.append(img_to_array(load_img(f\"{fake_img_dir}/{img}\", target_size=input_shape)) / 255.0)\n",
    "    Y.append(1)\n",
    "\n",
    "\n",
    "# normalization\n",
    "X = np.array(X)\n",
    "X = X.reshape(-1, 128, 128, 3)\n",
    "Y = to_categorical(Y, 2)\n",
    "# Y = np.array(Y)\n",
    "\n",
    "# saving processed data arrays\n",
    "os.chdir(output_dir)\n",
    "np.save('X.npy',X)\n",
    "np.save('Y.npy',Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b65b2957",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:30:29.794904Z",
     "iopub.status.busy": "2024-05-03T04:30:29.794596Z",
     "iopub.status.idle": "2024-05-03T04:30:30.489581Z",
     "shell.execute_reply": "2024-05-03T04:30:30.488490Z"
    },
    "papermill": {
     "duration": 0.704945,
     "end_time": "2024-05-03T04:30:30.491757",
     "exception": false,
     "start_time": "2024-05-03T04:30:29.786812",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "data shape: (6000, 128, 128, 3); labels shape: (6000, 2)\n",
      "X_train shape: (4800, 128, 128, 3)\n",
      "Y_train shape: (4800, 2)\n",
      "X_val shape: (1200, 128, 128, 3)\n",
      "Y_val shape: (1200, 2)\n"
     ]
    }
   ],
   "source": [
    "# load saved data files\n",
    "os.chdir(working_dir)\n",
    "X = np.load('X.npy')\n",
    "Y = np.load('Y.npy')\n",
    "\n",
    "# train test split\n",
    "X_train, X_val, Y_train, Y_val = train_test_split(X, Y, test_size = 0.2, random_state=5)\n",
    "\n",
    "# info\n",
    "print(f\"\\ndata shape: {X.shape}; labels shape: {Y.shape}\")\n",
    "print(f\"X_train shape: {X_train.shape}\\nY_train shape: {Y_train.shape}\")\n",
    "print(f\"X_val shape: {X_val.shape}\\nY_val shape: {Y_val.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e5cf833",
   "metadata": {
    "papermill": {
     "duration": 0.006639,
     "end_time": "2024-05-03T04:30:30.506663",
     "exception": false,
     "start_time": "2024-05-03T04:30:30.500024",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Model Creation and Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1e11f61d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:30:30.521691Z",
     "iopub.status.busy": "2024-05-03T04:30:30.521380Z",
     "iopub.status.idle": "2024-05-03T04:30:30.531957Z",
     "shell.execute_reply": "2024-05-03T04:30:30.531293Z"
    },
    "papermill": {
     "duration": 0.020308,
     "end_time": "2024-05-03T04:30:30.533894",
     "exception": false,
     "start_time": "2024-05-03T04:30:30.513586",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications import InceptionResNetV2,InceptionV3\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Flatten, Input, InputLayer, GlobalAveragePooling2D, Activation, MaxPool2D\n",
    "from tensorflow.keras.models import Model, Sequential, load_model\n",
    "from tensorflow.keras import optimizers\n",
    "from tensorflow.keras.callbacks import ReduceLROnPlateau, EarlyStopping, ModelCheckpoint\n",
    "from keras import backend as K\n",
    "\n",
    "# transfer_model = InceptionV3(include_top=False, weights='imagenet', input_shape=input_shape, classes=2)\n",
    "# transfer_model.trainable = True\n",
    "# model = Sequential()\n",
    "# model.add(transfer_model)\n",
    "# model.add(GlobalAveragePooling2D(name='avg_2d_pool'))\n",
    "# model.add(Flatten(name='flatten_layer'))\n",
    "\n",
    "# model.add(Dense(100,activation='relu))\n",
    "# model.add(Dense(units=2, activation='softmax'))\n",
    "# model.compile(loss='binary_crossentropy',\n",
    "#               optimizer=optimizers.Adam(learning_rate=1e-5, beta_1=0.9, beta_2=0.999),\n",
    "#               metrics=['accuracy'])\n",
    "\n",
    "# model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "fa84375e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:30:30.548513Z",
     "iopub.status.busy": "2024-05-03T04:30:30.548257Z",
     "iopub.status.idle": "2024-05-03T04:30:31.368324Z",
     "shell.execute_reply": "2024-05-03T04:30:31.367442Z"
    },
    "papermill": {
     "duration": 0.82964,
     "end_time": "2024-05-03T04:30:31.370287",
     "exception": false,
     "start_time": "2024-05-03T04:30:30.540647",
     "status": "completed"
    },
    "tags": []
   },
   "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: \"BaseModel\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"BaseModel\"\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",
       "│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │         <span style=\"color: #00af00; text-decoration-color: #00af00\">4,864</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation (<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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)   │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_1 (<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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │       <span style=\"color: #00af00; text-decoration-color: #00af00\">102,464</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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_1 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │       <span style=\"color: #00af00; text-decoration-color: #00af00\">102,464</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_2 (<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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,464</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_3 (<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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_4 (<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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │         <span style=\"color: #00af00; text-decoration-color: #00af00\">4,624</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_4 (<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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (<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\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (<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\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │         <span style=\"color: #00af00; text-decoration-color: #00af00\">2,320</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_5 (<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\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ global_average_pooling2d        │ (<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",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePooling2D</span>)        │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten (<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\">16</span>)             │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (<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\">400</span>)            │         <span style=\"color: #00af00; text-decoration-color: #00af00\">6,800</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (<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\">400</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_6 (<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\">400</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<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\">512</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">205,312</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\">512</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_7 (<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\">512</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\">400</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">205,200</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_2 (<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\">400</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_8 (<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\">400</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\">2</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">802</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_9 (<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\">2</span>)              │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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",
       "│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │         \u001b[38;5;34m4,864\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation (\u001b[38;5;33mActivation\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m)   │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │       \u001b[38;5;34m102,464\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_1 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m64\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;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │       \u001b[38;5;34m102,464\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_2 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m64\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;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │        \u001b[38;5;34m18,464\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_3 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m32\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │         \u001b[38;5;34m4,624\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_4 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │         \u001b[38;5;34m2,320\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_5 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ max_pooling2d_5 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m16\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ global_average_pooling2d        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
       "│ (\u001b[38;5;33mGlobalAveragePooling2D\u001b[0m)        │                        │               │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m400\u001b[0m)            │         \u001b[38;5;34m6,800\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m400\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_6 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m400\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)            │       \u001b[38;5;34m205,312\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_7 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\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;34m400\u001b[0m)            │       \u001b[38;5;34m205,200\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m400\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_8 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m400\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;34m2\u001b[0m)              │           \u001b[38;5;34m802\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ activation_9 (\u001b[38;5;33mActivation\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m)              │             \u001b[38;5;34m0\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\">653,314</span> (2.49 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m653,314\u001b[0m (2.49 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\">653,314</span> (2.49 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m653,314\u001b[0m (2.49 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\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Not using transfer learning\n",
    "# made our own model!\n",
    "\n",
    "inputs=Input(shape=input_shape)\n",
    "x = Conv2D(64,5,padding='same')(inputs)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "x = Conv2D(64,5,padding='same')(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "x = Conv2D(64,5,padding='same')(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "\n",
    " #Decreasing Filters and MAxPool Layers\n",
    "x = Conv2D(32,3,padding='same',dilation_rate=2)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "# x = BatchNormalization()(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "x = Conv2D(16,3,padding='same',dilation_rate=2)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "# x = BatchNormalization()(x)\n",
    "x = Conv2D(16,3,padding='same',dilation_rate=2)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = MaxPool2D(strides=(2,2))(x)\n",
    "\n",
    " #Dense Layers\n",
    "x = GlobalAveragePooling2D()(x)\n",
    "x = Flatten()(x)\n",
    "x = Dense(400)(x)\n",
    "x = Dropout(0.5)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = Dense(512)(x)\n",
    "x = Dropout(0.5)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = Dense(400)(x)\n",
    "x = Dropout(0.5)(x)\n",
    "x = Activation(activation='relu')(x)\n",
    "x = Dense(2)(x)\n",
    "\n",
    "#Output\n",
    "out = Activation(activation='sigmoid')(x)\n",
    "\n",
    "# final model:\n",
    "model = Model(inputs,out,name='BaseModel')\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer=optimizers.Adam(learning_rate=1e-5, beta_1=0.9, beta_2=0.999),\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8bfb30ae",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:30:31.387882Z",
     "iopub.status.busy": "2024-05-03T04:30:31.387238Z",
     "iopub.status.idle": "2024-05-03T04:35:48.860693Z",
     "shell.execute_reply": "2024-05-03T04:35:48.859686Z"
    },
    "papermill": {
     "duration": 317.484482,
