Project_1_Food_Vision_Big.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Samkhan705279/foodvisionproject/blob/main/Project_1_Food_Vision_Big.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a1KwHWQD26BY",
        "outputId": "25fe8291-b70a-414a-84a3-bc5c3cf5866f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m498.0/498.0 MB\u001b[0m \u001b[31m2.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m42.6/42.6 kB\u001b[0m \u001b[31m5.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.8/5.8 MB\u001b[0m \u001b[31m97.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m462.3/462.3 kB\u001b[0m \u001b[31m34.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.4/1.4 MB\u001b[0m \u001b[31m74.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.9/4.9 MB\u001b[0m \u001b[31m94.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m781.3/781.3 kB\u001b[0m \u001b[31m53.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hTensorFlow version: 2.8.1\n"
          ]
        }
      ],
      "source": [
        "# Install TensorFlow 2.9.0 (\"-U\" stands for \"update\", \"-q\" stands for \"quiet\")\n",
        "!pip install -U -q tensorflow==2.8.1\n",
        "\n",
        "import tensorflow as tf\n",
        "print(f\"TensorFlow version: {tf.__version__}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OVaZwFnv0VM3",
        "outputId": "b06fdba0-476f-4e9d-d9e0-d5cedbb10cf1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Tue Jul 25 09:21:33 2023       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 525.105.17   Driver Version: 525.105.17   CUDA Version: 12.0     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   39C    P8    10W /  70W |      0MiB / 15360MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ],
      "source": [
        "!nvidia-smi"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K-IdaoD2ESm8",
        "outputId": "e99e4bb8-6038-4964-c648-94401702ae0a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-07-26 11:08:26--  https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/helper_functions.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 10246 (10K) [text/plain]\n",
            "Saving to: ‘helper_functions.py’\n",
            "\n",
            "helper_functions.py 100%[===================>]  10.01K  --.-KB/s    in 0s      \n",
            "\n",
            "2023-07-26 11:08:26 (76.5 MB/s) - ‘helper_functions.py’ saved [10246/10246]\n",
            "\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Get helper functions file\n",
        "import os\n",
        "\n",
        "if not os.path.exists(\"helper_functions.py\"):\n",
        "    !wget https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/helper_functions.py\n",
        "else:\n",
        "    print(\"[INFO] 'helper_functions.py' already exists, skipping download.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "UZNBIWSu5ekU"
      },
      "outputs": [],
      "source": [
        "# Import series of helper functions for the notebook (we've created/used these in previous notebooks)\n",
        "from helper_functions import create_tensorboard_callback, plot_loss_curves, compare_historys"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "At4062_ZErWE"
      },
      "source": [
        "# use tensorflow datasets to download data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pb7HR-Q7-R7x"
      },
      "outputs": [],
      "source": [
        "# !pip install --upgrade tensorflow-datasets\n",
        "# !pip install --upgrade protobuf\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "qtZlpkLsAjBx"
      },
      "outputs": [],
      "source": [
        "# import google.protobuf\n",
        "\n",
        "# print(google.protobuf.__version__)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "gyZ6chMUEy0w"
      },
      "outputs": [],
      "source": [
        "import tensorflow_datasets as tfds"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y-22pEOSORSx",
        "outputId": "a02a6aab-d4ed-4814-ee37-76f6bd4b8be5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "True\n"
          ]
        }
      ],
      "source": [
        "datasets_list=tfds.list_builders()\n",
        "print(\"food101\" in datasets_list)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 148,
          "referenced_widgets": [
            "50aaf069e6794e998bd281010bc5043b",
            "166a908eec7547daa239c2b32a7afec2",
            "ca8bf4ea67144dd0b58d8e79ec74edcf",
            "c3932637bf0e4dcba83a4dbfdff30e24",
            "68f5dbabcd5b408f8bb8f9bee92475c9",
            "56e20ee6e1124be0b787c5a5a3eca037",
            "e8aa3858be224cd0ac698a6ea5a489e9",
            "e1ac71bf85be420bb549bedd34a1ef4e",
            "95285302099b438bad41af674feaec56",
            "b59b3e2f5ce6484abf08dabfc7500653",
            "27a5885d2a3840c7a7e5d3cf41db53b5",
            "fa729f84dbe749188f3db0994974ecc2",
            "564787e186ae4859adb4089fccda05c3",
            "99b1edc4698b4e52bee727a10600f44b",
            "1f4399bfa4584f018477a2580f9a938c",
            "390243e5a48145f3a671f43ffe0ff1f1",
            "b3b982f760a0458383228ec1a44035e5",
            "bc29c4d9b55e45289225e5f62b2f5168",
            "c41e6c7a79cb4e58ad356afcbded7c37",
            "aad3a4379f9c487e9e7b92fe99f9a2dd",
            "427430e96fdf493dbe1a353e675e5106",
            "add28e4fe0874e2482b8c350bfbc4706",
            "04cadefe013e4252ba5560db93f9c87d",
            "04222ce145d94494a8c4c0c23fa11d18",
            "d02c36de95be4184a26540868d30b500",
            "695f9c09c3784a66bb4d61079797fdfd",
            "189dc7f8d9ad434ebeeb9c755fbb0219",
            "23418acc405d4ccbb03714dbd5e171e8",
            "a552a0e8af5d48968432dbe7542f351f",
            "e29be58af5de4cb5a649ca0aecb94e1a",
            "f386501578dc44788a8d62f325b31820",
            "9a74fdf9e66c42c8ba9918e09ce6882c",
            "8f88eb2ac7da41fca8e790dc3989bf3c",
            "cbef5539fbd44df29afcaae390008b4e",
            "70d9afda803f4e6988915edd92240e8b",
            "b1a976b8e7d54497b11f351b0020c75c",
            "435dca930eb646758e8ae99d3c2ac71c",
            "047a4a46d39e4dbd910cc2e0c32608d2",
            "6ac4130692314ddb8acd016c0303d445",
            "9b6174ca2b1a4ffdbc0e8ee78a3b9d59",
            "6239d34ff5f04772a38796edd35e37b2",
            "c4cf236757f94b5f8f414a506df1d2e0",
            "dd642f3bc07a4fd097ea5f39068fb408",
            "deba622183c64e5399bf6416dcbd496a",
            "30135294362b4c2e88151133fa003eee",
            "c2ff63e0263c46589eb11df7d4938b1c",
            "cbf6849683f24cf6acc73f6b8775b401",
            "6523f9271ee6468b809cc6fd5a25e7d6",
            "85149ff51ba74896ae15a2bdd1467034",
            "1fe54ec969ed4db3866b0985e18621b0",
            "4f0f2302b2d64abeae853a6f847e6473",
            "8ef0e1356e4c4af781560e6097272500",
            "7daeca23e1994c228d5d8ffef1ef3aab",
            "fca0d67882e548da92503f9cd6926b3d",
            "8edbcf1aae4c429eaaf29739d9dc4c48",
            "e0d17fba2b8a4d6eb893d13fec852f03",
            "93c3dbc7a118411d9974e4de6951b524",
            "2a37b5317f584e9aab196497d4e81b04",
            "fc9a3fc572a841b19b247b3b69807eb3",
            "7c18ebb5471e4cdf92de0887aa238808",
            "8dedf6ee94c44ff481bfb38aeb406047",
            "2317ad6ef3544ff18b2f622817a73667",
            "4029114fc4634900ad94d252337e25c0",
            "3b407e1d3776454bb54e9d12b720ff48",
            "334f0da01bab4a99abd5e5413491295f",
            "122ea049380d43eaa2870813981891df",
            "ccdaedaf51604a42b132ca2170e1a132",
            "666ebd2054c5439db6683716dcf0d5da",
            "9f7d382e97fa46a98825a38bd51a780e",
            "3c2a50d1a4d04a6b9719c64bc3dc954a",
            "f492e05817de463cbe70fc9da385012e",
            "f3f31742f5b741b6a6323d5a52bc5ec4",
            "175f6420ad844b4198e0b0dd97c6f53d",
            "cc398dfa5a264d6481277661dff6b96e",
            "05ee99df3e154c15869f35b9be8e0c9b",
            "db7a18c7b1264cb4ab5e3f605309de2e",
            "d4f14d9e1c874240bbea3288df4cffac",
            "d2b879d7843d432e8609d7414def04b4",
            "888a2e1eac574f828f0d5c7eaf3bc088",
            "ec5d92ef4de24f7fab17c91348a477b3",
            "bef6d3220d244b4b94aaec8aee29579d",
            "3c4d0a3c537e4793a13618b63aef8182",
            "1987ef9f5e15490083dbbded700b8f88",
            "8eaee9c1e19940e99c1ae094d167ccc6",
            "4b79d3cdc5f64df99a0b7ebeac674e1d",
            "71143da54b3c4b75908f9f3ae502c85b",
            "f70a7661959f44ceb0c88868a101cb51",
            "3bba5f3ddd4c4035a87033870d6165e9"
          ]
        },
        "id": "WS8eedsRPfLA",
        "outputId": "9dccdbf4-d3c7-407e-8900-f7728255b72e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading and preparing dataset 4.65 GiB (download: 4.65 GiB, generated: Unknown size, total: 4.65 GiB) to /root/tensorflow_datasets/food101/2.0.0...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Dl Completed...: 0 url [00:00, ? url/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "50aaf069e6794e998bd281010bc5043b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Dl Size...: 0 MiB [00:00, ? MiB/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "fa729f84dbe749188f3db0994974ecc2"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Extraction completed...: 0 file [00:00, ? file/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "04cadefe013e4252ba5560db93f9c87d"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating splits...:   0%|          | 0/2 [00:00<?, ? splits/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cbef5539fbd44df29afcaae390008b4e"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train examples...:   0%|          | 0/75750 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "30135294362b4c2e88151133fa003eee"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Shuffling /root/tensorflow_datasets/food101/2.0.0.incompleteZ55PQ3/food101-train.tfrecord*...:   0%|          …"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "e0d17fba2b8a4d6eb893d13fec852f03"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating validation examples...:   0%|          | 0/25250 [00:00<?, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "ccdaedaf51604a42b132ca2170e1a132"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Shuffling /root/tensorflow_datasets/food101/2.0.0.incompleteZ55PQ3/food101-validation.tfrecord*...:   0%|     …"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "d2b879d7843d432e8609d7414def04b4"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset food101 downloaded and prepared to /root/tensorflow_datasets/food101/2.0.0. Subsequent calls will reuse this data.\n"
          ]
        }
      ],
      "source": [
        "(train_data,test_data),ds_info=tfds.load(name=\"food101\",\n",
        "                                         split=[\"train\",\"validation\"],\n",
        "                                         shuffle_files=True,\n",
        "                                         as_supervised=True,\n",
        "                                         with_info=True\n",
        "                                         )"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-aV4cLaKSOuO",
        "outputId": "89281ad5-5af6-41ac-ce62-c98db6e4c1bf"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "FeaturesDict({\n",
              "    'image': Image(shape=(None, None, 3), dtype=uint8),\n",
              "    'label': ClassLabel(shape=(), dtype=int64, num_classes=101),\n",
              "})"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ],
      "source": [
        "# Features of Food101 TFDS\n",
        "ds_info.features"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2gfS0eD_a0du",
        "outputId": "cbf5c0c3-be78-424d-e0d6-934430cabef0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['apple_pie',\n",
              " 'baby_back_ribs',\n",
              " 'baklava',\n",
              " 'beef_carpaccio',\n",
              " 'beef_tartare',\n",
              " 'beet_salad',\n",
              " 'beignets',\n",
              " 'bibimbap',\n",
              " 'bread_pudding',\n",
              " 'breakfast_burrito']"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ],
      "source": [
        "# Get class names\n",
        "class_names = ds_info.features[\"label\"].names\n",
        "class_names[:10]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cx5G1-o66W51"
      },
      "source": [
        "# exploring the food 101 data from tensorflow dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "mOcaqR6S6gP5"
      },
      "outputs": [],
      "source": [
        "train_one_sample=train_data.take(1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qrDpBxPdCLpn",
        "outputId": "7de19c6a-32dc-49e5-c7a3-55b2e407fe29"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<TakeDataset element_spec=(TensorSpec(shape=(None, None, 3), dtype=tf.uint8, name=None), TensorSpec(shape=(), dtype=tf.int64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ],
      "source": [
        "train_one_sample"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "esLxznQZCQPI",
        "outputId": "2ed371c6-d2de-4bbd-962a-b1f99ada1eb8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "  image shape: (512, 512, 3)\n",
            "  image datatype: <dtype: 'uint8'>\n",
