{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KTxR9kIkXGpj"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import cv2\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tqdm import tqdm\n",
        "import os\n",
        "from sklearn.utils import shuffle\n",
        "from sklearn.model_selection import train_test_split\n",
        "from tensorflow.keras.applications import EfficientNetB0, InceptionResNetV2\n",
        "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, TensorBoard, ModelCheckpoint\n",
        "from sklearn.metrics import classification_report,confusion_matrix\n",
        "import ipywidgets as widgets\n",
        "import io\n",
        "from PIL import Image\n",
        "from IPython.display import display,clear_output\n",
        "from warnings import filterwarnings\n",
        "for dirname, _, filenames in os.walk('/kaggle/input'):\n",
        "    for filename in filenames:\n",
        "        print(os.path.join(dirname, filename))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "m9lTRS0lXITV"
      },
      "outputs": [],
      "source": [
        "colors_dark = [\"#1F1F1F\", \"#313131\", '#636363', '#AEAEAE', '#DADADA']\n",
        "colors_red = [\"#331313\", \"#582626\", '#9E1717', '#D35151', '#E9B4B4']\n",
        "colors_green = ['#01411C','#4B6F44','#4F7942','#74C365','#D0F0C0']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dHBaxAziXKl3",
        "outputId": "4d8746f7-f949-479e-854b-64b15acb3aea"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.10/dist-packages (1.5.16)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.10/dist-packages (from kaggle) (1.16.0)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from kaggle) (2023.11.17)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.8.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.31.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from kaggle) (4.66.1)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle) (8.0.1)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.0.7)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle) (6.1.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from bleach->kaggle) (0.5.1)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (3.6)\n"
          ]
        }
      ],
      "source": [
        "pip install kaggle"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/",
          "height": 56
        },
        "id": "psbfAjEDXNIH",
        "outputId": "656ad1b2-b2df-4aeb-e2b8-2bfe487283df"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Upload kaggle.json here\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-3592ce4c-cb79-4be2-9ccf-4dfa4243406d\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-3592ce4c-cb79-4be2-9ccf-4dfa4243406d\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "ename": "TypeError",
          "evalue": "ignored",
          "output_type": "error",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-1-fb86819d0be7>\u001b[0m in \u001b[0;36m<cell line: 3>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m   \u001b[0;32mfrom\u001b[0m \u001b[0mgoogle\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolab\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mfiles\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m   \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Upload kaggle.json here\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m   \u001b[0mfiles\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misfile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'IMDB Dataset.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/google/colab/files.py\u001b[0m in \u001b[0;36mupload\u001b[0;34m()\u001b[0m\n\u001b[1;32m     67\u001b[0m   \"\"\"\n\u001b[1;32m     68\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 69\u001b[0;31m   \u001b[0muploaded_files\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_upload_files\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmultiple\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     70\u001b[0m   \u001b[0;31m# Mapping from original filename to filename as saved locally.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     71\u001b[0m   \u001b[0mlocal_filenames\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/google/colab/files.py\u001b[0m in \u001b[0;36m_upload_files\u001b[0;34m(multiple)\u001b[0m\n\u001b[1;32m    161\u001b[0m   \u001b[0mfiles\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_collections\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdefaultdict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbytes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    162\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 163\u001b[0;31m   \u001b[0;32mwhile\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'action'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;34m'complete'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    164\u001b[0m     result = _output.eval_js(\n\u001b[1;32m    165\u001b[0m         'google.colab._files._uploadFilesContinue(\"{output_id}\")'.format(\n",
            "\u001b[0;31mTypeError\u001b[0m: 'NoneType' object is not subscriptable"
          ]
        }
      ],
      "source": [
        "! pip install -q kaggle\n",
        "import os\n",
        "if not os.path.isfile(os.path.expanduser('~/.kaggle/kaggle.json')):\n",
        "  from google.colab import files\n",
        "  print(\"Upload kaggle.json here\")\n",
        "  files.upload()\n",
        "\n",
        "if not os.path.isfile('IMDB Dataset.csv'):\n",
        "  !mkdir ~/.kaggle\n",
        "  !mv ./kaggle.json ~/.kaggle/\n",
        "  !chmod 600 ~/.kaggle/kaggle.json\n",
        "\n",
        "  dataset_name = 'sartajbhuvaji/brain-tumor-classification-mri'\n",
        "  zip_name = dataset_name.split('/')[-1]\n",
        "\n",
        "  !kaggle datasets download -d {dataset_name}\n",
        "  !unzip -q ./{zip_name}.zip -d .\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-8__nhUdXQeR",
        "outputId": "b87e2b99-0c27-46ac-c6a1-7d305115ec7e"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 826/826 [00:35<00:00, 23.48it/s]\n",
            "100%|██████████| 395/395 [00:10<00:00, 36.14it/s] \n",
            "100%|██████████| 822/822 [00:15<00:00, 53.90it/s] \n",
            "100%|██████████| 827/827 [00:15<00:00, 52.82it/s] \n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import cv2\n",
        "import numpy as np\n",
        "from tqdm import tqdm\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "labels = ['glioma_tumor', 'no_tumor', 'meningioma_tumor', 'pituitary_tumor']\n",
        "image_size = 150\n",
        "target_per_class = 1500\n",
        "\n",
        "X_train = []\n",
        "y_train = []\n",
        "\n",
        "# Define an ImageDataGenerator with data augmentation parameters\n",
        "datagen = ImageDataGenerator(\n",
        "    rotation_range=40,\n",
        "    width_shift_range=0.2,\n",
        "    height_shift_range=0.2,\n",
        "    shear_range=0.2,\n",
        "    zoom_range=0.2,\n",
        "    horizontal_flip=True,\n",
        "    fill_mode='nearest'\n",
        ")\n",
        "\n",
        "# Loop through each class\n",
        "for i in labels:\n",
        "    folderPath = os.path.join('/content', 'Training', i)\n",
        "\n",
        "    # Count the current number of images in the class\n",
        "    current_count = 0\n",
        "\n",
        "    # Loop through the images in the class\n",
        "    for j in tqdm(os.listdir(folderPath)):\n",
        "        img = cv2.imread(os.path.join(folderPath, j))\n",
        "        img = cv2.resize(img, (image_size, image_size))\n",
        "\n",
        "        # Apply data augmentation and add augmented images\n",
        "        for _ in range(5):  # Apply augmentation 5 times per image to reach the target\n",
        "            augmented_img = datagen.random_transform(img)\n",
        "            X_train.append(augmented_img)\n",
        "            y_train.append(i)\n",
        "            current_count += 1\n",
        "\n",
        "            # Stop when the target number of images is reached for this class\n",
        "            if current_count >= target_per_class:\n",
        "                break\n",
        "\n",
        "    # Fill the class with additional copies of existing images if needed\n",
        "    while current_count < target_per_class:\n",
        "        for j in range(len(X_train)):\n",
        "            if y_train[j] == i:\n",
        "                img = X_train[j]\n",
        "                X_train.append(img)\n",
        "                y_train.append(i)\n",
        "                current_count += 1\n",
        "                if current_count >= target_per_class:\n",
        "                    break\n",
        "\n",
        "X_train = np.array(X_train)\n",
        "y_train = np.array(y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3iV2dFeKXTUi",
        "outputId": "5046ba71-dc35-4d72-9b9e-a6616ff58b78"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "7670"
            ]
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(X_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HoKLFhs9XWS0",
        "outputId": "040f1e64-c65a-4ac0-806a-9cb65c6adc8f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Class 'glioma_tumor' has 2026 images.\n",
            "Class 'no_tumor' has 1595 images.\n",
            "Class 'meningioma_tumor' has 2022 images.\n",
            "Class 'pituitary_tumor' has 2027 images.\n"
          ]
        }
      ],
      "source": [
        "for label in labels:\n",
        "    count = np.sum(y_train == label)\n",
        "    print(f\"Class '{label}' has {count} images.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "-YeLtFpWXYtK",
        "outputId": "0c6e2581-7ccb-465c-f37f-3acbfbda565a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2000x2000 with 4 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "k=0\n",
        "fig, ax = plt.subplots(1,4,figsize=(20,20))\n",
        "fig.text(s='Sample Image From Each Label',size=18,fontweight='bold',\n",
        "             fontname='monospace',color=colors_dark[1],y=0.62,x=0.4,alpha=0.8)\n",
        "for i in labels:\n",
        "    j=0\n",
        "    while True :\n",
        "        if y_train[j]==i:\n",
        "            ax[k].imshow(X_train[j])\n",
        "            ax[k].set_title(y_train[j])\n",
        "            ax[k].axis('off')\n",
        "            k+=1\n",
        "            break\n",
        "        j+=1\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Maro8eNeXau9"
      },
      "outputs": [],
      "source": [
        "X_train, y_train = shuffle(X_train,y_train, random_state=101)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FMs8HrJKXdKc",
        "outputId": "c0258662-ab96-4fc3-f3f7-848950a1c71b"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(7670, 150, 150, 3)"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "X_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sQe8jMdEXfzQ"