     "end_time": "2024-05-03T04:35:48.862980",
     "exception": false,
     "start_time": "2024-05-03T04:30:31.378498",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "\u001b[1m 10/150\u001b[0m \u001b[32m━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m2s\u001b[0m 19ms/step - accuracy: 0.4713 - loss: 0.6935"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "I0000 00:00:1714710646.541668      79 device_compiler.h:186] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.4989 - loss: 0.6932\n",
      "Epoch 1: val_accuracy improved from -inf to 0.49417, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 37ms/step - accuracy: 0.4990 - loss: 0.6932 - val_accuracy: 0.4942 - val_loss: 0.6931\n",
      "Epoch 2/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5165 - loss: 0.6930\n",
      "Epoch 2: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.5162 - loss: 0.6930 - val_accuracy: 0.4942 - val_loss: 0.6931\n",
      "Epoch 3/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.4958 - loss: 0.6931\n",
      "Epoch 3: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.4959 - loss: 0.6931 - val_accuracy: 0.4942 - val_loss: 0.6931\n",
      "Epoch 4/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5095 - loss: 0.6931\n",
      "Epoch 4: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5094 - loss: 0.6931 - val_accuracy: 0.4942 - val_loss: 0.6930\n",
      "Epoch 5/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5136 - loss: 0.6930\n",
      "Epoch 5: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5134 - loss: 0.6930 - val_accuracy: 0.4942 - val_loss: 0.6930\n",
      "Epoch 6/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5024 - loss: 0.6929\n",
      "Epoch 6: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5023 - loss: 0.6929 - val_accuracy: 0.4942 - val_loss: 0.6929\n",
      "Epoch 7/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5105 - loss: 0.6929\n",
      "Epoch 7: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.5105 - loss: 0.6929 - val_accuracy: 0.4942 - val_loss: 0.6928\n",
      "Epoch 8/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5124 - loss: 0.6927\n",
      "Epoch 8: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5123 - loss: 0.6927 - val_accuracy: 0.4942 - val_loss: 0.6925\n",
      "Epoch 9/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5143 - loss: 0.6924\n",
      "Epoch 9: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5143 - loss: 0.6924 - val_accuracy: 0.4942 - val_loss: 0.6920\n",
      "Epoch 10/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5160 - loss: 0.6917\n",
      "Epoch 10: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5161 - loss: 0.6917 - val_accuracy: 0.4942 - val_loss: 0.6916\n",
      "Epoch 11/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5122 - loss: 0.6907\n",
      "Epoch 11: val_accuracy did not improve from 0.49417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.5123 - loss: 0.6907 - val_accuracy: 0.4942 - val_loss: 0.6890\n",
      "Epoch 12/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5281 - loss: 0.6888\n",
      "Epoch 12: val_accuracy improved from 0.49417 to 0.52750, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5281 - loss: 0.6888 - val_accuracy: 0.5275 - val_loss: 0.6848\n",
      "Epoch 13/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5482 - loss: 0.6840\n",
      "Epoch 13: val_accuracy improved from 0.52750 to 0.62167, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5484 - loss: 0.6840 - val_accuracy: 0.6217 - val_loss: 0.6763\n",
      "Epoch 14/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.5706 - loss: 0.6767\n",
      "Epoch 14: val_accuracy improved from 0.62167 to 0.66667, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.5709 - loss: 0.6767 - val_accuracy: 0.6667 - val_loss: 0.6613\n",
      "Epoch 15/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.6230 - loss: 0.6660\n",
      "Epoch 15: val_accuracy did not improve from 0.66667\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.6232 - loss: 0.6658 - val_accuracy: 0.6350 - val_loss: 0.6428\n",
      "Epoch 16/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.6448 - loss: 0.6507\n",
      "Epoch 16: val_accuracy improved from 0.66667 to 0.69833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.6450 - loss: 0.6506 - val_accuracy: 0.6983 - val_loss: 0.6114\n",
      "Epoch 17/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.6659 - loss: 0.6286\n",
      "Epoch 17: val_accuracy did not improve from 0.69833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.6662 - loss: 0.6285 - val_accuracy: 0.6683 - val_loss: 0.6091\n",
      "Epoch 18/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.7114 - loss: 0.5946\n",
      "Epoch 18: val_accuracy improved from 0.69833 to 0.73333, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.7114 - loss: 0.5945 - val_accuracy: 0.7333 - val_loss: 0.5463\n",
      "Epoch 19/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.7281 - loss: 0.5618\n",
      "Epoch 19: val_accuracy improved from 0.73333 to 0.78083, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.7281 - loss: 0.5618 - val_accuracy: 0.7808 - val_loss: 0.5016\n",
      "Epoch 20/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.7625 - loss: 0.5311\n",
      "Epoch 20: val_accuracy improved from 0.78083 to 0.80750, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.7625 - loss: 0.5309 - val_accuracy: 0.8075 - val_loss: 0.4698\n",
      "Epoch 21/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.7960 - loss: 0.4870\n",
      "Epoch 21: val_accuracy improved from 0.80750 to 0.86917, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.7961 - loss: 0.4868 - val_accuracy: 0.8692 - val_loss: 0.3808\n",
      "Epoch 22/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.8292 - loss: 0.4288\n",
      "Epoch 22: val_accuracy improved from 0.86917 to 0.89250, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.8293 - loss: 0.4287 - val_accuracy: 0.8925 - val_loss: 0.3185\n",
      "Epoch 23/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.8653 - loss: 0.3749\n",
      "Epoch 23: val_accuracy did not improve from 0.89250\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.8651 - loss: 0.3748 - val_accuracy: 0.8425 - val_loss: 0.3708\n",
      "Epoch 24/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.8642 - loss: 0.3581\n",
      "Epoch 24: val_accuracy improved from 0.89250 to 0.90000, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.8644 - loss: 0.3577 - val_accuracy: 0.9000 - val_loss: 0.2679\n",
