Image Classification Model Deployment.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l92t3N9JZk3N"
      },
      "source": [
        "# Proyek Akhir: Image Classification Model Deployment\n",
        "\n",
        "---\n",
        "\n",
        "### Dicoding Submission\n",
        "### Belajar Pengembangan Machine Learning\n",
        "\n",
        "---\n",
        "\n",
        "Kriteria submission:\n",
        "- Dataset yang akan dipakai bebas, namun minimal memiliki **1000 buah gambar**.\n",
        "- Dataset **tidak pernah** digunakan pada submission kelas machine learning sebelumnya.\n",
        "- Dataset harus dibagi menjadi **`80% train set`** dan **`20% test set`**.\n",
        "- Model harus menggunakan **model sequential**.\n",
        "- Model harus menggunakan **Conv2D Maxpooling Layer**.\n",
        "- **Akurasi** pada training dan validation set **minimal sebesar 80%**.\n",
        "- Menggunakan **Callback**.\n",
        "- Membuat **plot** terhadap akurasi dan loss model\n",
        "- Menulis kode untuk menyimpan model ke dalam format **TF-Lite**.\n",
        "\n",
        "---\n",
        "\n",
        "Saran dan Tips:\n",
        "- Dataset yang digunakan berisi lebih dari 2000 gambar.\n",
        "- Mengimplementasikan **Callback**.\n",
        "- Gambar-gambar pada dataset memiliki **resolusi yang tidak seragam.**\n",
        "\n",
        "---\n",
        "\n",
        "- **Bintang 3** : Semua ketentuan terpenuhi namun hanya mengikuti seperti apa yang ada pada modul.\n",
        "- **Bintang 4** : Semua ketentuan terpenuhi, dataset memiliki **minimal 2000** sampel gambar dan **minimal 3 kelas**. Serta akurasi pada training dan validation set **minimal 85%**.\n",
        "- **Bintang 5** : Semua ketentuan terpenuhi, dataset memiliki **minimal 10000 gambar**, resolusi gambar pada dataset tidak seragam. Serta akurasi pada training set dan validation set **minimal 92%**.\n",
        "\n",
        "---\n",
        "\n",
        "# Data Diri\n",
        "\n",
        "Nama: Andrew Benedictus Jamesie  \n",
        "E-mail: andrewbjamesie@yahoo.com  \n",
        "\n",
        "---\n",
        "---"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ub_CfCzuZbdb",
        "outputId": "c5c3cdc5-531f-4d93-f8ab-3b20e9be5640"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting split-folders\n",
            "  Downloading split_folders-0.5.1-py3-none-any.whl (8.4 kB)\n",
            "Installing collected packages: split-folders\n",
            "Successfully installed split-folders-0.5.1\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import zipfile\n",
        "import pathlib\n",
        "\n",
        "!pip install split-folders\n",
        "import splitfolders as sf\n",
        "\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.preprocessing import image\n",
        "from tensorflow.keras.applications import Xception\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import GlobalAveragePooling2D\n",
        "from tensorflow.keras.layers import Input\n",
        "from tensorflow.keras.layers import Conv2D\n",
        "from tensorflow.keras.layers import MaxPooling2D\n",
        "from tensorflow.keras.layers import Flatten\n",
        "from tensorflow.keras.layers import Dropout\n",
        "from tensorflow.keras.layers import Dense\n",
        "\n",
        "from google.colab import files\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L1Rv0axN_H-s",
        "outputId": "6c94a239-7786-454f-bf53-e503b77f9729"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading image-classification.zip to /content\n",
            "100% 2.33G/2.34G [00:11<00:00, 141MB/s]\n",
            "100% 2.34G/2.34G [00:11<00:00, 225MB/s]\n"
          ]
        }
      ],
      "source": [
        "# Kaggle username and key\n",
        "os.environ['KAGGLE_USERNAME'] = 'andrewbjamesie'\n",
        "os.environ['KAGGLE_KEY']      = 'c1cefea4cc184cb0d30547be43e5f107'\n",
        "\n",
        "# Download dataset from Kaggle\n",
        "!kaggle datasets download -d duttadebadri/image-classification"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6yYLEcZK8GIF"
      },
      "outputs": [],
      "source": [
        "# Unzip the downloaded zip file\n",
        "localZip = '/content/image-classification.zip'\n",
        "zipRef = zipfile.ZipFile(localZip, 'r')\n",
        "zipRef.extractall('image-classification')\n",
        "zipRef.close()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J3GaLzHD_CZ-",
        "outputId": "36ce6eee-c3f0-4d62-a4a9-87b40cfe2e99"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['art and culture', 'architecure', 'travel and  adventure', 'food and d rinks']"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "# List directory\n",
        "baseDir = 'image-classification/images/images'\n",
        "os.listdir(baseDir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F6t1rxO0GYKm",
        "outputId": "da48ab76-15e6-45dd-b042-fdf1331f6260"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['archi', 'fnb', 'art', 'travel']"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "# Rename directory\n",
        "os.rename(baseDir + '/art and culture',       baseDir + '/art')\n",
        "os.rename(baseDir + '/architecure',           baseDir + '/archi')\n",
        "os.rename(baseDir + '/food and d rinks',      baseDir + '/fnb')\n",
        "os.rename(baseDir + '/travel and  adventure', baseDir + '/travel')\n",
        "\n",
        "# List renamed directory\n",
        "os.listdir(baseDir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LihSRKxQH-D5",
        "outputId": "d755cfbf-ea15-40ab-eb5c-bee6d8a831ca"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Copying files: 35095 files [00:11, 2963.12 files/s]\n"
          ]
        }
      ],
      "source": [
        "# Split directory\n",
        "sf.ratio(\n",
        "    baseDir,\n",
        "    output = os.path.join('image-classification/image'),\n",
        "    seed   = None,\n",
        "    ratio  = (0.8, 0.2)\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c6rhgPJhOgnH"
      },
      "outputs": [],
      "source": [
        "# Train and val dir for each archi, art, fnb, travel\n",
