fish_classification .ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!apt install unzip\n",
        "\n",
        "!unzip -q \"/content/Dataset.zip\""
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ySE4nUKoDUvs",
        "outputId": "df5286fc-8c27-4654-f255-2d445ccaadf0"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Reading package lists... Done\n",
            "Building dependency tree... Done\n",
            "Reading state information... Done\n",
            "unzip is already the newest version (6.0-26ubuntu3.2).\n",
            "0 upgraded, 0 newly installed, 0 to remove and 45 not upgraded.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!apt install git\n",
        "!git clone \"https://github.com/rezapace/Machine-Learning-fish-classification\""
      ],
      "metadata": {
        "id": "Hw_YSo8yFk1e"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "66DuX3fdB3dd"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.applications import VGG16, ResNet50\n",
        "from tensorflow.keras.layers import GlobalMaxPooling2D, Dense\n",
        "from tensorflow.keras.models import Sequential, load_model\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\n",
        "from sklearn.metrics import confusion_matrix, classification_report"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def create_data_generators(train_dir, valid_dir, test_dir, target_size=(224, 224), batch_size=32, preprocessing_function=None):\n",
        "    train_datagen = ImageDataGenerator(\n",
        "        rotation_range=40,\n",
        "        horizontal_flip=True,\n",
        "        preprocessing_function=preprocessing_function\n",
        "    )\n",
        "    valid_datagen = ImageDataGenerator(preprocessing_function=preprocessing_function)\n",
        "    test_datagen = ImageDataGenerator()\n",
        "\n",
        "    train_generator = train_datagen.flow_from_directory(\n",
        "        train_dir, target_size=target_size, batch_size=batch_size,\n",
        "        class_mode='categorical', seed=42, shuffle=True\n",
        "    )\n",
        "    valid_generator = valid_datagen.flow_from_directory(\n",
        "        valid_dir, target_size=target_size, batch_size=batch_size,\n",
        "        class_mode='categorical', seed=42, shuffle=False\n",
        "    )\n",
        "    test_generator = test_datagen.flow_from_directory(\n",
        "        test_dir, target_size=target_size, batch_size=batch_size,\n",
        "        class_mode='categorical', seed=42, shuffle=False\n",
        "    )\n",
        "\n",
        "    return train_generator, valid_generator, test_generator\n",
        "\n",
        "# Definisikan direktori\n",
        "train_dir = \"/content/Dataset/Data_Train\"\n",
        "valid_dir = \"/content/Dataset/Data_Validasi\"\n",
        "test_dir = \"/content/Dataset/Data_Uji\"\n",
        "\n",
        "# Buat generator data\n",
        "vgg16_train, vgg16_valid, vgg16_test = create_data_generators(\n",
        "    train_dir, valid_dir, test_dir,\n",
        "    preprocessing_function=tf.keras.applications.vgg16.preprocess_input\n",
        ")\n",
        "resnet50_train, resnet50_valid, resnet50_test = create_data_generators(\n",
        "    train_dir, valid_dir, test_dir,\n",
        "    preprocessing_function=tf.keras.applications.resnet50.preprocess_input\n",
        ")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QEJcbcDBCI3i",
        "outputId": "ec3c0b53-3caa-447e-bdec-e98af8e516b1"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 340 images belonging to 2 classes.\n",
            "Found 86 images belonging to 2 classes.\n",
            "Found 40 images belonging to 2 classes.\n",
            "Found 340 images belonging to 2 classes.\n",
            "Found 86 images belonging to 2 classes.\n",
            "Found 40 images belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def create_model(base_model, num_classes=2):\n",
        "    model = Sequential([\n",
        "        base_model,\n",
        "        GlobalMaxPooling2D(),\n",
        "        Dense(128, activation=\"relu\"),\n",