      "Epoch 25/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.8953 - loss: 0.2990\n",
      "Epoch 25: val_accuracy improved from 0.90000 to 0.90250, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.8951 - loss: 0.2992 - val_accuracy: 0.9025 - val_loss: 0.2684\n",
      "Epoch 26/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.8925 - loss: 0.2814\n",
      "Epoch 26: val_accuracy improved from 0.90250 to 0.92833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.8926 - loss: 0.2813 - val_accuracy: 0.9283 - val_loss: 0.1959\n",
      "Epoch 27/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9134 - loss: 0.2520\n",
      "Epoch 27: val_accuracy improved from 0.92833 to 0.93083, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9135 - loss: 0.2519 - val_accuracy: 0.9308 - val_loss: 0.1791\n",
      "Epoch 28/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9182 - loss: 0.2338\n",
      "Epoch 28: val_accuracy improved from 0.93083 to 0.94000, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9183 - loss: 0.2338 - val_accuracy: 0.9400 - val_loss: 0.1700\n",
      "Epoch 29/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9184 - loss: 0.2213\n",
      "Epoch 29: val_accuracy did not improve from 0.94000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9186 - loss: 0.2211 - val_accuracy: 0.9400 - val_loss: 0.1572\n",
      "Epoch 30/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9184 - loss: 0.2195\n",
      "Epoch 30: val_accuracy improved from 0.94000 to 0.94167, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9184 - loss: 0.2196 - val_accuracy: 0.9417 - val_loss: 0.1654\n",
      "Epoch 31/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9221 - loss: 0.2166\n",
      "Epoch 31: val_accuracy improved from 0.94167 to 0.94833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9224 - loss: 0.2162 - val_accuracy: 0.9483 - val_loss: 0.1381\n",
      "Epoch 32/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9460 - loss: 0.1679\n",
      "Epoch 32: val_accuracy did not improve from 0.94833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9458 - loss: 0.1683 - val_accuracy: 0.9442 - val_loss: 0.1559\n",
      "Epoch 33/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9384 - loss: 0.1808\n",
      "Epoch 33: val_accuracy improved from 0.94833 to 0.95500, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9384 - loss: 0.1808 - val_accuracy: 0.9550 - val_loss: 0.1256\n",
      "Epoch 34/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9343 - loss: 0.1753\n",
      "Epoch 34: val_accuracy improved from 0.95500 to 0.95917, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9344 - loss: 0.1754 - val_accuracy: 0.9592 - val_loss: 0.1204\n",
      "Epoch 35/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9381 - loss: 0.1816\n",
      "Epoch 35: val_accuracy did not improve from 0.95917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9380 - loss: 0.1816 - val_accuracy: 0.9525 - val_loss: 0.1224\n",
      "Epoch 36/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9372 - loss: 0.1814\n",
      "Epoch 36: val_accuracy improved from 0.95917 to 0.96000, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9373 - loss: 0.1812 - val_accuracy: 0.9600 - val_loss: 0.1112\n",
      "Epoch 37/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9365 - loss: 0.1775\n",
      "Epoch 37: val_accuracy did not improve from 0.96000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9366 - loss: 0.1772 - val_accuracy: 0.9267 - val_loss: 0.1766\n",
      "Epoch 38/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9504 - loss: 0.1491\n",
      "Epoch 38: val_accuracy did not improve from 0.96000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9503 - loss: 0.1491 - val_accuracy: 0.9183 - val_loss: 0.1957\n",
      "Epoch 39/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9474 - loss: 0.1529\n",
      "Epoch 39: val_accuracy improved from 0.96000 to 0.96500, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 21ms/step - accuracy: 0.9474 - loss: 0.1528 - val_accuracy: 0.9650 - val_loss: 0.1007\n",
      "Epoch 40/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9496 - loss: 0.1410\n",
      "Epoch 40: val_accuracy did not improve from 0.96500\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9497 - loss: 0.1409 - val_accuracy: 0.9567 - val_loss: 0.1149\n",
      "Epoch 41/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9549 - loss: 0.1291\n",
      "Epoch 41: val_accuracy improved from 0.96500 to 0.96667, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 21ms/step - accuracy: 0.9549 - loss: 0.1290 - val_accuracy: 0.9667 - val_loss: 0.0950\n",
      "Epoch 42/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9569 - loss: 0.1286\n",
      "Epoch 42: val_accuracy improved from 0.96667 to 0.96833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9567 - loss: 0.1289 - val_accuracy: 0.9683 - val_loss: 0.0895\n",
      "Epoch 43/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9383 - loss: 0.1634\n",
      "Epoch 43: val_accuracy did not improve from 0.96833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9385 - loss: 0.1632 - val_accuracy: 0.9025 - val_loss: 0.2570\n",
      "Epoch 44/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9460 - loss: 0.1493\n",
      "Epoch 44: val_accuracy improved from 0.96833 to 0.97083, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 21ms/step - accuracy: 0.9462 - loss: 0.1491 - val_accuracy: 0.9708 - val_loss: 0.0873\n",
      "Epoch 45/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9551 - loss: 0.1314\n",
      "Epoch 45: val_accuracy improved from 0.97083 to 0.97167, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9550 - loss: 0.1317 - val_accuracy: 0.9717 - val_loss: 0.0884\n",
      "Epoch 46/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9592 - loss: 0.1245\n",
      "Epoch 46: val_accuracy did not improve from 0.97167\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9592 - loss: 0.1243 - val_accuracy: 0.9700 - val_loss: 0.0944\n",
      "Epoch 47/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9584 - loss: 0.1266\n",
      "Epoch 47: val_accuracy did not improve from 0.97167\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9584 - loss: 0.1264 - val_accuracy: 0.9717 - val_loss: 0.0866\n",
      "Epoch 48/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9666 - loss: 0.1005\n",
      "Epoch 48: val_accuracy improved from 0.97167 to 0.97417, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9665 - loss: 0.1007 - val_accuracy: 0.9742 - val_loss: 0.0806\n",