            "  target class from food101:64\n",
            "  class name: miso_soup\n",
            "\n",
            "       \n"
          ]
        }
      ],
      "source": [
        "for image,label in train_one_sample:\n",
        "  print(f\"\"\"\n",
        "  image shape: {image.shape}\n",
        "  image datatype: {image.dtype}\n",
        "  target class from food101:{label}\n",
        "  class name: {class_names[label.numpy()]}\n",
        "\n",
        "       \"\"\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zQ2O_BGwEM3M",
        "outputId": "1ecc8bc0-9c56-488e-f8b5-85c70d840a56"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(512, 512, 3), dtype=uint8, numpy=\n",
              "array([[[71, 60, 56],\n",
              "        [70, 59, 55],\n",
              "        [70, 59, 55],\n",
              "        ...,\n",
              "        [65, 60, 67],\n",
              "        [65, 60, 67],\n",
              "        [64, 59, 66]],\n",
              "\n",
              "       [[62, 51, 47],\n",
              "        [64, 53, 49],\n",
              "        [66, 55, 51],\n",
              "        ...,\n",
              "        [65, 60, 67],\n",
              "        [65, 60, 67],\n",
              "        [64, 59, 66]],\n",
              "\n",
              "       [[61, 50, 46],\n",
              "        [65, 54, 50],\n",
              "        [68, 57, 53],\n",
              "        ...,\n",
              "        [66, 61, 68],\n",
              "        [65, 60, 67],\n",
              "        [63, 58, 65]],\n",
              "\n",
              "       ...,\n",
              "\n",
              "       [[63, 64, 66],\n",
              "        [64, 65, 67],\n",
              "        [67, 67, 69],\n",
              "        ...,\n",
              "        [45, 42, 49],\n",
              "        [45, 42, 49],\n",
              "        [45, 42, 49]],\n",
              "\n",
              "       [[64, 64, 66],\n",
              "        [64, 64, 66],\n",
              "        [65, 65, 67],\n",
              "        ...,\n",
              "        [44, 41, 48],\n",
              "        [44, 41, 48],\n",
              "        [44, 41, 48]],\n",
              "\n",
              "       [[62, 62, 64],\n",
              "        [62, 62, 64],\n",
              "        [62, 62, 64],\n",
              "        ...,\n",
              "        [43, 40, 47],\n",
              "        [42, 39, 46],\n",
              "        [42, 39, 46]]], dtype=uint8)>"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ],
      "source": [
        "\n",
        "image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QsdOEdQjG0rk",
        "outputId": "adbdf4d2-2e49-4c56-f6ea-98c803636d1e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(), dtype=uint8, numpy=0>,\n",
              " <tf.Tensor: shape=(), dtype=uint8, numpy=255>)"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "tf.reduce_min(image),tf.reduce_max(image)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "y-9NqG1OHTxj",
        "outputId": "6cef26cb-1d4f-463e-afac-bf8623dbd45e"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# plot an image from tensorflow dataset\n",
        "import matplotlib.pyplot as plt\n",
        "plt.imshow(image)\n",
        "plt.title(class_names[label.numpy()])\n",
        "plt.axis=False;"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w26BFRN7H8pp"
      },
      "source": [
        "## create preprocessing function for our datasets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "5In08nRQJCeS"
      },
      "outputs": [],
      "source": [
        "\n",
        "def preprocess_img(image,label,img_shape=224):\n",
        "  image=tf.image.resize(image,[img_shape,img_shape])\n",
        "  # image=image/255.\n",
        "  return tf.cast(image,tf.float32),label"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R5qbdXhsMMks",
        "outputId": "ce5fb043-751d-46d7-d3a0-e9e66ee0e0a2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Image before preprocessing:\n",
            " [[[12 13  7]\n",
            "  [12 13  7]\n",
            "  [13 14  8]\n",
            "  ...\n",
            "  [21 11  0]\n",
            "  [21 11  0]\n",
            "  [21 11  0]]\n",
            "\n",
            " [[12 13  7]\n",
            "  [11 12  6]\n",
            "  [11 12  6]\n",
            "  ...\n",
            "  [21 11  0]\n",
            "  [21 11  0]\n",
            "  [21 11  0]]]...,\n",
            "Shape: (512, 512, 3),\n",
            "Datatype: <dtype: 'uint8'>\n",
            "\n",
            "Image after preprocessing:\n",
            " [[[11.586735   12.586735    6.586735  ]\n",
            "  [11.714286   12.714286    6.714286  ]\n",
            "  [ 8.857142    9.857142    4.8571424 ]\n",
            "  ...\n",
            "  [20.714308   11.142836    1.2857144 ]\n",
            "  [20.668371   10.668372    0.        ]\n",
            "  [21.         11.          0.        ]]\n",
            "\n",
            " [[ 2.3571415   3.3571415   0.1428566 ]\n",
            "  [ 3.1530607   4.153061    0.07653028]\n",
            "  [ 3.0561223   4.0561223   0.        ]\n",
            "  ...\n",
            "  [26.071407   18.071407    7.0714073 ]\n",
            "  [24.785702   14.785702    4.7857018 ]\n",
            "  [22.499966   12.499966    2.4999657 ]]]...,\n",
            "Shape: (224, 224, 3),\n",
            "Datatype: <dtype: 'float32'>\n"
          ]
        }
      ],
      "source": [
        "# Preprocess a single sample image and check the outputs\n",
        "preprocessed_img = preprocess_img(image, label)[0]\n",
        "print(f\"Image before preprocessing:\\n {image[:2]}...,\\nShape: {image.shape},\\nDatatype: {image.dtype}\\n\")\n",
        "print(f\"Image after preprocessing:\\n {preprocessed_img[:2]}...,\\nShape: {preprocessed_img.shape},\\nDatatype: {preprocessed_img.dtype}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gRhPI5VQNbVz"
      },
      "source": [
        "# batch and prepare dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "8J0P8SEQFrQk"
      },
      "outputs": [],
      "source": [
        "#preprocess\n",
        "train_data=train_data.map(map_func=preprocess_img,num_parallel_calls=tf.data.AUTOTUNE)\n",
        "train_data=train_data.shuffle(buffer_size=1000).batch(batch_size=32).# Creating learning rate reduction callback\n",
        "reduce_lr = tf.keras.callbacks.ReduceLROnPlateau(monitor=\"val_loss\",\n",
        "                                                 factor=0.2, # multiply the learning rate by 0.2 (reduce by 5x)\n",
        "                                                 patience=2,\n",
        "                                                 verbose=1, # print out when learning rate goes down\n",
        "                                                 min_lr=1e-7)\n",
        "#map preprocessing functon to test data\n",
        "\n",
        "test_data=test_data.map(map_func=preprocess_img,num_parallel_calls=tf.data.AUTOTUNE).batch(32).prefetch(tf.data.AUTOTUNE)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yr8-xFSkKSMY",
        "outputId": "87699a48-d761-49f7-998c-2fa976643a83"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<PrefetchDataset element_spec=(TensorSpec(shape=(None, 224, 224, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))>,\n",
              " <PrefetchDataset element_spec=(TensorSpec(shape=(None, 224, 224, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))>)"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "train_data,test_data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "5Tm-1jyeLFRY"
      },
      "outputs": [],
      "source": [
        "# create modelling callbacks\n",
        "from helper_functions import create_tensorboard_callback\n",
        "\n",
        "#create model checkpoint callbacks\n",
        "checkpoint_path=\"model_checkpoints/cp.ckpt\"\n",
        "model_checkpoint=tf.keras.callbacks.ModelCheckpoint(checkpoint_path,moniter=\"val_acc\",\n",
        "                                                    save_best_only=True,\n",
        "                                                    save_weights_only=True,\n",
        "                                                    verbose=0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "NMHqv2zuR117"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras import mixed_precision"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "Dc9X8F41HILY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3e4b5441-0310-4bae-8fed-aef00cb4392b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TkHtF_F-Y2mN",
        "outputId": "0aa7bd3c-2263-4ce8-f65d-71fca0adb12f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Policy \"mixed_float16\">"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "mixed_precision.set_global_policy(\"mixed_float16\")\n",
        "mixed_precision.global_policy()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DUaxNrsBZb2T"
      },
      "source": [
        "#build feature exxtraction model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "id": "buXenezabuDW"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras import layers\n",
        "from tensorflow.keras.layers.experimental import preprocessing\n",
        "\n",
        "input_shape=(224,224,3)\n",
        "base_model=tf.keras.applications.EfficientNetB0(include_top=False)\n",
        "base_model.trainable=False\n",
        "inputs=layers.Input(shape=input_shape,name=\"input_layer\")\n",
        "x=base_model(inputs,training=False)\n",
        "x=layers.GlobalAveragePooling2D()(x)\n",
        "x=layers.Dense(len(class_names))(x)\n",
        "outputs=layers.Activation(\"softmax\",dtype=tf.float32,name=\"softmax_float32\")(x)\n",
        "model=tf.keras.Model(inputs,outputs)\n",
        "#compile\n",
        "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
        "              optimizer=tf.keras.optimizers.Adam(),\n",
        "              metrics=[\"accuracy\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DNDJow0Vd4hc",
        "outputId": "9804a5c4-9f4b-4830-b828-7af3dd8e7ddd"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_layer (InputLayer)    [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " efficientnetb0 (Functional)  (None, None, None, 1280)  4049571  \n",
            "                                                                 \n",
            " global_average_pooling2d_1   (None, 1280)             0         \n",
            " (GlobalAveragePooling2D)                                        \n",
            "                                                                 \n",
            " dense (Dense)               (None, 101)               129381    \n",
            "                                                                 \n",
            " softmax_float32 (Activation  (None, 101)              0         \n",
            " )                                                               \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,178,952\n",
            "Trainable params: 129,381\n",
            "Non-trainable params: 4,049,571\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FI0Dwoeteeru",
        "outputId": "0839c435-6fd8-4004-c9df-9e0fbffdc7df"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "input_layer True float32 <Policy \"float32\">\n",
            "efficientnetb0 False float32 <Policy \"mixed_float16\">\n",
            "global_average_pooling2d_1 True float32 <Policy \"mixed_float16\">\n",
            "dense True float32 <Policy \"mixed_float16\">\n",
            "softmax_float32 True float32 <Policy \"float32\">\n"
          ]
        }
      ],
      "source": [
        "for layer in model.layers:\n",
        "  print(layer.name,layer.trainable,layer.dtype,layer.dtype_policy)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-lggXrzJf4Ef",
        "outputId": "32e1bc23-e731-4b04-aebe-d5dee888f764"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 input_2 False float32 <Policy \"float32\">\n",
            "1 rescaling_1 False float32 <Policy \"mixed_float16\">\n",
            "2 normalization_1 False float32 <Policy \"mixed_float16\">\n",
            "3 stem_conv_pad False float32 <Policy \"mixed_float16\">\n",
            "4 stem_conv False float32 <Policy \"mixed_float16\">\n",
            "5 stem_bn False float32 <Policy \"mixed_float16\">\n",
            "6 stem_activation False float32 <Policy \"mixed_float16\">\n",
            "7 block1a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "8 block1a_bn False float32 <Policy \"mixed_float16\">\n",
            "9 block1a_activation False float32 <Policy \"mixed_float16\">\n",
            "10 block1a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "11 block1a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "12 block1a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "13 block1a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "14 block1a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "15 block1a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "16 block1a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "17 block2a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "18 block2a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "19 block2a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "20 block2a_dwconv_pad False float32 <Policy \"mixed_float16\">\n",