      },
      "outputs": [],
      "source": [
        "X_train,X_test,y_train,y_test = train_test_split(X_train,y_train, test_size=0.2,random_state=101)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E0VDtV9kXhsf"
      },
      "outputs": [],
      "source": [
        "y_train_new = []\n",
        "for i in y_train:\n",
        "    y_train_new.append(labels.index(i))\n",
        "y_train = y_train_new\n",
        "y_train = tf.keras.utils.to_categorical(y_train)\n",
        "\n",
        "\n",
        "y_test_new = []\n",
        "for i in y_test:\n",
        "    y_test_new.append(labels.index(i))\n",
        "y_test = y_test_new\n",
        "y_test = tf.keras.utils.to_categorical(y_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ikBO8BnAXj4Y",
        "outputId": "297fba81-933b-41fc-f742-6ff8880f73ce"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/densenet/densenet121_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "29084464/29084464 [==============================] - 0s 0us/step\n"
          ]
        }
      ],
      "source": [
        "from tensorflow.keras.applications import DenseNet121\n",
        "densenet = DenseNet121(weights='imagenet', include_top=False, input_shape=(image_size, image_size, 3))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wHVl0sx_ZPqe"
      },
      "outputs": [],
      "source": [
        "model = densenet.output\n",
        "model = tf.keras.layers.GlobalAveragePooling2D()(model)\n",
        "model = tf.keras.layers.Dropout(rate=0.5)(model)\n",
        "model = tf.keras.layers.Dense(4,activation='softmax')(model)\n",
        "model = tf.keras.models.Model(inputs=densenet.input, outputs = model)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "83PuA0v9ZQ4x",
        "outputId": "c53b6b8d-2279-4cd7-9976-54bddc8e62a1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                Output Shape                 Param #   Connected to                  \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)        [(None, 150, 150, 3)]        0         []                            \n",
            "                                                                                                  \n",
            " zero_padding2d (ZeroPaddin  (None, 156, 156, 3)          0         ['input_1[0][0]']             \n",
            " g2D)                                                                                             \n",
            "                                                                                                  \n",
            " conv1/conv (Conv2D)         (None, 75, 75, 64)           9408      ['zero_padding2d[0][0]']      \n",
            "                                                                                                  \n",
            " conv1/bn (BatchNormalizati  (None, 75, 75, 64)           256       ['conv1/conv[0][0]']          \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " conv1/relu (Activation)     (None, 75, 75, 64)           0         ['conv1/bn[0][0]']            \n",
            "                                                                                                  \n",
            " zero_padding2d_1 (ZeroPadd  (None, 77, 77, 64)           0         ['conv1/relu[0][0]']          \n",
            " ing2D)                                                                                           \n",
            "                                                                                                  \n",
            " pool1 (MaxPooling2D)        (None, 38, 38, 64)           0         ['zero_padding2d_1[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_0_bn (BatchNo  (None, 38, 38, 64)           256       ['pool1[0][0]']               \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block1_0_relu (Activ  (None, 38, 38, 64)           0         ['conv2_block1_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block1_1_conv (Conv2  (None, 38, 38, 128)          8192      ['conv2_block1_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block1_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block1_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block1_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block1_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block1_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_concat (Conca  (None, 38, 38, 96)           0         ['pool1[0][0]',               \n",
            " tenate)                                                             'conv2_block1_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv2_block2_0_bn (BatchNo  (None, 38, 38, 96)           384       ['conv2_block1_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block2_0_relu (Activ  (None, 38, 38, 96)           0         ['conv2_block2_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block2_1_conv (Conv2  (None, 38, 38, 128)          12288     ['conv2_block2_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block2_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block2_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block2_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block2_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block2_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_concat (Conca  (None, 38, 38, 128)          0         ['conv2_block1_concat[0][0]', \n",
            " tenate)                                                             'conv2_block2_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv2_block3_0_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block2_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block3_0_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block3_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block3_1_conv (Conv2  (None, 38, 38, 128)          16384     ['conv2_block3_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block3_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block3_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block3_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block3_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block3_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_concat (Conca  (None, 38, 38, 160)          0         ['conv2_block2_concat[0][0]', \n",
            " tenate)                                                             'conv2_block3_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv2_block4_0_bn (BatchNo  (None, 38, 38, 160)          640       ['conv2_block3_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block4_0_relu (Activ  (None, 38, 38, 160)          0         ['conv2_block4_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block4_1_conv (Conv2  (None, 38, 38, 128)          20480     ['conv2_block4_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block4_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block4_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block4_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block4_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block4_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block4_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block4_concat (Conca  (None, 38, 38, 192)          0         ['conv2_block3_concat[0][0]', \n",
            " tenate)                                                             'conv2_block4_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv2_block5_0_bn (BatchNo  (None, 38, 38, 192)          768       ['conv2_block4_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block5_0_relu (Activ  (None, 38, 38, 192)          0         ['conv2_block5_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block5_1_conv (Conv2  (None, 38, 38, 128)          24576     ['conv2_block5_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block5_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block5_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block5_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block5_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block5_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block5_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block5_concat (Conca  (None, 38, 38, 224)          0         ['conv2_block4_concat[0][0]', \n",
            " tenate)                                                             'conv2_block5_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv2_block6_0_bn (BatchNo  (None, 38, 38, 224)          896       ['conv2_block5_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block6_0_relu (Activ  (None, 38, 38, 224)          0         ['conv2_block6_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block6_1_conv (Conv2  (None, 38, 38, 128)          28672     ['conv2_block6_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block6_1_bn (BatchNo  (None, 38, 38, 128)          512       ['conv2_block6_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv2_block6_1_relu (Activ  (None, 38, 38, 128)          0         ['conv2_block6_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv2_block6_2_conv (Conv2  (None, 38, 38, 32)           36864     ['conv2_block6_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block6_concat (Conca  (None, 38, 38, 256)          0         ['conv2_block5_concat[0][0]', \n",
            " tenate)                                                             'conv2_block6_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " pool2_bn (BatchNormalizati  (None, 38, 38, 256)          1024      ['conv2_block6_concat[0][0]'] \n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " pool2_relu (Activation)     (None, 38, 38, 256)          0         ['pool2_bn[0][0]']            \n",
            "                                                                                                  \n",
            " pool2_conv (Conv2D)         (None, 38, 38, 128)          32768     ['pool2_relu[0][0]']          \n",
            "                                                                                                  \n",
            " pool2_pool (AveragePooling  (None, 19, 19, 128)          0         ['pool2_conv[0][0]']          \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block1_0_bn (BatchNo  (None, 19, 19, 128)          512       ['pool2_pool[0][0]']          \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block1_0_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block1_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block1_1_conv (Conv2  (None, 19, 19, 128)          16384     ['conv3_block1_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block1_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block1_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block1_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block1_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block1_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_concat (Conca  (None, 19, 19, 160)          0         ['pool2_pool[0][0]',          \n",
