SIPaKMed_GAN_Train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "94b7228833c20503",
   "metadata": {
    "jupyter": {
     "is_executing": true
    }
   },
   "outputs": [],
   "source": [
    "import cv2\n",
    "import os\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "\n",
    "def FormatDataset(dataset_src, dataset_dest, classes):\n",
    "    \n",
    "    new_cropped_dest = [os.path.join(dataset_dest, cls, 'CROPPED') for cls in classes]\n",
    "    new_complete_dest = [os.path.join(dataset_dest, cls, 'COMPLETE') for cls in classes]\n",
    "\n",
    "    \n",
    "    cropped_src = [os.path.join(dataset_src, f'im_{cls}', 'CROPPED') for cls in classes]\n",
    "    complete_src = [os.path.join(dataset_src, f'im_{cls}') for cls in classes]\n",
    "\n",
    "    for (dest1, dest2) in zip(new_cropped_dest, new_complete_dest):\n",
    "        os.makedirs(dest1, exist_ok=True)\n",
    "        os.makedirs(dest2, exist_ok=True)\n",
    "\n",
    "    \n",
    "    for (src, new_dest) in zip(cropped_src, new_cropped_dest):\n",
    "        if os.path.exists(src): \n",
    "                filename, file_ext = os.path.splitext(file)\n",
    "                if file_ext == '.bmp':\n",
    "                    img_des = os.path.join(new_dest, filename + '.jpg')\n",
    "                    img = cv2.imread(os.path.join(src, file))\n",
    "                    img = cv2.resize(img, (64, 64))\n",
    "                    img = cv2.copyMakeBorder(img, 1, 1, 1, 1, cv2.BORDER_CONSTANT, value=0)\n",
    "                    img = cv2.blur(img, (2, 2))\n",
    "                    cv2.imwrite(img_des, img)\n",
    "        else:\n",
    "            print(f\"Directory not found: {src}\")\n",
    "\n",
    "    \n",
    "    for (src, new_dest) in zip(complete_src, new_complete_dest):\n",
    "        if os.path.exists(src):  \n",
    "            for file in os.listdir(src):\n",
    "                filename, file_ext = os.path.splitext(file)\n",
    "                if file_ext == '.bmp':\n",
    "                    img_des = os.path.join(new_dest, filename + '.jpg')\n",
    "                    img = cv2.imread(os.path.join(src, file))\n",
    "                    img = cv2.resize(img, (256, 256))\n",
    "                    img = cv2.copyMakeBorder(img, 2, 2, 2, 2, cv2.BORDER_CONSTANT, value=0)\n",
    "                    img = cv2.blur(img, (2, 2))\n",
    "                    cv2.imwrite(img_des, img)\n",
    "        else:\n",
    "            print(f\"Directory not found: {src}\")\n",
    "\n",
    "src = '/home/iamsanko/Downloads/Test 1'\n",
    "dest = './CervicalCancer'\n",
    "classes = [\"Dyskeratotic\", \"Koilocytotic\", \"Metaplastic\", \"Parabasal\", \"Superficial-Intermediate\"]\n",
    "FormatDataset(src, dest, classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "fbd1dc0ddba698f5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.672093172Z",
     "start_time": "2024-09-27T13:05:04.543967Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Dyskeratotic': 34, 'Koilocytotic': 14, 'Metaplastic': 10, 'Parabasal': 7, 'Superficial-Intermediate': 126}\n"
     ]
    }
   ],
   "source": [
    "root_dir = \"./CervicalCancer\"\n",
    "classes = [\"Dyskeratotic\",\"Koilocytotic\",\"Metaplastic\",\"Parabasal\",\"Superficial-Intermediate\"]\n",
    "\n",
    "def GetDatasetSize(path, classes, main = \"CROPPED\"):\n",
    "    num_of_image = {}\n",
    "    for cls in classes:\n",
    "        num_of_image[cls] = len(os.listdir(os.path.join(path, cls, main)));\n",
    "    return num_of_image;\n",
    "\n",
    "print(GetDatasetSize(root_dir, classes, \"COMPLETE\"));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "980bccb99ec883c9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.735696461Z",
     "start_time": "2024-09-27T13:43:28.125666Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Contents of ./CervicalCancer/Dyskeratotic/CROPPED: ['004.jpg', '012.jpg', '029.jpg', '011.jpg', '005.jpg', '025.jpg', '009.jpg', '006.jpg', '016.jpg', '007.jpg', '033.jpg', '027.jpg', '034.jpg', '028.jpg', '019.jpg', '026.jpg', '010.jpg', '013.jpg', '031.jpg', '015.jpg', '017.jpg', '032.jpg', '014.jpg', '001.jpg', '024.jpg', '018.jpg', '030.jpg', '003.jpg', '022.jpg', '020.jpg', '008.jpg', '023.jpg', '002.jpg', '021.jpg']\n",
      "Total images found in ./CervicalCancer/Dyskeratotic/CROPPED: 34\n",
      "Dyskeratotic :\n",
      "Total images:  34\n",
      "Training:  23\n",
      "Validation:  5\n",
      "Testing:  6\n",
      "\n",
      "Contents of ./CervicalCancer/Koilocytotic/CROPPED: ['004.jpg', '012.jpg', '011.jpg', '005.jpg', '009.jpg', '006.jpg', '007.jpg', '010.jpg', '013.jpg', '014.jpg', '001.jpg', '003.jpg', '008.jpg', '002.jpg']\n",
      "Total images found in ./CervicalCancer/Koilocytotic/CROPPED: 14\n",
      "Koilocytotic :\n",
      "Total images:  14\n",
      "Training:  9\n",
      "Validation:  2\n",
      "Testing:  3\n",
      "\n",
      "Contents of ./CervicalCancer/Metaplastic/CROPPED: ['004.jpg', '005.jpg', '009.jpg', '006.jpg', '007.jpg', '010.jpg', '001.jpg', '003.jpg', '008.jpg', '002.jpg']\n",
      "Total images found in ./CervicalCancer/Metaplastic/CROPPED: 10\n",
      "Metaplastic :\n",
      "Total images:  10\n",
      "Training:  7\n",
      "Validation:  1\n",
      "Testing:  2\n",
      "\n",
      "Contents of ./CervicalCancer/Parabasal/CROPPED: ['004.jpg', '005.jpg', '006.jpg', '007.jpg', '001.jpg', '003.jpg', '002.jpg']\n",
      "Total images found in ./CervicalCancer/Parabasal/CROPPED: 7\n",
      "Parabasal :\n",
      "Total images:  7\n",
      "Training:  4\n",
      "Validation:  1\n",
      "Testing:  2\n",
      "\n",
      "Contents of ./CervicalCancer/Superficial-Intermediate/CROPPED: ['122_02.jpg', '010_04.jpg', '053_02.jpg', '057_09.jpg', '094_09.jpg', '046_06.jpg', '076_02.jpg', '023_04.jpg', '070_05.jpg', '044_02.jpg', '020_03.jpg', '107_04.jpg', '003_02.jpg', '125_02.jpg', '061_01.jpg', '069_01.jpg', '029_08.jpg', '025_01.jpg', '080_04.jpg', '094_04.jpg', '032_10.jpg', '113_05.jpg', '043_07.jpg', '121_02.jpg', '073_04.jpg', '125_04.jpg', '048_02.jpg', '050_01.jpg', '017_17.jpg', '083_01.jpg', '020_13.jpg', '014_07.jpg', '106_05.jpg', '035_02.jpg', '019_06.jpg', '017_02.jpg', '047_01.jpg', '089_01.jpg', '061_05.jpg', '115_01.jpg', '120_01.jpg', '126_05.jpg', '057_11.jpg', '056_02.jpg', '093_04.jpg', '112_03.jpg', '066_01.jpg', '060_04.jpg', '082_03.jpg', '002_05.jpg', '086_01.jpg', '010_02.jpg', '008_03.jpg', '106_11.jpg', '045_05.jpg', '122_06.jpg', '017_21.jpg', '057_08.jpg', '044_04.jpg', '023_08.jpg', '061_09.jpg', '083_10.jpg', '100_01.jpg', '090_02.jpg', '098_08.jpg', '101_04.jpg', '061_07.jpg', '049_01.jpg', '076_04.jpg', '005_03.jpg', '107_06.jpg', '062_06.jpg', '050_04.jpg', '007_01.jpg', '051_04.jpg', '116_05.jpg', '069_03.jpg', '045_06.jpg', '086_03.jpg', '032_08.jpg', '094_06.jpg', '032_01.jpg', '066_09.jpg', '056_04.jpg', '048_04.jpg', '072_02.jpg', '081_03.jpg', '104_02.jpg', '121_04.jpg', '082_04.jpg', '095_02.jpg', '036_03.jpg', '108_06.jpg', '102_02.jpg', '055_03.jpg', '020_10.jpg', '029_09.jpg', '064_02.jpg', '037_01.jpg', '099_09.jpg', '002_02.jpg', '125_03.jpg', '068_05.jpg', '057_04.jpg', '083_07.jpg', '073_06.jpg', '116_03.jpg', '102_08.jpg', '116_02.jpg', '059_04.jpg', '087_02.jpg', '104_05.jpg', '063_06.jpg', '058_01.jpg', '050_02.jpg', '106_10.jpg', '035_06.jpg', '010_07.jpg', '017_05.jpg', '056_07.jpg', '008_05.jpg', '067_10.jpg', '056_01.jpg', '075_05.jpg', '067_03.jpg', '034_05.jpg', '029_10.jpg', '023_01.jpg', '046_04.jpg', '013_04.jpg', '031_02.jpg', '068_06.jpg', '017_20.jpg', '004_11.jpg', '080_12.jpg', '086_02.jpg', '098_09.jpg', '122_04.jpg', '106_08.jpg', '023_11.jpg', '057_02.jpg', '012_01.jpg', '006_01.jpg', '069_06.jpg', '032_09.jpg', '035_05.jpg', '020_07.jpg', '028_05.jpg', '034_01.jpg', '106_06.jpg', '124_02.jpg', '045_07.jpg', '039_02.jpg', '078_07.jpg', '025_05.jpg', '081_04.jpg', '108_08.jpg', '121_06.jpg', '036_02.jpg', '118_03.jpg', '037_02.jpg', '012_04.jpg', '097_12.jpg', '080_03.jpg', '017_18.jpg', '068_08.jpg', '094_10.jpg', '008_02.jpg', '074_02.jpg', '063_05.jpg', '006_06.jpg', '113_02.jpg', '093_03.jpg', '032_07.jpg', '123_07.jpg', '061_10.jpg', '079_03.jpg', '071_07.jpg', '054_05.jpg', '016_03.jpg', '099_07.jpg', '058_02.jpg', '067_05.jpg', '109_02.jpg', '023_10.jpg', '056_05.jpg', '046_01.jpg', '018_02.jpg', '074_01.jpg', '017_11.jpg', '017_08.jpg', '019_03.jpg', '082_07.jpg', '126_06.jpg', '119_09.jpg', '104_04.jpg', '017_15.jpg', '017_03.jpg', '043_04.jpg', '118_06.jpg', '069_02.jpg', '082_06.jpg', '051_03.jpg', '001_03.jpg', '111_01.jpg', '075_03.jpg', '001_02.jpg', '015_02.jpg', '058_03.jpg', '062_02.jpg', '110_02.jpg', '104_06.jpg', '004_05.jpg', '080_09.jpg', '119_01.jpg', '116_01.jpg', '043_01.jpg', '106_04.jpg', '063_02.jpg', '023_02.jpg', '080_02.jpg', '094_01.jpg', '007_05.jpg', '017_04.jpg', '052_02.jpg', '017_14.jpg', '004_13.jpg', '081_02.jpg', '074_03.jpg', '020_06.jpg', '084_02.jpg', '041_03.jpg', '062_04.jpg', '096_09.jpg', '118_02.jpg', '034_03.jpg', '084_03.jpg', '066_08.jpg', '078_05.jpg', '126_01.jpg', '071_05.jpg', '024_01.jpg', '035_01.jpg', '075_01.jpg', '005_04.jpg', '051_01.jpg', '116_04.jpg', '098_04.jpg', '010_09.jpg', '111_03.jpg', '099_06.jpg', '064_03.jpg', '114_08.jpg', '103_01.jpg', '007_14.jpg', '095_03.jpg', '098_01.jpg', '054_01.jpg', '059_06.jpg', '102_06.jpg', '107_02.jpg', '096_07.jpg', '067_02.jpg', '080_06.jpg', '108_04.jpg', '085_03.jpg', '062_07.jpg', '101_01.jpg', '011_03.jpg', '028_08.jpg', '088_02.jpg', '071_06.jpg', '123_08.jpg', '006_02.jpg', '118_08.jpg', '118_01.jpg', '021_05.jpg', '031_03.jpg', '073_01.jpg', '026_01.jpg', '069_04.jpg', '004_02.jpg', '100_06.jpg', '039_01.jpg', '124_05.jpg', '063_04.jpg', '115_03.jpg', '045_11.jpg', '068_09.jpg', '032_06.jpg', '114_07.jpg', '118_09.jpg', '042_02.jpg', '070_02.jpg', '081_05.jpg', '024_03.jpg', '028_06.jpg', '105_02.jpg', '031_01.jpg', '099_02.jpg', '083_04.jpg', '097_03.jpg', '044_03.jpg', '007_02.jpg', '009_01.jpg', '010_08.jpg', '020_11.jpg', '032_05.jpg', '004_04.jpg', '122_05.jpg', '068_01.jpg', '097_08.jpg', '062_01.jpg', '077_01.jpg', '088_01.jpg', '014_09.jpg', '080_08.jpg', '107_03.jpg', '001_04.jpg', '023_06.jpg', '096_06.jpg', '019_08.jpg', '100_05.jpg', '060_05.jpg', '014_03.jpg', '115_04.jpg', '102_07.jpg', '043_05.jpg', '118_10.jpg', '020_01.jpg', '017_07.jpg', '122_09.jpg', '048_01.jpg', '072_04.jpg', '126_03.jpg', '066_02.jpg', '007_10.jpg', '071_03.jpg', '004_06.jpg', '013_02.jpg', '040_01.jpg', '010_06.jpg', '020_04.jpg', '126_04.jpg', '108_02.jpg', '018_05.jpg', '119_05.jpg', '020_05.jpg', '075_02.jpg', '009_04.jpg', '002_01.jpg', '032_03.jpg', '057_05.jpg', '018_06.jpg', '005_05.jpg', '017_22.jpg', '010_03.jpg', '083_06.jpg', '030_03.jpg', '092_01.jpg', '060_03.jpg', '052_01.jpg', '057_10.jpg', '007_13.jpg', '022_01.jpg', '097_02.jpg', '120_02.jpg', '083_09.jpg', '007_16.jpg', '007_11.jpg', '114_02.jpg', '048_03.jpg', '095_07.jpg', '036_04.jpg', '080_07.jpg', '065_03.jpg', '099_01.jpg', '092_02.jpg', '073_02.jpg', '037_05.jpg', '111_05.jpg', '097_06.jpg', '063_01.jpg', '106_03.jpg', '047_04.jpg', '001_05.jpg', '008_04.jpg', '021_01.jpg', '030_04.jpg', '100_07.jpg', '064_01.jpg', '010_10.jpg', '053_01.jpg', '101_03.jpg', '120_05.jpg', '019_07.jpg', '098_07.jpg', '017_12.jpg', '055_04.jpg', '098_02.jpg', '118_04.jpg', '041_01.jpg', '115_02.jpg', '003_01.jpg', '025_03.jpg', '119_02.jpg', '028_04.jpg', '108_03.jpg', '055_02.jpg', '100_04.jpg', '017_01.jpg', '094_07.jpg', '017_16.jpg', '102_04.jpg', '008_09.jpg', '067_04.jpg', '124_03.jpg', '008_01.jpg', '072_01.jpg', '061_06.jpg', '054_02.jpg', '029_07.jpg', '112_02.jpg', '014_08.jpg', '009_10.jpg', '126_02.jpg', '035_03.jpg', '004_07.jpg', '006_05.jpg', '106_02.jpg', '060_07.jpg', '078_02.jpg', '018_04.jpg', '013_01.jpg', '112_07.jpg', '045_08.jpg', '033_03.jpg', '115_05.jpg', '102_03.jpg', '025_04.jpg', '037_03.jpg', '030_02.jpg', '051_05.jpg', '038_01.jpg', '101_06.jpg', '102_01.jpg', '121_08.jpg', '011_01.jpg', '096_05.jpg', '124_07.jpg', '061_08.jpg', '018_01.jpg', '123_04.jpg', '027_05.jpg', '008_07.jpg', '055_05.jpg', '098_05.jpg', '119_04.jpg', '010_11.jpg', '059_07.jpg', '027_01.jpg', '005_01.jpg', '046_07.jpg', '106_13.jpg', '057_01.jpg', '036_01.jpg', '019_09.jpg', '008_06.jpg', '122_01.jpg', '117_03.jpg', '096_02.jpg', '112_01.jpg', '110_01.jpg', '058_07.jpg', '080_01.jpg', '123_06.jpg', '029_11.jpg', '045_09.jpg', '111_04.jpg', '099_04.jpg', '050_05.jpg', '015_01.jpg', '064_04.jpg', '120_03.jpg', '033_02.jpg', '104_01.jpg', '082_05.jpg', '016_02.jpg', '087_03.jpg', '030_05.jpg', '105_04.jpg', '019_01.jpg', '011_02.jpg', '040_03.jpg', '065_01.jpg', '119_10.jpg', '004_01.jpg', '073_05.jpg', '083_05.jpg', '084_01.jpg', '040_02.jpg', '098_03.jpg', '041_04.jpg', '071_01.jpg', '027_04.jpg', '059_02.jpg', '115_06.jpg', '101_05.jpg', '078_06.jpg', '106_12.jpg', '118_07.jpg', '117_01.jpg', '056_03.jpg', '079_01.jpg', '109_03.jpg', '023_05.jpg', '009_09.jpg', '066_11.jpg', '104_03.jpg', '097_09.jpg', '023_12.jpg', '059_05.jpg', '003_03.jpg', '062_05.jpg', '078_04.jpg', '093_02.jpg', '046_03.jpg', '009_05.jpg', '078_01.jpg', '095_05.jpg', '028_02.jpg', '032_02.jpg', '097_11.jpg', '122_03.jpg', '065_04.jpg', '045_04.jpg', '112_04.jpg', '009_03.jpg', '083_11.jpg', '067_06.jpg', '107_05.jpg', '013_05.jpg', '101_02.jpg', '016_01.jpg', '113_06.jpg', '028_07.jpg', '082_01.jpg', '075_04.jpg', '058_05.jpg', '061_02.jpg', '124_06.jpg', '026_02.jpg', '096_03.jpg', '047_03.jpg', '013_03.jpg', '046_09.jpg', '047_05.jpg', '122_08.jpg', '071_04.jpg', '045_01.jpg', '015_03.jpg', '099_03.jpg', '076_03.jpg', '026_05.jpg', '113_01.jpg', '112_06.jpg', '126_07.jpg', '021_02.jpg', '023_07.jpg', '034_04.jpg', '121_01.jpg', '027_07.jpg', '062_03.jpg', '069_05.jpg', '049_04.jpg', '072_03.jpg', '073_03.jpg', '002_06.jpg', '019_02.jpg', '125_01.jpg', '103_05.jpg', '029_02.jpg', '043_02.jpg', '029_05.jpg', '103_07.jpg', '067_08.jpg', '029_03.jpg', '045_03.jpg', '017_19.jpg', '058_04.jpg', '087_01.jpg', '124_04.jpg', '015_04.jpg', '091_01.jpg', '106_09.jpg', '050_06.jpg', '109_01.jpg', '111_02.jpg', '066_10.jpg', '037_04.jpg', '050_03.jpg', '007_06.jpg', '046_02.jpg', '077_02.jpg', '114_05.jpg', '053_03.jpg', '093_01.jpg', '004_08.jpg', '095_04.jpg', '064_05.jpg', '069_07.jpg', '114_04.jpg', '103_06.jpg', '086_04.jpg', '094_05.jpg', '046_08.jpg', '066_03.jpg', '025_02.jpg', '070_03.jpg', '085_01.jpg', '090_01.jpg', '112_08.jpg', '108_07.jpg', '004_09.jpg', '068_03.jpg', '123_02.jpg', '057_06.jpg', '102_05.jpg', '027_06.jpg', '067_07.jpg', '017_10.jpg', '103_02.jpg', '014_02.jpg', '108_05.jpg', '047_02.jpg', '053_05.jpg', '053_06.jpg', '080_05.jpg', '014_05.jpg', '083_02.jpg', '058_06.jpg', '065_02.jpg', '014_04.jpg', '056_06.jpg', '061_03.jpg', '070_04.jpg', '031_04.jpg', '007_08.jpg', '063_03.jpg', '068_02.jpg', '097_05.jpg', '105_01.jpg', '119_07.jpg', '068_04.jpg', '066_04.jpg', '123_01.jpg', '053_04.jpg', '046_05.jpg', '035_07.jpg', '036_05.jpg', '023_09.jpg', '076_05.jpg', '055_07.jpg', '007_12.jpg', '080_10.jpg', '094_03.jpg', '030_01.jpg', '071_02.jpg', '021_06.jpg', '059_03.jpg', '123_03.jpg', '006_04.jpg', '019_04.jpg', '004_12.jpg', '051_02.jpg', '014_06.jpg', '044_01.jpg', '050_07.jpg', '007_15.jpg', '072_05.jpg', '017_06.jpg', '097_04.jpg', '004_10.jpg', '049_02.jpg', '007_03.jpg', '069_10.jpg', '061_04.jpg', '017_13.jpg', '105_03.jpg', '057_07.jpg', '055_06.jpg', '021_07.jpg', '034_02.jpg', '114_03.jpg', '004_03.jpg', '049_03.jpg', '124_01.jpg', '078_03.jpg', '103_04.jpg', '009_02.jpg', '009_06.jpg', '006_03.jpg', '080_11.jpg', '113_03.jpg', '107_01.jpg', '021_08.jpg', '096_01.jpg', '008_08.jpg', '095_06.jpg', '125_05.jpg', '006_07.jpg', '055_01.jpg', '007_07.jpg', '042_03.jpg', '053_07.jpg', '114_06.jpg', '050_08.jpg', '010_05.jpg', '029_12.jpg', '094_08.jpg', '014_01.jpg', '083_08.jpg', '029_04.jpg', '121_07.jpg', '108_01.jpg', '099_05.jpg', '066_07.jpg', '020_08.jpg', '112_05.jpg', '018_03.jpg', '067_01.jpg', '029_06.jpg', '024_02.jpg', '012_02.jpg', '081_06.jpg', '021_04.jpg', '046_10.jpg', '028_01.jpg', '032_04.jpg', '023_03.jpg', '026_04.jpg', '026_03.jpg', '100_03.jpg', '069_09.jpg', '005_02.jpg', '095_01.jpg', '067_09.jpg', '070_01.jpg', '052_03.jpg', '010_01.jpg', '066_06.jpg', '083_03.jpg', '121_05.jpg', '042_01.jpg', '059_01.jpg', '123_05.jpg', '079_02.jpg', '012_03.jpg', '021_09.jpg', '042_04.jpg', '002_03.jpg', '002_04.jpg', '066_05.jpg', '060_06.jpg', '031_05.jpg', '094_02.jpg', '118_05.jpg', '041_02.jpg', '097_07.jpg', '109_04.jpg', '045_02.jpg', '076_01.jpg', '114_01.jpg', '045_10.jpg', '119_06.jpg', '117_04.jpg', '019_05.jpg', '021_03.jpg', '009_07.jpg', '119_08.jpg', '022_02.jpg', '054_04.jpg', '098_06.jpg', '107_07.jpg', '122_07.jpg', '043_03.jpg', '035_04.jpg', '057_03.jpg', '017_09.jpg', '106_01.jpg', '097_01.jpg', '060_01.jpg', '007_04.jpg', '103_03.jpg', '027_03.jpg', '085_02.jpg', '099_08.jpg', '096_04.jpg', '111_06.jpg', '020_02.jpg', '106_07.jpg', '009_08.jpg', '097_10.jpg', '007_09.jpg', '096_08.jpg', '060_02.jpg', '068_07.jpg', '020_09.jpg', '081_01.jpg', '028_03.jpg', '069_08.jpg', '033_01.jpg', '101_07.jpg', '121_03.jpg', '065_05.jpg', '117_02.jpg', '001_01.jpg', '029_01.jpg', '012_05.jpg', '016_04.jpg', '119_03.jpg', '120_04.jpg', '100_02.jpg', '020_12.jpg', '124_08.jpg', '043_06.jpg', '027_02.jpg', '054_03.jpg', '113_04.jpg', '082_02.jpg', '019_10.jpg']\n",
      "Total images found in ./CervicalCancer/Superficial-Intermediate/CROPPED: 831\n",
      "Superficial-Intermediate :\n",
      "Total images:  831\n",
      "Training:  581\n",
      "Validation:  125\n",
      "Testing:  125\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import shutil\n",
    "def TrainValTestSplit(root_dir, classes_dir, main=\"CROPPED\", val_ratio=0.15, test_ratio=0.15):\n",
    "    output_dir = \"/home/iamsanko/Downloads/Test 1\"\n",
    "\n",
    "    for cls in classes_dir:\n",
    "        src = os.path.join(root_dir, cls, main)\n",
    "\n",
    "        \n",
    "        if not os.path.exists(src):\n",
    "            print(f\"Warning: The folder {src} does not exist.\")\n",
    "            continue\n",
    "\n",
    "       \n",
    "        print(f\"Contents of {src}: {os.listdir(src)}\")\n",
    "\n",
    "       \n",
    "        allFileNames = os.listdir(src)\n",
    "        print(f\"Total images found in {src}: {len(allFileNames)}\")  # Debugging line\n",
    "\n",
    "        if len(allFileNames) == 0:\n",
    "            print(f\"No images found in {src}. Check the directory and file formats.\")\n",
    "            continue\n",
    "\n",
    "        \n",
    "        np.random.shuffle(allFileNames)\n",
    "        train_FileNames, val_FileNames, test_FileNames = np.split(\n",
    "            np.array(allFileNames),\n",
    "            [int(len(allFileNames) * (1 - (val_ratio + test_ratio))), int(len(allFileNames) * (1 - test_ratio))]\n",
    "        )\n",
    "\n",
    "        train_FileNames = [os.path.join(src, name) for name in train_FileNames.tolist()]\n",
    "        val_FileNames = [os.path.join(src, name) for name in val_FileNames.tolist()]\n",
    "        test_FileNames = [os.path.join(src, name) for name in test_FileNames.tolist()]\n",
    "\n",
    "      \n",
    "        os.makedirs(os.path.join(output_dir, 'train', cls), exist_ok=True)\n",
    "        os.makedirs(os.path.join(output_dir, 'val', cls), exist_ok=True)\n",
    "        os.makedirs(os.path.join(output_dir, 'test', cls), exist_ok=True)\n",
    "\n",
    "      \n",
    "        print(cls, ':')\n",
    "        print('Total images: ', len(allFileNames))\n",
    "        print('Training: ', len(train_FileNames))\n",
    "        print('Validation: ', len(val_FileNames))\n",
    "        print('Testing: ', len(test_FileNames))\n",
    "\n",
    "        for name in train_FileNames:\n",
    "            shutil.copy(name, os.path.join(output_dir, 'train', cls))\n",
    "\n",
    "        for name in val_FileNames:\n",
    "            shutil.copy(name, os.path.join(output_dir, 'val', cls))\n",
    "\n",
    "        for name in test_FileNames:\n",
    "            shutil.copy(name, os.path.join(output_dir, 'test', cls))\n",
    "        print()\n",
    "\n",
    "root_dir = \"./CervicalCancer\"  \n",
    "classes_dir = [\"Dyskeratotic\", \"Koilocytotic\", \"Metaplastic\", \"Parabasal\", \"Superficial-Intermediate\"]  # Classes\n",
    "TrainValTestSplit(root_dir, classes_dir)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "9fa058f1d34140ff",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.736707934Z",
     "start_time": "2024-09-27T13:43:29.725598Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Dyskeratotic': 34, 'Koilocytotic': 14, 'Metaplastic': 10, 'Parabasal': 7, 'Superficial-Intermediate': 126}\n"
     ]
    }
   ],
   "source": [
    "root_dir = \"./CervicalCancer\"\n",
    "classes = [\"Dyskeratotic\",\"Koilocytotic\",\"Metaplastic\",\"Parabasal\",\"Superficial-Intermediate\"]\n",
    "\n",
    "def GetDatasetSize(path, classes, main = \"CROPPED\"):\n",
    "    num_of_image = {}\n",
    "    for cls in classes:\n",
    "        \n",
    "        num_of_image[cls] = len(os.listdir(os.path.join(path, cls, main)));\n",
    "    return num_of_image;\n",
    "\n",
    "print(GetDatasetSize(root_dir, classes, \"COMPLETE\"));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "5b8702b17eee04b2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.737298561Z",
     "start_time": "2024-09-27T13:46:55.444508Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dyskeratotic :\n",
      "Total images:  34\n",
      "Training:  23\n",
      "Validation:  5\n",
      "Testing:  6\n",
      "\n",
      "Koilocytotic :\n",
      "Total images:  14\n",
      "Training:  9\n",
      "Validation:  2\n",
      "Testing:  3\n",
      "\n",
      "Metaplastic :\n",
      "Total images:  10\n",
      "Training:  7\n",
      "Validation:  1\n",
      "Testing:  2\n",
      "\n",
      "Parabasal :\n",
      "Total images:  7\n",
      "Training:  4\n",
      "Validation:  1\n",
      "Testing:  2\n",
      "\n",
      "Superficial-Intermediate :\n",
      "Total images:  831\n",
      "Training:  581\n",
      "Validation:  124\n",
      "Testing:  126\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import shutil\n",
    "\n",
    "def TrainValTestSplit(root_dir, classes_dir, main=\"CROPPED\", val_ratio=0.15, test_ratio=0.15, output_dir=''):\n",
    "    \n",
    "    os.makedirs(os.path.join(output_dir, 'train'), exist_ok=True)\n",
    "    os.makedirs(os.path.join(output_dir, 'val'), exist_ok=True)\n",
    "    os.makedirs(os.path.join(output_dir, 'test'), exist_ok=True)\n",
    "\n",
    "    for cls in classes_dir:\n",
    "        \n",
    "        class_train_dir = os.path.join(output_dir, 'train', cls)\n",
    "        class_val_dir = os.path.join(output_dir, 'val', cls)\n",
    "        class_test_dir = os.path.join(output_dir, 'test', cls)\n",
    "\n",
    "        os.makedirs(class_train_dir, exist_ok=True)\n",
    "        os.makedirs(class_val_dir, exist_ok=True)\n",
    "        os.makedirs(class_test_dir, exist_ok=True)\n",
    "\n",
    "        \n",
    "        src = os.path.join(root_dir, cls, main)\n",
    "\n",
    "        \n",
    "        if not os.path.exists(src):\n",
    "            print(f\"Source directory {src} does not exist. Skipping class: {cls}\")\n",
    "            continue\n",
    "\n",
    "        allFileNames = os.listdir(src)\n",
    "        np.random.shuffle(allFileNames)\n",
    "\n",
    "        \n",
    "        total_files = len(allFileNames)\n",
    "        train_size = int(total_files * (1 - (val_ratio + test_ratio)))\n",
    "        val_size = int(total_files * val_ratio)\n",
    "\n",
    "        train_FileNames, val_FileNames, test_FileNames = np.split(\n",
    "            np.array(allFileNames),\n",
    "            [train_size, train_size + val_size]\n",
    "        )\n",
    "\n",
    "        \n",
    "        train_FileNames = [os.path.join(src, name) for name in train_FileNames.tolist()]\n",
    "        val_FileNames = [os.path.join(src, name) for name in val_FileNames.tolist()]\n",
    "        test_FileNames = [os.path.join(src, name) for name in test_FileNames.tolist()]\n",
    "\n",
    "        \n",
    "        print('Total images: ', total_files)\n",
    "        print('Training: ', len(train_FileNames))\n",
    "        print('Validation: ', len(val_FileNames))\n",
    "        print('Testing: ', len(test_FileNames))\n",
    "\n",
    "        for name in train_FileNames:\n",
    "            shutil.copy(name, class_train_dir)\n",
    "\n",
    "        for name in val_FileNames:\n",
    "            shutil.copy(name, class_val_dir)\n",
    "\n",
    "        for name in test_FileNames:\n",
    "            shutil.copy(name, class_test_dir)\n",
    "\n",
    "        print()\n",
    "\n",
    "# Train / Validation / Test Split\n",
    "root_dir = \"./CervicalCancer\"  \n",
    "classes_dir = [\n",
    "    \"Dyskeratotic\",\n",
    "    \"Koilocytotic\",\n",
    "    \"Metaplastic\",\n",
    "    \"Parabasal\",\n",
    "    \"Superficial-Intermediate\"\n",
    "]  # Classes\n",
    "output_dir = './output'  \n",
    "TrainValTestSplit(root_dir, classes_dir, output_dir=output_dir)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cee21e5ce54a2d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "# Building the CNN model\n",
    "class VGG16Model(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(VGG16Model, self).__init__()\n",
    "        self.conv_layers = nn.Sequential(\n",
    "            nn.Conv2d(3, 16, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(16, 16, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "\n",
    "            nn.Conv2d(16, 32, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(32, 32, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "\n",
    "            nn.Conv2d(32, 64, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(64, 64, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(64, 64, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "\n",
    "            nn.Conv2d(64, 128, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(128, 128, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(128, 128, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "\n",
    "            nn.Conv2d(128, 256, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(256, 256, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(256, 256, kernel_size=3, padding='same'),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.MaxPool2d(kernel_size=2, stride=2),\n",
    "        )\n",
    "        self.fc_layers = nn.Sequential(\n",
    "            nn.Flatten(),\n",
    "            nn.Linear(256 * 2 * 2, 2048),  # taking input size 64x64\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Linear(2048, 2048),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Linear(2048, 5),\n",
    "            nn.Softmax(dim=1)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.conv_layers(x)\n",
    "        x = self.fc_layers(x)\n",
    "        return x\n",
    "\n",
    "\n",
    "model = VGG16Model()\n",
    "\n",
    "\n",
    "print(model)\n",
    "\n",
    "# The loss function and optimizer\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "af90d45eb78e1129",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.738195770Z",
     "start_time": "2024-09-27T13:55:28.791611Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torchvision.transforms as transforms\n",
    "from torchvision import datasets\n",
    "from torch.utils.data import DataLoader, random_split\n",
    "import numpy as np\n",
    "\n",
    "# The Generator\n",
    "class Generator(nn.Module):\n",
    "    def __init__(self, input_dim, output_dim):\n",
    "        super(Generator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(input_dim, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, output_dim),\n",
    "            nn.Tanh()\n",
    "        )\n",
    "\n",
    "    def forward(self, z):\n",
    "        return self.model(z)\n",
    "\n",
    "class Discriminator(nn.Module):\n",
    "    def __init__(self, input_dim):\n",
    "        super(Discriminator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(input_dim, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, 1),\n",
    "            nn.Sigmoid()\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.model(x)\n",
    "\n",
    "\n",
    "z_dim = 100  # Latent dimension\n",
    "data_dim = 64 * 64 * 3  # Image size for cells (64x64 RGB)\n",
    "generator = Generator(z_dim, data_dim)\n",
    "discriminator = Discriminator(data_dim)\n",
    "\n",
    "# Loss and optimizers\n",
    "criterion = nn.BCELoss()\n",
    "optimizer_G = optim.Adam(generator.parameters(), lr=0.0002, betas=(0.5, 0.999))\n",
    "optimizer_D = optim.Adam(discriminator.parameters(), lr=0.0002, betas=(0.5, 0.999))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "6f7a09b3f196e290",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.738669483Z",
     "start_time": "2024-09-27T13:55:48.403653Z"
    }
   },
   "outputs": [],
   "source": [
    "root_dir = \"./CervicalCancer\"  \n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),\n",
    "    transforms.ToTensor(),\n",
    "])\n",
    "\n",
    "dataset = datasets.ImageFolder(root=root_dir, transform=transform)\n",
    "train_size = int(0.6 * len(dataset))\n",
    "val_size = int(0.2 * len(dataset))\n",
    "test_size = len(dataset) - train_size - val_size\n",
    "\n",
    "train_dataset, val_dataset, test_dataset = random_split(dataset, [train_size, val_size, test_size])\n",
    "\n",
    "train_loader = DataLoader(train_dataset, batch_size=64, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=64, shuffle=False)\n",
    "test_loader = DataLoader(test_dataset, batch_size=64, shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "4e88d096f1ae6cb7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.739161831Z",
     "start_time": "2024-09-27T13:56:00.692456Z"
    }
   },
   "outputs": [],
   "source": [
    "def train_gan(num_epochs):\n",
    "    for epoch in range(num_epochs):\n",
    "        for i, (images, _) in enumerate(train_loader):\n",
    "            batch_size = images.size(0)\n",
    "            images = images.view(batch_size, -1)\n",
    "\n",
    "            real_labels = torch.ones(batch_size, 1)\n",
    "            fake_labels = torch.zeros(batch_size, 1)\n",
    "\n",
    "            # Train Discriminator\n",
    "            optimizer_D.zero_grad()\n",
    "            outputs = discriminator(images)\n",
    "            d_loss_real = criterion(outputs, real_labels)\n",
    "            d_loss_real.backward()\n",
    "\n",
    "            z = torch.randn(batch_size, z_dim)\n",
    "            fake_images = generator(z)\n",
    "            outputs = discriminator(fake_images.detach())\n",
    "            d_loss_fake = criterion(outputs, fake_labels)\n",
    "            d_loss_fake.backward()\n",
    "            optimizer_D.step()\n",
    "\n",
    "            # Train Generator\n",
    "            optimizer_G.zero_grad()\n",
    "            outputs = discriminator(fake_images)\n",
    "            g_loss = criterion(outputs, real_labels)\n",
    "            g_loss.backward()\n",
    "            optimizer_G.step()\n",
    "\n",
    "        print(f'Epoch [{epoch}/{num_epochs}], d_loss: {d_loss_real.item() + d_loss_fake.item()}, g_loss: {g_loss.item()}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "1f2dba7080b700b8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.739613529Z",
     "start_time": "2024-09-27T14:24:39.265891Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [0/5], d_loss: -0.0, g_loss: -0.0\n",
      "Epoch [1/5], d_loss: -0.0, g_loss: -0.0\n",
      "Epoch [2/5], d_loss: -0.0, g_loss: -0.0\n",
      "Epoch [3/5], d_loss: -0.0, g_loss: -0.0\n",
      "Epoch [4/5], d_loss: -0.0, g_loss: -0.0\n",
      "Epoch [0/5], Loss: 0.6597757339477539\n",
      "Epoch [1/5], Loss: 0.4483407735824585\n",
      "Epoch [2/5], Loss: 0.5041589140892029\n",
      "Epoch [3/5], Loss: 0.03810190036892891\n",
      "Epoch [4/5], Loss: 0.27435240149497986\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class CNNModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(CNNModel, self).__init__()\n",
    "        self.conv1 = nn.Conv2d(3, 16, kernel_size=3, padding='same')\n",
    "        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "        self.fc1 = nn.Linear(16 * 32 * 32, 256) \n",
    "        self.fc2 = nn.Linear(256, 5)  # 5 classes\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.pool(F.relu(self.conv1(x)))\n",
    "        x = x.view(-1, 16 * 32 * 32) \n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x\n",
    "\n",
    "#train the CNN model\n",
    "cnn_model = CNNModel()\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.Adam(cnn_model.parameters(), lr=0.001)\n",
    "\n",
    "def train_cnn(num_epochs):\n",
    "    for epoch in range(num_epochs):\n",
    "        cnn_model.train()\n",
    "        for images, labels in train_loader:\n",
    "            optimizer.zero_grad()\n",
    "            outputs = cnn_model(images)\n",
    "            loss = criterion(outputs, labels)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "        print(f'Epoch [{epoch}/{num_epochs}], Loss: {loss.item()}')\n",
    "\n",
    "# Train the GAN and CNN\n",
    "train_gan(num_epochs=5)\n",
    "train_cnn(num_epochs=5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "4041cfdd4b797858",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.740045334Z",
     "start_time": "2024-09-27T15:16:28.881467Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class CNNModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(CNNModel, self).__init__()\n",
    "\n",
    "        \n",
    "        self.conv1 = nn.Conv2d(3, 16, kernel_size=3, padding='same')\n",
    "        self.conv2 = nn.Conv2d(16, 16, kernel_size=3, padding='same')\n",
    "        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "\n",
    "        self.conv3 = nn.Conv2d(16, 32, kernel_size=3, padding='same')\n",
    "        self.conv4 = nn.Conv2d(32, 32, kernel_size=3, padding='same')\n",
    "        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "\n",
    "        self.conv5 = nn.Conv2d(32, 64, kernel_size=3, padding='same')\n",
    "        self.conv6 = nn.Conv2d(64, 64, kernel_size=3, padding='same')\n",
    "        self.conv7 = nn.Conv2d(64, 64, kernel_size=3, padding='same')\n",
    "        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "\n",
    "        self.conv8 = nn.Conv2d(64, 128, kernel_size=3, padding='same')\n",
    "        self.conv9 = nn.Conv2d(128, 128, kernel_size=3, padding='same')\n",
    "        self.conv10 = nn.Conv2d(128, 128, kernel_size=3, padding='same')\n",
    "        self.pool4 = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "\n",
    "        self.conv11 = nn.Conv2d(128, 256, kernel_size=3, padding='same')\n",
    "        self.conv12 = nn.Conv2d(256, 256, kernel_size=3, padding='same')\n",
    "        self.conv13 = nn.Conv2d(256, 256, kernel_size=3, padding='same')\n",
    "        self.pool5 = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "\n",
    "        \n",
    "        self.flat = nn.Flatten()\n",
    "        self.fc1 = nn.Linear(256 * 2 * 2, 2048)  \n",
    "        self.fc2 = nn.Linear(2048, 2048)\n",
    "        self.output = nn.Linear(2048, 5)  \n",
    "\n",
    "    def forward(self, x):\n",
    "        \n",
    "        x = F.relu(self.conv1(x))\n",
    "        x = F.relu(self.conv2(x))\n",
    "        x = self.pool1(x)\n",
    "\n",
    "        x = F.relu(self.conv3(x))\n",
    "        x = F.relu(self.conv4(x))\n",
    "        x = self.pool2(x)\n",
    "\n",
    "        x = F.relu(self.conv5(x))\n",
    "        x = F.relu(self.conv6(x))\n",
    "        x = F.relu(self.conv7(x))\n",
    "        x = self.pool3(x)\n",
    "\n",
    "        x = F.relu(self.conv8(x))\n",
    "        x = F.relu(self.conv9(x))\n",
    "        x = F.relu(self.conv10(x))\n",
    "        x = self.pool4(x)\n",
    "\n",
    "        x = F.relu(self.conv11(x))\n",
    "        x = F.relu(self.conv12(x))\n",
    "        x = F.relu(self.conv13(x))\n",
    "        x = self.pool5(x)\n",
    "\n",
    "        x = self.flat(x)\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = F.relu(self.fc2(x))\n",
    "        x = self.output(x)\n",
    "        return x\n",
    "\n",
    "cnn_model = CNNModel()\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.Adam(cnn_model.parameters(), lr=0.001)\n",
    "\n",
    "def train_cnn(num_epochs, train_loader):\n",
    "    for epoch in range(num_epochs):\n",
    "        cnn_model.train()\n",
    "        for images, labels in train_loader:\n",
    "            optimizer.zero_grad()\n",
    "            outputs = cnn_model(images)\n",
    "            loss = criterion(outputs, labels)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "        print(f'Epoch [{epoch}/{num_epochs}], Loss: {loss.item()}') ```````````````"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "57f4def738aaba28",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.740484496Z",
     "start_time": "2024-09-27T15:18:35.315720Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------------------------\n",