            "21 block2a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "22 block2a_bn False float32 <Policy \"mixed_float16\">\n",
            "23 block2a_activation False float32 <Policy \"mixed_float16\">\n",
            "24 block2a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "25 block2a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "26 block2a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "27 block2a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "28 block2a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "29 block2a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "30 block2a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "31 block2b_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "32 block2b_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "33 block2b_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "34 block2b_dwconv False float32 <Policy \"mixed_float16\">\n",
            "35 block2b_bn False float32 <Policy \"mixed_float16\">\n",
            "36 block2b_activation False float32 <Policy \"mixed_float16\">\n",
            "37 block2b_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "38 block2b_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "39 block2b_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "40 block2b_se_expand False float32 <Policy \"mixed_float16\">\n",
            "41 block2b_se_excite False float32 <Policy \"mixed_float16\">\n",
            "42 block2b_project_conv False float32 <Policy \"mixed_float16\">\n",
            "43 block2b_project_bn False float32 <Policy \"mixed_float16\">\n",
            "44 block2b_drop False float32 <Policy \"mixed_float16\">\n",
            "45 block2b_add False float32 <Policy \"mixed_float16\">\n",
            "46 block3a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "47 block3a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "48 block3a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "49 block3a_dwconv_pad False float32 <Policy \"mixed_float16\">\n",
            "50 block3a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "51 block3a_bn False float32 <Policy \"mixed_float16\">\n",
            "52 block3a_activation False float32 <Policy \"mixed_float16\">\n",
            "53 block3a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "54 block3a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "55 block3a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "56 block3a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "57 block3a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "58 block3a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "59 block3a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "60 block3b_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "61 block3b_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "62 block3b_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "63 block3b_dwconv False float32 <Policy \"mixed_float16\">\n",
            "64 block3b_bn False float32 <Policy \"mixed_float16\">\n",
            "65 block3b_activation False float32 <Policy \"mixed_float16\">\n",
            "66 block3b_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "67 block3b_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "68 block3b_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "69 block3b_se_expand False float32 <Policy \"mixed_float16\">\n",
            "70 block3b_se_excite False float32 <Policy \"mixed_float16\">\n",
            "71 block3b_project_conv False float32 <Policy \"mixed_float16\">\n",
            "72 block3b_project_bn False float32 <Policy \"mixed_float16\">\n",
            "73 block3b_drop False float32 <Policy \"mixed_float16\">\n",
            "74 block3b_add False float32 <Policy \"mixed_float16\">\n",
            "75 block4a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "76 block4a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "77 block4a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "78 block4a_dwconv_pad False float32 <Policy \"mixed_float16\">\n",
            "79 block4a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "80 block4a_bn False float32 <Policy \"mixed_float16\">\n",
            "81 block4a_activation False float32 <Policy \"mixed_float16\">\n",
            "82 block4a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "83 block4a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "84 block4a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "85 block4a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "86 block4a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "87 block4a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "88 block4a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "89 block4b_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "90 block4b_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "91 block4b_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "92 block4b_dwconv False float32 <Policy \"mixed_float16\">\n",
            "93 block4b_bn False float32 <Policy \"mixed_float16\">\n",
            "94 block4b_activation False float32 <Policy \"mixed_float16\">\n",
            "95 block4b_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "96 block4b_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "97 block4b_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "98 block4b_se_expand False float32 <Policy \"mixed_float16\">\n",
            "99 block4b_se_excite False float32 <Policy \"mixed_float16\">\n",
            "100 block4b_project_conv False float32 <Policy \"mixed_float16\">\n",
            "101 block4b_project_bn False float32 <Policy \"mixed_float16\">\n",
            "102 block4b_drop False float32 <Policy \"mixed_float16\">\n",
            "103 block4b_add False float32 <Policy \"mixed_float16\">\n",
            "104 block4c_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "105 block4c_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "106 block4c_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "107 block4c_dwconv False float32 <Policy \"mixed_float16\">\n",
            "108 block4c_bn False float32 <Policy \"mixed_float16\">\n",
            "109 block4c_activation False float32 <Policy \"mixed_float16\">\n",
            "110 block4c_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "111 block4c_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "112 block4c_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "113 block4c_se_expand False float32 <Policy \"mixed_float16\">\n",
            "114 block4c_se_excite False float32 <Policy \"mixed_float16\">\n",
            "115 block4c_project_conv False float32 <Policy \"mixed_float16\">\n",
            "116 block4c_project_bn False float32 <Policy \"mixed_float16\">\n",
            "117 block4c_drop False float32 <Policy \"mixed_float16\">\n",
            "118 block4c_add False float32 <Policy \"mixed_float16\">\n",
            "119 block5a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "120 block5a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "121 block5a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "122 block5a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "123 block5a_bn False float32 <Policy \"mixed_float16\">\n",
            "124 block5a_activation False float32 <Policy \"mixed_float16\">\n",
            "125 block5a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "126 block5a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "127 block5a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "128 block5a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "129 block5a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "130 block5a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "131 block5a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "132 block5b_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "133 block5b_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "134 block5b_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "135 block5b_dwconv False float32 <Policy \"mixed_float16\">\n",
            "136 block5b_bn False float32 <Policy \"mixed_float16\">\n",
            "137 block5b_activation False float32 <Policy \"mixed_float16\">\n",
            "138 block5b_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "139 block5b_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "140 block5b_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "141 block5b_se_expand False float32 <Policy \"mixed_float16\">\n",
            "142 block5b_se_excite False float32 <Policy \"mixed_float16\">\n",
            "143 block5b_project_conv False float32 <Policy \"mixed_float16\">\n",
            "144 block5b_project_bn False float32 <Policy \"mixed_float16\">\n",
            "145 block5b_drop False float32 <Policy \"mixed_float16\">\n",
            "146 block5b_add False float32 <Policy \"mixed_float16\">\n",
            "147 block5c_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "148 block5c_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "149 block5c_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "150 block5c_dwconv False float32 <Policy \"mixed_float16\">\n",
            "151 block5c_bn False float32 <Policy \"mixed_float16\">\n",
            "152 block5c_activation False float32 <Policy \"mixed_float16\">\n",
            "153 block5c_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "154 block5c_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "155 block5c_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "156 block5c_se_expand False float32 <Policy \"mixed_float16\">\n",
            "157 block5c_se_excite False float32 <Policy \"mixed_float16\">\n",
            "158 block5c_project_conv False float32 <Policy \"mixed_float16\">\n",
            "159 block5c_project_bn False float32 <Policy \"mixed_float16\">\n",
            "160 block5c_drop False float32 <Policy \"mixed_float16\">\n",
            "161 block5c_add False float32 <Policy \"mixed_float16\">\n",
            "162 block6a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "163 block6a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "164 block6a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "165 block6a_dwconv_pad False float32 <Policy \"mixed_float16\">\n",
            "166 block6a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "167 block6a_bn False float32 <Policy \"mixed_float16\">\n",
            "168 block6a_activation False float32 <Policy \"mixed_float16\">\n",
            "169 block6a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "170 block6a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "171 block6a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "172 block6a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "173 block6a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "174 block6a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "175 block6a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "176 block6b_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "177 block6b_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "178 block6b_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "179 block6b_dwconv False float32 <Policy \"mixed_float16\">\n",
            "180 block6b_bn False float32 <Policy \"mixed_float16\">\n",
            "181 block6b_activation False float32 <Policy \"mixed_float16\">\n",
            "182 block6b_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "183 block6b_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "184 block6b_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "185 block6b_se_expand False float32 <Policy \"mixed_float16\">\n",
            "186 block6b_se_excite False float32 <Policy \"mixed_float16\">\n",
            "187 block6b_project_conv False float32 <Policy \"mixed_float16\">\n",
            "188 block6b_project_bn False float32 <Policy \"mixed_float16\">\n",
            "189 block6b_drop False float32 <Policy \"mixed_float16\">\n",
            "190 block6b_add False float32 <Policy \"mixed_float16\">\n",
            "191 block6c_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "192 block6c_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "193 block6c_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "194 block6c_dwconv False float32 <Policy \"mixed_float16\">\n",
            "195 block6c_bn False float32 <Policy \"mixed_float16\">\n",
            "196 block6c_activation False float32 <Policy \"mixed_float16\">\n",
            "197 block6c_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "198 block6c_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "199 block6c_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "200 block6c_se_expand False float32 <Policy \"mixed_float16\">\n",
            "201 block6c_se_excite False float32 <Policy \"mixed_float16\">\n",
            "202 block6c_project_conv False float32 <Policy \"mixed_float16\">\n",
            "203 block6c_project_bn False float32 <Policy \"mixed_float16\">\n",
            "204 block6c_drop False float32 <Policy \"mixed_float16\">\n",
            "205 block6c_add False float32 <Policy \"mixed_float16\">\n",