            " tenate)                                                             'conv3_block1_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block2_0_bn (BatchNo  (None, 19, 19, 160)          640       ['conv3_block1_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block2_0_relu (Activ  (None, 19, 19, 160)          0         ['conv3_block2_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block2_1_conv (Conv2  (None, 19, 19, 128)          20480     ['conv3_block2_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block2_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block2_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block2_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block2_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block2_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_concat (Conca  (None, 19, 19, 192)          0         ['conv3_block1_concat[0][0]', \n",
            " tenate)                                                             'conv3_block2_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block3_0_bn (BatchNo  (None, 19, 19, 192)          768       ['conv3_block2_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block3_0_relu (Activ  (None, 19, 19, 192)          0         ['conv3_block3_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block3_1_conv (Conv2  (None, 19, 19, 128)          24576     ['conv3_block3_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block3_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block3_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block3_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block3_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block3_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_concat (Conca  (None, 19, 19, 224)          0         ['conv3_block2_concat[0][0]', \n",
            " tenate)                                                             'conv3_block3_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block4_0_bn (BatchNo  (None, 19, 19, 224)          896       ['conv3_block3_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block4_0_relu (Activ  (None, 19, 19, 224)          0         ['conv3_block4_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block4_1_conv (Conv2  (None, 19, 19, 128)          28672     ['conv3_block4_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block4_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block4_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block4_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block4_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block4_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_concat (Conca  (None, 19, 19, 256)          0         ['conv3_block3_concat[0][0]', \n",
            " tenate)                                                             'conv3_block4_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block5_0_bn (BatchNo  (None, 19, 19, 256)          1024      ['conv3_block4_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block5_0_relu (Activ  (None, 19, 19, 256)          0         ['conv3_block5_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block5_1_conv (Conv2  (None, 19, 19, 128)          32768     ['conv3_block5_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block5_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block5_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block5_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block5_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block5_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block5_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block5_concat (Conca  (None, 19, 19, 288)          0         ['conv3_block4_concat[0][0]', \n",
            " tenate)                                                             'conv3_block5_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block6_0_bn (BatchNo  (None, 19, 19, 288)          1152      ['conv3_block5_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block6_0_relu (Activ  (None, 19, 19, 288)          0         ['conv3_block6_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block6_1_conv (Conv2  (None, 19, 19, 128)          36864     ['conv3_block6_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block6_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block6_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block6_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block6_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block6_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block6_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block6_concat (Conca  (None, 19, 19, 320)          0         ['conv3_block5_concat[0][0]', \n",
            " tenate)                                                             'conv3_block6_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block7_0_bn (BatchNo  (None, 19, 19, 320)          1280      ['conv3_block6_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block7_0_relu (Activ  (None, 19, 19, 320)          0         ['conv3_block7_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block7_1_conv (Conv2  (None, 19, 19, 128)          40960     ['conv3_block7_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block7_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block7_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block7_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block7_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block7_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block7_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block7_concat (Conca  (None, 19, 19, 352)          0         ['conv3_block6_concat[0][0]', \n",
            " tenate)                                                             'conv3_block7_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block8_0_bn (BatchNo  (None, 19, 19, 352)          1408      ['conv3_block7_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block8_0_relu (Activ  (None, 19, 19, 352)          0         ['conv3_block8_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block8_1_conv (Conv2  (None, 19, 19, 128)          45056     ['conv3_block8_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block8_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block8_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block8_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block8_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block8_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block8_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block8_concat (Conca  (None, 19, 19, 384)          0         ['conv3_block7_concat[0][0]', \n",
            " tenate)                                                             'conv3_block8_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block9_0_bn (BatchNo  (None, 19, 19, 384)          1536      ['conv3_block8_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block9_0_relu (Activ  (None, 19, 19, 384)          0         ['conv3_block9_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block9_1_conv (Conv2  (None, 19, 19, 128)          49152     ['conv3_block9_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block9_1_bn (BatchNo  (None, 19, 19, 128)          512       ['conv3_block9_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv3_block9_1_relu (Activ  (None, 19, 19, 128)          0         ['conv3_block9_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv3_block9_2_conv (Conv2  (None, 19, 19, 32)           36864     ['conv3_block9_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block9_concat (Conca  (None, 19, 19, 416)          0         ['conv3_block8_concat[0][0]', \n",
            " tenate)                                                             'conv3_block9_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv3_block10_0_bn (BatchN  (None, 19, 19, 416)          1664      ['conv3_block9_concat[0][0]'] \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block10_0_relu (Acti  (None, 19, 19, 416)          0         ['conv3_block10_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block10_1_conv (Conv  (None, 19, 19, 128)          53248     ['conv3_block10_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block10_1_bn (BatchN  (None, 19, 19, 128)          512       ['conv3_block10_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block10_1_relu (Acti  (None, 19, 19, 128)          0         ['conv3_block10_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block10_2_conv (Conv  (None, 19, 19, 32)           36864     ['conv3_block10_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block10_concat (Conc  (None, 19, 19, 448)          0         ['conv3_block9_concat[0][0]', \n",
            " atenate)                                                            'conv3_block10_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv3_block11_0_bn (BatchN  (None, 19, 19, 448)          1792      ['conv3_block10_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block11_0_relu (Acti  (None, 19, 19, 448)          0         ['conv3_block11_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block11_1_conv (Conv  (None, 19, 19, 128)          57344     ['conv3_block11_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block11_1_bn (BatchN  (None, 19, 19, 128)          512       ['conv3_block11_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block11_1_relu (Acti  (None, 19, 19, 128)          0         ['conv3_block11_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block11_2_conv (Conv  (None, 19, 19, 32)           36864     ['conv3_block11_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block11_concat (Conc  (None, 19, 19, 480)          0         ['conv3_block10_concat[0][0]',\n",
            " atenate)                                                            'conv3_block11_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv3_block12_0_bn (BatchN  (None, 19, 19, 480)          1920      ['conv3_block11_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block12_0_relu (Acti  (None, 19, 19, 480)          0         ['conv3_block12_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block12_1_conv (Conv  (None, 19, 19, 128)          61440     ['conv3_block12_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block12_1_bn (BatchN  (None, 19, 19, 128)          512       ['conv3_block12_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv3_block12_1_relu (Acti  (None, 19, 19, 128)          0         ['conv3_block12_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv3_block12_2_conv (Conv  (None, 19, 19, 32)           36864     ['conv3_block12_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv3_block12_concat (Conc  (None, 19, 19, 512)          0         ['conv3_block11_concat[0][0]',\n",
            " atenate)                                                            'conv3_block12_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " pool3_bn (BatchNormalizati  (None, 19, 19, 512)          2048      ['conv3_block12_concat[0][0]']\n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " pool3_relu (Activation)     (None, 19, 19, 512)          0         ['pool3_bn[0][0]']            \n",