      "Epoch 49/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9659 - loss: 0.0972\n",
      "Epoch 49: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9658 - loss: 0.0974 - val_accuracy: 0.9683 - val_loss: 0.0871\n",
      "Epoch 50/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9640 - loss: 0.1018\n",
      "Epoch 50: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9639 - loss: 0.1020 - val_accuracy: 0.9592 - val_loss: 0.1096\n",
      "Epoch 51/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9550 - loss: 0.1266\n",
      "Epoch 51: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9550 - loss: 0.1267 - val_accuracy: 0.9583 - val_loss: 0.1117\n",
      "Epoch 52/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9565 - loss: 0.1191\n",
      "Epoch 52: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9566 - loss: 0.1189 - val_accuracy: 0.9642 - val_loss: 0.0995\n",
      "Epoch 53/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9632 - loss: 0.1128\n",
      "Epoch 53: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9631 - loss: 0.1126 - val_accuracy: 0.9708 - val_loss: 0.0783\n",
      "Epoch 54/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9650 - loss: 0.1021\n",
      "Epoch 54: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9649 - loss: 0.1022 - val_accuracy: 0.9150 - val_loss: 0.2216\n",
      "Epoch 55/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9688 - loss: 0.1012\n",
      "Epoch 55: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9687 - loss: 0.1012 - val_accuracy: 0.9717 - val_loss: 0.0876\n",
      "Epoch 56/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9514 - loss: 0.1317\n",
      "Epoch 56: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9516 - loss: 0.1314 - val_accuracy: 0.9742 - val_loss: 0.0766\n",
      "Epoch 57/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9638 - loss: 0.1082\n",
      "Epoch 57: val_accuracy did not improve from 0.97417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9636 - loss: 0.1085 - val_accuracy: 0.9408 - val_loss: 0.1554\n",
      "Epoch 58/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9589 - loss: 0.1134\n",
      "Epoch 58: val_accuracy improved from 0.97417 to 0.97833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9589 - loss: 0.1135 - val_accuracy: 0.9783 - val_loss: 0.0645\n",
      "Epoch 59/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9645 - loss: 0.0998\n",
      "Epoch 59: val_accuracy did not improve from 0.97833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9644 - loss: 0.1000 - val_accuracy: 0.9758 - val_loss: 0.0658\n",
      "Epoch 60/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9734 - loss: 0.0889\n",
      "Epoch 60: val_accuracy did not improve from 0.97833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9733 - loss: 0.0890 - val_accuracy: 0.9775 - val_loss: 0.0653\n",
      "Epoch 61/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9709 - loss: 0.0843\n",
      "Epoch 61: val_accuracy did not improve from 0.97833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9709 - loss: 0.0843 - val_accuracy: 0.9692 - val_loss: 0.0953\n",
      "Epoch 62/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9755 - loss: 0.0802\n",
      "Epoch 62: val_accuracy did not improve from 0.97833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9754 - loss: 0.0801 - val_accuracy: 0.9542 - val_loss: 0.1169\n",
      "Epoch 63/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9704 - loss: 0.0899\n",
      "Epoch 63: val_accuracy improved from 0.97833 to 0.97917, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9705 - loss: 0.0897 - val_accuracy: 0.9792 - val_loss: 0.0573\n",
      "Epoch 64/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9763 - loss: 0.0737\n",
      "Epoch 64: val_accuracy did not improve from 0.97917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9761 - loss: 0.0740 - val_accuracy: 0.9783 - val_loss: 0.0566\n",
      "Epoch 65/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9681 - loss: 0.0893\n",
      "Epoch 65: val_accuracy did not improve from 0.97917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9681 - loss: 0.0894 - val_accuracy: 0.9358 - val_loss: 0.1641\n",
      "Epoch 66/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9646 - loss: 0.0953\n",
      "Epoch 66: val_accuracy did not improve from 0.97917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9646 - loss: 0.0952 - val_accuracy: 0.9650 - val_loss: 0.1026\n",
      "Epoch 67/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9695 - loss: 0.0842\n",
      "Epoch 67: val_accuracy did not improve from 0.97917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9696 - loss: 0.0841 - val_accuracy: 0.9783 - val_loss: 0.0682\n",
      "Epoch 68/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9553 - loss: 0.1153\n",
      "Epoch 68: val_accuracy improved from 0.97917 to 0.98167, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9555 - loss: 0.1148 - val_accuracy: 0.9817 - val_loss: 0.0584\n",
      "Epoch 69/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9784 - loss: 0.0670\n",
      "Epoch 69: val_accuracy did not improve from 0.98167\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9783 - loss: 0.0671 - val_accuracy: 0.9692 - val_loss: 0.0923\n",
      "Epoch 70/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9788 - loss: 0.0719\n",
      "Epoch 70: val_accuracy improved from 0.98167 to 0.98417, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9788 - loss: 0.0718 - val_accuracy: 0.9842 - val_loss: 0.0513\n",
      "Epoch 71/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9759 - loss: 0.0775\n",
      "Epoch 71: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9759 - loss: 0.0774 - val_accuracy: 0.9808 - val_loss: 0.0565\n",
      "Epoch 72/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9774 - loss: 0.0683\n",
      "Epoch 72: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9773 - loss: 0.0684 - val_accuracy: 0.9642 - val_loss: 0.1059\n",
      "Epoch 73/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9489 - loss: 0.1315\n",
      "Epoch 73: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9492 - loss: 0.1309 - val_accuracy: 0.9642 - val_loss: 0.0912\n",
      "Epoch 74/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9796 - loss: 0.0632\n",
      "Epoch 74: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9795 - loss: 0.0633 - val_accuracy: 0.9842 - val_loss: 0.0479\n",
      "Epoch 75/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9829 - loss: 0.0566\n",