            "206 block6d_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "207 block6d_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "208 block6d_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "209 block6d_dwconv False float32 <Policy \"mixed_float16\">\n",
            "210 block6d_bn False float32 <Policy \"mixed_float16\">\n",
            "211 block6d_activation False float32 <Policy \"mixed_float16\">\n",
            "212 block6d_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "213 block6d_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "214 block6d_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "215 block6d_se_expand False float32 <Policy \"mixed_float16\">\n",
            "216 block6d_se_excite False float32 <Policy \"mixed_float16\">\n",
            "217 block6d_project_conv False float32 <Policy \"mixed_float16\">\n",
            "218 block6d_project_bn False float32 <Policy \"mixed_float16\">\n",
            "219 block6d_drop False float32 <Policy \"mixed_float16\">\n",
            "220 block6d_add False float32 <Policy \"mixed_float16\">\n",
            "221 block7a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "222 block7a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "223 block7a_expand_activation False float32 <Policy \"mixed_float16\">\n",
            "224 block7a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "225 block7a_bn False float32 <Policy \"mixed_float16\">\n",
            "226 block7a_activation False float32 <Policy \"mixed_float16\">\n",
            "227 block7a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "228 block7a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "229 block7a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "230 block7a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "231 block7a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "232 block7a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "233 block7a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "234 top_conv False float32 <Policy \"mixed_float16\">\n",
            "235 top_bn False float32 <Policy \"mixed_float16\">\n",
            "236 top_activation False float32 <Policy \"mixed_float16\">\n"
          ]
        }
      ],
      "source": [
        "for x,layer in enumerate(model.layers[1].layers):\n",
        "  print(x,layer.name,layer.trainable,layer.dtype,layer.dtype_policy)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OWqwnBJyjo-s"
      },
      "source": [
        "# fit the feature extraction model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eUoZHpJ-llvb",
        "outputId": "07aa83b5-b4d1-4609-e75e-b64f33bd707b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving TensorBoard log files to: training_logs/efficientnetb0_101_classes_all_data_feature_extract/20230726-113705\n",
            "Epoch 1/3\n",
            "2368/2368 [==============================] - 230s 94ms/step - loss: 1.8195 - accuracy: 0.5577 - val_loss: 1.2307 - val_accuracy: 0.6798\n",
            "Epoch 2/3\n",
            "2368/2368 [==============================] - 252s 105ms/step - loss: 1.2929 - accuracy: 0.6672 - val_loss: 1.1272 - val_accuracy: 0.6949\n",
            "Epoch 3/3\n",
            "2368/2368 [==============================] - 226s 94ms/step - loss: 1.1432 - accuracy: 0.7025 - val_loss: 1.0745 - val_accuracy: 0.7058\n"
          ]
        }
      ],
      "source": [
        "# Turn off all warnings except for errors\n",
        "tf.get_logger().setLevel('ERROR')\n",
        "\n",
        "# Fit the model with callbacks\n",
        "history_101_food_classes_feature_extract = model.fit(train_data,\n",
        "                                                     epochs=3,\n",
        "                                                     steps_per_epoch=len(train_data),\n",
        "                                                     validation_data=test_data,\n",
        "                                                     validation_steps=int(0.15 * len(test_data)),\n",
        "                                                     callbacks=[create_tensorboard_callback(\"training_logs\",\n",
        "                                                                                            \"efficientnetb0_101_classes_all_data_feature_extract\"),\n",
        "                                                                model_checkpoint])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xyy2aZx3nQHP",
        "outputId": "42fc8b2d-dcf5-4063-b94a-2df03da01f78"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "790/790 [==============================] - 62s 78ms/step - loss: 1.0878 - accuracy: 0.7065\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.0877867937088013, 0.7064950466156006]"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ],
      "source": [
        "results_feature_extract_model=model.evaluate(test_data)\n",
        "results_feature_extract_model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sGCOvHa8wuif"
      },
      "outputs": [],
      "source": [
        "# # Create save path to drive\n",
        "save_dir = \"drive/MyDrive/tensorflowmodels/food_vision/07_efficientnetb0_feature_extract_model_mixed_precision/\"\n",
        "os.makedirs(save_dir) # Make directory if it doesn't exist\n",
        "\n",
        "# # Save model\n",
        "model.save(save_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "2TqKdlRz6kwc"
      },
      "outputs": [],
      "source": [
        "#    Download the saved model from Google Storage\n",
        "model_path =\"drive/MyDrive/tensorflowmodels/food_vision/07_efficientnetb0_feature_extract_model_mixed_precision/\"\n",
        "loaded_model = tf.keras.models.load_model(model_path)\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for layer in loaded_model.layers:\n",
        "  print(layer.name,layer.dtype,layer.trainable,layer.dtype_policy)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3tkTWeuZ4Vnc",
        "outputId": "4c83cd0b-aabd-49c6-83c2-13ead44c9a33"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "input_layer float32 True <Policy \"float32\">\n",
            "efficientnetb0 float32 False <Policy \"mixed_float16\">\n",
            "global_average_pooling2d float32 True <Policy \"mixed_float16\">\n",
            "dense float32 True <Policy \"mixed_float16\">\n",
            "softmax_float32 float32 True <Policy \"float32\">\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for x,layer in enumerate(loaded_model.layers[1].layers[:20]):\n",
        "  print(x,layer.name,layer.trainable,layer.dtype,layer.dtype_policy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SaH4YxzG4hHP",
        "outputId": "fb5858a8-8b0a-4fd6-d14c-0755484bb8cf"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 input_1 True float32 <Policy \"float32\">\n",
            "1 rescaling False float32 <Policy \"mixed_float16\">\n",
            "2 normalization False float32 <Policy \"mixed_float16\">\n",
            "3 stem_conv_pad False float32 <Policy \"mixed_float16\">\n",
            "4 stem_conv False float32 <Policy \"mixed_float16\">\n",
            "5 stem_bn False float32 <Policy \"mixed_float16\">\n",
            "6 stem_activation False float32 <Policy \"mixed_float16\">\n",
            "7 block1a_dwconv False float32 <Policy \"mixed_float16\">\n",
            "8 block1a_bn False float32 <Policy \"mixed_float16\">\n",
            "9 block1a_activation False float32 <Policy \"mixed_float16\">\n",
            "10 block1a_se_squeeze False float32 <Policy \"mixed_float16\">\n",
            "11 block1a_se_reshape False float32 <Policy \"mixed_float16\">\n",
            "12 block1a_se_reduce False float32 <Policy \"mixed_float16\">\n",
            "13 block1a_se_expand False float32 <Policy \"mixed_float16\">\n",
            "14 block1a_se_excite False float32 <Policy \"mixed_float16\">\n",
            "15 block1a_project_conv False float32 <Policy \"mixed_float16\">\n",
            "16 block1a_project_bn False float32 <Policy \"mixed_float16\">\n",
            "17 block2a_expand_conv False float32 <Policy \"mixed_float16\">\n",
            "18 block2a_expand_bn False float32 <Policy \"mixed_float16\">\n",
            "19 block2a_expand_activation False float32 <Policy \"mixed_float16\">\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        " #Check loaded model performance (this should be the same as results_feature_extract_model)\n",
        "results_loaded_saved_model = loaded_model.evaluate(test_data)\n",
        "results_loaded_saved_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2PcrjzNO9_GM",
        "outputId": "004424f2-fca7-48ea-9709-49d73c9b22be"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "790/790 [==============================] - 74s 80ms/step - loss: 1.0901 - accuracy: 0.7035\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.090087890625, 0.7035247683525085]"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "np.isclose(results_feature_extract_model, results_loaded_saved_model).all()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qh2x8hlY-Ebl",
        "outputId": "a11d3375-4319-4792-d6bd-94b513d45b03"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## prepare our model layers for finetuning"
      ],
      "metadata": {
        "id": "P2POJVSUAVlM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Download the saved model from Google Storage\n",
        "!wget https://storage.googleapis.com/ztm_tf_course/food_vision/07_efficientnetb0_feature_extract_model_mixed_precision.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g-gdwW8UAcW_",
        "outputId": "d6733dca-ea35-42ac-b042-2698d9174482"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-07-26 11:51:27--  https://storage.googleapis.com/ztm_tf_course/food_vision/07_efficientnetb0_feature_extract_model_mixed_precision.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 108.177.13.128, 74.125.26.128, 172.217.193.128, ...\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|108.177.13.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 16976857 (16M) [application/zip]\n",
            "Saving to: ‘07_efficientnetb0_feature_extract_model_mixed_precision.zip.1’\n",
            "\n",
            "\r          07_effici   0%[                    ]       0  --.-KB/s               \r07_efficientnetb0_f 100%[===================>]  16.19M  --.-KB/s    in 0.08s   \n",
            "\n",
            "2023-07-26 11:51:27 (195 MB/s) - ‘07_efficientnetb0_feature_extract_model_mixed_precision.zip.1’ saved [16976857/16976857]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Unzip the SavedModel downloaded from Google Stroage\n",
        "!mkdir downloaded_gs_model # create new dir to store downloaded feature extraction model\n",
        "!unzip 07_efficientnetb0_feature_extract_model_mixed_precision.zip -d downloaded_gs_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DOKtuX3TAf_c",
        "outputId": "d6cb0623-323e-4a15-8f16-1d71a848e2cd"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  07_efficientnetb0_feature_extract_model_mixed_precision.zip\n",
            "   creating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/\n",
            "   creating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/variables/\n",
            "  inflating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/variables/variables.data-00000-of-00001  \n",
            "  inflating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/variables/variables.index  \n",
            "  inflating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/saved_model.pb  \n",
            "   creating: downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision/assets/\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load and evaluate downloaded GS model\n",
        "loaded_gs_model = tf.keras.models.load_model(\"downloaded_gs_model/07_efficientnetb0_feature_extract_model_mixed_precision\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lzkrYEptAvEz",
