depixelator_training.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1lZXN11cY5xi"
      },
      "source": [
        "## GETTING THE DATASETS"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4X-1lbxMY-Jd",
        "outputId": "d8286327-ff95-44cb-e7e0-3e0098a1abf2"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LKgUTC0yksvd"
      },
      "source": [
        "- `dataset.zip` is the COCO 2017 val dataset, cropped to 256x256 and pixelated"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aWZs4QsUkWJ5"
      },
      "outputs": [],
      "source": [
        "!unzip /content/drive/MyDrive/DEPIXELATOR_004/dataset.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": true,
        "id": "x3KHey_skh_1"
      },
      "outputs": [],
      "source": [
        "!mv /content/content/drive/MyDrive/DEPIXELATOR_003/x/* /content/x/\n",
        "!mv /content/content/drive/MyDrive/DEPIXELATOR_003/y/* /content/y/"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uf9-lPwLqasq",
        "outputId": "ed7453d8-ab3a-4cb3-9a64-34de29b14f37"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "7996\n",
            "7996\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "print(len(os.listdir('/content/x/')))\n",
        "print(len(os.listdir('/content/y/')))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JJ__DIFE404y"
      },
      "source": [
        "## IMPORTING THE LIBRARIES"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7P8eIT4xksve"
      },
      "source": [
        "upgrading these libraries to the latest version"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mSSv2VwyfC7U"
      },
      "outputs": [],
      "source": [
        "%pip install --upgrade pip\n",
        "%pip install --upgrade tensorflow\n",
        "%pip install --upgrade scikit-learn\n",
        "%pip install --upgrade pydot\n",
        "%pip install --upgrade matplotlib\n",
        "\n",
        "## IN CASE YOU ARE FACING ISSUES WHILE LOADING MODELS OR IMPORTING LIBRARIES YOU CAN\n",
        "## USE THE VERSIONS OF THESE LIBRARIES ON WHICH OUR MODEL WAS CREATED\n",
        "\n",
        "# %pip install --upgrade pip==24.1.1\n",
        "# %pip install --upgrade ipykernel==5.5.6\n",
        "# %pip install --upgrade numpy==1.25.2\n",
        "# %pip install --upgrade pandas==2.0.3\n",
        "# %pip install --upgrade tensorflow==2.16.2\n",
        "# %pip install --upgrade keras==3.4.1\n",
        "# %pip install --upgrade scikit-learn==1.5.0\n",
        "# %pip install --upgrade matplotlib==3.9.0\n",
        "# %pip install --upgrade pydot==2.0.0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "b33UJlAEYyQO"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import tensorflow as tf\n",
        "import tensorflow.keras.layers as tfl\n",
        "from tensorflow.keras.applications import VGG19, vgg19\n",
        "import matplotlib.pyplot as plt"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hOl25vcFY5xo"
      },
      "source": [
        "## DATASET OBJECT CREATION"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "avBxf7eoksvg"
      },
      "source": [
        "getting the `pixelated_image_paths` and corresponding `ground_truth_image_paths` as lists"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fm3FEK41rQgz"
      },
      "outputs": [],
      "source": [
        "pixelated_dir = './x/'\n",
        "ground_truth_dir = './y/'\n",
        "\n",
        "pixelated_images = [os.path.join(pixelated_dir, fname) for fname in sorted(os.listdir(pixelated_dir))]\n",
        "ground_truth_images = [os.path.join(ground_truth_dir, fname.split('_')[0] + '.png') for fname in sorted(os.listdir(pixelated_dir))]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y31Wzi91YyQU",
        "outputId": "24146d55-44cc-4923-bcba-ab0e8e9abceb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "7996\n",
            "7996\n",
            "./x/01235o_real.png\n",
            "./y/01235o.png\n"
          ]
        }
      ],
      "source": [
        "# CHECKING\n",
        "\n",
        "print(len(pixelated_images))\n",
        "print(len(ground_truth_images))\n",
        "\n",
        "for file in ground_truth_images:\n",
        "   if not os.path.exists(file):\n",
        "      print(file, 'doesnt exist')\n",
        "\n",
        "print(pixelated_images[1234])\n",
        "print(ground_truth_images[1234])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nWl0YImOksvg"
      },
      "source": [
        "function for loading the image to have values between [0,1]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "V6N5R1ruYyQY"
      },
      "outputs": [],
      "source": [
        "def load_image(file_path):\n",
        "    image = tf.io.read_file(file_path)\n",
        "    image = tf.image.decode_png(image, channels=3)\n",
        "    image = tf.image.convert_image_dtype(image, tf.float32)\n",
        "    return image\n",
        "\n",
        "def load_data(pixelated_path, ground_truth_path):\n",
        "    pixelated_image = load_image(pixelated_path)\n",
        "    ground_truth_image = load_image(ground_truth_path)\n",
        "    return pixelated_image, ground_truth_image"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l0FwQ1-Zksvg"
      },
      "source": [
        "train, test, validation dataset creations"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iGI_wcixYyQZ"
      },
      "outputs": [],
      "source": [
        "dataset = tf.data.Dataset.from_tensor_slices((pixelated_images, ground_truth_images))\n",
        "\n",
        "total_size = dataset.cardinality()\n",
        "dataset = dataset.shuffle(buffer_size=total_size)\n",
        "\n",
        "train_size = int(0.75 * int(total_size))\n",
        "val_size = int(0.10 * int(total_size))\n",
        "\n",
        "train_dataset = dataset.take(train_size)\n",
        "rem = dataset.skip(train_size)\n",