      "        Layer (type)               Output Shape         Param #\n",
      "================================================================\n",
      "            Conv2d-1           [-1, 16, 64, 64]             448\n",
      "            Conv2d-2           [-1, 32, 64, 64]           4,640\n",
      "         MaxPool2d-3           [-1, 32, 32, 32]               0\n",
      "            Conv2d-4           [-1, 64, 32, 32]          18,496\n",
      "         MaxPool2d-5           [-1, 64, 16, 16]               0\n",
      "            Conv2d-6          [-1, 128, 16, 16]          73,856\n",
      "         MaxPool2d-7            [-1, 128, 8, 8]               0\n",
      "           Dropout-8            [-1, 128, 8, 8]               0\n",
      "            Linear-9                   [-1, 64]         524,352\n",
      "          Dropout-10                   [-1, 64]               0\n",
      "           Linear-11                    [-1, 5]             325\n",
      "================================================================\n",
      "Total params: 622,117\n",
      "Trainable params: 622,117\n",
      "Non-trainable params: 0\n",
      "----------------------------------------------------------------\n",
      "Input size (MB): 0.05\n",
      "Forward/backward pass size (MB): 2.75\n",
      "Params size (MB): 2.37\n",
      "Estimated Total Size (MB): 5.17\n",
      "----------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class CNNModel(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(CNNModel, self).__init__()\n",
    "\n",
    "        # Convo layers for the CNN\n",
    "        self.conv1 = nn.Conv2d(3, 16, kernel_size=3, padding=1)  # Input shape: (64, 64, 3)\n",
    "        self.conv2 = nn.Conv2d(16, 32, kernel_size=3, padding=1)\n",
    "        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "        self.conv3 = nn.Conv2d(32, 64, kernel_size=3, padding=1)\n",
    "        self.conv4 = nn.Conv2d(64, 128, kernel_size=3, padding=1)\n",
    "        self.dropout = nn.Dropout(0.25)\n",
    "        self.flatten_size = 128 * 8 * 8  \n",
    "        self.fc1 = nn.Linear(self.flatten_size, 64)\n",
    "        self.fc2 = nn.Linear(64, 5)  \n",
    "\n",
    "    def forward(self, x):\n",
    "        # Convolutional layers with ReLU and MaxPooling\n",
    "        x = F.relu(self.conv1(x))\n",
    "        x = F.relu(self.conv2(x))\n",
    "        x = self.pool(x)  # Output size: (32, 32, 32)\n",
    "\n",
    "        x = F.relu(self.conv3(x))\n",
    "        x = self.pool(x)  # Output size: (64, 16, 16)\n",
    "\n",
    "        x = F.relu(self.conv4(x))\n",
    "        x = self.pool(x)  # Output size: (128, 8, 8)\n",
    "        x = self.dropout(x)\n",
    "        x = x.view(-1, self.flatten_size)\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = self.dropout(x)\n",
    "        x = torch.sigmoid(self.fc2(x))  # Sigmoid\n",
    "\n",
    "        return x\n",
    "cnn_model = CNNModel()\n",
    "\n",
    "\n",
    "optimizer = torch.optim.Adam(cnn_model.parameters(), lr=0.001)\n",
    "criterion = nn.BCELoss()  \n",
    "\n",
    "from torchsummary import summary\n",
    "summary(cnn_model, (3, 64, 64))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "b9ed314d280d3b57",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:10.740901599Z",
     "start_time": "2024-09-27T15:22:41.383445Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-27 20:52:52.287723: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:485] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2024-09-27 20:52:53.237043: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:8454] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2024-09-27 20:52:53.577877: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1452] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2024-09-27 20:52:55.584889: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2024-09-27 20:53:03.223044: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    },
    {
     "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: \"functional\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"functional\"\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>┃<span style=\"font-weight: bold\"> Connected to      </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer         │ (<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\">3</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,472</span> │ input_layer[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalization │ (<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\">256</span> │ conv2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation          │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_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\">16</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,160</span> │ max_pooling2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_1        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │ activation_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_2        │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">16,640</span> │ activation_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">16,640</span> │ max_pooling2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">1,024</span> │ conv2d_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">1,024</span> │ conv2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_3        │ (<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\">0</span> │ add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]         │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">16,448</span> │ activation_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_4        │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_6 (<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\">36,928</span> │ activation_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_5        │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_7 (<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\">16,640</span> │ activation_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">1,024</span> │ conv2d_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │ activation_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_6        │ (<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\">0</span> │ add_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_8 (<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\">16,448</span> │ activation_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_8[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_7        │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_9 (<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\">36,928</span> │ activation_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span> │ conv2d_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_8        │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_10 (<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\">16,640</span> │ activation_8[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">1,024</span> │ conv2d_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │ activation_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_9        │ (<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\">0</span> │ add_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_11 (<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\">128</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">32,896</span> │ activation_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_10       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_12 (<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\">128</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ activation_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_11       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_13 (<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\">512</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">66,048</span> │ activation_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_14 (<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\">512</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">131,584</span> │ activation_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_14[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│                     │                   │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_12       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_15 (<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\">128</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">65,664</span> │ activation_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_15[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_13       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_16 (<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\">128</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ activation_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_16[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_14       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_17 (<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\">512</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">66,048</span> │ activation_14[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_17[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_15       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_18 (<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\">128</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">65,664</span> │ activation_15[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_18[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_16       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_19 (<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\">128</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ activation_16[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_19[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_17       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_20 (<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\">512</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">66,048</span> │ activation_17[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_20[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_15[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_18       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_21 (<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\">128</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">65,664</span> │ activation_18[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_21[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_19       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_22 (<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\">128</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │ activation_19[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">128</span>) │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ conv2d_22[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_20       │ (<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\">128</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_23 (<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\">512</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">66,048</span> │ activation_20[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_23[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_18[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_21       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ add_6[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_24 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">131,328</span> │ activation_21[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_24[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_22       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_25 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_22[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_25[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_23       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_26 (<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\">263,168</span> │ activation_23[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_27 (<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\">525,312</span> │ activation_21[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_26[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_27[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_24       │ (<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\">0</span> │ add_7[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_28 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">262,400</span> │ activation_24[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_28[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_25       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_29 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_25[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_29[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_26       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_30 (<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\">263,168</span> │ activation_26[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_30[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ activation_24[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_27       │ (<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\">0</span> │ add_8[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_31 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">262,400</span> │ activation_27[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_31[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_28       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_32 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_28[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_32[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_29       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_33 (<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\">263,168</span> │ activation_29[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_33[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ activation_27[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_30       │ (<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\">0</span> │ add_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]       │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_34 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">262,400</span> │ activation_30[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_34[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_31       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_35 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_31[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_35[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_32       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_36 (<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\">263,168</span> │ activation_32[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_36[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_10 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ activation_30[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_33       │ (<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\">0</span> │ add_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_37 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">262,400</span> │ activation_33[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_37[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_34       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_38 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_34[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_38[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_35       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_39 (<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\">263,168</span> │ activation_35[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_39[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_11 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ activation_33[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_36       │ (<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\">0</span> │ add_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_40 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">262,400</span> │ activation_36[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_40[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_37       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_41 (<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\">256</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │ activation_37[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">256</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ conv2d_41[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_38       │ (<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\">256</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_42 (<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\">263,168</span> │ activation_38[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">4,096</span> │ conv2d_42[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_12 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │ activation_36[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_39       │ (<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\">0</span> │ add_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_43 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │    <span style=\"color: #00af00; text-decoration-color: #00af00\">524,800</span> │ activation_39[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_43[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_40       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_44 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │ activation_40[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_44[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_41       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_45 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,050,624</span> │ activation_41[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_46 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,099,200</span> │ activation_39[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">8,192</span> │ conv2d_45[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">8,192</span> │ conv2d_46[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_13 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_42       │ (<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\">0</span> │ add_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_47 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,049,088</span> │ activation_42[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_47[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_43       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_48 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │ activation_43[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_48[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_44       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_49 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,050,624</span> │ activation_44[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">8,192</span> │ conv2d_49[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_14 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │ activation_42[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_45       │ (<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\">0</span> │ add_14[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_50 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,049,088</span> │ activation_45[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_50[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_46       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_51 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>) │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,359,808</span> │ activation_46[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">512</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ conv2d_51[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_47       │ (<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\">512</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_52 (<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\">2</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2</span>,      │  <span style=\"color: #00af00; text-decoration-color: #00af00\">1,050,624</span> │ activation_47[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">8,192</span> │ conv2d_52[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_15 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</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\">0</span> │ batch_normalizat… │\n",
       "│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │ activation_45[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_48       │ (<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\">0</span> │ add_15[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (<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\">8,192</span> │ activation_48[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_49       │ (<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\">0</span> │ batch_normalizat… │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</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\">8192</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_49[<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\">64</span>)        │    <span style=\"color: #00af00; text-decoration-color: #00af00\">524,352</span> │ flatten[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">5</span>)         │        <span style=\"color: #00af00; text-decoration-color: #00af00\">325</span> │ dense[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</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┃\u001b[1m \u001b[0m\u001b[1mConnected to     \u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
       "│ input_layer         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m3\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
       "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d (\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;34m9,472\u001b[0m │ input_layer[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ max_pooling2d       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ activation[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
       "│ (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_1 (\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;34m4,160\u001b[0m │ max_pooling2d[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_1        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_2 (\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;34m36,928\u001b[0m │ activation_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_2        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\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;34m16,640\u001b[0m │ activation_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_4 (\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;34m16,640\u001b[0m │ max_pooling2d[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add (\u001b[38;5;33mAdd\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_3        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]         │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_5 (\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;34m16,448\u001b[0m │ activation_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_4        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_6 (\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;34m36,928\u001b[0m │ activation_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_5        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_7 (\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;34m16,640\u001b[0m │ activation_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_1 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │ activation_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_6        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ add_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_8 (\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;34m16,448\u001b[0m │ activation_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_7        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_9 (\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;34m36,928\u001b[0m │ activation_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │        \u001b[38;5;34m256\u001b[0m │ conv2d_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_8        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_10 (\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;34m16,640\u001b[0m │ activation_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_2 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m256\u001b[0m)              │            │ activation_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_9        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ add_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_11 (\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;34m128\u001b[0m) │     \u001b[38;5;34m32,896\u001b[0m │ activation_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_10       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_12 (\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;34m128\u001b[0m) │    \u001b[38;5;34m147,584\u001b[0m │ activation_10[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_11       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_13 (\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;34m512\u001b[0m) │     \u001b[38;5;34m66,048\u001b[0m │ activation_11[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_14 (\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;34m512\u001b[0m) │    \u001b[38;5;34m131,584\u001b[0m │ activation_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_14[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_3 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │                   │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_12       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ add_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_15 (\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;34m128\u001b[0m) │     \u001b[38;5;34m65,664\u001b[0m │ activation_12[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_15[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_13       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_16 (\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;34m128\u001b[0m) │    \u001b[38;5;34m147,584\u001b[0m │ activation_13[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_16[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_14       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_17 (\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;34m512\u001b[0m) │     \u001b[38;5;34m66,048\u001b[0m │ activation_14[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_17[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_4 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_12[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_15       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ add_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_18 (\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;34m128\u001b[0m) │     \u001b[38;5;34m65,664\u001b[0m │ activation_15[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_18[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_16       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_19 (\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;34m128\u001b[0m) │    \u001b[38;5;34m147,584\u001b[0m │ activation_16[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_19[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_17       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_20 (\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;34m512\u001b[0m) │     \u001b[38;5;34m66,048\u001b[0m │ activation_17[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_20[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_5 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_15[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_18       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ add_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_21 (\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;34m128\u001b[0m) │     \u001b[38;5;34m65,664\u001b[0m │ activation_18[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_21[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_19       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_22 (\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;34m128\u001b[0m) │    \u001b[38;5;34m147,584\u001b[0m │ activation_19[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │        \u001b[38;5;34m512\u001b[0m │ conv2d_22[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_20       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_23 (\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;34m512\u001b[0m) │     \u001b[38;5;34m66,048\u001b[0m │ activation_20[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_23[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_6 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │                   │            │ activation_18[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_21       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ add_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_24 (\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;34m256\u001b[0m) │    \u001b[38;5;34m131,328\u001b[0m │ activation_21[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_24[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_22       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_25 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_22[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_25[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_23       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_26 (\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;34m263,168\u001b[0m │ activation_23[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_27 (\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;34m525,312\u001b[0m │ activation_21[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_26[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_27[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_7 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_24       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_28 (\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;34m256\u001b[0m) │    \u001b[38;5;34m262,400\u001b[0m │ activation_24[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_28[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_25       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_29 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_25[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_29[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_26       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_30 (\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;34m263,168\u001b[0m │ activation_26[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_30[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_8 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ activation_24[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_27       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_31 (\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;34m256\u001b[0m) │    \u001b[38;5;34m262,400\u001b[0m │ activation_27[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_31[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_28       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_32 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_28[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_32[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_29       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_33 (\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;34m263,168\u001b[0m │ activation_29[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_33[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_9 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ activation_27[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_30       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_34 (\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;34m256\u001b[0m) │    \u001b[38;5;34m262,400\u001b[0m │ activation_30[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_34[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_31       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_35 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_31[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_35[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_32       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_36 (\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;34m263,168\u001b[0m │ activation_32[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_36[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_10 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ activation_30[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_33       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_37 (\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;34m256\u001b[0m) │    \u001b[38;5;34m262,400\u001b[0m │ activation_33[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_37[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_34       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_38 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_34[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_38[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_35       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_39 (\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;34m263,168\u001b[0m │ activation_35[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_39[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_11 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ activation_33[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_36       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_40 (\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;34m256\u001b[0m) │    \u001b[38;5;34m262,400\u001b[0m │ activation_36[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_40[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_37       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_41 (\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;34m256\u001b[0m) │    \u001b[38;5;34m590,080\u001b[0m │ activation_37[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │      \u001b[38;5;34m1,024\u001b[0m │ conv2d_41[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_38       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m256\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_42 (\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;34m263,168\u001b[0m │ activation_38[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ conv2d_42[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_12 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m1024\u001b[0m)             │            │ activation_36[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_39       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_12[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1024\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_43 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │    \u001b[38;5;34m524,800\u001b[0m │ activation_39[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_43[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_40       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_44 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │  \u001b[38;5;34m2,359,808\u001b[0m │ activation_40[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_44[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_41       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_45 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │  \u001b[38;5;34m1,050,624\u001b[0m │ activation_41[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_46 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │  \u001b[38;5;34m2,099,200\u001b[0m │ activation_39[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │      \u001b[38;5;34m8,192\u001b[0m │ conv2d_45[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │      \u001b[38;5;34m8,192\u001b[0m │ conv2d_46[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_13 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │ batch_normalizat… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_42       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_13[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_47 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │  \u001b[38;5;34m1,049,088\u001b[0m │ activation_42[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_47[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_43       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_48 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │  \u001b[38;5;34m2,359,808\u001b[0m │ activation_43[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_48[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_44       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_49 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │  \u001b[38;5;34m1,050,624\u001b[0m │ activation_44[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │      \u001b[38;5;34m8,192\u001b[0m │ conv2d_49[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_14 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │ activation_42[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_45       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_14[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_50 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │  \u001b[38;5;34m1,049,088\u001b[0m │ activation_45[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_50[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_46       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_51 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │  \u001b[38;5;34m2,359,808\u001b[0m │ activation_46[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │      \u001b[38;5;34m2,048\u001b[0m │ conv2d_51[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_47       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m512\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │                   │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ conv2d_52 (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │  \u001b[38;5;34m1,050,624\u001b[0m │ activation_47[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │      \u001b[38;5;34m8,192\u001b[0m │ conv2d_52[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ add_15 (\u001b[38;5;33mAdd\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│                     │ \u001b[38;5;34m2048\u001b[0m)             │            │ activation_45[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_48       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ add_15[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │      \u001b[38;5;34m8,192\u001b[0m │ activation_48[\u001b[38;5;34m0\u001b[0m]… │\n",