        "outputId": "8114ab81-7b98-4c9b-f082-a23af8ad417a"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Importing a function (__inference_block1a_activation_layer_call_and_return_conditional_losses_158253) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_activation_layer_call_and_return_conditional_losses_191539) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_expand_activation_layer_call_and_return_conditional_losses_196076) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_activation_layer_call_and_return_conditional_losses_195780) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_activation_layer_call_and_return_conditional_losses_196153) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_model_layer_call_and_return_conditional_losses_180010) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_stem_activation_layer_call_and_return_conditional_losses_191136) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_expand_activation_layer_call_and_return_conditional_losses_160354) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_expand_activation_layer_call_and_return_conditional_losses_195703) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_expand_activation_layer_call_and_return_conditional_losses_159392) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_activation_layer_call_and_return_conditional_losses_191213) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_se_reduce_layer_call_and_return_conditional_losses_193678) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_se_reduce_layer_call_and_return_conditional_losses_194051) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_expand_activation_layer_call_and_return_conditional_losses_158768) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_se_reduce_layer_call_and_return_conditional_losses_191907) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_se_reduce_layer_call_and_return_conditional_losses_162720) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_activation_layer_call_and_return_conditional_losses_194708) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_se_reduce_layer_call_and_return_conditional_losses_196195) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_expand_activation_layer_call_and_return_conditional_losses_194258) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_188022) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_activation_layer_call_and_return_conditional_losses_161995) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_183149) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_activation_layer_call_and_return_conditional_losses_158824) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_activation_layer_call_and_return_conditional_losses_159787) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_expand_activation_layer_call_and_return_conditional_losses_158482) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_se_reduce_layer_call_and_return_conditional_losses_158588) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_se_reduce_layer_call_and_return_conditional_losses_195449) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_se_reduce_layer_call_and_return_conditional_losses_194377) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_expand_activation_layer_call_and_return_conditional_losses_162615) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_activation_layer_call_and_return_conditional_losses_192238) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_se_reduce_layer_call_and_return_conditional_losses_160121) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_expand_activation_layer_call_and_return_conditional_losses_192860) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_activation_layer_call_and_return_conditional_losses_191865) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_expand_activation_layer_call_and_return_conditional_losses_160016) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_se_reduce_layer_call_and_return_conditional_losses_194750) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_169029) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_170771) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_activation_layer_call_and_return_conditional_losses_159448) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_expand_activation_layer_call_and_return_conditional_losses_194631) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_se_reduce_layer_call_and_return_conditional_losses_192979) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_activation_layer_call_and_return_conditional_losses_193263) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_expand_activation_layer_call_and_return_conditional_losses_160977) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_expand_activation_layer_call_and_return_conditional_losses_162953) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_se_reduce_layer_call_and_return_conditional_losses_159836) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_se_reduce_layer_call_and_return_conditional_losses_191581) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_activation_layer_call_and_return_conditional_losses_158539) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_se_reduce_layer_call_and_return_conditional_losses_162382) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_expand_activation_layer_call_and_return_conditional_losses_196449) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_top_activation_layer_call_and_return_conditional_losses_163238) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_expand_activation_layer_call_and_return_conditional_losses_162277) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_expand_activation_layer_call_and_return_conditional_losses_192487) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_se_reduce_layer_call_and_return_conditional_losses_191255) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_activation_layer_call_and_return_conditional_losses_163009) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_activation_layer_call_and_return_conditional_losses_194335) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_expand_activation_layer_call_and_return_conditional_losses_193559) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_expand_activation_layer_call_and_return_conditional_losses_159730) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_se_reduce_layer_call_and_return_conditional_losses_161759) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_expand_activation_layer_call_and_return_conditional_losses_192161) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_se_reduce_layer_call_and_return_conditional_losses_193305) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_activation_layer_call_and_return_conditional_losses_160748) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_activation_layer_call_and_return_conditional_losses_161371) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_activation_layer_call_and_return_conditional_losses_192937) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_se_reduce_layer_call_and_return_conditional_losses_196568) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_expand_activation_layer_call_and_return_conditional_losses_191788) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_expand_activation_layer_call_and_return_conditional_losses_159106) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_se_reduce_layer_call_and_return_conditional_losses_159497) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_expand_activation_layer_call_and_return_conditional_losses_161315) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_184891) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_model_layer_call_and_return_conditional_losses_178256) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_activation_layer_call_and_return_conditional_losses_161710) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_expand_activation_layer_call_and_return_conditional_losses_161653) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_se_reduce_layer_call_and_return_conditional_losses_159212) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_stem_activation_layer_call_and_return_conditional_losses_158197) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_189764) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_se_reduce_layer_call_and_return_conditional_losses_192606) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_activation_layer_call_and_return_conditional_losses_195081) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_activation_layer_call_and_return_conditional_losses_162333) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_se_reduce_layer_call_and_return_conditional_losses_160797) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_activation_layer_call_and_return_conditional_losses_194009) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_se_reduce_layer_call_and_return_conditional_losses_195822) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_activation_layer_call_and_return_conditional_losses_161033) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_expand_activation_layer_call_and_return_conditional_losses_195330) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_activation_layer_call_and_return_conditional_losses_159163) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_se_reduce_layer_call_and_return_conditional_losses_160459) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_activation_layer_call_and_return_conditional_losses_195407) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_se_reduce_layer_call_and_return_conditional_losses_163058) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_se_reduce_layer_call_and_return_conditional_losses_192280) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_activation_layer_call_and_return_conditional_losses_162671) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference__wrapped_model_152628) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_se_reduce_layer_call_and_return_conditional_losses_162044) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_se_reduce_layer_call_and_return_conditional_losses_158873) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_activation_layer_call_and_return_conditional_losses_160410) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_expand_activation_layer_call_and_return_conditional_losses_195004) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_activation_layer_call_and_return_conditional_losses_192564) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_se_reduce_layer_call_and_return_conditional_losses_161082) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_se_reduce_layer_call_and_return_conditional_losses_161420) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_activation_layer_call_and_return_conditional_losses_193636) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_top_activation_layer_call_and_return_conditional_losses_196775) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_activation_layer_call_and_return_conditional_losses_160072) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_expand_activation_layer_call_and_return_conditional_losses_161939) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_expand_activation_layer_call_and_return_conditional_losses_193932) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_expand_activation_layer_call_and_return_conditional_losses_193186) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_se_reduce_layer_call_and_return_conditional_losses_158302) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_se_reduce_layer_call_and_return_conditional_losses_195123) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_expand_activation_layer_call_and_return_conditional_losses_191462) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_activation_layer_call_and_return_conditional_losses_196526) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_expand_activation_layer_call_and_return_conditional_losses_160692) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Get a summary of our downloaded model\n",
        "loaded_gs_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mAPcEKp-Av1b",
        "outputId": "eb8d8dcf-72e1-4b85-e6e2-798dd9206b53"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_layer (InputLayer)    [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " efficientnetb0 (Functional)  (None, None, None, 1280)  4049571  \n",
            "                                                                 \n",
            " pooling_layer (GlobalAverag  (None, 1280)             0         \n",
            " ePooling2D)                                                     \n",
            "                                                                 \n",
            " dense (Dense)               (None, 101)               129381    \n",
            "                                                                 \n",
            " softmax_float32 (Activation  (None, 101)              0         \n",
            " )                                                               \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,178,952\n",
            "Trainable params: 129,381\n",
            "Non-trainable params: 4,049,571\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# How does the loaded model perform?\n",
        "results_loaded_gs_model = loaded_gs_model.evaluate(test_data)\n",
        "results_loaded_gs_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a23vNsfHA2zC",
        "outputId": "2388201c-9b4b-40ff-933e-392f3c0c6056"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "790/790 [==============================] - 74s 90ms/step - loss: 1.0881 - accuracy: 0.7066\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.088100552558899, 0.7066138386726379]"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Are any of the layers in our model frozen?\n",
        "for layer in loaded_gs_model.layers:\n",
        "    layer.trainable = True # set all layers to trainable\n",
        "    print(layer.name, layer.trainable, layer.dtype, layer.dtype_policy) # make sure loaded model is using mixed precision dtype_policy (\"mixed_float16)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0M86GnlqA3SO",
        "outputId": "45e2e465-d037-4b62-afe9-3c89102636fa"
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "input_layer True float32 <Policy \"float32\">\n",
            "efficientnetb0 True float32 <Policy \"mixed_float16\">\n",
            "pooling_layer True float32 <Policy \"mixed_float16\">\n",
            "dense True float32 <Policy \"mixed_float16\">\n",