        "imageDir = 'image-classification/image'\n",
        "\n",
        "trainDirArchi  = os.path.join(imageDir, 'train/archi')\n",
        "trainDirArt    = os.path.join(imageDir, 'train/art')\n",
        "trainDirFnb    = os.path.join(imageDir, 'train/fnb')\n",
        "trainDirTravel = os.path.join(imageDir, 'train/travel')\n",
        "\n",
        "valDirArchi    = os.path.join(imageDir, 'val/archi')\n",
        "valDirArt      = os.path.join(imageDir, 'val/art')\n",
        "valDirFnb      = os.path.join(imageDir, 'val/fnb')\n",
        "valDirTravel   = os.path.join(imageDir, 'val/travel')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cNpbRKvbPhz_",
        "outputId": "df81bef1-8c19-4184-bdee-3d052246d7f2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Set      : 28075\n",
            "Validation Set : 7020\n"
          ]
        }
      ],
      "source": [
        "# Count train and val image\n",
        "trainSet = (\n",
        "      len(os.listdir(trainDirArchi))\n",
        "    + len(os.listdir(trainDirArt))\n",
        "    + len(os.listdir(trainDirFnb))\n",
        "    + len(os.listdir(trainDirTravel))\n",
        ")\n",
        "\n",
        "valSet = (\n",
        "      len(os.listdir(valDirArchi))\n",
        "    + len(os.listdir(valDirArt))\n",
        "    + len(os.listdir(valDirFnb))\n",
        "    + len(os.listdir(valDirTravel))\n",
        ")\n",
        "\n",
        "print(f'Train Set      : {trainSet}')\n",
        "print(f'Validation Set : {valSet}')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ubKM6y-oQPFd",
        "outputId": "e2253f26-560e-4778-eb7a-c46e58291e9f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['archi', 'fnb', 'art', 'travel']\n",
            "['archi', 'fnb', 'art', 'travel']\n"
          ]
        }
      ],
      "source": [
        "# List directory of train and validation image\n",
        "trainDir = os.path.join(imageDir, 'train')\n",
        "valDir   = os.path.join(imageDir, 'val')\n",
        "\n",
        "print(os.listdir(trainDir))\n",
        "print(os.listdir(valDir))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iCMz6-iAR9Rl"
      },
      "outputs": [],
      "source": [
        "# Image Augmentation for duplicating image\n",
        "trainDatagen = ImageDataGenerator(\n",
        "    rescale            = 1./255,\n",
        "    rotation_range     = 30,\n",
        "    # width_shift_range  = 0.2,\n",
        "    # height_shift_range = 0.2,\n",
        "    shear_range        = 0.2,\n",
        "    zoom_range         = 0.2,\n",
        "    horizontal_flip    = True,\n",
        "    # vertical_flip      = True,\n",
        "    fill_mode          = 'nearest',\n",
        "    # validation_split   = 0.2\n",
        ")\n",
        "\n",
        "valDatagen = ImageDataGenerator(\n",
        "    rescale         = 1./255\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bLtucg0dZuWI",
        "outputId": "99a124f7-87b8-48d5-f15a-940a03969720"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 28073 images belonging to 4 classes.\n",
            "Found 7020 images belonging to 4 classes.\n"
          ]
        }
      ],
      "source": [
        "# Preperation the training and validation data with .flow_from_directory()\n",
        "trainGen = trainDatagen.flow_from_directory(\n",
        "    trainDir,\n",
        "    target_size = (200, 200),\n",
        "    batch_size  = 50,\n",
        "    shuffle     = True,\n",
        "    color_mode  = 'rgb',\n",
        "    class_mode  = 'categorical',\n",
        "    # save_format = 'jpeg'\n",
        ")\n",
        "\n",
        "valGen = valDatagen.flow_from_directory(\n",
        "    valDir,\n",
        "    target_size = (200, 200),\n",
        "    batch_size  = 50,\n",
        "    shuffle     = True,\n",
        "    color_mode  = 'rgb',\n",
        "    class_mode  = 'categorical',\n",
        "    # save_format = 'jpeg'\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseModel = Xception(weights=\"imagenet\", include_top=False, input_shape=(200, 200, 3))\n",
        "\n",
        "baseModel.trainable = False\n",
        "\n",
        "baseModel.summary()\n",
        "print(f'Base Model Layer : {len(baseModel.layers)}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fYXZMkUtmYz1",
        "outputId": "8a0b438c-ef1d-4b5c-98a1-0ba7433012b6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/xception/xception_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "83689472/83683744 [==============================] - 0s 0us/step\n",
            "83697664/83683744 [==============================] - 0s 0us/step\n",
            "Model: \"xception\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(None, 200, 200, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " block1_conv1 (Conv2D)          (None, 99, 99, 32)   864         ['input_1[0][0]']                \n",
            "                                                                                                  \n",
            " block1_conv1_bn (BatchNormaliz  (None, 99, 99, 32)  128         ['block1_conv1[0][0]']           \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " block1_conv1_act (Activation)  (None, 99, 99, 32)   0           ['block1_conv1_bn[0][0]']        \n",
            "                                                                                                  \n",
            " block1_conv2 (Conv2D)          (None, 97, 97, 64)   18432       ['block1_conv1_act[0][0]']       \n",
            "                                                                                                  \n",
            " block1_conv2_bn (BatchNormaliz  (None, 97, 97, 64)  256         ['block1_conv2[0][0]']           \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " block1_conv2_act (Activation)  (None, 97, 97, 64)   0           ['block1_conv2_bn[0][0]']        \n",
            "                                                                                                  \n",