      "Epoch 75: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9828 - loss: 0.0569 - val_accuracy: 0.9825 - val_loss: 0.0536\n",
      "Epoch 76/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9625 - loss: 0.1002\n",
      "Epoch 76: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9626 - loss: 0.1000 - val_accuracy: 0.9683 - val_loss: 0.0916\n",
      "Epoch 77/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9751 - loss: 0.0815\n",
      "Epoch 77: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9752 - loss: 0.0813 - val_accuracy: 0.9575 - val_loss: 0.1058\n",
      "Epoch 78/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9767 - loss: 0.0676\n",
      "Epoch 78: val_accuracy did not improve from 0.98417\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9767 - loss: 0.0678 - val_accuracy: 0.9758 - val_loss: 0.0684\n",
      "Epoch 79/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9698 - loss: 0.0877\n",
      "Epoch 79: val_accuracy improved from 0.98417 to 0.98667, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9699 - loss: 0.0875 - val_accuracy: 0.9867 - val_loss: 0.0420\n",
      "Epoch 80/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9786 - loss: 0.0609\n",
      "Epoch 80: val_accuracy did not improve from 0.98667\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9784 - loss: 0.0613 - val_accuracy: 0.9750 - val_loss: 0.0714\n",
      "Epoch 81/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9813 - loss: 0.0517\n",
      "Epoch 81: val_accuracy did not improve from 0.98667\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9813 - loss: 0.0518 - val_accuracy: 0.9783 - val_loss: 0.0560\n",
      "Epoch 82/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9865 - loss: 0.0476\n",
      "Epoch 82: val_accuracy did not improve from 0.98667\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9865 - loss: 0.0476 - val_accuracy: 0.9525 - val_loss: 0.1178\n",
      "Epoch 83/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9792 - loss: 0.0628\n",
      "Epoch 83: val_accuracy improved from 0.98667 to 0.98833, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9792 - loss: 0.0628 - val_accuracy: 0.9883 - val_loss: 0.0386\n",
      "Epoch 84/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9775 - loss: 0.0633\n",
      "Epoch 84: val_accuracy did not improve from 0.98833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9774 - loss: 0.0636 - val_accuracy: 0.9775 - val_loss: 0.0598\n",
      "Epoch 85/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9807 - loss: 0.0657\n",
      "Epoch 85: val_accuracy did not improve from 0.98833\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9808 - loss: 0.0655 - val_accuracy: 0.9767 - val_loss: 0.0630\n",
      "Epoch 86/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9812 - loss: 0.0592\n",
      "Epoch 86: val_accuracy improved from 0.98833 to 0.98917, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9811 - loss: 0.0596 - val_accuracy: 0.9892 - val_loss: 0.0364\n",
      "Epoch 87/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9853 - loss: 0.0492\n",
      "Epoch 87: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9852 - loss: 0.0493 - val_accuracy: 0.9300 - val_loss: 0.1881\n",
      "Epoch 88/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9768 - loss: 0.0700\n",
      "Epoch 88: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9768 - loss: 0.0703 - val_accuracy: 0.9675 - val_loss: 0.0981\n",
      "Epoch 89/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9805 - loss: 0.0540\n",
      "Epoch 89: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9805 - loss: 0.0540 - val_accuracy: 0.9767 - val_loss: 0.0608\n",
      "Epoch 90/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9835 - loss: 0.0490\n",
      "Epoch 90: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9835 - loss: 0.0491 - val_accuracy: 0.9883 - val_loss: 0.0341\n",
      "Epoch 91/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9816 - loss: 0.0576\n",
      "Epoch 91: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9816 - loss: 0.0576 - val_accuracy: 0.9808 - val_loss: 0.0502\n",
      "Epoch 92/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9723 - loss: 0.0741\n",
      "Epoch 92: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9724 - loss: 0.0740 - val_accuracy: 0.9800 - val_loss: 0.0428\n",
      "Epoch 93/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9782 - loss: 0.0589\n",
      "Epoch 93: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9782 - loss: 0.0589 - val_accuracy: 0.9858 - val_loss: 0.0348\n",
      "Epoch 94/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9804 - loss: 0.0589\n",
      "Epoch 94: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9804 - loss: 0.0590 - val_accuracy: 0.9683 - val_loss: 0.0873\n",
      "Epoch 95/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9843 - loss: 0.0468\n",
      "Epoch 95: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9843 - loss: 0.0470 - val_accuracy: 0.9867 - val_loss: 0.0353\n",
      "Epoch 96/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9790 - loss: 0.0525\n",
      "Epoch 96: val_accuracy did not improve from 0.98917\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9791 - loss: 0.0524 - val_accuracy: 0.9642 - val_loss: 0.1084\n",
      "Epoch 97/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9810 - loss: 0.0504\n",
      "Epoch 97: val_accuracy improved from 0.98917 to 0.99000, saving model to /kaggle/working/deepfake.keras\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9809 - loss: 0.0505 - val_accuracy: 0.9900 - val_loss: 0.0323\n",
      "Epoch 98/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9832 - loss: 0.0478\n",
      "Epoch 98: val_accuracy did not improve from 0.99000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9832 - loss: 0.0478 - val_accuracy: 0.9767 - val_loss: 0.0557\n",
      "Epoch 99/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9573 - loss: 0.1243\n",
      "Epoch 99: val_accuracy did not improve from 0.99000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 19ms/step - accuracy: 0.9576 - loss: 0.1234 - val_accuracy: 0.9900 - val_loss: 0.0302\n",
      "Epoch 100/100\n",
      "\u001b[1m148/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 18ms/step - accuracy: 0.9895 - loss: 0.0404\n",
      "Epoch 100: val_accuracy did not improve from 0.99000\n",
      "\u001b[1m150/150\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 20ms/step - accuracy: 0.9895 - loss: 0.0403 - val_accuracy: 0.9867 - val_loss: 0.0317\n"
     ]
    }
   ],
   "source": [