        "        Dense(128, activation=\"relu\"),\n",
        "        Dense(num_classes, activation=\"softmax\")\n",
        "    ])\n",
        "    return model\n",
        "\n",
        "# Buat model VGG16\n",
        "base_vgg16 = VGG16(include_top=False, input_shape=(224, 224, 3), weights='imagenet')\n",
        "base_vgg16.trainable = False\n",
        "vgg16_model = create_model(base_vgg16)\n",
        "\n",
        "# Buat model ResNet50\n",
        "base_resnet50 = ResNet50(include_top=False, input_shape=(224, 224, 3), weights='imagenet')\n",
        "base_resnet50.trainable = False\n",
        "resnet50_model = create_model(base_resnet50)\n"
      ],
      "metadata": {
        "id": "reBjivotCMG0"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def train_model(model, train_data, valid_data, epochs=10, checkpoint_path=None):\n",
        "    model.compile(loss='categorical_crossentropy', optimizer='Adam', metrics=['accuracy'])\n",
        "\n",
        "    callbacks = []\n",
        "    if checkpoint_path:\n",
        "        checkpoint = ModelCheckpoint(checkpoint_path, save_best_only=True, monitor='val_loss', mode='min', verbose=1)\n",
        "        callbacks.append(checkpoint)\n",
        "\n",
        "    history = model.fit(\n",
        "        train_data,\n",
        "        epochs=epochs,\n",
        "        validation_data=valid_data,\n",
        "        callbacks=callbacks\n",
        "    )\n",
        "    return history\n",
        "\n",
        "# Latih model VGG16\n",
        "vgg16_history = train_model(vgg16_model, vgg16_train, vgg16_valid,\n",
        "                            checkpoint_path='/content/model_vgg16_GlobalMax_aug.h5')\n",
        "\n",
        "# Latih model ResNet50\n",
        "resnet50_history = train_model(resnet50_model, resnet50_train, resnet50_valid,\n",
        "                               checkpoint_path='/content/model_resnet50_GlobalMax_aug.h5')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d3Rv_njvCMxT",
        "outputId": "d8a9b121-f357-4cf0-d8ed-6b7e8926d064"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 14.2503 - accuracy: 0.5824\n",
            "Epoch 1: val_loss improved from inf to 3.82502, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 23s 1s/step - loss: 14.2503 - accuracy: 0.5824 - val_loss: 3.8250 - val_accuracy: 0.7442\n",
            "Epoch 2/10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "11/11 [==============================] - ETA: 0s - loss: 4.0840 - accuracy: 0.7647\n",
            "Epoch 2: val_loss improved from 3.82502 to 3.46653, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 594ms/step - loss: 4.0840 - accuracy: 0.7647 - val_loss: 3.4665 - val_accuracy: 0.7326\n",
            "Epoch 3/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 2.1785 - accuracy: 0.8324\n",
            "Epoch 3: val_loss improved from 3.46653 to 1.18621, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 497ms/step - loss: 2.1785 - accuracy: 0.8324 - val_loss: 1.1862 - val_accuracy: 0.8837\n",
            "Epoch 4/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.7962 - accuracy: 0.9000\n",
            "Epoch 4: val_loss improved from 1.18621 to 0.93973, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 7s 607ms/step - loss: 0.7962 - accuracy: 0.9000 - val_loss: 0.9397 - val_accuracy: 0.9419\n",
            "Epoch 5/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.7730 - accuracy: 0.9029\n",
            "Epoch 5: val_loss did not improve from 0.93973\n",
            "11/11 [==============================] - 5s 492ms/step - loss: 0.7730 - accuracy: 0.9029 - val_loss: 1.4554 - val_accuracy: 0.8721\n",
            "Epoch 6/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.8077 - accuracy: 0.8765\n",
            "Epoch 6: val_loss did not improve from 0.93973\n",
            "11/11 [==============================] - 6s 571ms/step - loss: 0.8077 - accuracy: 0.8765 - val_loss: 1.1661 - val_accuracy: 0.9186\n",
            "Epoch 7/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.7614 - accuracy: 0.9029\n",