            " block2_sepconv1 (SeparableConv  (None, 97, 97, 128)  8768       ['block1_conv2_act[0][0]']       \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block2_sepconv1_bn (BatchNorma  (None, 97, 97, 128)  512        ['block2_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block2_sepconv2_act (Activatio  (None, 97, 97, 128)  0          ['block2_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block2_sepconv2 (SeparableConv  (None, 97, 97, 128)  17536      ['block2_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block2_sepconv2_bn (BatchNorma  (None, 97, 97, 128)  512        ['block2_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d (Conv2D)                (None, 49, 49, 128)  8192        ['block1_conv2_act[0][0]']       \n",
            "                                                                                                  \n",
            " block2_pool (MaxPooling2D)     (None, 49, 49, 128)  0           ['block2_sepconv2_bn[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization (BatchNorm  (None, 49, 49, 128)  512        ['conv2d[0][0]']                 \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " add (Add)                      (None, 49, 49, 128)  0           ['block2_pool[0][0]',            \n",
            "                                                                  'batch_normalization[0][0]']    \n",
            "                                                                                                  \n",
            " block3_sepconv1_act (Activatio  (None, 49, 49, 128)  0          ['add[0][0]']                    \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block3_sepconv1 (SeparableConv  (None, 49, 49, 256)  33920      ['block3_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block3_sepconv1_bn (BatchNorma  (None, 49, 49, 256)  1024       ['block3_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block3_sepconv2_act (Activatio  (None, 49, 49, 256)  0          ['block3_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block3_sepconv2 (SeparableConv  (None, 49, 49, 256)  67840      ['block3_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block3_sepconv2_bn (BatchNorma  (None, 49, 49, 256)  1024       ['block3_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_1 (Conv2D)              (None, 25, 25, 256)  32768       ['add[0][0]']                    \n",
            "                                                                                                  \n",
            " block3_pool (MaxPooling2D)     (None, 25, 25, 256)  0           ['block3_sepconv2_bn[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization_1 (BatchNo  (None, 25, 25, 256)  1024       ['conv2d_1[0][0]']               \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " add_1 (Add)                    (None, 25, 25, 256)  0           ['block3_pool[0][0]',            \n",
            "                                                                  'batch_normalization_1[0][0]']  \n",
            "                                                                                                  \n",
            " block4_sepconv1_act (Activatio  (None, 25, 25, 256)  0          ['add_1[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block4_sepconv1 (SeparableConv  (None, 25, 25, 728)  188672     ['block4_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block4_sepconv1_bn (BatchNorma  (None, 25, 25, 728)  2912       ['block4_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block4_sepconv2_act (Activatio  (None, 25, 25, 728)  0          ['block4_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block4_sepconv2 (SeparableConv  (None, 25, 25, 728)  536536     ['block4_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block4_sepconv2_bn (BatchNorma  (None, 25, 25, 728)  2912       ['block4_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_2 (Conv2D)              (None, 13, 13, 728)  186368      ['add_1[0][0]']                  \n",
            "                                                                                                  \n",
            " block4_pool (MaxPooling2D)     (None, 13, 13, 728)  0           ['block4_sepconv2_bn[0][0]']     \n",
            "                                                                                                  \n",
            " batch_normalization_2 (BatchNo  (None, 13, 13, 728)  2912       ['conv2d_2[0][0]']               \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " add_2 (Add)                    (None, 13, 13, 728)  0           ['block4_pool[0][0]',            \n",
            "                                                                  'batch_normalization_2[0][0]']  \n",
            "                                                                                                  \n",
            " block5_sepconv1_act (Activatio  (None, 13, 13, 728)  0          ['add_2[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block5_sepconv1 (SeparableConv  (None, 13, 13, 728)  536536     ['block5_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block5_sepconv1_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block5_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block5_sepconv2_act (Activatio  (None, 13, 13, 728)  0          ['block5_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block5_sepconv2 (SeparableConv  (None, 13, 13, 728)  536536     ['block5_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block5_sepconv2_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block5_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block5_sepconv3_act (Activatio  (None, 13, 13, 728)  0          ['block5_sepconv2_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block5_sepconv3 (SeparableConv  (None, 13, 13, 728)  536536     ['block5_sepconv3_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block5_sepconv3_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block5_sepconv3[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " add_3 (Add)                    (None, 13, 13, 728)  0           ['block5_sepconv3_bn[0][0]',     \n",
            "                                                                  'add_2[0][0]']                  \n",
            "                                                                                                  \n",