            "                                                                                                  \n",
            " pool3_conv (Conv2D)         (None, 19, 19, 256)          131072    ['pool3_relu[0][0]']          \n",
            "                                                                                                  \n",
            " pool3_pool (AveragePooling  (None, 9, 9, 256)            0         ['pool3_conv[0][0]']          \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block1_0_bn (BatchNo  (None, 9, 9, 256)            1024      ['pool3_pool[0][0]']          \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block1_0_relu (Activ  (None, 9, 9, 256)            0         ['conv4_block1_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block1_1_conv (Conv2  (None, 9, 9, 128)            32768     ['conv4_block1_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block1_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block1_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block1_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block1_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block1_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_concat (Conca  (None, 9, 9, 288)            0         ['pool3_pool[0][0]',          \n",
            " tenate)                                                             'conv4_block1_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block2_0_bn (BatchNo  (None, 9, 9, 288)            1152      ['conv4_block1_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block2_0_relu (Activ  (None, 9, 9, 288)            0         ['conv4_block2_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block2_1_conv (Conv2  (None, 9, 9, 128)            36864     ['conv4_block2_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block2_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block2_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block2_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block2_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block2_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_concat (Conca  (None, 9, 9, 320)            0         ['conv4_block1_concat[0][0]', \n",
            " tenate)                                                             'conv4_block2_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block3_0_bn (BatchNo  (None, 9, 9, 320)            1280      ['conv4_block2_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block3_0_relu (Activ  (None, 9, 9, 320)            0         ['conv4_block3_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block3_1_conv (Conv2  (None, 9, 9, 128)            40960     ['conv4_block3_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block3_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block3_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block3_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block3_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block3_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_concat (Conca  (None, 9, 9, 352)            0         ['conv4_block2_concat[0][0]', \n",
            " tenate)                                                             'conv4_block3_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block4_0_bn (BatchNo  (None, 9, 9, 352)            1408      ['conv4_block3_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block4_0_relu (Activ  (None, 9, 9, 352)            0         ['conv4_block4_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block4_1_conv (Conv2  (None, 9, 9, 128)            45056     ['conv4_block4_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block4_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block4_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block4_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block4_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block4_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_concat (Conca  (None, 9, 9, 384)            0         ['conv4_block3_concat[0][0]', \n",
            " tenate)                                                             'conv4_block4_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block5_0_bn (BatchNo  (None, 9, 9, 384)            1536      ['conv4_block4_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block5_0_relu (Activ  (None, 9, 9, 384)            0         ['conv4_block5_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block5_1_conv (Conv2  (None, 9, 9, 128)            49152     ['conv4_block5_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block5_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block5_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block5_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block5_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block5_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_concat (Conca  (None, 9, 9, 416)            0         ['conv4_block4_concat[0][0]', \n",
            " tenate)                                                             'conv4_block5_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block6_0_bn (BatchNo  (None, 9, 9, 416)            1664      ['conv4_block5_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block6_0_relu (Activ  (None, 9, 9, 416)            0         ['conv4_block6_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block6_1_conv (Conv2  (None, 9, 9, 128)            53248     ['conv4_block6_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block6_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block6_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block6_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block6_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block6_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_concat (Conca  (None, 9, 9, 448)            0         ['conv4_block5_concat[0][0]', \n",
            " tenate)                                                             'conv4_block6_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block7_0_bn (BatchNo  (None, 9, 9, 448)            1792      ['conv4_block6_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block7_0_relu (Activ  (None, 9, 9, 448)            0         ['conv4_block7_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block7_1_conv (Conv2  (None, 9, 9, 128)            57344     ['conv4_block7_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block7_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block7_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block7_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block7_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block7_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block7_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block7_concat (Conca  (None, 9, 9, 480)            0         ['conv4_block6_concat[0][0]', \n",
            " tenate)                                                             'conv4_block7_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block8_0_bn (BatchNo  (None, 9, 9, 480)            1920      ['conv4_block7_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block8_0_relu (Activ  (None, 9, 9, 480)            0         ['conv4_block8_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block8_1_conv (Conv2  (None, 9, 9, 128)            61440     ['conv4_block8_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block8_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block8_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block8_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block8_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block8_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block8_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block8_concat (Conca  (None, 9, 9, 512)            0         ['conv4_block7_concat[0][0]', \n",
            " tenate)                                                             'conv4_block8_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block9_0_bn (BatchNo  (None, 9, 9, 512)            2048      ['conv4_block8_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block9_0_relu (Activ  (None, 9, 9, 512)            0         ['conv4_block9_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block9_1_conv (Conv2  (None, 9, 9, 128)            65536     ['conv4_block9_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block9_1_bn (BatchNo  (None, 9, 9, 128)            512       ['conv4_block9_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv4_block9_1_relu (Activ  (None, 9, 9, 128)            0         ['conv4_block9_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv4_block9_2_conv (Conv2  (None, 9, 9, 32)             36864     ['conv4_block9_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block9_concat (Conca  (None, 9, 9, 544)            0         ['conv4_block8_concat[0][0]', \n",
            " tenate)                                                             'conv4_block9_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv4_block10_0_bn (BatchN  (None, 9, 9, 544)            2176      ['conv4_block9_concat[0][0]'] \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block10_0_relu (Acti  (None, 9, 9, 544)            0         ['conv4_block10_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block10_1_conv (Conv  (None, 9, 9, 128)            69632     ['conv4_block10_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block10_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block10_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block10_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block10_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block10_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block10_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block10_concat (Conc  (None, 9, 9, 576)            0         ['conv4_block9_concat[0][0]', \n",
            " atenate)                                                            'conv4_block10_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block11_0_bn (BatchN  (None, 9, 9, 576)            2304      ['conv4_block10_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block11_0_relu (Acti  (None, 9, 9, 576)            0         ['conv4_block11_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block11_1_conv (Conv  (None, 9, 9, 128)            73728     ['conv4_block11_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block11_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block11_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block11_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block11_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block11_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block11_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block11_concat (Conc  (None, 9, 9, 608)            0         ['conv4_block10_concat[0][0]',\n",