    "# es = EarlyStopping(monitor='val_loss', min_delta=0, patience=2, verbose=1, mode='auto')\n",
    "checkpoint_callback = ModelCheckpoint(filepath='/kaggle/working/deepfake.keras', monitor='val_accuracy', save_best_only=True, verbose=1)\n",
    "\n",
    "\n",
    "train_logs = model.fit(X_train, Y_train, batch_size=32, epochs=100, \n",
    "                    validation_data = (X_val, Y_val), verbose = 1, callbacks=[checkpoint_callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b8a202f9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:35:49.776564Z",
     "iopub.status.busy": "2024-05-03T04:35:49.776211Z",
     "iopub.status.idle": "2024-05-03T04:35:50.005423Z",
     "shell.execute_reply": "2024-05-03T04:35:50.004499Z"
    },
    "papermill": {
     "duration": 0.664229,
     "end_time": "2024-05-03T04:35:50.007601",
     "exception": false,
     "start_time": "2024-05-03T04:35:49.343372",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "K.clear_session()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c64cca27",
   "metadata": {
    "papermill": {
     "duration": 0.433863,
     "end_time": "2024-05-03T04:35:50.896413",
     "exception": false,
     "start_time": "2024-05-03T04:35:50.462550",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Performance Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a24669d",
   "metadata": {
    "papermill": {
     "duration": 0.488808,
     "end_time": "2024-05-03T04:35:51.819684",
     "exception": false,
     "start_time": "2024-05-03T04:35:51.330876",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Training visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3cd44e25",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:35:52.682584Z",
     "iopub.status.busy": "2024-05-03T04:35:52.681736Z",
     "iopub.status.idle": "2024-05-03T04:35:53.363273Z",
     "shell.execute_reply": "2024-05-03T04:35:53.362302Z"
    },
    "papermill": {
     "duration": 1.114199,
     "end_time": "2024-05-03T04:35:53.365512",
     "exception": false,
     "start_time": "2024-05-03T04:35:52.251313",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sn\n",
    "\n",
    "\n",
    "fig, axs = plt.subplots(1, 2, figsize=(10, 4))\n",
    "\n",
    "axs[0].plot(train_logs.history['loss'], label='loss')\n",
    "axs[0].plot(train_logs.history['val_loss'], label='val_loss')\n",
    "axs[0].legend()\n",
    "axs[0].set_xlabel('Epochs')\n",
    "axs[0].set_ylabel('Score')\n",
    "axs[0].set_title('Loss Logs')\n",
    "\n",
    "axs[1].plot(train_logs.history['accuracy'], label='accuracy', linewidth=2)\n",
    "axs[1].plot(train_logs.history['val_accuracy'], label='val_accuracy', linewidth=2)\n",
    "axs[1].legend()\n",
    "axs[1].set_xlabel('Epochs')\n",
    "axs[1].set_ylabel('Score')\n",
    "axs[1].set_title('Accuracy Logs')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0efcd6ac",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:35:54.238515Z",
     "iopub.status.busy": "2024-05-03T04:35:54.238165Z",
     "iopub.status.idle": "2024-05-03T04:35:54.627300Z",
     "shell.execute_reply": "2024-05-03T04:35:54.626234Z"
    },
    "papermill": {
     "duration": 0.827274,
     "end_time": "2024-05-03T04:35:54.629876",
     "exception": false,
     "start_time": "2024-05-03T04:35:53.802602",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "best_model = load_model('/kaggle/working/deepfake.keras')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b55b2fd1",
   "metadata": {
    "papermill": {
     "duration": 0.448386,
     "end_time": "2024-05-03T04:35:55.565381",
     "exception": false,
     "start_time": "2024-05-03T04:35:55.116995",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Performance Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6a3f70c0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:35:56.461738Z",
     "iopub.status.busy": "2024-05-03T04:35:56.461236Z",
     "iopub.status.idle": "2024-05-03T04:35:58.499500Z",
     "shell.execute_reply": "2024-05-03T04:35:58.495631Z"
    },
    "papermill": {
     "duration": 2.489109,
     "end_time": "2024-05-03T04:35:58.502775",
     "exception": false,
     "start_time": "2024-05-03T04:35:56.013666",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m38/38\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step\n",
      "Accuracy Score: 0.99\n",
      "Recall Score: 0.98851\n",
      "Precision Score: 0.99176\n",
      "\n",
      "True positive =  602\n",
      "False positive =  7\n",
      "False negative =  5\n",
      "True negative =  586\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# checking metrics: Accuracy, Reacall, Precision\n",
    "from sklearn.metrics import precision_score, recall_score, accuracy_score, confusion_matrix\n",
    "pred = best_model.predict(X_val).round()\n",
    "\n",
    "predicted = [np.argmax(p) for p in pred]\n",
    "actual = [np.argmax(p) for p in Y_val]\n",
    "\n",
    "print(f\"Accuracy Score: {accuracy_score(predicted,actual).round(5)}\")\n",
    "print(f\"Recall Score: {recall_score(predicted,actual).round(5)}\")\n",
    "print(f\"Precision Score: {precision_score(predicted,actual).round(5)}\\n\")\n",
    "\n",
    "\n",
    "def print_confusion_matrix(y_true, y_pred):\n",
    "    cm=confusion_matrix(y_true,y_pred)\n",
    "    cm[0][0],cm[1][1]=cm[1][1],cm[0][0]\n",
    "    print('True positive = ', cm[0][0])\n",
    "    print('False positive = ', cm[0][1])\n",
    "    print('False negative = ', cm[1][0])\n",
    "    print('True negative = ', cm[1][1])\n",
    "    print('\\n')\n",
    "    cm=pd.DataFrame(cm,index=['True_predicted','False_predicted'],columns=['True_Actual','False_Actual'])\n",
    "\n",
    "    plt.subplots(figsize=(4,4))\n",
    "    sn.heatmap(cm,cmap='BuGn',annot=True, fmt='.0f',cbar=False)\n",
    "    plt.show()\n",
    "    \n",
    "print_confusion_matrix(actual,predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4f2ce0a3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:35:59.439034Z",
     "iopub.status.busy": "2024-05-03T04:35:59.438165Z",
     "iopub.status.idle": "2024-05-03T04:35:59.737591Z",
     "shell.execute_reply": "2024-05-03T04:35:59.736508Z"
    },
    "papermill": {
     "duration": 0.784697,
     "end_time": "2024-05-03T04:35:59.739776",
     "exception": false,
     "start_time": "2024-05-03T04:35:58.955079",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC curve\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "\n",
    "\n",
    "def plot_roc(actual,predicted):\n",
    "    # Calculate the ROC curve\n",
    "    fpr,tpr,thresholds = roc_curve(actual,predicted)\n",