            " block6_sepconv1_act (Activatio  (None, 13, 13, 728)  0          ['add_3[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block6_sepconv1 (SeparableConv  (None, 13, 13, 728)  536536     ['block6_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block6_sepconv1_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block6_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block6_sepconv2_act (Activatio  (None, 13, 13, 728)  0          ['block6_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block6_sepconv2 (SeparableConv  (None, 13, 13, 728)  536536     ['block6_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block6_sepconv2_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block6_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block6_sepconv3_act (Activatio  (None, 13, 13, 728)  0          ['block6_sepconv2_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block6_sepconv3 (SeparableConv  (None, 13, 13, 728)  536536     ['block6_sepconv3_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block6_sepconv3_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block6_sepconv3[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " add_4 (Add)                    (None, 13, 13, 728)  0           ['block6_sepconv3_bn[0][0]',     \n",
            "                                                                  'add_3[0][0]']                  \n",
            "                                                                                                  \n",
            " block7_sepconv1_act (Activatio  (None, 13, 13, 728)  0          ['add_4[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block7_sepconv1 (SeparableConv  (None, 13, 13, 728)  536536     ['block7_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block7_sepconv1_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block7_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block7_sepconv2_act (Activatio  (None, 13, 13, 728)  0          ['block7_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block7_sepconv2 (SeparableConv  (None, 13, 13, 728)  536536     ['block7_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block7_sepconv2_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block7_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block7_sepconv3_act (Activatio  (None, 13, 13, 728)  0          ['block7_sepconv2_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block7_sepconv3 (SeparableConv  (None, 13, 13, 728)  536536     ['block7_sepconv3_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block7_sepconv3_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block7_sepconv3[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " add_5 (Add)                    (None, 13, 13, 728)  0           ['block7_sepconv3_bn[0][0]',     \n",
            "                                                                  'add_4[0][0]']                  \n",
            "                                                                                                  \n",
            " block8_sepconv1_act (Activatio  (None, 13, 13, 728)  0          ['add_5[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block8_sepconv1 (SeparableConv  (None, 13, 13, 728)  536536     ['block8_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block8_sepconv1_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block8_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block8_sepconv2_act (Activatio  (None, 13, 13, 728)  0          ['block8_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block8_sepconv2 (SeparableConv  (None, 13, 13, 728)  536536     ['block8_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block8_sepconv2_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block8_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block8_sepconv3_act (Activatio  (None, 13, 13, 728)  0          ['block8_sepconv2_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block8_sepconv3 (SeparableConv  (None, 13, 13, 728)  536536     ['block8_sepconv3_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block8_sepconv3_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block8_sepconv3[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " add_6 (Add)                    (None, 13, 13, 728)  0           ['block8_sepconv3_bn[0][0]',     \n",
            "                                                                  'add_5[0][0]']                  \n",
            "                                                                                                  \n",
            " block9_sepconv1_act (Activatio  (None, 13, 13, 728)  0          ['add_6[0][0]']                  \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block9_sepconv1 (SeparableConv  (None, 13, 13, 728)  536536     ['block9_sepconv1_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block9_sepconv1_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block9_sepconv1[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block9_sepconv2_act (Activatio  (None, 13, 13, 728)  0          ['block9_sepconv1_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block9_sepconv2 (SeparableConv  (None, 13, 13, 728)  536536     ['block9_sepconv2_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block9_sepconv2_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block9_sepconv2[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " block9_sepconv3_act (Activatio  (None, 13, 13, 728)  0          ['block9_sepconv2_bn[0][0]']     \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " block9_sepconv3 (SeparableConv  (None, 13, 13, 728)  536536     ['block9_sepconv3_act[0][0]']    \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " block9_sepconv3_bn (BatchNorma  (None, 13, 13, 728)  2912       ['block9_sepconv3[0][0]']        \n",
            " lization)                                                                                        \n",
            "                                                                                                  \n",
            " add_7 (Add)                    (None, 13, 13, 728)  0           ['block9_sepconv3_bn[0][0]',     \n",
