Model_1_on_Food_101_data.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyPw9Yiy8GKf94G4LCveqmMO",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/149189/Experiments_on_Food101_dataset/blob/main/Model_1_on_Food_101_data.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers, models\n",
        "import zipfile\n",
        "import os\n",
        "from pathlib import Path"
      ],
      "metadata": {
        "id": "t-skhv36s5zn"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def unzip_data(zip_path, extract_to=\".\"):\n",
        "    with zipfile.ZipFile(zip_path, \"r\") as zip_ref:\n",
        "        zip_ref.extractall(extract_to)"
      ],
      "metadata": {
        "id": "o_rb_IyFs7K_"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://storage.googleapis.com/ztm_tf_course/food_vision/10_food_classes_1_percent.zip\n",
        "unzip_data(\"10_food_classes_1_percent.zip\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "odhpjI-Ss9kG",
        "outputId": "c2ab8d7f-84a6-4a9a-cfeb-15d572bc9fc4"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-08-16 15:03:22--  https://storage.googleapis.com/ztm_tf_course/food_vision/10_food_classes_1_percent.zip\n",
            "Resolving storage.googleapis.com (storage.googleapis.com)... 74.125.137.207, 142.250.101.207, 142.251.2.207, ...\n",
            "Connecting to storage.googleapis.com (storage.googleapis.com)|74.125.137.207|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 133612354 (127M) [application/zip]\n",
            "Saving to: ‘10_food_classes_1_percent.zip’\n",
            "\n",
            "10_food_classes_1_p 100%[===================>] 127.42M   175MB/s    in 0.7s    \n",
            "\n",
            "2024-08-16 15:03:23 (175 MB/s) - ‘10_food_classes_1_percent.zip’ saved [133612354/133612354]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_dir_1_percent = Path(\"10_food_classes_1_percent/train\")\n",
        "test_dir = Path(\"10_food_classes_1_percent/test\")"
      ],
      "metadata": {
        "id": "e3lK8-wctHqn"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data_1_percent = tf.keras.utils.image_dataset_from_directory(\n",
        "    train_dir_1_percent,\n",
        "    image_size=(224, 224),\n",
        "    batch_size=32\n",
        ")\n",
        "\n",
        "test_data = tf.keras.utils.image_dataset_from_directory(\n",
        "    test_dir,\n",
        "    image_size=(224, 224),\n",
        "    batch_size=32\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ruZDQtavtKSW",
        "outputId": "e622e902-65e5-45e6-e0c4-e1e13cdbb63c"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 70 files belonging to 10 classes.\n",
            "Found 2500 files belonging to 10 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_augmentation = tf.keras.Sequential([\n",
        "    layers.RandomFlip(\"horizontal_and_vertical\"),\n",
        "    layers.RandomRotation(0.2),\n",
        "    layers.RandomZoom(0.2),\n",
        "    layers.RandomHeight(0.1),\n",
        "    layers.RandomWidth(0.1),\n",
        "], name=\"data_augmentation\")"
      ],
      "metadata": {
        "id": "sOqq3EDEtMaO"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "base_model = tf.keras.applications.EfficientNetB0(include_top=False)\n",
        "base_model.trainable = False"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tSsz6JbFtOiv",
        "outputId": "62cd8004-a200-40ac-de23-71a7755ad9a0"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/keras-applications/efficientnetb0_notop.h5\n",
            "\u001b[1m16705208/16705208\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = layers.Input(shape=(224, 224, 3), name=\"input_layer\")\n",
        "x = data_augmentation(inputs)\n",
        "x = base_model(x, training=False)\n",
        "x = layers.GlobalAveragePooling2D(name=\"global_average_pooling_layer\")(x)\n",
        "outputs = layers.Dense(10, activation=\"softmax\", name=\"output_layer\")(x)\n",
        "model_1 = models.Model(inputs, outputs)"
      ],
      "metadata": {
        "id": "ElsxU-uZtQfm"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile the model with sparse_categorical_crossentropy\n",
        "model_1.compile(\n",
        "    loss=\"sparse_categorical_crossentropy\",\n",
        "    optimizer=tf.keras.optimizers.Adam(),\n",
        "    metrics=[\"accuracy\"]\n",
        ")\n"
      ],
      "metadata": {
        "id": "t0SAQWzotSoH"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=\"logs/feature_extraction\")\n",
        "checkpoint_callback = tf.keras.callbacks.ModelCheckpoint(\"best_model.keras\", save_best_only=True)"
      ],
      "metadata": {
        "id": "rA-lbDpvtUV_"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = model_1.fit(\n",
        "    train_data_1_percent,\n",
        "    epochs=5,\n",
        "    validation_data=test_data,\n",
        "    validation_steps=int(0.25 * len(test_data)),\n",
        "    callbacks=[tensorboard_callback, checkpoint_callback]\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TeFbY1pgtWa_",
        "outputId": "cc6a228e-80cb-4769-f5b3-c0573d4fc879"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m26s\u001b[0m 4s/step - accuracy: 0.1411 - loss: 2.2836 - val_accuracy: 0.1250 - val_loss: 2.2849\n",
            "Epoch 2/5\n",
            "\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 1s/step - accuracy: 0.2191 - loss: 2.0998 - val_accuracy: 0.1842 - val_loss: 2.1573\n",
            "Epoch 3/5\n",
            "\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 970ms/step - accuracy: 0.3686 - loss: 1.9278 - val_accuracy: 0.2533 - val_loss: 2.0672\n",
            "Epoch 4/5\n",
            "\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 914ms/step - accuracy: 0.4187 - loss: 1.8614 - val_accuracy: 0.3438 - val_loss: 1.9643\n",
            "Epoch 5/5\n",
            "\u001b[1m2/3\u001b[0m \u001b[32m━━━━━━━━━━━━━\u001b[0m\u001b[37m━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 95ms/step - accuracy: 0.5547 - loss: 1.6852 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/lib/python3.10/contextlib.py:153: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n",
            "  self.gen.throw(typ, value, traceback)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2s/step - accuracy: 0.5702 - loss: 1.6677 - val_accuracy: 0.3382 - val_loss: 1.8993\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "results_feature_extraction_model = model_1.evaluate(test_data)\n",
        "results_feature_extraction_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oSecE43FtaAX",
        "outputId": "11b9df7c-f7c1-4d3d-881d-606bf04995cf"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m79/79\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 77ms/step - accuracy: 0.3521 - loss: 1.8930\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.8965716361999512, 0.3587999939918518]"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/helper_functions.py"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yzMmnUOfuDYg",
        "outputId": "4bdea11f-7c4b-4e3c-adc5-0c9e194b6799"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-08-16 15:05:01--  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",
            "2024-08-16 15:05:02 (93.1 MB/s) - ‘helper_functions.py’ saved [10246/10246]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from helper_functions import create_tensorboard_callback, plot_loss_curves, unzip_data, compare_historys, walk_through_dir"
      ],
      "metadata": {
        "id": "KxfqRiwWuHpv"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_loss_curves(history)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 927
        },
        "id": "hsP6iyUluLnv",
        "outputId": "9b115786-544a-47f6-ee33-a1781f779e80"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "px2YBw5wuOHI"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}