       "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ activation_49       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m, \u001b[38;5;34m2\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
       "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m2048\u001b[0m)             │            │                   │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8192\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ activation_49[\u001b[38;5;34m0\u001b[0m]… │\n",
       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │    \u001b[38;5;34m524,352\u001b[0m │ flatten[\u001b[38;5;34m0\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;34m5\u001b[0m)         │        \u001b[38;5;34m325\u001b[0m │ dense[\u001b[38;5;34m0\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\">24,120,581</span> (92.01 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m24,120,581\u001b[0m (92.01 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\">24,063,365</span> (91.79 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m24,063,365\u001b[0m (91.79 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\">57,216</span> (223.50 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m57,216\u001b[0m (223.50 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, BatchNormalization, Activation, Add, Flatten, Dense\n",
    "from tensorflow.keras.models import Model\n",
    "\n",
    "input_shape = (64, 64, 3)\n",
    "num_classes = 5\n",
    "\n",
    "def identity_block(X, filters):\n",
    "    F1, F2, F3 = filters\n",
    "\n",
    "    X_shortcut = X\n",
    "    X = Conv2D(F1, kernel_size=(1, 1), strides=(1, 1), padding='valid')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X = Activation('relu')(X)\n",
    "    X = Conv2D(F2, kernel_size=(3, 3), strides=(1, 1), padding='same')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X = Activation('relu')(X)\n",
    "    X = Conv2D(F3, kernel_size=(1, 1), strides=(1, 1), padding='valid')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X = Add()([X, X_shortcut])\n",
    "    X = Activation('relu')(X)\n",
    "\n",
    "    return X\n",
    "\n",
    "\n",
    "def convolutional_block(X, filters, stride=2):\n",
    "    F1, F2, F3 = filters\n",
    "\n",
    "    X_shortcut = X\n",
    "    X = Conv2D(F1, kernel_size=(1, 1), strides=(stride, stride), padding='valid')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X = Activation('relu')(X)\n",
    "    X = Conv2D(F2, kernel_size=(3, 3), strides=(1, 1), padding='same')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X = Activation('relu')(X)\n",
    "    X = Conv2D(F3, kernel_size=(1, 1), strides=(1, 1), padding='valid')(X)\n",
    "    X = BatchNormalization()(X)\n",
    "    X_shortcut = Conv2D(F3, kernel_size=(1, 1), strides=(stride, stride), padding='valid')(X_shortcut)\n",
    "    X_shortcut = BatchNormalization()(X_shortcut)\n",
    "    X = Add()([X, X_shortcut])\n",
    "    X = Activation('relu')(X)\n",
    "    return X\n",
    "\n",
    "\n",
    "input_tensor = Input(shape=input_shape)\n",
    "X = Conv2D(64, kernel_size=(7, 7), strides=(2, 2), padding='same')(input_tensor)\n",
    "X = BatchNormalization()(X)\n",
    "X = Activation('relu')(X)\n",
    "X = MaxPooling2D((3, 3), strides=(2, 2), padding='same')(X)\n",
    "X = convolutional_block(X, filters=[64, 64, 256], stride=1)\n",
    "X = identity_block(X, filters=[64, 64, 256])\n",
    "X = identity_block(X, filters=[64, 64, 256])\n",
    "X = convolutional_block(X, filters=[128, 128, 512])\n",
    "X = identity_block(X, filters=[128, 128, 512])\n",
    "X = identity_block(X, filters=[128, 128, 512])\n",
    "X = identity_block(X, filters=[128, 128, 512])\n",
    "# train diff\n",
    "X = convolutional_block(X, filters=[256, 256, 1024])\n",
    "X = identity_block(X, filters=[256, 256, 1024])\n",
    "X = identity_block(X, filters=[256, 256, 1024])\n",
    "X = identity_block(X, filters=[256, 256, 1024])\n",
    "X = identity_block(X, filters=[256, 256, 1024])\n",
    "X = identity_block(X, filters=[256, 256, 1024])\n",
    "#train diff 2\n",
    "X = convolutional_block(X, filters=[512, 512, 2048])\n",
    "X = identity_block(X, filters=[512, 512, 2048])\n",
    "X = identity_block(X, filters=[512, 512, 2048])\n",
    "# factor creator \n",
    "X = BatchNormalization()(X)\n",
    "X = Activation('relu')(X)\n",
    "X = Flatten()(X)\n",
    "X = Dense(64, activation='relu')(X)\n",
    "X = Dense(num_classes, activation='sigmoid')(X)\n",
    "model = Model(inputs=input_tensor, outputs=X)\n",
    "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b4c6e6e324cdca2e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.607012217Z",
     "start_time": "2024-09-28T04:52:19.542255Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n",
      "Downloading: \"https://download.pytorch.org/models/vgg16-397923af.pth\" to /home/iamsanko/.cache/torch/hub/checkpoints/vgg16-397923af.pth\n",
      "100%|██████████| 528M/528M [02:11<00:00, 4.20MB/s] \n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torchvision import models\n",
    "\n",
    "\n",
    "vgg16_model = models.vgg16(pretrained=True)\n",
    "num_classes = 5\n",
    "vgg16_model.classifier[6] = nn.Linear(in_features=4096, out_features=num_classes)\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.Adam(vgg16_model.parameters(), lr=0.001)\n",
    "def calculate_accuracy(output, target):\n",
    "    _, predicted = torch.max(output, 1)\n",
    "    correct = (predicted == target).sum().item()\n",
    "    total = target.size(0)\n",
    "    accuracy = correct / total\n",
    "    return accuracy\n",
    "\n",
    "\n",
    "def train_model(num_epochs, train_loader):\n",
    "    for epoch in range(num_epochs):\n",
    "        vgg16_model.train() \n",
    "        total_loss = 0\n",
    "        total_accuracy = 0\n",
    "        for images, labels in train_loader:\n",
    "            optimizer.zero_grad() \n",
    "\n",
    "            # Forward passer\n",
    "            outputs = vgg16_model(images)\n",
    "\n",
    "            # Loss calculation\n",
    "            loss = criterion(outputs, labels)\n",
    "            total_loss += loss.item()\n",
    "\n",
    "            # Backward passing \n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            accuracy = calculate_accuracy(outputs, labels)\n",
    "            total_accuracy += accuracy\n",
    "        avg_loss = total_loss / len(train_loader) \n",
    "        avg_accuracy = total_accuracy / len(train_loader)\n",
    "        print(f'Epoch [{epoch+1}/{num_epochs}], Loss: {avg_loss:.4f}, Accuracy: {avg_accuracy:.4f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "28ce694aa4549edd",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.607842129Z",
     "start_time": "2024-09-28T05:03:56.112209Z"
    }
   },
   "outputs": [],
   "source": [
    "import torchvision.transforms as transforms\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision.datasets import ImageFolder\n",
    "\n",
    "train_transforms = transforms.Compose([\n",
    "    transforms.RandomResizedCrop(64),   \n",
    "    transforms.RandomHorizontalFlip(),  \n",
    "    transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1),  \n",
    "    transforms.RandomAffine(degrees=0, shear=0.2),  \n",
    "    transforms.ToTensor(),               \n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])  \n",
    "])\n",
    "val_transforms = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),         \n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n",
    "])\n",
    "test_transforms = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),         g\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n",
    "])\n",
    "train_dataset = ImageFolder(root='/home/iamsanko/Downloads/Test 1/train', transform=train_transforms)\n",
    "val_dataset = ImageFolder(root='/home/iamsanko/Downloads/Test 1/val', transform=val_transforms)\n",
    "test_dataset = ImageFolder(root='/home/iamsanko/Downloads/Test 1/test', transform=test_transforms)\n",
    "train_loader = DataLoader(dataset=train_dataset, batch_size=32, shuffle=True)\n",
    "val_loader = DataLoader(dataset=val_dataset, batch_size=32, shuffle=False)\n",
    "test_loader = DataLoader(dataset=test_dataset, batch_size=32, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6039920cf23255cc",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.609366443Z",
     "start_time": "2024-09-28T05:05:18.429747Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1657 images belonging to 5 classes.\n"
     ]
    }
   ],
   "source": [
    "import torchvision.transforms as transforms\n",
    "from torchvision.datasets import ImageFolder\n",
    "from torch.utils.data import DataLoader\n",
    "train_transforms = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),         \n",
    "    transforms.RandomHorizontalFlip(),   \n",
    "    transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2), \n",
    "    transforms.RandomAffine(degrees=0, shear=0.2),  \n",
    "    transforms.ToTensor(),              \n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) \n",
    "])\n",
    "train_dataset = ImageFolder(root='./train', transform=train_transforms)\n",
    "print(f\"Found {len(train_dataset)} images belonging to {len(train_dataset.classes)} classes.\")\n",
    "train_loader = DataLoader(dataset=train_dataset, batch_size=100, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "baef4128e30504f5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.609865175Z",
     "start_time": "2024-09-28T05:07:38.549155Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Dyskeratotic': 0, 'Koilocytotic': 1, 'Metaplastic': 2, 'Parabasal': 3, 'Superficial-Intermediate': 4}\n"
     ]
    }
   ],
   "source": [
    "train_dataset = ImageFolder(root='./train', transform=train_transforms)\n",
    "print(train_dataset.class_to_idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "600bb969ecdfcb4f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.610301930Z",
     "start_time": "2024-09-28T05:13:05.448090Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 867 images belonging to 5 classes.\n",
      "{'Dyskeratotic': 0, 'Koilocytotic': 1, 'Metaplastic': 2, 'Parabasal': 3, 'Superficial-Intermediate': 4}\n"
     ]
    }
   ],
   "source": [
    "from torchvision.datasets import ImageFolder\n",
    "from torch.utils.data import DataLoader\n",
    "import os\n",
    "\n",
    "test_transforms = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),        \n",
    "    transforms.ToTensor(),             \n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])  \n",
    "])\n",
    "test_dataset = ImageFolder(root='./test', transform=test_transforms)\n",
    "test_loader = DataLoader(test_dataset, batch_size=100, shuffle=False)\n",
    "num_images = len(test_dataset)\n",
    "num_classes = len(test_dataset.classes)\n",
    "\n",
    "print(f\"Found {num_images} images belonging to {num_classes} classes.\")\n",
    "print(test_dataset.class_to_idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2f4abe66a9bf8582",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.610708577Z",
     "start_time": "2024-09-28T08:02:07.808699Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pytorch-lightning in /home/iamsanko/anaconda3/lib/python3.11/site-packages (2.4.0)\r\n",
      "Requirement already satisfied: torch>=2.1.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (2.4.1)\r\n",
      "Requirement already satisfied: tqdm>=4.57.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (4.65.0)\r\n",
      "Requirement already satisfied: PyYAML>=5.4 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (6.0)\r\n",
      "Requirement already satisfied: fsspec[http]>=2022.5.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (2023.4.0)\r\n",
      "Requirement already satisfied: torchmetrics>=0.7.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (1.4.2)\r\n",
      "Requirement already satisfied: packaging>=20.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (23.1)\r\n",
      "Requirement already satisfied: typing-extensions>=4.4.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (4.11.0)\r\n",
      "Requirement already satisfied: lightning-utilities>=0.10.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from pytorch-lightning) (0.11.7)\r\n",
      "Requirement already satisfied: requests in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from fsspec[http]>=2022.5.0->pytorch-lightning) (2.31.0)\r\n",
      "Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from fsspec[http]>=2022.5.0->pytorch-lightning) (3.8.5)\r\n",
      "Requirement already satisfied: setuptools in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from lightning-utilities>=0.10.0->pytorch-lightning) (68.0.0)\r\n",
      "Requirement already satisfied: filelock in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (3.9.0)\r\n",
      "Requirement already satisfied: sympy in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (1.11.1)\r\n",
      "Requirement already satisfied: networkx in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (3.1)\r\n",
      "Requirement already satisfied: jinja2 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (3.1.2)\r\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (9.1.0.70)\r\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.3.1)\r\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (11.0.2.54)\r\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (10.3.2.106)\r\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (11.4.5.107)\r\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.0.106)\r\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (2.20.5)\r\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (12.1.105)\r\n",
      "Requirement already satisfied: triton==3.0.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch>=2.1.0->pytorch-lightning) (3.0.0)\r\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>=2.1.0->pytorch-lightning) (12.6.68)\r\n",
      "Requirement already satisfied: numpy>1.20.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torchmetrics>=0.7.0->pytorch-lightning) (1.24.3)\r\n",
      "Requirement already satisfied: attrs>=17.3.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (22.1.0)\r\n",
      "Requirement already satisfied: charset-normalizer<4.0,>=2.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (2.0.4)\r\n",
      "Requirement already satisfied: multidict<7.0,>=4.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (6.0.2)\r\n",
      "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (4.0.2)\r\n",
      "Requirement already satisfied: yarl<2.0,>=1.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (1.8.1)\r\n",
      "Requirement already satisfied: frozenlist>=1.1.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (1.3.3)\r\n",
      "Requirement already satisfied: aiosignal>=1.1.2 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2022.5.0->pytorch-lightning) (1.2.0)\r\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from jinja2->torch>=2.1.0->pytorch-lightning) (2.1.1)\r\n",
      "Requirement already satisfied: idna<4,>=2.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from requests->fsspec[http]>=2022.5.0->pytorch-lightning) (3.4)\r\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from requests->fsspec[http]>=2022.5.0->pytorch-lightning) (1.26.16)\r\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from requests->fsspec[http]>=2022.5.0->pytorch-lightning) (2024.8.30)\r\n",
      "Requirement already satisfied: mpmath>=0.19 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from sympy->torch>=2.1.0->pytorch-lightning) (1.3.0)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install pytorch-lightning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "dce48ed38cdc924c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.611340655Z",
     "start_time": "2024-09-28T08:12:26.728969Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://download.pytorch.org/whl/cu118\r\n",
      "Requirement already satisfied: torch in /home/iamsanko/anaconda3/lib/python3.11/site-packages (2.4.1)\r\n",
      "Requirement already satisfied: torchvision in /home/iamsanko/anaconda3/lib/python3.11/site-packages (0.19.1)\r\n",
      "Requirement already satisfied: torchaudio in /home/iamsanko/anaconda3/lib/python3.11/site-packages (2.4.1+cu118)\r\n",
      "Requirement already satisfied: filelock in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.9.0)\r\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (4.11.0)\r\n",
      "Requirement already satisfied: sympy in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (1.11.1)\r\n",
      "Requirement already satisfied: networkx in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.1)\r\n",
      "Requirement already satisfied: jinja2 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.1.2)\r\n",
      "Requirement already satisfied: fsspec in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (2023.4.0)\r\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (9.1.0.70)\r\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.3.1)\r\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (11.0.2.54)\r\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (10.3.2.106)\r\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (11.4.5.107)\r\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.0.106)\r\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (2.20.5)\r\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: triton==3.0.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.0.0)\r\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch) (12.6.68)\r\n",
      "Requirement already satisfied: numpy in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torchvision) (1.24.3)\r\n",
      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torchvision) (9.4.0)\r\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from jinja2->torch) (2.1.1)\r\n",
      "Requirement already satisfied: mpmath>=0.19 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from sympy->torch) (1.3.0)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5710df4ac61ebc59",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T09:30:11.611830515Z",
     "start_time": "2024-09-28T08:12:37.161567Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu113\r\n",
      "Requirement already satisfied: torch in /home/iamsanko/anaconda3/lib/python3.11/site-packages (2.4.1)\r\n",
      "Requirement already satisfied: torchvision in /home/iamsanko/anaconda3/lib/python3.11/site-packages (0.19.1)\r\n",
      "Requirement already satisfied: torchaudio in /home/iamsanko/anaconda3/lib/python3.11/site-packages (2.4.1+cu118)\r\n",
      "Requirement already satisfied: filelock in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.9.0)\r\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (4.11.0)\r\n",
      "Requirement already satisfied: sympy in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (1.11.1)\r\n",
      "Requirement already satisfied: networkx in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.1)\r\n",
      "Requirement already satisfied: jinja2 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.1.2)\r\n",
      "Requirement already satisfied: fsspec in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (2023.4.0)\r\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (9.1.0.70)\r\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.3.1)\r\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (11.0.2.54)\r\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (10.3.2.106)\r\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (11.4.5.107)\r\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.0.106)\r\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (2.20.5)\r\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (12.1.105)\r\n",
      "Requirement already satisfied: triton==3.0.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torch) (3.0.0)\r\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch) (12.6.68)\r\n",
      "Requirement already satisfied: numpy in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torchvision) (1.24.3)\r\n",
      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from torchvision) (9.4.0)\r\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from jinja2->torch) (2.1.1)\r\n",
      "Requirement already satisfied: mpmath>=0.19 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from sympy->torch) (1.3.0)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1a5497a2bee8e956",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T12:35:41.380759Z",
     "start_time": "2024-09-28T12:35:15.901999Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:05:27.492660: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:485] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2024-09-28 18:05:28.282096: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:8454] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2024-09-28 18:05:28.501803: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1452] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2024-09-28 18:05:30.156155: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2024-09-28 18:05:36.222539: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "physical_devices = tf.config.list_physical_devices('GPU')\n",
    "if physical_devices:\n",
    "    tf.config.experimental.set_memory_growth(physical_devices[0], True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "95b23c1edfcf575c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T12:38:20.822546Z",
     "start_time": "2024-09-28T12:38:20.816357Z"
    }
   },
   "outputs": [],
   "source": [
    "class PyDataset(tf.keras.utils.Sequence):\n",
    "    def __init__(self, **kwargs):\n",
    "        super().__init__(**kwargs)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "eede6d55c48e7e9e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T12:39:19.424277Z",
     "start_time": "2024-09-28T12:39:11.929952Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:09:19.240887: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 102760448 exceeds 10% of free system memory.\n",
      "2024-09-28 18:09:19.281185: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 102760448 exceeds 10% of free system memory.\n",
      "2024-09-28 18:09:19.301670: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 102760448 exceeds 10% of free system memory.\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications import VGG16\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Dense, Flatten\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "\n",
    "base_model = VGG16(weights='imagenet', include_top=False, input_shape=(224, 224, 3))\n",
    "x = base_model.output\n",
    "x = Flatten()(x)  \n",
    "x = Dense(1024, activation='relu')(x)  \n",
    "predictions = Dense(5, activation='softmax')(x)\n",
    "vgg16_model = Model(inputs=base_model.input, outputs=predictions)\n",
    "for layer in base_model.layers:\n",
    "    layer.trainable = False\n",
    "vgg16_model.compile(\n",
    "    optimizer=Adam(learning_rate=0.0001), \n",
    "    loss='categorical_crossentropy',      \n",
    "    metrics=['accuracy']                  \n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8037f386d6fc959e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T12:40:13.305805Z",
     "start_time": "2024-09-28T12:40:13.298923Z"
    }
   },
   "outputs": [],
   "source": [
    "'''vgg16_model.compile(\n",
    "    optimizer='adam', \n",
    "    loss='categorical_crossentropy',  \n",
    "    metrics=['accuracy']\n",
    ")''' #test file \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cf5e719abe5665d3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-28T19:50:14.225733Z",
     "start_time": "2024-09-28T12:50:54.691299Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1657 images belonging to 5 classes.\n",
      "Found 877 images belonging to 5 classes.\n",
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:20:58.850608: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 102760448 exceeds 10% of free system memory.\n",
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/trainers/data_adapters/py_dataset_adapter.py:122: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n",
      "  self._warn_if_super_not_called()\n",
      "2024-09-28 18:21:00.226292: W external/local_tsl/tsl/framework/cpu_allocator_impl.cc:83] Allocation of 19267584 exceeds 10% of free system memory.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9579 - loss: 1.8113\n",
      "Epoch 1: val_accuracy improved from -inf to 0.93171, saving model to cervical_cancer_best_model_2.keras\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m520s\u001b[0m 10s/step - accuracy: 0.9581 - loss: 1.7952 - val_accuracy: 0.9317 - val_loss: 0.3559\n",
      "Epoch 2/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m4:41\u001b[0m 6s/step - accuracy: 0.9688 - loss: 0.1577"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:29:43.384791: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n",
      "/home/iamsanko/anaconda3/lib/python3.11/contextlib.py:155: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n",
      "  self.gen.throw(typ, value, traceback)\n",
      "2024-09-28 18:29:46.040882: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 2: val_accuracy did not improve from 0.93171\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 61ms/step - accuracy: 0.9688 - loss: 0.1577 - val_accuracy: 0.9231 - val_loss: 0.3952\n",
      "Epoch 3/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7s/step - accuracy: 0.9818 - loss: 0.1190\n",
      "Epoch 3: val_accuracy improved from 0.93171 to 0.97569, saving model to cervical_cancer_best_model_2.keras\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m519s\u001b[0m 10s/step - accuracy: 0.9819 - loss: 0.1187 - val_accuracy: 0.9757 - val_loss: 0.0821\n",
      "Epoch 4/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m4:36\u001b[0m 6s/step - accuracy: 0.9688 - loss: 0.0766\n",
      "Epoch 4: val_accuracy did not improve from 0.97569\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 51ms/step - accuracy: 0.9688 - loss: 0.0766 - val_accuracy: 0.9231 - val_loss: 0.2836\n",
      "Epoch 5/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:38:32.896763: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9873 - loss: 0.0603\n",
      "Epoch 5: val_accuracy improved from 0.97569 to 0.98032, saving model to cervical_cancer_best_model_2.keras\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m502s\u001b[0m 10s/step - accuracy: 0.9873 - loss: 0.0604 - val_accuracy: 0.9803 - val_loss: 0.0563\n",
      "Epoch 6/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m4:40\u001b[0m 6s/step - accuracy: 1.0000 - loss: 0.0051\n",
      "Epoch 6: val_accuracy improved from 0.98032 to 1.00000, saving model to cervical_cancer_best_model_2.keras\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 211ms/step - accuracy: 1.0000 - loss: 0.0051 - val_accuracy: 1.0000 - val_loss: 0.0111\n",
      "Epoch 7/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9919 - loss: 0.0295\n",
      "Epoch 7: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9919 - loss: 0.0295 - val_accuracy: 0.9884 - val_loss: 0.0400\n",
      "Epoch 8/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.4901e-08\n",
      "Epoch 8: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 1.4901e-08 - val_accuracy: 1.0000 - val_loss: 2.0180e-05\n",
      "Epoch 9/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 18:55:31.012616: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9934 - loss: 0.0185\n",
      "Epoch 9: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m489s\u001b[0m 10s/step - accuracy: 0.9934 - loss: 0.0186 - val_accuracy: 0.9792 - val_loss: 0.0612\n",
      "Epoch 10/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:26\u001b[0m 7s/step - accuracy: 0.9688 - loss: 0.0494\n",
      "Epoch 10: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 0.9688 - loss: 0.0494 - val_accuracy: 1.0000 - val_loss: 0.0000e+00\n",
      "Epoch 11/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9919 - loss: 0.0276\n",
      "Epoch 11: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9918 - loss: 0.0281 - val_accuracy: 0.9618 - val_loss: 0.3703\n",
      "Epoch 12/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:13\u001b[0m 6s/step - accuracy: 0.9688 - loss: 0.0792\n",
      "Epoch 12: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 48ms/step - accuracy: 0.9688 - loss: 0.0792 - val_accuracy: 1.0000 - val_loss: 0.0197\n",
      "Epoch 13/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9861 - loss: 0.0868\n",
      "Epoch 13: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m490s\u001b[0m 10s/step - accuracy: 0.9862 - loss: 0.0863 - val_accuracy: 0.9769 - val_loss: 0.1298\n",
      "Epoch 14/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.3245e-06\n",
      "Epoch 14: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 1.0000 - loss: 2.3245e-06 - val_accuracy: 1.0000 - val_loss: 2.0632e-06\n",
      "Epoch 15/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9924 - loss: 0.0267\n",
      "Epoch 15: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 0.9924 - loss: 0.0268 - val_accuracy: 0.9965 - val_loss: 0.0149\n",
      "Epoch 16/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:32\u001b[0m 7s/step - accuracy: 1.0000 - loss: 2.6077e-08\n",
      "Epoch 16: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 2.6077e-08 - val_accuracy: 1.0000 - val_loss: 0.0000e+00\n",
      "Epoch 17/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 19:28:49.235316: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9983 - loss: 0.0067\n",
      "Epoch 17: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9983 - loss: 0.0068 - val_accuracy: 1.0000 - val_loss: 0.0018\n",
      "Epoch 18/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 0.0040\n",
      "Epoch 18: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 0.0040 - val_accuracy: 1.0000 - val_loss: 4.9425e-06\n",
      "Epoch 19/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9984 - loss: 0.0050\n",
      "Epoch 19: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9983 - loss: 0.0053 - val_accuracy: 0.9664 - val_loss: 0.0889\n",
      "Epoch 20/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.0107e-05\n",
      "Epoch 20: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 5.0107e-05 - val_accuracy: 0.8462 - val_loss: 0.7476\n",
      "Epoch 21/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9919 - loss: 0.0168\n",
      "Epoch 21: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9920 - loss: 0.0167 - val_accuracy: 0.9988 - val_loss: 0.0036\n",
      "Epoch 22/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.1995e-05\n",
      "Epoch 22: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 1.0000 - loss: 1.1995e-05 - val_accuracy: 1.0000 - val_loss: 1.8850e-04\n",
      "Epoch 23/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7s/step - accuracy: 0.9972 - loss: 0.0185\n",
      "Epoch 23: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m499s\u001b[0m 10s/step - accuracy: 0.9972 - loss: 0.0184 - val_accuracy: 1.0000 - val_loss: 3.3409e-04\n",
      "Epoch 24/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.7471e-05\n",
      "Epoch 24: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 3.7471e-05 - val_accuracy: 1.0000 - val_loss: 2.0412e-05\n",
      "Epoch 25/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.8960e-04\n",
      "Epoch 25: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 4.0212e-04 - val_accuracy: 0.9965 - val_loss: 0.0080\n",
      "Epoch 26/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.2070e-06\n",
      "Epoch 26: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 53ms/step - accuracy: 1.0000 - loss: 1.2070e-06 - val_accuracy: 1.0000 - val_loss: 3.3195e-06\n",
      "Epoch 27/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9983 - loss: 0.0036\n",
      "Epoch 27: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m490s\u001b[0m 10s/step - accuracy: 0.9983 - loss: 0.0037 - val_accuracy: 1.0000 - val_loss: 7.2982e-04\n",
      "Epoch 28/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.7636e-04\n",
      "Epoch 28: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 55ms/step - accuracy: 1.0000 - loss: 1.7636e-04 - val_accuracy: 1.0000 - val_loss: 1.2838e-07\n",
      "Epoch 29/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9987 - loss: 0.0087\n",
      "Epoch 29: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9987 - loss: 0.0087 - val_accuracy: 1.0000 - val_loss: 3.2931e-04\n",
      "Epoch 30/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.7508e-04\n",
      "Epoch 30: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 56ms/step - accuracy: 1.0000 - loss: 2.7508e-04 - val_accuracy: 1.0000 - val_loss: 1.0371e-05\n",
      "Epoch 31/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.9256e-04\n",
      "Epoch 31: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m490s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 3.8963e-04 - val_accuracy: 1.0000 - val_loss: 1.7546e-04\n",
      "Epoch 32/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.1781e-07\n",
      "Epoch 32: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 54ms/step - accuracy: 1.0000 - loss: 5.1781e-07 - val_accuracy: 1.0000 - val_loss: 9.6921e-06\n",
      "Epoch 33/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 20:35:41.599536: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 7.0752e-05\n",
      "Epoch 33: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m490s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 7.0325e-05 - val_accuracy: 1.0000 - val_loss: 1.5394e-04\n",
      "Epoch 34/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 0.0152\n",
      "Epoch 34: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 60ms/step - accuracy: 1.0000 - loss: 0.0152 - val_accuracy: 1.0000 - val_loss: 6.3273e-07\n",
      "Epoch 35/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9980 - loss: 0.0040\n",
      "Epoch 35: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m490s\u001b[0m 10s/step - accuracy: 0.9980 - loss: 0.0040 - val_accuracy: 0.9977 - val_loss: 0.0046\n",
      "Epoch 36/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 9.4620e-06\n",
      "Epoch 36: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 9.4620e-06 - val_accuracy: 1.0000 - val_loss: 2.8427e-06\n",
      "Epoch 37/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9977 - loss: 0.0051\n",