            "                                                                  'add_6[0][0]']                  \n",
            "                                                                                                  \n",
            " block10_sepconv1_act (Activati  (None, 13, 13, 728)  0          ['add_7[0][0]']                  \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block10_sepconv1 (SeparableCon  (None, 13, 13, 728)  536536     ['block10_sepconv1_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block10_sepconv1_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block10_sepconv1[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block10_sepconv2_act (Activati  (None, 13, 13, 728)  0          ['block10_sepconv1_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block10_sepconv2 (SeparableCon  (None, 13, 13, 728)  536536     ['block10_sepconv2_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block10_sepconv2_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block10_sepconv2[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block10_sepconv3_act (Activati  (None, 13, 13, 728)  0          ['block10_sepconv2_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block10_sepconv3 (SeparableCon  (None, 13, 13, 728)  536536     ['block10_sepconv3_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block10_sepconv3_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block10_sepconv3[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " add_8 (Add)                    (None, 13, 13, 728)  0           ['block10_sepconv3_bn[0][0]',    \n",
            "                                                                  'add_7[0][0]']                  \n",
            "                                                                                                  \n",
            " block11_sepconv1_act (Activati  (None, 13, 13, 728)  0          ['add_8[0][0]']                  \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block11_sepconv1 (SeparableCon  (None, 13, 13, 728)  536536     ['block11_sepconv1_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block11_sepconv1_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block11_sepconv1[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block11_sepconv2_act (Activati  (None, 13, 13, 728)  0          ['block11_sepconv1_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block11_sepconv2 (SeparableCon  (None, 13, 13, 728)  536536     ['block11_sepconv2_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block11_sepconv2_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block11_sepconv2[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block11_sepconv3_act (Activati  (None, 13, 13, 728)  0          ['block11_sepconv2_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block11_sepconv3 (SeparableCon  (None, 13, 13, 728)  536536     ['block11_sepconv3_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block11_sepconv3_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block11_sepconv3[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " add_9 (Add)                    (None, 13, 13, 728)  0           ['block11_sepconv3_bn[0][0]',    \n",
            "                                                                  'add_8[0][0]']                  \n",
            "                                                                                                  \n",
            " block12_sepconv1_act (Activati  (None, 13, 13, 728)  0          ['add_9[0][0]']                  \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block12_sepconv1 (SeparableCon  (None, 13, 13, 728)  536536     ['block12_sepconv1_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block12_sepconv1_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block12_sepconv1[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block12_sepconv2_act (Activati  (None, 13, 13, 728)  0          ['block12_sepconv1_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block12_sepconv2 (SeparableCon  (None, 13, 13, 728)  536536     ['block12_sepconv2_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block12_sepconv2_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block12_sepconv2[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block12_sepconv3_act (Activati  (None, 13, 13, 728)  0          ['block12_sepconv2_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block12_sepconv3 (SeparableCon  (None, 13, 13, 728)  536536     ['block12_sepconv3_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block12_sepconv3_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block12_sepconv3[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " add_10 (Add)                   (None, 13, 13, 728)  0           ['block12_sepconv3_bn[0][0]',    \n",
            "                                                                  'add_9[0][0]']                  \n",
            "                                                                                                  \n",
            " block13_sepconv1_act (Activati  (None, 13, 13, 728)  0          ['add_10[0][0]']                 \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block13_sepconv1 (SeparableCon  (None, 13, 13, 728)  536536     ['block13_sepconv1_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block13_sepconv1_bn (BatchNorm  (None, 13, 13, 728)  2912       ['block13_sepconv1[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block13_sepconv2_act (Activati  (None, 13, 13, 728)  0          ['block13_sepconv1_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block13_sepconv2 (SeparableCon  (None, 13, 13, 1024  752024     ['block13_sepconv2_act[0][0]']   \n",
            " v2D)                           )                                                                 \n",
            "                                                                                                  \n",
            " block13_sepconv2_bn (BatchNorm  (None, 13, 13, 1024  4096       ['block13_sepconv2[0][0]']       \n",
            " alization)                     )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_3 (Conv2D)              (None, 7, 7, 1024)   745472      ['add_10[0][0]']                 \n",