            " atenate)                                                            'conv4_block11_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block12_0_bn (BatchN  (None, 9, 9, 608)            2432      ['conv4_block11_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block12_0_relu (Acti  (None, 9, 9, 608)            0         ['conv4_block12_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block12_1_conv (Conv  (None, 9, 9, 128)            77824     ['conv4_block12_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block12_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block12_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block12_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block12_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block12_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block12_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block12_concat (Conc  (None, 9, 9, 640)            0         ['conv4_block11_concat[0][0]',\n",
            " atenate)                                                            'conv4_block12_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block13_0_bn (BatchN  (None, 9, 9, 640)            2560      ['conv4_block12_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block13_0_relu (Acti  (None, 9, 9, 640)            0         ['conv4_block13_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block13_1_conv (Conv  (None, 9, 9, 128)            81920     ['conv4_block13_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block13_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block13_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block13_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block13_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block13_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block13_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block13_concat (Conc  (None, 9, 9, 672)            0         ['conv4_block12_concat[0][0]',\n",
            " atenate)                                                            'conv4_block13_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block14_0_bn (BatchN  (None, 9, 9, 672)            2688      ['conv4_block13_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block14_0_relu (Acti  (None, 9, 9, 672)            0         ['conv4_block14_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block14_1_conv (Conv  (None, 9, 9, 128)            86016     ['conv4_block14_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block14_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block14_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block14_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block14_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block14_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block14_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block14_concat (Conc  (None, 9, 9, 704)            0         ['conv4_block13_concat[0][0]',\n",
            " atenate)                                                            'conv4_block14_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block15_0_bn (BatchN  (None, 9, 9, 704)            2816      ['conv4_block14_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block15_0_relu (Acti  (None, 9, 9, 704)            0         ['conv4_block15_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block15_1_conv (Conv  (None, 9, 9, 128)            90112     ['conv4_block15_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block15_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block15_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block15_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block15_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block15_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block15_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block15_concat (Conc  (None, 9, 9, 736)            0         ['conv4_block14_concat[0][0]',\n",
            " atenate)                                                            'conv4_block15_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block16_0_bn (BatchN  (None, 9, 9, 736)            2944      ['conv4_block15_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block16_0_relu (Acti  (None, 9, 9, 736)            0         ['conv4_block16_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block16_1_conv (Conv  (None, 9, 9, 128)            94208     ['conv4_block16_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block16_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block16_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block16_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block16_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block16_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block16_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block16_concat (Conc  (None, 9, 9, 768)            0         ['conv4_block15_concat[0][0]',\n",
            " atenate)                                                            'conv4_block16_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block17_0_bn (BatchN  (None, 9, 9, 768)            3072      ['conv4_block16_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block17_0_relu (Acti  (None, 9, 9, 768)            0         ['conv4_block17_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block17_1_conv (Conv  (None, 9, 9, 128)            98304     ['conv4_block17_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block17_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block17_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block17_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block17_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block17_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block17_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block17_concat (Conc  (None, 9, 9, 800)            0         ['conv4_block16_concat[0][0]',\n",
            " atenate)                                                            'conv4_block17_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block18_0_bn (BatchN  (None, 9, 9, 800)            3200      ['conv4_block17_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block18_0_relu (Acti  (None, 9, 9, 800)            0         ['conv4_block18_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block18_1_conv (Conv  (None, 9, 9, 128)            102400    ['conv4_block18_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block18_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block18_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block18_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block18_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block18_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block18_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block18_concat (Conc  (None, 9, 9, 832)            0         ['conv4_block17_concat[0][0]',\n",
            " atenate)                                                            'conv4_block18_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block19_0_bn (BatchN  (None, 9, 9, 832)            3328      ['conv4_block18_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block19_0_relu (Acti  (None, 9, 9, 832)            0         ['conv4_block19_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block19_1_conv (Conv  (None, 9, 9, 128)            106496    ['conv4_block19_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block19_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block19_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block19_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block19_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block19_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block19_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block19_concat (Conc  (None, 9, 9, 864)            0         ['conv4_block18_concat[0][0]',\n",
            " atenate)                                                            'conv4_block19_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block20_0_bn (BatchN  (None, 9, 9, 864)            3456      ['conv4_block19_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block20_0_relu (Acti  (None, 9, 9, 864)            0         ['conv4_block20_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block20_1_conv (Conv  (None, 9, 9, 128)            110592    ['conv4_block20_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block20_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block20_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block20_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block20_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block20_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block20_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block20_concat (Conc  (None, 9, 9, 896)            0         ['conv4_block19_concat[0][0]',\n",
            " atenate)                                                            'conv4_block20_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block21_0_bn (BatchN  (None, 9, 9, 896)            3584      ['conv4_block20_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block21_0_relu (Acti  (None, 9, 9, 896)            0         ['conv4_block21_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block21_1_conv (Conv  (None, 9, 9, 128)            114688    ['conv4_block21_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block21_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block21_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block21_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block21_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block21_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block21_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block21_concat (Conc  (None, 9, 9, 928)            0         ['conv4_block20_concat[0][0]',\n",
            " atenate)                                                            'conv4_block21_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block22_0_bn (BatchN  (None, 9, 9, 928)            3712      ['conv4_block21_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block22_0_relu (Acti  (None, 9, 9, 928)            0         ['conv4_block22_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block22_1_conv (Conv  (None, 9, 9, 128)            118784    ['conv4_block22_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block22_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block22_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block22_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block22_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block22_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block22_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block22_concat (Conc  (None, 9, 9, 960)            0         ['conv4_block21_concat[0][0]',\n",