        "val_dataset = rem.take(val_size)\n",
        "test_dataset = rem.skip(val_size)\n",
        "\n",
        "batch_size = 16\n",
        "train_dataset = train_dataset.map(load_data, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .batch(batch_size, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        "val_dataset   =   val_dataset.map(load_data, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .batch(batch_size, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .prefetch(buffer_size=tf.data.AUTOTUNE)\n",
        "test_dataset  =  test_dataset.map(load_data, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .batch(batch_size, num_parallel_calls=tf.data.AUTOTUNE)\\\n",
        "                             .prefetch(buffer_size=tf.data.AUTOTUNE)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CvHzTti4Y5xp"
      },
      "source": [
        "## CREATE MODEL\n",
        "\n",
        "Full model architecture (for a visual representation see output)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rhAWxODCWew8",
        "outputId": "1df6ae61-2db3-419a-8ce2-00a3d3318dcf"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">75</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalization │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_1  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">656</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\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_1        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_2  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,336</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_3  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">8,768</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_3        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_4  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">33,920</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_4        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_5  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │    <span style=\"color: #00af00; text-decoration-color: #00af00\">133,376</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_5        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_6  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │    <span style=\"color: #00af00; text-decoration-color: #00af00\">528,896</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: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,096</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_6        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1024</span>)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_7  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │    <span style=\"color: #00af00; text-decoration-color: #00af00\">533,504</span> │ up_sampling2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_7        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │ activation_5[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_1     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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\">UpSampling2D</span>)      │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_8  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │    <span style=\"color: #00af00; text-decoration-color: #00af00\">135,680</span> │ up_sampling2d_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_8        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │            │ activation_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_2     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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\">UpSampling2D</span>)      │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_9  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">35,072</span> │ up_sampling2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_9        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_9[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "│                     │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │ activation_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_3     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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\">UpSampling2D</span>)      │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_10 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,344</span> │ up_sampling2d_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_10       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_10[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
              "│                     │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)         │            │ activation_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_4     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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\">UpSampling2D</span>)      │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_11 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,624</span> │ up_sampling2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_11       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_11[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
              "│                     │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │ activation_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_12 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">800</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\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_12       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_12[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