    "    # Calculate the area under the ROC curve (AUC)\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    # Plot the ROC curve\n",
    "    plt.figure()\n",
    "    plt.plot(fpr, tpr, color='darkorange', lw=2, label='ROC curve (area = %0.2f)' % roc_auc)\n",
    "    plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "    plt.xlim([0.0, 1.0])\n",
    "    plt.ylim([0.0, 1.05])\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "    plt.legend(loc=\"lower right\")\n",
    "    plt.show()\n",
    "    \n",
    "\n",
    "plot_roc(actual,predicted)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5f8a32d9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:36:00.649320Z",
     "iopub.status.busy": "2024-05-03T04:36:00.648305Z",
     "iopub.status.idle": "2024-05-03T04:36:01.503955Z",
     "shell.execute_reply": "2024-05-03T04:36:01.503028Z"
    },
    "papermill": {
     "duration": 1.299119,
     "end_time": "2024-05-03T04:36:01.505920",
     "exception": false,
     "start_time": "2024-05-03T04:36:00.206801",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 128, 128, 3)\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 792ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# random image testing\n",
    "input_arr = img_to_array(load_img('/kaggle/input/ffhq-face-data-set/thumbnails128x128/00001.png', target_size=(128,128,3))) / 255.0\n",
    "input_arr = np.expand_dims(input_arr, axis=0)\n",
    "print(input_arr.shape)\n",
    "\n",
    "np.argmax(best_model.predict(input_arr).round()[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2610622",
   "metadata": {
    "papermill": {
     "duration": 0.431443,
     "end_time": "2024-05-03T04:36:02.385133",
     "exception": false,
     "start_time": "2024-05-03T04:36:01.953690",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c1aad154",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-03T04:36:03.312035Z",
     "iopub.status.busy": "2024-05-03T04:36:03.311131Z",
     "iopub.status.idle": "2024-05-03T04:36:03.508277Z",
     "shell.execute_reply": "2024-05-03T04:36:03.507492Z"
    },
    "papermill": {
     "duration": 0.634281,
     "end_time": "2024-05-03T04:36:03.510514",
     "exception": false,
     "start_time": "2024-05-03T04:36:02.876233",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "model.save('/kaggle/working/deepfake.h5')\n",
    "saved_model = tf.keras.models.load_model('/kaggle/working/deepfake.h5')"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "gpu",
   "dataSources": [
    {
     "datasetId": 379454,
     "sourceId": 735991,
     "sourceType": "datasetVersion"
    },
    {
     "datasetId": 1789402,
     "sourceId": 2919163,
     "sourceType": "datasetVersion"
    }
   ],
   "dockerImageVersionId": 30698,
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.10.13"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 499.589736,
   "end_time": "2024-05-03T04:36:06.460184",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-05-03T04:27:46.870448",
   "version": "2.5.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "1a1a271621ec4dfd8e543545cdef3ebe": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "1ae6cf195c584a748b6b5d26cfcf98b7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "1cdc58a6b39f4b609c4c9690c55f8132": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "1e9cf89d1fcc4653a5ca1b19c194d968": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "260cacb7114c4c5a9edd363f7c5d95f6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "2c0973d160854b43a4f8cb66574237ae": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "2c4709dd8dea46f3b54e11e350c96fca": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_1a1a271621ec4dfd8e543545cdef3ebe",
       "placeholder": "​",
       "style": "IPY_MODEL_e0941e9b7bc241c2b88e196ae11042e9",
       "value": "100%"
      }
     },
     "32ee2fcd79b64f9b88caa9d32ffb7165": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_d7aa8c0541194c169b17ab90a9ec59c0",
        "IPY_MODEL_e5b16ab07bd1480d9551d22f65e824ea",
        "IPY_MODEL_89acbdcb437a4e33ab6b0657d65431e1"
       ],
       "layout": "IPY_MODEL_646f34289be04c2bb35314257999c326"
      }
     },
     "34420adb61314f55aa0f8772895b7825": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "3d9cca1ca2b14372a3228f6ff5ff54dc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "4630a33f5fef433c928ced6a2c2593b0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "646f34289be04c2bb35314257999c326": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "6d46fbff4f454bae8a386e2890cd6aee": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "89acbdcb437a4e33ab6b0657d65431e1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_3d9cca1ca2b14372a3228f6ff5ff54dc",
       "placeholder": "​",
       "style": "IPY_MODEL_1e9cf89d1fcc4653a5ca1b19c194d968",
       "value": " 3000/3000 [00:17&lt;00:00, 176.67it/s]"
      }
     },
     "b5ee82844b614cf6aba62ca60616a3e5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_2c4709dd8dea46f3b54e11e350c96fca",
        "IPY_MODEL_d8242c797ca343dfaa4a555e04888193",
        "IPY_MODEL_bd789247e85a4d9d8cfb57d4b77e6c19"
       ],
       "layout": "IPY_MODEL_cb522dc1a40a40c4b1a8f8977c9ba767"
      }
     },
     "b712f7052e194c26be4b451d959eeba5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "bd789247e85a4d9d8cfb57d4b77e6c19": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_1ae6cf195c584a748b6b5d26cfcf98b7",
       "placeholder": "​",
       "style": "IPY_MODEL_b712f7052e194c26be4b451d959eeba5",
       "value": " 3000/3000 [02:07&lt;00:00, 22.67it/s]"
      }
     },
     "cb522dc1a40a40c4b1a8f8977c9ba767": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "d7aa8c0541194c169b17ab90a9ec59c0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_6d46fbff4f454bae8a386e2890cd6aee",
       "placeholder": "​",
       "style": "IPY_MODEL_4630a33f5fef433c928ced6a2c2593b0",
       "value": "100%"
      }
     },
     "d8242c797ca343dfaa4a555e04888193": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_2c0973d160854b43a4f8cb66574237ae",
       "max": 3000.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_34420adb61314f55aa0f8772895b7825",
       "value": 3000.0
      }
     },
     "e0941e9b7bc241c2b88e196ae11042e9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "e5b16ab07bd1480d9551d22f65e824ea": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_1cdc58a6b39f4b609c4c9690c55f8132",
       "max": 3000.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_260cacb7114c4c5a9edd363f7c5d95f6",
       "value": 3000.0
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}