            " atenate)                                                            'conv4_block22_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block23_0_bn (BatchN  (None, 9, 9, 960)            3840      ['conv4_block22_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block23_0_relu (Acti  (None, 9, 9, 960)            0         ['conv4_block23_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block23_1_conv (Conv  (None, 9, 9, 128)            122880    ['conv4_block23_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block23_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block23_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block23_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block23_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block23_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block23_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block23_concat (Conc  (None, 9, 9, 992)            0         ['conv4_block22_concat[0][0]',\n",
            " atenate)                                                            'conv4_block23_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv4_block24_0_bn (BatchN  (None, 9, 9, 992)            3968      ['conv4_block23_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block24_0_relu (Acti  (None, 9, 9, 992)            0         ['conv4_block24_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block24_1_conv (Conv  (None, 9, 9, 128)            126976    ['conv4_block24_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block24_1_bn (BatchN  (None, 9, 9, 128)            512       ['conv4_block24_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv4_block24_1_relu (Acti  (None, 9, 9, 128)            0         ['conv4_block24_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv4_block24_2_conv (Conv  (None, 9, 9, 32)             36864     ['conv4_block24_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv4_block24_concat (Conc  (None, 9, 9, 1024)           0         ['conv4_block23_concat[0][0]',\n",
            " atenate)                                                            'conv4_block24_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " pool4_bn (BatchNormalizati  (None, 9, 9, 1024)           4096      ['conv4_block24_concat[0][0]']\n",
            " on)                                                                                              \n",
            "                                                                                                  \n",
            " pool4_relu (Activation)     (None, 9, 9, 1024)           0         ['pool4_bn[0][0]']            \n",
            "                                                                                                  \n",
            " pool4_conv (Conv2D)         (None, 9, 9, 512)            524288    ['pool4_relu[0][0]']          \n",
            "                                                                                                  \n",
            " pool4_pool (AveragePooling  (None, 4, 4, 512)            0         ['pool4_conv[0][0]']          \n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block1_0_bn (BatchNo  (None, 4, 4, 512)            2048      ['pool4_pool[0][0]']          \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block1_0_relu (Activ  (None, 4, 4, 512)            0         ['conv5_block1_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block1_1_conv (Conv2  (None, 4, 4, 128)            65536     ['conv5_block1_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block1_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block1_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block1_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block1_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block1_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_concat (Conca  (None, 4, 4, 544)            0         ['pool4_pool[0][0]',          \n",
            " tenate)                                                             'conv5_block1_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block2_0_bn (BatchNo  (None, 4, 4, 544)            2176      ['conv5_block1_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block2_0_relu (Activ  (None, 4, 4, 544)            0         ['conv5_block2_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block2_1_conv (Conv2  (None, 4, 4, 128)            69632     ['conv5_block2_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block2_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block2_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block2_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block2_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block2_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_concat (Conca  (None, 4, 4, 576)            0         ['conv5_block1_concat[0][0]', \n",
            " tenate)                                                             'conv5_block2_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block3_0_bn (BatchNo  (None, 4, 4, 576)            2304      ['conv5_block2_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block3_0_relu (Activ  (None, 4, 4, 576)            0         ['conv5_block3_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block3_1_conv (Conv2  (None, 4, 4, 128)            73728     ['conv5_block3_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block3_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block3_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block3_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block3_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block3_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_concat (Conca  (None, 4, 4, 608)            0         ['conv5_block2_concat[0][0]', \n",
            " tenate)                                                             'conv5_block3_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block4_0_bn (BatchNo  (None, 4, 4, 608)            2432      ['conv5_block3_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block4_0_relu (Activ  (None, 4, 4, 608)            0         ['conv5_block4_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block4_1_conv (Conv2  (None, 4, 4, 128)            77824     ['conv5_block4_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block4_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block4_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block4_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block4_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block4_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block4_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block4_concat (Conca  (None, 4, 4, 640)            0         ['conv5_block3_concat[0][0]', \n",
            " tenate)                                                             'conv5_block4_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block5_0_bn (BatchNo  (None, 4, 4, 640)            2560      ['conv5_block4_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block5_0_relu (Activ  (None, 4, 4, 640)            0         ['conv5_block5_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block5_1_conv (Conv2  (None, 4, 4, 128)            81920     ['conv5_block5_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block5_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block5_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block5_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block5_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block5_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block5_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block5_concat (Conca  (None, 4, 4, 672)            0         ['conv5_block4_concat[0][0]', \n",
            " tenate)                                                             'conv5_block5_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block6_0_bn (BatchNo  (None, 4, 4, 672)            2688      ['conv5_block5_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block6_0_relu (Activ  (None, 4, 4, 672)            0         ['conv5_block6_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block6_1_conv (Conv2  (None, 4, 4, 128)            86016     ['conv5_block6_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block6_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block6_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block6_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block6_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block6_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block6_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block6_concat (Conca  (None, 4, 4, 704)            0         ['conv5_block5_concat[0][0]', \n",
            " tenate)                                                             'conv5_block6_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block7_0_bn (BatchNo  (None, 4, 4, 704)            2816      ['conv5_block6_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block7_0_relu (Activ  (None, 4, 4, 704)            0         ['conv5_block7_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block7_1_conv (Conv2  (None, 4, 4, 128)            90112     ['conv5_block7_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block7_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block7_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block7_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block7_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block7_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block7_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block7_concat (Conca  (None, 4, 4, 736)            0         ['conv5_block6_concat[0][0]', \n",
            " tenate)                                                             'conv5_block7_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block8_0_bn (BatchNo  (None, 4, 4, 736)            2944      ['conv5_block7_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block8_0_relu (Activ  (None, 4, 4, 736)            0         ['conv5_block8_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block8_1_conv (Conv2  (None, 4, 4, 128)            94208     ['conv5_block8_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block8_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block8_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block8_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block8_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block8_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block8_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block8_concat (Conca  (None, 4, 4, 768)            0         ['conv5_block7_concat[0][0]', \n",
            " tenate)                                                             'conv5_block8_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block9_0_bn (BatchNo  (None, 4, 4, 768)            3072      ['conv5_block8_concat[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block9_0_relu (Activ  (None, 4, 4, 768)            0         ['conv5_block9_0_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block9_1_conv (Conv2  (None, 4, 4, 128)            98304     ['conv5_block9_0_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block9_1_bn (BatchNo  (None, 4, 4, 128)            512       ['conv5_block9_1_conv[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv5_block9_1_relu (Activ  (None, 4, 4, 128)            0         ['conv5_block9_1_bn[0][0]']   \n",