      "Epoch 37: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9977 - loss: 0.0051 - val_accuracy: 1.0000 - val_loss: 0.0012\n",
      "Epoch 38/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 0.0026\n",
      "Epoch 38: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 54ms/step - accuracy: 1.0000 - loss: 0.0026 - val_accuracy: 1.0000 - val_loss: 2.8406e-05\n",
      "Epoch 39/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9955 - loss: 0.0171\n",
      "Epoch 39: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9954 - loss: 0.0172 - val_accuracy: 1.0000 - val_loss: 1.7734e-04\n",
      "Epoch 40/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 0.9688 - loss: 0.0337\n",
      "Epoch 40: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 53ms/step - accuracy: 0.9688 - loss: 0.0337 - val_accuracy: 1.0000 - val_loss: 2.1749e-04\n",
      "Epoch 41/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9901 - loss: 0.0376\n",
      "Epoch 41: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9900 - loss: 0.0379 - val_accuracy: 0.9780 - val_loss: 0.1101\n",
      "Epoch 42/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.1548e-07\n",
      "Epoch 42: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 1.1548e-07 - val_accuracy: 1.0000 - val_loss: 1.7056e-06\n",
      "Epoch 43/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9945 - loss: 0.0324\n",
      "Epoch 43: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m491s\u001b[0m 10s/step - accuracy: 0.9945 - loss: 0.0321 - val_accuracy: 0.9838 - val_loss: 0.0504\n",
      "Epoch 44/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.2339e-06\n",
      "Epoch 44: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 5.2339e-06 - val_accuracy: 1.0000 - val_loss: 9.3487e-05\n",
      "Epoch 45/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9991 - loss: 0.0013\n",
      "Epoch 45: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9991 - loss: 0.0013 - val_accuracy: 1.0000 - val_loss: 6.1659e-04\n",
      "Epoch 46/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.7546e-06\n",
      "Epoch 46: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 57ms/step - accuracy: 1.0000 - loss: 1.7546e-06 - val_accuracy: 1.0000 - val_loss: 1.2416e-05\n",
      "Epoch 47/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.9814e-04\n",
      "Epoch 47: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m495s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 5.9550e-04 - val_accuracy: 0.9954 - val_loss: 0.0085\n",
      "Epoch 48/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.3155e-07\n",
      "Epoch 48: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 3.3155e-07 - val_accuracy: 1.0000 - val_loss: 3.8880e-06\n",
      "Epoch 49/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9967 - loss: 0.0079\n",
      "Epoch 49: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m495s\u001b[0m 10s/step - accuracy: 0.9967 - loss: 0.0080 - val_accuracy: 1.0000 - val_loss: 4.1958e-04\n",
      "Epoch 50/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.0547e-07\n",
      "Epoch 50: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 3.0547e-07 - val_accuracy: 1.0000 - val_loss: 2.9619e-06\n",
      "Epoch 51/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9983 - loss: 0.0055\n",
      "Epoch 51: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m495s\u001b[0m 10s/step - accuracy: 0.9983 - loss: 0.0056 - val_accuracy: 0.9838 - val_loss: 0.0739\n",
      "Epoch 52/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.4901e-06\n",
      "Epoch 52: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 1.4901e-06 - val_accuracy: 1.0000 - val_loss: 2.0705e-05\n",
      "Epoch 53/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9931 - loss: 0.0339\n",
      "Epoch 53: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m495s\u001b[0m 10s/step - accuracy: 0.9931 - loss: 0.0339 - val_accuracy: 1.0000 - val_loss: 0.0011\n",
      "Epoch 54/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.9321e-05\n",
      "Epoch 54: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 55ms/step - accuracy: 1.0000 - loss: 5.9321e-05 - val_accuracy: 1.0000 - val_loss: 4.9355e-05\n",
      "Epoch 55/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9987 - loss: 0.0037\n",
      "Epoch 55: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 0.9987 - loss: 0.0038 - val_accuracy: 1.0000 - val_loss: 0.0012\n",
      "Epoch 56/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.3314e-05\n",
      "Epoch 56: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 1.3314e-05 - val_accuracy: 1.0000 - val_loss: 1.5327e-04\n",
      "Epoch 57/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9981 - loss: 0.0030\n",
      "Epoch 57: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 0.9981 - loss: 0.0030 - val_accuracy: 0.9965 - val_loss: 0.0090\n",
      "Epoch 58/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.6508e-07\n",
      "Epoch 58: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 3.6508e-07 - val_accuracy: 1.0000 - val_loss: 7.7027e-07\n",
      "Epoch 59/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9970 - loss: 0.0136\n",
      "Epoch 59: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 0.9970 - loss: 0.0136 - val_accuracy: 1.0000 - val_loss: 6.4049e-04\n",
      "Epoch 60/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.3709e-06\n",
      "Epoch 60: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 1.3709e-06 - val_accuracy: 1.0000 - val_loss: 5.9191e-05\n",
      "Epoch 61/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9991 - loss: 0.0058\n",
      "Epoch 61: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 0.9990 - loss: 0.0058 - val_accuracy: 0.9977 - val_loss: 0.0072\n",
      "Epoch 62/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:21\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.0489e-07\n",
      "Epoch 62: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 2.0489e-07 - val_accuracy: 1.0000 - val_loss: 1.7496e-05\n",
      "Epoch 63/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9991 - loss: 0.0025\n",
      "Epoch 63: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m495s\u001b[0m 10s/step - accuracy: 0.9991 - loss: 0.0025 - val_accuracy: 0.9815 - val_loss: 0.0536\n",
      "Epoch 64/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:18\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.0115e-05\n",
      "Epoch 64: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 3.0115e-05 - val_accuracy: 1.0000 - val_loss: 1.4305e-05\n",
      "Epoch 65/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-09-28 22:49:32.668399: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence\n",
      "\t [[{{node IteratorGetNext}}]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9911 - loss: 0.0522\n",
      "Epoch 65: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m497s\u001b[0m 10s/step - accuracy: 0.9911 - loss: 0.0519 - val_accuracy: 1.0000 - val_loss: 0.0044\n",
      "Epoch 66/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.9200e-04\n",
      "Epoch 66: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 1.9200e-04 - val_accuracy: 1.0000 - val_loss: 6.8925e-04\n",
      "Epoch 67/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9978 - loss: 0.0029\n",
      "Epoch 67: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9978 - loss: 0.0029 - val_accuracy: 1.0000 - val_loss: 3.4799e-04\n",
      "Epoch 68/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 8.3777e-05\n",
      "Epoch 68: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 8.3777e-05 - val_accuracy: 1.0000 - val_loss: 1.4635e-04\n",
      "Epoch 69/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9992 - loss: 0.0028\n",
      "Epoch 69: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9992 - loss: 0.0028 - val_accuracy: 1.0000 - val_loss: 3.1516e-04\n",
      "Epoch 70/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 6.2154e-05\n",
      "Epoch 70: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 6.2154e-05 - val_accuracy: 1.0000 - val_loss: 1.0087e-06\n",
      "Epoch 71/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9929 - loss: 0.0289\n",
      "Epoch 71: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 0.9930 - loss: 0.0286 - val_accuracy: 0.9954 - val_loss: 0.0100\n",
      "Epoch 72/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 4.8164e-06\n",
      "Epoch 72: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 1.0000 - loss: 4.8164e-06 - val_accuracy: 1.0000 - val_loss: 0.0000e+00\n",
      "Epoch 73/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 0.0014\n",
      "Epoch 73: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 0.0014 - val_accuracy: 1.0000 - val_loss: 2.7939e-04\n",
      "Epoch 74/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.0936e-06\n",
      "Epoch 74: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 2.0936e-06 - val_accuracy: 1.0000 - val_loss: 7.9573e-05\n",
      "Epoch 75/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 4.4669e-05\n",
      "Epoch 75: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 4.7399e-05 - val_accuracy: 1.0000 - val_loss: 1.4848e-04\n",
      "Epoch 76/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.5354e-04\n",
      "Epoch 76: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 1.5354e-04 - val_accuracy: 1.0000 - val_loss: 2.1107e-05\n",
      "Epoch 77/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9994 - loss: 0.0010\n",
      "Epoch 77: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9994 - loss: 0.0010 - val_accuracy: 1.0000 - val_loss: 6.6910e-04\n",
      "Epoch 78/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 7.4506e-09\n",
      "Epoch 78: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 55ms/step - accuracy: 1.0000 - loss: 7.4506e-09 - val_accuracy: 1.0000 - val_loss: 8.3810e-06\n",
      "Epoch 79/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 0.9996 - loss: 4.6275e-04\n",
      "Epoch 79: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 0.9996 - loss: 4.6752e-04 - val_accuracy: 1.0000 - val_loss: 3.6612e-04\n",
      "Epoch 80/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 8.2700e-05\n",
      "Epoch 80: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 8.2700e-05 - val_accuracy: 1.0000 - val_loss: 2.7847e-05\n",
      "Epoch 81/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 6.3479e-05\n",
      "Epoch 81: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 6.3583e-05 - val_accuracy: 1.0000 - val_loss: 1.1989e-04\n",
      "Epoch 82/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.3882e-05\n",
      "Epoch 82: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 3.3882e-05 - val_accuracy: 1.0000 - val_loss: 1.5909e-05\n",
      "Epoch 83/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.8791e-05\n",
      "Epoch 83: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 5.8557e-05 - val_accuracy: 1.0000 - val_loss: 5.3284e-05\n",
      "Epoch 84/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.0117e-07\n",
      "Epoch 84: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 2.0117e-07 - val_accuracy: 1.0000 - val_loss: 6.5107e-07\n",
      "Epoch 85/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 5.1437e-05\n",
      "Epoch 85: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 5.1395e-05 - val_accuracy: 1.0000 - val_loss: 1.3745e-05\n",
      "Epoch 86/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.4529e-07\n",
      "Epoch 86: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 1.4529e-07 - val_accuracy: 1.0000 - val_loss: 3.9431e-07\n",
      "Epoch 87/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.2901e-05\n",
      "Epoch 87: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 1.2919e-05 - val_accuracy: 1.0000 - val_loss: 1.2328e-05\n",
      "Epoch 88/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.8312e-07\n",
      "Epoch 88: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 2.8312e-07 - val_accuracy: 1.0000 - val_loss: 1.2746e-06\n",
      "Epoch 89/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 3.1647e-05\n",
      "Epoch 89: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 3.1341e-05 - val_accuracy: 1.0000 - val_loss: 9.0650e-06\n",
      "Epoch 90/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 7.4506e-08\n",
      "Epoch 90: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 7.4506e-08 - val_accuracy: 1.0000 - val_loss: 4.2182e-07\n",
      "Epoch 91/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.6338e-05\n",
      "Epoch 91: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 1.6732e-05 - val_accuracy: 1.0000 - val_loss: 7.2420e-06\n",
      "Epoch 92/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 7.9720e-07\n",
      "Epoch 92: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 51ms/step - accuracy: 1.0000 - loss: 7.9720e-07 - val_accuracy: 1.0000 - val_loss: 5.7771e-07\n",
      "Epoch 93/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.3956e-04\n",
      "Epoch 93: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 1.3771e-04 - val_accuracy: 1.0000 - val_loss: 1.2149e-05\n",
      "Epoch 94/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 4.7683e-07\n",
      "Epoch 94: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 50ms/step - accuracy: 1.0000 - loss: 4.7683e-07 - val_accuracy: 1.0000 - val_loss: 2.1091e-07\n",
      "Epoch 95/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.9076e-06\n",
      "Epoch 95: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m494s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 2.9034e-06 - val_accuracy: 1.0000 - val_loss: 1.1690e-05\n",
      "Epoch 96/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.6077e-07\n",
      "Epoch 96: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 1.0000 - loss: 2.6077e-07 - val_accuracy: 1.0000 - val_loss: 1.1646e-06\n",
      "Epoch 97/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.5402e-06\n",
      "Epoch 97: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m493s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 2.9207e-06 - val_accuracy: 1.0000 - val_loss: 1.4856e-05\n",
      "Epoch 98/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:20\u001b[0m 6s/step - accuracy: 1.0000 - loss: 2.6077e-08\n",
      "Epoch 98: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 1.0000 - loss: 2.6077e-08 - val_accuracy: 1.0000 - val_loss: 1.7056e-06\n",
      "Epoch 99/100\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.0185e-05\n",
      "Epoch 99: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m492s\u001b[0m 10s/step - accuracy: 1.0000 - loss: 1.0216e-05 - val_accuracy: 1.0000 - val_loss: 1.8227e-05\n",
      "Epoch 100/100\n",
      "\u001b[1m 1/51\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m5:19\u001b[0m 6s/step - accuracy: 1.0000 - loss: 1.1548e-07\n",
      "Epoch 100: val_accuracy did not improve from 1.00000\n",
      "\u001b[1m51/51\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 49ms/step - accuracy: 1.0000 - loss: 1.1548e-07 - val_accuracy: 1.0000 - val_loss: 7.8861e-07\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1./255,            \n",
    "    shear_range=0.2,           \n",
    "    zoom_range=0.2,            \n",
    "    horizontal_flip=True       \n",
    ")\n",
    "val_datagen = ImageDataGenerator(rescale=1./255)  \n",
    "train_data = train_datagen.flow_from_directory(\n",
    "    '/home/iamsanko/Downloads/Test 1/train',  \n",
    "    target_size=(224, 224),      \n",
    "    batch_size=32,               \n",
    "    class_mode='categorical'     \n",
    ")\n",
    "val_data = val_datagen.flow_from_directory(\n",
    "    '/home/iamsanko/Downloads/Test 1/val',    \n",
    "    target_size=(224, 224),      \n",
    "    batch_size=32,               \n",
    "    class_mode='categorical'     \n",
    ")\n",
    "cnn = vgg16_model.fit(\n",
    "    train_data,\n",
    "    steps_per_epoch=train_data.samples // train_data.batch_size,  \n",
    "    epochs=100,                                   \n",
    "    validation_data=val_data,                     \n",
    "    validation_steps=val_data.samples // val_data.batch_size,  \n",
    "    callbacks=callbacks_list                      \n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "457dc542f06f7897",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T00:59:44.625651Z",
     "start_time": "2024-09-29T00:59:07.123112Z"
    }
   },
   "outputs": [],
   "source": [
    "'''from tensorflow.keras.models import load_model\n",
    "\n",
    "# Load the best saved model\n",
    "model = load_model(\"cervical_cancer_best_model_2.keras\")'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "157f77cc92c9ee6a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:06:37.144343Z",
     "start_time": "2024-09-29T01:03:36.693948Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 867 images belonging to 5 classes.\n",
      "\u001b[1m28/28\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m173s\u001b[0m 6s/step - accuracy: 0.9919 - loss: 0.0215\n",
      "The accuracy of your model is = 99.19 %\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "test_datagen = ImageDataGenerator(rescale=1.0/255)\n",
    "test_data = test_datagen.flow_from_directory(\n",
    "    '/home/iamsanko/Downloads/Test 1/test',  \n",
    "    target_size=(224, 224),                  \n",
    "    batch_size=32,                           \n",
    "    class_mode='categorical'                 \n",
    ")\n",
    "accuracy = model.evaluate(test_data)[1]\n",
    "print(f\"The accuracy of your model is = {accuracy*100:.2f} %\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "899fca134afaa99e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:07:42.063774Z",
     "start_time": "2024-09-29T01:07:41.014274Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['accuracy', 'loss', 'val_accuracy', 'val_loss'])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "h =  cnn.history;\n",
    "h.keys() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6c2e27b84d13c43c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:12:34.875191Z",
     "start_time": "2024-09-29T01:12:28.552868Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: matplotlib in /home/iamsanko/anaconda3/lib/python3.11/site-packages (3.7.2)\r\n",
      "Requirement already satisfied: contourpy>=1.0.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (1.0.5)\r\n",
      "Requirement already satisfied: cycler>=0.10 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (0.11.0)\r\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (4.25.0)\r\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (1.4.4)\r\n",
      "Requirement already satisfied: numpy>=1.20 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (1.24.3)\r\n",
      "Requirement already satisfied: packaging>=20.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (23.1)\r\n",
      "Requirement already satisfied: pillow>=6.2.0 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (9.4.0)\r\n",
      "Requirement already satisfied: pyparsing<3.1,>=2.3.1 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (3.0.9)\r\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from matplotlib) (2.8.2)\r\n",
      "Requirement already satisfied: six>=1.5 in /home/iamsanko/anaconda3/lib/python3.11/site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\r\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiwAAAGxCAYAAABBZ+3pAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABxTElEQVR4nO3deXgTdf4H8HeS5ujd0htaeiBHsYBQECggglpETo+1eIAguKLsyuHZBURRQdyVdVelCognCj8RXdGq1HVBsChQAbkPobSUlh70PpI2md8fk0mTJmmTtE0Lvl/Pk6fN5NtkMpTOO5/vMTJBEAQQERERdWLyjt4BIiIiopYwsBAREVGnx8BCREREnR4DCxEREXV6DCxERETU6TGwEBERUafHwEJERESdHgMLERERdXoMLERERNTpMbAQERFRp8fAQkRERJ0eAwsRERF1egwsRFe5M2fOYNasWejZsye8vLzQrVs3TJo0CYcPH7ZqW1ZWhscffxxxcXFQq9UIDQ3FbbfdhhMnTpjaaLVaLF++HPHx8dBoNAgKCsKYMWOQmZlpdx8WLFgAb29vVFRUWD2WkpKCsLAw1NfXAwB++OEH3HjjjQgKCoKnpye6d++OO++8EzU1NXaff+rUqYiOjobBYLB6bOjQoRg0aFCzx6ioqAiPPvoo+vbtCx8fH4SGhmLs2LHYtWuXVVtH3r/BYMDrr7+O6667Dp6enggICMCwYcPw5ZdfNrsfRGQfAwvRVe7ixYsICgrCyy+/jG+//RZvvvkmPDw8MHToUJw8edLUrrKyEiNHjsTbb7+NWbNmYdu2bXjrrbfQq1cv5OfnAwAaGhowfvx4vPDCC5g4cSI+//xzvPfee0hKSkJOTo7dfXjwwQdRU1OD//u//7PYXlZWhv/85z+4//77oVQqkZ2djQkTJkClUmHDhg349ttv8fLLL8Pb2xs6na7Z58/JycEPP/xgsf3EiRPYu3cvZs2a1ewxunz5MgBg2bJl+Prrr/Huu+8iLi4ON954I3bs2GFq5+j7nzlzJubPn48hQ4Zg8+bN2LRpEyZPnozs7Oxm94OImiEQ0R9KQ0ODoNPphJ49ewoLFy40bV++fLkAQMjIyLD7sx988IEAQFi3bp3Trzto0CAhKSnJYtuaNWsEAMLhw4cFQRCELVu2CACEgwcPOvXc9fX1QlhYmHDvvfdabH/qqacElUolFBcXO/V8DQ0NQn19vXDTTTcJt99+u2m7I+//xx9/FAAIixcvduo1iah5rLAQXeUaGhqwYsUK9O3bFyqVCh4eHlCpVDh9+jSOHz9uavfNN9+gV69euPnmm+0+1zfffAONRoMHH3zQ6f2YNWsWMjMzLao67777LoYMGYKEhAQAwHXXXQeVSoU///nPeP/993H27FmHntvDwwP3338/tm7divLycgCAXq/Hhx9+iClTpiAoKMh0LMxvgiCYnuOtt97CoEGDoNFo4OHhAaVSif/+979Wx6il9//NN98AAObNm+fgkSEiRzCwEF3lFi1ahKVLl2Lq1KnYtm0bfvnlF+zbtw8DBgxAbW2tqV1RUREiIyObfa6ioiJ07doVcrnzfzruu+8+qNVqvPfeewCAY8eOYd++fRbdNT169MD333+P0NBQzJs3Dz169ECPHj3wr3/9q8Xnf/DBB1FXV4dNmzYBAL777jvk5+ebnj87OxtKpdLitnPnTgDA6tWr8cgjj2Do0KH47LPP8PPPP2Pfvn249dZbrY5RS++/qKgICoUC4eHhTh8jImpGR5d4iKh9BQYGCjNnzrTa3q1bN2H06NGm+8OHDxd69erV7HPdc889gkajEfR6vUv7cvfddwtdu3YVGhoahCeeeELQaDRCaWmpzbYNDQ3Czz//LNx3330CAOGTTz5p8fmvv/56YejQoYIgCMJdd91lei1BEAStVivs27fP4lZRUSEIgiAMHDhQuPHGG62eb8SIEUJ0dLTpviPvPzU1VQAgXLx4scX9JSLHscJCdJWTyWRQq9UW277++mvk5eVZbBs/fjxOnTplNXC1aZu6ujpTlcRZs2bNwsWLF5Geno6PPvoIt99+OwICAmy2VSgUGDp0KN58800AwK+//urQ8//yyy/YvXs3tm3bhgceeAAKhQIAoFKpMHjwYIubr68vANvH6LfffsOePXsstjny/sePHw8ASEtLa3F/ichxMkEw68QloqvOAw88gM2bN+Pll19G//79kZWVhb///e9Qq9Xo0aOHaRZMZWUlhg8fjvPnz+OZZ57B9ddfj9raWuzcuRMTJ07EmDFj0NDQgOTkZOzevRsLFy7EmDFjYDAY8MsvvyA+Ph7Tpk1rdl8MBgOio6MBABcuXMD27dtxyy23mB5/66238MMPP2DChAno3r076urqsGHDBmzZsgXfffcdkpOTm33+8vJyREREICgoCBcuXMDJkyfRq1evFo/RsmXL8MILL2Dp0qUYPXo0Tp48ieXLl8Pb2xsNDQ2m2T2Ovv8ZM2bgo48+wkMPPYSJEydCrVbjwIED8PLywl//+tcW94eIbOjoEg8Rta/S0lJh9uzZQmhoqODl5SWMHDlS2LVrlzB69GiLLiGp7fz584Xu3bsLSqVSCA0NFSZMmCCcOHHC1Ka2tlZ49tlnhZ49ewoqlUoICgoSxo4dK2RmZjq0P3/7298EAEJUVJRV18qePXuE22+/XYiOjhbUarUQFBQkjB49Wvjyyy8dfr/33nuvAEAYMWKEwz+j1WqFJ554QujWrZug0WiEQYMGCV988YXwwAMPWHQJCYJj71+v1wv//Oc/hYSEBEGlUgn+/v7C8OHDhW3btjm8T0RkiRUWIiIi6vQ4hoWIiIg6PQYWIiIi6vQYWIiIiKjTY2AhIiKiTo+BhYiIiDo9BhYiIiLq9Dw6egfaisFgwMWLF+Hr6wuZTNbRu0NEREQOEAQBlZWVLV6n66oJLBcvXkRUVFRH7wYRERG5IDc3t9kLsF41gUW6Jkhubi78/Pw6eG+IiIjIERUVFYiKijKdx+25agKL1A3k5+fHwEJERHSFaWk4BwfdEhERUafHwEJERESdHgMLERERdXoMLERERNTpMbAQERFRp8fAQkRERJ0eAwsRERF1egwsRERE1OkxsBAREVGnx8BCREREnZ7TgeXHH3/EpEmT0LVrV8hkMnzxxRct/szOnTuRmJgIjUaDuLg4vPXWW1ZtPvvsM/Tt2xdqtRp9+/bF559/7uyuERER0VXK6cBSXV2NAQMG4I033nCo/blz53Dbbbdh1KhROHDgAP72t7/hsccew2effWZqs2fPHqSkpGD69Ok4dOgQpk+fjrvvvhu//PKLs7tHREREVyGZIAiCyz8sk+Hzzz/H1KlT7bZ5+umn8eWXX+L48eOmbXPnzsWhQ4ewZ88eAEBKSgoqKirwzTffmNrceuutCAwMxCeffGLzebVaLbRarem+dLXH8vLyK+bih0fSPoShqgr9n3zEbpsTm7ah5tARDFqZav+JDh4E/vc/4K9/BTzsXM/y9Gng88+BefMAb2+bTUrP5uDE39cgYckC+HYLt9mmMr8QJ598Dtd4yxDgqbTZRqc34FhhDXwfeQg9bhxqs40gCPjolxz0jfBDYnSg7X2ursaBZ/+OoMKL6B7kZbsNgLPFVTgf3QcRf/0zeof5Wl1AK/dyDU6+/ym6/bwTfcKtH5cUVtahoFyL3mE+UCsVNttcqqjD+ZJqGFz+X9PI4OmFqKfnI2pQX5uPVxSX4viSl9HLUIlAL5XNNnpBwIn8CoT4qhHqq2n+BSdOBG6+2e7Dh/61AYLBgOsWzrHb5uiHWyH/6iv0DPWFh8L6OAqCgAultTih8MOeKTMgKGwfx5CcMxi640sMCPex+TwAUK1twNEqoNeKJQiIsv37WH6hAKf+9iL6egPeatu/+w0GAUcvlqN7Fy/7x9Eg4MSlSij+9Cf0uWu87TZ6Aw6krkDXolx09fe02QYQfx8L+g/B9Y8/BA+F9WdCg0HAL2kbIRRcwrDnF0Eut/3+j2zdDmzejGvDfez+zpbX1iP3cg3iQrzhpbL9/ourtLhQWoteYb7wUtn+96jWNeB0WT2ilj6JoPhrbLe5XI7f5i+BrLzM5uPO8tMoEdXFE74ay78jgiCgtKYeeWW1kAGIDPSEv6fS6hiU19bjQmkNKusa2mR/yDHdl/8NXa+Lb9PnrKiogL+/f8vnb6EVAAiff/55s21GjRolPPbYYxbbtm7dKnh4eAg6nU4QBEGIiooSVq9ebdFm9erVQvfu3e0+77JlywQAVrfy8nLX3oyb1VbVCLUeKkEPmXDx+O822+gb9EKhbxdBAIQD6T/af7IRIwQBEITt2+23ufdesc369Xab7Jo6SxAA4dTgGwTBYLDZ5vjYieLzOHD7OX6YYLDzPNsO5QnRT38lDHj+O6Gyrt7ywYYGQdiwQagLC3f4tQRAmDZthZC08r/C4s9/E/5zME9Y8fUx4aZXdwgjH14v1CmUTj2Xu275fsHCrz/9ZnV88gvLhf29Brft6wUH2/23LysuE7RyD0EnVwjHT+TabHOpvEYo8fJz+PWeGD9fiH76K+vbU9uEYyExDj/P5jsesfl7ZDAYhP+b+nCbHqPLGl9h6/aDVq9Vq2sQ1s19wann+vedC4SfzhRZPM/BnFJh1YPPC3rIBAEQlqz+j1Cra7B6vQ/3ZDt1jNrqtm/ASLv/Z3fcPbfD/p/w1nluxz//zu7fEVeVl5cLjpy/7XwcbzsFBQUICwuz2BYWFoaGhgYUFxcjIiLCbpuCggK7z5uamopFixaZ7ksVlivF+SNn0LtBBwA4+/1PiOgTZ9Xm7NGzuKbyMgDg0I8Hcd34Ubaf7Nw58eulS/ZfUHrs/Hm7TYTsbABAz/0/Atu2AZMnWzbYuRN9fvgKBsiwfshU6JUq3BQfil5hvgCAy9U6bDt0EQGX8jDl+E54XS5C5u8lGHFNsNVrvZ8pvlZZTT02/nweD4/uIT7w/ffAE08Ahw5BDSDXPwxf9RkFmYcCdwyKRKiv2vQcF8tq8cXBPCRcPI0bsg/ghe/TcGtkX3z0cy0++jnH1G79f9dCra/H4bAe2BmXiJv7hKJPRGOKr9I2YMv+C6jSNcBDLkODsXwS4qvGyB7BuFRZh33nLqPeIEAGoGeYj9WnQld0/yEd3QpzUTF1ErZ//g2SR4ifWk7ll+PsbXfi1lP7UaNU4/1Bk+DrrcZdiZHQmFV+DuaWYveZEtP9OwZ2Q9cAG5/8GxqAV14BiouB6mqbFbbsw2cwwCB+Uv38s11I/ds9Vm02fnMQC2sqAABrht0Fg0yOyEBPDIsLwu9FVTiUWwaDAPQuycUtp/bgqePpCH3sYaDJJ+PoX39CfFE2qpUafDJ0CqYnxUDtYfnJv6CiDme2/Rcjzx+C/NQpbNqXi3uu727R5pO9uVCfPgUA+DluIIam3GqzErHt0EWcv1wDAAj0UuHOQd0sjuOvOaXI/L0E409mosflC6h98hn88+//woKbe0Imk+FytQ5/fWsHVn/0TwDAt72Goyz6GvwpMRIKs+pIXb0en+zNQWDRRUw9thPzPvsX5jVosPFPf8IjN/bAh3vOI2fr13jv/16AHOLv2NlfDmO6ZzDWzRiMAC8VBEHAP7afxJs/nMGxsnwAwO5bp2HkIOu/D9XaBrxr/H8kCfRSYVTPYJTX6PDzucvQNhgAABoPOeoaDNB4yDGhXwQijL8nF0prkH64AIJOh4f3bsWgQz9hd3omRk0YYfG8Z88Xot+2j8XjNXYKtKG2K16OEgSgoLwWBRV1NquVCpkMUV08YRCAvNJa6AXrRnIZ0DXAE6G+mqa/YtSOYnvEdNhrt3tgAWD1R0Qw/vKZb7fVxl4ZFADUajXUarXdxzu7guO/o7fx+4q9vwKYbtUm+4dMSMXZnKO/o0FvsC4x6/WNYaSiwv4LSo/ZCTW1Oj1UJUWm+w1/+Ss8br4Z8DJ2xdTXo+GRR+EBYOPA8ch85BnsOFmEVQAWj4vHgKgAPPTBfpQPrceEqmxMOb4TftpqvJ6ZbRVYjl4sx77sUtP9dbvO4YGkGGheWw089ZT4+r5+WJV4FzYOmYxr40KwL7sU7/lp8MW8EQj31yC7uBq3r/kJpaPqcWecF0YtvhvXFOfiG9mveH/YXfj1fBn6hPti2qWDuP7MXkCpxN4X/41/nBHwL4UMG+cMw/WxXVCja8C0t/fgiEcFrgn1wacPD8cXB/OwOuNUY6nZD0A3YFD3ADw/OQH9Iv3tH2cn1J0+i9Lrh6FX0XmUT7sL7723Bb1jQnFy+sOYefC/aJArcHH9h/gwrwsultchvUcQ3n/weigVcmQcu4Q/f7gfQlegq78GF8vrUDQ8Gs9PSbB+IUEAXn8dqK0V//3jrE9++SfPYYDx++z9R/B7URV6hPiYHi+t1iHz+/1YCEDbJRh1y1/CWzt/h854QoQ3gBhgbJ9QTBrVFbiuD0KyT+NJWQ4wbpzli/1zAQDgu2ET8GLSdOhu7Y1Hb7Tshljy/n54nZdj5PlDiCq/hAe/OoaR1wQjqov4+5h7uQYvfn0M75WLv88br70ZXo8+hf6RAZbHuF6PJ5dvR129Af6eSpTX1uPbnsHYMHMIlAo5vj1SgEc2ZkHoBsTdfwd6zJuGaYe+w+0ffoWLZTfhkRt7YM77+3HP/6UhrOoyqrvHYtndf8OlehnyxlyDx5N7m17rmU0H8IX8IuKCvZB87GN4vbser331D8zw8sfEw/0QX3gWmz97EWp9YxdGXO1lfJhdijvTMrH+gSF4/YfT2PprHvzrquBVL3Z5PzXkPux6boJFOAKA//ySg394HEb/SH/cN7Q7Xvn2JEqqxQ9B8AcQAcRH+OH5ydeiR4g3Hnx/Pw7lluEVDzn+lXIddHoDnvj0EOpHCLg+pgtuNJSg9/6duPziKujGfQGVR+PfmsxnX8X9tRUoCumKQds/A+x09TmrvKYeP54uwg8nCnGyoBKDogMwtk8ohscFw9PYfVWtbcBPZ4rxv5OFOJRbjr5d/XBTn1CM7BncJh8c6MrR7tOaw8PDrSolhYWF8PDwQFBQULNtmlZdrialp7NN33scPWKzTeXeX03f+14uRObvJdaNiorE0AK0KrAcvViO4OoyAEC9XAGP3Bzg5ZcbG7z+OjyOH0OJpx/+c+cj2PDAEMwaEQMAeCn9OKat3YPy2noM7B6Al2aNBAD41VXj++OXcKG0xuK1PsgUqzy3XhuObgGeKK7SYvO+XGD3brHBnXdiwUtbsG7oHZh0fSzemTkEPUN9UFBRhzkf7ENBeR0efH8fSmvqMSDSHy/OvAGyV14BAPRc8ypeHNIF6fNHYfXk3rj+teXicy5ahFkPjsdt/cJRrxfw8If7ca64Ggs2HcSRvAp08VZhwwNDEOitwqwRsfjfEzciZXAUZDIg2EeNV/80AFvmJrVZWAEATc84+O34HrVevhhy4RgiHnkQO+c8gZl7xAHp2rfX4ZoZf8I7M4fAW6VA5u8lWPrFERy9WI75mw5AEIB7h3bHS7f3AwCkHymA3tbHVZkMCDd+IrZTtSz5vbEiFVl2CWv+97vF4+9mZiO46CIAQNWzBxbd0gv/XTQayX3F/6PRQV5454HB2DBzCKJ7dAMeekj8wX/8w/KFDh8GvvsOkMvh/ZRYIV2/6xxqdI0n8WMXK8TfmwDxuXtUFaFap8eTWw7BYBBgMAh44tNDqNHp0aNKDNm5AeHYcbIITe09dxl19QaE+2mwcc5QeCoV2HW6GM99eRSHL5RjwWbxOM4YHo3bHk0BZsyAHAJeyHgLn+07j5tX74TH8aOYlfUlAMD77TV49k+JAIA1O37HodwyAMC3R/LxxcGLkMuAV+++Dl7r3gJuvx1qfQPe/fxF3Hz6F2z87Dn46WqA0aOB++4DADzW2xPhfhr8XlSNm17dga2/5kEhl+GVoV0AAJe9/HFRJ8OBnFKr9/b9cfH/8rhrw5EypDt+eOJGzEyKgUIug7+nEi9MuRbb/jIC18d2QZCPGpseGoab40OhazDg0Y9/xfxNB1GvFzChXwQ+mH09olY8Kz7f/u/wf99kmV7np5OXkLTtI/HOgoVtFlYAwN9LiUkDuuKfKdchff4ovDi1H8b2CTOFFUAcm5R8bThW3tEf6fNH4R9/GoDx/SIYVv6A2j2wDB8+HBkZGRbbtm/fjsGDB0OpVDbbJikpqb13r8PUnjM7QeSeQVGl1uJxQRCgPtYYZEKrSvHVbxetnyg/v/H7ykr7L9hCYDmYW4bgavGP4r9GGLsDXnkFOHMGuHgRWLYMALBq9EwkDuoBuVyGZZOuxZIJYjeGQQCS+4bh4znDENA1FADgp6uBwSBYdM+U1ejwxcE8AMCcUbGYe6PYFfTWzt9haBBPWheG34iv8nSQy4BHbuwBP40SG2YOQRdvFY7kVWDsqztwtqgaXf01WDdjsPjH7YEHgOHDxS6PJ54QX2zVKrG7LDISWLIEcrkMr/7pOgyI9EdpTT0m/HsXth+7BJVCjrXTEy0G9gb7qLHqrv746emx2PXUGNyZGGl3cGRrKAb0hyZ9GxqUKow7/TOe2fEeAKB+5cvwnjMLgPgp+fV7B0IuAzbty8Wf3tqDGp0eI68JxvOTr8WIa4Lh76lEUaUWv5yzEWoBICJC/Gr++2KmJvuC6fuo8kv44mAeco3dKBV19Xjvp3OIMlYzZLGxYrsuXlg7YzB2PTUGGQtH46Z4sw8Y8+eLJ7bvvwcOHWrcvnq1+PXOO3HT+GHo3sULl6t1+GRvrqnJmzvOAAD6DBdrPkFlhfCTG/Dz2ct4f0823svMxi/nLsNfrkeX8mIAYtfhzlPWgUUKMTf2DkFCN3/8a9p1kMmAjb/kIGXtHtTVGzC6VwienWgc+PzKK4CfH/oXnMb0IxkwGAT8Y+d6eBgMwO23A7feign9IzBpQFfoDQIe//QQ8spqsfhz8f/q3NE9MLB7oPjeP/4YGDUKXnXVWL/1BXSpuAz06wd88QVwjVhRCikrxOfzktA7zBcGAfBSKbD+gcEY518PAKgOFf/dMo5Z/r+t1jZg9xnxvd9sPO7+nko8N/laZD4zFj89MxbTh8dYVGQ9VQq8dX8i7hvaHVIPy+yRsXj9noHQKBXwunkMLvfpB02DDuWrX0dptQ56g4DvX1mPuNKLqPX2Q8hjc62OMZG7OB1YqqqqcPDgQRw8eBCAOG354MGDyMkRT0qpqamYMWOGqf3cuXNx/vx5LFq0CMePH8eGDRvwzjvv4AnppAJg/vz52L59O1atWoUTJ05g1apV+P7777FgwYLWvbvOLC/P9O01JbnIOmV5IjlfUoPYi42fckOrSvDtkYLGErzkolmIaUWF5cjZQvhrqwEAHw28DbvjBgFarXjieeIJoKoKhyP74NP+N+PGXqGmn5szKg4fzR6Kl25PQNr9iWJ4CAgAACgMenjV12HzvhzU1YtVoM37cqFtMJhmB/0pURyXkl9eh0uXxdfPOCWedCcP6IqYYHG8RVQXL6ydngiVQo4and74h30IQv2MM2PkcuDNN8WvmzYB69c3VohWrwZ8xO4NT5UC62YMRld/DWp04j69cld/DI7pYvO4dA3wtPi01x5ko0fDY9MnEIxdoIYFC6B8+imLNmP7hGHJBPGkWqPTo0eIN968bxCUCjlUHnLceq1YQfnqN9uBpLkKi8EgwGAWZK5rKIXeICBtp/j79+Ge86ioa0CCThxPBWNgkUR18bLoPgAAREcDd90lfv/qq+LXixeBjRvF7x9/HB4KOR41Bta1P/6Ouno9zhRWIf2wuC8zpl4PeHpCJghYPlAcJ7Xq2xNY9e0JAMDy63whEwQYvLxQ4uWPAzmlKK+pt9iNnacKAQCje4UAAJKvDUfq+D6m49grzAev3zuw8cQeFga88AIA4Nk9H+Fj7T70//0g4OkJ/POfpuddPvlahPiqcaawChP+vQsl1Tr0CffF/Jt7Nr64RgN8+SWQYOym694d+PZb8f+HNN4uNxcR/p749JHhWHxbPD5/dATG9A4FcsUAp4wRx+00DSy7ThdB12BA9y5e6BXmY/FYmJ8GPnZmTHko5HhxagL+Ne06rLlvEJZO7NsYxGUyBCx5BgCQsvdLvJl+GJ/uz8X47eLYFcx92PT/iKgjOB1Y9u/fj4EDB2LgwIEAgEWLFmHgwIF49lmxnJifn28KLwAQGxuL9PR07NixA9dddx1eeOEF/Pvf/8add95papOUlIRNmzbh3XffRf/+/fHee+9h8+bNGDrU9pTYK121tgFeRY1/gJQGPc7tzrJos/9UAXqUNH7q7VZbhoo6sS/XgiOBRa8XKw+A3cCSezIbAGDw8IAqOAhLb3oYBqUSSE8HPhFPps/cNBdeaqXVNOSRPYNx39Doxj52jQZQidNHe6n1KK2px5eHLkJvEPDhz2J30MykGMhkMmiUCvz5BnFMxYXiKgDAwYtipWjeGMtxDYNjuuDf9wzEdVEBSLs/EX27Npn+NnAg8Oij4vcPPSQGrptvbjxxGoX6afDOzCEYHB2IZZP6YurAbraPmzvdcQdk330HvP025K++ajVQFQBmjYjBolt6YVhcF2yYOQT+ZtPKJ/QXP4l/e6QADXqD1c82F1jyymrhX3HZdL93rfg7tmX/BZwtqsI7u8VB3UNRLjZoEljsevxx8esnn4gB/Y03gPp6YMQIwPh/+45BkYjw1+BShRZbsi5gzY4zEIzVuj4R/kBMDABgsp8OI68JRl29AdoGA0b1DMZkP7EqKY+NRa9wsUKx60xjlSX3cg1+L6qGQi5Dktk4qodGxWH+TT0xPC4I7zwwBH5NuxYefRTo3x+KsjIk/cvYpbhkiRjCjAK9VVhp7Iorq6mHh1yGV+8eYDV4GAEBYpVp1Srgxx+Brl3F7WaBBRCn+D50Qxx6h/tabO/SuweUChnOFlfjTGGV6Wm3GwPMLX3Dmh3rZ4tMJsOU67rhtn4RVo/J7/4TtBHdEFxTjpp338dX67/A9ReOQe+hhOeiBU69DlFbczqw3HjjjRAEwer23nvvAQDee+897Nixw+JnRo8ejV9//RVarRbnzp3D3LnWZcW77roLJ06cgE6nw/Hjx3HHHXe49IauBGcKqxBeZVm6r953wOL+ucwDUBka+/W71ZYBALY17RYyL/HbCyxVjX/oUFNjeR9ASZUWujzjiSw0DKN6h+Fcl27Yc/ssU5vDk+7B0fBrkHRNsPWn6aZkMlOV5Z5eYqh4PzMbP5woxIXSWgR4KTH5uq6m5vcO7Y4u3io06MQBg3q5AuMTwtHTOPvI3K0J4fhi3gjTJ2YrL7wAhBgfUyrFwaY2/qDHR/hhyyNJmDXCwZOvO9xyC/DnP4tVIhtkMhkeu6knNv15OKKDLGf6JPUIQhdvFS5X67DnrI1uoWa6hI7nVyC0qjGweOblYmhMIHR6A+5f/wsuV+vQvYsXQkuMP+toYBkyBLjhBnGW0sqVgLTCtVl1VeUhx1zjDLF///c0/nNQ/P3+y9hrLF5Lfj4bq+7qjwAvJQK8lFh1Z3/IjLPaEBtr+n3YaTaOReoiGtQ9wCLcyWQyLLylFz758zDTIF4LHh5itU7Ss2dj+DJzc98wTBsiBo+Ft/TCtV3tjG8KCxMHk5sFnqaBxYpxuyqmO4b3EMOWVGVp0BvwwwmxcnRL3zYe56dUQv2EOLbowV8+xz0//p+4/Z57GsMWUQfhtYQ6wMlLlQivFD/F1vfrDwDwO30c5bWN5ezaLHHAbW33GACAb3kJ5AY9Mo5eMnWvAHCswtJ0e5Mqy28XyhFcUwYAkIeHYXRv8Y//qoF3AH37ArGx+OcN4iymG3vbCQpN+Yt/vMdHaqD2kOPoxQo8v+0oACBlcJTFtFIvlQdmj4yF3CBWBvQyuVV1xWEBAeIneblcHHfTp49rz3OF8VDIcWuCsVvokI1uoWYqLCcLKhFSbTaos6YGCweI/34Xy+sAAI/eEGsREBwmnejffBMoLRXHbkyaZNEkZUgUgn3UKKzUQm8QMLpXSONsH+m1zp1DtwBPfL9oNL5fNFqcvi1N54+NxWhjN+XOU0WmWYhSYLmxdyicNnKkuMiiSiUGLTszElfc3g/fL7rB1LXlMCmwVFTY/n8rBZmoKFMoyTgm/tvtP1+Kspp6BHgpMdjeooutMWcO9L5+uObyBUw4+RMAQPGEdWAjcjcGlg5wuqAC4ZXip2DlhNsAAH0KzyHrvPgp92JZLULPietLeCTfAsjlkBkMiJfXoVLbgB/NBxe2QWA5YDbgFmFhGHVNMOQy4LeyBuT/+DPKDx7BjyVimLihp4OBxVhh8dNWY/IA8ZPZhdJayGTA/cOirZrPGB4NNcTX6Nu9CxK6tWI2zt13i1WkxYtdf44r0ERjif/bozbGOjUTWE40DSwAhqICA6ICAAAR/hrc0U0pdrHJ5eJYDId3aiLQq1fj/YXWs0w0SgUeGtUYgkzVFcAisADiYOhgH2N4MAtQQ2ID4alUoLBSi+P5ldA1GJBp7D61W41ryeuvA+XlwNixdpvI5TJcE2p/9WS7fHxM/0dw4YL149K2qCjcHC8GrgO5ZSiq1JoqLWP7hNpcSbfV/Pyg+PNDprvCLbcA/fu3/esQOYmBpQPknb0Itd5YTbn1VgBAnyJx5gMA7Mu+jPhC8Q+0cnCi6WQzNUz8o/j1YbNP0I50CbUQWA7lliHEOKUZYWEI9FaZTlY/ni3FT+fLoTcI6BHibbuEbov0x7isDA8kxZg239QnzOZz+GqUiPIXT0S3D7YONE7ztL90+tVqaFwQgn3UKK+tx0+/NxnrJHUJ2aqwXCwzTWlHD7FSIMvOxrMT+6JHiDeWTboWqlzjgoORkWJXm6PkckBa4DEoCJg502az+4dFI6lHEO4d2h1DzAdANwksFswqLGoPBZJ6iMsk7DxVhKzzpajW6RHso0LfCD/rn3WETCaOx2ovkZHi16bdQoJgEVgi/D3Rr5s/BAH47/FLpsByS3w7Lvswf77pMh8ysy48oo7EwNIBKs5kAwDqA7sAgwdDkMkQWl2Kk7+JszJ+OXcZfYrENujf33SyGesvjmn5/phZt1ArKyyCIODQhbLGT9jGtW+kT6U7ThZhx0mxv9yp0rpZYEno5o+RxqqN+SfppoI04ifvyBDrsSvUMoVchtv62ekWkiosly4BhsbqS129HuW5+fAQDOIsJWmg+7lzSIwOxH8fv1HsajILB0578EFxbNGnnzYuRNiEt9oDHz80DCuMA1lNHAwsAExdmTtOFmKHcXbQDb1C2mU6epuwN46luBioqxMDUzdxQLjULbR211nkXK6BykOOG1ytHDm6b598Avz73+LYKqJOgIHFzSrq6uGRL4YMeWQk4O2NhjjxU63+0G+o0TXg5OGzCJMGQfbrZxrsFldfgW4BnqjW6fG/E4Xi7B/zT8wuBJacyzUoq6lHaI1xBogxsEjhZPfpYtNYAKdK62aBBQDenp6I/z5+I4bGBdn/GeM6LG25MNUfzcT+4u/K9mMF0DaYjXUKNYbNhgagpHFQ7ulLVQiqMobVkBBxgClgHRBaE1iUSnGWzZgxzv+s9HpFRZaDxauqxBO7WRvp9zPrfCm+O1Jgsa1TshdYpPthYabZdlJgOVskzvYb0SPI7sUe28xdd4kXVOW699RJMLC42elLlQgzzhBSRIklYeV14gJZvS6dQ8axS1AdFwen6uPixL5uY2CR5eebpq+mHykQ/4ibfVpGTU3jSd9cM4HloHGlzu4NxkXnjIGlXzd/BHopUaltwKUKLTRKOa6Ptb1WiU1NAou32gOxwbavEm0irdhr74rT1KLB0YEI81OjsukUeKUSCDZO7TULuScKKkzVNVl4uP2KRmsCS2sEBDT+LkljVsz3p0sXwHh11+ggb8QEeaHBICC7pAYyGTDK0TFXHaGlwCJ1GQHoE+6LyMDGbs5b+rbuWj5EVyIGFjc7danKNOBWKvdKA9r6FGYjbcfviC/MBgAoBhiv7mI2JXWoMTRkF1c3dgcFmVUtbK1260BgCa+1rLAo5DKLP/bD4oIsZva0qElgcQgrLK0ml8tMY0DOFVteEsHW1OYTBZUIlSosnTGwmL+m+T7Z2R/zbssBkQHo4q1q771zXUuBxexirjKZzGIKszQQl+iPhIHFzU6ZTWluGljii87hREEl+hSds9huWv/g4kXTH+DL1brGE090dOPgQFvdQtI2aX0Ss8AiXQvF33hVaJhdv8l8CvONzpbWXQksrLC0iSDT74jl5R5szRQyr7BYBJacnMZ/D+CKCSzmXUCdujsIaAwkTWcJmQ24NTd5QFfIZcDIa4IbV3gm+gNhYHGzU5cqGxeNaxJYehafh8KgtxxwCzR+Mr540TSls7hKC0Fa3r9rV1NZvNnAIo1PMJ6w6vUGHLlYAQ99A1TlZeJjZoFlVM8QU/f1aGfXsmCFpcN08RZ/R0qqdJYP2AgsFmuwRESIv0tKpfhvIZ04GxoaP/V38sAyLC4IauPChqMdXTOoo5jPEpIu7iPdB6wCy8Dugdi+8Aa8ed8gN+0gUefCj7JuZrNLKCYGgo8P1FVV6Fmcg95FximkTSss+fmmCou2wYD63DyoAPFE4+cHFBa2HFgyM00VlpMF4noV1+iNgxkVCovupRBfNf5593Woq9e3PP6kKSmwlJc7/jOssLSJIB/xd6SkuklgadIlVFSpRXGVDqHVxupaeLj4OxAdLV708tw58fvcXPHfRq1ufA53ciKweKoU+Pc9A3GxrBYDjVPzOy0psFRXi8E+0LgInJ3AAgDXhHIGHf1x8czgRqXVOhRVahtnAEmBRS6HrF8/YM8ejD+ZKa7R4uUFxInX2DEFlkuX4KUA1B5yaBsM0F0wBhZnKyxVVUBNDQ4Yu4OG+RgrGyEhVsvCu3ydHVZYOkyQebehuSYVlhMF4u9FlLbC8vHYWDGwSINcpXAQHW33sgHtyng9IZuDbqXHzIy79goZkOrlJX5AKCkRQ0rTwGI26JaI2CXkVqcuVULdoEOXWuMJoptZGDBWU6Ye+594v1+/xpODFCQMBsiKikwnJP0FY5eQVGEBTOHkpzPFyL1cY7EN3bo1jnW5dMk0fmWg2nhiC2vDhaiMS/NzDIv7dXEwsJwsEAdohxuvU2WqnjStaHTk+JWm+yMI4q2j96mtNB14azA0XsndRoWF6I+MgcWNThVWNV5kTqMRp2RKjIEluqzA4j4AseIgnWwuXkSQcRyLzLieS9MKy+lLlbhv/S/4y8e/mrYBEEOEFErMAksfGINNWwYW8wqLef98c1hhaRNBZuOcLDRZ7fZ4vhhYAivMuoSAzhdYpCpKRYV4PaKSksY1WWxUWK4oTQPLpUviFa3lcl5skKgJBhY3OlXQZIaQ+YJM0hRme/fNBt5Kn6A9Co2zfbp2BXyNfdsVFfjduLiUtMiUKbD4+ZlCSe2FizhTJP7Rj5bGsIS3YSldCiwNDeL6MI5ghaVNSBW4yroGy2sKSf++xjEsJy9VQF2vhbq60vLxzhZYvLwaw/S5c437ExHRvkvnu0PTmULS14gI/j8gaoKBxY3EKc1NBtxKEhIs7ze92JjZwNsgbxXkBj00JcaLIDbpEioyfrKu1DagVqe3GVjKzuVCEIBALyV8yoz71JYVFi+vxj+4jnYLscLSJvw9lVAYl6MvrTHrFpICSXk5GqqqcepSVeMMIY2m8XeoswUW89c2DyxXencQYH09oWYG3BL90TGwuIkgCM0HFn9/y/J2vybXVDFbiyXIR4WgmnLIDQaxdBwaahlYKhu7AoqrtI2BxdfXFEq0F8RP2WF+msZ1WdoysMhkzg+8ZYWlTcjlMgR6iRcotOgW8vcXZ/oAuHDiHHQNBkTpjL8bERGNFT8pCFy8KF6h2eyqyB3GPLB0hv1pK027hBhYiOxiYHGT4iodSmvqrddgMSdVVbp3bzzZSyy6hNSNY2HCwsQTvEVgqTP9WGFFrc0KS0O+OI4hxFfduC5HWwYWwPnAwgpLmwkyrsViMfBWJjP9Hl04fhYAcJ1HrfiYeXdgcDDg7S2OPTp5snGBws4SWK6mCou9wMIZQkRWGFjc5PQlcZxAnK5M3NBcYGnaHQRYdQmZpkZLQcZOheVyYVnj9YbMAguM419CfdupwgKwwtKBWpopVHw6BwDQB9UW2wGIwUYKAzt2iF99fS0Hibvb1R5YLlwQAyIrLER28czgJieNgSWqxjhmwFZgeeQR4OhR4MknrR9rsjy/qcIibbcTWMovGSs6Mpn4qdkYSpRF4viXUD915wgsgtAYrFhhabUu0uJxdla71V24CITGNq7B0nRBuNhY4MgR4IcfGu935FV7m05tNt92JZP+DtTViVefZmAhsouBxU2kGTshFU2uI2Sua1dg61bbT2C2SmmQj+MVlsrCksbHZTJTKNFcFvcjzFMhThMFOjawmF+3hhWWVpNmCpU0vZ6Q8fdFVngJCAWCzC98aK5phaWjw4H0+uaLx3X0PrUFtVr8f3fpklhlsXMdISJil5Db5JfXQSYY4HPZOLPH2T5qs9Vug9SKxsDSpMIimM0SAoDq4lKLx6VQ4lMuhpTIhmrxE6tcLo5daEvOBBZp/ArACksbsDmGBTAFE3WRWFXzLyu22G4ihQHp0godHQ6iosTfUa1WvCkUV884D+l9ZGc3XoGdgYXICgOLmxRU1KJLTQUUDfUWgx8dZrbabVBNGUKM13/RhhqrIlJgKa9Avb5xoba6EtuBxbu2CuoGHcK1xhNScHDbBwVptVtHrifECkubaqlLyLtUDCpeUoC2F1js3Xc3pdLyJN69+9XzeyK9r337xP8HHh5tX+0kugowsLhJQXld46JxoaHiH2BnmK1261VSiAhjhaUiwHhFWrMKizldabnF4wgIgKAST2ZBNWUIrioTt7fHH0hWWDpMS9cTCjL+/iiLC8XttsawNHe/I5jvQ2fYn7YiBZbMTPFr1678P0BkAwNLG6ir1+NgbhkMBttL0Gsb9Ciu0jU/pdkRxu4fWX4+wowLfl32N3bjGAOJvKoSMsFgGh+pL2sSWGQyCCGhAIDg6jIEmk+Pbmscw9JhGsew2L5ic2hVKYI9PSCTprR39gpL033oDPvTVqTAsnev5X0issDA0gZWph/H1Dd/wndHC2w+fqlcHFPSrbrJVZqdJX0KvnABXYyDJYt8jVd4NQYSmSDAS1eH2CBvAIChvElgAaAzBpYoXSXUJcaqT0cHFvMKS0dcEfgqE2TqEmoy6NYYTIKry9BToW087qGhlu18fcUrCUs6wzV7rvbAUltreZ+ILPDM0AYO55VbfG0qv1z8Q9Sj3vh4KyssOHgQCsEAvUyOApUxiHh6msrIProaxEeI273qjNfxMQsstYHiiShGX9l+U5oB1yosCkXHTp+9SnQxDrqtqGtAvd7sekLGf2eVoQEDqowLwgUHA8ZuQgtSKAgJAXx82nN3HXO1BxZ794kIAANLm7hQWmvxtamCCnHl2e61zazB4ggpsGRlAQCKvQNQrDWejGQyUyjx0dWiW6AnAr2U8NVZB5YKP3EBsEhdJwosXOW2TQV4KmG8nBBKzbuFVCrU+gUAABKKxNVu7V70UgoFnSUcMLAQ/aExsLRSXb0ehcZ1T3JLbV+VOL9cDCzSQNlWdwkdPgwAKPQOtBxUKQUWbQ1CfNQI8VXDR2sdWEp9xcASUVfRGFja8krNElcqLBy/0ibkcplptdviJjOFyv3FClvchdPiBgaWjmV+HSfg6pmuTdTGGFhaKa+ssapit8JiDCxB5S6uwSKRKiw68QR0yaeL5bRVswpLiK8xsNiosBR5BQCAeKVed1VYBNsDkk1YYWlz9pbnL/ERA2vXsyfEDfYCy913i5eJmDGj3fbRKRERwD33iLeradqvUmk5S4sVFiKb+HG2lcxDSlGlFnX1emiUliddaQyLf4kxHLS2wmJU6NPFciVT8wqLrxohPrYrLBfV4veBlZfb78KHQGNgqa8XBxR6edlvywpLm+tiZ7XbAq8AXAvAL9tYYbG3JlBiInDoUDvuoZNkMuDjjzt6L9pHVBQXjSNqASssrZR72bIbyFaVJb+8Dp66OqiqxOsJtXoMi1GhTxebXUK+uhqE+qoR6qeBr67W4jEAyFX6iptKi8TrlwDtE1i8vRsrJi11C7HC0uak1W6bLh6XqxYX9JNLx7w9ugPJOVJIUanEQc5EZIWBpZWaBpQLNsax5JfXNa7B4u1tER6cEhJicUK/5BNkcTLS+4pBpKUKy1mFOOPDOy9HvOCgTNY+fyRlMsfHsbDC0uakqc3mobauXo8L6ia/fwwsHU8KLJGRnNZPZIdL/zPWrFmD2NhYaDQaJCYmYteuXc22f/PNNxEfHw9PT0/07t0bH3zwgVWb1157Db1794anpyeioqKwcOFC1NXVubJ7btU0oOQ2CTC6BgOKq7QIrzRbNM7VabsKhUUlpNDHctBtnae49op/fS38PZV2x7Cchtg1I5NCQlBQ+wUFaXl+VljcrouNxeMKyutQ5B1o2dDZy0RQ25MCC7uDiOxyOrBs3rwZCxYswOLFi3HgwAGMGjUK48ePR05Ojs32aWlpSE1NxXPPPYejR4/i+eefx7x587Bt2zZTm40bN+KZZ57BsmXLcPz4cbzzzjvYvHkzUlNTXX9nbiIFlK7+GgDWAaawsg6CAHSraeUMIYlZt9AlnyDU1utRoxNP9tVqMbCECFrIZDKE2pglVFevxwWZBvVys2DQngMYpQpLS9cTYoWlzTUuz984hiW/vA6FxkG3JqywdLzx44EePYD77uvoPSHqtJwOLKtXr8bs2bMxZ84cxMfH47XXXkNUVBTS0tJstv/www/x8MMPIyUlBXFxcZg2bRpmz56NVatWmdrs2bMHI0aMwL333ouYmBgkJyfjnnvuwf79+11/Z26SZwwow3qIU0WbdhFJM4R61JeJG9owsJQal+WXuoUqlZ4AgGCDeIKyVWEpqtRCkMlx2ThTCIB7AgsrLG7XxcYYlksVNiosDCwdr08f4MwZ4KGHOnpPiDotpwKLTqdDVlYWkpOTLbYnJycjU7pwVxNarRYajcZim6enJ/bu3Yv6+noAwMiRI5GVlYW9xmtpnD17Funp6ZgwYYLdfdFqtaioqLC4uVuNrsG0xsXwOGNgaTIIV1qDJUpnHHDb2vK79PNyORAqjjuRuoXKjIElUC++ZogKUOvFIKD1Eqsv0poxZX5mJ63OEFhYYWlztsawFDQNLGp1478REVEn5lRgKS4uhl6vR1iTE1xYWBgKCmxfR2fcuHFYv349srKyIAgC9u/fjw0bNqC+vh7Fxhkq06ZNwwsvvICRI0dCqVSiR48eGDNmDJ555hm7+7Jy5Ur4+/ubblEd0PebZ6ym+Gk80LerWMFoWmGRpjQHCmI4a/US51KFJSwMgX5iQJGmrV5WiMHQv77W4isAFMvET9tFlWKYqQ4wu05MZwgsrLC0OVsXQCwor0O5xgcNHsarhYeH81IIRHRFcGnQrazJHzhBEKy2SZYuXYrx48dj2LBhUCqVmDJlCmbOnAkAUBhPTjt27MBLL72ENWvW4Ndff8XWrVvx1Vdf4YUXXrC7D6mpqSgvLzfdcnNzXXkrrSKFk8hAL0QGigNZS6p1pjElQGOFxV8KLM2tReIIqcISEWFV8i+SiScoadyKrFKs6lQrNSisFl9fqrDUdQlufM7OEFhYYWlz0qDb8tp60/WECsrrAJkMdUHGWWHsDiKiK4RTgSU4OBgKhcKqmlJYWGhVdZF4enpiw4YNqKmpQXZ2NnJychATEwNfX18EB4snzaVLl2L69OmYM2cO+vXrh9tvvx0rVqzAypUrYTAYbD6vWq2Gn5+fxc3dpKX4IwM94e+phJ9GPNmaV1mkMSw+UmDx9Gzdi44ZIwaMO+80G1QpBpZCiAHGS1sttjV2k1WpvVBkDCqFFeJXfYjZ1Xk7Q2BhhaXNBXiprK4nJF3XqiHU+G/OwEJEVwinAotKpUJiYiIyMjIstmdkZCApKanZn1UqlYiMjIRCocCmTZswceJEyI3rDdTU1Ji+lygUCgiCAKGlJd07kBRMorqIVROpymI+U0iqsHjp2yiwXHMNkJ8P/O1vVkuv50Ms82tqmwQWlReKqoyBxdglZBFSOkNgYYWlzSnkMgR6WXYLXTIGFpkUVDilmYiuEE6fHRYtWoTp06dj8ODBGD58ONauXYucnBzMnTsXgNhVk5eXZ1pr5dSpU9i7dy+GDh2K0tJSrF69GkeOHMH7779ves5JkyZh9erVGDhwIIYOHYozZ85g6dKlmDx5sqnbqDOSVrmNDPQ0fT2WX4Hcy9YVFq8G49TS1nYJAaYxB9KgSmngb55eDCyq6iqxnTGwVKo9Gyssxq/KrmafrDtDYGGFpV108VahpFqHy9U66A2C6d/f45o4IANAdHTH7iARkYOcDiwpKSkoKSnB8uXLkZ+fj4SEBKSnpyPa+IcvPz/fYk0WvV6PV199FSdPnoRSqcSYMWOQmZmJmJgYU5slS5ZAJpNhyZIlyMvLQ0hICCZNmoSXXnqp9e+wHZkqLMbKilRpkSosDXqDqaKhrjcGltZWWMyYr7MhCALON4iBRVFVKV5s0KzCUtikS0gTaTa9uj27BVhh6VCNV2zWorhKC71BgEIug+ZvqUBcDDB7dsfuIBGRg1w6Ozz66KN49NFHbT723nvvWdyPj4/HgQMHmt8JDw8sW7YMy5Ytc2V3OowUTCK7NFZYxO1ikCms1MIgAB5yGTx0bR9YpEG3l6t1qKhrQKlxlpCsvh7Qam2PYTF+9e5uFlhCzcaztDWudNuhgn0af0ekal+orxqKyG7AE0905K4RETmFH2ddVKVtQGmNOC5FGrsiVVqkwbjS+JUwPw1kNcZxLW3RJWRk3iVUVKlFtcpsvZuKCssxLJVaNOgNpinQ/v3jxZU1Y2PFy9u3F1ZYOpT5OCfz30cioisNzw4ukqorgV5K+JQUAtu3I+rG24yPiRUW6RNthL8GqDWOa2mXLiGdaQXbarUXvLU1FoGl0lhhKanWQRAAuQwICvIHTp5s/wutObo0Pyss7cL8ekLSgNtwBhYiugLxsqAuunC5cQ0WPPssMGsWon9IBwCU1dSjsq7etGhceDsFFulkVFuvR85lcWaQdAFEWxUW6YQV7KOGQi4Tw0F7LxomBRatFmjuYpassLQLqQpXUqU1TWkO92dgIaIrDwOLi8zXYEF+PgBAU1KEQC+xe+VCaa1lhaUduoR81B5QKcR/whMF4iJxOm/jSrrmgUXtCZ3egDOF4uyhUD91m+1Dyzvp01jFaa5biBWWdhHkbT2GhYGFiK5EDCwusliDRQojNTVma7HUmsYMhPtpGqsLbVhhkclkpk/QJ/LFwKL38RUfNAssDd7itqMXxfuhvm48Ycnljg28ZYWlXZh3CRWUs0uIiK5cDCwusliDxSywRBlnDOVerjF1CUWanx/aMLAAjSekk5fEwCL4WgcWub+4CvDRi+I4klBfN1ZYAMcG3rLC0i7ML4AodQly0C0RXYn4cdZFFmuw2KmwmLqEVGar9bZTYJFWu5X5GasZZoFFGRgAADhmqrB0wsDCCku7kAZml9XUo1YnHuMIdgkR0RWIFRYXXSi1U2ExrsVyvqQal4xrnoQrjYHFw6PNpxBL62xIPAKsA4u6S4C4qU6sYoS4+xM2KywdJsBLZRpXrW0Qr8vFMSxEdCViYHFBeW296eTfrUlgkSosB3PLTKuKBimMF3Bs4+oK0FhhkaiM4cQ8sHgGB1q0CfFhheWPwvx6QgDg76mERslQSERXHgYWF0jVlWAfFbxUHk0CixhKpIvNhfqqoagzTmluwxlCkqaBRRMUIH5jFli8Q7pYtHHrLCGAFZYOZv47wu4gIrpSMbC4QLq4YbdAL/GaPdXGqyNXV5sqLJL2WjROEmR2MlLIZdAEGasply+bgpR/aJDFz7h9DAtnCXUo898RDrgloisVA4sLLMav1Nc3nmxrauCpUiDYx/wTrWf7Bhaz7p1gHxXkUjjIyzNtDwi3rLCEdMZBt6ywtJsgs99HTmkmoisVA4sLbM4QAkzfm1dZwttp0TiJebk/xFcN+IlTmE2BRa1GcJCfqU2AlxJqDzeHAo5h6VDmvyNh7BIioisUA4sLbM4QAswCS2MlxZ1dQqG+GuvA4udnMcjW7d1BgGPXE2KFpd1IV/UGOIaFiK5cDCwusLnKLWC/wtKuXUJmFRYfswqLtLKunx8CvVTwkItzW926yq2EFZYOFcwuISK6CjCwOEkQBFNgsVdhkVa7BZpUWNqhS8j8ekIWXUISPz/I5TLTei0dWmHhGJYO0YWDbonoKsDA4qSymnpUaY1rsATY6xJqDCYR/mZt2qHCIpPJTCcke4HF9BiAEHdPaQZYYelgnNZMRFcDBhYn7cu+DECsVGiUisYpzQCg1QJ6Pbp3EQOLQi4Tg0I7dgkBQLCveEIK81MD0rWEJMb7YcagEtZZu4RYYWk30hgmjVKOAK+2XWmZiMhd+HHWCVXaBjy/7RgAYPKAruJG8wqL8X5MkA9mjYhBuJ8GSoW83QPLX8Zcg68PF+CGXiEADJYPGisss0fGQeUhx8T+Ee2yD82SAktdnXjT2AhNrLC0m2tCfTB9WDTiQrwhk9bpJyK6wvDs4IR/fHcSeWW1iAz0xMJbeokbbQQWma8vlk261mIbgHYZwwIAtyZE4NYEsyCi0VgMugWA4T2CMLxHkI2fdgNfX0AmExfZKy+3HVhYYWk3MpkML0xN6OjdICJqFXYJOWh/9mW8vycbAPDyHf3hrTZmPRuBxUo7V1ismI9jaTqmpSPI5Y37Ya9biBUWIiJqBgOLA+rq9Xjqs98gCMDdgyMxsmdw44MMLI5paRwLKyxERNQMBhYHvP7DaZwtqkaIrxqLb+tr+aAjgaWdu4SsXImBhRUWIiJqBgNLC45eLMdbO88CAF6YkgD/prMsWGFxjDRuRau1/TgrLERE1AwGlmbU6w14astv0BsETOgXgVsTwq0bmU9rtnUfYGABAKUx6NXX236cFRYiImoGA0sLbr02HCG+ajw3+VrbDdgl5BgpiEiVlKZYYSEiombw42wzlAo5/npTT8wZFQdPlZ0TKbuEHCNVWOwFFlZYiIioGaywOMBuWAEYWBwlBRF7XUKssBARUTMYWFqLXUKOaalLiBUWIiJqBgNLa0lhxMfH8r45VlhaHnTLCgsRETWDgaW1pIASHGx531xHBRa53H1VnZawwkJERK3AwNJa0jRmKbDYmtbcUV1Cfn7iNXw6A1ZYiIioFRhYWqulCkt9fWP1wN0VFl9f97yeI1hhISKiVnApsKxZswaxsbHQaDRITEzErl27mm3/5ptvIj4+Hp6enujduzc++OADqzZlZWWYN28eIiIioNFoEB8fj/T0dFd2z72kgBISYnlfInUHAe4LLEOGAIMHAw8+6J7Xc0RL05pZYSEiomY4/XF28+bNWLBgAdasWYMRI0bg7bffxvjx43Hs2DF0797dqn1aWhpSU1Oxbt06DBkyBHv37sVDDz2EwMBATJo0CQCg0+lwyy23IDQ0FFu2bEFkZCRyc3Ph25kqBPa0FFjM70vL07c3Hx9g3z73vJajWprWzAoLERE1w+mzw+rVqzF79mzMmTMHAPDaa6/hu+++Q1paGlauXGnV/sMPP8TDDz+MlJQUAEBcXBx+/vlnrFq1yhRYNmzYgMuXLyMzMxNK4yfx6Ohol9+UW7XUJWQ+4LazjCfpCKywEBFRKzjVJaTT6ZCVlYXk5GSL7cnJycjMzLT5M1qtFpomlQVPT0/s3bsX9cZP219++SWGDx+OefPmISwsDAkJCVixYgX00qduO89bUVFhcXO7hgZApxO/dySw/JGxwkJERK3gVGApLi6GXq9HWFiYxfawsDAUFBTY/Jlx48Zh/fr1yMrKgiAI2L9/PzZs2ID6+noUFxcDAM6ePYstW7ZAr9cjPT0dS5YswauvvoqXXnrJ7r6sXLkS/v7+pltUVJQzb6VtmI9PsRdY3D1DqLPitYSIiKgVXBp0K2vStSEIgtU2ydKlSzF+/HgMGzYMSqUSU6ZMwcyZMwEACuPJyWAwIDQ0FGvXrkViYiKmTZuGxYsXIy0tze4+pKamory83HTLzc115a20jjSFWSYDAgMtt0lYYRHxas1ERNQKTgWW4OBgKBQKq2pKYWGhVdVF4unpiQ0bNqCmpgbZ2dnIyclBTEwMfH19EWysSkRERKBXr16mAAMA8fHxKCgogE7qcmlCrVbDz8/P4uZ25tUTb2/LbRIGFhErLERE1ApOBRaVSoXExERkZGRYbM/IyEBSUlKzP6tUKhEZGQmFQoFNmzZh4sSJkMvFlx8xYgTOnDkDg8Fgan/q1ClERERApVI5s4vu5UhgYZeQiBUWIiJqBae7hBYtWoT169djw4YNOH78OBYuXIicnBzMnTsXgNhVM2PGDFP7U6dO4aOPPsLp06exd+9eTJs2DUeOHMGKFStMbR555BGUlJRg/vz5OHXqFL7++musWLEC8+bNa4O32I6kMOLt3RhIWGGxjRUWIiJqBac/zqakpKCkpATLly9Hfn4+EhISkJ6ebpqGnJ+fj5ycHFN7vV6PV199FSdPnoRSqcSYMWOQmZmJmJgYU5uoqChs374dCxcuRP/+/dGtWzfMnz8fTz/9dOvfYXsyr55IgaW2FjAYxOv4SPcBBpaWpjWzwkJERM1w6ezw6KOP4tFHH7X52HvvvWdxPz4+HgcOHGjxOYcPH46ff/7Zld3pOLYCCwDU1VlXXP7oXUItTWtmhYWIiJrBawm1hnkYMa+gmHcLscIi4rWEiIioFRhYWkOawuzlJVYG1GrL7QADi4RXayYiolZgYGmNpt09tgbesktIxAoLERG1AgNLazQNI7amNrPCImKFhYiIWoGBpTXMpzUDtissDCyi5iosgiDOrDJvR0REZIaBpTXYJeS45ios5he5ZIWFiIhsYGBpDUcCCyssouYqLOaBhRUWIiKygYGlNcxnCZl/ZWCx1tzCcebbWGEhIiIbGFhaw16Fxda05j96l1BzC8exwkJERC1gYGkNR2YJSd//0SsszXUJscJCREQtYGBpDY5hcRwH3RIRUSswsLSGM9Oa2SUkfm2uwiKXAzKZ+/aJiIiuGAwsreHMtGZWWMSvzVVYWF0hIiI7GFhag11CjnOkwsIBt0REZAcDS2s4M635j94l1Ny0ZlZYiIioBQwsrdHStGaDAairE79nhUX8aqtLiBUWIiJqAQNLa7Q0rVkKKwADCyssRETUCgwsrjKvntjrEpK6gwAGFlZYiIioFRhYXGUeRuxNa5a+KpU8GTtyLSFWWIiIyA4GFleZD6zVaMSv9iosf/TqCmDZJSQIlo+xwkJERC1gYHGV+foqcuNhtBdY/ugzhADLMGK+sq35fVZYiIjIDgYWVzWd0mz+fdMuIVZYGissgPU4FlZYiIioBQwsrmo6Q8j8eynMsEuokXkYaTqOhRUWIiJqAQOLq2wFFvNpzYLALiFz5hWWpoGFFRYiImoBA4urmquwCAKg1bJLyJx59aRplxArLERE1AIGFlfZCizmwaSmhl1C5mSyxkDCCgsRETmJgcVVUmCRuoEAsdtD6vowDyzsEhLZu2IzKyxERNQCBhZX2aqwmN+vqWGXUFP2Fo9jhYWIiFrAwOIqW9Oaze+zS8gaKyxEROQiBhZX2auwSF1E1dXsEmqKFRYiInIRA4ur2CXkPFZYiIjIRQwsrnIksLBLyBIrLERE5CIGFlc5E1jYJSQyvwCiOVZYiIioBS4FljVr1iA2NhYajQaJiYnYtWtXs+3ffPNNxMfHw9PTE71798YHH3xgt+2mTZsgk8kwdepUV3bNfWxNawbYJdQcqYLCawkREZGTnD5DbN68GQsWLMCaNWswYsQIvP322xg/fjyOHTuG7t27W7VPS0tDamoq1q1bhyFDhmDv3r146KGHEBgYiEmTJlm0PX/+PJ544gmMGjXK9XfkLuwScp69LiFWWIiIqAVOV1hWr16N2bNnY86cOYiPj8drr72GqKgopKWl2Wz/4Ycf4uGHH0ZKSgri4uIwbdo0zJ49G6tWrbJop9frcd999+H5559HXFyca+/GnZyZ1swuIZG9QbessBARUQucCiw6nQ5ZWVlITk622J6cnIzMzEybP6PVaqHRaCy2eXp6Yu/evag3O3EtX74cISEhmD17tkP7otVqUVFRYXFzK0emNbNLyBIrLERE5CKnAktxcTH0ej3CwsIstoeFhaGgoMDmz4wbNw7r169HVlYWBEHA/v37sWHDBtTX16O4uBgA8NNPP+Gdd97BunXrHN6XlStXwt/f33SLiopy5q20HruEnMcKCxERucilQbcymcziviAIVtskS5cuxfjx4zFs2DAolUpMmTIFM2fOBAAoFApUVlbi/vvvx7p16xAcHOzwPqSmpqK8vNx0y83NdeWtuI6zhJzHCgsREbnIqY+0wcHBUCgUVtWUwsJCq6qLxNPTExs2bMDbb7+NS5cuISIiAmvXroWvry+Cg4Px22+/ITs722IArsFgEHfOwwMnT55Ejx49rJ5XrVZDrVY7s/ttiwvHOc/etGZWWIiIqAVOVVhUKhUSExORkZFhsT0jIwNJSUnN/qxSqURkZCQUCgU2bdqEiRMnQi6Xo0+fPjh8+DAOHjxouk2ePBljxozBwYMH3d/V4yhHpjWzS8iSvWnNrLAQEVELnP5Iu2jRIkyfPh2DBw/G8OHDsXbtWuTk5GDu3LkAxK6avLw801orp06dwt69ezF06FCUlpZi9erVOHLkCN5//30AgEajQUJCgsVrBAQEAIDV9k6FXULO40q3RETkIqfPECkpKSgpKcHy5cuRn5+PhIQEpKenIzo6GgCQn5+PnJwcU3u9Xo9XX30VJ0+ehFKpxJgxY5CZmYmYmJg2exNuJwicJeQKXkuIiIhc5NJH2kcffRSPPvqozcfee+89i/vx8fE4cOCAU8/f9Dk6nbo6MbQA9iss5eWAcSwOA4sRKyxEROQiXkvIFVLlBLAOI1JgMU7Zttnmj4oVFiIichEDiyukwKJSWVcFmgYWmQzoyNlMnQkrLERE5CIGFlfYG79ivs18hpCdNWr+cFhhISIiFzGwuMLelGbAOsSwO6gRKyxEROQiBhZXOFJhsXf/j8zewnGssBARUQsYWFxh70rNgHXVhRWWRvYWjmOFhYiIWsDA4gpnKiwMLI14LSEiInIRA4srmgssSqXliZddQo14tWYiInIRA4srmgssMpnldlZYGrHCQkRELmJgaUlGBrBkCVBS0ritucDSdDsDSyNWWIiIyEU8Q7RkwQLg2DFgyBBgyhRxW3PTmgHLwMIuoUassBARkYtYYWnJqFHi1x9/bNzGCotr7E1rZoWFiIhawMDSkhtuEL+aB5bmpjUDlpUXBpZG9qY1s8JCREQtYGBpiVRh+fVXoLJS/N6ZCgu7hBqxwkJERC5iYGlJVBQQGwsYDMCePeI2dgm5hhUWIiJyEQOLI5p2CzGwuIbXEiIiIhcxsDii6cBbdgm5hldrJiIiFzGwOEKqsPzyC1BX59y0ZlZYGrU0rZkVFiIisoOBxRHXXAOEhwM6HbB3b8sVFs4Ssq2lheNYYSEiIjsYWBwhk1mOY2lpWjO7hGxjhYWIiFzEwOIoKbDs2sVBt65qaVozKyxERGQHP9I6Shp4+9NPgEYjfs/A4pyWpjWzwkJERHbwDOGohAQgIAAoK2OXkKtamtbMCgsREdnBLiFHyeWNVRYJKyzOaWlaMyssRERkBwOLM6RxLBJOa3YOKyxEROQiBhZnOFphMQ8y7BJqxAoLERG5iIHFGYMGNQYQD4/GE3BTrLDYxgoLERG5iIHFGUolkJQkft9c5YSBxTZWWIiIyEUMLM6SxrEwsDiPFRYiInIRA4uzxo4Vv4aE2G8TFCTOKurShSdhc/YWjmOFhYiIWsAzhLOSkoB33xXXZbEnOBj49FMxsFAjewvHscJCREQtYGBxlkwGzJzZcrs77mj3Xbni8FpCRETkIpe6hNasWYPY2FhoNBokJiZi165dzbZ/8803ER8fD09PT/Tu3RsffPCBxePr1q3DqFGjEBgYiMDAQNx8883Yu3evK7tGnRmv1kxERC5yOrBs3rwZCxYswOLFi3HgwAGMGjUK48ePR05Ojs32aWlpSE1NxXPPPYejR4/i+eefx7x587Bt2zZTmx07duCee+7B//73P+zZswfdu3dHcnIy8vLyXH9n1PlIFRSDQbwBgCCwwkJERC2SCYIgOPMDQ4cOxaBBg5CWlmbaFh8fj6lTp2LlypVW7ZOSkjBixAj8/e9/N21bsGAB9u/fj927d9t8Db1ej8DAQLzxxhuYMWOGQ/tVUVEBf39/lJeXw8/Pz5m3RO5SVgYEBorfa7WASiWGFSmoFBeLA5aJiOgPw9Hzt1MVFp1Oh6ysLCQnJ1tsT05ORmZmps2f0Wq10EhXNzby9PTE3r17Ud+0a8CopqYG9fX16NLMoFWtVouKigqLG3Vy5hUUqRtIqq40fZyIiMiMU4GluLgYer0eYWFhFtvDwsJQUFBg82fGjRuH9evXIysrC4IgYP/+/diwYQPq6+tRXFxs82eeeeYZdOvWDTfffLPdfVm5ciX8/f1Nt6ioKGfeCnUE85WBpcBiPgCXY1iIiMgOlwbdymQyi/uCIFhtkyxduhTjx4/HsGHDoFQqMWXKFMw0zrJR2DhBvfLKK/jkk0+wdetWq8qMudTUVJSXl5tuubm5rrwVcifzCopUXWOFhYiIHOBUYAkODoZCobCqphQWFlpVXSSenp7YsGEDampqkJ2djZycHMTExMDX1xfBwcEWbf/xj39gxYoV2L59O/r379/svqjVavj5+VncqJOTm/26scJCREROcCqwqFQqJCYmIiMjw2J7RkYGkqRr7NihVCoRGRkJhUKBTZs2YeLEiZCbncD+/ve/44UXXsC3336LwYMHO7NbdKWQyaynNptXWBhYiIjIDqdr8IsWLcL06dMxePBgDB8+HGvXrkVOTg7mzp0LQOyqycvLM621curUKezduxdDhw5FaWkpVq9ejSNHjuD99983Pecrr7yCpUuX4uOPP0ZMTIypguPj4wMfH5+2eJ/UWXh4iGGlaYVFJrOswBAREZlxOrCkpKSgpKQEy5cvR35+PhISEpCeno7o6GgAQH5+vsWaLHq9Hq+++ipOnjwJpVKJMWPGIDMzEzExMaY2a9asgU6nw1133WXxWsuWLcNzzz3n2jujzkmpBGprrSssHL9CRETNcHodls6K67BcIYKCgMuXgWPHgPh44Px5ICYG0GjEIENERH8o7bIOC1Gr2RvDwgoLERE1g4GF3KvpBRB5HSEiInIAAwu5l1RhabrSLSssRETUDAYWci8pmEhdQqywEBGRAxhYyL2adgmxwkJERA5gYCH3ajrolhUWIiJyAAMLuRcrLERE5AIGFnIvVliIiMgFDCzkXqywEBGRCxhYyL1YYSEiIhcwsJB7scJCREQuYGAh92q6cBwrLERE5AAGFnKvpgvHscJCREQOYGAh9+K1hIiIyAUMLORevFozERG5gIGF3IsVFiIicgEDC7kXKyxEROQCBhZyL1ZYiIjIBQws5F5NpzWzwkJERA5gYCH3ajqtmRUWIiJyAAMLuRdXuiUiIhcwsJB78VpCRETkAgYWci9WWIiIyAUMLORerLAQEZELGFjIvVhhISIiFzCwkHuxwkJERC5gYCH3YoWFiIhcwMBC7tV04ThWWIiIyAEMLOReTReOY4WFiIgcwMBC7sVrCRERkQsYWMi9eLVmIiJyAQMLuRcrLERE5AIGFnIvVliIiMgFLgWWNWvWIDY2FhqNBomJidi1a1ez7d98803Ex8fD09MTvXv3xgcffGDV5rPPPkPfvn2hVqvRt29ffP75567sGnV2rLAQEZELnA4smzdvxoIFC7B48WIcOHAAo0aNwvjx45GTk2OzfVpaGlJTU/Hcc8/h6NGjeP755zFv3jxs27bN1GbPnj1ISUnB9OnTcejQIUyfPh133303fvnlF9ffGXVOTac1s8JCREQOkAmCIDjzA0OHDsWgQYOQlpZm2hYfH4+pU6di5cqVVu2TkpIwYsQI/P3vfzdtW7BgAfbv34/du3cDAFJSUlBRUYFvvvnG1ObWW29FYGAgPvnkE4f2q6KiAv7+/igvL4efn58zb4ncads2YPJk4PrrgV9+AWbPBjZsAFasAFJTO3rviIjIzRw9fztVYdHpdMjKykJycrLF9uTkZGRmZtr8Ga1WC41GY7HN09MTe/fuRb1xHMOePXusnnPcuHF2n1N63oqKCosbXQG40i0REbnAqcBSXFwMvV6PsLAwi+1hYWEoKCiw+TPjxo3D+vXrkZWVBUEQsH//fmzYsAH19fUoLi4GABQUFDj1nACwcuVK+Pv7m25RUVHOvBXqKLyWEBERucClQbcymcziviAIVtskS5cuxfjx4zFs2DAolUpMmTIFM2fOBAAozE5SzjwnAKSmpqK8vNx0y83NdeWtkLuxwkJERC5wKrAEBwdDoVBYVT4KCwutKiQST09PbNiwATU1NcjOzkZOTg5iYmLg6+uL4OBgAEB4eLhTzwkAarUafn5+Fje6ArDCQkRELnAqsKhUKiQmJiIjI8Nie0ZGBpKSkpr9WaVSicjISCgUCmzatAkTJ06EXC6+/PDhw62ec/v27S0+J12BWGEhIiIXOH2WWLRoEaZPn47Bgwdj+PDhWLt2LXJycjB37lwAYldNXl6eaa2VU6dOYe/evRg6dChKS0uxevVqHDlyBO+//77pOefPn48bbrgBq1atwpQpU/Cf//wH33//vWkWEV1FWGEhIiIXOB1YUlJSUFJSguXLlyM/Px8JCQlIT09HdHQ0ACA/P99iTRa9Xo9XX30VJ0+ehFKpxJgxY5CZmYmYmBhTm6SkJGzatAlLlizB0qVL0aNHD2zevBlDhw5t/TukzoUVFiIicoHT67B0VlyH5Qpx/DjQty8QFAQUFwPjxgHbtwMffABMn97Re0dERG7WLuuwELWaVEnhtYSIiMgJDCzkXryWEBERuYCBhdyLV2smIiIXMLCQe7HCQkRELmBgIfeSKiyCIFZXWGEhIiIHMLCQe5kHk4YGVliIiMghDCzkXlKFBRDHsbDCQkREDmBgIfdihYWIiFzAwELu1TSwsMJCREQOYGAh95LLxRsgdgmxwkJERA5gYCH3M5/azAoLERE5gIGF3M988ThWWIiIyAEMLOR+rLAQEZGTGFjI/VhhISIiJzGwkPuxwkJERE5iYCH3kyosXIeFiIgcxMBC7idVU7jSLREROYiBhdzPvEuIFRYiInIAAwu5n/mgW1ZYiIjIAQws5H6ssBARkZMYWMj9pAqLVtu4jRUWIiJqBgMLuZ8UTurqGrexwkJERM1gYCH3kyos5oGFFRYiImoGAwu5HyssRETkJAYWcj+OYSEiIicxsJD7scJCREROYmAh92saWGQyQM5fRSIiso9nCXK/poNuWV0hIqIWMLCQ+zWtsHD8ChERtYCBhdyPFRYiInISAwu5HyssRETkJAYWcj9WWIiIyEkMLOR+rLAQEZGTXAosa9asQWxsLDQaDRITE7Fr165m22/cuBEDBgyAl5cXIiIiMGvWLJSUlFi0ee2119C7d294enoiKioKCxcuRJ35Oh109WgaWFhhISKiFjgdWDZv3owFCxZg8eLFOHDgAEaNGoXx48cjJyfHZvvdu3djxowZmD17No4ePYpPP/0U+/btw5w5c0xtNm7ciGeeeQbLli3D8ePH8c4772Dz5s1ITU11/Z1R59W0S4gVFiIiaoHTgWX16tWYPXs25syZg/j4eLz22muIiopCWlqazfY///wzYmJi8NhjjyE2NhYjR47Eww8/jP3795va7NmzByNGjMC9996LmJgYJCcn45577rFoQ1cRVliIiMhJTgUWnU6HrKwsJCcnW2xPTk5GZmamzZ9JSkrChQsXkJ6eDkEQcOnSJWzZsgUTJkwwtRk5ciSysrKwd+9eAMDZs2eRnp5u0aYprVaLiooKixtdIVhhISIiJzl1piguLoZer0dYWJjF9rCwMBQUFNj8maSkJGzcuBEpKSmoq6tDQ0MDJk+ejNdff93UZtq0aSgqKsLIkSMhCAIaGhrwyCOP4JlnnrG7LytXrsTzzz/vzO5TZ8EKCxEROcmlQbcymcziviAIVtskx44dw2OPPYZnn30WWVlZ+Pbbb3Hu3DnMnTvX1GbHjh146aWXsGbNGvz666/YunUrvvrqK7zwwgt29yE1NRXl5eWmW25uritvhToCpzUTEZGTnKqwBAcHQ6FQWFVTCgsLraoukpUrV2LEiBF48sknAQD9+/eHt7c3Ro0ahRdffBERERFYunQppk+fbhqI269fP1RXV+PPf/4zFi9eDLmNC+Op1Wqo1Wpndp86C05rJiIiJzlVYVGpVEhMTERGRobF9oyMDCQlJdn8mZqaGqvAoTB+ohYEodk2giCY2tBVRKqwaLXiV1ZYiIioBU5/tF20aBGmT5+OwYMHY/jw4Vi7di1ycnJMXTypqanIy8vDBx98AACYNGkSHnroIaSlpWHcuHHIz8/HggULcP3116Nr166mNqtXr8bAgQMxdOhQnDlzBkuXLsXkyZNN4YauIqywEBGRk5w+U6SkpKCkpATLly9Hfn4+EhISkJ6ejujoaABAfn6+xZosM2fORGVlJd544w08/vjjCAgIwNixY7Fq1SpTmyVLlkAmk2HJkiXIy8tDSEgIJk2ahJdeeqkN3iJ1Ohx0S0RETpIJV0mfS0VFBfz9/VFeXg4/P7+O3h1qztq1wMMPN94fORJoYbVkIiK6Ojl6/ua1hMj9mnYBscJCREQtYGAh95MG3Uo4hoWIiFrAwELuxwoLERE5iYGF3I8VFiIichIDC7kfKyxEROQkBhZyP1ZYiIjISQws5H6ssBARkZMYWMj9mgYWVliIiKgFDCzkfk27hFhhISKiFjCwkPuxwkJERE5iYCH3Y4WFiIicxMBC7scKCxEROYmBhdyPFRYiInISAwu5HyssRETkJAYWcj+uw0JERE5iYCH340q3RETkJAYWcj9WWIiIyEkMLOR+rLAQEZGTGFjI/VhhISIiJzGwkPuxwkJERE5iYCH3Y4WFiIicxMBC7scKCxEROYmBhdyvaUWFFRYiImoBAwu5n0xmGVJYYSEiohYwsFDHMO8WYoWFiIhawMBCHcO8qsIKCxERtYCBhToGKyxEROQEBhbqGKywEBGRExhYqGOwwkJERE5gYKGOwQoLERE5gYGFOoZ5SGGFhYiIWsDAQh3DvEuIFRYiImqBS4FlzZo1iI2NhUajQWJiInbt2tVs+40bN2LAgAHw8vJCREQEZs2ahZKSEos2ZWVlmDdvHiIiIqDRaBAfH4/09HRXdo+uBKywEBGRE5wOLJs3b8aCBQuwePFiHDhwAKNGjcL48eORk5Njs/3u3bsxY8YMzJ49G0ePHsWnn36Kffv2Yc6cOaY2Op0Ot9xyC7Kzs7FlyxacPHkS69atQ7du3Vx/Z9S5scJCREROcPpMsXr1asyePdsUOF577TV89913SEtLw8qVK63a//zzz4iJicFjjz0GAIiNjcXDDz+MV155xdRmw4YNuHz5MjIzM6E0nsiio6NdekN0hWCFhYiInOBUhUWn0yErKwvJyckW25OTk5GZmWnzZ5KSknDhwgWkp6dDEARcunQJW7ZswYQJE0xtvvzySwwfPhzz5s1DWFgYEhISsGLFCuj1erv7otVqUVFRYXGjKwgrLERE5ASnAktxcTH0ej3CwsIstoeFhaGgoMDmzyQlJWHjxo1ISUmBSqVCeHg4AgIC8Prrr5vanD17Flu2bIFer0d6ejqWLFmCV199FS+99JLdfVm5ciX8/f1Nt6ioKGfeCnU0VliIiMgJLg26lclkFvcFQbDaJjl27Bgee+wxPPvss8jKysK3336Lc+fOYe7cuaY2BoMBoaGhWLt2LRITEzFt2jQsXrwYaWlpdvchNTUV5eXlpltubq4rb4U6CissRETkBKfOFMHBwVAoFFbVlMLCQquqi2TlypUYMWIEnnzySQBA//794e3tjVGjRuHFF19EREQEIiIioFQqoTD7pB0fH4+CggLodDqoVCqr51Wr1VCr1c7sPnUmrLAQEZETnKqwqFQqJCYmIiMjw2J7RkYGkpKSbP5MTU0N5HLLl5GCiSAIAIARI0bgzJkzMBgMpjanTp1CRESEzbBCVwGudEtERE5wukto0aJFWL9+PTZs2IDjx49j4cKFyMnJMXXxpKamYsaMGab2kyZNwtatW5GWloazZ8/ip59+wmOPPYbrr78eXbt2BQA88sgjKCkpwfz583Hq1Cl8/fXXWLFiBebNm9dGb5M6HV5LiIiInOD0R9uUlBSUlJRg+fLlyM/PR0JCAtLT003TkPPz8y3WZJk5cyYqKyvxxhtv4PHHH0dAQADGjh2LVatWmdpERUVh+/btWLhwIfr3749u3bph/vz5ePrpp9vgLVKnxAoLERE5QSZI/TJXuIqKCvj7+6O8vBx+fn4dvTvUknvuATZtEr//7TegX7+O3R8iIuoQjp6/eS0h6hissBARkRMYWKhjcAwLERE5gYGFOgYrLERE5AQGFuoYrLAQEZETGFioY7DCQkRETmBgoY7BlW6JiMgJDCzUMXgtISIicgIDC3UMVliIiMgJDCzUMVhhISIiJzCwUMdghYWIiJzAwEIdgxUWIiJyAgMLdQxWWIiIyAkMLNQxzAOLnL+GRETUPJ4pqGNIXUIKBSCTdey+EBFRp8fAQh1DqrBw/AoRETmAgYU6hnmFhYiIqAUMLNQxWGEhIiInMLBQx2CFhYiInMDAQh2DFRYiInICAwt1DFZYiIjICQws1DFYYSEiIicwsFDHkIIKKyxEROQABhbqGFKXECssRETkAAYW6hhduohfAwI6dDeIiOjKwI+31DH69wfefRe47rqO3hMiIroCMLBQx5DJgJkzO3oviIjoCsEuISIiIur0GFiIiIio02NgISIiok6PgYWIiIg6PQYWIiIi6vQYWIiIiKjTY2AhIiKiTs+lwLJmzRrExsZCo9EgMTERu3btarb9xo0bMWDAAHh5eSEiIgKzZs1CSUmJzbabNm2CTCbD1KlTXdk1IiIiugo5HVg2b96MBQsWYPHixThw4ABGjRqF8ePHIycnx2b73bt3Y8aMGZg9ezaOHj2KTz/9FPv27cOcOXOs2p4/fx5PPPEERo0a5fw7ISIioquW04Fl9erVmD17NubMmYP4+Hi89tpriIqKQlpams32P//8M2JiYvDYY48hNjYWI0eOxMMPP4z9+/dbtNPr9bjvvvvw/PPPIy4uzrV3Q0RERFclpwKLTqdDVlYWkpOTLbYnJycjMzPT5s8kJSXhwoULSE9PhyAIuHTpErZs2YIJEyZYtFu+fDlCQkIwe/Zsh/ZFq9WioqLC4kZERERXJ6cCS3FxMfR6PcLCwiy2h4WFoaCgwObPJCUlYePGjUhJSYFKpUJ4eDgCAgLw+uuvm9r89NNPeOedd7Bu3TqH92XlypXw9/c33aKiopx5K0RERHQFcWnQrUwms7gvCILVNsmxY8fw2GOP4dlnn0VWVha+/fZbnDt3DnPnzgUAVFZW4v7778e6desQHBzs8D6kpqaivLzcdMvNzXXlrRAREdEVwKmrNQcHB0OhUFhVUwoLC62qLpKVK1dixIgRePLJJwEA/fv3h7e3N0aNGoUXX3wRly5dQnZ2NiZNmmT6GYPBIO6chwdOnjyJHj16WD2vWq2GWq023RcEAQDYNURERHQFkc7b0nncHqcCi0qlQmJiIjIyMnD77bebtmdkZGDKlCk2f6ampgYeHpYvo1AoTDvXp08fHD582OLxJUuWoLKyEv/6178c7uqprKwEAHYNERERXYEqKyvh7+9v93GnAgsALFq0CNOnT8fgwYMxfPhwrF27Fjk5OaYuntTUVOTl5eGDDz4AAEyaNAkPPfQQ0tLSMG7cOOTn52PBggW4/vrr0bVrVwBAQkKCxWsEBATY3N6crl27Ijc3F76+vna7p1xRUVGBqKgo5Obmws/Pr82el6zxWLsPj7V78Xi7D4+1+7TVsRYEAZWVlaZMYI/TgSUlJQUlJSVYvnw58vPzkZCQgPT0dERHRwMA8vPzLdZkmTlzJiorK/HGG2/g8ccfR0BAAMaOHYtVq1Y5+9LNksvliIyMbNPnNOfn58dffjfhsXYfHmv34vF2Hx5r92mLY91cZUUiE1rqNPqDq6iogL+/P8rLy/nL3854rN2Hx9q9eLzdh8fafdx9rHktISIiIur0GFhaoFarsWzZMosZSdQ+eKzdh8favXi83YfH2n3cfazZJURERESdHissRERE1OkxsBAREVGnx8BCREREnR4DCxEREXV6DCxERETU6TGwtGDNmjWIjY2FRqNBYmIidu3a1dG7dEVbuXIlhgwZAl9fX4SGhmLq1Kk4efKkRRtBEPDcc8+ha9eu8PT0xI033oijR4920B5fPVauXAmZTIYFCxaYtvFYt628vDzcf//9CAoKgpeXF6677jpkZWWZHufxbhsNDQ1YsmQJYmNj4enpibi4OCxfvtx04VyAx9pVP/74IyZNmoSuXbtCJpPhiy++sHjckeOq1Wrx17/+FcHBwfD29sbkyZNx4cKF1u+cQHZt2rRJUCqVwrp164Rjx44J8+fPF7y9vYXz58939K5dscaNGye8++67wpEjR4SDBw8KEyZMELp37y5UVVWZ2rz88suCr6+v8NlnnwmHDx8WUlJShIiICKGioqID9/zKtnfvXiEmJkbo37+/MH/+fNN2Huu2c/nyZSE6OlqYOXOm8Msvvwjnzp0Tvv/+e+HMmTOmNjzebePFF18UgoKChK+++ko4d+6c8Omnnwo+Pj7Ca6+9ZmrDY+2a9PR0YfHixcJnn30mABA+//xzi8cdOa5z584VunXrJmRkZAi//vqrMGbMGGHAgAFCQ0NDq/aNgaUZ119/vTB37lyLbX369BGeeeaZDtqjq09hYaEAQNi5c6cgCIJgMBiE8PBw4eWXXza1qaurE/z9/YW33nqro3bzilZZWSn07NlTyMjIEEaPHm0KLDzWbevpp58WRo4cafdxHu+2M2HCBOHBBx+02HbHHXcI999/vyAIPNZtpWlgceS4lpWVCUqlUti0aZOpTV5eniCXy4Vvv/22VfvDLiE7dDodsrKykJycbLE9OTkZmZmZHbRXV5/y8nIAQJcuXQAA586dQ0FBgcVxV6vVGD16NI+7i+bNm4cJEybg5ptvttjOY922vvzySwwePBh/+tOfEBoaioEDB2LdunWmx3m8287IkSPx3//+F6dOnQIAHDp0CLt378Ztt90GgMe6vThyXLOyslBfX2/RpmvXrkhISGj1sXf6as1/FMXFxdDr9QgLC7PYHhYWhoKCgg7aq6uLIAhYtGgRRo4ciYSEBAAwHVtbx/38+fNu38cr3aZNm/Drr79i3759Vo/xWLets2fPIi0tDYsWLcLf/vY37N27F4899hjUajVmzJjB492Gnn76aZSXl6NPnz5QKBTQ6/V46aWXcM899wDg73Z7ceS4FhQUQKVSITAw0KpNa8+dDCwtkMlkFvcFQbDaRq75y1/+gt9++w27d++2eozHvfVyc3Mxf/58bN++HRqNxm47Huu2YTAYMHjwYKxYsQIAMHDgQBw9ehRpaWmYMWOGqR2Pd+tt3rwZH330ET7++GNce+21OHjwIBYsWICuXbvigQceMLXjsW4frhzXtjj27BKyIzg4GAqFwioRFhYWWqVLct5f//pXfPnll/jf//6HyMhI0/bw8HAA4HFvA1lZWSgsLERiYiI8PDzg4eGBnTt34t///jc8PDxMx5PHum1ERESgb9++Ftvi4+ORk5MDgL/bbenJJ5/EM888g2nTpqFfv36YPn06Fi5ciJUrVwLgsW4vjhzX8PBw6HQ6lJaW2m3jKgYWO1QqFRITE5GRkWGxPSMjA0lJSR20V1c+QRDwl7/8BVu3bsUPP/yA2NhYi8djY2MRHh5ucdx1Oh127tzJ4+6km266CYcPH8bBgwdNt8GDB+O+++7DwYMHERcXx2PdhkaMGGE1Rf/UqVOIjo4GwN/ttlRTUwO53PL0pVAoTNOaeazbhyPHNTExEUql0qJNfn4+jhw50vpj36ohu1c5aVrzO++8Ixw7dkxYsGCB4O3tLWRnZ3f0rl2xHnnkEcHf31/YsWOHkJ+fb7rV1NSY2rz88suCv7+/sHXrVuHw4cPCPffcw+mIbcR8lpAg8Fi3pb179woeHh7CSy+9JJw+fVrYuHGj4OXlJXz00UemNjzebeOBBx4QunXrZprWvHXrViE4OFh46qmnTG14rF1TWVkpHDhwQDhw4IAAQFi9erVw4MAB03IejhzXuXPnCpGRkcL3338v/Prrr8LYsWM5rdkd3nzzTSE6OlpQqVTCoEGDTNNvyTUAbN7effddUxuDwSAsW7ZMCA8PF9RqtXDDDTcIhw8f7ridvoo0DSw81m1r27ZtQkJCgqBWq4U+ffoIa9eutXicx7ttVFRUCPPnzxe6d+8uaDQaIS4uTli8eLGg1WpNbXisXfO///3P5t/oBx54QBAEx45rbW2t8Je//EXo0qWL4OnpKUycOFHIyclp9b7JBEEQWlejISIiImpfHMNCREREnR4DCxEREXV6DCxERETU6TGwEBERUafHwEJERESdHgMLERERdXoMLERERNTpMbAQERFRp8fAQkRERJ0eAwsRERF1egwsRERE1On9P5LayTHmIukBAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "!pip install matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "plt.plot(h['accuracy'])\n",
    "plt.plot(h['val_accuracy'] , c = \"red\")\n",
    "plt.title(\"acc vs v-acc\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99d8ff4a",
   "metadata": {},
   "source": [
    "# Ploting Accuracy In Training Set & Validation Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "bf3d56d7bed3c008",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:12:35.513836Z",
     "start_time": "2024-09-29T01:12:35.346921Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(h['accuracy'])\n",
    "plt.plot(h['val_accuracy'] , c = \"red\")\n",
    "plt.title(\"acc vs v-acc\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3518d02",
   "metadata": {},
   "source": [
    "# Ploting Loss In Training Set & Validation Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c9d6cd08d4d21b46",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:12:53.314221Z",
     "start_time": "2024-09-29T01:12:53.184432Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGxCAYAAACwbLZkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABVfUlEQVR4nO3deXhTZdoG8DtLk3QvLbRQKKVlERQBaQHZxAVxgNFxXEBcEAFHREXkc0Nn3BXHmXFQR1BUxIUBXNBBrWgR2VGgtOz71lJaSgvd2zTL+/1xcrK06ZI07UnJ/buuXA0nJ8npoZC7z/u871EJIQSIiIiIFKJW+gCIiIgosDGMEBERkaIYRoiIiEhRDCNERESkKIYRIiIiUhTDCBERESmKYYSIiIgUxTBCREREimIYISIiIkUxjBC1IUuWLIFKpcLJkyeVPpQ2g+eMyP8xjBAREZGiGEaIiIhIUQwjRBeBxYsXo3///jAYDIiOjsaf//xnHDhwwGWf48eP44477kB8fDz0ej3i4uJw3XXXISsry77P2rVrcfXVVyMmJgbBwcHo2rUrbr31VlRWVtb73jfffDMSExNhtVrrPDZkyBAMHDjQ/ucvv/wSQ4YMQWRkJEJCQpCcnIypU6c2+L1dccUVGDlyZJ3tFosFnTt3xi233NLg8+uj5DkjIlcMI0Rt3Lx58zBt2jRcdtllWLlyJd566y3s3r0bQ4cOxZEjR+z7jRs3DhkZGXjjjTeQnp6OhQsX4oorrkBxcTEA4OTJkxg/fjx0Oh0WL16M1atX4/XXX0doaChqamrqff+pU6ciOzsba9euddl+8OBBbNu2Dffddx8AYOvWrZg4cSKSk5OxfPly/PDDD3juuedgNpsb/P7uu+8+bNq0yeV7AYCff/4ZZ86csb++J5Q+Z0RUiyCiNuPjjz8WAMSJEyeEEEJcuHBBBAcHi3Hjxrnsl52dLfR6vbjzzjuFEEIUFhYKAGL+/Pn1vvZXX30lAIisrCyPjslkMom4uDj7e8mefPJJodPpRGFhoRBCiH/+858CgCguLvbo9QsLC4VOpxPPPPOMy/YJEyaIuLg4YTKZGny+P54zInLFyghRG7Z161ZUVVVhypQpLtsTEhJw7bXX4pdffgEAREdHo3v37vjHP/6BN998E5mZmXWGVQYMGACdToe//OUv+OSTT3D8+PEmHYNWq8Xdd9+NlStXoqSkBIA0hPLZZ5/hT3/6E2JiYgAAgwYNAgBMmDABX3zxBXJzc5v0+jExMbjxxhvxySef2I/5woUL+N///ofJkydDq9VCCAGz2exyq48/nDMicsUwQtSGFRUVAQA6depU57H4+Hj74yqVCr/88gtuuOEGvPHGGxg4cCA6dOiAWbNmoaysDADQvXt3rFmzBrGxsXjooYfQvXt3dO/eHW+99VajxzF16lRUV1dj+fLlAICffvoJeXl5LkMoV111Fb799luYzWZMnjwZXbp0Qd++fbFs2bImvX5ubi7S09MBAMuWLYPRaLQHivXr1yMoKMjlVt9UXn85Z0TkROnSDBE1Xe0hh7S0NAFAfPHFF3X2HTt2rOjYsaPb1zl06JB4+eWXhUajEQ888ECdx81ms/jtt9/EXXfdJQCIZcuWNXpsgwcPFkOGDBFCCHHbbbeJ+Ph4YTab3e5bXV0t1q1bJ2644QYBQGzZsqXB1zabzSI+Pl5MnDhRCCFEamqq/b2EEKK0tFRs377d5WY0GoUQ/n3OiEjCMELUhtTX/3DTTTe57JeTkyP0er246667Gny9AQMGiEGDBtX7eHFxsQAgnnjiiUaPbeHChQKA2Lhxo9Dr9WLu3LmNPicrK0sAEO+++26j+z711FNCr9eLDRs2CADi/fffb/Q5Qvj3OSMiiVaJagwR+UZUVBT+9re/4ZlnnsHkyZMxadIkFBUV4cUXX4TBYMDzzz8PANi9ezcefvhh3H777ejZsyd0Oh3Wrl2L3bt34+mnnwYAvPfee1i7di3Gjx+Prl27orq6GosXLwYAjB49utFjmTRpEubMmYNJkya5DKHInnvuOZw+fRrXXXcdunTpguLiYrz11lsICgrCqFGjGn39qVOn4u9//zvuvPNOBAcHY+LEiR6eLYk/nTMislE6DRFR09X+LV/24Ycfin79+gmdTiciIyPFn/70J7Fv3z7742fPnhVTpkwRvXv3FqGhoSIsLEz069dP/Pvf/7YPpWzdulX8+c9/FomJiUKv14uYmBgxatQosWrVqiYf35133ikAiOHDh9d57Pvvvxdjx44VnTt3FjqdTsTGxopx48aJjRs3Nvn1hw0bJgA0Wr1w5u/njIiEUAkhhLJxiIiIiAIZZ9MQERGRohhGiIiISFEMI0RERKQohhEiIiJSFMMIERERKYphhIiIiBTVJhY9s1qtOHPmDMLDw6FSqZQ+HCIiImoCIQTKysoQHx8Ptbr++kebCCNnzpxBQkKC0odBREREXsjJyUGXLl3qfbxNhJHw8HAA0jcTERGh8NEQERFRU5SWliIhIcH+OV6fNhFG5KGZiIgIhhEiIqI2prEWCzawEhERkaIYRoiIiEhRHoeRDRs24MYbb0R8fDxUKhW+/fbbRp+zfv16pKSkwGAwIDk5Ge+99543x0pEREQXIY/DSEVFBfr374///Oc/Tdr/xIkTGDduHEaOHInMzEw888wzmDVrFr7++muPD5aIiIguPh43sI4dOxZjx45t8v7vvfceunbtivnz5wMA+vTpgx07duCf//wnbr31VrfPMRqNMBqN9j+XlpZ6ephERETURrR4z8jWrVsxZswYl2033HADduzYAZPJ5PY58+bNQ2RkpP3GNUaIiIguXi0eRvLz8xEXF+eyLS4uDmazGYWFhW6fM3fuXJSUlNhvOTk5LX2YREREpJBWWWek9vxiIYTb7TK9Xg+9Xt/ix0VERETKa/HKSMeOHZGfn++yraCgAFqtFjExMS399kREROTnWjyMDB06FOnp6S7bfv75Z6SmpiIoKKil356IiIj8nMdhpLy8HFlZWcjKygIgTd3NyspCdnY2AKnfY/Lkyfb9Z8yYgVOnTmHOnDk4cOAAFi9ejI8++giPP/64b74DIiIiatM87hnZsWMHrrnmGvuf58yZAwC49957sWTJEuTl5dmDCQAkJSUhLS0Njz32GN59913Ex8fj7bffrndaLxEREQUWlZC7Sf1YaWkpIiMjUVJS4tML5X2dcRp7ckvwh74dcWUy+1eIiIh8qamf3wF9bZr1h89hyZaT2H+Gi6oREREpJaDDiFYtTS22WP2+OERERHTRCugworGFETPDCBERkWICOoxoNXJlxKrwkRAREQWugA4jcmXEZGFlhIiISCkBHUa0aunbZ88IERGRcgI8jLBnhIiISGkBHUY07BkhIiJSXECHEVZGiIiIlBfQYUTDnhEiIiLFBXQYYWWEiIhIeQEdRuyLnlnYM0JERKSUgA4jrIwQEREpL7DDiIY9I0REREoL7DDCyggREZHiAjqMyD0jFi4HT0REpJiADiOsjBARESkvoMOIvTLCFViJiIgUE9BhRKthZYSIiEhpAR1G5BVYzewZISIiUkxAh5Eg+zANwwgREZFSAjqM2FdgZc8IERGRYgI6jMg9I6yMEBERKSegw4i9Z4RhhIiISDEBHUa07BkhIiJSXECHEblnxMSr9hIRESkmoMMIKyNERETKC+gwouFy8ERERIoL6DASpJG+fVZGiIiIlBPQYYSVESIiIuUFdBhhzwgREZHyAjqM2CsjnE1DRESkmIAOI1o1e0aIiIiUFtBhRGNbDt7EMEJERKSYgA4j7BkhIiJSHsMIpDAiBAMJERGREgI8jDi+fVZHiIiIlBHQYUTuGQG41ggREZFSAjqMyMM0ACsjRERESgnoMKJRszJCRESktMAOIyqnMMKFz4iIiBQR0GFErVZBLo5wmIaIiEgZAR1GAEBru3Ivh2mIiIiUwTDChc+IiIgUFfBhxH6xPIYRIiIiRQR8GHFURtjASkREpISADyMaNXtGiIiIlBTwYUSujJgtDCNERERKCPgwwp4RIiIiZQV8GNFq2DNCRESkJIYRDtMQEREpimHE1sDKdUaIiIiUEfBhhD0jREREygr4MOLoGWEYISIiUkLAhxG5MmLiVXuJiIgUEfBhhNemISIiUlbAhxH2jBARESnLqzCyYMECJCUlwWAwICUlBRs3bmxw/6VLl6J///4ICQlBp06dcN9996GoqMirA/a1IA1n0xARESnJ4zCyYsUKzJ49G88++ywyMzMxcuRIjB07FtnZ2W7337RpEyZPnoxp06Zh3759+PLLL7F9+3ZMnz692QfvC6yMEBERKcvjMPLmm29i2rRpmD59Ovr06YP58+cjISEBCxcudLv/b7/9hm7dumHWrFlISkrCiBEj8MADD2DHjh3NPnhf4FV7iYiIlOVRGKmpqUFGRgbGjBnjsn3MmDHYsmWL2+cMGzYMp0+fRlpaGoQQOHv2LL766iuMHz++3vcxGo0oLS11ubUUVkaIiIiU5VEYKSwshMViQVxcnMv2uLg45Ofnu33OsGHDsHTpUkycOBE6nQ4dO3ZEVFQU3nnnnXrfZ968eYiMjLTfEhISPDlMj8grsHI5eCIiImV41cCqUqlc/iyEqLNNtn//fsyaNQvPPfccMjIysHr1apw4cQIzZsyo9/Xnzp2LkpIS+y0nJ8ebw2wSVkaIiIiUpfVk5/bt20Oj0dSpghQUFNSplsjmzZuH4cOH44knngAA9OvXD6GhoRg5ciReeeUVdOrUqc5z9Ho99Hq9J4fmNfaMEBERKcujyohOp0NKSgrS09Ndtqenp2PYsGFun1NZWQm12vVtNBoNAKmiojR5OXhWRoiIiJTh8TDNnDlz8OGHH2Lx4sU4cOAAHnvsMWRnZ9uHXebOnYvJkyfb97/xxhuxcuVKLFy4EMePH8fmzZsxa9YsDB48GPHx8b77Trykka/ay54RIiIiRXg0TAMAEydORFFREV566SXk5eWhb9++SEtLQ2JiIgAgLy/PZc2RKVOmoKysDP/5z3/wf//3f4iKisK1116Lv//97777LppBy54RIiIiRamEP4yVNKK0tBSRkZEoKSlBRESET1/7hVX7sGTLSTx8TQ88fsMlPn1tIiKiQNbUz++AvzYNKyNERETKCvgwopEbWC2cTUNERKSEgA8jrIwQEREpi2FEzav2EhERKYlhhJURIiIiRQV8GJF7RrgCKxERkTICPoywMkJERKSsgA8jGvaMEBERKSrgw4i9MsLl4ImIiBQR8GFEYx+mYc8IERGREgI+jMiVEQ7T+JAQwIwZwCuvKH0kRETUBjCMaKRTwAZWH8rNBd5/n2GEiIiahGGElRHfq66WvhqNUpWEiIioAQEfRjRsYPU9k8lxn704RETUiIAPI6yMtADnMOJ8n4iIyI2ADyNyZcTE3+B9xzmAmM3KHQcREbUJAR9GtBpWRnyOlREiIvJAwIcReQVW9oz4ECsjRETkgYAPI0HsGfE9VkaIiMgDAR9GuAJrC6ipcdxnZYSIiBoR8GGEPSMtgJURIiLyQMCHEXvPCMOI7zCMEBGRBwI+jPCqvS2ADaxEROSBgA8jjp4RhhGfYWWEiIg8EPBhxLECKxtYfYaVESIi8gDDCK/a63usjBARkQcYRrjOiO9xai8REXkg4MMIe0ZaACsjRETkgYAPI6yMtAD2jBARkQcCPoxonMKIEAwkPsHKCBEReSDgw4hW7TgFHKrxEYYRIiLyQMCHEY1tOXiAQzU+w2EaIiLyQMCHEblnBGBlxGdYGSEiIg8wjDiFEQuXhPcNTu0lIiIPBHwY0bhURrgKq0+wMkJERB4I+DCiUqlcZtSQD7BnhIiIPBDwYQTgwmc+x8oIERF5gGEEjr4RM3tGfIOVESIi8gDDCJwrI+wZ8QlWRoiIyAMMI+CS8D7HMEJERB5gGAGg1UingT0jPsKpvURE5AGGEbAy4nOsjBARkQcYRsDZND7HBlYiIvIAwwicKyNsYPUJVkaIiMgDDCNwVEZMnNrrG6yMEBGRBxhGAGjV0mlgz4iPsDJCREQeYBgBe0Z8jrNpiIjIAwwjAII07BnxKVZGiIjIAwwjcKqMsGfENxhGiIjIAwwjYM+Iz7GBlYiIPMAwAvaM+BwrI0RE5AGGEQBaDS+U51OsjBARkQcYRsCeEZ9jZYSIiDzAMAJem8bnOLWXiIg8wDACRwMre0Z8hJURIiLyAMMIAI2GlRGfEQKwWBx/ZmWEiIgawTACxzANKyM+ULsSwsoIERE1gmEEjgZWrsDqAwwjRETkIa/CyIIFC5CUlASDwYCUlBRs3Lixwf2NRiOeffZZJCYmQq/Xo3v37li8eLFXB9wStLxqr+/UDh8cpiEiokZoPX3CihUrMHv2bCxYsADDhw/H+++/j7Fjx2L//v3o2rWr2+dMmDABZ8+exUcffYQePXqgoKAAZj/6kNJwBVbfYWWEiIg85HEYefPNNzFt2jRMnz4dADB//nz89NNPWLhwIebNm1dn/9WrV2P9+vU4fvw4oqOjAQDdunVr3lH7GHtGfMh5Wi/AyggRETXKo2GampoaZGRkYMyYMS7bx4wZgy1btrh9zqpVq5Camoo33ngDnTt3Rq9evfD444+jqqqq3vcxGo0oLS11ubUk9oz4ECsjRETkIY8qI4WFhbBYLIiLi3PZHhcXh/z8fLfPOX78ODZt2gSDwYBvvvkGhYWFmDlzJs6fP19v38i8efPw4osvenJozRKkYWXEZ9gzQkREHvKqgVWlUrn8WQhRZ5vMarVCpVJh6dKlGDx4MMaNG4c333wTS5Ysqbc6MnfuXJSUlNhvOTk53hxmk9l7RtjA2nysjBARkYc8qoy0b98eGo2mThWkoKCgTrVE1qlTJ3Tu3BmRkZH2bX369IEQAqdPn0bPnj3rPEev10Ov13tyaM3CnhEfYmWEiIg85FFlRKfTISUlBenp6S7b09PTMWzYMLfPGT58OM6cOYPy8nL7tsOHD0OtVqNLly5eHLLv2S+Ux56R5mNlhIiIPOTxMM2cOXPw4YcfYvHixThw4AAee+wxZGdnY8aMGQCkIZbJkyfb97/zzjsRExOD++67D/v378eGDRvwxBNPYOrUqQgODvbdd9IMvFCeD8nhQx62Y2WEiIga4fHU3okTJ6KoqAgvvfQS8vLy0LdvX6SlpSExMREAkJeXh+zsbPv+YWFhSE9PxyOPPILU1FTExMRgwoQJeOWVV3z3XTSTfG0aM3tGmk+e2hsSAlRUsDJCRESN8jiMAMDMmTMxc+ZMt48tWbKkzrbevXvXGdrxJ6yM+JAcPhhGiIioiXhtGgBa22waNrD6gHMYAaSr+LIXh4iIGsAwAkCrYWXEZ+Qw4twPxOoIERE1gGEEnE3jU7UrIwCbWImIqEEMI3BaZ4QNrM3HyggREXmIYQSOFVjZM+ID7sIIKyNERNQAhhFwNo1PyVN7DQbAFvJYGSEiooYwjIA9Iz4lB4+gIEBrmznOyggRETWAYQSOq/ayMuIDzmEkKMh1GxERkRsMI2DPiE8xjBARkYcYRsCeEZ/iMA0REXmIYQROPSOc2tt8rIwQEZGHGEbgtM4IG1ibj5URIiLyEMMInGfTsDLSbPLUXp2OlREiImoShhHw2jQ+xcoIERF5iGEETlftZc9I87FnhIiIPMQwAscwDSsjPsDKCBEReYhhBI5hGvaM+AArI0RE5CGGETivM8LZNM3GMEJERB5iGIHTCqzsGWk+DtMQEZGHGEbgvM4Iw0izcWovERF5iGEEbGD1KVZGiIjIQwwj4AqsPsWeESIi8hDDCACtRjoNVgFYWR1pHlZGiIjIQwwjcAzTAIBFMIw0CysjRETkIYYROIZpAPaNNBsrI0RE5CGGEbhWRkwW9o00izybhpURIiJqIoYRsDLiU3Lw4NReIiJqIoYRuFZGuNZIM3GYhoiIPMQwAkClUnGtEV9hAysREXmIYcSGq7D6CCsjRETkIYYRG/vF8nh9muZhZYSIiDzEMGKj4SqsvsHKCBEReYhhxEZehZXDNM3Eqb1EROQhhhEbe2WEwzTN4zy1l5URIiJqAoYRGy1n0/gGe0aIiMhDDCM27BnxEXdhhJURIiJqAMOITZCtZ4SVkWawWqUb4NrAysoIERE1gGHERsN1RprPOXRwmIaIiJqIYcSGPSM+UDuMsIGViIiagGHERq6M8Kq9zSBP6wVYGSEioiZjGLFhZcQHnEOHVsvKCBERNQnDiA17RnzAeSaNSsXKCBERNQnDiI1Wzdk0zeYcRgBWRoiIqEkYRmxYGfGB2mGElREiImoChhEbrUbuGWEDq9fqCyOsjBARUQMYRmy0vDZN89U3TMPKCBERNYBhxEbDnpHmc75ir/NXhhEiImoAw4iNXBkxMYx4z/mKvQAbWImIqEkYRmw0cs8IFz3zHhtYiYjICwwjNlrOpmk+Tu0lIiIvMIzYaLgCa/OxMkJERF5gGLEJsjWwsjLSDKyMEBGRFxhGbOw9Iwwj3mNlhIiIvMAwYsOeER+ob2ovKyNERNQAhhEb+3LwnE3jvfqm9rIyQkREDWAYsdGygbX5OExDREReYBix0bCBtfnqa2C1WqUbERGRGwwjNqyM+EB9lRGAfSNERFQvr8LIggULkJSUBIPBgJSUFGzcuLFJz9u8eTO0Wi0GDBjgzdu2KPmqvWb+Bu+9+iojAMMIERHVy+MwsmLFCsyePRvPPvssMjMzMXLkSIwdOxbZ2dkNPq+kpASTJ0/Gdddd5/XBtiRWRnygocoI+0aIiKgeHoeRN998E9OmTcP06dPRp08fzJ8/HwkJCVi4cGGDz3vggQdw5513YujQoV4fbEuy94xYGEa8VntqLysjRETUBB6FkZqaGmRkZGDMmDEu28eMGYMtW7bU+7yPP/4Yx44dw/PPP9+k9zEajSgtLXW5tTSuM+IDtaf2ajSASuX6GBERUS0ehZHCwkJYLBbExcW5bI+Li0N+fr7b5xw5cgRPP/00li5dCq3zb8oNmDdvHiIjI+23hIQETw7TK/I6I723rwPWrWvx97so1R6mcb7PyggREdXDqwZWlfzbro0Qos42ALBYLLjzzjvx4osvolevXk1+/blz56KkpMR+y8nJ8eYwPaLVqNC9KAf3vzkHuPlmQLBC4jF3YYQLnxERUSOaVqqwad++PTQaTZ0qSEFBQZ1qCQCUlZVhx44dyMzMxMMPPwwAsFqtEEJAq9Xi559/xrXXXlvneXq9Hnq93pNDazaNWoXxBzdBLQRQUgIYjYDB0KrH0OY1VBlhGCEionp4VBnR6XRISUlBenq6y/b09HQMGzaszv4RERHYs2cPsrKy7LcZM2bgkksuQVZWFoYMGdK8o/ehILUa4w86TVGuqFDuYNqqhiojHKYhIqJ6eFQZAYA5c+bgnnvuQWpqKoYOHYpFixYhOzsbM2bMACANseTm5uLTTz+FWq1G3759XZ4fGxsLg8FQZ7vSok4cxiWFTtOTKyqAmBjlDqgtYmWEiIi84HEYmThxIoqKivDSSy8hLy8Pffv2RVpaGhITEwEAeXl5ja454o+6rv3BdQMrI56rPbUXYGWEiIgapRLC/zs1S0tLERkZiZKSEkRERPj+DYRAWfdeCD9x1LFt+3YgNdX373UxmzQJWL4cmD8fePRRaVtyMnDiBLB1K3DllYoeHhERta6mfn7z2jQAsHcvwk8chVGjRVFke2lbZaWyx9QWsWeEiIi8wDACAF98AQBYn5yK4ohoaRuHaTzHnhEiIvICw4gQ9jDyfe8RqNIFS9v9NYxYLMDIkdKQiL/homdEROQFhpHdu4HDh2HRG/BL98GoDrKtb+KvYeT0aWDTJqk3w9/afbjoGREReYFhZMUKAEDx1dehQh+CqiDbQmf+Gkaqqhz3q6uVOw533M2mYWWEiIgaEdhhxGmI5sL4PwMAqnS2yoi/NrA6H5e/HSMrI0RE5IXADiOZmcCxY0BwMMquk65EXNmWKiPO9/1B7av2AmxgJSKiRgV2GLFVRTB+PNTh4QDaWBhpS5URDtMQEVE9AjuM7N8vfZ04ERq1dNXhKq3tt3qGEc9xai8REXnB4+XgLyqrVgEHDgDdukFbIn1YVmrbUGXEX4dpWBkhIiIPBHYYAYA+fQAA2jLpw7Lc36f2sjJCREQXmcAepnGiVUunwl4Z8bcPepk/z6bh1F4iIvICw4iN3DNSrm1DlZG2NEzDyggREdWDYcRGq5HCSEVbCiP+VhlpaGovKyNERFQPhhEbuTJSpuFsGq+xMkJERF5gGLFx9Iy0ocpIWximYQMrERE1gmHERq6M2Bc987eqg6ytVkY4TENERPVgGLHRyoue6drQOiP+FEYsFsdVhFkZISIiDzCM2MgNrPbKSFUVYLUqeET1cA4g/jRMI0/rBTi1l4iIPMIwYiP3jFTJPSOAf1UeZP5aGXGufLCBlYiIPMAwYmMbpUF1kNO0VH8cqmkLYYRTe4mIyAMMIzYqlQpatQpCpYY1JETa6O9hxJ+GaeQwolIBGo1jOysjRETUCIYRJ/KMGiGHEX+qPMj8vTLiPETj/GdWRoiIqB4MI0609jASKm3w98pIWwgjrIwQEVEjGEacyJURDtN4obHKCMMIERHVg2HESZBGOh1WQ7C0wR/DiL9etdfdFXsBLnpGRESNYhhxIldGLMFtZJiGlREiIroIMIw4kXtGLGxg9Zy7K/YCbGAlIqJGMYw40WjkyoifDtOYza4f6v4YRtjASkREHmIYcSKvwmoO9tMG1trDMm1pmIaVESIiqgfDiBO5Z8RsaCNhxGTyn4oDKyNEROQlhhEn9p6RxmbTmEzKVCXkYRm101+bv1RHWBkhIiIvMYw4kSsjpsbCyJAhQM+egNHYSkdmIwePyEhp2XXnbUprbGovKyNERFQPhhEnWts6I2Y5jLhrEK2uBjIzgdxc4PTpVjw6OIJHcLB0A/yniZVTe4mIyEsMI07kYZqahiojFy447peUtMJROXEOI/42/bi+qb1c9IyIiBqhVfoA/Il9mEbfQAOrv4QR+cPfX4ZpWBkhIiIvMYw4sVdG9Hppg7+GkZAQx4e7v1VG2MBKREQeYhhx0qQG1uJix/3S0pY/KGfuKiP+HkbYwEpERI1gGHEiV0aM+gaaQ5WsjMjHExzsqDS0lWEaVkaIiKgeDCNONLb1O2r0baCBVf5w95fKCKf2EhGRlxhGnARpalVG/DmMWCzSfX8JI7xQHhEReYlhxIncM2LUGaQN/hxGrFbXbUpjzwgREXmJYcSJtnYYka/94vwB69zA6g9hxN8qI5zaS0REHuKiZ07knpEqOYwAdasj/lAZCQnx30XP6quMWCyAEK17TERE1CYwjDiRKyMmTRCg0Ugba3/Y+8tsGjmM+PswjfOf2TdCRERuMIw40dgaWM1WAKGh0kZ/rIy0xWvTAAwjRETkFsOIE7kyYrFa/T+M+NswTWNTewH2jRARkVtsYHWitfWMmK3C8WHfUBhRcgVWlcp1m9Iam9oLsDJCRERuMYw40WrkyohwXxkxmVz/XF4uNWbK/SUtzV0Y8ZfKSH3DNGp13X2IiIiccJjGibzOiLm+MOI8rVfWmtWRtjibRqXiwmdERNQghhEncs+I2eLUM+L8YS8P0UREAPKVfVuzb8Rdz4i/DdPUDiMAFz4jIqIGMYw4abQyIoeRdu2AyEjpfmuGEeepvW1lNo3zNlZGiIjIDYYRJ47ZNI2EkagoZcJIW5xNA7AyQkREDWIDqxNNY7Np5J6Rdu0cH7pKhRH5A74tDNNwSXgiImoAw4iTJldG2rVzzKBROoz4S2Wkvqm9AIdpiIioQQwjTrQaNz0j7hpY27VzTK1VKozIH/D+FkY4TENERB5iGHHS6AqszmFEvuhba4URk8lRWQgJcdyvqpKORQ5HSmEDKxEReYlhxIncM2KyNKGB1WKR7rfWOiPOvSHBwa4f7NXVjtk1SmFlhIiIvOTVbJoFCxYgKSkJBoMBKSkp2LhxY737rly5Etdffz06dOiAiIgIDB06FD/99JPXB9ySXHpG3DWwKjm11zmMGAyu4cMfhmpYGSEiIi95HEZWrFiB2bNn49lnn0VmZiZGjhyJsWPHIjs72+3+GzZswPXXX4+0tDRkZGTgmmuuwY033ojMzMxmH7yvNXkFViXDiMEgDclotY5mUX+YUcOpvURE5CWPw8ibb76JadOmYfr06ejTpw/mz5+PhIQELFy40O3+8+fPx5NPPolBgwahZ8+eeO2119CzZ0989913zT54X3Ncm6YJPSMREdL91g4jzhURf1r4jJURIiLykkdhpKamBhkZGRgzZozL9jFjxmDLli1Neg2r1YqysjJER0fXu4/RaERpaanLrTXYKyOWJsymUaoy4hxG/Gnhs4am9rIyQkREDfAojBQWFsJisSAuLs5le1xcHPLz85v0Gv/6179QUVGBCRMm1LvPvHnzEBkZab8lJCR4cphe09oaWP1yBVbni+TJ/On6NFz0jIiIvORVA6uq1jRSIUSdbe4sW7YML7zwAlasWIHY2Nh695s7dy5KSkrst5ycHG8O02PahnpGLBZH8PCXygiHaYiI6CLg0dTe9u3bQ6PR1KmCFBQU1KmW1LZixQpMmzYNX375JUaPHt3gvnq9Hnr5qritSGNf9MxadzaN81BRu3aOZtbWCiPOF8mT+eMwDRtYiYjIQx5VRnQ6HVJSUpCenu6yPT09HcOGDav3ecuWLcOUKVPw3//+F+PHj/fuSFuB1l3PSEWFtKiYPEQTEiL1RciVkbIywGpt+YNrqGekrQzTsDJCRERueLzo2Zw5c3DPPfcgNTUVQ4cOxaJFi5CdnY0ZM2YAkIZYcnNz8emnnwKQgsjkyZPx1ltv4corr7RXVYKDgxEpf6D7CY27a9MIARiNrs2rgCOMCCEFkpb+Xvx5mEYIVkaIiMhrHoeRiRMnoqioCC+99BLy8vLQt29fpKWlITExEQCQl5fnsubI+++/D7PZjIceeggPPfSQffu9996LJUuWNP878CG3DayAVB1xbl4FpPU+goKkD9iSEmXCiL8M0zhXPFgZISIiD3m1HPzMmTMxc+ZMt4/VDhjr1q3z5i0U4bLomUYD6PVSVcQ5jMiVEZVKCiCFha2zJLw/D9M4Vzw4tZeIiDzk1Wyai1WQxmmYBnDtG6kdRoDWnVHjbmqvvwzTOIcMVkaIiMhDDCNOHJURW0Oq84wa56XgZUqEEX8cpmksjLAyQkREDWAYcSL3jJgtTayMtOaS8A1N7fWXYRq1WrrVxkXPiIioAQwjTlx6RgDXJeFrN7ACyldG/G2Yxl1VxHk7h2mIiMgNhhEnWnUb6Bnxx2Gahq7YC3CYhoiIGsQw4qROz0hbCSP+MkzDyggREXmBYcSJtvZsGn9sYPXn2TTupvUCrIwQEVGDGEachBuk3+BNFoHCcmPbqYz4SxhhZYSIiLzAMOIkTK9Fr7gwAEDGqQvuG1iVCiNtYTYNe0aIiMgLDCO1pCRGA6gVRsrLHcM07mbTKLUCq78N07AyQkREXmAYqSU1Uap87Dh53hFGzp4FLBbpPodp6mJlhIiImoFhpJbUblLY2JtbCpPB9sF/+rT0VadzDQP+EkaUHqZpbGovFz0jIqIGMIzU0jU6BO3D9KixWHGmRppdYw8j7dpJF8iTKR1GOExDREQXAYaRWlQqlX2o5pj8GZ+bK311HqIBHMvBl5YCQrTsgbmb2utvwzSc2ktERF5gGHFDHqo5XG7rE5E/7J2bVwFHZcRikab/tqSGKiNms7If9KyMEBFRMzCMuJFiq4zsK7G4PlC7MhIaCmg00v2WHKoxmRwf5O56RoC6fSMXLkiNt62BDaxERNQMDCNuXBYfCb1WjXNWresDtcOIStU6V+51DhrOYUSvd/SwOA/VCAEMHAj06dM6QzisjBARUTMwjLih06rRPyEKVVq96wO1wwjQOk2szmHEYHDcV6kc4cR5n3PngJMnperIyZMtd1wyVkaIiKgZGEbqkZrYDpVBBteNSocRg8F1Ng/gvon11CnHfbn5tiU1dWovKyNEROQGw0g9Uru1Q5WuVhip3cAKtG4YcR6ikblba8Q5jJw503LHJWNlhIiImoFhpB4pXaM9q4y05JLw7qb1ytytNaJUGKlvai8rI0RE1ACGkXpEhgShc+cY141KDdO4u0iezN0wTXa2474/VEa4AisRETWAYaQBfXt0ct2gdM8Ih2moKaqqgM8/BwoLlT4SIqImYRhpQGq3aFQGOc2o8ccw4k/DNGxg9Q8ffQTccw/w4otKHwkRUZMwjDQgtVutGTXuGlhbc52Rpg7TtHYYaWw2DSsjrWv/ftevRER+jmGkAV2jQ2DUOQUAf6yM1B6mKSuT1heR5eUBVmvLHRvAyoi/kcOocyglIvJjDCMNUKlUQKj0YW9Va4Dw8Lo7tWYYacpsGrl5NTxcWpPEZAKKilru2AD2jPgbOYTk5LR8ECUi8gGGkUYER0nDMCX6UJRUufnN3l8qI3IYkT+IkpOBDh2k+y09VMOpvf5DCMfPQE1N612fiIioGRhGGtGuQxQAoNgQinfWHqm7g79M7ZUDi/xBlJgIxMdL91srjLAyorziYqC83PFnDtUQURvAMNIIVWgoAKBUH4ZPtp7EqaIK1x2UrozUHqZxDiOdO0v3W3pJePaM+I/a4YNhhIjaAIaRxtjCiDq6HUwWgb+vPuj6uPMKrEK0zDF4Mkwj94x07dp6lZGmXpvGZGq5c0SS2uHDeQE8IiI/xTDSGFsYSezRBWoVkLYnHxmnzjsel8OIyQRUV7fMMXgym0aJYRq58hIb6/5xeZgGACyWlj2WQMfKCBG1QQwjjbGFkYhOHTAhNQEA8MoPByDk3/DDwhxX0m2poRpvh2laK4wcPix9veQS9487V0w4VNOy5L9/eeYXwwgRtQEMI42RP2AvuwxzxvRCiE6DzOxifL87T9quVrf8wmcNTe21bdt7NA8Zh/MdwaO1wsj5845lx3v0cL+Pc2WETawtSx6WGTFC+sowQkRtAMNIY2bOBPbuBR56CLHhBswY1R0A8PqPB3G21DYs09JNrA3MphG2bdWlFfjbuz9JPRl6vTSttzXCiFwV6dxZqhK5w8pI65HDx8iR0lf2jBBRG8Aw0hi1GrjsMukrgPtHJqNzVDByi6twy4ItOFpQ3vJhpIFhmowCKRAFm4yIOCeFDktCgnS8chg5e7blQoAcRnr1qn8fjcZxn5WRliWHEbkyUlLSsjO9fOmhh6QQ5XzRRyIKCAwjHgrWabDs/iuR1D4UucVVuO29LSjX24ZPWjmMWK0Cy/ZKQySxWgv6mosBAPu07WA0W6TqiEYjrcJZUNAyx9aUMKJSOYZqWBlpOVVVjr/nSy8FYmKk+21hqKa6GnjvPWDTJmDrVqWPhohaGcOIF7rGhOCrGUPRv0skiitN2FFsW3K7lcPI6n35OFQmvXeM2ooHu0kf+Pt17TD36z0QajXQsaO0c0sN1TTWvCrjwmctTx6SCQ0FoqOlviGgbYSR/fsdS9fv3q3ssRBRq2MY8VJMmB7L/nIlrrmkA0qCpMrIoUM5LfNmbsKI1Srw1pojqNZKS7CrqyoRU5QPAMiLjMXKzFy8s/Zoy/eNNKUyAnDhs9bgPJNKpWpbYcQ5gDCMEAUchpFmCNFpsWhyKtp3lq4Bs3d/M8OIEMC77wIbNrhudzObJm1vHg6dLYMmzGnRM9uHzpVXXwEAmL/mMIyxLVgZsVqBI7Yl8psaRlgZaTlyZUQOIV27um73ZwwjRAGNYaSZgjRq9O4trT9Sca4IZkszrpK6aRPw8MPAHXe4rlRaqzJisQrMXyOFgFuHX+LYxxZGhl47ED1jw2AVQEFYtPR4SywJf+aMFIK0WqBbt4b35TBNy3OujDh/bWuVkX37WEEjCjAMIz7QrlN7AEBwZQUOnS3z/oU2b5a+5uW5foDUmtr7/e4zOFpQjgiDFndc7VSROHFC+pqYiP4JUQCAU4Z20raWqIwcOiR9TU6ufyl4GYdpWp5TGKkwmttuGKmuBo4eVe5YiKjVMYz4gNpWFbi04Dgys4u9f6HffnPc//13x31bZeRIqRkfbjyOv/8oXR/n/pHJiGgX4djPYpF6Bbp0Qf8u0nTjg2rb2h8tEUaa2i8CsDLSGmyhY5slFJe/8BO+Oa9x2e63zp4Fzp2TfnYvv1zadjEN1RQUAH37As8+q/SREPkthhFfuP56CJUKlxUcx7GsQ969hhCuAcQWTM6cK7Vfz+W2T7Lwyg8HcKakGrHhekwZ3k36kNfpHM+LjweCgtCvSxQAIMNsa3pVOoywMtLybKHjnaMmWAXwVYHtn3d+PmA0uu575gxw1VXAZ5+18kG6IQePnj2BK6903XYxWLlSGnp6+22GcaJ6MIz4QocOKO03EAAQ+svP3r1GTo70oSGzBZMnP3MEFFVwMK6+pAP+Or4PvntkBMINtg945ym/ttJ8707h0GnUOB5kW5BN6TDCykjLMpshTp8GABwJltYX2Vqqsq/Qi5xazdWLFwMbNwIvvdSaR+meHDz69ZNuztsuBuvWSV/Ly4GdOxU9FCJ/xTDiI7o/3QgA6J+1Cecrajx/AXmIpoM0Mwc7dyLr6FkcPHHWvsvvL4/HkvsGY/rIZMRFGBzPdb5mjS2M6LUa9OkUjrNyA2thYd3fjpurqWuMAKyMtLQzZ6CyWGBSa1AcFYOYUB2sUKGyUxfp8dpDNWvWSF+PHlV+GMddGNmzR7nj8SUhgPXrHX92vk9EdgwjPhJ8800AgOGnsrDrSJ77nTIzgeJi94/JQzS33y4tWGU0In15OgxmW7AJDoY+SOv+uW7CCAD06xKFYkM4zFpbEHCuvDRXTY2jYZaVEcVdOCg1fOaFt8cj1/fGtb1jAQBn28VJO9RuiHZe5fSXX1rrMN1zDiNyz8jJk21nGfuGHDni+u+OYYTILYYRXxkwACXtOiDEZETRD+l1H1+1Chg4ELj7bvfPlysjV14JDBkCAKjcsAkGk62a4ea6NHZuhmkAoF+XSEClwvlIabaPT4dqTpyQellCQ4FOnRrfn5WRFvXtt1sAAMUd4vGXq5IxJFkaqjkqz6ZyDiObNklhUiZXSZRgMkmrrwJSGGnXDuhiq+bs3avccfmKHD5ipXCITZv4b4DIDYYRX1GpcG7kdQCAsLW1+kaEAF59Vbr/44/S7AFnNTWOseQhQ+xNfP1yD2FQnG04pqEw4lwZkRe6AjDANr33dHALTO+Vp/X26iXNgmgMFz1rMT/uycO5fdK6M10HXIIgjRpDkqThuT2aKGkn54XP5EpIz57S17VrXde1aU2HD0s//2FhjiDdQN+IyWJFaXUb+hmS+0Xuv1+6oGZpKZCVVWe3yhoz7vrwN/zt24sggBF5gWHEhww3S30jl2ZuhMV58bONG4Ft26T7VivwzTeuT9y9W1pboV07oGdPVA9MBQAMyDuEW3rbLnbW1DDiVBlJ7hCGUJ0GeaEtEEY8aV4FOEzTQnafLsYz3+xBlxLpAnlRfaSA0aVdMOIjDciJsP1G7lwZkSshTz8t/VydPSvN9lCC3Bty+eX2K2M3FEZmLcvE4FfX4GB+aSsdYDM494tce610RWLA7VBN2p58bD5ahM9+O4Xj58pb8SCJ/APDiA91uvVG1GiC0PVCPk5sduqaf+MNAMAFQzgAQHz5pesT5X6RIUMAlQrfBnUGACRdyEOKpkJ6zIthGo1ahb6dIx2rsNYKI+VGMwpKq5v43dXiaRjhMI3PrT98Dncs+g0XKk3oXXNe2mj7+1epVBiSHIPcCFtDtBxGzp+XepcAYOxYxwekUkM1zv0isnrCyMnCCvy4Nx/VJis+2eLna6cAwPHj0srHQUFStXPUKGm7mzCycudpp/stsFoykZ9jGPEhTUQ4Dl4iTfEt+epbaeO+fcAPP8CqUmHWTU9I29atc1zqHXDpF7FaBd7fcwHHoqVxc/UG239cTamMREdL5W4n/ROicDbMVl1xWhLeahW4Y9FWXPWPX3HEm1VjWRlR1DeZpzFtyXZU1lgwokd7DLDaKgVOYXRIUjTOyJWRnBypKvfrr9Jv7JdeKvX6jB4tPa5UE2tDYWTPHseVfAGs2OGYnrwqK1daZdafyUM0gwdL/0blMLJxo33tIADILa7C1uNF9j9/k5kLq1WhYTMihTCM+FjhKKlvJELuG/nnPwEAq3sOxcakgdgT1x0qqxX49lvHk5wqI2sPFuBEYQX2JfSWtsn/oTkPxdQmP+bULyLr3yUK+eG2MOJUGfl5fz725pai2mTFwnXHPPkWJU5hpMZsxd7cRmY+sDLiE0IILNpwDI+t2AWzVeBPA+Kx+N5UqHNqXSQPwJDkGOSHx8CsUkshMD/fETrkEHKd9POK9euVCYruwkivXtJCfmVl9oqOyWLFVxlS9UCnUaOixoLvd7fQlah9Ra6AXH219PWKK4DwcGlGnVPV59vMXAgBXNE1CuEGLXKLq/CbUzghCgQMIz4WfPOfAABJB3YCBw5ALF0KAFg05FZMHpqItN4jAADGZSukJxQVOa58O3gwFm+Wpsvqhg+TtsmNok0ZpnH6IJL16xJpX2vEmiv95y2EwAKnAPK/XWeQc76y6d9kWZl0/RwA1h498cBnO/DHdzbh060n638OKyM+8cHG43gtTbocwPQRSfj3hAHQlVxwXL8oIcG+b7eYEERHhjjC6KlTjuEYOYQMGCBV1MrKgO3bW+m7sLlwwbEYmzylF5CCa58+0n3bh/bagwU4V2ZE+zAdHrm2BwBg+fZmXiW7pclhRK6IaLXAiBEujwkh8LVtiGbS4K74Y794AMDXHKqhAMMw4mN9RgzAsegu0FotsNx6G1QmE35P6Iv83v3x7Pg+OD16PAAgaMN6aSEyubG1Vy/8Xiyw5VgRNGoVBk64wfWFGwoj4VIvirsr53ZpFwxjB2mtCesZ6T+4LceKsPt0CQxBagxIiILFKvDhxuNN/ybl8BQbiyX7i/HroXMAgH/8dAiF5fUsrNbEykiNuRlXPb7I7TtTgn/8JIXTJ/9wCf76x0uhVqsc/SBxcYDBsRieSqXCkKRo5EY4TSs9ckRqFB01CjnnK2GBSmquBFp/qEZuXk1MlGaaOKvVN7LCFjxuTemCOwZ3hVatQmZ2MQ7lN+PClC3p5ElpBpNWCwwb5theq29k1+kSHD9XAUOQGmP7dsStA6V+sR/35vn/MBSRDzGM+FhUiA4ZfYcCADQHpPUT3h98Cx68ujv0Wg3G3Dgce+O6Q221wLxypb1fxJQ6CI9/tQsAMCG1C2KHDXINIA2FkalTgUmTgBkz6jykUqkQ1zsJAKAtLQUqKrBgnbRA1msVu/CfXSugNxmxfHtO/UGiNtsQTWW37njddtG+yOAglFWb8cbqg+6f04TKyBc7ctD/xZ8xL+1A047DT+Wcr8SMzzKwYnt24zs3kdFswZwVu2CyCIy5NA4PjurueNDpar21DUmOwRm5ifXjj6Wvgwbhs/0XMPKNXzFp0W8wX32NtL0Fm1h/3JOH73bVGlZxN0QjcwojeSVVWHdI6rGamJqADuF6jO4jBexl23x3jms7U1yF13886F1PlTy8mpoqrcUjk8PIhg2A1YqvbUNPf7isI8INQUhJbIduMSGorLFg9V4fLlJI5OcYRlpA0ajr7fcPx3TFvv7DMXGQVD6/4bKOWN9P+g/p/JL/2vtF0kITkXO+Cp2jgvHMuD5SJSElxfGiDYWRSy8F/vtfoHdvtw/36tEZFUHSb8wHdh7C5qNFmLh3DW7591x0+eAdLFnzFkw1JnxsGyJqlG3oaD3aocZixeg+sVg8RZqO/MWO08jMvlDnKRdqpIa8/PPupy3+frwIz6zcgyqTBe9vOI6lv7eB2RJuFFfW4N6Pt2H1vnw89fUe/OOngxA+WMPjzZ8P49DZMsSE6vDaLZdD5by2SwNh5MqkaORG2iojB6SQl5syHC9+JwXlbSfP4y2V7XlbtwIVFc0+1to+2XISDy7diUeWZbpW4OTKiC14VNVYHIHYKYx8ueM0rEJqyE3uIDVoTxws/Xv6JjMX1SZHM6ivFJUbcdeHv+O99cdwx6LfcLLQw/NSu19ElpIihZPz51GTtQvf2fpebhkoNayrVCr7/ZWZp9FS1h0qwNtrDqOyhtUX8g8MIy0gaszVKNVJTaUfDP4zHrimJwxB0uXcdVo1gidNAADE/L4J2CKtnPlBjfSB8Y/b+zkugGdbiRVAw2GkEf27Rtn7Rn5YvQNXH9uB1358x/740N0b8Pwvi/DplpMoc7OglKV2Z7+tMrIrOBax4Xq8cVt/pCRG41bbf6LPr9rnMhvg10MFWHNUCijLNx/D5qOFLi+Xc74SDy7dCbNVIDFGOm/P/28fth5z08QnhFRNulA38Hjq8NkynL7gQa9MI6pNFtz/6Q4cP1eBCINUCXr312OYu3IPzBbvh5+2nTiPRbYP8Xm3XI72YXrXHRoIIz1iw1DcwXWF3JeqOsJsFRjYNQoqFfDOKYGKjp2lqtWmTZ4dnMUCLF9e7xDPd7vO4IXvHGuYvPLDAcc0VqfKyI6T53HVP37F0Hm/SEHUFkbE0aP4drM0LHjHYEc/zFU9O6BzVDBKqkz4aZ9rBcFiFSgoa2TK+oYNwMyZjmnOTiprzJj6yQ6csAWQogopYNZbObS6+but3S8iCwoChg8HABz9Mg3FlSbERegxvEd7+y5/vkIaqtlyrAhniqsa/j48JITA599th+mPN+L2Pw/Dvx/5Z9MrokQtyKswsmDBAiQlJcFgMCAlJQUbN25scP/169cjJSUFBoMBycnJeO+997w62LZiQPc4PDF+Nt4adgc2DPkD7hziOstl7C1XYX9sEjRWC1BWhmqtDgc7dMOUYd0wrLvjPyX75dSBZoWRfl2i7GuNRK5ZjXf/97r03vfcI32QALh35w+4a/0KLP3dUfbeeqwIf3p3M3r99Ufc9J9NeC3tANYePIsLWdKHy4noePxrQn9Eh+oAAE+NvQRhei12ny7BF7ZpmF9lnMb0T3bACOk3eavJjCkfb7PPhKgwmnH/pztwvqIGfTtHYPWjV+Gm/vEwWwUeXJqB7CKnsLBnj/Sb5tChQI8ewKJFbj8IzhRXYdGGY/gm87Tb35oLSqvx2IosjPn3Blz9j3V45fv9zV7V02oVePzLXdh+8gLC9Vp8OWMYXr/lcqhVUqPlI59uQ/W5+mdImCxW/C8rFy+s2ocvd+Qg1/YhVG404/++zIIQwO0pXTDmso51n5xddyaNTKVSIaJXD/ufjUE6rIvugT6dIrB0+pX4v+ulFXTT4i6TdvBkqGbfPqkhc9IkaXbOxIkuqwtvPHIOc76Qjv2eKxMxbYQ0XPjEV7vx6/58e2XkO2sMJn3wG86VGWGyCDz7zV7M3XwWokMHqKxWhB47jAiDFmP7OkKVRq3C7alS+HUeqll/+Bz+MH8DBr/6Cx5aurNuY3ZFBfDII1JIWLhQmnb78sv24UOTxYqHlu7ErpxiRIUEYflfrkRCdDBOFVXivo+3u/ZxFBbi7PSZMIWE4uzgEaj5fZvj7+PECUCjsQcPF7aAUrVmLQDg5is6Q6NWSSvRms1IiA7BlcnREEKq/HjCZLFi9d58fJVxus4vFharwJK/f4bRd/4B1x/dhk7lRXh20VysGT8Zp/KLPXofIl+r58pr9VuxYgVmz56NBQsWYPjw4Xj//fcxduxY7N+/H13dTC09ceIExo0bh/vvvx+ff/45Nm/ejJkzZ6JDhw649dZbffJN+JteceHYfPlV+KnXMPz12l72qoisU2Qwtl31B1z61UIAwJ64HugaF4mn/lBrmMW5MtLQ1N5GtA/TY2d0LJAD3L/Ntvrr6NHAhx9KUyjz8oDHHsPT65fgb//piL09nsa/0w9j7YF8xJafR5KxEkeqY7H7dAkWrT+G3Uel31RTrh+CkT072N8nNtyA2aN74pUfDuCNnw7hTEk13v5F2jexo9SgeGPhARw4uBXPflyGc7cNxm/Hi3Awvwztw/RYdE8qgnUavHFbP5wsqsDu0yWY/ul2fH3nZQh//VXgnXcc6zOcPw888ID0Pbz7LjBoEDKzL2Dx5pNI25Nnr+a8/P0B3Dm4K+6+MhHRoTp8vPkE3v7lCCpqpNcxWwU+3HQC32bl4okbLsFtKQnSB4OH3vjpEL7fnYcgjQrv35OCSzqG45IOIUjeux0n3vkQY97aDMPUMhRengLVpDsQPeUuqDp1QrnRjOXbsvHx5pP2ACLrFhOCyOAg+/Ddczde6v7NG6iMAECXAY6rKm+LvxTBEaFYdE8KgnUaPHRND+zNLcXG/f1w+66fYfp5DYL+Ie1rtQoUVhgRotMiTO/0X4XRCLz2GjBvnvQhHhoqrSD8xRdAejrwr39h9+ib8cBnGTBZBMZf3gkv3HQZVADOV9Tgm8xcvPZuGq6pqIApSIfZOytgUWsw/vJO6NMpHP9KP4xl20/jjuiu6H/uHHqfO4GBt4yu8+9oQmoC3vrlCH47fh5r9p/F57+fwjpbMzUA/LAnD+kHzmL6iCTMvKYHwn7fAuuU+6A+IVWZTib0Qrecw8Bzz8H6v1VQfbIEc/eb8euhczAEqfHRvYOQktgOn04dglsXbsGe3BLM+DwDH97WB9nPzUP8oncQVy1VT+K2bwauHIIj1/4RCcNTYABgHTgQm/Kq8eu6bGw9VoQO4XpcfUks/tB/MDoDSNq3A+EjKnDv8c3AzXOB1aulf4+jR+PxPkPwSGkcvs4Ixcyrpf6g8xU1OGkL58ntQ9HO9ksAAJwtrcZ/f8/Gsm3ZKCiTKh3P/0+DPw/sjMlDu6FrlAE/3jsH93yxAFphxYWu3aG+ehQiP12MOzZ8gazBe1Hx5QpcOqSv+5+xZhJCuA4tEtWiEh4OaA8ZMgQDBw7EwoUL7dv69OmDm2++GfPmzauz/1NPPYVVq1bhwAFHU+KMGTOwa9cubHW+cqgTo9EIo9Pl7ktLS5GQkICSkhJERER4criKWb4tG1k5xXjhpsvq/CcKABnpvyNljFT5+HDQzRj41WIM7NrOdSchpIuGnTkD/P3vwJNPen08a8bfg9FpnwMAKi+9HCFbNwFO59Ly2Bxo5v8bNWotNiRdgYSSs+hafBbBZqe/h6j2yI6MRd9T+2FVqWAuLYcuzDUkmSxWjHtrI44UOHpDHrgqGU/t/Q7qp56yb7NChX1xyTgakwAtBEYkt0M7vUaqdAQFoQoapB+9gDIrMPb4NkSXSiuMbrz8Kvxr9DRce3Qbpqd/jJDqSlhVKvzefxTyTCqEGysRVlOJOFGDSqhwXhOMcl0wKgyhQHAIaoxG6MxmxBpU6B8bghqrwK6CSlwwASaNFmFhwWhn0EJttUJttUBttUiX3lFrAa0aQqOFSqOBymqFSlihslhgNpmRXVQBi0qNIT1jkRwXIX1gp6XZp0DXZlGpcahPCo5pwxFVch4dKi4gtqoY4TVVKAmNRL4hAudColAY0g7VQTpc37sDOoYGSWHMYpF+NuR/uitXSlN7s7KA/v3rvNeBo3no01OaMvr3q6dg+KJ/YERPRwWu3GjGfa9/hy9fuEU6l6mjYamsBCoroa8xwqzRwGQIhggNgyYiHL2O7kZsrtRftOuKkVhy1xOIKCvGtE9eQ9eTUgNzRuLlOBTVGV1CVBjeJQwaoxEQAla9HltzK1BztgDXHM/AnrjuuOm+t/DkDb0xY1QyVCoV1h0qwKxlmXj0h4WYtuN/OBKTgE59eyKs9II0A62qCoiJAdq3x85KDY6Y9TCYaxBTWYz2lSXobC5HiLEKF/RhyNeF4YIhHCq9DiMOS5WL3PAOeHrsI9jY7QrcdGA9Xv55ISKNFajR6rCq90hohBUj4vTooDJJlRS9HiUaHbbkG1Gu1mHUyZ2ILZd+Hg/EJmPLhPsRv+kX3JD1C9Rw/Hf60dDb8PJVU+r8fQRZTNj91h0INhlh0mgRZKm/b+NYdGfUhEeiHBpUqbSo0Whh1mhhVmmg0WkRHGyAJkiDcyXVUFkt0FotCNYAWo0aF8wqmDRamDRaXFKSh9QTUoP86XG3oMuKT4CwMBT/dwW006cjrKocF4IjcHDkHyDMJqlKYzIBZgtUQVqotFqogrRQy7PihICwCsBqhYCAUKkhVGpY1WpArYa1pgaqsnJoKsqhrapAkLEaZr0B5pBQiNAwIDwcKl0QhNEIVBulfy+mGkCjhdDpAL0OKp0eCAqS/p0JYbtZAZVKej+1GlCrIKCCSj7vzh9nKhUAlW1/lcvj8l4ql31bhvMHbH3v0pR9GqNqRm+afH7aP3Q/evxhVCN7e6a0tBSRkZGNf34LDxiNRqHRaMTKlStdts+aNUtcddVVbp8zcuRIMWvWLJdtK1euFFqtVtTU1Lh9zvPPPy8g/f243EpKSjw5XL9mtVrF4c49hQDE18/9p/4dJ0yQPnY++qhZ77fp/14WAhAF0XFC5ObW3cFiEceu+6P8Eee4aTRCREbW2W7t3bv+9zpyTiQ+9b1IfOp78cGGY9JGs1mIr74S4sEHpefWfp9GbkejO4u7J7xkf93Ep74XqQ99Kr6+7BqPX6tVb1FRQkybJnK/+k6888kvYvFts0Rm/CW+fx+9XojSUrd/HxaLVeRGxQkBiK8+/M7tPsfPlYt9Hbs3+f0KQqLEg396WiQ++Z397yP5if+JV66eKiq1+ia/zpcDx4p1hwrqHM+Jc+Vi3l1/9fl5+m+/MWLci6vEMyt3i8+2nhSPf5ElbnxquVibnOLR6+RExYlVT7whzl6oEEIIYTRZxM9LV4ttlwyy73PXhJfF4FfTxVNf7RLf7zojPthwTNz1wW+ixzM/iF+TnN6vVy8h/vpXIXbtEmL7diFeekmIYcOERaX26fderQkSR195Uwir1eVcl+8/JI4lev5vkreL77b99QX1/r/urZKSEtGUz2+PKiNnzpxB586dsXnzZgxzmjv/2muv4ZNPPsEheYEuJ7169cKUKVPwzDPP2Ldt2bIFw4cPx5kzZ9DJzeXnL4bKSFOc+j0LeWlrkfK3RxGkrVs9ASCNP3/3HTBliusUQQ8ZS0qR9fJ89Jh2J2L69HC7j6WqGrv++jraR4Wg6+D+Ul9G165S092FC8CxY8DRo9IxXX+9tKJkPX49VIBQnRaDbVePrSMvDweWrYImPw+94qOk8XXbb1Uwm6XfzGpqkFtQgsPqcOT96XZERoWjXWgQIoODUG2y4HyFCRcqaqD7fSsSMreiV/c4hHeIkSo+4eHS65SWAqWlOJtzFoVnL6BHQjT0IcFSOVxnK3Pb3quirAInzhTDBECoNbabGlaoIMxmwGyGsFggLFYItQpCLR2zVa1G+3ADencIkVbXlSsXgwYBN9zgeB+bapMFh7fuQuWXK9HeoEZy3x5Qx3cCOnaUjr2wUFot9exZ6WY0Os6PRiPdVCrHb3MqlfRe8nVm3Dj4/VqUHD6BwY9Nrbdcfnj9dhQu/QIh4aGIiI5AZEwkImMiYKquQfG5CygtLEFFUTHKhBqHx90GVUwM9EEa6LVqQADVZguqTRboT53EpRvS0Dc+AsHhodLaJ/L6J9XVQHU1qssqcCC/DLGPPojOl3Z3ezwV5VXY89wb6BmlQ0y3zkCHDkD79tJrnT8PFBZCnDuHw/tOIiQyDAm9uwGxsdItLExa6bSwEOazBdi35wQKe/dFr1vGIiHatZonhMDxc+U4/cFn6Jx7HD26d5J+fsLCpOFRoxEoLwcqKnAq+xzygiNx2eMzEB4ZVueYrVaBXUv/h+Jj2Yi9fzIujY+sc77LjWbsXJ8Jw08/YuDdN0E7oL/b38xLTucjY+UaRKnM6GhQo30QoLOYAJMJNdVGFJVU4XxpJSoqjEjsEIa46DDHz4cQUmWjpgalJeU4drYMHe67C11GDHJ7rk2VVch6/l+wns2HRq+HRq+D2qCHWquFucYEi8kEs9EEi7EGAKBSq6BSq6WbyvZbudUqVQytVqh1QdC3i4ShXRRCoiMRHBWOypJylBdeQFVRMaqLSyBqTNAEG6A16KENCUZQsB4WkwWWqiqYq6phqTbCWmOW/q3ZKiHyh5X8PiphlaqpKucKiAqAgMoqICAAqwAgAKjqLYBIn4ICjdclPNvH3fvVfq+G92mquq/l+snudNy139Bpx7jp9yD5umHwpaZWRrwKI1u2bMHQoUPt21999VV89tlnOHiw7hoTvXr1wn333Ye5c+fat23evBkjRoxAXl4eOnZ005Dn5TdDRERE/qOpn98ezaZp3749NBoN8vNdp9IVFBQgLi7O7XM6duzodn+tVouYmBhP3p6IiIguQh6FEZ1Oh5SUFKSnp7tsT09Pdxm2cTZ06NA6+//8889ITU1FkNwMRURERAHL43VG5syZgw8//BCLFy/GgQMH8NhjjyE7OxszbEuRz507F5MnT7bvP2PGDJw6dQpz5szBgQMHsHjxYnz00Ud4/PHHffddEBERUZvl8TojEydORFFREV566SXk5eWhb9++SEtLQ6JtjYO8vDxkZzsWIUpKSkJaWhoee+wxvPvuu4iPj8fbb7990a4xQkRERJ7xeJ0RJbCBlYiIqO1pkQZWIiIiIl9jGCEiIiJFMYwQERGRohhGiIiISFEMI0RERKQohhEiIiJSFMMIERERKYphhIiIiBTl8QqsSpDXZSstLVX4SIiIiKip5M/txtZXbRNhpKysDACQkJCg8JEQERGRp8rKyhAZGVnv421iOXir1YozZ84gPDwcKpXKZ69bWlqKhIQE5OTkcJn5FsZz3bp4vlsPz3Xr4bluPb4610IIlJWVIT4+Hmp1/Z0hbaIyolar0aVLlxZ7/YiICP5gtxKe69bF8916eK5bD8916/HFuW6oIiJjAysREREpimGEiIiIFBXQYUSv1+P555+HXq9X+lAuejzXrYvnu/XwXLcenuvW09rnuk00sBIREdHFK6ArI0RERKQ8hhEiIiJSFMMIERERKYphhIiIiBTFMEJERESKCugwsmDBAiQlJcFgMCAlJQUbN25U+pDavHnz5mHQoEEIDw9HbGwsbr75Zhw6dMhlHyEEXnjhBcTHxyM4OBhXX3019u3bp9ARXxzmzZsHlUqF2bNn27fxPPtWbm4u7r77bsTExCAkJAQDBgxARkaG/XGeb98wm83461//iqSkJAQHByM5ORkvvfQSrFarfR+ea+9s2LABN954I+Lj46FSqfDtt9+6PN6U82o0GvHII4+gffv2CA0NxU033YTTp083/+BEgFq+fLkICgoSH3zwgdi/f7949NFHRWhoqDh16pTSh9am3XDDDeLjjz8We/fuFVlZWWL8+PGia9euory83L7P66+/LsLDw8XXX38t9uzZIyZOnCg6deokSktLFTzytmvbtm2iW7duol+/fuLRRx+1b+d59p3z58+LxMREMWXKFPH777+LEydOiDVr1oijR4/a9+H59o1XXnlFxMTEiO+//16cOHFCfPnllyIsLEzMnz/fvg/PtXfS0tLEs88+K77++msBQHzzzTcujzflvM6YMUN07txZpKeni507d4prrrlG9O/fX5jN5mYdW8CGkcGDB4sZM2a4bOvdu7d4+umnFTqii1NBQYEAINavXy+EEMJqtYqOHTuK119/3b5PdXW1iIyMFO+9955Sh9lmlZWViZ49e4r09HQxatQoexjhefatp556SowYMaLex3m+fWf8+PFi6tSpLttuueUWcffddwsheK59pXYYacp5LS4uFkFBQWL58uX2fXJzc4VarRarV69u1vEE5DBNTU0NMjIyMGbMGJftY8aMwZYtWxQ6qotTSUkJACA6OhoAcOLECeTn57uce71ej1GjRvHce+Ghhx7C+PHjMXr0aJftPM++tWrVKqSmpuL2229HbGwsrrjiCnzwwQf2x3m+fWfEiBH45ZdfcPjwYQDArl27sGnTJowbNw4Az3VLacp5zcjIgMlkctknPj4effv2bfa5bxNX7fW1wsJCWCwWxMXFuWyPi4tDfn6+Qkd18RFCYM6cORgxYgT69u0LAPbz6+7cnzp1qtWPsS1bvnw5du7cie3bt9d5jOfZt44fP46FCxdizpw5eOaZZ7Bt2zbMmjULer0ekydP5vn2oaeeegolJSXo3bs3NBoNLBYLXn31VUyaNAkAf7ZbSlPOa35+PnQ6Hdq1a1dnn+Z+dgZkGJGpVCqXPwsh6mwj7z388MPYvXs3Nm3aVOcxnvvmycnJwaOPPoqff/4ZBoOh3v14nn3DarUiNTUVr732GgDgiiuuwL59+7Bw4UJMnjzZvh/Pd/OtWLECn3/+Of773//isssuQ1ZWFmbPno34+Hjce++99v14rluGN+fVF+c+IIdp2rdvD41GUyfJFRQU1EmF5J1HHnkEq1atwq+//oouXbrYt3fs2BEAeO6bKSMjAwUFBUhJSYFWq4VWq8X69evx9ttvQ6vV2s8lz7NvdOrUCZdeeqnLtj59+iA7OxsAf6596YknnsDTTz+NO+64A5dffjnuuecePPbYY5g3bx4AnuuW0pTz2rFjR9TU1ODChQv17uOtgAwjOp0OKSkpSE9Pd9menp6OYcOGKXRUFwchBB5++GGsXLkSa9euRVJSksvjSUlJ6Nixo8u5r6mpwfr163nuPXDddddhz549yMrKst9SU1Nx1113ISsrC8nJyTzPPjR8+PA6U9QPHz6MxMREAPy59qXKykqo1a4fTRqNxj61l+e6ZTTlvKakpCAoKMhln7y8POzdu7f5575Z7a9tmDy196OPPhL79+8Xs2fPFqGhoeLkyZNKH1qb9uCDD4rIyEixbt06kZeXZ79VVlba93n99ddFZGSkWLlypdizZ4+YNGkSp+X5gPNsGiF4nn1p27ZtQqvVildffVUcOXJELF26VISEhIjPP//cvg/Pt2/ce++9onPnzvapvStXrhTt27cXTz75pH0fnmvvlJWViczMTJGZmSkAiDfffFNkZmbal7RoynmdMWOG6NKli1izZo3YuXOnuPbaazm1t7neffddkZiYKHQ6nRg4cKB9+il5D4Db28cff2zfx2q1iueff1507NhR6PV6cdVVV4k9e/Yod9AXidphhOfZt7777jvRt29fodfrRe/evcWiRYtcHuf59o3S0lLx6KOPiq5duwqDwSCSk5PFs88+K4xGo30fnmvv/Prrr27/f7733nuFEE07r1VVVeLhhx8W0dHRIjg4WPzxj38U2dnZzT42lRBCNK+2QkREROS9gOwZISIiIv/BMEJERESKYhghIiIiRTGMEBERkaIYRoiIiEhRDCNERESkKIYRIiIiUhTDCBERESmKYYSIiIgUxTBCREREimIYISIiIkX9Pw0QCsF1OlC8AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(h['loss'])\n",
    "plt.plot(h['val_loss'] , c = \"red\")\n",
    "plt.title(\"loss vs v-loss\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86c41a08",
   "metadata": {},
   "source": [
    "# Checking the file location is present or not"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "541dcd593616de8e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:22:46.418465Z",
     "start_time": "2024-09-29T01:22:44.358365Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1s/step\n",
      "Parabasal\n"
     ]
    }
   ],
   "source": [
    "# checking the file location is present or not [DEMO]\n",
    "'''\n",
    "import numpy as np\n",
    "from tensorflow.keras.preprocessing import image\n",
    "\n",
    "def cancerPrediction(path):\n",
    "    classes_dir = [\"Dyskeratotic\", \"Koilocytotic\", \"Metaplastic\", \"Parabasal\", \"Superficial-Intermediate\"]\n",
    "    img = image.load_img(path, target_size=(224, 224))\n",
    "    norm_img = image.img_to_array(img) / 255.0\n",
    "    input_arr_img = np.array([norm_img])\n",
    "    pred = np.argmax(model.predict(input_arr_img))\n",
    "    print(classes_dir[pred])\n",
    "path = \"/home/iamsanko/Downloads/Test 1/test/Parabasal/007.jpg\"\n",
    "cancerPrediction(path)'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "81757c91932891ac",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-29T01:29:01.915180Z",
     "start_time": "2024-09-29T01:28:56.864108Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing: 004.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 163ms/step\n",
      "Koilocytotic\n",
      "Processing: 012.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 178ms/step\n",
      "Koilocytotic\n",
      "Processing: 011.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 161ms/step\n",
      "Koilocytotic\n",
      "Processing: 005.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 223ms/step\n",
      "Koilocytotic\n",
      "Processing: 009.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205ms/step\n",
      "Koilocytotic\n",
      "Processing: 006.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 168ms/step\n",
      "Koilocytotic\n",
      "Processing: 007.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 177ms/step\n",
      "Koilocytotic\n",
      "Processing: 010.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 164ms/step\n",
      "Koilocytotic\n",
      "Processing: 013.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 149ms/step\n",
      "Koilocytotic\n",
      "Processing: 014.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 166ms/step\n",
      "Koilocytotic\n",
      "Processing: 001.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 165ms/step\n",
      "Superficial-Intermediate\n",
      "Processing: 003.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213ms/step\n",
      "Koilocytotic\n",
      "Processing: 008.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 165ms/step\n",
      "Koilocytotic\n",
      "Processing: 002.jpg\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 161ms/step\n",
      "Koilocytotic\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "def batchCancerPrediction(directory):\n",
    "    for filename in os.listdir(directory):\n",
    "        path = os.path.join(directory, filename)\n",
    "        print(f\"Processing: {filename}\")\n",
    "        cancerPrediction(path)\n",
    "local_directory = \"/home/iamsanko/Downloads/Test 1/test/Koilocytotic\"\n",
    "batchCancerPrediction(local_directory)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a90fe797bf55572b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:00:07.309923467Z",
     "start_time": "2024-09-29T14:31:20.751376Z"
    },
    "jupyter": {
     "is_executing": true
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[I 2024-09-29 20:01:24,916] A new study created in memory with name: no-name-fb477780-8693-4213-ad69-c9f5e7800b65\n",
      "[I 2024-09-29 20:05:03,648] Trial 0 finished with value: 0.928164196123147 and parameters: {'filters': 36, 'dropout_rate': 0.07492744765636328, 'dense_units': 72, 'learning_rate': 0.004478865574974525}. Best is trial 0 with value: 0.928164196123147.\n",
      "[I 2024-09-29 20:07:23,812] Trial 1 finished with value: 0.935005701254276 and parameters: {'filters': 57, 'dropout_rate': 0.3249980585697148, 'dense_units': 87, 'learning_rate': 5.664953772898972e-05}. Best is trial 1 with value: 0.935005701254276.\n",
      "[I 2024-09-29 20:09:11,284] Trial 2 finished with value: 0.8916761687571265 and parameters: {'filters': 37, 'dropout_rate': 0.10920142955492323, 'dense_units': 119, 'learning_rate': 2.009319907890419e-05}. Best is trial 1 with value: 0.935005701254276.\n",
      "[I 2024-09-29 20:12:36,974] Trial 3 finished with value: 0.95096921322691 and parameters: {'filters': 79, 'dropout_rate': 0.16795102523253103, 'dense_units': 114, 'learning_rate': 0.00013193677226920296}. Best is trial 3 with value: 0.95096921322691.\n",
      "[I 2024-09-29 20:17:21,747] Trial 4 finished with value: 0.8916761687571265 and parameters: {'filters': 69, 'dropout_rate': 0.14996438961195696, 'dense_units': 60, 'learning_rate': 1.2694986028786546e-05}. Best is trial 3 with value: 0.95096921322691.\n",
      "[I 2024-09-29 21:19:52,283] Trial 18 finished with value: 0.9475484606613455 and parameters: {'filters': 86, 'dropout_rate': 0.4844921692014557, 'dense_units': 81, 'learning_rate': 0.0001618608324599456}. Best is trial 9 with value: 0.9555302166476625.\n",
      "[I 2024-09-29 21:24:42,190] Trial 19 finished with value: 0.9201824401368301 and parameters: {'filters': 111, 'dropout_rate': 0.31841951386171846, 'dense_units': 102, 'learning_rate': 0.003257705244928196}. Best is trial 9 with value: 0.9555302166476625.\n",
      "[I 2024-09-29 21:26:39,324] Trial 20 finished with value: 0.9133409350057012 and parameters: {'filters': 55, 'dropout_rate': 0.3930833370540597, 'dense_units': 72, 'learning_rate': 0.009257062224634746}. Best is trial 9 with value: 0.9555302166476625.\n",
      "[I 2024-09-29 21:30:24,000] Trial 21 finished with value: 0.9555302166476625 and parameters: {'filters': 96, 'dropout_rate': 0.20322969902439014, 'dense_units': 103, 'learning_rate': 0.00020326856503089563}. Best is trial 9 with value: 0.9555302166476625.\n",
      "[I 2024-09-29 21:34:04,615] Trial 22 finished with value: 0.9623717217787914 and parameters: {'filters': 96, 'dropout_rate': 0.23117714794397584, 'dense_units': 107, 'learning_rate': 0.0003091416059328105}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:37:26,593] Trial 23 finished with value: 0.9475484606613455 and parameters: {'filters': 81, 'dropout_rate': 0.21838094028590022, 'dense_units': 108, 'learning_rate': 9.191064657945529e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:41:08,921] Trial 24 finished with value: 0.9612314709236032 and parameters: {'filters': 93, 'dropout_rate': 0.18492727624278937, 'dense_units': 125, 'learning_rate': 0.00030075077337221356}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:44:08,473] Trial 25 finished with value: 0.9395667046750285 and parameters: {'filters': 78, 'dropout_rate': 0.27494553021885343, 'dense_units': 126, 'learning_rate': 0.00028277395509755676}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:48:38,729] Trial 26 finished with value: 0.95096921322691 and parameters: {'filters': 105, 'dropout_rate': 0.17652471713519363, 'dense_units': 128, 'learning_rate': 0.00042329865273641647}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:53:35,759] Trial 27 finished with value: 0.9566704675028507 and parameters: {'filters': 116, 'dropout_rate': 0.24107526744685043, 'dense_units': 110, 'learning_rate': 8.546757667117646e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 21:58:27,700] Trial 28 finished with value: 0.9452679589509693 and parameters: {'filters': 116, 'dropout_rate': 0.24392131534916267, 'dense_units': 111, 'learning_rate': 9.326723397402912e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:03:54,951] Trial 29 finished with value: 0.9543899657924744 and parameters: {'filters': 127, 'dropout_rate': 0.08928537275168255, 'dense_units': 92, 'learning_rate': 7.809177978273874e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:08:11,981] Trial 30 finished with value: 0.9087799315849487 and parameters: {'filters': 103, 'dropout_rate': 0.04154382136853085, 'dense_units': 107, 'learning_rate': 2.6095896505450114e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:11:49,317] Trial 31 finished with value: 0.9521094640820981 and parameters: {'filters': 92, 'dropout_rate': 0.2570615425380756, 'dense_units': 119, 'learning_rate': 0.0003184716348490172}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:16:39,224] Trial 32 finished with value: 0.95096921322691 and parameters: {'filters': 115, 'dropout_rate': 0.3113512179550506, 'dense_units': 83, 'learning_rate': 0.000637014386263623}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:18:39,449] Trial 33 finished with value: 0.9486887115165337 and parameters: {'filters': 60, 'dropout_rate': 0.18383328031556284, 'dense_units': 72, 'learning_rate': 0.00012826703279488535}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:22:07,252] Trial 34 finished with value: 0.9589509692132269 and parameters: {'filters': 85, 'dropout_rate': 0.23692408177128113, 'dense_units': 117, 'learning_rate': 0.0002590854393097567}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:26:22,045] Trial 35 finished with value: 0.9475484606613455 and parameters: {'filters': 101, 'dropout_rate': 0.1196149719931196, 'dense_units': 119, 'learning_rate': 7.168113247566171e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:31:41,929] Trial 36 finished with value: 0.9623717217787914 and parameters: {'filters': 119, 'dropout_rate': 0.2438911508798292, 'dense_units': 124, 'learning_rate': 0.00012557383891169982}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:34:34,931] Trial 37 finished with value: 0.9384264538198404 and parameters: {'filters': 75, 'dropout_rate': 0.15338197681304785, 'dense_units': 124, 'learning_rate': 0.00014077601958500474}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:37:56,131] Trial 38 finished with value: 0.9498289623717218 and parameters: {'filters': 88, 'dropout_rate': 0.18542383400104348, 'dense_units': 115, 'learning_rate': 0.0002074209902210208}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:41:47,925] Trial 39 finished with value: 0.9475484606613455 and parameters: {'filters': 91, 'dropout_rate': 0.08412255952210101, 'dense_units': 121, 'learning_rate': 4.5929278350432676e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:44:01,483] Trial 40 finished with value: 0.9441277080957811 and parameters: {'filters': 64, 'dropout_rate': 0.004686865365463921, 'dense_units': 115, 'learning_rate': 0.0006962222797518744}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:49:02,036] Trial 41 finished with value: 0.9612314709236032 and parameters: {'filters': 119, 'dropout_rate': 0.24051473388390213, 'dense_units': 128, 'learning_rate': 0.00012027521713306355}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:54:23,186] Trial 42 finished with value: 0.9566704675028507 and parameters: {'filters': 123, 'dropout_rate': 0.28690978707529813, 'dense_units': 127, 'learning_rate': 0.0002907658268399043}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 22:59:39,947] Trial 43 finished with value: 0.9452679589509693 and parameters: {'filters': 121, 'dropout_rate': 0.23588226316466718, 'dense_units': 122, 'learning_rate': 0.0001120680391277298}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:05:17,061] Trial 44 finished with value: 0.9521094640820981 and parameters: {'filters': 112, 'dropout_rate': 0.19919936456737655, 'dense_units': 117, 'learning_rate': 0.00020906037788352128}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:10:29,063] Trial 45 finished with value: 0.9486887115165337 and parameters: {'filters': 105, 'dropout_rate': 0.13743855377134187, 'dense_units': 123, 'learning_rate': 5.638723327428649e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:14:05,970] Trial 46 finished with value: 0.9589509692132269 and parameters: {'filters': 86, 'dropout_rate': 0.26434936387488756, 'dense_units': 112, 'learning_rate': 0.0004362726268278419}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:18:31,201] Trial 47 finished with value: 0.9475484606613455 and parameters: {'filters': 100, 'dropout_rate': 0.15173037387329225, 'dense_units': 128, 'learning_rate': 0.0009818589311795694}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:21:18,810] Trial 48 finished with value: 0.9395667046750285 and parameters: {'filters': 74, 'dropout_rate': 0.22461220175498403, 'dense_units': 106, 'learning_rate': 3.145222970608124e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:26:27,634] Trial 49 finished with value: 0.9144811858608894 and parameters: {'filters': 119, 'dropout_rate': 0.1655060364557169, 'dense_units': 118, 'learning_rate': 1.2218828708872364e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:31:01,340] Trial 50 finished with value: 0.9304446978335233 and parameters: {'filters': 107, 'dropout_rate': 0.28829536474123363, 'dense_units': 124, 'learning_rate': 1.759715598273131e-05}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:34:20,380] Trial 51 finished with value: 0.9498289623717218 and parameters: {'filters': 83, 'dropout_rate': 0.2636366784972327, 'dense_units': 113, 'learning_rate': 0.0004213800770075098}. Best is trial 22 with value: 0.9623717217787914.\n",
      "[I 2024-09-29 23:38:05,661] Trial 52 finished with value: 0.9635119726339795 and parameters: {'filters': 93, 'dropout_rate': 0.2570099534225515, 'dense_units': 95, 'learning_rate': 0.00037734730659259646}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:41:51,001] Trial 53 finished with value: 0.9589509692132269 and parameters: {'filters': 93, 'dropout_rate': 0.20381388896089123, 'dense_units': 91, 'learning_rate': 0.00018100835640164593}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:43:14,937] Trial 54 finished with value: 0.9338654503990877 and parameters: {'filters': 26, 'dropout_rate': 0.2966505617360274, 'dense_units': 96, 'learning_rate': 0.0002671793862905596}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:45:50,943] Trial 55 finished with value: 0.9498289623717218 and parameters: {'filters': 69, 'dropout_rate': 0.23550206965740905, 'dense_units': 104, 'learning_rate': 0.0006138857071990256}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:50:40,650] Trial 56 finished with value: 0.9498289623717218 and parameters: {'filters': 99, 'dropout_rate': 0.33443789911489913, 'dense_units': 116, 'learning_rate': 0.0001587212788235099}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:53:41,769] Trial 57 finished with value: 0.9486887115165337 and parameters: {'filters': 78, 'dropout_rate': 0.1934154265813266, 'dense_units': 86, 'learning_rate': 0.00024855086199189464}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:57:21,045] Trial 58 finished with value: 0.9623717217787914 and parameters: {'filters': 90, 'dropout_rate': 0.21287969092051226, 'dense_units': 121, 'learning_rate': 0.0003683986427296574}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-29 23:59:15,440] Trial 59 finished with value: 0.9498289623717218 and parameters: {'filters': 52, 'dropout_rate': 0.16640442236151057, 'dense_units': 95, 'learning_rate': 0.0011247114196793625}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-30 00:03:53,927] Trial 60 finished with value: 0.9395667046750285 and parameters: {'filters': 111, 'dropout_rate': 0.25499503218688735, 'dense_units': 109, 'learning_rate': 0.00040254354058706215}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-30 00:07:28,883] Trial 61 finished with value: 0.9498289623717218 and parameters: {'filters': 89, 'dropout_rate': 0.21753451528618253, 'dense_units': 121, 'learning_rate': 0.0005196324060595649}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-30 00:11:10,166] Trial 62 finished with value: 0.9600912200684151 and parameters: {'filters': 94, 'dropout_rate': 0.2325656873894942, 'dense_units': 125, 'learning_rate': 0.00033667808278448384}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-30 00:14:50,125] Trial 63 finished with value: 0.9578107183580388 and parameters: {'filters': 96, 'dropout_rate': 0.2710361801089008, 'dense_units': 125, 'learning_rate': 0.00011496108199357419}. Best is trial 52 with value: 0.9635119726339795.\n",
      "[I 2024-09-30 00:20:09,930] Trial 64 finished with value: 0.9669327251995439 and parameters: {'filters': 125, 'dropout_rate': 0.2260519806580994, 'dense_units': 128, 'learning_rate': 0.0005190067073972647}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:25:22,322] Trial 65 finished with value: 0.9600912200684151 and parameters: {'filters': 126, 'dropout_rate': 0.20502719443025788, 'dense_units': 76, 'learning_rate': 0.0008296033105080783}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:30:37,932] Trial 66 finished with value: 0.9327251995438997 and parameters: {'filters': 123, 'dropout_rate': 0.2980163463744471, 'dense_units': 120, 'learning_rate': 0.0003540863504423562}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:35:35,517] Trial 67 finished with value: 0.9635119726339795 and parameters: {'filters': 118, 'dropout_rate': 0.2560874331029203, 'dense_units': 127, 'learning_rate': 0.000565014223275562}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:40:13,246] Trial 68 finished with value: 0.9418472063854048 and parameters: {'filters': 116, 'dropout_rate': 0.2155232074900933, 'dense_units': 42, 'learning_rate': 0.001574149685374991}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:44:41,944] Trial 69 finished with value: 0.9521094640820981 and parameters: {'filters': 112, 'dropout_rate': 0.1768515887524874, 'dense_units': 101, 'learning_rate': 0.0005428887034467868}. Best is trial 64 with value: 0.9669327251995439.\n",
      "[I 2024-09-30 00:49:11,603] Trial 70 finished with value: 0.9680729760547321 and parameters: {'filters': 108, 'dropout_rate': 0.36821133628071573, 'dense_units': 122, 'learning_rate': 0.0007581239747499647}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 00:53:21,974] Trial 71 finished with value: 0.9486887115165337 and parameters: {'filters': 103, 'dropout_rate': 0.4335354318953007, 'dense_units': 123, 'learning_rate': 0.0008485379170458113}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 00:57:45,041] Trial 72 finished with value: 0.9418472063854048 and parameters: {'filters': 109, 'dropout_rate': 0.33221103970053434, 'dense_units': 112, 'learning_rate': 0.00115757087745662}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:03:05,837] Trial 73 finished with value: 0.9521094640820981 and parameters: {'filters': 128, 'dropout_rate': 0.4006369015175401, 'dense_units': 120, 'learning_rate': 0.0006656599381026769}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:08:13,658] Trial 74 finished with value: 0.936145952109464 and parameters: {'filters': 124, 'dropout_rate': 0.36550749583483394, 'dense_units': 126, 'learning_rate': 0.0005757260640255193}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:12:43,924] Trial 75 finished with value: 0.9475484606613455 and parameters: {'filters': 114, 'dropout_rate': 0.2491911366175053, 'dense_units': 64, 'learning_rate': 0.0017703286777412711}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:17:38,431] Trial 76 finished with value: 0.9258836944127709 and parameters: {'filters': 118, 'dropout_rate': 0.4665072380849823, 'dense_units': 128, 'learning_rate': 0.0003617360005014548}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:21:49,143] Trial 77 finished with value: 0.9475484606613455 and parameters: {'filters': 103, 'dropout_rate': 0.2771721047035187, 'dense_units': 115, 'learning_rate': 0.000499794919510603}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:25:24,463] Trial 78 finished with value: 0.9600912200684151 and parameters: {'filters': 96, 'dropout_rate': 0.1923939208582162, 'dense_units': 105, 'learning_rate': 0.0007962536288272269}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:29:46,613] Trial 79 finished with value: 0.9338654503990877 and parameters: {'filters': 108, 'dropout_rate': 0.11459867938550546, 'dense_units': 122, 'learning_rate': 0.00023324532152241598}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:33:18,896] Trial 80 finished with value: 0.9612314709236032 and parameters: {'filters': 91, 'dropout_rate': 0.22543756096303477, 'dense_units': 109, 'learning_rate': 0.00017385786634709134}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:38:33,380] Trial 81 finished with value: 0.9589509692132269 and parameters: {'filters': 122, 'dropout_rate': 0.25019299827431757, 'dense_units': 128, 'learning_rate': 0.00014098083653679903}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:43:25,008] Trial 82 finished with value: 0.9543899657924744 and parameters: {'filters': 119, 'dropout_rate': 0.21218209340393374, 'dense_units': 125, 'learning_rate': 0.00031464203495212015}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:48:22,016] Trial 83 finished with value: 0.9464082098061574 and parameters: {'filters': 119, 'dropout_rate': 0.3127087260719452, 'dense_units': 119, 'learning_rate': 0.00010045720340689696}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:53:09,928] Trial 84 finished with value: 0.9486887115165337 and parameters: {'filters': 113, 'dropout_rate': 0.13099036635968656, 'dense_units': 123, 'learning_rate': 6.730324576098347e-05}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 01:58:29,947] Trial 85 finished with value: 0.9407069555302167 and parameters: {'filters': 126, 'dropout_rate': 0.2620075605346062, 'dense_units': 118, 'learning_rate': 0.001017904614881016}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:02:36,446] Trial 86 finished with value: 0.9600912200684151 and parameters: {'filters': 100, 'dropout_rate': 0.22666623518719523, 'dense_units': 126, 'learning_rate': 0.00038493651575048014}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:06:55,759] Trial 87 finished with value: 0.95096921322691 and parameters: {'filters': 105, 'dropout_rate': 0.24175401557855783, 'dense_units': 114, 'learning_rate': 0.00046293911102456264}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:10:08,196] Trial 88 finished with value: 0.95096921322691 and parameters: {'filters': 81, 'dropout_rate': 0.17327684901064908, 'dense_units': 128, 'learning_rate': 0.00225582170518172}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:15:02,819] Trial 89 finished with value: 0.9566704675028507 and parameters: {'filters': 117, 'dropout_rate': 0.4993156628937555, 'dense_units': 123, 'learning_rate': 0.00022319459090763586}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:19:28,139] Trial 90 finished with value: 0.9372862029646523 and parameters: {'filters': 110, 'dropout_rate': 0.2990092968139772, 'dense_units': 88, 'learning_rate': 0.0002887702726127275}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:22:59,572] Trial 91 finished with value: 0.9521094640820981 and parameters: {'filters': 90, 'dropout_rate': 0.20837011060643842, 'dense_units': 110, 'learning_rate': 0.0001780293861349764}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:26:23,962] Trial 92 finished with value: 0.9612314709236032 and parameters: {'filters': 85, 'dropout_rate': 0.224289988121886, 'dense_units': 117, 'learning_rate': 0.00017871769209327657}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:30:29,473] Trial 93 finished with value: 0.9532497149372862 and parameters: {'filters': 98, 'dropout_rate': 0.18703677061417984, 'dense_units': 121, 'learning_rate': 0.0001411536039141129}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:34:03,253] Trial 94 finished with value: 0.9589509692132269 and parameters: {'filters': 91, 'dropout_rate': 0.27107352747479646, 'dense_units': 126, 'learning_rate': 0.0002564821424470358}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:37:42,166] Trial 95 finished with value: 0.9578107183580388 and parameters: {'filters': 93, 'dropout_rate': 0.2476760954016146, 'dense_units': 113, 'learning_rate': 0.00012621077202003915}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:41:06,660] Trial 96 finished with value: 0.9475484606613455 and parameters: {'filters': 87, 'dropout_rate': 0.2280533894969374, 'dense_units': 99, 'learning_rate': 0.0004681458472036123}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:46:22,509] Trial 97 finished with value: 0.9578107183580388 and parameters: {'filters': 125, 'dropout_rate': 0.1592674277345052, 'dense_units': 96, 'learning_rate': 0.00016540857907340042}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:51:29,959] Trial 98 finished with value: 0.9635119726339795 and parameters: {'filters': 121, 'dropout_rate': 0.20120998336085347, 'dense_units': 124, 'learning_rate': 0.0007366783226851301}. Best is trial 70 with value: 0.9680729760547321.\n",