            "                                                                                                  \n",
            " block13_pool (MaxPooling2D)    (None, 7, 7, 1024)   0           ['block13_sepconv2_bn[0][0]']    \n",
            "                                                                                                  \n",
            " batch_normalization_3 (BatchNo  (None, 7, 7, 1024)  4096        ['conv2d_3[0][0]']               \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " add_11 (Add)                   (None, 7, 7, 1024)   0           ['block13_pool[0][0]',           \n",
            "                                                                  'batch_normalization_3[0][0]']  \n",
            "                                                                                                  \n",
            " block14_sepconv1 (SeparableCon  (None, 7, 7, 1536)  1582080     ['add_11[0][0]']                 \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block14_sepconv1_bn (BatchNorm  (None, 7, 7, 1536)  6144        ['block14_sepconv1[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block14_sepconv1_act (Activati  (None, 7, 7, 1536)  0           ['block14_sepconv1_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " block14_sepconv2 (SeparableCon  (None, 7, 7, 2048)  3159552     ['block14_sepconv1_act[0][0]']   \n",
            " v2D)                                                                                             \n",
            "                                                                                                  \n",
            " block14_sepconv2_bn (BatchNorm  (None, 7, 7, 2048)  8192        ['block14_sepconv2[0][0]']       \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " block14_sepconv2_act (Activati  (None, 7, 7, 2048)  0           ['block14_sepconv2_bn[0][0]']    \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 20,861,480\n",
            "Trainable params: 0\n",
            "Non-trainable params: 20,861,480\n",
            "__________________________________________________________________________________________________\n",
            "Base Model Layer : 132\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = Sequential([\n",
        "    baseModel,\n",
        "    GlobalAveragePooling2D(),\n",
        "    Dense(4, activation='softmax')\n",
        "])\n",
        "\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QpkwYNtimdLz",
        "outputId": "0bdd9313-f85d-4713-aba3-9a2353c73dc4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " xception (Functional)       (None, 7, 7, 2048)        20861480  \n",
            "                                                                 \n",
            " global_average_pooling2d (G  (None, 2048)             0         \n",
            " lobalAveragePooling2D)                                          \n",
            "                                                                 \n",
            " dense (Dense)               (None, 4)                 8196      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 20,869,676\n",
            "Trainable params: 8,196\n",
            "Non-trainable params: 20,861,480\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Qu1scqtlyogc"
      },
      "outputs": [],
      "source": [
        "# Model architecture\n",
        "# pre_trained_model = ResNet152V2(\n",
        "#     weights      = 'imagenet',\n",
        "#     include_top  = False,\n",
        "#     input_tensor = Input(shape=(150, 150, 3))\n",
        "# )\n",
        "\n",
        "# for layer in pre_trained_model.layers:\n",
        "#     layer.trainable = False\n",
        "\n",
        "# model_output = pre_trained_model.output\n",
        "\n",
        "# m = Flatten(name='flatten')(model_output)\n",
        "# m = Dropout(0.5)(m)\n",
        "# m = Dense(128, activation='relu')(m)\n",
        "# m = Dense(4,   activation='softmax')(m)\n",
        "\n",
        "# model = Sequential(pre_trained_model.input, m)\n",
        "\n",
        "# from tensorflow.keras.applications import EfficientNetB3\n",
        "# from tensorflow.keras.layers import BatchNormalization\n",
        "\n",
        "# model = Sequential([\n",
        "#     EfficientNetB3(weights=\"imagenet\", include_top=False, pooling='max', input_shape=(200, 200, 3), input_tensor=Input(shape=(200, 200, 3))),\n",
        "#     BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001),\n",
        "#     Dropout(0.25),\n",
        "#     Flatten(),\n",
        "#     Dense(256, activation='relu'),\n",
        "#     Dense(128, activation='relu'),\n",
        "#     Dense(4,   activation='softmax')\n",
        "#     # Conv2D(64,  (3, 3), activation='relu', input_shape=(200, 200, 3)),\n",
        "#     # MaxPooling2D(2, 2),\n",
        "#     # Dropout(0.25),\n",
        "#     # Conv2D(64,  (3, 3), activation='relu'),\n",
        "#     # MaxPooling2D(2, 2),\n",
        "#     # Conv2D(128,  (3, 3), activation='relu'),\n",
        "#     # MaxPooling2D(2, 2),\n",
        "#     # Conv2D(256,  (3, 3), activation='relu'),\n",
        "#     # MaxPooling2D(2, 2),\n",
        "#     # Conv2D(512,  (3, 3), activation='relu'),\n",
        "#     # MaxPooling2D(2, 2),\n",
        "#     # Dropout(0.25),\n",
        "#     # Flatten(),\n",
        "#     # Dense(512, activation='relu'),\n",
        "#     # Dense(4,   activation='softmax')\n",
        "# ])\n",
        "\n",
        "# model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wogwdFo5382V"
      },
      "outputs": [],
      "source": [
        "model.compile(\n",
        "    optimizer = 'adam',\n",
        "    # optimizer = tf.optimizers.Adam(lr=1e-4),\n",
        "    # optimizer = tf.keras.optimizers.SGD(lr=1e-4, momentum=0.9),\n",
        "    loss      = 'categorical_crossentropy',\n",
        "    metrics   = ['accuracy']\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lAdXGo1m27Fr"
      },
      "outputs": [],
      "source": [
        "# Stop training callback\n",
        "class stopCallback(tf.keras.callbacks.Callback):\n",
        "    def on_epoch_end(self, epoch, logs={}):\n",