            "softmax_float32 True float32 <Policy \"float32\">\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check the layers in the base model and see what dtype policy they're using\n",
        "for layer in loaded_gs_model.layers[1].layers[:20]:\n",
        "    print(layer.name, layer.trainable, layer.dtype, layer.dtype_policy)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aIPS2ipRBFrH",
        "outputId": "556fb8e9-3245-4cb4-c1b6-611b885f81b1"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "input_1 True float32 <Policy \"float32\">\n",
            "rescaling True float32 <Policy \"mixed_float16\">\n",
            "normalization True float32 <Policy \"float32\">\n",
            "stem_conv_pad True float32 <Policy \"mixed_float16\">\n",
            "stem_conv True float32 <Policy \"mixed_float16\">\n",
            "stem_bn True float32 <Policy \"mixed_float16\">\n",
            "stem_activation True float32 <Policy \"mixed_float16\">\n",
            "block1a_dwconv True float32 <Policy \"mixed_float16\">\n",
            "block1a_bn True float32 <Policy \"mixed_float16\">\n",
            "block1a_activation True float32 <Policy \"mixed_float16\">\n",
            "block1a_se_squeeze True float32 <Policy \"mixed_float16\">\n",
            "block1a_se_reshape True float32 <Policy \"mixed_float16\">\n",
            "block1a_se_reduce True float32 <Policy \"mixed_float16\">\n",
            "block1a_se_expand True float32 <Policy \"mixed_float16\">\n",
            "block1a_se_excite True float32 <Policy \"mixed_float16\">\n",
            "block1a_project_conv True float32 <Policy \"mixed_float16\">\n",
            "block1a_project_bn True float32 <Policy \"mixed_float16\">\n",
            "block2a_expand_conv True float32 <Policy \"mixed_float16\">\n",
            "block2a_expand_bn True float32 <Policy \"mixed_float16\">\n",
            "block2a_expand_activation True float32 <Policy \"mixed_float16\">\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Setup EarlyStopping callback to stop training if model's val_loss doesn't improve for 3 epochs\n",
        "early_stopping = tf.keras.callbacks.EarlyStopping(monitor=\"val_loss\", # watch the val loss metric\n",
        "                                                  patience=3) # if val loss decreases for 3 epochs in a row, stop training\n",
        "\n",
        "# Create ModelCheckpoint callback to save best model during fine-tuning\n",
        "checkpoint_path = \"fine_tune_checkpoints/\"\n",
        "model_checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_path,\n",
        "                                                      save_best_only=True,\n",
        "                                                      monitor=\"val_loss\")"
      ],
      "metadata": {
        "id": "Hj3l3zQ4BkTE"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating learning rate reduction callback\n",
        "reduce_lr = tf.keras.callbacks.ReduceLROnPlateau(monitor=\"val_loss\",\n",
        "                                                 factor=0.2, # multiply the learning rate by 0.2 (reduce by 5x)\n",
        "                                                 patience=2,\n",
        "                                                 verbose=1, # print out when learning rate goes down\n",
        "                                                 min_lr=1e-7)"
      ],
      "metadata": {
        "id": "tOXI3JygCldU"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile the model\n",
        "loaded_gs_model.compile(loss=\"sparse_categorical_crossentropy\", # sparse_categorical_crossentropy for labels that are *not* one-hot\n",
        "                        optimizer=tf.keras.optimizers.Adam(0.0001), # 10x lower learning rate than the default\n",
        "                        metrics=[\"accuracy\"])"
      ],
      "metadata": {
        "id": "dB7zUkwKEVvb"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Start to fine-tune (all layers)\n",
        "history_101_food_classes_all_data_fine_tune = loaded_gs_model.fit(train_data,\n",
        "                                                        epochs=100, # fine-tune for a maximum of 100 epochs\n",
        "                                                        steps_per_epoch=len(train_data),\n",
        "                                                        validation_data=test_data,\n",
        "                                                        validation_steps=int(0.15 * len(test_data)), # validation during training on 15% of test data\n",
        "                                                        callbacks=[create_tensorboard_callback(\"training_logs\", \"efficientb0_101_classes_all_data_fine_tuning\"), # track the model training logs\n",
        "                                                                   model_checkpoint, # save only the best model during training\n",
        "                                                                   early_stopping, # stop model after X epochs of no improvements\n",
        "                                                                   reduce_lr]) # reduce the learning rate after X epochs of no improvements"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pf9uSz5LExC5",
        "outputId": "abdbb12e-91ca-4641-9909-d6ddfa46b961"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving TensorBoard log files to: training_logs/efficientb0_101_classes_all_data_fine_tuning/20230726-120244\n",
            "Epoch 1/100\n",
            "2368/2368 [==============================] - 452s 180ms/step - loss: 0.9228 - accuracy: 0.7531 - val_loss: 0.7870 - val_accuracy: 0.7802 - lr: 1.0000e-04\n",
            "Epoch 2/100\n",
            "2368/2368 [==============================] - 388s 163ms/step - loss: 0.5777 - accuracy: 0.8393 - val_loss: 0.7999 - val_accuracy: 0.7805 - lr: 1.0000e-04\n",
            "Epoch 3/100\n",
            "2368/2368 [==============================] - ETA: 0s - loss: 0.3290 - accuracy: 0.9069\n",
            "Epoch 3: ReduceLROnPlateau reducing learning rate to 1.9999999494757503e-05.\n",
            "2368/2368 [==============================] - 398s 166ms/step - loss: 0.3290 - accuracy: 0.9069 - val_loss: 0.8368 - val_accuracy: 0.7926 - lr: 1.0000e-04\n",
            "Epoch 4/100\n",
            "2368/2368 [==============================] - 389s 162ms/step - loss: 0.0847 - accuracy: 0.9797 - val_loss: 0.9548 - val_accuracy: 0.7982 - lr: 2.0000e-05\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# # Save model to Google Drive (optional)\n",
        "loaded_gs_model.save(\"/content/drive/MyDrive/tensorflowmodels/food_vision/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/\")"
      ],
      "metadata": {
        "id": "z4lXWmZiNTOh"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Download and evaluate fine-tuned model from Google Storage\n",
        "!wget https://storage.googleapis.com/ztm_tf_course/food_vision/07_efficientnetb0_fine_tuned_101_classes_mixed_precision.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nsp_Rf2TNqME",
        "outputId": "0f3a0999-1109-46ff-e5c7-61c66901cb3a"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-07-26 12:40:10--  https://storage.googleapis.com/ztm_tf_course/food_vision/07_efficientnetb0_fine_tuned_101_classes_mixed_precision.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 173.194.217.128, 173.194.218.128, 108.177.11.128, ...\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|173.194.217.128|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 46790356 (45M) [application/zip]\n",
            "Saving to: ‘07_efficientnetb0_fine_tuned_101_classes_mixed_precision.zip’\n",
            "\n",
            "07_efficientnetb0_f 100%[===================>]  44.62M  66.2MB/s    in 0.7s    \n",
            "\n",
            "2023-07-26 12:40:11 (66.2 MB/s) - ‘07_efficientnetb0_fine_tuned_101_classes_mixed_precision.zip’ saved [46790356/46790356]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Unzip fine-tuned model\n",
        "!mkdir downloaded_fine_tuned_gs_model # create separate directory for fine-tuned model downloaded from Google Storage\n",
        "!unzip 07_efficientnetb0_fine_tuned_101_classes_mixed_precision -d downloaded_fine_tuned_gs_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pt65ZuJINrp7",
        "outputId": "2ad3710d-12bc-40b0-a22b-b9facd286ff1"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  07_efficientnetb0_fine_tuned_101_classes_mixed_precision.zip\n",
            "   creating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/\n",
            "   creating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/variables/\n",
            "  inflating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/variables/variables.data-00000-of-00001  \n",
            "  inflating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/variables/variables.index  \n",
            "  inflating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/saved_model.pb  \n",
            "   creating: downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision/assets/\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load in fine-tuned model from Google Storage and evaluate\n",
        "loaded_fine_tuned_gs_model = tf.keras.models.load_model(\"downloaded_fine_tuned_gs_model/07_efficientnetb0_fine_tuned_101_classes_mixed_precision\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IEnGw_pnN2WM",
        "outputId": "e7ac129e-5580-4071-9797-65309282b727"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Importing a function (__inference_block3b_expand_activation_layer_call_and_return_conditional_losses_443625) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_activation_layer_call_and_return_conditional_losses_412189) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_se_reduce_layer_call_and_return_conditional_losses_409120) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_expand_activation_layer_call_and_return_conditional_losses_446895) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_activation_layer_call_and_return_conditional_losses_442329) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_421687) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_activation_layer_call_and_return_conditional_losses_411851) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_se_reduce_layer_call_and_return_conditional_losses_446637) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_activation_layer_call_and_return_conditional_losses_447353) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_expand_activation_layer_call_and_return_conditional_losses_445426) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_se_reduce_layer_call_and_return_conditional_losses_442371) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_419945) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_se_reduce_layer_call_and_return_conditional_losses_442703) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_activation_layer_call_and_return_conditional_losses_445173) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_440876) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_se_reduce_layer_call_and_return_conditional_losses_447774) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_expand_activation_layer_call_and_return_conditional_losses_413771) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_se_reduce_layer_call_and_return_conditional_losses_413200) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_expand_activation_layer_call_and_return_conditional_losses_410834) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_se_reduce_layer_call_and_return_conditional_losses_411277) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_expand_activation_layer_call_and_return_conditional_losses_410210) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_se_reduce_layer_call_and_return_conditional_losses_445215) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_activation_layer_call_and_return_conditional_losses_446974) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_se_reduce_layer_call_and_return_conditional_losses_443414) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_expand_activation_layer_call_and_return_conditional_losses_413433) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_stem_activation_layer_call_and_return_conditional_losses_442250) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_activation_layer_call_and_return_conditional_losses_410890) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_se_reduce_layer_call_and_return_conditional_losses_413876) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_top_activation_layer_call_and_return_conditional_losses_414056) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_activation_layer_call_and_return_conditional_losses_409981) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_435905) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_activation_layer_call_and_return_conditional_losses_410266) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_expand_activation_layer_call_and_return_conditional_losses_409924) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_se_reduce_layer_call_and_return_conditional_losses_410939) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_activation_layer_call_and_return_conditional_losses_444794) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_se_reduce_layer_call_and_return_conditional_losses_412862) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_se_reduce_layer_call_and_return_conditional_losses_445926) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference__wrapped_model_403250) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_se_reduce_layer_call_and_return_conditional_losses_445547) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_expand_activation_layer_call_and_return_conditional_losses_447274) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_expand_activation_layer_call_and_return_conditional_losses_412471) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_se_reduce_layer_call_and_return_conditional_losses_412238) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_expand_activation_layer_call_and_return_conditional_losses_411510) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_activation_layer_call_and_return_conditional_losses_445884) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_se_reduce_layer_call_and_return_conditional_losses_411615) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_se_reduce_layer_call_and_return_conditional_losses_410315) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_expand_activation_layer_call_and_return_conditional_losses_444336) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_model_layer_call_and_return_conditional_losses_429172) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_se_reduce_layer_call_and_return_conditional_losses_409406) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_se_reduce_layer_call_and_return_conditional_losses_412577) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_434163) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_top_activation_layer_call_and_return_conditional_losses_447985) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_expand_activation_layer_call_and_return_conditional_losses_412757) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_expand_activation_layer_call_and_return_conditional_losses_446184) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_activation_layer_call_and_return_conditional_losses_411566) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_activation_layer_call_and_return_conditional_losses_444415) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_expand_activation_layer_call_and_return_conditional_losses_442914) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_expand_activation_layer_call_and_return_conditional_losses_409300) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_activation_layer_call_and_return_conditional_losses_443372) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_se_reduce_layer_call_and_return_conditional_losses_447395) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5a_expand_activation_layer_call_and_return_conditional_losses_445094) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_activation_layer_call_and_return_conditional_losses_413489) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_expand_activation_layer_call_and_return_conditional_losses_410548) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_activation_layer_call_and_return_conditional_losses_412813) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_activation_layer_call_and_return_conditional_losses_444083) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_se_reduce_layer_call_and_return_conditional_losses_444836) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_activation_layer_call_and_return_conditional_losses_447732) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_expand_activation_layer_call_and_return_conditional_losses_411795) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_expand_activation_layer_call_and_return_conditional_losses_446516) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_se_reduce_layer_call_and_return_conditional_losses_444125) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_expand_activation_layer_call_and_return_conditional_losses_444715) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_se_reduce_layer_call_and_return_conditional_losses_443035) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_activation_layer_call_and_return_conditional_losses_442661) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_model_layer_call_and_return_conditional_losses_430926) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_expand_activation_layer_call_and_return_conditional_losses_444004) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_expand_activation_layer_call_and_return_conditional_losses_413095) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_se_reduce_layer_call_and_return_conditional_losses_411900) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_se_reduce_layer_call_and_return_conditional_losses_447016) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_activation_layer_call_and_return_conditional_losses_412528) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_se_reduce_layer_call_and_return_conditional_losses_443746) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_activation_layer_call_and_return_conditional_losses_442993) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_expand_activation_layer_call_and_return_conditional_losses_443293) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_se_reduce_layer_call_and_return_conditional_losses_446305) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6b_activation_layer_call_and_return_conditional_losses_446595) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_activation_layer_call_and_return_conditional_losses_413827) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3b_activation_layer_call_and_return_conditional_losses_443704) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_activation_layer_call_and_return_conditional_losses_409357) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6c_activation_layer_call_and_return_conditional_losses_413151) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block3a_se_reduce_layer_call_and_return_conditional_losses_410030) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4b_se_reduce_layer_call_and_return_conditional_losses_444457) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block1a_activation_layer_call_and_return_conditional_losses_409071) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_expand_activation_layer_call_and_return_conditional_losses_445805) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_activation_layer_call_and_return_conditional_losses_411228) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2a_expand_activation_layer_call_and_return_conditional_losses_442582) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_se_reduce_layer_call_and_return_conditional_losses_410654) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block7a_expand_activation_layer_call_and_return_conditional_losses_447653) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6a_activation_layer_call_and_return_conditional_losses_446263) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block6d_se_reduce_layer_call_and_return_conditional_losses_413538) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5c_expand_activation_layer_call_and_return_conditional_losses_412133) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_se_reduce_layer_call_and_return_conditional_losses_409691) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_expand_activation_layer_call_and_return_conditional_losses_409586) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block5b_activation_layer_call_and_return_conditional_losses_445505) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4a_activation_layer_call_and_return_conditional_losses_410605) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_stem_activation_layer_call_and_return_conditional_losses_409015) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block2b_activation_layer_call_and_return_conditional_losses_409642) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_block4c_expand_activation_layer_call_and_return_conditional_losses_411172) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n",
            "WARNING:absl:Importing a function (__inference_efficientnetb0_layer_call_and_return_conditional_losses_439134) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Get a model summary (same model architecture as above)\n",
        "loaded_fine_tuned_gs_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7xPIQjK-OQd4",
        "outputId": "f2768927-656b-466e-c754-ed26304afb03"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_layer (InputLayer)    [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " efficientnetb0 (Functional)  (None, None, None, 1280)  4049571  \n",
            "                                                                 \n",
            " pooling_layer (GlobalAverag  (None, 1280)             0         \n",
            " ePooling2D)                                                     \n",
            "                                                                 \n",
            " dense (Dense)               (None, 101)               129381    \n",
            "                                                                 \n",
            " softmax_float32 (Activation  (None, 101)              0         \n",
            " )                                                               \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,178,952\n",
            "Trainable params: 4,136,929\n",
            "Non-trainable params: 42,023\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Note: Even if you're loading in the model from Google Storage, you will still need to load the test_data variable for this cell to work\n",
        "results_downloaded_fine_tuned_gs_model = loaded_fine_tuned_gs_model.evaluate(test_data)\n",
        "results_downloaded_fine_tuned_gs_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bfZPS_YQOVAV",
        "outputId": "397e7f6c-bdee-4fb9-c8a8-0c096d75549b"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "790/790 [==============================] - 70s 85ms/step - loss: 0.9072 - accuracy: 0.8015\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.9072173833847046, 0.8014653325080872]"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L0qbjlm2UlG-",
        "outputId": "6d6dfc56-b2cc-4b9f-be34-1421a7549590"
      },
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PrefetchDataset element_spec=(TensorSpec(shape=(None, 224, 224, 3), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# Get the class names\n",
        "\n",
        "class_names[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YzsyW77bPylg",
        "outputId": "7d4b41a0-51bf-4a12-e6fb-6a437bb70dd3"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['apple_pie',\n",
              " 'baby_back_ribs',\n",
              " 'baklava',\n",
              " 'beef_carpaccio',\n",
              " 'beef_tartare',\n",
              " 'beet_salad',\n",
              " 'beignets',\n",
              " 'bibimbap',\n",
              " 'bread_pudding',\n",
              " 'breakfast_burrito']"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "dWNdQNmHWBob"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": [],
      "mount_file_id": "1o1fHppceThjA8ZlZP6ApOAK4XnE90BhJ",
      "authorship_tag": "ABX9TyNKhWtcfP/zjcJv3S6Qvfj/",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "50aaf069e6794e998bd281010bc5043b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_166a908eec7547daa239c2b32a7afec2",
              "IPY_MODEL_ca8bf4ea67144dd0b58d8e79ec74edcf",
              "IPY_MODEL_c3932637bf0e4dcba83a4dbfdff30e24"
            ],
            "layout": "IPY_MODEL_68f5dbabcd5b408f8bb8f9bee92475c9"
          }
        },
        "166a908eec7547daa239c2b32a7afec2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_56e20ee6e1124be0b787c5a5a3eca037",
            "placeholder": "​",
            "style": "IPY_MODEL_e8aa3858be224cd0ac698a6ea5a489e9",
            "value": "Dl Completed...: 100%"
          }
        },
        "ca8bf4ea67144dd0b58d8e79ec74edcf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e1ac71bf85be420bb549bedd34a1ef4e",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_95285302099b438bad41af674feaec56",
            "value": 1
          }
        },
        "c3932637bf0e4dcba83a4dbfdff30e24": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b59b3e2f5ce6484abf08dabfc7500653",
            "placeholder": "​",
            "style": "IPY_MODEL_27a5885d2a3840c7a7e5d3cf41db53b5",
            "value": " 1/1 [08:54&lt;00:00, 190.11s/ url]"
          }
        },
        "68f5dbabcd5b408f8bb8f9bee92475c9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "56e20ee6e1124be0b787c5a5a3eca037": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e8aa3858be224cd0ac698a6ea5a489e9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e1ac71bf85be420bb549bedd34a1ef4e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "95285302099b438bad41af674feaec56": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "b59b3e2f5ce6484abf08dabfc7500653": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "27a5885d2a3840c7a7e5d3cf41db53b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fa729f84dbe749188f3db0994974ecc2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_564787e186ae4859adb4089fccda05c3",