              "│                     │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)         │            │ activation[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_13 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">192</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\">SeparableConv2D</span>)   │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">12</span> │ separable_conv2d… │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_13       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</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: #00d7ff; text-decoration-color: #00d7ff\">None</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ activation_13[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n",
              "│                     │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │ input_layer[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_14       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</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>)        │ <span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</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;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ -                 │\n",
              "│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │         \u001b[38;5;34m75\u001b[0m │ input_layer[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │         \u001b[38;5;34m64\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_1  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m656\u001b[0m │ activation[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m128\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_1        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_2  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m2,336\u001b[0m │ activation_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m256\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_2        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_3  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m8,768\u001b[0m │ activation_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m512\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_3        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_4  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │     \u001b[38;5;34m33,920\u001b[0m │ activation_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m1,024\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_4        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_5  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │    \u001b[38;5;34m133,376\u001b[0m │ activation_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m2,048\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_5        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_6  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │    \u001b[38;5;34m528,896\u001b[0m │ activation_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1024\u001b[0m)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m4,096\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1024\u001b[0m)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_6        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1024\u001b[0m)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1024\u001b[0m)       │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_7  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │    \u001b[38;5;34m533,504\u001b[0m │ up_sampling2d[\u001b[38;5;34m0\u001b[0m]… │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m2,048\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_7        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add (\u001b[38;5;33mAdd\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │ activation_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_1     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\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;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_8  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │    \u001b[38;5;34m135,680\u001b[0m │ up_sampling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m1,024\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_8        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\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;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │ activation_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_2     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\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;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_9  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │     \u001b[38;5;34m35,072\u001b[0m │ up_sampling2d_2[\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m512\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_9        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add_2 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │ activation_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_3     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\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;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_10 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m9,344\u001b[0m │ up_sampling2d_3[\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m256\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_10       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add_3 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_10[\u001b[38;5;34m0\u001b[0m]… │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │ activation_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ up_sampling2d_4     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\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;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_11 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │      \u001b[38;5;34m2,624\u001b[0m │ up_sampling2d_4[\u001b[38;5;34m…\u001b[0m │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m128\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_11       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add_4 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_11[\u001b[38;5;34m0\u001b[0m]… │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │ activation_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_12 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m800\u001b[0m │ add_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │         \u001b[38;5;34m64\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_12       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add_5 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_12[\u001b[38;5;34m0\u001b[0m]… │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m)         │            │ activation[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ separable_conv2d_13 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │        \u001b[38;5;34m192\u001b[0m │ add_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]       │\n",
              "│ (\u001b[38;5;33mSeparableConv2D\u001b[0m)   │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │         \u001b[38;5;34m12\u001b[0m │ separable_conv2d… │\n",
              "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_13       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n",
              "│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ add_6 (\u001b[38;5;33mAdd\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ activation_13[\u001b[38;5;34m0\u001b[0m]… │\n",
              "│                     │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │ input_layer[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
              "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
              "│ activation_14       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;45mNone\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)        │ \u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\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\">1,437,415</span> (5.48 MB)\n",
              "</pre>\n"
            ],
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m1,437,415\u001b[0m (5.48 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\">1,431,329</span> (5.46 MB)\n",
              "</pre>\n"
            ],
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m1,431,329\u001b[0m (5.46 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\">6,086</span> (23.77 KB)\n",
              "</pre>\n"
            ],
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m6,086\u001b[0m (23.77 KB)\n"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "def build_net(input_shape=(None, None, 3)):\n",
        "\n",
        "  inputs = tfl.Input(input_shape)\n",
        "\n",
        "  def translate_encoder(x, filters, kernel_size=3, activation='relu'):\n",
        "    x = tfl.SeparableConv2D(filters=filters, kernel_size=kernel_size, strides=(1, 1), padding='same', activation=None, use_bias=False)(x)\n",
        "    x = tfl.BatchNormalization(epsilon=1e-3, momentum = 0.99)(x)\n",
        "    return tfl.Activation(activation=activation)(x)\n",
        "\n",
        "  def encoder_block(x, filters, kernel_size=3, strides=2, padding='same', activation='relu'):\n",
        "    x = tfl.SeparableConv2D(filters=filters, kernel_size=kernel_size, strides=strides, padding=padding, activation=None, use_bias=False)(x)\n",
        "    x = tfl.BatchNormalization(epsilon=1e-3, momentum = 0.99)(x)\n",
        "    return tfl.Activation(activation=activation)(x)\n",
        "\n",
        "  def decoder_block(x, filters, skip, kernel_size=3, strides=2, padding='same', activation='relu'):\n",
        "    x = tfl.UpSampling2D(size=(strides, strides), interpolation='nearest')(x)\n",
        "    x = tfl.SeparableConv2D(filters=filters, kernel_size=kernel_size, padding=padding, activation=None, use_bias=False)(x)\n",
        "    x = tfl.BatchNormalization(epsilon=1e-3, momentum = 0.99)(x)\n",
        "    x = tfl.Activation(activation=activation)(x)\n",
        "    return tfl.Add()([x, skip])\n",
        "\n",
        "  def translate_decoder(x, filters, skip, kernel_size=3, activation='relu'):\n",
        "    x = tfl.SeparableConv2D(filters=filters, kernel_size=kernel_size, strides=(1, 1), padding='same', activation=None, use_bias=False)(x)\n",
        "    x = tfl.BatchNormalization(epsilon=1e-3, momentum = 0.99)(x)\n",
        "    x = tfl.Activation(activation=activation)(x)\n",
        "    return tfl.Add()([x, skip])\n",
        "\n",
        "  t0 = translate_encoder(inputs, 16)\n",
        "  t1 = translate_encoder(t0, 32)\n",
        "\n",
        "  e1 = encoder_block(t1, 64)\n",
        "  e2 = encoder_block(e1, 128)\n",
        "  e3 = encoder_block(e2, 256)\n",
        "  e4 = encoder_block(e3, 512)\n",
        "  e5 = encoder_block(e4, 1024)\n",
        "\n",
        "  d5 = decoder_block(e5, 512, e4)\n",
        "  d4 = decoder_block(d5, 256, e3)\n",
        "  d3 = decoder_block(d4, 128, e2)\n",
        "  d2 = decoder_block(d3,  64, e1)\n",
        "  d1 = decoder_block(d2,  32, t1)\n",
        "\n",
        "  t2 = translate_decoder(d1, 16, t0)\n",
        "  t3 = translate_decoder(t2,  3, inputs, activation='linear')\n",
        "  outputs = tfl.Activation(activation='sigmoid')(t3)\n",
        "\n",
        "  return tf.keras.Model(inputs, outputs)\n",
        "\n",
        "model = build_net()\n",
        "\n",
        "model.summary()\n",
        "# tf.keras.utils.plot_model(model, to_file='./depixelator_004_2.png', show_layer_activations=True, show_layer_names=True, show_shapes=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ExyaPc05Y5xq"
      },
      "source": [
        "## CUSTOM FUNCTIONS"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "faFBr9Ua6LFI"
      },
      "source": [
        "loading the VGG19 Model for perceptual loss function"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qxMIuIWDYyQc",
        "outputId": "24311545-b221-4192-e5cb-f3dc575dec73"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg19/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "\u001b[1m80134624/80134624\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n"
          ]
        }
      ],
      "source": [
        "vgg = VGG19(include_top=False, weights='imagenet', input_shape=(None, None, 3))\n",
        "layer_names = ['block3_conv4', 'block4_conv1', 'block4_conv2']\n",
        "outputs = [vgg.get_layer(name).output for name in layer_names]\n",
        "vgg_model = tf.keras.Model(inputs=vgg.input, outputs=outputs)\n",
        "vgg_model.trainable = False"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rzpG7oXuu8r3"
      },
      "source": [
        "custom loss and metric functions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZxzKn7QnYyQd"
      },
      "outputs": [],
      "source": [
        "@tf.keras.utils.register_keras_serializable()\n",
        "def compute_perceptual_loss(y_true, y_pred):\n",
        "    y_true = vgg19.preprocess_input(y_true * 255.0)\n",
        "    y_pred = vgg19.preprocess_input(y_pred * 255.0)\n",
        "\n",
        "    true_features = vgg_model(y_true)\n",
        "    pred_features = vgg_model(y_pred)\n",
        "\n",
        "    loss = 0\n",
        "    for t, p in zip(true_features, pred_features):\n",
        "        loss += tf.reduce_mean(tf.abs(t - p))\n",
        "\n",
        "    return loss\n",
        "\n",
        "@tf.keras.utils.register_keras_serializable()\n",
        "def compute_psnr(y_true, y_pred):\n",
        "    return tf.image.psnr(y_true, y_pred, max_val=1.0)\n",
        "\n",
        "@tf.keras.utils.register_keras_serializable()\n",
        "def compute_ssim(y_true, y_pred):\n",
        "    return tf.image.ssim(y_true, y_pred, max_val=1.0)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6w1GJ-5l6cH8"
      },
      "source": [
        "compiling the model with custom loss and metrics"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vllNUkoEYyQe"
      },
      "outputs": [],
      "source": [
        "model.compile(\n",
        "    optimizer=tf.keras.optimizers.RMSprop(learning_rate=1e-3),\n",
        "    loss=compute_perceptual_loss,\n",
        "    metrics=[tf.keras.metrics.MeanSquaredError(), compute_psnr, compute_ssim]\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CX4T5XhK633a"
      },
      "source": [
        "## TRAINING"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FJHfp76Cksvi"
      },
      "source": [
        "initial testing"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K2AGFBu2YyQf",
        "outputId": "ed73fc65-71ab-433f-8184-ce94e2ae4296"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m147s\u001b[0m 2s/step - compute_psnr: 11.5335 - compute_ssim: 0.4884 - loss: 896.2933 - mean_squared_error: 0.0779\n"
          ]
        }
      ],
      "source": [
        "perceptual_loss0, pixel_loss0, psnr0, ssim0  = model.evaluate(val_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q3WbOxs2YyQg",
        "outputId": "facc7e11-e90d-4d6e-a0fa-3381a775a3ee"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "initial perceptual loss: 903.24\n",
            "initial pixel loss: 0.08\n",
            "initial PSNR: 11.45\n",
            "initial SSIM: 0.49\n"
          ]
        }
      ],
      "source": [
        "print(\"initial perceptual loss: {:.2f}\".format(perceptual_loss0))\n",
        "print(\"initial pixel loss: {:.2f}\".format(pixel_loss0))\n",
        "print(\"initial PSNR: {:.2f}\".format(psnr0))\n",
        "print(\"initial SSIM: {:.2f}\".format(ssim0))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WgXCmFOMvMiV"
      },
      "source": [
        "required callbacks"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d-7VMbD4YyQh"
      },
      "outputs": [],
      "source": [
        "early_stopping = tf.keras.callbacks.EarlyStopping(\n",
        "    monitor=\"val_loss\",\n",
        "    min_delta=1,\n",
        "    patience=3,\n",
        "    verbose=1,\n",
        "    mode=\"auto\",\n",
        "    baseline=None,\n",
        "    restore_best_weights=True,\n",
        "    start_from_epoch=0,\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "202IyNcSvzMk"
      },
      "source": [
        "### TRAINING (1-12 epochs)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lRM2cXmFYyQi",
        "outputId": "393bb07a-d321-43ac-ab88-e07ee0a080c1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m705s\u001b[0m 2s/step - compute_psnr: 15.3638 - compute_ssim: 0.6081 - loss: 742.3759 - mean_squared_error: 0.0374 - val_compute_psnr: 17.2536 - val_compute_ssim: 0.6655 - val_loss: 678.9356 - val_mean_squared_error: 0.0198\n",
            "Epoch 2/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m525s\u001b[0m 1s/step - compute_psnr: 19.3434 - compute_ssim: 0.7103 - loss: 555.5149 - mean_squared_error: 0.0129 - val_compute_psnr: 19.9661 - val_compute_ssim: 0.7240 - val_loss: 538.2248 - val_mean_squared_error: 0.0107\n",
            "Epoch 3/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m571s\u001b[0m 1s/step - compute_psnr: 21.3720 - compute_ssim: 0.7221 - loss: 533.9988 - mean_squared_error: 0.0088 - val_compute_psnr: 22.3604 - val_compute_ssim: 0.7437 - val_loss: 503.8843 - val_mean_squared_error: 0.0067\n",
            "Epoch 4/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m535s\u001b[0m 1s/step - compute_psnr: 22.9596 - compute_ssim: 0.7355 - loss: 511.4113 - mean_squared_error: 0.0064 - val_compute_psnr: 22.7788 - val_compute_ssim: 0.7428 - val_loss: 500.0814 - val_mean_squared_error: 0.0064\n",