      "[I 2024-09-30 02:56:23,464] Trial 99 finished with value: 0.9555302166476625 and parameters: {'filters': 120, 'dropout_rate': 0.19466029550239966, 'dense_units': 124, 'learning_rate': 0.0007653289315357687}. Best is trial 70 with value: 0.9680729760547321.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best trial:\n",
      "  Value: 0.9680729760547321\n",
      "  Params: \n",
      "    filters: 108\n",
      "    dropout_rate: 0.36821133628071573\n",
      "    dense_units: 122\n",
      "    learning_rate: 0.0007581239747499647\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader\n",
    "import optuna\n",
    "class SimpleCNN(nn.Module):\n",
    "    def __init__(self, filters, dropout_rate, dense_units):\n",
    "        super(SimpleCNN, self).__init__()\n",
    "        self.conv1 = nn.Conv2d(3, filters, kernel_size=3, padding=1)\n",
    "        self.conv2 = nn.Conv2d(filters, filters * 2, kernel_size=3, padding=1)\n",
    "        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)\n",
    "        self.dropout = nn.Dropout(dropout_rate)\n",
    "        self.fc1 = nn.Linear(filters * 2 * 16 * 16, dense_units)  \n",
    "        self.fc2 = nn.Linear(dense_units, 5) \n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.pool(nn.functional.relu(self.conv1(x)))\n",
    "        x = self.pool(nn.functional.relu(self.conv2(x)))\n",
    "        x = x.view(-1, self.num_flat_features(x))\n",
    "        x = self.dropout(x)\n",
    "        x = nn.functional.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x\n",
    "\n",
    "    def num_flat_features(self, x):\n",
    "        size = x.size()[1:]  \n",
    "        num_features = 1\n",
    "        for s in size:\n",
    "            num_features *= s\n",
    "        return num_features\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),  # Resize to 64x64\n",
    "    transforms.ToTensor(),\n",
    "])\n",
    "train_data = datasets.ImageFolder(root='/home/iamsanko/Downloads/Test 1/train', transform=transform)\n",
    "val_data = datasets.ImageFolder(root='/home/iamsanko/Downloads/Test 1/val', transform=transform)\n",
    "\n",
    "train_loader = DataLoader(train_data, batch_size=64, shuffle=True)\n",
    "val_loader = DataLoader(val_data, batch_size=64, shuffle=False)\n",
    "def objective(trial):\n",
    "    filters = trial.suggest_int(\"filters\", 16, 128)\n",
    "    dropout_rate = trial.suggest_float(\"dropout_rate\", 0.0, 0.5)\n",
    "    dense_units = trial.suggest_int(\"dense_units\", 32, 128)\n",
    "    learning_rate = trial.suggest_float(\"learning_rate\", 1e-5, 1e-2, log=True)\n",
    "\n",
    "    model = SimpleCNN(filters, dropout_rate, dense_units)\n",
    "    optimizer = optim.Adam(model.parameters(), lr=learning_rate)\n",
    "    criterion = nn.CrossEntropyLoss()\n",
    "    model.train()\n",
    "    for epoch in range(10):  #defined the ranges upto 10+- \n",
    "        for inputs, labels in train_loader:\n",
    "            optimizer.zero_grad()\n",
    "            outputs = model(inputs)\n",
    "            loss = criterion(outputs, labels)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "    model.eval()\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    with torch.no_grad():\n",
    "        for inputs, labels in val_loader:\n",
    "            outputs = model(inputs)\n",
    "            _, predicted = torch.max(outputs.data, 1)\n",
    "            total += labels.size(0)\n",
    "            correct += (predicted == labels).sum().item()\n",
    "\n",
    "    accuracy = correct / total\n",
    "    return accuracy\n",
    "study = optuna.create_study(direction=\"maximize\")\n",
    "study.optimize(objective, n_trials=100)\n",
    "\n",
    "print(\"Best trial:\")\n",
    "trial = study.best_trial\n",
    "print(\"  Value: {}\".format(trial.value))\n",
    "print(\"  Params: \")\n",
    "for key, value in trial.params.items():\n",
    "    print(\"    {}: {}\".format(key, value)) # took 11hrs & 28mito run\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "51e216f8ffbf9381",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:47:43.851073Z",
     "start_time": "2024-09-30T02:47:43.639717Z"
    }
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import os\n",
    "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
    "import numpy as np\n",
    "\n",
    "def load_images_from_directory(data_dir, target_size=(32, 32)):\n",
    "    images = []\n",
    "    for filename in os.listdir(data_dir):\n",
    "        img_path = os.path.join(data_dir, filename)\n",
    "        if os.path.isfile(img_path):\n",
    "            img = load_img(img_path, target_size=target_size, color_mode='grayscale')\n",
    "            img_array = img_to_array(img)\n",
    "            img_array = (img_array - 127.5) / 127.5  # Normalize to [-1, 1]\n",
    "            images.append(img_array)\n",
    "    return np.array(images)\n",
    "\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "X_train = load_images_from_directory(data_dir)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2311032b239269b5",
   "metadata": {},
   "source": [
    "## Building a generator checker\n",
    " With the help of Fn. generator from the previous cell's has the model and which was needed to add the latest dimention"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "2097715077bbbf70",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:47:54.179539Z",
     "start_time": "2024-09-30T02:47:54.110384Z"
    }
   },
   "outputs": [],
   "source": [
    "def build_generator(latent_dim):\n",
    "    model = tf.keras.Sequential()\n",
    "    model.add(tf.keras.layers.Dense(128 * 8 * 8, input_dim=latent_dim))\n",
    "    model.add(tf.keras.layers.LeakyReLU(alpha=0.2))\n",
    "    model.add(tf.keras.layers.Reshape((8, 8, 128)))\n",
    "    model.add(tf.keras.layers.UpSampling2D())\n",
    "    model.add(tf.keras.layers.Conv2D(128, (5, 5), padding='same'))\n",
    "    model.add(tf.keras.layers.LeakyReLU(alpha=0.2))\n",
    "    model.add(tf.keras.layers.UpSampling2D())\n",
    "    model.add(tf.keras.layers.Conv2D(64, (5, 5), padding='same'))\n",
    "    model.add(tf.keras.layers.LeakyReLU(alpha=0.2))\n",
    "    model.add(tf.keras.layers.Conv2D(1, (5, 5), padding='same', activation='tanh'))\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "2640f41433454ae9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:48:03.259994Z",
     "start_time": "2024-09-30T02:48:03.253980Z"
    }
   },
   "outputs": [],
   "source": [
    "def build_discriminator(img_shape):\n",
    "    model = tf.keras.Sequential()\n",
    "    model.add(tf.keras.layers.Conv2D(64, (5, 5), padding='same', input_shape=img_shape))\n",
    "    model.add(tf.keras.layers.LeakyReLU(alpha=0.2))\n",
    "    model.add(tf.keras.layers.Dropout(0.3))\n",
    "    model.add(tf.keras.layers.Conv2D(128, (5, 5), padding='same'))\n",
    "    model.add(tf.keras.layers.LeakyReLU(alpha=0.2))\n",
    "    model.add(tf.keras.layers.Dropout(0.3))\n",
    "    model.add(tf.keras.layers.Flatten())\n",
    "    model.add(tf.keras.layers.Dense(1, activation='sigmoid'))\n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "e915d47c60a37920",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:48:28.405617Z",
     "start_time": "2024-09-30T02:48:15.621735Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n",
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/activations/leaky_relu.py:41: UserWarning: Argument `alpha` is deprecated. Use `negative_slope` instead.\n",
      "  warnings.warn(\n",
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "# Dimensions\n",
    "img_shape = (32, 32, 1)  \n",
    "latent_dim = 100\n",
    "generator = build_generator(latent_dim)\n",
    "discriminator = build_discriminator(img_shape)\n",
    "discriminator.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "discriminator.trainable = False\n",
    "gan_input = tf.keras.Input(shape=(latent_dim,))\n",
    "generated_image = generator(gan_input)\n",
    "gan_output = discriminator(generated_image)\n",
    "gan = tf.keras.Model(gan_input, gan_output)\n",
    "gan.compile(loss='binary_crossentropy', optimizer='adam')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "c5a2c16483319cb8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:49:59.419105Z",
     "start_time": "2024-09-30T02:49:59.374724Z"
    }
   },
   "outputs": [],
   "source": [
    "def train_gan(epochs, batch_size):\n",
    "    for epoch in range(epochs):\n",
    "        idx = np.random.randint(0, X_train.shape[0], batch_size)\n",
    "        real_images = X_train[idx]\n",
    "\n",
    "        noise = np.random.normal(0, 1, (batch_size, latent_dim))\n",
    "        fake_images = generator.predict(noise)\n",
    "\n",
    "        real_labels = np.ones((batch_size, 1))\n",
    "        fake_labels = np.zeros((batch_size, 1))\n",
    "\n",
    "        d_loss_real = discriminator.train_on_batch(real_images, real_labels)\n",
    "        d_loss_fake = discriminator.train_on_batch(fake_images, fake_labels)\n",
    "        d_loss = 0.5 * np.add(d_loss_real, d_loss_fake)\n",
    "        noise = np.random.normal(0, 1, (batch_size, latent_dim))\n",
    "        valid_labels = np.ones((batch_size, 1))\n",
    "        g_loss = gan.train_on_batch(noise, valid_labels)\n",
    "        print(f\"{epoch}/{epochs} [D loss: {d_loss[0]:.4f}, acc.: {d_loss[1]:.2f}] [G loss: {g_loss:.4f}]\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78be8357",
   "metadata": {},
   "source": [
    "# Check the shape is got inn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "237dcebf455b40f3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:50:58.411483Z",
     "start_time": "2024-09-30T02:50:58.400870Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape: (0,)\n"
     ]
    }
   ],
   "source": [
    "print(\"X_train shape:\", X_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "d45082e9ce67b188",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:52:09.370170Z",
     "start_time": "2024-09-30T02:52:09.364209Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded images shape: (0,)\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
    "\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "X_train = []\n",
    "\n",
    "# Load images\n",
    "for filename in os.listdir(data_dir):\n",
    "    img_path = os.path.join(data_dir, filename)\n",
    "    if img_path.endswith('.png') or img_path.endswith('.jpg'):\n",
    "        img = load_img(img_path, target_size=(64, 64)) \n",
    "        img_array = img_to_array(img)\n",
    "        X_train.append(img_array)\n",
    "X_train = np.array(X_train)\n",
    "\n",
    "print(\"Loaded images shape:\", X_train.shape) \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "b484d092f8cacff2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:52:13.451717Z",
     "start_time": "2024-09-30T02:52:13.446738Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded images shape: (0,)\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
    "import os\n",
    "import numpy as np\n",
    "\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "X_train = []\n",
    "\n",
    "# Load images\n",
    "for filename in os.listdir(data_dir):\n",
    "    img_path = os.path.join(data_dir, filename)\n",
    "    if img_path.endswith('.png') or img_path.endswith('.jpg'):\n",
    "        try:\n",
    "            img = load_img(img_path, target_size=(64, 64))  # Adjust size as needed\n",
    "            img_array = img_to_array(img)\n",
    "            X_train.append(img_array)\n",
    "        except Exception as e:\n",
    "            print(f\"Error loading image {img_path}: {e}\")\n",
    "\n",
    "# Convert list to numpy array\n",
    "X_train = np.array(X_train)\n",
    "print(\"Loaded images shape:\", X_train.shape)  # Should show the number of loaded images\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "3552862810e84c6a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:52:45.750558Z",
     "start_time": "2024-09-30T02:52:45.743459Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files in directory: ['val', 'train', 'test']\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "print(\"Files in directory:\", os.listdir(data_dir))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "60d2ee9a3a392ce7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:55:10.321367Z",
     "start_time": "2024-09-30T02:55:10.317981Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files in directory: ['val', 'train', 'test']\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "files = os.listdir(data_dir)\n",
    "print(\"Files in directory:\", files)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e46b9d5b",
   "metadata": {},
   "source": [
    "# Checking the directory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "7946e3a400126a83",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:55:50.272935Z",
     "start_time": "2024-09-30T02:55:50.268533Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Directory exists.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "if os.path.exists(data_dir):\n",
    "    print(\"Directory exists.\")\n",
    "else:\n",
    "    print(\"Directory does not exist.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "add6bfbe843ae6f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T02:59:35.083360Z",
     "start_time": "2024-09-30T02:59:34.382215Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded images shape: (896, 64, 64, 3)\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import cv2\n",
    "import numpy as np\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "target_size = (64, 64)  \n",
    "images = []\n",
    "for subdir, _, files in os.walk(data_dir):\n",
    "    for file in files:\n",
    "        if file.endswith(('.jpg', '.jpeg')): \n",
    "            file_path = os.path.join(subdir, file)\n",
    "            img = cv2.imread(file_path)  \n",
    "            if img is not None:\n",
    "                img_resized = cv2.resize(img, target_size)  \n",
    "                images.append(img_resized)\n",
    "X_train = np.array(images)\n",
    "\n",
    "print(\"Loaded images shape:\", X_train.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "9df99fb1ef62d810",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:00:52.877854Z",
     "start_time": "2024-09-30T03:00:46.513936Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set shape: (537, 64, 64, 3)\n",
      "Validation set shape: (179, 64, 64, 3)\n",
      "Test set shape: (180, 64, 64, 3)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_splitX_train, X_temp = train_test_split(X_train, test_size=0.4, random_state=42)  # 60% train, 40% temp\n",
    "X_val, X_test = train_test_split(X_temp, test_size=0.5, random_state=42)  # 20% val, 20% test\n",
    "\n",
    "print(f'Training set shape: {X_train.shape}')\n",
    "print(f'Validation set shape: {X_val.shape}')\n",
    "print(f'Test set shape: {X_test.shape}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "b2f77886540c1b16",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:01:20.439781Z",
     "start_time": "2024-09-30T03:01:13.279224Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n",
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/activations/leaky_relu.py:41: UserWarning: Argument `alpha` is deprecated. Use `negative_slope` instead.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Reshape, Flatten, LeakyReLU\n",
    "\n",
    "def build_generator(latent_dim):\n",
    "    model = Sequential()\n",
    "    model.add(Dense(256, input_dim=latent_dim))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(512))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(1024))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(64 * 64 * 3, activation='tanh'))  \n",
    "    model.add(Reshape((64, 64, 3)))\n",
    "    return model\n",
    "\n",
    "latent_dim = 100\n",
    "generator = build_generator(latent_dim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "9b80b7b91d3560c4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:01:31.124022Z",
     "start_time": "2024-09-30T03:01:29.415338Z"
    }
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Reshape, Flatten, LeakyReLU\n",
    "\n",
    "def build_generator(latent_dim):\n",
    "    model = Sequential()\n",
    "    model.add(Dense(256, input_dim=latent_dim))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(512))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(1024))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dense(64 * 64 * 3, activation='tanh'))  # Assuming images are 64x64x3\n",
    "    model.add(Reshape((64, 64, 3)))\n",
    "    return model\n",
    "\n",
    "latent_dim = 100\n",
    "generator = build_generator(latent_dim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "28ed004374ba4f94",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:01:42.692861Z",
     "start_time": "2024-09-30T03:01:41.922728Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/iamsanko/anaconda3/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "from keras.layers import Conv2D, Dropout\n",
    "\n",
    "def build_discriminator():\n",
    "    model = Sequential()\n",
    "    model.add(Conv2D(64, kernel_size=3, strides=2, input_shape=(64, 64, 3), padding='same'))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dropout(0.25))\n",
    "    model.add(Conv2D(128, kernel_size=3, strides=2, padding='same'))\n",
    "    model.add(LeakyReLU(alpha=0.2))\n",
    "    model.add(Dropout(0.25))\n",
    "    model.add(Flatten())\n",
    "    model.add(Dense(1, activation='sigmoid'))\n",
    "    return model\n",
    "\n",
    "discriminator = build_discriminator()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f61bf1a18f96b0a9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:45:36.160571Z",
     "start_time": "2024-09-30T03:45:32.872068Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from torchvision import datasets, transforms\n",
    "\n",
    "\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((128, 128)),  # Resize images to 128x128\n",
    "    transforms.ToTensor()\n",
    "])\n",
    "\n",
    "# Load your dataset (Assuming you're using image data from a folder)\n",
    "dataset = datasets.ImageFolder(root='/home/iamsanko/Downloads/Test 1/output', transform=transform)\n",
    "\n",
    "# Split the dataset\n",
    "X_train, X_temp = train_test_split(dataset, test_size=0.4, random_state=42)  # 60% train, 40% temp\n",
    "X_val, X_test = train_test_split(X_temp, test_size=0.5, random_state=42)  # 20% val, 20% test\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ee935762c1d332cf",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T03:51:56.087223Z",
     "start_time": "2024-09-30T03:51:55.685803Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch \n",
    "from sklearn.model_selection import train_test_split\n",
    "from torchvision import datasets, transforms\n",
    "\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((128, 128)),  \n",
    "    transforms.ToTensor() \n",
    "])\n",
    "dataset = datasets.ImageFolder(root='/home/iamsanko/Downloads/Test 1/output', transform=transform)\n",
    "train_size = int(0.6 * len(dataset))\n",
    "val_test_size = len(dataset) - train_size\n",
    "val_size = int(0.5 * val_test_size)\n",
    "test_size = val_test_size - val_size\n",
    "X_train, X_val, X_test = torch.utils.data.random_split(dataset, [train_size, val_size, test_size])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d0a32cafa1904dc4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:38:07.394212Z",
     "start_time": "2024-09-30T04:35:07.081708Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/200], d_loss: 1.2584, g_loss: 0.5565\n",
      "Epoch [2/200], d_loss: 1.5970, g_loss: 0.4319\n",
      "Epoch [3/200], d_loss: 0.9119, g_loss: 1.9698\n",
      "Epoch [4/200], d_loss: 1.5119, g_loss: 1.3555\n",
      "Epoch [5/200], d_loss: 1.2815, g_loss: 1.7102\n",
      "Epoch [6/200], d_loss: 0.7097, g_loss: 1.7032\n",
      "Epoch [7/200], d_loss: 0.6447, g_loss: 1.2189\n",
      "Epoch [8/200], d_loss: 0.9148, g_loss: 0.6256\n",
      "Epoch [9/200], d_loss: 1.3164, g_loss: 0.3703\n",
      "Epoch [10/200], d_loss: 1.6234, g_loss: 0.3002\n",
      "Epoch [11/200], d_loss: 0.9819, g_loss: 0.7527\n",
      "Epoch [12/200], d_loss: 0.4128, g_loss: 2.1738\n",
      "Epoch [13/200], d_loss: 0.4211, g_loss: 1.5668\n",
      "Epoch [14/200], d_loss: 0.9881, g_loss: 0.6622\n",
      "Epoch [15/200], d_loss: 1.7048, g_loss: 0.3602\n",
      "Epoch [16/200], d_loss: 0.9059, g_loss: 1.1275\n",
      "Epoch [17/200], d_loss: 0.4375, g_loss: 2.2173\n",
      "Epoch [18/200], d_loss: 0.6386, g_loss: 1.2223\n",
      "Epoch [19/200], d_loss: 0.8271, g_loss: 0.7742\n",
      "Epoch [20/200], d_loss: 1.1085, g_loss: 0.5793\n",
      "Epoch [21/200], d_loss: 0.9667, g_loss: 0.6743\n",
      "Epoch [22/200], d_loss: 0.6981, g_loss: 1.0408\n",
      "Epoch [23/200], d_loss: 0.5985, g_loss: 1.1569\n",
      "Epoch [24/200], d_loss: 0.7476, g_loss: 0.9193\n",
      "Epoch [25/200], d_loss: 1.1548, g_loss: 0.5992\n",
      "Epoch [26/200], d_loss: 1.0351, g_loss: 0.8928\n",
      "Epoch [27/200], d_loss: 0.6340, g_loss: 1.5501\n",
      "Epoch [28/200], d_loss: 1.7183, g_loss: 0.6890\n",
      "Epoch [29/200], d_loss: 0.5672, g_loss: 1.9863\n",
      "Epoch [30/200], d_loss: 0.4614, g_loss: 1.8965\n",
      "Epoch [31/200], d_loss: 1.5933, g_loss: 0.6251\n",
      "Epoch [32/200], d_loss: 1.2277, g_loss: 0.9463\n",
      "Epoch [33/200], d_loss: 0.7299, g_loss: 1.7758\n",
      "Epoch [34/200], d_loss: 2.2137, g_loss: 1.6112\n",
      "Epoch [35/200], d_loss: 3.4230, g_loss: 2.7872\n",
      "Epoch [36/200], d_loss: 0.2536, g_loss: 3.0122\n",
      "Epoch [37/200], d_loss: 0.2336, g_loss: 3.3827\n",
      "Epoch [38/200], d_loss: 0.3406, g_loss: 3.0900\n",
      "Epoch [39/200], d_loss: 0.9292, g_loss: 2.0511\n",
      "Epoch [40/200], d_loss: 0.4264, g_loss: 2.9867\n",
      "Epoch [41/200], d_loss: 0.8929, g_loss: 2.7671\n",
      "Epoch [42/200], d_loss: 3.7552, g_loss: 1.4248\n",
      "Epoch [43/200], d_loss: 3.2196, g_loss: 2.2502\n",
      "Epoch [44/200], d_loss: 0.3120, g_loss: 3.1943\n",
      "Epoch [45/200], d_loss: 0.3425, g_loss: 2.4407\n",
      "Epoch [46/200], d_loss: 0.4929, g_loss: 2.0127\n",
      "Epoch [47/200], d_loss: 0.5315, g_loss: 1.5058\n",
      "Epoch [48/200], d_loss: 0.8248, g_loss: 0.9882\n",
      "Epoch [49/200], d_loss: 1.5476, g_loss: 0.5727\n",
      "Epoch [50/200], d_loss: 1.8039, g_loss: 0.6790\n",
      "Epoch [51/200], d_loss: 1.1835, g_loss: 1.3078\n",
      "Epoch [52/200], d_loss: 0.6249, g_loss: 1.6355\n",
      "Epoch [53/200], d_loss: 0.5748, g_loss: 1.7416\n",
      "Epoch [54/200], d_loss: 0.9227, g_loss: 1.7506\n",
      "Epoch [55/200], d_loss: 1.0481, g_loss: 1.4312\n",
      "Epoch [56/200], d_loss: 1.3212, g_loss: 1.3489\n",
      "Epoch [57/200], d_loss: 1.1663, g_loss: 1.5630\n",
      "Epoch [58/200], d_loss: 1.7709, g_loss: 1.7075\n",
      "Epoch [59/200], d_loss: 0.6104, g_loss: 2.1796\n",
      "Epoch [60/200], d_loss: 0.8211, g_loss: 2.7088\n",
      "Epoch [61/200], d_loss: 1.4980, g_loss: 2.9642\n",
      "Epoch [62/200], d_loss: 3.3516, g_loss: 4.2987\n",
      "Epoch [63/200], d_loss: 3.8911, g_loss: 4.6231\n",
      "Epoch [64/200], d_loss: 1.5159, g_loss: 4.5597\n",
      "Epoch [65/200], d_loss: 0.4671, g_loss: 5.4029\n",
      "Epoch [66/200], d_loss: 0.1481, g_loss: 4.1405\n",
      "Epoch [67/200], d_loss: 0.3119, g_loss: 3.8672\n",
      "Epoch [68/200], d_loss: 0.5082, g_loss: 3.7899\n",
      "Epoch [69/200], d_loss: 0.7270, g_loss: 4.0878\n",
      "Epoch [70/200], d_loss: 0.2659, g_loss: 4.4464\n",
      "Epoch [71/200], d_loss: 0.2573, g_loss: 4.1718\n",
      "Epoch [72/200], d_loss: 0.3667, g_loss: 3.9858\n",
      "Epoch [73/200], d_loss: 0.7885, g_loss: 3.6541\n",
      "Epoch [74/200], d_loss: 1.2641, g_loss: 2.5802\n",
      "Epoch [75/200], d_loss: 0.8565, g_loss: 2.8697\n",
      "Epoch [76/200], d_loss: 0.2984, g_loss: 3.5959\n",
      "Epoch [77/200], d_loss: 0.2585, g_loss: 3.8008\n",
      "Epoch [78/200], d_loss: 0.1990, g_loss: 4.0546\n",
      "Epoch [79/200], d_loss: 0.2584, g_loss: 3.9474\n",
      "Epoch [80/200], d_loss: 0.3072, g_loss: 3.8304\n",
      "Epoch [81/200], d_loss: 0.3954, g_loss: 3.9367\n",
      "Epoch [82/200], d_loss: 1.6195, g_loss: 4.2388\n",
      "Epoch [83/200], d_loss: 0.3248, g_loss: 4.7726\n",
      "Epoch [84/200], d_loss: 0.2010, g_loss: 4.1188\n",
      "Epoch [85/200], d_loss: 0.4348, g_loss: 5.1402\n",
      "Epoch [86/200], d_loss: 1.0243, g_loss: 4.7342\n",
      "Epoch [87/200], d_loss: 0.7555, g_loss: 3.7614\n",
      "Epoch [88/200], d_loss: 0.9317, g_loss: 3.4994\n",
      "Epoch [89/200], d_loss: 1.3864, g_loss: 2.3622\n",
      "Epoch [90/200], d_loss: 1.2886, g_loss: 2.5612\n",
      "Epoch [91/200], d_loss: 1.7761, g_loss: 2.2456\n",
      "Epoch [92/200], d_loss: 0.5602, g_loss: 3.1664\n",
      "Epoch [93/200], d_loss: 0.8451, g_loss: 5.0544\n",
      "Epoch [94/200], d_loss: 3.9859, g_loss: 3.1723\n",
      "Epoch [95/200], d_loss: 4.4836, g_loss: 3.1020\n",
      "Epoch [96/200], d_loss: 0.6714, g_loss: 2.8501\n",
      "Epoch [97/200], d_loss: 0.0916, g_loss: 4.5309\n",
      "Epoch [98/200], d_loss: 0.1278, g_loss: 4.0209\n",
      "Epoch [99/200], d_loss: 0.2066, g_loss: 3.8886\n",
      "Epoch [100/200], d_loss: 0.3236, g_loss: 3.5853\n",
      "Epoch [101/200], d_loss: 0.4168, g_loss: 3.3820\n",
      "Epoch [102/200], d_loss: 0.5719, g_loss: 2.9685\n",
      "Epoch [103/200], d_loss: 0.5514, g_loss: 3.0567\n",
      "Epoch [104/200], d_loss: 0.3721, g_loss: 3.6385\n",
      "Epoch [105/200], d_loss: 0.4543, g_loss: 3.3028\n",
      "Epoch [106/200], d_loss: 0.7074, g_loss: 3.3651\n",
      "Epoch [107/200], d_loss: 1.7240, g_loss: 2.5954\n",
      "Epoch [108/200], d_loss: 2.0583, g_loss: 2.3941\n",
      "Epoch [109/200], d_loss: 0.1730, g_loss: 3.3920\n",
      "Epoch [110/200], d_loss: 0.1189, g_loss: 4.5255\n",
      "Epoch [111/200], d_loss: 0.1477, g_loss: 3.6610\n",
      "Epoch [112/200], d_loss: 0.3898, g_loss: 3.3112\n",
      "Epoch [113/200], d_loss: 1.0517, g_loss: 3.0300\n",
      "Epoch [114/200], d_loss: 2.2513, g_loss: 2.6525\n",
      "Epoch [115/200], d_loss: 0.4936, g_loss: 3.0145\n",
      "Epoch [116/200], d_loss: 0.7870, g_loss: 3.6439\n",
      "Epoch [117/200], d_loss: 0.7117, g_loss: 3.0318\n",
      "Epoch [118/200], d_loss: 0.4705, g_loss: 4.5210\n",
      "Epoch [119/200], d_loss: 0.7653, g_loss: 3.1878\n",
      "Epoch [120/200], d_loss: 1.0824, g_loss: 2.6769\n",
      "Epoch [121/200], d_loss: 1.3331, g_loss: 1.7340\n",
      "Epoch [122/200], d_loss: 1.6808, g_loss: 1.1455\n",
      "Epoch [123/200], d_loss: 0.7568, g_loss: 1.7693\n",
      "Epoch [124/200], d_loss: 0.4801, g_loss: 1.9042\n",
      "Epoch [125/200], d_loss: 1.0620, g_loss: 1.4091\n",
      "Epoch [126/200], d_loss: 2.2352, g_loss: 0.8691\n",
      "Epoch [127/200], d_loss: 3.1052, g_loss: 0.7024\n",
      "Epoch [128/200], d_loss: 0.7401, g_loss: 1.6332\n",
      "Epoch [129/200], d_loss: 0.3625, g_loss: 2.4124\n",
      "Epoch [130/200], d_loss: 1.1769, g_loss: 1.0909\n",
      "Epoch [131/200], d_loss: 1.7059, g_loss: 0.6920\n",
      "Epoch [132/200], d_loss: 0.8209, g_loss: 1.2992\n",
      "Epoch [133/200], d_loss: 0.9193, g_loss: 0.9863\n",
      "Epoch [134/200], d_loss: 1.3166, g_loss: 0.6733\n",
      "Epoch [135/200], d_loss: 1.0690, g_loss: 1.2045\n",
      "Epoch [136/200], d_loss: 0.5990, g_loss: 1.5336\n",
      "Epoch [137/200], d_loss: 0.8933, g_loss: 1.3640\n",
      "Epoch [138/200], d_loss: 1.0667, g_loss: 1.3011\n",
      "Epoch [139/200], d_loss: 1.1794, g_loss: 1.2388\n",
      "Epoch [140/200], d_loss: 1.2107, g_loss: 1.0625\n",
      "Epoch [141/200], d_loss: 1.1468, g_loss: 1.2370\n",
      "Epoch [142/200], d_loss: 0.5812, g_loss: 1.6781\n",
      "Epoch [143/200], d_loss: 0.5881, g_loss: 1.5880\n",
      "Epoch [144/200], d_loss: 0.6180, g_loss: 1.6604\n",
      "Epoch [145/200], d_loss: 1.0391, g_loss: 1.1154\n",
      "Epoch [146/200], d_loss: 1.5234, g_loss: 0.6862\n",
      "Epoch [147/200], d_loss: 1.3511, g_loss: 0.7910\n",
      "Epoch [148/200], d_loss: 0.9034, g_loss: 1.1973\n",
      "Epoch [149/200], d_loss: 0.5798, g_loss: 1.3297\n",
      "Epoch [150/200], d_loss: 0.6072, g_loss: 1.3049\n",
      "Epoch [151/200], d_loss: 1.3470, g_loss: 0.7686\n",
      "Epoch [152/200], d_loss: 1.6165, g_loss: 0.5580\n",
      "Epoch [153/200], d_loss: 1.1379, g_loss: 0.8810\n",
      "Epoch [154/200], d_loss: 0.6161, g_loss: 1.2557\n",
      "Epoch [155/200], d_loss: 0.5007, g_loss: 1.4291\n",
      "Epoch [156/200], d_loss: 0.4339, g_loss: 1.5946\n",
      "Epoch [157/200], d_loss: 0.5965, g_loss: 1.3066\n",
      "Epoch [158/200], d_loss: 0.9053, g_loss: 1.0147\n",
      "Epoch [159/200], d_loss: 1.0907, g_loss: 0.6975\n",
      "Epoch [160/200], d_loss: 1.2013, g_loss: 0.7913\n",
      "Epoch [161/200], d_loss: 0.9142, g_loss: 0.9658\n",
      "Epoch [162/200], d_loss: 0.6750, g_loss: 0.9706\n",
      "Epoch [163/200], d_loss: 0.5929, g_loss: 1.2173\n",
      "Epoch [164/200], d_loss: 0.5671, g_loss: 1.4587\n",
      "Epoch [165/200], d_loss: 0.7040, g_loss: 1.3820\n",
      "Epoch [166/200], d_loss: 0.8788, g_loss: 1.1091\n",
      "Epoch [167/200], d_loss: 1.6800, g_loss: 0.8085\n",
      "Epoch [168/200], d_loss: 1.5651, g_loss: 0.6887\n",
      "Epoch [169/200], d_loss: 1.2333, g_loss: 0.8902\n",
      "Epoch [170/200], d_loss: 0.9324, g_loss: 1.0781\n",
      "Epoch [171/200], d_loss: 0.9606, g_loss: 0.7290\n",
      "Epoch [172/200], d_loss: 0.9867, g_loss: 0.7744\n",
      "Epoch [173/200], d_loss: 0.6794, g_loss: 0.9374\n",
      "Epoch [174/200], d_loss: 0.8886, g_loss: 1.2911\n",
      "Epoch [175/200], d_loss: 1.4218, g_loss: 1.0261\n",
      "Epoch [176/200], d_loss: 3.2138, g_loss: 0.9725\n",
      "Epoch [177/200], d_loss: 1.0212, g_loss: 1.1128\n",
      "Epoch [178/200], d_loss: 0.7512, g_loss: 1.6548\n",
      "Epoch [179/200], d_loss: 1.4678, g_loss: 0.9197\n",
      "Epoch [180/200], d_loss: 1.5433, g_loss: 0.6757\n",
      "Epoch [181/200], d_loss: 1.4959, g_loss: 0.5893\n",
      "Epoch [182/200], d_loss: 1.0969, g_loss: 0.7837\n",
      "Epoch [183/200], d_loss: 0.8685, g_loss: 1.0002\n",
      "Epoch [184/200], d_loss: 0.7138, g_loss: 1.1697\n",
      "Epoch [185/200], d_loss: 0.6522, g_loss: 1.2380\n",
      "Epoch [186/200], d_loss: 0.8576, g_loss: 1.1202\n",
      "Epoch [187/200], d_loss: 0.8872, g_loss: 0.9503\n",
      "Epoch [188/200], d_loss: 1.1454, g_loss: 0.7857\n",
      "Epoch [189/200], d_loss: 1.2331, g_loss: 0.8647\n",
      "Epoch [190/200], d_loss: 0.9799, g_loss: 1.0509\n",
      "Epoch [191/200], d_loss: 0.9664, g_loss: 1.0744\n",
      "Epoch [192/200], d_loss: 0.8974, g_loss: 0.9522\n",
      "Epoch [193/200], d_loss: 0.8030, g_loss: 1.2124\n",
      "Epoch [194/200], d_loss: 0.3765, g_loss: 1.9123\n",
      "Epoch [195/200], d_loss: 0.3755, g_loss: 1.7548\n",
      "Epoch [196/200], d_loss: 0.7389, g_loss: 1.4132\n",
      "Epoch [197/200], d_loss: 1.0951, g_loss: 1.2648\n",
      "Epoch [198/200], d_loss: 1.0221, g_loss: 1.1150\n",
      "Epoch [199/200], d_loss: 1.0343, g_loss: 1.0428\n",
      "Epoch [200/200], d_loss: 1.5780, g_loss: 0.8892\n",
      "Training complete.\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision import datasets, transforms\n",
    "class Generator(nn.Module):\n",
    "    def __init__(self, noise_dim):\n",
    "        super(Generator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(noise_dim, 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(256, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, 1024),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(1024, 28 * 28),\n",
    "            nn.Tanh()\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.model(x).view(-1, 1, 28, 28)  # Output as 1 channel 28x28 image\n",
    "\n",
    "# Define Discriminator\n",
    "class Discriminator(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Discriminator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(28 * 28, 512),  # Input size = 28*28 (grayscale image)\n",
    "            nn.LeakyReLU(0.2),\n",
    "            nn.Linear(512, 256),\n",
    "            nn.LeakyReLU(0.2),\n",
    "            nn.Linear(256, 1),\n",
    "            nn.Sigmoid()\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = x.view(x.size(0), -1)  # [batch_size, 784]\n",
    "        return self.model(x)\n",
    "\n",
    "# Hyperparameters\n",
    "batch_size = 64\n",
    "noise_dim = 100\n",
    "lr = 0.0002\n",
    "num_epochs = 200\n",
    "\n",
    "# Data preparation\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize(28),\n",
    "    transforms.Grayscale(num_output_channels=1),  \n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.5], [0.5]) \n",
    "])\n",
    "\n",
    "dataset = datasets.ImageFolder(root='/home/iamsanko/Downloads/Test 1/output', transform=transform)\n",
    "train_size = int(0.6 * len(dataset))\n",
    "val_size = int(0.2 * len(dataset))\n",
    "test_size = len(dataset) - train_size - val_size\n",
    "\n",
    "train_dataset, val_dataset, test_dataset = torch.utils.data.random_split(dataset, [train_size, val_size, test_size])\n",
    "\n",
    "train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)\n",
    "test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)\n",
    "generator = Generator(noise_dim=noise_dim)\n",
    "discriminator = Discriminator()\n",
    "criterion = nn.BCELoss()\n",
    "optimizer_g = torch.optim.Adam(generator.parameters(), lr=lr)\n",
    "optimizer_d = torch.optim.Adam(discriminator.parameters(), lr=lr)\n",
    "def generate_noise(batch_size, noise_dim):\n",
    "    return torch.randn(batch_size, noise_dim)\n",
    "for epoch in range(num_epochs):\n",
    "    for real_images, _ in train_loader:\n",
    "        batch_size = real_images.size(0)\n",
    "        real_labels = torch.ones(batch_size, 1)\n",
    "        fake_labels = torch.zeros(batch_size, 1)\n",
    "        optimizer_d.zero_grad()\n",
    "        outputs = discriminator(real_images).view(batch_size, -1) \n",
    "        d_loss_real = criterion(outputs, real_labels)\n",
    "        noise = generate_noise(batch_size, noise_dim)\n",
    "        fake_images = generator(noise)\n",
    "        outputs = discriminator(fake_images.detach()).view(batch_size, -1)  \n",
    "        d_loss_fake = criterion(outputs, fake_labels)\n",
    "        d_loss = d_loss_real + d_loss_fake\n",
    "        d_loss.backward()\n",
    "        optimizer_d.step()\n",
    "        optimizer_g.zero_grad()\n",
    "        noise = generate_noise(batch_size, noise_dim)\n",
    "        fake_images = generator(noise)\n",
    "        outputs = discriminator(fake_images).view(batch_size, -1) \n",
    "        g_loss = criterion(outputs, real_labels)\n",
    "        g_loss.backward()\n",
    "        optimizer_g.step()\n",
    "    print(f'Epoch [{epoch+1}/{num_epochs}], d_loss: {d_loss.item():.4f}, g_loss: {g_loss.item():.4f}')\n",
    "print(\"Training complete.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f043a74db820e532",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:45:34.249101Z",
     "start_time": "2024-09-30T04:45:34.146141Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x100 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def show_generated_images(generator, num_images, noise_dim):\n",
    "    noise = generate_noise(num_images, noise_dim)\n",
    "    fake_images = generator(noise)\n",
    "    fake_images = fake_images * 0.5 + 0.5  # Denormalize to [0, 1]\n",
    "\n",
    "    fig, axes = plt.subplots(1, num_images, figsize=(num_images, 1))\n",
    "    for i in range(num_images):\n",