        "        if (logs.get('accuracy') > 0.92 and logs.get('val_accuracy') > 0.92):\n",
        "            print('\\nAccuracy and Validation Accuracy reach > 92%')\n",
        "            # self.model.stop_training = True\n",
        "\n",
        "stopTraining = stopCallback()\n",
        "\n",
        "# ReduceLROnPlateau callback\n",
        "reduceLROP   = tf.keras.callbacks.ReduceLROnPlateau(monitor='val_loss', patience=3)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "\n",
        "**Cannot connect to GPU backend**\n",
        "\n",
        "\n",
        "You cannot currently connect to a GPU due to usage limits in Colab. [Learn more](https://research.google.com/colaboratory/faq.html#usage-limits)\n",
        "\n",
        "If you are interested in priority access to GPUs and higher usage limits, you may want to check out [Colab Pro](https://colab.research.google.com/signup?utm_source=dialog&utm_medium=link&utm_campaign=gpu_assignment_failure).\n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "abvGBg84kcls"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "**Error ketika training di model.fit()**\n",
        "\n",
        "\n",
        "```\n",
        "Epoch 1/30\n",
        "32/32 - 358s - loss: 1.5903 - accuracy: 0.2189 - val_loss: 1.9429 - val_accuracy: 0.1125 - lr: 1.0000e-04 - 358s/epoch - 11s/step\n",
        "Epoch 2/30\n",
        "\n",
        "/usr/local/lib/python3.7/dist-packages/PIL/Image.py:960: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n",
        "  \"Palette images with Transparency expressed in bytes should be \"\n",
        "\n",
        "/usr/local/lib/python3.7/dist-packages/PIL/TiffImagePlugin.py:788: UserWarning: Corrupt EXIF data.  Expecting to read 4 bytes but only got 0. \n",
        "  warnings.warn(str(msg))\n",
        "\n",
        "32/32 - 304s - loss: 1.1308 - accuracy: 0.2762 - val_loss: 1.6228 - val_accuracy: 0.1438 - lr: 1.0000e-04 - 304s/epoch - 9s/step\n",
        "Epoch 3/30\n",
        "32/32 - 305s - loss: 0.9584 - accuracy: 0.3191 - val_loss: 1.4837 - val_accuracy: 0.2187 - lr: 1.0000e-04 - 305s/epoch - 10s/step\n",
        "\n",
        "```\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "Pfk8afV2gD_j"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a0fYBmSK1xk5",
        "outputId": "59647a2e-682b-47d6-95be-f2371353a753"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/PIL/Image.py:960: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n",
            "  \"Palette images with Transparency expressed in bytes should be \"\n",
            "/usr/local/lib/python3.7/dist-packages/PIL/TiffImagePlugin.py:788: UserWarning: Corrupt EXIF data.  Expecting to read 4 bytes but only got 0. \n",
            "  warnings.warn(str(msg))\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "562/562 - 427s - loss: 0.3557 - accuracy: 0.8815 - val_loss: 0.2708 - val_accuracy: 0.9093 - 427s/epoch - 760ms/step\n",
            "Epoch 2/10\n",
            "562/562 - 402s - loss: 0.2674 - accuracy: 0.9108 - val_loss: 0.2512 - val_accuracy: 0.9150 - 402s/epoch - 715ms/step\n",
            "Epoch 3/10\n",
            "562/562 - 399s - loss: 0.2497 - accuracy: 0.9160 - val_loss: 0.2393 - val_accuracy: 0.9174 - 399s/epoch - 711ms/step\n",
            "Epoch 4/10\n",
            "562/562 - 402s - loss: 0.2377 - accuracy: 0.9186 - val_loss: 0.2358 - val_accuracy: 0.9181 - 402s/epoch - 715ms/step\n",
            "Epoch 5/10\n",
            "562/562 - 401s - loss: 0.2308 - accuracy: 0.9204 - val_loss: 0.2322 - val_accuracy: 0.9185 - 401s/epoch - 713ms/step\n",
            "Epoch 6/10\n",
            "562/562 - 401s - loss: 0.2278 - accuracy: 0.9217 - val_loss: 0.2301 - val_accuracy: 0.9198 - 401s/epoch - 713ms/step\n",
            "Epoch 7/10\n",
            "562/562 - 400s - loss: 0.2207 - accuracy: 0.9247 - val_loss: 0.2355 - val_accuracy: 0.9192 - 400s/epoch - 711ms/step\n",
            "Epoch 8/10\n",
            "562/562 - 400s - loss: 0.2188 - accuracy: 0.9230 - val_loss: 0.2331 - val_accuracy: 0.9199 - 400s/epoch - 712ms/step\n",
            "Epoch 9/10\n",
            "562/562 - 399s - loss: 0.2148 - accuracy: 0.9263 - val_loss: 0.2270 - val_accuracy: 0.9192 - 399s/epoch - 709ms/step\n",
            "Epoch 10/10\n",
            "562/562 - 398s - loss: 0.2113 - accuracy: 0.9279 - val_loss: 0.2262 - val_accuracy: 0.9184 - 398s/epoch - 709ms/step\n"
          ]
        }
      ],
      "source": [
        "# Model training with .fit()\n",
        "epoch = 10\n",
        "\n",
        "history = model.fit(\n",
        "    trainGen,\n",
        "    # steps_per_epoch  = 32,\n",
        "    epochs           = epoch,\n",
        "    validation_data  = valGen,\n",
        "    # validation_steps = 5,\n",
        "    verbose          = 2,\n",
        "    callbacks        = [stopTraining, ]\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_6SChH2J2C5o",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        },
        "outputId": "5f376607-9af8-427d-d6b6-8fb3408e5af0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "# Visualize accuracy and loss plot\n",
        "accuracy     = history.history['accuracy']\n",
        "val_accuracy = history.history['val_accuracy']\n",
        "\n",
        "loss         = history.history['loss']\n",
        "val_loss     = history.history['val_loss']\n",
        "\n",
        "plt.figure(figsize = (12, 4))\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(range(epoch), accuracy,     label='Training Accuracy')\n",
        "plt.plot(range(epoch), val_accuracy, label='Validation Accuracy')\n",
        "plt.title('Training and Validation Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(range(epoch), loss,     label='Training Loss')\n",
        "plt.plot(range(epoch), val_loss, label='Validation Loss')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend(loc='upper right')\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "rpuZPevl9klA",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "60383161-03a1-481e-ba4c-50df16882264"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'archi': 0, 'art': 1, 'fnb': 2, 'travel': 3}\n"
          ]
        }