            "Epoch 5/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m535s\u001b[0m 1s/step - compute_psnr: 23.8155 - compute_ssim: 0.7429 - loss: 483.8384 - mean_squared_error: 0.0055 - val_compute_psnr: 25.4694 - val_compute_ssim: 0.7481 - val_loss: 463.1321 - val_mean_squared_error: 0.0042\n",
            "Epoch 6/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m526s\u001b[0m 1s/step - compute_psnr: 24.6499 - compute_ssim: 0.7489 - loss: 466.5933 - mean_squared_error: 0.0048 - val_compute_psnr: 25.5532 - val_compute_ssim: 0.7491 - val_loss: 444.9970 - val_mean_squared_error: 0.0042\n",
            "Epoch 7/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m534s\u001b[0m 1s/step - compute_psnr: 25.0509 - compute_ssim: 0.7553 - loss: 452.5810 - mean_squared_error: 0.0045 - val_compute_psnr: 25.6272 - val_compute_ssim: 0.7497 - val_loss: 448.3964 - val_mean_squared_error: 0.0041\n",
            "Epoch 8/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m535s\u001b[0m 1s/step - compute_psnr: 25.5013 - compute_ssim: 0.7556 - loss: 445.1950 - mean_squared_error: 0.0042 - val_compute_psnr: 26.6025 - val_compute_ssim: 0.7594 - val_loss: 445.2373 - val_mean_squared_error: 0.0037\n",
            "Epoch 9/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m535s\u001b[0m 1s/step - compute_psnr: 25.4713 - compute_ssim: 0.7550 - loss: 441.0043 - mean_squared_error: 0.0042 - val_compute_psnr: 26.9061 - val_compute_ssim: 0.7567 - val_loss: 416.4450 - val_mean_squared_error: 0.0035\n",
            "Epoch 10/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m562s\u001b[0m 1s/step - compute_psnr: 25.6139 - compute_ssim: 0.7538 - loss: 437.8643 - mean_squared_error: 0.0041 - val_compute_psnr: 27.6248 - val_compute_ssim: 0.7838 - val_loss: 419.3857 - val_mean_squared_error: 0.0032\n",
            "Epoch 11/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m526s\u001b[0m 1s/step - compute_psnr: 25.8316 - compute_ssim: 0.7608 - loss: 428.1790 - mean_squared_error: 0.0040 - val_compute_psnr: 26.9824 - val_compute_ssim: 0.7539 - val_loss: 429.7457 - val_mean_squared_error: 0.0036\n",
            "Epoch 12/12\n",
            "\u001b[1m375/375\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m562s\u001b[0m 1s/step - compute_psnr: 25.8767 - compute_ssim: 0.7632 - loss: 424.9778 - mean_squared_error: 0.0040 - val_compute_psnr: 27.2648 - val_compute_ssim: 0.7645 - val_loss: 421.2108 - val_mean_squared_error: 0.0033\n",
            "Epoch 12: early stopping\n",
            "Restoring model weights from the end of the best epoch: 9.\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
          ]
        }
      ],
      "source": [
        "history = model.fit(train_dataset,\n",
        "                    epochs=12,\n",
        "                    validation_data=val_dataset,\n",
        "                    callbacks=[early_stopping])\n",
        "\n",
        "model.save('/content/drive/MyDrive/DEPIXELATOR_004/depixelator_004_2.keras')\n",
        "model.save('/content/drive/MyDrive/DEPIXELATOR_004/depixelator_004_2.h5')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lxgz7qn0v1vI"
      },
      "source": [
        "### RESULTS"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kfheZ_vBYyQj"
      },
      "outputs": [],
      "source": [
        "def history_results(history, par1='loss', par2='val_loss', title='Perceptual Loss'):\n",
        "    plt.plot(history.history[par1])\n",
        "    plt.plot(history.history[par2])\n",
        "    plt.title(title)\n",
        "    plt.ylabel(par1)\n",
        "    plt.xlabel('epoch')\n",
        "    plt.legend(['train', 'val'], loc='upper left')\n",
        "    plt.grid()\n",
        "    plt.show()\n",
        "    return"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "7Ig0tJtKYyQj",
        "outputId": "853324d6-bff4-4bd8-c976-8b1c9b835d8e"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "history_results(history)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "0efSsBdJohsG",
        "outputId": "a4fe965e-9c39-4b0f-c0a7-d462f0654ff4"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "history_results(history, 'mean_squared_error', 'val_mean_squared_error', 'Pixel Loss')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "XxruBNK2YyQj",
        "outputId": "be9995c1-9594-4556-bb85-485b3071a00a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "history_results(history, 'compute_psnr', 'val_compute_psnr', 'PSNR')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "X72BH_DQYyQk",
        "outputId": "ee90faa5-88b7-4eb0-9772-be046b6e9a2c"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "history_results(history, 'compute_ssim', 'val_compute_ssim', 'SSIM')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kNdsIBdN1LDy",
        "outputId": "c5d46ab7-4722-48bf-a316-e6f43a4ea944"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[1m75/75\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m48s\u001b[0m 639ms/step - compute_psnr: 27.6186 - compute_ssim: 0.7796 - loss: 386.0658 - mean_squared_error: 0.0032\n"
          ]
        }
      ],
      "source": [
        "perceptual_loss_test, pixel_loss_test, psnr_test, ssim_test = model.evaluate(test_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QyQO68anohsI",
        "outputId": "4893ad70-b9b0-4e17-90dc-42cd7c54c613"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "perceptual loss: 399.97\n",
            "pixel loss: 0.00\n",
            "PSNR: 27.25\n",
            "SSIM: 0.77\n"
          ]
        }
      ],
      "source": [
        "print(\"perceptual loss: {:.2f}\".format(perceptual_loss_test))\n",
        "print(\"pixel loss: {:.2f}\".format(pixel_loss_test))\n",
        "print(\"PSNR: {:.2f}\".format(psnr_test))\n",
        "print(\"SSIM: {:.2f}\".format(ssim_test))"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.12.4"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}