    "        axes[i].imshow(fake_images[i].squeeze(0).cpu().detach().numpy(), cmap='gray')\n",
    "        axes[i].axis('off')\n",
    "    plt.show()\n",
    "show_generated_images(generator, num_images=5, noise_dim=noise_dim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "28d25a2847d655c9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:46:11.826325Z",
     "start_time": "2024-09-30T04:46:11.673580Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "def generate_and_show_images(generator, noise_dim, num_images=5):\n",
    "    generator.eval() \n",
    "    noise = generate_noise(num_images, noise_dim).to(device) \n",
    "    with torch.no_grad():\n",
    "        fake_images = generator(noise).cpu() \n",
    "    fake_images = (fake_images * 0.5 + 0.5).clamp(0, 1) \n",
    "    fig, axes = plt.subplots(1, num_images, figsize=(15, 5))\n",
    "    for i in range(num_images):\n",
    "        img = fake_images[i].permute(1, 2, 0)  \n",
    "        axes[i].imshow(img)\n",
    "        axes[i].axis('off')\n",
    "    plt.show()\n",
    "generate_and_show_images(generator, noise_dim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "ab9edd93c7e42693",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:47:15.094425Z",
     "start_time": "2024-09-30T04:47:15.069474Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "def generate_and_save_images(generator, noise_dim, num_images, save_dir):\n",
    "    if not os.path.exists(save_dir):\n",
    "        os.makedirs(save_dir)\n",
    "    generator.eval()\n",
    "    noise = torch.randn(num_images, noise_dim).to(device)\n",
    "    with torch.no_grad():\n",
    "        fake_images = generator(noise).cpu()\n",
    "    for i in range(num_images):\n",
    "        img = fake_images[i].squeeze().numpy()\n",
    "        plt.imsave(os.path.join(save_dir, f\"generated_image_{i+1}.png\"), img, cmap='gray')\n",
    "num_images = 10\n",
    "noise_dim = 100\n",
    "save_dir = \"/home/iamsanko/Downloads/generated_images\"\n",
    "generate_and_save_images(generator, noise_dim, num_images, save_dir)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "3eddf7014a9816f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:56:08.562803Z",
     "start_time": "2024-09-30T04:56:08.392382Z"
    }
   },
   "outputs": [],
   "source": [
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "# Set paths\n",
    "train_dir = '/home/iamsanko/Downloads/Test 1/output/train'\n",
    "val_dir = '/home/iamsanko/Downloads/Test 1/output/val'\n",
    "test_dir = '/home/iamsanko/Downloads/Test 1/output/test'\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),  # Resize to 64x64 (standard GAN input)\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.5], [0.5])  # Normalize between [-1, 1] for Tanh output\n",
    "])\n",
    "train_dataset = datasets.ImageFolder(train_dir, transform=transform)\n",
    "val_dataset = datasets.ImageFolder(val_dir, transform=transform)\n",
    "test_dataset = datasets.ImageFolder(test_dir, transform=transform)\n",
    "train_loader = DataLoader(train_dataset, batch_size=64, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=64, shuffle=False)\n",
    "test_loader = DataLoader(test_dataset, batch_size=64, shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "7583553818697409",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:56:18.324313Z",
     "start_time": "2024-09-30T04:56:18.320142Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "\n",
    "class Generator(nn.Module):\n",
    "    def __init__(self, noise_dim):\n",
    "        super(Generator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(noise_dim, 256),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(256, 512),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(512, 1024),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(1024, 64*64*1),  \n",
    "            nn.Tanh()  \n",
    "        )\n",
    "    def forward(self, z):\n",
    "        img = self.model(z)\n",
    "        img = img.view(img.size(0), 1, 64, 64)  # Reshape to image dimensions\n",
    "        return img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1b0e5084ae3f9aca",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:57:36.325468Z",
     "start_time": "2024-09-30T04:57:36.320568Z"
    }
   },
   "outputs": [],
   "source": [
    "class Discriminator(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Discriminator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(64*64*1, 1024),\n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(1024, 512),\n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(512, 256),\n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(256, 1),\n",
    "            nn.Sigmoid() \n",
    "        )\n",
    "\n",
    "    def forward(self, img):\n",
    "        img_flat = img.view(img.size(0), -1)\n",
    "        validity = self.model(img_flat)\n",
    "        return validity\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "db2b4006f84f31f9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:59:41.506078Z",
     "start_time": "2024-09-30T04:59:41.501258Z"
    }
   },
   "outputs": [],
   "source": [
    "class Generator(nn.Module):\n",
    "    def __init__(self, noise_dim):\n",
    "        super(Generator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(noise_dim, 256),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(256, 512),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(512, 1024),\n",
    "            nn.ReLU(True),\n",
    "            nn.Linear(1024, 64*64*3),  # Change this to 64*64*3 for RGB\n",
    "            nn.Tanh()  # Output range [-1, 1]\n",
    "        )\n",
    "    def forward(self, z):\n",
    "        img = self.model(z)\n",
    "        img = img.view(img.size(0), 3, 64, 64)  # Change to 3 channels\n",
    "        return img\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "1a083dac3225848a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T04:59:48.283571Z",
     "start_time": "2024-09-30T04:59:48.279306Z"
    }
   },
   "outputs": [],
   "source": [
    "class Discriminator(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Discriminator, self).__init__()\n",
    "        self.model = nn.Sequential(\n",
    "            nn.Linear(64*64*3, 1024), \n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(1024, 512),\n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(512, 256),\n",
    "            nn.LeakyReLU(0.2, inplace=True),\n",
    "            nn.Linear(256, 1),\n",
    "            nn.Sigmoid()  \n",
    "        )\n",
    "    def forward(self, img):\n",
    "        img_flat = img.view(img.size(0), -1)  \n",
    "        validity = self.model(img_flat)\n",
    "        return validity\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "518f9ef4f65d4b13",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T05:07:36.434794Z",
     "start_time": "2024-09-30T04:59:59.480669Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/100] | D Loss: 0.9842427968978882 | G Loss: 0.7698355317115784\n",
      "Epoch [2/100] | D Loss: 1.7141879796981812 | G Loss: 1.0247533321380615\n",
      "Epoch [3/100] | D Loss: 0.17828479409217834 | G Loss: 2.9229345321655273\n",
      "Epoch [4/100] | D Loss: 0.6865127682685852 | G Loss: 0.9690119624137878\n",
      "Epoch [5/100] | D Loss: 0.47508054971694946 | G Loss: 1.0982474088668823\n",
      "Epoch [6/100] | D Loss: 0.2184382826089859 | G Loss: 1.8230900764465332\n",
      "Epoch [7/100] | D Loss: 0.4037191569805145 | G Loss: 1.765008568763733\n",
      "Epoch [8/100] | D Loss: 0.10448507964611053 | G Loss: 3.2011473178863525\n",
      "Epoch [9/100] | D Loss: 0.03805321827530861 | G Loss: 5.258705139160156\n",
      "Epoch [10/100] | D Loss: 0.03365657106041908 | G Loss: 6.265377044677734\n",
      "Epoch [11/100] | D Loss: 0.09067761898040771 | G Loss: 6.335056781768799\n",
      "Epoch [12/100] | D Loss: 0.07083749026060104 | G Loss: 8.40924072265625\n",
      "Epoch [13/100] | D Loss: 0.03873594105243683 | G Loss: 9.325901985168457\n",
      "Epoch [14/100] | D Loss: 0.11320313811302185 | G Loss: 10.841694831848145\n",
      "Epoch [15/100] | D Loss: 0.08297958225011826 | G Loss: 9.437055587768555\n",
      "Epoch [16/100] | D Loss: 0.49983668327331543 | G Loss: 3.632936716079712\n",
      "Epoch [17/100] | D Loss: 2.1649768352508545 | G Loss: 0.25169840455055237\n",
      "Epoch [18/100] | D Loss: 0.30186861753463745 | G Loss: 1.6397660970687866\n",
      "Epoch [19/100] | D Loss: 0.5401976704597473 | G Loss: 1.2034833431243896\n",
      "Epoch [20/100] | D Loss: 0.8552388548851013 | G Loss: 1.270056128501892\n",
      "Epoch [21/100] | D Loss: 0.3622024953365326 | G Loss: 2.522953987121582\n",
      "Epoch [22/100] | D Loss: 0.12098810821771622 | G Loss: 3.4084205627441406\n",
      "Epoch [23/100] | D Loss: 0.6483169794082642 | G Loss: 1.6977639198303223\n",
      "Epoch [24/100] | D Loss: 0.4470050632953644 | G Loss: 3.3078033924102783\n",
      "Epoch [25/100] | D Loss: 1.2112650871276855 | G Loss: 4.083232402801514\n",
      "Epoch [26/100] | D Loss: 0.2284863293170929 | G Loss: 11.274840354919434\n",
      "Epoch [27/100] | D Loss: 0.11744863539934158 | G Loss: 11.515589714050293\n",
      "Epoch [28/100] | D Loss: 0.19990159571170807 | G Loss: 7.115450382232666\n",
      "Epoch [29/100] | D Loss: 0.03770437091588974 | G Loss: 7.170186519622803\n",
      "Epoch [30/100] | D Loss: 0.07091613858938217 | G Loss: 7.672579288482666\n",
      "Epoch [31/100] | D Loss: 0.02089580148458481 | G Loss: 5.586132526397705\n",
      "Epoch [32/100] | D Loss: 0.23096729815006256 | G Loss: 3.6336872577667236\n",
      "Epoch [33/100] | D Loss: 0.16974076628684998 | G Loss: 3.6414356231689453\n",
      "Epoch [34/100] | D Loss: 0.19182609021663666 | G Loss: 4.36651086807251\n",
      "Epoch [35/100] | D Loss: 0.04953290522098541 | G Loss: 3.874737501144409\n",
      "Epoch [36/100] | D Loss: 0.09526391327381134 | G Loss: 4.078271865844727\n",
      "Epoch [37/100] | D Loss: 0.38411080837249756 | G Loss: 2.5900967121124268\n",
      "Epoch [38/100] | D Loss: 0.6724213361740112 | G Loss: 1.1485809087753296\n",
      "Epoch [39/100] | D Loss: 0.9530499577522278 | G Loss: 2.084998369216919\n",
      "Epoch [40/100] | D Loss: 0.8338083028793335 | G Loss: 4.217459201812744\n",
      "Epoch [41/100] | D Loss: 0.007836022414267063 | G Loss: 15.801929473876953\n",
      "Epoch [42/100] | D Loss: 0.07857814431190491 | G Loss: 5.810476303100586\n",
      "Epoch [43/100] | D Loss: 0.02747730165719986 | G Loss: 8.315506935119629\n",
      "Epoch [44/100] | D Loss: 0.42336684465408325 | G Loss: 3.466933250427246\n",
      "Epoch [45/100] | D Loss: 1.311592698097229 | G Loss: 1.7151910066604614\n",
      "Epoch [46/100] | D Loss: 0.09188102185726166 | G Loss: 4.140201568603516\n",
      "Epoch [47/100] | D Loss: 0.5864598751068115 | G Loss: 2.782756805419922\n",
      "Epoch [48/100] | D Loss: 0.6772530674934387 | G Loss: 3.1216847896575928\n",
      "Epoch [49/100] | D Loss: 0.09803149849176407 | G Loss: 4.7636332511901855\n",
      "Epoch [50/100] | D Loss: 0.07418456673622131 | G Loss: 5.320276260375977\n",
      "Epoch [51/100] | D Loss: 0.5676308870315552 | G Loss: 5.473681926727295\n",
      "Epoch [52/100] | D Loss: 0.1132475733757019 | G Loss: 11.247570037841797\n",
      "Epoch [53/100] | D Loss: 0.37349507212638855 | G Loss: 4.138914585113525\n",
      "Epoch [54/100] | D Loss: 0.5567697286605835 | G Loss: 2.1781623363494873\n",
      "Epoch [55/100] | D Loss: 0.27797895669937134 | G Loss: 3.021085500717163\n",
      "Epoch [56/100] | D Loss: 0.5388233065605164 | G Loss: 2.8729541301727295\n",
      "Epoch [57/100] | D Loss: 0.41423699259757996 | G Loss: 2.440147638320923\n",
      "Epoch [58/100] | D Loss: 0.18251515924930573 | G Loss: 4.913730144500732\n",
      "Epoch [59/100] | D Loss: 2.1595709323883057 | G Loss: 9.220807075500488\n",
      "Epoch [60/100] | D Loss: 0.007740905042737722 | G Loss: 14.512016296386719\n",
      "Epoch [61/100] | D Loss: 1.7421976327896118 | G Loss: 6.5170979499816895\n",
      "Epoch [62/100] | D Loss: 0.40804535150527954 | G Loss: 5.105948448181152\n",
      "Epoch [63/100] | D Loss: 0.6556383967399597 | G Loss: 4.52007532119751\n",
      "Epoch [64/100] | D Loss: 1.3565893173217773 | G Loss: 5.22204065322876\n",
      "Epoch [65/100] | D Loss: 0.013808505609631538 | G Loss: 9.873432159423828\n",
      "Epoch [66/100] | D Loss: 0.8218103647232056 | G Loss: 12.557013511657715\n",
      "Epoch [67/100] | D Loss: 29.030458450317383 | G Loss: 12.004844665527344\n",
      "Epoch [68/100] | D Loss: 0.6218430399894714 | G Loss: 22.937253952026367\n",
      "Epoch [69/100] | D Loss: 0.060837794095277786 | G Loss: 46.66337585449219\n",
      "Epoch [70/100] | D Loss: 5.135209083557129 | G Loss: 0.05634501203894615\n",
      "Epoch [71/100] | D Loss: 3.9485130310058594 | G Loss: 0.48431113362312317\n",
      "Epoch [72/100] | D Loss: 0.03369690105319023 | G Loss: 5.415870189666748\n",
      "Epoch [73/100] | D Loss: 0.6128984689712524 | G Loss: 4.382452964782715\n",
      "Epoch [74/100] | D Loss: 0.8842580318450928 | G Loss: 1.5505037307739258\n",
      "Epoch [75/100] | D Loss: 0.08696217089891434 | G Loss: 4.132904529571533\n",
      "Epoch [76/100] | D Loss: 0.3830776512622833 | G Loss: 2.604299783706665\n",
      "Epoch [77/100] | D Loss: 0.41991788148880005 | G Loss: 2.4432060718536377\n",
      "Epoch [78/100] | D Loss: 1.7535051107406616 | G Loss: 2.0473270416259766\n",
      "Epoch [79/100] | D Loss: 0.4686053991317749 | G Loss: 2.9461476802825928\n",
      "Epoch [80/100] | D Loss: 0.7730993628501892 | G Loss: 3.910832405090332\n",
      "Epoch [81/100] | D Loss: 2.308781147003174 | G Loss: 0.8572402000427246\n",
      "Epoch [82/100] | D Loss: 0.31576547026634216 | G Loss: 2.6479604244232178\n",
      "Epoch [83/100] | D Loss: 0.8003783226013184 | G Loss: 3.1227729320526123\n",
      "Epoch [84/100] | D Loss: 0.3389246463775635 | G Loss: 2.6616737842559814\n",
      "Epoch [85/100] | D Loss: 0.31600379943847656 | G Loss: 2.0093908309936523\n",
      "Epoch [86/100] | D Loss: 0.5151837468147278 | G Loss: 2.150949716567993\n",
      "Epoch [87/100] | D Loss: 0.8751004934310913 | G Loss: 1.6313329935073853\n",
      "Epoch [88/100] | D Loss: 0.17720571160316467 | G Loss: 2.7374675273895264\n",
      "Epoch [89/100] | D Loss: 0.11576597392559052 | G Loss: 2.9700887203216553\n",
      "Epoch [90/100] | D Loss: 0.1664949655532837 | G Loss: 2.689694404602051\n",
      "Epoch [91/100] | D Loss: 0.1595844179391861 | G Loss: 2.289189100265503\n",
      "Epoch [92/100] | D Loss: 0.402098685503006 | G Loss: 1.7613908052444458\n",
      "Epoch [93/100] | D Loss: 0.7576771974563599 | G Loss: 1.288012981414795\n",
      "Epoch [94/100] | D Loss: 0.5189253687858582 | G Loss: 1.9342390298843384\n",
      "Epoch [95/100] | D Loss: 0.417006254196167 | G Loss: 2.2866697311401367\n",
      "Epoch [96/100] | D Loss: 0.07471394538879395 | G Loss: 3.5495922565460205\n",
      "Epoch [97/100] | D Loss: 0.5797431468963623 | G Loss: 3.0666961669921875\n",
      "Epoch [98/100] | D Loss: 0.7687367796897888 | G Loss: 6.462123394012451\n",
      "Epoch [99/100] | D Loss: 1.6756280660629272 | G Loss: 1.8161325454711914\n",
      "Epoch [100/100] | D Loss: 2.5609726905822754 | G Loss: 0.653850793838501\n"
     ]
    }
   ],
   "source": [
    "import torch.optim as optim\n",
    "import torch\n",
    "noise_dim = 100\n",
    "generator = Generator(noise_dim).to(device)\n",
    "discriminator = Discriminator().to(device)\n",
    "criterion = nn.BCELoss()\n",
    "optimizer_g = optim.Adam(generator.parameters(), lr=0.0002)\n",
    "optimizer_d = optim.Adam(discriminator.parameters(), lr=0.0002)\n",
    "epochs = 100\n",
    "for epoch in range(epochs):\n",
    "    for i, (real_images, _) in enumerate(train_loader):\n",
    "        real_images = real_images.to(device)\n",
    "        batch_size = real_images.size(0)\n",
    "\n",
    "        # Labels\n",
    "        real_labels = torch.ones(batch_size, 1).to(device)\n",
    "        fake_labels = torch.zeros(batch_size, 1).to(device)\n",
    "        optimizer_d.zero_grad()\n",
    "\n",
    "        # Real images\n",
    "        outputs_real = discriminator(real_images)\n",
    "        d_loss_real = criterion(outputs_real, real_labels)\n",
    "\n",
    "        # Fake images\n",
    "        noise = torch.randn(batch_size, noise_dim).to(device)\n",
    "        fake_images = generator(noise)\n",
    "        outputs_fake = discriminator(fake_images.detach())  # Detach so generator is not updated\n",
    "        d_loss_fake = criterion(outputs_fake, fake_labels)\n",
    "\n",
    "        # Total discriminator loss\n",
    "        d_loss = d_loss_real + d_loss_fake\n",
    "        d_loss.backward()\n",
    "        optimizer_d.step()\n",
    "        optimizer_g.zero_grad()\n",
    "        outputs_fake = discriminator(fake_images)\n",
    "        g_loss = criterion(outputs_fake, real_labels)\n",
    "        g_loss.backward()\n",
    "        optimizer_g.step()\n",
    "    print(f\"Epoch [{epoch+1}/{epochs}] | D Loss: {d_loss.item()} | G Loss: {g_loss.item()}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "fbb3db2562e1825e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T05:13:15.495519Z",
     "start_time": "2024-09-30T05:13:15.210792Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import torch\n",
    "def generate_images(generator, num_images, noise_dim):\n",
    "    noise = torch.randn(num_images, noise_dim).to(device)\n",
    "    fake_images = generator(noise).cpu().detach()\n",
    "    return fake_images\n",
    "def save_generated_images(fake_images, save_path):\n",
    "    if not os.path.exists(save_path):\n",
    "        os.makedirs(save_path)\n",
    "    for i, img in enumerate(fake_images):\n",
    "        img = img.permute(1, 2, 0).numpy()  \n",
    "        img = (img + 1) / 2  \n",
    "        plt.imsave(f\"{save_path}/image_{i+1}.png\", img)\n",
    "num_images = 10\n",
    "generated_images = generate_images(generator, num_images, noise_dim)\n",
    "save_path = \"/home/iamsanko/Downloads/generated_images\"\n",
    "save_generated_images(generated_images, save_path)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f74ca2258a3b0202",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T05:19:17.482960Z",
     "start_time": "2024-09-30T05:19:17.473127Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set size: 537\n",
      "Validation set size: 179\n",
      "Test set size: 180\n"
     ]
    }
   ],
   "source": [
    "import torch \n",
    "from torch.utils.data import DataLoader, random_split\n",
    "from torchvision import datasets, transforms\n",
    "data_dir = '/home/iamsanko/Downloads/Test 1/output'\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),  \n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.5], [0.5])  \n",
    "])\n",
    "my_dataset = datasets.ImageFolder(root=data_dir, transform=transform)\n",
    "dataset_size = len(my_dataset)\n",
    "train_size = int(0.6 * dataset_size)  # 60% for training\n",
    "val_size = int(0.2 * dataset_size)    # 20% for validation\n",
    "test_size = dataset_size - train_size - val_size \n",
    "train_dataset, val_dataset, test_dataset = random_split(my_dataset, [train_size, val_size, test_size])\n",
    "batch_size = 64 \n",
    "train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)\n",
    "test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)\n",
    "print(f\"Training set size: {len(train_loader.dataset)}\")\n",
    "print(f\"Validation set size: {len(val_loader.dataset)}\")\n",
    "print(f\"Test set size: {len(test_loader.dataset)}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8e2ea846e2e4c4d5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T05:43:41.731822Z",
     "start_time": "2024-09-30T05:43:41.724654Z"
    }
   },
   "outputs": [],
   "source": [
    "import torchvision.transforms as transforms\n",
    "from torchvision.datasets import ImageFolder\n",
    "from torch.utils.data import DataLoader]\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((64, 64)),  \n",
    "    transforms.ToTensor(),         \n",
    "])\n",
    "train_dataset = ImageFolder(root='/home/iamsanko/Downloads/Test 1/output/train', transform=transform)\n",
    "val_dataset = ImageFolder(root='//home/iamsanko/Downloads/Test 1/output/val', transform=transform)\n",
    "test_dataset = ImageFolder(root='/home/iamsanko/Downloads/Test 1/output/test', transform=transform)\n",
    "train_data_loader = DataLoader(train_dataset, batch_size=64, shuffle=True)\n",
    "val_data_loader = DataLoader(val_dataset, batch_size=64, shuffle=False)\n",
    "test_data_loader = DataLoader(test_dataset, batch_size=64, shuffle=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "336e0679b51e44a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T06:02:50.635190Z",
     "start_time": "2024-09-30T06:02:47.147907Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/5], Loss: 1.3668\n",
      "Epoch [2/5], Loss: 1.3509\n",
      "Epoch [3/5], Loss: 1.3196\n",
      "Epoch [4/5], Loss: 1.2726\n",
      "Epoch [5/5], Loss: 1.2536\n",
      "Training complete!\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "class SimpleNN(nn.Module):\n",
    "    def __init__(self, input_size, num_classes):\n",
    "        super(SimpleNN, self).__init__()\n",
    "        self.fc1 = nn.Linear(input_size, 128)\n",
    "        self.fc2 = nn.Linear(128, num_classes)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = torch.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x\n",
    "input_size = 10 \n",
    "num_classes = 4 \n",
    "num_samples = 100  \n",
    "X = torch.randn(num_samples, input_size)\n",
    "y = torch.randint(0, num_classes + 1, (num_samples,))\n",
    "y[y == 4] = 3\n",
    "dataset = TensorDataset(X, y)\n",
    "dataloader = DataLoader(dataset, batch_size=16, shuffle=True)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = SimpleNN(input_size, num_classes).to(device)\n",
    "criterion = nn.CrossEntropyLoss()  \n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
    "\n",
    "# Training loop\n",
    "num_epochs = 5\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    model.train()  \n",
    "    total_loss = 0\n",
    "\n",
    "    for inputs, labels in dataloader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device)  \n",
    "\n",
    "        # Forward pass\n",
    "        outputs = model(inputs)\n",
    "\n",
    "        # Calculate loss\n",
    "        loss = criterion(outputs, labels)\n",
    "\n",
    "        # Backward pass and optimization\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        total_loss += loss.item()\n",
    "\n",
    "    # Average loss for the epoch\n",
    "    avg_loss = total_loss / len(dataloader)\n",
    "    print(f'Epoch [{epoch + 1}/{num_epochs}], Loss: {avg_loss:.4f}')\n",
    "\n",
    "print(\"Training complete!\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "387c723adad8dd3c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T06:04:25.203210Z",
     "start_time": "2024-09-30T06:04:22.769651Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/5], Loss: 1.3229, Val Loss: 1.4607, Val Accuracy: 0.1333\n",
      "Epoch [2/5], Loss: 1.2498, Val Loss: 1.4654, Val Accuracy: 0.1333\n",
      "Epoch [3/5], Loss: 1.2068, Val Loss: 1.4774, Val Accuracy: 0.1333\n",
      "Epoch [4/5], Loss: 1.1723, Val Loss: 1.4922, Val Accuracy: 0.1333\n",
      "Epoch [5/5], Loss: 1.1595, Val Loss: 1.5084, Val Accuracy: 0.2000\n",
      "Test Loss: 1.2765, Test Accuracy: 0.3333\n",
      "Training and testing complete!\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Example neural network\n",
    "class SimpleNN(nn.Module):\n",
    "    def __init__(self, input_size, num_classes):\n",
    "        super(SimpleNN, self).__init__()\n",
    "        self.fc1 = nn.Linear(input_size, 128)\n",
    "        self.fc2 = nn.Linear(128, num_classes)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = torch.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x\n",
    "\n",
    "\n",
    "input_size = 10  \n",
    "num_classes = 4  \n",
    "num_samples = 100  \n",
    "X = torch.randn(num_samples, input_size)\n",
    "y = torch.randint(0, num_classes + 1, (num_samples,))\n",
    "y[y == 4] = 3 \n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Create DataLoaders\n",
    "train_dataset = TensorDataset(X_train, y_train)\n",
    "val_dataset = TensorDataset(X_val, y_val)\n",
    "test_dataset = TensorDataset(X_test, y_test)\n",
    "\n",
    "train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=16)\n",
    "test_loader = DataLoader(test_dataset, batch_size=16)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = SimpleNN(input_size, num_classes).to(device)\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
    "\n",
    "# Training loop\n",
    "num_epochs = 5\n",
    "for epoch in range(num_epochs):\n",
    "    model.train()\n",
    "    total_loss = 0\n",
    "    for inputs, labels in train_loader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device)  # Move data to device\n",
    "        # Forward pass\n",
    "        outputs = model(inputs)\n",
    "        # Calculate loss\n",
    "        loss = criterion(outputs, labels)\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        total_loss += loss.item()\n",
    "    avg_loss = total_loss / len(train_loader)\n",
    "    model.eval() \n",
    "    val_loss = 0\n",
    "    correct = 0\n",
    "    with torch.no_grad():\n",
    "        for inputs, labels in val_loader:\n",
    "            inputs, labels = inputs.to(device), labels.to(device) \n",
    "            outputs = model(inputs)\n",
    "            loss = criterion(outputs, labels)\n",
    "            val_loss += loss.item()\n",
    "            _, predicted = torch.max(outputs.data, 1)\n",
    "            correct += (predicted == labels).sum().item()\n",
    "    avg_val_loss = val_loss / len(val_loader)\n",
    "    val_accuracy = correct / len(val_dataset)\n",
    "\n",
    "    print(f'Epoch [{epoch + 1}/{num_epochs}], Loss: {avg_loss:.4f}, Val Loss: {avg_val_loss:.4f}, Val Accuracy: {val_accuracy:.4f}')\n",
    "\n",
    "# Testing phase\n",
    "model.eval()\n",
    "test_loss = 0\n",
    "correct = 0\n",
    "\n",
    "with torch.no_grad():\n",
    "    for inputs, labels in test_loader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device) \n",
    "        outputs = model(inputs)\n",
    "        loss = criterion(outputs, labels)\n",
    "        test_loss += loss.item()\n",
    "        _, predicted = torch.max(outputs.data, 1)\n",
    "        correct += (predicted == labels).sum().item()\n",
    "avg_test_loss = test_loss / len(test_loader)\n",
    "test_accuracy = correct / len(test_dataset)\n",
    "print(f'Test Loss: {avg_test_loss:.4f}, Test Accuracy: {test_accuracy:.4f}')\n",
    "print(\"Training and testing complete!\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "38d53824b572ead8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T06:27:49.174741Z",
     "start_time": "2024-09-30T06:27:46.870349Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/5], Train Loss: 1.3989, Validation Loss: 1.3531, Validation Accuracy: 0.4667\n",
      "Epoch [2/5], Train Loss: 1.3680, Validation Loss: 1.3323, Validation Accuracy: 0.4667\n",
      "Epoch [3/5], Train Loss: 1.3294, Validation Loss: 1.3165, Validation Accuracy: 0.5333\n",
      "Epoch [4/5], Train Loss: 1.3124, Validation Loss: 1.2979, Validation Accuracy: 0.4667\n",
      "Epoch [5/5], Train Loss: 1.2914, Validation Loss: 1.2838, Validation Accuracy: 0.4000\n",
      "Test Loss: 1.3337, Test Accuracy: 0.3333, Tes Validation: 0.3333\n",
      "Training and testing complete!\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score \n",
    "class SimpleNN(nn.Module):\n",
    "    def __init__(self, input_size, num_classes):\n",
    "        super(SimpleNN, self).__init__()\n",
    "        self.fc1 = nn.Linear(input_size, 128)\n",
    "        self.fc2 = nn.Linear(128, num_classes)\n",
    "    def forward(self, x):\n",
    "        x = torch.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        return x\n",
    "input_size = 10\n",
    "num_classes = 4\n",
    "num_samples = 100\n",
    "X = torch.randn(num_samples, input_size)\n",
    "y = torch.randint(0, num_classes, (num_samples,))\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "train_dataset = TensorDataset(X_train, y_train)\n",
    "val_dataset = TensorDataset(X_val, y_val)\n",
    "test_dataset = TensorDataset(X_test, y_test)\n",
    "train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=16)\n",
    "test_loader = DataLoader(test_dataset, batch_size=16)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = SimpleNN(input_size, num_classes).to(device)\n",
    "criterion = nn.CrossEntropyLoss()  \n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
    "num_epochs = 5\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    model.train()  \n",
    "    total_loss = 0\n",
    "    for inputs, labels in train_loader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device)  \n",
    "        # Forward pass\n",
    "        outputs = model(inputs)\n",
    "        # Calculate loss\n",
    "        loss = criterion(outputs, labels)\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        total_loss += loss.item()\n",
    "    avg_loss = total_loss / len(train_loader)\n",
    "    model.eval() \n",
    "    val_loss = 0\n",
    "    correct = 0\n",
    "    with torch.no_grad():\n",
    "        for inputs, labels in val_loader:\n",
    "            inputs, labels = inputs.to(device), labels.to(device)  # Move data to device\n",
    "            outputs = model(inputs)\n",
    "            loss = criterion(outputs, labels)\n",
    "            val_loss += loss.item()\n",
    "            _, predicted = torch.max(outputs.data, 1)\n",
    "            correct += (predicted == labels).sum().item()\n",
    "    avg_val_loss = val_loss / len(val_loader)\n",
    "    val_accuracy = correct / len(val_dataset)\n",
    "    print(f'Epoch [{epoch + 1}/{num_epochs}], Train Loss: {avg_loss:.4f}, '\n",
    "          f'Validation Loss: {avg_val_loss:.4f}, Validation Accuracy: {val_accuracy:.4f}')\n",
    "model.eval() \n",
    "test_loss = 0\n",
    "test_correct = 0\n",
    "y_test_pred = []  \n",
    "with torch.no_grad():\n",
    "    for inputs, labels in test_loader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device) \n",
    "        outputs = model(inputs)\n",
    "        loss = criterion(outputs, labels)\n",
    "        test_loss += loss.item()\n",
    "        _, predicted = torch.max(outputs.data, 1)\n",
    "        test_correct += (predicted == labels).sum().item()\n",
    "        y_test_pred.extend(predicted.cpu().numpy())\n",
    "avg_test_loss = test_loss / len(test_loader)\n",
    "test_accuracy = test_correct / len(test_dataset)\n",
    "test_validation_accuracy = accuracy_score(y_test.cpu().numpy(), y_test_pred)\n",
    "print(f'Test Loss: {avg_test_loss:.4f}, Test Accuracy: {test_accuracy:.4f}, Tes Validation: {test_validation_accuracy:.4f}')\n",
    "print(\"Training and testing complete!\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "b547b40a692200f9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T06:51:56.750992Z",
     "start_time": "2024-09-30T06:51:36.065579Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GAN Inception Score: 1.0586 ± 0.0153\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn.functional as F\n",
    "from torchvision import models, transforms\n",
    "from torchvision.utils import save_image\n",
    "from PIL import Image\n",
    "import numpy as np\n",
    "from scipy.stats import entropy\n",
    "def inception_score(images, splits=10):\n",
    "    model = models.inception_v3(pretrained=True, transform_input=False).eval()\n",
    "    if torch.cuda.is_available():\n",
    "        model = model.cuda()\n",
    "    preprocess = transforms.Compose([\n",
    "        transforms.Resize((299, 299)),  \n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),  \n",
    "    ])\n",
    "    all_probs = []\n",
    "    with torch.no_grad():\n",
    "        for img in images:\n",
    "            img = img.permute(1, 2, 0).cpu().numpy()  \n",
    "            img = (img * 255).astype(np.uint8) \n",
    "            img = Image.fromarray(img)  \n",
    "            img = preprocess(img).unsqueeze(0)  \n",
    "\n",
    "            if torch.cuda.is_available():\n",
    "                img = img.cuda()\n",
    "            prob = F.softmax(model(img), dim=1).cpu().numpy()\n",
    "            all_probs.append(prob)\n",
    "    all_probs = np.concatenate(all_probs, axis=0)  \n",
    "    scores = []\n",
    "    for i in range(splits):\n",
    "        part = all_probs[i * (len(all_probs) // splits): (i + 1) * (len(all_probs) // splits)]\n",
    "        p_y = np.mean(part, axis=0)  \n",
    "        score = np.exp(np.mean(entropy(part, p_y)) )\n",
    "        scores.append(score)\n",
    "    return np.mean(scores), np.std(scores)\n",
    "N = 100  # Number of generated images\n",
    "C = 3    # Number of channels (RGB)\n",
    "H = 64   # Height of the image\n",
    "W = 64   # Width of the image\n",
    "generated_images = torch.rand(N, C, H, W)\n",
    "mean_score, std_score = inception_score(generated_images)\n",
    "print(f'GAN Inception Score: {mean_score:.4f} ± {std_score:.4f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ade5154580278d5a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:44:53.628857Z",
     "start_time": "2024-09-30T08:42:08.425454Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GAN Inception Score: 2.4375 ± 0.2811\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from PIL import Image\n",
    "import torch\n",
    "import torchvision.transforms as transforms\n",
    "from torchvision.models import inception_v3\n",
    "import torch.nn.functional as F\n",
    "from torch.autograd import Variable\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "base_path = '/home/iamsanko/Downloads/Test 1/output'\n",
    "train_path = os.path.join(base_path, 'train')\n",
    "val_path = os.path.join(base_path, 'val')\n",
    "test_path = os.path.join(base_path, 'test')\n",
    "inception_model = inception_v3(pretrained=True, transform_input=False).to(device)\n",
    "inception_model.eval()\n",
    "preprocess = transforms.Compose([\n",
    "    transforms.Resize((299, 299)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
    "])\n",
    "def load_images_from_folder(folder):\n",
    "    images = []\n",
    "    for subdir in os.listdir(folder):\n",
    "        subdir_path = os.path.join(folder, subdir)\n",
    "        if os.path.isdir(subdir_path):\n",
    "            for filename in os.listdir(subdir_path):\n",
    "                img_path = os.path.join(subdir_path, filename)\n",
    "                if img_path.endswith(\".jpg\") or img_path.endswith(\".png\") or img_path.endswith(\".bmp\"):\n",
    "                    try:\n",
    "                        img = Image.open(img_path).convert('RGB')\n",
    "                        img_tensor = preprocess(img).unsqueeze(0) \n",
    "                        images.append(img_tensor)\n",
    "                    except Exception as e:\n",
    "                        print(f\"Error loading image {img_path}: {e}\")\n",
    "    if images:\n",
    "        return torch.cat(images) \n",
    "    return None\n",
    "def inception_score(images, splits=10):\n",
    "    if images is None or len(images) == 0:\n",
    "        print(\"No images loaded, cannot calculate Inception Score.\")\n",
    "        return None\n",
    "    all_scores = []\n",
    "    for i in range(0, len(images), 32):  \n",
    "        img_batch = images[i:i+32]\n",
    "        img_batch = img_batch.to(device)\n",
    "        with torch.no_grad():\n",
    "            output = inception_model(img_batch)\n",
    "        softmax_output = F.softmax(output, dim=1).cpu().numpy() \n",
    "        all_scores.append(softmax_output)\n",
    "    all_scores = np.concatenate(all_scores, axis=0)\n",
    "    scores = []\n",
    "    num_images = len(all_scores)\n",
    "    split_size = num_images // splits\n",
    "    for k in range(splits):\n",
    "        part = all_scores[k * split_size: (k + 1) * split_size, :]\n",
    "        p_y = np.mean(part, axis=0)\n",
    "        scores.append(np.exp(np.mean(np.sum(part * (np.log(part) - np.log(p_y)), axis=1))))\n",
    "    return np.mean(scores), np.std(scores)\n",
    "train_images = load_images_from_folder(train_path)\n",
    "val_images = load_images_from_folder(val_path)\n",
    "test_images = load_images_from_folder(test_path)\n",
    "all_images = torch.cat([train_images, val_images, test_images]) if train_images is not None and val_images is not None and test_images is not None else None\n",
    "if all_images is not None:\n",
    "    mean_score, std_dev = inception_score(all_images)\n",
    "    print(f'GAN Inception Score: {mean_score:.4f} ± {std_dev:.4f}')\n",
    "else:\n",
    "    print(\"Could not load images from the specified directories.\")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}