            " ation)                                                                                           \n",
            "                                                                                                  \n",
            " conv5_block9_2_conv (Conv2  (None, 4, 4, 32)             36864     ['conv5_block9_1_relu[0][0]'] \n",
            " D)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block9_concat (Conca  (None, 4, 4, 800)            0         ['conv5_block8_concat[0][0]', \n",
            " tenate)                                                             'conv5_block9_2_conv[0][0]'] \n",
            "                                                                                                  \n",
            " conv5_block10_0_bn (BatchN  (None, 4, 4, 800)            3200      ['conv5_block9_concat[0][0]'] \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block10_0_relu (Acti  (None, 4, 4, 800)            0         ['conv5_block10_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block10_1_conv (Conv  (None, 4, 4, 128)            102400    ['conv5_block10_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block10_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block10_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block10_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block10_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block10_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block10_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block10_concat (Conc  (None, 4, 4, 832)            0         ['conv5_block9_concat[0][0]', \n",
            " atenate)                                                            'conv5_block10_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block11_0_bn (BatchN  (None, 4, 4, 832)            3328      ['conv5_block10_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block11_0_relu (Acti  (None, 4, 4, 832)            0         ['conv5_block11_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block11_1_conv (Conv  (None, 4, 4, 128)            106496    ['conv5_block11_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block11_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block11_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block11_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block11_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block11_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block11_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block11_concat (Conc  (None, 4, 4, 864)            0         ['conv5_block10_concat[0][0]',\n",
            " atenate)                                                            'conv5_block11_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block12_0_bn (BatchN  (None, 4, 4, 864)            3456      ['conv5_block11_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block12_0_relu (Acti  (None, 4, 4, 864)            0         ['conv5_block12_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block12_1_conv (Conv  (None, 4, 4, 128)            110592    ['conv5_block12_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block12_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block12_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block12_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block12_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block12_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block12_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block12_concat (Conc  (None, 4, 4, 896)            0         ['conv5_block11_concat[0][0]',\n",
            " atenate)                                                            'conv5_block12_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block13_0_bn (BatchN  (None, 4, 4, 896)            3584      ['conv5_block12_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block13_0_relu (Acti  (None, 4, 4, 896)            0         ['conv5_block13_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block13_1_conv (Conv  (None, 4, 4, 128)            114688    ['conv5_block13_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block13_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block13_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block13_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block13_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block13_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block13_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block13_concat (Conc  (None, 4, 4, 928)            0         ['conv5_block12_concat[0][0]',\n",
            " atenate)                                                            'conv5_block13_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block14_0_bn (BatchN  (None, 4, 4, 928)            3712      ['conv5_block13_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block14_0_relu (Acti  (None, 4, 4, 928)            0         ['conv5_block14_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block14_1_conv (Conv  (None, 4, 4, 128)            118784    ['conv5_block14_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block14_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block14_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block14_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block14_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block14_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block14_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block14_concat (Conc  (None, 4, 4, 960)            0         ['conv5_block13_concat[0][0]',\n",
            " atenate)                                                            'conv5_block14_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block15_0_bn (BatchN  (None, 4, 4, 960)            3840      ['conv5_block14_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block15_0_relu (Acti  (None, 4, 4, 960)            0         ['conv5_block15_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block15_1_conv (Conv  (None, 4, 4, 128)            122880    ['conv5_block15_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block15_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block15_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block15_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block15_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block15_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block15_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block15_concat (Conc  (None, 4, 4, 992)            0         ['conv5_block14_concat[0][0]',\n",
            " atenate)                                                            'conv5_block15_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " conv5_block16_0_bn (BatchN  (None, 4, 4, 992)            3968      ['conv5_block15_concat[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block16_0_relu (Acti  (None, 4, 4, 992)            0         ['conv5_block16_0_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block16_1_conv (Conv  (None, 4, 4, 128)            126976    ['conv5_block16_0_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block16_1_bn (BatchN  (None, 4, 4, 128)            512       ['conv5_block16_1_conv[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv5_block16_1_relu (Acti  (None, 4, 4, 128)            0         ['conv5_block16_1_bn[0][0]']  \n",
            " vation)                                                                                          \n",
            "                                                                                                  \n",
            " conv5_block16_2_conv (Conv  (None, 4, 4, 32)             36864     ['conv5_block16_1_relu[0][0]']\n",
            " 2D)                                                                                              \n",
            "                                                                                                  \n",
            " conv5_block16_concat (Conc  (None, 4, 4, 1024)           0         ['conv5_block15_concat[0][0]',\n",
            " atenate)                                                            'conv5_block16_2_conv[0][0]']\n",
            "                                                                                                  \n",
            " bn (BatchNormalization)     (None, 4, 4, 1024)           4096      ['conv5_block16_concat[0][0]']\n",
            "                                                                                                  \n",
            " relu (Activation)           (None, 4, 4, 1024)           0         ['bn[0][0]']                  \n",
            "                                                                                                  \n",
            " global_average_pooling2d (  (None, 1024)                 0         ['relu[0][0]']                \n",
            " GlobalAveragePooling2D)                                                                          \n",
            "                                                                                                  \n",
            " dropout (Dropout)           (None, 1024)                 0         ['global_average_pooling2d[0][\n",
            "                                                                    0]']                          \n",
            "                                                                                                  \n",
            " dense (Dense)               (None, 4)                    4100      ['dropout[0][0]']             \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 7041604 (26.86 MB)\n",
            "Trainable params: 6957956 (26.54 MB)\n",
            "Non-trainable params: 83648 (326.75 KB)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kJ6WngB2ZVTT"
      },
      "outputs": [],
      "source": [
        "model.compile(loss='categorical_crossentropy',optimizer = 'Adam', metrics= ['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kqgPOfv1ZYEc"
      },
      "outputs": [],
      "source": [
        "tensorboard = TensorBoard(log_dir = 'logs')\n",
        "checkpoint = ModelCheckpoint(\"effnet.h5\",monitor=\"val_accuracy\",save_best_only=True,mode=\"auto\",verbose=1)\n",
        "reduce_lr = ReduceLROnPlateau(monitor = 'val_accuracy', factor = 0.3, patience = 2, min_delta = 0.001,\n",
        "                              mode='auto',verbose=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n8-SWDkfZaM8",
        "outputId": "57c74442-88e3-4c8f-d4bf-28ed72f9361a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.5833 - accuracy: 0.8104\n",