            "Epoch 7: val_loss improved from 0.93973 to 0.83206, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 5s 488ms/step - loss: 0.7614 - accuracy: 0.9029 - val_loss: 0.8321 - val_accuracy: 0.9535\n",
            "Epoch 8/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.6093 - accuracy: 0.9265\n",
            "Epoch 8: val_loss did not improve from 0.83206\n",
            "11/11 [==============================] - 5s 478ms/step - loss: 0.6093 - accuracy: 0.9265 - val_loss: 0.8380 - val_accuracy: 0.9651\n",
            "Epoch 9/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.2749 - accuracy: 0.9412\n",
            "Epoch 9: val_loss did not improve from 0.83206\n",
            "11/11 [==============================] - 6s 579ms/step - loss: 0.2749 - accuracy: 0.9412 - val_loss: 0.9241 - val_accuracy: 0.9302\n",
            "Epoch 10/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.2556 - accuracy: 0.9559\n",
            "Epoch 10: val_loss improved from 0.83206 to 0.74478, saving model to /content/model_vgg16_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 528ms/step - loss: 0.2556 - accuracy: 0.9559 - val_loss: 0.7448 - val_accuracy: 0.9419\n",
            "Epoch 1/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 3.6303 - accuracy: 0.5794\n",
            "Epoch 1: val_loss improved from inf to 0.88801, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 16s 932ms/step - loss: 3.6303 - accuracy: 0.5794 - val_loss: 0.8880 - val_accuracy: 0.7093\n",
            "Epoch 2/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.6244 - accuracy: 0.7471\n",
            "Epoch 2: val_loss improved from 0.88801 to 0.21644, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 534ms/step - loss: 0.6244 - accuracy: 0.7471 - val_loss: 0.2164 - val_accuracy: 0.9302\n",
            "Epoch 3/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.3179 - accuracy: 0.8706\n",
            "Epoch 3: val_loss did not improve from 0.21644\n",
            "11/11 [==============================] - 6s 557ms/step - loss: 0.3179 - accuracy: 0.8706 - val_loss: 0.3243 - val_accuracy: 0.8605\n",
            "Epoch 4/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.3055 - accuracy: 0.8529\n",
            "Epoch 4: val_loss improved from 0.21644 to 0.13108, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 520ms/step - loss: 0.3055 - accuracy: 0.8529 - val_loss: 0.1311 - val_accuracy: 0.9419\n",
            "Epoch 5/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.1494 - accuracy: 0.9471\n",
            "Epoch 5: val_loss improved from 0.13108 to 0.09080, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 7s 619ms/step - loss: 0.1494 - accuracy: 0.9471 - val_loss: 0.0908 - val_accuracy: 0.9767\n",
            "Epoch 6/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.0749 - accuracy: 0.9735\n",
            "Epoch 6: val_loss improved from 0.09080 to 0.08557, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 7s 601ms/step - loss: 0.0749 - accuracy: 0.9735 - val_loss: 0.0856 - val_accuracy: 0.9651\n",
            "Epoch 7/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.0816 - accuracy: 0.9706\n",
            "Epoch 7: val_loss improved from 0.08557 to 0.08317, saving model to /content/model_resnet50_GlobalMax_aug.h5\n",
            "11/11 [==============================] - 6s 537ms/step - loss: 0.0816 - accuracy: 0.9706 - val_loss: 0.0832 - val_accuracy: 0.9651\n",
            "Epoch 8/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.0468 - accuracy: 0.9853\n",
            "Epoch 8: val_loss did not improve from 0.08317\n",
            "11/11 [==============================] - 6s 569ms/step - loss: 0.0468 - accuracy: 0.9853 - val_loss: 0.0922 - val_accuracy: 0.9767\n",
            "Epoch 9/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.0582 - accuracy: 0.9824\n",
            "Epoch 9: val_loss did not improve from 0.08317\n",
            "11/11 [==============================] - 5s 464ms/step - loss: 0.0582 - accuracy: 0.9824 - val_loss: 0.0883 - val_accuracy: 0.9651\n",
            "Epoch 10/10\n",