              "IPY_MODEL_99b1edc4698b4e52bee727a10600f44b",
              "IPY_MODEL_1f4399bfa4584f018477a2580f9a938c"
            ],
            "layout": "IPY_MODEL_390243e5a48145f3a671f43ffe0ff1f1"
          }
        },
        "564787e186ae4859adb4089fccda05c3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b3b982f760a0458383228ec1a44035e5",
            "placeholder": "​",
            "style": "IPY_MODEL_bc29c4d9b55e45289225e5f62b2f5168",
            "value": "Dl Size...: 100%"
          }
        },
        "99b1edc4698b4e52bee727a10600f44b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c41e6c7a79cb4e58ad356afcbded7c37",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_aad3a4379f9c487e9e7b92fe99f9a2dd",
            "value": 1
          }
        },
        "1f4399bfa4584f018477a2580f9a938c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_427430e96fdf493dbe1a353e675e5106",
            "placeholder": "​",
            "style": "IPY_MODEL_add28e4fe0874e2482b8c350bfbc4706",
            "value": " 4764/4764 [08:54&lt;00:00, 27.81 MiB/s]"
          }
        },
        "390243e5a48145f3a671f43ffe0ff1f1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b3b982f760a0458383228ec1a44035e5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "bc29c4d9b55e45289225e5f62b2f5168": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c41e6c7a79cb4e58ad356afcbded7c37": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "aad3a4379f9c487e9e7b92fe99f9a2dd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "427430e96fdf493dbe1a353e675e5106": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "add28e4fe0874e2482b8c350bfbc4706": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "04cadefe013e4252ba5560db93f9c87d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_04222ce145d94494a8c4c0c23fa11d18",
              "IPY_MODEL_d02c36de95be4184a26540868d30b500",
              "IPY_MODEL_695f9c09c3784a66bb4d61079797fdfd"
            ],
            "layout": "IPY_MODEL_189dc7f8d9ad434ebeeb9c755fbb0219"
          }
        },
        "04222ce145d94494a8c4c0c23fa11d18": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_23418acc405d4ccbb03714dbd5e171e8",
            "placeholder": "​",
            "style": "IPY_MODEL_a552a0e8af5d48968432dbe7542f351f",
            "value": "Extraction completed...:  99%"
          }
        },
        "d02c36de95be4184a26540868d30b500": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e29be58af5de4cb5a649ca0aecb94e1a",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f386501578dc44788a8d62f325b31820",
            "value": 1
          }
        },
        "695f9c09c3784a66bb4d61079797fdfd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9a74fdf9e66c42c8ba9918e09ce6882c",
            "placeholder": "​",
            "style": "IPY_MODEL_8f88eb2ac7da41fca8e790dc3989bf3c",
            "value": " 101008/101008 [08:54&lt;00:00, 1020.44 file/s]"
          }
        },
        "189dc7f8d9ad434ebeeb9c755fbb0219": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "23418acc405d4ccbb03714dbd5e171e8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a552a0e8af5d48968432dbe7542f351f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e29be58af5de4cb5a649ca0aecb94e1a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "f386501578dc44788a8d62f325b31820": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9a74fdf9e66c42c8ba9918e09ce6882c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8f88eb2ac7da41fca8e790dc3989bf3c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cbef5539fbd44df29afcaae390008b4e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_70d9afda803f4e6988915edd92240e8b",
              "IPY_MODEL_b1a976b8e7d54497b11f351b0020c75c",
              "IPY_MODEL_435dca930eb646758e8ae99d3c2ac71c"
            ],
            "layout": "IPY_MODEL_047a4a46d39e4dbd910cc2e0c32608d2"
          }
        },
        "70d9afda803f4e6988915edd92240e8b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6ac4130692314ddb8acd016c0303d445",
            "placeholder": "​",
            "style": "IPY_MODEL_9b6174ca2b1a4ffdbc0e8ee78a3b9d59",
            "value": "Generating splits...: 100%"
          }
        },
        "b1a976b8e7d54497b11f351b0020c75c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6239d34ff5f04772a38796edd35e37b2",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_c4cf236757f94b5f8f414a506df1d2e0",
            "value": 2
          }
        },
        "435dca930eb646758e8ae99d3c2ac71c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_dd642f3bc07a4fd097ea5f39068fb408",
            "placeholder": "​",
            "style": "IPY_MODEL_deba622183c64e5399bf6416dcbd496a",
            "value": " 2/2 [03:53&lt;00:00, 107.41s/ splits]"
          }
        },
        "047a4a46d39e4dbd910cc2e0c32608d2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "6ac4130692314ddb8acd016c0303d445": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9b6174ca2b1a4ffdbc0e8ee78a3b9d59": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6239d34ff5f04772a38796edd35e37b2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c4cf236757f94b5f8f414a506df1d2e0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "dd642f3bc07a4fd097ea5f39068fb408": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "deba622183c64e5399bf6416dcbd496a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "30135294362b4c2e88151133fa003eee": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_c2ff63e0263c46589eb11df7d4938b1c",
              "IPY_MODEL_cbf6849683f24cf6acc73f6b8775b401",
              "IPY_MODEL_6523f9271ee6468b809cc6fd5a25e7d6"
            ],
            "layout": "IPY_MODEL_85149ff51ba74896ae15a2bdd1467034"
          }
        },
        "c2ff63e0263c46589eb11df7d4938b1c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1fe54ec969ed4db3866b0985e18621b0",
            "placeholder": "​",
            "style": "IPY_MODEL_4f0f2302b2d64abeae853a6f847e6473",
            "value": "Generating train examples...: 100%"
          }
        },
        "cbf6849683f24cf6acc73f6b8775b401": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8ef0e1356e4c4af781560e6097272500",
            "max": 75750,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7daeca23e1994c228d5d8ffef1ef3aab",
            "value": 75750
          }
        },
        "6523f9271ee6468b809cc6fd5a25e7d6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fca0d67882e548da92503f9cd6926b3d",
            "placeholder": "​",
            "style": "IPY_MODEL_8edbcf1aae4c429eaaf29739d9dc4c48",
            "value": " 75664/75750 [02:04&lt;00:00, 698.75 examples/s]"
          }
        },
        "85149ff51ba74896ae15a2bdd1467034": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "1fe54ec969ed4db3866b0985e18621b0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4f0f2302b2d64abeae853a6f847e6473": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8ef0e1356e4c4af781560e6097272500": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7daeca23e1994c228d5d8ffef1ef3aab": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "fca0d67882e548da92503f9cd6926b3d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8edbcf1aae4c429eaaf29739d9dc4c48": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e0d17fba2b8a4d6eb893d13fec852f03": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_93c3dbc7a118411d9974e4de6951b524",
              "IPY_MODEL_2a37b5317f584e9aab196497d4e81b04",
              "IPY_MODEL_fc9a3fc572a841b19b247b3b69807eb3"
            ],
            "layout": "IPY_MODEL_7c18ebb5471e4cdf92de0887aa238808"
          }
        },
        "93c3dbc7a118411d9974e4de6951b524": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8dedf6ee94c44ff481bfb38aeb406047",
            "placeholder": "​",
            "style": "IPY_MODEL_2317ad6ef3544ff18b2f622817a73667",
            "value": "Shuffling /root/tensorflow_datasets/food101/2.0.0.incompleteZ55PQ3/food101-train.tfrecord*...: 100%"
          }
        },
        "2a37b5317f584e9aab196497d4e81b04": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4029114fc4634900ad94d252337e25c0",
            "max": 75750,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_3b407e1d3776454bb54e9d12b720ff48",
            "value": 75750
          }
        },
        "fc9a3fc572a841b19b247b3b69807eb3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_334f0da01bab4a99abd5e5413491295f",
            "placeholder": "​",
            "style": "IPY_MODEL_122ea049380d43eaa2870813981891df",
            "value": " 75529/75750 [00:43&lt;00:00, 2268.45 examples/s]"
          }
        },
        "7c18ebb5471e4cdf92de0887aa238808": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "8dedf6ee94c44ff481bfb38aeb406047": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2317ad6ef3544ff18b2f622817a73667": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "4029114fc4634900ad94d252337e25c0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3b407e1d3776454bb54e9d12b720ff48": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "334f0da01bab4a99abd5e5413491295f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "122ea049380d43eaa2870813981891df": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ccdaedaf51604a42b132ca2170e1a132": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_666ebd2054c5439db6683716dcf0d5da",
              "IPY_MODEL_9f7d382e97fa46a98825a38bd51a780e",
              "IPY_MODEL_3c2a50d1a4d04a6b9719c64bc3dc954a"
            ],
            "layout": "IPY_MODEL_f492e05817de463cbe70fc9da385012e"
          }
        },
        "666ebd2054c5439db6683716dcf0d5da": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f3f31742f5b741b6a6323d5a52bc5ec4",
            "placeholder": "​",
            "style": "IPY_MODEL_175f6420ad844b4198e0b0dd97c6f53d",
            "value": "Generating validation examples...: 100%"
          }
        },
        "9f7d382e97fa46a98825a38bd51a780e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cc398dfa5a264d6481277661dff6b96e",
            "max": 25250,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_05ee99df3e154c15869f35b9be8e0c9b",
            "value": 25250
          }
        },
        "3c2a50d1a4d04a6b9719c64bc3dc954a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_db7a18c7b1264cb4ab5e3f605309de2e",
            "placeholder": "​",
            "style": "IPY_MODEL_d4f14d9e1c874240bbea3288df4cffac",
            "value": " 25244/25250 [00:59&lt;00:00, 480.31 examples/s]"
          }
        },
        "f492e05817de463cbe70fc9da385012e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "f3f31742f5b741b6a6323d5a52bc5ec4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "175f6420ad844b4198e0b0dd97c6f53d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cc398dfa5a264d6481277661dff6b96e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "05ee99df3e154c15869f35b9be8e0c9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "db7a18c7b1264cb4ab5e3f605309de2e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d4f14d9e1c874240bbea3288df4cffac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d2b879d7843d432e8609d7414def04b4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_888a2e1eac574f828f0d5c7eaf3bc088",
              "IPY_MODEL_ec5d92ef4de24f7fab17c91348a477b3",
              "IPY_MODEL_bef6d3220d244b4b94aaec8aee29579d"
            ],
            "layout": "IPY_MODEL_3c4d0a3c537e4793a13618b63aef8182"
          }
        },
        "888a2e1eac574f828f0d5c7eaf3bc088": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1987ef9f5e15490083dbbded700b8f88",
            "placeholder": "​",
            "style": "IPY_MODEL_8eaee9c1e19940e99c1ae094d167ccc6",
            "value": "Shuffling /root/tensorflow_datasets/food101/2.0.0.incompleteZ55PQ3/food101-validation.tfrecord*...:  99%"
          }
        },
        "ec5d92ef4de24f7fab17c91348a477b3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4b79d3cdc5f64df99a0b7ebeac674e1d",
            "max": 25250,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_71143da54b3c4b75908f9f3ae502c85b",
            "value": 25250
          }
        },
        "bef6d3220d244b4b94aaec8aee29579d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f70a7661959f44ceb0c88868a101cb51",
            "placeholder": "​",
            "style": "IPY_MODEL_3bba5f3ddd4c4035a87033870d6165e9",
            "value": " 24985/25250 [00:04&lt;00:00, 4617.67 examples/s]"
          }
        },
        "3c4d0a3c537e4793a13618b63aef8182": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": "hidden",
            "width": null
          }
        },
        "1987ef9f5e15490083dbbded700b8f88": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8eaee9c1e19940e99c1ae094d167ccc6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "4b79d3cdc5f64df99a0b7ebeac674e1d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "71143da54b3c4b75908f9f3ae502c85b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "f70a7661959f44ceb0c88868a101cb51": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3bba5f3ddd4c4035a87033870d6165e9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}