            "Epoch 1: val_accuracy improved from -inf to 0.47394, saving model to effnet.h5\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3079: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "173/173 [==============================] - 1769s 10s/step - loss: 0.5833 - accuracy: 0.8104 - val_loss: 5.6671 - val_accuracy: 0.4739 - lr: 0.0010\n",
            "Epoch 2/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.2741 - accuracy: 0.9062\n",
            "Epoch 2: val_accuracy improved from 0.47394 to 0.59121, saving model to effnet.h5\n",
            "173/173 [==============================] - 1645s 9s/step - loss: 0.2741 - accuracy: 0.9062 - val_loss: 1.2483 - val_accuracy: 0.5912 - lr: 0.0010\n",
            "Epoch 3/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.2067 - accuracy: 0.9292\n",
            "Epoch 3: val_accuracy improved from 0.59121 to 0.73779, saving model to effnet.h5\n",
            "173/173 [==============================] - 1568s 9s/step - loss: 0.2067 - accuracy: 0.9292 - val_loss: 0.8337 - val_accuracy: 0.7378 - lr: 0.0010\n",
            "Epoch 4/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.1666 - accuracy: 0.9395\n",
            "Epoch 4: val_accuracy did not improve from 0.73779\n",
            "173/173 [==============================] - 1535s 9s/step - loss: 0.1666 - accuracy: 0.9395 - val_loss: 1.3857 - val_accuracy: 0.7085 - lr: 0.0010\n",
            "Epoch 5/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.1097 - accuracy: 0.9629\n",
            "Epoch 5: val_accuracy did not improve from 0.73779\n",
            "\n",
            "Epoch 5: ReduceLROnPlateau reducing learning rate to 0.0003000000142492354.\n",
            "173/173 [==============================] - 1560s 9s/step - loss: 0.1097 - accuracy: 0.9629 - val_loss: 1.7182 - val_accuracy: 0.6612 - lr: 0.0010\n",
            "Epoch 6/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0483 - accuracy: 0.9855\n",
            "Epoch 6: val_accuracy improved from 0.73779 to 0.96417, saving model to effnet.h5\n",
            "173/173 [==============================] - 1566s 9s/step - loss: 0.0483 - accuracy: 0.9855 - val_loss: 0.0856 - val_accuracy: 0.9642 - lr: 3.0000e-04\n",
            "Epoch 7/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0201 - accuracy: 0.9946\n",
            "Epoch 7: val_accuracy did not improve from 0.96417\n",
            "173/173 [==============================] - 1572s 9s/step - loss: 0.0201 - accuracy: 0.9946 - val_loss: 0.1832 - val_accuracy: 0.9479 - lr: 3.0000e-04\n",
            "Epoch 8/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0100 - accuracy: 0.9975\n",
            "Epoch 8: val_accuracy improved from 0.96417 to 0.97557, saving model to effnet.h5\n",
            "173/173 [==============================] - 1616s 9s/step - loss: 0.0100 - accuracy: 0.9975 - val_loss: 0.0849 - val_accuracy: 0.9756 - lr: 3.0000e-04\n",
            "Epoch 9/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0092 - accuracy: 0.9967\n",
            "Epoch 9: val_accuracy did not improve from 0.97557\n",
            "173/173 [==============================] - 1611s 9s/step - loss: 0.0092 - accuracy: 0.9967 - val_loss: 0.1857 - val_accuracy: 0.9577 - lr: 3.0000e-04\n",
            "Epoch 10/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0110 - accuracy: 0.9960\n",
            "Epoch 10: val_accuracy did not improve from 0.97557\n",
            "\n",
            "Epoch 10: ReduceLROnPlateau reducing learning rate to 9.000000427477062e-05.\n",
            "173/173 [==============================] - 1591s 9s/step - loss: 0.0110 - accuracy: 0.9960 - val_loss: 0.1813 - val_accuracy: 0.9609 - lr: 3.0000e-04\n",
            "Epoch 11/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0044 - accuracy: 0.9989\n",
            "Epoch 11: val_accuracy improved from 0.97557 to 0.98371, saving model to effnet.h5\n",
            "173/173 [==============================] - 1601s 9s/step - loss: 0.0044 - accuracy: 0.9989 - val_loss: 0.0933 - val_accuracy: 0.9837 - lr: 9.0000e-05\n",
            "Epoch 12/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0027 - accuracy: 0.9996\n",
            "Epoch 12: val_accuracy did not improve from 0.98371\n",
            "173/173 [==============================] - 1574s 9s/step - loss: 0.0027 - accuracy: 0.9996 - val_loss: 0.0743 - val_accuracy: 0.9821 - lr: 9.0000e-05\n",
            "Epoch 13/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0037 - accuracy: 0.9991\n",
            "Epoch 13: val_accuracy did not improve from 0.98371\n",
            "\n",
            "Epoch 13: ReduceLROnPlateau reducing learning rate to 2.700000040931627e-05.\n",
            "173/173 [==============================] - 1589s 9s/step - loss: 0.0037 - accuracy: 0.9991 - val_loss: 0.0949 - val_accuracy: 0.9772 - lr: 9.0000e-05\n",
            "Epoch 14/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0028 - accuracy: 0.9993\n",
            "Epoch 14: val_accuracy did not improve from 0.98371\n",
            "173/173 [==============================] - 1588s 9s/step - loss: 0.0028 - accuracy: 0.9993 - val_loss: 0.0868 - val_accuracy: 0.9805 - lr: 2.7000e-05\n",
            "Epoch 15/15\n",
            "173/173 [==============================] - ETA: 0s - loss: 0.0035 - accuracy: 0.9993\n",
            "Epoch 15: val_accuracy did not improve from 0.98371\n",
            "\n",
            "Epoch 15: ReduceLROnPlateau reducing learning rate to 8.100000013655517e-06.\n",
            "173/173 [==============================] - 1537s 9s/step - loss: 0.0035 - accuracy: 0.9993 - val_loss: 0.0869 - val_accuracy: 0.9805 - lr: 2.7000e-05\n"
          ]
        }
      ],
      "source": [
        "history = model.fit(X_train,y_train,validation_split=0.1, epochs =15, verbose=1, batch_size=32,\n",
        "                   callbacks=[tensorboard,checkpoint,reduce_lr])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e3x6aTB61xDU",
        "outputId": "42eece7d-4d8b-45a6-eefa-4b21ea1016e5"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3079: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        }
      ],
      "source": [
        "model.save(\"brain.h5\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 422
        },
        "id": "cWAkANf5Zir6",
        "outputId": "e289ad02-e757-4714-b985-a341e010c342"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1400x600 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Extracting metrics from the history object\n",
        "train_acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "train_loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "# Generate list of epoch numbers\n",
        "epochs = range(1, len(train_acc) + 1)\n",
        "\n",
        "# Create subplots\n",
        "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 6))\n",
        "\n",
        "# Plot accuracy\n",
        "ax1.plot(epochs, train_acc, label='Training Accuracy', marker='o')\n",
        "ax1.plot(epochs, val_acc, label='Validation Accuracy', marker='x')\n",
        "ax1.set_title('Training and Validation Accuracy')\n",
        "ax1.set_xlabel('Epoch')\n",
        "ax1.set_ylabel('Accuracy')\n",
        "ax1.legend()\n",
        "\n",
        "# Plot loss\n",
        "ax2.plot(epochs, train_loss, label='Training Loss', marker='o')\n",
        "ax2.plot(epochs, val_loss, label='Validation Loss', marker='x')\n",
        "ax2.set_title('Training and Validation Loss')\n",
        "ax2.set_xlabel('Epoch')\n",
        "ax2.set_ylabel('Loss')\n",
        "ax2.legend()\n",
        "\n",
        "# Show the plots\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cqNYNxKwZm5-",
        "outputId": "96fa2463-e702-4ef6-d315-ee1227815b87"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "48/48 [==============================] - 88s 2s/step\n"
          ]
        }
      ],
      "source": [
        "pred = model.predict(X_test)\n",
        "pred = np.argmax(pred,axis=1)\n",
        "y_test_new = np.argmax(y_test,axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Or__8wfoZo9r",
        "outputId": "544b8092-66c8-47f3-822e-64e6735ce04a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.98      0.97      0.98       406\n",
            "           1       0.97      0.99      0.98       314\n",
            "           2       0.95      0.97      0.96       384\n",
            "           3       1.00      0.98      0.99       430\n",
            "\n",
            "    accuracy                           0.98      1534\n",
            "   macro avg       0.97      0.98      0.98      1534\n",
            "weighted avg       0.98      0.98      0.98      1534\n",
            "\n"
          ]
        }
      ],
      "source": [
        "print(classification_report(y_test_new,pred))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kA53Gv-nZqu-"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HOzPro3JZsap",
        "outputId": "6a21ead6-f6e8-4058-951f-92627498c696"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "48/48 [==============================] - 92s 2s/step\n",
            "Testing Accuracy: 97.52%\n"
          ]
        }
      ],
      "source": [
        "pred = model.predict(X_test)\n",
        "pred = np.argmax(pred, axis=1)\n",
        "\n",
        "# Assuming y_test is one-hot encoded, if it's not, you don't need the next line\n",
        "y_test_new = np.argmax(y_test, axis=1)\n",
        "\n",
        "# Calculate the accuracy\n",
        "test_accuracy = accuracy_score(y_test_new, pred)\n",
        "\n",
        "# Print the accuracy\n",
        "print(f\"Testing Accuracy: {test_accuracy * 100:.2f}%\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "NJ422NV34s-K"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras.models import load_model\n",
        "\n",
        "# Replace 'brain.h5' with the actual model file name\n",
        "model = load_model('/content/brain.h5')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d1Z-PldUfvtj"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k86V3FKwhG1E",
        "outputId": "e8ad2ba3-0b25-4e46-bbe6-8b734fdc7498"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 146ms/step\n",
            "[[1.2034530e-08 4.5189158e-10 9.9999928e-01 6.8947082e-07]]\n",
            "Predictions: 2\n"
          ]
        }
      ],
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "# Load and preprocess the image for prediction\n",
        "image_path = '/content/Training/meningioma_tumor/m3 (202).jpg'\n",
        "image = cv2.imread(image_path)\n",
        "image = cv2.resize(image, (150, 150))  # Resize the image to match the input shape of the model\n",
        "image = np.expand_dims(image, axis=0)  # Add an extra dimension for batch size\n",
        "\n",
        "# Make predictions\n",
        "predictions = model.predict(image)\n",
        "print(predictions)\n",
        "# Assuming the model outputs probabilities for different classes\n",
        "print(\"Predictions:\", np.argmax(predictions))\n"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}