            "11/11 [==============================] - ETA: 0s - loss: 0.0352 - accuracy: 0.9941\n",
            "Epoch 10: val_loss did not improve from 0.08317\n",
            "11/11 [==============================] - 6s 580ms/step - loss: 0.0352 - accuracy: 0.9941 - val_loss: 0.0880 - val_accuracy: 0.9651\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_training_history(history, title):\n",
        "    plt.figure(figsize=(12, 4))\n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.plot(history.history['accuracy'], label='train')\n",
        "    plt.plot(history.history['val_accuracy'], label='validation')\n",
        "    plt.title(f'{title} - Accuracy')\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.legend()\n",
        "\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.plot(history.history['loss'], label='train')\n",
        "    plt.plot(history.history['val_loss'], label='validation')\n",
        "    plt.title(f'{title} - Loss')\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "    plt.tight_layout()\n",
        "    plt.show()\n",
        "\n",
        "plot_training_history(vgg16_history, 'VGG16')\n",
        "plot_training_history(resnet50_history, 'ResNet50')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 797
        },
        "id": "Q8XhBktPCPUe",
        "outputId": "6e555df9-7da0-474e-9f1f-2e687dfebc53"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def evaluate_model(model, test_data, class_labels):\n",
        "    predictions = model.predict(test_data)\n",
        "    predicted_classes = np.argmax(predictions, axis=1)\n",
        "    true_classes = test_data.classes\n",
        "\n",
        "    cm = confusion_matrix(true_classes, predicted_classes)\n",
        "    plt.figure(figsize=(8, 6))\n",
        "    sns.heatmap(cm, annot=True, fmt='d', xticklabels=class_labels, yticklabels=class_labels, cmap='Blues')\n",
        "    plt.xlabel('Predicted')\n",
        "    plt.ylabel('Actual')\n",
        "    plt.title('Confusion Matrix')\n",
        "    plt.show()\n",
        "\n",
        "    report = classification_report(true_classes, predicted_classes, target_names=class_labels)\n",
        "    print(report)\n",
        "\n",
        "class_labels = list(vgg16_test.class_indices.keys())\n",
        "\n",
        "print(\"VGG16 Model Evaluation:\")\n",
        "evaluate_model(vgg16_model, vgg16_test, class_labels)\n",
        "\n",
        "print(\"ResNet50 Model Evaluation:\")\n",
        "evaluate_model(resnet50_model, resnet50_test, class_labels)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "I08HhNXQCQki",
        "outputId": "76cd236d-7ca1-427c-8ba9-d3113e8c90a9"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "VGG16 Model Evaluation:\n",
            "2/2 [==============================] - 3s 3s/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "    cakalang       0.75      0.90      0.82        20\n",
            "     tongkol       0.88      0.70      0.78        20\n",
            "\n",
            "    accuracy                           0.80        40\n",
            "   macro avg       0.81      0.80      0.80        40\n",
            "weighted avg       0.81      0.80      0.80        40\n",
            "\n",
            "ResNet50 Model Evaluation:\n",
            "2/2 [==============================] - 2s 1s/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "    cakalang       0.86      0.90      0.88        20\n",
            "     tongkol       0.89      0.85      0.87        20\n",
            "\n",
            "    accuracy                           0.88        40\n",
            "   macro avg       0.88      0.88      0.87        40\n",
            "weighted avg       0.88      0.88      0.87        40\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "vgg16_model.save(\"/content/model_vgg16_GlobalMax_aug.h5\")\n",
        "resnet50_model.save(\"/content/model_resnet50_GlobalMax_aug.h5\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W0zohO5ECSCI",
        "outputId": "bccc8dc5-650f-4147-950e-623aeeb6ab25"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        }
      ]
    }
  ]
}