      ],
      "source": [
        "print(trainGen.class_indices)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Y1oDI50T29Ju",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 397,
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          }
        },
        "outputId": "ae5f7c48-ea6a-4268-b436-7853dc241464"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-5cb532a5-8534-45ac-8585-6737e529f512\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-5cb532a5-8534-45ac-8585-6737e529f512\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving 04_Monasterio_de_Palazuelos_absides_exterior_ni.jpg to 04_Monasterio_de_Palazuelos_absides_exterior_ni (2).jpg\n",
            "04_Monasterio_de_Palazuelos_absides_exterior_ni.jpg\n",
            "Art and Culture\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "uploaded = files.upload()\n",
        "\n",
        "# Predicting images\n",
        "for up in uploaded.keys():\n",
        "    path = up\n",
        "    img  = image.load_img(path, target_size = (200, 200))\n",
        "\n",
        "    imgplot = plt.imshow(img)\n",
        "    x       = image.img_to_array(img)\n",
        "    x       = np.expand_dims(x, axis = 0)\n",
        "\n",
        "    images  = np.vstack([x])\n",
        "    classes = model.predict(images, batch_size = 10)\n",
        "    print(up)\n",
        "\n",
        "    # if classes[0][0] == 1:\n",
        "    #     print('Architecture')\n",
        "    # elif classes[0][1] == 1:\n",
        "    #     print('Food and Drinks')\n",
        "    # elif classes[0][2] == 1:\n",
        "    #     print('Art and Culture')\n",
        "    # elif classes[0][3] == 1:\n",
        "    #     print('Travel and Adventure')\n",
        "    # else:\n",
        "    #     print('Unclassified')\n",
        "\n",
        "    if classes == 0:\n",
        "        print('Architecture')\n",
        "    elif classes == 1:\n",
        "        print('Art and Culture')\n",
        "    elif classes == 2:\n",
        "        print('Food and Drinks')\n",
        "    elif classes == 3:\n",
        "        print('Travel and Adventure')\n",
        "    else:\n",
        "        print('Unclassified')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6PVxFKNs_IcE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2e02bcb6-86a8-436c-aecd-4763397653b6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Buffer deduplication procedure will be skipped when flatbuffer library is not properly loaded\n"
          ]
        }
      ],
      "source": [
        "# Save model to SavedModel format\n",
        "exportDir = 'saved_model/'\n",
        "tf.saved_model.save(model, exportDir)\n",
        "\n",
        "# Convert model to TF-Lite format\n",
        "converter    = tf.lite.TFLiteConverter.from_saved_model(exportDir)\n",
        "tflite_model = converter.convert()\n",
        "\n",
        "# tflite_model_file = pathlib.Path('vegs.tflite')\n",
        "# tflite_model_file.write_bytes(tflite_model)\n",
        "\n",
        "# Save the model\n",
        "with open('model.tflite', 'wb') as t:\n",
        "    t.write(tflite_model)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GB7I8zBg2FZA"
      },
      "source": [
        "---\n",
        "\n",
        "\n",
        "References:\n",
        "\n",
        "[(Dicoding) Latihan: Deploy Model ML Menggunakan TensorFlow Lite](https://www.dicoding.com/academies/185/tutorials/10239)\n",
        "\n",
        "[(Dicoding) Error Saat Memakai Transfer Learning](https://www.dicoding.com/academies/185/discussions/117337)\n",
        "\n",
        "[(Dicoding) Kesalahan Hasil Prediksi, Sudah Pakai Pre-trained Model (Resnet, Learning Rate) ](https://www.dicoding.com/academies/185/discussions/139707)\n",
        "\n",
        "[(Dicoding) ValueError: Unknown loss function: categotical_crossentropy](https://www.dicoding.com/academies/185/discussions/131992)\n",
        "\n",
        "[(Dicoding) Cara Meningkatkan Akurasi?](https://www.dicoding.com/academies/185/discussions/128712)\n",
        "\n",
        "[(Dicoding) logits and labels must be broadcastable: logits_size=[32,2] labels_size=[32,3]](https://www.dicoding.com/academies/185/discussions/130702)\n",
        "\n",
        "[(Colab) Pendeteksi_Kebersihan_Lingkungan2.ipynb](https://colab.research.google.com/drive/1XYbPUen-IO6MKYtqRQMg5ypVupCsI_y7?usp=sharing)\n",
        "\n",
        "[(Colab) ProyekAkhirMLmenengah.ipynb](https://colab.research.google.com/drive/1pa06QfSueDDrjdOlp1TQUGj0DUA8dR6Q?usp=sharing)\n",
        "\n",
        "[(Colab) Rev2 Final_Submission_ItMG.ipynb](https://colab.research.google.com/drive/1b-RoPVn_ikdcWjKlyL09xA1ZA8NBVAAb?usp=sharing)\n",
        "\n",
        "[(Colab) Submission3MachineLearningIntermediate.ipynb](https://colab.research.google.com/drive/13LMkMCtvtnyD_lsz2BfP2yxY-jZEx9Fi)\n",
        "\n",
        "[(GitHub) Submission3MachineLearningIntermediate.ipynb](https://github.com/wilsonoey60/Kumpulan-Submission-Dicoding/blob/main/Bintang%205/Belajar%20Pengembangan%20Machine%20Learning/Submission%203/Submission3MachineLearningIntermediate.ipynb)\n",
        "\n",
        "\n",
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "b7DJnF57lGLv"
      },
      "source": [
        "---\n",
        "\n",
        "\n",
        "Dataset:\n",
        "\n",
        "[(Kaggle) Google Scraped Image Dataset](https://www.kaggle.com/datasets/duttadebadri/image-classification)\n",
        "\n",
        "[(Kaggle) Google Image Classification Xception Model](https://www.kaggle.com/code/israakhalil/google-image-classification-xception-model)\n",
        "\n",
        "[(Kaggle) EfficientNet Model F1 Score = 94%](https://www.kaggle.com/code/gpiosenka/efficientnet-model-f1-score-94)\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "Alternative Dataset:\n",
        "\n",
        "[(Kaggle) Vegetable Image Dataset](https://www.kaggle.com/datasets/misrakahmed/vegetable-image-dataset)\n",
        "\n",
        "[(Kaggle) Food-11 Image Dataset](https://www.kaggle.com/datasets/trolukovich/food11-image-dataset)\n",
        "\n",
        "\n",
        "---"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}